Citation
2013: Proceedings of the 16th International Congress of Speleology, Czech Republic, Brno, July 21-28, 2013 Volume 3

Material Information

Title:
2013: Proceedings of the 16th International Congress of Speleology, Czech Republic, Brno, July 21-28, 2013 Volume 3
Alternate Title:
International Congress of Speleology
Alternate Title:
ICS Proceedings
Publication Date:
Language:
English

Subjects

Genre:
Conference Proceeding
serial ( sobekcm )

Notes

General Note:
Produced by the Organizing Committee of the 16th International Congress of Speleology. Published by the Czech Speleological Society and the SPELEO2013 and in the co-operation with the International Union of SpeleologyEdited by Michal Filippi, Pavel Bosák Contents Preface Session: Karst And Caves In Carbonate Rocks, Salt And GypsumKarst Hydrogeology Of The Haney Limestone, South-Central Kentucky / Sarah M. Arpin, Christopher G. Groves -- An Introduction To Kalahroud Cave, North Of Esfahan, Iran / Shirin Bahadorinia, Sayed Hassen Hejazi, Alireza Nadimi, Derek C. Ford -- Cave System Of Kita GaćesÌŒina - DrazÌŒenova Puhaljka. The Longest Cave In The Dinaric Karst / Teo BarisÌŒić, Darko BaksÌŒić, Dalibor Paar -- Caves Of El Peñon, Cordillera Oriental, Colombia / Martin Bochud, Roman Hapka, Jean-Marc Jutzet -- Recent Survey And Exploration In Lechuguilla Cave, New Mexico, Usa / Peter Bosted, John Lyle-- -- Pleistocene Sea Level Changes As Revealed By Flank Margin Caves In Telogenetic Limestones In Sicily And Sardinia (Italy) / Ilenia Maria D'angeli, Jo De Waele, Rosario Ruggieri, Laura Sanna -- Some Scientific Features Of The Puerto Princesa Underground River: One Of The New 7 Wonders Of Nature (Palawan, Philippines) / Antonio De Vivo, Leonardo Piccini, Paolo Forti, Giovanni Badino -- Subterranean Glacial Spillways: An Example From The Karst Of South Wales, U.K. / Andrew R. Farrant, Michael J. Simms, Steven R. Noble -- Project Namak: Some Of The Most Spectacular Findings In The Iranian Salt Karst / Michal Filippi, JirÌŒí Bruthans, OndrÌŒej Jager, Mohammad Zare, Naser Asadi -- Karst Development In The Glaciated And Permafrost Regions Of The Northwest Territories, Canada / Derek C. Ford -- Little Limestone Lake: A Beautiful Marl Lake In Manitoba, Canada / Derek Ford -- Caves And Karst Hydrogeology Of Jerusalem, Israel / Amos Frumkin -- Caves Under Dubrovnik Airport In Croatia / Mladen GarasÌŒić -- Some Information About The Deepest Caves Known In Croatian Karst Area / Mladen GarasÌŒić, Davor GarasÌŒić -- Hypogene Speleogenesis And Speleothems Of Sima De La Higuera Cave (Murcia, South-Eastern Spain) / Fernando Gázquez, José-María Calaforra . -- The Application Of Gis Methods In Morphometrical Analysis Of Dolines On Limestone And Dolomite Bedrock / Petra GostincÌŒar -- Origin Of Atypical Calcite Speleothems Filling Fissures In Sandstones / MichaÅ‚ Gradziński, Marek Duliński, Helena Hercman, Andrzej Górny, StanisÅ‚aw Przybyszowski -- Uplift Evidence From Karst Morphology: Preliminary Evidence From Blambangan Peninsula Karst, Indonesia / Eko Haryono-- Valley Incision In The Nízke Tatry Mts. (Slovakia) Estimated Based On Paleomagnetic And Radiometric Cave Sediment Datings / Jaroslav Kadlec, Pavol Bella, Kristýna CÌŒízÌŒková, Darryl E. Granger, Helena Hercman, Peter Holúbek, Martin Chadima, Monika OrvosÌŒová, Petr Pruner, Petr Schnabl, Stanislav SÌŒlechta -- Magnetic Fabric And Mineralogy Of Cave Deposits In Botovskaya Cave (Eastern Siberia, Russian Federation) / Jaroslav Kadlec, Helena Herman, Martin Chadima, Lenka Lisá, Hedi Oberhänsli, Alexandr Osintsev -- Case Studies Of Fluorescent Groundwater Tracing In Recent Cave Research / Benjamin V. Miller, Chris Groves, Jason S. Polk, Robert N. Lerch -- The Jaj Plateau (Lebanon): Typical High Altitude Mediterranean Karst / Fadi H. Nader, Hughes Badaoui, Marc Metni, Chadi Chaker, Habib Helou, Johnny Tawk -- A Conceptual Model Of Speleogenesis In Greece / Christos Pennos, Stein-Erik Lauritzen -- Complex Epikarst Hydrologeology And Contaminant Transport In A South-Central Kentucky Karst Landscape / Jason S. Polk, Sean Vanderhoff, Chris Groves, Benjamin Miller, Carl Bolster -- Karst Hydrogeological Observations In Cao Bang Province (Vietnam): The Tra Linh-Thang Hen Lake Area / Gheorghe M. Ponta, Bogdan P. Onac, Nyguen Xuan Nam -- Incidences Of The Tectonics In The Karstification Of Chalk Limestones In The Western Paris Basin: Example From The Petites Dales Cave (Saint Martin Aux Buneaux, France) / Joël Rodet, Kun Ma, Jean-Pierre Viard -- Ceiling Channel And Input Karst. Example Of The Petites Dales Cave, Normandy, France / Joël Rodet, Laurent Magne, Jean-Pierre Viard -- Gulls, Gull-Caves And Cambering In The Southern Cotswold Hills, England / Charles Self, Andrew Farrant -- The Role Of Fold-And-Thrust Structure In The Large Shafts And Chambers Development: Case Study Of The Polish Tatra Mts. / Jacek SzczygieÅ‚ -- Hypogenic Caves Of Sicily (Southern Italy) / Marco Vattano, Philippe Audra, Fabrizio Benvenuto, Jean-Yves Bigot, Jo De Waele, Ermanno Galli, Giuliana Madonia, Jean-Claude Nobécourt -- Sediments At The Münnich Passage In The Baradla Cave (Hungary): Mineralogical And Petrological Study/ Gábor Vid, István Berényi Üveges, Orsolya Viktorik, Tibor Németh, Zsolt BendoÌ‹,Sándor Józsa, Judit Berényi Üveges -- Endokarsts And Cryptokarsts In Cretaceous Coarse And Highly Porous Chalk At The Belgian-Dutch Border / Luc Willems, Joël Rodet -- The Tupper Glacier Sink - Raspberry Rising Cave System, Glacier National Park, Canada: A Remarkable Example In Stripe Karst / Charles Yonge, Nicholaus Vieira, Adam Walker -- Cave And Karst Prospection In Ras Al-Khaimah Mountains, Northern United Arab Emirate / Nadja Zupan Hajna, Asma Al Farraj Al Ketbi, Franci GabrovsÌŒek, Metka PetricÌŒ, Tadej Slabe, Martin Knez, Janez Mulec -- Influence Of The Pleistocene Glaciations On Karst Development In The Dinarides - Examples From Velebit Mt. (Croatia) / Neven BocÌŒić, Sanja Faivre, Marijan KovacÌŒić, Nada HorvatincÌŒić -- Contribution Of The Karstic Environment On The Origin Of The Collophanites Of Itataia Uranium-Phosphorus Deposit, Borborema Province, Brazil / José Adilson Dias Cavalcanti, Cesar Ulisses Vieira Veríssimo, Maria Dulcinea M.R. Bessa, Clovis Vaz Parente -- Na Javorce Cave - A New Discovery In The Bohemian Karst (Czech Republic): Unique Example Of Relationships Between Hydrothermal And Common Karstification / JirÌŒí Dragoun, Karel ZÌŒák, JirÌŒí Vejlupek, Michal Filippi, JirÌŒí Novotný, Petr DobesÌŒ -- Field Measurements Of Gypsum Denudation Rate In Kulogorskaya Cave System / Nikolay Franz, Sergey Sorokin, Alexandra Alexeeva, Irina Inshina, Olga Novysh, Anton Kazak -- Geology And Structure Of Parian Cave, Isfahan, Iran / Ghassem Ghaderi, Leila Karimi -- Preliminary Studies On Corrosive Forms And Clastic Deposits In The Śnieżna Cave (The Tatra Mts., Poland) / Ditta Kicińska -- Geochemical And Stable Isotope Characterization Of Drip Water From The Postojna Cave, Slovenia / Magda Mandić, Andrej Mihevc, Albrecht Leis, Ines Krajcar Bronić -- The Skalistý Potok Cave In The Relationship To The Relief Of The South Part Of Jasovská Plateau (Slovak Karst) After 25 Years Of Research / Alena Petrvalská, Zdenko Hochmuth -- Can Conduit Volumes Obtained From Artificial Tracer Tests Be Trusted? / Anna VojteÌŒchová, JirÌŒí Bruthans, OndrÌŒej Jäger, FrantisÌŒek KrejcÌŒa -- The Tectonic Control Of An Underground River Network, Agia Triada Cave (Karystos, Greece) / Emmanuel Vassilakis, Kyriaki Papadopoulou-Vrynioti. Session: Karst And Caves In Other Rocks, PseudokarstCave Formation Initiated By Dissolution Of Carbonate Cement In Quartzose Sandstones / JirÌŒí AdamovicÌŒ, Radek MikulásÌŒ, TomásÌŒ Navrátil, Jan Mertlík -- Arenitic Caves In Venezuelan Tepuis: What Do They Say About Tepuis Themselves? / Roman Aubrecht, TomásÌŒ Lánczos, Ján Schlögl, LukásÌŒ VlcÌŒek, Branislav SÌŒmída -- Infrared Thermographic Survey Of Pseudokarst Sites In The Fysch Belt Of Outer West Carpathians (Czech Republic) / Ivo BaronÌŒ, David BecÌŒkovský, Lumír MícÌŒa -- Kahuenaha Nui (Hawaii): A Cave Developed In Four Different Lava Flows / Ingo Bauer, Stephan Kempe, Peter Bosted -- Genetic Types Of Non-Solution Caves / Pavel Bella, Ludovít Gaál -- The Keokeo Lava Tube System In Hawaii / Peter Bosted, Tomislav Gracanin, Veda Hackell, Ann Bosted, Ingo Bauer, Stephan Kempe -- Origin Of "Rock Cities", Pillars And Cleft-Conduits In Kaolinite-Bonded Sandstone: New Insight From Study In Sandstone Quarry Where Landforms Recently Evolve / JirÌŒí Bruthans, Jan Soukup, Jana Schweigstillová, Jana Vaculíková, Daniel Smutek, Alan L. Mayo, LukásÌŒ Falteisek -- Origins Of The Karstic Sediment Filling Of The Northwestern Paris Basin Caves / S. Chédeville, J. Rodet, B. Laignel, D. Todisco, E. Dupuis, G. Girot, G. Hanin -- Sandstone Caves In The Sydney Basin: A Review Of Their Cultural And Natural Heritage / John R Dunkley -- Karst Evolution In Sandstone: The Chapada Dos Guimarães Site, Brazil / Rubens Hardt, Joël Rodet, Sergio Dos Anjos Ferreira Pinto -- Three Types Of Crevice-Type Caves In The Area Of Czech Flysch Carpathians / Jan Lenart, TomásÌŒ Pánek -- Exhumation Of Palaeokarst By Vadose Weathering: A Telling Example From Eastern Australia / R. Armstrong L. Osborne -- An Introduction To Sri Lankan Gneiss And Granite Caves / R. Armstrong L. Osborne, Pathmakumara Jayasingha, Wasantha S. Weliange -- Endo- And Exo-Karstic Features In Banded Iron Formation And Canga, Serra Da Piedade, Quadrilátero Ferrífero (Brazil) / Manuela CorreÌ‚a Pereira, Joel Georges Marie Andre Rodet, André Augustos Rodrigues Salgado -- Englacial Caves Of Glaciar Perito Moreno And Glaciar Ameghino, Patagonia (Argentina) / Leonardo Puccini, Marco Mecchia -- Speleogenesis And Speleothems Of The Guacamaya Cave, Auyan Tepui, Venezuela / Francesco Sauro, Joyce Lundberg, Jo De Waele, Nicola Tisato. Ermanno Gall -- Gobholo Cave: A Long Granite Cave In Swaziland (Southern Africa) / Manuela Scheuerer, Johannes E. K. Lundberg, Rabbe Sjöberg -- Some Significant Caves Found In The Non-Karstic Rocks Of Hungary / István Eszterhás, George Szentes -- Types Of Non-Karst Caves In Polish Outer Carpathians - Historical Review And Perspectives / Jan Urban, WÅ‚odzimierz Margielewski -- Caves In Sandstone Deposits Of The Southern Italian Apennines / Gianni Campanella, Mario Parise, Angela Rizzi, Mariangela Sammarco, Antonio Trocino -- Karst Developed In Siliciclastic Rocks At Serra Do Espinhaço Meridional, Minas Gerais (Brazil) / Alessandra Mendes Carvalho Vasconcelos, Fernanda Cristina Rodrigues De Souza, Joel Rodet, Cristiane Valéria De Oliveira, André Augusto Rodrigues Salgado Session: SpeleogenesisLag And Transfer Time Inferred From Melting Cycles Record In The Coulomp Karst Spring (Alpes De Hauteprovence, France) / Philippe Audra, Jean-Claude Nobécourt -- Dissolution Rates, Carbon Dioxide Dynamics, And Geomorphological Feedbacks In Open Channel Cave Streams / Matthew D. Covington, Mitja PrelovsÌŒek, Franci GabrovsÌŒek -- A Methodology To Estimate The Age Of Caves In Northern Latitudes, Using Toerfjellhola In Norway As An Example / Trevor Faulkner -- How Do Aperture Sizes In Limestone Vary For The Onsets Of Turbulent Flow And First-Order Dissolution Kinetics? / Trevor Faulkner -- Investigations Into The Potential For Hypogene Speleogenesis In The Cumberland Plateau Of Southeast Kentucky, U.S.A. / Lee J. Florea -- The Role Of Base Level Incision And Transient Recharge In Vertical Organisation Of Karst Network / Franci GabrovsÌŒek, Philipp Häuselmann, Philippe Audra -- Hypogenic Speleogenesis In The Crimean Fore-Mountains (The Black Sea Region, South Ukraine) And Its Role In The Regional Geomorphology / Alexander B. Klimchouk, Gennadiy N. Amelichev, Elizaveta I. Tymokhina, Sergiy V. Tokarev -- Paragenesis: The "Royal Mark" Of Subglacial Speleogenesis / Stein-Erik Lauritzen -- Glacier Ice-Contact Speleogenesis / Stein-Erik Lauritzen, Rannveig Øvrevik Skoglund -- Groundwater / Surface Water Mixing In The Underground Hydrologic System Of The Demänovská Dolina Valley (Slovakia) / Peter Malík, Dagmar Haviarová, Jaromír SÌŒvasta, MilosÌŒ Gregor, Anton Auxt -- Karst Processes And Carbon Flux In The Frasassi Caves Italy / Marco Menichetti -- Miocene - Pliocene Age Of Cave SnezÌŒna Jama Na Raduhi, Southern Alps, Slovenia / Andrej Mihevc, Ivan HorácÌŒek, Petr Pruner, Nadja Zupan Hajna, Stanislav CÌŒermák, Jan Wagner, Pavel Bosák -- Sedimentary Study And U-Th Datations Contribution To The Morphodynamic Reconstitution Of The Junction Chamber (Kassarat Cave, Nabay, Lebanon): A Geomorphological Approach For Palaeohydrological Records Analysis / C. Nehme, J.-J. Delannoy, S. Jaillet, J. Adjizian-Gerard, John Hellstrom, T. Comaty, M. Arzouni, P. Matta -- A Genetic Classification Of Caves In Lower Austria / Pauline Oberender, Lukas Plan -- Effects Of Co2-Depleted Vadose Seepage In Karst /Arthur N. Palmer, Margaret V. Palmer -- Detailed Morphologic Analysis Of Palaeotraun Gallery Using A Terrestrial Laser Scan (Dachstein-Mammuthöhle, Upper Austria) / Lukas Plan, Andreas Roncat, Gregor Marx -- Influence Of The Plio-Pleistocene Tectonics On The Evolution Of The Purgatorio Polje (North-Western Sicily) Rosario Ruggieri, Giuseppe Napoli , Pietro Renda -- Speleogenesis Of A 3D-Maze Cave (Hermannshöhle, Lower Austria) / Andrea Schober, Lukas Plan, Denis Scholz, Christoph Spötl -- Evolution Of Cave Provalata: Remarkable Example Of Hypogenic Thermal Carbonic And Sulfuric Acid Speleogenesis / Marjan Temovski, Philippe Audra -- Flank Margin Caves On A Passive Continental Margin: Naracoorte And Other Southern Australian Examples / Susan White -- Tritium And H, O And C Stable Isotopes As A Tool For Tracking Of Water Circulation In The Niedźwiedzia Cave System (Sudetes, Poland) / MichaÅ‚ Gąsiorowski, Helena Hercman. Session: Cave MineralogyHypogene Sulfuric Acid Speleogenesis And Rare Sulfate Minerals (Fibroferrite, Jarosite Subgroup) Baume Galiniere Cave (Alpes-De-Haute-Provence, France) / Philippe Audra, Fernando Gázquez, Fernando Rull, Jean-Yves Bigot, Hubert Camus -- Rare Sulfates (Mirabilite, Eugsterite) In The Dry Microclimate Of Chamois Cave (Alpes-De-Haute-Provence, France) / Philippe Audra, Jean-Claude Nobécourt -- Mineralogy And Speleogenesis Of The Corona 'E Sa Craba Quartzite Cave (Southwest Sardinia) / Eleonora Baldoni, Jo De Waele, Ermanno Galli, Mauro Messina, Bogdan P. Onac, Laura Sanna, Francesco Sauro, Mauro Villani -- Genesis And Evolution Of Calcite Bubbles In Gypsum Caves / Massimo Ercolani, Katia Poletti, Paolo Forti -- Strontianite From Serizjan Cave (Iran): A New Occurrence Of This Rare Cave Mineral With A Discusion On Its Genesis / José Maria Calaforra, Jo De Waele, Paolo Forti, Ermanno Galli, Masoud Ebadi -- Skeleton Crystals Of Cryogenic Gypsum From Kungur Ice Cave, Ural Mountains, Russia / Olga Kadebskaya, Ilya Tchaikovsky -- Allophane Moonmilk From A Granite-Gneiss Cave In Norway: A Biogenic Intermediate In Deep Weatehring? / Stein-Erik Lauritzen -- Calcite Speleothems In Pseudokarst Tjuv-Antes Grotta, Northern Sweden / Johannes E. K. Lundberg, Magnus Ivarsson, Therese Sallstedt, Rabbe Sjöberg, Juan Ramón Vidal Romaní, Love Dalén, Stina Höglund -- Dating Diagenised Aragonite Speleothems / Rebeca Martín-García, Ana M. Alonso-Zarza, Silvia Frisia -- The Hypogene Origin Of Diana Cave (Romania) And Its Sulfuric Acid Weathering Environment / Bogdan P. Onac, Cristina M. Pușcaș, Herta S. Effenberger, Ioan Povară, Jonathan G. Wynn -- The Life Cycle Of Speleothems, With Possible Implications For Radiometric Dating / Charles Self -- Speleothems In Cavities Developed In Magmatic Rocks / Juan Ramón Vidal-Romaní, Jorge Sanjurjo-Sánchez, Marcos Vaqueiro-Rodríguez, Laura González-López, María José López-Galindo -- Evolution Of Guano Under Different Environmental Conditions: A Mineralogical Approach / Alexandra M. Giurgiu, Bogdan P. Onac, Tudor Tămaş, Joan J. Fornós -- A Contribution To The Mineralogy Of Monte Guisi Cave (Sw Sardinia, Italy) / Cristian Moldovan, Monica I. Călugăr, Bogdan P. Onac, Jo De Waele, Angelo Naseddu, Joan J. Fornós -- New Localities Of Coarsely Crystalline Cryogenic Cave Carbonates In Slovakia / Monika OrvosÌŒová, LukásÌŒ VlcÌŒek, Karel ZÌŒák
Restriction:
Open Access - Permission by Publisher
General Note:
See Extended description for more information.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
K26-00116 ( USFLDC DOI )
k26.116 ( USFLDC Handle )
16321 ( karstportal - original NodeID )
978-80-87857-09-0 ( ISBN )

USFLDC Membership

Aggregations:
Karst Information Portal

Postcard Information

Format:
Serial

Downloads

This item has the following downloads:


Full Text
Description
Produced by the Organizing Committee of the 16th
International Congress of Speleology. Published by the Czech
Speleological Society and the SPELEO2013 and in the
co-operation with the International Union of
SpeleologyEdited by Michal Filippi, Pavel Bosak
Contents
Preface
Session: Karst And Caves In Carbonate Rocks, Salt And
GypsumKarst Hydrogeology Of The Haney Limestone,
South-Central Kentucky / Sarah M. Arpin, Christopher G. Groves
--
An Introduction To Kalahroud Cave, North Of Esfahan, Iran
/ Shirin Bahadorinia, Sayed Hassen Hejazi, Alireza Nadimi,
Derek C. Ford --
Cave System Of Kita Gacesina Drazenova Puhaljka. The
Longest Cave In The Dinaric Karst / Teo Barisic, Darko
Baksic, Dalibor Paar --
Caves Of El Penon, Cordillera Oriental, Colombia /
Martin Bochud, Roman Hapka, Jean-Marc Jutzet --
Recent Survey And Exploration In Lechuguilla Cave, New
Mexico, Usa / Peter Bosted, John Lyle--
--
Pleistocene Sea Level Changes As Revealed By Flank Margin
Caves In Telogenetic Limestones In Sicily And Sardinia (Italy)
/ Ilenia Maria D'angeli, Jo De Waele, Rosario Ruggieri, Laura
Sanna --
Some Scientific Features Of The Puerto Princesa
Underground River: One Of The New 7 Wonders Of Nature (Palawan,
Philippines) / Antonio De Vivo, Leonardo Piccini, Paolo Forti,
Giovanni Badino --
Subterranean Glacial Spillways: An Example From The Karst
Of South Wales, U.K. / Andrew R. Farrant, Michael J. Simms,
Steven R. Noble --
Project Namak: Some Of The Most Spectacular Findings In
The Iranian Salt Karst / Michal Filippi, Jiri Bruthans,
Ondrej Jager, Mohammad Zare, Naser Asadi --
Karst Development In The Glaciated And Permafrost Regions
Of The Northwest Territories, Canada / Derek C. Ford --
Little Limestone Lake: A Beautiful Marl Lake In Manitoba,
Canada / Derek Ford --
Caves And Karst Hydrogeology Of Jerusalem, Israel / Amos
Frumkin --
Caves Under Dubrovnik Airport In Croatia / Mladen
Garasic --
Some Information About The Deepest Caves Known In
Croatian Karst Area / Mladen Garasic, Davor Garasic --
Hypogene Speleogenesis And Speleothems Of Sima De La
Higuera Cave (Murcia, South-Eastern Spain) / Fernando Gazquez,
Jose-Maria Calaforra --
The Application Of Gis Methods In Morphometrical Analysis
Of Dolines On Limestone And Dolomite Bedrock / Petra Gostincar
--
Origin Of Atypical Calcite Speleothems Filling Fissures
In Sandstones / Micha Gradzinski, Marek Dulinski, Helena
Hercman, Andrzej Gorny, Stanisaw Przybyszowski --
Uplift Evidence From Karst Morphology: Preliminary
Evidence From Blambangan Peninsula Karst, Indonesia / Eko
Haryono--
Valley Incision In The Nizke Tatry Mts. (Slovakia)
Estimated Based On Paleomagnetic And Radiometric Cave Sediment
Datings / Jaroslav Kadlec, Pavol Bella, Kristyna Cizkova,
Darryl E. Granger, Helena Hercman, Peter Holubek, Martin
Chadima, Monika Orvosova, Petr Pruner, Petr Schnabl,
Stanislav Slechta --
Magnetic Fabric And Mineralogy Of Cave Deposits In
Botovskaya Cave (Eastern Siberia, Russian Federation) /
Jaroslav Kadlec, Helena Herman, Martin Chadima, Lenka Lisa,
Hedi Oberhansli, Alexandr Osintsev --
Case Studies Of Fluorescent Groundwater Tracing In Recent
Cave Research / Benjamin V. Miller, Chris Groves, Jason S.
Polk, Robert N. Lerch --
The Jaj Plateau (Lebanon): Typical High Altitude
Mediterranean Karst / Fadi H. Nader, Hughes Badaoui, Marc
Metni, Chadi Chaker, Habib Helou, Johnny Tawk --
A Conceptual Model Of Speleogenesis In Greece / Christos
Pennos, Stein-Erik Lauritzen --
Complex Epikarst Hydrologeology And Contaminant Transport
In A South-Central Kentucky Karst Landscape / Jason S. Polk,
Sean Vanderhoff, Chris Groves, Benjamin Miller, Carl Bolster --
Karst Hydrogeological Observations In Cao Bang Province
(Vietnam): The Tra Linh-Thang Hen Lake Area / Gheorghe M.
Ponta, Bogdan P. Onac, Nyguen Xuan Nam --
Incidences Of The Tectonics In The Karstification Of
Chalk Limestones In The Western Paris Basin: Example From The
Petites Dales Cave (Saint Martin Aux Buneaux, France) / Joel
Rodet, Kun Ma, Jean-Pierre Viard --
Ceiling Channel And Input Karst. Example Of The Petites
Dales Cave, Normandy, France / Joel Rodet, Laurent Magne,
Jean-Pierre Viard --
Gulls, Gull-Caves And Cambering In The Southern Cotswold
Hills, England / Charles Self, Andrew Farrant --
The Role Of Fold-And-Thrust Structure In The Large Shafts
And Chambers Development: Case Study Of The Polish Tatra Mts. /
Jacek Szczygie --
Hypogenic Caves Of Sicily (Southern Italy) / Marco
Vattano, Philippe Audra, Fabrizio Benvenuto, Jean-Yves Bigot,
Jo De Waele, Ermanno Galli, Giuliana Madonia, Jean-Claude
Nobecourt --
Sediments At The Munnich Passage In The Baradla Cave
(Hungary): Mineralogical And Petrological Study/ Gabor Vid,
Istvan Berenyi Uveges, Orsolya Viktorik, Tibor Nemeth,
Zsolt Bendo,Sandor Jozsa, Judit Berenyi Uveges --
Endokarsts And Cryptokarsts In Cretaceous Coarse And
Highly Porous Chalk At The Belgian-Dutch Border / Luc Willems,
Joel Rodet --
The Tupper Glacier Sink Raspberry Rising Cave System,
Glacier National Park, Canada: A Remarkable Example In Stripe
Karst / Charles Yonge, Nicholaus Vieira, Adam Walker --
Cave And Karst Prospection In Ras Al-Khaimah Mountains,
Northern United Arab Emirate / Nadja Zupan Hajna, Asma Al
Farraj Al Ketbi, Franci Gabrovsek, Metka Petric, Tadej Slabe,
Martin Knez, Janez Mulec --
Influence Of The Pleistocene Glaciations On Karst
Development In The Dinarides Examples From Velebit Mt.
(Croatia) / Neven Bocic, Sanja Faivre, Marijan Kovacic,
Nada Horvatincic --
Contribution Of The Karstic Environment On The Origin Of
The Collophanites Of Itataia Uranium-Phosphorus Deposit,
Borborema Province, Brazil / Jose Adilson Dias Cavalcanti,
Cesar Ulisses Vieira Verissimo, Maria Dulcinea M.R. Bessa,
Clovis Vaz Parente --
Na Javorce Cave A New Discovery In The Bohemian Karst
(Czech Republic): Unique Example Of Relationships Between
Hydrothermal And Common Karstification / Jiri Dragoun, Karel
Zak, Jiri Vejlupek, Michal Filippi, Jiri Novotny, Petr
Dobes --
Field Measurements Of Gypsum Denudation Rate In
Kulogorskaya Cave System / Nikolay Franz, Sergey Sorokin,
Alexandra Alexeeva, Irina Inshina, Olga Novysh, Anton Kazak --
Geology And Structure Of Parian Cave, Isfahan, Iran /
Ghassem Ghaderi, Leila Karimi --
Preliminary Studies On Corrosive Forms And Clastic
Deposits In The Sniezna Cave (The Tatra Mts., Poland) / Ditta
Kicinska --
Geochemical And Stable Isotope Characterization Of Drip
Water From The Postojna Cave, Slovenia / Magda Mandic, Andrej
Mihevc, Albrecht Leis, Ines Krajcar Bronic --
The Skalisty Potok Cave In The Relationship To The
Relief Of The South Part Of Jasovska Plateau (Slovak Karst)
After 25 Years Of Research / Alena Petrvalska, Zdenko Hochmuth
--
Can Conduit Volumes Obtained From Artificial Tracer Tests
Be Trusted? / Anna Vojtechova, Jiri Bruthans, Ondrej
Jager, Frantisek Krejca --
The Tectonic Control Of An Underground River Network,
Agia Triada Cave (Karystos, Greece) / Emmanuel Vassilakis,
Kyriaki Papadopoulou-Vrynioti.
Session: Karst And Caves In Other Rocks, PseudokarstCave
Formation Initiated By Dissolution Of Carbonate Cement In
Quartzose Sandstones / Jiri Adamovic, Radek Mikulas,
Tomas Navratil, Jan Mertlik --
Arenitic Caves In Venezuelan Tepuis: What Do They Say
About Tepuis Themselves? / Roman Aubrecht, Tomas Lanczos,
Jan Schlogl, Lukas Vlcek, Branislav Smida --
Infrared Thermographic Survey Of Pseudokarst Sites In The
Fysch Belt Of Outer West Carpathians (Czech Republic) / Ivo
Baron, David Beckovsky, Lumir Mica --
Kahuenaha Nui (Hawaii): A Cave Developed In Four
Different Lava Flows / Ingo Bauer, Stephan Kempe, Peter Bosted
--
Genetic Types Of Non-Solution Caves / Pavel Bella,
Ludovit Gaal --
The Keokeo Lava Tube System In Hawaii / Peter Bosted,
Tomislav Gracanin, Veda Hackell, Ann Bosted, Ingo Bauer,
Stephan Kempe --
Origin Of "Rock Cities", Pillars And Cleft-Conduits In
Kaolinite-Bonded Sandstone: New Insight From Study In Sandstone
Quarry Where Landforms Recently Evolve / Jiri Bruthans, Jan
Soukup, Jana Schweigstillova, Jana Vaculikova, Daniel
Smutek, Alan L. Mayo, Lukas Falteisek --
Origins Of The Karstic Sediment Filling Of The
Northwestern Paris Basin Caves / S. Chedeville, J. Rodet, B.
Laignel, D. Todisco, E. Dupuis, G. Girot, G. Hanin --
Sandstone Caves In The Sydney Basin: A Review Of Their
Cultural And Natural Heritage / John R Dunkley --
Karst Evolution In Sandstone: The Chapada Dos Guimaraes
Site, Brazil / Rubens Hardt, Joel Rodet, Sergio Dos Anjos
Ferreira Pinto --
Three Types Of Crevice-Type Caves In The Area Of Czech
Flysch Carpathians / Jan Lenart, Tomas Panek --
Exhumation Of Palaeokarst By Vadose Weathering: A Telling
Example From Eastern Australia / R. Armstrong L. Osborne --
An Introduction To Sri Lankan Gneiss And Granite Caves /
R. Armstrong L. Osborne, Pathmakumara Jayasingha, Wasantha S.
Weliange --
Endo- And Exo-Karstic Features In Banded Iron Formation
And Canga, Serra Da Piedade, Quadrilatero Ferrifero (Brazil)
/ Manuela Correa Pereira, Joel Georges Marie Andre Rodet,
Andre Augustos Rodrigues Salgado --
Englacial Caves Of Glaciar Perito Moreno And Glaciar
Ameghino, Patagonia (Argentina) / Leonardo Puccini, Marco
Mecchia --
Speleogenesis And Speleothems Of The Guacamaya Cave,
Auyan Tepui, Venezuela / Francesco Sauro, Joyce Lundberg, Jo De
Waele, Nicola Tisato. Ermanno Gall --
Gobholo Cave: A Long Granite Cave In Swaziland (Southern
Africa) / Manuela Scheuerer, Johannes E. K. Lundberg, Rabbe
Sjoberg --
Some Significant Caves Found In The Non-Karstic Rocks Of
Hungary / Istvan Eszterhas, George Szentes --
Types Of Non-Karst Caves In Polish Outer Carpathians -
Historical Review And Perspectives / Jan Urban, Wodzimierz
Margielewski --
Caves In Sandstone Deposits Of The Southern Italian
Apennines / Gianni Campanella, Mario Parise, Angela Rizzi,
Mariangela Sammarco, Antonio Trocino --
Karst Developed In Siliciclastic Rocks At Serra Do
Espinhaco Meridional, Minas Gerais (Brazil) / Alessandra
Mendes Carvalho Vasconcelos, Fernanda Cristina Rodrigues De
Souza, Joel Rodet, Cristiane Valeria De Oliveira, Andre
Augusto Rodrigues Salgado
Session: SpeleogenesisLag And Transfer Time Inferred
From Melting Cycles Record In The Coulomp Karst Spring (Alpes
De Hauteprovence, France) / Philippe Audra, Jean-Claude
Nobecourt --
Dissolution Rates, Carbon Dioxide Dynamics, And
Geomorphological Feedbacks In Open Channel Cave Streams /
Matthew D. Covington, Mitja Prelovsek, Franci Gabrovsek --
A Methodology To Estimate The Age Of Caves In Northern
Latitudes, Using Toerfjellhola In Norway As An Example / Trevor
Faulkner --
How Do Aperture Sizes In Limestone Vary For The Onsets Of
Turbulent Flow And First-Order Dissolution Kinetics? / Trevor
Faulkner --
Investigations Into The Potential For Hypogene
Speleogenesis In The Cumberland Plateau Of Southeast Kentucky,
U.S.A. / Lee J. Florea --
The Role Of Base Level Incision And Transient Recharge In
Vertical Organisation Of Karst Network / Franci Gabrovsek,
Philipp Hauselmann, Philippe Audra --
Hypogenic Speleogenesis In The Crimean Fore-Mountains
(The Black Sea Region, South Ukraine) And Its Role In The
Regional Geomorphology / Alexander B. Klimchouk, Gennadiy N.
Amelichev, Elizaveta I. Tymokhina, Sergiy V. Tokarev --
Paragenesis: The "Royal Mark" Of Subglacial Speleogenesis
/ Stein-Erik Lauritzen --
Glacier Ice-Contact Speleogenesis / Stein-Erik Lauritzen,
Rannveig vrevik Skoglund --
Groundwater / Surface Water Mixing In The Underground
Hydrologic System Of The Demanovska Dolina Valley (Slovakia)
/ Peter Malik, Dagmar Haviarova, Jaromir Svasta, Milos
Gregor, Anton Auxt --
Karst Processes And Carbon Flux In The Frasassi Caves
Italy / Marco Menichetti --
Miocene Pliocene Age Of Cave Snezna Jama Na Raduhi,
Southern Alps, Slovenia / Andrej Mihevc, Ivan Horacek, Petr
Pruner, Nadja Zupan Hajna, Stanislav Cermak, Jan Wagner,
Pavel Bosak --
Sedimentary Study And U-Th Datations Contribution To The
Morphodynamic Reconstitution Of The Junction Chamber (Kassarat
Cave, Nabay, Lebanon): A Geomorphological Approach For
Palaeohydrological Records Analysis / C. Nehme, J.-J. Delannoy,
S. Jaillet, J. Adjizian-Gerard, John Hellstrom, T. Comaty, M.
Arzouni, P. Matta --
A Genetic Classification Of Caves In Lower Austria /
Pauline Oberender, Lukas Plan --
Effects Of Co2-Depleted Vadose Seepage In Karst /Arthur
N. Palmer, Margaret V. Palmer --
Detailed Morphologic Analysis Of Palaeotraun Gallery
Using A Terrestrial Laser Scan (Dachstein-Mammuthohle, Upper
Austria) / Lukas Plan, Andreas Roncat, Gregor Marx --
Influence Of The Plio-Pleistocene Tectonics On The
Evolution Of The Purgatorio Polje (North-Western Sicily)
Rosario Ruggieri, Giuseppe Napoli Pietro Renda --
Speleogenesis Of A 3D-Maze Cave (Hermannshohle, Lower
Austria) / Andrea Schober, Lukas Plan, Denis Scholz, Christoph
Spotl --
Evolution Of Cave Provalata: Remarkable Example Of
Hypogenic Thermal Carbonic And Sulfuric Acid Speleogenesis /
Marjan Temovski, Philippe Audra --
Flank Margin Caves On A Passive Continental Margin:
Naracoorte And Other Southern Australian Examples / Susan White
--
Tritium And H, O And C Stable Isotopes As A Tool For
Tracking Of Water Circulation In The Niedzwiedzia Cave System
(Sudetes, Poland) / Micha Gasiorowski, Helena Hercman.
Session: Cave MineralogyHypogene Sulfuric Acid
Speleogenesis And Rare Sulfate Minerals (Fibroferrite, Jarosite
Subgroup) Baume Galiniere Cave (Alpes-De-Haute-Provence,
France) / Philippe Audra, Fernando Gazquez, Fernando Rull,
Jean-Yves Bigot, Hubert Camus --
Rare Sulfates (Mirabilite, Eugsterite) In The Dry
Microclimate Of Chamois Cave (Alpes-De-Haute-Provence, France)
/ Philippe Audra, Jean-Claude Nobecourt --
Mineralogy And Speleogenesis Of The Corona 'E Sa Craba
Quartzite Cave (Southwest Sardinia) / Eleonora Baldoni, Jo De
Waele, Ermanno Galli, Mauro Messina, Bogdan P. Onac, Laura
Sanna, Francesco Sauro, Mauro Villani --
Genesis And Evolution Of Calcite Bubbles In Gypsum Caves
/ Massimo Ercolani, Katia Poletti, Paolo Forti --
Strontianite From Serizjan Cave (Iran): A New Occurrence
Of This Rare Cave Mineral With A Discusion On Its Genesis /
Jose Maria Calaforra, Jo De Waele, Paolo Forti, Ermanno Galli,
Masoud Ebadi --
Skeleton Crystals Of Cryogenic Gypsum From Kungur Ice
Cave, Ural Mountains, Russia / Olga Kadebskaya, Ilya
Tchaikovsky --
Allophane Moonmilk From A Granite-Gneiss Cave In Norway:
A Biogenic Intermediate In Deep Weatehring? / Stein-Erik
Lauritzen --
Calcite Speleothems In Pseudokarst Tjuv-Antes Grotta,
Northern Sweden / Johannes E. K. Lundberg, Magnus Ivarsson,
Therese Sallstedt, Rabbe Sjoberg, Juan Ramon Vidal Romani,
Love Dalen, Stina Hoglund --
Dating Diagenised Aragonite Speleothems / Rebeca
Martin-Garcia, Ana M. Alonso-Zarza, Silvia Frisia --
The Hypogene Origin Of Diana Cave (Romania) And Its
Sulfuric Acid Weathering Environment / Bogdan P. Onac, Cristina
M. Puscas, Herta S. Effenberger, Ioan Povara, Jonathan G.
Wynn --
The Life Cycle Of Speleothems, With Possible Implications
For Radiometric Dating / Charles Self --
Speleothems In Cavities Developed In Magmatic Rocks /
Juan Ramon Vidal-Romani, Jorge Sanjurjo-Sanchez, Marcos
Vaqueiro-Rodriguez, Laura Gonzalez-Lopez, Maria Jose
Lopez-Galindo --
Evolution Of Guano Under Different Environmental
Conditions: A Mineralogical Approach / Alexandra M. Giurgiu,
Bogdan P. Onac, Tudor Tamas, Joan J. Fornos --
A Contribution To The Mineralogy Of Monte Guisi Cave (Sw
Sardinia, Italy) / Cristian Moldovan, Monica I. Calugar,
Bogdan P. Onac, Jo De Waele, Angelo Naseddu, Joan J. Fornos --
New Localities Of Coarsely Crystalline Cryogenic Cave
Carbonates In Slovakia / Monika Orvosova, Lukas Vlcek,
Karel Zak



PAGE 1

ProceedingsVOLUME 3 Edited by Michal Filippi Pavel Bosk16thINTERNATIONAL CONGRESS OF SPELEOLOGY

PAGE 3

Czech Republic, Brno July 21 28, 2013201316thINTERNATIONAL CONGRESS OF SPELEOLOGYProceedingsVOLUME 3Edited by Michal Filippi Pavel Bosk

PAGE 4

16thINTERNATIONAL CONGRESS OF SPELEOLOGY Czech Republic, Brno July 21 28, 2013Cover photos (some photos were adjusted/cropped) Top left Specific carbonate speleothem decorations in the Ghost Chamber, Sima de la Higuera Cave. Photo by V. Ferrer. For details see the paper by F. Gzquez and J.-M. Calaforra. Top right A challenging exploration in the Cueva de los Cristales. Mexico. Photo by La Venta Exploring Team and Speleoresearch & Films. For details see the paper by F. Gzquez et al. Bottom left An example of a microcrystalline grained halite speleothem, the Octopus formation in the 3N Cave, Qeshm Island, Iran. Photo by NAMAK team. For details see the paper by Filippi et al. Bottom right Internal skeleton structure of cryogenic gypsum crystals caused by the presence of partitions oriented parallel to the faces. For details see the paper by Kadebskaya and Tchaikovsky. Produced by the Organizing Committee of the 16thInternational Congress of Speleology. Published by the Czech Speleological Society and the SPELEO2013 and in the co-operation with the International Union of Speleology. Design by M. Filippi and SAVIO, s.r.o. Layout by SAVIO, s.r.o. Printed in the Czech Republic by H.R.G. spol. s r.o. The contributions were not corrected from language point of view. Contributions express author(s) opinion. Recommended form of citation for this volume: Filippi M., Bosk P. (Eds), 2013. Proceedings of the 16thInternational Congress of Speleology, July 21, Brno. Volume 3, p. 499. Czech Speleological Society. Praha.VOLUME 3ProceedingsISBN 978-80-87857-09-0 2013 Czech Speleological Society, Praha, Czech Republic. Individual authors retain their copyrights. All rights reserved. No part of this work may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any data storage or retrieval system without the express written permission of the copyright owner. All drawings and maps are used with permission of the authors. Unauthorized use is strictly prohibited. KATALOGIZACE V KNIZE NRODN KNIHOVNA R International Congress of Speleology (16. : Brno, esko) 16th International Congress of Speleology : Czech Republic, Brno July 21,2013 : proceedings. Volume 3 / edited by Michal Filippi, Pavel Bosk. -[Prague] : Czech Speleological Society and the SPELEO2013 and in the co-operation with the International Union of Speleology, 2013 ISBN 978-80-87857-09-0 (bro.) 551.44 551.435.8 551.435.88-021.252 549 speleology karstology karst pseudokarst mineralogy proceedings of conferences speleologie karsologie kras pseudokras mineralogie sbornky konferenc 551 Geology, meteorology [7] 551 Geologie. Meteorologie. Klimatologie [7]

PAGE 5

Scientific CommitteeChairmanPavel Bosk (Czech Republic) Karst and PseudokarstVice-ChairmanMichal Filippi (Czech Republic) Karst and PseudokarstMembers Ji Adamovi (Czech Republic) Pseudokarst Philippe Audra (France) Speleogenesis Jean-Pierre Bartholeyns(France) Management and Protection Aaron Bird (USA) Exploration Didier Cailhol (France) Speleogenesis Matt Covington (USA) Modelling in Karst and Caves Robert Eavis (USA) Exploration Anette S. Engel (USA) Geomicrobiology Luk Faltejsek (Czech Republic) Biospeleology Derek Ford (Canada) Climate and Paleoclimate Franci Gabrovek (Slovenia) Modelling Mladen Garai(Croatia) Survey, Mapping and Data Processing Martin Golec (Czech Republic) Archeology and Paleontology Christiane Grebe(Germany) Management and Protection Nadja Zupan Hajna(Slovenia) Extraterrestrial Karst Ivan Horek (Czech Republic) Biospeleology Stephan Kempe (Germany) History Aleksander A. Klimchouk(Ukraine) Speleogenesis Ji Kyselk (Czech Republic) Exploration Peter Matthews (Australia) Survey, Mapping and Data Processing Iona Meleg (France) Management and Protection Mario Parise (Italy) Artificial Underground Bohdan P. Onac (USA) Mineralogy Yavor Shopov (Bulgaria) Climate and Paleoclimate The names of the Committee members are given along with their home countries and fields of research they represented as convenors.

PAGE 7

ContentsPreface 10Session: Karst and Caves in Carbonate Rocks, Salt and Gypsum 1313KARST HYDROGEOLOGY OF THE HANEY LIMESTONE, SOUTH-CENTRAL KENTUCKY Sarah M. Arpin, Christopher G. Groves ..........................................................................................................................................................15 AN INTRODUCTION TO KALAHROUD CAVE, NORTH OF ESFAHAN, IRAN Shirin Bahadorinia, Sayed Hassen Hejazi, Alireza Nadimi, Derek C. Ford ......................................................................................16 CAVE SYSTEM OF KITA GAEINA DRAENOVA PUHALJKA. THE LONGEST CAVE IN THE DINARIC KARST Teo Barii, Darko Baki, Dalibor Paar ........................................................................................................................................................17 CAVES OF EL PEON, CORDILLERA ORIENTAL, COLOMBIA Martin Bochud, Roman Hapka, Jean-Marc Jutzet ..........21 RECENT SURVEY AND EXPLORATION IN LECHUGUILLA CAVE, NEW MEXICO, USA Peter Bosted, John Lyles ............27 PLEISTOCENE SEA LEVEL CHANGES AS REVEALED BY FLANK MARGIN CAVES IN TELOGENETIC LIMESTONES IN SICILY AND SARDINIA (ITALY) Ilenia Maria DAngeli, Jo De Waele, Rosario Ruggieri, Laura Sanna ............................29 SOME SCIENTIFIC FEATURES OF THE PUERTO PRINCESA UNDERGROUND RIVER: ONE OF THE NEW 7 WONDERS OF NATURE (PALAWAN, PHILIPPINES) Antonio De Vivo, Leonardo Piccini, Paolo Forti, Giovanni Badino ......................................................................................................35 SUBTERRANEAN GLACIAL SPILLWAYS: AN EXAMPLE FROM THE KARST OF SOUTH WALES, U.K. Andrew R. Farrant, Michael J. Simms, Steven R. Noble ..........................................................................................................................42 PROJECT NAMAK: SOME OF THE MOST SPECTACULAR FINDINGS IN THE IRANIAN SALT KARST Michal Filippi, Ji Bruthans, Ondej Jager, Mohammad Zare, Naser Asad ....................................................................................48 KARST DEVELOPMENT IN THE GLACIATED AND PERMAFROST REGIONS OF THE NORTHWEST TERRITORIES, CANADA Derek C. Ford ........................................................................................................................................................................................54 LITTLE LIMESTONE LAKE: A BEAUTIFUL MARL LAKE IN MANITOBA, CANADA Derek Ford ..................................................55 CAVES AND KARST HYDROGEOLOGY OF JERUSALEM, ISRAEL Amos Frumkin ..........................................................................60 CAVES UNDER DUBROVNIK AIRPORT IN CROATIA Mladen Garai ..............................................................................................66 SOME INFORMATION ABOUT THE DEEPEST CAVES KNOWN IN CROATIAN KARST AREA Mladen Garai, Davor Garai ........................................................................................................................................................................72 HYPOGENE SPELEOGENESIS AND SPELEOTHEMS OF SIMA DE LA HIGUERA CAVE (MURCIA, SOUTH-EASTERN SPAIN) Fernando Gzquez, Jos-Mara Calaforra ..........................................................................78 THE APPLICATION OF GIS METHODS IN MORPHOMETRICAL ANALYSIS OF DOLINES ON LIMESTONE AND DOLOMITE BEDROCK Petra Gostinar ..........................................................................................................................................................84 ORIGIN OF ATYPICAL CALCITE SPELEOTHEMS FILLING FISSURES IN SANDSTONES Micha Gradziski, Marek Duliski, Helena Hercman, Andrzej Grny, Stanisaw Przybyszowski ......................................89 UPLIFT EVIDENCE FROM KARST MORPHOLOGY: PRELIMINARY EVIDENCE FROM BLAMBANGAN PENINSULA KARST, INDONESIA Eko Haryono ......................................................................................................................................................................90 VALLEY INCISION IN THE NZKE TATRY MTS. (SLOVAKIA) ESTIMATED BASED ON PALEOMAGNETIC AND RADIOMETRIC CAVE SEDIMENT DATINGS Jaroslav Kadlec, Pavol Bella, Kristna kov, Darryl E. Granger, Helena Hercman, Peter Holbek, Martin Chadima, Monika Orvoov, Petr Pruner, Petr Schnabl, Stanislav lechta ....................................................................94 MAGNETIC FABRIC AND MINERALOGY OF CAVE DEPOSITS IN BOTOVSKAYA CAVE (EASTERN SIBERIA, RUSSIAN FEDERATION) Jaroslav Kadlec, Helena Herman, Martin Chadima, Lenka Lis, Hedi Oberhnsli, Alexandr Osintsev ....................................................................................................................................................................................................96 CASE STUDIES OF FLUORESCENT GROUNDWATER TRACING IN RECENT CAVE RESEARCH Benjamin V. Miller, Chris Groves, Jason S. Polk, Robert N. Lerch ......................................................................................................99 THE JAJ PLATEAU (LEBANON): TYPICAL HIGH ALTITUDE MEDITERRANEAN KARST Fadi H. Nader, Hughes Badaoui, Marc Metni, Chadi Chaker, Habib Helou, Johnny Tawk ....................................................105 A CONCEPTUAL MODEL OF SPELEOGENESIS IN GREECE Christos Pennos, Stein-Erik Lauritzen ..................................107 2013 ICS Proceedings

PAGE 8

COMPLEX EPIKARST HYDROLOGEOLOGY AND CONTAMINANT TRANSPORT IN A SOUTH-CENTRAL KENTUCKY KARST LANDSCAPE Jason S. Polk, Sean Vanderhoff, Chris Groves, Benjamin Miller, Carl Bolster ........110 KARST HYDROGEOLOGICAL OBSERVATIONS IN CAO BANG PROVINCE (VIETNAM): THE TRA LINH-THANG HEN LAKE AREA Gheorghe M. Ponta, Bogdan P. Onac, Nyguen Xuan Nam ................................................................................116 INCIDENCES OF THE TECTONICS IN THE KARSTIFICATION OF CHALK LIMESTONES IN THE WESTERN PARIS BASIN: EXAMPLE FROM THE PETITES DALES CAVE (SAINT MARTIN AUX BUNEAUX, FRANCE) Jol Rodet, Kun Ma, Jean-Pierre Viard ..........................................................................................................................................................121 CEILING CHANNEL AND INPUT KARST. EXAMPLE OF THE PETITES DALES CAVE, NORMANDY, FRANCE Jol Rodet, Laurent Magne, Jean-Pierre Viard ..........................................................................................................................................126 GULLS, GULL-CAVES AND CAMBERING IN THE SOUTHERN COTSWOLD HILLS, ENGLAND Charles Self, Andrew Farrant ............................................................................................................................................................................132 THE ROLE OF FOLD-AND-THRUST STRUCTURE IN THE LARGE SHAFTS AND CHAMBERS DEVELOPMENT: CASE STUDY OF THE POLISH TATRA MTS. Jacek Szczygie ..............................................................................................................137 HYPOGENIC CAVES OF SICILY (SOUTHERN ITALY) Marco Vattano, Philippe Audra, Fabrizio Benvenuto, Jean-Yves Bigot, Jo De Waele, Ermanno Galli, Giuliana Madonia, Jean-Claude Nobcourt ..................................................144 SEDIMENTS AT THE MNNICH PASSAGE IN THE BARADLA CAVE (HUNGARY): MINERALOGICAL AND PETROLOGICAL STUDY Gbor Vid, Istvn Bernyi veges, Orsolya Viktorik, Tibor Nmeth, Zsolt Bend, Sndor Jzsa, Judit Bernyi veges ..............................................................................................................................................................150 ENDOKARSTS AND CRYPTOKARSTS IN CRETACEOUS COARSE AND HIGHLY POROUS CHALK AT THE BELGIAN-DUTCH BORDER Luc Willems, Jol Rodet ..............................................................................................................................1 54 THE TUPPER GLACIER SINK RASPBERRY RISING CAVE SYSTEM, GLACIER NATIONAL PARK, CANADA: AREMARKABLE EXAMPLE IN STRIPE KARST Charles Yonge, Nicholaus Vieira, Adam Walker ........................................158 CAVE AND KARST PROSPECTION IN RAS AL-KHAIMAH MOUNTAINS, NORTHERN UNITED ARAB EMIRATE Nadja Zupan Hajna, Asma Al Farraj Al Ketbi, Franci Gabrovek, Metka Petri, Tadej Slabe, Martin Knez, Janez Mulec ..............................................................................................................................................................................................................164 INFLUENCE OF THE PLEISTOCENE GLACIATIONS ON KARST DEVELOPMENT IN THE DINARIDES EXAMPLES FROM VELEBIT MT. (CROATIA) Neven Boi, Sanja Faivre, Marijan Kovai, Nada Horvatini ......................................170 CONTRIBUTION OF THE KARSTIC ENVIRONMENT ON THE ORIGIN OF THE COLLOPHANITES OF I TATAIA URANIUM-PHOSPHORUS DEPOSIT, BORBOREMA PROVINCE, BRAZIL Jos Adilson Dias Cavalcanti, Cesar Ulisses Vieira Verssimo, Maria Dulcinea M.R. Bessa, Clovis Vaz Parente ....................................................................173 NA JAVORCE CAVE A NEW DISCOVERY IN THE BOHEMIAN KARST (CZECH REPUBLIC): UNIQUE EXAMPLE OF RELATIONSHIPS BETWEEN HYDROTHERMAL AND COMMON KARSTIFICATION Ji Dragoun, Karel k, Ji Vejlupek, Michal Filippi, Ji Novotn, Petr Dobe ............................................................................179 FIELD MEASUREMENTS OF GYPSUM DENUDATION RATE IN KULOGORSKAYA CAVE SYSTEM Nikolay Franz, Sergey Sorokin, Alexandra Alexeeva, Irina Inshina, Olga Novysh, Anton Kazak ........................................185 GEOLOGY AND STRUCTURE OF PARIAN CAVE, ISFAHAN, IRAN Ghassem Ghaderi, Leila Karimi ....................................190 PRELIMINARY STUDIES ON CORROSIVE FORMS AND CLASTIC DEPOSITS IN THE NIENA CAVE (THE TATRA MTS., POLAND) Ditta Kiciska ..............................................................................................................................................194 GEOCHEMICAL AND STABLE ISOTOPE CHARACTERIZATION OF DRIP WATER FROM THE POSTOJNA CAVE, SLOVENIA Magda Mandi, Andrej Mihevc, Albrecht Leis, Ines Krajcar Broni ........................................................................196 THE SKALIST POTOK CAVE IN THE RELATIONSHIP TO THE RELIEF OF THE SOUTH PART OF JASOVSK PLATEAU (SLOVAK KARST) AFTER 25 YEARS OF RESEARCH Alena Petrvalsk, Zdenko Hochmuth ..............................201 CAN CONDUIT VOLUMES OBTAINED FROM ARTIFICIAL TRACER TESTS BE TRUSTED? Anna Vojtbchov, Ji Bruthans, Ondej Jger, Frantiek Kreja ........................................................................................................206 THE TECTONIC CONTROL OF AN UNDERGROUND RIVER NETWORK, AGIA TRIADA CAVE (KARYSTOS, GREECE) Emmanuel Vassilakis, Kyriaki Papadopoulou-Vrynioti ..........................................................................................................................208 Session: Karst and Caves in Other Rocks, Pseudokarst 2151 CAVE FORMATION INITIATED BY DISSOLUTION OF CARBONATE CEMENT IN QUARTZOSE SANDSTONES Ji Adamovi, Radek Mikul, Tom Navrtil, Jan Mertlk ................................................................................................................217 2013 ICS Proceedings

PAGE 9

ARENITIC CAVES IN VENEZUELAN TEPUIS: WHAT DO THEY SAY ABOUT TEPUIS THEMSELVES? Roman Aubrecht, Tom Lnczos, Jn Schlgl, Luk Vlek, Branislav mda ..........................................................................221 INFRARED THERMOGRAPHIC SURVEY OF PSEUDOKARST SITES IN THE FYSCH BELT OF OUTER WEST CARPATHIANS (CZECH REPUBLIC) Ivo Barot, David Bekovsk, Lumr Ma ..........................................................................227 KAHUENAHA NUI (HAWAII): A CAVE DEVELOPED IN FOUR DIFFERENT LAVA FLOWS Ingo Bauer, Stephan Kempe, Peter Bosted ................................................................................................................................................231 GENETIC TYPES OF NON-SOLUTION CAVES Pavel Bella, Ludovt Gal ........................................................................................237 THE KEOKEO LAVA TUBE SYSTEM IN HAWAII Peter Bosted, Tomislav Gracanin, Veda Hackell, Ann Bosted, Ingo Bauer, Stephan Kempe ..............................................243 ORIGIN OF ROCK CITIES, PILLARS AND CLEFT-CONDUITS IN KAOLINITE-BONDED SANDSTONE: NEW INSIGHT FROM STUDY IN SANDSTONE QUARRY WHERE LANDFORMS RECENTLY EVOLVE Ji Bruthans, Jan Soukup, Jana Schweigstillov, Jana Vaculkov, Daniel Smutek, Alan L. Mayo, Luk Falteisek ..........247 ORIGINS OF THE KARSTIC SEDIMENT FILLING OF THE NORTHWESTERN PARIS BASIN CAVES S. Chdeville, J. Rodet, B. Laignel, D. Todisco, E. Dupuis, G. Girot, G. Hanin ............................................................................253 SANDSTONE CAVES IN THE SYDNEY BASIN: A REVIEW OF THEIR CULTURAL AND NATURAL HERITAGE John R Dunkley ........................................................................................................................................................................................................259 KARST EVOLUTION IN SANDSTONE: THE CHAPADA DOS GUIMARES SITE, BRAZIL Rubens Hardt, Jol Rodet, Sergio dos Anjos Ferreira Pinto ..............................................................................................................264 THREE TYPES OF CREVICE-TYPE CAVES IN THE AREA OF CZECH FLYSCH CARPATHIANS Jan Lenart, Tom Pnek ....................................................................................................................................................................................268 EXHUMATION OF PALAEOKARST BY VADOSE WEATHERING: A TELLING EXAMPLE FROM EASTERN AUSTRALIA R. Armstrong L. Osborne ....................................................................................................................................................................................275 AN INTRODUCTION TO SRI LANKAN GNEISS AND GRANITE CAVES R. Armstrong L. Osborne, Pathmakumara Jayasingha, Wasantha S. Weliange ........................................................................280 ENDOAND EXO-KARSTIC FEATURES IN BANDED IRON FORMATION AND CANGA, SERRA DA PIEDADE, QUADRILTERO FERRFERO (BRAZIL) Manuela Corra Pereira, Joel Georges Marie Andre Rodet, Andr Augustos Rodrigues Salgado ....................................286 ENGLACIAL CAVES OF GLACIAR PERITO MORENO AND GLACIAR AMEGHINO, PATAGONIA (ARGENTINA) Leonardo Puccini, Marco Mecchia ................................................................................................................................................................292 SPELEOGENESIS AND SPELEOTHEMS OF THE GUACAMAYA CAVE, AUYAN TEPUI, VENEZUELA Francesco Sauro, Joyce Lundberg, Jo De Waele, Nicola Tisato. Ermanno Galli ........................................................................298 GOBHOLO CAVE: A LONG GRANITE CAVE IN SWAZILAND (SOUTHERN AFRICA) Manuela Scheuerer, Johannes E. K. Lundberg, Rabbe Sjberg ........................................................................................................305 SOME SIGNIFICANT CAVES FOUND IN THE NON-KARSTIC ROCKS OF HUNGARY Istvn Eszterhs, George Szentes ..................................................................................................................................................................308 TYPES OF NON-KARST CAVES IN POLISH OUTER CARPATHIANS HISTORICAL REVIEW AND PERSPECTIVES Jan Urban, Wodzimierz Margielewski ..........................................................................................................................................................314 CAVES IN SANDSTONE DEPOSITS OF THE SOUTHERN ITALIAN APENNINES Gianni Campanella, Mario Parise, Angela Rizzi, Mariangela Sammarco, Antonio Trocino ..................................................320 KARST DEVELOPED IN SILICICLASTIC ROCKS AT SERRA DO ESPINHAO MERIDIONAL, MINAS GERAIS (BRAZIL) Alessandra Mendes Carvalho Vasconcelos, Fernanda Cristina Rodrigues de Souza, Joel Rodet, Cristiane Valria de Oliveira, Andr Augusto Rodrigues Salgado ........................................................................................................................326 Session: Speleogenesis 3331LAG AND TRANSFER TIME INFERRED FROM MELTING CYCLES RECORD IN THE COULOMP KARST SPRING (ALPES DE HAUTEPROVENCE, FRANCE) Philippe Audra, Jean-Claude Nobcourt ..................................................................335 DISSOLUTION RATES, CARBON DIOXIDE DYNAMICS, AND GEOMORPHOLOGICAL FEEDBACKS IN OPEN CHANNEL CAVE STREAMS Matthew D. Covington, Mitja Prelovek, Franci Gabrovek ......................................................340 A METHODOLOGY TO ESTIMATE THE AGE OF CAVES IN NORTHERN LATITUDES, USING TOERFJELLHOLA INNORWAY AS AN EXAMPLE Trevor Faulkner ........................................................................................................................................342 2013 ICS Proceedings

PAGE 10

HOW DO APERTURE SIZES IN LIMESTONE VARY FOR THE ONSETS OF TURBULENT FLOW AND FIRST-ORDER DISSOLUTION KINETICS? Trevor Faulkner ................................................................................................................................................349 INVESTIGATIONS INTO THE POTENTIAL FOR HYPOGENE SPELEOGENESIS IN THE CUMBERLAND PLATEAU OFSOUTHEAST KENTUCKY, U.S.A. Lee J. Florea ....................................................................................................................................356 THE ROLE OF BASE LEVEL INCISION AND TRANSIENT RECHARGE IN VERTICAL ORGANISATION OF KARST NETWORK Franci Gabrovek, Philipp Huselmann, Philippe Audra ..............................................................................................362 HYPOGENIC SPELEOGENESIS IN THE CRIMEAN FORE-MOUNTAINS (THE BLACK SEA REGION, SOUTH UKRAINE) AND ITS ROLE IN THE REGIONAL GEOMORPHOLOGY Alexander B. Klimchouk, Gennadiy N. Amelichev, Elizaveta I. Tymokhina, Sergiy V. Tokarev ..................................................................................................................................................364 PARAGENESIS: THE ROYAL MARK OF SUBGLACIAL SPELEOGENESIS Stein-Erik Lauritzen ............................................366 GLACIER ICE-CONTACT SPELEOGENESIS Stein-Erik Lauritzen, Rannveig vrevik Skoglund ............................................368 GROUNDWATER / SURFACE WATER MIXING IN THE UNDERGROUND HYDROLOGIC SYSTEM OF THE DEMNOVSK DOLINA VALLEY (SLOVAKIA) Peter Malk, Dagmar Haviarov, Jaromr vasta, Milo Gregor, Anton Auxt ..............................................................................370 KARST PROCESSES AND CARBON FLUX IN THE FRASASSI CAVES ITALY Marco Menichetti ............................................376 MIOCENE PLIOCENE AGE OF CAVE SNENA JAMA NA RADUHI, SOUTHERN ALPS, SLOVENIA Andrej Mihevc, Ivan Horek, Petr Pruner, Nadja Zupan Hajna, Stanislav ermk, Jan Wagner, Pavel Bosk ..........379 SEDIMENTARY STUDY AND U-TH DATATIONS CONTRIBUTION TO THE MORPHODYNAMIC RECONSTITUTION OF THE JUNCTION CHAMBER (KASSARAT CAVE, NABAY, LEBANON): A GEOMORPHOLOGICAL APPROACH FOR PALAEOHYDROLOGICAL RECORDS ANALYSIS C. Nehme, J.-J. Delannoy, S. Jaillet, J. Adjizian-Gerard, John Hellstrom, T. Comaty, M. Arzouni, P. Matta ......................384 A GENETIC CLASSIFICATION OF CAVES IN LOWER AUSTRIA Pauline Oberender, Lukas Plan ........................................390 EFFECTS OF CO2-DEPLETED VADOSE SEEPAGE IN KARST Arthur N. Palmer, Margaret V. Palmer ................................393 DETAILED MORPHOLOGIC ANALYSIS OF PALAEOTRAUN GALLERY USING A TERRESTRIAL LASER SCAN (DACHSTEIN-MAMMUTHHLE, UPPER AUSTRIA) Lukas Plan, Andreas Roncat, Gregor Marx ........................................399 INFLUENCE OF THE PLIO-PLEISTOCENE TECTONICS ON THE EVOLUTION OF THE PURGATORIO POLJE (NORTH-WESTERN SICILY) Rosario Ruggieri, Giuseppe Napoli Pietro Renda ........................................................................402 SPELEOGENESIS OF A 3D-MAZE CAVE (HERMANNSHHLE, LOWER AUSTRIA) Andrea Schober, Lukas Plan, Denis Scholz, Christoph Sptl ............................................................................................................408 EVOLUTION OF CAVE PROVALATA: REMARKABLE EXAMPLE OF HYPOGENIC THERMAL CARBONIC AND SULFURIC ACID SPELEOGENESIS Marjan Temovski, Philippe Audra ............................................................................................412 FLANK MARGIN CAVES ON A PASSIVE CONTINENTAL MARGIN: NARACOORTE AND OTHER SOUTHERN AUSTRALIAN EXAMPLES Susan White ........................................................................................................................................................418 TRITIUM AND H, O AND C STABLE ISOTOPES AS A TOOL FOR TRACKING OF WATER CIRCULATION IN THE NIEDbWIEDZIA CAVE SYSTEM (SUDETES, POLAND) Micha Gnsiorowski, Helena Hercman ............................421 Session: Cave Mineralogy 423HYPOGENE SULFURIC ACID SPELEOGENESIS AND RARE SULFATE MINERALS (FIBROFERRITE, JAROSITE SUBGROUP) BAUME GALINIERE CAVE (ALPES-DE-HAUTE-PROVENCE, FRANCE) Philippe Audra, Fernando Gzquez, Fernando Rull, Jean-Yves Bigot, Hubert Camus ............................................................425 RARE SULFATES (MIRABILITE, EUGSTERITE) IN THE DRY MICROCLIMATE OF CHAMOIS CAVE (ALPES-DE-HAUTE-PROVENCE, FRANCE) Philippe Audra, Jean-Claude Nobcourt ................................................................432 MINERALOGY AND SPELEOGENESIS OF THE CORONA E SA CRABA QUARTZITE CAVE (SOUTHWEST SARDINIA) Eleonora Baldoni, Jo De Waele, Ermanno Galli, Mauro Messina, Bogdan P. Onac, Laura Sanna, Francesco Sauro, Mauro Villani ..........................................................................................................................................437 GENESIS AND EVOLUTION OF CALCITE BUBBLES IN GYPSUM CAVES Massimo Ercolani, Katia Poletti, Paolo Forti ............................................................................................................................................443 STRONTIANITE FROM SERIZJAN CAVE (IRAN): A NEW OCCURRENCE OF THIS RARE CAVE MINERAL WITH ADISCUSION ON ITS GENESIS Jos Maria Calaforra, Jo De Waele, Paolo Forti, Ermanno Galli, Masoud Ebadi ......449 SKELETON CRYSTALS OF CRYOGENIC GYPSUM FROM KUNGUR ICE CAVE, URAL MOUNTAINS, RUSSIA Olga Kadebskaya, Ilya Tchaikovsky ..............................................................................................................................................................454 2013 ICS Proceedings

PAGE 11

ALLOPHANE MOONMILK FROM A GRANITE-GNEISS CAVE IN NORWAY: A BIOGENIC INTERMEDIATE IN DEEP WEATEHRING? Stein-Erik Lauritzen ..............................................................................................................................................................459 CALCITE SPELEOTHEMS IN PSEUDOKARST TJUV-ANTES GROTTA, NORTHERN SWEDEN Johannes E. K. Lundberg, Magnus Ivarsson, Therese Sallstedt, Rabbe Sjberg, Juan Ramn VidalRoman, Love Daln, Stina Hglund ................................................................................................................................................................................461 DATING DIAGENISED ARAGONITE SPELEOTHEMS Rebeca Martn-Garca, Ana M. Alonso-Zarza, Silvia Frisia ........465 THE HYPOGENE ORIGIN OF DIANA CAVE (ROMANIA) AND ITS SULFURIC ACID WEATHERING ENVIRONMENT Bogdan P. Onac, Cristina M. Puca, Herta S. Effenberger, Ioan Povarf, Jonathan G. Wynn ............................................470 THE LIFE CYCLE OF SPELEOTHEMS, WITH POSSIBLE IMPLICATIONS FOR RADIOMETRIC DATING Charles Self ........474 SPELEOTHEMS IN CAVITIES DEVELOPED IN MAGMATIC ROCKS Juan Ramn Vidal-Roman, Jorge Sanjurjo-Snchez, Marcos Vaqueiro-Rodrguez, Laura Gonzlez-Lpez, Mara Jos Lpez-Galindo ..................................................................................................................................................................................479 EVOLUTION OF GUANO UNDER DIFFERENT ENVIRONMENTAL CONDITIONS: A MINERALOGICAL APPROACH Alexandra M. Giurgiu, Bogdan P. Onac, Tudor Tfmar, Joan J. Forns ............................................................................................483 A CONTRIBUTION TO THE MINERALOGY OF MONTE GUISI CAVE (SW SARDINIA, ITALY) Cristian Moldovan, Monica I. Cflugfr, Bogdan P. Onac, Jo de Waele, Angelo Naseddu, Joan J. Forns ........................486 NEW LOCALITIES OF COARSELY CRYSTALLINE CRYOGENIC CAVE CARBONATES IN SLOVAKIA Monika Orvoov, Luk Vlek, Karel k ................................................................................................................................................490Partners, Sponsors 496 Authors Index 498 2013 ICS Proceedings

PAGE 12

Dear readers, the Proceedings volumes you are holding in your hands were issued within the 16thInternational Congress of Speleology (hereafter 16thICS) on July 21, 2013 in Brno, Czech Republic. Let us welcome you to its reading. In total, over 320 contributions (over 250 oral presentations and over 70 posters) by more than 750 authors have been received to be included within the Congress Proceedings. This represents over 2,300 received e-mails and a similar number of responses during the last 6 months, approximately 4,300 electronic files and over 1,450 printed pages of the text. To put it simply, really, really much interesting stuff concerned with cave and karst subject. The authors guidelines stipulated that the particular contributions should not exceed 6 pages of text and we were delighted to find that most authors prepared contributions close to this upper limit. Only very few contributions did not exceed one page of text. This illustrates a clear willingness of the cavers and karst scientist to share their discoveries and research conclusions. The presented contributions (abstracts/papers) stand for both oral and poster presentations as indicated in the headings. Contributions in each session are arranged alphabetically by the last name of the first author. All contributions were reviewed from the viewpoint of technical quality and scientific content by members of the scientific committee and invited reviewers. The authors had the opportunity to revise their papers in response to reviewers comments and we were pleased to see that the reviews have improved the clarity and readability of the contributions. However, profound improvement of the English language could not be arranged due to the shortage of time and insufficient human resources; the authors themselves are therefore responsible for the linguistic level of their contributions. Thirteen thematically different sessions and six special sessions were scheduled within the call for your contributions to cover the whole range of subjects to be discussed within the wide scope of the 16thICS. The low number of contributions for some of these detailed sessions necessitated their merging with others. As a result, eleven original and three joint sessions are presented within the Proceedings. The contributions were grouped into three separate volumes. The purpose of this arrangement was that each particular Volume is filled with a certain logical hierarchy of topics, and that related topics are presented together. It was also the intention that the content of each Volume is topically balanced and contains both generally interesting (popular) topics with rich photographic documentation and hardcore scientific topics dominated by tables and plots. Volume I starts with three plenary lectures representing three global topics related to 16thICS subject. Further it contains papers concerned with history of research (session History of Speleology and Karst Research), archeology and paleontology (sessions Archaeology and Paleontology in Caves), topics focused on management and preservation of caves and karst areas and other social-related aspects (sessions Protection and Management of Karst, Education; Karst and Caves: Social Aspects and Other Topics). In the last mentioned session you can also find a small part devoted to extraterrestrial karst. Volume I is ended by a relatively large portion of biology-oriented papers placed within the session Biospeleology, Geomicrobiology and Ecology. Volume II contains the traditionally heavily attended session Exploration and Cave Techniques and by the related session Speleological Research and Activities in Artificial Underground. These exploration topics are, we believe, logically supplemented with contributions from the field of Karst and Cave Survey, Mapping and Data Processing. The content of the second Volume is completed with a somewhat more specialized session Modelling in Karst and Cave Environments and with session Cave Climate and Paleoclimate Record. The last mentioned session probably better fits to the end of Volume III, but it was placed into Volume II in order to reach balance in the extent of the individual volumes. Volume III also starts with traditional, heavily attended topics organized in two sessions: Karst and Caves in Carbonate Rocks, Salt and Gypsum and Karst and Caves in Other Rocks, Pseudokarst. These topics are supplemented by the related session Speleogenesis. This last volume of the Proceedings is ended by the study of cave minerals, included in a specific session Cave Minerals. It is clear already from the previous ICS meetings that the range of the published topics becomes wider and wider, including localities in the whole world but also owing to the access to high-quality spacecraft images from other planets. The range of the instrumental, analytical and software methods employed in cave and karst research is remarkable and shows that the topic of cave & karst exploration attracts an ever increasing number of researchers even from already established scientific disciplines. Let us also say a few words about the selection of the cover photos for the Proceedings volumes. The idea was to select such photos which would best represent all topics (especially those enjoying the highest interest) in each particular volume and be of high technical quality. Since we believe the cover page is a place for a serious presentation of the inner content, we made our selection from photos used in the presented papers. In one case the additional photo was requested to get a better representation of the topic. For our purpose, we decided to place several photos on the cover page of each volume. We hope that you enjoy them. We wish to take this opportunity to apologize for the all mistakes which might have possibly originated within the operations with different versions of the manuscripts and other related files and e-mails which passed through our computers. We believe that everybody find their interesting reading in the Proceedings and we wish that the whole publication (Volumes IIII) becomes a valuable record of the 16thmeeting of enthusiasts addicted to the fascination of the underground world.Preface10 2013 ICS Proceedings

PAGE 13

Finally we wish to thank all the authors for their contributions. Enormous thanks belong to the reviewers and especially convenors (members of the scientific committee) of the particular sessions for their time and effort in the improvement of the overall message of the texts. We also wish to thank Michal Molhanec who significantly helped with the on-line form for the contribution submission, to Ji Adamovi who repeatedly helped us with the improvement of our English, and to Jan Spruina, Zdentk Motyka, Jana Holubcov, and Renata Filippi who contributed to the preparation of the Proceedings. After the few introductory words, lets now enjoy the papers from localities all over the world, presenting all forms of activities in karst, caves and other related surface and subsurface environments! Michal Filippi and Pavel Bosk Proceedings editors11 2013 ICS Proceedings

PAGE 14

12

PAGE 15

Karst and Caves in Carbonate Rocks, Salt and GypsumSession: 2013 ICS Proceedings 13

PAGE 16

14

PAGE 17

KARST HYDROGEOLOGY OF THE HANEY LIMESTONE, SOUTH-CENTRAL KENTUCKYSarah M. Arpin, Christopher G. Groves Hoffman Environmental Research Institute, Department of Geography and Geology, Western Kentucky University, Bowling Green, KY 42101, USA, sarah.arpin759@topper.wku.edu, chris.groves@wku.edu South-central Kentucky has one of the worlds most intensively studied karst areas, with most work focusing on the Mammoth Cave System and its aquifers. However, slightly higher in the stratigraphic section than Mammoth Cave, the Haney Limestone is an important but less well studied carbonate aquifer. This research provides the first comprehensive hydrogeologic description of the Haney Limestone, synthesizing patterns of the cave and karst features developed within. Joints are the most dominant control on passage development in the Haney Limestone within the study area; when considered together the orientation of these joints is consistent with the orientation of regional joint sets. There is evidenc e that the spaces along these joints in some cases may have been were enlarged by stress release fracturing. Bedding planes and the presence of insoluble rock at the base of the Haney also exert significant control on conduit development in the Haney Limestone. The strong influence of joints on conduit development in the Haney Limestone contrasts with the major caves of the St. Louis, Ste. Genevieve, and Girkin Limestones below, in which bedding planes are a dominant influence. Most of the caves of the study area developed in the Haney Limestone are single-conduit caves that receive water through direct, allogenic sources. Cave entrances are frequently perennial spring resurgences; whereas the presence of active streams suggests that the caves are a function of the current landscape, acting as drains for localized recharge areas. The hydrology of the Haney Limestone plays an important, if localized, role in the regional hydrology of south-central Kentucky; integrated into the current system of surface and subsurface drainage of the regional karst landscape. Evidence supports the idea that caves of the Haney Limestone are, geologically, relatively recent phenomena. A majority of the caves in the study area are hydrologically active, the water resurging from the sampled springs is undersaturated with respect to limestone, and the caves are developed along potential stress release fractures associated with small, relatively young valleys. This suggests that caves in the Haney Limestone were not directly influenced by the incision of the Green River, like Mammoth Cave, but that cave development is a much more recent process. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings15

PAGE 18

AN INTRODUCTION TO KALAHROUD CAVE, NORTH OF ESFAHAN, IRANShirin Bahadorinia1, Sayed Hassen Hejazi1, Alireza Nadimi2, Derek Ford3 1Department of Geology, Islamic Azad University, Khorasgan Branch, Esfahan, Iran, sbahadorinia@gmail.com2Department of Geology, Isfahan University, Esfahan, Iran3School of Geography and Earth Sciences, McMaster University, Canada Kalahroud Cave (Lat. 33 22 21 N, Long. 51 34 35 W, at 2,300 m a.s.l. near the village of Kalahroud, Esfahan Province, Iran) is an interesting and complex cave formed in Cretaceous limestones. It is located on the southwestern boundary of high mountains of the Urmiyeh-Dokhtarmagmatic beltof Iran, which is a dramatically faulted, active tectonic area. The cave has formed in the foot wall of the Kalahroud Thrust. The limestones are thick to massively bedded but contain thinner shales, are locally tectonized, and are overlain (capped) by insoluble strata. Stratal dip is approximately 15. The cave entrance is in cliffs ~25 m above the floor of a strike-aligned river valley with a channel active only in rare storms. The region is semi-arid, with only 51.8 mm average annual precipitation and a mean temperature of 22.1 C. The explored cave is ~200 m in length and has three morphologically distinct parts. From the entrance a steep breakdown passage extends broadly down dip to a depth of ~50 m. There the form changes to two large solution notch chambers beveling the dipping strata, with multiple notches, solution pockets and pendants, tapering ramiform extensions upwards and laterally, and blind pits in the floors. Two constricted joint-guided tributaries enter at the notching level that display irregular phreatic form, including slow flow scalloping, and terminations in blind pits. There were a-periodic pools many metres in diameter in the main chambers, indicated by stranded calcite raft debris and many conulites built up to precise paleo-water surfaces. On a visit shortly after rains in April 2012 one small pool with some raft accretion was found. There is a variety of chemical and clastic sediments in the cave. Precipitates include the calcite rafts, widespread popcorn and cave coral on walls and pits, anthodites, frostwork associated with ferruginous shale bands, and dogtooth spar. Very significantly, cemented rinds of highly weathered bedrock overlain by dense layered calcite are the oldest chemical deposits, reminiscent of those in Wind Cave, South Dakota, USA. ICPMS U-Series dating (courtesy of the Department of Earth Sciences, Oxford University, UK) shows that the oldest rafts in the largest pool are ~ 9,000 y B.P. in age, and the uppermost are modern. Clastic sediments are limited to breakdown and colluvial debris flows in the entrance passage, abundant clay fines in the notch chambers and tributaries. There do not appear to be any vadose stream deposits. From field observations, two main factors in cave development were tectonic uplift and deformation facilitating groundwater circulation, and dissolution with prominent rest levels (notching). Surface river entrenchment has drained and truncated the upper cave and there is only net deposition in the notch chambers. It appears that Kalahroud Cave is a hypogene cave, but further studies are required to prove this. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings16

PAGE 19

CAVE SYSTEM OF KITA GAEINA DRAENOVA PUHALJKA THELONGEST CAVE IN THE DINARIC KARSTTeo Barii, Darko Baki, Dalibor Paar The Speleological Committee of the Croatian Mountaineering Association, Kozar eva 22, Zagreb, Croatia teobarisic@gmail.com, baksic@sumfak.hr, dpaar@phy.hr The Croatian cavers have continued to explore the Crnopac Massif, the most northern, the highest and the most karstified part of the south-eastern Velebit mountain ridge known thanks to its underground karst phenomena. In more than 8 years of intensive research Cave system of Kita Ga eina Drazenova puhaljka reached the length of 23,334 m, depth of 737 m and became the Croatian and Dinaric karst longest cave. Speleologists have done some geomorphological research, microclimatic and radon concentracion measurements, collecting and analysing some animal species. The whole area shows enormous speleological and tourism potential.1. IntroductionThe Dinaric karst form a mountain chain in the Southern Europe, spanning areas of Slovenia, Croatia, Bosnia and Herzegovina, Serbia, Albania and Montenegro. They extend for 645 km along the coast of the Adriatic Sea (northwestsoutheast), from the Julian Alps in the northwest down to the ar-Korab Massif. The highest mountain of the Dinaric Karst is Mount Prokletije, located on the border of eastern Montenegro and northern Albania (2,692 m a.s.l.). The Dinaric Karst is the fifth most rugged and extensively mountainous area of Europe after the Caucasus Mountains, Alps, Pyrenees and Scandinavian Mountains. The Dinarides are known as a place of origin of term karst, the German name for Kras, a region in Slovenia partially extending into Italy, where the first scientific research of karst topography was made. This includes use of Croatian karst phenomena terms such as ponor, dolina, polje, jama, etc., in international geological vocabulary. Dinaric karst is also the top of hotspots of worlds subterranean biodiversity. The Postojnska Cave, described for the first time in 17thcentury is one of the famous and oldest tourist caves in the World. Since that time it was also known as the longest Dinaric cave. In last few decades, number of speleological explorations in whole area grows. Hundreds of cavers in each country have found new caving attractive zones with deep and long caves. The Crnopac Massif in Croatia is the most northern, the highest and the most karstified part of the south-eastern Velebit mountain ridge (part of Dinaric coastal mountainous karst). Numerous surface and underground karst phenomena are characteristic, including Cave system of Kita Ga eina Draenova puhaljka, since 2011 the Dinaric longest cave.2. Crnopac MassifThe Crnopac Massif is situated between higher terrain of the Graac Polje on the north and Zrmanja and Krupa river valleys on the South and East. The massif is built of thick upper Triassic, Jurassic and Cretaceous carbonate deposits. Due to dominantly carbonate structure of this part of Velebit, waters flow subterraneous through the Crnopac Massif generally southwards. Underground waters get water from sinking streams that come from karst polje in Lika and Gra ac areas and appear in springs on right side of Zrmanja and Krupa river valleys. In such conditions, polygenetic multilevel caves have developed. Speleogenesis of the caves in the Crnopac Massif probably have lasted continuously from the beginning of the massif uplift (Upper Miocene). Mechanical properties of the Oligocene to Lower Miocene carbonate breccias (Jelar breccias) play a significant role in the cave morphology. Low frequency of cracks and joints in these massive breccias enables the preservation of underground passages and chambers of very large dimensions. The most important caves of the Crnopac Massif are Muniaba (9,318 m, -510 m, volume 2.3 millions of m3). Cave system Kita Ga eina Draenova puhaljka (23,334 m, -737 m, volume 1.3 millions of m3), Burinka In higher parts of the massif, mentioned rocks are covered with Oligocene to Lower Miocene carbonate breccias, probably deposited on the flanks of tectonically uplifted areas during older tectonic phase (Korbar 2009). Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings17 Figure 1. Geology of Crnopac Massif.

PAGE 20

Draenova puhaljka pit was found in 2005 (the SS eljezni ar), and 2009 connected to Kita Ga eina. In 2010, the system became the longest cave in Croatia and 2011 the longest in Dinarides. Eighty-one explorations conducted by cavers of the almost every caving club in Croatia brought the total length of 23,334 m and placed system on the list of 200 World longest caves. Caving accidents in 2011 and 2012 caused two most complex cave rescue operations in Croatia ever with two blissfully happy ends.4. Cave system Kita Ga eina Draenova puhaljka (KGDP)Cave system of KGDP is a network of multiphase cave passages, remnants of various levels of paleodrainage systems that conducted waters from the higher terrain to the base level springs on the south and south-east. Older phreatic and epiphreatic channels are frequently crossed by younger invasion vadose shafts, which provide entrances and also connections between different levels. These are three main levels in the system. The highest one approximately at 800 m is developed in Jelar breccias. The middle one is approximately at 600 m, similar to Cerova ke caves and the Muniaba Cave. Even not in breccias, this level also characteristics large passages, result of simultaneous collapsing and removing of collapsed material by big underground water flows during the hydrologically active phase of their development. The lowest level (250 m) is inclined according to carbonate layer inclination (-20 to -35, SW). Lower entrance (Kita Ga eina) is at 920 m, and the higher one (Draenova puhaljka) at 985 m a.s.l. The lowest point (-737 m) is at 248 m a.s.l. what means almost 300 m lower than the Graac Polje (545) and its sinkholes. All water in system is the rainwater or melted water from ice and snow caves in the Crnopac Massif. Permanent water stream Graac Polje drainage system havent been found yet. (930 m, -290 m, volume 1.1 millions of m3), Gornja (Upper Cave) (1,292 m) and Donja (Lower Cave) Cerova ka peina (2,779 m), and Muda Labudova (2,490 m, -680 m).3. History of explorationsWhile the Upper Cerova ka Cave entrance was known to local people, Lower cave entrance was found during the railway building in 1913. New found cave brought large quantities of archaeological findings, sites of the cave bear, cave lion and amazing speleothem forms. Both caves in a total length of 4 km become most popular tourist cave site in Croatia. Modern cave explorations with rope techniques started in late seventies. Year after year just a few cavers succeeded to rich summit plateau. Members of Speleological section (SS) eljezni ar (Zagreb) in cooperation with others clubs have found impressive chambers in Burinka (160 90 90 m), Muniaba (185 60 70 m) and Veliko Grotlo pit (100 85 130 m). War in Croatia stopped further caving activity. Explorations continued few years after the end of the war. Members of SS eljezni ar have found dozens of new pits and caves (Michelangelo pit -274 m, Alibabina jama -218 m, Muda Labudova pit -680 m). Today, there are more than 250 known caves on the Crnopac. The SS PDS Velebit (Zagreb) continued exploration of Muniaba channels with massive use of alpine techniques; drilling, bolting and spending a lot of time in underground camps what led to the total length of more than 9 km. Using the cave diving techniques, the DDISKF Dinaridi explored Kusa iI Cave (3,010 m) and Krnjeza spring (sump, -106 m) in Zrmanja and Krupa river valleys the outputs of hydrologic system. In 2004, cavers from the SS St. Mihovil (ibenik) found lower entrance of Cave system of Kita Ga eina Draenova puhaljka. Cavers from the SS PDS Velebit and SS Mosor (Split) join them and new cave rapidly grew up.Figure 2. Projected section of large caves in Crnopac Massif: NS. Figure 3. Cave system of Kita Ga eina Draenova puhaljka 3D view. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings18

PAGE 21

The KGDP cave system in almost all parts and many surrounding caves characteristics strong air flow. In permanent air pressure during the summer it tends to came out from the entrances and during the winter its changes the direction. Cold air flows lowers the temperature and create ice speleothemes in areas close to the entrances. There must be some higher entrances placed close to the top of the Crnopac Massif (1,402 m a.s.l.) and connected to the lowest parts of the system. Measured temperatures in system vary from 3.7 to 5.4 C.5. Biospeleological researchBiospeleologists collected KGDP cave system fauna material only twice and collected 20 different species. First interpretation of results showed that organisms are similar to surrounding caves that were visited much more times by the biospeleologists. Some data still wait to be explained and there are some indications that there must be at least one new speciae. Cavers were collecting remains of cave bats and it is confirmed that at least 8 different bats lived in the cave system.6. Measurements of radon concentrationMeasurements of radon concentration were performed during 6 months in the upper level of the system at 7 locations. Concentration range, between 514 and 1,220 Bq.m-3and it is higher during summer. These values are comparable with the results of measurements in the deep pits of the northern Velebit (Paar et al. 2008) and significantly lower than the concentration of radon in umberak caves (Paar et al. 2005). In complex caves like Kita Ga eina, there is no simple connection of radon concentration with ground surface climate conditions. Future investigations will try to connect radon dynamic with cave microclimate and geology.Figure 4. Plans of largest caves on Crnopac Mt. 7. Nature Protection and tourism potentialExisting known values, diversity of forest cover and extraordinary landscapes with existing near tourist centers makes exceptional tourism potential. Opening of the forest road construction has made the area more accessible but in karst areas hitherto untouched forests have been felled. Speleological research results presentation initiates idea of establishing a more nature protection zone Speleological Park Crnopac what become one of the Zadar County goals (SPARC project). For now, cavers have succeed to stop intensive logging8. ConclusionCroatian cavers have found more than 250 caves, dozens kilometers of subterranean channels and impressive chambers showing Crnopac massif remarkable geomorphologic phenomena. In more than 8 years of intensive research Cave system of Kita Ga eina Drazenova puhaljka reached the length of 23,334 m, depth of -737 m and became longest cave in Croatia and Dinaric Karst. The whole area shows high speleological and tourism potential Despite undertaken geomorphological research, microclimatic measurements, radon, physico-chemical analysis of water, the collection and analysis of animal species, we can say that the systematic scientific investigation of the whole area is still in its infancy.AcknowledgmentsWe would like to thank to Velebit Nature Park, The Speleological Committee of the Croatian Mountaineering Association, Croatian Mountain Rescue Service. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings19

PAGE 22

ReferencesBarii T, Baki D, Paar D, Rnjak G, 2010. Kita Ga eina Draenova puhaljka Speleological Explorations 2004 and Some Characteristics of Cave System, 1. Croatian speleological Congress, Pore Bo i N, 2009. Cerova ke caves and other karst phenomena of the Crnopac Massif. International Interdisciplinary Scientific Conference Sustainability of the Karst Environment Dinaric karst and other karst regions, Plitvi ka jezera. Kova -Konrad P, Jali V, 2006. Prvi sifon pilje Kusa 1. Speleolog, Vol 54., 20, Zagreb (in Croatian). Kova evi A, 2008. Kusa nad Manastirskim lukama Kusa 2. Speleolog, Vol. 56, 61, Zagreb (in Croatian). Kuhta M, Stroj A, 2005. The Speleogenesis of the caves in Crnopac Mt. Area. Proceedings of the 14th International Congress of Speleology, 46, Hellenic Speleological Society, Athens. Kuhta M, Borovec M, Bosner N, 2002. Speleoloka istraivanja Crnopca u 2002. i 2003. godini. Speleolog, Vol. 50, 48, Zagreb (in Croatian). Luki O, 1988. Speleoloka istraivanja Crnopca na Velebitu. Speleolog, Vol. 36, 14, Zagreb (in Croatian). Paar D, Ujevi M, Baki D, Lackovi D, op A., Radoli V, 2008. Physical and Chemical Research in Velebita pit (Croatia), Acta Carsologica 37/2, 273.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings20

PAGE 23

CAVES OF EL PEON, CORDILLERA ORIENTAL, COLOMBIAMartin Bochud1,2, Roman Hapka1, Jean-Marc Jutzet1 1Splo Club des Pralpes fribourgeoises, 1700 Fribourg, Switzerland, info@scpf.info,2GeoAzimut Srl, Rte de la Fonderie 8, 1700 Fribourg, Switzerland, martin.bochud@geoazimut.com Abstract. In 2011, to celebrate their 40thanniversary, the Pralpes Fribourgeoises Caving Club (SCPF) of the Swiss Society of Speleology set out to explore the Colombian depths of South America. Following the discoveries made in 2011 in the El Peon karstic field located in the state of Santander, a second expedition was launched in JanuaryFebruary 2012. The Colombia reveals a country with a rich speleological heritage. As proof 10 km of galleries were discovered in a few weeks by a Swiss-Colombian team: the labyrinthine karst found at over 2,500 m, pastures punctuated by thick forests, tropical mists, imposing verticals where cave-dwelling birds of prey circle, mysterious rumbling underground rivers and fossil galleries with beautiful concretions. This article summarizes the discoveries made in 2011 and makes a first attempt to define the geological and hydrogeological context of the area. Abstrait. En 2011, loccasion de ses 40 ans dexistence, le Splo-Club des Pralpes Fribourgeoises (SCPF) de la Socit Suisse de Splologie sest lanc la conqute des abmes sud-amricains de Colombie. Suite aux dcouvertes ralises en 2011 dans le massif karstique dEl Penon, situ dans ltat de Santander, une seconde expdition a vu le jour en janvier fvrier 2012. La Colombie se rvle un pays au riche patrimoine splologique. Pour preuves, 10 km de galeries ont t dcouvertes en quelques semaines par une quipe suisso-colombienne: karst labyrinthique plus de 2,500 m daltitude, pturages entrecoups dune paisse fort, brumes tropicales, grandes verticales imposantes o tournoient les rapaces cavernicoles, rivires souterraines mystrieuses et grondantes et galeries fossiles magnifiquement concrtionnes. Cet article rsume les dcouvertes faites en 2011 et fait un premier essai dinterprtation gologique et hydrogologique de la zone explore.1. IntroductionIn 2011, the Pralpes fribourgeoises caving club (SCPF) decided to start an exploration of the Colombian karst. Together with Colombian cavers, they find the unexplored and interesting El Peon karstic area. After a successful first expedition in FebruaryMarch 2011 and the very friendly welcome of the local people, a second expedition was setup at the beginning of 2012. During both expeditions, more than 10 km of galleries were discovered and documented by the Swiss-Colombian team. The karst is located at an altitude of over than 2,500 m under pasture, thick forest and tropical mists. The climate of the area is humid to very humid. The high annual precipitations between 2.5 to 3 m combined to the favourable geology are ideal for the development of major cave systems. The anticline formed by the limestones of the Rosablanca Formation is drilled by imposing vertical pits and traversed by imposing decorated horizontal galleries. The speleological potential proves to be important and the two expeditions to date in this region have only scratched the surface.2. Short history of caving in ColombiaThe Colombian caves situated on the high central plateaux or in the low south-eastern forests were known to the Indian civilisations that used them as burial places and refuges. The first descriptions of caves, often picturesque made by foreign travellers, date from the mid-19thcentury. In 1878 Alexander von Humboldt published a work on the rifts and caves of the Cordillera. Various caves are mentioned in the first half of the 20thcentury but it was in 1940 that Luis Cuervo Marquez described the famous Hoyo del Aire (a pit cave of 120 m), remapped during Speleo Colombia in 2011 and various other caves. During the next 30 years various scientific authors became interested in Colombian karst but it was not until 1975 that a Polish caving expedition explored 24 caves using modern equipment. In 1977 a single reconnaissance expedition of the Colombian karst was carried out by the Speleological Group of Nice (France; Hof 1977). Three months of on-site presence resulted in an impressive work of exploration and data compilation extending over almost the entire country. More than 100 caves and pit caves were visited, explored and mapped. The Hermosura region, close to El Penon, drew attention for the first time, and two large pit caves including the Hoyo del Aguila (-149 m, 105 m pit) were explored. Other French and American incursions, more or less related to mineral and oil research, were occasionally reported later, without it being possible to find published material. For example, the list of the largest caves in the world indicates that the longest Colombian cave is the Hermosura Sistema (4,926 m, -193 m). However, the bibliographic references, the geographical location and topography currently cannot be found.3. Geography and geologyThe El Peon Massif (cities of El Peon and Bolivar, see Fig. 1) is located in the Santander State in the Eastern Cordillera of the Andes in Colombia about 160 km North of Bogota city. The area is sparsely populated and the few localities found here have only FWD cars or even no car access. The community of El Penon is only 50 years old and the access road is only a few decades old. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings21

PAGE 24

Structurally, the El Peon area is located on a main anticline structure with secondary folds forming the plateau where the caves are located. The main fold axis is dipping slightly to the NNE. As described by Coletta et al. (1990). The VelezEl Peon area is a large pop-up structure bordered by two opposite verging thrust fault involving Jurrassic and pre-Jurassic structure. The west verging thrusts are located to the West of the El Peon Area. The Cretaceous-Jurassic deposits thrust over Tertiary deposits. Geographical, hydrological and geological settings points on a speleological potential of more than 2,500 m. Indeed, the highest altitude in the SE of El Peon is about 3,000 m and the limestone deposits almost reach the altitude of 385 m where the two rivers draining most of the massif join in the west of the massif. The studied area lies at altitudes between 3,000 m in the SW and lower than 1,000 m in the NE. The southern part is bounded by a huge cliff on the Panama deep incised valley. From the SW to NE, the massif dips slowly down. In the NW, it dips fast to 400 m and then reaches the Magdalena Plain at 100 m. 3.1. Climate The climate is humid and annual precipitation is about 2.25 to 3 m. The mean annual temperature is about 22 C (Pabn-Caicedo et al. 2001). 3.2. Hydrology hydrogeology The karstic hydrogeological systems of the area have been little studied at present. Based on maps and satellite images, some watersprings seem to spread out along the WNW cliff of the Massif but they were still not observed. No major karst spring has been found yet. On the El Peon Massif, small rivers are present. They flow out of cave and they generally rapidly disappear by infiltration in doline or directly in other cave. The El Peon area is surrounded by the Rio Dorada to the S and W and by the Rio Horta to the NW and N. They seem to drain the main part of the water of the massif and they both join at an altitude of 385 m. In caves, rivers, lakes or impenetrable siphons can be observed at different levels. The lack of flooding cue in caves despite the high precipitation indicates that the water flow is certainly rapidly spread over the whole massif. 3.3. Geology Lithologically, the Cretaceous limestone extends over several hundred kilometers between cities of Bogota and Bucamaranga at altitudes between 50 and 3,000 m. The karst and all discovered caves of El Peon and surrounding areas are located in the Lower Cretaceous Rosablanca Formation (Late ValanginianEarly Hauterivian). The thickness of the whole formation can reach 400 m (Guerrero 2002). As described on Figure 2, it is composed of a succession of grey dolomites, grey limestones, brown limestones, marls, sandstone and finally limestone on the top (Guerrero 2002; Mendoza-Parada et al. 2009; Toussaint 1996). Based on our observations in the Panama Valley, the caves of El Peon area seem to be located in the lowermiddle part of the Rosablanca Formation in the grey and brown limestones. In some area, the thickness of this lower part of the formation can reach 200 m (Guerrero 2002).Figure 1. Location of El Peon Massif and Santander Department in Colombia. Figure 2. Rosablanca Formation in the Mesa de los Santos area (Toussaint 1996).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings22

PAGE 25

4. Area surveys and explorationThe two expeditions in 2011 and 2012 are the first of this kind to be carried out in this region of Colombia. This is the first systematic exploration and survey of the predominantly vertical high altitude karst caves. After a few weeks of survey and exploration, the surface of the karst field has barely been scratched. The dead end sumps, the impenetrable squeezes or cave-ins at depths exceeding 200 m, did not take anything from the pleasure of caving. In fact, these discoveries have succeeded one another at a frantic pace of nearly one cave per day. Future expeditions, in connection with a better hydrogeological knowledge of the region and search for the access to a larger system, will take cavers deep into the depth of the Rosablanca Limestones. Nine caves were explored and mapped in 2011, totalling 4,200 m of development. In 2012, 15 caves were mapped, totalling 4,500 m. About ten other important caves (pits of 100 m and more, areas of large cave entrances) were located and partially explored during area surveys. The Table 1 summarizes the discovered caves and their main characteristics.5. Field observation and discussion5.1. Cave morphology There are predominately three types of caves: shallow pits 50 to 150 m deep, sometimes giving access to large rooms or horizontal galleries; horizontal caves sometimes with 2 levels; and more or less impressive hollow shelters in cliffs leading sometimes to underground streams. With a depth of -219 m, the Hoyo de los Ocelotes is the deepest cave at present in the area whereas the Cueva de los Carracos is the cave with the most important development (1,500 m). The entrance pit of the Hoyo de la Neblina is the deepest of the area with its 147 m. Based on cave survey data (statistics based on measures of more than 20 m), two main gallery directions were emphased: N50 and N140 (Figure A). It is interesting to note that they both correspond to morphologic lineaments visible on satellite images. Numerous caves finish on a mud plug on a level corresponding to the base of the high pits. The mud level seems to correspond to the contact of limestone with a lower and less permeable geological layer at the base of the Rosablanca Formation that allows the accumulation of erosional products. In the Hoyo de los Ocelotes, two pits zones linked by a horizontal gallery were discovered and they both have a mud zone. These two levels could be explained by the presence of a vertical fault that moves down the Rosablanca Formation and allowed the formation of the second and deeper pits zone. The fault can be observed in a part of the gallery that link both zone. This part is a straight, high roofed and narrow (~2 m) gallery that follows directly the first mud zone. 5.2. Tectonic properties Only few measures were made until now. They are represented on the steoreogram of Figure B. Beddings are concordant with the main anticline structure that has an NNESSW fold axis. Faults can be separated in 2 groups: vertical faults dipping to the north at 58/89 and Figure 3. Pit in the Tronera cave (Jutzet 2012). Figure 4. Horizontal gallery in the Cueva de los Carracos (Bochud 2012). Figure 5. A) direction rosediagram of cave survey data of more than 20 m (150 measures); B) Lower hemispher equal area projection stereogram of faults and bedding great circles.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings23

PAGE 26

to the south at 222/87 and inclined faults (probably inverse) dipping to the south at about 223/36. The N140 gallery direction can be related to the faulting. The second main gallery direction N50 is, as said before related to a surface geomorphologic lineation but presently cannot be linked neither with the bedding nor the measured faults. It is probably linked to a conjugate fault system. Another observation can be made on the alignment of pits higher than 100 m on a vertical cross-section. On the Fig.7, five caves are represented in a 3 dimensional cross-section on a WE direction. High pits from 60 to 100 m can be linked with a synclinal structure that can be related to the structure of the limestones layer. Moreover, the high of the pit can be related to the thickness of the LowerMiddle part of the Rosablanca Formation that is estimated between 100 and 200 m. 5.3. Biospeleology and paleontology To date, no systematic work has been done on the El Peon area. Some interesting and punctual observations will be related. In several caves, the typical cave bird Gucharo ( Steatornis caripensis) can be observed. It is a night bird navigating by echolocation in the same way as bats. Some crabs were also observed until the bottom of the Hoyo de los Ocelotes at -219 m. Some bats and spiders were also seen. An interesting discovery was made in the Hoyo de los Ocelotes: a cranium with two canines (Fig. 6). As a first assumption, it was identified as an ocelot and it gave the name of the cave. Afterwards, the picture was sent to the specialist Michel Blant at the Swiss Institute of Speleology and karst Research and he determined that it was a Choloepus didactylus (Southern two-toed sloth).6. ConclusionsThe two expeditions Colombia 2011 and 2012 were carried out in the El Peon Massif in the Santander Department in Colombia. The karst morphology is well developed in this humid area with more than 2.5 m of precipitations per year. The El Peon area is structurally located on an anticline structure with a flattened hinge. Twenty-four caves were explored and documented in a relatively small area during two expeditions totalizing about 10 km of galleries. Geologically they developed in the limestones of lowermiddle part of the Lower Cretaceous Rosablanca Formation. The caves morphology seems to be closely related the structural deformation of the area. Although the deepest discovered caves stop presently at depth between 150 m, the speleological potential is estimated at 2,500 m. In term of scientific observations, no systematic work has been done until now except the detailed survey of the caves. But the few observations indicate that the El Peon area is very interesting in term of caves and karst morphology but also in term of biospeleology and paleontology. Future expeditions, in connection with a better hydrogeological knowledge of the region and search for the access to a larger system, will take us certainly deeper into the depth of the Rosablanca Limestones. Figure 6. The cranium Choloepus didactylus (Southern two-toed sloth discovered in the Hoyo de los Ocelotes (Bochud 2012). Figure 7. 3-dimensional cross-sections with 5 caves. It emphazises the alignment of high pits along a syncline formed by the limestones of the lower and middle parts of the Rosablanca Formation.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings24

PAGE 27

AcknowledgmentsA great thank to the participants of the 2 expeditions for their motivation and the good ambiance. A special thanks to Jesus Fernandez for the organisation of the expedition.ReferencesColletta B, Hebrard F, Letouzey J, Werner P, Rudkiewicz, J-L, 1990. Tectonic style and crustal structure of the Eastern Cordillera (Colombia) from a balanced cross-section, in: Letouzey, J. (Ed.), Petroleum and tectonics in Mobile Belts. Technip, Paris, 81. Guerrero J, 2002. A proposal on the Classification of the Systmes Tracks: Application to the Allostratigraphy and Sequence Stratigraphy of the Cretaceous Colombian Basin. Part 1: Berriasian to Hauterivian. Geologia Colombiana 27, 3. Hof B, 1977. Recherches splologiques en Colombie. Fdration Franaise de Splologie, G.S. Nice. Mendoza-Parada JE, Moreno-Murillo JM, Rodriguez-Orjuela G, 2009. Rosablanca Formation Karstic Systems lower Cretaceous in Vlez, Santander Province, Colombia. Geologia Colombiana 34, 36. Pabn-Caicedo JD, Eslava-Ramrez JA, Gmez-Torres RE, 2001. Generalidades de la distribucin espacial y temporal de la temperatura del aire y de la precipitation en Colombia. Meteorologia Colombiana 4, 47. Toussaint J-F, 1996. Evolucin geolgica de Colombia durante el Cretcico. Universidad Nacional de Colombia, Medellin. Figure 8. Map with the discovered caves in the El Peon area. Numbers are related to the Table.Caves of the la Hermosura area a re not represented on the map (B1, B2, B3, B4 and B5).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings25

PAGE 28

NoName (place) LengthDepthAltitudeYearComment [m][m][m a.s.l.] EP1Cueva de los Carracos (El Peon)1,500(-81,2,4882011Underground river with fossil gallery. +4)Explored by local people. EP2Hoyo del Caballo (El Peon)267(-149,2,5932011Swallow hole with river and sump. +10) EP3Cueva del Atun (El Peon)764-1592,5932011Stops on a sump in a big room. Sinking river at the entrance. EP4Hoyo de la Limpieza (El Peon)343-1472,6122011Swallow hole with river and sump. EP5Hoyo de la Basura (El Peon)25-152,6172011Swallow hole with river. Garbage. EP6Cueva de los Elechos (El Peon)50-52,6362011Small horizontal cave. EP7Hoyo de Pepero (El Peon)321-1672,6012011Swallow hole with river. EP8Hoyo de la Neblina (El Peon)715-2172,6862011Swallow hole with river and big room. EP9Cueva del Hipocampo (El Peon)913(-15,2011Cave with a river. +28)Numerous unexplored galleries. EP10Cueva del Caracol (El Peon)324-362,4102012Collapsed sinkhole with river and sump. EP11Cueva del Krypton (El Peon)~ 40~ -402011Swallow hole (no survey). EP12Cueva sin Fin (El Peon)~ 50~ -502011Swallow hole (no survey). EP13Cueva de la Casa Virgen (El Peon)102-782,5532011Swallow hole with river. EP14Hoyo de la Bollas de Oro (El Peon)159-1312,4222012Swallow hole with sinking river. EP15Cueva de las Escuillas (El Peon)310-532,3742011Collapsed sinkhole. EP16Hoyo de los Ocelotes (El Peon)812-2192,5442012Swallow hole with 2 entrances and a river. EP17Hoyo de las Professoras 1 (El Peon)130-922,6332012Swallow hole. EP18Hoyo de las Professoras 2 (El Peon)79-402,6362012Swallow hole. EP19Cueva de la Tronera (El Peon)1203-1672011Collapsed sinkhole swallow hole with river EP20Cueva de las Gallinas (El Peon)179302012Cave with a river. EP21Cueva de las Golondrinas5002,4122012Huge entrance. (Cruces, El Peon) EP22Cueva Grande (Cruces, El Peon)~400~ -402,4842012Huge entrance. EP23Cueva de los Murcielagos~ 300~ -202,4142012Cave with a river. (Cruces, El Peon) EP24Hoyo de los Golondrinas 1 (El Peon)77-472012Swallow hole on a fault. EP25Hoyo de los Golondrinas 2 (El Peon)94-482012Swallow hole on a fault. B1Hoyo del Campesino / del Aguila~ 120~ -1201977?Big pit, probably explored by Hof 1977. (La Hermosura, Bolivar) B2Hoyo de la Campesina~ 100~ -1001977?Big pit, probably explored by Hof 1977. (La Hermosura, Bolivar) B3Cueva de la Puerta de los Cerros524-2051,8352011Cave with a large gallery. (La Hermosura, Bolivar) B4Hoyo Horrible~ 100~ -1001977?Big pit, probably explored by Hof 1977. (La Hermosura, Bolivar) B5Cueva de Los Ossos 1-2-3~ 1,000~ -501,1302012Cave with 3 entrances (La Hermosura, Bolivar) and a river (not explored).Table 1. List of discovered caves in El Peon-Bolivar area (department of Santander, Colombia). Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings26

PAGE 29

RECENT SURVEY AND EXPLORATION IN LECHUGUILLA CAVE, NEW MEXICO, USAPeter Bosted1, John Lyles2 1PO Box 6254, Ocean View HI 96737 USA, peter@cavepics.com2PO Box 95, Los Alamos NM 87644 USA, jtml@vla.com Since the last ICS in 2009, considerable progress has been made in the survey and exploration of Lechuguilla Cave, located in Carlsbad Caverns National Park, New Mexico, USA. The most important new discovery is an area called Oz, reached by a long aid climb from Emerald City, in the western portion of the cave. Many smaller discoveries were made as well, bringing the total length to over over 218 km. In addition, a big effort was put into improving the survey quality of passages explored in the early years (1988). Most of the poor survey loop closures were fixed, and quadrangle maps were updated with improved detail.1. IntroductionLechuguilla Cave was known in the early 1900s as a relatively short cave with a 30 m entrance pit leading to a rich deposit of bat guano, which was mined for a time. After several unsuccessful attempts, cavers finally pushed through a breakdown choke in 1986 to discover a vast and beautiful cave (Reames, 1989). Over 100 km were explored and surveyed by the early 1990s, with the 200 km mark being passed in 2008. The cave is most famous not for its great length and volume, but rather for the spectacular cave formations, such as the long selenite crystals in the Chandelier Ballroom, or the impressive flowstone columns in Tower Place (Widmer, 1998). The overall structure of this single-entrance cave is large long borehole trunk passages that connect extremely complex three-dimensional mazes together. In recent years, the number of week-long exploration and survey expeditions has been reduced in number to just one every two to four months. Nonetheless, new discoveries remain to be made, with about 4 to 6 km of newly discovered passages being mapped each year. In parallel, there has been an increased effort to improve the quality of early surveys done in the frenzy years of 1988. This effort includes removing survey blunders, improving survey accuracy, and improving sketching quality. Since the beginning of 2009, there have been 26 week-long expeditions, with almost 18 km of new passage surveyed, and over 15 km of passage re-surveyed.2. New discoveriesMost of the newly discovered passages in the past few years have been relatively small, and have resulted from pushing smaller passages than was the norm in the early days of exploration. This mopping up process was greatly facilitated by cartographers drawing up detailed maps of their sections of the cave, making it easier to identify which passages had been surveyed already. Some of the most productive areas included Southern Climes in the Far West, the Chandelier Maze and Voids in the South, and the Outback in the Far East. Some spectacular example of cave formations were found, such as the Wasps Nest in the newly discovered Coral Sea area of the Far East, and individual gypsum flowers and selenite needles in Southern Climes each over 110 cm long. A large effort went into climbing domes. One such example was the six-day effort to aid climb the Capitol Dome above Mt. Vernon, in the Far East branch of the cave. Unfortunately, the passages at the top ended after only a short distance. Figure 1. Plan and profile line plots of Lechuguilla Cave. Green lines show passages mapped prior to 2009, while red lines show new discoveries or passages that have been re-mapped since 2009.More successful was the climb of the Kansas Twister, a 125-m-tall dome in Emerald City. The first 30 m were very difficult, and were finally conquered using a sling-shot to shoot a line over a bridge in 2010. The sling-shot technique proved un-successful for reaching the second bridge, so in 2012 a team climbed straight up the overhanging wall in very poor quality rock, reaching the summit five days later. On the last day of the expedition, a large new area was found, including big rooms, massive flowstone formations, and lakes. It was named Oz, following the naming theme of Emerald City below. Many leads remain to be explored, as of this writing. As can be seen in Figure 1, the Oz area is very high up in the cave stratigraphy, nearly matching the elevation of the entrance. Maps and photographs of this new area will be shown in the oral presentation.3. Blunder fixingThe method used for finding possible survey blunders used a computer code that simultaneously closes all the loops in a given section of the cave. The basic method is a leastsquares fit that minimizes that total corrections (added Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings27

PAGE 30

quadraticly, and called 2) to all of the segments, where a segment is defined as a sequence of survey shots that connect one junction to another. The corrections in each case are weighted by the expected errors based on station placement uncertainties of 5 cm, and compass and inclinometer readings with random errors of 2 degrees. Once all the segments were closed simultaneously, a search for likely blunders was made by removing one segment at a time, and re-closing the remaining loops, starting with the one with the largest contribution to 2. Using the junction locations defined by all other segments, a search is done for the removed segment to see what could make it close with a contribution to 2of less than two units. The following possibilities were examined: reversed inclinometer reading, reversed compass reading, both reversed, compass off by 100, inclinometer read on percent scale instead of degree scale, digits on distance reversed, and either starting or ending station being different that the one noted in the survey book. If any of these possibilities was considered to be probable, an examination of the survey notes was made to see if there had been a data entry error, or if the notes contradicted the sketch. In about half of the cases, the blunder was found in this manner. In the case where the problem could not be resolved by an examination of the survey notes, a report was generated for cavers to field check the problem. Generally, we re-measured all the front and back-sight shots in each bad loop, using the Disto-X instrument. We used the Auriga program, running on a PDA using Palm OS, to enter the new loop data and make sure that it closed well. Sometimes, we found a major blunder early on, and didnt need to re-shoot the entire loop. During 2008, we were able to reduce the fraction of loops that almost certainly contained a blunder from over 6% to about 2%. The area of the cave that we have concentrated on the most, the Far West, currently has no very bad loop closures, with closures worse than four standard deviations. The most common blunders were found to be (in order from most common to least common): a) One or more stations with the wrong station number written on them (in this project, every station is labelled with flagging tape). b) Front and back sights were reversed, and the sketcher used the reversed numbers to make the sketch. c) Compass and/or inclinometer readings were off by more than 2 for many shots in a sequence. d) Distance reading was wrong for one reason or another. e) Shot simply didnt make any sense upon field checking (for example, shot goes through solid wall). f) Station marker was moved from its original location.4. Future prospectsA big effort is currently underway to produce high-quality, detailed quadrangle maps of the entire cave surveyed to date. In support of this effort, we will attempt to identify and remove most of the remaining survey blunders, and continue to re-sketch areas for which the original survey notes were poor. In parallel, exploration of new areas such as Oz will continue. With hundreds of leads remaining in the cave, prospects are good for eventually passing the 250 km mark for the overall surveyed length of Lechuguilla Cave.AcknowledgmentsWe would like to acknowledge the efforts of the recent cave explorers and cartographers. Among the most active in the past four years are: Stan Allison, Andy Armstrong, Bonny Armstrong, Mark Andrich, Jeff Bartlett, Hazel Barton, Peter Bosted, Cathy Borer, Derek Bristol, Paul Burger, Daniel Chailloux, Andrea Croskley, Jen Foote, Art Fortini, James Hunter, Jean Krejka, Luc LeBlanc, Heather Levy, Vivian Loftin, John Lyles, Ron Miller, Annick Normandin, Carol Vesely, Simeon Warner, James Wells, and Max Wisshak. We also are very grateful to the National Park Service for their support, which includes a place to stay while working in the Park, maintaining the gates and ropes in the cave, providing supplies, entering the data into the computer, and archiving all of the survey notes. We especially acknowledge the support of Stan Allison, Cave Resource Technician at the Park.ReferencesReames S, Fish L, Burger P, Kambesis P, 1989. Deep Secrets: The Discovery and Exploration of Lechuguilla Cave. 412, Cave Books. ISBN-10: 0939748282. Widmer U, 1998. Lechuguilla, Jewel of the Underground. 168, Speleo Projects. ISBN-10:3908495016.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings28

PAGE 31

PLEISTOCENE SEA LEVEL CHANGES AS REVEALED BY FLANK MARGIN CAVES IN TELOGENETIC LIMESTONES IN SICILY AND SARDINIA (ITALY)Ilenia Maria DAngeli1, Jo De Waele1, Rosario Ruggieri 2, Laura Sanna 3 1Department of Biological, Geological and Environmental Sciences, Bologna University, Via Zamboni 67, 40126 Bologna, Italy, ileniamaria.dangeli@studio.unibo.it; jo.dewaele@unibo.it2Centro Ibleo di Ricerche Speleo-Idrogeologiche, Via Carducci 165, Ragusa, Italy University of Nova Gorica, info@cirs-ragusa.org3Department of Natural Sciences and Land, Sassari University, Via Piandanna 4, 07100 Sassari, Italy, speleokikers@tiscali.it Coastal karst areas often host many indices of past sea level changes, such as marine terraces, fossiliferous sediments, tidal notches and coastal caves. Tectonic movements can then displace these ancient coastlines vertically. The interplay between rising or falling sea level and uplifting or subsidence can be very complex and difficult to unravel. The combination of a detailed knowledge of marine terraces and the study of some flank margin caves located at various altitudes have allowed to reconstruct the speleogenetic history of the coastal plain of Cornino-Custonaci (NW Sicily). Along the centraleastern coast of Sardinia, instead, the detailed study of the Fico Cave has allowed to recognise it as a flank margin cave developed on five levels, related to Pleistocene sea level highstands. These studies show that this type of mixing corrosion caves is much more widespread than previously thought also in telogenetic limestones. These caves, being excellent sea level markers, might help coastal geomorphologists to understand more on both sea level rise and fall and tectonic movements in coastal areas.1. IntroductionGeomorphological evidences of past sea level changes are best preserved in coastal karst areas around the world. Carbonate shorelines exhibit some typical landforms created at or close to sea level, like tidal notches (Furlani et al. 2011), coastal caves (Mylroie and Carew 1988), biocorrosion (e.g., lithophaga borings) or bioconstruction (e.g., vermetid reefs) morphologies. These sea level markers, combined with other data derived from marine terraces and fossiliferous continental or marine sediments, have been used worldwide for sea level reconstructions (Dorale et al. 2010). It is well-known that more or less horizontally developed solutional caves in coastal areas often indicate the position of sea level at the time of their formation (Mylroie and Carew 1988; Florea et al. 2007). Among the different types of coastal caves, flank margin caves are the most reliable sea level indicators. In fact, flank margin caves form at the top and the bottom of the freshwater lens in coastal karst areas by mixing dissolution (Mylroie and Carew 1990), showing clear evidence of widening at the fresh-salt water boundary, often very close to sea level. Flank margin caves develop easily in young immature (eogenetic) limestones, where primary porosity facilitates the ingression of salt water and its mixing with seeping fresh water. Many of these caves have been reported from carbonate islands such as the Bahamas and Bermuda (Mylroie et al. 1995), Guam (Mylroie et al. 2001), Isla de Mona (Puerto Rico) (Frank et al. 1998), and the Mariana Islands (Jenson et al. 2006), or from eolian calcarenites such as the case of Kangaroo Island, Australia (Mylroie and Mylroie 2009). In more recent times flank margin caves have also been found in older and mature limestones, where mixing occurs along secondary permeability pathways such as fractures and bedding planes. Nice examples are described from New Zealand (Mylroie et al. 2008), and Croatia (Otonicar et al. 2010). Where fresh water discharge on the carbonate coast is greater, caves with a ramiform pattern will form. When fresh water discharge largely overrules the salt water entering the rock mass, caves will have very extensive stream passages developing inland. In this case speleogenesis is mainly controlled by normal dissolution and erosion processes, and cave altitudes do not necessarily reflect sea levels. The influence of mixing processes is visible only in the cave passages closest to the sea, fading slowly going landwards. Figure 1. Location of the study areas: 1) San Vito Lo Capo, Sicily; 2) Fico Cave, Sardinia.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings29

PAGE 32

Where fresh water discharge is slightly more important than salt water intrusion, ramiform caves develop, with passages modified by mixing corrosion processes also far inland. This type of caves, halfway between normal epigenic stream caves and flank margin caves, can be reliable sea level indicators only in their parts close to the coast (Smart et al. 2006). In this paper we describe three interesting Pleistocene flank margin caves in telogenetic limestones: two are located respectively at 100 and 70 m a.s.l. on the border of the Cornino-Custonaci plain in W Sicily, the other is situated at 14 m a.s.l. in the Gulf of Orosei, E Sardinia (Fig. 1). The timing of formation of these caves is based on a series of correlations with dated marine terraces, geomorphological observations, and dated speleothems and their phreatic organic overgrowths.2. Study areas2.1. Cornino-Custonaci plain The Cornino-Custonaci coastal plain is located on the western edge of the San Vito Lo Capo peninsula, halfway between Trapani and Palermo (W Sicily). The geology of the area is characterised by Mesozoic dolostones and limestones, and Eocene-Miocene limestones, marly limestones, marls and clays. Since Upper Pliocene the region is slowly uplifting (Nigro and Renda 2002). Extensive both continental, transitional, and marine Pleistocene deposits are present in the coastal plain, mostly composed of calcarenites (Di Maggio et al. 1999). The Fantasma and Rocca Rumena I caves are located in the Cretaceous limestones that form cliffs surrounding the coastal plain (Fig. 2). Seven marine terraces are present between altitudes of 0 and 105 m a.s.l. (Di Maggio et al. 1999). The age of these terraces has been estimated based on stratigraphical and paleontological evidences. The marine terrace of VIthorder, with upper boundary at 13 m a.s.l., contains the typical warm fauna with Strombus bubonius and can be attributed to MIS 5e. A speleothem sampled at 8 m a.s.l. in a notch related to this terrace at San Vito Lo Capo has given a U/Th age of 19,625 5,300 years BP, confirming this terrace to be older than LGM (Antonioli et al. 1999). Based on these observations, terrace VII should thus be related to MIS 5a or 5c, terrace V to MIS 7, terrace IV to MIS 9, terrace III to MIS 11, terrace II to MIS 13, and finally the highest terrace to MIS 15 or older. The presence of Elephas falconeri in some continental deposits at elevations above 40 m a.s.l., roughly corresponding to terraces III or IV, confirms their age older than MIS 9. A speleothem that covered the notch at 42 m a.s.l. at San Vito Lo Capo is beyond the U/Th radiometric dating method and indicates terrace III to be older than 300 ka BP (Antonioli et al. 1999). 2.2. Gulf of Orosei The Gulf of Orosei is characterised by the outcropping of Mesozoic dolostones and limestones that overly a Palaeozoic Variscan basement composed of granites and phyllites (Dieni and Massari 1985) (Fig. 3). Figure 2. Geological map of the Cornino-Custonaci plain. Figure 3. Geological map of the Gulf of Orosei.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings30

PAGE 33

This carbonate sequence forms a monocline ridge dipping gently eastwards to the centre of the Gulf, showing two transcurrent arcuate fault systems directed NNW-SSE and NNE-SSW (Pasci 1997). These Mesozoic rocks are locally covered with alluvial conglomerates and quartz sands, related to an intense erosion-deposition cycle caused by an uplifting phase of Middle Pliocene age (Massari and Dieni 1973). In the same period Pliocene basalts with K-Ar ages of 3.56 .23 to 1.99 .08 million years (Savelli and Pasini 1973) were put in place. Early Pleistocene alluvial sands and conglomerates are foundupon these basalts (Dieni and Massari 1966). During Pleistocene stratified slope-waste deposits (boulis ordonnes), and aeolian sands visible in karst pockets and coastal cave entrances have deposited. The Fico Cave is located at 14 metres height along the coastal limestone cliffs, about 20 km N of Santa Maria Navarrese. Detailed geomorphological studies have been carried out along the coastline, and especially in the Bue Marino Cave located 8 km to the North. Pleistocene basalts filling phreatic conduits in Bue Marino Cave (Mahler 1979), the Golgo shaft and Su Molente Cave testify the start of speleogenesis to be pre-Quaternary (De Waele 2004). Several tidal notches are well developed along the entire carbonate coast of the Gulf and indicate present and past mean sea levels. Such tidal notches are well described from depths of 10 m below sea level to heights of 10.5 m above sea level (Antonioli et al. 1999; Carobene and Pasini 1982). The relict tidal notch, that has a decreasing height from North to South between 10.5 and 7.7 m a.s.l., is attributed to the isotopic stage 5e (125,000 B.P.) and its continuity along the 37 km of limestone coast testifies a relative tectonic stability since, exception given for the slight N-S tilting (Antonioli et al. 1999). At the Fico Cave, although not visible anymore, its height would have been 8.5 m a.s.l.3. Cave morphology3.1. Fantasma Cave Fantasma (or Monte Cofano I) Cave opens at 198 m a.s.l. in the area of Contrada Marcato Gnarosa, close to marble quarries (Fig. 4) (Ruggieri and Messina Panfalone 2011). The cave follows two main fault systems, NESW with inclination of 75toward the SE, on which the entire first part of the cave is developed, and the second with WNWESE direction and dipping 80 toward the SSW that influences the deepest part of the cave. The first part is developed along a rapidly descending low and wide fissure with many indices of recent tectonic activity (broken columns and deflected stalagmites). At 120 m depth (around 80 m a.s.l.) the cave changes its morphology, with rounded passages, cupola, and a large ellyptical room called Sala del Fantasma, which centre is occupied by a large heap of old bat guano, while large white and corroded speleothems contrast the dark background. The northern part of this chamber is formed along a WNWESE fault with reddish mylonite creating a perfectly straight wall up to 10 m high. The roof and walls of this room have extremely well rounded shapes, and at around 70 m a.s.l. the walls are perforated by rounded tubes, forming a perfectly horizontal notch. Nowhere in the cave there are fluvial sediments, and the walls are always smooth and do not reveal any scallops or similar forms. The southern sector of the Fantasma room, however, hosts some evidences of phreatic (paragenetic) and vadose morphologies. 3.2. Rocca Rumena I Cave Rocca Rumena I is a small mostly horizontally developed cave formed at the foot of the 20 m high Pleistocene coastal cliff (Ruggieri and Messina Panfalone 2011). Its entrance opens at 100 m a.s.l. in correspondence of a not well-defined notch and a clear level of lithophaga borings. It is composed of a series of small chambers and passages with rounded forms, small tubes and pillars (Fig. 5). The passages develop parallel to the cliff wall, mostly along fractures and bedding planes, but tend to tighten gradually going away from the cliff. The roof and part of the walls are covered with organogenic crusts, mostly scleractinian corals that also fill part of the lithophaga holes. Also speleothems are covered with these corals and are perforated by lithophaga. One stalactite in particular shows three growth hiatuses, suggesting submersion at least three Figure 4. Map of the Fantasma Cave. Figure 5. Map of the Rocca Rumena I Cave.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings31

PAGE 34

times (Ruggieri et al. 2012). The corals have been dated using 87Sr/86Sr ratio and appear to be 1,100 200 ka years old (Antonioli et al. 2012). 18Omeasurements on the continental stalactite layers and comparison with marine and continental records suggest this speleothem to be formed during MIS 27, around a million years ago (Antonioli et al. 2012). 3.3. Fico Cave The Fico Cave has three entrances (respectively at -7, 0 and 14 m), and is characterised by a large passage running parallel to the coastline, and by a much smaller mostly submerged branch running perpendicular (Fig. 6). The development of most passages is clearly influenced by fractures directed NNWSSE and NESW. Water flow is present only along the small submerged branch, containing sediments deriving from the weathering of granites and carbonatic rocks, while the main branch has only few sediments mostly of autigenic origin. Nowhere, except from the active submerged branch, signs of phreatic flow have been observed (e.g., scallops). Close to the entrance some sands of aeolian and marine origin have been found at altitudes up to 7 m a.s.l. Cave passages tend to diminish in size the farher they develop from the coastline, ending abruptly on blind walls. Five Cave levels can clearly be recognised: the main branch at 14 m a.s.l., three fossil levels at 22, 40 and 50 m a.s.l. respectively, and the active submerged branch below present sea level. The level at 22 m a.s.l. has a perfectly flat ceiling, no sediments except for residual clays, and a nice corrosion notch on its walls, all testifying a relatively stable sea level. A stalagmite at 8 m a.s.l. in the Ramo del Lago shows organic overgrowths, testifying a submersion during its growth. Figure 6. Map of the Fico Cave.4. Results and conclusions4.1. Cornino-Custonaci plain Palaeontological and geomorphological evidences, combined with absolute datings on speleothems and corals, allow for a relatively detailed reconstruction of the speleogenetic phases and relative sea level changes in the Cornino-Custonaci plain. Rocca Rumena I Cave formed during an interglacial period, with relatively high sea level, around MIS 29. The Fantasma Cave, instead, is an enlarged (Middle Pleistocene) fault intercepted by a flank margin cave at 80 m a.s.l., located at around 500 m of the Pleistocene coastline, with the main level of mixing corrosion, corresponding to the longest stable position of the sea level, located at 70 m a.s.l.. Based on tectonic uplift rates calcolated using the Rocca Rumena date and the MIS 5e (terrace VI) (0.1 m/ka) (Antonioli et al., 2012) and other geomorphological constraints (Di Maggio et al. 1999), the horizontal maze-like chamber and conduits of Fantasma Cave might thus have been created by the mixing corrosion of fresh and salt water probably around 800,000 years ago (ca. MIS 19) (Fig. 7). 4.2. Gulf of Orosei The horizontal levels of Fico Cave have been formed during high sea level stands prior to MIS 5e. Using estimated uplift rates somewhat higher than those revealed by the Tyrrhenian notch (0.02 m/ka) and those estimated by Carobene and Pasini (0.03 m/ka), and considering that Figure 7. Evolution of the Cornino-Custonaci plain.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings32

PAGE 35

during MIS 7 the sea level highstand stayed around 10 m below present, the active submerged level would have formed during MIS 7 (ca. 250 ka), the main level at 15 m a.s.l. during MIS 9 (ca. 350 Ka), the 22 m a.s.l. one during MIS 11 (ca. 425 ka) and the upper levels would be older than MIS 21 (>865 Ka).AcknowledgmentsField work at Custonaci was made comfortable thanks to Davide Vito Messina Panfalone, while access to Fico Cave was guaranteed by the cavers of the Societ Speleologica Baunese.ReferencesAntonioli F, Montagna P, Caruso A, Ruggieri R, Lo Presti V, Silenzi S, Frank N, Douville E, Pierre C, 2012. Investigation of marine and continental layers in a stalactite older than 1 million years (Custonaci, north-western sector of Sicily), SLALOM 2012, Athens 19 March 2012, 57. Antonioli F, Silenzi S, Vittori E, Villani C, 1999. Sea level changes and tectonic mobility: precise measurements in three coastlines of Italy considered stable during the last 125 ky. Physics and Chemistry of the Earth (A), 24, 337. Carobene L, Pasini G, 1982. Contributo alla conoscenza del Pleistocene superiore e dellOlocene del Golfo di Orosei (Sardegna orientale). Bollettino della Societ Adriatica di Scienze, 64, 5. De Waele J, 2004. Geomorphologic evolution of a coastal karst: the Gulf of Orosei (Central-East Sardinia, Italy). Acta Carsologica, 33, 37. Di Maggio C, Incandela A, Masini F, Petruso D, Renda P, Simonelli C, Boschian G, 1999. Oscillazioni eustatiche, biocronologia dei depositi continentali quaternari e neotettonica nella Sicilia nord-occidentale (Penisola di San Vito Lo Capo Trapani). Il Quaternario, 12, 25. Dieni I, Massari F, 1966. Il neogene e il Quaternario dei dintorni di Orosei (Sardegna). Memorie della Societ Italiana di Scienze Naturali e del Museo Civico di Storia Naturale di Milano, 15, 91. Dieni I, Massari F, 1985. Mesozoic of Eastern Sardinia: 19th European Micropaleontological Colloquium-Guide Book, Cagliari, 1 October 1985, 66. Dorale JA, Onac BP, Fornos JJ, Gines J, Gines A, Tuccimei P, Peate DW, 2010. Sea-Level Highstand 81,000 Years Ago in Mallorca. Science, 327, 860. Florea LJ, Vacher HL, Donahue B., Naar D., 2007. Quaternary cave levels in peninsular Florida. Quaternary Science Reviews, 26, 1344. Frank EF, Mylroie JE, Troester J, Calvin Alexander E, Carew JL, 1998. Karst development and speleogenesis, Isla de Mona, Puerto Rico. Journal of Cave and Karst Studies, 60(2), 73. Furlani S, Cucchi F, Biolchi S, Odorico R., 2011. Notches in the northern Adriatic Sea: genesis and development. Quaternary International, 232, 158. Jenson JW, Keel TM, Mylroie JR, Mylroie JE, Stafford KW, Taborosi D, Wexel C, 2006. Karst of the Mariana Islands: The interaction of tectonics, glacio-eustasy, and freshwater/seawater mixing in island carbonates. In: RS Harmon and CM Wicks (eds.). Perspectives on Karst Geomorphology, Hydrology, and Geochemistry. Geological Society of America Special Paper, 404, 129. Mahler A, 1979. Verkarstung der Karbonatgebiete am Golfo di Orosei (Sardinien). Geologischer Palaeontologischer Mitteilungen Innsbruck, 7, 1. Massari F, Dieni I, 1973. La Formazione fluvio-lacustre di Nuraghe Casteddu ed i suoi rapporti con i basalti di OroseiDorgali (Sardegna). Memorie della Societ Geologica Italiana, 12, 377. Mylroie JE, Carew JL, 1988. Solution Conduits as Indicators of Late Quaternary Sea-Level Position. Quaternary Science Reviews, 7, 55. Mylroie JE, Carew JL, 1990. The flank margin model for dissolution cave development in carbonate platforms. Earth Surface Processes and Landforms, 15, 413. Mylroie JE, Mylroie JR, 2009. Caves as Sea Level and Uplift Indicators, Kangaroo Island, South Australia. Journal of Cave and Karst Studies, 71, 32. Mylroie JE, Carew JL, Vacher HL, 1995. Karst development in the Bahamas and Bermuda. In: HA Curran and B White (eds.). Terrestrial and Shallow Marine Geology of the Bahamas and Bermuda.Geological Society of America Special Paper, 300, 251. Mylroie JE, Jenson JW, Taborosi D, Jocson JMU, Vann DT, Wexel C, 2001. Karst features of Guam in terms of a general model of carbonate island karst. Journal of Cave and Karst Studies, 63, 9. Mylroie JE, Mylroie JR, Nelson CS, 2008. Flank margin cave development in telogenetic limestones of New Zealand. Acta Carsologica, 37, 15. Nigro F., Renda P., 2002. From Mesozoic extension to Tertiary collision: deformation patterns in the units of the NorthWestern Sicilian chain. Bollettino della Societ Geologica Italiana, 121, 87. Figure 8. Evolution of the Fico Cave.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings33

PAGE 36

Pasci S, 1997. Tertiary transcurrent tectonics of North-Central Sardinia. Bulletin de la Socit Gologique de France, 168, 301. Ruggieri R, Messina Panfalone D, 2011. Dentro e fuori la montagna. Priulla, Palermo 2011, 182. Ruggieri R, Messina Panfalone D, Antonioli F, Rosso A, San Filippo R, Maniscalco R, 2012. The Rumena Flank Margin cave and its implications in support of paleogeographic evolution of the Monti di Capo San Vito (north western Sicily). Rendiconti Online Societ Geologica Italiana, 21, 632. Savelli C, Pasini G, 1973. Preliminary results of K-Ar dating of basalts from Eastern Sardinia and the Gulf of Orosei. Giornale di Geologia, 39, 303. Smart PL, Beddows PA, Coke J, Doerr S, Smith S, Whitaker FF, 2006. Cave development on the Caribbean coast of the Yucatan Peninsula, Quintana Roo, Mexico. In: RS Harmon and CM Wicks (eds.). Perspectives on Karst Geomorphology, Hydrology, and Geochemistry. Geological Society of America Special Paper, 404, 105.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings34

PAGE 37

SOME SCIENTIFIC FEATURES OF THE PUERTO PRINCESA UNDERGROUND RIVER: ONE OF THE NEW 7 WONDERS OF NATURE (PALAWAN, PHILIPPINES)Antonio De Vivo1, Leonardo Piccini1,2, Paolo Forti1, Giovanni Badino1,3 1La Venta Esplorazioni Geografiche Via Priamo Tron, 35F 31100, Treviso Italy, t.devivo@laventa.it2Department of Earth Science Universit di Firenze3Nuclear Physic National Institute Universit di Torino This paper describes some of the most significant scientific issues of interest of a cave system in the island of Palawan, Philippines. The Puerto Princesa Underground River and the surrounding protected area represent a unique karst environment under several points of view. Its geographical position in the tropical area and its closeness to the sea, making it a true underground estuary, create unique conditions in the fields of hydrology, meteorology, mineralogy and biology. These aspects are analyzed as elements contributing to the comprehension of the exceptionality of this karst phenomenon. 1. IntroductionPalawan covers an area of 12,000 km2 and is located between 11 and 12 latitude N, and 117 and 120 longitude E, in the south-western part of the Philippines archipelago, not far from Borneo. The island is narrow, elongated and mostly mountainous. Two important NS depressions, corresponding to valleys or lowlands, divide the island into three tectonic sectors. Along the depression which divides the northern from the central sector, we find the Saint Paul Dome karst ridge (Fig. 1). Here, since 1991, the Puerto Princesa Subterranean River National Park was established in order to protect a unique underground system and the surrounding area. Originally covering only 3,900 ha, in 1999 the park was expanded to over 38,700 ha, and in the same year it was inscribed in the list of Unesco natural World Heritage Sites. Formerly managed by the Dept. for Environment and Natural Resources, since 1993 it falls under the rule of the City Government of Puerto Princesa. The first recorded explorations in the underground River date back to 1930, witnessed by some writings on the walls of the outflow and the area called Rockpile. In 1973 a first sketch of the main gallery of the cave was carried out by the Hungarian geologist Balazs (Balazs 1976). In the early 80s two expeditions from Australia explored the whole main branch and some side tributaries, pushing the total development of the cave to over 8 km (Hayllar 1980, 1981). Between 1986 and 1992 several expeditions were organized by the Italian Speleological Society, leading to the discovery of giant fossil branches which brought the total development to over 20 km (Piccini and Rossi 1994). The first decade of the new millennium has seen a methodic research project carried out by the Association La Venta from Italy, also assisted by interesting explorations by the Gaia Exploring Club from the Philippines (De Vivo et al. 2009). The last La Venta expedition, in 2011, has brought the development of the cave to 32 km (Piccini and De Vivo 2012). In 2012 the PPUR has been elected on of the New 7 Wonders of Nature.2. Geological and morphological overviewThe St. Paul area is located east of Ulugan Bay, about 50 km NE of Puerto Princesa. The length of the ridge, placed between the Babuyan River valley to the E and the Cabayugan River Valley to the W, is about 10 km and its average width is 4 km (Fig. 2). The carbonate rock covers an area of about 35 km2and is made of massive to roughly bedded, light to dark grey limestone showing levels rich in fossils. Such rock formation, more than 400 m thick, lays over sedimentary (mudstones, sandstones and marls) and volcanic rocks, dating back to OligoceneMiocene and covering an older metamorphic basement (Hashimoto 1973). The limestone outcrop is shaped as a NNESSW-elongated, asymmetric ridge sloping down to the west. From a structural point of view it consists of a multiple NW dipping homoclinal relief limited by NESW-oriented faults. Such lineaments have controlled both the general shape of the mountain and the karst landforms, determining the lining up of dolines and the development of the major caves. Figure 1. An aerial view of the Saint Paul Dome (1,028 m a.s.l.) from SE (photo L. Piccini, La Venta).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings35

PAGE 38

The main access to this cave is from the sea. On the side of a beach, a water stream carves a shallow canal through the coralline sand. Going upstream, after a few dozen metres you reach a small lagoon bounded by a calcareous cliff, at the foot of which a portal opens the way into the subterranean river. The main underground stream runs in a large tunnel, with lateral minor diverting passages and in some places upper old relict galleries joining with the active one through large chambers. This cave represents the main resurgence of the Cabayugan River, that disappears into the limestone 7 km upstream. Another marine spring cave is located a few hundred metres further E and collects the water sinking in the closed basins occurring along the east side of the karst in its northern sector. This cave is named the Little Underground River. Based upon the known hydrologic setting of the area at least two drainage systems can be identified, one drained by the Underground River and one by the Little Underground River. However, the occurrence of submarine and presently unknown springs cannot be excluded. As the entrance of the Underground River on the coast is at sea level, the tides affect a large part of the cave up to about 6 km inside the mountain (Forti et al. 1993). Along the whole navigable part, the salt water lies under a thin sheet of fresh water, just a few centimetres thick. This condition is only found during the dry season and in absence of rain. During the floods the cave is cleared of salt water, which later returns slowly once the flood has passed. Mixing phenomena between fresh and salt waters take place inside the cave causing particular processes of rock corrosion, which lead to the formation of water level notches (Mylroie and Carew 1988). In the past, sea level was often lower than now, so the PPUR has not always been a marine cave. Indeed, despite the occurrence of corrosion produced by the mixing of fresh water with saline water, the speleogenesis of the PPUR is mainly due to solution by continental water and to mechanical erosion by suspended load during floods (Fig. 3). Only in its downstream part, mixing corrosion has produced typical forms of coastal caves, such as waterline notches, spongework and lateral conduits. From this perspective, the system may be considered a classic example of an underground estuary. The landscape of the area is a typical tropical karst consisting of towers, cones, pinnacles, and large depressions, occurring mainly in the northern and southern sectors of the ridge. Large closed depressions (cockpits and dolines) cover about 10% of the total limestone surface. Major depressions occur in the form of elongated blind valleys on the east side of the northern zone, and are mainly developed on clastic rocks. Steep slopes and calcareous cliffs characterise the central part of the St. Paul Ridge, while to the north the landscape consists of an irregular plateau with several dolines. Large and deep depressions occur along the eastern limit of the limestone outcrop and can actually be considered as small blind valleys (Fig. 2). Some of the eastern sinking streams, located at 290 and 250 m a.s.l., feed two active Nagbituka caves), which consist of large tunnels, mainly vadose in origin, descending to the north along the contact between limestone and sandstone. At present, Nagbituka 1 is the deepest cave in the Philippines (Laumanns 2010). The southern part of the St. Paul Ridge has a different morphology and is characterised by two mountains, which have plain summit surfaces gently descending toward the NW, intersected by large and deep elongated depressions and large sinkholes. Along the coastline, a notch, located about 6 m above the present sea level, testifies the sea-level highstand which occurred about 120,000 years ago, when the mountain glaciers and polar ice-sheets had a much smaller extension than today. A similar notch can also be found inside the Underground River, up to 4 km from the entrance and at the same elevation (Piccini and Iandelli 2011).3. Underground HydrologyThe Underground (or Subterranean) River is a cave stream about 8 km long which collects the water from a wide surface basin, diverting it to the sea through huge galleries flooded by water. Figure 2. Geological sketch map of Saint Paul karst area. Figure 3. The Underground River just downstream the inflow, where the effect of tides does not occur. The lower limit of flowstone on the left wall indicates the water level during flood (photo by R. De Luca, La Venta).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings36

PAGE 39

This cave with its numerous branches houses, in fact, one of the worlds largest and most important underground ecosystems. There are hundreds of thousands of swiftlets ( salangane) nests, and almost as many bats. Seemingly, the salangane and bats have reached an agreement about the timing of their entry and exit from the cave, in general, when the bats begin to exit, almost all the swallows are back on their nests. Because of the great amount of organic matter that swallows and bats bring daily into the cave, there are many other animals present, made up of reptiles (snakes), fishes, crustaceans, and insects (Piccini and Rossi 1994). A fact that has always been surprising is that the cave runs all along the north-western side of the Saint Paul ridge, this means that most of the karst surface lies on the right-hand (eastern) side of the cave; despite this the Underground River has no consistent tributary from the right side. Why? Where does the water infiltrated all over the karst area flow to? How does it reach the sea? The first part of the cave, close to the seaward exit, presents a very complex pattern. Several branches divert from the main gallery. One of these lateral branches, just 150 m upstream of the outlet, was explored by the Filipino cavers of the Gaia Exploring Club in 2003. A large gallery, which they reached after a narrow flooded corridor, goes towards the E and closes at a huge stalagmite formation, 15 m high. Beyond this obstacle, which stopped the exploration of the Gaia Club, a second dry and huge system of galleries exists: it is like entering another cave, a parallel system of new amazing galleries that are flooded only during very rainy seasons (Fig. 4). The first part consists of a high gallery, about 10 m wide and 40 m high. The floor is formed by a sand deposit. On the right an upper sector hosts a hall with very nice helictites. On the left we reach a part of the passage where the floor is covered by extraordinary calcite crystals. This branch probably represents the high inactive level of the main drainage pathway of the cave (Fig. 5). In several points it is possible to descend in little branches that have typical epiphreatic morphologies until you reach a network of small horizontal often waterfilled tunnels, which lay at sea level. We can argue that the main tunnels collecting the infiltration water of the karst area lay now below the sea level, due to the present high-stand condition and that a new epiphreatic system is now developing. During the rainy season, the water level rises and inundates the upper galleries.4. GeomorphologyThe Underground River has a long and multiphase evolution history which is strictly linked with the sea level fluctuation and with the uplift phases (Piccini and Iandelli 2011). Features that indicate former water levels are present along the PPUR up to 5 or 6 km upstream from the coastal spring. In the last sector of the navigable path, where the ceiling of the main tunnel rises up to 20 m or more, there are two evident old corrosion notches carved by persistent levels of water. The upper notch is at 11 m above present mean sea level (pmsl). The second notch is at 7 m above the pmsl. Some of the lateral branches contain alluvial terraces consisting of sands and gravels, which are related to this second high-stand notch. This circumstance allows us to correlate the lower notch to the marine one, visible on the coastal cliff at about 7 m above sea level (Maeda et al. 2004; Omura et al. 2004), which dates back to the last interglacial phase (MIS 5e, about 125,000 years before the present time; Linsley 1996). The morphology of the active level of the PPUR is clearly adjusted to the current sea level, but we have to consider that in the last 500,000 years, the sea was most of the time lower than now (mainly 50 m lower; Haq and Hardenbol 1987). This implies that the PPUR has functioned as a vadose through-cave affected by fresh water flow, with a substantial load of insoluble material, forming a subterranean canyon that is buried by the alluvial sediments that form the current riverbed. The PPUR profile also shows several large passages at an elevation of mainly 50 m above the pmsl. This level consists of large inactive tunnels parallel to the current river containing thick alluvial deposits covered by flowstones, which in places almost completely fill the conduits. In the upstream sector of the cave, erosion forms may be found, indicating a long phase of vadose entrenchment. This ancient underground river could reasonably be dated back to the Early Pleistocene, as suggested by the extrapolation of the recent low uplift rate of the coastal zone (Piccini and Iandelli 2011). Several morphologic features, such as the presence of corrosion notches at 12.4 m above the pmsl, and the huge Figure 4. The huge canyon of 150 Years Gallery formed by downcutting of a large epi-phreatic tube (photo by A. Romeo, La Venta).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings37

PAGE 40

speleothems corroded and interbedded with alluvial deposits, suggest that this lower and presently active level experienced more than two sea level highstands and could have formed during most of the Middle-Late Pleistocene.5. Speleothems and cave mineralsHuge stalagmites and flowstones are abundant in several parts of the cave together with some rare speleothems. One of these speleothems is calcite grass, an extremely rare kind of helictite, consisting of bended and twisted calcite monocrystals growing from the cave floor and covering several tens of square metres of the 150 Years Gallery (Fig. 6). From the mineralogical point of view, recent and still ongoing investigations are evidencing that the number of cave minerals is much higher than supposed. Most of the minerals of the system are concentrated within the black crusts, which are widespread along the main galleries of the PPUR. These crusts, often completely detached from the cave wall, are related to periods of strong biogenic reactions induced by the mineralization of guano in water filled conditions. The same process was active in all the Figure 5. Survey of the PPUR in plan view.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings38

PAGE 41

other caves of the St. Paul karst: in two of these cavities the process also allowed peculiar black cave pearls to develop (Forti et al. 1991). In general, the crusts consist of several very thin layers of different colour: beside the far more dominant black ones, other layers are present from black to reddish and from yellow to white. Each colour corresponds to one or more compounds. In the black layers most of the minerals are manganese compounds, the reddish are characterised by iron-manganese minerals, while the white and yellow ones consist of gypsum and phosphates, many of which are amorphous. Up to now eleven different minerals have been detected in the PPUR: calcite (CaCO3), gypsum (CaSO4H2O), apatite [Ca5(PO4)3(C,F,Cl,O,OH)], variscite [AlPO4H2O)], strengite [(Fe, Al)PO4H2O)], manganite [MnO(OH)], rhodochrosite (MnCO3) pyrolusite (MnO2), robertsite [Ca6Mn9(PO4)9O6(H2O)6(H2O)], janggunite [Mn5x(Mn,Fe)1-xO8(OH)6] and serrabrancaite [MnPO4H2O)]. The first 8 minerals were already known from the cavern environment while the last three (robertsite, janggunite and serrabrancaite) are new cave minerals, being restricted to the single St. Paul karst: in particular serrabrancaite is very rare even outside caves, being found before only once in Brazil.6. The Sirenian fossilDuring the scientific explorations carried out by La Venta Esplorazioni Geografiche inside the Puerto Princesa Underground River (FebruaryMarch 2011), several partially articulated bones (ribs and dorsal vertebrae) exposed for approximately 10 cm of their length, were observed about 3 m above the river (Fig. 7). The bones emerge from the rock wall due to the differential dissolution of the limestone with respect to bones and were not extracted to preserve this rare paleontological site. A detailed comparison of the bones with several available photos of closely related taxa, and with the fossil specimens present in many museums all over the world, has allowed us to state that they belong to a sirenid. The specimen discovered on the Island of Palawan represents the first from the Philippines and the easternmost occurrence in the region. The partially articulated thoracic elements include a well preserved vertebra with exposed neural spine and processes, partial centrum and complete articulation with the head of the ribs. Several bone fragments are well preserved in situ, showing broad and re-curved ribs typical of sirenians. The specimen is approximately 60 cm wide, whereas the vertebra is 10 cm wide and 15 cm high including the centrum. It is therefore possible to estimate a total length for the individual of 180 cm. This sirenid is obviously coeval with the hosting rock, the St. Paul Limestone, which is Oligo-Miocene in age.7. Underground meteorologyThe Underground River is a cave into which an external river flows and which transports all the meteorological events from the outside deep into the rocks: precipitation peaks, large biological load and temperature jumps. The peculiarity which makes this cave so unique is that it reaches the sea level already within the mountain, in the core of Mt. Saint Paul. The flows of fresh water from upstream and salt water from downstream are not connected to each other and, on the contrary, for long distances they dont mix, the former flowing on top of the latter. This creates pockets of cooler air in contact with the water and, seasonally, generates flows of relatively warm and humid air which fills the galleries. Micro-meteorological niches and extremely complicated ecosystems are thus created, each having their own seasons, which at the moment we have only barely glimpsed. The island of Palawan lies directly on the thermal equator, therefore the UR cave is at the maximum temperature possible for a cave created by external water flows. Figure 6. The crystallized pavement in the 150 Years Gallery (photo by A. Romeo La Venta). Figure 7. The bones of a Sirenian fossil along the main active gallery of PPUR (photo N. Russo La Venta).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings39

PAGE 42

Then there is the wind, which is very noticeable when you pass through the portal to visit the cave on the bancas. The galleries are crossed by an important airflow, with peaks of up to 150 m2.s-1. The origin of air circulation within caves is usually the temperature difference between inside and outside. But, unique also in this, the PPUR is affected by a subtle factor which is insignificant elsewhere: the variations in the humidity of the outside air. The cave has a very peculiar characteristic. On one hand it is very hot, as it opens in an area where the average temperature is especially high in fact the highest possible on the other, it opens in an area where the climate is super-oceanic, that is characterised by a small in fact, the smallest possible thermal excursion between day and night and summer and winter. The result is therefore that the temperature difference between inside and outside is always small, just a few degrees difference, and it is very ineffective in creating an underground air circulation. In these conditions, a variation in the outside air density due to humidity variations can assume a decisive role. So, one of the powers of the underground storm is the humidity of the outside air, which in other places doesnt manage to move anything.8. Final notesThe Puerto Princesa Underground River represents a unique karst environment under several points of view. To start with, being an underground estuary, the PPUR hydrology is strongly affected by the sea tides, that reach up to 6 km inside the cave, with salt/fresh water mixture effects heavily influencing the cave genesis. On the mineralogical front, the huge fossil levels above the main gallery, particularly the recently explored 150 Years Gallery, contain extremely interesting and rare cave minerals and speleothems, some of which new to science and still under study. Absolutely unique is the presence of a sirenid fossil englobed in the wall of Gods Highway, along the caves main active branch. The paleontological specimen may supply interesting information on the geological history of the area and the cave genesis. Among others, the biological aspect is also extremely significant. The presence of huge colonies of bats and swallows and their guano production, together with the biological mass carried inside the cave by the underground river, has allowed the development of a complex and rich trophic network. From the ecological point of view, the cave presents three distinct ecosystems, each characterised by the different nature and abundance of the trophic resources. These features make the PPUR system an extraordinary natural laboratory to study the evolutionary processes and the ecology of hypogean environments. At present, the PPUR has become the most important show cave of the Philippines, with over 150,000 visitors per year. Its inscription in the New 7 Wonders of Nature will unavoidably increase the anthropic pressure on this cave. If the presence of the tides inside the cave is sufficient to guarantee the stability of its microclimate, the same cannot be guaranteed regarding its ecosystem, which might undergo serious alterations. It is thus absolutely necessary that the cave managing authorities pay particular attention to the problems that might arise from its tourist overexploitation.AcknowledgementsPartners: Societ Speleologica Italiana, Club Alpino Italiano, City of Puerto Princesa, Puerto Princesa Subterranean River National Park. Sponsor: Ferrino, Dolomite, Chelab, GT Line, Allemano Metrology, Intermatica, Amphibious, New Foods. Special thanks to: Mayor Edward Hagedorn, Rebecca Labit, Anthony Russell Abaya, Jamas Mendoza and the PPSRNP Staff.ReferencesBalazs D, 1976. Karst Types in the Philippines. Proc. 6thInternational Congress of Speleology, vol. 2, Praha (1973): 13. De Vivo A, Piccini L, Mecchia M, 2009. Recent explorations in the St. Paul karst (Palawan, Philippines). Proc. XVth International Congress of Speleology, Kerrville, Texas (USA), July 19 2009, vol. 3: 1786. Forti P, Gorgoni C, Piccini L, Rossi A, 1991. Studio mineralogico e genetico delle pisoliti nere della Lyon Cave (Palawan Filippine). Le Grotte dItalia, 4(15), 59. Forti P, Piccini L, Rossi G, Zorzin R, 1993. Note preliminari sullidrodinamica del sistema carsico di St. Paul (Palawan, Filippine). Bulletin Societ Geographique de Liege 29: 37. Haq BU, Hardenbol J, Vail PR, 1987. Chronology of fluctuating sea level since the Triassic. Science, 235, 1156167. Hashimoto ST, 1973. Geologic Structure of North Palawan and its bearing on the Geological History of the Philippines. Geology and Paleontology of Southeast Asia, 13, 145. Hayllar T, 1980. A description of the St Paul Cave, Palawan, Philippines. The Journal of the Sydney Speleological Society, 24 (7), 153. Hayllar T, 1981. Caving on Palawan. The Journal of the Sydney Speleological Society, 25 (12), 215. Laumanns M, 2010. Philippines. In: Laumanns M, Prize L, Atlas of the Great Caves and the Karst of Southeast Asia. Berliner Hhlenkundliche Berichte (BHB) 40: 203. Linsley BK, 1996. Oxygen isotope evidence of sea level and climatic variations in the Sulu Sea over the past 150,000 years. Nature, 380, 234. Maeda Y, Siringana F, Omurab A, Berdina R, Hosonob Y, Atsumib S, Nakamurac T, 2004. Higher-than-present Holocene mean sea levels in Ilocos, Palawan and Samar, Philippines. Quaternary International, 11516, 15. Mylroie JE, Carew JL, 1988. Solution conduits as indicators of Late Quaternary sea level position. Quaternary Science Reviews, 7, 55. Omura A, Maeda Y, Kawana T, Siringan FP, Berdi RD, 2004. U-series dates of Pleistocene corals and their implications to the paleo-sea levels and the vertical displacement in the Central Philippines. Quaternary International, 11516, 3.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings40

PAGE 43

Piccini L, Rossi G, (Eds.), 1994. Le esplorazioni speleologiche italiane nellIsola di Palawan, Filippine Italian caving exploration in the island of Palawan, Philippines. Speleologia, 31, 5 (ItalianEnglish). Piccini L, De Vivo A, 2012. Il carso del Monte Saint Paul (Palawan Filippine). Speleologia, 66, 46. Piccini L, Iandelli N, 2011. Tectonic uplift, sea level changes and evolution of a coastal karst: the Saint Paul Mountan (Palawan, Philippines). Earth Surface Processes and Landforms, 36, 594.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings41

PAGE 44

SUBTERRANEAN GLACIAL SPILLWAYS: AN EXAMPLE FROM THE KARST OF SOUTH WALES, UKAndrew R Farrant1, Michael J Simms2, Steven R Noble1 1British Geological Survey, Keyworth, Nottingham, NG12 5GG, UK, arf@bgs.ac.uk2Department of Geology, National Museums Northern Ireland, Cultra, Holywood, Co. Down, BT18 0EU, Northern Ireland, michael.simms@nmni.com Many karst areas in the UK have been glaciated one or more times during the last 0.5 Ma, yet there are few documented examples of caves in these regions being affected by glacial processes other than erosion. The karst of South Wales is one area where sub or pro-glacial modification of pre-existing caves is thought to occur. Evidence from the Ogof Draenen cave system suggests that caves can sometimes act as subterranean glacial underspill channels for melt-water. This cave, one of the longest in Britain with a surveyed length of over 70 km, underlies the interfluve between two glaciated valleys. Sediment fills and speleo-morphological observations indicate that melt-water from a high level glacier in the Afon Lwyd valley (>340 m a.s.l.) filled part of the cave and over-spilled into the neighbouring Usk valley, temporarily reversing nonglacial groundwater flow directions in the cave. It is suggested that this may have occurred during a Middle Pleistocene glaciation. (Note: Welsh terms used in this paper: Ogof = Cave, Afon = River, Cwm = Valley, Mynydd = Mountain).1. IntroductionMost of the upland karst areas in the north and west of the UK have been glaciated several times during the past 0.5 Ma, particularly during the Marine Isotope Stage (MIS) 12 Anglian glaciation (equivalent to the Elsterian of northwest Europe) and the MIS 2-4 Devensian glaciation (equivalent to the Weichselian), but also during Middle Pleistocene glaciations. Glaciations can have profound and complex effects upon karst landforms and their underlying aquifers, and may destroy, inhibit, preserve, or stimulate karst development (Ford 1987). Sub-glacial water flow can be considerable, especially in active, warm based glaciers, and at the margins of glaciers and ice sheets. Where these are in contact with karstified aquifers, there is scope for significant input of allogenic melt water into pre-existing cave systems (Lauritzen 1984, 1986). The impact of glaciations on caves in the UK is poorly understood, with few documented examples of caves being affected by glacial activity except through valley incision and erosion. One example of a cave system being affected by glacial activity is Dale Barn Cave in the Yorkshire Dales karst of northwest England, which is thought to have been reactivated by sub glacial melt waters (Murphy et al. 2001). This largely phreatic system connects two adjacent glacial valleys, Chapel-le-Dale and Kingsdale near Ingleton, Yorkshire. Speleothems dated to 343.4 (+86.0/-47.7) ka within the Illusion Pot section of the cave suggest that the conduit was drained during or prior to Marine Isotope Stage10. However, scalloping on these speleothems suggest it underwent a second, later phreatic episode, with water forced uphill from Chapel-le-Dale to Kingsdale, a distance of approximately 1.5 km. Sediments preserved within the conduit contain distinctive lithologies derived from the Chapel-le-Dale end of the cave, and resemble those described from sub-glacial eskers. The reactivation of the conduit is most plausibly explained by sub-glacial drainage near the snout of the Chapel-le-Dale glacier flowing through to the higher, but unglaciated Kingsdale valley. In the karst of South Wales, there is further evidence for glacial modification of pre-existing cave systems through melt water recharge. Copious amounts of sediment have been introduced into some caves by glacial melt-water, notably those under Mynydd Llangattock (Smart and Gardner 1988). This has led to ponding and localized paragenesis; blocking some passages, reactivating others and in some cases, facilitating the development of new conduits (Farrant and Smart 2011). In Ogof Agen Allwedd, Simms and Hunt (2007) provide evidence of sediment influx, glacial flooding and impoundment and suggest that glacial damming and recharge from melt-water might have been a significant factor in its development. The discovery of a major new cave system Ogof Draenen in 1994 (Farrant and Simms 2011) a few kilometers to the southeast of Mynydd Llangattock has provided additional evidence for the utilization of caves by glacial melt water.2. Ogof DraenenOgof Draenen [National Grid Reference SO 24631 1179] is a complex multiphase intrastratal cave system located six kilometres south-west of Abergavenny, South Wales (Figure1). It is currently one of the longest cave systems in the UK at more than 70 km in length and spanning over 150 m in elevation (Chelsea Speleological Society and Stevens 1997; Waltham et al. 1997). The cave is developed within Lower Carboniferous limestones on the northeastern margin of the South Wales coalfield. The limestones are overlain by Upper Carboniferous siliciclastics including the Twrch Sandstone Formation (Millstone Grit) and the Coal Measures, a cyclical sequence of sandstones and mudstones with some coal seams (Barclay 1989). The geological structure is relatively simple, with dips to the south-west at between 5 and 20. Ogof Draenen underlies the interfluve between the deeply incised Usk valley and its tributaries to the north and east, and the smaller Afon Lwyd valley to the west. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings42

PAGE 45

The cave has had a long and complex evolutionary history (Farrant and Simms 2011; Simms and Farrant 2011). It essentially comprises three vertically stacked, genetically separate, cave systems linked by phreatic under-captures, shaft drains, chance passage intersections and invasive vadose inlets. The present autogenic catchment is very small as the limestone forms only a relatively narrow outcrop along the steep scarp of the Usk valley. Consequently, recharge throughout the caves history has been predominantly allogenic, derived largely from numeroussmall streams draining the Upper Carboniferous clastics that cap the escarpment above the cave. The higher level passages in the cave system (those that lie up dip to the east of the Score-Gilwern Passage conduit) are considered to have drained south, initially to resurgences at c. 360 m above sea-level (asl) in the Usk valley (Farrant and Simms 2011), and then to progressively lower resurgences in the Afon Lwyd valley at 360 m a.s.l. Subsequent cave development occurred in response to valley incision in the Clydach Gorge to the north, effectively reversing the hydraulic gradient. This change facilitated the development of a lower level series of passages (the Score-Gilwern Passage conduit) that flowed northwest to a former resurgence in the Clydach Gorge at 320 m a.s.l. (Figure 2). Subsequent valley incision in the Afon Lwyd valley caused a second reversal in flow direction, this time to the south, creating the present Beyond a Choke stream-way, which resurges 10 km to the south near Pontypool at 120 m a.s.l. This streamway is fed by a series of small inlets and stream sinks along the margin of the overlying sandstone cover.3. Cave sedimentsObservation of the sediment fills in and around the northern end of the main streamway and its tributaries (Gilwern Passage, Upstream Passage, The Score and Pen-y-Galchen Passages; Figure 2) suggested that two distinct sediment Figure 1. Nextmap hill-shaded surface model image of the northeastern part of the South Wales coalfield and the Usk valley, showing the location of Ogof Draenen. Figure 2. Outline centre-line survey of Ogof Draenen, adapted from surveys by Chelsea Speleological Society and Stevens (1997). A. The northwestern part of the cave. B. Inset of area around the cave entrance. The black passages are those developed during the ScoreGilwern Passage conduit phase of development, whilst the Beyond a Choke streamway (dark grey) represents the final phase of cav e development. Directions of water flow are those when the passage was formed. The rest of the cave is shaded pale grey .Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings43

PAGE 46

can be observed throughout the upper reaches of the streamway (Upstream Passage) and surrounding passages. Sedimentary structures are often picked out by conspicuous, very distinctive, dark grey or black laminae, comprised of coal, carbonaceous or manganese stained material. These cross-bedded sands are more typical of the channel deposits of Bosch and White (2007), but locally are capped by laminated sediments of the slack-water facies. Excellent exposures occur in Gilwern, Upstream and Peny-Galchen passages. At the southern end of Gilwern Passage (307 m a.s.l.), >2 m of fineto medium-, locally coarse-grained, cross-bedded sand with coal-rich laminae overlain by a few centimetres of fine grained, laminated silts can be seen. Cross bedding here and ripple marks 200m farther downstream indicate northward flow. Significantly, no other deposits are known in the rest of Gilwern Passage and its northerly continuation. A similar sequence can be seen in Upstream Passage, 50 m north of Cairn Junction at 321 m a.s.l. Here up to two metres of fineto medium-grained dark grey sands overlying boulders are capped by 1 m of laminated silts (Figure 4). Another 100 m upstream at the same elevation, plaques of cross bedded sands (Figure 5) can be seen high up on the walls, at least 4 m above the current floor and within a couple of metres of the roof. The 0.5 m high northerly dipping foresets are picked out by the dark grey and black carbonaceous or manganiferous laminae. A short distance further on the large passage ends in a sediment choke, again comprised of fine-grained dark grey sands with ripple cross-lamination showing northward flow. Remnants of similar, but coarser sand again with northerly dipping cross beds, and sometimes cemented by calcite can be seen on the walls of Pen-y-Galchen Passage. Similar coal-rich sediments occur further south in The Score, an inlet passage off White Arch Passage, at 313 m asl. This passage is the upstream continuation of the Gilwern Passage conduit (Farrant and Simms 2011), and contains an abundant sandy fill throughout, as does its southerly continuation (Crystal Mole passage and Pontypool or Bust). In the Entrance Series, a conspicuous tide-mark is present on the passage walls up to an elevation of c. 325 m, indicating the maximum sediment fill level. facies occur in this area. These comprise coarse, poorly sorted sands and gravels in the active streamway; and relict, finer-grained, silt, sand and sandy gravel preserved at higher levels. To characterize these sediments, samples were collected from over 30 sites and subjected to clast size, lithology and facies analyses (Pash 2003; Trowbridge 2003). Clast lithology data, together with two surface streams for comparison is shown in Table 1, and particle size cumulative frequency graphs in Figure 3. Sediments in the active streamway are dominated by poorly sorted, sandy gravels comprising mostly allogenic mudstone and sandstone clasts, often with a dark grey manganese patina. Most of the clasts are angular to sub rounded. Angular clasts of limestone, derived from passage collapse and breakdown (largely caused by vadose incision and secondary gypsum growth in fractures) are common, but not appear to be undergoing significant transport. These are typical of the thalweg facies of Bosch and White (2007). A second distinct facies occurs as sediment banks and remnant fills preserved up to 22 m above the present stream-way and in neighbouring relict passages. This consists of fineto medium-grained, moderately sorted, pale grey, brown and black, cross-bedded sand, silty sand and laminated clays. Minor amounts of coarse sand and gravel comprising mudstone and quartz occur in places, but few large clasts are present. Remnants of this once extensive fillTable 1. Clast lithologies for the 2,000,350 mm particle size range (Mean %) for the present streamway, Gilwern Passage and two surface sites. The Total Sandstone is a combination of Millstone Grit and Sandstone. The glacial till samples were collected from Forgeside, near Blaenavon [NGR SO 2471 0876], whilst the Coal Measures sample was taken from a tributary feeding the River Clydach at [NGR SO 2156 1254]. Data from Pash, (2003). BeyondGlacialCoal a ChokeGilwernT illMeasures Lithology (Mean %)Stream-Passage(surface)(surface) Mudstone (Shale)67.458.928100 Sandstone16.220.738.40 Millstone Grit7.111.525.10 Quartz6.77.56.50 Limestone1.10.700 Total Sandstone23.332.263.50 Carbonaceous clasts1.40.720 Figure 3. Cumulative frequency plots for the streamway and Gilwern Passage sediments. Figure 4. Desiccated, cracked laminated silts overlying finegrained silty sand, draped over breakdown, Upstream Passage (photo M J Simms).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings44

PAGE 47

Figure 5. Cemented remnants of cross-bedded coaly sands preserved several metres above the floor of Upstream Passage. Cross bedding indicates flow to the left (upstream). Height of face shown is about three metres. Photo M J. Simms.It is clear that these passages were largely filled with sediment in the past. The sedimentary structures preserved in the deposits within Upstream Passage and its tributaries indicate flow upstream into progressively smaller vadose inlet passages. This is in the opposite direction to the present stream and the regional hydraulic gradient. Moreover, these sediments overlie extensive breakdown indicating they were deposited after considerable vadose incision and collapse. The sands bear no resemblance to the poorly sorted, manganese stained, fine-medium grained fluvial sand and gravel currently in transport. Today, even in extreme flood conditions, water levels in Upstream Passage rarely exceed a metre in depth, barely enough for the stream to be visible above the boulder floor and 4 m below the relict cross bedded sands observed on the passage walls. Very little sediment is transported during these floods; indeed, many of the gravel banks in the inlet streams are cemented with a manganese and iron oxide coating. Significantly, no relict sediment deposits occur in the southerly continuation of the streamway downstream of the junction with Gilwern Passage. This part of the streamway (Beyond a Choke) is a late-stage under-capture off the relict MS&D-Score-Gilwern Passage conduit (Farrant and Simms 2011), and only carries a small amount of sandy gravel in the active stream. Unlike Upstream Passage, no sediments are preserved above stream level. The point of capture is clearly marked at the southern end of Gilwern Passage where the stream that flows down Upstream Passage swings south into a smaller, lower level passage, whilst the roof tube swings north into Gilwern Passage. It is clear that these finer grained, carbonaceous and manganiferous sands seen in Gilwern Passage, The Score and Upstream Passage are genetically distinct from those currently in transport in the active stream. Although both are fluvial sediments, there is a striking difference between them in composition, fabric and volume of sediment in transport. As such they must have been brought into the cave system under very different hydrological conditions. The presence of fine fragments of sandstone and mudstone, together with abundant quartz clearly indicates an allogenic source, most probably from the overlying Upper Carboniferous Coal Measure sandstones and the Twrch Sandstone Formation. The cross bedding preserved within these relict sediments indicate that when they were emplaced, hydraulic gradients were locally reversed to that of the stream that formed the passage. This must have been a temporary reversal, as these sediments have since been largely flushed out and the former hydraulic gradients restored. Moreover, the time during which the sediment was emplaced must have been short lived as there is no significant evidence for dissolution, scalloping or paragenetic modification of the passage morphology (Farrant and Smart 2011). No pendants, notches, wall anastomoses or half tubes have been identified in Upstream Passage or in Gilwern Passage. Clearly, fluvial transport under present climatic conditions cannot account for these anomalous sediments. An alternative explanation is that the sediments were emplaced during glacial or pro-glacial conditions when glacial melt water was able to transport significant amounts of sediment into the cave. This hypothesis has been invoked for the extensive sediment fills in the Mynydd Llangattock caves, notably Ogof Agen Allwedd (Smart and Gardner 1989; Simms and Hunt 2007), a few km to the north across the Clydach Gorge (Figure 1).4. Discussion4.1. Age of the sediments Although these relict cave sediments have not been dated directly, two strands of evidence suggest that they are of considerable antiquity and predate the last Devensian (MIS2-4) glaciation. Firstly, the lack of fine-grained relict sediments in the Beyond a Choke streamway is in marked contrast to the abundant sediment fills in the passages associated with the relict MS&D-Score-Gilwern Passage conduit. This suggests that the finer grained sediments were deposited prior to the development of the streamway. This passage, which is up to 30 m high in places, is too large to have developed since the end of the last glaciation. It currently flows south to resurgences in the Afon Lwyd valley and probably developed during and subsequent to the last interglacial (MIS 5). Secondly, much of the sediment fill has been flushed out, leaving remnants preserved up to 6 m above the floor. Whilst sediment removal can be a very efficacious process, it is hard to imagine this amount of sediment being removed since the end of the last glaciation. A speleothem in Gilwern Passage was taken for Uranium series dating at the NERC Isotope Geosciences laboratory, at the British Geological Survey, Keyworth. The sample (OD-07-96E) was dissolved in hydrochloric acid, and the uranium and thorium were co-precipitated with Fehydroxide and separated by ion exchange (AG1x8 and UTEVA). Uranium isotopes were measured by TIMS (Triton) and Thorium by MC-ICP-MS (Nu HR). The sample yielded an age of 302.9 2.8 ka (Table 2) which suggests the passage was vadose by this time. However, it is not clear whether the sediment fill predates speleothem deposition or represents a later stage reactivation of the passage. Either way, it is probable that the sediments predate the last interglacial (the most likely period when the present streamway developed), and possibly as old as the Anglian (Elsterian) glaciation (MIS 12). Further U-series dating work is currently in progress to constrain the age better. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings45

PAGE 48

4.2. Glacial geomorphology Mid and South Wales was glaciated on at least two occasions during the Pleistocene; during the Anglian (Elsterian) and more recently during the Devensian (Weichselian) glaciation (Barclay 1989). A further three Middle Pleistocene glaciations have been postulated, during MIS 16, 10 and 6 (Saalian glaciations) (Lee et al. 2012). Whilst the extent of some of these Middle Pleistocene glaciations across Britain may be controversial, it seems likely that upland parts of Wales would have been at least partially glaciated during these events. During any of these glaciations, it is probable that Ogof Draenen would have lain at or close to the southern margin of the ice sheet at various times (Ehlers and Gibbard 2004). These ice caps probably had several spreading centres, principally in mid and north Wales, but a local dispersion centre was also likely over the Brecon Beacons. These reconstructions are supported by the presence of till up to elevations of 445 m asl on the flanks of the Usk valley and its tributaries (Barclay 1989), and glacial striae on the Llangattock plateau. The uplands of Mynydd Llangattock, Gilwern Hill and Mynydd Garnclochdy were probably occupied by ice caps (as they were during the Late Glacial Maximum where glacial till forms an extensive sheet at c. 350 m around Brynmawr). To the south the ice was funneled into a series of valley glaciers, including the Afon Lwyd glacier. Locally derived gravelly till (of presumed Devensian (MIS 2-4) age) over 10 m thick is present in the Forgeside borehole [SO 2504 0828; 345 m asl], and thin pre-Devensian tills extend south as far as Pontypool. To the north, the major glacier in the Usk valley was largely confined to the present valley (Barclay 1989). During the Devensian, the ice surface was not much above 250 m in the Abergavenny area, terminating in a complex series of moraines around Kemys Commander [SO 340 040] south of Abergavenny. Patches of morainic material demonstrate that the small north-east facing cirques on The Blorenge contained small glaciers or snow patches, but there is little evidence for a major glacier in Cwm Llanwernarth during this time. We suggest that during either the Anglian (Elsterian) glaciation, or possibly during MIS 10 and 6, similar conditions prevailed (Figure 6). It is possible that a comparable glacial setting may have occurred during the early Devensian (MIS 4), but the size and maturity of the Beyond a Choke streamway suggests that the sediments were deposited prior to this time. The presence of a glacier in the Afon Lwyd valley at elevations above 350 m, and an ice surface less than 250 m in the lower Usk valley, coupled with open, relict cave passages extending through the intervening mountain provided suitable conditions for the reactivation of these passages by glacial melt-water. Sediment laden melt water flowed into the cave via inlets along the eastern margin of the Afon Lwyd valley around Blaenavon (>320 m a.s.l.). From these and other inlets, water flowed north via a currently sediment choked passage (Pontypool or Bust) into The Score and then into the start of Gilwern Passage and the surrounding area. In doing so, it deposited finegrained sand and silt up to an elevation of c. 320 m asl. With outlets to the north (the Clydach Gorge) blocked by glacial ice, sediment or collapse, and the present streamway not yet in existence, the only available outlet was into Cwm Llanwernarth, a small tributary valley to the Usk. Although this valley probably contained a small cirque glacier, the glacier surface was considerably lower than the Afon Lwyd (Figure 7). Consequently, the water then flowed out upstream, via a series of passages at the eastern end of Upstream Passage, including Pen-y-Galchen Passage. These former inlet passages had previously been truncated by valley incision at the head of Cwm Llanwernarth, but because they form the lowest overspill point in the cave system, they were subsequently reactivated as series of temporary resurgences.Table 2. U-series data. U and Th total procedural blanks were <26pg and <8 pg, respectively. Data reduced using an in-house spreadsheet and Isoplot with decay constants from Cheng et al. (2000). SampleU ppm232Th ppm[230Th/232Th] OD-07-96E25.90.00149.088E+04 [230Th/238U]2%[234U/238U]2% 1.6370.221.5660.04 08-48Age (ka)[234U/238U]initial0.000 302.9.8 2.335.010 Figure 6. Proposed glacial setting during periods of subterranean glacial under-spill through Ogof Draenen. Contours at 100 m intervals. Glaciated areas are pale grey. Figure 7. Schematic cross section between the Afon Lwyd valley to the west (left) and the Cwm Llanwernarth valley to the east.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings46

PAGE 49

5. ConclusionsWithin Ogof Draenen, several distinct sediment facies can be observed in the network of passages around Gilwern Passage, Upstream Passage, The Score and the present streamway. In particular, the deposition of a distinctive suite of fine grained sediments that infilled parts of the cave to depths of over 20 m is ascribed to the influx of sediment laden glacial melt water, either during the Anglian (Elsterian) glaciation or possibly during MIS 10 or 6. This melt water, derived from the base of a glacier in the Afon Lwyd valley flowed into the lower part of Ogof Draenen via many pre-existing inlets. As the level of glacial ice in the neighbouring Usk valley was significantly lower, this melt-water was able to flow north through the cave (in the opposite direction to normal interglacial drainage), over spilling through various truncated inlet passages in the headwall of the Cwm Llanwernarth cirque to form a series of temporary springs at c. 320 m. The cave thus acted as a subterranean glacial spillway, transferring water from one catchment to another.AcknowledgementsWe would like to thank the many people who have contributed, in particular Laura Trowbridge and Jon Pash who did the sedimentological analyses, John Hunt, John Stevens, members of the University of Bristol and Chelsea Spelaeological Societies and Oxford University Caving Club, all the cave surveyors and the Pwll Ddu Cave Management Committee. Farrant and Noble publish with the approval of the Executive Director, of the British Geological Survey (NERC).ReferencesBarclay WJ, 1989. Geology of the South Wales Coalfield, Part II, the country around Abergavenny. Memoir for 1:50,000 geological sheet 232. British Geological Survey, HMSO, London. Bosch R, White WB, 2007. Lithofacies and transport of clastic sediments in karstic aquifers. In: ID Sasowsky, J Mylroie, R, Bosch and WB White (Eds.). Studies of Cave Sediments. Springer, Netherlands, 1. Chelsea Spelaeological Society, Stevens, J, 1997. Ogof Draenen. Chelsea Spelaeological Society Cave Survey (3 Sheets). Cheng H, Edwards RL, Hoff J, Gallup, CD, Richards, DA, Asmerom, Y. The half-lives of uranium-234 and thorium-23. Chemical Geology, 169(1), 17. Ehlers J, Gibbard PL, (Eds.), 2004. Quaternary Glaciations Extent and Chronology, Part I: Europe. Elsevier, Amsterdam, 488. Farrant AR, Simms MJ, 2011. Ogof Draenen: speleogenesis of a hydrological see-saw from the karst of South Wales.Cave and Karst Science, 38(1), 31. Farrant AR, Smart, PL, 2011. Role of sediment in speleogenesis: sedimentation and paragenesis. Geomorphology, 134 (1) 79. Ford DC, 1987. Effects of glaciations and permafrost upon the development of karst in Canada. Earth Surface Processes and Landforms, 12(5), 507. Lee JR, Busschers FS, Sejrup HP, 2012. Pre-Weichselian Quaternary glaciations of the British Isles, The Netherlands, Norway and adjacent marine areas south of 68N: implications for long-term ice sheet development in northern Europe. Quaternary Science Reviews, 44, 213. Lauritzen S-E, 1984. Evidence of subglacial karstification in Glomdal, Svartisen, Norway. Norsk Geografisk Tidsskrift, 38, 169. Lauritzen S-E, 1986. Kvithole at Fauske, northern Norway: an example of ice contact speleogenesis. Norsk Geografisk Tidsskrift, 66, 153. Murphy PJ, Smallshire R, Midgley C, 2001. The sediments of Illusion Pot, Kingsdale, UK: Evidence for sub-glacial utilisation of a karst conduit in the Yorkshire Dales? Cave and Karst Science, 28, 29. Pash J, 2003. Sediment dynamics in Ogof Draenen (South Wales) An assessment of past flow regimes. BSc (Hons) dissertation, Department of Geographical Science, University of Bristol. Simms MJ, Hunt JB, 2008. Flow capture and reversal in the Agen Allwedd Entrance Series, South Wales: evidence for glacial flooding and impoundment. Cave and Karst Science, 34(2), 69. Simms MJ, Farrant AR, 2011. Landscape evolution in southeast Wales: evidence from aquifer geometry and surface topography associated with the Ogof Draenen cave system.Cave and Karst Science, 38(1), 7. Smart PL, Gardener CG, 1989. The Mynydd Llangattwg cave systems. In TD Ford (Ed.). Limestones and Caves of Wales. Cambridge Univrsity Press, Cambridge, 124. Trowbridge L, 2003. Ogof Draenen: the sedimentary dynamics and development stages. BSc (Hons) dissertation, Department of Geography, Aberystwyth University. Waltham AC, Simms MJ, Farrant AR, Goldie HS, 1997. Karst and caves of Great Britain. Geological Conservation Review Series 12. Joint Nature Conservancy Council, Peterborough.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings47

PAGE 50

PROJECT NAMAK: SOME OF THE MOST SPECTACULAR FINDINGS IN THE IRANIAN SALT KARSTMichal Filippi1, Ji Bruthans2, Ondej Jager3, Mohammad Zare4, Naser Asad5 1Institute of Geology ASCR, v. v. i., Rozvojov 269, 165 00 Prague 6, Czech Republic, filippi@gli.cas.cz2Charles University in Prague, Albertov 6, 128 43 Prague, Czech Republic, bruthans@natur.cuni.cz3AQH s. r. o., Frdlanstk 1310/23, 182 00 Prague, Czech Republic4Departmnet of Earth Sciences, Shiraz University, Shiraz, Iran5Department of Geology, Faculty of Science, Sistan and Baluchestan University, Zahedan, Iran NAMAK Project (namak means salt in Persian language) is an informal association of geologists and speleologists who cooperate on exploration and scientific research of the salt karst in southern and southwestern Iran. During the Project (from 1998 until 2013) about 16 salt diapirs were visited and more than 60 caves were discovered, 30 of which were mapped. The most exciting scientific and speleological discoveries were made in the Namakdan, Hormoz and Jahani salt diapirs; however, many other remarkable discoveries have been made on other several Iranian salt diapirs. This contribution summarizes the most interesting findings achieved during the ten expeditions carried out by the NAMAK team.1. IntroductionThe NAMAK team studies Iranian salt karst since 1998. Up to now the 10 full-scale expeditions and several shorter trips were carried out between 1998 and 2010. Members of the NAMAK team assisted during film documentaries made by the BBC, Iranian TV and the Czech TV. The discovery and documentation of the 3N Cave by the NAMAKteam positively influenced declaring a part of Qeshm Island as a National Geopark, which was also recognized as part of the UNESCO Network of Geoparks in the year 2006. The southern and southwestern part of the Zagros Mts. in Islamic Republic of Iran (provinces Hormozgan, Fars and Bushehr)represents an area with the largest number of piercing salt diapirs in the world. Numerous diapirs teem with a variability of well-developed karst phenomena, including countless salt caves with brine springs and rich secondary halite deposits. However, the area was not previously subjected to some systematic geological research, except the tectonic and structural studies (see references in Talbot and Alavi, 1996). New knowledge from other geological fields has been obtained within the last circa 10 years. Characterization of the karst phenomena and notes to the speleogenesis in this area were presented by Bosk et al. (2002). Evolution of the surface morphology in relation to the karstification on some coastal salt diapirs was described by Bruthans et al. (2008, 2009, 2010). Overview of the various types of secondary cave deposits was presented by Filippi et al. (2011). Also the speleological findings will be summarized in detail soon (Filippi et al. in prep.). The presented contribution extracts some of the most spectacular findings made by the NAMAK team.2. Geography and geology of the areaFour salt diapirs were studied in detail during the NAMAK expeditions: Hormoz and Namakdan diapirs on islands in the Persian Gulf (Hormozgan Province), Kuh-e Jahani diapir near Firuzabad (Fars Province) and Namak diapir (Busher Province) in the Zagros Range. However some other salt diapirs (e.g., Namak, Gach, Saadad Abad, Mesijune have been visited (for location see the Fig. 1 in Bruthans et al. 2009). The climate in the visited areas is arid or semi-arid. Main precipitation occurs from November April. The annual average temperature is 27 C in the Persian Gulf region and 20 C at mountainous diapirs. Temperature can reach 50 C in the summer. During our expeditions (usually in the period from January to May), temperatures varied between 15 and 40 C. The humidity on the surface varied from 50% to 98%. Strong precipitation events occur sporadically every several years, usually during the winter season. Geologically salt karst areas are situated in the eastern part of the Zagros Mountain Range and the Persian Gulf Platform, which is composed of elongated whaleback or box-shaped anticlinal mountains, generally NWSEtrending (Bosk et al. 1998). The relief is young; the principal folding started only in Middle Pliocene. Rock salt composing the diapirs belongs to the Hormoz Complex, which was deposited during Upper Precambrian to Middle Cambrian times, i.e. more than half a billion years ago (Bosk et al. 1998). The salt deposits have been covered by other sedimentary rocks, which attained a thickness of several kilometers. Due to high pressure in such depths, tectonic factors, and due to the high plasticity and low density of salt, the deposits started to rise towards the surface in form of salt diapirs (plugs, domes). Salt diapirs in Iran are represented by more or less circular, cylindrical, etc. bodies with diameter between ca. 1 to 18 km. The rock salt is folded into tight to isoclinal folds, with alternating white to grayish and reddish salt with varicolored carbonate, siliciclastic, iron oxides and volcanosedimentary rocks and rarely anhydrite and gypsum. Blocks of exotic (igneous and sedimentary rocks) transported by halokinesis from depth reach diameters up to 2 km across in many diapirs.3. MethodsTopographical maps (1:25,000 and 1:50,000), satellite images (presented by the Google Earth utility), and aerial Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings48

PAGE 51

photographs were used as a basis for morphological and geological surveys. Global positioning system (GPS) instruments, with a horizontal accuracy of m, were used for field orientation and identification of sampling sites. Altitudes of the important points were determined using precise barometric altimeters (BARIGO 43, GARMIN eTrex Summit) with accuracy better than m. Some points with small elevation differences were measured using a total station (Leica). Cave passages were mapped by compass, clinometer, and a tape or a laser rangefinder. Polished sections were prepared for the mineralogical study of the speleothems followed by the scanning electron microscope (SEM) and X-ray diffraction spectroscopy (XRD).4. Karstology of the Iranian salt karstIranian salt karst is very a dynamic landscape, especially during and after heavy rains. In such time intense transport of sediments and abundant rock falls occur on the surface and also in caves. One would expect that salt caves will originate on salt diapirs with large areas of exposed rock salt, but thats not the case. Caves usually do not occur in bare salt diapirs. Runoff water reaches very rapidly saturation with respect to halite and looses ability to dissolve rock salt. Runoff water leaves the bare salt diapirs via dense net of surface streambeds and canyons in lower reaches. Fortunately for cavers, surfaces of the salt diapirs in Zagros Mountains are mostly covered by various types of surficial (residual) deposits (detail see in Bruthans et al. 2009). Regardless of the type of surficial deposit, its thickness and areal extent, its presence is the key factor governing the degree of karstification and cave development on salt diapirs in Iran (Bruthans et al. 2000). Thickness of surficial deposits above cave does not influence the cave itself; more important is the thickness of surficial deposit in recharge area of the cave. The surficial deposits act as impermeable cover that enables accumulation of large amount of fresh water. Also large blocks of exotic rocks act similarly as impermeable surficial deposits. Thickness of overburden above the cave usually influences intensity of breakdowns when caves in shallow subsurface are less stable compared to those in greater depth. The two very important geomorphological and/or karstological features affected by thickness of the surficial deposits (overburden of halite) are as follows (for details see Bruthans et al. 2000, 2009 and Filippi et al. in prep.): Density (and character) of recharge points, which has a negative correlation with thickness of overburden. Surface of diapirs with no surficial deposits is typical by presence of steep canyons and gullies, sharp Rillenkarren on rock exposures. Thin overburden is permeable for water at numerous places and therefore prevents development of extensive drainage networks that transmit and concentrate fresh water. As a result, dense fields of small dolines and ponors develop. On the contrary, thick impermeable overburden enables evolution of landscape with well developed drainage network terminated in valleys and deep canyons with scarcely distributed large ponors (followed by large caves) and also with large dolines. The rate of lowering the surface of salt karst. There is relatively very low denudation rate of surface in case of the thick overburden; salt dissolution is concentrated into cave passages. On the contrary, the denudation rate of ground surface in case of bare salt surfaces is very fast (centimeters per year) and could simply exceed the rate of halokinesis (Bruthans et al. 2008).5. Speleological finds5.1. The longest/largest salt cave Absolute majority of caves discovered in the Iranian salt karst reach first tens to few hundreds of meters in their lengths. So far, there are only three caves exceeding one kilometer: the White Foam Cave (1,262 m) on Jahani diapir, the Ghar-e Daneshyu (Students Cave; 1,909 m) on Hormoz diapir and finally the 3N Cave (6,580 m) on Namakdan diapir. The 3N Cave is probably the most impressive and remarkable salt cave in Iran. The 3N Cave is located in the southeastern part of the Namakdan diapir in the western part of the Qeshm Island (Persian Gulf). Since the detailed description of the cave was presented in English and in French by Bruthans et al. (2006) and Filippi et al. (2006), respectively, we present here only general description and underline some most important knowledge about this unique cave. Figure 1. Comparison of the first and the second worlds longest salt caves.Generally, the 3N Cave consists of a singular large meandering passage (Fig. 1) opened to a valley, via lower entrance (resurgence partly filled by brine) 8 m wide and 2 m high. A large salt lake, usually 100 m long (changing based on presence/absence of natural dams formed by salt precipitates) and up to 1 m deep, is at intervals forming approximately 160 m from this entrance. The roof is lowered here to only 20 cm above the water level. Very high salt concentration (up to approximately 330 g.l-1of dissolved salt) made it easy to swim because of Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings49

PAGE 52

of large a volume of incoming water during heavy rains. Collapses from the ceiling are very often and may endanger possible visitors. One of the largest underground spaces was documented also in the 3N Cave. The space called as the Megadomes consists of two connected relatively low halls (from 2 up to 8 m in the height), which together occupy an area of 200 m by 100 m (Fig. 3A). In general, these halls have flat ceilings only locally interrupted by fallen rocks; the walls often consist of fluvial sedimentary infill which points at their possible larger extent. The Megadomes are situated at a confluence of streams from two main sinkholes. The main stream is coming from the Big Ponor part and the second from other important tributary, the neighboring Upper Entrance Cave. Unfortunately, the connection with this cave is impossible due to a complete sedimentary infill. The Coffee Bar Hall in Eugens Cup of Coffee Cave on the Mesijune salt diapir is the largest and most impressive cave hall found in the Iranian salt karst and probably also worldwide (Figs. 2A, 3B). It has an extent of 70 30 35 m (length, width and height). The Eugens Cup of Coffee Cave is situated close to the northern margin of the diapir and is developed in apparently unstable environment in a rock salt covered by relatively thin surficial sedimentary overburden. The relatively short cave (the length 235 m) consists of a large meander passage with common width of 5 to 10 m and the similar height and already mentioned large hall. The cave has a flat bottom which is periodically flooded by rain water coming from the surficial valley. The large volume of sweet water coming quickly into the cave after the heavy rains is the main reason for development of such a large underground space located close to the cave inlet. Fallen rocks from the ceiling are quickly dissolved and thus the space remains partly empty (from the bottom to the ceiling). The inactive part of the passage is progress towards the surface. The lack of fresh water causes successive infilling of the space by rock fragments and fluvial sediments in many other documented underground spaces. 5.3. The deepest shaft Based on vertical cross-section, there are two general groups of caves in the Iranian salt karst: (i) more or less sub-horizontal caves, and (ii) vertical shafts. The first group of caves is typically situated as endings of larger elongated valleys with intermittent streams. Shafts are typical for short circular valleys and dolines, commonly without a significant water supply. Important factor for the shaft development seems to be sub-surficial subrosion, especially in the early stadium of their development. One of the one of the largest and deepest shafts was found in the western part of the Jahani salt diapir and was named the Goats Shaft, according to numerous goats grazing all around. The shaft has a typical morphology for such environment a short streambed on the bottom of circular funnel-shaped doline ends in a vertical shaft (Figs. 2B, 3D). The doline and first meters of the shaft are created in unstable only partly consolidated surficial sediment with large blocks of exotic rocks. The shaft with an oval crosscut and a flask vertical profile is mostly developed in rock salt. Narrow similar buoyancy effect to that of the Dead Sea. However, the oversaturation with salt is badly etching to skin scratches and open cuts, which are unavoidable when exploring caves. About 700 m from the southern entrance, the Hangar chamber (35 20 16 m) is situated. Behind this point, the width of the mostly muddy and collapseprone passage decreases from 10 m to only 6 m. One kilometer from the entrance, several collapse domes occur. Three shafts connecting the cave with the surface are situated behind them, all about 40 m in height. Since the year 2000 the NAMAK Project has used one of these shafts as a main entrance to the cave. North of the shaft mounts the draught inside the cave dies out and makes progress extremely uncomfortable with a temperature of 28 C, high humidity and no air draught. Also, the general character of the cave changes: the common arched cross-sections caused by breakdown occurring south of the shafts turn into passages with flat ceiling without any breakdown resembling to a large underground garage. This is probably related with a larger depth of this part of the cave below surface. 150 m long low crawl named the Bend is very uncomfortable part of the cave. After it, the passage increases in size. On the right side a small inlet creates a permanent brine shower called Waterfall, the 10 m high vertical chimney with a window into approximately 170 m long narrow meandering passage ending in dangerous narrow collapsed space. However, in 2010, this inlet was hidden by a rock fall, so its recent accessibility is questionable. The main passage named the Namakdan Highway is 40 m wide and up to 6 m high. The most challenging and endless long Azadis Crawlway starts after this large sapces. After approximately 800 m creeping over fine mud one reaches Megadomes (see the next chapter). Fortunately, a bypass of Azadis Crawlway was discovered in 2004 providing a shortcut dug through an older +6 m higher level. This shortcut was named Motashakkeram, what means Thanks a lot in Farsi. Behind the Megadomes, the main passage changes into a sandy to gravelly crawlway called Active Passage. The Big Ponor Cave entrance is the main inflow into the cave (for details see next section). 5.2. The largest underground spaces Width and height of accessible parts in most of the visited caves usually range from one to several meters. However, sometimes a large corridors and underground halls may develop. The Big Ponor is one of the largest natural semiunderground spaces in the rock salt. This cave (actually part of the 3N Cave the main ponor of this cave) contain an incredibly large entrance hall consisting of a huge portal (35 15 m) and an entrance hall (approximately 50 m by 50 m, 15 m high). The hall continues by dozens of low manholes in sedimentary infill leading in different directions and getting (with one exception) inaccessibly low after several tens of meters. Besides the mapped parts, there are many more inlets leading in westerly direction. This makes sense as Big Ponor is actually developed in a 200 m (probably more) wide underground alluvial fan where sheet floods model the cave. It actually represents an underground alluvial of a meandering underground river channel. Bottom of this great hall changes very significantly in time because Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings50

PAGE 53

terrace composed of large blocks of rocks that are cemented by halite and gypsum occur 30 meters below the shaft mouth. Shaft bottom covered by rock boulders can be reached after another 40 m. The shaft continued by a sloping cave starting with relatively extensive dome with an unstable ceiling. After approximately 120 m the cave terminates in inaccessible riverbeds in sediment and debris. 5.4. Caves formed in two different rocks Rare examples of caves may develop in areas where salt glaciers (overflow) cover the surroundings country rocks (mostly limestone). Bottoms of such caves and partly also walls are formed in limestone and ceilings consist of rock salt. Cave of Deliverance in western part of the Khurgu salt diapir (about 2 km SEE from the south portal of the road tunnel situated on the main road from Bandar-e Abbas to Yazd) is one of the best developed examples of such caves (for cave map see Bosk et al. 2002). The cave drains a huge blind canyon, which is developed on the border of salt diapir and a limestone anticline. The small southern cave branch drains a funnel-shaped doline developed on contact between limestone and salt. Originally, the cave was exclusively developed in salt on the contact with limestone bedrock. After having been enlarged enough to enable the transport of a clastic load, the vadose canyon entrenched into the limestone. An inclined (35) contact between salt and limestone is visible in the small southern branch of cave. 5.5. The most remarkable doline/canyon Iranian salt karst is rich in various types of blind valleys, shafts, abysses and dolines. Mostly they are represented by funnel-shaped depression with gentle slopes. Sometimes these depressions are followed by vertical shafts continued by relatively short steeply inclined inlet cave. For unusual morphology and complex genesis, the Puzzle Canyon on the Jahani salt diapir is quite different and unique. The branched labyrinth-type canyon with up to 13 m high, but only 0.5.5 m wide walls is situated on a flat bottom of the doline 100 200 m wide (Fig. 3C). The bottom of the doline is formed by fine deposits which settled there by numerous floods after heavy rains before the swallow hole in the bottom of doline was activated. The canyon developed due to headword erosion of sediment during heavy rains when streams enter the doline from several sides and rapidly carve several canyon branches and sink into the Puzzle Canyon Cave (Fig. 2C). The cave starts with a 15 m deep vertical step followed by relatively high and apparently unstable underground space formed in sediment. Fallen blocks of the sediments and rock fragments are common there. Gradually decreasing height of passages ends the passability of the cave on a distance 128 m from the entrance. 5.6. The highest waterfall Majority of the visited caves are more or less subhorizontal. However, few caves located on margins of salt diapirs with significant positive morphology are Figure 2. Maps of the selected speleological objects.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings51

PAGE 54

characteristic by a relatively steep slope and/or sharp vertical steps. Waterfall Cave on the southeastern margin of the Jahani salt diapir is the most interesting locality from this point of view. This cave is situated in an area consisting of a valley interrupted by several caves and drained by relatively permanent stream of brine. The cave is usually entered from the lower entrance (resurgence), that is accessible from the valley on the south-eastern margin of the diapir. The 15 m high step occurs at the hinder part of the cave, approximately 30 m from the inlet entrance. The flow rate is usually small up to several liters per second; however, during heavy rains it rises significantly. Splashing and aerosol brine causes precipitation of interesting halite speleothems on the walls around the waterfall (see Filippi et al. 2011). 5.7. The greatest speleothems Speleothems consisting of a microcrystalline to grained porous halite matter are the absolutely prevailing type of secondary precipitates in the Iranian salt karst. The initial microcrystalline deposits start to form on halite-/rockssaturated sites. The microcrystalline halite matter is able to create quite a large speleothems like sinters, stalactites and stalagmites (Fig. 3E). Their initial formation is primarily related to the presence of dripping water; however, evaporation of capillary water is the key factor governing their growth. Usually, the microcrystalline stalactites form only when the water feed decreases so that capillary transport prevails over gravitational dripping. This is the stage when the primary forms finish their development. In fact, microcrystalline stalactites are mostly the second stage in the secondary halite deposition (for details see Filippi et al. 2011). The largest microcrystalline stalactites and stalagmites reach up to approximately 4 m in length and up to approximately 30 cm in width. The stalactites are usually not vertical but are wildly curved or zigzag shaped. This is caused by the shape of the original form and by successive development, i.e. periodic changes in the shape at the tip of the stalactite, and by outer factors, such as the air draft direction or spatial distribution of the atmospheric humidity. Straw stalactites are another common speleothem type, which may reach notable proportions. Straw stalactites grow directly from rock salt or from older and destroyed secondary halite precipitates. This is caused by reopening of the central channels inside the grained stalactites after the rains. Straw stalactites display more or less uniform diameter usually of 0.5 up to 1 cm and a smooth surface composed of compact glazy halite. In ideal cases, the halite straws are straight and regular, like in carbonate caves. More often, they are irregular, slightly curved and combined with microcrystalline (grained) growths or rarely with filamentary helictites. Straw stalactites commonly have a length of 0.1 up to 1 m, but can reach up to approx. 3.5 m, as was documented in the White Foam Cave. 5.8. The most unusual speleothem A variety of secondary halite deposits (speleothems) was distinguished in the Iranian salt karst. The speleothems were classified into several groups, on the basis of the site and mechanism of their origin (Filippi et al. 2011). Some speleothems are similar to those known from carbonate karst, others are different. Macrocrystalline skeletal speleothem is surely one of the most interesting and most unusual halite forms (Fig. 3F). Halite macrocrystalline stalactites have been documented earlier (see references in Filippi et al. 2011). However, even the discovery of rich occurrences of macrocrystalline forms in the Iranian salt karst helped to understand their formation. Macrocrystalline skeletal forms (stalactites, stalagmites and wall/bottom crusts) are connected to irregularly splashing brine around waterfalls and on cave ceilings with strong dripping. Rich but more or less random water supply, accompanied by gravityand capillaryinduced water movement are the main prerequisites for growth of these forms. Compared to some other forms, macrocrystalline speleothems exhibit no internal feeding tube and the brine is distributed on their surface. Euhedral to skeletal, mostly irregular and sometimes hopper-shaped crystals are clustered into twigs forming the whole treelike formations. Macrocrystalline speleothems consist of a central columns and side twigs. The side twigs typically occur in groups of three, mutually forming an angle of 120, with an angle of ~70 to the central column. The directions of the central and side columns thus correspond to the four body diagonals of a cube. Each twig is terminated with a cube when all the cubes are equally oriented, again documenting that the whole form is actually a single crystal, albeit with a fractal appearance (a fractal cube). As the form grows more and more, the central column and the twigs can be completely buried under the growing cubes. For other details about this unusual form see Filippi et al. (2011).6. Important notesWe ask all possible visitors of the Iranian salt karst to be considerate and protect them. Project NAMAK has its own website where you can find other information: http://home.gli.cas.cz/namak/. All important speleological findings are summarized in the work of Filippi et al. (in prep.).AcknowledgmentsMany thanks to the Heads of Shiraz University and Head of the Qeshm Free Zone Organization. The authors thank to participants of the NAMAK Project: M. Audy, R. Bouda, M. Gerl, E. Janouek, J. Kamas, M. Kolava, J. Kukaka, M. Novk, Mike, T. Svoboda, S. lechta, J. md, M. Va ek and Z. Vilhelm. Research wassupported by projects no. KJB315040801, RVO67985831, MSM00216220855, and MZP0002579801.ReferencesBosk P, Jaro J, Spudil J, Sulovsk P, Vclavek V, 1998. Salt plugs in the Eastern Zagros, Iran: results of regional geological reconnaissance. GeoLines 7, 1.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings52

PAGE 55

Bosk P, Filippi M, Bruthans J, Svoboda T, md J, 2002. Karstogenesis and speleogenesis in salt plugs: Case study from the SE Zagros Mts., Iran. Proc. of the Middle-East Speleology Symposium, April 20.. 2001, Lebanon, 62. Bruthans J, Filippi M, Zare M, Asadi N, Vilhelm Z, 2006. 3N Cave (6580 m): longest salt cave in the world the NSS news. National Speleological Society 64, 10. Bruthans J, Asadi N, Filippi M, Vilhelm Z, Zare M, 2008. Erosion rates of salt diapirs surfaces: An important factor for development of morphology of salt diapirs and environmental consequences (Zagros Mts., SE Iran). Environmental Geology, 53(5), 1091. Bruthans J, Filippi M, Asadi N, Zare M, lechta S, Churkov Z, 2009. Surficial deposits on salt diapirs (Zagros Mts. and Persian Gulf Platform, Iran): Characterization, evolution, erosion and influence on landscape morphology. Geomorphology. 107, 195. Bruthans J, Filippi M, Zare M, Chur kov Z, Asadi N, Fuchs M, Adamovi J, 2010. Evolution of salt diapir and karst morphology during the last glacial cycle: effects of sea-level oscillation, diapir and regional uplift, and erosion (Persian Gulf, Iran). Geomorphology. 121, 291. Filippi M, Bruthans J, Vilhelm Z, Zare M, Asadi N, 2006. La grotte 3N, Iran (6 580 m): le nouveau record du monde de la plus longue cavit dans le sel. Spelunka, 104, 15. Filippi M, Bruthans J, Palatinus L, Zare M, Asadi N, 2011. Secondary halite deposits in the Iranian salt karst: general description and origin. International Journal of Speleology. 40(2), 141. Filippi et al. (in preparation). Project NAMAK. (working title). Talbot CJ, Alavi M, 1996. The past of a future syntaxis across the Zagros, in Alsop. In: Alsop GI, Blundell DJ, Davison I (eds) Salt Tectonics, Geol. Soc. Amer. Spec. Paper, 100, 89. Figure 3. A) The Megadomes in the 3N Cave; B) Upper part of the Coffee Bar Hall in the Eugens Cup of Coffee Cave; C) The Puzzl e Canyon; D) Goat Shaft and a caver on a step in -30 m; E) An example of a microcrystalline grained halite speleothem, the Octopu s formation in the 3N Cave; F) An example ofa macrocrystalline skeletal halite speleothem; narrows point at persons in the photos.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings53

PAGE 56

KARST DEVELOPMENT IN THE GLACIATED AND PERMAFROSTREGIONS OF THE NORTHWEST TERRITORIES, CANADADerek Ford School of Geography and Earth Sciences, McMaster University, Canada dford@mcmaster.ca The Northwest Territories of Canada are ~1.2 million km2in area and appear to contain a greater extent and diversity of karst landforms than has been described in any other region of the Arctic or sub-Arctic. The Mackenzie River drains most of the area. West of the River, the Mackenzie Mountains contain spectacular highland karsts such as Nahanni (Lat. 62 N) and Canol Road (Lat. 65 N) that the author has described at previous International Speleological Congresses. This paper summarizes samples of the mountain and lowland karst between Lats. 64N that are located east of the River. The Franklin Mountains there are east-facing cuestas created by over-thrusting from the west. Maximum elevations are ~1,000 m a.s.l., diminishing eastwards where the cuestas are replaced by undeformed plateaus of dolomite at 300 m asl that overlook Great Bear Lake. In contrast to the Mackenzie Mountains (which are generally higher) all of this terrain was covered repeatedly by Laurentide Continental glacier ice flowing from the east and southeast. The thickness of the last ice sheet was >1,200 m. It receded c.10,000 years ago. Today permafrost is mapped as widespread but discontinuous below 350 m a.s.l. throughout the region, and continuous above that elevation. The vegetation is mixed taiga and wetlands at lower elevations, becoming tundra higher up. Access is via Norman Wells (population 1,200), a river port at 65 37N, 12648W, 67 m a.s.l.: its mean annual temperature is -6.4 C (January mean -20 C, July +14 C) and average precipitation is ~330 mm.y-1, 40 % falling as snow. In the eastern extremities a glacial spillway divides the largest dolomite plateau into Mahony Dome and Tunago Dome. The former (~800 km2)has a central alvar draining peripherally into lakes with overflow sinkholes, turloughs, dessicated turloughs, and stream sinks, all developed post-glacially in regular karst hydrologic sequences. Tunago Dome is similar in extent but was reduced to scablands by a sub-glacial mega-flood from the Great Bear basin; it is a mixture of remnant mesas with epikarst, and wetlands with turloughs in flood scours. Both domes are largely holokarstic, draining chiefly to springs at 160 m a.s.l. in the spillway. The eastern limit of overthrusting is marked by narrow ridges created by late-glacial hydration of anhydrite at shallow depth in interbedded dolostones and sulphate rocks. Individual ridges are up to 60 km long, 500,000 m wide, 50 m in height. They impound Lac Belot (300 km2), Tunago Lake (120 km2) and many lesser lakes, all of which are drained underground through them. In the main overthrust structures, the Norman Range (Franklin Mountains) is oriented parallel with the direction of Laurentide ice flow. It displays strongly scoured morphology with elongate sinkholes on its carbonate benches. In contrast, the Bear Rock Range is oriented across the ice flow, has multiple cuestas, is deeply furrowed and holokarstic but preserves pinnacle karst on higher ground due to karst-induced polar thermal (frozen-down) conditions at the glacier base there. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings54

PAGE 57

LITTLE LIMESTONE LAKE: A BEAUTIFUL MARL LAKE IN MANITOBA, CANADADerek Ford School of Geography and Earth Sciences, McMaster University, Canada, dford@mcmaster.ca Marl lakes are those accumulating fine-grained bottom sediments that include at least 15% CaCO3. They are found worldwide. The most visually attractive, however, have higher proportions of CaCO3, with crystallites precipitating in the water to give it a rich and opaque duck-egg blue colouration. From the literature, such lakes are largely limited to recently glaciated carbonate rock terrains. Most are also shallow, with much or all of the water column being in the photic zone. Little Limestone Lake, (Lat. 53N, Long. 99W in the province of Manitoba) is the finest example that the author has seen. It stands out sharply from neighbouring lakes in summertime colour satellite imagery due to the intensity and uniformity of its colour. The lake occupies a shallow glacial trough scoured in a plain of flat-lying cyclothem dolomites. It is ~12 km long, 1 km wide, rarely >7 m deep. Including bordering wetlands, it occupies ~45% of the area of an elongated, narrow topographic basin. Recharge is through impoverished boreal forest with little soil cover; it discharges chiefly as springs and seeps along and below the shore. Mean annual temperature is ~0 C, and precipitation is ~475 mm.y1. Previous studies of springs in the surrounding region showed ground waters to be simple bicarbonate composition, with TDS = 230 mg.l-1(Ca 40 mg.l-1, Mg 30 mg.l-1). Grab sampling at 27 sites throughout the lake found the waters de-gassed to 125 mg.l-1, placing them in the mid-range of one hundred marl lakes investigated in more detail in the British Isles. Ca was reduced to 25 mg.l-1, while Mg was stable at 30 mg.l-1. There were 2 mg.l-1of free CO3in two fully analysed samples, indicating that plankton photosynthesis might be occurring. However, samples of the bottom marl were predominantly inorganic in their composition. Little Limestone Lake is visually spectacular because it is almost entirely groundwater-fed, with a ratio of recharge area to lake area that is low. It has no large, chemically equilibrated, surface streams entering it. In contrast, the dozens of nearby lakes (similar, larger or smaller in size) are regularly flushed by channelled storm water and, although they also produce some carbonate marl, cannot maintain high densities of crystallites in suspension. Little Limestone Lake was placed under legislated protection as a provincial park in June 2011.1. Marl lakesSediments accumulating on lake floors may be divided into: (1) clastics, which are fragments of insoluble rocks carried into the lakes by rivers or the wind; (2) organics, consisting of animal or vegetal material carried in by air or water currents from surrounding land, and the remains of fish, molluscs, plankton, etc. created in the water body itself; (3) precipitates and evaporites, crystalline or micro-crystalline minerals created by organic or inorganic chemical processes along the lake shore, on the bottom, or in the water column and mechanically settled out of it. At the global scale, mixtures of clastic and organic sediments in varying proportions are overwhelmingly predominant in the worlds lakes. Amongst the chemical class evaporite lakes and ponds are more numerous than precipitate lakes (including marl lakes), and probably quantitatively predominant in bulk terms because much greater tonnages are being deposited in them each year. Lakes in which carbonate precipitates make up a significant proportion of the sediment accumulating on the floors are thus comparatively rare. In a popular sedimentological classification of them mixtures of 25% calcium carbonate with clay or silt or non-carbonate organic debris are considered to be marl (Schurrenberger et al. 2003). They have the feel and appearance of soft mud or ooze. Some authorities would extend marl to include sediments with up to 95% calcium carbonate provided that they remain soft, i.e. unconsolidated. Anessential feature of the most attractive marl lakes everywhere is that there are also calcium carbonate micro-crystals (crystallites) in suspension in sufficient concentration in the water column to scatter sunlight and create strong and opaque blue colouration. Lakes where marl contributes between 5% and 95% of the bottom sediments (marl lakes sensu lato) are found in every region of the world where there are large quantities of limestone and/or dolomite to contribute the essential calcium ions, and the climate is humid enough to maintain the lakes and keep the concentrations of dissolved solids in them well below the levels where evaporite deposition begins. Most climatically humid regions and the wetter fringes of the sub-humid/semi-arid regions qualify. However, it is recognised that a disproportionately large number of marl lakes are to be found in the glaciated regions, including perhaps a majority of the visually very attractive. Most marl lakes that have been described in the scientific literature are located in the glaciated regions of northern Europe and North America. Following Pentecost (2005, 2009) there are two different paths to the creation and deposition of the marl, biogenic and abiogenic (inorganic). Processes in a given lake may be completely dominated by one of them, or be a mixture of both that may differ in its proportions as the seasons progress. The abiogenic is the simple process of limestone or dolomite inorganic dissolution, with crystallite precipitation following when CO2degasses from the Ca, Mg, HCO3solution due to warming or loss of pressure or both. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings55

PAGE 58

Wholly biogenic deposits may arise because photosynthesis in fresh waters creates many species of microscopic algae and bacteria. Pentecost (2009) focuses on the common green alga Chara but there are a host of alternatives. In the first step: 2HCO3 CH2O (photosynthesis)+ O2+ CO3 2-. There is abundant HCO3 _in karst lake waters, e.g., McConnaughey et al. (1994) found that macrophytes, chiefly Chara, extracted ~23 grams C/m2during the growth season in a Minnesota marl lake. With abundant Ca2+also being present via the inorganic path, step 2 proceeds: Ca2++ CO3 2CaCO3 (calcite). calcite begins to form in strongly alkaline surface regions of the internodal Chara cells and subsequently envelops almost the entire plant. During the later stages of calcification additional inorganic precipitation may be taking place but this has been little studied (Pentecost 2009). The inorganic calcite probably accretes to the outer surface of the alga in the manner that it accretes to larger plants to form tufa deposits in many streams and lakes around the world. The algae remain very tiny, however, suspended in the water and scattering incident light, settling slowly downwards only if conditions become very calm.2. Little Limestone Lake and its settingLittle Limestone Lake (Figs. 1, 2; Lat. 53N, Long. 99 19W in the province of Manitoba, Canada) is the finest example of a marl lake that the author has seen in many years spent in glaciated karst regions. It stands out sharply from the neighbouring lakes in summertime colour satellite imagery due to the intensity and uniformity of its duck-egg blue colour. Viewed from the shore or in a boat it is equally colourful. For these reasons the Manitoba chapter of the Canadian Parks and Wilderness Society, a national conservation body, argued for its protection for many years. The Conservation Branch of the provincial government requested an independent evaluation by a specialist (Ford 2010), which resulted in a park protection area being proclaimed in 2011. Little Limestone Lake is one amongst hundreds of small lakes in the Interior Lowlands geologic region of North America. It is located a few kms northwest of the much larger Lake Winnipeg, and drains to Hudson Bay (Fig. 1). The bedrocks are platformal dolomites of Silurian age totalling approximately 80 m in thickness that dip very gently westwards (Bezys 1991; Bezys and Kobylecki 2003). Local relief upon them is never more than 70 m, created chiefly by glacial scour, or the deposition of glacial moraines during the recession of the last Laurentide Continental Icesheet ~11,000 years ago. A sample bedrock section exposed in a road cut a few km south of the lake is shown in Figure 3. The strata are a cyclic succession of regular beds varying from medium (10 cm) to massive (>100 cm) in thickness. The thicker beds are mechanically stronger and so survived glacier scour more readily. Clintand-grike solutional pavement (epikarst) has developed on them in post-glacial times (Ford 1987; Ford and Williams 2007). They also have a high density of vugs, which can increase the matrix permeability of the rock and thus the quantities of dolomite taken into solution by groundwater permeating through it. Examples of the cliffs on massive dolomites and epikarst on thinner beds along the shores of Little Limestone Lake are shown in Figures 4 and 5. In Figure 2 the boundaries of the Lake basin are drawn conventionally to follow the surface topographic divide because no karstic diversions of groundwater flow are known. The basin is ~20 km in length, 5 km wide. It has an area of ~91 km2, of which 36 km2is occupied by the Lake itself. Pond A (~2 km2; Fig. 2) is a former arm of the lake that is now separated by wetlands but still drains into it; it has negligible ground water supplies of its own and its water is clear, not opaque and blue. The dry land area that is believed to contribute to the lake is thus ~53 km2. Wetlands at or very close to the lake level occupy ~3.5 km2of this and the balance consists of gentle side slopes and broad plateaus. The lake is at ~267 m above sea level and the highest ground is around 290 m a.s.l. Figure 1. Location of Little Limestone Lake, Manitoba. Figure 2. Sketch map of Little Limestone Lake. The watershed is shown by dashed line.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings56

PAGE 59

There is no bathymetric map of Little Limestone Lake and the bottom is always obscured by the suspended load in summer. The geologic and physiographic setting indicates that the lake is likely to be shallow everywhere. A few measured depths suggest that much of it has a flat floor at 7.0 to 7.5 m. This is comparatively deep for a marl lake but not as great as the deepest mentioned in Pentecosts 2009 sample of 100 such lakes in the British Isles. At 7.5 m the waters are fully in the photic zone. Sample cream-coloured carbonate oozes were dredged from the bottom there. At its north end Little Limestone Lake is drained by a regular stream channel that passes through a wetland into a lake about one metre lower (266 m a.s.l.). A small springfed stream flows into the southern end. To the west William Lake, a much larger water body, is at 265 m a.s.l. To the east and south there are extensive wetlands and smaller dolomite plateaus draining 18 km eastwards into Lake Winnipeg at 217 m a.s.l. The important consequence is that the Little Limestone Lake basin, although it functions as a water-retaining shallow trough, can be envisioned as an elongated hydrological ridge or high perched between drainage to lower lake levels around it.The Lake plus Pond A and associated wetlands occupy nearly 46% of the basin, a very large proportion. Recharge from the remaining ~54% is almost entirely by groundwater that has passed through the dolomite. The basin is not prone to flushing by mud-laden flood waters from proportionally greater recharge areas, as the majority of other lakes in the region are. Under the natural conditions that have prevailed for at least some thousands of years here a delicate physical, chemical and (probably) biogenic balance has been able to build up and establish the remarkable marl blue colouration. The current climatic conditions are Cool Continental (Koppen type D). The mean annual temperature is ~1.0 C and precipitation is 475 mm.y-1. The natural vegetation is Boreal Forest (coniferous). Nearly flat plateau and bench surfaces in bedrock are predominant in the recharge areas. There is thin and discontinuous glacial till cover on them due to erosion by over-wash immediately after the ice receded. There is a nearly 100% cover of mosses, lichens, flowers, juniper and other low shrubs, and forest litter. The trees are widely spaced, slow-growing or stunted in many places, reflecting the edaphic drought that occurs where the dolomite surfaces are efficiently drained by epikarst (Figure 6). Forest fires have broken up the karst on thinner beds, increasing the efficiency of groundwater recharge. The greatest straightline distance that groundwater must flow from these uplands to the lake is no more than 3 kms, giving hydraulic gradients (depth/length gradients) of 5 m.km-1or more. There are some seepages in the cliffs and at the water line in the lake but most ground water discharge is below that, perhaps because industrious beavers have raised the lake level by ~0.5 m in recent decades.3. Water chemistryThe water chemical characteristics of the Lake itself have not been studied in detail. However, McRitchie (1994, 1995) carried out systematic sampling at springs and seeps in the dolomite escarpment immediately south of it, plus two water samples from springs on William Lake. Seventy water samples were analysed for the standard water quality variables plus trace concentrations of economic minerals and of contaminants. All of the waters were simple calcium bicarbonate ground waters typical of clean dolomite terrains. Total hardness ranged between 230 and 300 mg.l-1; Ca 40 mg.l-1, Mg 30. The waters were saturated or supersaturated with respect to calcium carbonate under standard atmospheric conditions. Na, K, SO4and Cl were negligible in amount. At each site there was one-time sampling only and in most cases the water was taken from the spring itself before it had had any opportunity to de-gas its CO2content or make other adjustments to open air conditions. McRitchie (1994) described what happened downstream, however precipitates quickly appeared, as sand-sized particles at first but rapidly shrinking in size downstream and ceasing to form after 200 m of flow in the open channels. In the ponds at some springs there were organicrich mats of material. Downstream there was only cream-grey carbonate ooze, very like that collected from the bottom of Little Limestone Lake. The oozes were almost entirely CaCO3, chiefly of inorganic origin although one of four samples may have contained as much as 10% organics (i.e. from photosynthesis, as described above). Figure 3. A typical section of the cyclic dolomite beds at Little Limestone Lake. Figure 4. Cliffed shore with groundwater outlets at and below the waterline. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings57

PAGE 60

In September 2010 the author took one near-shore bulk water sample from the surface of Little Limestone Lake and a second from -2.0 m in the centre. Comprehensive standard analyses (ALS Laboratory,Winnipeg) found the two samples nearly identical, indicating that at least the upper two metres of the lake are uniform in composition. Na, K, Cl and SO4were negligible in amount. Total hardness (as CaCO3) was 200 mg.l-1. This is significantly less than the values between 230 and 300 mg.l-1that Ritchie obtained at the springs and seeps from the dolomites. It suggests that the Lake waters have de-gassed CO2and precipitated large amounts of their dissolved load. This explanation is substantiated by the Ca: Mg ratios of 26: 33. In the large majority of published dolomite karst water compositions the ratio is the reverse of this (60: 40 to 70: 30 in favour of Ca2+until vigorous precipitation begins). The detection of 2 mg.l-1of CO3 2-also indicates de-gassing, at a pH above 8.3. Determination of free CO3 in the lake waters is a hint that photosynthesis was occurring. Water temperature and electrical conductivity were measured at a depth of ~1.0 m at 27 different points in the Lake. The Krawczyk and Ford (2006) best fit equation for clean bicarbonate waters (salts + sulphates + nitrates + phosphates <10%; electrical conductivity <600 S) was used to calculate total hardness as mg.l-1CaCO3:TH = 0.53 EC -1.6 (R2= 0.93; n = 2300)(1) The principal finding was the uniformity of shallow water chemistry everywhere on the Lake proper. Despite a range of temperatures from 12.8 to 17.4 oC during the two days of sampling, EC was in the narrow range of 240 S with one exception. From Equation 1, estimated CaCO3concentrations of 126 mg.l-1place the Lake waters comfortably in the mid-range that Pentecost (2009) has cited in his sample of more than one hundred marl lakes and ponds in the British Isles. The one exception was the small spring-fed stream draining the southern epikarst. It entered the Lake at a temperature of 11 C, EC = 296 S, and was clear (no blue opacity). Twenty metres further out into the Lake sampling a few minutes later yielded T = 14.4 C, EC = 247 S. This suggested rapid warming with net precipitation of ~25 mg.l-1CaCO3. Concerning the stability of the density and hue of the colour, my previous visits to the Lake (1984, 2004) were of short duration during warm sunny afternoons which led me to wonder whether there might be a daily blooming of crystallites in response to solar warming, such as occurs in knee-deep water on the Bahamas Banks for example. It is now appreciated that this is not the case; the crystallites can have a duration of at least many weeks in suspension during the ice-free period on the Lake, giving sufficient time for winds to mix them thoroughly across the surface. Further, many authorities contend that lakes that are as shallow as typical marl lakes do not develop strong thermal or other stratification (Pentecost 2009); the physical and chemical conditions are broadly similar from surface to the bottom. From the physical chemistry some increase in the rate of precipitation is to be expected at the warmest times of the day: at Little Limestone Lake other observers have reported their qualitative impressions that there is some increase in density (opacity) then.4.Conclusions and recommendationsLittle Limestone Lake has outstanding blue (marl laketype) colouration because its elongated, small and shallow, basin is delicately perched between much larger basins draining to east and west respectively and thus is protected from flushing or clastic or organic detrital input by large surface streams during any strong thaw or rain flood events. The ratio of recharge area to lake area is low. Almost all annual recharge is from the combination of seasonal snowmelt and perennial groundwater flow directed through the enclosing dolomites. From current knowledge it appears that the water chemical behaviour that determines the striking blue colouration of the Lake during the summer is quite robust. The bulk of CaCO3crystallite precipitation appears to be of physical origin but the contribution of photosynthesis is not adequately understood at present. The location is remote and the local population is very small. However, there are nickel exploration prospects in Canadian Shield rocks beneath the dolomites around William Lake that touch the western side of Little Limestone Lake. Physical threats to the lake could arise if Figure 5. Shoreline recharge directly from the epikarst. Figure 6. Recently burned forest on medium-bedded dolomite. The epikarst surface has been broken up by the heat, by toppling and by frost.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings58

PAGE 61

the volumes of water flowing into it were substantially increased or diminished by any exploration or other development. Both could impact the colouration and its chemical stability in ways that are difficult to forecast. Chemical hazards could include acidification or introduction of other contaminants that kill off the pertinent biota if biogenic activity is a crucial contributor to the carbonate precipitation. Heavy gasoline-powered boat traffic could also pose a threat. It is hoped that the creation of the provincial park will deal with these problems.AcknowledgementsThe author is indebted to Roger Turenne and Ron Thiessen (Manitoba chapter, Canadian Parks and Wilderness Society), Ken Schykulski (Manitoba Conservation) and Chief Buck of the Mosakahiken First Nation for thweir encouragement and support in the field.ReferencesBezys RK, 1991. Stratigraphic mapping (NTS 63F, 63K) and corehole programme 1991. Manitoba Energy and Mines, Report of Activities 1991, 61. Bezys RK, Kobylecki AJ, 2003. Preliminary karst inventory of areas north and south Grand Rapids, Manitoba (NTS 63B and 63G). Manitoba Energy and Mines, Report of Activities 2003, 213. Ford DC, 1987. Effects of Glaciations and Permafrost upon the Development of Karst in Canada. Earth Surface Processes and Landforms, 12(5), 507. Ford DC, 2010. Final Report upon Field Studies and Review at Little Limestone Lake Park Reserve. Parks, Natural Areas Branch, Manitoba Conservation. 48. Ford DC, Williams PW, 2007. Karst Hydrogeology and Geomorphology. Chichester: John Wiley & Sons, Ltd. xiii, 563. Krawczyk WE, Ford DC, 2006. Correlating Specific Conductivity with Total Hardness in Limestone and Dolomite Karst Waters. Earth Surface Processes and Landforms. 31, 221. McConnaughey TA, Labaugh JW, Rosenberry DO, Striegl RG, Reddy MA, Schuster PF, Carter V, 1994. Carbon budget for a groundwater-fed lake: calcification supports summer photosynthesis. Limnology and Oceanography, 39, 1319. McRitchie WD 1994. GS-29. Spring water and marl geochemical investigations, Grand Rapids Uplands (NTS 63G). Manitoba Energy and Mines, Mineral Division, Report of Activities 1994, 148. McRitchie WD, 1995. GS-22. Spring water and marl geochemical investigations, Grand Rapids Region, 1995 Status Report (NTS 63G). Manitoba Energy and Mines, Mineral Division, Report of Activities 1995, 10919. Pentecost A, 2009. The marl lakes of the British Isles. Freshwater Reviews, 2009(2), 167. Schurrenberger D, Russell J, Kerry Kelts, 2003. Classification of lacustrine sediments based on sedimentary components. Journal of Paleolimnology, 29, 141.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings59

PAGE 62

CAVES AND KARST HYDROGEOLOGY OF JERUSALEM, ISRAELAmos Frumkin Israel Cave Research Center, Geography Department, The Hebrew Universit of Jerusalem, Israel 91905, msamos@mscc.huji.ac.il The city of Jerusalem, Israel, is growing for ~4,000 years on karst terrain. Lacking closed depressions, surface topography seems fluvial, but karst is well demonstrated by speleology and subsurface hydrology. Several caves in the city were truncated by construction works, including an 800 m long river cave (longest limestone river cave in Israel), and a 200 140 90 m isolated chamber cave (largest chamber cave in Israel). Caves are being discovered at a growing rate, as construction works dig deeper into the subsurface in the crowded city. Some of them are eventually destroyed by the construction works; only presently accessible caves are discussed here. The hydrogeology and hydrochemistry of the Gihon, Jerusalems main karst spring, was studied in order to understand its behavior, as well as urbanization effects on karst groundwater resources. High-resolution monitoring of the spring discharge, temperature and electrical conductivity, as well as chemical and bacterial analysis demonstrate a rapid response of the spring to rainfall events and human impact. A complex karst system is inferred, including conduit flow, fissure flow and diffuse flow. Electrical conductivity is high compared to nearby springs located at the town margins, indicating considerable urban pollution in the Gihon area. The previously cited pulsating nature of the spring does not exist today. This phenomenon may have ceased due to additional water sources from urban leakage and irrigation feeding the spring. The urbanization of the recharge area thus affects the spring water dramatically, both chemically and hydrologically.1. IntroductionFamous as a sacred city for the western world religions, it is rarely mentioned that Jerusalem is built on karst. The city was founded just above the Gihon karst spring. This paper briefly presents the karst underlying the city. In addition to the well-developed karst features, human impact on the karst is also discussed. The data comes from ongoing research by Israel Cave Research Center (ICRC) which monitors closely the new caves and karst phenomena found during urban development. The paper deals with presently accessible features, neglecting smallisolated chambers and vadose shafts which have been destroyed or filled during construction works.2. Setting2.1. Topography Jerusalem is located on hilly surface on both sides of the main water divide of Israel, ~600 m above sea level (Fig. 1). The Jerusalem Hills (20 km around Jerusalem) is a structural and topographic saddle within the plateau-like Judean Mountains, at the center of the Judea and Samaria Mountain range. The present topography demonstrates an uplifted Tertiary erosion surface entrenched by fluviokarstic wadies (rarely flowing streams) which drain westward to the Mediterranean and eastward to the Dead Sea. Closed depressions are not observed within the city, but do exist a few km to the north (Frumkin 1993). 2.2. Geology Jerusalem is underlain by Late Cretaceous rocks. Most of the city is built on Judea Group, dominated by shallow, epiric marine carbonates, on which karst is well-developed (Gill 1997; Picard 1956, Sneh and Avni 2011). The eastern suburbs of the city were recently built on the overlying Mt. Scopus Group, dominated by chalk and some chert deposited in a deeper southern basin of the Neo-Tethys Sea. The dominant rocks are well-bedded to massive limestones, dolomites and chalk, with smaller amounts of marl and chert. Stratal dip across Jerusalem ranges from 5 to 15 to the southeast, associated with several stages of uplift and folding. Jerusalem was last inundated by marine water probably during the early Eocene. Following regional regression and exposure during the late Eocene, an erosion surface has cut the dipping strata, exposing older layers to the west and younger to the east of the city. Periodic uplift and mild folding was accompanied by continuous karstification and erosion, truncating hundreds of meters of the bedrock (Frumkin 1993). Figure 1. Location topographic map with roads and mentioned caves.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings60

PAGE 63

The section of Judea Group underlying Jerusalem is divided into six sub-units (Sneh and Avni 2011): (A) Cenomanian Kefar Shaul Formation (Dir Yasini, 10 m thick) chalk and well-bedded limestone; (B) Cenomanian-Turonian Weradim Formation (Mizi Yahudi-Ahmar, 55 m thick) karstified massive to bedded crystalline, dense dolomite; (C) Turonian Shivta Formation (Meleke, 20 m thick) karstified massive, porous biosparitic limestone; (D) Nezer Formation (Mizi Hilu, 40 m thick) well-bedded dense biomicritic limestone; (E) SantonianCampanian Menuha Formation (60 m thick) biomicritic chalk with some chert; (F) Campanian Mishash Formation (90 m thick) mainly chalk and chert. 2.3. Climate Jerusalem is located at the boundary between the temperate Mediterranean climatic belt to the west and the rain-shadow Judean Desert to the east. Mean annual temperature is 18 C, and precipitation averages 550 mm.y-1, falling during the cool winter (OctoberMay), when potential evaporation is ~2 mm/day. The summer (JuneSeptember) is hot and dry with potential evaporation ~7 mm/day (Goldriech 1998). During DecemberJanuary about 2/3 of the annual rainfall occurs, typically concentrated in rainfall events lasting from a few hours to several days. Less commonly, some precipitation falls as snow. August is the hottest month with an average temperature of 24 C, while the winter average is 10 C. About 30% of precipitation flows and infiltrates to the subsurface, and the rest is mostly evapotranspirated. Natural surface runoff amount to few %, but has increased due to urbanization (Benami-Amiel et al. 2010). Speleothem stable isotopes indicate that glacial periods were wetter and cooler than interglacials (Frumkin et al. 1999; 2000). Increased dust input has contributed to local terra rossa soils, mainly during glacial periods (Frumkin and Stein 2004). 2.4. The urbanization of Jerusalem During the 2ndmillennium Before the Common Era (BCE), Jerusalem occupied ~0.04 km2on the hill above the Gihon Spring. The area increased to ~1 km2 around 700 BCE, and to ~2 km2 during the Roman period (1stcentury CE). Following the Arab conquest the city contracted to ~0.85km2, within the presently walled Old City (BenAryeh 1977), covering only a small portion of the Gihon Spring catchment. This was followed by rapid growth outside the walls of the Old City, since the mid 19thcentury. Population has increased from 4,700 in 1525 CE, through 45,000 in 1896, to 933,000 in December 2011. Today, Jerusalem covers ~130 km2(Kaplan et al. 2000). Open area accounts for 65% of the land-use within the municipal area of the city. Land uses in the catchment include small industrial areas, archaeological parks, public administration buildings, religious compounds and mostly residential quarters. The main anthropogenic activities that pose threat to the quality of groundwater are small industry and human residence where sewage infrastructure is old.3. Caves in JerusalemFew small natural caves were known within the present city borders prior to human intervention. Most known karst caves were truncated by construction works since ancient times. Thousands of years ago, the City of David water systems have truncated some phreatic and vadose voids of local nature. Of these, Warren Shaft is the largest vadose cave, and the Gihon Spring Cave is the largest phreatic cave. Both caves are connected to (but not part of) the famous Siloam Tunnel (Frumkin et al. 2003). In general, most ancient water systems did not follow karst voids (Shimron and Frumkin 2011), but were artificially excavated (Frumkin and Shimron 2006). Underlying the Old City, the large ancient underground quarry dubbed Tsidkiyahu Cave (also called King Solomon Quarry) truncated some karst voids, mostly vadose shafts (Fig. 2). Modern construction has truncated many caves, some of which much larger than known before. The longest one is the 800 m long Hauma Cave (Figs. 3, 4), truncated by an artificial shaft 80 m below surface. It is the longest limestone river cave at the northern edge of the SaharaArabia desert belt in the Levant, and the 8thlongest cave in Israel (Langford and Frumkin 2013). It was explored by the ICRC upstream until becoming a narrow tube-like conduit, and downstream down to a sump. Figure 2. Vadose shaft with a trace of meandering canyon at its ceiling. Tsidkiyahu Cave, view upward.Photo by author.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings61

PAGE 64

The cave is mostly a low-gradient dip passage perched on Kefar Shaul Formation, with a ten m vertical segment. The cave is mostly a vadose canyon with remnants of an original phreatic tube. Some 16 vadose shafts are connected to the stream passage, forming domepits up to 32 m above stream level. The largest volume chamber under Jerusalem (and in Israel) is Atarot Cave, at Atarot industrial zone (Fig. 5). It is a 200 140 m large isolated chamber cave (Frumkin and Fischhendler 2005), with 90 m maximal depth. It was formed in Shivta Formation probably under phreatic (hypogenic?) conditions, but large scale collapse features conceal the original solutional morphology. The chamber intersects some vadose shafts. The Atarot Cave is welldecorated with speleothems (Fig. 6), but unfortunately, it was partly rubble-filled by construction contractors before it was known to speleologists and authorities. Hotsvim Cave (Fig. 7) is another chamber cave, truncated by a building at Har-Hotsvim industrial zone. This cave is a symbol of cave conservation in Jerusalem, as the building plan above it was altered in order to preserve the cave. Among the vadose shafts truncated by construction works, the Jerusalem-West Cave is one of the most studied in the city. This small shaft system in the western suburb of HarNof was blasted by a contractor; consequently its speleothems were used for paleoclimatic reconstruction of the last 220,000 years (Frumkin et al. 1999; 2000). This exceptional, almost continuous record has shown that regional climate was much wetter and dustier during glacial periods (Frumkin and Stein 2004). Relict karst features in Jerusalem include a wide range of dissolution voids filled with breccias of chert fragments, derived from collapse, erosion and disintegration of Menuha and Mishash Formations chert layers. Some of these voids reach deep into the Judea Group rocks. These voids are probably associated with deep karstification of the Tertiary Judean erosion surface. An age of at least early Pleistocene is indicated by archaeozoologic deposit in one such pit at nearby Bethlehem (Hooijer 1958). Their geology has been properly described following their discovery (Picard 1956; Shaw 1961), but unfortunately, they were misinterpreted in later publications (e.g., Horowitz 1970; Sass and Freund 1977).4. Gihon Spring hydrologyGihon is the largest karst spring in Jerusalem (Fig. 7), and the reason for its original location. Gihon recharge zone is within the older parts of Jerusalem, on Judea Group outcrops (CenomanianTuronian carbonates). The Spring emerges 635 m asl within a small phreatic cave, 3 m below the current Qidron Valley level, whose lower channel is filled with thick debris. The spring hydrology was studied by Benami-Amiel et al. (2010). We cleared out the Gihon Spring cave from debris. It was found that the water emerges into the cave from an underwater (~130 cm below water surface) vertical fissure ~1.5 cm wide in Mizi Ahmar dolomite (Gill, 1997). Discharge, temperature and electric conductivity (EC) were measured continuously during 2004/5. Discharge during the Figure 3. Vadose canyon in Hauma Cave.Photo by author. Figure 4. Hauma Cave vadose river cave, survey by Boaz Langford. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings62

PAGE 65

dry summer is relatively stable, ~0.01.03 m3.s-1. During the investigated hydrological year (2004/5; 564 mm precipitation) spring discharge started at 0.025 m3.s-1(October), peaked at 0.164 m3.s-1(February), and recovered by the end of the year (September) to 0.031 m3.s-1. During rainstorms and soon afterwards, quick-flow response time Figure 5. Atarot Cave survey the largest chamber cave in Israel. In gray areas the fill reaches the ceiling. Figure 6. A stalagmite in Atarot Cave. Figure 7. Hotsvim Cave survey.The average EC is relatively high (1.54 mS.cm-1) compared with other springs around Jerusalem, especially those in nearby natural open areas (~0.6 mS.cm-1). The EC fluctuates considerately: Maximal EC (2.39 mS.cm-1) was recorded at the end of the dry season and minimal EC (1.278 mS.cm-1) on February at the peak of the rainy season. Temperature of the water is more stable between 19.03 C during wintertime (February) up to 19.42 C at the end of the dry season (September). Water temperature and EC minima are closely associated with rainfall peaks. The EC and temperature values began to decline 9 hours after rainfall began, and reached ranged from 3.5 to 89 hours, depending on rainstorm type, rainfall amount, and the previous saturation of the system with water. A century ago Vincent (1911) observed discharge pulses (ebb and flow) in the Gihon Spring: water emerges accompanied by loud echoing noise heard 1 min before the water rises and during the whole period of the strongest flow water rushed out unexpectedly every two or three hours, running for twelve or fifteen minutes at a time. Today the Gihon Spring is not pulsating any more, possibly due to anthropogenic impact. We suggest that water sources added to the recharge (from irrigation and pipe losses) within the city may continuously fill the underground siphon feeding the spring. External water import to Jerusalem has increased substantially overtime with the development and expansion of the city. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings63

PAGE 66

Due to the quick flow component of karst springs they are vulnerable to rapid pollution. Indeed, a major pollution event occurred on May 2002. This event occurred due to leakage from a sewage pipeline about 1.2 km north of the spring.5. ConclusionsAlthough the topography of Jerusalem seems fluvial, its karstic nature is revealed by underground features and, to a minor extent, by subaerial morphology. The underground karst voids incloude a river cave, vadose shafts (Fig. 9) and ancient isolated chambers. The discussed caves were found due to truncation by anthropogenic construction works. The Gihon karst spring reflects typical karst hydrology with increasing human interference by pollution and modification of the flow regime. Modern infrastructure is increasingly moving to the subsurface; karst features are intercepted at an accelerated rate and should be closely monitored and preserved wherever possible; the ICRC is doing its best in both these respets. minimum values 8 hours later. After about 8 days, both values recovered to their initial values. Despite large air temperature variations, water temperature fluctuates only slightly, indicating the dominance of rock temperature on fissure water. Water temperature is higher than average air temperature, explained by the geothermal gradient of the Judea Group stratum and the water residence time in the host rock. The Gihon Spring rather small temperature fluctuations correlate with the EC. As expected, arrival of fresh rainwater results in lower solute concentration, EC and temperature. The saturation index (SI) of the water indicates that the water is always under-saturated with respect to dolomite (and evaporites). Super-saturation with respect to calcite was recorded in some samples, suggesting possible calcite precipitation. Historic calcite tufa deposits of the Gihon waters (Frumkin and Shimron 2006) indicate that supersaturation with respect to calcite is a recurring phenomenon. These results match the characteristics of the aquifer: the quick flow component of the water is often under-saturated, although baseflow waters can become super-saturated with respect to calcite, especially under free-flow conditions in the Siloam Tunnel downstream of the spring, where rapid CO2degassing occurs. Figure 8. The study of Gihon Spring. Figure 9. Vadose shaft truncated by construction works. Close and similar to Jerusalem West Cave.Photo by author.AcknowledgementsI thank Boaz Langford, Ronit Benami-Amiel, Shmulik Avidan, and other members and volunteers of the Israel Cave Research Center who took part in studying individual karst features. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings64

PAGE 67

ReferencesBenami-Amiel R, Grodek T, Frumkin A, 2010. Characterization of the hydrogeology of the sacred Gihon Spring, Jerusalem: A deteriorating urban karst spring. Hydrogeology Journal 18, 1465. Ben-Aryeh Y, 1977. A City Reflected in its Times: Jerusalem in the Nineteenth Century, the Old City, Yad Ben-Zvi Press, Jerusalem (in Hebrew). Frumkin A, 1993. Karst origin of the upper erosion surface in the Northern Judean Mountains, Israel. Israel Journal of Earth Sciences 41, 169. Frumkin A, Ford DC, Schwarcz HP, 1999, Continental oxygen isotopic record of the last 170,000 years in Jerusalem: Quaternary Research, 51, 3, 317. Frumkin A, Ford DC, Schwarcz HP, 2000, Paleoclimate and vegetation of the last glacial cycles in Jerusalem from a speleothem record: Global Biogeochemical Cycles, v. 14, 3, 863. Frumkin, A. Stein M, 2004, The Sahara East Mediterranean dust and climate connection revealed by strontium and uranium isotopes in a Jerusalem speleothem: Earth and Planetary Science Letters 217, 451. Frumkin A, Fischhendler I, 2005. Morphometry and distribution of isolated caves as a guide for phreatic and confined paleohydrological conditions. Geomorphology 67, 457. Frumkin A, Shimron A, Rosenbaum J, 2003. Radiometric Dating of the Siloam Tunnel, Jerusalem. Nature 425, 169. Frumkin A, Shimron A 2006. Tunnel engineering in the Iron Age: geoArchaeology of the Siloam Tunnel, Jerusalem. Journal of Archaeological Science 33, 227. Gill D, 1997. The Geology of the City of David and its Ancient Subterranean Waterworks, Qedem, 35, Monographs of Institution of Archaeology, Hebrew University of Jerusalem, IV, 1. Goldreich Y, 1998. The Climate of Israel: Observations, Research and Applications, Bar Ilan University, Ramat Gan (in Hebrew). Hooijer DA, 1958. An early Pleistocene mammalian fauna from Bethlehem. The Bulletin of the British Museum (natural history) 3(8), 26. Horowitz A, 1970. Outrops of Hazeva and Taqiye formations in the Jerusalem area and their possible significance. Israel Journal of Earth Sciences 19, 35. Kaplan M, Kimhi I, Choshen M, 2000. The Jerusalem Hills and the Judea Coastal Plain: Policy for Land Conservation and Sustainable Development, Keter Press, Jerusalem (in Hebrew). Langford B, Frumkin A, 2013. The longest limestone caves of Israel. Proceedings of the 16thInternational Congress of Speleology, Brno. Picard LA, 1956. Geology, The Book of Jerusalem: Jerusalem, its Natural Conditions, History and development from the Origins to the Present Day, Avi-Yonah (ed.) vol. 1, The Bialik Institute, Jerusalem, 35 (in Hebrew). Sass E, Freund R, 1977. Deeply incised Senonian unconformity in Jerusalem and its implication on the early evolution of the Judean hills. Israel. Israel Journal of Earth Sciences 26, 10811. Shaw SH, 1961. Geological report on the elephant pit, Bethlehem. The Bulletin of the British Museum (natural history) 5, 4, 87. Shimron AE, Frumkin A, 2011. The Why, How, and When of the Siloam Tunnel Reevaluated: A Reply to Sneh,Weinberger, and Shalev. BASOR 364, 53. Sneh A, Avni Y, 2011. Geological map of Jerusalem, 1:50,000. Israel Geological Survey, Jerusalem. Vincent LH, 1911. Underground Jerusalem: Discoveries on the Hill of Ophel (19091), Horace Cox, London.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings65

PAGE 68

of the Dubrovnik Airport. It should however be noted that during airport construction in early 1960 as many as three greater speleological sites were subjected to speleological study in the area presently occupied by the control tower and terminal building. Two speleological sites are vertical with the depths of 31 m and 10.50 m, while the remaining site is complex and measures about 200 m long, and about 30 m deep. All this, in addition to the discovery of new caves, level of weathering of Senonian limestones in this area, and intensive tectonic situation, has pointed to the need to study the newly discovered cavern systems in more detail, and to propose the best possible method for their improvement. The decision was made to conduct the speleological study and to investigate possibilities for penetrating into the deeper parts of the caverns, and also to prepare geophysical profiles so as to cover the surface weathering zone, and to present the area all the way down to compact rocks. To facilitate the analysis of geological situation, the cavern systems were marked with K-1 (cavern No. 4), K-2 (widened space around the opening measuring 0.57 x 0.33 m Fig.1), and K-3 (area in which the material caved in, but the entrance to cavern was not found). The zones K-1, K-2 (Fig. 2) and K-3 were connected by geophysical profiles (Fig. 6) which were then analysed in form of a geophysical presentation. Three profiles were prepared. The speleologists descended into the K-1, while in K-2 the cavern system could not be penetrated even after the widening. No search of cave entrance was conducted in K-3. The cavern was formed in Upper Cretaceous (Senonian) limestones and dolomites and the layer thickness varies from 20 to 50 cm. Limestones alternating with dolomites are quite frequent in this zone (Markovi 1966, 1975). Limestones are light to dark grey in colour, and brown intercalations of carbonate calcitic binder are locally found between the layers. In the speleological site, and in its immediate vicinity, the inclination of carbonate formationsCAVES UNDER DUBROVNIK AIRPORT IN CROATIAMladen Garai University of Zagreb, Faculty of Civil Engineering, Department of Geotechnics and Geolog., Ka ieva 26, HR-10000 Zagreb, Croatia, mgarasic@grad.hr Croatian Speleological Federation, Society for the Research, Surveying and Filming on Karst Phenomena (DISKF) Zagreb, Nova Ves 66, HR-10000 Zagreb, Croatia, dgarasic@lycos.com Croatian Academy of Sciences and Arts (HAZU), Committee for Karst, Zrinjski trg 11, HR-10000 Zagreb, Croatia, mgarasic@zg.t-com.hr Six speleological sites are known to exist at the premises of the Dubrovnik Airport in Croatia. This is a highly weathered area that has been in the focus of attention of speleologists ever since the airport was built in 1961/62. Two vertical caves with the depth of 31 m and 10.5 m were discovered at that time. Both are now situated right underneath the new control tower of the Dubrovnik Airport. A tunnel entrance to the cave that has been known to local population for a long time is situated in the immediate vicinity of the control tower. In late 1950s, the cave entrances were closed with concrete because of a military airport construction, but a tunnel was built to enable access to the cave. The cave is about 200 m long and it fully occupies the space underneath the concrete runways of the Dubrovnik Airport. Thanks to efforts made by speleologists in 2010 the cave was adapted to enable tourist visits, and it is now the worlds only tourist cave underneath an operating airport. During apron extension activities in May 2012, three additional speleological sites were discovered and examined, together with other previously discovered caves, from the standpoint of geophysics, geology and speleology.1. Detailed speleological investigation (research and exploration)Three cavern systems were discovered at the site of the Dubrovnik Airports future apron on May 2012. Out of these systems, only the cavern No. 1 (K-1) complies with size requirements for speleological sites according to classification proposed by the UIS (Union Internationale de Splologie International Union of Speleology) where it is specified that speleological sites are only those sites that can be entered by humans. The entrance to the cavern K-1 was discovered during construction works conducted on March 2012. The original entrance measuring 0.30 0.30m was subsequently widened to 1.09 0.82 m (Fig. 1). After the initial opening of the cavern, its depth was measured, according to the supervising engineer, by lowering the measurement rope (Garai and Cvetkovi 2012). It was established that the cavern is about 22 m deep. The width of this vertical cavern (Fig. 3) was not established at that time. Former geophysical measurements by the GPR (ground penetrating radar) and exploratory drilling did not greatly point to the existence of cavern systems in this part Figure 1. Narrow passages in Cave 4 (or K-1) at Dubrovnik airport.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings66

PAGE 69

varies between 10 and 15, while the formations strike toward the northeast (from 55 to 58). Joints observed in the cave and in its immediate vicinity are up to ten cm in width, they are planar and smooth, and are locally filled with very fine-grained clayey material and with carbonate calcitic binder, in crystalline or amorphous form (Garai 1986, 1991). Intensive joint and fault zones varying from 3 to 4 m in width are situated in the area. Together with lithological and hydrogeological conditions, these zones have greatly influenced the genesis (formation) of the cavern. According to speleological survey, these are relatively young but presently inactive neotectonic faults, as locally proven by thin dripstone coatings that freely cover fault paraclases, especially in top parts of the cavern and, at that, they are neither fractured nor deformed. From the tectonic standpoint, the apron area is a part of the paraautochthon. The para-autochthon occurs in a coastal part, and is formed of Maastrichtian and Middle Eocene limestones and dolomites, preceded by Eocene and Lower Oligocene flysch deposition. Endogenetic movements during geological past, as well as exogenetic processes, have caused formation of the present day structure. Thus, orogenetic movements that occurred during the Helvetian phase resulted in thrusting of the geotectonic unit High Karst over the para-autochthon, from the northeast direction. Then the intensive faulting took place in the area under study, and significant faults of general NESW-trend (Slivnica Fault), and NS-trend (Zubak Fault), were formed. Structure inclinations and further reverse overthrusting occurred along these formations. Such past movements caused formation of blocks on which, especially next to transverse faults, we even now experience movements, which results in a pronounced seismic activity in the area under study. The Konavle Zone, as well as its wider surroundings, is characterized by linear folds of northwest-southeast strike, but transverse deformations of secondary significance are not so frequent in this area. Strikes are of the NESW orientation and are manifested as transverse faults with frequent relative lowering of one of the blocks. From the engineering-geology standpoint, the following can be said about Upper Cretaceous limestones or dolomites: the uniaxial compressive strength of the substratum (limestones at cavern level) amounts to 100 MPa, the RQD value differs, but is mostly higher than 75 percent, on an average discontinuities are spaced at 50 to 200 cm intervals, and are locally filled with clay. Cave systems sometimes exceed 20 m in length, and are normally 10 m long. Joints vary from 0.1 to 1 mm in aperture. In top parts, joints are rough and undulating, while they are smooth in lower parts of the cavern.2. Hydrogeology of cavesFrom the hydrogeological standpoint, it can be stated that a permanently active (but not primary) underground stream has not been found at this speleological site (caves), although the dripping water is probably present. The water flow in the cavern varies depending on climatic conditions on the ground surface. The water reaches the cavern via joints directly from the ground surface (to a lesser extent) or in deeper parts via joints and paraclases from other parts of Cretaceous carbonates (in most cases). Figure 2. Cave 5 (or K-2) at Dubrovnik airport. Figure 3. Vertical entrance in Cave 4 (or K-1) at Dubrovnik airport. Figure 4. Cave 4 (or K-1) speleothems near bottom of the cave.Even today, the dripping water exerts an intense mechanical and chemical action on the surrounding rocks (erosion and corrosion), thus creating speleothems (stalagmites, dripstone crusts Figs. 4, 5) The weathering depth in the area of this speleological site, despite the fact that these are practically impermeable rocks, is estimated at 300 to 500 m, and the zone of vertical circulation varies from 50 to 150 m. It is followed by the zone of horizontal circulation in which the ground water is carried via Cretaceous limestones toward submarine springs in the Adriatic Sea (Cavtat Bay and Molunat, ilipi coast). As the Upper Cretaceous (Senonian) and Paleocene (Danian) carbonate rocks are permeable from the hydrogeological standpoint, and this due to intensive Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings67

PAGE 70

secondary (joint) porosity, it may be expected that several similar caverns that can not be accessed from the ground surface are situated in the continuation of this joint, or parallel to it in a zone with genetically correspondent joints formed in the area of the same or similar fault due to general subduction. These caves might even be connected with the mentioned hydrogeological system in deeper zones of this part of the Konavle Zone. They are probably situated more than 30 meters beneath the apron level.3. SpeleogenesisThis speleological site (cavern) has been formed by widening of the fault paraclase with the strike of 51 and 120, and with a noticeable rotation of paraclase in intensely weathered Upper Cretaceous and Paleocene limestones and dolomites. The initial water that is significant in the speleogenesis of the cavern, has penetrated through surface joints, and has widened these joints through its corrosive and erosive action to the present day dimensions. The bottom part of the cavern was formed through regressive palaeo-influence of ground water. The processes occur even today, but only in the parts of the cavern that are at least some fifty meters below the apron level. A continuous water stream has not been noticed, i.e. only the speleothems were registered. Cavern dimensions may be even greater, but the narrowness of canals has prevented speleologists from advancing any further. This speleogenesis is related to the weathering processes that occurred after general uplift of the mountainous massif Cukali Zone or High Karst Zone starting at the end of Cretaceous and continuing to the present time.4. Geophysical investigations and surveyingThe geophysical survey was conducted using the seismic refraction method (three profiles: Profile 1, Profile 2 and Profile 3, 385 m in total length Fig. 7, Fig. 9): The physical disposition of profiles and test sites is presented in the following Figure 6. The test method was selected taking into account the geological structure of the terrain, and geotechnical nature of the problem with regard to design requirements. In this test method, the emphasis was placed on definition of longitudinal wave velocities. Based on seismic wave velocities and their spatial distribution, as determined during the testing, the following information relevant for geotechnical design can be determined and estimated: depth of layers, soil stiffness at small deformations, position of faults and fault zones, and ground water level. Seismic tests are conducted in order to define velocity of elastic waves across the depth. Wave velocities are directly linked with elastic stiffness properties of materials they are passing through. Tests are conducted at the ground surface and they belong to the group of non-destructive tests. Seismic methods are based on the propagation of elastic waves through soil or rock. Waves are initiated at the ground surface through generation of impulses or controlled vibrations. At the contact between layers the waves are reflected or refracted, and then they travel back to the ground surface. The wave refraction or reflection is operated according to the Snells law on propagation of light rays through a stratified medium known from optics. The wave coming from the top layer at an angle i1 from the vertical, and at speed v1, which arrives at the boundary with the bottom layer, is partly reflected at the same angle and Figure 5. Cave 4 (or K-1) speleothems at the first step. Figure 6. Situation of the Caves Nos. 4, 5, 6 (or K-1, K-2, K-3) at the Dubrovnik Airport.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings68

PAGE 71

goes back to the surface at the speed of v1, and is partly refracted at an angle i2 and goes to the underground at the speed of v2. If v2>v1 the wave refracts at an angle greater than that of the incoming wave and, if v2
PAGE 72

urovi a Cave is located in the area of the Airport Dubrovnik (Kovaevi 2006; Buzjak 2006), at its southern edge, near the control tower. Due to the intensive tourist traffic during the summer season and nearby tourist centres of Dubrovnik and Cavtat, airport management is interested in its use as a show cave.The wider area is characterized by karst proceses and forms, mainly dolines. The cave entrance is located on a small karst plateau developed in the Upper Cretaceous limestone and breccia where the Airport Dubrovnik was built 1960. The depth of the cave is 25 m (129 m above sea level and 49 m above the Konavosko polje level). The cave volume is about 9,000 m3. During the airport construction the caves natural entrance was closed and a 37 m entrance artificial tunnel was built. Its entrance is closed by an iron door. It is located 154 m above sea-level. According to literature, the cave was first visited and explored by the researchers of the cave fauna in the first half of the 20thcentury. First detailed speleological research was performed in 2001. The cave was formed in the Upper Cretaceous limestone and breccia beds along the fissures which controlled the passage directions (Kova evi et al. 2006). As observed during recent researches the dripping water is abundant in wet seasons. It is very interesting because the passages are below the airport where the surface is covered with concrete and asphalt. It is proved by a very branched fissure water circulation in vadose zone where the cave is Figure 10. Entrance of Cave under Dubrovnik Airport (Airport director and author of paper). Figure 11. Djurovica Cave under the Dubrovnik Airport after reparation for tourists visiting. Figure 12. Touristic part of Djurovica Cave in big channels. Figure 13. Touristic Stairways in Djurovica Cave under Dubrovnik Airport.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings70

PAGE 73

located today. Water temperature and pH were measured in two small cave pools. It can be concluded that urovi a cave is very suitable for tourist use (Fig. 11, Fig. 12, Fig. 13). In addition to this longest horizontal urovia Cave, five additional vertical speleological sites have also been found within the airport premises. Two of them, 31 m and 10.5 m deep, are situated under the control tower. Three more caves, one of which is 22 m deep, were discovered during apron extension works in May 2012. The most recent GPR methods and geoseismic and geoelectric methods were used in the survey and analysis of these speleological sites. According to speleological analyses, it is highly probable that several significant speleological sites (caves) are situated near the already discovered ones, and so future surveys will be oriented in that direction.ReferencesBuzjak N, 2006. Speleomorfoloke i hidroloke zna ajke urovia pilje ( ilipi, juna Dalmacija). Hrvatski geografski glasnik, 68/2, 57, Zagreb. COST project 620, 2004. Vulnerability and risk mapping for the protection of carbonate (karst) aquifers. EUR 20912 EN, Final report. Directorate-General Science, Research and Development, Brussels, Belgium. Garai M, 1986. Hidrogeologija i morfogeneza speleolokih objekata u hrvatskom kru. Disertacija, 1, Sveu ilite u Zagrebu, Zagreb. Garai M, 1991. Morphological and Hydrogeological Classification of Speleological structures (Caves and Pits) in the Croatian Karst area. Geoloki vjesnik, vol. 44, 289, fot. 3, sl. 4, Zagreb. Garai M, Cvetkovi M, 2012. Izvje e o rezultatima speleoloke prospekcije novootvorenih kaverni I rezultati geofizi kih ispitivanja stijenske mase u neposrednoj blizini kaverni. Faculty of Civil Engineering, University of Zagreb. 1, Zagreb. Kova evi T, Bre ak D, 2006. Elaborat o speleolokim, geomorfolokim, meteorolokim, paleontolokim, arholokim I biospeleolokim istraivanjima, te geodetskom I fotografskom snimanju urovi a jame-pilje ispod platforme Zra ne luke Dubrovnik u ilipima. DISKF Zbornik radova, Proceedings, 177, Zagreb. Markovi B, 1966. Osnovna geoloka karta OGK, list Dubrovnik, K34. Savezni geoloki zavod, Beograd. Markovi B, 1975. Tuma za OGK, list Dubrovnik, K34, 1, Beograd.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings71

PAGE 74

SOME INFORMATION ABOUT THE DEEPEST CAVES KNOWN IN CROATIAN KARST AREAMladen Garai 1, 2, 3, Davor Garai2 1University of Zagreb, Faculty of Civil Engineering, Department of Geotechnics and Geolog., Ka ieva 26, HR-10000 Zagreb, Croatia, mgarasic@grad.hr2Croatian Speleological Federation, Society for the Research, Surveying and Filming on Karst Phenomena (DISKF)Zagreb, Nova Ves 66, HR-10000 Zagreb, Croatia. dgarasic@lycos.com3Croatian Academy of Sciences and Arts (HAZU), Committee for Karst, Zrinjski trg 11, HR-10000 Zagreb, Croatia. mgarasic@zg.t-com.hr The investigation of 3 caves explored more than 1,000 meters in depth in the Dinaric karst area in Croatia, has been in progress for a considerable period of time. These are complex speleological features situated in the longest mountain range of the Dinara karst, i.e. at the Northern Velebit mountain range. In fact, these caves have been studied for over two decades now. The first one is a cave system of Lukina jama (Lukes Cave) Trojama cave, which has been investigated until the depth of 1,421 meters (Jali 2007; mda 1993). Its total length is 3,731 m and a new expedition will soon continue to investigate this pit through speleodiving in siphons. The second greatest cave by depth is Slova ka jama (Slovak Cave), 1,320 meters in depth, with cave chanals measuring 5,677 m in total length. The third greatest cave by depth is the Cave system of Velebita, reaching down to 1,026 m in depth, with the chanal length of 3,176 m (Baki 2006a, b). However, another 3 speleological sites, which can rightly be added to those deeper than 1,000 m, have recently been discovered. These are three caverns that were discovered during construction of the Sveti Ilija Tunnel that passes through Mt. Biokovo, in the Dinaric karst area. These caverns undoubtedly point to the link with the ground surface, while the rock overburden above the tunnel in the zone where the caverns were discovered ranges from 1,250 and 1,350 m. Bats from the ground surface were found in the caverns and, according to measurements, they are situated in the depth from 200 and 300 m below the tunnel level. This would mean that the depth of these newly found caves ranges from 1,450 and 1,650 m, when observed from the ground surface. There are several hundreds of known caves in Biokovo, and the deepest ones discovered so far are Jama Mokre noge (Wet Feet Cave) 831 m in depth, and Jama Amfora (Amphora Cave) 788 m in depth (Bockovac 1999; Baki et al. 2002; Lackovi et al. 2001). The investigations and surveys have been still in progress.1. IntroductionThe speleological team formed of experts from the Faculty of Civil Engineering University of Zagreb, and the Society for the Research, Surveying and Filming on Karst Phenomena (DISKF) from Zagreb, conducted a detailed speleological investigation of the newly opened caverns, vertical spelological sites (caves), situated in the so called service tube (tunnel) of the Sveti Ilija Tunnel at KM 1+415; KM 1+193, and KM 1+637 on the route of the Zagvozd Baka Voda link road in Dalmatia (Garai 2009a, b, c). According to the request placed by the client, the objective of the investigation was to prepare a documentation that will contribute to better understanding and possible repair of the speleological site crossing the tunnel line. Already at the time of discovery of the first cavern in tunnel, a good correspondence was noted with the results of statistical processing of tectonic elements and occurrences of speleological sites, which has been conducted since 1991 upon discovery of each speleological feature during tunnel construction in Croatia. Over one thousand of caverns studied in almost fifty tunnels in Croatian karst can statistically be approximated, through their morphology and geological parameters, to the prognosis that is dependent on tectonic, lithostratigraphic and hydrogeological conditions. The Sveti Ilija Tunnel passing through the Biokovo Mountain in the so called External karst belt is characterized by the following features. The entrance to the tunnel at the north side of Biokovo starts at KM 3+971, and the southside exit is located at KM 8+220. The main tunnel tube is 4,250 m long. The tunnel consists of the main tube and the service tube, which are connected to one another via crosslinks destined for the passage of pedestrians and emergency vehicles. The service tube is destined for the rescue and evacuation of participants in traffic in emergency situations. That is why it is first of all destined for fire-fighting vehicles and ambulances, for which an undisturbed access and manoeuvring space must be ensured. It is also used as an evacuation route for pedestrians. The main tunnel tube is 4,249 m long, while the service tube in 4,255 m long. The traffic will be operated (in 2013) via the main tunnel tube only, as the service tube must not be used for regular traffic. The driving speed will be limited to 80 km.h-1. In the beginning, the tunnel tube is in the right-side curve (600 meters in radius, with transition curve 200 m in length). The main tunnel tube has two driving lanes (each 3.25 m in width) and each of these lanes has two marginal strips (0.3 m). Thus, the total pavement width amounts to 7.7 m. When two elevated pedestrian sidewalks are added (each 0.75 m in width), the highest width of the tunnel tube is 9.81 m, while the overhead clearance in the mid-pavement is 6.85 m. The width of one driving lane in the service tube is dependent on the fire-fighting vehicle width, and amounts to 3 m. There are two footways for individual rescue of pedestrians (each 0.8 m wide). The maximum width is 6.06 m, and the overhead clearance in the mid-pavement is 5.26 m. The cross sectional profile of the footway totals Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings72

PAGE 75

1.6 m, the greatest width is 3.5 m, and the overhead clearance is 3.75 m. As the pavement structure in tunnels can not be realized in phases, it is dimensioned for the twenty-year design life. Thus, in the main tube, the pavement structure is composed of the following layers: asphalt concrete 5 cm in thickness, bituminised base course 10 cm in thickness, and mechanically compacted base course 25 cm in thickness. This all rests on the subgrade or embankment formed of stone materials. In the service tube, the pavement structure is formed of the bituminised base course (wearing course) 6 cm in thickness, and of the mechanically compacted base course 30 cm in thickness. These layers also rest on the subgrade or embankment formed of stone materials. In zones where invert is planned, the pavement structure lies on concrete subgrade. The north-side portal of the tunnel is situated at 360 m a.s.l., and the south-side one is at 224 m a.s.l. It should be noted that this tunnel has the highest overburden when compared to other road tunnels in Croatia. It locally reaches 1,380 m above the tunnel line. Interestingly, although no cavern was discovered during the works in the main tubes, as many as three vertical caverns were discovered in the main pressure and highest overburden zone of the service tube. In addition, after discovery and study of the first cavern in the service tube of the Sveti Ilija Tunnel at KM 1+415, it was concluded that, considering the geological situation and former statistical study, an another cavern should be located about 200 to 250 m away from this one. The speleologists have made the corresponding note in the diary and thus warned the contractor about possible problems. Soon thereafter, the second cavern was found at KM 1+193, some 222 m away from the first one. By comparing geological parameters, tectonics and statistical processing data, the speleologists have once again predicted an another cavern which should be situated 200 to 250 m away. The corresponding note was also made in the technical diary and the speleological report. Several weeks later, a cavern was discovered at KM 1+637, i.e. precisely 222 m from the second cavern. Thus, something that no one could predict with absolute certainty has actually happened. The distance between each of the 3 caverns discovered in the service tube is 222 m, which is really amazing.2. GeologyThe rocks in which the caverns were formed belong to the Lower Cretaceous BaremianAptian limestones K1 4,5, and these formations vary between 15 and 75 cm in layer thickness. The limestones are light to dark grey in colour, and brown intercalations of carbonate calcitic binder can locally be observed between the layers. In the immediate vicinity of the cave, and in the site itself, the inclination of carbonate formations varies between 45 and 55 and they strike toward the northeast (from 42 to 48 ). The fissures noted in the site and in its immediate vicinity are up to 10 cm wide, they are planar and smooth, very rarely filled with clayed material, and are very firm. They are more often filled with carbonate calcitic binder in crystalline or amorphous form. Intensive joint and fault zones varying from 3 to 4 m wide are situated in each the area where the three speleological sites were discovered. Together with lithological and hydrogeological conditions, these zones have greatly influenced the genesis (formation) of the caverns under study. Mylonitic fault zones were not registered. According to speleogeological survey results, these are relatively young but presently inactive neotectonic faults, as locally proven by thin dripstone coatings that freely cover fault paraclases, especially in top parts of the cavern and, at that, they are neither fractured nor deformed (Marin i et al. 1969a, b; Rai et al. 1968a, b; Marin i et al. 1972a, b). The strike of the dominant fault is 105, and the paraclase is fully vertical. The corrosion of the surrounding rocks is highly intensive. In case of dolomitic limestones, it is however less pronounced than in case of pure limestones. In fact, it is most probable that dolomitic limestones are transformed, in their upper part, into calcareous dolomites. Pure limestones and dolomitic limestones could be expected closer to the reverse (thrust) contact with older formations. These limestones, and also late-diagenetic dolomitic limestones, are crystalline, and are formed of dolomite grains distributed in form of a mosaic. Grains are filled with calcitic inclusions, and with some clay and limonite. Fossil remains have not been discovered in the cavern. Lithologically, these sediments are Baremian-Aptian limestones and locally dolomitic limestones undergoing transition into younger Upper Cretaceous formations. Dolomites are less often found and, when found, they are in form of a continuous package in the upper and top parts of the sedimentary column, or in form of intercalations in limestones, or as the dominant member with limestone intercalations. They are mostly coarse grained, with many calcitic micro-grained inclusions of zonal distribution, and Figure 1. Map situation of tunnel Sveti Ilija. Figure 2. Geological cross section through Mt.Biokovo tunnel.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings73

PAGE 76

their concentration reduces going from the centre toward periphery. They therefore correspond to metasomatic dolomites with the dolomitisation process completed, and so relicts of primary micritic structure are not visible. According to their type, these formations are calcitic dolomites (Blakovi 1998). The tunnel runs through central lower parts of all three caverns. Thus, if viewed from the tunnel axis, the caverns spread vertically toward the ground surface, but also vertically downwards. Dolomitic intercalations locally appear as shelves in vertical cave segments. Both the top and bottom parts of the cavern are of so called vertical or cave shape. They are characterized by generally large cross-sections, and follow the fault paraclase of EWtrend (105). The initial water, significant in the speleogenesis of these caves, had sunk through surface fissures (some of which were later on covered with clay) and has over time extended these caves to present dimensions through extensive corrosive and erosive action. These processes are still going on and, in this respect, it was noted that an intermittent stream runs through lower parts of the caverns. This steam enters the west part of the cavern through fissures and reaches the scree material at the cave bottom, and from there it continues to sink vertically further down into the underground. These caverns show signs of karstification by gravity in their upper parts, while there are traces of regressive karstification, i.e. aggressive karstification towards the ground surface, in the bottom parts of the caverns. Rare corrosive forms were noted. The caves may be greater in size, but this could not have been determined as further passage was prevented by collapsed and unstable blocks through which the water passes intermittently. This undoubtedly points to the link with a (presently unknown) greater underground space. The difference in temperature between the discovered part of the cavern (+8.9 to +9.1 C) and the undiscovered part of the cavern (+8.5 C) points to the possible presence of a permanent ground water body. From the engineering-geology standpoint, the following can be said about these Lower Cretaceous limestones or dolomitic limestones to possibly calcareous dolomites: uniaxial compressive strength of the substratum (cave-level limestones) amounts to 150 MPa, the RQD value differs, but is mostly higher than 70 percent, on an average discontinuities are spaced at 50 to 200 cm intervals, and are locally filled with clay. Cave systems sometimes exceed 20 m in length, and their usual length varies from 10 to 15 m. The joint aperture varies from 0.1 to 1 mm. In top parts joints are rough and undulating, while they are smooth in lower parts of the caverns. According to the RMR classification, the rocks in the zone where the caves were discovered can be classified as a good rock mass (category III). This is a rock complex that was classified in the prognostic geological documentation as the second geotechnical unit.Figure 3. Cavern at km 1+415 in Biokovo tunnel. Figure 4. Cave profile and plan of cavern at km 1+193 in Biokovo tunnel. Figure 5. Cavern at km 1+193 in Biokovo tunnel. Figure 6. Deeper part in Cavern km 1+637 in Biokovo tunnel. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings74

PAGE 77

3. Methods and ResultsState of the art speleological methods were used for measuring vertical distances and geological parameters. A special attention was paid to the radon concentration measurement, and to chemical analyses of ground water. According to current speleological site classifications (Garai 1986, 1991), the caves discovered at KM 1+415, KM 1+193, and KM 1+637, are large-size vertical speleological sites (caves) of an elbow shaped morphological type, characterized by the occurrence of the so called false bottoms, with heigh differences of about 297 m, 268 m and 203.5 m, respectively. This is the zone of the highest tunnel overburden (from 1,300 to 1,380 m). This means that the deepest parts of these caverns (accessed by bats from the ground surface rather than from the tunnel) sometimes extend to 200 m below the tunnel line. Therefore, from the standpoint of geology, these sites can be classified among the deepest speleological sites discovered so far in Dinara karst in Croatia (Garai 1986, 1989, 1991, 1995). The depths from the ground surface range from 1,350 to possible 1,650 m. Some speleologists will rightfully argue that no one has as yet descended into these caverns from the ground surface, and that their relative depth is smaller. Nevertheless, the objective geological depth established in this karst complex is certainly one of the greatest in Croatia. This has also been revealed by measurement of ground water properties and temperature, 222Rn and 218Po. It was established that these properties are comparable to those measured at other deepest speleological sites in Croatia. This will however be demonstrated during subsequent investigations. From the hydrogeological standpoint, it can be stated that intermittent and maybe continuously active underground streams run through these speleological sites. The presence of dripping water has also been established. This dripping water exerts an intensive chemical action on the surrounding rocks (corrosion), which results in formation of underground karrens. Smaller parts of these caverns are covered with speleothems (dripstone formations). The palaeo-hydrogeological function of these caves is characterized by sinkholes on the ground surface (although their entrances are most probably now caved-in or yet undiscovered), and perhaps by a part of cave canals of a swallow hole (ponor) in the Biokovo hinterland (which is less probable because of intensive neotectonic uplift of this mountainous mass). Inaccessible bottom, northern and eastern parts of the cavern might be linked with the speleological system of caverns passing through Biokovo and participating inFigure 7. Cavern at km 1+415 in Biokovo tunnel. Figure 8. Cave profile of one part of cavern at km 1+415 in Biokovo tunnel. Figure 9. Cave plan of cavern at km 1+415 in Biokovo tunnel. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings75

PAGE 78

formation of submarine springs in the sea below Biokovo. In fact, the ground water found in these speleological sites has to pass either below or through the Sveti Ilija Tunnel, as it re-emerges from numerous submarine springs in the coastal area of the Makarska (from sites of Dubac to Podgora and Dranica; Alfirevi 1969). The depth of karstification (weathering) in the zone of this speleological site is estimated at several hundreds meters, with an estimated maximum of 1,500 m, while the zone of vertical circulation ranges from 500 to 1,700 m. This is followed by the zone of inclined or horizontal circulation through which the water is carried toward the Adriatic Sea. Impermeable Triassic and maybe even Palaeozoic clastite or dolomite formations are situated in the substratum. As Triassic dolomites are from the hydrogeological standpoint permeable, due to intensive secondary porosity (jointing), it can reasonably be expected that several similar and genetically correspondent joints, formed in the anticline due to general subduction, are situated in continuation of these sites (or are parallel to them). These possible caves are not accessible from the surface and might be linked to the mentioned hydrogeological system in the lower parts of Biokovo. A note should also be made of the Pavlinovi i cave situated some ten kilometres to the northeast of the tunnel, where the ground water oscillation of as many as 236 m has been registered. This information is particularly relevant for the tunnel zone approaching the impermeable substratum. In fact, such increase in the ground water level shows that sometimes in rainy season the capacity of underground chanals becomes insufficient to transport the water through Biokovo toward the sea, and so the water level rises enormously once this temporary hydrogeological barrier is reached.4. ConclusionBased on the position, morphology and other geological characteristics of three newly-investigated vertical caves situated in the service tube of the Sveti Ilija Tunnel, it has been established that the speleogenesis of these caves can be compared with that of other deepest caves in Croatia. Although the caves were not accessed from the ground surface, i.e. their entrances are about 1,300 m below the ground surface, they have most certainly been formed under conditions of intensive weathering and can as such be analysed from the geological point as well. An excellent correspondence of their occurrence shows that statistical processing of caverns located in tunnels is finally giving some results in correlation with the tectonics. It should also be noted that a certain trace of karstification and dripstone formation has been found in Croatian karst in a deep borehole in the Adriatic Sea, where a speleothem was found at the depth of 3,125 m in a smaller cavern.ReferencesAlfirevi S, 1969. Jadranske vrulje u vodnom reimu dinarskog primorskog kra i njihova problematika. Kr Jugoslavije 6, Zagreb. Baki D, Lackovi D, 2002. Jama Amfora, -614 m najdublja jama Biokova. Velebiten 36, 16, Zagreb. Baki D, Paar D, 2006a. Croatia and the Deep Caves of Northern Velebit. Alpine Karst, vol 2., ed. J. and T. Oliphant, 105, Cave books, Dayton, USA. Baki D, 2006b. Ekspedicija Velebita 2005. Velebiten 43, 2, SO PDS Velebit, Zagreb. Blakovi I, 1998. The two stages of structural formation of the coastal belt of the External Dinarides. Geologica Croatica, vol 51/1, Zagreb. Figure 10. Cave profile and plan of cavern at km 1+637 in Biokovo tunnel. Figure 11. Cave plan of cavern at km 1+637 in Biokovo tunnel.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings76

PAGE 79

Bockovac 1999. Jama Amfora. Speleozin, 12, 7, 14, Karlovac. Garai M, 1986. Hydrogeology and Morphogenesis of the Caves in Croatin Karst. Ph.D. Thesis, 1, University of Zagreb, Faculty of Geology, Zagreb. Garai M, 1989. New conception of the morphogenesis and hydrogeology of the speleological objects in karst area in Croatia (Yugoslavia). 10. International Congress of Speleology, Procceedings, vol. 1, 234, s1. 8, Budapest, Hungary. Garai M, 1991. Morphological and Hydrogeological Classification of Speleological structures (Caves and Pits) in the Croatian Karst area. Geoloki vjesnik, vol. 44, 289, Zagreb. Garai M, 1995. Speleogeneza u okviru hidrogeologije kra i procesa karstifikacije. 1. Hrvatski geoloki kongres, Opatija, Zbornik radova, Proceedings, 177, Zagreb. Garai M, 2009a. Izvje e o speleolokom istraivanju, fotografskom i topografskom snimanju, te hidrogeolokim opservacijama u novootvorenoj jami (kaverni) u tunelu Sveti Ilija (servisna tunelska cijev), na stacionai km 1+193, na trasi spojne ceste Zagvozd Baka voda u Dalmaciji. University of Zagreb, 1, Zagreb. Garai M, 2009b. Izvje e o speleolokom istraivanju, fotografskom i topografskom snimanju, te hidrogeolokim opservacijama u novootvorenoj jami (kaverni) u tunelu Sveti Ilija (servisna tunelska cijev), na stacionai km 1+415, na trasi spojne ceste Zagvozd Baka voda u Dalmaciji. University of Zagreb, 1, Zagreb. Garai M, 2009c. Izvje e o speleolokom istraivanju, fotografskom i topografskom snimanju, te hidrogeolokim opservacijama u novootvorenoj jami (kaverni) u tunelu Sveti Ilija (servisna tunelska cijev), na stacionai km 1+637, na trasi spojne ceste Zagvozd Baka voda u Dalmaciji. University of Zagreb, 1, Zagreb. Jali B, 2007. Jamski sustav Lukina jama Trojama. 1, HPS, Zagreb. Lackovi D, Baki D, 2001. Izvje e sa speleolokog istraivanja jame Amfora u Parku prirode Biokovo. 16, Park prirode Biokovo, Makarska. Marini S, Maga N, Benek , 1972a. Osnovna geoloka karta 1:100,000, list Plo e, K 33-35. Institut za geoloka istraivanja Zagreb, Savezni geoloki zavod Beograd. Maga N, Marini S, Benek , 1972b. Osnovna geoloka karta 1:100,000, tuma za list Ploe, K 33-35. Institut za geoloka istraivanja Zagreb, Savezni geoloki zavod Beograd. Marin i S, Korolija Majcen 1969a. Osnovna geoloka karta 1:100,000, list Omi, K 33-22. Institut za geoloka istraivanja Zagreb, Savezni geoloki zavod Beograd. Marin i S, Korolija Majcen 1969b. Osnovna geoloka karta 1:100,000, tuma za list Omi, K 33-22. Institut za geoloka istraivanja Zagreb, Savezni geoloki zavod Beograd. Rai V, Ahac A, Pape J, 1968a. Osnovna geoloka karta 1:100,000, list, Imotski K 33-23. Institut za geoloka istraivanja Sarajevo, Savezni geoloki zavod Beograd. Rai V, Pape J, 1968b. Osnovna geoloka karta 1:100,000, tuma za list Imotski, K 33-23. Institut za geoloka istraivanja Sarajevo, Savezni geoloki zavod Beograd. mda B, 1993. Velebit Speleoforum 51, Brno.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings77

PAGE 80

HYPOGENE SPELEOGENESIS AND SPELEOTHEMS OF SIMA DE LA HIGUERA CAVE (MURCIA, SOUTH-EASTERN SPAIN)Fernando Gzquez1,2, Jos-Mara Calaforra1 1Water Resources and Environmental Geology Research Group, Dept. of Hydrogeology and Analytical Chemistry University of Almera, Crta.Sacramento s/n, 04120 La Caada de San Urbano, Almera, Spain, f.gazquez@ual.es, jmcalaforra@ual.es2Unidad Asociada UVA-CSIC al Centro de Astrobiologa, University of Valladolid, Parque Tecnolgico Boecillo, 47151, Valladolid (Spain) Sima de la Higuera Cave (Pliego, south-eastern Spain) has been recently adapted for speleological use. Nevertheless, knowledge of the hypogenic origin of this cavity is still quite limited. The peculiar genetic mechanisms could provide added value if the cave is exploited for speleotourism. By studying geomorphological features and speleothem characteristics, it has been possible to deduce the predominant speleogenetic mechanism (whether hypogenic or epigenic) that controlled the evolution of this cave. The hypogenic mechanism that gave rise to this cavity was associated with upflow of CO2-rich hydrothermal fluid from depth, and was unconnected to meteoric water seepage. In this paper we describe some of the geomorphological evidence and unusual speleothems in Sima de la Higuera Cave. Large scallops are found on the upper level (-74 m); these are related to the mechanism of hypogenic speleogenesis and generally indicate the direction of ascending flow. There are also corrosion crusts made of micritic calcite. In addition, bubble trails related to bubbles of rising CO2have been identified. Centimetric calcite spar speleothems frequently fill fractures in the host rock. Other typical hypogenic speleothems occur in this cave, including calcite raft cones, folia, cave clouds, tower coral and calcite raft deposits, all suggesting the influence of thermal water during the caves formation. Furthermore, the first reported occurrence of calcite raft double-tower cones has been described in this cave; their origin is linked to water table oscillations in Paradise Chamber (-82 m). At the deepest level (-110 m), Mn-Fe oxyhydroxides occur as a black coating totally covering the cave walls, usually over subaerial boxwork formations. The wide variety of speleothems unconnected to meteoric water seepage make Sima de la Higuera Cave one of the most unusual hypogenic caves in Spain.1. IntroductionSeepage of meteoric water in karstic terrains is the most common mechanism for cave development. However, besides this kind of cavity (epigenic caves), there is another type of genesis linked to rising hydrothermal fluids, usually rich in dissolved CO2and/or H2S. These hypogenic caves form as a result of circulation of thermal water, usually from depth, and are unconnected with superficial flow (Palmer 2011). The term hypogenic does not refer specifically to extremely deep caves but rather to the origin of the fluids responsible for these caves development (Klimchouk 2009). This kind of system is represented by around 5% of the cavities worldwide (Forti 1996; Forti et al. 2002). Geomorphological features and specific speleothems are the most useful tools for identifying whether the origin of a cave was hypogenic or epigenic (Audra et al. 2002). The morphologies generated by hypogene mechanisms depend on the characteristics of the host rock, the temperature at which dissolution occurs, and the nature of the gases in solution. For instance, H2S-rich thermal water gives rise to acid attack that is more efficient than corrosion produced exclusively by CO2(Forti et al. 2002). Furthermore, in H2Srich systems, the water-rock interaction frequently results in gypsum precipitation (Palmer and Palmer, 2012). Other morphological features typical of hypogenic conditions are bubble trails, bell-shaped condensationcorrosion domes, scallops and widespread corrosion pockets (Forti 1996). On the other hand, the speleothems and cave minerals formed under subaqueous conditions from a solution highly saturated in calcium carbonate can provide the evidence to support a hypogenic origin of caves. For example, speleothems such as large bisphenoidal calcite crystals (Lundberg et al. 2000), calcite raft cones (Audra et al. 2002), cave clouds and folia (Audra et al. 2009; Davis 2012), tower coral and calcite raft deposits (Hill and Forti 1997) all commonly occur in hypogenic caves. In the current paper, we describe and examine the hypogenic geomorphological features of Sima de la Higuera Cave (Murcia, South-eastern Spain), which has been recently adapted for speleological use. Together with the mineralogical and geochemical characteristics of some of its speleothems, these have enabled a preliminary model of the evolution of this cave to be established.2. Geological settingSima de la Higuera (Fig Tree Cave) is located in the Sierra de Espua, in the municipal district of Pliego (Murcia Region). Its entrance lies 485 m a.s.l. and its mouth is crowned by a large fig tree that gives the cave its name. Speleological exploration of the cave began in 1997, although there is evidence that it was discovered earlier than this date (Club Cuatro Picos and Club Pliego Espua 2001; Ferrer 2010). Its surveyed length is 5,500 m and its deepest part is 156 m below the cave entrance (82 m below the base of the entrance sinkhole) (Fig. 1B). The cave lies in Oligo-Miocene detrital and marly limestone (Fig. 1A). The carbonate sequence is quite fractured due to NW-SE pressure that has given rise to a series of joints and Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings78

PAGE 81

faults that subsequently determined the caves morphology, particularly its deeper levels. Significant hydrothermal springs currently arise in the vicinity of the cave, with temperatures ranging from 30 to 50 C. The cave contains strong evidence of a hypogenic origin. In this study, we use the term hypogenic as postulated by Palmer (2011), who suggested that hypogenic caves form due to the upward flow of deep-seated water, or by solutional aggressivity generated at depth below the ground surface. In Sima de la Higuera Cave the hypogene speleogenetic mechanisms are evidenced by the presence of different types of speleothems and geomorphological features typical of hypogenic caves, such as calcite raft cones, tower cones, cave clouds (mammillary crusts), folia, specific corrosion forms, cupola, condensation domes and scallops. Further evidence comes from the fact the ambient cave temperature is higher than the annual mean outside temperature of 13.8 C. The current cave temperature oscillates between 18.6 C and 21.7 C, increasing slightly in the deeper parts, and this indicates a significant positive thermal anomaly. Relative humidity of the cave air is between 87.5 and 90% (Club Cuatro Picos and Club Pliego Espua 2001). Figure 1. A. Location and geological setting of Sima de la Higuera Cave. Geological cartography modified from Kampschuur et al. (1972); B. Location of the main hypogenic geomorphological features (squares) and speleothem formations (circles) in Sima de la Higuera Cave (topography by the Cuatro Picos and PliegoEspua caving clubs (2001): 1. Scallops, corrosion domes and alteration crusts; 2. Bubble trails; 3. Bubble grooves on mammillary crusts; 4. Boxwork and ferromanganese coatings; 5. Calcite spars infilling fractures; 6. Tower coral; 7. Calcite raft cones; 8. Folia and cave clouds (mammillary crusts); 9. Piles of calcite rafts.Although the evidence points to deep hydrothermal water flowing through the caves in the past, present-day water inflow is entirely from infiltration of meteoric water. There are only a few vadose speleothems generated from dripwater (stalactites, stalagmites, etc.) in the shallowest levels, around -74 m, and above the level of Bath Chamber.3. Methodology3.1. Inventory of speleological features and sampling The speleothems and the speleogenetic forms of the Sima de la Higuera Cave were inventoried and photographed for classification and analysis. The locations of these features were included on the topographical map drawn by the Cuatro Picos and Pliego-Espua caving clubs (2001). The cave cones of the Paradise Chamber were inventoried and positioned relative to a topographic station located 85.2 m below the cave entrance. Distances were measured using a laser-distance meter Disto A3 of Leica Geosystems AG and an upgrade kit (DistoX) which adds a 3-axis compass, clinometer and a Bluetooth connection. This wireless instrument was connected to a PDA device where data were stored. Calibration was performed using Palm OS software designed by Luc Leblanc and adapted for the topographical Auriga software. A fragment of a dark-coloured boxwork blade (SHG) was taken from the roof of the Manganese Gallery, situated in one of the deeper levels of Sima de la Higuera Cave, at the -110 m level (Fig.2H, I). The sample comprised a mineral lamina, 5 mm thick with a sugary texture, whose outer surface was covered by dull greyish-blue deposits, rough in texture. One sample of a calcite raft consisting of thin, brownish calcite laminae (CR-01) was collected from the Paradise Chamber (-85 m). Lastly, a sample of powdered raft calcite was taken from the Four Paths Chamber (-117 m) where the piles of white raft calcite reach up to 2 m high (CR-02) (Fig. 2E). 3.2. Analytical methodology SEM microphotographs were taken using a HITACHI S-3500 instrument in high vacuum mode. The samples were previously dried and coated with graphite to increase electron transmissivity. The elemental chemistry was determined by EDX (Energy Dispersive X-ray Spectroscopy) microprobe at two points with different typology over the boxwork sample (Fig. 5). Microanalyses employed the same instrument coupled to an Oxford INCA 7210 X-ray detector. The diameter of the beam was approximately 1 m. (Table 1). Carbon concentration was not measured due to masking by the graphite coating. Quantitative chemical analysis of samples was done using X-ray fluorescence (wavelength dispersive XRF) with a BRUKER S4 Pioneer instrument. A subsample of dark material of the boxwork from Manganese Gallery (CMN) was extracted using a needle for subsequent mineralogical analysis by XRD (X-Ray Diffraction). The mineralogy of the internal crystalline Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings79

PAGE 82

following infiltration of meteoric water through the carbonate rock overlying the cave. Dripstones like stalagmites and stalactites are scarce in this cave level; however, some examples can be observed around the lakes. Typical features of phreatic speleogenesis such as scallops, cupolas and bell-shaped condensation domes appear in this area (Fig. 3A). Furthermore, dissolution forms like bubble trails (Fig. 3C) and alteration calcite crusts (Fig. 3B) are seen above the -85 m level. (2) Beyond this point, the cave morphology changes considerably, with larger galleries and chambers, such as Ghost Chamber and Paradise Chamber, which occupy an intermediate level (-85 m). This cave level accommodates speleothems such as calcite spars that fill fractures in the host rock, cave clouds, folia, tower coral and calcite raft cones (Fig. 2), all of them indicative of epiphreatic conditions during their precipitation. Of particular note are the calcite raft cones present in Ghost Chamber. In Paradise Chamber, 92 cave cones have been inventoried (Fig. 5) displaying two different morphologies. Thirty-seven of them can be considered as tower cones (or simple-tower cones), whilst the remaining fifty-five cones have a notch in the middle and look like two cones, one superimposed over the other(Fig. 2F). (3) Lastly, the deepest levels include labyrinthine galleries (three-dimensional maze caves) that are smaller in size and typical of hypogenic caves (Klimchouk 2009). Features like pendants, related to phreatic dissolution-corrosion have been identified in this level. Boxwork formations and ferromanganesic deposits also appear, particularly in Manganese Gallery (-110 m level) (Gzquez et al. 2012) (Fig. 2H). The SEM microphotographs enabled two visibly different zones to be identified (Fig. 4B). The first comprises submillimetric euhedral calcite crystals, some of which have a sphenoidal habit with well-defined faces and edges. The laminae was also determined by XRD of a powdered sample. The calcite raft samples (CR-01 and CR-02) were crushed and analyzed by XRD. Mineral analysis using XRD used a BRUKER APEX CCD area detector. The geochemical and mineralogical analyses were performed in the Servicios Centrales de Investigation of the University of Almera (Spain).4. ResultsObservations led us to identify three zones with different morphological appearances in Sima de la Higuera Cave: (1) The mouth of the cave gives access to a subvertical sinkhole 74 m deep, which is developed along the length of a diaclase running E-W, which finally opens out in Junction Chamber. This chamber and the galleries that communicate with it form one of the upper levels of the cave, which also run E-W. On this level appear several, small perched lakes (Coral Lake and Bath Chamber). The temperature of the lake water is 19.8 0.5 C, similar to the cave air (20.2 .2 C), while pH (8.10), conductivity (550 S/cm) and HCO3 -concentration (220 mg/l) are typical of cave infiltration water. This fact suggests that the only water inflow to these lakes come from dripwater Figure 2. Speleothems linked to the hypogenic origin of Sima de la Higuera Cave. A. Tower coral (floor), cave clouds (wall) and folia (ceiling) in the Bath Chamber level; B. Folia; C. Calcite spar crystals coated with ferromanganese oxyhydroxides; D. Calcite raft cones and cave clouds in Ghost Chamber; E. Piles of calcite rafts in Four Paths Chamber; F. Double-tower raft cones in Paradise Chamber; G. Volcano cone in Ghost Chamber; H and I. Boxwork and ferromanganese coatings in Manganese Gallery (Photos: Vctor Ferrer). Figure 3. Features related to the hypogenic origin of Sima de la Higuera Cave: A. Scallop in the entrance shaft; B. Alteration calcite crusts; C. Bubble trails and boxwork; D. Bubble grooves on cave clouds in Ghost Chamber; E. Corrosion forms due to acid dissolution appear in several places inside the cave. (Photos: Vctor Ferrer).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings80

PAGE 83

calcite mineralogy of this lamina was confirmed using XRD. EDX analysis revealed traces of Mn, Fe and Si, in addition to Ca, C and O (CMNcal, Table 1). Over this mineral lamina appear botryoidal structures. EDX analysis of their chemical composition indicated Mn concentrations of up to 38.2 % wt, and Fe of up to 10.3 % wt. Other elements like Na, Al and Si were present in quantities below 3 % wt. The XRD analyses revealed that this patina is composed of todorokite (NaMn6O12H2O) (22 %), pyrolusite (MnO2) (16 %), in addition to the calcite (42 %) and amorphous phases (20 %), probably poorly crystalline Fe oxyhydroxides (CMNmn, Table 1). Finally, the samples of calcite raft analyzed were composed exclusively of calcite, though the XRF analyses revealed traces of Si, Mg, P, S and other minor elements. The composition of samples taken from Paradise Chamber (CR-01) and Four Paths Chamber (CR-02) were similar (Table 1).5. Discussion5.1. Speleogenetic forms and dissolution-corrosion features Sima de la Higuera Cave hosts a great number of morphological features and speleothems, indicating that its origin is not linked to seepage of meteoric water. Typical features of hypogenic speleogenesis, such as white micritized rind and alteration crusts, appear in the upper levels and in the entrance sinkhole (Fig. 3B). The origin of these elements is usually connected to interactions between carbonate host rock and acid hydrothermal water (Palmer and Palmer 2012). Furthermore, forms typical of phreatic speleogenesis, such as the large scallop (Fig. 3A) identified along the diaclase that gives access to the main galleries, suggest slow upward flows during the speleogenetic stages. Besides, cupola and corrosion domes in the upper levels of the aquifer indicate interactions between the carbonate rock and the thermal water in its gaseous phase (due to lower hydrostatic pressure) here. Bubble trails have been identified on the ceiling in the intermediate levels, for instance, around Bath Chamber (Fig. 3C). Such features also formed due to acid aggressivity of thermal water; however in this case, gases (CO2and/or H2S) were released as bubbles that rose along preferential trails (Forti 1996) that eventually carved channels on the cave ceiling. Other features related to rising CO2bubbles were observed on the ceiling of Ghost Chamber (-85 m). Grooves several centimeters deep appear on the mammillary crust (cave clouds) that covers the ceiling of this chamber (Fig. 3D). When this cave level was under water, CO2degassing caused bubble trails on the hemispherical surface of the cave clouds. The continuous trail of rising bubbles led to dissolution-corrosion of the calcite crusts. Thermal water accessed the intermediate cave level mostly through a large fracture in the floor of Paradise Chamber (-98 m). This diaclase seems to have acted in the past as a feeder of deeper thermal water into this chamber. The cave forms found below the level of Paradise Chamber differ substantiallyfrom those in the upper levels. The are no dissolution features due to CO2bubbles. Instead, pendants, a cave form typical of phreatic dissolution, indicate a different mechanism of carbonate dissolution. The difference was determined by the different hydrostatic pressure of the solution containing the dissolved CO2. As the thermal water flowed toward the upper levels, the hydrostatic pressure fell and the volume of the gaseous phase increased. As a result, larger bubbles formed and these followed preferential trails that carved vertical dissolution grooves and channels. Other dissolution-corrosion forms, in this case generated under vadose conditions, appear in the lower cave levels. Dissolution-corrosion mechanisms affected the cave walls when the water table dropped and left the cave exposed. Wet CO2-rich air entered from lower levels and condensed on the slightly cooler cave walls and ceiling. Dissolution of the host rock by the condensed water increased as a result of the high CO2partial pressure in the cave atmosphere (Sarbu and Lascu 1997). Current CO2levels in the cave reach 2000 ppm and may have been even higher in the past due to degassing of hydrothermal water below the cave galleries. Dissolution-corrosion forms are particularly striking in Manganese Gallery (-110 m). Boxwork and ferromanganese coatings on the cave walls have been described (Fig. 2H, I) whose origin lie in (1) the Figure 4. Secondary electron images of (A) calcite rafts from Four Paths Chamber and (B) ferromanganese coatings on the boxwork of Manganese Gallery.ElementsCR-01XRFCR-02XRFCMNcalEDXCMNmnEDX(% wt)(-117 m)(-98 m)(-110 m)(-110 m) Ca52.851.838.12.3 O22.222.450.834.8 Cn.an.an.an.a Mg0.1110.110n.d0.7 Sr0.06160.0603n.dn.d Mn0.00320.00512.138.2 Fe0.06210.1731.510.3 Si0.3240.7780.72.4 Al0.1490.465n.d2.8 K0.03410.104n.dn.d P0.2610.0961n.dn.d S0.1030.0918n.dn.dTable 1. Elemental composition of the samples analyzed by X-ray fluorescence (XRF) and EDX microprobe. Analytical errors ranged from .33 for oxygen to .13 for aluminum. Errors for Fe and Mn were better than .2 (n.a = not analyzed; n.d = non detected). Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings81

PAGE 84

precipitation of sparitic calcite veins in the fissures of the carbonate host rock when the cave was submerged in thermal water; and (2) the corrosion of carbonates by acid generated from CO2diffusion into the condensed water, and oxidation of reduced Fe-Mn under aerobic conditions. The acid attack preferentially dissolved the carbonate host rock (which has a microcrystalline structure), while the veins of sparitic calcite were more resistant to corrosion. As a result, the calcite blades now project into the cave in the form called boxwork (Gzquez et al. 2012). On a microscopic scale, the Fe-Mn oxyhydroxides are botryoidal structures over a visibly altered sugary-textured calcite substrate (Fig. 4B). Oxidation of manganese and iron from the host rock gave rise to protons that acidified the medium, so lead to corrosion of the calcite beneath (Gzquez et al. 2012). 5.2. Hydrothermal speleothems Sima de la Higuera Cave contains a wide variety of speleothems of subaqueous origin which were generated from a solution highly saturated in calcium carbonate. As a general rule, this type of speleothem precipitates out from hydrothermal solutions that have a high CO2 content, frequently related to dissolution of the host rock during the genesis of the cave. Nevertheless, some cases have been reported where subaqueous speleothems (calcite raft cones, folia, etc.) have been generated from cold water in epigenic caves (Davis 2012). Notwithstanding, the relatively high ambient temperature of Sima de la Higuera Cave (up to 6 C higher than the annual mean temperature outside the cave) clearly points to this cavity retaining residual heat from an earlier hypogenic stage. Speleothems in Sima de la Higuera are not randomly distributed; rather, there is a gradation from the base of the cave towards shallower levels. This fact suggests that speleothem precipitation was controlled by the saturation state in calcite of the water, which in turn was conditioned by the hydrostatic pressure of the solution. Thus, crystallization of larger calcite crystals (Fig. 2C) was favoured in the lower cave levels, where slower CO2degassing caused low calcite supersaturation leading to slower calcite precipitation. As the hydrothermal fluid rose, CO2degassing intensified, resulting in other kinds of phreatic speleothems, such as cave clouds and folia (Fig. 2A, B) in the intermediate levels of Sima de la Higuera Cave. These type of speleothems usually form near the water surface (Audra et al. 2009; Davis 2012), where CO2degassing is more active and calcite precipitation faster than in the levels beneath. In some places, crusts of microcrystalline calcite cover the walls and ceiling of the intermediate levels (Fig. 2D). More recently, the groundwater level gradually fell and the water table intercepted the intermediate cave levels. In such a situation, precipitation of calcite raft laminae was favoured on the water surface due to intense evaporation and CO2degassing. Significant temperature and pCO2differences between the thermal water and the cave atmosphere triggered rapid precipitation of calcite rafts. This process would have started in the upper levels; nowadays however, calcite raft deposits are scarce in this shallower level as they would have been eroded during subsequent vadose stages. By contrast, there are abundant calcite raft piles in deeper levels, like Paradise Chamber and Four Paths Chamber (Fig. 2E). On a microscopic scale, the calcite rafts in Sima de la Higuera Cave display well-developed faceted crystals (Fig. 4A). This kind of dogtooth calcite crystal often precipitates at a low-medium temperature (Lundberg et al. 2000). A similar trace element composition of the calcite rafts at -82 and -110 m depth points to the precipitation of calcite laminae with similar characteristics. When the weight of the crystalline laminae is greater than the surface tension can support, rafts of calcite sink and accumulate on the pool bottom. Since the drips fall consistently at one point over a long period, piles of rafts lamina accumulate on the pool bed forming a cone-shaped speleothem, dubbed cave cones (Hill and Forti 1997). In Sima de la Higuera Cave, cave cones are particularly significant in the Ghost and Paradise Chambers (Fig. 2D, F, G). A new variety of cave raft cones has been recently discovered in Paradise Chamber, called calcite raft doubletower cones (Fig. 2F) (Gzquez and Calaforra 2012). They represent 60% of the cave cones in this chamber (Fig. 5). The unusual shape of these speleothems is explained by an intermediate phase of rapid calcite precipitation of uncemented calcite rafts covering cave cones formed in previous stages. When conditions favouring slow calcite raft formation were restored, new cave cones were formed exactly on top of some of the earlier cones (Gzquez and Calaforra 2012). The phreatic/epiphreatic conditions ceased when the water table definitively left the cave. Under the new vadoseFigure 5. Spatial distribution of cave cones in Paradise Chamber of Sima de la Higuera Cave, distinguishing simple-tower cones (n=37, in black) and double-tower cones (n=55, in grey). Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings82

PAGE 85

conditions, condensation and corrosion processes took over as dripwater became the main source of water to the cave. In some cases, a subtype of cave cone was formed as the water dripping onto the cave cones, being subsaturated in calcite, gradually drilled a hole in the apex of the cone, giving rise to the so-called volcano cones (Fig. 2G). Finally, tower coral speleothems appear near Bath Chamber (Fig. 2A). These subaqueous speleothems, less than 10 cm high, formed in a shallow pool where evaporation caused the near-surface water to be slightly more saturated with respect to calcite than the water nearer the pool bottom (Hill and Forti 2007). This mechanism favored the vertical coneshaped structure of these speleothems.6. ConclusionsThe geomorphological characteristics and subaqueous speleothems of Sima de la Higuera Cave are related to rising hydrothermal water. Speleothems in Sima de la Higuera are not randomly distributed over the vertical cave profile but show a gradation from the base to the top of the cave, suggesting that calcite precipitation and dissolutioncorrosion forms were determined by the hydrostatic pressure of the solution.AcknowledgmentsThe authors are grateful for the support of Andrs Ros, Pepe Liza and the Speleoclub G40 during the field work. Financial support was made available through the Spanish Science Grant AP-2007-02799, the Project RLS Exomars Science (AYA2011-30291-C02-02;Ministry of Science and Innovation, Spain and FEDER funds of EU). The authors thank Vctor Ferrer for his kind permission to use photographs to illustrate this paper. Sarah Steines is also acknowledged for improving our English.ReferencesAudra Ph, Bigot JY, Mocochain L, 2002. Hypogenic caves in Provence (France). Specific features and sediments. Acta Carsologica, 31(3), 33. Audra Ph, Mocochain L, Bigot JY, Nobcourt, JC, 2009. The association between bubble trails and folia: a morphological and sedimentary indicator of hypogenic speleogenesis by degassing, example from Adaouste Cave (Provence, France). International Journal of Speleology, 38(2), 93. Club Cuatro Pico, Club Pliego Espua, 2001. Sima de la Higuera. El mayor complejo subterrneo topografiado de la Regin de Murcia. Subterrnea, 16, 35. Davis DG, 2012. In defense of a fluctuating-interface, particleaccretion origin of folia. International Journal of Speleology, 41(2), 65. Ferrer V, 2010. La Sima de la Higuera (Pliego-Murcia). 80. Forti P, 1996. Thermal karst systems. Acta Carsologica, 25, 9917. Forti P, Galdenzi S, Sarbu SM, 2002. The hypogenic caves: a powerful tool for the study of seeps and their environmental effects. Continental shelf research, 22, 2373. Gzquez F, Calaforra JM, Rull F, 2012. Boxwork and ferromanganese coatings in hypogenic caves: an example from Sima de la Higuera Cave (Murcia, SE Spain). Geomorphology, 11718, 158. Gzquez F, Calaforra, JM, 2012. Origin of the double-tower raft cones in hypogenic caves. Earth Surface Processes and Landforms (accepted). Hill CA, Forti P, 1997. Cave minerals of the World 2. National Speleological Society, Huntsville. 461. Klimchouk AB, 2009. Morphogenesis of hypogenic caves. Geomorphology, 106, 10017. Kampschuur W, Langeberg CW, Montenat Ch, Pignatelli R, Egeler CG, 1972. Mapa Geolgico de Espaa 1:50,000, hoja n 933 (Alcantarilla). IGME, Madrid. Lundberg J, Ford DC, Hill CA, 2000. A preliminary U-Pb date on cave spar, Big Canyon, Guadalupe Mountains, New Mexico. Journal of Cave and Karst Studies, 62(2), 144. Palmer AN, 2011. Distinction between epigenic and hypogenic maze caves. Geomorphology, 134, 9. Palmer MV, Palmer AN, 2012. Petrographic and isotopic evidence for late-stage processes in sulfuric acid caves of the Guadalupe Mountains, New Mexico, USA. International Journal of Speleology, 41(2), 231. Sarbu SM, Lascu C, 1997. Condensation corrosion in Movile Cave. Journal of Karst and Cave Studies, 59, 99.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings83

PAGE 86

The analysis of morphological characteristics of dolines was conducted on tree case study areas: the Karst, the Logatec Plateau and Bela Krajina (Figure 1). The Karst Plateau (the Classical Karst) is situated in western part of Slovenia on elevation 220 m a.s.l., with an average yearly temperature 10.9 C and 1,399 mm of precipitation (1971; ARSO 2011).The case study area is situated on bedded granular Cretaceous dolomite (2.4 sq km) and platy laminated Cretaceous limestone (1.9 sq km) (Figure 2). The Logatec Plateau is situated in central part of Slovenia on elevation 530 m a.s.l., with an average yearlyTHE APPLICATION OF GIS METHODS IN MORPHOMETRICAL ANALYSIS OF DOLINES ON LIMESTONE AND DOLOMITE BEDROCKPetra Gostin ar Karst Research Institute ZRC SAZU, Titov trg 2, SI-6230 Postojna, Slovenia, petra.gostincar@zrc-sazu.si The dolines are considered to be a diagnostic karst landform. In this paper a GIS method, which was developed for the automatic delineation and classification of karst depressions (Obu 2010), was used for a study of morphometrical characteristics of dolines on limestone and dolomite bedrock. The spatial extent of auto-delineated dolines was compared to the manually delineated dolines from topographic maps. The analysis was conducted on three case study areas. The morphometrical characteristics of auto-delineated dolines show that the dolines on dolomite are larger in area but shallower. Their slopes are not as steep as the slopes of the limestone dolines and they are less regular-shaped. On the other hand, the dolines on limestone are smaller in area but deeper and their slopes are steeper. Their shape is more circular. This method, when used on DEM of higher resolution (e.g., LIDAR DEM) and a greater number of case study areas, could be used as an efficient tool for determining the characteristics of dolines on various bedrock and even as a part of computer-based detection of bedrock (geological mapping) in karst areas.1. IntroductionGeomorphometry is the science of quantitative land surface analysis. It offers a modern and analytical approach to the geomorphological analysis of the landscape with the use of digital elevation model (DEM) data and parametrization software. With the rapid increase of sources of DEMs today the methods of geomorphometry are more and more attractive to conduct (Pike et al. 2009). For the analysis of karst surface the methods of general geomorphometry alone (e.g., Woods method (Wood 1996), method of Surface Specific points (Peucker and Douglas 1975), the use of supervised and unsupervised classification (MacMillan et al. 2000)) or methods of specific geomorphometry applicable to the fluvial or any other geomorphic system (MacMillan and Shary 2009) do not provide good results as they are not adjusted for detection of specific karst landforms. The dolines are considered to be a diagnostic karst landform (Ford and Williams 2007). A number of attempts of auto-delineation of dolines and similar features (e.g., craters) with the use of DEMs and satellite images have been described in the literature (Portugal et al. 2004; Guimares et al. 2005; Matsumoto et al. 2005; Krgli et al. 2007; Siart et al. 2009; Suma et al. 2010; etc.). In this paper a GIS method, which was developed for the automatic delineation and classification of karst depressions (Obu 2010), was used for a study of morphometrical characteristics of dolines on limestone and dolomite bedrock. The results of GIS auto-delineation using DEM and manual delineation of dolines on topographic maps were compared. Later on, a comparison of characteristics of dolines on dolomite and bedrock obtained by autodelineation was made.2. Characterisation of study areaIn Slovenia karstic i.e. carbonate rocks (mainly limestones and dolomites) cover 43% of the surface (Figure 1) approximately 35% of the surface is on limestone and about 8% on dolomite (Gams 2003). Limestone and dolomite areas in Slovenia interchange on small distances as the area is fractured into numerous smaller segments due to sedimentary and tectonic modifications. Because of chemical and physical characteristics of the two rock types, karst on limestone and dolomite is different. Dolines are a common surface karst feature of Slovenian (Dinaric) karst and are present on both, dolomite and limestone areas. Figure 1. Distribution of limestone and dolomite in Slovenia with marked case study areas.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings84

PAGE 87

temperature 8.7 C and 1,587 mm of precipitation (1971; ARSO 2011). The case study area is situated on Jurassic dolomite (0.7 sq km) and Jurassic limestone (0.6 sq km) (Figure 3). Bela Krajina is situated in the south-eastern part of Slovenia on elevation 260 m a.s.l., with an average yearly temperature 10.4 C and 1,257 mm of precipitation (1971; ARSO 2011). The studied area is situated on Jurassic dolomite (1.4 sq km) and Cretaceous limestone (1.3 sq km) (Figure 4). Figure 2. Case study area 1: Karst. Figure 3. Case study area 2: Logatec Plateau. Figure 4. Case study area 3: Bela Krajina.3. Method3.1. Auto-delineation of depressions on karst (Obu 2010) A GIS method for automatic delineation and classification of karst depressions has been developed by Obu (2010). The delineation of karst depressions is conducted on the basis of the lowest elevation of depression watershed border cells. This elevation represents the level on which water starts to flow out of the depression when it is theoretically filled with water. The method introduces the existence of karst depressions on different levels. In our research the depression detected on level 1 were used for the morphological analysis. For the analysis ArcGIS software (the Spatial Analyst extension) was used, as it supports the Python programming language in which the script was written. The second part of the developed method is karst depression classification based on a number of attributes calculated for each depression. These attributes (e.g., area, circumference, depth, volume, average diameter, the length of the shortest axis through the centroid of a doline, the length of the longest axis through the centroid of a doline, roundness, the highest elevation, the lowest elevation, mean elevation, lowest slope, highest slope, mean slope, mean curvature, area of flattened part of doline, percentage of flattened floor comparing to the area of doline etc.) are stored in a database. A database with morphological characteristics for each delineated doline was used for comparison of morphological characteristics of dolines on neighbouring dolomite and limestone areas. Auto-delineation of dolines with GIS was conducted on DEM with a resolution 12.5 m (provided by Surveying and Mapping authority of Republic of Slovenia GURS). It was produced in 2005 with a method of data integration. An average height accuracy of the DEM is 3.2 m (Podobnikar 2008). 3.2. Manual delineation The dolines were also manually delineated (digitized) on topographic maps with a scale of 1:5,000 (provided by Surveying and Mapping authority of Republic of Slovenia GURS). The contour interval of these topographic maps is 10 m. With the manual delineation of the actual extent of dolines on case study areas the method of auto-delineation of depressions could be evaluated. At the beginning the comparison between the manually delineated and auto-delineated depressions was made.4. ResultsWhen comparing the number of delineated dolines it was calculated that auto-delineation was able to detect from 45% up to 87% of dolines comparing to manual delineation (Figure 5). In most cases the smaller depressions were not Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings85

PAGE 88

detected with GIS as the resolution of DEM 12.5 m is not suitable for their recognition. Details of the relief smaller than 12.5 m (cell size) cannot be detected as well as those up to 17.7 m (length of the diagonal of cells).This is especially visible at case study area of Logatec Plateau (Figure 3) where a large number of smaller depressions went undetected. Obu (2010) determined that only dolines with diameter greater than 30 to 50 meters are successfully recognised on a DEM 12.5 m. In many cases the spatial extent of manually delineated and auto-delineated dolines does not overlap entirely (see Figures 2, 3, and 4).Table 1. Comparison of number of manuallyand auto-delineated dolines. of dolines on dolomite is slightly smaller (0.6 on limestone; 0.7 on dolomite). The mean area parameter shows that the dolines on dolomite cover an approximately 30% larger area than dolines on limestone, and the percentage of flattened floor comparing to the area of doline is on average slightly bigger (dolomite: 3.4%; limestone: 3.2%). The morphometrical characteristics of auto-delineated dolines show that the dolines on dolomite are larger in area but shallower. Their slopes are not as steep as the slopes of the limestone dolines and they are less regular-shaped. On the other hand, the dolines on limestone are smaller in area but deeper and their slopes are steeper. Their shape is more circular. Figure 6. Circumference (mean, m)(auto-delineation). Figure 7. Depth (mean, m)(auto-delineation). Figure 8. Slope (mean, ) (auto-delineation). Manual delineationAuto-delineation Case studydetectedDolinesDetectedDolines areadolines/ km2dolines/ km2Dolomite2475513230 1 dolomite104435121 2 dolomite39593451 3 dolomite104744733 Limestone43111324364 1 limestone1809610757 2 limestone711114672 3 limestone1801409070A number of attributes of auto-delineated depressions were calculated. The results of analysis show a great variation when comparing the three case study areas. There is more variation of the parameters when comparing individual case study areas than the dolines according to the rock type. However, some general characteristics of the dolines on limestone and dolomite were determined. The mean volume of auto-delineated dolines on dolomite (3,375 m3) is slightly bigger than of those on limestone (3,208 m3); however the maximum volume is bigger on limestone (61,373 m3) than on dolomite (53,683 m3). On the other hand the average depth is smaller on dolines on dolomite bedrock (1.5 m; average depth of dolines on limestone is 1.8 m), and the mean circumference is much larger (213.3 m on dolomite; 180.8 m on limestone). The average slope of dolomite dolines (3.5) is smaller than of those on limestone (4.2). Additionally, the mean roundnessFigure 5. Number of detected dolines.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings86

PAGE 89

Table 2. Some morphometrical characteristics of auto-delineated dolines. Study area A B C D E F G H I Dolomite 213.3 1.5 7.1 3,374 53,683 49.9 0.6 3.5 3.4 1 dolomite 268.1 2.0 6.1 4,604 53,683 62.1 0.6 3.8 3.5 2 dolomite 157.4 0.8 3.9 798.0 4,500 37.5 0.7 2.4 3.8 3 dolomite 194.1 1.6 7.1 3,904 41,639 45.4 0.6 4.1 3.1 Limestone 180.8 1.8 10.6 3,208 61,373 43.1 0.7 4.2 3.2 1 limestone214.0 2.4 10.6 5,573 61,373 51.4 0.7 4.6 3.8 2 limestone164.1 1.1 4.2 1,156 11,953 38.5 0.6 3.0 2.7 3 limestone149.7 1.3 6.1 1,443 22,260 35.5 0.7 4.3 2.8 A: Circumference (m) B: Depth (mean, m) C: Depth (max, m) D: Volume (mean, m3) E: Volume (max, m3) F: Diameter (mean, m) G: Roundness (mean) H: Slope (mean, ) I: % of flattened floor comparing to the area of doline (mean) 5. Discussion and conclusionsEven though the GIS method for auto-delineation of dolines used for our research has a number of drawbacks, the general morphometrical characteristic of dolines on limestone and dolomite bedrock were calculated. There is a problem of determining the extent of dolines which directly influences the results of morphometrical analysis. The results of auto-delineation were compared with the manual delineation. When analyzing a landscape with smaller dolines the auto-delineation recognizes less than a half of existing dolines. However, the method is much more reliable in case of larger dolines. The comparison clearly showed that when analyzing the morphometrical characteristic of auto-delineated dolines we must take into account that a large percentage of existing dolines could have been undetected. When conducting a GIS analysis the results are limited by the quality of the data. For more accurate results LIDAR DEM should be used. For the doline study 12 points/m2, vertical accuracy of few dm, and positional accuracy half a meter would be needed (Triglav ekada 2011). The methodology also ignores the influence of the strong structural control which results in the morphology of the dolines e.g., elongation, depth, slope, etc. For more accurate calculation of the morphometrical parameters of dolines on different bedrock more case study areas would be needed too. Despite the presented drawbacks of the method, the proposed technique, when used on DEM of higher resolution and a greater number of case study areas, could be used as an efficient tool for determining the characteristics of dolines (or any other depressions, as a matter of fact) on various bedrock. Moreover, this data could be used for computer-based detection of the bedrock (geological mapping) in karst areas only with the analysis of the morphometrical analysis of the local dolines.ReferencesARSO, 2011. Opazovalne postaje (Observation stations), http://meteo.arso.gov.si/met/sl/climate/observation-stations/ ar J, 1986. Geoloke osnove oblikovanja krakega povrja (Geological bases of karst surface formation). Acta Carsologica, 14/15, 31. Digital Elevation model 12,5 12,5 m. Surveying and Mapping authority of Republic of Slovenia (GURS), Ljubljana, Slovenia. Ford D, Williams P, 2007. Karst Hydrogeology and Geomorphology. John Wiley & Sons, Chichester. Gams I, 2003. Kras v Sloveniji v prostoru in asu (Karst in Slovenia in time and space). Zaloba ZRC, Ljubljana. Geological map of SFRY: rnomelj, 1984. 1:100,000. Federal Geological Survey, Beograd. Geological map of SFRY: Postojna, 1967. 1:100,000. Federal Geological Survey, Beograd. Geological map of Slovenia (digital map). 1:100,000. Geological Survey of Slovenia (GeoZS), Ljubljana, Slovenia. Guimares RF, de Carvalho Jnior OA, de Souza Martins E, Ferreira de Carvalho AP, Trancoso Gomes RA, 2005. Detection of karst depression by aster image in the Bambui Group, Brazil. Proc. SPIE 5983, Brugge, College of Europe Press, 328. Jurkovek B, 2008. Geological map of the Northern part of the Trieste-Komen plateau 1:25,000. Geological Survey of Slovenia, Ljubljana, Slovenia. Krgli SO, Dypvik H, Etzelmller B, 2007. Automatic detection of circular depressions on digital elevation data in the search for potential Norwegian impact structures. Norwegian Journal of Geology, 87, 157. MacMillan RA, Pettapiece WW, Nolan SC, Goddard TW, 2000. A generic procedure for automatically segmenting landforms into landform elements using DEMs heuristic rules and fuzzy logic. Fuzzy Sets and Systems, 113, 81. MacMillan RA, Shary PA, 2009. Landforms and landform elements in geomorphometry. Geomorphometry Concepts, Software, Applications. Elsevier, Oxford, 227. Matsumoto N, Asada N, Demura H, 2005. Automatic Crater Recognition on Digital Terrain Model. Lunar and Planetary Science, 34, http://www.lpi.usra.edu/meetings/lpsc2005/pdf/ 1995.pdf Obu J, 2010. Prepoznavanje kotanj na podlagi digitalnega modela viin (Recognition of karst depressions on basis of digital elevation model). B.Sc. Thesis. Faculty of Arts, Department of Geography, Ljubljana, Slovenia. Peucker TK., Douglas DH, 1975. Detection of Surface-Specific Points by Local Parallel Processing of Discrete Terrain Elevation Data. Computer graphics and image processing 4, 4. New York. Pike RJ, Evans I, Hengl T, 2009. Geomorphometry: a Brief Guide. In Geomorphometry Concepts, Software, Applications. Elsevier, Oxford, 3. Podobnikar T, 2008. Nadgradnja modela reliefa Slovenije z visokokakovostnimi podatki. Geodetski vestnik, 52/4, 834.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings87

PAGE 90

Portugal RS, De Souza Filho CR, Bland PA, 2004. Automatic Crater Detection Using DEM and Circular Coherency Analysis A Case Study on South American Craters. 67th Annual Meteorological Society Meeting, www.lpi.usra.edu/meetings/ metsoc2004/pdf/5096.pdf Siart C, Bubenzer O, Eitel B, 2009. Combining digital elevation data (SRTM/ASTER), high resolution satellite imagery (Quicbird) and GIS for geomorphological mapping. A multicomponent case study on Mediterranean karst in Central Crete. Geomorphology, 112, 1, 106. Suma A, Prieto FJG, de Cosmo PD, 2010. Detection and mapping of karst depressions through remote sensing approach: an example from Sierra de Libar. Classical karsts, 18thkarstological school, Postojna, Slovenia. Triglav ekada M., 2011. Monost uporabe zranega laserskega skeniranja (LIDAR) za geomorfoloke tudije (Possibilities of aerial laser scanning (LIDAR) usage for geomorphic studies). Geografski vestnik, 83/2, 81. Wood JD, 1996. The Geomorphological Characterisation of Ditigal Elevation models: Ph.D. Thesis, University of Leicester, London, United Kingdom.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings88

PAGE 91

ORIGIN OF ATYPICAL CALCITE SPELEOTHEMS FILLING FISSURES IN SANDSTONESMicha Gradzi ski1, Marek Duliski2, Helena Hercman3, Andrzej Grny4, Stanisaw Przybyszowski5 1Institute of Geological Sciences, Jagiellonian University, Oleandry 2a, 30-063 Krakw, Poland, michal.gradzinski@uj.edu.pl2Faculty of Physics and Applied Computer Science, AGH University of Science and Technology, al. Mickiewicza 30, 30-059 Krakow,marek.dulinski@fis.agh.edu.pl3Institute of Geological Sciences, Polish Academy of Sciences, Twarda 51/55, 00-818 Warszawa, Poland, hhercman@twarda.pan.pl4Andrzej Grny, Geological Museum, Faculty of Geology, Geophysics and Environment Protection, AGH Academy of Science and Technology, Aleja Mickiewicza 30, 30-059 Krakw, Poland, a.gorn@op.pl5Stanisaw Przybyszowski, Kleczany Stone Resources Company Ltd, 33-394 Kleczany 176, Poland Atypical calcite speleothems developed in fissures in the Cergowa sandstones were found in the Kleczany Quarry (Polish Western Carpathians). They represent flowstone and stalactites, rafts and various sparry crusts. Such speleothems, especially phreatic ones, are uncommon in the Outer Carpathians composed mainly of siliciclastic rocks of flysch type, with only limited content of calcium carbonate. The analyzed speleothems grew in vadose and phreatic conditions as well as at the air-water interface. Phreatic speleothems and thin rafts comprise calcite crystals of eccentric morphology, as for example gothic-arch crystals, macrotrilate crystals, singleand multi-stepped crystals. They were formed due to disequilibrium conditions probably resulting from the presence of foreign ions in parent water. Flowstones were fed with thin film of seepage water flowing down the fissure walls. Based on their stable isotope composition, the majority of speleothems form two clusters. The first of them is characterized by 18O values between -9.8 and -8.5 and of 13C values between -5.7 and -0.6. The second cluster of samples yields 18O values between -9.4 and -7.3 and values of 13C from -11.5 to -9.7. The speleothems originated between 230 +14, -13 ka and Holocene, mainly during warm periods. Phreatic speleothems, including massive rafts, precipitated from ascending water of deep circulation whereas vadose and water-table speleothems crystallized from local infiltration water charged with soil CO2. Mixing of both waters in a shallow phreatic zone is plausible near the water-table. It affected the morphology of crystal composing the thin-rafts. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings89

PAGE 92

UPLIFT EVIDENCE FROM KARST MORPHOLOGY: PRELIMINARY EVIDENCE FROM BLAMBANGAN PENINSULA KARST, INDONESIAEko Haryono Karst Research Group, Fac. of Geography, Gadjah Mada University, Yogyakarta, Indonesia, e.haryono@geo.ugm.ac.id This paper aims at exploring morphological evidence of uplift history from karst morphology. Special interest is attributed to the relationship between phases of uplift on one hand and karst landform development on the other hand. Morphology variables involved in this research are karst hill distribution, karst hill density, karst valley drainage density, and valley width. Morphology data were acquired mainly from panchromatic aerial photograph (at the scale of 1: 50,000) incorporated with field survey. Field survey was conducted to collect cave entrance altitude. The result shows that karst hill dispersion, karst valley density, karst valley depth, and karst valley width were proven having been differentiated as a result of upliftin g phases. The oldest terrace is characterized by scatter karst hills with no karst valley networks; middle terrace is recognized from deep incised valley with high valley density, whereas the younger terrace is characterized by nearly flat morphology with no karst valley and karst hills. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings90 1. IntroductionKarst development in the convergence plate margin in many localities has been reported being governed by uplift history of karst host rock. Uplift attains its important roles in the karst development through the fact that karstification will take place only if the karst rocks have been uplifted to the sub-aerial environment. Uplift has been found having close association with speleogenesis (Mylroie and Carew 1995; Taboroi et al. 2003), epikarst development (Klimchouk 2004), dry valley terrace development (Sartono et al. 1978; Urushibara 1995; Urushibara 1997). Being driven by uplift, karst landforms in vice versa record the uplift history through their morphology, the so-called morphological evidences. Multilevel cave systems associated with past sea water level were documented in many areas being resulted from uplift phases (Urhushibara 1997; Taboroi et al. 2003). The caves associated with sea water level still stands were developed through corrosion at the mixing zone either between descending meteoric water and phreatic water or between phreatic water and saline water near shoreline. Marine and river terraces are also ubiquities documented in association with uplift history (Brahmantyo et al. 1998; Urushibara and Yoshino 1997; Sutikno and Tanudirdjo 2006).Series of these terraces are found in Gunungsewu Karst of Java-Indonesia. Evidences of uplift from cave levels and terraces have been much documented. However, uplift evidences from dry valley and karst hill morphology is less explored. Karst valley and karst hills in this case imprint the time that is consumed by karstification. It has been suggested by Lehman (1936) from his observation of Gunungsewu Karst that karstification is initiated by the development of dry valleys. As karst develops, dry valleys are disorganized to form conical karst morphology. Based on this theory the research documented herein tries to explore the morphological evidences of uplift from dry valley and karst hill morphometry.2. The Study AreaBlambangan Peninsula is the eastern-most fragment of carbonate formation which is physiographically situated in the Southern Zone or Plateau Zone of Java Island (Fig. 1). The southern zone is bordered in the north by steep escarpment. Not like generally Southern Zone where north of the escarpment are occupied by fluvio-vulcanic material and volcanoes, the north escarpment of Blambangan Peninsula shares a border with Bali Strait. In the west, Blambangan Peninsula is separated from coastal-alluvial plain of Segara Anakan by escarpment. To the south, the plateau of Blambangan Peninsula gradually lowers with several terraces.Figure 1. Blambangan Peninsula, the study area.

PAGE 93

Blambangan Peninsula constitutes mainly coral limestone of Punung Formation (Achdan and Bachri 1993). The limestone underlies volcanic breccias and conglomerate of Wuni Formation. In general, the limestone is massive, composed mainly of packstone and rudstone. The limestone outcrop is uplifted between 0 and 325 meters above mean sea level with thew highest segment situated in the western part. The limestone of Blambangan Peninsula was deposited during late Miocene up to Middle Miocene (Achdan and Bachri 1993; Bemmelen 1977; Balazs 1968). The prevailing contemporary climate in the Blambangan Peninsula is strongly influenced by the Northwest and Southeast monsoon, which result in wet season from October to April and a dry season between May and September. The average annual rainfall is 1,350 mm.3. MethodsKarst morphology data were acquired mainly from 1:30,000 panchromatic aerial photograph interpretation incorporated with field survey. Interpretation was conducted complementary under Topcon Mirror Stereoscope and under computer (on screen interpretation). First, aerial photographs were examined under mirror stereoscope to get 3D view and then delineation was conducted directly on screen used in this over geometrically corrected aerial photograph. Geometric correction was employed in Arc View through polynomial procedure. Minimal of nine control points were employed to every single photographs. Field survey was conducted to collect data of caves entrance altitude and to observe the nature of karst morphology. Dip and strike of bedding plane were also measured during field survey.4. Results and DiscussionThree difference pattern of karst morphology can be observed from aerial photographs of Blambangan Peninsula. From the higher to the lower part, pattern of karst morphology are described as follows: Upper terrace, characterized by scattered karst hills with extensive corrosion plain, The middle terrace, characterized by well organized deep incised dry valley, The lower terrace, characterized by shallow valley indicating less develop karst area (Figure 2). Those three karst morphology differentiation must have resulted from three major uplifts. The phases of uplift took place in Blambangan Peninsula are confirmed by the occurrence three difference levels of caves (Table 1). Most of the caves found in the area are horizontal caves. Since there is no clear field evidence that cave occurrence is associated with bedding plane, the development of caves in the Blambangan Peninsula must have been associated with uplift through the formation of cave at former water levels. Development of horizontal cave close to water table have Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings91 Figure 2. Morphological features of Blambangan Peninsula.

PAGE 94

been observed in many karst areas by previous works Mylroie and Carew 1995; Taborosi et al, 2003) through a mixing water mechanism. This result echoes the previous works in Gunungsewu Karst (Urushibara 1995; Urushibara and Yoshino 1997) and in Karangbolong Karst (Brahmantyo 2005) suggesting that since Middle Pliocene, Southern Zone of Java Island have experienced three major uplifts. However karst morphology differentiation related to uplift history is more distinguishable in this area, compare to Gunungsewu Karst and Karangbolong Karst. 4.1. Morphology in the upper terrace Karst morphology of upper terrace of Blambangan Peninsula is characterized by scattered karst hills with extensive corrosion plain. Scattered karst hills with extensive corrosion plain in the upper terrace indicate that karstification had taken place for much longer period than those during the lower terrace formation. Referring to karst development model (Grund 1914; Lehmann 1936; Aref et al. 1987), karst morphology of upper terrace is considered being in the stage of old karst landform development. The existence of corrosion plain in this area indicates that karstification must had took place intensively before the lower terraces were uplifted. The other evidence of intensive karstification in the upper terrace is the occurrence of degraded caves in the highest terrace (third level caves). The caves found in the uppermost terraces are generally collapsed big chamber and shallow. Major speleothem found within the caves are column. In some localities like in Trisula Cave (242 m m.s.l.) huge column are situated in the outermost part of cave entrance. 4.2. Karst morphology in the middle terrace Karst morphology of middle terrace in Blambangan Peninsula is characterized by well organized deep incised dry valley network. It indicates that karst morphology of the middle terrace is not developed yet. Referring to the karst development model of Lehmann (1936), karst morphology of middle terrace is considered in the stage of incipient karst. In this stage, fluvial processes are still prominent in sculpturing karst landform. It is confirmed by evidence that running water takes place during rainy season. No blind valley or pocket valley are found in the middle terrace. 4.3. Karst morphology in the lower terrace Morphology of the lower terrace does not indicate any karst landform. The lower terrace is the latest parts of the carbonate formation in the area which was exposed to the sub aerial environment. No karst remnant is found indicating that the lower terrace is a newly uplifted limestone formation rather than a planation surface. The1Goa Istana 94 2Goa Basori 60 3Goa Lowo46 4Goa Baru25 5Goa Jengger45 6Goa L40 7Goa Dampit38 8Goa Dobol 348 9Goa Dobol 231 10Goa Dobol 123 11Goa Kaji72 12Goa Satrio74 13Goa Gajah Mungkur80 14Goa Garuda75 15Goa Mayangkara109 16Goa Padepokan110 17Goa Mangkleng138 18Goa 45144 19Goa Rajawali136 20Goa Angkrik152 21Goa Trisula242 22Goa Putri247 23Goa Gentong175FIRST LEVEL CAVE SECOND LEVEL THIRD LEVELStatic pool Underground river Dry Dry Dry Dry Dry Dry Dry Dry Drip water Drip water Dry Drip water Dry Dry Drip water Dry Dry Dry Dry Dry Dry Altitude NoCaves(metersCaveWater occurrence from msl)Level Table 1. Major Caves in Blambangan Peninsula. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings92

PAGE 95

lower terrace is separated from the middle terrace by eastwest direction structural terrace.5. Concluding remarksUplift history of Java Island was documented very well in Blambangan Peninsula, either from morphological evidence as discuss here or structural evidence. Three major uplifts as describe from different part of the island are clearly documented in this area. Different evidence from cave sediment seems also very interesting topic for further research, especially for dating purpose, neo-tectonic, or paleo-climate reconstruction.ReferencesAref El MM, Kadrah AMA, Lotfy ZH, 1987. Karst topography and karstification processes in the Eocene limestone plateau of El Bahariya, Z. Geomorphologie N.F, (31)1, 45. Achdan A, Bachri S, 1993. Geologic Map of The Blambangan Quadrangle-East Java, Centre for Geological Research and Development, Bandung. Balazs D, 1968. Karst Regions in Indonesia, Karszt es Barlangkutatas, Volume 5, Budapest. Bemmelen RWVan, 1970. The Geology of Indonesia, Vol IA, Second edition, Martinus Nijhoff, The Hgue. Brahmantyo B, Puradimaja DJ, Bandono, Sadisun IA, 1998. Interpretasi kelurusan dari citra SPOT dan hubungannya dengan pola pengaliran bawah tanah pada Perbukitan Karst Gunung Sewu, Jawa Tengah bagian selatan, Buletin Geologi, 28 (1), 37. Brahmantyo B, 2005. Perkembangan bentangalam karst Gombong Selatan, dengan geologi sebagai faktor kendali, Disertasi, Tidak dipublikasikan, ITB, Bandung. Gillieson D, 1996, Caves: Processes, Development, and Management, Blackwell, Oxford. Grund A, 1914. Der geographische Zyklus im Karst, Z. Ges Erdkunde, 52, 621. Mylroie JE, Carew JL, 1995. Karst Development on carbonate Islands, in Budd, D.A. Sallet A.H. dan Harris, P.A. (eds), Unvconfromity in Carbonat strata-Their Reconition and the significance of Associated Porosity, AAPG Meomoir, 63, 55. Pannekoek AJ, 1949. Outline of The Geomorphology of Java, Leiden. Taboroi D, Jenson JW, Mylroie JE, 2003. Zones of enhanced dissolution and associated cave morphology in an uplifted carbonate island karst aquifer, northern Guam, Mariana Islands, Speleogenesis and Evolution of Karst Aquifers. 1 (4), December 2003, 16. Klimchouk A, 2004. Towards defining, delimiting and classifying epikarst: Its origin, processes and variants of geomorphic evolution, in Jones, WK, Culver, D.C. and Herman, J. (Eds.). 2004. Epikarst. Karst Water Institute special publication, 9, 23. Lehmann H, 1936. Morphologiche Studien auf Java, Gogr, Abh, 3, Stutgart. Sartono S, Hidayat S, Zaim J, Nababan UP, Djubiantono T, 1878. Undak Sungai Baksoko Berdasarkan Analisis Foto Udara, Proyek Pendendalian dan Penggalian Purbakala, Dep P dan K. Sutikno, Tanudirdjo D, 2006. Kajian Georkeologi Kawasan Gunungsewu sebagai Dasar Model Pelestarian Lingkungan Karst, Laporan Penelitian, UGM, Yogyakarta. Urushibara-Yoshino K, 1995. Environmental change in the karst areas on the Island of Java, Proceedings of The International Symposium on Paleoenvironmental Change in TropicalSubtropical Monsoon Asia, Research Centre for Regioanal geography, Hiroshima University. Hiroshima. Urushibara-Yoshino K, dan Yoshino M, 1997. Palaeoenvironmental change in Java Island and its surounding areas, Journal of Quarternary Science, 12 (5), 435.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings93

PAGE 96

VALLEY INCISION IN THE NZKE TATRY MTS. (SLOVAKIA) ESTIMATED BASED ON PALEOMAGNETIC AND RADIOMETRIC CAVE SEDIMENT DATINGSJaroslav Kadlec1, Pavol Bella2, Kristna kov1, Darryl E. Granger3, Helena Hercman4, Peter Holbek5, Martin Chadima1,6, Monika Orvoov5, Petr Pruner1, Petr Schnabl1, Stanislav lechta1 1Institute of Geology AS CR, v.v.i., Rozvojov 269, 165 00 Praha 6, kadlec@gli.cas.cz2Sprva slovenskch jask Hodova 11, 031 01 Liptovsk Mikul3Earth and Atmospheric Sciences, Purdue University, West Lafayette, Indiana 479074Institute of Geological Sciences PAS, Twarda 51/55, 00-818, Warsaw5Slovensk mzeum ochrany prrody a jaskyniarstva, kolsk 4, 031 01 Liptovsk Mikul6Agico spol. s r.o., Je n 29a, 621 00 Brno Up to eleven horizontal cave levels occur at different altitudes in Jnska, Demnovsk and Monick karst valleys in the Nzke Tatry Mts. Most of the caves are filled with allochthonous sediments transported from the area formed mostly by granite. The cave levels were filled with fluvial sediments in dependence on the valleys incision caused by Neogene and Pleistocene uplift of the mountain range. The fluvial sediments are intercalated with, or capped, by flowstone layers in the caves. The paleomagnetic polarities measured both in clastic and chemogenic sediments indicate the age of deposition. Based on obtained polarity data we are able to distinguish cave sediments deposited during the Brunhes, Matuyama and Gauss chrons. The paleomagnetic interpretation was partly verified by U-series datings of flowstones preserved in the sedimentary sections. Except for the horizontal cave levels located in the karst valleys, additional large cave systems were found at extremely high altitudes in the Nzke Tatry Mts. 600 m above the lowest horizontal cave level.1. IntroductionAn estimation of deposition time of inner cave sediments is possible based on measurement of paleomagnetic polarity record preserved in fine clastic and chemogene deposits. The paleomagnetic interpretation can be verified by U-series dating of speleothems intercalated into the sedimentary sections (e.g., Bosk and Pruner 2011).2. Geography and geologyTriassic carbonate rocks forming the northern slopes of the Nzke Tatry Mts. in Slovakia are hosting cave systems developed in eleven levels (Droppa 1966). Cave passages located in different altitudes were filled with fluvial, flood and chemogenic deposits during the Demnovsk and Jansk karst valleys incision. Except for the horizontal cave levels located in these karst valleys, additional cave systems were developed at high altitudes in the Nzke Tatry Mts. in Krakova Hoa, Kozie Chrbty, and Ohnit Plateau.3. MethodsPaleomagnetic polarities recorded in both fine clastic and chemogenic cave deposits were acquired by thermal or alternating field demagnetization. Directions of magnetization vector were measured using both spinner and superconducting rock magnetometers. Magnetostratigraphic inter-pretations were verified by U-series (230Th/234U and234U/238U)dating of speleothems deposited in the sedimentary sections. The paleomagnetic age of sediments preserved in the high altitude cave systems was verified based on cosmogenic isotope activity (10Be and 27Al) measured in quartz pebbles.4. Results and discussionBased on obtained magnetostratigraphic pattern we are able to distinguish cave sediments deposited during the Gauss (2.588.58 Ma), Matuyama (0.781.588 Ma), and Brunhes (<0.781 Ma) paleomagnetic chrons. The paleomagnetic record extracted from the high altitude cave systems indicates deposition between 3.1 and 1.75 Ma ago. The sediments had to be deposited during a period of tectonic stability when only shallow valleys were developed on both sites of the mountain range. The tectonic uplift in this area was accelerated since 1.7 Ma. The valley incision rate can be estimated based on paleomagnetic polarity and elevation data as follows: 6 cm.ka-1during 1.7.1 Ma; 32 cm.ka-1during 1.1.78 Ma; and 4 cm.ka-1since 0.78 Ma. The remarkable increase of the valley incision during the latest Early Pleistocene corresponds with results of valley incision study performed in the Swiss Alps reaching twice to four times faster rate in comparison with ther Nizke Tatry Mts. (Haeuselmann et al 2007). The periods of alluvial agradation in the caves were also correlated with terraces deposited by the Vh River running 10 km North from the karst area. Paleomagnetic polarity of the fluvial deposits filling the largest 4thcave level show the age of sediments younger than 0.78 Ma which is in agreement with stratigraphic conclusion of the most extensive Vh River terrace to the early Mid-Pleistocene (Droppa 1966).5. ConclusionsThe cave levels in the Nzke Tatry Mts. were filled with fluvial sediments in dependence of Neogene and Pleistocene uplift of the mountain range. Sediments preserved in the large cave systems developed in extremely Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings94

PAGE 97

high altitudes in the Nzke Tatry Mts. Were deposited between 3.1 and 1.75 Ma ago (paleomagnetic chrons Gauss and Matuyama). Sediments deposited in cave levels developed in Demnovsk, Jnsk and Monica karst valleys in a relative elevation above 185 m are older than 2.588 Ma (Gauss chron).Fluvial sediments found in theFigure 1. Magnetostratigraphic pattern of cave sediments in the Nzke Tatry Mts. karst valleys. Black (white) squares represent normal (reverse) magnetization of sample, black-and-white squares represent transitional character of sample. DC Demnovsk Jaskya Slobody Cave. cave levels in the relative elevations 30 m were deposited 2.588.781 Ma ago (Matuyama chron). Fluvial sediments deposited in lower elevation cave levels are younger than Brunhes/Matuyma paleomagnetic boundary (0.781 Ma). After relative tectonic stability period between 3.1 and 1.7 Ma the tectonic uplift was accelerated since 1.7 Ma. The valley incision rate can be estimated based on paleomagnetic polarity and elevation data as follows: 6 cm.ka-1during 1.7.1 Ma; 32 cm.ka-1during 1.1.78 Ma; and 4 cm.ka-1since 0.78 Ma.AcknowledgmentsThe research is supported by grants No. IAA3013201 and IAA30013001. The institutional funding is provided by the Institute of Geology AS CR, v. v. i. (AV0Z30130516).ReferencesBosk P, Pruner P, 2011. Magnetic rekord in cave sediments. In Petrovsk E, Herrero-Bervera E, Harinarayana T, Ivers D (Eds.) The Earths magnetic interir, 343. Droppa A, 1966. The correlation of some horizontal caves with river terraces. Stud. Speleol., 1(4), 186. Haeuselmann P, Granger DE, Jeannin P-Y, Lauritzen S-E, 2007. Abrupt glacial valley incision at 0.8 Ma dated from cave deposits in Switzerland. Geology, 35(2), 143.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings95

PAGE 98

MAGNETIC FABRIC AND MINERALOGY OF CAVE DEPOSITS IN BOTOVSKAYA CAVE (EASTERN SIBERIA, RUSSIAN FEDERATION)Jaroslav Kadlec1, Helena Herman2, Martin Chadima1,3, Lenka Lis1, Hedi Oberhnsli4, Alexandr Osintsev5 1Institute of Geology AS CR, v.v.i., Rozvojova 269, CZ-165 00 Praha 6, Czech Republic, kadlec@gli.cas.cz2Institute of Geological Sciences, PAN, Warszawa, Poland3Agico spol. s r.o., Jen 29a, 621 00 Brno4GeoForschungsZentrum Potsdam, Germany5Speleoclub Arabica, Irkutsk, Russia The Botovskaya Cave is a typical example of a two-dimensional maze with a total length of explored passages exceeding 60 km, which represents the longest limestone cave system in the Russian Federation. The clastic cave sediments filling the cave passages differ in both mineral and mineral magnetic properties and were deposited under different hydrological conditions. The older portion of the clastic cave fills was derived from overlying sandstones, whereas the properties of younger cave sediments show closer affinity to the soils and weathering products originating on the sandstone plateau above the cave. The cave sediments underwent repeated periods of deposition and erosion during the Tertiary and Pleistocene.1. IntroductionThe aim of this project is an examination of the Botovskaya Cave deposits using methods operating with magnetic and heavy minerals and with quartz grain exoscopy. Obtained mineral characteristics were used for correlations from the point of view of sediment source and mode of transportation into the cave passages. Radiometric and paleomagnetic datings of the cave carbonate bed allowed us to estimate the age of both depositional and post-depositional processes.2. Geography and geologyThe Botovskaya Cave is located on the Angarsko-Lensky Plateau of the southern Siberian Craton about 500 km north of Irkutsk City. The area reaches altitudes of 1,100 m a.s.l., and belongs to the Zhigalovo District of the Irkutsk Area. The plateau is dissected by river valleys up to 400 m deep. Cave entrances lie at a relative elevation of 310 m above the Lena River level, in a valley of the Garevogo Creek, the left tributary of the Boty River, which joins the Lena River. The cave system, dipping gently to the north, has developed in an Early Ordovician limestone formation with a thickness of 6 to 12 m. The limestone bed is underlain by Middle and Late Cambrian sandstone, siltstone, marl and gypsum and overlain by Middle Ordovician sandstone, limestone and argillite. The cave system developed under confined (artesian karst) settings (Klimchouk 2000, 2003). The speleogenesis of the Botovskaya Cave system was interpreted by Filippov (2000) and is due to two different processes: (i) corrosion involving meteoric water, and (ii) ascending deep circulating artesian water (below groundwater level) spanning the time period between Late Mesozoic and Early Neogene.3. MethodsThe studied sections of the cave deposits were documented with special reference to lithology, sedimentary structures and aggradation and erosion event records. Mineral magnetic characteristics such as low field bulk magnetic susceptibility (MS) and anhysteretic remanent magnetization (ARM) together with anisotropy of magnetic susceptibility (AMS) help to find the source of the cave fills and estimate a mode of sediment transport to the cave passages. The character of quartz grain surfaces indicates transportation and post-depositional history of clastic sediments. Exoscopic observations were performed on quartz grains larger than 0.25 mm separated from either clastic cave deposits or from the Ordovician sandstone bedrock after wet sieving and boiling in HCl. Heavy minerals were separated after wet sieving from the grain-size fraction of 0.25.063 mm using tetrabromethane and observed in Canadian balsam. At least 300 grains of transparent heavy minerals were determined in each sample. The flowstone bed used for the paleomagnetic polarity measurements was also dated by the 230Th/234U radiometric method.4. Results4.1. Magnetic fabric and mineral magnetic characteristics The MS values slightly increase from bottom sand sediment beds to the above lying clay dominating sediments (see Fig. 2). The highest MS values were measured in the modern topsoil collected above the Medeo Entrance. The ARM values show steep increase in the top beds. The AMS parameter is sensitive to presence of very fine superparamagnetic minerals originating during pedogenesis or forest fires. The magnetic fabric of the sediments is mostly oblate. The magnetic lineation directions in the top sedimentary beds show concentration in NW to SW directions with the mean direction tending to the WSW. The poles to magnetic foliation are usually concentrated around the center of the projection. The magnetic lineation directions measured in the bottom sedimentary beds show almost random distribution accompanied by large dispersion of poles to magnetic foliation. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings96

PAGE 99

4.2. Quartz surfaces and heavy mineral content Three types of microstructures were observed on quartz grain surfaces: (i) precipitation of SiO2on the surface of grains in lace-like patterns; (ii) corrosive etched microstructures, and (iii) overgrowth with quartz crystals. The bedrock Ordovician sandstone forming the ceiling of the cave passage above Section 2 contains rounded quartz grains about quartz grains about 1 mm in diameter. The grain surfaces show weak dissolution and precipitation of SiO2. The studied samples of the bedrock and cave deposits show monotonous association of stable heavy minerals. Grains of transparent heavy minerals are rounded, while the opaque mineral grains are mostly angular. Beds in the upper portions contain greater amounts of garnet, similar to modern soil. 4.3. Flowstone datings Radiometric dating of the flowstone bed by 230Th/234U method reveals that the age of three dated samples exceeds 350 ka (see also Breitenbach et al. 2005). The flowstone age is probably higher than 1.2 Ma based on 234U/238U ratio. Based on the paleomagnetic data, it is evident that the flowstone records normal polarity of the Earth magnetic field from the time of the carbonate deposition.5. DiscussionThe bottom beds of the cave sequences reveal similar lithological, magnetic and heavy metal properties. Basal sand deposited on the bedrock bottom shows a lower MS, similar to the bedrock sandstone exposed in the cave. The exoscopic quartz grain characteristics are similar, too. The deposits filling the bottom part of cave corridors were derived from the local bedrock formed by Early Ordovician sandstones. The younger cross-bedded and laminated, mostly clayey sands (beds No. 1 in Fig. 2) were deposited after partial erosion of the basal deposits. The deposits reveal slightly higher MS values. The exoscopic observation shows cemented quartz grain aggregates indicating the source in the local cross-bedded sandstone. The inner structures of these laminated sediments show a frequent alternation between local aggradation and erosion events. The sediments were deposited during heavy precipitation events, when water penetrated into the cave corridors from the surface through the swallow holes and along open vertical cracks. The water escape structuresFigure 1. Botovskaya Cave (The Old World) map (adopted from Gbel and Breitenbach 2003) with indication of studied sections and dated flowstone. Figure 2. Section 1 SW face 1 clay, sporadic angular clasts of sandstone; 2 sand; 3 clay; 4 sandy clay to clayey sand; 5 clay; 6 sandy clay; 7 slightly sandy clay; 8 sand; 9 sand, fine-grained; 10 clayey sand, laminated; 11 clayey sand; 12 sand; 13 clayey sand; 14 sand with sandy aggregates; 15 sand; 16 carbonate cementation along fissures; 17 bedrock wall; black squares with numbers collected samples. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings97

PAGE 100

were formed during the compaction of sediments. The combination of radiometric and paleomagnetic polarity datings suggests a likely age for the carbonate bed before 2.588 Ma (e.g., the boundary Matuyama and Gauss paleomagnetic chrons). In the time of the flowstone deposition the artesian aquifer regime was already disrupted due to surrounding valley incision and vadose conditions dominated in the cave system. The later intensive erosion removed clastic sediments from under the flowstone bed and destroyed the entire flowstone. This erosional event had to be triggered by unusually heavy precipitation, which entered the bedrock through vertical ruptures opened both in sandstone and limestone due to differential gravityinduced subsidence of the massif during the Pleistocene.6. Conclusions1. The sections in detrital cave sediments in Botovskaya Cave (in The Old World part) evidence periodical sediment deposition. Sediments of the cave fill are of two different types: the older, bottom sands are derived from weathered bedrock sandstones and were probably horizontally transported over a short distance probably not exceeding first hundreds of meters. The overlying sediments dominated by clay and clay/sand were transported vertically with precipitation waters from the surface above the cave. 2. If the bottom sand was transported horizontally through the cave by flowing water, it must have taken place before the incision of the present deep valleys, probably in the Late TertiaryAcknowledgementsThe project was supported by the INTAS Program No. 03-51-4152: Speleothems and other cave sediments from Siberia: An archive from the boreal climate zone with the potential for climate reconstruction on an annual to decadal basis (SPELEOARCH) and it is a part the research project of the Institute of Geology AS CR, v.v.i. No. AV0Z30130516.ReferencesBreitenbach S, Fernandez D, Adkins J. Mingram B, Oberhnsli H, Haug G 2005. Speleothem records older than 500 ka from Southern Siberia. Kalamos, Hellas, Hellenic Speleological Society, 614. Filippov AG, 2000. Speleogenesis of Botovskaya Cave, Eastern Siberia, Russia. In: A.B. Klimchouk, D.C. Ford, A.N. Palmer and W. Dreybrodt (Eds.). Speleogenesis, Evolution of Karst Aquifers. Huntsville, Ala., National Speleological Society, 282. Gbel E, Breitenbach S, 2003. Die Peschera Botovskaja, die lngste Hhle Russlands. Mitteilungen des Verbandes der Deutchen Hhlen und Karstforscher, 49(2), 42. Klimchouk A 2000. Speleogenesis under deep-seated and confined settings. In: A.B. Klimchouk, D.C. Ford, A.N. Palmer and W. Dreybrodt (Eds.). Speleogenesis, Evolution of Karst Aquifers. Huntsville, Ala., National Speleological Society, 244. Klimchouk A, 2003. Conceptualisation of speleogenesis in multistorey artesian systems: a model of transverse speleogenesis, http://www.speleogenesis.info/archive/publication.php?PubI D524&Type5publication&keyword5Klimchouk Figure 3. View on the AngaraLena Plateau formed by Lower Paleozoic sedimentary sequences. A white strip marked by arrow represents the limestone bed intercalated between sand stones probably containing undiscovered cave systems.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings98

PAGE 101

CASE STUDIES OF FLUORESCENT GROUNDWATER TRACING INRECENT CAVE RESEARCHBenjamin V. Miller1, Chris Groves1, Jason S. Polk1, Robert N. Lerch2 1Western Kentucky University, Department of Geography and Geology, 1906 College Heights Blvd., Bowling Green, KY42101, United States, benjamin.miller@wku.edu2USDA-Agricultural Research Service, Cropping Systems & Water Quality Research Unit, Agricultural Engineering Building, University of Missouri, Columbia, MO 65211, United States Groundwater studies have utilized fluorescent dyes in order to map groundwater flow paths and confirm hydrologic connections amongst karst features. These techniques are now well established so that common dyes and common monitoring techniques can be used to apply these methods to a wide range of applications and in a variety of locations. This paper examines four case studies where groundwater tracing with fluorescent dyes to better understanding the hydrologic characterization of cave and karst systems. A five-year project delineated recharge areas for two significant cave systems; Onondaga Cave and Cathedral Cave; which are publicly owned and manage cave systems. Dye tracing was used to delineate recharge areas for streams located in Carroll Cave, a hydrologically complex and biologically significant cave system in central Missouri. P-Bar Cave, located in the Bighorn Mountains of Wyoming, is a large stream swallet draining approximately 104 km2into the main cave entrance. Groundwater tracing techniques were applied at P-Bar Cave to determine resurgences for the stream at springs located 9.6 kilometers down canyon. Crumps Cave, a large cave located on the Pennyroyal Sinkhole Plain of south-central Kentucky is a site where epikarstic research on contaminant transport is taking place. Tracing with fluorescent dyes has examined the extreme variances in travel times between flow from the surface into the epikarstic aquifer and travel from in-cave perennial waterfalls to the resurgence for the cave stream 35 kilometers away. Groundwater tracing continues to be a useful tool in understanding the complex flow systems found in karst areas and provides useful information for both researchers and agency land managers. 1. Common groundwater tracing methodsFluorescent dyes have been used in groundwater tracing to delineate groundwater basins and to compliment hydrologic research in cave and karst regions. In conducting groundwater tracing in a karst area one of the initial tasks is to identify all recharge and discharge features during a Karst Hydrologic Inventory (KHI). This KHI helps to identify potential monitoring locations, at features draining the aquifer or in local surface streams, and potential dye injection locations, at features recharging the aquifer (Crawford Hydrology Lab 2010). Once discharge features are identified, a monitoring method is chosen depending on given objectives for the trace. Charcoal receptors are the most common method for monitoring groundwater features and consist of a mesh packet containing 3 grams of crushed coconut charcoal (Crawford Hydrology Lab, 2010). These are attached to wire or nylon string and are placed directly in the flow of discharge features. Regular changing of these receptors at given time intervals aids in determining travel from injection location to recovery site. Grab samples can also be collected in glass vials for analysis of current concentrations of dye. Autosamplers can be used for the collection of these grab samples at specific time intervals and when combined with discharge data quantitative tracing can be used to examine the amount of dye recovered versus the amount of dye injected. Technological advances have also made portable field fluorometers, which can analyze concentrations of specific dyes and record this data to a memory device for later download (Meus et al. 2006). Tracers used in groundwater tracing commonly use five common dyes; fluorescein, eosine oj, rhodamine wt, sulphorhodamine b, and tinopal ob (Aley and Fletcher 1976; Jones 1984). Picking a specific dye for trace must be determined by background levels in a study area and the type of features tracers are injected into. Injection methods also vary depending on the hydrologic conditions present at a given location. Dry streambeds or sinkholes may require the use of a dry-set where powdered dye is placed inside of a PVC pipe and then staked to the base of a stream channel to await injection of the tracer following a storm event. If road access is available fire tanker trucks may also be used to create a temporary stream into which dye can be injected. Finally if flowing streams are located at an injection site then the dye can be directly poured into the stream for an injection. However, in all cases extreme care must be taken to prevent any contamination of dyes amongst injection locations and specifically amongst monitoring sites. Several examples of case studies from groundwater traces in various locations in the continental United States (Fig. 1) are presented below. Each of these examples utilized groundwater tracing to achieve different objectives and used different methods.2. Case Studies of Groundwater Tracing in Cave Studies2.1. Onondaga Cave and Cathedral Cave, Missouri, USA Onondaga Cave and Cathedral Cave are two large cave systems located in central Missouri in the Ozarks ecoregion. Both caves are preserved by the Missouri Division of State Parks and are the centerpieces of Onondaga Cave State Park Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings99

PAGE 102

2.2. Carroll Cave, Missouri, USA Carroll Cave is an extensive stream cave system, currently 28 km in length and still being explored. The cave is developed within the Ordovician Gasconade Dolomite and located in the northwest portion of the Wet Glaize Creek watershed. The cave is known to have dense populations of the Southern Cavefish ( Typhlicthys subterraneus) and the Grotto Salamander ( Eurycea spelaea ) both stygobitic (OCSP). The caves share many common aspects, including a high biodiversity of largely aquatic dominated biota, a large quantity of speleothems, and the fact that perennial streams flow through the caves, which drain losing streams in the surrounding area. A project was initiated in 2003 to begin delineation of recharge areas for both caves in order to guide cave management, future land acquisition, and to determine possible threats from land use. A total of 15 monitoring sites were established in surface and subsurface streams, as well as other groundwater discharge features, utilizing charcoal packets as the primary monitoring method. Over the next five years a total of 21 dye injections resulted in six positive traces to Onondaga Cave and three positive trace to Cathedral Cave (Fig. 2). This information was then used to delineate recharge areas using GIS (Fig. 3), resulting in a 24 km2recharge are for Onondaga Cave and a 2.9 km2recharge area for Cathedral Cave (Miller and Lerch 2011). The majority of the recharge areas for both caves are located outside of the park, however current land use in the recharge areas is dominated by deciduous forest. Information from the traces has been applied to the Parks management in the form of a cave management plan and now gives agency land managers a target area for possible threats to water quality in the cave streams. Figure 1. Map of USA showing location of case studies discussed in paper. Figure 2. Map of positive groundwater traces conducted in the Onondaga Cave area.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings100

PAGE 103

species of special concern within the state of Missouri. The hydrologic behavior of the cave is controlled by three independent cave streams; Carroll River, Thunder River, and Confusion Creek/New River. Thunder River, the largest of the cave streams, has an average daily flow of 150 L/s and contains multiple tributaries along the 11 km course of the cave stream. Dye tracing confirmed the resurgence for two of the cave streams is Toronto Springs, a multiple outlet alluviated springs system located 4 kilometers north of the cave (Figure 4). The dye tracing project which was took place from 2008 delineated a recharge area of 18.4 km2for Carroll Cave (Fig. 5) and identified a total of eight resurgences for two of the cave streams; Thunder River and Confusion Creek; at Toronto Springs. Figure 3. Recharge areas for Onondaga Cave and Cathedral Cave.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings101

PAGE 104

Additional traces in area losing streams indicated a mixing zone (of surface and subsurface streams) which recharge the 13 springs at Toronto Springs (Miller 2010). Use of the recharge area has allowed for land use analysis (Fig. 6) and indicates a land use dominated by deciduous forest and grasslands used as pasture for cattle. 2.3. P Bar Cave, Wyoming, USA P Bar Cave is located in the western Bighorn Mountains in north-central Wyoming. The cave is co-owned and managed by the US Forest Service and the Bureau of Land Management. Formed in Ordovician Bighorn Dolomite, the cave is a large swallet located near the contact with igneous basement rocks, and is recharged by Wet Medicine Lodge Creek, which drains a 104 km2area consisting largely of granitic uplands. The cave was mapped in the 1970s to a total of 2.9 kilometers in length and additional hydrologic work identified a series of springs 9.6 kilometers downcanyon, which were speculated to be the resurgences for the cave system based on micaceous sands found downstream of the spring outlets (Huntoon 1985). A project to remap the cave was started in 2009 and in 2012 this work was combined with some groundwater tracing to confirm which springs and how many springs were hydrologically connected to the cave system. A KHI identified eight springs in lower Wet Medicine Lodge Canyon (Fig. 7) that were then monitored using charcoal receptors, two springs were also located immediately downstream of the main cave entrances which sank into the streambed after a few hundred meters. Rhodamine WT was injected into Wet Medicine Lodge immediately downstream of the upper springs yet above the stream sink and fluorescein was injected into the main cave entrance to see if the water flowing into the cave resurged at the same springs as the water sinking below the upper springs. The water which enters the main entrance was traced to the second spring, located immediately downstream of the cave entrances, but bypassed the uppermost spring. This indicates that the lower spring is fed by groundwater sources independent from the Figure 5. Recharge area for Carroll Cave showing positive groundwater traces. Figure 4. Map of Toronto Springs, showing springs hydrologically connected to Carroll Cave.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings102

PAGE 105

cave waterfalls to the final resurgences at Graham Springs located over 35 kilometers to the west. Variances in travel times of the epikarstic traces largely depend on the hydrologic conditions of the epikarstic aquifer influenced primarily by rain events and the storage capacity of the aquifer. Traces through the epikarst to the in-cave waterfall, a distance of less than 150 meters, may take as less than a few hours if aided by rainfall simulator or as much as several weeks if conducted during a drier period. Traces from the perennial waterfalls inside the cave to Graham Springs also vary, but traces conducted in base flow conditions took approximately six days to travel 35 kilometers. These traces highlight the variability in travel main swallet. The water sinking downstream of the upper springs was positively traced to all eight of the springs located 9.6 kilometers down canyon. Travel time from injection site to resurgences took less than seven days and flow velocities in the area at least 56 m/hr but likely exceed this rate. This information was reported to agency personnel and will be incorporated into the management of the water resources of this area, especially since Wet Medicine Lodge Creek is utilized for irrigation a short distance downstream of the resurgences. 2.4. Crumps Cave, Kentucky, USA Crumps Cave is a two-kilometer-long cave located in southcentral Kentucky on the Pennyroyal Sinkhole Plain. The cave is formed in the Carboniferous St. Louis Limestone and is located in the headwaters of the Graham Springs recharge area, a 360 km2basin recharging one of the largest springs in Kentucky (Hess et al. 1989). The cave is owned and managed for karst research and education by the Hoffman Environmental Research Institute at Western Kentucky University. Hydrologic research examining the characteristics of the epikarstic aquifer has used groundwater tracing to compliment geochemical analyses to aid in determining flow rates from the surface environment to monitored perennial waterfalls located throughout the cave. These injections have been conducted using augured holes into the soil column (Fig. 8) as well as artificial rainfall simulators. Additionally tracer tests have been conducted from the inFigure 6. Land Use Land Cover in the Carroll Cave recharge area. Figure 7. Conducting a Karst hydrologic Inventory in Lower Wet Medicine Lodge Canyon, Wyoming.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings103

PAGE 106

times between the epikarstic aquifer and the conduitflow aquifers feeding the large springs of the sinkhole plain. Traces continue to examine these variances and aid in understanding contaminant transport in a karst area where land use is dominated by row crop agriculture.3. ConclusionsGroundwater tracing is a useful tool in understanding the hydrology of a cave system or karst area, which can be applied in a variety of different situations. By identifying recharge and discharge features in a karst area one can determine good monitoring locations and good dye injection locations. Monitoring for the traces typically use charcoal receptors, however other options exist which may work better depending on ones objectives and budgets. Case studies highlighted in this paper used groundwater tracing to delineate groundwater recharge areas for significant caves, determine resurgences for cave streams, and to compliment hydrologic research on contaminant transport in karst systems. Many other applications of groundwater tracing exist in cave studies and many karst areas still contain a large amount of work which can be done to better understand the ongoing hydrologic processes.ReferencesAley T, Fletcher MW, 1976. Water tracers cookbook. Missouri Speleology, 3(16), 1. Crawford Hydrology Laboratory, 2010., Karst Groundwater Investigation Procedures. Bowling Green, Kentucky. www.dyetracing.com Hess JW, Wells SG, Quinlan JF, White WB, 1989. Hydrogeology of the south-central Kentucky karst. In: White, W.B. and White, E.L. (Eds.). Karst Hydrology. Concepts from the Mammoth Cave Area Van Nostrand Reinhold New York. P. 15. Huntoon PW, 1985. Gradient Controlled Caves, Trapper Medicine Lodge Area, Bighorn Basin, Wyoming. Groundwater, 23(4), 443. Jones WK, 1984. Dye tracer tests in karst areas. The NSS Bulletin, Vol. 46, 3. Miller B, Lerch R, 2011. Delineating recharge areas for Onondaga and Cathedral Caves using groundwater tracing techniques. Missouri Speleology, 51(2), 1. Miller BV, 2010. The Hydrology of the Carroll Cave-Toronto Springs System: Identifying and Examining Source Mixing through Dye Tracing, Geochemical Monitoring, Seepage Runs, and Statistical Methods. Masters Thesis, Western Kentucky Univ. Bowling Green, Kentucky. Meus P, Kss W, Schnegg PA, 2006. Background and detection of fluorescent tracers in karst groundwater. Karst, climate change and groundwater. Hidrogeologia y Aguas Subterrneas, 18, 65. Figure 8. Injecting fluorescein dye into drilled hole in the soil column.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings104

PAGE 107

THE JAJ PLATEAU (LEBANON): TYPICAL HIGH ALTITUDE MEDITERRANEAN KARSTFadi H. Nader, Hughes Badaoui, Marc Metni, Chadi Chaker, Habib Helou, Johnny Tawk Splo-Club du Liban, P.O. Box: 70-923 Antelias, Lebanon, fadi.nader@gmail.com The high Plateau of Jaj in Mount Lebanon displays an excellent open-air museum for a typical Mediterranean Karst landscape (Fig. 1). It exposes the Jurassic carbonate rocks (limestone and hydrothermal dolomites), locally called the Kesrouane Formation over a relief ranging from about 1,300 to 1,955 m above sea level (Qornet el Alieh, being the highest point). Structurally speaking, this area is part of the Qartaba folded (asymmetrical) structure with relatively gentle dipping towards the west and steep limbs towards the east. The Jurassic rocks were covered by volcanic deposits that were eroded later on upon the emergence of Mount Lebanon. This could explain not finding deep caves most of them are filled by volcanic weathered material. Such a high elevated Jurassic landscape, with well evolved karstified weathered surface cannot be overlooked by speleologists and cave explorers! Even if, the plateau is bounded by major karst features like the Balaa sinkholes (three potholes with average depth of 200 m), undertaking field search for caves at altitudes averaging 1,700 m, with no water sources or enough shades (except the few majestic cedars trees) and on rough terrain has proven to be less attractive (Fig. 2). Decades ago, cavers from the Splo-Club du Liban have discovered and documented around 20 caves on the plateau, mostly pitches; some of which hosted ice that was thought to remain from year to year. Recent systematic exploration has started again and was planned together with the 3rdMiddle East Speleology Symposium (2011). This presentation will expose some geomorphological peculiarities of the Jaj Plateau, and the exploration work that is undergoing with its first results. It will also highlight the need to protect this special karstic area.Figure 1. Simplified topographic map of Lebanon indicating the location of the Jaj Plateau (square). The index map shows the lo cation of Lebanon in the Eastern Mediterranean region. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings105

PAGE 108

Figure 2. View from a cave on the Jaj Plateau showing the bare-karst landscape. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings106

PAGE 109

A CONCEPTUAL MODEL OF SPELEOGENESIS IN GREECEChristos Pennos1Stein-Erik Lauritzen2 1School of Geology, Department of Physical Geography, Aristotle University of Thessaloniki, GR-54124, Greece, pennos@geo.auth.gr2Department of Geology, University of Bergen, N-5007 Bergen, Norway, Stein.Lauritzen@geo.uib.no Over 70 % of Greeces territory consists of carbonate rocks (limestone and marble). This almost homogenous geology in correlation with the special tectonic regime of the broader region of the east Mediterranean makes the region an ideal site for cave development. The exact number of Greek caves is still unknown since there is no formal cadastral. However, there is an estimation of more than 10,000 caves and rockshelters. Here, we attempt to make a correlation between the cave dimensions and their locality. First, a creation of a cadastral took place using data provided by cavers and caving clubs. Statistical analyses of depth, length and altitude of the caves were performed in order to define the key factor for the development of the caves and to create a general conceptual model about the speleogenesis in Greece. The combination of the above revealed a clear difference on the speleogenesis between the caves in the outer and the inner part of the Aegean orogeny.1. IntroductionThe lithological structure of Greece consists mainly of carbonate rocks in combination with the Mediterranean climate, favors the development of karst phenomena to a large extent. A rough estimation of the total number of karst caves in Greece includes more than 10,000 caves. During the last decades the development of cave exploration in Greece along with the increasing interest of many European cavers has led to numerous caving expeditions. However, little attention has been paid to the study of caves from a scientific point of view. In this work we attempt for the first time to create a general conceptual speleogenetic model for the Greek caves based on cave statistics of their characteristics, such as depth, width and length. A correlation between cave locality and the tectonic regime is tested.2. Geological setting.The Greek territory consists mainly by carbonate rocks (marbles and limestones). In general, northern Greece consists of Paleozoic marbles and metamorphic rocks since it is part of the old continental crust around which the closing of the Tethys Sea took place during the Alpine orogenesis. In contrast, the rest of the Greek mainland as well as most of the Greek islands are mainly built up of Mesozoic limestones that represent the shallow and deep marine deposits of the Tethys Sea. The general structure is characterized by a series of stacked nappes, ~ 5 km thick (composite), consisting of the upper crust that decoupled from the present subducted continental and oceanic lithosphere of the AdriaticAfrican Plate (van Hinsbergen et al.; 2005, 2010; Jolivet and Brun 2010). These nappes (or mega-units) were thrusted and stacked in a north-to-south direction since the Cretaceous (Faccenna et al. 2003; van Hinsbergen et al. 2005; Jolivet and Brun 2010) and form a strongly shortened representation of the paleogeographical distribution of continental ribbons and deep basins that existed in the western Neo-Tethys (e.g., Dercourt et al. 1986; Barrier and Vrielynck 2008; Stampfli and Hochard 2009). The Aegean Sea and the surrounding areas belong to the active continental boundary of the AlpineHimalayan belt (Fig. 1) and as a result suffer a large-scale active deformation, stemming from the subduction of the eastern Mediterranean lithosphere under the Aegean Sea, along the Hellenic Arc (Papazachos and Comninakis 1969). Consequently, the continental and coastal parts of Greece share the common characteristics of the backarc extensional tectonics, expressed by the presence of a strong deformational pattern, volcanic activity and the development of fault bounded grabens, lying in accordance with the dominant NS extensional stress field. However, as illustrated in Figure 1, the northern part of Greece is additionally influenced by a subsidiary rightlateral shear because of the coexistence with the North Aegean Trough, a dextral strike-slip structure associated with a series of strong earthquakes (McKenzie 1972). Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings107 3. MethodologyIn order to investigate the key factors that influenced the development of Greek caves and to set up a general model concerning the speleogenesis in Greece, a proper cadastral was created. Parameters concerning cave dimensions such as depth, length and altitude, as well as the locality of almost 4,000 Greek caves are included. The cadastral is based on an enriched version of the SPELEO club cadastral (Theodosiadis 2011), while further data provided from theFigure 1. Geotectonic Regime of the Aegean (from van Hinsbergen and Schmid 2012).

PAGE 110

In contrast as it seen in the diagram between the depth and the length of caves (Fig. 3) there is appositive correlation between these two parameters. This is because the measured length of the cave is a function of the surveyed depth, deep caves tend to be long and vice versa Furthermore, the diagram created to investigate the relationship between the depth and the altitude at which the caves appear (Fig. 4) shows that the majority of the deeper caves occur at an altitude over 400 m. There is also an upper boundary of cave altitude, controlled by the topography. The lower boundary (slope 0.91) is connected to the internal relief; deep caves are found in terrain of great relief. In order to further investigate the factors that influenced Greek cave development a map showing the positions of the studied caves and the tectonic setting of Greece with the main tectonic lines was created (Fig. 5). In this map, it becomes evident that the deepest caves tend to be located in areas adjacent to the subduction arc. This is explained by the fact that at these areas compressive stress results in constant uplift and in intense (tensional) fracturing of the bedrock at high altitudes. The continuous raising of these Greek cavers K. Adamopoulos (SELAS caving club) and G. Sotiriadis (Proteas caving club). A statistical analysis of the relation between various speleometric parameters was conducted and explored through correlation plots. Moreover, ArcGIS software was used to create a thematic map with the position of the forty deepest caves in order to clarify the relation of the present tectonic regime and cave development.4. Results and discussionStatistical analysis revealed that there is no clear correlation between the altitude and the length of the studied caves (Fig. 2). This is probably due to the different geological and hydrological criteria that prevailing in the study area (Greece) and they are controlled by the regional geology, tectonic activity and the landscape. regions (and erosion) lead to continuous lowering of the phreatic zone and, as a consequence, to the creation of a deep vadose zone. This fact, in combination with the strong fracturing favors the creation of deep caves.5. ConclusionsThe exploratory statistical analysis of the cave parameters in relation to the spatial distribution of the deep caves suggests that the major factor controlling the speleogenesis in the area is the tectonic activity of the region. The tectonic setting of the broader Greek territory results in the fracturing of the carbonate bedrock in accentuated topographic relief, creating ideal conditions for vadose speleogenesis (most deep caves of the World are essentially vadose). The high tectonic activity is responsible for the continuous lowering of the phreatic zone creating a high vadose zone in which the deep caves occurred.AcknowledgmentsThe authors would like to thank the cavers Thomas Theodosiadis (SPELEO club), George Sotiriadis (Proteas club) for providing data from their personal archives and Kostas Adamopoulos (SELAS club) for providing us numerous unpublished data from his archive as well as for the discussions on the topic. Special thanks to Dr. Sofia Pechlivanidou and Charikleia Garlaouni for their valuable conversations during the preparation of the manuscript. Finally, the authors would like to express their gratitude to Pavel Bosak and Michal Filippi for their remarks and recommendations. Figure 2.Altitude length plot. Figure 3.Length depth plot. Figure 4.Altitude depth plot. Figure 5.Map of the region with the main tectonic lines. The positions of the forty deepest Greek caves are depicted with dots.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings108

PAGE 111

ReferencesAdamopoulos K, 2005. The deepest and the longest caves in Greece, Proceedings of 14thInternational Congress of Speleology, 21 August, 2005, Athens, Kalamos, Greece. Adamopoulos K, 2011. Lefka Ori, The biggest Greek caves in Crete island (in Greek). Proceedings of the 14thPanhellenic speleological meeting, Chania, Crete. Barrier E, Vrielynck B, 2008. MEBE Atlas of Paleotectonic maps of the Middle East, Commission for the Geological Map of the World. Dercourt J, Zonenshain LP, Ricou L-E, Kazmin VG, Le Pichon X, Knipper AL, Grandjacquet C, Sbortshikov IM, Geyssant J, Lepvrier C, Pechersky DH, Boulin J, Sibuet J-C, Savostin LA, Sorokhtin O, Westphal M, Bazhenov ML, Lauer JP, Biju-Duval B, 1986. Geological evolution of the tethys belt from the atlantic to the pamirs since the LIAS, Tectonophysics, Volume 123, Issues 1, Pages 241, ISSN 0040-1951, 10.1016/00401951(86)90199-X. Faccenna C, Jolivet L, PiromalloC, Morelli A, 2003. Subduction and the depth of convection of the Mediterranean mantle, Journal of Geophysical Research, 108(B2), 2099. Jolivet L, Brun J-P, 2010. Cenozoic geodynamic evolution of the Aegean, International Journal of Earth Sciences, 99, 109. McKenzie D, 1972. Active Tectonics of the Mediterranean Region. Geophysical Journal of the Royal Astronomical Society, 30: 109. doi: 10.1111/j.1365-246X.1972.tb02351.x. Papazachos B, Comninakis P, 1970. Geophysical features of the Greek island arc and eastern Mediterranean ridge, Com. Ren. Des Sceances de la Conference Reunie a Madrid, 1969, Vol. 16, 74. Stampfli G, Hochard MC, 2009. Plate tectonics of the Alpine realm, in Ancient orogens and modern analogues. Geological Society, London, Special Publications, edited by J. B. Murphy, et al., 8911. Theodosiadis T, 2011. SPELEO cave cadastral, SPELEO club, Athens, Greece. http://arxeio.speleo.gr van Hinsbergen, DJJ, Hafkenscheid E, Spakman W, Meulenkamp JE, Wortel MJR, 2005. Nappe stacking resulting from subduction of oceanic and continental lithosphere below Greece, Geology, 33(4), 325. van Hinsbergen, DJJ, Kaymakci N, Spakman W, Torsvik TH, 2010. Reconciling the geological history of western Turkey with plate circuits and mantle tomography, Earth and Planetary Science Letters, 297, 674. van Hinsbergen, DJJ, Schmid SM, 2012. Map view restoration of AegeanWest Anatolian accretion and extension since the Eocene, Tectonics, 31, TC5005, doi:10.1029/2012TC003132.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings109

PAGE 112

COMPLEX EPIKARST HYDROLOGEOLOGY AND CONTAMINANT TRANSPORT IN A SOUTH-CENTRAL KENTUCKY KARST LANDSCAPEJason S. Polk1, Sean Vanderhoff1, Chris Groves1, Benjamin Miller1, and Carl Bolster2 1Hoffman Environmental Research Institute, Western Kentucky University, Bowling Green, KY 42101 United States, jason.polk@wku.edu2USDA-Agricultural Research Service, Bowling Green, KY, 42101 United States The movement of autogenic recharge through the shallow epikarstic zone in soil-mantled karst aquifers is important in understanding recharge areas and rates, storage, and contaminant transport processes. The groundwater in agricultural karst areas, such as Kentuckys Pennyroyal Plateau, which is characterized by shallow epikarst and deeper conduits flow, is susceptible to contamination from organic soil amendments and pesticides. To understand the storage and flow of autogenic recharge and its effects on contaminant transport on water flowing to a single epikarst drain in Crumps Cave on Kentuckys Mississippian Plateau, we employed several techniques to characterize the nature and hydrogeology of the system. During 2010, water samples and geochemical data were collected every four hours before, during, and between storm events from a waterfall in Crumps Cave to track the transport and residence time of epikarst water and organic soil amendments during variable flow conditions. Geochemical data consisting of pH, specific conductivity, temperature, and discharge were collected continuously at 10-minute intervals, along with rainfall amounts. In addition, stable isotope data from rainfall, soil water, and epikarst water were collected weekly and during storm events to examine storage and recharge behavior of the system. The changes in geochemistry indicate simultaneous storage and transport of meteoric water through epikarst pathways into the cave, with rapid transport of bacteria occurring through the conduits that bypass storage. The isotopic data indicate that recharge is rapidly homogenized in the epikarst, with storage varying throughout the year based on meteorological conditions. Results indicate current best management practices in agricultural karst areas need to be revisited to incorporate areas that do not have surface runoff, but where contaminants are transported by seepage into local aquifers.1. IntroductionKentuckys subtropical climate and fertile soil provide extensive agricultural lands for row crops. A common agricultural practice in the area is to apply animal waste as an organic soil amendment for soil nutrient enhancement. If these amendments are not completely exhausted through crop utilization, they can become pollutants and enter the groundwater system. In Kentucky, 55% of the land area is characterized by highly soluble carbonate rocks within which karst landscapes form (Currens 2002). The resulting karst landscape/aquifer systems, typically with high permeability, are characterized by the development of features such as sinkholes, caves, and large springs. Because much of the recharge entering these systems moves rapidly under turbulent flow, and in many cases as sinking streams with little physical filtration, groundwater in these karst aquifers is often highly susceptible to contamination from agricultural practices, among other sources of pollution (White 1988; Drew and Holtzl 1999). This research was designed to better understand the fate and transport of agricultural contaminants in the well-developed karst aquifer/landscape systems of south-central Kentucky by conducting field experiments associated with actual field-scale agriculture at the Crumps Cave Educational Preserve, and aimed to answer the following research questions: (1) if aquifer recharge influences manure transport at this representative site, is there significant retardation of flow and storage of water and/or fecal bacteria in the soil/epikarst zone before it enters the main part of the aquifer?; and, if so 2) what is the timing of flow through this shallow part of the epikarst flow system?; 3) how does that effect the introduction of fecal bacteria into the main part of the aquifer?; and 4) where is the primary storage for contaminants and bacteria in the soil-epikarst setting?2. Study AreaCrumps Cave is located beneath a portion of the extensive sinkhole plain of the Pennyroyal Plateau within the Mississippian Plateaus Section of the Interior Low Plateaus Physiographic Province in Waren County, KY, USA (Figure 1) (Groves et al. 2005). There is about two km of horizontal cave passages beneath several agricultural fields, with the cave floor averaging 25 m below the surface. The recharge area lays within the Graham Springs groundwater basin (Ray and Currens 1998, 2000) which discharges at Wilkins Bluehole on the Barren River, 18 km southwest. It is the second largest spring in Kentucky (Ray and Blair 2005). The site is underlain by Crider silt loam, Pembroke silt loam and Baxter gravelly silt loam soils (Soil Survey Staff NRCS 2011). These soils are moderately permeable, well-drained soils, reddish in color with chert fragments in their lower portions. The thickness of the soils varies throughout the study area. Auger hole tests show the thickness before encountering chert fragments ranges from 0.3 meters. The entrance to Crumps Cave is a collapse sinkhole that has partially collapsed. The cave passages have formed within the highest part of the Mississippian-aged St. Louis limestone, with a local dip of 1 to the west (Richards 1964). The bedded Lost River Chert lies between the ground surface and the cave below, and locally appears to operate as a leaky perching layer. Water tends to reach the Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings110

PAGE 113

3. MethodsThe movement of autogenic recharge through the shallow epikarstic zone in soil-mantled karst aquifers is important in understanding recharge areas and rates, storage, and contaminant transport processes. To understand the storage and flow of autogenic recharge flowing to a single epikarst drain in Crumps Cave on Kentuckys Mississippian Plateau, we employed several methods to characterize the nature of the system. 3.1. Storm Event Contaminant Sampling From 2010, on the surface at Crumps Cave, 110 meters from the sink entrance, a HOBO U-30tmweather station was used to collect weather data. A rain gauge tipping bucket collected rainfall amounts every ten minutes. The weather station also collected temperature, dew point, solar radiation, relative humidity, wind speed and direction, and soil moisture content with ten-minute resolution. Inside the cave, a 208 liter barrel with circular holes drilled into its side to measure discharge was placed under WF1 and a conical tarp directs virtually all flow of the epikarst drain into the barrel (Figure 3). A procedure based on Bernoullis law relates the WF1 discharge rate (L/s) to the water level (stage height) in the barrel (White 1988). The water level is measured by a pressure transducer inside a stilling well at ten-minute resolution. At WF1, two Campbell Scientific CR10x data loggers were used to collect geochemical and discharge data for the waterfall. Data Logger One (DL1) recorded data from one pH probe, one dual specific conductance and temperature probe, and a pressure transducer probe placed in the discharge barrel at WF1. Data Logger Two (DL2) recorded data from two pH probes and a duel specific conductivity probe. Both data loggers collected data every two minutes and recorded the average every ten minutes for temperature, specific conductance (SpC), and pH. Stage height from the pressure transducer was also recorded every ten minutes. During the farming season of late winter through spring, three fluorescent dye traces (sulphorhodamine B, fluorescein, and eosine, respectively) took place to track transport and residence time of water from storm events and epikarstic waters. The dyes were chosen for their spectrum wavelength so as to be able to recognize each individual dye as it came through WF1 from the surface. The traces were performed in a location on the edge of the property in an area that has previously been established as having a hydrological surface connection to WF1. ISCO 3700 portable water samplers were placed in the cave at WF1 to collect water samples to analyze for dye, bacteria, cations and anions. Samples were collected in 1000 mL polypropylene bottles every four hours during storm events occurring within the study period. During a portion of the winter and spring sampling period weekly samples of fecal coliform bacteria (FC) were taken. Samples were also collected weekly for the analysis of dye and collected within 24 hours of analysis time for total coliform, E. coli, cations and anions. cave at distinct locations, mainly at perennial or intermittent waterfalls emerging from the cave ceiling through fractures, draining the epikarstic zone to the east of the cave and flowing westward down the dip of the rock (Bolster et al. 2005). Six perennial in-cave waterfalls are located within the entrance area of the cave. These waterfalls are focused on the east side of the cave, but some flow from different parts of the ceiling. Waterfall One (WF1) is approximately 4.5 m tall and is located 40 m from the entrance. It is the closest waterfall to the entrance and has perennial flow (Figure 2). It is the focus of the monitoring and research described herein. The climate of Warren County is classified as a humid subtropical climate on the Kppen climate classification scale (Cfa). Its humid summers reach an average high temperature of 31 C and its mild to cool winters average a high of 7 C (NOAA 2011). The average annual total precipitation is around 1,294 millimeters. Of this, about 721 millimeters, or 56 percent, usually falls in April through October. May has the highest average rainfall with 136 millimeters (NOAA 2011). The growing season for most crops falls in the April through October range. Hess (1974) estimated that mean-annual potential evaporation is 800 mm, varying from near zero to over 100 mm/mo. Land use above and surrounding Crumps Cave is dominated by agriculture (Figure 1). Row cropping, which usually rotates between corn, soy, and wheat, surrounds the Crumps Cave property to the east and north. West of the property is a residential property at which a bed and breakfast operation is run. Northeast of the property land is currently being used for cattle grazing.Figure 1. Location of Crumps Cave, Warren County, KY, USA. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings111

PAGE 114

3.2. Epikarst Storage and Recharge First, we performed base flow separation of several discrete storms, where stormflow was integrated to measure individual storm volumes, and a nominal recharge area parameter, was calculated for each storm event by determining the measured amount of rainfall (thickness) and setting the drainage volume from the in-cave waterfall during the event to be equal to this recharge amount, and then dividing the resulting recharge volume by the presumably uniform measured rainfall depth. In this manner, we are taking the volume of discharge at the waterfall and dividing it by the amount of rainfall during the event (volume/thickness) to calculate an estimated area over which the rainfall presumably fell during the event, or in this case an average recharge area on the surface that Figure 2. Map of Crumps Cave in south-central Kentucky.contributes to the water flowing from the waterfall. A second independent proxy of recharge and epikarst storage conditions from isotopic analysis of precipitation and epikarst water supports this finding. Rainfall amounts above the cave and discharge of the water flowing from the drain below were measured every ten minutes in 2010. Weekly precipitation and cave waterfall samples were collected for stable isotope (O and H) analysis, with higher resolution sampling of the waterfall during storm events. A third measure of the character and response of the epikarst was the three dye-traces performed during the study period, which resulted in a breakthrough curve for each storm event that caused the dye to move through WF1, which provided an estimated threshold for epikarst storage to be flushed out and the travel time of surface water to WF1 Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings112

PAGE 115

4. Results and DiscussionIdentifying the hydrological characteristics of the soilepikarst zone in response to storm events is an important task for understanding the fate of agricultural contaminants. Recognizing the mechanisms of movement and storage is central for determining the fate of these pollutants. The movement of fecal coliform bacteria through the soilepikarst is dictated by the amount and intensity of storm events. During the study period, the majority of storm events provoked a response at WF1 indicated by the increased discharge. The data show contaminants that move through the epikarstic system correlate with significant rainfall events and rainfall amount. The dye traces and SpC data support this statement and further the understanding of soil-epikarst hydrology and contaminant transport. The various storm event data and dye traces combined with fecal coliform analysis indicate rapid response of the epikarst drain to precipitation events (Figures 5, 6). High levels of contaminants, including E. Coli were found during the peak of each storm event, with maximum numbers (greater than 200 colonies/100mL) being reached shortly after surface application, but lasting for many months. under various antecedent storage conditions (Figure 4). Figure 3. Discharge barrel and data logger setup at WF1 during storm event.For the epikarst storage data, values of range from 843 m2to 11,200 m2, with lower values likely resulting from water entering epikarst storage that does not reach the drain during that storm response. Values of may thus provide way to quantify varying epikarst storage input. Data also suggest that the actual recharge area is on the order of 100 100 m. Using between-storm baseflow discharges and individual values of for storm events, we calculated unit-baseflow (UBF baseflow discharge per unit recharge area, calculated from dividing total discharge during baseflow conditions by calculated area from ), and show that calculated values for the epikarst drain are 2 orders of magnitude higher than published UBF values for regional springs (Worthington 2007). This gives quantitative evidence that storage is more concentrated in the epikarst than the regional aquifer as a whole. Precipitation isotope 18O values ranged from -5.7 to -16, while the cave waterfall 18O values averaged -6.3 (ranging between -5.2 to -7.3), indicating that despite intense storm events with highly variable 18O values, a rapid homogenization of meteoric recharge water with epikarst storage water occurs in the system, further supporting that substantial storage occurs in the epikarst zone through diffuse flowpaths. From the seasonal data, there are two conditions that occur in the soil that dictate the transport of fecal coliform and E. coli. First, there is a threshold for rain intensity and rain amount that push the bacteria (Pasquarell and Boyer 1995) and dye through the soil-epikarst system (Figure 6). Additionally, diffuse flow through conduits adds to the movement of bacteria through the soil-epikarst system. Significant storm events infiltrate the soils and create a high hydraulic head that rapidly pushes the bacteria through main conduits of the epikarst (WF1). This causes a quick drop in SpC and a rise in discharge simultaneously. After the head is lowered, discharge decreases, SpC will slowly rise back toward pre-storm levels and FC counts will decrease. Often, the SpC does not return to previous base flow levels, likely due to the dilution of storage water by rainfall. However, during periods of higher storage, it appears as though continuing recharge and hydraulic pressure pushes out additional storage waters after storm event recovery, and there is a rise in SpC during the falling limb of thedischarge curve. During time in-between storms, waters percolate through diffuse conduits as evident by the lower fecal coliform and E. coli counts and steady rise of SpC, which indicates water that is in contact with the bedrock for a longer duration, thus dissolving more carbonate rock. This information is vital to understanding and improving best management practices in agriculture on karst terrains, particularly with regard to water quality degredation from the application of organic soil amendments. The results indicate that rapid, continuous contamination can occur from the use of manure and fertilizers in karst areas, and this must be addressed through better policy creation.AcknowledgmentsWe gratefully acknowledge USDA ARS for funding support for this project and Western Kentucky University for Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings113

PAGE 116

Figure 4. Dye trace 2 during the study period (dots), indicating the WF1 response to a storm event as shown by how the dye move d through with the storm pulses. Figure 5. Example of discharge response in WF1 during storm events from 2011. contributing resources to site management. Crumps Cave site purchased using a Kentucky Heritage Land Conservation Funds. Much thanks to WKU Hoffman Institute staff and students, Nick Lawhon, Sarah Arpin, Micah Ruth, Dan Nedvidek, Celia Davis, Lee Anne Bledsoe, Beth Tyrie, Kegan McClanahan, Laura Osterhoudt, Gabe Russell, and Dalene Smith for assistance in the field and lab.ReferencesBolster C, Grover C, Meiman J, Fernandez-Cortez A, Carrie Crockett, 2006. Practical limits of high-resolution evaluation of carbonate chemistry within karst flow systems. Proceedings of the Conference on Limestone Hydrogeology. Currens JS, 2002. Kentucky is Karst Country: What you should know about sinkholes and springs. Kentucky Geological Survey. Information Circular004, Series 12. Drew D, Holtzl H, 1999. Karst Hydrogeology and Human Activities. Impacts, Consequences and Implications. Rotterdam, Brookfield: Balkema. Groves C, Bolster C, Meiman J, 2005. Spatial and temporal variations in epikarst storage andflow in south central Kentuckys Pennyroyal Plateau sinkhole plain. Proceedings of the USGS Karst Interest Group Conference. Hess J, 1974. Hydrochemical Investigations of the central Kentucky Karst Aquifer System. PhD. thesis, Department of Geosciences, The Pennsylvania State University. NOAA. National Climatic Data Center http://www.ncdc.noaa.gov/ oa/ncdc.html. Accessed 11/23/2011. Pasquarell GC, Boyer DG, 1995. Agricultural Impacts on Bacterial Water Quality in Karst Groundwater. Journal of Environmental Quality 24(5): 959. Ray JA, Currens JC, 1998. Mapped karst groundwater basins in the Beaver Dam 30 60 Minute Quadrangle, Kentucky Geological Survey. Ray JA, Currens JC, 2000. Mapped karst groundwater basins in the Campbellsville 30 60 Minute Quadrangle, Kentucky Geological Survey. Ray JA, Blair RJ, 2005. Large perennial springs of Kentucky: Their identification, base flow, catchment, and classification. in Beck, B.F. (ed) Sinkholes and the Engineering and Environmental Impact of Karst. Geotechnical Special Publication 144. American Society of Civil Engineers, Reston, Virginia, 410. Richards PW, 1964. Geologic map of the Smiths Grove quadrangle, Kentucky. US Geological Survey Geologic Quadrangle Map GQ 357. Soil Survey Staff NRCS. 2011. United States Department of Agriculture. Web Soil Survey. http://websoilsurvey.nrcs. usda.gov/. Accessed (11/7/2011).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings114

PAGE 117

Figure 6. Fecal coliform (dots) counts at WF1 during a storm event from April 25 May 10, 2011. White WB, 1988. Geomorphology and Hydrology of Karst Terrains. Oxford University Press: New York. Worthington S, 2007. Groundwater residence times in unconfined carbonate aquifers. Journal of Cave and Karst Studies 69(1): 94.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings115

PAGE 118

KARST HYDROGEOLOGICAL OBSERVATIONS IN CAO BANG PROVINCE (VIETNAM): THE TRA LINH-THANG HEN LAKE AREAGheorghe M. Ponta1, Bogdan P. Onac2, Nyguen Xuan Nam3 1PELA GeoEnvironmental, 1009 23rdAvenue, Tuscaloosa, AL 35401, USA, gponta@yahoo.com2Department of Geology, University of South Florida, 4202 E. Fowler Ave., SCA 528, Tampa, FL 33620 USA, bonac@usf.edu3Vietnam Institute of Geoscience and Mineral Resource, nam_nguyen_xuan@hotmail.com In March 2012, the Geokarst Adventure Association under the auspices of the French Speleological Federation and the Romanian Speleological Federation along with the Vietnam Institute of Geoscience and Mineral Resource organized an expedition to explore the karst systems located in the Cao Bang Province, Northeastern Vietnam. The results include karst geological and hydrogeological observations and preliminary geochemical data based on 14 water samples. The values measured for the specific conductance are typical for caves waters (between 232.6 and 371 S/cm). Total alkalinity as CaCO3ranges between 92.5 and 186 mg/l (gour 68), and the total hardness as CaCO3 varies between 85 and 326 mg/l. The highest pH value (8.29), specific conductance (371 S/cm) and concentration of calcium (97.93 mg/l) were recorded at Thang Hen Lake. The lowest temperature of 16.4 C, specific conductivity of 166.9 S/cm, and total alkalinity as CaCO3of 68 mg/l were recorded in the water sample collected in the gour of Bang Ga T09 cave. The waters are calcium bicarbonate type. The local meteoric water line (LMWL) for the investigated region is D = 7.8318O + 14.694 and shows a higher intercept and slope value than the global meteoric water line (GMWL), suggesting major changes in the origin of precipitation during the seasonal rain cycles.1. Geographical SettingsThe Socialist Republic of Vietnam is situated in the easternmost part of the Indochinese Peninsula (Fig. 1a). It covers a total area of 331,210 km2of which 40% is represented by mountains, mostly forested. Carbonate rocks are exposed over 60,000 km2, representing 18.12% of countries surface (Clements 2006). The Province of Cao B ngis located in the northeastern Vietnam, 270 kilometers from Hanoi, on the border with China. It has a surface of 6,724.6 km2 and a population of 632,450 (Fig. 1a). The topography of the region is characterized by mountain ranges with elevation over 900 m above sea level (a.s.l.), and karstic plateaus developed between 500 and 700 m a.s.l. Tropical climate is characteristic to the area. The temperature varies from 5 C in December and January to 37 C in July and August. The karst of northern Vietnam was extensively explored over the last 15 years and several cave exploration reports, karst geology and hydrology papers are available: Brouquisse (1998/1999), Holroyd and his team (2003, 2005, 2010), Limbert and his team (1999, 2007), Nyuyen Thi Thuy (2007) and Italian-French-Vietnamese Caving Project (2007).2. GeologyThe Cao Bang province is underlain by a variety of rocks ranging in age from Cambrian to Quaternary. The majority of the karst features identified in the field and water samples collected in March 2012 are hosted in the B c S n formation of CarboniferousPermian age (C-P bs) which consists of siliceous shales, shaly limestones, and limestones. The limestone unit is up to 800 m thick. The Bac Son limestone is finely crystalline, light gray to dark gray. Bedding is generally 30 to 50 cm thick, oriented NW SE, dipping 19 to 25. Groundwater movement occurs along solutionally enlarged fractures, cavities, joints, and bedding planes. The area underlain by limestones is extensive. The topography consists of large poljes, surrounded by prominent limestones pinnacles and tower karst.3. Karst hydrogeological observations in the Tra Linh Thang Hen lake areaThe Tra Linh Thang Hen Lake area is formed by a sequence of poljes and karst windows developed along the Tran Linh River (Q ~400 l/s in March 2012). The limestone massifs are traverse by caves that carry rivers from one polje to the next. In some areas two levels of cave passages are developed, one as a stream passage at the present flood plain level, and a fossil one at 50 m relative altitude, marking a former flood plain level. The Tra Linh River is collecting its waters a few kilometers north of the border with China. Fourteen km downstream, the river is disappearing underground at the base of a limestone massif, reappearing 500 m downstream in an underwater cave, which ends with a large spring. Hundred meters downstream from the cave entrance an 8 m high waterfall is formed. Downstream, the Tra Linh River is meandering through a large polje, covered with alluvial deposits, sinking for the second time in a cave/swallet, which is penetrable for about 100 m (Fig. 2). The end sump is clogged with alluvial deposits, trash and tree branches. The sump was not explored due concerns that flash flood waters moved unexploded ordinances from the Vietnam War era into the cave. One km north from the cave, a stream with a flow of about 10 l/s is disappearing underground through an impenetrable swallet. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings116

PAGE 119

The Tra Linh River continues the underground pathway, reappearing for about 100 m, in two separate karst windows (Karst Window 105 and Tra Linh Sinking Stream Ponor 107 on Fig. 1b), and later is recharging the Thang Hen Lake through a stream cave. The lake is a large swallet, located at the bottom of a polje, where the Tra Linh River is sinking underground for the fifth time, recharging the main spring located at 374 m elevation, 5 km away. In March 2012, the lake was about 300 m long and 150 wide. Around the lake several dry swallets were identified. The main spring, with a discharge of 200 l/s represents the partial outlet of the system. Along the river, additional groundwater sources recharge the stream. About 1 km downstream from the spring, the yield of the river reaches 1,000 l/s. The type of the tower karst landscape develop along the Tra Linh River is Residual Hills on a Planed Limestone Surface (Ford and Williams 2007). During the raining season, the water level can rise up to 50 m in the area, the poljes mention above being flooded and interconnected, forming a large lake. A cross section along Tra Linh River, traversing tower karst landscape is shown in Figure 3. Figure 1. Location of the study area within the Cao Bang Province in the northern part of Vietnam.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings117

PAGE 120

4. Sampling and methodsFourteen water samples were collected and analyzed from the following areas: Thang Hen Mountain Lake (7), Trung Han (2), Bao Lac (2), Bang Ga (2; one spring and one cave pool water), and one rain water sample. Only the seven sampling point in the Tra Linh Thang Hen Lake area are shown on Figure 1b. The water samples were analyzed for anions, metals, and stable isotopes. Field parameters, including temperature, pH, specific conductance, and salinity, were measured with an YSI 63 instrument. A digital titrator (Hach Model 16900) was used to determine alkalinity and total hardness as CaCO3and carbon dioxide in the field (Table 1). Anion analyses were performed at the University of Alabama, whereas the metals and stable isotopes at the Department of Geology, University of South Florida. Anions were analyzed using a Dionex DX 600 Ion Chromatograph, trace metals analysis was achieved using a Perkin-Elmer Elan DRC II Quadrupole inductively coupled plasma mass spectrometer (ICP-MS) analytical instrument. Standards used were formulated stock standards with metals in concentrations from 1,000 mg/l to 10,000mg/l. US EPA Test Methods 200.7 and 6010B were performed to complete the metals testing and analysis. The stable isotope analyses were conducted on a Thermo Fisher Scientific (Finnigan) Delta V Isotope Ratio Mass Spectrometer. Figure 2. Polje along the Tra Linh River with a fossil cave marking the former flood plain level (Photo Gheorghe Ponta with Geo karst Adventure). Table 1. Field parameter data. 1: Trung Han cave system; 2: Bao Lac; 3: Thang Hen cave system; mean sea level (MSL).Sample Name Units1/Trung Han Spring 1 HQ 313/13/201213:50DateTimeElevation m (MSL)475Discharge (Q) l/s1007.54ph Temperature (T) C21.70Specific Conductance S/cm352.30Salinity ppt0.20Alkalinity as CACO3mg/l156.00Carbon Dioxide mg/l115.60Total Hardness as CACO3mg/l326.001/Trung Han Spring 2 HQ 323/13/201215:144432508.2519.80339.600.20186.0092.60274.002/Bao Lac HQ14 Spring3/15/201214:22219207.3120.10232.600.10192.5074.4085.002/Bao Lac Sump in Hang Kanh Xuan BL14 Cave 3/15/201216:09249207.4720.40241.200.1092.5040.60120.003/Tra Linh Sinking Stream3/19/201215:456434007.7623.50279.400.10119.0032.40147.003/Tra Linh Water Fall Cave TH 07 3/18/201214:026352507.1118.60300.500.10118.0048.80163.003/Tra Linh Sinking Stream Ponor 045 3/18/201211:446252506.5119.10297.600.10124.0040.203/Karst Window 1053/21/201214:476033007.7321.60296.600.10131.0045.603/Tra Linh Sinking Stream Ponor 107 3/21/201216:005931007.7821.50296.500.10125.0049.00188.003/Thang Hen Lake3/21/201211:115904008.2920.60371.000.20168.0042.60151.003/Tra Linh Spring3/19/201213:10374507.7021.70325.400.20119.0091.80193.00Bang Ga Gour Cave TR 093/20/201214:466980.00017.8816.40166.900.1068.0045.40123.00Bang Ga 098 Spring TR 043/20/201217:2460257.2020.40318.700.20136.0074.00273.00 Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings118

PAGE 121

5. Water quality dataLaboratory results for anions and cations are provided in Table 2. Sampling locations are shown on Figure 1b. The elevations of sampling points range between 219 m and 698 m. The estimated flow rates of the sampled springs range between 20 l/s and 400 l/s. The pH value ranges between 6.51 and 8.29, typical for karst waters. Temperatures ranged between 18.6 and 23.5 C, which correspond to mean annual air temperature in the area. The values we measured for the specific conductance are typical for caves waters (between 232.6 and 371 S/cm). Total alkalinity as CaCO3 ranges between 92.5 and 186 mg/l (gour 68), and the total hardness as CaCO3 varies between 85 and 326 mg/l. Calcium concentration ranges between 48.60 and 97.94 mg/l, with the highest values at the Thang Hen Lake, whereas the magnesium concentration was found to fluctuate between 0.30 and 5.76 mg/l. Carbon dioxide concentration ranges between 32.40 to 115.60 mg/l. The highest pH value (8.29), specific conductance (371 S/cm) and concentration of calcium (97.93 mg/l) were recorded at Thang Hen Lake. The lowest temperature of 16.4 C, specific conductivity of 166.9 S/cm, and Total Alkalinity as Ca CO3 of 68 mg/l were recorded in the water sample collected in the gour of Bang Ga T09 Cave. The waters are bicarbonate calcium type.6. Stable isotopesIsotope fractionation accompanying evaporation from the ocean and condensation during atmospheric transport of water vapour causes spatial and temporal variations in the deuterium and 18O composition of precipitation (Dansgaard 1964). Regional-scale processes such as water vapor transport patterns across landmasses and the average rainout history of the air masses precipitating at a given place controls the isotopic composition of local precipitation (Table 3). Figure 3. Cross section along Tra Linh River, traversing tower karst landscape.Sample ID DateTime Total Calcium Total Magnesium Total Manganese Total Potassium Total Sodium Total Strontium ChlorideNitratePhosphateSulfateTable 2. Summary of the water quality data. 1: Trung Han cave system; 2: Bao Lac; 3: Thang Hen cave system. As, Cd, Cr, Zn, F, and Br were not detected. CATIONS (METALS) ANIONS Units mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/lmg/l1 Trung Han Spring 1 HQ 31 3/13/2012 13:50 95.42 2.45 0.02 0.55 1.80 0.05 1.27 3.54 0.18 4.40 1 Trung Han Spring 2 HQ 32 3/13/2012 15:14 94.37 3.13 0.020.62 2.09 0.05 1.23 3.05 0.09 4.712 Bao Lac HQ14 Spring3/15/2012 14:22 57.02 4.25 0.00 0.59 2.98 0.08 0.73 3.64 0.02 4.923 Thang Hen Lake 3/21/2012 11:11 97.94 1.15 0.01 0.42 1.89 0.06 1.70 2.77 0.03 5.88 3 Tra Linh Spring 3/19/2012 13:10 86.87 5.76 0.37 1.07 3.78 0.08 1.48 3.21 0.00 5.61 Bang Ga Gour Cave TR 09 3/20/2012 14:46 84.78 5.68 0.20 1.15 3.98 0.08 0.65 1.33 0.00 8.30 Bang Ga 098 Spring TR 04 3/20/2012 17:08 0.00 0.00 0.00 0.00 0.00 0.00 0.78 3.76 0.11 5.60 3/15/2012 15:37 57.67 4.38 0.01 0.67 3.73 0.09 0.62 3.94 0.00 4.942 Bao Lac Sump in Hang Kanh Xuan BL14 Cave3 Tra Linh Sinking Stream 3/19/2012 15:45 57.67 4.38 0.01 0.67 3.73 0.09 1.61 2.01 0.00 5.79 3Tra Linh Water Fall Cave TH 07 3/18/2012 14:02 82.23 5.77 0.05 0.91 3.51 0.07 1.60 2.06 0.01 5.76 3/18/2012 11:44 75.31 5.35 0.05 0.97 3.28 0.07 1.61 2.79 0.01 5.763 Tra Linh Sinking Stream Ponor 0453 Karst Window 105 3/21/2012 14:47 75.31 5.35 0.05 0.97 3.28 0.07 1.67 2.28 0.05 5.77 3/21/2012 16:00 48.60 0.30 0.00 0.18 1.70 0.02 1.64 2.56 0.01 5.823 Tra Linh Sinking Stream Ponor 107 Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings119

PAGE 122

The 18O values of the water samples collected over 8 days ranged from -9 to -7.62. The D values in the cave waters, springs, and rivers ranged from -55 to -42. The only rainfall sample collected on March 13, 2012 stands out with its very positive values for both 18O and D (see Table 3). The relationship between 18O and 2H in worlds fresh surface / cave waters is predicted by the GMWL defined by Craig (1961) as: D = 818O + 10 (). The local meteoric water line (LMWL) for the investigated region is D = 7.8359x + 14.694 and shows a higher intercept and slope value than the GMWL, suggesting major changes in the origin of precipitation during the seasonal rain cycles. Although we called LMWL, we are fully aware that the isotopic compositions undergoes variations on event-based and therefore, our LMWL solely reflects the conditions during those 8 days of sampling, therefore, the results need to be considered as preliminary and used with caution if comparisons with other regions or data sets are sought.Table 3. Summary of the stable isotope data. 1: Trung Han cave system; 2: Bao Lac; 3: Thang Hen cave system. Sample 18O D d-excess ()()()1Trung Han Spring 1HQ 31 -7.86 -49.12 13.741Trung HanSpring 2HQ 32 -7.78 -46.19 16.062Bao LacSpring -8.93 -55.23 16.222Bao LacSump in Hang Kanh Xuan Cave -9.00 -52.54 19.493Tra Linh sinking stream -7.76 -45.34 16.733Tra Linh Water Fall Cave TH 07-7.63-42.3518.663Tra Linh Sinking Stream Ponor 045-7.98-44.2919.523Karst Window 105-7.66-53.967.313Tra Linh Sinking Stream Ponor 107-7.62-54.156.853Thang Hen Lake-7.71-48.5113.163Tra Linh Spring-8.08-45.2719.37 Bang Ga gour-8.39-42.4924.60 Bang Ga Spring T4-8.24-50.7915.10 Rain Water1.3726.8015.85 More samples of precipitation and fresh groundwater are needed to construct a real LMWL for this region, and to calculate the deuterium excess value, defined as: d = D 818O (). This parameter is a valuable indicator of the source area of the water vapor. Values around +10 are typical for most continental meteoric waters, whereas values well above +10 suggest more evaporated moisture being added to the atmosphere (Rozanski et al. 1992).AcknowledgmentsWe are grateful to the following organizations that provided us with the hydrogeological analyses: Institute of Geosciences and Mineral Resources and PELA GeoEnvironmental and to members of the Geokarst Adventure for their help during the fieldwork.ReferencesBrouquisse F, 1998/1999. Donnes physico-chimiques, Federation Francaise de Speleologie, Societa Speleologica Italiana (Unpublished report). Cao Bang 2007, Italo-french-vietnamese caving project in Vietnam (Unpublished report). Clements R, Sodhi NS, Schilthuizen M, Ng KLP, 2006. Limestone karsts of Southeast Asia: imperiled arks of biodiversity. BioScience, 56(9), 733. Craig H, 1961. Isotopic variations in meteoric waters. Science, 133, 1702. Dansgaard W, 1964. Stable isotopes in precipitation. Tellus, XVI (4), 436. Ford DC, Williams PW, 2007. Karst Hydrogeology and Geomorphology. John Wiley & Sons, London, 562. Holroyd M. and team, 2003, 2005, 2010. Vietnam Expedition (Unpublished reports). Limbert H. and team, 1999, 2007, 2009. Vietnam Caving Expedition. http://www.vietnamcaves.com/report-2009 Nguyen Thi Thuy, 2007. Application of stable isotope geochemistry in a study of paleoclimate and environment. Case study in Son La, Vietnam. Master Dissertation, Universiteit Gent, Vrije Universitei Brussels, Belgium. Rozanski K, Aragus-Aragus L, Gonfiantini R, 1992. Relation between long-term trends of oxygen-18 isotope composition of precipitation and climate. Science, 258, 981. Topographic map of Na Khoang, 1:50,000. Topographic map of Cao Bang, 1:50,000.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings120

PAGE 123

INCIDENCES OF THE TECTONICS IN THE KARSTIFICATION OFCHALK LIMESTONES IN THE WESTERN PARIS BASIN: EXAMPLE FROM THE PETITES DALES CAVE (SAINT MARTIN AUX BUNEAUX, FRANCE)Jol Rodet1, Kun Ma, Jean-Pierre Viard3 1UMR 6143 CNRS-M2C-Universit de Rouen, Laboratoire de Gologie, CNEK, 76821 Mont Saint Aignan, France, joel.rodet@univ-rouen.fr2Universit de Rouen, UMR 2795 CNRS-IDEES, Institut de Gographie, 76821 Mont Saint Aignan, France3Centre Normand dEtude du Karst et des Cavits du Sous-sol, 97 rue du 8 mai, 27400 Montaure, France The classical approach to study the karstification attributes a major role to the structure in the establishment of concentrate d drainage of groundwater. This structure, essentially tectonics and stratigraphy, serves to guide the water, which gradually opens up these discontinuities to build a network, from the introduction to the resurgence. This too idealistic view does not reflect the complexity of the establishment of a karst system. Indeed, experience shows that some bedrocks contain karst drains in the absence of any cracking. Whats more, some conduits can go through the structural elements without undergoing any morphological changes. In the chalk of Western Paris Basin, the Petites Dales Cave proves an excellent observatory. We have conducted a study on the relationship between the main conduit, restitution collector of the underground system, and observable fissures in the roof and walls of the conduit. Along a drain of 421 m, we counted 374 fissures, the total length of which being a little more than 867 m. Examination of the orientation of the drain and fissures reveals four types of relationship: (1) parallel (2) oblique, (3) perpendicular and (4) no joints. No correlation could be established between the development of the collector and the presence of fissures, other than very occasionally or during episodes of overflow. In fact, the relationship between fissure and karstic conduit cannot be established, therefore it is necessary to introduce other factors in the speleogenesis, such as porosity of the chalky bedrocks, and the direct effect of the hydraulic gradient.1. IntroductionTypically, it is assumed that the development of karst drains depends on the structure, especially the tectonics. Structural cracking is the main factor of permeability to water in a limestone (Jakucs 1977). Explorations in the chalky limestone of the Paris Basin showed that this was not always the case (Rodet 1992). So we started the study on the relationship between drain and fissures in one of the largest caves of Normandy, the Petites Dales Cave. Petites Dales Cave opens in the eponymous valley, 1.2 km from its outlet on the coast of the English Channel (Fig. 1). Small cavity of only 62 m of development at its speleological recognition in 1966, this site became one of the most important underground excavation sites in France since 1991 (Rodet et al. 2007). To date, with more than 710 m of explored galleries, the cavity is the largest in Seine Maritime. It hosts a multidisciplinary karst research program (Laignel et al. 2004; Rodet et al. 2006), being referenced as the chalk karst of Normandy Region (Rodet and Viard 2009). As part of this research program, we studied specifically the relationship between visible tectonics and directions of drain to verify the applicability of the concept of the drainage network establishment based on the tectonic frame (Hauchard et al. 2002, 2008) for cavities in chalk. Our approach is limited to observable fissures in the main gallery of the cave between the entrance (topographic point# 1 pt. 1) and the impact of the first solution pipe (topographic point # 43).2. MethodologyThe topographic map of the Petites Dales Cave was surveyed from a network of fixed and materialized stations, which are therefore easy to locate (Fig. 2). The topography of the collector was completely established along a length of 460 m, till the foot of the second solution pipe where we have not found upstream gallery. The impact of trepanation of the two solution pipes makes it illusory the study of fissures upstream of the pt. 43, which is 421 m from the entrance (Rodet et al. 2009). As a result, we limited our study area between the pt. 1 (entrance) and the pt. 43. From this topography, we carried out a campaign of systematic survey, in situ, of (i) the orientation of the Figure 1. Location of the study area.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings121

PAGE 124

Figure 2. Survey of the Petites Dales Cave, with the tectonic network. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings122 sections of the collector, (ii) the length of these sections, (iii) the orientation of visible fissures in the roof and the walls of the sections, (iv) the length of the fissures. Every joint was oriented (az degree) and measured with a tape measure. For fissures in the form of broken line or curve, we measured the orientation between the two extreme points. Each joint was numbered and plotted on the spot on a 1/100 topographic map. The measures were then introduced into a spreadsheet. By combining these data, we can define drain / joint relation models to assess the impact of visible tectonicson the karst development, and the impact of hydraulic gradient supported by the porosity of the bedrocks. The first element to consider is the spatial organization of the 42 segments which compose the drain being studied, according to their orientation and accumulated length of development by azimuth class. The second element is the tectonic network survey, according to its frequency, directional distribution and accumulated length of development by class. Finally, the comparison of the two types of data aims to understand the relationship between tectonics and karst drain under discussion. As indicated by the reference point pt. 43, the 42 oriented segments of gallery were grouped by class of 5 (Fig. 3a). The 374 inventoried joints were grouped according to their magnetic orientation (az value) by class of five degrees.3. ResultsThe cavity was examined along the 421 m of the drain. 374 joints, with a total length of 867.70 m of fissures (f), were identified. 3.1. Collector directions The distribution of the 42 directions is very dispersed: no class contains more than 4 sections and thus reaches 10%. It is relatively uniform across all classes, with a slight concentration in two large sets (11 and 106), in the middle of less relevant values.Concerning the cumulative length of the conduits (Fig. 3b), two sets are slightly reinforced. The first set is bipolar with a first peak around 11/15, a second around 31/50, and more specifically between 40 and 50, clearly in SW/NE direction. The second set is more uniform with a maximum in the 136/140 Class, the maximum being further accentuated as to the cumulative of NW/SE segment lengths. Both directions, and especially values between 120 and 150, correspond to lineaments and faults of the Pays de Caux (Hauchard et al. 2002, 2008). 3.2. Joints in the collector The output gallery is marked regularly with topographic stations used as a benchmark to locate oriented gallery segments and fissures identified in the roof and the walls of the gallery. The total development of drain is 421 m (420.90 m), in which 374 joints are identified with a total fissure length of 867.70 m. Thus, the average length of fissure is 2.32 m, the number of fissures per linear meter of drain is 0.89, and the length of fissure (mf) per meter of drain is 2.06 m. For morphological reasons (Fig. 2), we divided the main gallery into four sectors, namely:

PAGE 125

3.2.3. From pt 24 to pt 37 In this third sector, the length of conduit is 136.40 m, which represent 32.41% of the total length of conduit. We identified 118 joints, or 31.55% of all fissures identified throughout the gallery. We obtain a density of 0.87 fissure per meter of gallery. The total length of joints rises to 267.50 m, which means a ratio of 1.96 m of fissure per meter of drain. 3.2.4. From pt 37 to pt 43 In fourth sector, the length of conduit is 64.40 m, which represent 15.30% of the total length of conduit. We identified 88 joints, or 23.53% of all fissures identified throughout the gallery. We obtain a density of 1.37 fissures per meter of gallery. The total length of joints rises to 67.30m, which gives a ratio of 1.05 m of fissure per meter of drain. 3.2.5. Summary of all sectors There is therefore a medium density of joints in the first sector (0.85 f/ml). It falls in the second sector to 0.66 f/ml, returns to the average of the first sector in the third sector (0.87 f/ml), and increases significantly in the fourth sector (1.37 f/ml). We also note that the joint density is not in the pair with the joint length, since the ratio decreases from the entrance to the end where the lowest ratio (1.05 mf/ml) accompanies the highest fissure density (1.37 f/ml). Only in the two intermediate parts the overall average is approached. Is this increasing ratio from the end to the entrance reflecting a possible relaxation effect of the massif induced by the opening of the Petites Dales Valley? 3.3. The directions of joints They are 374 inventoried joints grouped according to their magnetic orientation (Fig. 4). There is an overwhelming concentration of orientations between 100 E and 145 E. The rest is trivial: a small class of 15 and an even smaller one near 180 (Fig. 4a). This NW/SE orientation dominance is even stronger. It concentrates around 115 when we focuse not on the number of joints but their cumulative development (Fig. 4b).4. DiscussionWhat do the results mean? 4.1. Drain/joint directional relations The relations between the direction of the drain and the joints reveal four cases. 4.1.1. The joints accompany or frame the drain We can suppose that there is a relationship between joint and drain, as a lot of studies shows in numerous conventional caverns in limestone with low permeability or porosity (White, 1988). Certainly the combination of fissures and hydraulic gradient provides the best conditions for the development of karst. In the main gallery, it represents 148.4 m, therefore 35.26% of the gallery. 4.1.2. The joints intersect obliquely the drain The joints intersect obliquely the drain without thereby deviating the gallery. We must therefore conclude that the 3.2.1. From pt 01 to pt 11 In this first sector, the length of conduit is 119.30 m, which represent 28.34% of the total length of gallery. We identified 101 joints or 27.1% of all fissures identified throughout the gallery. We obtain a density of 0.85 fissure per meter of gallery. The total length of joints rises to 336.80 m, which means a ratio of 2.82 m of fissure per meter of drain. 3.2.2. From pt 11 to pt 24 In this second sector, the length of conduit is 100.80 m, which represent 23.95% of the total length of conduit. We identified 67 joints, or 17.91% of all fissures identified throughout the gallery. We obtain a density of 0.66 fissures per meter of gallery. The total length of joints rises to 196.10 m, which means a ratio of 1.95 m of fissure per meter of drain. Figure 3. Distribution of the segments in the main gallery. A:number of segments by 5 sets. B: cumulative length of the segments by the same sets. Figure 4. Distribution of the fissures in the main gallery. A: number of joints by 5 sets. B: cumulative length of joints distributed in the same classes.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings123

PAGE 126

oblique fissures do not affect directly the conduit. In the collector, it represents 130.2 m and therefore 30.93% of the gallery. 4.1.3. The joints intersect perpendicularly the drain The joints intersect perpendiculary the drain without impacting the morphology of the walls. Joint does not influence the flow. As in the previous case, only the phases of flooding with a very reduced flow speed affect the morphology of the drain by the establishment of equilibrium chimneys. In the collector, it represents 86.8 m and therefore 20.62% of the gallery. 4.1.4. The joints are absent from the drain In the collector, there is no joint along 55.5 m, that is to say, 13.19% of the gallery has no relationship with tectonics (Fig. 5). In total, 365.40 m of the gallery exhibit fissures in the ceiling or walls, that represent 86.81% of the total length of the collector, but in which only 148.4 m, or 35.26%, show a clear relationship between tectonic and karst drain. We conclude that tectonics is not the only responsible for the establishment of all sections of the collector. Other factors involve either stratigraphy (favorable interbedding) or porosity (with direct impact of the hydraulic gradient). Some authors, supporters of the all tectonics, suggest the existence of a network of micro-cracks, unproven to date: curiously this network should be large enough to guide the drain, but at the same time it would not be revealed by the flooding phases 4.2. Drain and joint The last approach was made by comparing the drains and joints. It shows significant disparities, where there are areas with many joints or significant accumulated metrics of joints, and areas almost without joint. We have established a ratio between length of joint (mf) and length of gallery (mlg), which reveals three classes, depending on whether the ratio is close to the average ratio of the cavity, higher, or on the contrary, lower: 4.2.1. Gallery with a low ratio of fissure The ratio is low (<1 mf/mlg). 11 of 42 sections provide a ratio of less than 1 m of fissure per lineal meter of drain. This represents 26% of the sections. 4.2.2. Gallery with a ratio close to the average The ratio is close to the average of joint (2.06 mf/mlg) of the whole gallery, between 1 and 3 m of fissure per meter of drain. This represents 25 of the 42 sections of the collector, i.e. 60% of the sections. 4.2.3. Gallery with a strong relationship to tectonics (>3 mf/mlg) Only 6 sections provide a ratio greater than 3 between the meter of fissure and the linear meter of the gallery. Those sections represent 14% of the segments of the collector. Strangely, the sectors with a high ratio of fissure are those which were strongly influenced by episodes of significant flooding associated with sudden and torrential floods from the downstream solution pipe (pt. 44). Therefore, it relates Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings124 to the morphology after the establishment of the drain collector, thus the tectonic factor cannot be applied as a determinant factor to the morphogenesis of the former drain. The last graphic (Fig. 6) oppose the cumulative length of drains grouped into orientation classes by 5 to the cumulative length of fissures, which are also grouped by class of 5. There is a wide difference in the distribution of the orientations between the development of drains and the development of joints. In particular, note that while the cumulative lengths of conduits (classes 116 and 121) are from 7 m to 24 m, giving a ratio of 3.5, the cumulative lengths of joints are equal (175.9 m and 177.2 m), giving a difference of 1.3 m or a ratio of only 1.007. Moreover, it should be noted that values equivalent to the linear meters of drains (24 m for 121, and 28 m for 41) are completely opposite to the those of joint (177.2 m and 2.1 m), i.e. a ratio of 84. The lack of relationship between classes of drains and those of fissures, shows very clearly in this case that the development of drains does not depend on the tectonics of bedrock, but on other speleogenetic factors and constraints which must to be identified.5. ConclusionThe collector of the Petites Dales Cave offers several sectors without joint (Fig. 5). Sections where there are oblique (Fig. 7) or perpendicular joints which do not affect the morphology of the collector drain are not linked to the structure of the bedrock. It is only in the presence of joints / drain correlation that tectonics can be considered with Figure 5. Segment of the main gallery of the Petites Dales, without joint, between pt. 2 and pt. 3 (photo by D. Guillemette).

PAGE 127

certainty to affect the karstification. In theory, it is one out four cases. Concerning the development of collector of the Petites Dales Cave, it represents less than 150 m of the drain, that is to say, just over 35% of the main drain. Clearly, porosity, by allowing the dispersion of the flow, promotes the expression of the hydraulic gradient. This dimension reduces, sometimes even annihilates the impact of the structure. Thus, it is in the phases of saturating flooding, when the flow velocity is minimized by effect of damming, that the weaknesses of the bedrocks manifest, especially through joints. As a result, the correlation or the non-correlation between the direction of the drains and the structure of the bedrock reflects the impact of bedrock on drains. For a long time, chalk was compared to a sponge and thus the existence of karst within it was denied. Today, no one denies any more the obvious, but the chalk, porous limestone, has its specificities, both morphologically and hydrologically, which are shown by the development of the karst.ReferencesHauchard E, Laignel B, Delahaye D, 2002. Proposition dun nouveau schma structural du Nord-Ouest du bassin de Paris reposant sur lanalyse fractale de la morphologie des rseaux de thalwegs et les donnes rcentes de la gologie rgionale. C. R. Geoscience, 334, 295. Figure 6. Dual graph of distribution by class of orientation (5). A: cumulated length of the segments in the main gallery. B: cumulated length of the joints observed in the main gallery. We note the absence of correlation between A and B. Figure 7. Segment of the main gallery, with obliquily intersection of joints, between pt. 24 and pt. 25 (photo by J. Rodet). Hauchard E, Laignel B, 2008. Evolution morphotectonique de la marge nord-occidentale du Bassin de Paris. Zeitschrift fr Geomorphologie, 52(4), 463. Jakucs L, 1977. Morphogenetics of karst regions variants of karst evolution. Adam Hilger, Bristol: 284. Laignel B, Dupuis E, Rodet J, Lacroix M, Massei N, 2004. An example of sedimentary filling in the chalky karst of the Western Paris Basin: Characterization, origins and hydrosedimentary behaviour. Zeitschrift fr Geomorphologie N.F., 48 (2), 219. Rodet J, 1992. La craie et ses karsts. Ed. Groupe Seine CNRS Caen & CNEK, Elbeuf, 560. Rodet J, Laignel B, Dupuis E, Brocard G, Massei N, Viard J-P, 2006. Contribution of a sedimentary study to the karstic evolution concept of a chalk cave of the Western Paris Basin (Normandy, France). Geologica Belgica, 9 (3), 287. Rodet J, Viard J-P, Poudras J, 2007. A la dcouverte de la grotte des Petites Dales (Saint Martin aux Buneaux, Seine Maritime, France). Splo-Tract, 6, 38. Rodet J, Viard J-P, 2009. La grotte des Petites Dales, un patrimoine normal? non! normand! Spelunca, 114, 28. Rodet J, Willems L, Brown J, Ogier-Halim S, Bourdin M, Viard J-P, 2009. Morphodynamic incidences of the trepanning of the endokarst by solution pipes. Examples of chalk caves in Western Europe (France and Belgium). Proceedings of the 15thInternational Congress of Speleology, Kerrville (Texas, USA), 19 July 2009, UIS, vol. 3, contributed papers: 1657. White WB, 1988. Geomorphology and hydrology of karst terrains. Oxford University Press, 464.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings125

PAGE 128

CEILING CHANNEL AND INPUT KARST. EXAMPLE OF THE PETITES DALES CAVE, NORMANDY, FRANCEJol Rodet1, Laurent Magne2, Jean-Pierre Viard3 1CNRS UMR 6143-Laboratoire de Gologie, Universit de Rouen and CNEK, 76821 Mont Saint Aignan, France, joel.rodet@univ-rouen.fr2Hydrogeologist engineer and CNEK, 94430 Chennevires sur Marne, France, lmim.ber@gmail.com3Centre Normand dEtude du Karst et des Cavits du Sous-sol CNEK, 27400 Montaure, France, follainjanine@orange.fr Within the endokarstic forms, the ceiling channel is often put forward because of its morphodynamic significance. Generally, the ceiling channel demonstrates a breaking into the initial drainage and a reduced stream flow resumes in a very congested drain. This top incision plays a new collection role on a filled drain which is also hydrologically disconnected, and gradually the channel becomes more and more important downstream. Studies made for more than twenty years in the Petites Dales Cave (Saint Martin aux Buneaux, France) revealed that the ceiling channel excavated in the roof of the collector-gallery presents an atypical development, which does not correspond to the classic scheme because it is much more developed upstream than downstream. This progressive morphology shows a huge variation of dimension linked to abrupt changes in the collectors orientation. These elements help to affirm that the ceiling channel of the main drain of the Petites Dales Cave is a form of introduction whose function is the absorption of brutal introductions by the water chasms which hydrate the underground network.1. IntroductionIn karstic geomorphology, the ceiling channel is an important indicator about the evolution of the drain, even the system (Jennings 1985). In a classic approach, it deals in fact with an erratic drainage of restitution which develops at the top of a filled drain, temporarily non-operational (Renault 1968). As a consequence, the channel becomes bigger downstream, in link with the collector function of the conduit (Slabe 1995). But sometimes, the channel can work differently, with another kind of drainage: the absorption of an introduction as a consequence of the trepanation of the restitution drain by a pothole or a solution pipe (Rodet et al 2009). In the latter case, the channel becomes smaller and smaller as it develops downstream, by reducing the flow absorbed by the restitution drain. This is what we focused on in the main gallery of the Petite Dales Cave. The Petites Dales Cave is an important chalk cavity in the Western Paris Basin (Fig. 1) because of its dimensions and moreover because of the section of its main gallery, from 2 to 5 m large for 10 m high, almost completely filled (Rodet et al. 2007). The Norman speleologists work, under the direction of Jean-Pierre Viard, by a partial emptying of the soil runoff landfill since 1991, revealed day after day the importance of this cavity whose size was 62 m when speleologists discovered it in 1966, now there are more than 710 m of developed length (Rodet and Viard 2009). Poorly developed downstream, the channel, which develops at the roof of the collector, increases upstream. Several hypothesis were envisaged, that the discoveries linked to the cleaning, easily rejected, until the base of solution pipes was crosscheck by exploration. We have also studied the conditions of contact between introduction and restitution dynamics, leading to the notion of trepanation (Rodet et al 2009). In this context, the ceiling channel is at least significant.2. MethodologyThe Petite Dales Cave benefits from a good organization: its topographic stations are easy to find (Fig. 2). Referring to this topographic network we directly checked the height and width of the ceiling channel in the main gallery between the topographic points (pt.) 2 and 43. To this finality we used a flexible panta-decametre, a level and an electronic laser meter. The data were noted on a notebook and registered in our office computer. We excluded (i) the downstream first segment of the gallery, between the entry (pt. 1) and the confluence (pt. 2) because the channel is absolutely absent, erased by the calibration of the collector during the reactivation of the siphon and (ii) the segment between pt. 42 and pt. 43 because the channel exactly corresponds to the topographic drain, because the underlying collector has not been cleared. The topographic statement is composed of 259 stations which represent a metric measure if sectors without channel Figure 1. Location of the study area.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings126

PAGE 129

are excluded. Each station has been subject to 3 to 8 measures, totaling more that 1,000 measures. With this morphometric statement we have realized more than 600 photographic cross-sections, set to the topographic network. A longitudinal cross-section was then designed by a computer, enabling a study of the latter. This approach was completed by perpendicular sections from the morphometric statements and the photographic cross-sections. Figure 2. Cave survey with topographic stations (pt) and cross sections (A to G; see Fig. 4).3. ResultsAll data have been gathered and studied, and this enabled us to build a longitudinal profile between pt 2 and pt 42 (Fig. 3). It can be noticed that the channel does not evolve in a regular way downstream to upstream and upstream to downstream. Then this is not a regular linear flow. Then what is its function? If it is a concentration drainage (output or restitution karst), its most evolved parts should contain significant lateral contributions (tributaries). If it is an introduction drainage (input karst), sudden entries leading to flows and sedimentary disruption inside the collector. The longitudinal profile reveals 4 sectors from upstream to downstream which can be linked to the cave plan (Fig. 2) 3.1. The upstream ceiling channel and the Espace des Six chamber Between upstream (pt. 42) and 315 m from the entry (pt.35) we can observe a 2 m high ceiling channel, well distinguished and almost rectilinear, easily penetrable by a man. It is rooted to the base of a solution pipe (pt. 44). This channel, almost separated from the support-gallery (Fig. 4A) leads in a perpendicular way to the most elevated sector of the collector, the Espace des Six chamber whose volume is imposing and seems to hang above the collector (Fig. 4B). Its roof rises more than 8 m above the ground of the collector. 3.2. The central part Between pt. 35/315 m (downstream of the Espace des Six chamber and pt. 27/240 m, the sector is low and sinuous, in which the channel is more discreet and above all discontinuous (Fig. 4C). It is rarely higher than 0.5 to 1 m with plenty of small equilibrium chimneys. It is not easily penetrable by a man. 3.3. The great curve Between pt. 27/240 m and pt. 22, 190 m, the channel rises again between 1 and 2 m above the collector and shows a width which can merge with the width of the collector (Fig. 4D). We can observe a high density of equilibrium chimneys, which can rise higher than 5 m. Between pt. 27 and pt. 24, the drain is almost straight and presents the longest straight line since the Espace des Six chamber. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings127

PAGE 130

This area offers a large volume, which can be compared to the Espace des Six chamber, with smaller dimensions. Between pt. 24 and pt. 22, the channel remains high but becomes narrower. Moreover, pt. 22 presents the last dome which rises higher than 5 m above the rolling ground of the collector. 3.4. The downstream ceiling channel Beyond this dome, from pt. 22/190 m to outside, the channel tends to reduce as well as presenting more frequent and larger interruptions (Fig. 4E). It rarely goes higher than 0.50 m and gets too narrow for a man to fit in (Fig. 4F). The dissolution pockets do not rise higher than 3 m above the rolling ground, except for 3 exceptions which are not higher than 4 m. From pt. 4/60 m, the channel becomes discrete to anecdotal (Fig. 4G). Between pt. 2 and pt. 1, on the last 31 m, the collector has been re-calibrated and the channel has been completely erased. The ceiling half-tube is therefore much larger in its upcollector part then in its down-collector part. Why is that?4. DiscussionTwo aspects are particularly noticeable: (i) what does the size reduction of the channel from upstream to downstream mean?, and (ii) why do we observe a great irregularity in the longitudinal profile? The reduction from upstream to downstream of the dimension of a drain often means a reduction of the flow which digs it. If the flow decreases, it means that the drainage scatters instead of concentrating itself, which means that it is not a restitution drainage which collects underground waters, but, on the contrary, an introduction drainage from the surface, which is absorbed along the drain. Figure 3. Longitudinal profile of the main gallery. Between 374 m and 319 m, a well developed ceiling channel with downstream a large reservoir; between 319 m and 187 m, a moderately developed ceiling channel with downstream a smaller reservoir; between 187m and 31 m, a punctually developed ceiling channel without reservoir, only equilibrium chimneys. Figure 4. Selected cross sections of the main gallery in the Petites Dales Cave. The black section represents the gallery and the grey section, the development of the ceiling channel. A: pt. 38; B: between pt. 36 and pt. 37 (Espace des Six dam chamber); C: bet ween pt. 30 and pt. 31; D: between pt. 24 and pt. 25 (dam-equilibrium chimneys); E: pt. 19; F: between pt. 12 and pt. 13; G: between pt. 5 and pt. 6 (lower point of the main gallerys roof). See the Figure 2.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings128

PAGE 131

Consequently it is an introduction and we have to identify the source, meaning, the input point. Right up pt. 44, the clearance work has shown the base of a solution pipe. In a spectacular way, and without a doubt, this solution pipe has a morphodynamic link with the ceiling channel which starts there, and is, therefore, the source of this introduction channel. This is proved by the morphological continuity (Rodet et al 2009), confirmed by the sedimentary continuity (Laignel et al 2004; Rodet et al 2006). We can particularly notice the complexity, midstudy, of the fluvial sedimentary deposits, strongly upset by many torrential introduction phases which have affected this part of the cavity. This torrential dynamic is included in the ceiling channel, of which the morphology strongly contrasts with the rest of the cavity. Three particularly noticeable morphological elements illustrate invasive torrential dynamics from the solution pipe. They demonstrate an adaptation of the cavity to these sudden and abrupt external inputs, and their effects on the restitution networks. 4.1. The upstream channel of the Espace des Six We can notice that the ceiling channel is an introduction pipe coming from the pt. 44/45 solution pipe which joins tangentially the main gallery previously calibrated by the earlier trepanation of the upstream pt. 47 solution pipe. We can observe the junction of a superior drain which quickly loses its erosive and transport dynamic (the drainage loses speed), as well as its drainage mass, which is partially absorbed by the underlying completely filled collector. This explains why the ceiling channel appears more distinctly above the collector as we get closer to the Espace des Six chamber, particularly with an input with tends to isolate both drains (Figs. 4A, 5). Upstream, in the draw point 404 (pt. 42), channel and collector are fused with the same width, and it is more difficult to show their limit. This evolution can explain the number and the size of big chalk blocks seen in the draw point 404, which can result in a destruction of the intermediary floor: no water circulation could explain the movement of such volumes, without leaving traces on the walls. On the contrary, we have to consider a stock of introduction water of which the geochemical activity would dissolve the bedrock. The blocks are almost in situ and underline the destruction of the floor which isolated the ceiling channel coming from the pt. 44 solution pipe and the calibrated gallery coming from the upstream solution pipe (pt. 47). 4.2. The Espace des Six chamber Located right downstream, the Espace des Six chamber (pt. 36/37) is a noticeable point, nonesuch in the whole Petites Dales Cave. Indeed, nowhere else in the cavity can we observe such a large volume. We can also note that the enlargement concerns the top part and not the whole volume (Figs. 4B, 6). Another important element: the inferior third is narrow and was infilled, the intermediary third widens and was infilled too, and the superior third is wide and was empty, except for chalk blocks fallen from the walls and from the roof, underlining submergence phases. It clearly appears that this large volume hanging above the collector is an underground tank, stocking important water inputs during sudden introductions. Figure 5. Between pt. 37 and pt. 39, the channel seems to be separate over the main drain by a platform in the bedrock (photo by J. Rodet). Figure 6. Between pt. 36 and pt. 37, the Espace des Six, large dam chamber perched on the more narrow main gallery (photo by L. Magne).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings129

PAGE 132

4.3. Point 24 turning point It is the third point morphologically remarkable that can be observed from the upstream part of the main gallery. It clearly shows a high contrast between the upstream side of the drain, almost rectilinear for about 30 m, and downstream which contains not less than four right angle bends on a few dozen meters, except the extremely tight curve of point 24 sensu stricto, which almost enables a half turn to the flow thanks to a 45 angle. If the channel worked like a collection and restitution water drain, the abrupt changes in direction would not have such an incidence. What can be seen? Downstream: close and tight turns, in a quite reduced drain, highly congested by sediment filling and with a narrow and high ceiling channel. Upstream: a long straight line, one of the most important in the cavity. The roof presents a larger channel which is however largely disemboweled by a succession of equilibrium chimneys, that are high (about 6 m), more and more numerous reaching point 24, giving a beginning coalescence which strangely enough reminds of the draft of a reservoir similar to the Espace des Six chamber (Figs. 4D, 7). The difference is essentially due to the positioning of the joints. If in the Espace des Six chamber, the parallel positioning of the drain and of the fissures facilitates the development of a large volume, at point 24, the joints located upstream are oblique compared to the axis of the drain. They generate a succession of equilibrium chimneys, approximately ten along 20 m. The dynamic responsible for the morphology is the one of a rising which hastily flow out from the upstream through the sinkhole and the solution pipe and hammers the wall of the sudden bend of point 24. It leads to a sudden stopping of the dripping and thus the quasi-instant loss of its sedimentary load which settles and partially or totally blocks the drain downstream. Whereas water accumulates and generates a pressure to the ceiling which is responsible for the digging of equilibrium chimneys on the weak points that the fissures are. Their coalescence draws a chamber typed tank. From these different elements, it can be deduced that the ceiling channel of the Petites Dales Cave is a feature which results from concentrated introductions from the surface, in a restitution drain. The episodic and torrential dynamic leads to transport solid elements which deposit and settle out as soon as the stream slows down because of the well-defined turnings of the drain, creating an obstacle to the flow and phases of submergence upstream with diggings of dam chambers. Both the drain morphology and the sedimentation figures in the fillings confirm this model of speleogenesis.5. ConclusionThe geometry of such a kind of ceiling channel is not compatible with the classic model of restitution drain of the underground waters. The decreasing section of the ceiling channel toward downstream shows indeed that competence is lost and that the latter is opposed to the notion of flow concentration. Then the examination of the longitudinal profile shows a relation between the development of a volume reservoir and brutal orientation changing of the flow. Such an association (upstream: reservoir/downstream: brutal direction changing) shows the process of a very concentrated flow penetrating a pre-existing drain. Unlike the common image of the ceiling channel, which is an erratic circulation collector, the example of the Petites Dales Cave shows a model of concentrated introduction of surface water in a restitution network. This aspect seems it has never been studied through a karstological approach, in spite of the consultation of a wide bibliography, although it gives plenty of information about the dynamics of underground waters, and enables a better understanding of the proccesses of some physical and chemical characters, such as for instance the turbidity in emergences and catchment areas.AcknowledgmentsThe authors would like to thank the French Federation of Speleology to have classified the Petites Dales Cave in the Underground Heritage Conservatory Agency, and all those who donated time and energy to allow this big clearing site in the absence of which this study could not have existed.ReferencesJennings JN, 1985. Karst geomorphology. Basil Blackwell, Oxford: 293. Laignel B, Dupuis E, Rodet J, Lacroix M, Massei N, 2004. An example of sedimentary filling in the chalky karst of the Western Paris Basin: Characterization, origins and hydrosedimentary behaviour. Zeitschrift fr Geomorphologie N.F., 48 (2): 219. Renault P, 1968. Contribution ltude des actions mcaniques et sdimentologiques dans la splogense. Annales de Splologie, tomes 22 (1967): 5, 209 and 23 (1968): 259, 529, 317. Rodet J, Laignel B, Dupuis E, Brocard G, Massei N, Viard J-P, 2006. Contribution of a sedimentary study to the karstic evolution concept of a chalk cave of the Western Paris Basin (Normandy, France). Geologica Belgica, 9 (3): 287. Figure 7. Flooding reservoir resulting of coalescence of several equilibrium chimneys developed on the structure, obliquely to the axis of the drain, between pt. 26 and pt. 24 (photo by J. Rodet).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings130

PAGE 133

Rodet J, Machado MMM, Ruchkys de Avezedo 2012. Thoughts about the karstic heritage management and its enhancement: examples from Normandy and Brazil. Proceedings of the 13thNational Congress of Speleology, Muotathal (CH), 29 sept. oct. 2012: 305, abstract, PPT. Rodet J and Viard J-P, 2009. La grotte des Petites Dales, un patrimoine normal? non! normand!. Spelunca, 114: 28. Rodet J, Viard J-P, Poudras J, 2007. A la dcouverte de la grotte des Petites Dales (Saint Martin aux Buneaux, Seine Maritime, France). Splo-Tract, 6: 38. Rodet J, Willems L, Brown J, Ogier-Halim S, Bourdin M, Viard J-P, 2009. Morphodynamic incidences of the trepanning of the endokarst by solution pipes. Examples of chalk caves in Western Europe (France and Belgium). Proceedings of the 15thInternational Congress of Speleology, Kerrville (Texas, USA), 19 July 2009, UIS, vol. 3, contributed papers: 1657. Slabe T, 1995. Cave rocky relief and its speleogenetical significance. Znanstvenoraziskovalni Center SAZU, 128.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings131

PAGE 134

GULLS, GULL-CAVES AND CAMBERING IN THE SOUTHERN COTSWOLD HILLS, ENGLANDCharles Self1, Andrew Farrant2 1University of Bristol Spelaeological Society, UK., self@globalnet.co.uk2British Geological Survey, Keyworth, Nottingham, NG12 5GG, UK, arf@bgs.ac.uk The study of fissures and caves formed by mass movement is important not only because they are a significant geohazard, but because such caves can be repositories for palaeogeographic information. In southern Britain, such open joints in solid strata are known as gulls. In the southern Cotswold Hills, a sequence of interbedded limestones and mudstones has been deeply dissected by the River Avon and its tributaries. Mass movement and cambering has opened up narrow gulls and larger gull-caves to a far greater extent than had previously been realized. Gull-cave surveys and the mapping of gulls inside the extensive abandoned Box Freestone Mine have allowed the spatial distribution of these features to be studied. We found that the direction of extension is not always in the direction of topographic gradient; locally it can be influenced by the stratal dip direction. In some areas, abrupt changes in the orientation of valley sides have allowed mass movement in two directions, creating rectilinear networks of gull-caves. A quantitative assessment of the valley-ward extension of the strata caused by gulling was made in Box Freestone Mine. Extension may often exceed 5%, and locally may reach 9% in discrete zones. Evidence of preferential dissolution along the NWSE joint set in both gulls and gull-caves suggests former groundwater flow to the north-west, not to the present River Avon valley. This indicates that the River Avon has subsequently captured the dip slope streams that once fed the headstreams of the River Thames. The greatly enhanced flow of the River Avon after this capture caused rapid over-deepening of the valley, which triggered the original cambering and thus formation of the gull-caves themselves.1. IntroductionThe Cotswold Hills, located in the south-western part of the English Midlands, forms a significant escarpment about 100 km long and generally 20 km wide, reaching a maximum elevation of just over 300 m near Cheltenham. They comprise a sequence of interbedded limestones and mudstones of Middle Jurassic (Bathonian) age, with a steep scarp slope facing to the west and a shallow dip slope to the east (Barron et al. 2011). Rivers rising on the dip slope flow east towards the River Thames and the North Sea, scarp streams flow west to the River Severn and the Atlantic Ocean. The exception to this is the (Bristol) River Avon which rises on the dip slope of the southern part of the Cotswold Hills, then cuts through the escarpment via the Claverton Gorge near Bath; it continues west, picking up scarp stream tributaries, and passes through the cities of Bath and Bristol before reaching the River Severn. This paper describes part of the southern Cotswold Hills to the east of Bath which has been deeply dissected by the River Avon and the By Brook, a major tributary which joins from the east (Fig. 1). In the valley floor, the rivers have incised through the Middle Jurassic sequenceinto Lower Jurassic strata of the Lias Group, which are mainly mudstones with subordinate limestones. The lower part of the valley sides are cut in the Inferior Oolite Formation, a rubbly oolitic limestone up to 23 m thick. Above this is the Fullers Earth Formation, a series of calcareous mudstones with occasional beds of flaggy limestone, approximately 46 m thick. The Fullers Earth mudstones are over-consolidated, highly plastic clays prone to mass movement. Capping the interfluves is the Chalfield Oolite; a succession of largely matrix-free oolitic limestones 35 m thick. Succeeding these limestones on the dip slope is the Forest Marble Formation, a sequence of coarse bioclastic limestones and mudstones. The regional dip is about 2 to the south-east. These Jurassic mudstones and limestones are all prone to mass movement. The valley sides of the River Avon and its tributaries have extensively foundered, with significant land slipping in and around the city of Bath. These mass movements include both rotational landslips and extensive cambering. Gulls are common in the more competent oolitic limestones. The term gull (derived from gully) is an old quarrymans term, used to describe open joints in solid strata (Fitton 1836). They are particularly well developed in the Chalfield Oolite where the major joints have opened as the stata has extended valley-ward. When large enough to be explored by cavers, they are termed gull-caves; the anthropomorphological element of this definition is important because it means that they are accessible to direct study. Gull-caves are different from normal dissolutionally widened fissures and caves, and can be identified by their distinct morphology. Gulls are typically narrow, parallelsided, joint orientated rifts, often with symmetrically opposing wall morphologies (fit features of Self 1986), but where there has been vertical as well as lateral movement, bedding planes or other discontinuities may also have parted. In the Bath area, the Jurassic oolitic limestones are excellent building stones, as they can be cut as freestones which then harden on exposure to air. Consequently they have been extensively mined. Several large abandoned pillar and stall workings, with several hundred kilometres of surveyed passages occur in the Bath area. The effects of cambering can be seen in many of these mines. Gulls are common and there are also gull-caves large enough to be accessible for direct study.Moreover, these extensive stone mines enable the spatial extent of the gulls to be mapped out over wide areas, and can be used to identify zones of maximum extension. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings132

PAGE 135

2. The formation of cambers and gull cavesGulls form on steep hillsides as a result of mass movement, when well-jointed strata are unsupported on their downhill side. In sedimentary rocks, extension takes place along bedding planes with bed-over-bed sliding and the opening of joints (Hawkins and Privett 1981). Gulls are particularly common in flat-lying or gently inclined strata affected by cambering. Cambers are caused by the gravitational lowering of outcropping or near-surface strata towards an adjacent valley. They occur where competent and permeable rocks overlie incompetent and impermeable beds such as clays. The competent beds develop a local dip towards the valleys, swathing the hill-tops and draping the valley sides (Hollingworth et al. 1944). The incompetent material is extruded from beneath the cap-rock, initially as a result of stress relief. Parks (1991) has suggested that as a camber develops, the competent cap-rock breaks up into joint-bounded blocks above a basal shear plane in the underlying material. A Quaternary cold stage with permafrost conditions is then required, since the underlying strata (if it is mudstone) is much more susceptible to creep when frozen. Thawing at the end of the glacial cycle increases the water content of the mudstone, potentially saturating it and drastically reducing its shear strength. This causes it to behave as a plastic fluid and the competent cap-rock migrates in the direction of slope, opening the joints to form gulls. The Parks model shows how gulls that are open to the surface can form in a thin cap-rock. In the Cotswold Hills, the limestone cap-rock is much more substantial and the gulls and gull caves generally have intact roofs. They have formed in the lower part of the limestone strata, with little or no mass movement having taken place in the upper part. This requires not only a basal shear plane, but also an upper parting/ sliding plane within the limestone sequence. A possible mechanism, involving the sequential unloading of joint-bounded blocks, was suggested by Self (1986). As extension occurred, individual blocks were able to settle slightly and then move laterally over the mudstone. The blocks move in the same direction but independently of each other, with neighbouring blocks supporting the overlying strata in turn, creating a gull network that propagates away from the valley. Other features seen in the caves and mines of the southern Cotswolds appear to contradict the Parks model. Significant camber angles are limited in extent, only affecting the strata closest to the hillside margins, whereas gulls continue to occur deep within the stone mines. A possible contributing factor is pyrite oxidation in the upper horizons of the Fullers Earth mudstone, with chemical leaching of the calcite cement greatly reducing its shear strength (Brown 1991) and allowing mass movement at very shallow camber Figure 1. Map of the Avon valley around Bath. The topography is based on a NextMap digital terrain model, with a 5 m spacing greyscale ramp. The inset map is a geological map of the Box area (based on the British Geological Survey digital geological 1: 50,000 scale data). Gull caves are marked with a star and numbered: 1. Bury Wood Camp [ST 8162 7397]. 2. Guys Rift [ST 8450 7372]. 3. The Rocks Rift [ST 7896 7057]. 4. Henrys Hole [ST 8360 6944]. 5. Murhill Rift [ST 7956 6073]. 6. Gully Wood Cave No. 5 [ST 7937 6600]. 7. Sallys Rift [ST 7941 6506]. 8. Gully Wood Cave No. 4 [ST 7946 6500]. 9. Gortons Rift. 10. Bathampton Down and Bath University. Detailed descriptions of the gull caves are in Self and Boycott (2000).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings133

PAGE 136

This zone of lateral extension is the result of an abrupt change of direction of the River Avon valley just north of the cave, resulting in mass movement in two different directions (Self 2008). The cave has several examples of fit features, whereby a ledge on one wall matches an overhang at the same level on the opposite wall. In both Sallys Rift and Gully Wood Cave No. 4, there are localized deposits of gravel containing Cretaceous flints with occasional clasts of Carboniferous rocks; these are identical to local superficial deposits preserved in rare patches on the plateau above the cave, identified by Donovan (1995) as an early Quaternary deposit.4. Gulls in Box Freestone MineThe Box Freestone Mine complex was worked mostly during the late 19thcentury using cranes and a network of horse drawn tramways. Over 60 km of passages are still accessible in an area of less than one square kilometre. Gull fissures are pervasive throughout the system and reach into the remotest parts, more than 600 m from the escarpment. The fissures range in width from a few centimetres up to gull-caves over a metre wide and many tens of metres long. Locally they can extend up to 10 m above the level of the mine floor. These larger gulls are sometimes passable, but many have been used for storage of waste stone. A detailed survey of the gull fissures was made in two separate parts of the mine, around Jacks Entrance and in the Delta Rectangle area (see Fig. 1), in an attempt to determine their spatial extent. In the Jacks Entrance area, the majority of gulls are aligned on the 150 joint set and are fairly regularly spaced around 3 m apart (Fig. 3). Thirty-five gulls were recorded along a 200 m transect due east from the entrance, showing an average extension of the strata of just over 5% along the length of the passage (Fig. 4). However, in the middle part of the transect, there is a zone with much larger gulls, which give an extension value of 9% for this part of the survey. These zones of extension were first reported by Hawkins and Privett (1981) on a angles. Mass movement in the study area seems to have been triggered by the rapid over-deepening of the valley system resulting from the capture by the River Avon of dip slope streams of the palaeo-Thames drainage. High pore pressures in the mudstone need not have been caused by permafrost, but could have been a result of groundwater percolation from above. The stability of the gulls since their creation (with the growth of speleothem deposits) suggests that there was only one major episode of mass movement. A possible explanation is that mass movement ended when the valleys cut down into the more competent Inferior Oolite limestone, which gave stability to the hillsides above.3. The gull cavesTypically, cave entrances in the southern Cotswolds are found in the cliff faces of abandoned small valley-side quarries (Self and Boycott 2000). Some are single fissures a few metres long, while others form more extensive systems. The main jointing directions are NWSE and NESW, so the more complex gull caves tend to be rectilinear networks. NS and EW joints are also present in the study area, but they generally have not been opened by mass movement. The known caves all occur on northor west-facing (up-dip) slopes, probably because such slopes are less prone to rotational failure. Many more caves certainly exist, but these either do not intersect the surface or are buried under colluvium. Two main groups of gull-caves occur (Fig. 1). In the By Brook valley, Guys Rift has 42 m of surveyed passages. Archaeological material in the cave includes the remains of four human adults, three children and pottery of early Iron Age. Downstream, Henrys Hole is a narrow gull which gives access to a small area of independent mine workings, close to the vast Box Freestone Mine complex. The mine intersects other gulls, but one mined passage crosses (with an intact roof) beneath a substantial gull cave, 4 m tall and 40 cm wide. This is evidence for a high-level disturbance within the Chalfield Oolite. In the neighbouring valley, the Rocks Rift is entirely filled with sediment, including calcite flowstone fragments and limestone boulders; it had been previously excavated for 36 m to create a garden folly. The largest gull caves occur in the Claverton Gorge where the River Avon cuts through the Cotswolds escarpment (locations 5 on Fig. 1). Gortons Rift is a very deep gull accessed from within a small stone mine. The rift is 24 m deep from the mine level and must reach almost to the lithological boundary with the underlying Fullers Earth mudstones. Several caves can be found on the east side of the gorge. Gully Wood Cave No. 5 is a network of 50 m, developed on several levels whilst Gully Wood Cave No 4 is a 90 m long contour-aligned gull with impressive passage dimensions (Figure 2). Murhill Rift comprises a major passage 80 m long linked to a complex network of rifts and passages onthree levels close to the hillside, giving a total length of over 300 m. The largest cave in the region is Sallys Rift (Fig. 2). It has three entrances and is 365 m long. The passages are typically over 10 m tall and between 20 cm and 50 cm wide; the largest passage lies closest to the hillside and is 15 m tall and 2 m wide. The cave is a series of contour-aligned gulls linked by gulls aligned along a camber spreading axis which runs NE, directly into the hill side. Figure 2. The Gully Wood gull-caves, including Sallys Rift.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings134

PAGE 137

building site in cambered lower Jurassic strata. Cambering is obvious along the western edge of the mine, reaching a maximum of 11 close to a collapsed former entrance. However, the camber angle decreases sharply a short distance into the mine and within 200 m the bedding becomes essentially horizontal. In the Jacks Entrance area, the nearby hillside is aligned approximately north-south. Ordinarily, this valley orientation might have caused both the NW and NE joint sets to open. However the regional dip of the strata, around 2 to the southeast, prevents this. Opening of the NE joint set would require cambering up-dip, whereas for the NW joint set cambering is in the much easier direction of strike of the strata. The measured camber directions are not in the direction of slope, which is to the west, but consistently have a dip component which explains the dominance of NW aligned gulls. The Delta Rectangle area lies almost directly beneath the summit of Box Hill, about 400 m from the nearest escarpment edge. Numerous gull fissures are present but they are more widely spaced than those seen in the southern part of the mine. The gulls are typically 5 to 20 cm wide and both joint sets have opened, with no clear direction of extension (Fig. 5). A 200 m east-west transect recorded seventeen gulls with ca. 1% extension of the strata along the length of the passage. Throughout the mine, the NW joint set is heavily corroded with extensive dissolutional fretting while the walls of the NE joint set are smooth. This suggests that preferred orientation of groundwater flow was in a NWSE direction.5. The development of the Claverton GorgeThe Chalfield Oolite is an important aquifer, but it is only weakly karstic and groundwater travels freely along the joints. On the south side of the By Brook and the east side of the Claverton gorge of the River Avon, dissolutional etching is very pronounced in the NWSE joint set (Self 1995). This etching of the joint walls is the result of slow groundwater movement which pre-dates the onset of cambering (and de-watering of the strata). Drainage could not have been down-dip to the south-east because the Forest Marble forms an impermeable cap-rock. The outlet was therefore up-dip. Up-dip springs contribute to the present drainage of the By Brook valley. The significance of the up-dip palaeodrainage of the NW joints in Sallys Rift is that the nearby River Avon flows in this direction. The Claverton Gorge (and the entire River Avon valley system upstream) could not have existed at this time, otherwise the groundwater movement would have been towards this valley using the conjugate joint set. This suggests that the proto-River Avon was an aggressive scarp Figure 3. Gulls in the Jacks Entrance area of the Box Freestone Mine. Figure 4. Transects through Box Freestone Mine: 1. Owen Bishop Road close to Jacks Entrance and the hillside (Jacks Entrance to the west); and 2. Delta Rectangle Passage in the interior of the mine beneath the interfluve. Two zones of extension are apparent on Owen Bishop Road in the Jacks Entrance area.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings135

PAGE 138

Figure 5. Gulls in the Delta rectangle area of the Box Freestone Mine.stream tributary to the By Brook, while the Cotswold dip slope drainage was originally east to the River Thames. Eventually, head-ward erosion by the River Avon allowed it to break through into dip-slope territory and capture these former Thames headstreams. With a greatly enhanced flow, the River Avon rapidly over-deepened its valley and formed the steep-sided Claverton Gorge. The tributary valleys also cut down to this new base level and the entire valley system was primed for cambering.ReferencesBarron AJM, Sheppard TH, Gallois RW, Hobbs PRN, Smith NJP, 2011. Geology of the Bath district: a brief explanation of the geological map sheet 265 Bath. Nottingham, UK, British Geological Survey, 35. Brown CJ, 1991. A geotechnical study of abandoned mineworkings in the Bath area. Ph.D. thesis, Bristol Univ. UK. Donovan DT, 1995. High level drift deposits east of Bath. Proceedings of the University of Bristol Spelaeological Society, 20, 109. Fitton WH, 1836. Observations on some of the strata between the Chalk and the Oxford Oolite in the South-East of England. Transactions of the Geological Society, 2(4), 103. Hawkins AB, Privett KD, 1981. A building site on cambered ground at Radstock, Avon. Quarterly Journal of Engineering Geology, London, 14, 151. Hollingworth SE, Taylor JH, Kellaway GA, 1944. Large scale superficial structures in the Northampton Ironstone Field. Quarterly Journal of the Geological Society, 99, 1. Parks CD, 1991. A review of the mechanisms of cambering and valley bulging. In: A Forster, MG Culshaw, JC Cripps, JA Little, CF Moon (Eds). Quaternary EngineeringGeology. Geological Society, London, 7, 373. Self CA, 1986 (for 1985). Two gull caves from the Wiltshire/Avon border. Proceedings of the University of Bristol Spelaeological Society, 17, 153 Self CA, 1995. The relationship between the gull cave Sallys Rift and the development of the River Avon east of Bath. Proceedings of the University of Bristol Spelaeological Society, 20, 91. Self CA, 2008 (for 2007). Cave passages formed by a newly recognized type of mass movement: a gull tear. Proceedings of the University of Bristol Spelaeological Society, 24, 101. Self CA, Boycott A, 2000 (for 1999). Landslip caves of the Southern Cotswolds.Proceedingsof the University of Bristol Spelaeological Society, 21, 197.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings136

PAGE 139

THE ROLE OF FOLD-AND-THRUST STRUCTURE IN THE LARGE SHAFTS AND CHAMBERS DEVELOPMENT: CASE STUDY OF THE POLISH TATRA MTS.Jacek Szczygie Faculty of Earth Science University of Silesia, B dziska 60, 41-200 Sosnowiec, Poland, j_szczygiel@tlen.pl The Tatra Mts. are characterized by a geological structure typical to fold-and-thrust alpine belt. This paper focuses on four caves belonging to the largest in the Tatra Mts.: the Ptasia Studnia System, the Ma a w Muowej Cave, the Kozia Cave, the niena Studnia Cave. The caves are developed in the Czerwone Wierchy Unit. A deep shaft or a series of shafts and a large chamber characterize each of these caves. All these large underground features are located within hinge zones of the main synclines of the dziary or Organy Units (the sub-units of the Czerwone Wierchy Unit). The Organy Syncline has chevron geometry, the dziary Syncline is characterized by a concentric geometry. Large chambers are formed in hinge zones of chevron folds and in the hinge zones closer to the core of concentric folds. A strong compression in this zone has resulted in a large number of sub-structures (folds, faults, fractures). This zone is relatively narrow which results in the occurence of forms of similar horizontal and vertical dimensions. Chambers are characterized by breakdown features. The extent of chamber is correlated to the strike of strata and faults perpendicular to beds. Deep shafts are developed in the outer part of the concentric folds. The shafts based on long steep stratification surface loosen and dislocated each other by flexular slip, which is an effect of the local extension zone.1. IntroductionThe relationships between the geological structure of the massif and the development of karst features discussed by Grodzicki (1970), Hauselmann et al. (1999), Klimchouk and Ford (2000), Tognini and Bini (2001) and others. The shape and direction of cave passages, as well as volume and extent depend on litology, tectonic and geometry of specific structures. The geometry and extent of specific structures are particularly important in case of a fold-and-thrust orogens. Groundwater flows by the most convenient and the easiest way. Overthrusts and folds can either facilitate waterflow or prevent it (Goldscheider 2005). Presented examples from the Tatra Mts. Illustrate the controling role of the structures geometry of the major folds on the large shafts and chambers.2. Study areaThe Tatra Mts. are the northernmost part of the Central Western Carpathians, they are composed of a crystalline core and are overcast by autochthonous sedimentary cover, Hight-Tatric Nappe (Czerwone Wierchy Unit, Giewont Unit) and Sub-Tatric Nappes (Krina and Cho nappes; Fig. 1A). Approximately 1,200 caves have been described in the Tatras so far, of which just over 800 are in Poland (Gradziski et al. 2009). Karst in Polish Tatra Mts. occurs mainly in the Czerwone Wierchy Massif. In this paper the presented data comes from the four caves located in the N part of the Czerwone Wierchy Massif. These are Ptasia Studnia System, Ma a w Mu owej Cave, Kozia Cave, nie na Studnia Cave. Large forms of caverns have developed in those caves. 2.1. Geological setting The caves are located in the Czerwone Wierchy Unit (Fig. 1B) which consists of a sedimentary sequence from Triassic to Lower Cretaceous. Simplified profile of Mesozoic succession is (based on Kota ski 1963; Lefeld et al. 1985; Piotrowska et al. 2009): Early Triassic Campilian limestone and dolomite with shale interbeds (the Myophoria Beds); Middle Triassic limestone and dolomite with bioturbate beds; Middle Jurassic Doggerian crinoida limestone (Smolegowa Formation Bajocian), red nodular limestone (Krupianka Formation Batonian); Late Jurassic to Lower Cretaceous (Hauterivian) MalmoNeokomnian thick-bedded limestone (Raptawicka Turnia Formation); Lower Cretaceous (Barremian, Aptian) Urgonian organodetritic thick-bedded limestone (Wysoka Turnia Formation) Lower Cretaceous (Albian-Cenomanian) green marly shale, (Zabijak Formation) The Czerwone Wierchy Unit is composed of two sub-units: more northern Organy and lower southern dziary, separated by Organy Fault extending latitudinally along the major unit, from the Ko cieliska Valley to Ma a ka Valley (Kotaski 1963). In spite of numerous studies the character of the Organy and dziary units is not clearly defined. Kota ski (1963, 1965) and Bac et al. (1984) presented these units as a syncline. Grodzicki (1978) based on measurements of the caves ascertained that the Organy Unit is made up of two Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings137

PAGE 140

elements, the lower Up az Mietusi Anticline and the upper Organy Syncline. Then Grodzicki and Karda (1989) based on geological measurements from the caves interpreted geological structure as a duplex style. Jurewicz (2005) confirmed that the Tatra nappes have been formed in a manner similar to duplexes, but also showed a much more complicated history and dynamics of the Tatra massif proving primary southern inclined of the major overthrust plane. 2.2. Characteristic of caves The Ptasia Studnia System has 3 entrances: Ptasia Studnia (1,627 m a.s.l.), Lodowa Litworowa Cave (1,576 m a.s.l.) and Nad Dachem Cave (1,522 m a.s.l.). All within the N wall of the Kozi Ridge (Fig. 1B). The cave is a network of passages of different nature and origin: phreatic pipe sometimes remodeled by vadose meanders, tectonic conduit, deep shafts of probably invasive (vadose). On the 1,480 m a.s.l. Dante Chamber is located, one of the largest cave rooms in the Tatras measuring approximately 12 35 m and 50 m height. The cave has a length of 6,000 m and 352m of denivelation (Lorczyk 2010). The Kozia Cave has two entrances located in Kozi Ridge (Fig. 1B) at 1,850 m a.s.l. The cave consists of two parallel main conduits with many branches. One of the meanders at a depth of approximately 245 m goes into aven type, reaching a depth of -389 m. The second conduit at the upper part continues as a tectonic corridor. Below a depth of 100 m is a typical vadose passage and reaches a depth of -263 m. The connective passages of two conduits are a level at a depth of approximately 100 m. Kozia Cave is 3,470 m long (Wi niewski and Kotarba 2010). The Ma a w Mu owej Cave is located at NE slope of the Ciemniak (Fig. 1B). Entrance is situated at 1,757 m a.s.l. Ma a w Mu owej Cave is 555 m deep and 3,863 m long (Antkiewicz and Lorczyk 2010). The cave has two main conduits which separate at a depth 60 m. The first conduit is a vertical type and contain the biggest chamber in Tatra Mts. of dimensions 85 35 90 m. The second conduit is of Aven type too, but only to 300 m deep. At a depth of -160 m begins the Czesanka Shaft with a depth of 130 m. Further passages run 800 m westwards in a straight line to a depth of 555 m. The niena Studnia cave is known only by one entrance. It is located at a height of 1,747 m a.s.l. at the NE slope of the Maoczniak (Fig. 1B). In the cave there are two basic types of corridors. Sub horizontal corridors of phreatic or Figure 1. Location of study area; A Main tectonic units of the Polish Tatra Mts.(after Kota ski 1963); B Schematic tectonic map of the study area with caves location (after Piotrowska et al. 2009); 1 Ma a w Muowej Cave; 2 Ptasia Studnia Cave; 3 Lodowa Litworowa Cave; 4 Nad Dachem Cave; 5 and 6 Kozia Cave; 7 niena Studnia; I Fakro Chamber; II Czesank Shaft; III Large Shaft; IV Dante Chamber; V Chamber Under Colossal Boulder; VI chamber with Wielki K amca lake; VII Wazeliniarzy Shaft.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings138

PAGE 141

tectonic origin, with the general course W-E and the extent of 900 m. Old inactive conduits are developed between 1,350 and 1,200 m a.s.l., the known active conduit (zone of sump) extends from the 1,100 m to 1,030 m asl. The second type is vertical. Within parts from the entrance to the level -500 the deepest shaft is located in Tatra Mts. Wazeliniarzy Shaft (200 m deep). The cave has 763 m of denivelation and length of 12,050 m (Fuja et al. 2010).3. MethodsFieldwork has covered measurements of bedding planes, fractures, faults and tectogliphs. Observations of relationship between measured structures and the course of the cave have been made. Rock samples for stratigraphic position of studied units were collected. All data are presented as structural plans of caves and geological crosssections. Structural studies were carried out on the basis of the detailed structural analysis in order to recognize geometric, kinematic and dynamic features of minor tectonic deformation structures represented by: joints, mesofaults (with accompanying minor structures on the fault surfaces and shears) and veins. The studies were based mainly on the geometric analysis of various mesoscale tectonic structures: their morphology, spatial orientations and superposition. Collected data were summarized in the form of structural diagrams (great circles and contours) made in equal-area Lambert-Schmidt projection on the lower hemisphere using SpheriStat software. Rose diagrams presenting strike directions and dip angles of studied structures were prepared in TectonicsFP software.4. Results of geological observations4.1. Ptasia Studnia System The Ptasia Studnia System has been investigated by Grodzicki and Karda (1989). They have recognized at the Organy Fault plane depth of 15 m below Ptasia Studnia Cave entrance. Below documented hinge of the Organy Syncline around 1,450 m a.s.l. Detailed revision confirms the presence of overthrust separates the two units at a depth of approximately 15 m. In the depth the cave develops mainly in Middle Traissic limestone and dolomites. The Nad Dachem Cave developed in limestone of the Raptawicka Turnia Formation up to 1,450 m a.s.l. where there is the boundary between the limestone of the Raptawicka Turnia Formation, Middle Jurassic limestone and Middle Triassic rocks (Fig. 2). Beds in the inverted limb of the Organy Syncline oscillate from 55to 35, locally to 20. The dip direction of bedding is S and SSE less frequently SSW. In the lower limb of the syncline the dip of bedding are between 55and 3to NE direction (Fig. 2). Axial surface orientation is 109/22. The syncline is a chevron type fold with gently inclined axial surface. Dominated joints in the cave are orientated NNESSW, very steeply occurs The NESW-trending and steep re slightly less frequent. There are also sub-latitudinal and NNWSSE trending fractures (Fig. 2). Comparison of fracture diagrams of the upper and lower fold limbs shows similarities, which proves the formation of these fractures after the folding. Three faults are documented in the cave. They are in the Canyon (160/70, about 0.5 m dip separation, normal fault; Fig. 2), in the Dante Chamber (210/50), which probably extends to the Chamber Under Colossal Boulder, and in the chamber with Wielki K amca lake (120/80). 4.2. Kozia Cave Geology of the Kozia Cave has been generally described by Grodzicki and Karda (1989). They recognized the cave developed in inverted series of the Middle Triassic limestone and dolomites. They described that beds are inclined to S and dip increase with depth, and the cave developed in the vertical joints. The research shows that in the western conduit, dip of beds to a depth of approximately 140 m oscillates between 40 and 25to the SSE, and then grow to a depth of 245 m where achieves 60. Here, the cave changes the charater of meander to aven (shafts serie; Fig. 3), and dip further increase to 75to the SSE and at the bottom reaches 85to NNW. In the eastern conduit the dip of bedding is between 32 and 45to the SSE, and at a depth of about 200 m grows to 60. At the bottom dip is 65to the SSE. The cave is dominated by very steep NE SW striking joints. This set can be identified with the same set occurring in the Ptasia Studnia System which is younger than folds. Meridional steeply fractures, very steep latitudinal fractures and of WNWESE-oriented fractures are of similar participation in the cave. Meridional and WNWESE trending fractures often occur together (Fig.3). Figure 2. Geological cross-section of the Ptasia Studnia System; upper contour diagram is projection of stratification in the cave, lower is showing fracture; line of the cross-section on the Figure 1B.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings139

PAGE 142

An angle of 60between them and flat fractures plane can point that this fractures are complementary shears system. 1 direction (about 335) which bisects 60angle is the same as the direction of the main tension of folding. At the depth of -160 to -190 a series of parallel faults of the two cave conduits has been found. Faults planes are gently eastwards inclined (15). In the W conduit observed drag indicates left-rotation. In the E conduit the fault is rejuvenated after the cave development becouse the passages are dislocated by right-rotation. The presence of this fault is correlated with interpretative geological crosssection through Krzesanica by Bac-Moszaszwili and Nowicki (2006). 4.3. Maa w Muowej Cave The geological setting of the Ma a w Mu owej cave has been described by Bac-Moszaszwili and Nowicki (2006). Their conclusions were based on the surface studies. Preliminary and incomplete studies have been carried out by Recielski and next interpreted by Grodzicki (2008). The Ma a w Mu owej Cave developed in the Middle Triassic limestone and dolomites banding up to 300 m of depth. The beds are arranged in concentric brachysyncline with sub-horizontal axes and axial surface inclined 37 toward N. The upper limb is oriented middling 170/45, the lower 5/65 (Fig. 4). This characteristic is not similar to the interpretations of Bac et al. (1984), Grodzicki and Karda (1989), Bac-Moszaszwili and Nowicki (2006). The closest is the interpretation of Kota ski (1965). About 300 m under the cave entrance passages are formed mainly at the contact between the autochthonous High Tatric sedimentary cover and the dziary Unit (Fig. 4). The first Unit consists of calcareous shale of the Zabijak Formation and limestone of the Wysoka Turnia Formation, the second one is formed by the Middle Triassic limestone and dolomites and the Early Triassic limestone with shale interbeds. The contact is tectonic (overthrust), the surface is inclined at an angle of 52 to 83toward the SW. Shales are oriented at contact consequently or rarely insequently. Triassic rocks contact the overthrust plan insequently, inclined 20 toward N. The contact is dislocated by strike-slip faults perpendicular to them. 4.4. niena Studnia Cave Geological research in the nie na Studnia Cave was not undertaken before. The geological structure of the cave was outlined in Bac-Moszaszwili and Nowicki (2006). They interpreted the surface data. Fieldwork in the cave still needs to be completed. Measurements were made in the main conduit to the bottom and in horizontal passages. It is expected to work further below the horizon -500. Paper presents preliminary interpretation of the data obtained so far. However, the main feature of this study is the Wazeliniarzy Shaft which is included in the research. Figure 3. Geological cross-section of the Kozia Cave; upper contour diagram is projection of stratification in the cave, lower is showing fracture; line of the cross-section on the Figure 1B; legend at the Figure 2. Figure 4.Geological cross-section of the Ma a w Muowej Cave; upper contour diagram is projection of stratification in the cave, lower is showing fracture; line of the cross-section on the Figure 1B; rest of the legend at the Figure 2.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings140

PAGE 143

Bac-Moszaszwili and Nowicki (2006) presented that the nie na Studnia cave, to 1,350 m a.s.l., developed into a inverted series of Triassic and Jurassic rocks, doubled by a fault. The vertical part of the cave, included the Wazeliniarzy Shaft, developed in the limestone of the Raptawicka Turnia Formation belonging to dziary Unit. The horizontal passages at the depth of -500 are developed along the Organy Fault and parts under -500 m are developed in the Triassic rocks of the Organy Unit. However, this cross-section was set too far to the west. The fieldwork shows that the layers are inverted. In the cave entrance the Middle Triassic limestone are inclined at an angle of 35toward the SE. At a depth of 60 m there is the boundary with Bajocian limestones, a few meters below the contact with the limestone of Raptawicka Turnia Formation. At a depth of 150 m the fault is observed, most likely the Organy Fault. Fault plane inclined 40toward S. Limestone interbeded by greenish shales was found, which probably are the Lower Triassic. Just below the cave is developed in banding Middle Triassic limestone inclined 50southward (Fig.5). From a depth of 200 m the cave is developed in the fault zone. Fault plane is initiated at bedding planes. Dip of fault plane increases from 55to 75to the S and SSE up to a depth of 270 m. Dislocation has been identified in the cave from the western to the eastern limits. The fault plane is mineralized and tectogliphs indicate normal dextral movement. This fault was rejuvenated in the recent past because it dislocated the passages, dip separation is equal to 22 cm. Apart the main dislocation, discontinuities in the cave in the upper part especially can be seen, where stratification is not too steep, the cave is formed rather along the cracks. Dominant fractures are NESW, NWSE, WNWESE mainly by steep and very steep planes (Fig. 5).5. Relations between large shafts and chambers and main foldsLarge volume features in the Ptasia Studnia System are presented at the depth of -100 to -160 especially. From Dante Chamber to the West to the chamber with Wielki K amca Lake to the East. The whole zone has cracked character. In the above-mentioned interval hinge zone of the Organy Syncline is located. Breakdown parts (bottom of the Dante Chamber, the Chamber Under Colossal Boulder, the Shaft with the Pirates Bridge). The Wielki K amca Lake covers the entire surface of the hall so it is hard to assess clearly the nature of the chamber bottom. However, the irregular shape of the plan of the hall and observations from the walls show a significant impact of the tectonics on chamber development. Cave parts between the Dante Chamber to the Chamber Under Colossal Boulder are probably partly conditioned by discontinuity 210/50, which direction correlates with the direction of the cave and a prominent fault visible at the roof of the Chamber Under Colossal Boulder. Below the depth of 254 m in the Kozia Cave the cave changes the nature from meander to aven, which correlates with increasing dip of layers. Caves along their whole length allude strongly to stratification, regardless of the type of passages (vadose or tectonic). Water penetrating the Kozia Cave used the conjugation of stratification and fractures. The bedding plane is influencing on inclination of passage, joints and shears on their direction. However, conduit from -245 m to the bottom (-389) is developed on the steep and vertical stratification, the fractures receded into the background. In this lower part the stratification determines not only inclination but also direction of passages. Fieldwork confirms the geological interpretation of Bac-Moszaszwili and Nowicki (2006). Thus it can be concluded that the increasingly steep layers are associated with the main syncline of the dziary, and its open interlimb angle is the result of the cave course in the outer part of the concentric fold. In the Maa w Muowej Cave the large volume parts, as the Fakro Chamber and the Czesanka Shaft, are located at the altitude interval of 1,650,470 m, where hinge zone of the dziary Syncline is located. The Fakro Chamber is developed in the N part closer to the core of the fold. This chamber is a breakdown type. Above the hall and on the walls minor folds are very visible, some of which are dislocated. The Czesanka Shaft is located in southern outer part of syncline. The shaft is developed at very steep and vertical bedding plan. Both forms have a latitudinal extent, referring to the strike of layers. Figure 5. Geological cross-section of the niena Studnia Cave; upper contour diagram is projection of stratification in the cave, lower is showing fracture; line of the cross-section on the Figure1B; legend at the Figures 2 and 4.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings141

PAGE 144

The deepest shaft in the Tatra Mts. the Wazeliniarzy shaft, which is 200 m deep, is located in the nie na Studnia Cave. Comparably to the Kozia Cave, it is associated with increasingly steep layers, which in this case were also moved relative to each other. Similarly as in the Kozia Cave gentle change of dip is likely due to the location of the cave in the outer zone of the fold. The direction of surface displacement of stratification are consistent with flexural slip (Fig. 6). All of these forms are located within the hinge zones. During folding sub-structures were formed in the manner that could have a major impact on the relationship discussed. Within the complex of folding rocks the highest compression zone and the local zone of extension can be distinguish. This zone is located closer to the core of the fold. Local zone of extension is located in the outer zone (Fig. 6). Zones are separated by a neautral surface (Ramsey and Huber 1987). In the contraction zone were developed faults and minor folds. In the extension zone were formed fractures perpendicular to strata. Flexural slip is a common phenomenon during folding (Ramsey and Huber 1987; Fig.6). It seems that in the case discussed above, in thin limestone layer, all types of movement relative to one to another is an easy way to discharge stress rocks through the use of existing surfaces.6. ConclusionsPresented folds are gently inclined or moderatly of an axial surface and interlimb angles are open or wight. The Organy Syncline in which the Ptasia Studnia System has evolved have chevron geometries and regardless of distance from the core, inclined of individual packages of rocks do not change. While the dziary Syncline has concentric geometries where closer to the core dip of layers is changing rapidly, it is a significant number of folds associated by minor faults and high density fractures. In the outer zone, layer on a very long stretch are very steeply and continuous of layers is interrupted by extensions fractures. The conducive role of hinge zone to the development of karst forms has been described by Tognini and Bini (2001). However, theirs research discussed the standing folds conditions where the cave galleries formed in extension zones and neutral surface is something of a barrier to karst processes. The geological situation of the Czerwone Wierchy Unit leads to the conclusion that the whole hinge zone will favour development of large chambers and deep shafts. Figure 6. Schematic of a minor deformations in major fold (based on Ramsey and Huber 1987, changed).Large chambers are formed in hinge zones of chevron folds and in the hinge zones closer to the core of concentric folds. Relatively large concentration of deformations in small vertical distance induces development of forms of similar horizontal and vertical dimensions. Chambers are characterized by brakdown feature. The extent of rooms is linked to the strike of layers and faults which are perpendicular to bedding. Deep shafts are developed at the outer part of the concentric folds. Shafts are based on long, very steep stratification surface which are loosen and dislocated each other by flexular slip.AcknowledgmentsStudies in caves would not be possible without permission of the authority of the Tatra National Park. The research has been funded by Grant for young scientist at the Faculty of Earth Sciences, University of Silesia.ReferencesAntkiewicz A, Lorczyk M. 2010. Jaskinia Ma a w Mu owej. Portal CBDG http://geoportal.pgi.gov.pl/jaskinie-pub/jaskinie (in Polish). Bac-Moszaszwili M, Jaroszewski W, Passendorfer E, 1984. On the tectonics of Czerwone Wierchy and Giewont area in the Tatra Mts. (Poland). Annales Societatis Geologorum Poloniae, 52(1/4), 67 (in Polish). Bac-Moszaszwili M, Nowicki T, 2006. Remarks on caves development in the Czerwone Wierchy Nappe in the Tatra Mts.. Przegld geologiczny, 54(1), 56 (in Polish). Fuja D, Filar F, Albrzykowski G, 2010. Jaskinia niena Studnia. Portal CBDG, http://geoportal.pgi.gov.pl/jaskinie-pub/jaskinie (in Polish). Goldscheider N. 2005. Fold structure and underground drainage pattern in the alpine karst system Hochifen-Gottesacker. Eclogae Geologicae Helvetiae, 98, 1. Gradzi ski M, Hercman H, Kici ska D, Barczyk G, Bella P, Holbek P, 2009. Karst in the Tatra Mountainsdevelopments of knowledge in the last thirty years. Przegl d Geologiczny, 57(8), 674 (in Polish). Grodzicki J, 1970. Rola tektoniki w genezie jaski masywu Czerwonych Wierchw. Speologia, 5, 33 (in Polish). Grodzicki J,1978. Nowe elementy strukturalne jednostki Organw mi dzy Dolin Ko cielisk i Dolin Mi tusi Kras i Speleologia, 2, 77 (in Polish). Grodzicki J, 2008. Remarks on caves development in the Czerwone Wierchy Nappe in the Tatra Mts.-how to fabrykowac fakty by potwierdzi w asna teorie. Proc. of 42. Speleological Symp. Tarnowskie Gry, Poland, 57. Grodzicki J, Karda R 1989. Tectonics of the Czerwone Wierchy (Massif Tatra Mts.) in the light on observations in caves. Annales Societatis Geologorum Poloniae, 59, 275. Huselmann P, Jeannin PY, Bitterli T, 1999 Reltionships between karst and tectonics: case-study of the cave system North of Lake Tun (Bern, Switzerland). Geodinamica Acta, 12(6), 377.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings142

PAGE 145

Jurewicz E, 2005. Geodynamic evolution of the Tatra Mts. and the Pieniny Klippen Belt (Western Carpathians): problems and comments. Acta Geologica Polonica, 55, 295. Klimchouk A, Ford D, 2000. Lithologic and structural controls of dissolutional cave development. In: AB Klimchouk, DC Ford, AN Palmer, W Dreybrodt (Eds.). Speleogenesis. Evolution of Karst Aquifers. National Speleological Society, Huntsville, Alabama, 54. Kotaski Z, 1963. Nowe elementy budowy masywu Czerwonych Wierchw. Acta Geologica Polonica, 13, 149 (in Polish). Lefeld J, Ga dzicki A, Iwanow A, Krajewski K, Wjcik K, 1985. Jurassic and Cretaceous lithostratigraphic units of the Tatra Mountains. Studia Geologica Polonica, 84, 1. Lorczyk M. 2010. Ptasia Studnia-Jaskinia Lodowa LitworowaJaskinia Nad Dachem. Portal CBDG; http://geoportal.pgi.gov.pl/jaskinie-pub/jaskinie (in Polish). Piotrowska K, Cymerman Z, R czkowski W, 2009. Detailed geological map of the Tatra Mts. in scale 1:10,000 sheet Czerwone. National Geological Institut-Polish Reserch Institut. Warszawa. Ramsay JG, Huber MI, 1987. The technigues of modern structural geology. Vol. 2:Folds and Fractures. Academic Press. London. Tognini P, Bini A, 2001. Efects of structural setting endokarst system geometry in the Valle del Nose (Como Lake, Northern Italy). Geologica Belgica, 4(3/4), 1971. Wi niewski WW, Kotarba S, 2010. Jaskinia Kozia. Portal CBDG, http://geoportal.pgi.gov.pl/jaskinie-pub/jaskinie (in Polish).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings143

PAGE 146

HYPOGENIC CAVES OF SICILY (SOUTHERN ITALY)Marco Vattano1, Philippe Audra2, Fabrizio Benvenuto3, Jean-Yves Bigot4, Jo De Waele3, Ermanno Galli5, Giuliana Madonia1, Jean-Claude Nobcourt6 1Department of Earth and Sea Sciences, University of Palermo, Via Archirafi 22, 90123 Palermo, Italy, marco.vattano@unipa.it, giuliana.madonia@unipa.it2PolytechNice-Sophia, Engineering School of Nice Sophia Antipolis University, & ICiTy (IMREDD), 930 route des Colles, 06903 Sophia-Antipolis, France, audra@unice.fr3Department of Biological, Geological and Environmental Sciences, Bologna University, Via Zamboni 67, 40126 Bologna, Italy, fabrizio.benvenuto2@studio.unibo.it; jo.dewaele@unibo.it4Association Franaise de Karstologie (AFK), 21 rue des Hospices, 34090 Montpellier, France, catherine.arnoux@club-internet.fr5Department of Earth Sciences, University of Modena and Reggio Emilia, Largo S. Eufemia 19, 41121 Modena, Italy, gallier@unimore.it6CRESPE, Le Hameau de lAra, 259 Bd Reine Jeanne, 06140 Vence, France, jcnobecourt@free.fr First results of a study on hypogenic caves in Sicily are presented. Inactive water-table sulphuric acid caves and 3D maze caves linked to rising of thermal waters rich in H2S were recognized. Cave patterns are guided by structural planes, medium and small scale morphological features are due mainly to condensation-corrosion processes. Calcite and gypsum represent the most common cave minerals. Different types of phosphates linked to the presence of large bat guano deposits were analyzed.1. IntroductionHypogenic caves are recognised as generated by water recharging from below independently of seepage from the overlying or immediately adjacent surface. Waters are often thermal and/or enriched in dissolved gases, the most common are CO2and H2S. Hypogenic caves can be thermal caves, sulphuric acid caves, and basal injection caves. They differ from epigenic caves in many ways, such as: speleogenetic mechanisms, morphological features, chemical deposits, and lack of alluvial sediments (Klimchouk 2007; Klimchouk and Ford 2009; Palmer 2011). Several studies were conducted to evaluate the hypogenic origin of a large number of caves (Klimchouk and Ford 2009; Stafford et al. 2009; Audra et al. 2010; Plan et al. 2012; Tisato et al. 2012). A significant contribution was given by the work of Klimchouk (2007) that systematically provided instruments and models to better understand and well define the hypogenic karst processes and landforms. Studies on hypogenic caves in Italy were carried out since the 90s in different karst systems, especially in the central and southern Appenines. These studies mainly concerned chemical deposits related to sulphuric ascending water and micro-biological action (Galdenzi and Menichetti 1995; Galdenzi 1997; Piccini 2000; Galdenzi and Maruoka 2003; Forti and Mocchiutti 2004; Galdenzi 2012; Tisato et al. 2012). This paper aims to describe preliminary the first results of a study conducted in some hypogenic caves in Sicily, highlighting their main features such as pattern, morphology, mineralogy and speleogenesis. These are Monte Inici karst system and Acqua Fitusa Cave. For this purpose topographic and geomorphological surveys were carried out, and about 40 samples of cave minerals were taken in different parts of the caves. In addition, a brief description of Monte Kronio karst system is given. It represents one of the most important hypogenic karst systems of Sicily, characterized by flows of hot air and vapor rising from below.2. Karst in SicilyKarst in Sicily is widespread and exhibits a great variety of surface and underground landforms related to the wide distribution of soluble rocks (Di Maggio et al. 2012). About 20% (more than 6,000 km2) of the land area consists of carbonates and evaporites, primarily gypsum (Fig. 1). Carbonate karst lies mainly in the northwestern and central sectors of the Apennine chain and the foreland area in southeastern Sicily; gypsum karst occurs chiefly in the central and southern areas of the island, though evaporite landscapes are also present in the northern and western parts of Sicily. Carbonate and gypsum karst systems develop under unconfined conditions and in most cases constitute epigenetic systems fed by meteoric waters. Hypogene caves are located only in carbonate rocks and are linked to the presence of deep thermal waters. Figure 1. Gypsum (white) and carbonate karst areas (grey) and localization of investigated hypogenic karst systems. 1. Monte Inici; 2. Acqua Fitusa. 3. Monte Kronio.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings144

PAGE 147

3. Hypogenic caves3.1. Acqua Fitusa Cave Acqua Fitusa Cave opens in the eastern section of the Sicani Mounts (west-central Sicily), along the north-eastern fault scarp of a N-S anticline, westward vergent, forming the Mt. La Montagnola (Fig. 1). The cave formed in the Upper Cretaceous Rudist breccias member of the Crisanti Fm., composed of conglomerates and reworked calcarenites with rudist fragments and benthic foraminifera (Catalano et al. 2011). At present it is inactive with a thermal spring occurring 300 m north and 30 m below the cave. The waters have a temperature of about 25 C, and are indicated as chlorinesulphate alkaline (Grassa et al. 2006 and references therein). During the spring-summer-early autumn the cavity hosts a large colony of bats, including Myotis myotis and Miniopterus schreibersii species (Mucedda pers. comm.), that produce significant amounts of guano. The first explorations of Acqua Fitusa were carried out in the early XXthcentury by some inhabitants of the neighboring villages, but the first human frequentations of the cavity have to go back to the Paleolithic and Chalcolithic periods, as evidenced by the discovery of numerous lithic fragments, remains of food and burials (Bianchini and Gambassini 1973). Lombardo et al. in 2007 gave a description of the cave and some studies concerned the hydrogeochemistry and isotopic composition of the nearby spring waters (Grassa et al. 2006 and references therein). According to the survey made in 2011, the cave consists in at least three stories of subhorizontal conduits, displaying a total length of 700 m, and a vertical range of 25 m (Fig. 2). The main passages are generally low and narrow and follow sets of joints oriented in ENE-WSW, E-W and N-S directions. Very small passages develop from these galleries making incipient mazes. The conduits breach the fault scarp in more points at different heights. The main entrance leads to a large chamber enlarged both by corrosion and breakdown processes, connected to an upper subhorizontal passage that likely formed during a past higher karst base level. Acqua Fitusa Cave represents a clear example of inactive water table sulphuric acid cave (Audra et al. 2009). Despite the small size, the cave is very interesting for the abundance and variety of morphologies and deposits formed at and above the water table where H2S degassing and thermal convection produced strong condensation-corrosion processes. The floor of some passages is breached for several meters by an inactive thermo-sulphuric discharge slot that can reach a depth of 7 m (Fig. 3). In some sections of the caves, notches with flat roof, linked to lateral corrosion of a water table with concentrated sulphuric acid, carve the walls at different heights, recording past stages of base-level change. Several forms of small and large sizes, generated by condensation-corrosion processes above the water table, can be observed along the ceiling and walls. Ceiling cupolas and large wall convection niches occur in the largest rooms of the cave. Here and in the upper gallery, pendants at junctions of more cupolas or between braided channels are widespread. Cave walls of many passages are carved at different heights by deep wall convection niches that in places form notches (Fig. 3). Condensation-corrosion channels similar to ceiling-half tubes carve the roof of some passages; replacements pockets due to corrosionsubstitution processes are widespread; boxwork created by differential condensation-corrosion were observed in the upper parts of the conduits. Figure 2. Acqua Fitusa Cave: plan view and extended profile (Survey: Ceresia, Inzerillo, Provenzano, Sausa, Scrima and Vattano, 2011). Figure 3. Acqua Fitusa. Passage with discharge slot at the floor and different levels of wall convection niches.The most abundant cave mineral is gypsum which displays different shapes and colours. Replacement gypsum crusts are common in many passages; the gypsum is located in large vertical fissures along the walls, it can partially cover wall convection notches, or replacement pockets. A gypsum body of about 50 cm of thickness was found on the floor of the biggest room in correspondence of which small ceiling cupolas and pendants are associated on the roof. Centimeter-sized euhedral gypsum crystals grew inside mud sediments. Finally, the walls of a feeding fissure are covered with a network of gypsum roots in which the Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings145

PAGE 148

3.2.1. Abisso dei Cocci Abisso dei Cocci is an inactive 3D maze cave formed in Lower Jurassic limestones and dolomitic limestones (Inici Fm.), arranged in decimetric westward dipping beds. The cave is one of the largest of Sicily, reaching a total length of over 2,053 m and a vertical range of -300/+61m. It consists of several stories of large subhorizontal galleries and chambers connected by deep shafts. In plan view the galleries are mainly oriented in NW-SE and NE-SW direction, parallel with the main tectonic discontinuity lines. Some conduits are gently inclined and follow the dip of bedding planes, whereas the shafts develop along vertical fissures or fault planes. Passages display sub-circular crosssections, sometimes with vadose entrenchments (Fig. 5). Several features linked to rising thermal water and air flow were recognized in many sections of the cave. Many passages are characterized by large convection wall niches, mega-scallops, ceiling cupolas, and ceiling spheres. In some cases adjacent passages are separated by partitions (Fig. 5). The cave lacks alluvial sediments; on the other hand chemical deposits are abundant mainly in the middle level of the cave, where dripping is still active. Here, a large variety of calcite speleothems, such as stalactites, stalagmites, flowstones, shelfstones, etc., occur. Calcite is present also in the shape of powder, thin crusts, and frostwork. Gypsum was recognized in the middle and deepest sections of the cave in the form of tabular or acicular colourless small crystals (Fig. 5). biological control is obvious. Further investigation on these apparently subaqueous gypsum speleothems is still ongoing. Phosphate minerals, such as apatite [Ca5(PO4)3(OH)], were found in the form of thin crusts near large deposits of bat guano. 3.2. Monte Inici Complex Monte Inici karst system is situated in northwestern Sicily (Fig. 1), in the eastern sector of the Trapani Mountains, and opens along the southeastern slope of Mt. Inici, a gently westward dipping monoclinal relief affected by NW-SE, NE-SW and NNW-SSE high angle faults. The karst system is composed of two caves, Grotta dellEremita (or Grotta del Cavallo) and Abisso dei Cocci, formed in Jurassic limestones and dolomitic limestones. Thermal waters emerge from three hot springs east and at lower altitude respect to the caves (Fig. 4). These are of chloride-sulphate alkaline-earth type and have a temperature respectively of 48.3 C (Gorga 1), 49.6 C (Gorga 2), and 44.2 C (Terme Segestane) (Grassa et al. 2006 and references therein). The caves preserve clear signs of prehistoric and medieval human presence, such as several lithic and bone fragments, food remains, ceramic finds typical of the Middle NeolithicMiddle Chalcolithic (Grotta dellEremita), and large amounts of medieval pottery (XIthXVthcentury), identified mainly within Abisso dei Cocci (Tusa 2004). These caves were explored and described in the early 90s. On the basis of morphological features a genesis linked to thermal waters was supposed (Biancone 1993; Messana 1994). Figure 4. Sketch of Monte Inici karst system and localization of the thermal zone. Figure 5. Abisso dei Cocci. Passages displaying sub-circular crosssection with vadose entrenchment characterised by uncovered ceiling and walls with cupolas and gypsum in the lower parts.Different types of phosphate minerals were found in several parts of the caves. Hydroxylapatite occurs as thin crusts or powder above the bedrock, and even forms true stalactites. Crandallite [CaAl3(PO4)2 (OH)5H2O] is present as soft white grains at the contact between weathered rock and bat guano, in association with green small spherical masses of montgomeryite [Ca4MgAl4(PO4)6(OH)4H2O]. Phosphates are linked to the presence of large fossil bat guano deposits occurring in many parts of the cave. 3.2.2. Grotta dellEremita Grotta dellEremita formed in Middle-Upper Jurassic reddish-gray limestones with ammonites (Buccheri Fm.), and in Lower Jurassic limestones and dolomitic limestones (Inici Fm.) in its deepest parts. The cave is a relict 3D maze cave, which reaches a total length of 2880 m, and a vertical range of -306 m.The air temperature, measured in Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings146

PAGE 149

December 2011, was 17.6 C in the passages near the entrance area, gradually increasing up to 21.0 C in deepest chamber of the cave. Grotta dellEremita shows many morphological and depositional features similar to Abisso dei Cocci, although some differences were recognized. The cave is made up of large subhorizontal passages and big chambers connected by deep shafts, which are guided by the main tectonic discontinuity planes. Some galleries are gently inclined following the dip of bedding planes and display sub-circular cross-sections. Along the walls of these passages, in correspondence of the bedding planes, several small conduits filled by well-cemented fine reddish sediment of continental nature, are visible. Large convection niches, mega-scallops, ceiling cupolas, ceiling spheres, and drip holes are widespread in the chambers and in the largest passages of the cavity (Fig. 6). A breccia consisting of decimetric carbonatic clasts in a reddish silt matrix characterized by thin laminas and decantation structures is exposed at walls and ceiling of some deep passages. Different chemical deposits were identified: besides calcite, occurring in the form of white cigar-shaped crystals grown under old bat guano, reddish laminae, or coralloids, gypsum was found as tabular or fibrous crystals. Phosphate minerals, such as hydroxylapatite and taranakite, occur mainly in form of powders or crusts near or above deposits of bat guano. Carbonate-apatite was recognised as crusts or small stalagmites (Messana 1994). As in Abisso dei Cocci clastic sediments are absent. 3.3. Monte Kronio karst system Monte Kronio karst system opens in the southern scarp of Mt. Kronio or Mt. San Calogero, north-east of Sciacca town (southern Sicily) (Fig. 1). Mt. Kronio consists of an imbricate fan system linked to ENE-striking, closely spaced imbricate thrust sheets, involving Triassic to Miocene platform and pelagic platform carbonate deposits (Monaco et al. 1996). The karst system is made up of a series of cavities characterized by rising of hot air and vapour flow at temperature of about 38 C, connected to the presence of thermal waters. These waters, emerging along the southern slope of Mt. Kronio at lower altitude respect to the cave entrances, are of chloride-sulphate alkaline type and have a temperature ranging between 32 and 55 C (Grassa et al. 2006 and references therein; Capaccioni et al. 2011). Actually they are used for aesthetic and therapeutic purposes. The caves were visited by man since the end of the Mesolithic for residential use, place of worship, necropolis, and from the Istcentury BC for thermal purposes. The first attempts to explore the caves date back to the end of the XVIIth century; since the 40s several exploration campaigns conducted by the Commission Grotte E. Boegan of Trieste identified and surveyed the cave system nowadays known. The explorations, carried out with great difficulty, due to the critical environmental conditions with temperatures of about 38 C and humidity of 100%, have allowed the discovery of an extended maze cave system about 200 m deep (Perotti 1994). The system is composed of more cavities, located at different altitude, characterized by subhorizontal passages connected by deep shafts or steep passages, but there is not always a passable connection between the several branches of the caves. Some galleries breach the southern scarp of Mt. Kronio through small openings some of which emit hot air, other ones aspire cold air from outside (Perotti 1994). The upper section of the system, known also as Stufe di San Calogero, consists of a series of chambers separated by man-made walls, connected to a maze of narrow passages. Some chambers are characterized by hot vapour flow rising from a deep shaft which connects these passages with two large hot galleries, oriented in NW-SE direction, where abundant archeological finds were discovered. A small cave (Grotta del Lebbroso), interested by rising of hot vapour flow, develops eastward at the same altitude, but to date no passable connection to the rest of the system was recognized. The deepest part of the cave is represented by a large and deep shaft (Pozzo Trieste), from which more levels of narrow passages, both hot and cold, develop breaching the southern cliff of Mt. Kronio (Fig. 7). A large breakdown deposit occurs at the bottom of the shaft. Walls and ceiling of the caves are weathered and important gypsum deposits, in form of powders or crusts, were observed (Perotti 1994). Figure 6. Grotta dellEremita. Passage along a bedding plane where filled protoconduits are visible. Walls and ceiling are characterised by large convection niches and cupolas. Figure 7. Monte Kronio. Forms linked to condensation-corrosion (on the right); weathered flat roof affected by condensation (on the left).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings147

PAGE 150

4. Discussion and ConclusionsThe study of the geological setting, cave pattern, medium and small scale morphological features, and the analysis of cave minerals allowed defining a hypogenic genesis for the investigated karst systems. All the caves developed along structural planes, such as bedding, fracture or fault planes, whose enlargement is dueto corrosion by H2S-rich thermal waters, and to condensation-corrosion processes by air flow in the cave atmosphere. Acqua Fitusa cave is an inactive water-table sulphuric acid cave linked to corrosion processes of carbonate rock with replacement of gypsum by H2S-rich thermal water (Fig. 8). Occurrence of notches with flat roof indicates lateral corrosion processes by sulphuric thermal water fed through discharge slots still open on the floor of some passages. Ceiling cupolas, large wall convection niches, condensation-corrosion channels, boxwork testify that enlargement of voids occurred mainly above the water table where H2S degassing in the cave atmosphere, oxidation of sulphides and thermal convection produced strong condensation-corrosion processes. In addition, large amounts of gypsum, replacement pockets, in places containing gypsum, suggest the corrosion of the carbonate rock occurred with replacement of gypsum according to the sulphuric acid speleogenesis (Galdenzi and Maruoka 2003; Audra et al. 2010 and references therein). Like other sulphuric acid systems (Galdenzi and Menichetti 1995) the different levels of passages record past stages of the water table, in relation to changes of the base-level. Grotta dellEremita and Abisso dei Cocci caves were identified as inactive 3D maze caves. Pattern and cross sections of the main passages of Abisso dei Cocci suggest the early speleogenetic phases to have occurred in phreatic conditions by rising thermal water which formed a well developed 3D maze system.An important role in the evolution and widening of the subterranean voids was played by air flow when the cave passages switched from phreatic to vadose conditions, as a consequence of the uplift phases of this sector of the Sicilian chain. In this case, the processes of corrosion by condensation from air flow rich in H2S, favoured on one hand the enlargement of the early voids with the formation of large megascallops and ceiling cupolas, on the other hand the deposition of gypsum in the lower parts of the passages (Fig. 9). The first speleogenetic phases in phreatic conditions of Grotta dellEremita are recorded, beside the pattern, by the presence of several filled anastomosed protoconduits visible at the bedding plane, along which the passages develop. The next phase of evolution of the system may have followed two possible ways: a) enlargement of voids by condensation-corrosion processes which destroyed the network of protoconduits and formed the main passages as testified by the presence of megascallops, ceiling cupolas, etc. (Fig. 10); b) corrosion processes by a deep sulphuric thermal water enlarging the early voids, destroying the protoconduits and forming the main passages in phreatic Figure 8. Acqua Fitusa. Genetic mechanism of cave passages due to H2S degassing in the cave atmosphere. Figure 9. Abisso dei Cocci. Enlargement of passages by widening of pre-existing forms due to condensation-corrosion. Figure 10. Grotta dellEremita. Genesis of passages along a bedding plane with formation of protoconduits enlarged by condensation-corrosion.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings148

PAGE 151

conditions. Successively, condensation-corrosion processes by acid sulphuric air flow enlarged the early phreatic voids with the formation of megascallops, ceiling cupolas etc., and deposition of gypsum. The question is open, future studies will clarify the issue. Finally, in all the investigated caves the presence of different types of phosphate minerals is linked to the large deposits of bat guano (Hill and Forti 1997). Monte Kronio system is an active hypogenic karst system and is unique in Sicily and probably in the world. It owes its peculiarity to the rising of hot air and vapour flow, linked to a deep thermal aquifer, still not identified within the system. Although this system is known since prehistoric times, yet little is known about its real development and speleogenetic mechanisms due to the harsh environmental conditions that make exploration extremely difficult. Multidisplinary studies are currently in progress.AcknowledgementsG. Ceresia, S. Inzerillo, A. Provenzano, L. Sausa, A. Scrima contributed to resurvey Acqua Fitusa cave. G. Ceresia, A. Provenzano, P. Tordjman and G. Valdese helped sampling. Mr. Mancuso helped visiting to Monte Inici karst system. Commissione E. Boegan and La Venta for facilitating the field activities in the Monte Kronio caves. Michal Philippi and Ji Bruthans for the review of this paper.ReferencesAudra P, DAntoni-Nobecourt J-C, Bigot J-Y, 2010. Hypogenic caves in France. Speleogenesis and morphology of the cave systems. Bulletin de la Socit Gologique de France, 181 (4), 327. Audra P, Mocochain L, Bigot J-Y, Nobcourt J-C, 2009. The pattern of hypogenic caves. Proceedings of the 15th International Congress of Speleology, Kerrville (TX), 2, 795. Bianchini G, Gambassini P, 1973. Le Grotte dellAcqua Fitusa (Agrigento) I Gli scavi e lindustria litica. Rivista di Scienze Preistoriche, 28 (1), 1. Biancone V, 1993. Monte Inici. Una sorpresa tutta siciliana! Speleologia, 29, 72. Capaccioni B, Vaselli O, Tassi F, Santo AP, Huertas AD, 2011. Hydrogeochemistry of the thermal waters from the Sciacca Geothermal Field (Sicily, southern Italy). Journal of Hydrology, 396, 292. Catalano R, Agate M, Avellone G, Basilone L, Gasparo Morticelli M, Gugliotta C, Sulli A, Valenti V, Gibilaro C, Pierini S, 2011. Walking along a crustal profile across the Sicily Fold and Thrust Belt. AAPG International Conference & Exhibition, Post conference field trip 4, 27 October 2011. Di Maggio C, Madonia G, Parise M, Vattano M, 2012. Karst in Sicily and its conservation. Journal of Cave and Karst Studies, 74(2), 157. Forti P, Mocchiutti A, 2004. Le condizioni ambientali che permettono levoluzione di speleotemi di zolfo in cavit ipogeniche: nuovi dati dalle grotte di Capo Palinuro (Salerno, Italia). Le Grotte dItalia, 4 (5), 39. Galdenzi S, 1997. Initial geological observations in caves bordering the Sibari plain (southern Italy). Journal of Cave and Karst Studies, 59 (2), 81. Galdenzi S, 2012. Corrosion of limestone tablets in sulfidic ground-water: measurements and speleogenetic implications. International Journal of Speleology, 41 (2), 149. Galdenzi S, Maruoka T, 2003. Gypsum deposits in the Frasassi Caves, central Italy. Journal of Cave and Karst Studies, 65, 111. Galdenzi S, Menichetti M, 1995. Occurrence of hypogenic caves in a karst region: examples from central Italy. Environmental Geology, 26, 39. Grassa F, Capasso G, Favara R, Inguaggiato S, 2006. Chemical and isotopic composition of waters and dissolved gases in some thermal springs of Sicily and adjacent volcanic islands, Italy. Pure and applied geophysics, 163, 781. Hill C, Forti P, 1997. Cave mineral of the world. National Speleological Society, Huntsville, 463. Klimchouk AB, 2007. Hypogene Speleogenesis: hydrogeological and morphogenetic perspective. National Cave and Karst Research Institute, Special Papers, 1, Carlsbad, NM, 106. Klimchouk AB, Ford DC (Eds) 2009. Hypogene Speleogenesis and Karst Hydrogeology of Artesian Basins. Ukrainian Institute of Speleology and Karstology, Special Paper, 1, Simferopol, 280. Lombardo G, Scium A, Sollano G, Vecchio E, 2007. La Grotta dellAcqua Fitusa e larea della Montagnola nel territorio di San Giovanni Gemini (Ag). Speleologia Iblea, 12, 125. Messana E, 1994. Il sistema carsico del gruppo montuoso di M. Inici (Castellammare del Golfo, TP). Bollettino dellAccademia Gioenia Scienze Naturali, 27 (348), 547. Monaco C, Mazzoli S, Tortorici L, 1996. Active thrust tectonics in western Sicily (southern Italy): the 1968 Belice earthquake sequence. Terra Nova, 8, 372. Palmer AN, 2011. Distinction between epigenic and hypogenic maze caves. Geomorphology, 134, 9. Perotti G., 1994. Kronio. Le stufe di San Calogero e il loro flusso vaporoso. Bollettino dellAccademia Gioenia Scienze Naturali, 27 (348), 435. Piccini L, 2000. Il carsismo di origine idrotermale del Colle di Monsummano (Pistoia Toscana). Le Grotte dItalia, 1 (5), 33. Plan L, Tschegg C, De Waele J, Sptl C, 2012. Corrosion morphology and cave wall alteration in an Alpine sulfuric acid cave (Kraushhle, Austria). Geomorphology, 169, 45. Stafford KW, Land L, Veni G (Eds.) 2009. Advances in Hypogene Karst Studies. National Cave and Karst Research Institute Symposium, 1, 182. Tisato N, Sauro F, Bernasconi SM, Bruijn R, De Waele J, 2012. Hypogenic contribution to speleogenesis in a predominant epigenic karst system: A case study from the Venetian Alps, Italy. Geomorphology, 151, 156. Tusa S, 2004. Grotta del Cavallo e la preistoria del comprensorio di Inici. In: CAI Palermo (Ed.). I Tesori di Monte Inici, 85.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings149

PAGE 152

sandy gravel, red clay). Ground-water tracing studies proved the presence of the second level under the main branch. There is no passage accessible for man between the two levels. The cave creek is temporary, flowing water occurs only during the time of great floods. The Mnnich Passage is located 1,300 m from the entrance at Aggtelek village. Its a man-made path named after Klmn Mnnich who recognised that by digging out a short section filled by sediments a convenient tourist route can be formed that exclose a difficultly accessible section called Nehz-t (Hard Road) (Figs 2, 3, 4). The height of the roof here is about 1.5.5 m now. Adding the thickness of the sediment (755 cm according to our sampling), the cavern of about 10 meters high can be assumed. It corresponds to the characteristic size of the cave at Aggtelek section. According to the findings (Bosak et. al. 2004) the age of the top part of the sediment at Mnnich Passage is 114 ka based on U/Th series dating.SEDIMENTS AT THE MNNICH PASSAGE IN THE BARADLA CAVE (HUNGARY): MINERALOGICAL AND PETROLOGICAL STUDYGbor Vid1, Istvn Bernyi veges1, Orsolya Viktorik1, Tibor Nmeth2,3, Zsolt Bend4, Sndor Jzsa4, Judit Bernyi veges1 1Niphargus Cave Research Association, Kaffka Pter 4., H-3758 Jsvaf Hungary, info@niphargus.hu2HAS Institute for Geological and Geochemical Research, Budarsi t 45., H-1112 Budapest, Hungary3Department of Mineralogy, Etvs Lornd University, Pzmny Pter stny. 1., H-1117 Budapest, Hungary4Department of Petrology and Geochemistry, Etvs Lornd University, Pzmny Pter stny. 1., H-1117 Budapest, Hungary Unconsolidated sediments at Mnnich Passage in the Baradla Cave were studied by sedimentological and mineralogical methods (grain size distribution, sorting by grain size fractions and studying them by, XRD and SEM). In silty sediments filling the cavern in 755 cm thickness quartz is the dominant mineral. Mica-illite, illite/smectite, kaolinite, limonite (goethite) are present in all layers. The mineral composition of the clay fraction, the presence of iron minerals and idiomorphic quartz and tourmaline crystals, as well as the absence of calcite suggests short transport pathway. This means that the fine sediments found at Mnnich Passage may originate from the soils and fine sediments of the surface area nearby.1. IntroductionOur research group has studied the sediments of Baradla Cave since 2002. The research was started at Mnnich Passage because this place seemed most promising to find sediments of several meters thick according to unpublished former studies (mid 1980s) of the Hungarian Geological Institute. Since 2002 seven other sites were sampled by drilling where sediments in various thickness were found. Based on these experiences a more detailed study using more sensitive methods with repeated sampling at this site can contribute to the understanding of the genesis of the cave particularly on the cyclic filling and depleting of the sediments (Bernyi et. al. 2006). Baradla Cave is located on the north-eastern part of Hungary at the border with the Slovak Republic in the mountains called Gmr-Torna Karst (Fig. 1). The cave is part of the Aggtelek National Park, belongs to the World Heritage and is protected by the Ramsar Convention. Baradla Cave is the longest known cave in Hungary with the total length of 27 km. It is developed in Triassic limestone that is covered by younger sediments (gravel, Figure 1. Location of Baradla Cave. Figure 2. Map of Baradla Cave. (Map from National Cave Register).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings150

PAGE 153

2. Materials and MethodsThe sediments were sampled by hand operated helical (D = 55 mm) soil sampler. Material forming a uniform layer was collected in one sample. If the uniform layer was thicker than 50 cm samples were taken at 50 cm intervals. Undisturbed samples were also taken from the wall of the passage (Fig. 4). Particle size distribution was determined by sieving combined with sedimentation method. Mineralogical composition was studied after wet sieving and heavy mineral separation (SPT method) by binocular microscopy. X-ray diffraction studies were carried out by a Phillips PW 1710 equipment with Cu K-alpha radiation 35 mA tube current and 45 kV acceleration voltage. Semi quantitative analysis was carried out on the disoriented powdered sample. Clay minerals were studied after separating the fraction less than 2 micrometers by sedimentation. Ethylene-glycol and heat (350 C and 550 C) treatments were also applied. SEM studies were carried out on the undisturbed dried samples. The samples were impregnated into two component clear epoxy adhesive (Araldite 2020) by vacuum impregnator. The measurements were carried out on the AMRAY 1830 I/T6 SEM/EDX.3. Results and discussionAt the entrance of the passage the thickness of the sediments is 755 cm. Table 1. presents the sediment characteristics in the borehole. The sediments found in the borehole and in the wall of the passage are uniform. The particle size distribution curves are presented in Figure 5. Figure 3. Map of Mnnich passage and its region (Orszg et al. 1989), left side: direction to Aggtelek, right side direction to Jsvaf. Legend: 1: Hard Road, 2: Hill of Mrea, 3: Wax street, 4: Mnnich passage, 5: Place of borehole. Figure 4. Baradla-Cave Mnnich-Passage (from Aggtelek side). Legend: Place of undisturbed sample. DepthMaterial 0 cmreddish brown silt 195 cmsilt with charcoal 202 cmlight brown silty sand 220 cmreddish brown silt 245 cmyellowish brown silt 400 cmyellowish brown silty sand 710 cmyellowish brown silty sand with black grains Table 1.Sediment characteristics in borehole. Figure 5. Particle size distribution curves measured in the sequence.According to the particle size distribution curves (Fig. 5) and the SEM study (Fig. 6) the particle size ranges between 10 micrometers and the smallest fraction is between 20 and 30 micrometers. Particle size characteristics are very similar to that of loess. The structure of the sediment is closely packed and only a few laminar voids can be found. Particles are angular and fragmented. Figure 6. Backscattered electron image of the undisturbed sample taken from the sidewall of the Mnnich-passage.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings151

PAGE 154

The results of SEM, XRD and microscope studies are in mutual agreement. The mineralogical composition of the sediments are: quartz 50%, 5% feldspars (both plagoiclase and K-feldspars), and 20% phyllosilicates (mica-illite, kaolinite, illite/smectite mixed layer mineral, smectite). Carbonates are not present in the sediments, except the layer where dropstone fragments were found. Goethite is present in less than 5% in the bottom part of the sequence. Limonite aggregates were also found. Binocular microscope study showed that the forms of limonite aggregates are laminar on the top of the sequence. Fine grained aggregates or nodules occur at the bottom. Tourmaline (dravite composition), monacite and xenotime were also detected by SEM-EDX. Figure 7. Part of bedded thin section sample from 0 cm depths, >250 m grains. Legend: 1: bone fragment, 2:calcite, 3: limonite aggregates.Dominant clay minerals are illite and kaolinite, their ratio is the same throughout the sequence. Smectite is present in a smaller amount in the form of illite/smectite. The amount of smectite increases upwards, and reaches its maximum in the top layer of the sequence (Fig. 8). The interlayer cation of the smectite is dominantly Ca and based on SEM-EDX it is rich in iron substitution in the octahedral sheet. It is noteworthy that the bottom layers do not contain chlorite, it occurs only in the uppermost layer in a small amount. In the bottom part goethite is detectable in the clay fraction which indicates its pedogenic origin. The shape and sorting of the particles indicate a short transport distance, they sedimented from a still or slowly percolating water. SEM investigation showed that the cracked parts of quartz, tourmaline monacite grains are next to each other. This indicates that the cracking took part after the sedimentation. The origin of a high pressure cracking these particles is unknown; perhaps frozen water on the surface of the sediment or in the pores can be responsible for that. Reddish clayey sediments and soils can be found at the surface of the Aggtelek-karst having goethite as dominant iron mineral, and kaolinite with smectite as characteristic clay minerals (Fekete et al. 2008). Other studies found hematite and kaolinite in red soils of the same region (Czirbus et al. 2010). The absence of carbonates in the sediment indicates that the sediment transported into the cave did not contain carbonates originally or they were eluviated (washed out and totally dissolved) similar to present day soils before the sediment entered the cave.4. SummaryBased on our mineralogical studies, the fine sediments found at Mnnich Passage can be originated from the soils and fine sediments of the surface area nearby. Short transport distance is indicated by the presence of idiomorphic quartz and tourmaline. To support this assumption a detailed study of the sediments on the surface is planned.AcknowledgementThe authors are grateful to the Aggteleki National Park, the Soliform Plc., Andrs Varga Etvs Lornd University and friends from the Nyphargus research association. Figure 8. XRD patterns of the clay fractions by depth. Legend: I/S = illite/smectite, Gt = goethite. Figure 9. SEM image of a cracked tourmaline. (Undisturbed sample taken from the sidewall of the Mnnich-passage).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings152

PAGE 155

ReferencesBernyi I., Bernyi J., Vid Gbor, 2006: Adalkok a Baradlabarlang fejl dsnek elmlethez ledkvizsglatok alapjn. Karszt s Barlang 2006 III (in Hungarian). Bosk P., Hercman H, Kadlec J, Mga J, Pruner P, 2004: Paleomagnetic and U-series dating of cave sediments in Baradla cave, Hungary. Acta Carsologica, 33(2), 219. Czirbus N, Nyilas T, Bozs G, Hetnyi M, 2010: Az aggtelekikarsztrl szrmaz vrsagyagos rendzina talaj szerves anyagnak geokmiai jellemzse Rock-Eval pirolzissel. Krpt-medencei Doktoranduszok Nemzetkzi Konferencija, Konferencia Ktet, 46 (in Hungarian). Fekete J, Csibi M, Stefanovits P, 2008: Magyarorszgi vrsagyagok jelent sge, fontosabb talajtani jellemz ik. Talajvdelem klnszm 585 (in Hungarian). National Cave Register, http://www.termeszetvedelem.hu/ index.php?pg=cave_5430-1 Orszg Gy, Vid Szilgyi F, Vgh Zs, Gyuricza Gy, Fruny E, Tth Zs, 1989: Baradla-barlang 1:1,000, Magyarorszg barlangtrkepei 7. ktet (in Hungarian).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings153

PAGE 156

Pure white chalk and impervious marls of Campanian age separate the exposed Maastrichtian carbonates from the underlying deeply weathered and kaolinized Carboniferous limestones (Felder and Bosch 1998). Above the Cretaceous layers, Oligocene marine sands were deposited. In some places these sands are preserved in dolines developed in the Cretaceous formations. During Pliocene time, the river Meuse drainage system came into existence and progressively incised down to its present level, as testified by peneplain surfaces and gravel terraces. The high terrace of the river is a Middle Pleistocene gravel formation. This coarse and loamy river deposits armour the underlying Cretaceous chalk and lead to a local relief inversion (Juvign 1976). During the Weichselian glaciation, aboutENDOKARSTS AND CRYPTOKARSTS IN CRETACEOUS COARSE AND HIGHLY POROUS CHALK AT THE BELGIAN-DUTCH BORDERLuc Willems1, Jol Rodet2 1Department of Geology, Sedimentary Petrology, Bd du Rectorat, B20, Sart Tilman, 4000 Lige, Belgium. L.Willems@ulg.ac.be2 UMR 6143 CNRS, Morphodynamique continentale et ctire, Laboratory of Geology, University of Rouen, 76821 Mont-Saint-Aignan Cedex, FRANCE Abstract. Since 2003, the study of several quarries at the Belgian-Dutch border has made it possible to identify numerous karsts essentially developed in coarse chalk (calcarenites) of the Maastricht Formation (Upper Cretaceous). This lithology is highly porous and is often considered unfavorable to karstification. However, caves, solution pipes, sponge networks, roof channel, pockets (alveoli) several meters in diameter developed inside without connection to fractures. These karsts belong to flooded karsts (caves and pockets) or to younger cryptokarsts (input karst type thousands of solution pipes). When the endokarsts dewater, the high porosity of calcarenites inhibits closed caves from evolving. Tubular solution pipes are produced by the seepage water under fluvial terrace gravels and can exceed 30 m deep under the surface plateau. Sometimes, they encounter caves which are consequently filled and fossilized. By this process, they preserve caves from further collapsing inside a crumbly lithology. Abstrait. Depuis 2003, ltude de diffrentes carrires, la frontire belgo-nerlandaise, a permis didentifier un grand nombre de karsts essentiellement localises dans des calcarnites de la Formation de Maastricht (Crtac suprieur). Cette lithologie trs poreuse est souvent considre peu karstifiable car elle devrait inhiber une concentration des altrations. Pourtant, grottes, racines du manteau daltration, rseaux en ponge, chenaux de vote, alvoles plurimtriques y sont dvelopps, hors fracturation. Ces morphologies sont rattaches soit des endokarsts noys (grottes, alvoles) soit des cryptokarsts plus rcent de type karsts dintroduction (plusieurs milliers de racines du manteau daltration). Au moment de lexondation des endokarsts, la porosit leve des calcarnites stoppe lvolution des grottes fermes. Des racines du manteau daltration, formes tubulaires gnres par les eaux dinfiltration sous un cailloutis de terrasse fluviale, peuvent senfoncer de plusieurs dizaines de m sous la surface du plateau. Elles peuvent recouper des grottes quelles colmatent et quelles prservent deffondrement ultrieur dans une lithologie trs friable.1. IntroductionCretaceous chalks and calcarenites (coarse chalks) at the Belgo-Dutch border (Fig. 1) develop numerous endokarsts and cryptokarsts without connection with fractures and in spite of a high porosity of the rock (Willems et al. 2004, 2005a, b; 2007a, 2007b, 2010; Lagrou et al. 2008; Rodet et al. 2009b). These cryptokarsts are essentially vertical tubes, solution pipes, which can reach more than 30 m deep under the plateau surface. In the area, they take place only under fluvial terrace gravels. The endokarsts are centimeters to several meters in size closed pockets which can be organized in subhorizontal blind tunnel-like caves which can be more than 60 m long. These caves had functioned as drains in the top of an aquifer.2. Physical environmentThe studied karsts are intersected by underground or surface quarries opened in the Gulpen and Maastricht Formations (Cretaceous), exposed with a thickness of 100 m. The upper-lying Maastricht Formation consists in macroporous calcarenites rich in flint nodules and subdivided by hardgrounds. Porosity between 37% and 44% (Willems et al. 2010) is determined near cavities. These values are much reduced at hardgrounds or fossiliferous beds, which have no influence on the studied karst morphologies. The underlying Gulpen Formation consists of very fine calcarenites grading downwards to chalk rich in flint beds. Figure 1. Localization of the studied area.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings154

PAGE 157

ten meters of loess was deposited, burying the pre-existing landscape (Haesaerts, 1984). High rates of underground water flow (2.2 to 13 km/day) have been reported (Dassargues and Montjoie 1993). They are similar to highly karstified aquifer (Harold 1937; Atkinson and Smith 1974; Rodet 1992; Maurice et al. 2006). They could be due to horizontal narrow pipes observed in drainage galleries (Van Den Broeck et al. 1910; Ministre de la Rgion wallonne and Universit de Lige 2006), but also to the caves and sponge networks found in the Maastricht Formation. Deep weathered nodes are also generated in chalks of the Gulpen Formation, tens of meters under the Meuse alluvial plain (Willems et al. 2007a, b; 2010). They could also favour high rates of flows.3. CryptokarstsThe cryptokarsts are essentially tubular solution pipes (raciness du manteaux daltration) (Fig. 2) developed under terrace gravels of the river Meuse. They have quite regular sections and their diameters vary between a few centimeters to over two meters. They may exceed 30 m deep under the plateau surface. The solution pipes contain Oligocene sands and/or sandy clays, and also rounded pebbles coming from the upper Meuse river terrace. The descent of superficial deposits into the Cretaceous strata supports evidence of deepening of the solution pipes. Their widening results from the coalescence of smaller ones. Juvign (1992) connected their genesis to lithological variation in terrace deposits. The coarse parts of the terraces produce local concentrations of seepage water and initiate these solution pipes (Rodet et al. 2009a). A local flint accumulation could trigger the same water concentration process. On the contrary, when sand deposits partially cover the Cretaceous formations, no evidence of solution pipes are found. The reason for this could be the dispersion of seepage water in this kind of lithology. Several observations show that these cryptokarsts stopped enlarging at around 75 ka BP, i.e. the Upper Weichselian (Juvign 1992; Juvign et al. 2008; Willems et al. 2010). Juvign (1992) suggests a permafrost, which inhibited the seepage water. Another hypothesis is a fast loess deposit on the terrace gravel playing the same role as a sand cover (Willems et al. 2010). The rapidity of loess deposits is supported by a noneroded tephra inside the loess. Some solution pipes encountered lower horizontal caves, their filling are withdrawn, and small swallow holes are generated (Willems et al. 2010).4. EndokarstsMost endokarsts consist of meters to several meters sized blind cavities with similar altitudes. They are located between 20 and 30 m below the plateau surface and between 10 and 20 m below the contact Oligocene/ Cretaceous (Willems et al. 2007, 2010). Two types of cavities can be identified: the first one consists in a corridor with an ellipsoidal section, developed in a parallel to the subhorizontal stratification. The second one is made of a single room with rounded walls, which are in turn generated by the coalescence of many centimetric to decametric pockets. There is an obvious continuum between sponge networks, single-room caves, and corridor caves (Fig 3). Endokarsts have the following characteristics: 1. The caves are formed outside the fractures. If fractures are present, they are poorly karstified and they dissect the caves. 2. Numerous small horizontal pipes are connected to the main cave without a preferential orientation. 3. The cave floor is made of mixed collapsed blocks and weathered calcarenite powder. There is a transition between this powder and the fresh bedrock without a clear limit. Traces of oxide migrations inside the weathered part of the floor attest to slow water flows. In some cases, the cave floor consists in pockets. 4. Some corridor caves are completely clogged up by superficial deposits (Tertiary sands, loess) (Fig. 3c) and show sedimentary fill typical of low energy flow. These observations show that some caves had functioned as drain in the top of an aquifer. 5. Trace of an ice-wedge is observed in the floor deposit of a cave (Willems et al. 2010). It demonstrates the top of an old permafrost tens meters under the plateau surface. Figures of hydromorphy under this structure (traces of iron hydroxide remobilization) could result from a perched aquifer due to the impermeable permafrost. It could explain the temporary concentrated flows from the seepage water and promotes drains and a slow moving of the brecciated materials of the floor. Figure 2. Solution pipes opened in the underground quarry (Caster, Belgium).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings155

PAGE 158

5. Discussion ConclusionAt least three stages are identified in evolution of the karsts: Firstly, the developments of caves, deep under the surface of the plateau, in phreatic zone, as evidenced by sponge networks, micro and macro pockets. Secondly, a dewatering of the caves and a widespread terminated of dissolution process. Thirdly, the development of fractures which dissect the caves or solution pipes which encounter them. In this stage, the caves can be completely filled and fossilized (Fig. 4). Various caves listed are sometimes several kilometers away from each other and have similar altitudes. Their genesis can be related to a paleo aquifer. The caves could also result in the evolution of deep weathering nodes (Fig. 5) such as those found twenty meters below the Meuse alluvial plain (Willems et al. 2007). These nodes are similar to ghostrocks found within the Tournaisian limestone (Quinif et al. 1994; Vergari 1998).The ghost-rock isin situ limestone weathering with residual weathered-rock (ghost-rock) that keeps its initial volume. The caves could be related to slow water circulations within the Cretaceous Formations and under the regional thalwegs (Willems et al. 2007 a, b) The dismantling of this paleo aquifer can be due to the downcutting of the Meuse River and its tributaries. Studies of karst in high porosity carbonate rocks are still few (Rodet 1992). However, the example of the Cretaceous chalk karstification at the Belgian-Dutch border shows that they can be very numerous. Their genesis results of a much more complex history than a simple fracture karstification. Their impact on water resources needs to be evaluated.ReferencesAtkinsons C, Smith DI, 1974. Rapid groundwater flow in fissures in the Chalk: An example from South Hampshire. Quarterly Journal of Engineering Geology, 7: 197. Dassargues A, Monjoie A, 1993. Hydrogeology of the chalk of North-West Europe. Chap. 8: Chalk as an aquifer in Belgium, Oxford University Press, 153.6. Some caves are encounter by solution pipes. The superficial materials (loess, terrace pebbles, etc.) which constitue the central part of the solution pipe partially dump in the caves. The deposit shows evidence of a process in an unsaturated zone. It shows that the cave was dewatered before it was encountured by solution pipes. Figure 3. Several types of endokarsts: a.) sponge network inside the calcarenite Petit Lanaye Suprieur (scale bar: 20 cm); b). single room cave Petit Lanaye Suprieur; c). corridor cave with Tertiary sands and Quaternary loess filling CRSOA cave Petit Lanaye infrieur Belgium (scale bar: 20 cm). Figure 4. Complex karst network in the quarry of RomontBelgium. Horizontal corridor cave (G) single room caves(A) encountered by solution pipes (R) which intersect previous forms.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings156

PAGE 159

Felder WM, Bosch PW, 1998. Geologie van de St. Pietersbergbij Maastricht. Grondboor & Hamer, 52: 53. Haesaerts P, 1984. Aspects de lvolution du paysage et de lenvironnement en Belgique au Quaternaire, in: Cahen, D., Haesaerts, P. (Eds.), Peuples chasseurs de la Belgique prhistorique dans leur cadre naturel, Bruxelles, 27. Harold C, 1937. The flow and bacteriology of underground water in the Lee Valley. Metropolitan Water Board 32nd Annual Report, London, England. 89, in Maurice et al. 2006. Juvign E, 1976. La stratigraphie du Quaternaire. Gomorphologie de la Belgique, in Hommage au Prof. P. Macar, Laboratoire de Gologie et Gographie physique, Universit de Lige, 169. Juvign E, 1992. Les formations cnozoques de la carrire C.B.R. du Romont (Eben/ Bassenge, Belgique). Ann. Soc. Gol. de Belgique, 115: 159. Juvign E, Tallier E, Haesaerts P, Pirson ST, 2008. Un nouveau stratotype du tphra de Rocourt dans la carrire de Romont (Eben/Bassenge, Belgique). Quaternaire, 19 (2), 2008, 133. Maurice LD, Atkinson TC, Barker JA, Bloomfield JP, Farrant AR, Williams AT, 2006. Karsticbehaviour of groundwater in the English Chalk. J. of Hydrology, 330: 63. Ministre de la Rgion wallonne (DGRNE) & Universit de Lige, GomacHydrogeologie, 2006. Carte hydrologique de Wallonie, Alleur-Lige 42/1-2,Notice explicative, 52. Quinif Y, Vergari A, Doremus P, Hennebert M, Charlet JM, 1994. Phnomnes karstiques affectant le calcaire du Hainaut. Bull. Soc. Bel. de Gologie, 102: 379. Rodet J, 1992. La craie et ses karsts, Caen, Centre normand dtude du Karst et des Cavits du Sous-sol. Elbeuf, & Groupe Seine-CNRS, 560. Rodet J, Brown J, Dupont J-P, 2009a. Development and function of a perched aquifer in the covering layers of the chalk limestones in the Paris Basin. Proceedings of the 15th International Congress of Speleology, Kerrville (Texas, USA), 19 july 2009, UIS, vol. 3, contributed papers: 1662. Rodet J, Willems L, Brown J, Ogier-Halm S, Bourdin M, Viard JP, 2009b. Morphodynamic incidences of the trepanning of the endokarst by solution pipes. Examples of chalk caves in Western Europe (France and Belgium). Proceedings of the 15th International Congress of Speleology, Kerrville (Texas, USA), 19 july 2009, UIS, vol. 3, contributed papers: 1657. Speleogenesis, 2012. Info KarstBase Glossary of Karst and Cave Terms. www.speleogenesis.info/directory/glossary/index.php Van Den Broeck E, Martel EA, Rahir E, 1910. Les Cavernes et les Rivires souterraines de la Belgique. Lamertin, Bruxelles, Vol.I, p. VIII. Vergari A, 1998. Nouveau regard sur la splogense: le pseudoendokarstduTournaisis (Hainaut, Belgique). Karstologia, 31/1: 12. Willems L, Rodet J, Massei N, Fournier M, Laignel B, DussartBaptista L, Schyns JCH, Ek C, 2004. Chalk karsts SaintPierre Mountain Basse-Meuse (Belgium). Poster MeuseRhine EuregioGeologists Meeting, Maastricht, 28 Mai 2004. Willems L, Rodet J, Massei N, Fournier M, Laignel B, DussartBaptista L, Schyns JCH, EK C, 2005a. Gense dun systme karstique dans la craie en Basse Meuse (Frontire belgonerlandaise). Poster, Coll. Intern. Karst et amnagement du territoire , Rgion wallonne, Jambes, Mai 2005. Willems L, Rodet J, Fournier M, Massei N, Laignel B, DussartBaptista L, Schyns JCH, Dusar M, Lagrou D, Ek C, 2005b. Karst system genesis in the chalk of the lower Meuse (BelgianDutchborder), 14e Congrs de lUIS, AthnesKalamos, 23 Aot 2005, CDRom Full Paper: 0 (6 p). Willems L, Rodet J, Fournier M, Laignel B, Dusar M, Lagrou D, Pouclet A, Massei N, Dussart-Baptista L, Compre PH, Ek C, 2007a. Polyphase karst system in Cretaceous chalk and calcarenite of the Belgian-Dutch border. Z.Geomorph. N.F., Berlin-Stuttgart, 51 (3): 361. Willems, L., Rodet, J, Ek, C., Dusar, M., Lagrou, D., Fournier, M., Laignel, B. & Pouclet, A., 2007b. Karsts des craies et calcarnites de la Montagne Saint-Pierre (Basse Meuse ligeoise). Bull. des Chercheurs de la Wallonie, 46: 171. Willems L, Rodet J, Ek C, Pirson ST, Juvign E, 2010. Karsts des calcarnites de la carrire du Romont (Eben Belgique). Bull. Chercheurs de la Wallonie, hors-srie No. 3, 115. Figure 5: deep weathering nodes ENCI quarry MaastrichtNetherlands.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings157

PAGE 160

THE TUPPER GLACIER SINK RASPBERRY RISING CAVE SYSTEM, GLACIER NATIONAL PARK, CANADA: A REMARKABLE EXAMPLE IN STRIPE KARSTCharles Yonge, Nicholaus Vieira, Adam Walker Alberta Karst Consulting, 1009 Larch Place, Canmore, Alberta, Canada, chas-karst@telus.net The large spring in Rogers Pass named by cavers as Raspberry Rising, and by Parks Canada as Cascade Cave, has been known of for over a hundred years. In the late 1960s, Derek Ford and others from the Karst Research Group at McMaster University inserted dye at the Tupper Sink, which takes the large outflow from the Tupper Glacier, recording an unprecedented underground flow time to Raspberry Rising (almost 500 m lower and 2 km away) of 53 minutes. Such a rapid flow-through indicates a very direct route through an open cave system, which is contained within a sub-vertical, narrow sheet of marble a classic Type 1 stripe-karst hydrology. The Tupper-Raspberry cave system is compared to 3 other significant stripe karst caves in the vicinity and to Raggejavri-Raigi Cave in Norway. A short (10 m) sump encountered 70 m into Raspberry Rising had been repeatedly dived from 1971 onward. Progress had been halted at a large waterfall, with only one attempt being made to climb it by mapole in 1981. Now while the systems proximity to the TransCanada Highway makes it accessible, high water levels in the summer necessitate winter travel into one of the Passs largest, most highly restricted avalanche zones (the Connaught Slide Path). Despite this however, the winters of 2011 and 2012 saw success in climbing the waterfall and the mapping of around 2.4 km of large, well-decorated passageways, with almost half of the distance to the Tupper Glacier Sink reached. Work is in progress.1. IntroductionThe large spring in Rogers Pass named by cavers as Raspberry Rising, and by Parks Canada as Cascade Cave, has been known of for over a hundred years. In the summer of 1966, Derek Ford and others from the Karst Research Group at McMaster University dye-traced the water at Tupper Sink, the source of which is the large Tupper Figure 1. Location of the TupperRaspberry Cave System.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings158

PAGE 161

Glacier, and getting a positive and rapid response at Raspberry Rising, 483 m lower and 1.94 km away, in 53 minutes (Ford 1967). An open, vadose cave system seemed likely, especially as both sink and spring appeared to be contained within a narrow, sub-vertical band of marble. Its proximity to the highway makes it very accessible (Figure1) although winter travel into one of the Passs largest avalanche zones (the Connaught Slide Path) is highly restricted. Once inside Raspberry Rising, the first serious obstacle preventing exploration comes 70 m from the entrance where there is a 10 m sump. This sump was first passed in 1971 by Mike Boon, but he was soon stopped by a high waterfall. While other dives followed the only progress made was by Mike Boon, Pete Lord and Randy Spahl in 1981 (see Yonge 2012 for the history of these dives). Boon ascended the lower part of the waterfall using a maypole, but was ultimately overwhelmed by the volume of falling water and Pete Lord had been able to climb ledges to the right of the waterfall, apparently finding 65 m of passage to a dig (this likely leads to the passage system above the waterfall found later). Thus Vieira, with us in support as far as the sump made it to the top of the waterfall via modern bolting methods on February 2nd, 2012; it is 25 m high and named the Nick Point. What he discovered beyond (The Dream Collector) was astounding; he saw that the passage ran ahead as a field of massive tumbled boulders with the stream running under them, 15 m wide and 20 m high. Later that winter and spring plus this winter, we conducted several diving and mapping trips into the system, penetrating 800 m into the mountain with some 2.4 km of survey completed. We also undertook basic photography of Figure 3. Steeply dipping beds at the Nick Point. Note thinner bands of schist within the marble. Figure 2. Copious speleothems in the T-R System. The curtain at right is ~2 m long. Figure 4. Location of stripe karst caves in the area.the very extensive speleothems (Figures 2 and 7) and made preliminary observations of the surprisingly sparse biological/organic material.2. Geology and Geomorphological SettingThe Tupper-Raspberry cave system (T-R) is confined to a 20 m to 60 m-wide marble band sandwiched between calcareous slates and garnet schists (the Precambrian Horsethief Creek Group of the Shuswap Metamorphic Complex (SMC): Poulton and Simony 1980). It varies in dip from near vertical to 45(Figure 3), giving a practically straight-line path between sink and rising. Furthermore the band extends eastwards to another sink from a small lake known as Dyegone, 371 m above Raspberry Rising (Figure1). Dye placed in this sink by the McMaster Group travelled to the rising in approximately 8 hours. The nearest caves of any significance, also in marble stripe karst, are White Rabbit, Cody and Nakimu Caves (Figure4 and Table 1). The closest, 11.5 km west of the T-R system, are Nakimu Caves (Rollins 2004) in the highly fractured, veined crystalline Badshot Limestone of Lower Cambrian age (Poulson and Simony 1980). Actually, it comprises a system of three caves with a common meltwater stream separated by collapsed sections, which yielded a total surveyed length of 5.49 km and and depth of 286 m. It is of the Type 3 perched stripe karst type. The next nearest cave of any significance, albeit at an early stage of exploration (<400 m mapped) is White Rabbit which is 100 km west Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings159

PAGE 162

of the T-R system. It is in a late Precambrian to early Cambrian marble of the Mount Grace Formation (Hoy and Pell 1985) with a possible length of >2 km and depth of ~400 m. Dips are low at an average of 23, qualifying it as a Type 3 stripe karst i.e. the marble band is narrow at ~50m and the cave appears perched on the lower impermeable unit. Cody Cave 180 km to the south is ~1.5 km long and ~60 m deep. It is in the Milford Group of Triassic age (Fyles 1967) and has some relationship to Nakimu lying in the same north-south metamorphic belt. It is an example of the Type 2 stripe karst being confined to development at the upper contact (Lauritzen 2001 and Table 1). The four caves (including the TR system Figure 4) are indicative of the rareness of speleogenic carbonate rocks in the area and the diversity of relationship (age of rock and setting) between them. All lie within the extensive Shushwap Metamorphic Complex of the Middle to Late Jurassic.3. Results and Discussion3.1. The Tupper-Raspberry Stripe Karst The Tupper-Raspberry System is a fine example of a Type1, sub-vertical marble stripe karst (the Scandanavian Caledonides or Norwegian Karst Type exemplifies the 3 types or classes of stripe karst defined in the Norwegian setting Lauritzen 2001 and 2004). The Tupper-Raspberry System meets the Type 1 criteria, consisting of a narrow outcrop of steeply dipping karst rock, where the allogenic contact perimeter (P) is very large relative to the area (A) of karst i.e. the outcrop is very narrow compared to its length. Lauritzen (2001) defines a quantity r based on the P/A where is the ratio of the length of the stripe to its width w (i.e. = /w) which replaces and w in the expression. We have used a modified expression to give a dimensionless integrity to the relationship by defining a ratio of r = P/A1/2 (Figure 6), yielding: r = P/A1/2= 2(1/2 + -1/2) (1) The right-hand side of the expression is the same as that used by Lauritzen (2001) and is a sigmoidal function with its differential having a maximum at = 3. dr/d = P/A1/2= -1/2 + -3/2(2) The modified r does not change the essential discussion given by Lauritzen (2001), and it can easily be shown that the minimum for P/A is a square (i.e. = 1). The dimensionless r = P/A1/2 also yields a minimum at = 1. The differential is a measure of the sensitivity of the perimeter to area of stripe karst in which values of 3> >30 which Lauritzen refers to as the region of development to a full stripe karst geometry. Thus based on the distance between Raspberry Rising and the known sink-points of 1.94 and 2.63 km, and an in-cave horizontal expression of the marble from 20 m (average 25 m), = 78 to 105 is indicative of a fully developed karst geometry > 30. (Note that White Rabbit, Cody and Nakimu Caves have = 44, 19 and 3 respectively, which defines all as stripe karst, although the latter represents an end member.) Narrow, steeply-dipping stripe karst gives rise to solution on both contact faces with greater karstification at the hanging wall. Furthermore, the confining walls of schists and slates containing sulphides and ferrous oxides, which tend to impregnate the enclosed marbles, allows sulphuric acid dissolution to take place (rust staining on the marble at the contact suggests this). However, because of metamorphism, primary voids tend to be absent, but the brittleness of the marble leads to prevalent fractures and joints for which extensive speleogenesis is required (Lauritzen 2001 and Figure 5). The TupperRaspberry system does show extensive fracturing with enormous, truck-sized breakdown blocks. The rotation of the marble band from near vertical at the rising to 45 at the midpoint back to around 80 at Tupper Sink suggest tectonic flexing of the unit leading to additional fracturing beyond that due to initial uplift and regional folding (Figures 5 and 7). This system can be compared to the celebrated Type 1 stripe karst cave at Hellemofjord, Norway known as Raggejavri-Raigi (RJR Lauritzen et al. 1991 and Table 1). RJR descends a subvertical vertical marble from a plateau to the base of the fjordal wall rising just below sea level (a dry, relict entrance allows for a through trip). Whereas the marble band in this case is oriented at a right-angle to the fjord wall, the T-R system is angled (Figure 1). However while similar in depth (RJR = 580 m), the distance from sink to rising is much less Figure 5. Person standing on breakdown in the main passage. Note bedding and jointing sets. Figure 6 (modified from Lauritzen, 2001). Eqn. (1) and its first derivative plotted, showing a maximum sensitivity of = 3. The sensitivity of P/ A with increasing dies out above = 30, i.e. stripe geometry is fully developed. Most stripe karsts display in the interval 10 < < 200.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings160

PAGE 163

for RJR at 0.79 km compared to 1.94 km for T-R. This gives RJR a high hydraulic gradient 0.73 compared to 0.25 for T-R, and results in some large pitches and likely rapid flowthrough times, although this has not yet been determined (Lauritzen, percomm.). Unlike RJR, where large chambers have consumed almost all of the surrounding rock (a width of between 20 m) the T-R system appears to wander between the confining walls. This occurs because thin discontinuous schist horizons in the marble limestone temporarily confine the passage width. Indeed the initial sump penetrates through a narrow slot in a thin schist unit dividing one massive marble unit from another. 3.2. Hydrology and Speleogenesis The rapid flow-through time for the Tupper-Raspberry link can be compared to some of the fastest vadose karstic flows anywhere. The summer T-R breakthrough of <53 minutes yields an average flow speed of 61 cm/s. Worthington et al (2000) have compiled linear karst groundwater velocities, based on a global data set of 3,015 trace experiments (compensated for a sinuosity of 1.2) and finds that only 0.5% of karstic flows have speeds greater than this. The high T-R hydraulic gradient (483 m/1,940 m = 0.25) combined with open passages in a linear, narrow, subvertical carbonate band leads to this rapid flow. Dygone Sink is at a lower altitude and is further away, giving a hydraulic gradient of 0.14 (371 m/2,630 m). With an 8-hour summer breakthrough, this yields a flowspeed of 9 cm/s; we suspect that additional sedimentary blocking and possible siphons are responsible for the slower flow. Despite this, Dygone falls within the top 10% of the fastest karst flow-through times. The combined flow of Tupper, Dygone and the many other inlets (both seen and unseen) yields summer flows of 0.71 cumecs. Based on increasing flows at Raspberry during a typical summers day; Ford (1967) estimated that 75% of the water comes from the Tupper Glacier. In winter, we have calculted average flows of around 0.05 cumecs or 7% of summer flow, but the relative contributions from the various sinks is unknown (we have already encountered many inlets so far). While the T-R system has obstructions at both ends a blocked sink at the top and sump at the bottom it (so far) appears open in between. Although we have reached a second 1m sump (duck) which leads to a series of cascades, including one pitch, and a third longer siphon, these are unlikely to hold up the flow. The average gradient (thalweg) of the cave thus far is ~11.5 (figure 8) which suggests a significant increase ahead (>300 m) to arrive at the Tupper Sink (an overall gradient of 14). Dygone has a gradient of 8, suggesting that it may actually be an earlier sink point. Alternatively, the current stream profile extrapolates to an area of depressions and one abandoned cave (Cavegone) between Tupper Sink and Dygone, which may be the original sink point (Figure 1; see discussion below). Higher gradient relict passages such as the Dream Collector and Freedom 65 suggest major inlets, and these may be provided by the sizeable valley system cutting the middle of the Cannaught Slide (informally named Grizzly Paradise figure 1). These latter, storeyed passages are expected in Type 1 stripe karst (Lauritzen 2001 and 2004), indicating a deep karstification of the marble unit. Passage below the Nick Point and out to the current entrance appears to be junvenile, so further exploration may yield a relict rising as a downward continuation of the main passage. Figure 7. Strike view of the Tupper-Raspberry System at 70 Note that the beds at the north end of the system at 45 splay the passage network. Figure 8. Elevation view on 227 of the Tupper-Raspberry System.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings161

PAGE 164

While Tupper is a massive sink compared to Dygone, the latter may actually be older or at least contemporaneous. Using a 1956, 1:50,000 topographic map and current Google Earth imagery, it can be seen that the Tupper Glacier has retreated 1.2 km (21 m/year). The glacier currently lies 1.85 km from Tupper Sink at an altitude of 2,100 m; in 1956 it was 0.65 km distant at 1920 m. This suggests that in postWisconsin times an extensive glacier lay above both sinks. We note that in the Western Cordilleras of Canada, karst and caves are more developed in south-facing valleys and cirques, where solar melting is considerably more extensive. In the case of Tupper, the west side of the valley cirque is shaded by the high Mount Tupper ridge, whereas the east side receives long daylight sun. We would then be expect to see an earlier karst develop on that side with Dygone (or at the aforementioned area of depressions) receiving meltwater at the contact for a longer time. Eventually the glacier would have retreated back over the cirque sill, depriving Dygone (and other area) of the main supply. Currently all of the glacial meltwater cascades over the low west side of the sill to sink entirely at Tupper (except in very high water when the residual flows down the overflow channel on the surface Figure 1). In respect of the marble band cutting across the front of the glacier, it is possible that the Tupper Sink had an anchoring effect on the glacier, causing it to pivot on the east side down the steep slope above Dygone thus enhancing the ice flow on that side and under a regime of greater melting. Anchoring of glaciers on karst by depriving them of basal, lubricating regelation water has been observed in Norway (Lauritzen 1996).4. ConclusionsWhile the Tupper-Raspberry System is still at an explorative stage, we can nevertheless conclude that it is a mature karst with a very high degree of development toward a full stripe geometry ( = 78 to 105, the latter if Dygone is included). In fact it represents an end-member. Furthermore, its storeyed or tiered network is typical of a mature, Type1 sub-vertical stripe karst. While we are uncertain of the relative age relationships of the two main sinks (Tupper and Dygone), it may be that the smaller Dygone and adjacent area of depressions represent an earlier phase of karstification. Work establishing a primarily baseline survey (Figure 9), but complemented with some spelothem dating, will be continued over the next few years.ReferencesFord DC, 1967. Sinking creeks of Mount Tupper: A remarkable groundwater system in Glacier National Park. Can, Geog, XI(1), 49. Hoy T, 1987. Geology of the Cottonbelt lead-zinc-magnetite layer, carbonatites and alkilic rocks in the Mount Grace Area, Frenchmans Cap Dome, southeastern BC. BC Ministry of Energy, Mines and Petroleum Resources. Bulletin 80, 43 pages. Lauritzen S-E, 2004. Stripe karst. In: J Gunn (Ed). Encyclopedia of Caves and Karst Science. Fitzroy Dearborn, NY & London, 705. Lauritzen S-E, 1981 Marble stripe karst of the Scandanavian Caladonides: An end-member in the contact karst spectrum. Acta Carsologica, 30(2), 47. Lauritzen S-E, 1996. Karst Landforms and Caves of Nordland, North Norway: Excursion Guide 2. In: JE Mylroie and S-E Lauritzen (Eds). Climate Change: the Karst Record, 29. Lauritzen S-E, Kyselk J, Lovlie R, 1991. A new survey of Raggejavri-Raigi and the Hellemofjord karst, Norway. Cave Science 18(3), Trans. British Cave Research Association, 131. Poulton TP, Simony PS, 1980. Stratigraphy, sedimentology and regional correlation of the Horsethief Creek Group (Hadrynian, Late Precambrian) in the northern Purcell and Selkirk Mountains, British Columbia. Canadian Journal of Earth Sciences, 17, 1708. Rollins J, 2004. Caves of the Canadian Rockies and Columbia Mountains. Rocky Mountain Books ISBN 0-921102-94-1, 336 pages. Table 1. Data on stripe karst caves discussed in text.* After Lauritzen et al (1991). Estimated length (m) 2,500+ 3,000+ 5,500 6,700 9,000 2,000+ Estimated depth (m) 398 60 285 483 371 580 Marble width (m) 50 110 300 20 20 20 Marble dip () 20 55 40 45 45 65 Sink elev. (m) 1,443 1,440 1,708 1,748 1,636 580 Resurge elev. (m) 1,045 1,380 1,423 1,265 1,265 0 Sink to resurg. (m) 2,220 2,040 910 1,940 2,630 790 Hydraulic gradient 0.18 0.03 0.31 0.25 0.14 0.73 Stripe Karst Type 3 Type 2 Type 3 Type 1 Type 1 Type 1 White Rabbit Cody Cave Nakimu Caves Tupper-Rasp. Dygone-Rasp. RJR Norway* Cave/System Figure 9. Surveying carefully around soda straws in a side passage (note use of laser for distance). Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings162

PAGE 165

Safford K, 2010. Following the White Rabbit. BC Caver, vol. 24(4), 18. Rogers JP, Coates D, 1980. Cody Cave survey. Vancouver Island Cave Exploration Group 10(8). Thompson P, 1976. Cave exploration in Canada. A special issue of the Canadian Caver, 183 pages. Worthington SRH, Ford DC, Beddows PA, 2000.Porosity and permeability enhancement in unconfined carbonate aquifers as a result of solution. In: A Klimchouk, DC Ford, AN Palmer, W Dreybrodt (Eds.). Speleogenesis: Evolution of Karst Aquifers. NationalSpeleological Society, Huntsville, AL, USA, 463. Yonge CJ, 2013. Exploration of the Tupper SinkRaspberry Rising Cave System. Canadian Caver #77. In press.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings163

PAGE 166

study areas were located close to the United Arab Emirates (UAE) Oman state border, where the Musandam Mountains are over 2,000 m high. The locations of our explorations are shown on map (Fig. 1).2. GeologyOman and the UAE are located on the NE margin of the Arabian plate (Ricateau and Riche 1980; Kusky et al. 2005). This plate is bounded to the south and SW by the active spreading axes of the Gulf of Aden and Red Sea. On the east and west its border is marked by transcurrent fault zones. On the north the plate is marked by a complex continentcontinent to continentocean collision boundary along the Zagros and Makran fold and thrust belts. During the Cretaceous spreading of the Tethyan Sea, Gondwana Land continued its dispersal, and the Arabian-African plate drifted northward about 10. These events combined with the opposite rotation of Eurasia and Africa initiated the closing of the Tethyan during the Late Cretaceous. At the early stages of closure, downwarping of the Arabian continental margin combined with the compressional forces of closure from the Eurasian plate initiated obduction of the Tethyan oceanic crust along preexisting transform faults, and still hot oceanic crust was detached along oblique northeast dipping thrust faults. The Musandam Mountains are built up of several major structural units ranging in age from Precambrian to Miocene. These include a pre-Permian Basement, the Hajar Unit, the Hawasina Nappes, the Samail Ophiolite and metamorphic sole, and the postnappe structural units. The main structure of the Musandam Mountains is a complex interaction of Late Cretaceous thrust sheets folded over the Musandam mid-Tertiary culmination, with two foreland basins developed in front of the rising mountains (Searle 1988). The Musandam Carbonates were deposited in shallow water on the Arabian continental margin. The Samail Ophiolite represents a portion of the Tethyan ocean crust formed at a spreading center of Middle Cretaceous age (Cenomanian). During the late Cretaceous the Musandam Carbonates and Semail Ophiolites wereCAVE AND KARST PROSPECTION IN RAS AL-KHAIMAH MOUNTAINS, NORTHERN UNITED ARAB EMIRATENadja Zupan Hajna1, Asma Al Farraj Al Ketbi2, Franci Gabrovek1, Metka Petri 1, Tadej Slabe1, Martin Knez1, Janez Mulec1 1Karst research Institute ZRC SAZU, Titov trg 2, Postojna, Slovenia, zupan@zrc-sazu.si2United Arab Emirates University, Geography Department, P.O. Box: 17771, Al Ain, United Arab Emirates, asma@uaeu.ac.ae In January 2011, we were visiting Ras Al-Khaimah Mountains in the Northern Emirate of United Arab Emirates with the purpose to find some new caves. There are several big caves known in Oman, but none really big in the UAE. Several areas in Ras Al-Khaimah emirate on the Musandam Peninsula were surveyed, particulary in the northern part of Oman Mountain range near the border with Oman. We found several small caves on the northern slopes of Mount Al Jeer, at the entrance to Wadi Haqil and at different locations in Wadi Al-Bih and Wadi Taweeyan and also in its tributary Wadi Maai. Unfortunately we didnt find any big caves, almost all were short horizontal caves on the wadi slopes and the longest one was not even formed by solution, but was an opening along the relaxation fissure.1. IntroductionThere were many studies done on caves and karst in different areas of the Arabian Peninsula. The most extensive research has been done in Saudi Arabia (Edgell 1990; Forti et al. 2005; Kempe and Dirks 2008) and Oman where the biggest caves of the region are known. In Oman for instance were done several studies on speleothems; especially on dating and climatic changes (Fleitmann and Matter 2009; Burns et al. 1998, 2001; Sadiq and Nasir 2002). But karst and caves in the United Arab Emirates are not very much explored. There have been only few research expeditions and the existing studies have been mainly undertaken by the Emirates Natural History Group in Jebel Hafit (Jeannin 1992; Waltham and Jeannin 1998; Fogg et al. 2002; Aspinall and Hellyer 2004). Here are presented preliminary research results of field observations in January 2011 when our group explored and mapped few new caves and other karst features on the western slops of Musandam Mountains, which are northern part of Oman Mountains, in Ras Al-Khaimah Emirate. Our Figure 1. Locations of our studied areas in Ras Al-Khaimah Emirate in UAE (source Google Earth).Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings164

PAGE 167

tectonically shifted over the Hawasina Unit (Glennie et al. 1974). The Musandam Mountains are separated from the rest of the Oman Mountains by the very distinct Dibba fault zone. The emergence and uplift of the Musandam Peninsula in the Miocene caused the initiation of the drainage network in the northern UAE. However, the UAEOman mountain range has only been a barrier for the last fifteen to twenty million years, which has also influenced the climatic conditions there. In the Pleistocene, huge terraces of alluvial sediments accumulated in wadis at the feet of the Musandam Mountains (Al-Farray and Harvey 2000). The limestone mountain ranges are dissected by wadis of various sizes ranging from smaller gorges to large valleys, which present a unique image of fluvial relatively densely dissected karst. Unique caves with characteristic rock relief forms developed in their slopes. Our studied areas were dominated by carbonate rocks of the Musandam Group (Lower, Middle, Upper Musandam limestone) of Upper Jurassic to Lower Cretaceous age, but carbonates of the Russ al Jibal Group (Bih dolomites, Haqil limestone, Ghail limestones) of Middle Permian to Middle Triassic age and of the Elphinstone Group (Milaha limestone, Ghalilah limestones and mudstones) of Upper Triassic to Lower Jurassic age may also be found (Hudson and Chattan 1995; Hudson 1960). Thirty-nine samples were taken at 6 locations in the wider surroundings of the city of Ras Al-Khaimah to make fiftynine microscopic thin-sections that were examined from the petrological and lithostratigraphic aspects. All samples were also subjected to calcimetric analyses. The main result was that the samples are made of shallow sea carbonates which were partly recrystallized and dolomitized.3. Hydrogeological characteristics and water qualityHydrogeological characteristics of the UAE are significantly influenced by low amount of precipitation and high potential evaporation. The annual precipitation ranges between 15 mm in very dry years to 600 mm in extremely wet conditions. Practically all rain water drains as surface flow in a form of seasonal floods, and according to some estimation only 2 to 3% of it recharges the groundwater (Rizk and Alsharhan 2003). Groundwater is stored in 4 main types of aquifers: karst in carbonate rocks, fissured in ophiolites, and intergranular in gravel and sand dunes. For the wadis karst aquifers of rocky slopes and intergranular aquifers of alluvial deposits are characteristic. The main source of recharge of the intergranular aquifers in Quaternary alluvial deposits, which are composed of gravel and sand with thin interbeds of silt and clay (Rizk and Alsharhan 2003), is surface water that drains in the wadis. The karst aquifers are composed of limestone and dolomite. They are recharged by infiltration of rain and discharged through karst springs at the contact with less permeable rocks or directly into the sea. During our field work the karst springs named Khatt and MeBreda, which differ significantly in their hydrogeological characteristics, were studied in more detail. For the comparison only, the basic physical and bacteriological parameters of samples taken at the pumping station in an intergranular aquifer and in a small collector of surface water in a rocky slope of the wadi Haqil were defined additionally. The Khatt spring with two main outflows Khatt north and Khatt south is located 15 km south of the town Ras AlKhaimah at the extreme east of the gravel plains at the foot hills of the Northern Oman Mountains. The gravel aquifer in the plain is overlying very poorly permeable rocks of Cretaceous age (marls and shales with varying admixtures of coarse detrital debris of chert, basic igneous rocks, and limestone). The karst recharge area of the spring is composed of dolomitic limestone and limestone of Jurassic and Cretaceous age. In the period from 1979 to 2004 the average annual discharge of the Khatt north spring ranged between 1 and 60 l.s-1, and of the Khatt south spring between 4 and 55 l.s-1(Wycisk et al. 2008). The oscillation of discharges clearly reflects the influence of the amount and distribution of precipitation. In dry years with low amount of rain the discharges were significantly lower. The spring with the water temperature of approximately 39 C is used for balneological purposes. In the MeBreda area in the rocky slope of the Wadi Shehah (a tributary to the Wadi Al-Bih) a karst spring is situated at the altitude of 710 m a.s.l. The name MeBreda is used also for the spring. At the point of a small, permanent outflow of water through a fissure in dolomite rock a group of local people built in 1925 two pools for collecting water. The location of the spring is most likely conditioned by the existence of a less permeable zone within the carbonate sequence. This enables the water storage in a small perched aquifer, which recharges a perennial spring with the discharge of several l.s-1. Water flowing out of the fissure is first collected in the upper pool (our local guide used it as drinking water), and then flows in the lower pool (water used by goats and for irrigation of a small plantation). Several small holes, which are artificially arranged as collectors of rain water, were seen in the rocky slope of the Wadi Haqil. One of them was partly filled with water in the time of our visit. Stored water is probably used by herds and goats as drinking water. In the nearby pumping station the groundwater is pumped from a gravel aquifer and used for irrigation of agricultural land, mostly for cultivation of date palms. At all described locations we measured physical and chemical parameters and did basic bacteriological analyses, because drinking water in arid areas is of great concern. Already the comparison of basic physical parameters shows very different types of analyzed waters. The EC value of groundwater from the gravel aquifer is very high due to salinization. In the small collector of rain water in the rocky slope of the Wadi Haqil stagnant water was sampled, which has been altered by some additional influences. The results show that although both, the MeBreda and Khatt springs, are recharged from karst aquifers, they differ significantly. In the MeBreda spring the anions arranged Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings165

PAGE 168

4. Characteristics of explored cavesEven though the geology of the area has been studied thoroughly, few references report on caves and karst phenomena in UAE (Jeannin 1992; Waltham and Jeannin 1998; Fogg et al. 2002). Our primary target was mountainous region in the NE part of the emirate Ras AlKhaimah, where some potential caves had been reported. Some of these caves came out to be small cavities or even rock shelters. As information on our work was well spread several more caves and areas with caves were reported to us by locals. Nevertheless, extended caves are not common for the area. Arid climate and local geology with relatively impure limestones and marls do not favor formation of big cave systems. Climate in Pleistocene however resulted in higher accumulation of clastic sediments in river terraces and diamicts at the wadis bottoms, probably also in some caves. Such sediments have been found in all visited wadis (Wadi Al-Bih, Wadi Taweeyan and Wadi Maai). We first explored caves on the very eastern border with Oman on the slopes of Mount Al-Jeer where the local people knew of several cave entrances that no one had previously entered. The surface of the rock was finely and shallowly corroded, and we found several solution pans with outflow channels. We examined many small passages according to their decreasing concentrations are HCO3 -> SO4 2-> Cl-> NO3 -, and in the Khatt spring HCO3 -> Cl-> SO4 2-> NO3 -. The springs have significantly different T, EC, and concentrations of anions. Relatively low EC, the chemical composition of water, and the ratio Ca2+/Mg2+=1.6 in the MeBreda spring indicate an outflow from a carbonate aquifer with a larger share of dolomite. Increased concentrations of nitrates are due to grazing goats around the spring. According to its high temperature of 38.5 C the Khatt spring can be characterized as a thermal spring with deep circulation of groundwater. For other springs in the northern part of the UAE the temperatures of springs range from 25 to 32 C (Wycisk et al. 2008). Viable microbes on Ridacount test plates were retrieved from all sampling sites except from water pumped from the underground which had high conductivity. Surprisingly, the traditionally collected rainwater in natural crack had low number of total bacterial count; however detected E. coli indicated probable fecal contamination of water from pasturing goats. Considering two ISO standards SIST EN ISO 9308-1 (coliforms and E. coli) and SIST EN ISO 6222 (total bacteria) none of the water from any of the tested sites matched the ISO criteria for direct human consumption.Figure 2. A) The slopes above Wadi Maai with numerous opening to the small caves; B) entrance to the largest cave; C) plan view (left) and extended elevation (right) of the cave (Photos and map Karst Research Institute ZRC SAZU). Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings166

PAGE 169

along bedding planes that had developed in thin-layered limestone that alternated in places with layers of dolomite. All the passages ended after a few meters, and all narrowed characteristically from the entrance inwards. Their floors were covered with material from the walls of the passages, light grey clay, bones, and larger plant seeds. North East from Ras Al-Khaimah a large number of small cavities and caves have been found in a hill at the entrance of Wadi Haqil. Small ridge, which rises about 45 m high from the wadi bottom is intersected by numerous cavities, small conduits and roofless caves. Looking as whole, the hill looks like a block of limestone characterized by vuggy porosity. However, caves are short, cavities unlinked and isolated. There was only one cave longer than 10 m. It is a relatively simple channel with oval cross-section. Some of the channels are fully and some partially filled with yellow clastic sediments. Also few layers of flowstone between clastic sediments and many calcite crystals were detected. Few samples were after analyzed by x-ray diffraction method in the Laboratory of Physical Methods at Institute of Geology ASCR in Prague. All samples contained quartz, gypsum and kaolinite, almost all of them also calcite, palygorskite, smectite and illite. Calcite and gypsum represent precipitates from the cave (crusts, crystals, cements) and other minerals were brought to the cave most probably by water or wind from eroded rocks and/or soils from wadi catchment area. In caves in Wadi Haqil mousetailed bats (Rhinopomatidae) were observed which belong to the species Rhinopoma cystops. Along Wadi Taweeyan, for instance in the area of Jebel Al Khab and tributary wadi Maai (Fig. 2), many small cavities can be observed in several places. More or less all were short horizontal channels with some rest of fluvial sediments. In Wadi Al-Bih we checked two tributary wadis. First was wadi Shie Haft, where was 50 m above river located bigger cave entrance. But unfortunately after few meters cave finished. There was big accumulation of debris and clays in the cave bottom. A cave close to the top of MeBreda, above Wadi Shehah which is also tributary to Wadi Al-Bih, at 825 m above sea level, was very distinct from the others found. The cave follows a set of vertical relaxation fissures with absolutely no solutional features. A block of the massive, with well expressed fold, has separated from the plateau and is sliding into a valley (Fig. 3). The only sediment in the cave is breakdown material. We have measured about 70 m of the cave. Temperature in the cave was 17.5 C, relative humidity 57.8% and carbon dioxide concentration 380 ppm (January 27, 2011). 4.1. Rock relief of wadis and caves Characteristic rock relief dissects the circumference of the special caves forming in the walls of the wadis that were studied comprehensively for the first time. The slopes of the wadis are often dissected by large recesses, relatively rare subsoil forms, and karren with microrills and rain flutes. Karren-like rock surface also occurs between wadis, and rock forms are found on rocks that have broken off from the slopes as well as on and in the sediments covering the bottoms of the wadis. More extensive karren are found on the tops of the mountains. The rocky riverbeds of wadis were shaped by rapid water currents and corrosion at the contact with sediment. Rock forms are an important trace of the formation and development of caves and karst surfaces. Caves in wadis are located just above the sediment level and are therefore periodically flooded and dry at various levels above it. They also share similar rock relief. Ceiling pockets are found on the ceilings of niches that developed at the contacts of passages and at their sharp bends in areas of emphasized swirling of water. As a rule, their diameters do not exceed one meter but they are usually deeper, with a 1:3 ratio, and wider in the upper part. In places, pockets are also found on walls, but their lower parts are flat and their upper parts are vaulted. The lower parts were frequently covered by sediment that protected the rock from dissolving and erosion and therefore grew upwards and sideways. The walls of passages are smooth and rounded. The mechanical erosive action of water currents carrying material appears to play an important role in their formation. The ceilings of passages are usually covered by above-sediment channels that along with the paragenetically raised cross sections of passages indicate the frequent filling of caves with sediments and the flow of smaller quantities of water above them. The largest reach a depth of one meter, the consequence of the relatively long-term formation of caves of this kind and their filling with sediment. Along-sediment wall notches, up to one meter deep, indicate periodic partial filling of the caves and a long lasting sediment level in the cave. Along-sediment rock forms are the trace of relatively frequent filling of wadis with sediment and its removal.5. Discussion and ConclusionsIn our studied, very arid area, with an average annual precipitation only slightly above one hundred millimeters, of which only a small percentage recharges the groundwater, karst surface forms, caves, spring caves, and springs are rare. The described caves along with several smaller ones show some common characteristics that could point to their origin. The passages are generally round with well expressed ceiling channels, wall and ceiling pockets, and dead ends. There are no obvious flow paths along which caves developed. In some areas the vuggy porosity resembles that found in caves of hypogene origin. However, we did not observe any other typical indicators of hypogene origin. The dimensions of passages decrease from the entrance toward the interior of caves. Most caves end with solid walls with no leads continuing into the massif. Scallops are rare or non-existent. Detailed geological studies have shown that the significant factors for the karstification and the growth of the caves in the studied wadis were like everywhere: bedding planes and slips along them, older tectonically broken zones in the carbonate rocks, occurrence of large concentrations of calcite, contacts between dolomitized and non-dolomitized rocks, and last but not least, recent fissures and faults. Karst and Caves in Carbonate Rocks, Salt and Gypsum oral2013 ICS Proceedings167

PAGE 170

change of the sediment level. The majority of caves, including those currently located high above any sediment, were filled with sediment over a longer period. The walls on the sides of wadis commonly exhibit individual or linked cavities that resemble tafoni of various sizes. Some of these, although rare, are large enough to enter. The general morphology of caves can be linked to floodwater patterns (e.g., Palmer 2007). The flow patterns in bed sediments of wadis are known to be very complex. The carbonate bedrock at the side of the wadis is therefore exposed to dissolution by both superficially flowing water and water flowing within the sediments. The dissolution could occur along the sediment-bedrock contact resulting in cave geometry defined by the combined action of dissolution and sediment morphology. This suggests that caves could form as the wadis develop. However, this hypothesis needs further field and theoretical verification. At this stage we do not have firm arguments for any mechanism behind the origin of these caves; however, some of the characteristics mentioned above indicate that the caves originated from near-surface processes. We found and studied mainly clastic sediments. Speleothems are rare, probably due to the arid climate and the very low and diffuse infiltration from the surface. Sediment samples differ in content due to their origin in different rock. Along with fluvial cave sediments, eolian sediments and sediments of organic origin (guano, bones, seeds) are also present. Cave rock relief and surface rock relief indicate the major significance of along-sediment and subsoil formation of the wadis and the karst between them. We can observe the development of wadis through extensive filling with sediment and rocks from the slope, the shaping of rock at the contact with sediment, and the emptying or distinct Figure 3. Mountaine at MeBreda above Wadi Shehah A) broken and fissured slope above the cave; B) entrance to the cave marked by an arrow on Fig. 3A.; C) extended elevation of the cave; D) map of the cave. (Photos and maps Karst Research Institute ZRC SA ZU). Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings168

PAGE 171

The lack of fluvial sediments in the caves surely raises doubts about the genesis. One possible explanation of why there are not many caves in the studied area of Ras Al-Khaimah is that little water penetrates the steep karst slopes of wadis, which does not create ideal conditions for the development of caves, especially in recent climate conditions. The annual precipitation values here are very low and most of the precipitation falls in the form of downpours causing major surface runoff in the mountainous areas due to large inclination of slopes and the thin-layered impure limestone with intermediate marl layers.AcknowledgmentsIn January 2011, at the invitation of Prof. Dr. Asma AlFarraj, a team from the Karst Research Institute at the Research Centre of the Slovenian Academy of Sciences and Arts went to study the karst and caves in the mountainous regions of the northeastern Emirate of Ras Al-Khaimah in the United Arab Emirates. The research was also supported by Saud bin Saqr Al Qasimi, Sheikh of the RAK Emirate. The research was supported also by the research program Karst research financed by Slovenian Research Agency.ReferencesAl-Farraj A, Harvey AM, 2000. Desert pavement characteristics on wadi terrace and alluvial fan surfacesWadi Al-Bih UAE and Oman. Geomorphology, 35 (2000), 279. Aspinall S, Hellyer P, 2004. Jebel Hafit, A Natural History. Abu Dhabi. Burns SJ, Fleitmann D, Matter A, Neff U, Mangini A, 2001. Speleothem evidence from Oman for continental pluvial events during interglacial periods, Geology, 29, 623. Burns SJ, Matter A, Frank N, Mangini A, 1998. Speleothem-based paleoclimate record from northern Oman. Geology, 26, 499. Clesceri LS, Greenberg AE, Eaton AD, 1998. Standard methods for the examination of water and wastewater. 20thEdition, American Public Health Association, Washington. Edgell HS, 1990. Karst in Northeastern Saudi Arabia. J.K.S.A.U, Earth Science, Special issue: 1st Saudi symposium on Earth Science, Jeddah, 3, 81. Fleitmann D, Matter A, 2009. External geophysics, climate and environment, The speleothem record of climate variability in Southern Arabia. C. R. Geoscience 341, 633. Fogg T, Fogg P, Waltham T, 2002. Magharet Jebel Hafit a significant cave in the United Arab Emirates. Tribulus Journal of the Emirates Natural History Group. 12, 1, 5. Forti P, Galli E, Rossi A, Pint J, Pint S, 2005. Cave Minerals Of Some Limestone Caves Of Saudi Arabia. Hellenic Speleological Society. Research made within the MIUR 2002 Project Morphological and Mineralogical Study of speleothems to reconstruct peculiar karst environments Resp. 21 August 2005, Kalamos, Hellas. Glennie KW, Boeuff MGA, Hughes-Clarke MW, Moody-Stuart M, Pilaar WHF, Reinhart BM, 1974. Geology of the Oman Mountains. Kon. Ned. Geol. Minnhoukundia Genoot. Vern., 33, 423. Hudson RGS, Chattan M, 1959. The Musandam Limestone (Jurassic to lower Cretaceous) of Oman Arabia. Notes Memoirs. Moyen-Orient., 3, 69. Hudson RGS, 1960. The Permian and Trias of the Oman peninsula, Arabia. GeologicaMagazine, 97, 299. Jeannin P-Y, 1992. Expedition suisse aux Emirats Arabes Unis. Stalactite (Switzerland), 42(1), 47. Kempe K, Dirks H, 2008. Layla Lakes, Saudi Arabia: The Worldwide largest lacustrine Gypsum Tufa. Acta Carsologica, 37(1), 7. Kusky T, Cordula Robinson C, El-Baz F, 2005.Tertiary Quaternary faulting and uplift in the northern Oman Hajar Mountains. Journal of the Geological Society, 162, 871. Palmer AN, 2007. Cave Geology. Cave Books, Dayton, OH. Ricateau A, Riche PH, 1980. Geology of the Musandam peninsula (Sultanate of Oman) and its surroundings. Journal of petroleum geology, 2(3), 139. Rizk ZS, Alsharhan AS, 2003. Water resources in the United Arab Emirates. In: Alsharhan AS, Wood WW (Eds.) Water Resources Perspectives: Evaluation, Management and Policy, Elsevier, 245. Sadiq AM, Nasir SJ, 2002. Middle Pleistocene karst evolution in the State of Qatar, Arabian Gulf. Journal of Cave and Karst Studies, 64(2), 132. Searle MP, James NP, Calon TJ, Smewing JD, 1983. Sedimentological and Structural evolution of the Arabian continental margin in Musandam Mountains and Dibba zone, U.A.E. Geological Society of America, Bulletin, 94, 1381. Waltham AC, Jeannin PY, 1998. Forum: Caves in the United Arab Emirates. Journal Cave and Karst Science, 25(3), 149. Wycisk P, Al Assam M, Akram S, Al Mulla M, Schlesier D, Sefelnasr A, Al Suwaidi NB, Al Mehrizi MS, Ebraheem A, 2008. Three-dimensional geological and groundwater flow modelling of drought impact and recharge potentiality in Khatt springs area, Ras Al Khaimah Emirate, UAE. In: WASTA 8th Gulf Water Conf., Bahrain, Proceedings, 1.Karst and Caves in Carbonate Rocks, Salt and Gypsum oral 2013 ICS Proceedings169

PAGE 172

3. ResultsThe paper presents two examples of northern Velebit, and one which is in the border area of northern and central Velebit. In the area of Jezera ( Lakes in Croatian, ~1,400,500m a.s.l., Fig. 1) it was noticed that the moraine material (which is relatively impermeable) affected the appearance of small occasional streams. These streams cause cutting in small torrents and increase the surface erosion processes in karst environment.In the areas whereINFLUENCE OF THE PLEISTOCENE GLACIATIONS ON KARST DEVELOPMENT IN THE DINARIDES EXAMPLES FROM VELEBIT MT. (CROATIA)Neven Boi1, Sanja Faivre2, Marijan Kova i3, Nada Horvatini4 1University of Zagreb, Faculty of Science, Department of Geography, Maruli ev trg 19/II, HR-10000 Zagreb, Croatia, nbocic@geog.pmf.hr2University of Zagreb, Faculty of Science, Department of Geography, Maruli ev trg 19/II, HR-10000 Zagreb, Croatia, sfaivre@geog.pmf.hr3Marijan Kovai, University of Zagreb, Faculty of Science, Department of Geology, Horvatovac bb, HR-10000 Zagreb, Croatia, mkovacic@geol.pmf.hr4Nada Horvatini, Ruer Bokovi Institute, Division of experimental physics, Laboratory for Measurements of Low-level Radioactivity, Bijeni ka 54, HR-10000 Zagreb, Croatia, Nada.Horvatincic@irb.hr Traces of the Pleistocene glaciations in the Dinarides have attracted attention of many researchers. Thus in the Velebit Mt. area in Croatia numerous surface traces of ice influence were explored (destructive and accumulation). As Velebit is dominantly composed of carbonate karstified rocks the interaction between glacial and karst processes is expressed. Such phenomena are visible on the surface and in the karst underground. The aim of this paper is to explore this interaction on three examples in northern and central Velebit Mt.1. IntroductionVelebit Mt. is a part of the Dinaric karst mountain range. It is located in Croatia and extends to the length of 145 km along the northern part of the eastern coast of the Adriatic Sea. The highest peak is 1,757 m high. Velebit mostly consists of karstified carbonate rocks (limestone, dolomite and breccias consisting of carbonate rock fragments). Former glaciation of this region has long attracted researchers (e.g., Hranilovi 1901; Gavazzi 1903). Studies have shown that a large part of the area was affected by the Pleistocene glaciations. The best preserved glaciations traces in landscape are from the last glaciation period Wrm.Studies of these glaciations were more detailed in the southern Velebit area (e.g., Nikler 1973). Glaciation of north Velebit was first studied by Bauer (1935), while the central Velebit glaciations were less known. Research of the glaciations of northern and central Velebit were particularly intensified during the past twenty years (Bognar et al. 1991; Bognar and Faivre 2006; Veli et al. 2011). In those works they found morphological traces of different types of glaciers, and the surface morphology evidence of their propagation (denudation phenomena and material accumulation in moraines). These results have prompted questions about the impact of glaciation on the development of karst, especially the underground karst formations caves in this area. This question has been studied so far in the areas outside the Dinarides (e.g., Glover 1977; Kunaver 1983; Mylroie 1984; Audra 2000; Woodward and Goldberg 2001).2. Aims and methodsThis paper explores the impact of the Pleistocene glaciations on the karst development in northern and central Velebit. Special emphasis has been given to the research of the impact of the glaciations on the development and morphology of the caves in the area (Fig. 1). The main methods used in this study were geomorphological mapping and speleological research. Also, some sedimentological methods were used and dating of the flowstone crust was provided. It was sampled and dated by the 14C method in the Laboratory for Low-Level Radioactivity at Ruder Boskovic Institute. The measurements were carried out with a Liquid Scintillation Counter, Quantulus 1220, using the benzene synthesis method (Horvatin i et al. 2004). Figure 1. Location of the researching sites: 1 Jezera area; 2 Ledena jama u Lomskoj dulibi (Ledena Ice Cave); 3 tirova a Ice Cave.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings170

PAGE 173

the moraine cover is thinner or missing rekarstification of bedrock is visible, and occasional short streams sinking at the contact zone of the moraine material and bedrock (Jurassic carbonate rocks). In this way several small blind valleys were formed. In the glacial valley Lomska duliba (Fig. 1) is the Ledena jama (Ledena Ice Cave), 536 meters deep, characterized by ice accumulation thicker than 50 m. During this study we mapped the moraine forms along the valley and found that the same material plugged entrance part of the cave (Bo i et al. 2012a). The entranceto the caveislocatedon the route ofthe formerglacier at 1,256 m a.s.l.).Morphologyand sedimenttracesindicate thatthe water flow (from melting ice) was sinking into the cave.So, the entrance partof the cave was filled with glacial andfluvio-glacial materials.In this way, the formercanyonof the incomingsurface water flow is now completelyfilled (Fig. 2). Based on theresidue of moraine material one can concludethat the entrance part ofthe cavewasfilled with more quantity of that material and perhaps even thewhole entrance part of the cave was completelyfilled.In thecave channel thatis belowthese materials (at the depth of 60 m) the sample ofspeleothem was taken and datedwith 230Th/234Umethod. Therewas assessed theage of301,000 55,000yearsBP (Jelini et al., 2001). Itsuggeststhatglaciations older thanWrm also had the influence on the genesisof the cave. Figure 2. Ledena jama u Lomskoj dulibi (Ledena Ice Cave): A Entrance with moraine material, erratic and buried canyon, B Survey of the cave, GF/G glaciofluvial and glacial material, E erratics (cave topography according to Jelini et al. 2001).Great influence on the development of glaciation surface and underground karst forms was studied in tirova a (middle Velebit area). Here is the 351 m long tirova a Ice Cave with abundant erosional remnants of the fluvioglacial deposits that fully filed much of the known part of the cave in the past (Bo i et al. 2012b). Coarse-grain clastic sediments, generally rarely present, have been discovered in this cave (Fig. 3). They occur as erosional residues and pebbles are cemented with calcitic cement. This sediment is defined as a coarse-grained, moderately to weakly sorted, porous, polymictic ortoconglomerate, and formed by sedimentation resulting from polyphased torrential flows. We may presume that we are dealing with sediments of fluvioglacial origin, i.e. with material that was transported from the fluvioglacial fans, and which was finally sedimented in the caves channels. On the northern edge of the cave another sedimentary body has been found, 20 cm thick, which has been defined as limestone debris. The sediment is mainly composed of unrounded limestone fragments, rarely of sandstone fragments and pelitic pebbles. The material is moderately sorted. The upper part of this sediment body is covered with a flowstone layer. The presented results show that all the sedimentsare of different ages.Based on the age of the flowstone crust of 8,230 150 years BP, as well as on its position, it may be concluded that coarse-grain clastic sediment and limestone debris are older. Based on the sedimentological properties, primarily on their cementation properties, it can be assumed that limestone debris is significantly younger then the fluvioglacial fan material. This indicates that the fluvioglacial fan cannot be related to the Wrm/Holocene transition, although we may presume that the limestone debris was deposited at that time. During cold periods, glacial exaration produced a particularly significant quantity of sediments. Consequently, the underground karst conduits situated below the snow line during the Pleistocene were filled with high-energy sedimentary structures, glacial and fluvioglacial sediments. That reduced or even stopped their conduit function, and led to difficulties in the drainage of meteoric water, to lake formation within morphological depressions and to a rise in the local water-table level. With the denudation of these deposits, conduit reactivation and a drop in the underground water level occurred. Stabilisation during milder climatic conditions led to aggradation of the cave and to deposition of the flowstones from 8,230 150 years BP. Figure 3. The coarse-grain clastic sediments in the tirova a Ice Cave.4. ConclusionsIt can be concluded that the Pleistocene glaciation had multiple effects on the development of karst in this area. Moraine deposits, relatively impermeable, promote local development of surface runoff and erosion. In this way, they temporarily slow down the process of karstification below, but because of the contact process they accelerate karstification along their edges. Moraine-derived material can also be accumulated in the caves, directly or after resedimentation as fluvioglacial material. In this way it fills the underground channels and then reduces and hinders drainage function of these channels. These studies are being continued. Karst and Caves in Carbonate Rocks, Salt and Gypsum poster2013 ICS Proceedings171

PAGE 174

ReferencesBauer B, 1935. Uber des Nordlichen Velebit.Jahresber des Bundesrealgymnasium in Knittenfeld, 35. Bo i N, Faivre S, Kova i M, Horvatin i N, 2012a. Uloga pleistocenske oledbe na razvoj kra na podru ju Velebita. Znanstveno-stru ni skup Posebne vrijednosti dubokog kra, Krasno, 21.. 4. 2012. Book of Abstact, 27. Bo i N, Faivre S, Kova i M, Horvatin i N, 2012b. Cave development under the influence of Pleistocene glaciation in the Dinarides an example from tirova a Ice Cave (Velebit Mt., Croatia). Zeitschrift fr Geomorphologie, 56 ( 4), 409 433. Bognar A, Faivre S, Paveli J, 1991. Tragovi oledbe na Sjevernom Velebitu. Geografski glasnik, 53, 27. Bognar A, Faivre S, 2006. Geomorphological Traces of the Younger Pleistocene Glaciation in the Central Part of the Velebit Mt. Hrvatski geografski glasnik, 68(2), 19. Gavazzi A, 1903. Trag oledbe na Velebitu. Glasnik Hrvatskog naravoslovnog drutva, 14, 174. Glover RR, 1977. A conceptual model of cave development in a glaciated region. 7thInternational Congress of Speleology, Proceedings, 220. Horvatin i N, Barei J, Krajcar Broni I, Obeli B, 2004. Measurement of low 14C activities in Liquid Scintillation Counter in the Zagreb Radiocarbon Laboratory. Radiocarbon, 46, 10516. Hranilovi H, 1901. Geomorfoloki problemi iz Hrvatskog krasa, Glasnik Hrvatskog naravoslovnog drutva, XIII(1), 93. Jelini I, Horvatini N, Boi V, 2001. Ledena jama u Lomskoj dulibi. Senjski zbornik, 28, 5. Kunaver J, 1983. Geomorphology of the Kanin Mountains with special regard to the glaciokarst. Geografski zbornik, 22 201. Mylroie JE, 1984. Pleistocene climatic variation and cave development. Norsk Geografisk Tidsskrift, 38(3), 151. Nikler L, 1973. Novi prilog poznavanja oledbe Velebita. Geoloki vjesnik, 25, 10912. Veli J, Veli I, Kljajo D, 2011. Sedimentary bodies, forms and occurrences in the Tuderevo and Mirevo glacial deposits of northern Velebit (Croatia). Geologia Croatica, 64(1), 1. Woodward JC, Goldberg P, 2001. The sedimentary records in Mediterranean rockshelters and caves: archives of environmental change. Geoarchaeology: An International Journal 16, 327.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings172

PAGE 175

CONTRIBUTION OF THE KARSTIC ENVIRONMENT ON THE ORIGIN OF THE COLLOPHANITES OF ITATAIA URANIUM-PHOSPHORUS DEPOSIT, BORBOREMA PROVINCE, BRAZILJos Adilson Dias Cavalcanti1, Cesar Ulisses Vieira Verssimo2, Maria Dulcinea M.R. Bessa3, Clovis Vaz Parente4 1Geological Survey of Brazil, Av. Antnio Sales, 1418, Fortaleza-CE, Brazil, jose.adilson@cprm.gov.br2Federal University of Cear, Pici Campus, Fortaleza-CE, Brazil, verissim@ufc.br3Geological Survey of Brazil, Av. Antnio Sales, 1418, Fortaleza-CE, Brazil, dulcinea.bessa@cprm.gov.br4Federal University of Cear, Pici Campus, Fortaleza-CE, Brazil, clovis@ufc.br The U-P Itataia deposit, which occurs in the northern portion of the Borborema Province, northeastern Brazil, has the largest uranium reserves in the country. It is a phosphate deposit which has byproduct uranium. However, there is no uranium mineral. All uranium content is in the apatite structure in the microto cryptocrystalline masses called collophanites. The collophanites hosted in dolomitic marbles of Alcantil Formation of the Neoproterozoic age, belonging to the Cear Complex. These collophanites occur as massive bodies very porous filling brittle faults and fractures, fractures at the edges as part of calcite veins, the walls of open fractures with botrioidal habit, filling voids in epissienites dikes hosted in marble and as fragments filling karst features. The origin of the deposit remains enigmatic to this day. The genetic models for the source of the mineralizing fluids points to two main strands: hydrothermal with magmatogenic source or metamorphic-hydrothermal with sedimentogenic source, both with late supergene enrichment by karstification. The karstification phase is marked by a profile of supergene alteration that reaches up to 300 meters deep, with precipitation of collophanites in fracture zones in subaqueous environment of low temperature. The collophanites normally exhibit botryoidal habit characteristic of subaqueous environments. Its composed mainly of fluorapatite, but may contain newly formed calcite and fragments of preexisting minerals (feldspar, hornblende, biotite and opaque minerals). The content of P2O5may be above 25% and the contents of U3O8are averaged 0.23%.1. IntroductionThe Itataia uranium-phosphorus deposit was discovered in 1976 by NUCLEBRS during Canind Project, from a radiometric anomaly located in Santa Quitria, Ceara, northeastern Brazil. From the discovery were conducted geophysical surveys, geological mapping, 37,000 m of diamond drilling, 1,270 m of galleries and several wells. Total reserves of the deposit are estimated at 9 million tons of phosphate and 80,000 tonnes of uranium. In addition to the reserves of phosphate and uranium, the field also has potential for explotation about 32 million tonnes of marble. The richest ore has P2O5 content of 21.2%, intermediate 11.8% P2O5and the poor ore 2.5% P2O5is discarded in the tailings pile (Source: FIEC, 2008). From laboratory tests and pilot plant expected to yield a final concentrate of 34.4 % P2O5and 0.228% of U3O8(Vidal et al. 2005). The mineralization occur in the setentrional domain of the Borborema Province in metasedimentary sequence of passive margin of Neoproterozoic, metamorphismsed in amphibolite facies and individualized as Itataia Group which is formed by a sequence that is divided into four units, from bottom to top by: i) migmatites and gneisses, ii) quartzites, iii) migmatitic paragneisses, iv) marbles, Bt gneisses and calc silicates rocks (Figs. 1 and 2). Mendona etal. (1982) assigned the following names for these formations: Serra do Ce, Laranjeiras, Barrigas and Alcantil. The mains uranium-phosphorus bodies hosted in marbles of the Alcantil Formation. The Alcantil formation is composed for gray to white marble with green horizons of pure calcitic marbles or nearly pure, passing dolomitic marbles and even more impure types rich in pyroxene, hornblende, biotite-phlogopite, calcic plagioclase, garnet, quartz, titanite and scapolite, marking the gradual passage between the carbonate sequence and gneiss on the top. The presence of this lithological association could indicate that the mineralization is related with the deposition of marine or paralic restricted platform (Castro 2005). The Itataia phosphorus-uranium deposithascomplex geometry, buttwo bodieswere individualized. The main ore bodyofellipsoidal shapeoccurs at an elevation, with dimensions 800 m400 m anda depth of180 m, with east plunge.The otherore body is elongatedin the NW direction, withextensionof approximately800 mand depth ranging from100 mto200 m. Theorebodiesare collophanitesthat occurin different formsand associations, butpreferentiallyhostedinmarblesand calc silicates rocks (Castro etal. 2005). Regarding thetypology of collophanitesmain body, defined thefollowing types: solid; botrioidal, filling open fractures; fragmented, filling karst features(paleokarst); thinvein;and disseminatedinthehost rocksandepisyenites.2. Climate and Physiographic FeaturesIn Cear state, the tropical hot semi-arid occurs in 68% of its total area (IPECE 2008). Other climatic types also occur in the state in potions more restricted, as is the case in tropical hot semi-arid mild that occurs in the south, north and northwest of the state. Already, the tropical hot and humid and sub-humid tropical wet subquente, are restricted to the Baturit and Ibiapaba saws. The relief in the Cear were defined five geomorphological Karst and Caves in Carbonate Rocks, Salt and Gypsum poster2013 ICS Proceedings173

PAGE 176

performed at SGS Geosol, Brazil. We analyzed the major oxides (SiO2, TiO2, Al2O3, Fe2O3, MgO, CaO, MnO, Na2O, K2O P2O5) by fusion with lithium metaborate determination by ICP-ES and minor, trace and rare earth elements including (Ba, Be, Co, Cs, Ga, Hf, Nb, Rb, Sn, Sr, Ta, Th, U, V, W, Zr, Y, La, Ce, Pr, Nd, Sm, Eu, Gd, tb, Dy, Ho, Er, Tm, Yb, Lu) by fusion with lithium metaborate determination by ICP-MS. The elements Mo, Cu, Pb, Zn, Ni, As, Cd, Sb, Bi, Ag, Au, Hg, Tl and Se were held open by dilution with aqua regia and determination by ICP-MS.4. Results4.1. Field and Petrography The collophanites are epigenetics and confined mainly in marbles and secondarily in calc-silicate rocks belonging to metasedimentary sequence of sillimanite-garnet-biotite gneisses, marbles, quartzites and calc-silicate rocks, locally designated Itataia Group and regionally disigned Cear Complex (Fig. 2). It is a rock composed essentially of collophane (or microcristaline apatite), fine-grained, homogenous texture, reddish brown color, has cavities filled with powdery material and carbonates, exhibit massive, botryoidal and fragmented aspects (Figs. 3, 4, 5 and 6). These collophanites occur as massive lenticular lodes filling large cavities, and as veins and veinlets arranged in stockwork (Fig. 3), and filling open fractures, border of calcite veins in contact with marbles, and also in the main foliation planes and voids filling episyenites bodies. Marble filled fractures by collophanite of the main ore body are preferentially N5E and N40E (Alcntara e Silva, 2003). Another mode of occurrence is lithic fragments in breccias and filling cavities of karstic marbles (Fig. 6). Under the microscope, the collophanites has reddish brown, fine-grained (Fig. 7), light displays scores (carbonate and feldspar altered) and dark (amphibole, pyroxene and opaque minerals). It comes in a massive way, botryoidal, stockwork and veining. Often, displays or empty cavities filled with secondary material, such as calcite, clay or silica finely granulated. It consists essentially of a mass of microto cryptocrystalline brownish coloration of collophane that may contain feldspar, pyroxene, amphibole and accessory phases such as carbonate and titanite, apatite and opaque, subordinate. The other peculiar form of collophanite occurrence is filling voids in episyenites dikes. The episyenites rock is a coarsegrained to pegmatoide texture, vesicular, with pinkish to reddish mainly composed of albitized feldspar. The vacuoles, which in the literature are described, as a result of desilicification are usually filled by collophane and carbonate. In blade exhibit granular texture and consisting essentially of large crystals of albitized feldspars (80%) and subordinately for collophane. The opaque minerals occur in ancillary quantity and carbonate is secondary. The plagioclase feldspar is represented by albitized like crystals, altered to sericite and clay minerals. Apatite (collophane) has a dark color due to the presence of oxidizing material filling the interstices between larger crystals of feldspar and the edges of the cavities (Fig. 8). Calcite also occurs filling cavities and also as a product of alteration of plagioclase. units that are related to geological (lithology and structure), paleogeographic, climatic and environmental variables. These units are the coastal plain, the glacis pre-coastal, sertanejos highlands, waste massive and hinterland depressions (IPLANCE 1995). Making a correlation between morpho-structural compartments defined from SRTM image of the IBGE (www.ibge.gov.br) were earmarked by three compartments: hinterland depression with elevations below 400 m, the sertanejo plateau with elevations between 401 m and 600 m and massive waste with altitudes between 601 m and 850 m.3. MethodsThe field activities were conducted geological profiles, surface sampling and diamond core samples. Subsequently, thin and polished thin sections were prepared and were analyzed by optical microscopy and scanning electron microprobe. Geochemical whole rock analyzes were also performed for major oxides, trace and rare earth elements. The microprobe used was a Cameca SX50 with 4 WDS spectrometers and a spectrometer Kevex EDS, of the Mineralogy and Petrography Department, Braslia University. The whole rock geochemical analyzes were Figure 1. Setentrional portion of Borborema Province with the location of Itataia deposit (compiled and modified of Arthaud et al. 2008). Figure2. Geologic setting and localization of Itataia ore deposit (Modified of Angeiras, 1988).Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings174

PAGE 177

4.2. Mineral Chemistry Apatite is a generic term used to describe calcium phosphates with simplified formula Ca5(PO4)3X (where X= F, Cl, OH). Since a long time (1856), has been called fluorapatite, and hydroxyapatite cloroapatita, depending on the predominance of anion X-(Pasero et al. 2010). Currently, the IMA (International Mineralogy Association) has inserted several mineral species within the apatite supergroup, which was divided into 5 groups (Apatite, Hedyphane, Belovite, Britholite and Ellestadite). The apatite group is composed of apatite-(CaF), apatite-(CaCl) and apatite (CaOH), having the following idealized formula Ca5(PO4)3F, Ca5(PO4)3Cl and Ca5(PO4)3OH, respectively. The collophanite is a rock composed essentially of microcrystalline apatite. In Itataia deposit occur apatite(CaF), known as fluorapatite, with content of F between 4.1.6% and Cl <0.1% (Table 1). The content of MnO is less than 0.07% and Fe >1% and S (expressed as SO3%) <1%. The presence of Fe and S in the apatite may be related to the redox conditions of the mineralizing fluid, oxidizing the ore. SO occurs as SO4-2which is easily replaced by the (PO4)-3in the apatite (Pasero et al. 2010). In collophanites of Itataia SO4-2is relates with an increase of FeO. There is also a strong correlation between sodium (Na2O), strontium (SrO) and uranium (UO2). With increasing contents of Na2O and SrO, is also increased UO2. 4.3. Geochemistry The whole rock geochemical analyzes showed the phosphate content of collophanite that occurs almost always Figure 4. Detail of outcrop of marble of Alcantil Formation with collophanites filling fractures. Figure 5. Botryoidal collophanite on the top of main ore body of Itataia deposit. Figure 6. Paleokarst features in marble, such as conduits, filled with fragments colofanito in clay matrix.above 25% in the samples richest and around 20% in the samples poorest. The content of P2O5increasing with of increasing CaO and decreasing of SiO2. Already uranium (3,689,582 ppm) has positive correlation with Fe and Mg. The samples richer in MgO and Fe2O3, also have a higher content of uranium (Tables 2 and 3). In relation to REE, the collophanites are enriched by up to 100 times chondrite standard, planar displays, slightly enriched in light rare earth elements relative to heavy and has a mild anomaly of Eu (Fig. 9). In relation to the trace elements, have U and P positive anomalies and K, Zr, Ti negative (Fig. 10). Figure 3. Outcrop of marble of Alcantil Formation with collophanites filling fractures.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings175

PAGE 178

Figure 7. Microphotograph showing botryoidal collophanite associated with calcite veins hosted in marble. Figure 8. Microphotograph showing detail of botryoidal collophanite associated with calcite veins hosted in marble. Table 1.Mineral chemistry of apatite. Table 2.Lithogeochemistry of oxides in collophanites. Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings176

PAGE 179

5. Discussion and ConclusionsIn the study area, the intense magmatism that occurred in the period between the Neoproterozoic and Cambrian marked the evolutionary history of the Province Borborema. This story begins with the collision of cratons West Africa and So Luis, around 660 Ma, followed by magmatism of the Tamboril Santa Quitria Complex between 657 Ma, through high-grade metamorphism at 620 Ma, the anorogenic magmatism in around 470 Ma and the Rio Cear-Mirim magmatism related to the opening of the South Atlantic at 130 Ma (Brito Neves and Cordani 1991; Fetter 1999, Brito Neves et al. 2001, Hollanda et al. 2006). In this geological context of Brasiliano mobile belt that fits the Itataia deposit and other phosphorus-uranium occurrences of this region. The geological setting and structural control of the collophanites ore bodies are in brittle structures and infers different sources for mineralizing fluids. The genetic models existing on the deposit Itataia point to two main strands, a hydrothermal with magmatogenic source and metamorphic-hydrothermal with sedimentogenic source, both with supergene enrichment through the phenomenon of karstification. From the new studies, a new frame can be painted. In the proposed model, alkaline rocks associated with the final stages of consolidation Tamboril-Santa Quitria Complex (Neoproterozoic) are considered the primary source of phosphate and uranium. In the study area were described feldspathic affinity alkaline rocks rich apatite with contents of up to 10% P2O5and 0.08% U. These rocks have undergone processes of albitization, desquatzification (episyenitization) at temperatures between 550350 C (Cuney 2010), becoming the protore phosphate uranium-Table 3.Lithogeochemistry of REE in collophanites. rich. The processes of albitization and episyenitization correspond to the early stages of mineralizing process. In the study area were described sterile and mineralizaded albitites, and breccias locally enriched in phosphate (up to 15% P2O5) and uranium (2,300 ppm). The collophanites of Itataia deposit were to be formed later with the introduction of a hot spring, Rio Ceara-Mirim dykes, related to rifting on a global scale and activity hotspot during the opening of the South Atlantic and Equatorial. Based on thermo-geochronological studies of fission track in apatite, Netto et al. (1991) obtained the age of 91 Ma for the formation of collophanites associated with a Cretaceous thermal event. During this event, great faults of gravity and combined fractures that became the main sites of ore deposition. This hot spring associated with the exhumation of the land would have generated a convective system hydrothermal involving a mixture of meteoric and magmatic fluids, which removed the phosphate and uranium from alkaline rocks and depositing apatite rich in uranium as collophane at low temperatures (140 C), giving rise to collophanites in an environment subaqueous karst. Subsequently, a new phase of karstification reworked this material, fragments of reconcentrating collophanites in karst depressions. Figure 9. Diagram of rare earth elements for collophanites, according Boynton 1984. Figure 10. Diagram of trace elements for collophanites, according to Sun and McDonough (1989).Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings177

PAGE 180

AcknowledgmentsThanks INB (Brazilian Nuclear Industries) for having given piece of data CPRM and UFC. Collaboration of Braslia University, by Procad.ReferencesAlcntara e Silva JR, 2003. Caracterizao hidrogeolgica da Jazida de Itataia-CE. Master Science Thesis, Geology Department, Federal University of Cear, 150. Angeiras AG, 1988. Geology and metallogeny of the northeastern Brazil uranium-phosphorus province emphasizing the Itataia deposit. Ore Geology Reviews, Amsterdam, 3, 211. Boynton WV, 1984. Cosmoghemistry of the rare earth elements: meteorite studies. In: HENDERSON, P. (Ed.). Rare earth element geochemistry. Amsterdam: Elsevier, 1984. 6314. (Development in Geochemistry, 2). Brito Neves BB, Campos Neto MC, Van Schmus WR, Santos EJ, 2001. O sistema Paje-Paraiba e o macio So Jos do Campestre no leste da Borborema. Revista Brasileira de Geocincias, 31(2), 01. Brito Neves BB, Cordani UG, 1991. Tectonic Evolution of South America during the Late Proterozoic. Precambrian Research, 53, 23. Castro GL, Parente CV, Verssimo CUV, Sial NA, Garcia MGM, Santos, RV, Melo, RC, Santos, AA, 2005. Istopos de carbono e oxignio dos mrmores associados com o depsito fsforouranfero de Itataia, Cear. Revista Brasileira de Geocincias, 35(2), 199. Cuney M, 2010. Evolution of uranium fractionation process through time: driving the secular variation of uranium deposit types. Economic Geology, 105, 553. Fetter, AH, 1999. U/Pb and Sm/Nd Geocronological Constraints on the Crustal Framework and Geologic History of Cear State, NW Borborema Province, NE Brazil: Implications for the Assembly of Gondwana. Departament of Geology, Kansas University, Lawrence, PhD Thesis, 164. Hollanda, MHBM, Pimentel, MM, Oliveira, DC, Jardim de S, E.F, 2006. Lithosfere-asthenosphere interaction and the origin of Cretaceous tholeiitic magmatism in Northeastern Brazil: SrNd-Pb isotopic evidence. Lithos 86, 34. IBGE, 2009. Cidades @. In: www.ibge.gov.br/cidadesat/ topwindow.htm?1. Access in november, 2010. IPECE, 2008. Caracterizao territorial. Governo do Estado do Cear. In: http://www.ipece.ce.gov.br/atlas/capitulo1. Access in 2010. Braslia, 73. IPLANCE Instituto do Planejamento do Cear, 1995. Atlas do Cear, Governo do Estado, Fortaleza, 64. Mendona JCGS, Campos M, Braga APG, Souza EM, Favali JC, Leal JRLV, 1985. Jazida de Urnio de Itataia-CE. In: Principais Depsitos Minerais do Brasil, DNPM, v.1, 121. Netto AM, Meyer A, Cuney M, Poupeau G, 1991. A thermogeochronological study of the Itataia phospho-uraniferous deposit (Cear, Brazil) by analyses: genetic implications. In: Source Transport and Deposition of apatite fission track Metals, Pagel, M. and Leroy, J.L., eds., Proceedings of the 25 Years SGA Anniversary Meeting, 1991. Pasero M, Kampf AR, Ferraris C, Pekov IV, Rakovan J, White TJ, 2010. Nomenclature of the apatite supergroup minerals. European Journal of Mineralogy, 22, 163. Sun SS, McDonough WF, 1989. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: SUNDERS, AD; NORRY, MJ (Eds.). Magmatism in ocean basins. Oxfor: Geological Society of Lonfon; Blackwell, 1989. (Geological Society of London Special, 42). Vidal FWH, Sales FACB, Costa Roberto FA, Sousa JF, Mattos IC, 2005. Rochas e minerais industriais do Cear. FUNCAP, Fortaleza, 180.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings178

PAGE 181

NA JAVORCE CAVE A NEW DISCOVERY IN THE BOHEMIAN KARST (CZECH REPUBLIC): UNIQUE EXAMPLE OF RELATIONSHIPS BETWEEN HYDROTHERMAL AND COMMON KARSTIFICATIONJi Dragoun1, Karel k2, Ji Vejlupek1, Michal Filippi2, Ji Novotn1, Petr Dobe3 1Czech Speleological Society, ZO 1-11 Barrandien, Working Group Babyka, Trachtova 1/1129, 158 00 Prague 5, Czech Republic, drzahr@seznam.cz; vejlupj@seznam.cz; jinoli.centrum.cz2Institute of Geology AS CR, v. v. i., Rozvojov 269, CZ-165 00 Prague 6, Czech Republic, zak@gli.cas.cz; filippi@gli.cas.cz3Czech Geological Survey, Klrov 3, 118 21 Prague 1, Czech Republic, petr.dobes@geology.cz The Na Javorce Cave is located in the Bohemian Karst, Czech Republic, near the Karltejn castle, about 25 km SW of Prague. The cave was discovered as a result of extensive exploration including cave digging and widely employed capping of narrow sections. Exploration in the cave has already lasted 20 years. The cave is fitted with several hundred meters of fixed and rope ladders and several small fixed bridges across intra-cave chasms. Access to the remote parts of the cave is difficult because of long narrow crawl passages and deep and narrow vertical sections. The Na Javorce Cave became the deepest cave discovered to date in Bohemia with the discovery of its deepest part containing a lake in 2010. The cave was formed in vertically dipping layers of Lower Devonian limestone; it is 1,723 m long and 129 m deep, of which 9 m is permanently flooded (data as of December 2012). The cave is polygenetic, with several clearly separable evolutionary stages. Cavities discovered to date were mostly formed along the tectonic structures of two main systems. One of these systems is represented by vertical faults of generally N-S strike, which are frequently accompanied by vein hydrothermal calcite with crystal cavities. The second fault system is represented by moderately inclined faults (dip 27 to 45, dip direction to the W). Smaller tube-like passages of phreatic morphology connect the larger cavities developed along the two above-mentioned systems. The fluid inclusion data obtained for calcite developed along both fault systems in combination with C and O stable isotope studies indicate that the hydrothermal calcite was deposited from moderately saline fluids (0.5 to 8.7 wt. % NaCl equiv.) in the temperature range from 58 to 98 C. The fluids were NaCl-type basinal fluids, probably derived from the deeper clastic horizons of the Barrandian sedimentary sequence. The age of the hydrothermal processes is unknown; geologically it is delimited by the Permian and Paleogene. The hydrothermal cavities are small compared to cavities formed during the later stages of karstification. The majority of the known cavities were probably formed by corrosion by floodwater derived from an adjacent river. This process was initiated during the Late Oligocene to Early Miocene, as was confirmed by typical assemblage of heavy minerals identical in the surface river sediments and in clastic cave sediments. The morphology of most cavities is phreatic or epiphreatic, with only local development of leveled roof sections (Laugdecken). The phreatic evolution of the cave is probably continuing into the present in its deepest permanently flooded part, which exhibits a water level close to that of the adjacent Berounka River. Nevertheless, the chemistry of the cave lake differs from that of the river water. The cave hosts all the usual types of cave decoration (including locally abundant erratics). The most interesting speleothem type is cryogenic cave carbonate, which was formed during freezing of water in relation to the presence of permafrost during the Glacial period. The occurrence of cryogenic cave carbonate here indicates that the permafrost of the Last Glacial period penetrated to a depth of at least 65 m below the surface.1. Introduction and history of the cave explorationTwo small cave entrances, located high in the eastern slope of the Javorka Hill, close to the famous Karltejn castle, already attracted attention at the end of the nineteen forties and beginning of the nineteen fifties. Jaroslav Petrbok, a well-known karst researcher, archaeologist and zoologist focusing on terrestrial mollusks, performed excavations in both entrances. Radvan Horn and Ji Kovanda helped him during this work (both later became well-known geologists). Only a few meters of the entrance sections were explored, with finds of several ceramic fragments of the Hallstatt culture and an iron spear point dated to the Middle Ages (Petrbok and Horn 1950; Petrbok 1955). Allegedly Mesolithic finds, fragments of antlers, were later suggested not to be artifacts by Fridrich and Sklen (1976) and Sklen and Matouek (1992). This stage can be considered not to be important from the viewpoint of cave exploration. A new exploration phase was initiated by the members of the Czech Speleological Society, Unit 1-11 Barrandien (who are also the authors of this paper), in the upper entrance in March 1993. Clay sediments, with only small open spaces near the roof, filled the passages of the upper entrance behind the section explored archaeologically. The excavated material was transported using a special suspended mini-railway. The excavation performed in the complicated space of karstified tectonic structures with large unstable blocks of rock reached a length of 49 m in 1997. The work was interrupted here because of the instability of the walls of the excavated space. The next effort was focused on the lower entrance after an interruption of 3 years, again using a similar mini-railway suspended near the roof of the passage. Narrow sections had to be widened using capping. After 15 m of difficult work, the first open cavities with attractive decorations were reached in 2001 (Fig. 1). Karst and Caves in Carbonate Rocks, Salt and Gypsum poster2013 ICS Proceedings179

PAGE 182

The discovery initiated more intensive work, which enabled further uncovering of this morphologically complicated cave, which is very difficult to explore. It became clear that the cave developed in a structurally complex space, affected by low-temperature hydrothermal activity with deposition of calcite crystals in cavities. The cave generally consists of several types of cavities. The first type is represented by narrow, flat cavities developed along the NS faults with calcite veins. The cavities are up to 1 m wide (frequently only 0.3.4 m), but up to several tens meters long and commonly several tens of meters high in the direction along the faults. The second important type of cavities is represented by both narrow and larger corridors that developed along faults moderately dipping to the W. The two systems of cavities are interconnected by narrow, tube-like phreatic channels. Further exploration resulted in the interconnection of the upper and lower cave entrances in 2005. Together with further minor discoveries, the cave was explored to a length of 500 m that year. Difficult access to the above-mentioned connecting tubes between the vertical cavities required the construction of small hanging platforms, from which further exploration was possible. Difficult digging at the bottom of a narrow shaft called Zubat finally resulted in penetration into larger open spaces in 2006. The cave was explored to a depth of 104 m in that year (Figs. 2, 3). Further exploration was more rapid in the following several years. After cleaning of another connecting tube, a high parallel cavity was reached and, following difficult climbing up this chimney (in 2007), the upper levels of the cave have been gradually explored. This upper part of the cave also has rich decoration, including long soda straws and abundant erratics (Fig. 4). Any work in this distant cave section requires long access, which was facilitated by the construction of small bridges across the chasms. Cave digging in the narrow passages of the upper part of the cave enabled further discoveries in 2008 and 2009. The largest cavity, an inclined irregular passage called Seup, was discovered in this period. Exploration of vertical cavities in this part resulted in the discovery of the deepest cave section with a lake in 2010. The water level of the lake has lies at 120 m below the upper entrance (Fig. 5). Measurement of the water depth and first diving explorations demonstrated a water depth of 9 m, which corresponds to a total cave depth of 129 m. The cave became the deepest one discovered in Bohemia. Further cave exploration performed at several locations within the cave since 2010 did not result in new discoveries. Altogether, 498 one-day or two-day work shifts have been spent in the cave during the last 20 years. The cave became not only the deepest discovered in Bohemia, it is also the third longest cave discovered in the Bohemian Karst. Figure 1. Speleothem decoration in the cavities of the lower entrance (a cavity called Digital Chimney) discovered in 2001. Photo by J. Novotn. Figure 2. Simplified longitudinal E-W section (side view) of the Na Javorce Cave. The horizontal line at 285 m a.s.l. corresponds approximately to the base of the Neogene river terraces, and simultaneously to the base of the oldest Early Pleistocene river terraces in the adjacent valley. The sites of sampling of cryogenic cave carbonates (stars) and of hydrothermal vein fill (triangles) are also depicted. Based on mapping of J. Dragoun, J. Vejlupek and J. Novotn in 2010. Figure 3. Typical appearance of vertical cavities of the Na Javorce Cave. The photo is taken in a vertical cavity above the point of -104 m (cf. Fig. 2). Photo by J. Novotn.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings180

PAGE 183

2. Cave location, geology, hydrothermal calcite and cave decorationThe two cave entrances are located in the steep eastern slope of the Javorka Hill at elevations of 329.8 and 313.3 m a.s.l. The upper entrance opens 120 m above the water level of the Berounka River flowing at a distance of several hundred meters (the usual water level in the river is at 209.5m a.s.l.). The cave is located below the S and SE slopes of the hill close to its summit. This results in only a small geographical catchment of the cave under the presentday topography. The major lithology hosting the cave is Lower Devonian Kots Limestone, especially pure limestone layers in the upper part of the Lochkovian Stage. Nevertheless, cavities frequently also penetrate to the lower lithology, Kots Limestone with siliceous chert lenses and layers, and rarely also to the upper Lod nice Limestone and Dvorce-Prokop Limestone, both of which already belong to the Pragian Stage. Layers of all types of limestone dip almost vertically in this part of the Javorka Hill (the strike of the layers is almost EW, 80 to 85; the layer dip is around 90). The main Hercynian folding, producing a complicated structure including abundant large overthrusts, occurred in the Late Devonian and Early Carboniferous in the area. The vertically dipping sedimentary sequence is affected in the cave area by the faults of two main systems. Vertical faults of NS strike (the whole range of strikes is 155 to 180), represent the most important fault system. The limestone layering had a small effect on the cave morphology. The layering is roughly parallel to the EW simplified section (Fig. 2), while the NS faults and cavities developed along them are perpendicular to this section. The second fault system is represented by moderately inclined faults (dip 27 to 45, dip direction 240 to 270). Both entrance passages and the large Seup Passage developed along the faults and joints of this second system (Fig. 2). Calcite veining and crystal cavities were formed especially along the structures with NS strike. Locally they also penetrated into inclined faults. Calcite is coarsely crystalline, with scalenohedrons and rhombohedra with sizes up to 10 cm, of white, yellowish or brownish color. The evolution of calcite veins was irregular. Locally, they reach a thickness of several tens of cm, while at other places they are completely absent on the faults. Hydrothermal calcite was more resistant to corrosion during the later cave evolution, so that the calcite veins locally protrude out of the cavity walls, producing typical boxwork. Limestone alteration (inter-granular corrosion) locally produced rock with sandy disintegration and small accumulations of carbonate sand at several places. Cavities of tectonic and hydrothermal origin represent only a small portion of the known cavities. Most cavities developed during the later stages by common corrosion in the phreatic karst zone. Corrosion occurred along all the types of faults and calcite veins, so that the extent of earlier hydrothermal corrosion (if present) cannot be evaluated. Phreatic corrosion, a dominant process shaping the cave, is well-recorded in the morphology of the cavities. A small section of leveled roofs (Laugdecken), indicating long-term stagnation of the water level, developed only locally. There are several types of clastic cave sediments. The most interesting are quartz-dominated gravels with well-rounded pebbles and quartz-dominated sands of white to grey color. They occur in the upper part of the cave in a small quantities (up to several m3), especially in sections located close to the lower entrance. They represent the lowermost part of the cave sedimentary filling here. Since they show lithological similarity to the surface sediments of a river flowing not far from the cave during the Late Oligocene to Early Miocene, they have been subjected to study of the Figure 4. Speleothem decoration in the Radvansk Passage (distant upper part of the cave) with erratics and long soda straw. Photo by J. Novotn. Figure 5. Lake in the deepest part of the Na Javorce cave. Photo by J. Novotn.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings181

PAGE 184

Pseudosecondary fluid inclusions were observed in the studied samples along short healed microfractures. The inclusions were mostly of oval shape, up to 20 m in diameter, and exhibited consistent LVR ratios from 0.90 to 0.95. The Th values yielded valuable data in the range from 72 to 98 C, the salinity of the aqueous solution was calculated between 1.2 and 8.7 wt.% NaCl equiv., and NaCl was assumed to be the major compound of the aqueous solutions (Te = -21.5 C). Two generations of secondary inclusions were found in the studied calcite samples. They were observed either along longer healed microfractures, where they are of oval shape, up to 20 m in diameter, and with consistent LVR from 0.90 to 0.95, or along cleavage planes, where they are of irregular shape, from 10 to 200 m in diameter, and with variable LVR. The obtained Th values in the inclusions of the first generation ranged between 65 and 112 C, and the salinity of the solution was found to be from 0.2 to 5.6 wt.% NaCl equiv. The Th value of the second generation of secondary fluid inclusions was not determined because of variable LVR. The obtained Tm values yielded two intervals of salinity, from 1.9 to 1.9 wt.% NaCl equiv., and from 16.5 to 17.7 wt.% NaCl equiv. heavy mineral assemblage (see below). Quaternary speleothems can be found in the cave in a variety of morphological types, including long soda straws and locally abundant erratics (Fig. 4). Several chambers and corridors are well decorated, while the majority of the cave does not have any speleothems. Cryogenic cave carbonates formed by crystallization accompanied by water freezing occur at several sites (see the star symbols in Fig. 2; cf. k et al. 2012). They occur as accumulations of free crystals and crystal aggregates loosely deposited on the bottom of the cavities (Fig. 6 and the discussion below).3. Analytical methodsNew analytical research in the cave included study of the hydrothermal processes using fluid inclusion and stable isotope data on hydrothermal calcite, study of preQuaternary clastic sediments in the cave using the assemblage of heavy minerals, research on specific types of speleothems cryogenic cave carbonates (cf. k et al. 2012), and a pilot study on the chemistry of the lake water. Microthermometric analyses of fluid inclusion entrapped in hydrothermal vein calcite and calcite crystals in cavities were performed on cleavage chips (of about 300 m thick) using Chaixmeca apparatus (Poty et al. 1976). The apparatus was calibrated in the temperature range -100 to +400 C using Merck chemical standards, the melting point of distilled water, and phase transitions in natural pure CO2inclusions. The reproducibility of temperatures of homogenization up to +200 C was .0 C; the reproducibility of the temperatures of ice melting in fluid inclusions below 0 C was .2 C. The salinity of fluids was calculated according to Bodnar and Vityk (1994) and the composition of the salt system was evaluated after Borisenko (1977). C and O stable isotope composition of hydrothermal calcite was determined using the standard method of McCrea (1950). The isotopic composition of the CO2gas produced by decomposition of samples was measured using a Finnigan MAT 251 mass spectrometer. The heavy minerals were separated from a sieved 0.2 to 0.6 mm fraction of the cave sands using tetrabromethane with a density of 2.95g.cm-3. Methods used for the study of cryogenic cave carbonates are contained in k et al. (2012). Minerals were determined by X-ray powder diffraction using a Bruker D8 apparatus. The chemistry of water in the cave lake was determined using standard analytical methods (AAS, HPLC).4. Results and discussion4.1. Fluid inclusions in hydrothermal calcite The Jav1, Jav2 and Jav3 samples (see Fig. 2 for the sample location) contained relatively small amounts of primary fluid inclusions in 3D distribution. The inclusions were of irregular shape, from 5 to 60 m in diameter, and with variable liquid to vapor ratios (LVR = L/L+V). The most common (about 80%) were liquid-only inclusions, followed by liquid-rich two-phase inclusions with LVR of about 0.9 (about 15%), and inclusions with prevailing vapor phase were relatively rare. In a contrast, the Jav4 sample contained a very large number of tiny primary inclusions with variable LVR ratio in 3D distribution (sponge texture). The inclusions were of oval to irregular shape and from 5 to 20 m in diameter. Only fluids of the H2O-type were found in primary inclusions of all the studied samples. No fluid inclusions with a gaseous phase, such as CO2or CH4, were recorded. Due to the variable LVR, the homogenization temperatures (Th) were measured in clusters of inclusions with LVR of 0.90 to 0.95. The values of Th of the primary inclusions fluctuated from 58 to 91 C; the salinity of an aqueous solution, calculated from the melting temperature of the last ice crystal (Tm), varied from 0.5 to 7.0 wt.% NaCl equiv. The eutectic temperature (Te = -22.0 to -23.2) indicated that NaCl was the major component of the aqueous solution (Tab. 1).Table 1. Fluid inclusion data. FIA fluid inclusion assemblage; Th temperature of homogenization; Tm temperature of melting of the last ice crystal; Te eutectic temperature. No.FIAFIAThTm (C)salinityTe generationtype(C)(wt.%(C) NaCl eq.) Jav1primaryH2O58-2.1 to -4.43.6.0-23.2 pseudosec.H2O79-1.2 to -2.12.1.6 secondaryH2O65-0.1 to -0.40.2.7 Jav2primaryH2O84-1.1 to -3.31.9.4 pseudosec.H2O72-0.7 to -0.81.2.4-21.5 secondaryH2O7512-1.2 to -2.12.1.6 Jav3primaryH2O-0.3 to -1.20.5.1-22.0 pseudosec.H2O secondaryH2O8512-0.8 to -3.41.4.6-12.6 to -13.916.5.7-21.5 Jav4primaryH2O58-1.4 to -3.52.4.7-22.5 pseudosec.H2O74-2.6 to -5.64.3.7 secondaryH2O Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings182

PAGE 185

4.2. Carbon and oxygen stable isotopes in hydrothermal calcite and origin of fluids The carbon isotope data of the hydrothermal calcite ( 13C from -6.37 to -3.33 vs. VPDB) and the oxygen isotope data ( 18O from -10.74 to -9.20 VPDB) were obtained on identical hydrothermal calcite cleavage chips, which were also used for the fluid inclusion study. This enabled recalculation of the mineral C and O isotope data to the stable isotope parameters of the fluids. HCO3 -was assumed to be the dominant carbonaceous component of the fluids. The fluids depositing the calcite had 13C values in the range from -8.1 to -4.8 VPDB and 18O values in the range from 0 to +5 VSMOW. These characteristics are typical for basinal fluids. Meteoric waters of shallow circulation cannot be considered for the deposition of hydrothermal calcite because of the fluid 18O values and also their temperature and salinity. 4.3. Pre-Quaternary clastic cave sediments and the evolution of cavities in the phreatic zone Well size-sorted, white to gray, quartz-dominated sands from the passages of the lower cave entrance are locally accompanied by white to light-gray clays. The heavy mineral assemblage of the sands (2% andalusite, 5% secondary Fe-minerals, 30% ilmenite, 5% kyanite, 2% leukoxene, 1% magnetite, 10% rutile, 3% turmaline and 10% zircon) contained only stable minerals typical in the studied area for Late Oligocene and Early Miocene river sands. In contrast, minerals typical for Quaternary river terraces (amphiboles, pyroxenes, garnets, etc.) were completely missing in the studied cave sands. The relationship of the studied cave sands to an Oligocene/Miocene river can be considered as proved. The mineralogy of the spatially related lightcolored clays (kaolinite, illite/muscovite) supports the preQuaternary origin of the cavities hosting these clastic sediments. We assume that these clastic sediments penetrated into the cave together with floodwater injection from the Oligocene/Miocene river. The cavities were rapidly enlarged by corrosion by the river water, possibly with a contribution from mixing corrosion during mixing of the river and karst waters. As the Quaternary valley network started to develop, the upper parts of the cavities became drained, and the phreatic zone moved deeper and deeper. 4.4. Cryogenic cave carbonates and permafrost during the Glacial Freezing of mineralized karst water is inevitably accompanied by cryogenic mineral precipitation. The cryogenic cave carbonates exhibit typical modes of occurrence, morphology and geochemistry. The lowventilation caves can be cooled to freezing temperature only as a result of formation of permafrost. The occurrence and dating of cryogenic cave carbonates in these caves can be used for estimation of the permafrost depth of former glacials (k et al. 2012). The low ventilation of the Na Javorce Cave is reflected in the elevated content of CO2in the cave atmosphere (around 2%) and by its almost constant temperature. The occurrences of cryogenic cave carbonates in the Na Javorce Cave (stars in Fig. 2) have been studied and dated in detail by k et al. (2012). Site 1 yielded U-series ages of 50.04.39 and 28.56.17 ka BP, Site 2 an age of 14.98.17 ka BP and Site 3 an age of 9.19.07 ka BP. The data show that the rock massif and the cave were in a permafrost zone down to at least 65 m below the surface during the Last Glacial. 4.5. Present-day cave hydrogeology The cave lake (Fig. 4) contains low-mineralized karst water (sampling on 27 December 2010; composition in mg.L-1: Li+0.0042; NH4 +<0.02; Na+7.00; Mg2+29.94; Al <0.20; K+2.34; Ca2+71.9; Mn2+0.015; Fe 0.11; Zn2+<0.005; HCO3 -311.2; NO3 -13.6; F-0.09; SO4 2-20.2; Cl 15.84; pH 7.42; conductivity 503 S.cm-1). This water chemistry differs from that of the adjacent Berounka River (which exhibits a higher content of sulfate and chloride). It also differs from the water of karst springs with deep circulation, and from the water of a polluted creek flowing through the village of Karltejn. The water in the cave lake is therefore local infiltration water derived from the forested Javorka Hill itself. The present-day hydrogeological behavior of the lake is unknown. Detailed studies in the past few years clarified the hydraulic relationships between the river and adjacent caves (Vysok et al. 2012). Nevertheless, the cave lake in the Na Javorce Cave is farther away from the river than any of the lakes studied by Vysok et al. (2012).5. Conclusions overview of the cave evolutionThe cave evolution was initiated by tectonic processes. The inclined structures of western dip are probably older than the vertical tectonic structures of approximately S-N strike. Movements on these vertical structures locally opened small tectonic cavities. These cavities were later filled by hydrothermal calcite, containing abundant crystal cavities. Neither the tectonic phases nor the hydrothermal phase have been precisely dated. Dating is lacking for any of the abundant calcite veins of the Bohemian Karst. Geological observation indicates that the movement along vertical faults could have occurred from the Carboniferous to the Figure 6. Cryogenic cave carbonate of raft type with two generations of crystals, Na Javorce Cave, location 1 in Fig. 2. Photo by M. Filippi.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings183

PAGE 186

Cretaceous, and the hydrothermal phase could have occurred at any time between the Permian and the Paleogene. Movements along faults after the formation of hydrothermal calcite were either absent or only of small amplitude. The hydrothermal phase occurred at temperatures between 58 and 98 C. Moderately saline fluid (0.5 to 8.7 wt.% NaCl equiv.) with 18O in the range between 0 and +5 VSMOW and 13C in the range between -8.1 to -4.8 VPDB was most probably some kind of basinal water derived from a deeper part of the sedimentary sequence. The extent of hydrothermal corrosion is difficult to evaluate, since later karst corrosion in the phreatic zone overlaps the earlier hydrothermal corrosion. The upper parts of the Na Javorce Cave (above approximately 285 m a.s.l., which is the upper limit of the Quaternary river terraces) were formed by phreatic corrosion in the pre-Quaternary period, most probably during the Late Oligocene and Early Miocene. The relationship between the river of this age and the cave is reflected in the presence of quartz-dominated sands in the cave, which contain identical heavy mineral assemblage like the surface sediments. The surface fluvial sediments at a site located 2 km away from the cave are dated by paleontology (k et al. 2003). The principal processes of cave formation were probably floodwater injection and mixing corrosion. These processes resulted in irregular morphology of the cavities. The cavities below a level of 285 m a.s.l. could also have been formed (or their formation initiated) during the Oligocene or Miocene. In relation to the deep incision of the river during the Quaternary, the upper parts of the cave became drained and the phreatic evolution could continue only in deeper part of the cave. The present-day hydraulic relationship between the cave phreatic zone and the 400 m distant river is unknown (the cave lake was discovered in 2010 and has very difficult access). The water level in the cave lake and in the river are approximately equal (within the measuring error). It is clear that the cave was trapped in the Javorka Hill during the Quaternary. It has very limited geographic catchment area. The cave-lake water chemistry is different from those of the river and of a creek flowing through the town of Karltejn. The water chemistry probably reflects infiltration in the forested area of Javorka Hill itself. Water flow along the NS faults from a more northern area is another possibility. The Quaternary evolution of the cave is poorly understood. Sections with vadose morphology are generally rare in the cave (corrosion by drip-water is locally present) and the speleothems have not been dated yet, except the cryogenic cave carbonates of the Last Glacial. These cryogenic cave carbonates indicate that, during the Glacial, especially during the Last Glacial Maximum (between ca. 26.5 and 19.0 ka BP), the permafrost extended to a depth of at least 65 m below the surface. Cryogenic cave carbonates were found at several sites within the cave (Fig. 2) and were dated and studied in detail by k et al. (2012). The youngest age of the cryogenic cave carbonate of 9.19.07 ka BP was obtained for the sample with the deepest location within the cave. This possibly indicates that the relic permafrost lenses could have survived until the Early Holocene. Late Glacial and Early Holocene age have also been attributed to the bone finds in the entrance cave sections, transported into the cave by predators. The presence of animals and prehistoric humans was restricted to only a few meters behind both entrances, because of the sediment fill of the passages. Exploration work done in the cave since 1993 (intensive since 2001) resulted in the discovery of this interesting cave, which became the deepest cave of Bohemia in 2010. It is clear that the Na Javorce Cave has not yet revealed all its secrets.AcknowledgmentsThe research in the cave was performed within project P210/10/1760 of the Czech Science Foundation. The authors would like to thank Z. Tborsk for heavy mineral separation and determination, R. Skla for XRD mineral identification and I. Ja kov and B. ejkov for the C and O stable isotope determinations.ReferencesBodnar RJ, Vityk MO, 1995. Interpretation of microthermometric data for H2O-NaCl fluid inclusions. In: Fluid Inclusions in Minerals, Methods and Applications, B De Vivo and ML Frezzotti (Eds.). Virginia Tech, Blacksburg, VA, 117. Borisenko AS, 1977. Izutchenye solevogo sostava rastvorov gazovozhidkych vklyutcheniy v mineralach metodom kriometrii. Geologia i Geofyzika, 8, 16. Moscow (in Russian). Fridrich J, Sklen K, 1976. Die palolitische und mesolitische Hhlenbesiedlung des Bhmischen Karstes. Fontes Archaeologici Pragenses, 16, 1. Prague. McCrea JM, 1950. On the isotopic chemistry of carbonates and a paleotemperature scale. Journal of Chemical Physics, 18, 849. Poty B, Leroy J, Jachimowicz L, 1976. Measuring temperature under microscope Chaixmeca micro-thermometry apparatus. Bulletin de la Societe Francaise Mineralogie et de Cristallographie, 99(2), 182. Sklen K, Matouek V, 1994. Die Hhlenbesiedlung des Bhmischen Karstes vom Neolithikum bis zum Mittelalter. Fontes Archaeologici Pragenses, 20, 1. Prague. Vysok H, Bruthans J, k K, Mls J, 2012. Response of the karst phreatic zone to flood events in a major river (Bohemian Karst, Czech Republic) and its implication for the cave genesis. Journal of Cave and Karst Studies, 74(1), 65. k K, Tborsk Z, Lachmanov M, Pudilov M, 2001. Heavy mineral assemblages in allochthonous clastic cave sediments of the Bohemian Karst: A pilot study. esk kras, XXVI, 5. Beroun (in Czech with abstract in English). k K, Teodoridis V, Sakala J, 2003. Find of flora in Tertiary sediments near Karltejn. Geoscience Research Reports for 2002, 47. Prague (in Czech with abstract in English). k K, Richter DK, Filippi M, ivor R, Deininger M, Mangini A, Scholz D, 2012. Coarsely crystalline cryogenic cave carbonate a new archive to estimate the Last Glacial minimum permafrost depth in Central Europe. Climate of the Past, 8, 1821.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings184

PAGE 187

FIELD MEASUREMENTS OF GYPSUM DENUDATION RATE INKULOGORSKAYA CAVE SYSTEMNikolay Franz1, Sergey Sorokin2,3, Alexandra Alexeeva4, Irina Inshina4, Olga Novysh3, Anton Kazak3 1Arkhangels Speleological Association Labirinth, Arkhangels, Russia,franikol@mail.ru2Tver State University, Zhelyabova 33, Tver, Russia, sergey@tversu.ru3Arkhangels Speleological Association Labirinth, Arkhangels, Russia4Tver State University, Tver, Russia Results of measurements of karst denudation rate in natural conditions in Kulogorskaya cave system are reported. Taking into consideration relatively high rate of denudation in gypsum caves we have used simplified MEM measurement method. Observation of hydrological situation suggests that most of the dissolution in caves of the area occurs during a few days of spring flood. Here we report results of measurement at 62 points located at 6 sites for the period from 2003 to 2012. Data analysis gives us average retreat rate of 0.0448 mm/day, while point is underwater. Confidence interval for this value is 0.0379 to 0.0516 mm/day.1. IntroductionKulogorskaya cave system includes 7 caves, three of which constitute the largest cave system in Russia in a gypsum, with a total length of surveyed passages more than 17.5 km. Cave system is studied by speleologists from Arkhangels, St. Petersburg, Moscow, Tver and other cities under the guidance of Arkhangels Speleological Association Labirinth. This report presents the preliminary results of the measurements of dissolution of gypsum in the natural conditions, which are held from 2003.2. Geography and geologyGeographically Kulogorskaya cave system is located in the north of the East European Plain, in the watershed of Pinega (tributary of Northern Dvina) and Kuloi (Fig. 1). The climate here is quite severe, with low air temperatures and high humidity, rainfall exceeds evaporation. The multiyear average annual temperature is +0.2 C. The average Figure 1. Geographical position of Kulogorsky Cave Region.annual rainfall 560 mm. The coldest month January (average long-term temperature -13.0 C). The warmest month July (average long-term temperature +15.4 C). Karstifiable rocks of massif are represented by gypsum and dolomite of Permian age. Primary karst rock is gypsum, forming homogeneous strata ranging from 0.2 to 7 m, interleaved with dolomites. Average total width of sedimentary cover above caves is 20 m. Caverns in the massif are well represented in all hydrodynamic zones: vertical transfer zone, zone of seasonal fluctuations of water table levels and phreatic zone. Human-passable channels are located only at the zone of water table. In karst dissolution infiltrational, condensational and floodwaters are involved. Local erosion bases for caves in Kulogorskaya cave system is Pinega River and Pinega-Kuloy channel.3. Conditions of karst denudationThe most important role in the process of dissolution of gypsum walls in the water-table zone play floodwater penetrating the karst massif through numerous small ponors laid along the Kulogorsky ledge in the area of contact between the karst rocks and the floodplain alluvium. The penetration of floodwater into massif mainly occurs in spring (AprilMay) when Pinega waters rise by 4 meters above the winter low water (in the caves in this period water rises up to 2 meters, completely flooding almost all horizontal passages). During this period water flows in horizontal passages in water-table zone, which are typically dry in other seasons. Absorbed water temperature typically ranges around 1 C. Autumn floods in the river and caves are less affluent: the rise of water in the river up to 2 m, in caves up to 1 m. The inflow of water into the massif in this period is only through channels of phreatic zone (flooding of caves is caused by raising of the water-table level). Reverse runoff after autumn flood occurs on the same path in the opposite direction (the horizontal filtering through the thickness of the alluvium in the direction of erosion basis). Karst and Caves in Carbonate Rocks, Salt and Gypsum poster2013 ICS Proceedings185

PAGE 188

Rate of Karst Denudation in Russian) were fitted by us only in the passages of this zone.4. MethodsTo study the rate of dissolution of gypsum in the natural conditions since 2003, we have equipped 15 measurement sites in five caves. Almost all sites were equipped on the surfaces of the ceilings and walls of passages laid in the massif of white-and fine-grained gypsum with rare clusters of large gypsum crystals. Since the expected rate of dissolution of gypsum is higher than that of limestone, we used a simplified, compared to the classic version of the MEM method (Smith, 1977). Each SKD site includes from 1 to 4 rows of control points. Rows of control points may be oriented horizontally or vertically. Each row, in turn, includes from 4 to 7 observation points located with an interval of 3 cm along a straight line between the two fixed reference points. The main feature of the hydrogeology of Kulogorsky massif is absence of significant inverse flow of flood waters after passing through the peak of the spring flooding. In other words, floodwaters which penetrate the caves from the floodplain, never pour back by the same horizontal channels. Our current understanding is that at mean water period the water from caves is unloaded by slow horizontal filtering through the thickness of alluvium in the direction of the local base level of erosion. To determine the saturation of cave waters with gypsum, we measure its electrical conductivity, which is directly proportional to the salinity of water both according to the data of other researchers (Krawczyk and Ford 2006), and confirmed by our observations. In 2004 we conducted experiment in K-2 cave on the dissolution of gypsum powder in the water from underground lake (initial conductivity 2.1 mS) at natural cave temperature +2.2 C, which gives us maximal conductivity of 2.5 mS. But these values are not reachable in the natural underground water exchange, which is convincingly shown by the results of long-term observations of hydrochemical karst springs in the river Pinega: conductivity of discharged water is never more than 2.2 mS (88% of maximum mineralization). Since caves are inaccessible to humans during flood, we have used the tanks with automatic closure to determine the salinity of flood waters. These tanks have been installed by us in 2005 at a different distance from ponors. Analysis of the data showed, that conductivity of aggressive flood water, which penetrates the massif is 0.5.6 mS (20% of maximum). After passing for many hundreds of meters underground transit, these waters are saturated to 1.8.9 mS (72%). Then, in a few months of slow water level drop, conductivity of underground water reaches 2.2.3 mS (88%). Thus, during the transit time (in a few days at the peak of the spring flood) in conditions of heavy water exchange, floodwaters are saturated with the products of dissolution of sulphate rocks up to 50% approximately. Over the following months, during slow level decline, water gather additionally about 15% to reach the maximum observed mineralization of 2.2.3 mS, or 88% of the maximum possible in these climatic conditions. Since, as mentioned above, main floodwater flow occurs in the zone of water-table, we can assume that maximal denudation of cave walls happens in this zone. That is why all karst denudation rate measurement sites (SKD forFigure 2. Measurement method. Figure 3. Location of measurement sites.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings186

PAGE 189

Reference points are stainless steel screws fixed to the wall with plastic plugs to a depth of 50 mm. During the measurement special aluminum rod of square section with test holes every 3 cm is imposed on the heads of the screws (Fig. 2). Vernier caliper probe is inserted to each hole and distance between wall surface and outer surface of the rod is measured. On every SKD site we made a special passport with the tables of the measured values on each set of control points. We also made an exact measurement of height above sea level of each control point row.5. Results and discussionThis report presents the results of observations on six oldest SKD sites equipped in 2003 and affected by nine spring floods (2004). Locations of sites are shown in Fig. 3. Totally 62 points were analyzed. Obviously, in the existing hydrogeological environment hypsometric position of the observation point plays a significant role. Thus, the higher (in absolute elevation) is the reference point, the less time it is exposed to dissolving effects of flood waters, and vice versa. This is illustrated by a spring flood hydrograph of Pinega River in 2012 (Fig. 4), which is fairly typical. Therefore, in the analysis of the data, we took into account the absolute height of each of the control points. Taking into consideration available data on the dynamics of the spring floods of the river Pinega for 2004 years, it was possible to determine the estimated time (in days) of the interaction between gypsum surface and aggressive floodwater for each control point for the entire observation period. The resulting values of the total retreat (in mm) at each test point and the data on the duration of flooding of each of them allowed us to determine the retreat rate (Tables 1 and 2). We have used Smirnov-Grabbs criterion on the values of the retreat rate to identify outliers. Four points with exceptionally high rate where identified. They are shown with high rate comment in tables 1 and 4, and they are not used for any calculations. Using data from the remaining points we calculate mean retreat rate for every measurement row and site. Average retreat rate for all observation points is 0.0448 mm/day. The confidence interval for this value from 0.0379 to 0.0516 mm/day. The column conditions of Tables 1 and 2 provides an assessment of denudation conditions for each of the sites. In the sites marked Corrosion there is no signs of water flow. Erosion conditions are suitable for water flow and there are signs of such flow. It is interesting to note that the maximum retreat rate on surfaces of walls and the roof was measured at the SKD-8 site, equipped in one of the passages of Vodnaya (Water) Cave, where the nature of water deposits suggests floodflow velocities above 1 m/sec. and here, respectively, there is a very intense water exchange. From these data it is clear that the denudation rate of cave channels depends strongly on the local conditions, which is characteristic for gypsum karst (Klimchouk et al. 1996). On the example of SKD-2 site one can see that the dissolution rate can vary significantly even at points located in close proximity to each other. However, the obtained confidence limits give hope that average rate really characterizes the process of denudation within all massif. Another trend that can be observed in collected data, is higher rate of dissolution of horizontal surfaces (ceilings) than of vertical (walls). However, the existing data does not reliably establish that fact. We hope to clarify this point, after completing the measurements on the other sites. The resulting average speeds of karst denudation in real hydrogeological conditions of Kulogorskaya cave system can later be used to build a variety of mathematical models and reconstructions of paleogeography environments and stages of development of the caves of massif. Figure 4. Hydrograph of spring flood at Pinega river. Data from the state gauging station in Kulogory village in the immediate vicinity of the cave system (Federal State Unitary Enterprise Centre of Russian water works inventory and state water cadaster, 2012).Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings187

PAGE 190

Klimchuk A, Cicchi F, Calaforra JM, Aksem S, Finocchiaro F, Forti P, 1996. Dissolution of Gypsum from Field Observations. Gypsum Karst of the World. International Journal of Speleology, Theme issue, 1996, Vol 25, Issue 0, 37. Krawczyk WE, Ford DC, 2007. Correlating specific conductivity with total hardness in gypsum karst waters. Earth Surf. Process. Landforms, 32: 612. doi: 10.1002/esp.1409. Smith DI, 1977. The Micro Erosion Meter: Its application to the Weathering of Rock Surfaces. Conservation of Rock Art. Proceedings of the International Workshop on the Conservation of Rock Art, Perth, September 1977 (1978), 44.We continue to equip new measurement sites and are working on improvement of our techniques. Long-term monitoring of natural karst denudation processes will undoubtedly allow us to deepen our knowledge of the laws of development of not only Kulogorskaya cave system, but also of the entire northern sulfate karst.ReferencesFederal State Unitary Enterprise Centre of Russian water works inventory and state water cadaster. http://www.waterinfo.ru/ (in Russian). Table 1. Measurement data and statistics, part 1. SiteConditionsErosion SKD-2 Corrosion SKD-3 Erosion SKD-4Erosion SKD-5 Erosion SKD-6Row Sufrace Wall Wall Wall Wall Wall Wall Wall Wall Wall Wall PointPecularities high rate crystall crystall Height, m a.s.l. 16.69 16.66 16.60 16.57 16.54 16.36 16.33 16.30 16.21 16.02 15.99 15.96 15.93 15.90 15.35 15.32 15.29 15.26 15.23 15.20 15.57 15.57 15.57 15.57 16.27 16.27 16.27 16.27 16.27 15.18 15.18 15.18 15.18 15.18 15.34 15.34 15.34 15.34 15.70 15.70 15.70 15.70 15.94 15.94 15.94 15.94 15.94 15.94 1 2 4 5 6 1 2 3 6 1 2 3 4 5 1 2 3 4 5 6 1 2 3 4 1 2 3 4 5 1 2 3 4 5 1 2 3 5 1 2 3 4 1 2 3 4 5 6 Days underwater 25 25 25 25 25 36 36 36 36 46 46 46 46 46 79 79 79 79 79 79 51 51 51 51 33 33 33 33 33 80 80 80 80 80 72 72 72 72 50 50 50 50 44 44 44 44 44 44 Depth in 2004 mm 22.0 27.7 35.0 33.6 29.2 26.5 31.2 32.6 29.1 35.0 34.2 28.4 27.8 25.0 25.0 23.0 25.6 23.3 26.2 28.0 18.1 16.2 16.0 13.8 18.0 17.3 17.2 18.0 17.7 16.4 12.0 17.8 18.5 20.0 11.2 14.8 16.0 25.4 14.4 18.5 16.3 20.1 21.5 21.0 20.0 20.8 21.5 20.8 Depth in 2012 mm 23.7 29.4 36.4 36.8 31.3 28.2 32.1 33.1 30.3 38.4 36.0 31.2 30.5 27.9 27.3 25.1 28.5 26.6 29.3 31.0 19.9 16.8 16.9 14.5 19.2 18.3 18.3 18.8 20.8 17.3 13.1 19.5 19.7 21.0 13.0 16.8 19.2 29.3 16.5 20.1 18.6 21.4 22.5 23.0 21.4 22.3 23.0 22.7 Total retreat, mm 1.7 1.7 1.4 3.2 2.1 1.7 0.9 0.5 1.2 3.4 1.8 2.8 2.7 2.9 2.3 2.1 2.9 3.3 3.1 3.0 1.8 0.6 0.9 0.7 1.2 1.0 1.1 0.8 3.1 0.9 1.1 1.7 1.2 1.0 1.8 2.0 3.2 3.9 2.1 1.6 2.3 1.3 1.0 2.0 1.4 1.5 1.5 1.9 Retreat rate, mm/day 0.0680 0.0680 0.0560 0.1280 0.0840 0.0472 0.0250 0.0139 0.0333 0.0739 0.0391 0.0609 0.0587 0.0630 0.0291 0.0266 0.0367 0.0418 0.0392 0.0380 0.0353 0.0118 0.0176 0.0137 0.0364 0.0303 0.0333 0.0242 0.0939 0.0113 0.0138 0.0213 0.0150 0.0125 0.0250 0.0278 0.0444 0.0542 0.0420 0.0320 0.0460 0.0260 0.0227 0.0455 0.0318 0.0341 0.0341 0.0432 Avg. rate for row, mm/day 0.0690 0.0475 0.0299 0.0591 0.0352 0.0196 0.0330 0.0436 0.01480.0148 0.0378 0.0372 0.0365 0.03520.0352 Avg. rate for location, mm/day A B C D A B A A B A Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings188

PAGE 191

Table 2. Measurement data and statistics, part 2. SiteConditionsErosion, high-speed currents SKD-8 Row Sufrace Celling Wall Wall PointPecularities high rate high rate high rate Height, m a.s.l. 15.90 15.90 15.90 15.90 15.90 15.60 15.60 15.60 15.60 15.60 15.30 15.30 15.30 15.30 1 2 3 4 5 1 2 3 4 5 1 2 3 4 Days underwater 45 45 45 45 45 52 52 52 52 52 73 73 73 73 Depth in 2004 mm 14.2 15.2 15.2 14.2 15.0 14.3 15.5 21.0 22.6 19.6 20.1 16.1 13.2 16.0 Depth in 2012 mm 21.9 20.3 23.5 18.6 16.9 18.2 19.8 24.1 26.3 23.4 25.5 21.4 23.2 19.9 Total retreat, mm 7.7 5.1 8.3 4.4 1.9 3.9 4.3 3.1 3.7 3.8 5.4 5.3 10.0 3.9 Retreat rate, mm/day 0.1711 0.1133 0.1844 0.0978 0.0422 0.0750 0.0827 0.0596 0.0712 0.0731 0.0740 0.0726 0.1370 0.0534 Avg. rate for row, mm/day 0.0844 0.074 0.0723 0.0667 Avg. rate for location, mm/day A B C Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings189

PAGE 192

GEOLOGY AND STRUCTURE OF PARIAN CAVE, ISFAHAN, IRANGhassem Ghaderi, Leila Karimi Iran Speleology Society, 71438-54941, Gas S., Shiraz, Iran, ghaderi1348@yahoo.com, L_karimi2007@yahoo.com Parian Cave is located in Meimeh, Isfahan Province and formed in to the Triassic dolomites (Shotori Formation). The cave was discovered in 1999 and about 1,400 m of it has been surved. The cave developed above the water level (vadose zone) and has a very beautiful carbonate speleothems such as stalactites, stalagmites, columns, coralloids, flowstones, draperies, helectites, anthodites, frostworks, crusts, flowers, sodastraws, rimestones and shields. This survey is geological, morphological and structural viewpoints and the study results shows that the cracks and fissures have created enough porosity in the host stone and rainfalls have provided internally karst conditions. The cave passages follow discontinuity surfaces. They look are like a network of related underground spaces. Depending on the morphology of the passage and cave climate, every passage creating special speleothems and should be examined separately because each underground space has its special morphologic, hydrogeologic and hydrochemic conditions. The Parian Cave is a maze cave and has a pool with carbonate type elements.1. IntroductionVast regions in Iran including Zagros, Elburz and central Iran mountains consist of karst rocks which are important source of drinkable water for the population arid and semiarid climate of Iran. On the other hand, the hydrocarbon sources (oil and gas) are mostly located in karst regions in Iran and many silver, lead, zinc and bauxite sources are situated in similar locations. In this article the geology, geomorphology and tectonics of the Parian Cave were studied. The research is divided into two parts: (1) superficial and (2) subsurface examinations. The geologic map (1:2,000) and structural map including joints, faults, strata dip and strike and fold axes by the satellite photos are obtained. All discontinuous surfaces including fissures, faults, joints and beddinging in 14 places on the surface around the cave entrance were sampled and analyzed. The cave was mapped by compass and all the discontinuities were measured and analyzed by rockwork and finally the gained superficial and internal data were compared and the formation and development ways were analyzed.2. DiscussionThe Parian Cave is formed in Middle Teriassic dolomite. Lithology around the cave was mapped by meter and compass in the field (Tab. 1). Five main layers and eleven secondary layers were found in the travers from the valley bottom, hill summit up to the next valley. Limestones are light and dark colored with sandstone interbed deposited in shallow coastal environment. The Parian Cave is located in massive limestone layer. Dolomitic layers form cave walls. The change in the carbonate rock lithology and its impurities represents an important factor increasing potential for the karst formation in this region. Figure 1. Satellite photo of the Parian Cave. Figure 2. The Parian Cave in geologic map.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings190

PAGE 193

By virtue of the joint studies on 14 parts around the cave entrance, 3 sets of discontinuous surfaces are found. The joint surface properties included planar, semi-open, unfilled or filled partly with loam, and rough surface and moderately weathered. Countour and rose diagrams are shown three joint sets with northeast to southwest trend (Fig. 4).Table 1. Lithology of the layers measured during field. Figure 3. Geological section of The Parian Cave. Figure 4. The countour and rose diagram of discontinuous surfaces on the ground. Figure 5. The countour and rose diagram of discontinuous surfaces in the cave.Symbol Thickness Lithology Rash5 Alternation of green shale with layers of sandstone and thin layers of limestone and dolomite, sometimes ferrous Rash4 5 Light cream crushed limestone 7.4 Light cream limestone 2.9 Dark grey limestone 2.9 Alternation light and dark limestone 11.2 Cream crushed limestone 14 Alternative dark and sandy limestone 23.2 Brown ferruginous sandy limestone 7.4 Dark grey limestone 35.1 Alternation of grey limestone and red ferrous sandstone 20.7 Alternation of grey limestone and sandstone 10.8 Dark and cream limestone Rash3 184 Fissured cream limestone with fault and superficial karen Rash2 62 Green shale with sandstone and limestone interbeds Rash1 Light cream thickly-layered limestone with small caves Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings191

PAGE 194

The azimuth of the cave has the NESW trend and is completely similar to the discontinuities on the surface and inside the cave (Fig. 8). The cave has the negative or positive 20dip (Fig. 9) and also 70% of cave length is in the depths of 30 to 70 m (Fig. 10). The internal fissures were measured in the passages and chambers of cave (Fig. 5). The stereographic diagram shows that the cave fissures pattern is approximately similar to the earth surface; only the dip of discontinuities in the cave are less than the surface ones and have a 20 turning. The cave discontinuous surface properties are open and not filled. The Parian Cave was surveyed completely in four weeks by meter and compass and data drawn by the software compass. Figures 6 and 7 show the plan and profile drawn by software and Table 2 shows statistic analysis of the cave. Figure 6. The Parian Cave plan. Figure 7. The Parian Cave profile. Table 2. The specifications of the statistic analysis of The Parian Cave.Included Length1,401.5 m Total Surveyed1,401.5 m Horizontal Length1,185.7 m Cave Depth110.1 m Surface Length178.5 m Surface Width209.0 m Surface Area37,308.3 m2Cave Volume 99,409.4 m3Average Diameter8.4 Meters Wall Area39,593.5 m2Floor Area 10,349.5 m2Average Inclination25.8 Figure 8. Rose diagram of the cave passages. Figure 9. The dip diagram of The Parian Cave.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings192

PAGE 195

2. The discontinuous surfaces which developed on the earth and deep fissures transfer the rains into the cave and create the porosity necessary for karst development. The rose diagram in the cave is completely similar to the surface. Also, the properties of discontinuity are similar to the surface; only the cave fissures are open and unfilled completely. 3. The Parian Cave represents a set of related passages and corridors following the discontinuity pattern on the surface and in fact, it is a maze cave. 4. Total length of the cave is 1,401 m with an area of 37,000m2. The mean diameter of the disintegration passages is 8.4 namely the disintegration is effective. The lowest part of the cave is in 110 m depth from the entrance. 5. The mean slope is positive and negative 20 and the most development is the depth of 30 m. 6. Internal and External cave systems are effect to form Parian speleothems. Speleothems which were formed by dripping water and flowing water controlled by physicochemical conditions of outer cave system. Also Speleothems which were formed by percolating precipitation and speleothems which formed within pools controlled by physico-chemical conditions of external cave system.AcknowledgmentsThe study was done with the kind help and generous grace of Chakad Mountain Climbing Group and many spelunkers from Tehran, Isfahan and Shiraz to map and survey the cave. Also I continue to be indebted to Mr. Behrooz Mohammadi, Mr. Mansoor Khojasteh, Mr. Mahdi Hajhashemi, Ms Mehrnoosh Noorelahi, Mr. Mahdi Shahkarami, and Mr. Ahmad Emadzadeh for their kind cooperation.ReferencesDarvishzadeh A, 1991. Iran Geology, Emrooz Publications. Memarian H, 1995. Engineering Geology and Geotechnique, Tehran University Press. Ghobadi MH, 2007. Karst Engineering geology, Booali Sina University Press, Hamedan. Topographic maps with the scale 1:25,000, State Topography Organization. Kashan Geological Map with the scale: 1:250,000, Iran Oil National Company.This survey shows that two systems effect to form of speleothems. Internal cave system and external cave system. External cave system relates to near-surface hydrological cycle and internal cave system refers to every process that has happened in the cave. In this point of view, all of the speleothems in the Parian Cave have classified to two major groups. 1. Speleothems controlled by physico-chemical conditions of the External cave system that has recorded in infiltered water and goes down to cave and can be recognized in speleothems which were formed by dripping water(such as stalactite, stalagmite, column, sodastra) and flowing water (flowstone, drapery, shield). 2. Speleothems controlled by physico-chemical conditions of the internal cave system that related to microclimate conditions in cave and can be recognized in speleothems which were formed by seeping water(coralloid, helectite, frostwork, anthodite) and speleothems that formed under pools (crystal lining, gours).3. ConclusionsThe findings concerning the Isfahan Parian Cave studies are as follows: 1. The Parian Cave is developed mostly in the massive limestone layers. Dolomite layers are mostly in cave walls so the change in the lithology and rock impurities has been an important factor increasing karst evolution potentiality in one region and undeveloped potentiality in other ones. Figure 10. The passage length versus passage dip of The Parian Cave.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings193

PAGE 196

PRELIMINARY STUDIES ON CORROSIVE FORMS AND CLASTIC DEPOSITS IN THE NIENA CAVE (THE TATRA MTS., POLAND)Ditta Kici ska Institute of Geology, Adam Mickiewicz University, ul. Makw Polnych 16, 61-606 Pozna Poland, kicinska@amu.edu.pl The Wielka niena Cave is the deepest and the longest system in Poland, which consists from group of vertical shaft with horizontal parts among of which the highest situated on depth of 300 m (the part of Wodoci g and Ci gi Zakopia skie passages). New data on basis observation of corrosive forms and heavy minerals analyses indicate that some passages of the niena Cave could be formed on several stage and under phreatic or epiphreatic conditions.1. Introduction and geological settingsThe Wielka niena Cave System (denivelation of 824 m, length of 23 km) is situated in the Czerwone Wierchy Massif, which is surrounded on the northern part by the Maa ka, Mitusia and Kocieliska valleys. The system includes 5 caves: nie na, Wielka Litworowa, Nad Kotliny, Jasny Aven and Wilcza (Fig. 1). Preliminary studies were conducted in the niena Cave up to depth of 300 m where the horizontal part named Wodoci g and Ci gi Zakopia skie passages occur (Grodzicki 2002). The system is developed in Mesozoic limestones, which belong to the High Tatric sequence (Kota ski 1959). Three EW-trending carbonate belts (southern, middle and northern) separated by zones of non-karstic rocks occur in this sequence. The belts run across river valleys. The Wielka niena Cave System belongs to the middle belt, in which the longest and the deepest caves of Poland are located. This part of the Czerwone Wierchy Massif is discharged by an ascending Lodowe rd o Spring (in the Ko cieliska Valley).2. Investigations in the caveThe first observation on karst was published by Kowalski (1953), who claimed that the Lodowe Spring in the Kocieliska Valley drains the western part of the Czerwone Wierchy Massif between the Ko cieliska and Mi tusia valleys and possible the eastern part of Czerwone Wierchy Massif too. Zwoli ski (1955) suggested that the Lodowe Spring drains the massif between peaks of Ciemniak and Ma o czniak. The tracing of underground water in the niena Cave revealed flow to the Lodowe Spring during 6 days (Dbrowski and Rudnicki 1967). Investigations of origin and age of nie na Cave were made by Wjcik (1978), G azek et al. (1979), Grodzicki (1991, 2002) and G azek and Grodzicki (1996) who concluded that the main reason of development of cave is tectonic widening of fissures, which formed during deglaciations. Following researches were conducted by Kici ska (2002) and Pawowska Bielawska (2007) who origin horizontal passages related to be formed under phreatic or epiphreatic conditions. Observation and measurements of scallops were performed in the main passages of the Wodoci g and Ci gi Zakopia skie passages (toward Salka Rysi). Asymmetry of scallops indicate a west paleocurrents direction (towards the Ko cieliska Valley) (Fig. 2). Composition of heavy minerals in particular parts is varied. Predominance instable minerals in the Wodoci g passages and chemostable and mechanostable minerals in the Ci gi Zakopia skie passages provides that analyzed horizontal passages of niena Cave has been not formed on one stage (Fig. 3). The cross section of passages confirmed suggestion made by G azek et al. (1977) that the vertical elements developed on tectonic discontinuities during deglaciation of Pleistocene and Holocene age as the invasion caves (proglacial). Ellipsoidal cross section of the Wodoci g and Ci gi Zakopia skie passages, paleocurrents direction based on scallops and composition of heavy minerals show that the horizontal parts were formed under epiphreatic or phreatic conditions, what sheds a new light on the origin, evolution and probably age of the nie na Cave. To better understand the Wielka nie na Cave System more researches is needed such as analysis of spatial patterns of cave and cross-section of passages, analysis of corrosive forms and dating of speleothems. Figure 1. Location of niena Cave.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings194

PAGE 197

ReferencesBielawska-Paw owska P, 2007.Evolution of Wielka nie na Cave in the light of geomorphologic observations. Karst and Cryokarst, Uniwersytet lski, 155. D browski T, Rudnicki J., 1967. Wyniki bada przep yww krasowych w masywie Czerwonych Wierchw. Speleologia, 3, 1, 31. Figure 2. Schematic cross-section of Wielka nie na Cave System, areas where scallops were studied, the paleoflow directions and sampling places of clastic deposits. Figure 3. Relationship between instable (apatite, amphinole, epidot), chemostable (rutile, staurolite, disthene, tourmalines) and mechanostable (garnets, zircon) heavy minerals in the studied samples, classification diagram after Burkhardt (1978). G azek J, Grodzicki J, 1996. Kras i jaskinie. In: W: Mirek Z. (Ed.). Przyroda Tatrza skiego Parku Narodowego. Wyd. Inst. Ochr. Przyrody PAN, 139. Gazek J, Rudnicki J, Szynkiewicz A, 1977. Proglacjal caves a special genetic type of caves in glaciated areas. Proceedings 7thInternational Speleological Congress, Sheffield, England, Semptember, 1977. British Cave Research Association, Bridgewater. 215. Grodzicki J, 1991. Geneza i ewolucja jaski Tatr Zachodnich. In: Grodzicki J. (Ed.) Jaskinie Tatrza skiego Parku Narodowego. T. 1. Jaskinie Doliny Chocho owskiej i dolinek reglowych. Polskie Towarzystwo Przyjaci Nauk o Ziemi. 11. Grodzicki J, 2002. Jaskinie Tatrza skiego Parku Narodowego. Wielkie jaskinie Doliny Ma ej ki. T.9. Polskie Towarzystwo Przyjaci Nauk o Ziemi, Warszawa, 1. Kota ski Z, 1959. Profile stratygraficzne serii wierchowej Tatr Polskich. Biuletyn IG, 139, 1. Kowalski K, 1953. Jaskinie Tatr Polskich. Wyd. Pa stw. Muz. Archeol., 1. Wjcik Z, 1978. O wieku jaski tatrza skich. Prace Muzeum Ziemi, 28, 123. Zwoli ski S, 1961. W podziemiach tatrza skich. Wyd. Geol., 1.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings195

PAGE 198

GEOCHEMICAL AND STABLE ISOTOPE CHARACTERIZATION OF DRIP WATER FROM THE POSTOJNA CAVE, SLOVENIAMagdaMandi1, Andrej Mihevc2, Albrecht Leis3, Ines Krajcar Broni4 1Department of Physics, University of Rijeka, Rijeka, Croatia, mmandic@phy.uniri.hr2Karst Research Institute, Postojna, Slovenia3Joanneum Research Institute, Graz, Austria4Ruer Bokovi Institute, Zagreb, Croatia Comprehensive investigation of geochemical and isotopic characteristics of drip water in Postojna Cave (Classical Karst, SW Slovenia) is presented here. Sampling of drip water was performed at 9 locations within the cave from March 2010 to April 2011. Air and water temperature, pH value, drip rate, conductivity, ion concentrations, 18O of water and 13C of DIC were determined. All drip waters belong to the Ca HCO3type according to chemical composition and to the seepageflow type according to drip rate and its variability. Cave temperature is relatively constant during the year and reflects the average temperature in the region. Relatively constant 18O values of drip water reflect mean 18O of precipitation. A process of prior calcite precipitation (PCP) is proposed for explaining observed differences in some parameters at location 05 Podrti kapnik in comparison with all other locations.1. IntroductionThe Postojna Cave (Slovenia) is one of the most famous karst caves in the South-eastern Europe. The comprehensive investigation of speleothems and cave environment has been undertaken with the aim to map recent geochemical and stable isotope characteristics of the Postojna Cave which could help in providing insights into climate and environmental changes. In this work we present results that relate to various geochemical and isotopic parameters in drip water.2. Site descriptionThe Postojna Karst is situated in Slovenia on NW part of the Dinaric Karst (Figure 1). More than 60 caves are known in the Postojna Karst. The largest cave system (20.5 km long) is the Postojna Cave system. One of the caves in the system is the Postojna Cave, 10,399 m long (ebela 1997). The Postojna Cave was formed by the sinking Pivka River. There are two main levels of the cave passages: the upper, which is mostly dry, and the lower where the Pivka River flows. Modern entrance of the Postojna Cave is at 529 m a.s.l. It is 18 m above the Pivka River sinkhole. The entrance leads to the upper level of cave passages and larger halls which are also connected with lower river passage. In the whole length cross section is rounded and shows leveled ceilings. On the walls and ceiling of the cave interior there are marks and remains of sediments, indicating that in the cave evolution there were cycles of filling and flushing out the cave passages during the past (Brodar 1966).3. SamplingMonitoring in the Postojna Cave was performed in the period from March 2010 to April 2011. Sampling of drip water was done at 9 locations within the cave (Figure 2). Additionally, the Pivka River was sampled at two locations: at the entrance to the Postojna Cave and 2.5 km inside the cave. On the field were measured pH, conductivity, air and drip water temperature and drip rate. Laboratory chemical analysis of water samples included measurements of Ca2+, Mg2+, HCO3 -concentrations and twice a year measurement of Na+, K+,Cl-, SO4 2-and NO3 -concentrations. Stable isotope compositions of water ( 18O) and dissolved inorganic carbon DIC ( 13CDIC) were determined. Figure 1. Map of Europe with position of Slovenia (left); (right) map of Slovenia with Postojna Karst area (darker shadowed are a). The boundary of the Dinaric karst (dashed line) is shown (Gams 1974).Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings196

PAGE 199

4. MeasurementPhysical (temperature) and chemical parameters (pH and conductivity, both corrected to 25 C) in drip waters were measured in situ by the WTW multiline pH/Conductivity 340i Ph Meter. Chemical analyses of water samples were performed by certificated methods using ion chromatography. Stable isotope analyses were carried out in Laboratory Centre for Isotope Hydrology and Environmental Analytics, Joanneum Research Institute (Graz) by an on-line continuous-flow system (Gasbench II) linked to a Thermo Fisher Scientific DELTAplusXL isotope ratio mass spectrometer. The 18O values are reported relative to the VSMOW scale and 13CDICrelative to the VPDB scale. Calibration of the mass spectrometer was accomplished using in-house water and calcite reference materials whose stable isotopic composition had previously been calibrated (Sptl 2005).5. Results and discussionAir temperature (Fig. 3) within the Cave varies slightly around 10 C, which is close to the mean air temperature measured at the meteorological station during last decade (9.7 .5 C). Variations around the mean are larger at locations closer to the entrance to the Cave. Drip water temperature also slightly varies around 10 C, while the river water follows more closely the outside air temperature. Monthly amount of precipitation and drip rate are shown in Figure 4. According to drip rate and its variability, all drip waters belong to the seepage type (Smart and Friedrich 1987). The lowest drip rate is measured in summer (August 2010), and the highest in November 2010, about two months after abundant precipitation in September 2010. During winter period (February 2011), preceded by two months of low precipitation, the drip rate is again low. Low discharge in winter may be observed also because of snow cover during winter months. At location 08 Vrh Velike gore the highest drip rate is measured in November 2010 due to the high amount of precipitation during September 2010. In the time with low 01-F eb-10 01M ar-10 01-Apr-10 01-May-10 01-Jun-10 01-Jul-10 01-Aug-10 01Sep10 01Oct-10 01-Nov -10 01 -Dec-10 01-Jan-11 01-Fe b11 01-Mar-11-5 0 5 10 15 20 25 30 T (oC) 01 Slonova glava 02 Biospeleoloka postaja 03 Vodopad 04 Kongresna dvorana 05 Podrti kapnik 06 Stebrie 07 arobni vrt 08 Vrh Velike gore 09 Zgornji Tartar 10 Pivka Rive inside 11 Pivka River outside Postojna METEO Figure 3. Air temperatures on sampling locations in Postojna cave, together with data from Slovenian meteorological survey for Postojna during sampling period (step line).Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings197 Figure 2. Sampling locations in the Postojna Cave (left). List of names of the sampling locations and their numbers, and the sy mbol applied in this paper (right).

PAGE 200

Relation of calcium ion concentration and drip rate is different for different locations (Figure 5). At locations 07 arobni vrt, 08 Vrh Velike gore and 09 Zgornji Tartar where drip rate has large variations calcium concentration is stable. The larger variation of calcium ion concentration has been observed at location 02 Biospeleoloka postaja, where the drip rate is constant and low. The average concentration of calcium ion is plotted as a function of the caprock thickness in Figure 6. Higher calcium concentration (from 86 mg.l-1 to 89 mg.l-1) is determined at outer locations (01, 02, 03 and 04) with caprock thickness between 40 m and 90 m. At other (inner) locations (06, 07, 08, 09) with caprock thicknesses from 65 m to 85 m, Ca2+concentration is lower (73 to 82 mg.l-1). Therefore, concentration of calcium ion in the drip water is not determined only by caprock thickness, but also by geomorphology of the karst terrain itself, i.e. by the routes that drip water undergoes before reaching the drip site (Baker et al. 2000; Fairchild et al. 2000;Tooth and Fairchild 2003). Most dissolution of limestone occurs in the epikarst zone (up to 60 m terrain depth) (Ford and Williams 2007). In Figure 6, location 11 Pivka River inside is plotted only for comparison and demonstration that calcium ion content of water is not influenced by caprock thickness. Location 05 Podrti kapnik is an exception and will be discussed separately. Other chemical parameters are neither influenced by caprock thickness. At all locations, magnesium concentrations are lower than 1.5 mg.l-1, with the mean value of 0.85 .18 mg.l-1. Both sampling locations of the Pivka River show higher concentrations of magnesium ions (>3 mg.l-1) due to the Pivka River origin and flow path. Concentrations of other minor ions are negligible at all locations. The 18O values of drip water vary between -10 and -8 with practically now seasonal variations. The variations in precipitation amount the drip rate at location 08 Vrh Velike gore is low.That would appoint to short mean residence time(MRT)at location 08 Vrh Velike gore.The response of the drip rate to abundant precipitation is delayed for about 2 months at locations 07 arobni vrt, 08 Vrh Velike gore and 09 Zgornji Tartar, as opposed to location 02 Biospeleoloka postaja where high precipitation amount has negligible influence on drip rate, i.e. drip rate is relatively constant and not dependent on the amount of precipitation (Fig. 4). The pH values of the drip water show seasonal fluctuation lowest pH values are measured in August and September and highest during winter months (NovemberMarch). Calcium concentrations range between 74 and 89 mg.l-1(Fig. 5). Largest variation from mean value for calcium concentration is found at location 01 Slonova glava (89 mg.l-1), and at location 08 Vrh Velike gore deviation from mean value is the lowest (82 mg.l-1). Location 05 Podrti kapnik has the lowest mean value (65 mg.l-1). Mean HCO3 -concentrations range between 200 and 273 mg/L at all locations except for Podrti kapnik (152 mg.l-1). Jan-10 Feb-10 Mar-10 Apr-10 May -10 Jun10 Jul-10 Avg-10 Sept-10 Oct-1 0 No v -10 Dec-10 Ja n-11 F eb-11 Mar -11 A pr -1 10 50 100 150 200 250 300 350 400 450 500 550 600 Precipitation (L/ m2)Drip rate (min-1)0 50 100 150 200 250 300 350 400 450 500 550 600 Figure 4. Monthly precipitation amount () and drip rate. 30405060708090100110 60 65 70 75 80 85 90 95 Ca2+ ( mg/L)Caprock thickness (m) Figure 6. Mean values of calcium ion concentration vs. caprock thickness. There are two groups of locations. Full line square points out outer locations (0 04) and dashed line square inner locations (06). Location 05 does not belong to any of the two groups. 050100150200250300350400450500550 40 50 60 70 80 90 100 110 120 Ca2+ (mg/L)Drip rate (min-1) Figure 5. Dependence of calcium ion concentration on drip rate.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings198

PAGE 201

18O of drip water at all 9 locations are than that in local precipitation, indicating good mixing of waters in the epikarst zone and relatively long MRT of water, from several months (at locations 01, 02 and 08) to about 1 year at locations 03, 05 and 09 and reaching several years (locations 04, 06, 07). The 13CDICvalues measured in the Pivka River water at location 10 Pivka River inside and at location 11 Pivka River outside are approximately the same, i.e. -12.4 .0 and -12.4 .2, respectively. The drip water at location 05 Podrti kapnik (mean value -9.90 .26, ranging from -11.78 to -5.64) has more positive 13CDICvalues than other drip waters, where 13CDIC mean values range from -12.32 .28 at location 07 arobni vrt to -13.30 .33 at location 01 Slonova glava. Smallest variations have been measured at location 06 Stebri e (-13.06 .11). The highest average conductivity value has been measured at location 01 Slonova glava (454 S.cm-1), where maximum value of 556 S.cm-1is measured in November 2010. The lowest average value is found at location 05 Podrti kapnik (277 S.cm-1), where the minimum conductivity of 197 S.cm-1is measured in February 2011. At location 06 Stebri e minimum deviation from mean value is found 366 S.cm-1, i.e. this location shows most stable conductivity condition of all sampling locations. Correlation between mean values of HCO3 -and Ca2+ions for all locations is shown in Figure 7. Good correlation (R2 = 0.94) is in agreement with classical processes in karst where higher concentrations of bicarbonate HCO3 -in drip water causes higher concentrations of dissolved minerals, in the case of limestone, the calcium minerals. Mean conductivity and drip rate show linear relationship (R2= 0.51; Fig. 8). Conductivity is in direct correlation with concentration of dissolved minerals in the solution. Locations with higher drip rate show stronger response to the rain events. For drip water at location 08 Vrh Velike gore and 07 arobni vrt with short residence time conductivity is low because of shorter time for water rock interaction, i.e. for dissolution of the overlying limestone bedrock, which is opposite to the situation at location 02 Biospeleoloka postaja. The mean values of saturation index Isatvary from 1.94 .81 at location 01 Slonova glava to 6.19 .12 at location 04 Kongresna dvorana. The highest value (12.82) is observed at location 04 in February 2011 (Figure 9), and the lowest value of 0.58 at location 02. Line on the Figure9 presents equilibrium solution, i.e. Isat= 1. If Isat< 1 there is no possibility for carbonate precipitation, and if Isat> 1 then carbonate precipitates. Most of drip water Isatvalues lye in the region above the line, and only several Isatvalues are close to or slightly lower than 1. 140160180200220240260280 60 65 70 75 80 85 90 95 Ca2+ ( mg/L )HCO3 ( mg/L ) Figure 7. Correlation between mean values of HCO3-and Ca2+ion concentration for all sampling locations. 020406080100120140160 250 275 300 325 350 375 400 425 450 475 500 Conductivity ( S/cm)Drip rate (min-1) Figure 8. Correlations of mean annual conductivity and drip rate with linear fit. 7.07.27.47.67.88.08.28.48.6 0 3 6 9 12 15 IsatpH Figure 9. Relation between Isatand pH. Red line presents equilibrium solution, Isat= 1.Location 05 Podrti kapnik has been mentioned several times as an exception in comparison with other locations, e.g., concentrations of calcium and bicarbonate ions (Figures 6, 7) and conductivity (Figure 8) are lower, while the 13C values of DIC are on average 2 higher than at other studied locations. The differences are more pronounced during dry periods when the Mg/Ca ratio also increases. All these observations indicate the possibility of the loss of calcium and bicarbonate ions due to carbonate precipitation in some voids within the caprock above this locations caused by degassing of CO2from the percolating water. This process in known as prior calcite precipitation (PCP; Horvatini et al. 2003; Sptl et al. 2005; Fairchild et al. 2007). In the period after abundant precipitation when the water level is high, the voids are filled with water and do not allow CO2degassing, the mentioned parameters at Karst and Caves in Carbonate Rocks, Salt and Gypsum poster2013 ICS Proceedings199

PAGE 202

location 05 do not show different values compared to other locations, justifying our assumption of PCP.6. ConclusionThe one-year monitoring of various chemical and isotopic parameters in drip waters of the Postojna Cave leads to the following conclusions. Air temperature and drip water temperature reflect the mean annual temperature of the area, but the seasonal variations in the cave are much smaller. Similarly, the mean 18O values of drip waters reflect the 18O of the local precipitation, but without pronounced seasonal variations. According to its chemical composition, all drip waters belong to the same type (Ca HCO3 -). According to drip rate and its variability, all drip waters belong to the seepage type, although some differences in the drip rate response to the precipitation amount are observed among the studied locations. Location 05 Podrti kapnik is often an exception when compared to all other monitored locations by its lower Ca2+and HCO3 -concentrations, lower conductivity, higher Mg/Ca ratio, and higher 13CDICvalues. All these characteristics indicate the possibility of prior calcite precipitation on the way of the drip water from the epikarst zone to the cave.ReferencesBaker A, Genty D, and Fairchild I, 2000. Hydrological characterization of stalagmite drip waters at Grotte de Villars, Dordogne, by the analysis of inorganic species and luminescent organic matter. Hydrology and Earth System Sciences Discussions, 4(3), 439. Brodar S, 1966. Pleistocenski sedimenti in paleolitska najdi a v Postojnski jami. Acta carsologica, 4, 57. Fairchild I, Borsato A, Tooth A, Frisia S, Hawkesworth C and Huang Y, 2000. Controls on trace element (Sr-Mg) compositions of carbonate cave waters: Implications for speleothem climatic records. Chemical Geology 166, 255. Fairchild I, Frisia S, Borsato A and Tooth A, 2007. Speleothems. In Nash, D. and McLaren, S., eds., Geochemical Sediments and Landscapes. Wiley-Blackwell, Oxford. Ford D, and Williams F, 2007. Karst hydrogeology and geomorphology. John Wiley and Sons, West Sussex. Gams I, 1974. Kras zgodovinski, naravoslovni in geografski oris. Slovenska matica, Ljubljana, 359. Horvatini N, Krajcar Broni I, and Obeli B, 2003. Differences in the 14C age, 13C and 18O of Holocene tufa and speleothem in the Dinaric Karst. Palaeogeography, Palaeoclimatology, Palaeoecology 193(1), 139. Smart P, and Friedrich H, 1987. Water movement and storage in the unsaturated zone of a naturally karstified aquifer unsaturated zone of a maturaly karstified aquifer, Mendip Hills,England. In Zeitschrift fur Geomorphologie / Proceedings of Conference on Environmental Problems in Karst Terrains and their Solutions, volume 59. National Water Well Association, Bowling Green, Kentucky. Sptl C, 2005. A robust and fast method of sampling and analysis of 13C of dissolved inorganic carbon in ground waters. Isotopes in Environmental and Health Studies 41(3), 217. Sptl C, Fairchild I, and Tooth AF, 2005. Cave air control on dripwater geochemistry, Obir Caves (Austria): Implications for speleothem deposition in dynamically ventilated caves. Geochimica etCosmochimica Acta 69, 2451. ebela S, 1998. Tectonic structure of Postojnska jama Cave System. ZRC SAZU, Postojna. Tooth AF, and Fairchild IJ, 2003. Soil and karst aquifer hydrological controls on the geochemical evolution of speleothem-forming drip waters, Crag Cave, southwest Ireland. Journal of Hydrology 273, 51.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings200

PAGE 203

Pleiveck, Silick, Horn vrch, Doln vrch, Zdielska, Borianska and Jasovsk. The altitude difference between valley and plateau surface is 400 m on average. On the plateau surfaces, karst forms (exokarst) are well-developed. Lot of Avenand Lighthole-type abysses were found there. It is interesting that active karst hydrological systems with active flow are known only from the edge of the Slovak Karst.3. Skalist potok CaveBoth known and used entrances into the Skalist potok Cave (Rocky brook Cave) are man-made. The first was punctured by workers from the IGHP (Engineer-geological survey) in 1968 in the place of the periodic spring on the feet of the Jasovsk plateau in the altitude of 205 m a.s.l. The second one was punctured by cavers on the southern plateau slope in 2007 at 460 m a.s.l. Most of passages are overflowed by active underground river, 42 sumps are already mapped (with a length of about some meters up to 130 m). Finally, the cave is interesting speleodiving locality, but there are many vertical parts. These passages rank the cave to the deepest caves in the Slovak Karst and among Slovak caves it ranks the fourth. The topical length is 7,983m and denivelation 373 m.4. History of a speleological and speleodiving survey in the Skalist potok Cave1968: In the place of the periodic spring on the feet of the Jasovsk plateau, workers from the IGHP Co. from ilina gapped a 13 m long exploration tunnel and they find underwater parts of the natural cave. Cave passages were surveyed by Mikul Erds. 1982: Speleodiver J. Kucharovi from Tren n city overcame two sumps and reached a distance 258 m from the cave entrance, he tentatively surveyed the passages. 1985: renewal of research by regional caving club in Ruomberok city, later renamed a group Vchod East as a part of the Commission for speleodiving.THE SKALIST POTOK CAVE IN THE RELATIONSHIP TO THE RELIEF OF THE SOUTH PART OF JASOVSK PLATEAU (SLOVAK KARST) AFTER 25 YEARS OF RESEARCHAlena Petrvalsk, Zdenko Hochmuth Institute of geography, Faculty of Natural Sciences, P. J. afrik University, Jesenn 5, Koice 040 01, Slovakia, alena.petrvalska@upjs.sk zdenko.hochmuth@upjs.sk The Skalist potok Cave is situated on the south slope of the Jasovsk Plateau. In present, it is the longest and deepest cave in the Slovak Karst. It consists of three different parts, having different genesis and relationship to relief of slopes and plateau. We analyze location of cave corridors and relief above them. We can confirm that cave corridors do not have relation to the relief itself, they do not pass under dolines and they do not correspond to doline lines or lines of slope valleys. We deal with further perspectives of the research, possibilities and applications of new progressive methods that can help us solve some research problems (relationship of microclimate and discharge, neotectonic movements etc). 1. IntroductionThe Slovak Karst is situated in the eastern part of south Slovakia. It forms the largest continuous karst area in the Slovakia overreaching to Hungary where it is called the Aggtelek Karst. In Slovakia, it occupies an area of about 800 km2and is built up of karst rocks.2. Geological and geomorphological character of Slovak KarstThe Slovak Karst has a complicated geological structure with five tectonic units (nappes): Silicicum, Gemericum, Brka Nappe, Meliaticum and Turnaicum. A bottom part of nappes are built up of Lower Triassic non-karstifying rocks overlain by dominantlight grey and white Wetterstein Limestone, and Wetterstein Dolomite, Guttenstein Limestone and Dachstein Limestone (with megalodont fossils). The Slovak karst is characterized by uniform pediplanation surface. We expect that whole area of the Slovak Karst was in the time of tectonic calm in Miocene (Pannonian age) cut into the planation surface peneplain. New ideas on the evolustion of the Slovak Karst were presented by Jakl (2001), Gal and Bella (2005), Gal (2008), Hochmuth and Labunov (2008), Hochmuth and Petrvalsk (2010). After the uplift in the Rhodanian phase of Carpatian orogenesis, canyonand gorge-like valleys were created. They separate the area into several plateaus: Koniar, Figure 1. Slovak karst plateaus and the most important caves.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings201

PAGE 204

1986: In the Skalist potok Cave, speleodivers Hochmuth, Ko bik, imkovi and ur ek managed 11. sump in the distance of 600 m from the cave entrance. 1987: 21. sump was reached, the length of the cave is 1,500m. Detailed mapping with mining set (Meopta) was running. 1988: Discovery of the tributary flow in the section of 17. sump. 1989, February 11: Cavers overcame sump No. 17.11, longest in the cave with a length of about 130 m. In the same year they discovered new areas of the cave, ascendent dry cave with waterfalls and sinuous horizontal parts. 1991: In the ascending passage new research and mapping by classical methods started. The new highest point was attained (Vrchol 91 Peak 91) in the height of 441 m. 1992: Resumption in the passages against the underground river. The new highest point was discovered Vrchol peak 92 on the slope of a huge hall with waterfall. 1993: D. Hutan is going to help with new research in the cave. Climbing up the 31 m high waterfall and discovery of new subhorizontal parts with halls and sump No. 17.14. At this point the research was stopped. Denivelation of the cave is 287 m. 1994: A progress in the lower parts of the cave (Hochmuth, ur ek), overcome of the 31. sump and discovery of a long horizontal passage with sump No. 22. Damping of the speleodiving activity. 1999: Research of the Speleoaquanaut Club from Prague (the Czech Republic) and group from Kladno heading by D. Hut an. Overcome of sump 17.13 and discovery of next passages with sump and shallow inflow. In this time Peak 91 was overcome and the passage was near to the Jasovsk Plateau slope. 2000: Overcome of 22. sump and discovery of more than 1 km long new passages that are occurring nearby built areas of the Hj village. Caver began to perforate a tunnel from the surface to the Contact hall. 2002: A progress in the horizontal part, sumps No. 23 and 24 were overcome. 2007: Opening of the 2ndcave entrance on the plateau slope, revival of the speleodiving research in the upper parts. 2008: Sump No. 17.13 was overcome by M. Mahart and D. Hutan from Speleoaquanaut Prague. 2010: Gap of the crosscut above sump No. 17.13. 2011: The ascending part above Kladensk Passage and record denivelation 373 m was reached. 2012: The waterfall was overcome, new waterpassage with sumps Nos. 17.14, 17.15 and 17.16 was discovered, in the siphon 17.17 wasn`t attained. The cave length is under 8 km already. Many papers were published about this cave, for example Hochmuth (1989, 1992, 1994), Hochmuth, Danko and Hutan (2011), Hochmuth and Hut an (2012) etc.5. Genetic parts of the caveThe Skalist potok Cave has 3 genetically different parts. Lower subhorizontal part of the cave is situated between 164 m a.s.l. (the lowest localities are in sumps). It spreads into WSW and in some parts it generates meanders. In the straight direction the passages are more than 2,8 km long. Most of these passages are tectonically conditioned by fault area between the foothill of the plateau and the hollow basin. But typical are freatic, meander segment and flooded vertical well. A schematic profile of the cave is included in the paper Hochmuth (1994). Only a third of this part is overflowed by underground flow from a vertical passage. West heading passage is mainly dry, overflowed only in flood period of the year. It is interesting that this part of the cave is not increasing, but horizontal and every each siphon is in greater depth. Figure 2. Map of the Jasovsk Plateau and the Skalist potok Cave position.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings202

PAGE 205

6. Topical problems of researchThe research of the cave with classical methods, after the opening the upper parts of the cave, reached the top. At present time, there is research in three ways running: Observation of neotectonic movements with exploitation of TM 71 dilatomer, which was installed on the active zone. This fault formed north-south direction of the Upper level in the cave. Measurement of water discharge with automatic depth sensor which was installed on Thompson weir situated on the upper stream in the cave. It helps us understand relationship between rainfall, temperature and snow melting on the plateau surface. We proposed a solution to this question by divergention against the water flow longer time ago. A extension of this branch is problematic since there are permanent siphons (Nos. 23) situated, passages on the fault zones are tight and collapse-like. In sediments of this part, boulders of metamorphic rocks are occurring indicating transport by flood waters from the west-situated Hjska Valley. The surface flow of the Hjsky Brook contains these bulders as well. In the fine fraction of sediment we observed heavy minerals that outline the catchment area of the stream. There is an evidence of hydrological connection with underground stream in the Kunia Abyss (250 m) by tracers. The water passed from the siphon to the spring in 5 weeks in minimal discharge. It is possible to consider this part of the cave as relatively young developing in parallel with river plain of the Hjsky Brook and its alluvial cone. Vertical parts with meander passages. The passages that P. Bella called vadose canyons and other authors (P. Hipman, Z. Hochmuth) called caves of the mountain type, are known from the Alps and in Slovakia and Poland from higher mountains in the Carpathian range (the High and Low Tatras). Their existence in the Slovak karst region was surprising. The waterfalls of 2 m height alternate with meander passages of oval shape (profile) and with vertical canyons. This part heads generally to the north however not following tectonic line or lines. Some parts even avoid tectonic structures identified on the slope surface. Into these parts a new entrance was made to better access to the highest passages of the cave. Water channel is currently developing, in some parts completely without sediments and in other parts with corroded sinter crusts. These parts witness of accumulation phase in long-term period (paleomagnetic research of speleothem has been in progress). Vadose canyon is on some places 20 m high and higher parts have interesting crystal trappings and speloethem forms similar to helictits. Upper subhorizontal part is situated between 480 m a.s.l. There was an active underground flow with the existence of sumps in this part. Their position is not typical for such a high altitude. Passages are oriented northwards generally and east-west parts are tectonically conditioned by fault area. These passages are 200 m from the rim of the plateau and situated only 25 m under the plateau surface. This subhorizontal part consists of fossil freatic level 10 m above active flow, in the highest fossil passages an acoustic connection was established. Figure 4. Character of water channel in the upper parts. Photo by J. Stankovi Figure 3. Schematic profile of the Jasovsk Plateau (under the angle of 45).Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings203

PAGE 206

Research of sediments with classical mineralogical and petrographical methods to recognize a catchment area of non-karstic components. Prolongation research was centered on the area behind sumps Nos. 17.14.17 with major inflow. Direction of this passage to the NE only a few meters under the plateau surface is seemingly interesting and it will be necessary to work in here in the future. For practical survey in the cave it would be helpful to gap a next entrance from the plateau surface (near the Vrchol Top 2011). The survey in the down parts of the cave is possible only with change in research philosophy, methods and new generation of speleodivers. Another problem is a speleological connection with Kunia Abyss. It belongs to the catchments area of the Skalist potok spring and the research in past was in progress in both ways (from the Kunia Abyss bottom and in Skalist potok Cave from the sump No. 17.9). The revival of the research by group of new cave divers would be desirable.7. SummaryRelation of corridors in the Skalist potok Cave and its relief above them is shown on the map. This cave, similarly to other caves avoids to depressions on the relief. The best evolved parts of the cave are under the crest (covered by karren) of the Jasovsk plateau, however not under the slope valleys as we expected. The passing of spaces under the relief does not imply it. One part lying under the plateau relief nearby the slope and dry valley does not copy the line of the valleys (tectonic line of NS direction), but runs Figure 5. Fossil freatic passages in the upper parts. Photo by J. Stankovi Figure 6. Specific forms of stalactites and stalagmites formed under the conditions of flowing aerosol from the waterfalls. Photo by P. uster. Figure 7. Specific forms of stalactites and stalagmites formed under the conditions of flowing aerosol from the waterfalls. Photo by P. uster.through the valley bottom to the second site, and it continues there. In the cave one can find recrystalised sinter, which can, according to Hochmuth (in verb.) show the older age (pre-Quaternary), similar sinters were found in the older upper fossil level in the Drienoveck Cave. On the cave Karst and Caves in Carbonate Rocks, Salt and Gypsum poster2013 ICS Proceedings204

PAGE 207

wall, large manganese surface films (layers) under the sinters can be observed. It will be very interesting to run further research in this cave. The exploration of the Skalist potok Cave gradually sheds more light on understanding of this part of the Jasovsk Plateau. After more than 25 years of surveying the cave, many speleological problems remain unresolved. Moreover, prolongation of surveyed corridors in every direction is likely to be made. After opening the entrance in the upper parts of the cave, detailed mapping and research of the reachable parts we compiled enough useful facts to understand the main questions of the genesis. On the basis of these findings, we can identify several paleolevels 250 m above the bottom of the valley in the length of 2 km. They are drained by a water stream of unknown origin and source location. The pre-Quaternary age is indicated by old, corroded speleothems, large calcite crystals and manganese crusts. There is no apparent affinity towards superficial landforms, dolines, even though these were ground surveyed and mapped in detail. Only some corridors of the WE directions can be related with the fault directions in the marginal parts of the karst plateau. We suggest the horizontal level of the cave (i.e. its fossil part) was formed in the Upper Pliocene. At that time, the relative relief between the plateau and the valleys was about 50 m. This is indicated by the high age of the speleothems and Fe and Mn sediments which were not formed during the Quaternary. Tectonic uplift caused individualisation of particular plateaus in the subsequent evolution. For that reason, the cave area lost its original function and they became the current water collector only later on. This water changes the lower parts of the cave, destroys the speleothems and it creates waterfalls and abyssal sections by attempting to reach the current water level in the karst massive.AcknowledgmentsThis work has been funded by project Nr. 1/1251/12 with financial support of the VEGA grant agency.ReferencesGal 2008. Geodynamika a vznik jask Slovenskho krasu. Speleologia Slovaca, 1, OP SR, Sprva slovenskch jask Liptovsk Mikul, 166 (in Slovak). Gal Bella P, 2005. Vplyv tektonickch pohybov na geomorfologick vvoj zpadnej asti Slovenskho krasu. In Slovensk kras, XLIII, Liptovsk Mikul, 17 (in Slovak). Hochmuth Z, 1989. Vsledky speleologickho prieskumu jaskyne Skalist potok. Slovensk kras 1989, Martin, 3 (in Slovak). Hochmuth Z, 1992. Novie poznatky z prieskumu jaskyne Skalist potok a morfolgia ast objavench v rokoch 1989. Slovensk kras, XXX, Martin, 3 (in Slovak). Hochmuth Z, 1994. Skalist potok al prklad divergencie podzemnho toku. Spravodaj SSS, 4, XXV, Liptovsk Mikul, 15 (in Slovak). Hochmuth Z, Danko S, Hutan D, 2011. Prieskum, topografia a nrt genzy hornch ast jaskyne Skalist potok (Slovensk kras). Slovensk kras, Acta carsologica slovaca 49, 2, 111 (in Slovak). Hochmuth Z, Hutan D, 2012. Sifny v hornch astiach jaskyne Skalist potok (Slovensk kras). Speleofrum, SS, Brno, 31, 94 (in Slovak). Hochmuth Z, Labunov A, 2008. K priebehu podzemnch priestorov vznamnch jaskynnch systmov Slovenskho krasu vzh adom na s asn relief. In: Stav geomorfologickch vzkum v roce 2008, Brno, 15 (in Slovak). Hochmuth Z, Petrvalsk A, 2010. Povrchov a podzemn krasov formy Jasovskej planiny. In: Geografie pro ivot ve 21. stolet: Sbornk p sp vk z XXII. sjezdu esk geografick spole nosti po danho Ostravskou univerzitou v Ostrav 31. srpna 3. z 2010. Ostrava: Ostravsk univerzita v Ostrav, 15 (in Slovak). Jakl J, 2001. Vvoj relifu Slovenskho krasu v etape neotektonickho vyzdvihnutia zemia. Slovensk kras, XXXIX, Liptovsk Mikul, 7 (in Slovak).Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings205

PAGE 208

CAN CONDUIT VOLUMES OBTAINED FROM ARTIFICIAL TRACER TESTS BE TRUSTED?Anna Vojt chov1, Ji Bruthans1, Ondrej Jger2, Frantisek Krej a3 1Faculty of Science, Charles University in Prague, Albertov 6, Praha 2, Czech Republic, bruthans@natur.cuni.cz2AQH s.r.o., Frdlantsk 1310/23, 182 00 Praha 8, jager@aqh.cz3Management of Chnov Cave, Doln Ho ice 54, 391 55 Chnov, Czech Republic, krejca@caves.cz Phreatic loop in a natural carbonate cave was used to test the reliability of artificial-tracer tests for estimating the volume of a flooded karst conduit (Fig. 1). The volume of a phreatic tube was measured by filling a drained phreatic loop with a constant inflow over a known time period. The volume of the phreatic loop is 190 m3, and it was compared to independent calculations of conduit volumes based on values based on tracer breakthrough curves (Vojtechovska et al. 2010). The best results were for mean transit time, where tracer-test calculations yielded volumes very similar to the volume obtained by direct filling of the loop (Table 1). On the other hand, using the first-arrival time or peak time in the volume calculation resulted in considerable underestimation of the phreatic tubes volume, and these methods should be avoided except when breakthrough curves are affected by molecular diffusion. This demonstrates that volume estimation by tracer tests may be quite precise for common natural conduits, but results are strongly affected by the chosen breakthrough-curve parameter.AcknowledgmentsResearch was supported by institutional project No. MSM0021620855.ReferencesVojt chovsk A, Bruthans J, Krej a F, 2010. Comparison of conduit volumes obtained from direct measurements and artificial tracer tests. Journal of Cave and Karst Studies 72, 3: 156. Table 1.Times and corresponding calculated volumes (modified from Vojtchovsk et al. 2010). Time MinutesCorrespondingCompared to aftervolumereal volume injection(m3) (%) Real volume (pumping) 190 100 tAfirst arrival time 116 85 45 tPpeak time 176 129 68 tRmean residence time231 169 89 tCmean transit time290 212 112Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings206

PAGE 209

Figure 1. Map and longitudinal vertical section of sump in Chnov Cave (modified from Vojt chovsk et al. 2010).Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings207

PAGE 210

THE TECTONIC CONTROL OF AN UNDERGROUND RIVER NETWORK, AGIA TRIADA CAVE (KARYSTOS, GREECE)Emmanuel Vassilakis1, Kyriaki Papadopoulou-Vrynioti2 1National and Kapodistrian University of Athens, Department of Dynamics, Tectonics and Applied Geology, Panepistimioupolis Zographou, 15784 Athens, Greece, evasilak@geol.uoa.gr2National and Kapodistrian University of Athens, Department of Geography & Climatology, Panepistimioupolis Zographou, 15784 Athens, Greece, papadopoulou@geol.uoa.gr The water pathways of the underground river of Agia Triada (Karystos, Greece) and their generation are examined in this study. One of the longest caves explored in Greece is formed at heavily deformed metamorphic rocks and the suggested combined methodology, which includes traditional geological mapping, speleological exploration and remote sensing image interpretation, led us to the conclusion that the water flows along the hinge of a NESW-trending mega-fold. A number of faults that have been activated after the generation of the underground river, have altered its pathway by creating knick-points which host impressive subsurface waterfalls, the largest of which is about 20 m high. The extraction of morpho-lineaments from ortho-rectified satellite images revealed the importance of structures that were identified on the open surface mainly by high-resolution remote sensing data interpretation and are related to the cave development. This was made feasible with the use of the Geographic Information Systems as all the collected data were converted into layers for further interpretation. It proved to be very useful as the projection of the cave trace on the ortho-rectified data reveale d the underground linkage between two adjacent hydrological basins. This explained the unusual large quantities of water discharged by the Agia Triada spring. 1. IntroductionIt is quite often to study an underground karstic landform only regionally, since the major subject is the generation and operation of a cave. Especially in caves that act as underground river flows, the main question that arises is whether this subsurface network should be part of the surface drainage or not (Papadopoulou-Vrynioti 2002; Papadopoulou-Vrynioti and Kampolis 2011). In this paper we investigate the role of the tectonic structures at the local karstification and the influence of several tectonic structures at the underground connection between two separate subbasins at the area of Karystos (southern Evia, Greece), where the Agia Triada Cave has been found. It is classified as one of the longest caves in Greece (Petrochilou 1981). The existence of an impressive underground waterfall, among lakes and other unique speleological features attracted a great number of expeditions for investigating the cave since 1932 (Zervoudakis 1959; Avagianos 1981). The first fieldwork was carried out during 1994 by doing a speleological expedition as well as geological mapping at the open surface around the cave trace (Vassilakis and Vlachou 1995). Remote sensing techniques were also applied on an ortho-rectified multi spectral IKONOS satellite image and panchromatic aerial photographs by using a 25 m Digital Elevation Model, for increasing the mapping accuracy of the area around the Agia Triada spring which discharges the underground river. Moreover digital and fieldwork data were imported, combined and interpreted extensively in a Geographic Information System. The orientation of the cave mapping (by SPELEO Club speleologists) as well as the projection of the cave trace on the surface and combined with all kinds of the available data, proved to be very helpful as it was really important to relate the fieldwork results (geological and speleological) with the remote sensing data interpretation. The applied methodology seems to be ideal for cases like this one as underground rivers are more often related to tectonic structures that might have surface expression. In such cases remotely sensed datasets with proper interpretation can lead to the structures that were responsible for the cave generation.2. Geomorphological and geological settingThe discharge of the studied underground river is done by a karstic spring, which is located at an elevation of 250 m, next to the Agia Triada chapel just north of the coastal town of Karystos, at southern Evia, in Greece. The cave extends NE beneath the mountain of Ochi and great amount of precipitation water that is infiltrated through its permeable rocks cropping out on the surface, flows along a naturally constructed tunnel. The quantity of the water discharge is large enough for providing Karystos and the surrounding villages, domestic, potable water throughout the duration of the year. The cave is located in a hydrological basin which covers an area of about 20 km2discharging most of the WSW slopes of Ochi Mt. since its watershed reaches the highest peak of the mountain at nearly 1,400 m of elevation (Fig. 1). The hydrographic network reaches the 5thorder of Strahler classification (Strahler 1957). It comprises of two 4thorder branches, which divide the triangular shaped basin into two asymmetrically developed sub-basins. The westernmost main branch flows almost parallel and in a relatively small distance from the watershed margin, as well as the easternmost branch flows next to the eastern margin of the basin. The cave entrance is located almost at the central area of the basin and in the westernmost sub-basin. The projection of the underground river, which was mapped by SPELEO Karst and Caves in Carbonate Rocks, Salt and Gypsum poster2013 ICS Proceedings208

PAGE 211

Figure 2. Simplified geological map of southern Evia (upper plate). The white box shows the magnified area (lower plate) around the surface projection of the cave trace (blue line) where metamorphic rocks are cropping out. The cave entrance and the waterfall locations are also noted.3. MethodologyThe general idea of the applied methodology was to enhance the classic fieldwork of speleological exploration and geological mapping with the aid of the GIS and highresolution remote sensing datasets. 3.1. Fieldwork data The geological and structural mapping revealed that the entire area has been folded during the alpine period (late Jurassic early Cretaceous). Most of the folds that are observed in all kinds of scales have axis trending generally along the NESW orientation. This orientation is compatible with deep ductile deformation (Katsikatsos et al. 1976). It is expressed with isoclinal, overturned and recumbent folds with many orders of folding. A second deformation incident seems to have happened after the previous one as folding of the already folded rocks is observed. It is expressed by open folds with almost vertical axial plane and NNWSSE-trending axis. The most likely age of this deformation is during Oligocene (Papanikolaou 1978). Additionally, three main systems of faulting were identified throughout the wider area of southern Evia. Most of the faults trend in the NWSE direction but there is also a large number of faults trending either NESW or ENEWSW (Latsoudas and Triantafyllis 1993). Club speleologists during several explorations, onto the surface shows a subsurface connection between the two sub-basins. The main cave trace at least from the known mapping information has a SWNE orientation and it is clear that large amount of infiltrated water coming from precipitation on the eastern sub-basin flows through the underground river to the western sub-basin. The area comprises mainly of metamorphic rocks, parts of the Ochi unit (Papanikolaou 1986). Intercalations of marbles with sipolines and mica schists have been observed throughout the entire area and are geotectonically placed between a series of ophiolitic rocks and amphibolites on top and the basement rocks comprised of gneiss (Latsoudas and Triantafyllis 1993; Moustaka 2011) (Fig. 2). The formations that host the cave and can be identified on the surface along its projected trace (Fig. 2) are the sipolines and the schists. Quite often caves are guided by changes in lithology with passages developing along or close to the contact of carbonate rocks and underlying shales (Gillieson 1996). It is rather clear that in this case the underground river is developed on the contact between the permeable carbonate marbles and the impermeable schists. It is not quite clear if this contact is of tectonic or stratigraphic origin, since the host rocks are heavily deformed by various generations of folding. The deformation led to the generation of a dense network of discontinuities, which was identified either by fieldwork but also during the interpretation of the remote sensing data (see section 3.2). Our interpretation led us to the simplest model concerning the cave generation, arguing that the cave has been developed along a NESW-trending mega fold axis. Several open, erodible and karstified discontinuities were used by the water flow to offset its way to the discharge point at Agia Triada, but the general trend remains NESW and no major or minor fault of similar orientation has a surface expression near the projected cave trace. On the contrary quite a few NWSE-trending structures seem to crosscut the cave normally to its development, possibly affecting and altering the water pathways. Figure 1. Shaded relief map showing the geomorphology of the hydrological basin where the entrance of the cave is located.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings209

PAGE 212

3.2. Remote sensing data We combined a few remote sensing datasets for completing the required tasks and finally relate the fieldwork findings to the cave trace. Initially a medium resolution (25 m) DEM was constructed after digitizing the 20 m contours from scanned topographic maps of scale 1:50,000, for the entire area of southern Evia. The product was combined with a high-resolution ortho-photo mosaic constructed by panchromatic aerial photographs and both of the datasets were used for the ortho-rectification of a high-resolution satellite image captured by the IKONOS satellite. These tasks are necessary for merging the produced datasets with the cave trace projection, since the final interpretation needs to be performed in a large scale of the order of 1:5,000. The initial geological mapping was edited and finalized in a GIS where the basemap consisted of the four-spectralband IKONOS image (three bands in the visible and a fourth in the infrared spectra). Different band combinations lead to the production of several trueand pseudo-colour digital map compositions, aiming to reveal formations of similar mineral composition and especially the contacts between the various rock outcrops (Fig. 3). The orientation of all the mapped structures during fieldwork is in full agreement with the morpho-lineaments, which were extracted after the digital interpretation and combination of the DEM and the 1 m resolution orthorectified remote sensing data. These datasets were required for the construction of a morpho-lineament map for the upstream area of the underground river within a buffer zone of 1 km around the cave trace (Fig. 4). The morpholineaments are surface expressed lineament features, which might be either geological or geomorphological structures or neither of both. In any case these can be related to more or less significant structures that have affected the study area and left their imprints on the surface (Vassilakis 2006). The statistical analysis of 282 lineament features in total, which were identified on the produced pseudo-colour and true-colour images, show the exact main orientations with the field observations and measurements. It is more than clear by reading the rose diagram that the main orientation of the lineaments is along NESW (Fig. 4). Nevertheless, a secondary trend of NWSE is also identified. Even though both of the main calculated orientations are identical to those revealed by the field observations, it is clear that the remote sensing data interpretation altered the significance of the recorded orientation, by promoting the NESWtrending as the most significant. Moreover, the cave trace is developed along the very same orientation as the one that is suggested by the satellite image interpretation but on the other hand some sharp bendings seems to be related to the NWSE trends. 3.3. Speleological data The cave has been developed along the initial stratigraphic contact of the sipoline marbles and the schists that are found Figure 3. Pseudo-colour ortho-rectified image produced by digital interpretation of high spatial resolution (1 m) IKONOS satell ite image (2,4,1/R,G,B). The yellow line represents the projected cave trace on the surface.The green colours represent the vegetat ion cover whilst the bluish colours represent the uncovered rocks. The purple colours at the top represent an open excavation field for decorative stones (Plaka Karystou). Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings210

PAGE 213

all over the surrounding area. It is almost 1,800 m long and in spite of all the individual chambers, waterfalls, lakes and branches, it has all the characteristics of an underground river flowing along a naturally constructed tunnel. According to Boegli (1980), the classification the Agia Triada Cave is a contact large cave. The corrosion of the permeable carbonate rocks allowed the water to create a relatively high tunnel with a naturally created passage above the subsurface flow (Fig. 5). During the first few hundreds of meters no impressive stalagmitic decoration is being observed although the walls of the tunnel are usually covered by stalagmitic material, which in many cases make the passing through quite difficult. During the next part of the cave several lakes and highenergy water flow are observed until the arrival to a wide chamber with a large asbestite cover of 3 m wide and 6 m high, under which the river flows (Fig. 6). It seems that at this location there is a petrified waterfall since the water has managed to find a lower passage more recently. This section of the cave (420 m from the entrance) ends at a lake that covers an underwater passage, where diving equipment is necessary. There is also a very small passage next to this lake, which can be used for diverting it, after climbing on a 6 m high wall. It is an older branch of the river, which used to be active during the early stages of the cave generation and it was filled up with sandy material later. The widening of this passage was made recently by several exploring teams coordinated by the SPELEO Club. The next part of the cave is quite similar to the aforementioned one as the power of the water corrosion created another tunnel with many stalagmitic decorations hanging from its roof. After 150 m huge unstable boulders block the tunnel. Right after these there is a large opening where a deafening noise dominates. It is the noise of high water flow energy and it is caused by a 20 m high underground waterfall. The height of the chamber that hosts the impressive waterfall and the lake, which is formed at its bottom, is almost 40 m and its width about 20 m. The intercalations of the sipolines and the schists can be observed all over the walls of the chamber and it seems that this is a hinge of a mega-fold with several orders of minor asymmetric folding and the water path is created along the folding axis. The waterfall is formed normal to a knickpoint which seems to have been created by NWSEtrending fault. The cave is extending towards NE for at least 1,000 m beyond this chamber but reaching the mouth of the waterfall needs good climbing skills and equipment. Passing through this location a series of quite deep-formed lakes (>4 m) are observed and the way leads to another knick-point, which should be also overtaken by climbing. The final part of the cave extends into an impressively decorated NESWtrending tunnel with stalagmites and asbestitic covers until reaching the end of it about 1,800 m away from the entrance. It is worth to mention that at a close distance from the entrance, findings of Early Bronze Age were identified, as well as Final Neolithic material in large quantities (Nikolaidis and Tazartes 1998; Mavridis and Tankosi 2009). Figure 4. Morpho-lineament map at a buffer zone of 1 km upstream the cave trace (blue line). The statistical interpretation of the lineament features is shown at the inset rose-diagram. Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings211

PAGE 214

Figure 5. The underground river flows beneath the naturally created passage along the first few hundred meters after the cave entrance (Photo by Emm. Vassilakis). Figure 6. Petrified waterfalls above the contemporary water flow show a more recent migration of the underground river water path at a lower level (Photo by Emm. Vassilakis).4. ConclusionsThe combination of traditional field mapping, speleological exploration and innovative techniques of the GIS data manipulation and remote sensing image interpretation provided useful tools for generating an effective methodology. The underground river of Agia Triada is clearly controlled by tectonic structures. It is formed and developed along preexisting tectonic structures, which are dominant at the area of southern Evia. The main structure is a NESW-trending hinge of a mega-fold affecting marble sipolines and mica schists cropping out all over the wider area. The water pathway is placed on the contact between the permeable carbonate layers and the impermeable schists. An underground almost linear linkage between two sub-basins is created and this explains the unusually large quantities of water that are discharged at the Agia Triada spring. Several faults trending normally (NWSE) to the cave orientation have vertically offset the hinge resulting knickpoints of various scales and consequently impressive subsurface waterfalls. These morphological discontinuities are not visible at the open surface where erosional proceedures smoothened the relief.AcknowledgmentsThe authors would like to thank P. Stylianos, J. Speis and J. Froudarakis for their valuable assistance and participation during the speleological fieldwork. The IKONOS scene was kindly offered by TotalView Inc. The comments on the original manuscript by P. Bosak and M. Filippi were highly appreciated.ReferencesAvagianos G, 1981. Recent achievements in underwater speleology and in exploration of deep potholes. Bulletin H.S.S., 18, 553. Boegli A, 1980. Karst hydrology and physical speleology. Springer, Berlin, 284. Gillieson D, 1996. Caves: Processes, Development, and Management. Blackwell, Oxford, 324. Katsikatsos G, Mercier J, Vergely P, 1976. LEube mridionale: une double fentre polyphase dans les Hellnides internes (Grce). C. R. Acad. Sci. Paris, 283, 459. Latsoudas C, Triantafyllis E, 1993. Geological map of Greece, 1:50,000, Sheet Karystos-Platanistos. IGME, Athens. Mavridis F, Tankosi Z, 2009. The Ayia Triadha cave, southern Euboea: finds and implications of the earliest human habitation in the area (a preliminary report). Mediterranean, Archaeology and Archaeometry, 9, 47. Moustaka E, 2011. Structural analysis and geochemical stydy of minerals of the metamorphic rocks of Mt. Ochi, Southern Evia. MSc Thesis, Faculty of Geology & Geoenvironment. NKUA, Athens, 169. Nikolaidis S, Tazartes N, 1998. Ayia Triada cave (Karystos), Conf. on Human and Speleoenvironment, Athens, 179. Papadopoulou-Vrynioti K, 2002. Lost rivers associated with hazard pollution in Greece, 6th Int. Symposium on History of Speleology and Karstology. ALCADI, Gorizia, Italy, 55. Papadopoulou-Vrynioti K,.Kampolis I, 2011. The SelinitsaDrakos coastal karstic system in the Messinian Mani Peninsula (southwestern Greece) in relation to the terrestrial geoenvironment. Geologica Balcanica, 40(1), 75. Papanikolaou D, 1978. Contribution to the geology of Aegean Sea: The Island of Andros. Ann. Geol. Pays Hell., 29, 477. Papanikolaou D, 1986. Geology of Greece. (in Greek). Eptalofos Publ., Athens, 240.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings212

PAGE 215

Petrochilou A, 1981. The deepest and longest caves of the world and Greece. Bulletin H.S.S., 18, 375. Strahler AN, 1957. Quantitative analysis of watershed geomorphology. Trans. Amer. Geophys. Union, 38(6), 913. Vassilakis E, 2006. Study of the tectonic structure of Messara basin, central Crete, with the aid of remote sensing techniques and G.I.S. Ph.D. Thesis, NKUA, Athens, 546. Vassilakis E, Vlachou S, 1995. Operation mechanisms of southern Evia springs. BSc Thesis, Faculty of Geology. NKUA, Athens, 125. Zervoudakis I, 1959. Agia Triada cave, no 115. Bulletin H.S.S., 5, 76.Karst and Caves in Carbonate Rocks, Salt and Gypsum poster 2013 ICS Proceedings213

PAGE 216

214

PAGE 217

215 2013 ICS Proceedings Karst and Caves in Other Rocks, PseudokarstSession:

PAGE 218

216

PAGE 219

CAVE FORMATION INITIATED BY DISSOLUTION OF CARBONATE CEMENT IN QUARTZOSE SANDSTONESJi Adamovi1, Radek Mikul1, Tom Navrtil1, Jan Mertlk2 1Institute of Geology AS CR, v. v. i., Rozvojov 269, CZ-165 00 Praha 6, Czech Republic, adamovic@gli.cas.cz2Bohemian Paradise PLA Administration, A. Dvo ka 294, 511 01 Turnov, Czech Rep., mertlik@cesky-raj.info Besides cavities of irregular shape, European sandstones also feature symmetrical cavities of spherical, ellipsoidal or teardrop shapes. Most of them are tens of centimetres across but some reach as much as 2 m in diameter and may coalesce into large caves tens of metres in length. Their origin has not been clearly explained yet. Based on the field comparison between such cavities in quartzose sandstones and incompletely developed cavities in carbonate-cemented sandstones, it can be demonstrated that the symmetrical cavities form by carbonate dissolution within the limits of former carbonate concretions. The diagnostic features of post-concretionary cavities include their circular or elliptical cross-section, a uniform orientation of their long axes across the region, and the presence of a set of parallel vertical joints or grooves/ribs on their inner walls. In some sandstone areas (e.g., Petite Suisse area in Luxembourg, Koko n area in the Czech Republic), a wide variety of transitional forms can be found between the cavities and concretions forming positive relief on a vertical cliff face, depending on the position of the carbonate dissolution front in the present landscape.1. IntroductionSymmetrical cavities and caves in sandstone tens of centimetres to several metres in diameter were studied in the temperate zone of Europe: in the Petit Suisse area in Luxembourg and Germany, in the Koko n area and Kloko sk skly Cliffs in the Czech Republic, and in the Outer Flysch Belt of the Carpathians in Poland and Slovakia. The origin of many symmetrical (spherical or tunnel-like) caves in quartzose sandstone can be hardly explained by processes like salt weathering or stream erosion and must be attributed to processes of epigenetic carbonate dissolution.2. Preconditions of symmetrical cave originDepositional systems giving rise to thick packages of quartzose sandstones are characterized by high supply of detrital material from the source area and relatively dynamic sedimentary environments separating the deposited sand from finer particles which remain in suspended load. At tropical/subtropical climates with enhanced carbonate production, sand deposited in marine and lacustrine environments may become mixed with lime mud. At protected places, bottoms of seas/lakes are colonized by molluscs with calcareous shells. These may get preserved in situ or redeposited into shell lags during storm events. Under diagenetic conditions, lime mud and calcareous shells become recrystallized into sparitic cement (Adamovi 2005). Where nucleation centres are regularly dispersed in sandstone, newly formed cement grows into spherical concretions or elongate, cigar-shaped forms. Later during diagenesis, these forms get vertically compressed to a variable degree and their cross-sections become somewhat elliptical. Where carbonate crystallization is bound to a specific bedding plane (e.g., a shell lag), irregular but moreor-less continuous sheet-like bodies of cemented sandstone are formed. Under regional tectonic stress, cemented sandstone is prone to brittle deformation. This is caused by its higher elastic modulus compared to uncemented sandstone. A network of fractures forms within the concretions, parallel to the maximum principal stress (Bessinger et al. 2003; Quesada et al. 2009). With the emergence of the rock massif to near-surface conditions, pore spaces in the sandstones become flushed with meteoric waters. Dissolution of the sandstone compounds by infiltrating meteoric waters represents a natural process of acidification. The most reactive minerals such as calcium carbonate can buffer the acidity from deposition. At some sites the pH of atmospheric precipitation became extremely low due to the anthropogenic pollution. The low pH of precipitation in central Europe caused by high concentrations of acidifying compounds (especially SO4and NO3), can be further decreased upon infiltration into the sandstone by many processes, e.g., dissolution of organic material derived from the biomass (recent vegetation). Edmunds et al. (1992) reported pH values of interstitial waters ranging between 4.0.5 in the United Kingdom and pH values of 3.5.5 were reported in the Bohemian Switzerland National Park in N Bohemia (Va ilov et al. 2011; Navrtil et al. 2013 in press). The relatively slow buffering of infiltrating interstitial waters in low-Ca sandstones is usually due to their overall low buffering capacity. While resistant minerals such as quartz remain relatively stable under acidic conditions, carbonate cement (if present) will be readily dissolved.3. Areas of studyIn the Petit Suisse area in Luxembourg and Germany, the shallow marine Lower Jurassic Luxembourg Sandstone provides vast outcrops in the basin of the Sre (Sauer) River and its tributaries. Quartzose sandstones contain 0.1 to 2 m thick beds and concretions of calcite-cemented sandstone (Van den Bril and Swennen 2009). While most of them form positive relief on vertical cliff faces, others are partly or completely altered disintegrated into loose sand. The caves formed by calcite dissolution are strata-bound, max. 8 m long and 1.5 m high. Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings217

PAGE 220

Late Cretaceous quartzose sandstones of shallow marine origin are well exposed in the Koko n area N of Prague, Czech Republic. Intercalations of calcareous sandstone are preserved in marginal areas of siliciclastic deposition, at levels corresponding to highstand systems tracts (e.g., several metres above the Upper Turonian base). In the centre, symmetrical cavities as much as 2 m in diameter are displayed at the same levels, with occasional loosened cores of calcareous sandstone (Koko n, Vrto ez Gorge). At Jeovice, tube-shaped caves max. 10 m long and 2 m in diameter are a common phenomenon (Zimerman 1995; Adamovi and Mikul 2011). Quartzose sandstones of the same age (Upper Turonian) form a 140 m thick body in Kloko sk skly Cliffs near Turnov, Czech Republic. Spherical cavities 1 m to several metres in diameter concentrate at a level ca. 15 m below the top of the body. These cavities merge to form caves tens of metres long (Vtek 1987, Mertlk and Adamovi 2005). Spherical concretions max. 1 m across can be also occasionally found on vertical cliff faces, forming positive relief. The Ci kowice Sandstone in Slovakia and Poland (Paleocene to Lower Eocene) is a unit of quartzose sandstones and subarkoses belonging to the Outer Flysch Belt of the Carpathians. In Slovakia, spherical calcitecemented concretions max. 2.6 m in diameter are found in fresh exposures (Megonky) and on cuesta edges (Kloko ov), forming positive relief (Adamovi et al. 2010). Further east in the Beskidy Mts. (Ci kowice rock city, Prz dki rock city near Krosno), the calcareous cement is dissolved, leaving behind cavities and rock basins.4. Weathering forms observedThe presence of carbonate cement in packages of quartzose sandstone is associated with a range of weathering forms. The following typical forms were encountered. 1. Carbonate-cemented beds andconcretions forming positive relief on vertical cliff faces are a typical feature of the Luxembourg Sandstone, although they can be found at favourable sites in almost all areas. Cemented sandstone is characterized by a higher resistance to physical weathering and a lower total effective porosity than the ambient uncemented rock. Dense vertical fracturing is limited to the cemented part and maintains uniform strikes throughout the particular area. A thin (several cm) zone of onion-like scaling lines the outer edge of the concretions, particularly the spherical ones. 2. Concretions lined by a groove on their outer edge. The groove marks a zone more susceptible to granular disintegration and is sometimes composed of a series of honeycomb pits. Within the concretion, carbonate cement is either preserved or dissolved. The transition from #1 to #3 is associated with solutional widening of the vertical fractures, especially around their top ends. 3.Spherical or ellipsoidal cavities are typical for sandstone areas in the Czech Republic and for the Beskidy Mts. in Poland. Relics of carbonate-cemented sandstone are occasionally found inside the cavities. If not, their affinity to carbonate-cemented concretions can be deduced by the presence of dense vertical fractures on the inside walls, usually widened into morphologically prominent ribs, or the preserved onion-like weathering pattern on their outer edge. Figure 1. Calcite-cemented concretions in quartzose sandstone forming positive relief. Perekop near Berdorf, Luxembourg. 49.81794N, 6.37528E. Figure 2. Partly dissolved calcite-cemented concretion inside an elliptical cavity in quartzose sandstone at Perekop near Bedorf, Luxembourg. Note the dense vertical jointing within the concretion limits. 49.81816N, 6.37470E. Figure 3. A relict of a vertically jointed calcite-cemented concretion inside an elliptical cavity partly re-modelled by salt weathering. Quartzose sandstone at Koko n, Czech Republic. 50.42958N, 14.57294E.4. Symmetrical cavities with flattened bottoms and arched tops (teardrop-shaped cavities), partly filled with loose sand, can be attributed to this group based on their symmetrical shapes and topographic and/or stratigraphic cooccurrence with #1. Vertical ribs are sometimes present on the inside walls. Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings218

PAGE 221

5. Mineral composition of cementMineral cement was studied using optical microscopy (Olympus DP70 polarizing microscope) and X-ray diffraction method (Bruker D8 Discover powder diffractometer) at Institute of Geology AS CR, v.v.i. in Prague. Calcite was detected in samples of carbonate concretions from Perekop near Berdorf, Luxembourg, in a sample of a relict of a carbonate concretion from Koko n, Czech Republic, and in a sample from the core of a large spherical concretion at Megonky, Slovakia. Samples taken from concretions in the Kloko sk skly Cliffs, Czech Republic, show mostly no mineral cement; previous existence of carbonate or silica cement can be supposed. lowering of the base level, the acidification front is progressively sinking. Carbonate cement within the concretions in sandstone becomes dissolved and their quartz grains are set loose, creating sand nests inside a compact sandstone. When a former carbonate-cemented bed or a level of former concretions become reached by valley incision, the loose sand becomes evacuated, leaving behind open cavities. The sand nests in Late Cretaceous sandstones as a stage transitional to cavity formation were previously reported from Saxony (Seifert 1936) and Poland (Dumanowski 1961). Field evidence from the European sites suggests that the preservation potential for carbonate cement in quartzose sandstones is enhanced basically by three factors: 1, low precipitation. This is evidenced by the contrast between complete carbonate dissolution in the Gry Sto owe National Park in Poland with a total annual precipitation of max. 1,100 mm vs. weak carbonate dissolution in the Petite Suisse area in Luxembourg with a total annual precipitation of 700 mm; 2, local topography. The least carbonate dissolution is observed near the base level at sites with high relief dynamics, such as cliff bases in narrow valleys (Koko n); 3, the presence of carbonate-rich beds at topographically higher positions. This is best visible in the area around Turnov, Czech Republic where carbonate is presently precipitated in sandstones overlain by loess deposits (Trosky, P hrazy). Symmetrical cavities in most areas are larger in size than the observed carbonate concretions. Those with arched tops (form #4) seem to be enlarged by the process of salt weathering, as evidenced by the presence of honeycomb pits inside the cavities and by salt efflorescences. Their flat bottoms can be equally shaped by frost weathering. Some of the spherical cavities still possess a lining of concentric onion-like scales, which suggests only a slight or no enlargement after the concretion has been removed (gravitational detachment of big parts of concretions from the Cikowice Sandstone was observed with no solution involved). Other spherical cavities have been enlarged considerably and coalesced into caves tens of metres long. Their walls bear no honeycomb pits, which suggests a slow grain-by-grain disintegration (Kloko area). By dating calcitic speleothems on their walls, Bruthans et al. (2012) found that no enlargement occurred after the Late Glacial and attributed the growth of such cavities to their ventilation in the Glacial period.7. ConclusionThe variety of carbonate-solutional forms observed across Europe allows to conclude that the origin of symmetrical cavities in sandstone through dissolution of carbonate cement is a widespread phenomenon in humid temperate regions. After cement dissolution, sandstone within the limits of the original carbonate concretion or bed turns into loose sand which becomes evacuated upon subaerial exposure. The enlargement and coalescence of the cavities thus formed involves salt weathering, frost weathering and other processes leading to granular disintegration of their walls. In larger cavities/caves, a specific ventilation system may play a role. Figure 4. A cave with dense vertical jointing on its interior walls formed by dissolution of carbonate cement within the limits of a former concretion. Luxembourg Sandstone, Teufelsschlucht area, Germany. 49.84750N, 6.43826N. Figure 5. A cave in quartzose sandstone with vertical joints on its walls. The joints do not continue to the ambient rock, suggesting a carbonate-concretion precursor of the cave. ab jeskyn Cave near Jeovice, Koko n area, Czech Republic. 50.44203N, 14.44529E.6. Initiation and enlargement of cavitiesThe pore space of aquifers in sedimentary basins is affected by fluids of various pH and various red-ox conditions throughout the basin history. With the emergence of the aquifer to near-surface conditions, exposure to waters of meteoritic origin becomes a rule. The slightly acidic reaction of rainwater due to pCO2in the atmosphere becomes strongly acidic in areas where organic-rich soils develop on a quartzose sandstone bedrock. With ongoing erosion and Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings219

PAGE 222

Bessinger B, Cook NGW, Myer L, Nakagawa S, Nihei K, Benito P, Suarez-Rivera R, 2003. The role of compressive stresses in jointing on Vancouver Island, British Columbia. J. Struct. Geol., 25, 983. Bruthans J, Schweigstillov J, Jen P, Churkov Z, Bezdi ka P, 2012. 14C and U-series dating of speleothems in the Bohemian Paradise (Czech Republic): retreat rates of sandstone cave walls and implications for cave origin. Acta Geodyn. Geomater., 9(1), 93. Dumanowski B, 1961. Forms of spherical cavities in the Sto owe Mountains. Zeszyty nauk. Univ. Wroc aw., Ser. B, 8, 123. Edmunds WM, Kinniburgh DG, Moss PD, 1992. Trace metals in interstitial waters from sandstones acidic inputs to shallow groundwaters. Environmental Pollution, 77(2), 129. Mertlk J, Adamovi J, 2005. Some significant geomorphic features of the Kloko Cuesta, Czech Republic. Ferrantia, 44, 171. Navrtil T, Va ilov Z, Rohovec J, 2013.Mobilization of aluminum by the acid percolates within unsaturated zones of sandstones.Environmental Monitoring and Assessment. Online First. DOI 10.1007/s10661-013-3088-4. Quesada D, Leguillon D, Putot C, 2009. Multiple failures in or around a stiff inclusion embedded in a soft matrix under a compressive loading. Eur. J. Mechanics, ASolids, 28, 668. Seifert A, 1936. Sandnester im Turon-Sandstein der Schsischen Schweiz und ihre Bedeutung fr Verwitterungsformen (Wannen, Opferkessel und Hhlchenbildungen). Sitzungsberichte und Abhandlungen der Naturwissenschaftlichen Gesellschaft Isis in Dresden 1935, 136 (in German). Van den Bril K, Swennen R, 2009. Sedimentological control on carbonate cementation in the Luxembourg Sandstone Formation. Geol. Belgica, 12, 3. Va ilov Z, Navrtil T, Dobeov I, 2011. Recent atmospheric deposition and its effects on sandstone cliffs in Bohemian Switzerland National Park, Czech Republic. Water, Air and Soil Pollution220 (1), 117. Vtek J, 1987. Pseudokrasov tvary v pskovcch Kloko skch skal. eskosl. Kras, 38, 71 (in Czech). Zimerman V, 1995. Jeskyn Koko nska msto pr vodce. Mstn noviny Podbezdz, Kruh (in Czech). Figure 6. A succession of forms during the formation of symmetrical cavities by the dissolution of carbonate cement in quartzose sandstones.AcknowledgmentsThis research was supported by Project IAA300130806 of the Grant Agency of the Academy of Sciences of the Czech Republic and falls within the Research Plan RVO 67985831Z of the Institute of Geology AS CR.ReferencesAdamovi J, 2005. Sandstone cementation and its geomorphic and hydraulic implications. Ferrantia, 44, 21. Adamovi J, Mikul R, 2011. Vznik n kterch elipsoidlnch dutin rozpout nm karbontovho tmelu v pskovcch jizerskho souvrstv na Koko nsku. Zprvy o geologickch vzkumech v roce 2010, 9 (in Czech). Adamovi J, Mikul R, Clek V, 2010. Atlas pskovcovch skalnch m st esk a Slovensk republiky. Academia, Praha.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings220

PAGE 223

ARENITIC CAVES IN VENEZUELAN TEPUIS: WHAT DO THEY SAY ABOUT TEPUIS THEMSELVES?Roman Aubrecht1,2, Tom Lnczos3, Jn Schlgl1, Luk Vlek4, Branislav mda5 1Department of Geology and Paleontology, Faculty of Natural Sciences, Comenius University, Mlynsk dolina G, SK-842 15 Bratislava, Slovakia, aubrecht@fns.uniba.sk2Geophysical Institute, Slovak Academy of Sciences, Dbravsk cesta 9, SK-845 28Bratislava, Slovakia3Department of Geochemistry, Faculty of Natural Sciences, Comenius University, Mlynsk dolina G, SK-842 15 Bratislava, Slovakia, lanczos@t-zones.sk4Slovak Spe leological Society, Hodova 11, SK-031 01 Liptovsk Mikul, Slovakia, lukasvlcek@gmail.com5Speleoclub of Comenius University, Department of Geology and Paleontology, Faculty of Natural Sciences, Comenius University, Mlynsk dolina G, SK-842 15 Bratislava, Slovakia, brano.smida@gmail.com Geoscientific research was performed in the two largest sandstone cave systems in the world Charles Brewer Cave System on the Chimant Massif and Ojos de Cristal Cave System in the Mt. Roraima, both in South American table mountains (tepuis). These cave systems consist of subhorizontal caves. The research revealed that erosion of non-cemented layers and lateritization of arkosic arenites (hydrolysis of feldspars and micas) played a substantial role in their genesis. Softer beds in which the caves were initially formed show a lack of cementation or cementation by kaolinite. The hard overlying and underlying beds, as well as the finger-flow pillars which penetrate the uncemented arenite beds, are cemented by opal and quartz cements. The finger-flow pillars indicate that the main diagenetic phase was represented by descending silica-bearing fluids. The pillars originated when the fluid flow reached a coarse-grained arenitic bed, where the continuous fluid front splited to narrow channels. This caused lithification of the arenitic material in the channels and the rest of arenites in these beds escaped from lithification (softer beds) and was easily erodable. This unusual way of arenite lithification in tepuis infers new views on their genesis and on the geomorphological evolution of the north of South America. Tepuis were formed from hard quartzites and sandstones of the Matau Formation, which are underlain by arkoses of the Uaimapu Formation. These are the uppermost formations of the Roraima Supergroup which is the Paleoproterozoic detritic cover of the Archean Guyana Shield. From the speleogenetic and geomorphological observations it is evident that the main lithification phase of the Matau Formation which caused their hardening to quartzites was represented by descending silica-bearing fluids which did not penetrate to the underlying arkoses which remained almost unlithified. The question is: what was the source of these fluids? A new theory concerning the origin of tepuis is presented in this paper. According to this theory, tepuis originated in places where there was an intensive descending fluid flow, most likely emanating from surface water reservoirs, such as rivers or lakes. This continuous flow carried SiO2from the lateritized surface beds. Thus, the underlying part of the Roraima Supergroup was impregnated with SiO2and strongly lithified. These indurated parts of the formation remained as tepuis, while the remainder of the formation was removed by erosion. The softness of the underlying, non-lithified sediment below the tepuis caused undercutting of their margins thus maintaining the walls vertical.1. IntroductionDiscoveries of large arenitic caves in famous Venezuelan table mountains (tepuis) were unexpected. Caves in silicate rocks are not so ubiquitous as caves in much more soluble material, such as limestones or gypsum. The previously most accepted model for the genesis of the sandstone caves was based on the arenization concept presented first by Martini (1979). The term arenization involves the dissolution of the quartz cement in arenitic rocks, with subsequent erosion and winnowing of the loose sand material. However, recent data from the research in the largest sandstone cave systems in the world Charles Brewer Cave System on the Chimant Massif and Ojos de Cristal Cave System in the Mt. Roraima (Aubrecht et al. 2008, 2011, 2012) showed that the dominant role of quartz and/or quartz cements dissolution is questionable. Geological and geomorphological research showed that the most feasible way of speleogenesis is winnowing of unlithified or poorly lithified arenites which remained as isolated pockets among hard-lithified quartzites and sandstones. Another frequently observed phemomenon, which contributes to the formation of the cave systems is weathering of aluminosilicate minerals, i.e. lateritization. While the latter is related to the recently ongoing processes, the first mentioned speleogenetic factor was related to the processes which created the tepuis themselves. Analysis of this process and the consequences resulting from it are the main aim of this paper.2. Geology of the studied areaThe main morphological feature of Guyana Highlands, encompassing southern Venezuela, Northern Brazil and Guyana are its tepuis. These are table mountains characterized by steep cliff walls and relatively flat mesetas, composed of Precambrian quartzites and sandstones covering the Guyana Shield. More than 100 table mountains can be found in the area. They provide important habitats for a great variety of endemic flora and fauna. From a geological viewpoint, the caves and surface areas Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings221

PAGE 224

analyzed herein are situated in the Venezuelan Guyana, in the northern part of South America (southeastern Venezuela, Gran Sabana). This area comprises Archaean rocks of the Guyana Shield which is the northern part of the Amazonian Craton. The Guyana Shield has a Proterozoic sedimentary cover named Roraima Supergroup which is formed mainly by clastics derived from the northern Trans-Amazonian Mountains. Sedimentological studies showed that the depositional environments ranged from alluvial fans to fluvial braided-river deposits together with lacustrine, aeolian, tidal, shallow-marine deposits and some shallow water turbidites (Reis and Ynez 2001; Santos et al. 2003). Sandy continental deposits are dominant here. The thickness of the group ranges from 200 m to approximately 3,000 m, and it consists of the following lithostratigraphic units arranged in stratigraphic order: Arai Formation, SuapiGroup, Uaimapu Formation and the Matau Formation (Reis and Ynez 2001). Tepuis developed mainly in the uppermost, Matau Formation formed by quartzites and sandstones, whereas the underlying Uaimapu Formation is mainly formed by arkoses. Their age was determined as 1,873 Ma (late Paleo-Proterozoic). This was achieved by U-Pb analyses of zircons from a green ash-fall tuff of the Uaimapu Formation (Santos et al., 2003). Since most of the previous authors accepted the theory that quartz dissolution requires a long time, the recent landscape, including commencement of the cave-forming process, is considered to be inherited from the Mesozoic period (e.g., Cretaceous Galn and Lagarde 1988; Briceo et al. 1991; Piccini and Mechia 2009).3. MethodsOur geological, geomorphological and speleological observations were focused on phenomena relevant to the solution of the speleogenetic problems. This mainly centered on the differential physical and mineralogical properties of the various kinds of arenites on the tepuis surfaces and caves in the Matau Formation, and also on morphological aspects of the various stages of speleogenesis and its final manifestations on the surface. Along with geomorphological observations, petrological and mineralogical analyses were performed on the rocks of the Matau Formation. This mainly encompassed petrography of thin-sections of the arenites under polarized light, as well as SEM observations. Mineralogical composition of the samples was determined optically and also by X-ray diffraction analysis (XRD).4. Results of geomorphological, petrographical and speleogenetic research.Erosion and rockfalls, which recently prevail in the Charles Brewer and Ojos de Cristal cave systems (for position of the tepuis see Fig. 1) have concealed their true speleogenetic processes. The trigger and structural factors acting during initial stages of the cave evolution are now mostly obliterated in the mature parts of the cavern systems. Therefore, many of the most important clues resulted from caves which are still in their initial stages of evolution. The sandstone surfaces of tepuis are very uneven and bizarre, which we interpreted as originated due to an inhomogeneous lithification of the Proterozoic arenites. This is especially apparent in areas where arenite beds form overhangs. The overlying and underlying beds are hard, well-lithified sandstones to quartzites, so that sampling was possible only with strong hammering. However, the beds inbetween are considerably soft, consiting of sands or soft sandstones, so that it was almost impossible to take solid samples for petrographic microscopic study, even after digging 30 cm deep by hand. This contrast in hardness was also verified by Schmidt hammer measurements (Aubrecht et al. 2011, 2012). These soft beds are penetrated by perpendicular pillar-shaped bodies (Fig. 2). These are narrower in the middle, but they have funnel-like widening at either end, with the lower funnel often less developed than the upper one. They are relatively hard rocks, ranging from sandstones to quartzites. In our interpretation, the origin of these pillars is purely diagenetic and their presence proves that the softness of the beds is due to a poor lithification and, therefore, it is primary rather than being secondary, related to weathering (see the discussion between Sauro et al. 2013 and Aubrecht et al. 2013). The pillars are considered to originate by a finger flow mechanism (cf. Aubrecht et al. 2008, 2011, 2012, 2013). The main factors influencing diagenetic variability were the differing hydraulic properties of the sediment in different layers which influenced its hydraulic conductivity. Figure 1. Location of the examined tepuis. Figure 2. Well preserved finger-flow pillars in the uncollapsed main corridors of Cueva Caon Verde revealing their origin from descending silica-bearing fluid flow (note the similarity to leaking thick syrup).Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings222

PAGE 225

The diagenetic fluids most likely penetrated vertically from the overlying strata as a descending diagenetic fluid flow. In finer-grained sediments, these diagenetic fluids filled the intergranular spaces evenly and resulted in the formation of diagenetically well-lithified beds, resistant to weathering. In coarse-grained arenites with higher hydraulic conductivity below the fine-grained beds, the evenly distributed descending diagenetic front divided into fingers, where the fluid flow accelerated and formed separate, finger-like flows. This process has been described in detail by various authors working with transport processes in unsaturated zones of sandy aquifers and also in soils (Liu et al. 1994 and Bauters et al., 2000). A similar process is seen in snow penetrated by descending water as it leaks from melted snow above (Marsh 1988 Fig. 2). Liu et al. (1994) considered that when these finger flows are generated in originally dry sandstones they are conserved as the most preferred method of infiltrating solutions. This is the way pillars originated in the unlithified sands. The downward diminishing funnel shape of the upper part of the pillar originated from flow acceleration, which continued until it decelerated when approaching the less permeable bottom. This retardation process is manifested in the reversely oriented funnel shape in the lower part of the pillar. Observations in the Charles Brewer and Ojos de Cristal cave systems show that pillars are present in most of the caves and in their galleries, which are still in younger stages of their evolution. These usually possess low ceilings and strictly maintain one distinct layer. However, relic finger-flow pillars were also observed in the marginal, uncollapsed parts of the larger galleries. When several superimposed winnowed horizons evolve together, a second collapse stage follows, leading to formation of much larger subterranean spaces. Galleries in the Charles Brewer Cave System are typically 40 metres wide, but they can also be much larger. The largest chamber found in the cave is Gran Galera Karen y Fanny. This is 40 metres high, more than 355 metres long and 70 metres wide, giving a volume of approximately 400,000 cubic metres. The final stages of cave evolution often lead to huge collapses, and these are evident also on the tepui surfaces. This process most likely led to the creation of the large abysses in the Sarisariama Plateau (Sima Mayor and Sima Menor). Mineralogical and petrological analyses showed that softer beds in which the caves were initially formed lack cementation, or only kaolinite cementation is present, while the hard overlying and underlying beds and the finger-flow pillars are cemented by opal and quartz. Some workers, e.g., Sauro et al. (2013) argumented against differential diagenesis by the metamorphic overprint of the entire Matau Formation. They suggested that the metamorphism is evidenced by presence of pyrophyllite and pressure dissolution of quartz. Pressure dissolution also takes place in higher degrees of diagenesis and it can be neglected as a metamorphic indicator. Concerning pyrophyllite, although it is mentioned as a metamorphic indicator in the literature (anchizone to epizone) it does not explain presence of vast quantity of kaolinite in the soft arenitic beds that did not react with quartz to form pyrophylllite. Instead, pyrophyllite without kaolinite is present in hard lithified arenites (Aubrecht et al., 2013 Fig. 3). Literature studies revealed that there is some evidence that pyrophyllite may also originate by hydrothermal alteration at lower temperatures (e.g., Ehlman and Sand 1959; Bozkaya et al. 2007;Bauluz and Subas 2010). The most important information is that some part of the aluminosilicate mineral phases existing at atmospheric pressure and 25 C shows that H4SiO4concentration is the critical factor in kaolinite/pyrophyllite transition (Aubrecht et al. 2013 Fig.4). Here, increased H4SiO4concentration can theoretically cause the transformation of kaolinite to pyrophyllite without increased temperature or pressure (Aubrecht et al. 2013). Lateritization, which also contributes to cave-forming processes, is important, but as we are now focused to the very early diagenetic processes that affected the Matau Formation, it is beyond the scope of this paper.5. Interpretation of the descending silicabearing fluid flow: a new view on the origin of tepuisSummarizing the results of the speleogenetic research, several important points were discovered concerning the origin of the tepuis. Although the research elucidated many aspects of the speleogenetic process, it also created new questions and problems. The most conspicuous finding was that the Matau Formation is formed not only of quartzites but that its arenites show various degrees of lithification. Our research provided evidence of variability in vertical profiles. But what about the lateral variability? Are tepuis with a dominant presence of hard-lithified quartzites typical examples of the Matau Formation? What about the larger, missing portion of the formation which was removed by erosion? Why are tepuis usually isolated islands rising up from the flat Gran Sabana? And also, why there are no ruins of tepuis formed by accumulations of quartzite boulders dispersed throughout the Gran Sabana? Answers to these questions are currently purely theoretical as the missing, eroded portion of the Matau Formation can no longer be examined. However, knowledge gathered from our research of this formations remnants can be grouped under one common image which entails a new theory of the origin of tepuis. Finger-flow pillars in the arenites forming these tepuis indicate that the descending flow of silica-bearing diagenetic fluids provided induration to arenites even to very hard quartzites. This flow penetrated deeply enough to lithify hundreds of metres of arenites in a vertical profile, and the indurated rocks then protected less lithified portions of the formation below. Most of the tepuis are limited by vertical cliffs, and undercutting of these cliffs often occurs because the lower parts of the massifs are less lithified (see also Young et al. 2009 p. 58). Undercutting and the subsequent rockfall would be responsible for creation of the rock talus around tepuis (Montaas al pie del escarpado see Briceo and Schubert, 1992 Fig. 4.3). The talus then passes to flat country surrounding the tepuis, without retaining any remnants of quartzite boulder accumulations. However, closer inspection of the talus around Roraima indicates that Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings223

PAGE 226

it is formed by less lithified, soft arenites of the Roraima Supergroup, which most likely underlie the sandstones and quartzites, rather than by fallen quartzite blocks (Fig. 3). Erosion of these soft arenites causes undercutting of the Roraima cliffs and keeps them steep. Stage2 Hard, isolated beds were broken, and the flowing water which penetrated the poorly lithified beds started initial erosion. Lateritization then began in aluminosilicate-rich beds with the subsequent emptying of spaces by winnowing of sand and other products of lateritization. The empty spaces were then supported solely by finger-flow pillars. Stage 3 Empty spaces collapsed further, thus creating larger caves, and superior propagation of these collapses created large collapse depressions on the surface. Figure 3. A view on Mt. Roraima (and Mt. Kukenan left). The mountain consists of quartzites, which form steep cliffs (C) and surrounding talus (T). Figure 4. Schematic overview of the speleogenesis in the arenites of the Roraima Supergroup. Grey well-lithified arenites, pale poorly lithified arenites (caves formed by lateritization are omitted). AB The gradual diagenesis to arenites caused by descending silica-bearing fluids (Stage 1). C Two poorly-lithified horizons superimposed on each other. DE Flowing water penetrated the vertical cleft (on the right side) causing gradual winnowing of the poorly lithified sediment (Stage 2). F Two horizons remained empty (two superimposed initial caves), with lithified pillars offering the only support against collapse. GH Collapse of both floors, forming a large cave (Stage 3).All these observations suggest that the patchy distribution of tepuis in Gran Sabana was formed long ago by vertical lithification of the Matau Formation. This lithification required a voluminous source of soluble SiO2and sufficient fluids. Exactly as evident today from the recent lateritization, the best source of SiO2were the clay and rocks with micas and feldspars above the actually preserved Matau Formation. These rocks were easily affected by lateritization, which most likely occurred after the Late Carboniferous, when the northern-most part of South America reached the tropical zone (see Scotese 2001). The best source of fluids would undoubtedly have been water reservoirs on the surface, and thus the recent distribution of tepuis may have copied the distribution of ancient lakes and rivers. Alternative explanations for the erosion of the missing portions of the Roraima Supergroup include the inference of Galn et al. (2004 Fig. 1) that this erosion mainly affected tectonically disrupted parts of the Roraima Supergroup. However, this contradicts the lack of boulder accumulations on Gran Sabana. Moreover, the disrupted parts of the Roraima Supergroup had to be more widespread than the undissected ones, and this is considered most unlikely in this firm, rigid block of the Guyana Shield. This theory is currently based on a limited set of data and further research is necessary. Although new data may support or refute our theory, it is very satisfying to provoke future research in this area.6. Conclusions1. The following speleogenetic model can be inferred from the obtained geomorphological and geological results (Fig.4): Stage 1 The descending, SiO2-bearing diagenetic solutions caused complete lithification of some beds, whereas other beds with more coarse-grained arenites were only penetrated by narrow channels through which fluids flowed to completely fill some of the lower beds. This resulted in the contrasting diagenesis, where most beds turned to sandstone and quartzite while parts of other beds remained intact. 2. A new theory concerning the origin of tepuis can be proposed, based on the fact, that diagenesis by descending silica-bearing fluids appears to be the main phase that indurated the Matau Formation quartzites (Fig. 5). According to this theory, tepuis originated in places where there was an intensive descending fluid flow, most likely emanating from surface water reservoirs, such as rivers or lakes. This continuous flow carried SiO2from the lateritized surface beds. Thus, the underlying part of the Roraima Supergroup was impregnated with SiO2and strongly lithified. These indurated parts of the formation remained as tepuis, while the remainder of the formation was removed by erosion. The softness of the underlying, non-lithified sediment below the tepuis caused undercutting of their margins thus maintaing steep walls. Speleogenesis in the tepuis was a process which most likely began at a later stage, with initial incision of the valleys followed by erosion. Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings224

PAGE 227

AcknowledgementsThe authors are deeply indebted to Francesco Sauro and Jo De Waele (Bologna University, Italy) for their review and useful corrections and comments that helped to improve the text of this paper. The research was financed by grants APVV 1-0251-07 and VEGA 1/0246/08.ReferencesAubrecht R, Lnczos T, mda B, Brewer-Caras Ch, Mayoral F, Schlgl J, Audy M, Vlek L, Kovik Gregor M, 2008. Venezuelan sandstone caves: a new view on their genesis, hydrogeology and speleothems. Geologia Croatica, 61, 345. Aubrecht R, Lnczos T, Gregor M, Schlgl J, mda B, Li k P, Brewer-Caras Ch, Vlek L, 2011. Sandstone caves on Venezuelan tepuis: return to pseudokarst? Geomorphology, 132, 351. Aubrecht R, Barrio-Amoros C, Breure A, Brewer-Caras Ch, Derka T, Fuentes-Ramos OA, Gregor M, Kodada J, Kovik Lnczos T, Lee NM, Li k P, Schlgl J, mda B, Vlek L, 2012. Venezuelan tepuis their caves and biota. Acta Geologica Slovaca Monograph, ISBN 978-80-223-3349-8, 1. Aubrecht R, Lnczos T, Gregor M, Schlgl J, mda B, Li k P, Brewer-Caras Ch, Vl ek L, 2013. Reply to the Comment on Sandstone caves on Venezuelan tepuis: Return to pseudokarst? Geomorphology (DOI: 10.1016/j.geomorph.2012.11.017). Bauluz B, Subas I, 2010. Coexistence of pyrophyllite, I-S, R1 and NH4+-rich illite in Silurian black shales (Sierra de Albarracn, NE Spain): metamorphic vs. hydrothermal origin. Clay Minerals, 45, 3, 383. Bauters TWJ, Dicarlo DA, Steenhuis TS, Parlange J-Y, 2000. Soil water content dependent wetting front characteristics in sand. Journal of Hydrology, 231, 244. Bozkaya O, Yalcin H, Basibuyuk Z, Bozkaya G, 2007. Metamorphic-hosted pyrophyllite and dickite occurrences from the hydrous Al-silicate deposits of the Malatya-Puturge region, central eastern Anatolia, Turkey. Clays and Clay Minerals, 55, 4, 423. Briceo HO, Schubert C, 1992. Geomorfologa. In: Huber O. (Ed.): Chimant. Escudo de Guayana, Venezuela. Un Ensayo Ecolgico Tepuyano. Oscar Todtmann Editores, Caracas, 61. Briceo HO, Schubert C, Paolini J, 1991. Table-mountain geology and surficial geochemistry: Chimant massif, Venezuelan Guayana Shield. Journal of South American Earth Sciences, 3, 179. Ehlmann AJ, Sand LB, 1959. Occurrence of shales partially altered to pyrophyllite. Clays and Clay Minerals,6, 368. Faure G, 1991. Principles and applications of inorganic geochemistry: a comprehensive textbook for geology students. Macmillan, New York, 626. Galn C, Lagarde J, 1988. Morphologie et evolution des cavernes et formes superficielles dans les quartzites du Roraima. Karstologia, 11, 49. Galn C, Herrera FF, Carreo R, 2004. Geomorfologa e hidrologa del Sistema Roraima Sur, Venezuela, la mayor cavidad del mundo en cuarcitas: 10,8 km. Bol. SVE, 38, 2. Liu Y, Steenhuis TS, Parlange J-Y, 1994. Formation and persistence of fingered flow fields in coarse grained soils under different characteristics in sands. Journal of Hydrology, 159, 187. Martini JEJ, 1979. Karst in Black Reef Quartzite near Kaapsehoop, Eastern Transvaal. Annals of the Geological Survey of South Africa, 13, 115. Piccini L, Mecchia M, 2009. Solution weathering rate and origin of karst landforms and caves in the quartzite of Auyan-tepui (Gran Sabana, Venezuela). Geomorphology, 106, 15. Figure 5. The newly proposed model of the origin of the tepuis. A The Roraima Supergroup was originally capped by sediments rich in micas, feldspars or clay minerals which were prone to lateritization. This lateritization may have begun in the Late Carboniferous when the northern part of present South America reached tropical areas (Scotese, 2001). B The lateritization occurred mostly in the areas with excess fluids, such as rivers and lakes. The descending fluids brought silica from the lateritization zones downwards, causing additional cementation of the Matau Formation. This cementation was patchy, and concentrated only in the zones with sufficient water. CD In the later geomorphological evolution stages, the uncemented portions of the Roraima Supergroup were subjected to erosion and the cemented, quartzitic parts were preserved, together with the softer, uncemented parts protected below them. The steep cliffs of the tepuis are maintained by erosion of the softer, uncemented arenites below with subsequent undercutting of the quartzite layers.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings225

PAGE 228

Reis NJ, Ynez G, 2001. O Supergrupo Roraima ao longo da faixa fronteirica entre Brazil-Venezuela (Santa Elena del UairenRoraima Mountain). In Reis N.J. & Monteiro M.A.S. (Eds.): Contribuicao a geologia da Amazonia, Vol. 2: Manaus, Brazil, Soc. Brasil. Geol. 113. Santos JOS, Potter PE, Reis NJ, Hartmann LA, Fletcher IR, McNaughton NJ, 2003. Age, source, and regional stratigraphy of the Roraima Supergroup and Roraima-like outliers in northern South America based on U-Pb geochronology. Geological Society of America Bulletin, 115, 331. Sauro F, Piccini L, Mecchia M, De Waele J, 2012. Comment on Sandstone caves on Venezuelan tepuis: Return to pseudokarst? by R. Aubrecht, T. Lnczos, M. Gregor, J. Schlgl, B. Smda, P. Lisck, Ch. Brewer-Caras, L. Vlcek. Geomorphology 132, 351. Geomorphology. Scotese CR, 2001. Atlas of Earth History. PALEOMAP Project, Arlington, Texas, 1. Young RW, Wray RAL, Young ARM, 2009. Sandstone landforms. Cambridge University Press, 1.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings226

PAGE 229

INFRARED THERMOGRAPHIC SURVEY OF PSEUDOKARST SITES IN THE FYSCH BELT OF OUTER WEST CARPATHIANS (CZECH REPUBLIC)Ivo Baro1, 2, David Bekovsk3, Lumr M a3 1Speleological Society of Vienna and Lower Austria, Obere Donaustrae 97/1/61, A-1020 Wien, Austria2Geological Survey of Austria, Neulinggasse 38, 1030 Vienna, Austria, ivo.baron@geologie.ac.at3Faculty of Civil Engineering, Brno University of Technology, Kounicova 966/67a, 60200 Brno, Czech Republic The paper presents a test study of application of infrared thermography (IRT) for mapping thermal characteristics of pseudokarst crack caves related to deep-seated gravitational slope failures in the Flysch Belt of the Outer West Carpathians, and for locating new systems. The method was tested at four pseudokarst sites: K ov, Smrdut, Pustevny and Zryje in E Czech Republic. The IRT images excelently showed spatial pattern and thermal characteristics and gradients of the crack systems opened to the ground surface promising new cave discoveries. The method is, however, constrained by very specific meteorological conditions, such as cold weather with ambient air temperatures lower than the local annual mean rock temperature at depth, good visibility, lack of direct insolation, absence of thick snow cover, adequate sensor resolution and accuracy, and reasonable distance and sensing angle from the rock surface.1. IntroductionThermal gradients between deep rock and colder ground surface often help cavers identify unexplored caves by searching for melted snow cover in the winter season (Rinker 1975; Leinsk 1999; Baro 2002). The Infrared Thermography method (IRT) has been proposed as a useful tool for detecting karst cave openings (Rinker 1975; Wynne et al. 2008; Baro et al. 2012) and hydrological features in karst watersheds (Campbell et al. 1996). Here we present a test study of application of IRT to pseudokarst sites in E Czech Republic, which follows our previous work focused on detection of open cracks within deep-seated rockslides in Outer West Carpathians and Northern Calcareous Eastern Alps (Baro et al. 2012). The main aim of this study is testing the reliability of the method for pseudokarst crack caves within gravitational rock-slope failures.2. MethodsThe IRT detection method of pseudokarst caves relies on air circulation and ventilation though open joints and cracks when, in winter, the warmer subsurface communicates with the (colder) ground surface; crack caves could be identified as relatively warm areas using IRT then (Baro et al. 2012). For this study, we used a Flir B360 infrared thermal camera, which is a high-sensitivity camera with 320 240 IR resolution and thermal sensitivity (N.E.T.D) of 0.05 C at 30C. The sensing temperature range is -20 to 120 C, and the spectral range varies from 7.5 to 13 m (far-IR). IRT mapping in this study was conducted in beginning of cold season on 1stDecember, 2012, when it was overcast to cloudy weather with external temperatures ranging between 0 and -5 C. There was no or minimum snow cover at the time of the survey.3. Description of the K ov, Smrdut, Pustevny and Zryje SitesThe case-study sites comprise four pseudokarst localities related to deep-seated rock slope failures in the Flysch Belt of Outer West Carpathians in E Czech Republic. The K ov pseudokarst site has developed in a headscarp area of a Holocene rotational deep-seated rockslide at SE slopes of the K ov Hill (670 m a.s.l.) about 3.5 km west from Vsetn. The slope failure is about 700 m long, 300 m wide with estimated 8 million m3of affected bedrock and colluvium. The main scarp exhumed faces of several m thick sandstone beds of the Lukov Member of the Sol Formation. The sandstone cliff reaches height of up to 10 m, with a scree accumulation below; the whole scarp is about 30 m high. About 20 long and 12 m deep crack cave Zbojnick occurs at SW flank of the scarp. The Smrdut pseudokarst site comprises a deep-seated complex flow-like landslide combined with a deep-seated rockslide and rock sagging in the upper part. It is situated at SW slopes of the Smrdut Hill within competent sandstones and conglomerates of the Rusava Member and incompetent Figure 1. A) Location within the Czech Republic, and B) detailed location of the study sites within the topography of the centraleastern part of the Czech portion of the Outer West Carpathians. The black rectangle indicates the extent of B) (Source of SRTM data: USGS/NASA).Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings227

PAGE 230

claystones and shales of the Raztoka Member of the Ra a Unit. Our results of the radiocarbon dating of lacustrine deposits at the base of a palaeo-dam due to the landslide revealed a minimum age of the slope failure about 2055 14C years BP. Two smaller pseudokarst talus caves developed between conglomerate blocks in the flow part, and six other crack-caves developed within the rockslide and sagging part. The longest caves here are the Na Smrdut and Poscho ov Caves reaching maximum length of 56m (Wagner et al. 2009). Other melting spots developed on the scree accumulation (warm air originated within small caverns between stones and debris) or behind toppled boulders (Figure 1C and D). The spots related to toppled boulders indicate the future rock-fall hazard. The pseudokarst doline below the scarp was thermally neutral. The warm areas are, however, relatively very small, both in dimensions and thermal gradient. Therefore we do not suspect an existence of larger cave systems except of the Zbojnick Cave. Figure 2. Results of the IRT survey at the K ov Site: A) IRT image of the entrance of the Zbojnick Cave, B) BW photography of the same area, C) IRT image of a toppled rock column indicating open cracks behind it, D) BW photography of the same area. Figure 3. Results of the IRT survey at the Smrdut Site: A) IRT image of the entrance of the Vodn Cave, B) BW photography of the same area, C) IRT image of the entrance of the Poscho ov Cave, D) BW photography of the same area, E) IRT image of the entrance of an unknown undergroung space, F) BW photography of the same area, G) IRT image of the entrance of the Na Smrdut Cave, H) BW photography of the same area.The Pustevny Site is well known for the longest pseudokarst cave of the Czech Part of the Outer West Carpathians. The Cyrilka Cave has been recently prolonged to 535 m (J. Lenrt, pers. comm.) and it has developed in the crown of a large deep-seated translational consequent rockslide at the SE slopes of the Radho massif near Pustevny. The rockslide is about 2,400 m long, 1,500 m wide and about 20 m thick (Baro and Kaperkov 2007). The sliding surfaces are located within claystone intecalations between several m thick sandstone beds dipping about 9 to SSE. The last studied pseudokarst site at Zryje has, similarly to the previous site, developed in upper part of a large deepseated translational consequent rockslide at the S slopes of the Radho Several caves are known from this area (Wagner et al. 1990). The longest one is the Szalajka Cave discovered in 2004 (J. Szalai, pers. comm.).4. Resuls4.1. The K ov Site Several warmer areas were observed at this site; all of them were located along the cliff of the main-scarp. A blowhole with strong air current was related to the entrance the Zbojnick Cave (Figure 2A and B) as well as several warmer spots. 4.2. The Smrdut Site The IRT survey found several warmer areas related to (i) the highly-porous flow part containing blocks up to 10 m in diameter, e.g., at the Vodn Cave (Fig. 3A and B), loosened rock of an incipient deep-seated rockslide part with the Poscho ov Cave (Fig. 3C and D), several other unknown undergound crack caves (Fig. 3 E and F), and to the sagged part with the longest cave here, the Na Smrdut Cave (Fig. 3G and H). The IRT images excellently showed the spatial pattern and thermal characteristics and gradients of crack systems opened to the ground surface. The results promise new cave discoveries; however due to dimensions of the slope failure, they would not exceed several tens of meters. Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings228

PAGE 231

4.3. The Pustevny Site At the Pustevny Site, we surveyed only the area of the Cyrilka Cave. The site was in several places exposed to the sunshine, therefore the results were acceptable only from few a places. Despite that, thermal characteristics of the entrance of the Cyrilka cave (Fig. 4A and B) and the spatial pattern of its traces at the ground surface (Fig. 4C and D) were achieved by IRT.5. Discussion and ConclusionsThrough our research, we proved great advantages of the IRT ground-based survey for mapping thermal characteristics of pseudokarst crack caves related to deepseated gravitational slope failures in the Flysch Belt of the Outer West Carpathians, and for locating new cave systems. The method is, however, restricted to very specific meteorological conditions, similar to the IRT survey for mapping open cracks within unstable rock slopes and rock cliffs for engineering-geological purposes (Baro et al. 2012); the applicability of the IRT method is restricted to cold weather, when ambient air temperatures are lower than the local mean rock temperature at depth. Good visibility, lack of direct insolation, absence of thick snow cover, adequate sensor resolution and accuracy, reasonable distance from the rock surface, and sparse vegetation between the sensor and rock surface are main considerations of the method (Baro et al. 2012).AcknowledgmentsThe survey was supported by the project OP VAVpI CZ.1.05/2.1.00/03.0097 AdMaS at the Brno University of Technology. The authors thank to M. Filippi and other anonymous reviewer for careful reading of the manuscript and their comments, which helped improving its quality. Figure 4. Results of the IRT survey at the Pustevny Site: A) IRT image of the entrance of the Cyrilka Cave, B) BW photography of the same area, C) IRT image of a trace of the Cyrilka Cave at the ground surface, B) BW photography of the same area. Figure 5. Results of the IRT survey at the Zryje Site: A) IRT image of the entrance of the Szalajka Cave, B) BW photography of the same area,C) IRT image of a superficial trace of an unknown crack-cave system, D) BW photography of the same area. Figure 6. Results of the IRT survey at the Zryje Site: A), C), E) and G) IRT images of superficial traces of unknown crack-cave systems, B), D), F) and H) BW photographies of the respective areas.4.4. The Zryje Site We achieved the best results of the structural pattern of open cracks and the existing and potential cave systems at the Zryje Site. We identified all the known caves and their traces at the ground surface, (e.g., the Szalajka Cave, Fig. 5A and B), and also several up-to date unknown crack systems (Fig. 5C and D and Fig. 6). The site promises new speleological discoveries in near future. Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings229

PAGE 232

ReferencesBaro I, 2002. Location of underground spaces in selected slope deformations by observations of melted snow and ventaroles (in Czech). Speleofrum 21, 64. Baro I, Be kovsk D, M a L, 2012. Application of infrared thermography for mapping open fractures in deep-seated rockslides and unstable cliffs. Landslides, Springer Verlag. DOI 10.1007/s10346-012-0367-z. Baro I, Kaperkov D, 2007. Numerical analysis of Silesian Nappe gravitational disintegration by means of FDM (Case study: Radho -Pustevny). Zprvy o geologickch vzkumech v roce 2006, 49, Czech Geological Survey, Prague (in Czech). Campbell CW, Abd El Latif M, Foster JW, 1996. Application of Thermography to Karst Hydrology. Journal of Cave and Karst Studies 58(3), 163. Leinsk G, 1999. Comments to caving in winter season with focus on observing melted spots and ventaroles. Spravodaj SSS 1999 (4), 32. Liptovsk Mikul (in Slovak). Rinker JN, 1975. Airborne infrared thermal detection of caves and crevasses. Photogrammetric engineering and remote sensing 41, 1391. Wagner J, Demek J, Strnk Z, 1990. Caves of the MoravianSilesian Beskids and Surrounding. Knihovna SS 17: 1. Praha. Wagner J, Baro I, Bezdka P, 2009. Pseudokarst caves of the flysch belt of the Outer West Carpathians. In Hromas J. et al. (Eds.), Protected Areas of the Czech Republic, Volume XIV: Caves, Praha Brno, 536 (in Czech). Wynne JJ, Titus TN, Diaz GCh, 2008. On developing thermal cave detection techniques for earth, the moon and mars. Earth and Planetary Science Letters 272, 240.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings230

PAGE 233

KAHUENAHA NUI (HAWAII): A CAVE DEVELOPED IN FOUR DIFFERENT LAVA FLOWSIngo Bauer1, Stephan Kempe1, Peter Bosted2 1Institute of Applied Geosciences, University of Technology Darmstadt, Schnittspahnstr. 9, D-64287 Darmstadt, Germany, Kempe@geo.tu-darmstadt.de2Cave Conservancy of Hawaii: P.O. Box 7032 Ocean View, Hawaii 96737-7032 Exploration and survey of the Kahuenaha Nui Cave in March 2011 yielded astonishing insights into the processes that act to enlarge the tunnels of underground lava conduits (pyroducts). The cave is situated in lavas of the SW Rift of the Mauna Loa, Hawaii, within the area of the former Kahuku Ranch, south of the Belt Road at an altitude of 564 m a.s.l. On the geological map of Hawaii, these lavas belong to the stratigraphic group Qk2, dated to between 1500 and 3000 a BP. These flows are bordered by the 1868 pahoehoe flow in the E and the 1887 aaflow in the W and occupy ca. 70 km2. The cave survey yielded a total length of 1,850 m, a total vertical extent of 55 m, and an average slope of 5.7 (Bauer 2011). The cave features a main trunk that is up to 18 m wide and 11 m high. Its floor is in parts formed by terminal aa. Above this trunk passage, we explored numerous small to very small interconnected pahoehoe ducts. At the entrance puka (breakdown hole) and at a large open puka we were able to study the cave formation process. The trunk passage formed by eroding an underlying aarubble layer. In places even the underlying aacore layer has been cut into. Above, a stack of seven superimposed pahoehoe (phh) flows with small ducts occurs, forming the primary roof of the cave. The lava flowing in this stack of sheets managed to combine into one flow, eroding the main trunk underneath. After cave formation first an aaflow and then a thin pahoehoe flow transgressed the area. The caves roof partly collapsed, not only exposing the transgressed aabut also forming the two entrances. This cave forming mechanism is fundamentally different from the inflation and the crusting-over of channels mechanisms identified as pyroduct formation modes so far.1. IntroductionEven though many lava tubes are mapped on the island of Hawaii, especially on Mauna Loa, only little is known about the detailed processes taking place during the formation of lava caves. Volcanological literature does not deal in detail with near-surface transport processes of lava and their importance for building shield volcanoes. Description of lava tube forming processes are often limited to crustingover of channels. Despite this many lava cave forming processes are known today due to intense research in this field of study although we are far away from understanding all the features we can observe. In Kahuenaha Nui, a cave situated on the south-west flank of Mauna Loa, we were able to discover a new way to form large pyroducts. A number of small pyroducts in the stack of pahoehoe sheets of the primary roof drained into the main passage eroding an underlying aa-flow. Astonishingly intense erosion cut down not only through the underlying aa rubble but also the aa core beneath was deeply degraded. This results in a cave with its main volume build in older lavas.2. Geologic settingThe study area is located near Hawaiis South Point, on the southern flank of Mauna Loa, south of the Mamalahoa highway between Ocean View and Naalehu. The investigation area between the highway and the coast is 10 km long and covers roughly 70 km2. Our study concentrated on the Puu o Keokeo-Flow. The average surface slope is 3.2, decreasing towards the coast. Mauna Loa is the largest active volcano on the island of Hawaii. Eruptions can be discriminated in summit, riftzone, and flank eruptions. Two rift zones, the Northeast and Southwest rift zone cut through the volcanic edifice. Eruptions from the SW-Riftzone produced the 1868 and 1887 flows, bordering the Puu o Keokeo-Flow to the east and west. Kahuenaha Nui is situated in a series of old lavaflows also erupted from the SW-Riftzone of Mauna Loa. The lavas are not dated but supposed to be of late Holocene age, between 1500 and 3000 a BP. According to the geological map, these lavas belong to the stratigraphic group Qk2. The surface of the northern area is covered by a hummocky pahoehoe flow, the Puu o Keokeo-Flow. It transgressed an older aa-flow, appearing in patches on the surface. This flow is radiocarbon-dated to 1730 a BP. Mauka (Hawaiian for uphill) Mamalahoa highway aerial photographs allow tracking the Puu o Keokeo-Flow for another 3 km. Further uphill vegetation prevents identification. Lavas of the 1926 and 1950 flows cover the surface mauka. Considering the section shape of the SWRiftzone it is unlikely that the Puu o Keokeo lavas originate from a summit eruption. Thus it is probable that the source of the lava is a magma chamber below the SWrift zone. The surface above Kahuenaha Nui is covered with phhlava. In defined narrow areas transition from pahoehoe to aa can be observed, partially covered by pahoehoe from later events. The area is characterized by the appearance of numerous tumuli. They arise while already cooled surface lava bulges due to varying eruption rates. In the upper Puu o Keokeo-Flow no evidence for overflowing lava was found. Therefore one could assume that eruption rates fluctuated with low amplitude and high frequency. Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings231

PAGE 234

3. PetrologyThin sections from different Kahuenaha Nui samples were investigated and XRD-analysis was conducted on selected samples in order to characterize different lava flows (Table1). As expected the composition is tholeiite-basaltic. The pyroxene family is represented by diopside. Sample CS 5 stems from the lower aa-rubble of the basal aa-layer. The related aa-core is partially covered with lining and cannot be traced continuously. Albeit sample CS 5 can be correlated to CS 14 which belongs to the lower aa core at st. 13. Hematite can be verified in samples CS 5, CS 14, CS 17, and CS 18. It originates from oxidation of magnetite and indicates as reddish boundary layer temporally separated lava flows. During transgression of aa-lava oxygen reaches the boundary layer through the aa-rubble. Sample CS 18 belongs to a channel overflow during a late stage of flow activity. In the process lava flowed through a channel in the pyroduct floor, no longer occupying the whole width of the passage. The above placed gas space was connected with the surface through contraction cracks in the cave roof and probably by hot pukas, allowing rapid oxidation of the lava surface resulting in a reddish coloured surface. blocked by boulders arising from cold breakdown of the cave roof. The SE-cave wall exposes an aa-core pinching out to the S, overlain by a reddish hematized aa-rubble. The cave roof is situated in phh-sheets resulting from a younger eruption event. Even the NNW-wall exposes an aa-core, petering out northwards, leading to a thick layer of aa rubble overlain by phh. These observations give evidence that DH developed at the border of two aa-flows. The absence of a continuous aa-core allowed upward enlargement while the underlying original cave passage became blocked. The original cave floor is completely covered with boulders and is not exposed. Puka 2 is a wide open cold puka and gives comprehensive insight in the development of the cave. Figure 1.Stratigraphic column of Kahuenaha Nui at Puka 2.Gibbsite, thenardite, and gypsum are secondary minerals, resulting from alteration of lava due to chemical weathering. They occur in large amounts, covering the cave walls and breakdown.4. Cave descriptionThe area has numerous pukas and caves of which only a few have been investigated during our field trip. Two entrance pukas, both formed by cold breakdown give access to the cave. Entrance puka 2, with a width of 9 m and steep overhanging walls, is the bigger one and was found later, after the discovery of Entrance puka 1. Puka 2 exposes the layers responsible for cave formation. Entrance puka 1 led to the discovery of the cave while one of the authors, sitting near the wall suddenly felt a cold breath from behind, arising from an airflow sweeping through a small hole in the wall of the puka. A short dig led to Discovery Hall (DH; Fig. 6a, P1), representing the actually known far N-end of Kahuenana Nui. Makai opened the main passage. Here the original cave roof, represented by a stack of phh-sheets, can be seen. The mauka end of DH isTable 1.Results from XRD-analysis. Minerals without distinct evidence in italic letters. Sample Nr. LocationMineral composition CS 5Kahuenaha Nui, southern wall, St. K6, lower aarubble (layer 0 ar) anorthite diopside albite cristobalite hematite thenardite hornblende marcasite CS 14Kahuenaha Nui, st. 13, lower aacore (layer 1 ac) anorthite diopside high-albite hematite hornblende marcasite CS 16Kahuenaha Nui, entrance flow augite anorthite high-albite gypsum CS 17Kahuenaha Nui, southern wall, upper aa-rubble of lower aa-layer (layer 2 ar) anorthite albite diopsid cristobalite hematite magnesite thenardite gibbsite CS 18Kahuenaha Nui, st. 28b, porous overflow anorthite diopside albite hematite An alternating sequence of phhand aa-layers can be seen in the entrance puka 2 (Figs. 1, 6a, P2). The aa-rubble layers 12ar and 10ar are undercut while the more resistant aa-core layers 11ac and the phh-surface layers 13p and 14p form minor overhangs. The primary cave roof is located within the stack of phh-layers 3p to 9p but cannot be determined exactly. The development of Kahuenaha Nui is essentially determined by erosional downcutting of the cave forming phh-lava in underlying aa-layers 1ac and 2ar. Due to minor resistance of layer 1ar against downward erosion, one can expect fast downcutting through the slightly welded aa-rubble. Below layer 2ar the flowing phh even cut through the aa-core (layer 1ac). A cross section at st. 12 gives evidence that the phh-lava cut down through at least 3.5 m of aa. Further mauka lining covers layer 1ac while it Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings232

PAGE 235

downwells in breakdown from the ceiling. Even though the aa-layer below the cave-forming pahoehoe is not continuously exposed, samples from Discovery Hall and puka 2 show similar mineralogical composition (samples CS5 and CS 14), suggesting that both layers belong to the same flow. At least two transgressive events can be determined subsequent to the cave forming process. The primary cave roof is overlain by an aa-flow (layers 10ar, 11ac,12ar). Not less than two more phh-flows (13p, 14p) form the present surface. In the center of the cave, the main passage got so wide and deeply entrenched, that a secondary ceiling formed (Fig. 6b, P5). No evidence of hot pukas can be found. At the Ball Room, several small conduits unite to form a larger cavity (Fig. 6b, P6). The r-passage incorporates a number of small initially formed conduits. (Fig. 6c, P7). Several small tubes in different phh-sheets can be observed. No lavafall established at st. ra1 or st. rb2, giving evidence that the overlying conduits have already been drained earlier. However at st. r2a1 a lavaflow can be determined, draining a high-lying small pyroduct. Makai of the Misery Crawl the b-passage cuts an overlying passage. Partial lining can be found, indicating that the phh-stack collapsed, connecting both passages while at least the lower passage was still active. However no lavafall from the upper passage developed. The upper passage points towards the mauka end of Peters Challenge (Fig. 6c). This passage was cut from the active flow at st. r2a6 (Fig. 6c, P7). Figure 2. Outcrop of aa-core with overlying aa-rubble at st. 12. The phh-flow cut through both layers. View southwest. Photo by Stephan Kempe. Figure 3. Mauka view from Confluence Hall. Arrow points to one initial conduit in the phh stack that drained into the main passage. Photo by Peter and Ann Bosted. Figure 4. Collapse connects two simultaneously active and overlying passages, draining the upper passage. Photo by Peter and Ann Bosted.Not far behind st. 20 the main passage splits. While the right passage has numerous 0.2 m deep cataracts, a 1 m deep lavafall cut down into the left passage. Where the passages unite again several initial phh-conduits in the phh stack meet and their lava cut into the underlying aa (Figs. 3, 6a, P3). Makai (Hawaiian for downhill) of st. 34b7 a lavafall drops 2.5 m down into an underlying passage (Fig. 6b, P4). The separation between different conduits in the initial stack of phh collapsed and the cave-in was carried out by the lava in the remaining lower conduits. Occasionally the underlying passage was blocked by lavaballs which gave rise to the spill-over of lava in the overlying passage. Reddish oxidized lava can be observed around the collapse and can be followed makai. However the reddish lava cannot be traced mauka giving evidence for a post collapse event.5. Survey dataThe main trunks length is 458 m with a total extent of 1,850 m. The main passage is mostly covered with breakdown blocks. It reaches a maximum width of 18 m and a maximum height of 11 m (Table 2). Although the average slope is 5.4 just a few lavafalls contribute to this gradient. In comparison, lavafalls represent 25 % of the vertical extent of the Huehue-Flow (Oberwinder 1996). Total length Total depth Horizontal length Horizontal extension Highest survey point Lowest survey point Longest way to origin Area coverage Average slope Average surface slope Sinousity Maximum width Maximum height Striking direction Main trunk length 1,846 m -55 m 1,812 m 560 m 564 m a.s.l. 509 m a.s.l. 660 m 1.54 km25.7 4.0 1.18 m 18.9 m 11 m 224 (NESW) 458 m Table 2. Kahuenaha Nui survey data.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings233

PAGE 236

7. ConclusionsKahuenaha Nui is a new cave type that so far was not documented in detail. The cave involved four lava units from different eruptions. It eroded into an underlying aa, visible nowhere else in the vicinity. The lava that formed the cave was carried initially in several superimposed small conduits in a 3 to 4 m thick stack of surface pahoehoe sheets. After the cave formation, the phh stack was transgressed by a 3.5 m thick aa flow and a 2 m thick pahoehoe flow forming the flows visible at the surface. The confluence of lava from different conduits involves consequent downcutting into the lower bed below the phh stack. This type of conduit formation is most likely favoured where there are easily erodible layers below the active flow. However, once the flow is concentrated in this new conduit, it may also erode even through aa core layers.AcknowledgmentsThe authors wish to thank the members of the Hawaiian Volcano Observatory for their support. Ann Bosted, Ron Carlson, Don Coons, Harry Shick, Emily Davis, Bob South,6. DiscussionErosion of underlying rock beds plays an important role during the formation of pyroducts (Greeley et al. 1998; Kempe, 1997, 2002, 2009, 2012). Two aa flows are of significance for the genesis of Kahuenaha Nui. The first one (layers 1 ac and 2 ar, Fig. 1) is seated at the base of the pyroduct, the second transgressed during a later eruption (layers 10 ar, 11 ac, 12 ar). Initially a network of numerous small pyroducts established, basically following the surface morphology. These pyroducts interact while the phh flow branches at pre-existing tumuli. The parallel flows erode the underlying lava. Due to variations in flow velocity and resistance of the base layer one branch degrades the lower bed faster, draining other small pyroducts. These initial pyroducts can be seen at several places in Kahuenaha Nui. At the Deutsches Eck (Fig. 3, Fig. 6a, P3) numerous initial conduits can be observed, depicting a dense network of contemporaneous active pyroducts. The erosion rate increases as the main flow hits the underlying aa-rubble, resulting in fast lateral and downward extension. The small number of lavafalls gives evidence for fast downward erosion. Rather than local erosion by lavafalls the base layer denudates. Fast lateral erosion produces quite a number of lavaballs, causing temporary blockage of cave passages. The created afflux boosts the establishment of parallel, simultaneously active pyroducts. If there is enough time for the cave roof to consolidate, the newly generated flow can superimpose the older flow, while both trunks are still active. We name this superimposed trunk system. Fast downward erosion of basal aa-layers facilitates this process. The emerging gasfilled space above the flowing lava effects thermal insulation, resulting in fast cooling of the cave roof. Skylights between the different levels allow interaction between the two flows. Up to now three modes of lava tunnel formation have been described (Kempe 2010). Figure 5 shows the already known inflation mode (1a) and two modes of crusting over of channels (1c and 1d). Our investigations led to the conclusion that Kahuenaha Nui seems to represent a fourth type of genesis: Confluence of lavas from smaller conduits of superimposed sheets of pahoehoe (1d). While a first sheet of phh establishes a pyroduct it is overridden and buried by a second sheet, likewise establishing a pyroduct. This process happens several times. At the same time the lowest phh-flow erodes down. The tubes merge and drain into the main passage until the lowest tube contains all the flow. Lateral and downward erosion creates a large tunnel. In addition upward erosion enlarges the passage. Figure 5. Three modes of lava tunnel formation are already known (1a, 1c, 1d). 1a shows the inflationary type. 1c and 1d represent two modes of crusting over of channels, Kahuenaha Nui seems to represent a fourth type of genesis (1b).Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings234

PAGE 237

Mike Warner, Susan and Nick White, and John Wilson assisted during field work. XRD-analysis and polarizing microscopy was conducted at the Institute of Applied Geosciences, Technical University Darmstadt.ReferencesBauer I, 2011. Geologie, Petrographie und Pyroductgenese des Kahuku-Ranch-Gebiets, Big Island, Hawaii. Dipl. Thesis, Institute of Applied Geosciences, Technical University Darmstadt, Germany (unpubl.). Greeley R, Fagents SA, Harris RS, Kadel SD, Williams DA, 1998. Erosion by flowing lava, field evidence. J. Geophys. Res., 103 (B11), 27, 325, 345. Kempe S, 1997. Lavafalls: a major factor for the enlargement of lava tubes of the Ai-laau Shield phase, Kilauea, Hawaii. Proc. 12thInt. Congr. of Speleology, La Chaux-de-Fonds, Switzerland, 1, 445. Kempe S, 2002. Lavarhren (Pyroducts) auf Hawaii und ihre Genese. In: Angewandte Geowissenschaften in Darmstadt, Rosendahl W, Hoppe A (Eds.), Schriftenreihe der Deutschen Geologischen Gesellschaft, 15, 109. Kempe S, 2009. Principles of pyroduct (lava tunnel) formation. Proc. 15thInt. Congr. of Speleology, Kerrville, Texas, July 19, 2009, 669. Kempe S, Bauer I, Bosted P, Coons D, Elhardt R,2010. Inflationary versus crusted-over roofs of pyroducts (lava tunnels). Proc. 14th. Int. Symp. Volcanospeleol., Undara, Australia, 2010, 93. Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings235

PAGE 238

Kempe S, 2012. Volcanic rock caves. In: White WB, Culver DC (Eds.), Enzyclopedia of Caves, 2nded., Academic Press/ Elsevier, Amsterdam, 865. Oberwinder M, 1996. Genese und interne Struktur des oberen Teils des Lavastroms von 1801 (Huehue Flow, Hualalai, Hawaii). Dipl.-ThesisInstitute of Applied Geosciences, Technical University Darmstadt, Germany (unpubl.). Figure 6 a, b, c. Kahuenaha Nui cave map, split in three parts.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings236

PAGE 239

GENETIC TYPES OF NON-SOLUTION CAVESPavel Bella1,2, Ludovt Gal1 1State Nature Conservancy of the Slovak Republic, Slovak Caves Administration, Hodova 11, 031 01 Liptovsk Mikul, Slovakia, bella@ssj.sk, gaal@ssj.sk2Department of Geography, Pedagogical Faculty, Catholic University, Hrabovsk cesta 1, 034 01 Ruomberok, Slovakia, Pavel.Bella@ku.sk Many caves were originated by various mechanical processes in both non-soluble and soluble rocks. The genetic classification of the specific group of caves presents one of important and actual speleogenetic problematics. The paper gives a basic categorisation of non-solution caves according to their origin. The group of non-solution caves originated by endogenous processes consists of magmatic, volcanic and tectogene caves. Fluvial erosion caves, caves formed by moving glacier and deposition of moraine sediments, suffusion caves, constructional travertine caves, mudlow caves, wave cut caves, caves originated by deformation of rocks caused by their volumetric changes, caves originated by gravity disintegration of rocks, mechanically weathered caves, eolian caves, caves of biogenic origin, pyrogenic caves, and caves formed as a result of thetechnogenic activatisation of natural processesare distinguished within the group of non-solution caves originated by exogenous processes.1. IntroductionDuring last years much knowledge was obtained about the caves originated by non-solution processes. But their genetic classification is not completed until now. Only partial classifications of some genetic groups of non karstic caves were worked up, mostly without the unified terminology (Halliday 1960, 2004; Wood 1974; Grimes 1975; Vtek 1983; Andrej uk 1985; Sjberg 1986; Striebel 1999; Honda 1999; Bella and Gal 2007, and others). Based on clearly defined geological and geomorphologic criteria (action of endogenous or exogenous processes, type of geomorphic process and mechanism of cave origin) the basic categorisation of non-solution caves is presented in the paper (more detailly elaborated by Bella 2012). We do not deal with caves originated by solution in hardly soluble rocks, conglomerates or other lithified sedimentary rocks and by melting of ice.2. Caves originated by endogenous processesThis group of caves originated by natural processes induced by the inner dynamics of the Earth, i.e. magmatism, volcanism, earthquake or tectonic processes. 2.1. Magmatic caves They are smaller geode-like magmatic cavities created by intrusive magmatism under the Earths surface, often without natural entrance (cavities uncovered in the mines or quarries). They often contain many crystals of various minerals formed by slowly cooled solution in magma bubbles (Dubljanskij and Andrej uk 1989). 2.2. Volcanic caves This wide group includes the caves created by volcanic explosion, lava effusion and various volcanic rheogenetic, pneumatogenetic and pyrogenic processes. Roofed lava channels subcrustal lava caves originated by gradual growing of hardening lava from both sides of roof. The roofing happened also by connecting of flowed blocks of hardened lava and by lateral levees of inunded lava (Greeley and Hyde 1972; Atkinson et al. 1975; Calvari and Pinkerton 1999; Grimes 2008; Harris et al. 2009). Lava tube caves tubes or tunnels formed by outflowing of hot lava from the bellow solidified roof crust. If the liquid lava has been locked inside the tunnel, in these places large rooms originated. Straight tunnel has symmetric profile while asymmetric profile is resulted by the change of a direction of flowing lava (Greeley 1971; Wood 1974; Calvari and Pinkerton 1999; Halliday 2004b). Also lateral narrow subcrustal simple or branched tubes (lava overflow caves) can be formed on the side of main lava flow (Grimes 2008). Kempe (2009) distinguishes single-trunked, doubleor multitrunked, and superimposed-trunked systems of lava flows. Drainage tubes of active rift zones underground spaces looking like lava tube caves can occur within large dykes intruded along fractures, eruptive fissures or rift zones lava. The fissures are remodelled by intruded lava and the spaces have an oval shape (Harter 1976; Favre 1993; Okubo and Martel 1998; Halliday 2008). Lava rise caves some places of horizontal lava surface raised by pressure of accumulated hot lava about 0.5 m thereupon the low-lying underground spaces are created (Halliday 1998; Gadnyi 2008). Lava ridge caves smaller triangle-shaped caves occur in the lateral part of lava flows in consequence of decreasing in central zone of lava surface, also in the toe part of lava flow by pressure of flowing lava (Chitwood 1989; Gadnyi 2008). Lava tumulus caves occur under elliptic elevation of lava surface (tumulus) in consequence of decreasing of solidified lava. They can reach about 5 m of height (Ollier 1962; Chitwood 1989; Halliday 1998; Gadnyi 2008). Boulder caves in lava flows generally occur in the frontal part of aa lava between larger blocks of solidified lava surface. Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings237

PAGE 240

Volcanic crater shafts vertical shafts originated by decreasing lava into the diatremas and necks of crater structure. They have floater or bell-like shape (Stefnson 1992; Favre 1993). Volcanic exhalation caves smaller caves created by exhalation or explosion of ascending volcanic gases in scoria cone. In an ideal case the chimney, which flowed into small horizontal cave, is preserved (Eszterhs 2004; Gal and Balciar 2008). Chimneys of spatter cones and hornitos open vertical volcanic conduits formed by blowing of very liquid lava from primary fractures which creates spatter cones or hornitos (Ollier 1964; Licitra 1993). Lava blister caves lava bubbles of diameter about 1 m, rarely to 5 m originated by accumulation and pneumatic expansion of gas in lava (Ollier 1962). Polygenetic spatter cone caves in carbonatite volcanoes a rare case of cave origin in spatter cones in the summit crater of Ol Doinyo Lengai in Tanzania, cavities under spatter cones originated by thermal erosion, and meteoric water and condensation corrosion of natrocarbonatite lava (McFarlane et al. 2004). Volcanic pyrogenetic tree moulds formed by burning-out of trees in glowing lava (Finch 1931; Balzs 1974; Honda 1999; Gal 2003a; Bella and Gal 2007). According to the position of tree trunks they can be vertical, inclined or horizontal. Tree moulds are known as simple type, compound type with several trees and tree-break type. From a morphological point of view the mushroom type, plate type, bottle type, narusawa type (cedar-like cave after burning of gase) and hudoiwa type (with accumulated lava in a frontal part of tree) are distinguished (Tachihara et al. 2002). Volcanic eruptive fissure caves along the fissures originated by movement or quake eventually by pneumatic pressure during volcanic eruptions. Generally they have fissure character sporadically with marks of flowing lava (Ogawa 1986; Leotta and Liuzzo 1998; Halliday 2004b). 2.3. Tectogene caves The group includes the caves originated in consequence of endogenous tectonic movement and earthquake. They occur mostly along the fissure, rarely on the fold structure. Fault caves along the disjunctive fault and fissures formed by extension tectonic movement. They have linear character with narrow and long passages (Kiernan 1982; Dubljanskij and Andrej uk 1989). Rift caves fault caves in the margin part of the rift zone and trench fault along the mobile zones of the Earth (Davis 1998). Fold caves rare occurring caves on the flexion of folds (Kiernan 1982). Collapsed pit crater caves vertical circular or elliptic shafts along the active rift zone in the crossing place with structural or stratigraphic discontinuity. They occur mostly in areas of large shield volcanoes (Halliday 1998; Okubo and Martel 1998). Seizmotectonic fracture caves caves along the narrow fissures created in consequence of decompression seismic quake after melting of glacier and isostatic uplift in Scandinavia (Sjberg1986). Seismotectonic boulder caves underground spaces between rock blocks originated by disintegration of granite massive disturbed by seismic quake. The seismotectonic boulder caves can also be originated by earthquake-induced rockfall (Sjberg1986, 1987).3. Caves originated by exogenous processesThis group involves the caves, which originated by fluvial erosion, glacier moving, suffosion, marine abrasion, gravity disintegration (sliding and falling of rock blocks), mechanical weathering (fissure and boulder exfoliation), frost weathering, eolian processes (corrosion and deflation), biogenic way, pyrogenic processes and the caves of technogenic activatisation of natural processes. 3.1. Fluvial erosion caves The genetic group of caves originated by mechanical erosion of water flow that means abrading of rock walls and scouring of river channels. River bank erosion caves formedin the rock banks of river channel by lateral erosion mainly in meanders along horizontal bedding plane (Trimmel 1968; Striebel 1999; Day 2001). Waterfall erosion caves formed by backward erosion of falling water under occasional or constant waterfall (Striebel 1999; Day 2001). Fluvial channel erosion caves occur at the contact of rocks of different permeability and hardness or along fractures, the water course become engrosses in the floor (Kempe and Werner 2003). 3.2. Caves formed by moving glacier and deposition of moraine sediments This group includes the caves formed by mechanical deformation of glacier and underlying rocks, also subglacial caves formed as glacier slides over abrupt changes in the bedrock surface, and glacial boulder caves. Glacier crevasse caves formed by extension strain on the elevation of bedrock (V shape crevasse) or by compression in depression (A shape crevasse). The other diagonal crevasses can be formed in consequence of unequal moving of glacier in its central and side parts (White 1988). Crevasses formed by rotation of ice masses in the steep upper part of glacial cirquesare named bergschrunds(Battle and Lewis 1951). Crevasse caves under glacier in bedrock deformated and fractured by pressure of moving glacier, so-called glaciotectonic caves (Schroeder et al. 1986; Sjberg 1986, 1989; Schroeder 2004). Subglacial cavities formed as glacier slides over abrupt changes in the bedrock surface, the ice separated from bedrock in the lee of a downward step (Pulina et al. 2003; Fountain 2005). Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings238

PAGE 241

Glacial boulder caves irregular, chaotic cavities between large blocks of moraine or erratic boulders (Kiernan 1982; Sjberg 1986). 3.3. Suffosion caves Suffosion or piping caves formed by washing out of soft clastic material by infiltration or flowing underground water. They originated mainly in less consistent clastic rocks as clay, loess, silt, laterite, weathered sandstone and volcaniclastic rocks. Suffosion tunnel caves originated by suffosion and erosion generally on the contact of loose rocks and impermeable background (Clausen 1970; Clek 1997; Bella et al. 2005; Grimes and Spate 2008). Suffosion collapse shafts formed by decrease and subsequent collapse of hanging rock upper suffosion tunnels (Clausen 1970; Urban 2004). 3.4. Synsedimentary travertine caves Constructional caves originated by precipitation of travertine from solutions. Travertine crater caves spring travertine throats inside travertine craters (Bella 2005; Halliday 2011). Travertine fissure ridge caves rare occurring longitudinal narrow cavities along an effluence fissure (Pisarowicz 2003; Halliday 2011). Caves of constructive travertine waterfalls mostly longitudinal cavities along the foot of constructive travertine waterfalls(Trimmel 1968; Bella 2005). 3.5. Mudflow caves They present sporadic caves formed by outflowing of liquid mud from under and upper dried indurate crust on slopes of mud volcanoes (Kalenda 2010). 3.6. Wave-cut caves They are formed by mechanical wave abrasion at the foot of costal cliffs of sea or large lake formed by insoluble rocks (Moore 1954; Bunnell 1988). 3.7. Caves originated by deformation of rocks caused by their volumetric changes These are rarely occurring caves formed by enlarging of volume of evaporite caused by water absorbing. In consequence of hydratation of anhydrite to gypsum the rock can expand till 50 % whereby some hydratation raised subcrustal cavities is formed. Hydratation raised caves upward raised subcrustal cavities between hydrated gypsum upper anhydrite or subjacent beds. Some narrow cavities gypsum tumuli are also created by rising caused by precipitation or recrystallization of gypsum (Gorbunova 1978; Reinboth 1997; Callafora and Pulido-Bosch 1999). Subcrustal cavities of tepee structure upward cambered margin of expanded megapolygons of agglutinated crust of carbonate-evaporite coastal sediments in periodic inundated and shrinkages basins (Kendall and Warren 1987). Salt diapire fissure caves occur in the raising salt stump or diapire of hydrated anhydrite (Vendeville and Jackson 1992). 3.8. Caves originated by gravity disintegration of rocks Such caves occur in areas with vertically dissected relief and suitable geological conditions (rigid rocks with underlying soft or plastic beds). Crevice caves formed by gravity movement of marginal rigid blocks upper underlying plastic beds on the slopes (Campbell 1968; Vtek 1983;Self 1990; Turchinov 1990; Filippov 1997; Margielewski and Urban 2003; Halliday 2004 and others). Gal (2003b) distinguishes deflection crevice caves (formed in initial or opening phase by deflection of marginal blocks), inclination crevice caves (formed by backward inclination of blocks in relation to their rotation), decrease crevice caves (horizontal or subhorizontal caves formed by decreasing of blocks in the initial phase), displacement crevice caves (caves in wide crevice formed by gravity transfer of block in the sliding phase, mostly combined crevice-boulder caves are formed). The crevice caves are also formed on the coastal cliffs of a sea or lake and on the gravity disintegrated steep coast of carbonate platforms. The bank-margin fracture type of blue holes in Bahamas probably was formed during glacioeustatic decrease of sea level (Smart et al. 1988; Mylroie 2004). Falling-out caves small caves formed after gravity fallingout of released blocks. Disintegration and releasing can be supported by termomechanical weathering (Striebel 1999). Rockfall / rockslide boulder caves chaotic cavities formed by deposition of disintegrated large angular boulders from falling or sliding rock blocks at the base of cliffs on slopes, in the bottom of valleys or at the foot of coastal cliffs. Rockfall boulder caves are originated by sudden falling of rocks, rockslide boulder caves by sliding and subsequent disintregration of rock blocks (Kiernan 1982;Vtek 1983; Striebel 1999;Bella and Gal 2010 and others). Outside arid regions, in-valley boulder caves (so-called purgatory caves in California and the northeastern United States) are commonly featured by mechanical erosion of underground streams (Finlayson 1981;Halliday 2004a). 3.9. Caves originated by mechanical weathering This group involves caves originated by mechanical weathering and associated slope processes: caves in rock structures disintegrated by macroexfoliation, frost weathering caves, and caves formed by insolation and salt weathering. Exfoliation caves formed in consequence of segregation of rock squama mostly parallel with rock surface. Therefore the exfoliation cracks are often curved like an onion. Exfoliation caves are formed along the fissures (fissure exfoliation cave), by falling-out of block along the fissure (fissure-falling-out exfoliation cave), among boulders formed by exfoliation disintegration along the bedding or cubical jointing in situ (boulder exfoliation cave), wide and narrow simple exfoliation Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings239

PAGE 242

underground spaces (woolsack cave), weathered exfoliation caves and small deltoid caves of A-tents type (Ollier 1965; Jennings 1978; Striebel 1999; Twidale 2009). Frost weathering caves generally shelter type caves originated by break-up of host rock in consequence of frost (Schroeder 1979; Mitter 1983; Clek 1996; Striebel 1999). Belong them also weathered boulder caves in glacial and periglacial regions (Striebel 1999). Tree moulds formed by mechanical disintegration of silicified or carbonified trees covered by volcaniclastic beds or lahars (Gal 2003a; Bella and Gal 2007). Tafoni caves originated by insolation weathering along the contact plane of primary cubical jointing of granite and similar rocks. The favorable conditions of its origin are in repeated moistly and dry climatic areas (Martini 1978; Twidale 1982; Vidal Romani 2007). Salt weathering caves small caves originated by change of volume in consequence of crystallization of salt in poruses and cracks. They are as foot-cliffs shelter caves or caves formed along lithologically or tectonically weakened parts of rocks (Gavrilovi 1981; Clek 1998). 3.10. Eolian caves This group involves smaller caves originated by wind corrasion and deflation mainly in soft and loose rocks located mainly in arid and semiarid regions. Corrasion caves simple elliptic cavities originated by mechanical abrasion by wind (White 1988; Dubljanskij and Andrej uk 1989). Deflation caves short cavities generally with two entrances along which wind insufflate the soft particles of weathered rock. They occur most frequently in mountain ridges and elevation in open terrains (Dubljanskij and Andrej uk 1989). 3.11. Caves of biogenic origin Reef caves small irregular constructional cavities in barrier reefs and atolls formed by growing of corals under sea level (Trimmel 1968). Caves excavated by animals small cavities excavated mainly as hiding-place (salt ingestion caves). These caves occur in rock with salt because salt is searched by elephants, buffalos, antelopes and other animals (Carey and Middleton 1973; Lundquist and Varnedoe 2006). Biogenic tree moulds formed by biogenic disintegration of necrotic tree trunks by funguses and microorganisms. They are known mainly in travertine (Bella and Gal 2007; Gradzi ski 2008). 3.12. Pyrogenic caves of self-kindled fires These caves represent cavities after burned out of coal layers or other burnable raw material (Maximovich 1969; Dubljanskij and Andrej uk 1989). In volcanic rocks tree moulds formed by burning was described as volcanic pyrogenic tree moulds. 3.13. Caves of technogenic activatisation of natural processes This group involves caves originated by natural processes but in consequence of human activity, mostly mining, exploitation of raw material or violation of slope stability. Consequent crevice and breakdown caves formed by gravity block movement anthropogenically induced by rock instability (Eszterhs 1993). Technogenic pyrogenic caves created by underground burning of coal in order to obtain of gas for electric energy (Andrej uk 1985; Dubljanskij and Andrej uk 1989).4. ConclusionTotally 56 types of non-solution caves are distinguished and briefly characterised in this paper (21 types are related to endogenous processes, 35 types to exogenous processes). Various genetic types of non-solution caves point to a high diversity of their formation in varied environments with different lithological, structural-tectonic, geomorphologic and climatic conditions. Non-solution caves originated mostly in magmatic, volcanic and volcaniclastic rocks, crystalline shales and clastic sedimentary rocks, constructional caves also in travertine and coral limestone. Some specific types of nonsolution caves are known in gypsum and salt. A higher variety of cave formation by exogenous processes is linked with a higher heterogenity of the Earth-surface dynamics. Comparing with solution caves, an occurrence of non-solution caves is lower but their genetic diversity is higher. Several non-solution caves are important sites of nature and human history. Therefore they are protected as natural monuments or are included within protected areas.ReferencesAndrej uk VN, 1985. Klassifikacija podzemnych polostej. Izvestija VGO, 117, 4, 341 (in Russian). Atkinson A, Griffin TJ, Stephenson PJ, 1975. A major lava tube system from Undara Volcano, North Queensland. Bulletinof Volcanology, 39(2), 266. Balzs D, 1974. Lvaregek keletkezse, tpusai s formakincse. Fldrajzi Kzlny,2, 135. Battle WRB, Lewis WV, 1951. Temperature observation in bergschrunds and their relationship to cirque erosion. Journal of Geology, 59(6), 537. Bella P, 2005. Syngenetick travertnov jaskyne na Slovensku. Geomorphologia Slovaca, 5(2), 23 (in Slovak). Bella P, 2012. Genetick typy jask Verbum, Ruomberok (in Slovak). Bella P, Gal 2007. Tree mould caves within the framework of cave genetic classification. Nature Conservation, 63, 71. Bella P, Gal 2010. Medzibalvanov jaskyne terminolgia a genetick typy. Aragonit, 15(1), 3 (in Slovak). Bella P, Gal Inokura Y, 2005. Sufzne jaskyne vo vulkanoklastickch horninch v doline Nagatani pri Kagoime. Slovensk kras, 43, 67 (in Slovak). Bunnell D, 1988. Sea caves of Santa Cruz Island, Santa Barbara, California. McNally and Loftin Publishers.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings240

PAGE 243

Calaforra JM, Pulido-Bosch A, 1999. Genesis and evolution of gypsum tumuli. Earth Surface Processes and Landforns, 24(10), 919. Calvari S, Pinkerton H, 1999. Lava tube morphology on Etna and evidence for lava emplacement mechanism. Journal of Volcanology and Geothermal Research, 90(3), 263. Campbell NP, 1968. The Role of Gravity Sliding in the Development of Some Montana Caves. Bulletin of the National Speleological Society, 30(2), 25. Carey SD, Middleton AL, 1973. The Rock House An Unusual Mississippi Shelter Cave. NSS News, 31(11), 198. Clek V, 1996. Vznik previs a jeskynnch vklenk v braln asti Velk Fatry. Speleofrum, 15, 34 (in Slovak). Clek V, 1997. Sufozn podzemn systm ve spraov rokli v Zem chch u Kralup. Speleo, 25, 19 (in Czech). Clek V, 1998. Fyziklne-chemick procesy vzniku pskovcovho pseudokrasu. In: V Clek, J Kopeck (Eds.) Pskovcov fenomn: klma, ivot a relif. Knihovna SS, 32, Praha, 134 (in Czech). Clausen EN, 1970. Badland caves of Wyoming. Bulletin of the National Speleological Society, 32 (3), 59. Davis RA, 1998. Tectonic caves of Solai in the Kenyan Rift Valley. International Journal of Speleology 27B(1), 69. Day MJ, 2001. Sandstone caves in Wisconsin. Proceedings of the 13thInternational Congress of Speleology, 1, Brasilia, 88. Dubljanskij VN, Andrej uk V N, 1989. Speleologija (terminologia, svjazi s drugimi naukami, klassifikacija polostej). Ura skoe otdelenie AN SSSR, Kungur (in Russian). Eszterhs I, 1993. Konsequenzhhlen. Jahresbericht der Hhlenforschergruppe Rhein-Main, Frankfurt a. M., 43 (in German). Eszterhs I, 2004. Durch exhalation entstandene hhlen im Karpatenbecken.In: L Gal (Ed.) Proceedings of the 8thInternational Symposium on Pseudokarst, Tepl Vrch, 7. Favre G, 1993. Some observations on pits craters and relations with lava tubes. In: WR Halliday (Ed.) Proceeding of the 3rdInternational Symposium on Vulcanospeleology (Bend, Oregon), Washington, 37. Filippov AG, 1997. Gravity caves of the Siberian Platform. In: PY Jeannin (Ed.) Proceedings of the 12thInternational Congress of Speleology, 1, La Chaux-de-Fonds, 465. Finch RH, 1931. Lava tree casts and tree molds. The Volcano Letter (Hawaiian Volcano Observatory), 316, 1. Finlayson B, 1981. Underground streams on acid igneous rocks in Victoria. Helictite, 19(1), 5. Fountain AG, 2005. Glacier Caves. In: DC Culvier, WB White (Eds.) Encyclopedia of caves. Elsevier Academic Press, 271. Gal 2003a. Tree mould caves in Slovakia. International Journal of Speleology, 32, 1, 10711. Gal 2003b. Genetick typy rozsadlinovch jask na Slovensku. Slovensk kras, 41, 29 (in Slovak). Gal Balciar I, 2008. Caves in the youngest volcanic structure in Slovakia. Proceedings of the 10th International Symposium on Pseudokarst, Gorizia, 165. Gadnyi P, 2008. Caves under uplifted surface crusts of basalt lava flows. Proceedings of the 10thInternational Symposium on Pseudokarst, Gorizia, 119. Gavrilovi D, 1981. Genetic types of caves in Sahara. Evropejska regionalna konferencia po speleologija, Sbornik ot materiali, 2, Sofia, 137. Gorbunova KA, 1978. Peery gidratacii. Pe ery, 61 (in Russian). Gradzi ski M, 2008. Origin of a unique tree-mould type cave in travertine based on examples from the village L ky (Liptov, Slovakia). Slovensk kras, 46(2), 325. Greeley R, 1971. Observations of actively forming lava tubes and associated structures, Hawaii. Modern Geology, 2, 207. Greeley R, Hyde JH, 1972. Lava Tubes of the Cave Basalt, Mount St. Helens, Washington. Geological Society of America Bulletin, 83(8), 2397. Grimes KG, 1975. Pseudokarst: Definition and types. Proceedings of the 10thBiennial Conference of the Australian Speleological Federation (Brisbane, 27 December 1974), Sydney, 6. Grimes KG, 2008. Subcrustal Drainage Lava Caves; examples from Victoria, Australia. In: R Espinasa-Perea, J Pint (Eds.) Proceedings of the X, XI, and XII International Symposia on Vulcanospeleology. AMCS Bulletin, 19, 35. Grimes K, Spate A,2008. Laterite Karst. Australasian Cave and KarstManagement Association Journal, 73, 49. Halliday WR, 1960. Pseudokarst in The United States. Bulletin of the National Speleological Society, 22(2), 10913. Halliday WR, 1998. Hollow volcanic tumulus caves of Kilauea Caldera, Hawai County, Hawaii. International Journal of Speleology, 27B(1), 95. Halliday WR, 2004a. Talus caves. In: J Gunn (Ed.) Encyclopedia of Caves and Karst Sciences. Fitzroy Dearbon, New York London, 723. Halliday WR, 2004b. Volcanic caves. In: J Gunn (Ed.) Encyclopedia of Caves and Karst Sciences. Fitzroy Dearbon, New York London, 760. Halliday WR, 2008. What Is a Lava Tubes? In: R Espinasa-Perea, J Pint (Eds.) Proceedings of the X, XI, and XII International Symposia on Vulcanospeleology. AMCS Bulletin, 19, 48. Halliday WR, 2011. Karst and pseudokarst at Mammoth Hot Springs, Yellowstone National Park, USA. International Union of Speleology, Pseudokarst Commission Newsletter, 22, 10. Harris AL, Massimiliano F, Francesco M, Hamilton CW, 2009. Construction dynamics of a lava channel. Bulletin of Volcanology, 71, 4, 459. Harter JW, 1976. Morphologic Classification of Lava Tubes. In: WR Halliday (Ed.) Proceedings of the International Symposium on Vulcanospeleology (White Salmon, Washigton). NSS, Seattle, 74. Honda T, 1999. Classification of lava tree molds with/without remelted inner surface according to its formation process. In: N Barone, G Licitra (Eds.) Proceedings of the IXthInternational Symposium on Vulcanospeleology, Catania, Italy, 123. Chitwood LA, 1989. Inflated lava. Desert Ramblings, The Newsletter of the Oregon Natural Desert Association, 2(1), 1. Jennings JN, 1978. Genetic variety in a-tents and related features. Australian Geographer, 14(1), 34. Kalenda P, 2010. zerbajdn zem bahennch sopek a tak bahennch jeskyn. Speleo, 54, 16 (in Slovak). Kempe S, 2009. Principles of pyroduct (lava tunnel) formation. In: WB White (Ed.) Proceedings of the 15thInternational Congress of Speleology, 2, Kerrville, Texas, USA, 668.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings241

PAGE 244

Kempe S, Werner MS, 2003. The Kukaiau Cave, Mauna Kea, Hawaii, created by water erosion, a new Hawaiian cave type. Journal of Cave and Karst Studies, 65(1), 53. Kendall CGSt, Warren J, 1987. A review of the origin and setting of tepees and their associated fabrics. Sedimentology, 34(6), 1007. Kiernan K, 1982. Mechanically shaped pseudokarst: talus, joint and fault caves and their potential in Tasmania. Journal of Sydney Speleological Society, 26(3), 41. Leotta A, Liuzzoi M, 1998. The 1981 eruptive fissure on Mt. Etna: considerations on its exploration and genesis. International Journal of Speleology 27B(1), 147. Licitra GM, 1993. Essay on genetic classification of volcanic caves. In: WR Halliday (Ed.) Proceedings of the 3thInternational Symposium on Vulcanospeleology (Bend, Oregon), Vancouver, Washington, 118. Lundquist ChA, Varnedoe WW, 2006. Salt ingestion caves. International Journal of Speleology, 35(1), 13. Margielewski W, Urban J, 2003. Crevice-type caves as initial forms of rock landslide development in the Flysch Carpathians. Geomorphology, 54(3), 325. Martini IP, 1978. Tafoni weathering, with examples from Tuscany, Italy. Zeitschrift f r Geomorphologie, 22(1), 44. Maximovich GA, 1969. Peery podzemnych poarov. Peery 7(8), 87 (in Russian). McFarlane DA, Lundberg J, Belton F,2004. An Unusual Lava Cave From Ol Doinyo Lengai, Tanzania. Journal of Cave and Karst Studies, 66(3), 98. Mitter P,1983. Frost Features in the Karst Region of the West Carpathian Mountains. In: TL Pewe, J Brown (Eds.) Permafrost: Proceedings of the Fourth International Conference, Fairbanks July 17, 1983. Washington DC, 861. Moore DG, 1954. Origin and development of sea caves. Bulletin of the National Speleological Society, 16, 71. Mylroie JE, 2004. Blue holes of the Bahamas. In: J Gunn (Ed.) Encyclopedia of Caves and Karst Sciences. Fitzroy Dearbon, New York London, 155. Ogawa T, 1986. The formation of lava caves. Communications, 9thInternational Congress of Speleology, 2, Barcelona, 47. Okubo C, Martel S, 1998. Pit cra ter formation on Kilauea volcano, Hawaii. Journal of Volcanology and Geothermal Research, 86 (1), 1. Ollier CD, 1962. Tumuli and Lava Blisters of Victoria, Australia. Nature, 202, 1284. Ollier CD, 1964. Caves and related features at Mount Eccles. Victorian Naturalist, 81, 64. Ollier CD, 1965. Some features of granite weathering in Australia. Zeitschrift fr Geomorphologie, NF, 9, 285. Pisarowicz J, 2003. Beneath Yellowstone. Rocky Mountain Caving, 20(4), 12. Pulina M, ehk J, Schroeder J. 2003. Les cavits glaciaires sous le regard splologique. Karstologia, 42(2), 23. Reinboth F, 1997. Die Zwerglcher bei Walkenried am SdharzBemerkumgen zur Frage der Quellungshhlen. Die Hhle, 48(1), 1 (in German). Self C, 1990. A gravity sliding cave in Western England. In: J Wagner (Ed.) IV. symposium o pseudokrasu, sbornk refert Knihovna SS, 23, Praha, 139. Schroeder J, 1979. Dvelopement de cavits dorigine mcanique dans un karst froid (Nahanni, T. N. O., Canada). Annales de la Socit gologique de Belgique, 108, 59. Schroeder J, Beaupr M, Cloutier M, 1986. Ice-push caves in platform limestone of the Montreal area. Canadian Journal of Earth Sciences, 23(11), 1842. Schroeder J, 2004. Glaciotectonic cavities. In: AS Goudie (Ed.) Encyclopedia of Geomorphology. Routledge, London, 470. Sjberg R, 1986. A proposal for a classification system for Granitic Caves. Communications, 9thInternational Congress of Speleology, 2, Barcelona, 25. Sjberg R, 1987. Caves as indicators of neotectonics in Sweden. Zeitschrift fr Geomorphologie, NF, Suppl. 63, 141. Smart PL, Palmer RJ, Whitaker F, Wright VP 1988. Neptunian Dikes and Fissure Fills: An Overview and Account of Some Modern Examples. In: NP James, PW Choquette (Eds.) Paleokarst. Springer-Verlag, New York Berlin Heidelberg London Paris Tokyo, 149. Stefnson B, 1992. Prhnkarggur. In: GT Rea (Ed.) Proceeding of the 6thInternational Symposium on Vulcanospelogy, 197. Striebel T, 1999. Typen von Sandsteinhhlen und Granithhlen in der Umgebung von Bayreuth. Pseudokrasov zbornk, 1, Knihovna SS, 35, Praha, 51. Tachihara H, Sawa I, Kuroishikawa Y, Ogawa T, Honda T, Kim B, Makita T, Watanabe N, Ninata H, Nakau K, 2002. The shape classification and formation model by observation of lava treemold. Review of Osaka University of Economics and Law, 84, 46. Trimmel H, 1968. Hhlenkunde. Friedr. Vieweg und Sohn, Braunschweig. Turchinov I. 1990. Pseudokrasov jeskyn ve Skolevskch Beskydech (Ukrajinsk Karpaty). In: J Wagner (Ed.) IV. symposium o pseudokrasu, Knihovna SS, 23, Praha, 119 (in Czech). Twidale CR, 1982. Granite landforms. Elsevier, Amsterdam. Twidale CR, 2009. On the origin of A-tents (pop-ups), sheet structures, and associated forms. Progress in Physical Geography, 33(2), 147. Urban J, 2004. Morphological evolution of the pseudokarst forms in Quaternary loesses of Southern Poland a case study of Bugaj near Pinczw, Nida Basin. In: L Gal (Ed.) Proceedings of the 8th International Symposium on Pseudokarst, Tepl Vrch, 75. Vendeville BC, Jackson MPA, 1992. The rise of diapirs during thinskinned extension: Marine and Petroleum Geology, 9(4), 331. Vidal Roman JR, 2007. Origin of tafoni elastic deformation. International Conference on Granite Caves. Abstracts, La Corua, 20. Vtek J, 1983. Classification of Pseudokarst Forms in Czechoslovakia. International Journal of Speleology, 13(1), 1. White W B, 1988. Geomorphology and Hydrology of Karst Terrains. Oxford Univ. Press, Oxford New York. Wood C, 1974. The genesis and classification of lava tube caves. Transactions of the British Cave Research Association, 1(1), 15.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings242

PAGE 245

THE KEOKEO LAVA TUBE SYSTEM IN HAWAIIPeter Bosted1, Tomislav Gracanin2, Veda Hackell2, Ann Bosted1, Ingo Bauer3, Stephan Kempe3 1PO Box 6254, Ocean View, HI 96737 USA peter@cavepics.com2Houston, Texas, tnv@att.net3University of Technology Darmstadt, Schnittspahnstr. 9, D-64287 Darmstadt, Germany, kempe@geo.tu-darmstadt.de The Keokeo lava tube system lies within a lava flow that apparently originated in the Puu o Keokeo vent on the Southwest Rift zone of Mauna Loa, on the Big Island of Hawaii. The lava flowed 1,500,000 years ago from the source near 2,200m elevation all the way to the ocean, a distance of some 28 km. A large lava tube system has been discovered in most of the flow. The cave system is often braided into parallel branches, as well as being formed on up to four vertical layers. The system is sometimes segmented by ceiling collapse or lava sumps. As of this writing, about 22 km of passage has been surveyed in segments ranging from 10 m to 4 km long. The deepest passages are readily distinguished because the lava downcut through a thick layer of aa, which is seen when the wall linings peel away. In places, there is evidence of downcutting through two aa layers. There are many white and orange crusts, frostwork, and coatings throughout the system, which could be the subject of a mineralogical investigation. Microbial mats are fairly common as well. Ohia tree roots support a variety of cave-adapted species, including crickets, spiders, isopods, planthoppers, and moths. Other interesting resources in the system include fossil bird bones and occasional native Hawaiian water collection and shelter sites.1. IntroductionThe Big Island in the U.S. state of Hawaii is home to the majority of major lava tube (pyroduct) systems (e.g., Kempe 2012) in the world. The second longest system in the world, the Kipuka Kanohina system, originates in the SW rift zone of Mauna Loa, within the Kahuku Unit of Hawaii Volcanoes National Park (HAVO), and passes under the Hawaiian Ocean View Estates (HOVE) and Kula Kai subdivisions, reaching almost to the ocean. Recently, another major system has been identified, also originating in the SW rift zone. The approximate location of the system is shown as the thin and thick black lines in Figure 1. The system is formed in the Qk2 Mauna Loa flow which extends from the presumed source at Puu o Keokeo (altitude of 2,200 m), all the way down to the ocean, some 28 km distant. The estimated age of the flow is 1,500 to 3,000 aBP (Sherrod 2007).2. Exploration HistorySome of the lava tubes in the system were used by the original Polynesian settlers, possibly as long as 1,000 years ago. They likely explored the caves looking for good places to collect water in thisdry environment. They also used some entrance areas for temporary shelters and defense in times of war. In most of the 1800s and 1900s, the entire flow was on the property of the very large Kahuku Ranch. Ranch-hands and farm children sometimes visited the caves, although there is no written record of their activities. The upper portion of the Keokeo flow became part of Hawaii Volcanoes National Park (HAVO) in 2004. The lower portion was sold to private developers (Nani Kahuku Aina) in 2006. Systematic exploration of the Keokeo system began in 2011. Some entrances were found from satellite and aerial photos. The great majority of them were found on surface hikes, or by entering one entrance and coming out another one. To date, about 300 entrances have been identified. Of these, less than 100 have been explored and surveyed. For the entrances within HAVO, an initial reconnaissance of all of the passages accessible from a given entrance was made with archaeologist Dr. Tim Scheffler, in order to determine if sensitive cultural resources were present. To date, there have been about 110 field trips to the flow, of which roughly one quarter were to locate entrances, one quarter for archaeological reconnaissance, and half for detailed mapping and photography. About 22 km of passage has been mapped in segments ranging from less than 10 m to over 4 km in length. The segments are usually separated by trenches formed by the collapse of the original tube. In other cases, especially where the gradient is low, the lava completely filled the passages when it cooled, making for lava sumps.3. GeologyThe geology of one of the longest segments of the system, Kahuenaha Nui, was studied by Ingo Bauer as part of his Masters thesis (Bauer 2011). The cave survey of this segment yielded a total length of 1,850 m, a total vertical extent of 55 m and an average slope of 5.7. The cave features a main trunk that is up to 18 m wide and 11 m high. Its floor is in part formed by terminal aa. Above this trunk passage, there are numerous small to very small interconnected pahoehoe-floored passages. It was possible to study the cave formation process at the two entrance collapses (pukas in Hawaiian). The trunk passage appears to be formed by eroding an underlying aa rubble layer. In places even the underlying aa core layer has been cut into. Above, a stack of seven superimposed pahoehoe flows with small meandering passages occurs, forming the primary roof of the cave. The lava flowing in this stack of sheets managed to combine into one flow, eroding the main trunk underneath. After cave formation first an aa flow and then a thin pahoehoe flow transgressed the area. The caves roof partly collapsed, not only exposing the transgressed aa, but also forming the two entrance pukas. This cave-forming mechanism is fundamentally different from the standard Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings243

PAGE 246

lava tube formation modes inflation and the crustingover of channels (e.g., Kempe 2012).Figure 1. Geologic overview of the Keokeo flow on the south slope of Mauna Loa. The thick black lines show the approximate location of explored passages, while the thin black lines are likely extensions to the north and south. The distance from the top to bottom of the figure is about 30 km, and north is up. The Keokeo flow is colored light green. Older flows are yellow and green, while modern flows are colored blue. Highway 11 crosses the figure from left to right just below the middle. Figure 2. Illustration of wall-lining peeling away to reveal a thick layer of red aa rubble behind. Bob South is the model. Photo by Peter and Ann Bosted. Figure 3. Stephan Kempe (in red) and Tim Scheffler above the largest entrance of the system found so far. Photo by Peter and Ann Bosted. While detailed geologic studies have not been done in other segments of the system, the mode of down-cutting into an aa layer appears to persist throughout the main, lower level trunk passages. This is most often revealed by wall coatings, typically 0.2 to 1 m thick, that have peeled away, generally falling towards the center of the passage, and leaving a pile of exposed aa rubble behind. Often the aa rubble is stained orange or red, by oxidation of the iron in the basalt. One of the more dramatic examples is shown in Figure 2. Throughout the system, the larger entrances are often the best place to study the stratigraphy. Where there are clean ceiling collapses, it is sometimes possible to count the number of pahoehoe sheets and aa layers. The lava tube passages often contain white, orange, and sometimes pink or green minerals. These most commonly occur as crusts or coatings on the ceiling, walls, and sometimes on the floor. There is a noticeable correlation of mineral coatings with high wind velocities near major entrances and constrictions where the Venturi effect makes for strong air flow. The very dry climate of the region likely stops the minerals from being dissolved, because they are much more common here than in the wet areas of Hawaii. Based on sampling in the nearby Kanohina system, most of the white minerals are probably calcite, gypsum or opal, although epsomite and mirabilite can be found. A typical medium-size passage with white, orange, and red coatings, and an aa floor over smooth linings, is shown in Figure 4. In some areas, the minerals are in a frostwork morphology, as illustrated in Figure 5.4. BiologyThe Keokeo flow covers a large elevation range (from sea level to about 2,200 m), but the endemic Hawaiian Ohia tree can be found throughout almost the entire range. This tree is very important to the cave biology, because it is adapted to putting its roots down into lava tubes, often going right through to the floor below. Some of the roots are well over 10 m long and bundles up to 30 cm in diameter occur. An example of thick forest of Ohia tree roots is shown in Figure 6. Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings244

PAGE 247

The tree roots are an important source of nutrition and water for a host of cave-adapted species. While no professional biologist has studied this system as of this writing, we have sent photographs to biologist Fred Stone, who has identified various crickets, spiders, isopods, moths, and planthoppers. Another important source of the cave biology is the large number of goats that use the caves as night-time shelters. They supply nutrition by their excrements and the degradation of their dead bodies. The goats are generally found at lower elevations. At higher elevations, the most common mammal bones are Mouflon (a form of wild sheep), cattle, occasional horses, rats, and mongeese. A typical passage with an ungulate skull is shown in Figure 7. Well-preserved fossil bird bones are also sometimes found, especially in deep passages far from entrances, where they are protected from wheathering. Some of these may be from now-extinct species. A systematic study has not been done yet, however. One of the best-preserved examples is shown in Figure 8. This is most likely the head of one of the petrel varieties. Microbial mats are found throughout the system. None have been studied as of this writing. Finally, the deep entrances provide a haven for certain rare and endangered plants. The moister climate is a boon to the Figure 4. Ann Bosted in typical medium-sized passage with many mineral coatings. Photo by Peter and Ann Bosted. Figure 6. A thick forest of ohia tree roots. Photo by Peter and Ann Bosted. Figure 7. Typical passage with mouflon skull in the foreground. Photo by Peter and Ann Bosted. Figure 5. Delicate frostwork minerals on ceiling lava drips. Photo by Peter and Ann Bosted.Hapuu (tree fern). Large Koa trees can also be found in deep pukas where the steep sides have excluded cattle, which would otherwise eat and trample the young trees. An example of abundant native vegetation in a skylight puka is shown in Figure 9.5. ProspectsWith about 200 or more entrances to investigate, there is clearly a lot more work just to finish the exploration, mapping, and photo-documentation of the Keokeo system. Meanwhile, there are many opportunities for detailed studies of the geology, biology (both macro and micro), extinct birds, and archeology. Potential investigators would need to obtain a permit from the National Park Service or other land owners.AcknowledgmentsWe are extremely grateful to Tim Scheffler for archeological monitoring assistance during many more long field trips than originally anticipated. Thanks to the numerous cavers who helped with mapping and photography, especially Bob South. We thank the National Park Service for their support. Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings245

PAGE 248

ReferencesBauer I, 2011. Geologie, Petrographie und Pyroductgenese des Kahuku-Ranch-Gebiets, Big Island, Hawaii. MSc thesis, Inst. of Appl. Geosciences, Techn. Univ. Darmstadt, 139, unpublished.Sherrod E, David R, Sinton, JM, Watkins S, Brunt, KM, 2007. Geologic map of the State of Hawaii: U.S. Geological Survey Open-File Report 2007-1089, http://pubs.usgs.gov/of/2007/1089/ Kempe S, 2012. Volcanic rock caves. In: White W, Culver DC (Eds.), Encyclopedia of Caves, 2nded. Academic Press /Elsevier, Amsterdam, 865. Nani Kahuku Aina, LLC in Draft Environmental Impact Statement, PBR and Associates, Honolulu, 2011. http://oeqc.doh.hawaii.gov/Shared%20Documents/EA_and_EI S_Online_Library/Hawaii/2010s/2011-09-23-DEIS-KahukuVillage-Vol1.pdf Figure 8. Fossil bird head. Photo by Peter and Ann Bosted. Figure 9. An entrance skylight with native vegetation. Photo by Peter and Ann Bosted.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings246

PAGE 249

ORIGIN OF ROCK CITIES, PILLARS AND CLEFT-CONDUITS IN KAOLINITE-BONDED SANDSTONE: NEW INSIGHT FROM STUDY IN SANDSTONE QUARRY WHERE LANDFORMS RECENTLY EVOLVEJiri Bruthans1, Jan Soukup1, Jana Schweigstillova2, Jana Vaculikova1, Daniel Smutek3, Alan L. Mayo4, Lukas Falteisek1 1Faculty of Science, Charles University in Prague, Albertov 6, Praha 2, Czech Republic, bruthans@natur.cuni.cz2Institute of Rock Structure and Mechanics AS CR, v.v.i., V Holeovi kch 41, 182 09, Praha 8, Czech Republic,3Vodn zdroje Chrudim sro., U vodrny 137, Chrudim II, Czech Republic4Brigham Young University, Department of Geosciences, Provo, UT 84602, USA Various ideas exist on the origin of landforms like rock cities, pillars, clefts, rock shelters and lenticular hollows in kaoli nitebonded sandstone. Sandstone surfaces and processes were studied in St ele Quarry, the Czech Republic in Cretaceous marine quartz sandstone, where forms similar to landforms at natural exposures (clefts-conduits, lenticular hollows) are evolving at present time. The quarry offers a unique opportunity to characterize the erosion processes, which may form natural landforms prior to stabilization by case hardening. Based on measurements of relative erodibility, ambient and water-saturated tensile strength at natural and quarry exposures three distinct kinds of surfaces were distinguished: 1) Sandstone directly erodible by flowing water. 2) Sub-vertical fracture surfaces that are non-erodible and formed tectonically (slip faces of microfaults). 3) Case hardened surfaces that start to form after exposure. In favorable conditions case hardened surfaces become non-erodible and reach the full tensile strength in just 6 years. Flow in openings with a discharge 1 ml/s and hydraulic gradient > 0.05 exceeds the erosion threshold and initiates piping. In the first phase of conduit evolution, fast concentrated flow mobilizes erodible sandstone between sets of parallel fractures in the shallow phreatic zone. In second phase the conduit opening mainly expands vertically upward into the vadose zone by mass wasting of undercut sandstone slabs. Mass wasting is responsible for >90% of mobilized sandstone. Sides of the mature conduits are protected by non-erodible fracture surfaces. Natural landforms were probably formed rapidly by piping and possibly overland flow and fluidization during, or at, the end of the glacial periods when sandstone was not protected by case hardening yet. Erosion proceeded along highly fractured zones. The study does not support the existence of deep weathering along fractures proposed by some authors (etching).1. IntroductionThe weathering and erosion of sandstone result in the formation of interesting morphology features in many areas including the Colorado Plateau (United States) and the Bohemian Cretaceous Basin (Czech Republic, Germany) (Hrtel et al. 2007). The landscape of many areas in the northern Czech Republic, in Saxony, Germany, in Weald, SE England and elsewhere are developed in quartz sandstones bonded by kaolinite, whose tensile strength decreases strongly if sandstone is wetted (Bruthans et al. 2012b). The effect of wetting weakening is well known to local climbers who by regulation are prohibited from climbing in sandstone terrains for one or two days after rain events not to damage sandstone surfaces (HOROSVAZ 2011). Various landforms are known from these sandstones including caves, clefts separating individual sandstone pillars (castellated sandstones), rock overhangs (shelters) and also small-scale features like lenticular hollows and honeycombs (Hrtel et al. 2007; Fig. 5). Honeycombs are clearly evolving at present time, as they cover also few hundred years old quarry faces as well as sandstone blocks at cemeteries and walls of medieval castles (Hrtel et al. 2007). Based on radiocarbon and U-series dating of secondary carbonates in sandstone cave ceilings and archeological evidence, sandstone erosion in caves and sandstone overhangs ceased at least 8 kyr ago in Bohemian Paradise and wider surroundings of esk Lpa (Clek 2007; Bruthans et al. 2012a). Continuous peat bog sedimentation for the last 7.5 kyr in valleys directly below sandstone exposures in Adrpach region (60 km east of the studied locality) demonstrates that at least some sandstone exposures have not been eroded in this period (Kune and Jankovsk 2000; our study, unpublished). At present time the sandstone exposures and even sides of many overhangs are reinforced by case hardening, which makes sandstone exposures unerodible (Hrtel et al. 2007). Important question is how quickly case hardening is formed and what the properties of the sandstone were before the case hardening was formed. So far the evolution of sandstone landforms has been studied at natural exposures. Due to protection measures the studies were mostly limited to observation and documentation of landform shapes and taking small samples of rock crust for mineralogical analysis of salts and cements (e.g., Adamovi et. al. 2011). So far such studies failed to bring clear evidence for origin of landforms in clay-bonded sandstone except perhaps honeycombs. As a consequence, a wide variety of different ideas exist on processes responsible for evolution of landforms including deep weathering along fractures followed by the frost, salt and biogenic weathering (Clek 2010), dissolution of carbonate cement (Adamovi and Mikul 2011) and many others. Interestingly, some landforms (cleft-conduits, lenticular hollows) whose origin is discussed above are actively evolving at present time in St ele quarry (Fig. 5). In the quarry, Cretaceous Hrub Skla quartz sandstone is mined (for geotechnical and hydraulic parameters see Bruthans et al. 2012b). In this sandstone, a wide variety of natural forms Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings247

PAGE 250

occur in the so-called Bohemian Paradise (Hrtel et al. 2007). There are two strong advantages to study the landforms in the quarry: 1) All landforms are doomed to be destroyed by quarrying and therefore destructive methods can be fully applied; 2) Newly quarried surfaces are unprotected by case hardening, unlike natural surfaces. Possibility to observe presently evolving forms and potential to apply destructive techniques creates a unique opportunity to resolve the origin of landforms and study directly the erosion processes. Many fresh sandstone faces in the quarry are weak enough to be eroded by running water and rain, and the sandstone was even mined for decades by spraying a jet of pressurized water. The same sandstone, however, has to be mined by explosives when dry, and it has formed stable (up to 40 m high) vertical mining faces before the present safety regulations. The significant decrease in the strength of the wet sandstone compared to its dry state is known as wetting weakening (Lin et al. 2005). The aim of this study is to describe: 1) Sandstone surfaces in a quarry in terms of the erodibility, tensile strength and geomorphological functioning; 2) Erosion processes operating in the quarry; 3) Apply the findings from the quarry to understand the evolution of natural sandstone landforms. After hardening a tensile force was gradually increased perpendicular to sandstone surface until the sandstone beneath the epoxy failed. The maximum (pull off) force and effective area was measured and calculated as kPa. The ambient tensile strength (TSa) and saturated tensile strength (TSs) were measured (Bruthans et al. 2012b). The relative susceptibility to erosion by flowing water was measured by means of a water jet. A sprayer device equipped by a pressure gauge was used to create a jet of water 1 mm in diameter. The reservoir was pressurized to 180 kPa during all tests. The water jet was applied to tested points at the sandstone surface with a 10 cm-long stream of water for 5 seconds (Bruthans et al. 2012b). The depth of the water jet drill holes was measured by a caliper. The drilled depth is herein designated as the relative erodibility indicator (REI). The REI technique is based on a similar principle as the jet test developed by Hanson and Cook (2004).3. Results and discussion3.1. Characterization of conduits Up to several hundred meters long conduits developed in the quarry due to artificial water level decrease by 20 m (Fig. 5a). At several locations the underground conduits were mainly enlarging upward by mass wasting. At several locations the mass wasting broke through to the ground surface resulting in vertical shafts similar to natural clefts. A striking feature of the underground conduits is the fact that most of the sides of the passages do not display erosion features made by flowing water (Fig. 5a). Instead, conduit sides are smooth fracture surfaces. The measured stream gradient in major and minor conduits is ~0.01 and ~0.05, respectively with flow velocity ~40 cm/s (Fig. 1). Flow in the smallest channels transporting sand grains was investigated by studying the propagation of the leading edge of fluorescein dye. Small inflow trickles with flow rates as low as 1 ml/s had a maximum a velocity 8 cm/s in several cm-wide channels with gradients of 0.05.06. 3.2. Strength and erodibility of sandstone surfaces Surfaces measured by REI and TS included: 1) Fracture surfaces that are smooth and planar, 2) Case hardened surfaces that have a thin surface crust, 3) Water eroded surfaces that show traces of fluvial erosion, 4) Non-eroded surfaces that are a subset of fracture surfaces, which were in contact with flowing water but do not show traces of any erosion, and 5) other surfaces which include all other surfaces (mainly measurement at the quarry faces). More than 750 REI measurements were made, with drill hole depths ranging from <1 mm to 152 mm (Fig. 2). Ninety percent of the water eroded surfaces have REI >20mm and 93% of non-eroded surfaces have REI <20mm (Fig. 2). Therefore, the REI threshold for erodible sandstone and non-erodible sandstone is defined for the quarry sandstone as 20 mm. To estimate the percentage of erodible sandstone in the quarry, REI measurements were performed along a 240 m-long horizontal profile of the quarry face. On average 64% of Figure 1. Measured flow velocity in conduits compared to calculated flow velocity in sandstone pores as a function of channel gradient or hydraulic gradient. While piping is characterized by high flow velocity and low hydraulic gradient (conduit flow) the fluidization is characterized by steep hydraulic gradient and very low flow velocity (due to extreme friction in small pores). Modified from Bruthans et al. (2012b).2. MethodsThe geometry, locations and gradients of underground conduits were mapped by compass, precision clinometer, and a tape or a laser distance meter. Tracer tests in underground streams were performed with dissolved NaCl and fluorescent dye. Maximum flow velocity was measured by a tracer test (first arrival of dissolved tracer) and the leading edge of the breakthrough was photo-documented based on fluorescein dye. The tensile strength of surfaces was measured in the field. Metal plates were glued by epoxy to sandstone surfaces. Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings248

PAGE 251

sandstone in this profile is erodible (Bruthans et al., 2012b). REI measurements demonstrated that fracture surfaces are in many cases less erodible than the sandstone located several millimeters below the fracture surface. REI <20 mm shows 83% of fracture surfaces. The REI of 86% of case hardened surfaces <5 mm, which makes case hardened surfaces the least erodible surfaces in the quarry (Fig. 2). layer sandstone where most of the pores become filled by clay. Lichen hyphae were found in large numbers at case hardened surfaces (Fig. 5j). Sample of case hardened surface left for 5 days in hydrogen peroxide (30%) was disintegrated to loose sand, while control sample left in water was stable. As hydrogen peroxide decomposes organic matter, the hyphae organic fibers are probably responsible for stabilization of surfaces. Unlike fracture surfaces, case hardened surfaces start to develop after sandstone is exposed at ground surface. At the bottom of the quarry, parent surfaces had unlimited moisture supply and were directly exposed to sun radiation. The case hardened surfaces with firm surface crust (TSa 64 kPa) developed at places which were excavated just 6 years ago (Bruthans et al. 2012b). Figure 2. Histogram of REI. Values were rounded to tens. Category REI 100 contains also readings exceeding REI 100 mm. Modified from Bruthans et al. (2012b). Figure 3. Tensile strength of surfaces in the quarry and natural outcrop Apolena, both formed by the Hrub Skla sandstone.Surfaces of erodible sandstone have an ambient tensile strength (TSa) of 2.2 kPa (Fig. 3). The tensile strength of the sandstone decreases substantially when the sandstone is water saturated. The saturated tensile strength (TSs) of erodible sandstone is only < 0.6 kPa. Measurements suggest that the ambient tensile strength is at least about 2 orders of magnitude greater that the saturated tensile strength. Fracture surfaces and case hardened surfaces, which are not erodible, have a TSa of 27 and 64 kPa, respectively. The TSs of vertical fracture planes is between 0.9 and 38 kPa, which is less than fracture surface TSa. The TSs of fracture planes are nevertheless much greater than the those of erodible sandstone (Fig. 3). 3.3. Binding and microfabric of surfaces To explain the distinct difference in tensile strength and erodibility of sandstone surfaces in the quarry, the binding and sandstone microfabric were studied. Erodible sandstone shows a looser packing between grains compared to fracture surfaces. Neither silica nor calcite and iron-oxide cement were identified in any sample of erodible sandstone based on SEM and cathodo-luminescence inspection. Opal was not detected by FTIR in silt and clay separates from erodible sandstone. Only kaolinite was identified as binding agent in erodible sandstone (for details see Bruthans et al., 2012b). The decrease in tensile strength from relatively dry to the saturated state confirms that erodible sandstone is not bonded by permanent cement, but only by clay. Fracture surfaces are formed by zones of densely packed grains, which are several millimeters thick (Fig. 5h). Unerodible fracture surfaces were found at exposures as well as inside the conduits, tens of meters below the quarry surface. Fracture surfaces were thus non-erodible already prior to exposure at ground surface. Fracture surfaces correspond to slip faces or possibly deformation bands in sense of Antonellini et al. (1994). Case hardened surfaces are formed by a 0.5 mm thick 3.4. Erosion processes Field evidence proves the critical role of piping and mass wasting in conduit enlargement. Conduit flows were observed to erode by piping the most erodible sandstone surfaces at rates as high as several centimeters per minute. In one case a 3 m deep, 1 m wide, and 15 m long gully was cut into sandstone in less than 12 hours by water discharging from one spring (Bruthans et al. 2012b). Mass wasting of undercut sandstone blocks, along sub-vertical fractures, is the dominant mechanism in conduit enlargement because more than 90% of conduit volumes are excavated above the conduit stream surfaces. The potential role of fluidization in the enlargement of fracture openings is unclear. Fluidization may potentially contribute to erosion due to initial steep hydraulic gradients Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings249

PAGE 252

in the surroundings of new conduits. To test this possibility the critical hydraulic gradient for the onset of fluidization was measured in eight rectangular blocks of erodible sandstone. In two of the blocks, erosion due to fluidization occurred when the hydraulic gradient 0.7. In other blocks, no erosion was observed even under very steep hydraulic gradients (1.2.5). Lenticular hollows are relatively common in conduits and lake sides in the quarry (Fig. 5e). Active erosion was not observed under ambient conditions. In an attempt to better understand the process responsible for origin of lenticular hollows the overland flow and fluidization experiments were performed in the quarry. The experiments consisting of flowing water across erodible sandstone surfaces and via sandstone matrix in the quarry demonstrated that two different processes can create lenticular hollows: 1) Fluidization, and 2) Overland and/or pipe flow. 3.5.Evolution and propagation of conduits Incipient conduits probably propagate along bedding planes or subhorizontal cracks in several cm to few tens of cm thick sub-vertical slabs of sandstone. Conduits are formed by erosion of narrow sandstone slabs between sets of parallel subvertical fractures. Conduit width increases by mass wasting along parallel fractures (Fig. 4). 3.6. Origin of natural sandstone landforms To evaluate if the processes observed in St ele quarry are partially responsible for the sandstone landscape evolution observed elsewhere, exposures of the same sandstone were studied in the Apolena area 1.7 km from the quarry. In Apolena the sandstone forms pillars up to 30 m high, rock shelters, arches and lenticular hollows (Figs. 5 d, e). The pillars and parts of sandstone massif are separated by deep clefts. Outer faces of the sandstone outcrops are often defined by vertical fracture planes. Such landforms are a very common feature in sandstones in the Czech Republic and Saxony, Germany (Hrtel et al. 2007). At Apolena and other natural sandstone exposures (e.g., Adrpach), the features have similar appearance to those found in the quarry (Fig. 5). Some natural springs in the surroundings of the St ele quarry have feeding conduits of the same shape as that found in the quarry (Bruthans et al., 2012b). At Apolena fresh surfaces of sandstone boulders, which are the residual of pillar collapse, have similar values of TS and REI as fresh sandstone faces in St ele Quarry and are erodible. Case hardening in Apolena area is similar to the recent case hardening in St ele quarry. Hydraulic gradients of ~0.03 responsible for the onset of conduit development in the quarry surroundings (Bruthans et al. 2012b) could occur in the past under natural conditions. Observations made in the quarry are thus fully applicable to this natural sandstone outcrop. It may be possible to extrapolate the erosion processes observed in the St ele quarry to the landforms found in the Apolena area and elsewhere in the clay-bonded sandstone. Clefts were excavated probably by piping and consequent mass wasting of undercut sandstone slabs in erodible sandstone. Overland flow and fluidization may be considered too. Lateral enlargement of clefts by mass wasting of sandstone slabs was responsible for the formation of pillars and rock cities. Fracture planes were the only surfaces limiting erosion in erodible sandstone since case hardening was not yet formed. Unerodible parts of sandstone may have disintegrated by ice wedging. Rain water collected at ground surface was probably the major source of water feeding cleft conduits. The erosion was active during glacial periods, when sandstone was impermeable (except shallow subsurface in warm period of year) due to >30 m thick permafrost (k et al. 2012) and possibly also at glacial/interglacial transition. With the establishment of warmer climate and forest cover during the Holocene (Kune and Jankovsk 2000), the amount of water available for surface erosion and piping decreased considerably due to transpiration consumption of forest cover. Eroded surfaces were probably case hardened at this time, thus fossilizing the sandstone landforms into their present shapes. The postulated timing of erosion, formation of case hardening and consequent fossilization of landforms are in good accord with the results of dating studies (Clek 2007; Bruthans et al. 2012a, undisturbed peat bog sedimentation below sandstone exposures mentioned above).4. ConclusionsThree distinct kinds of sandstone surfaces can be defined based on field appearance, erodibility and tensile strength: A) Erodible sandstone is defined as sandstone with REI readings >20 mm based on field calibration. Erodible sandstone does not contain permanent cement. Sandstone is bonded by kaolinite, capillary forces and interlocking of sand grains. Erodible sandstone loses as much as 99% of its surface tensile strength when submerged below water. B) Sub-vertical fracture surfaces(slip faces of microfaults) are generally not erodible by piping or overland flow based on REI and field observations. Tensile strength is several times higher (TSa 27 kPa) compared to sandstone in the subsurface of a fracture plane. Unlike the erodible sandstone, fracture surfaces often keep considerable tensile Figure 4. Conceptual model of conduit propagation. A) Evolution in longitudinal vertical section; B) Horizontal plan of headward part of propagating conduit. Bedding planes provide a permeable path in a direction oblique to fractures; C) Evolution of conduit cross-section. Asterisk shows the original position of protoconduit. Modified from Bruthans et al. (2012b).Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings250

PAGE 253

strength if fully saturated. The non-erodible zone is several millimeters thick. Non-erodible fracture surfaces have clearly developed before the sandstone was exposed at ground surface and were recognized as a factor affecting shape of many sandstone landforms at natural outcrops. C) Case hardened surfaces cover most of the natural sandstone exposures, but also many exposures in the quarry, which are more than few years old. They evolve after the sandstone was exposed at ground surface, with a contribution of lichens. Case hardened surfaces consist of a 0.5 mm thick zone of sandstone where most of the pores become filled by kaolinite clay reinforced by lichen hyphae. Case hardened surface are not erodible by overland flow. Under optimum conditions (sun irradiation, moisture abundance), case hardening developed in the quarry within just 6 years reaching the same tensile strength as the case hardened surfaces at natural exposures (TS 23 kPa). The critical requisite for erosion of sandstone is the development of a zone of locally steep hydraulic gradient (~0.05), which resulted in an increase in flow velocity in small openings. The zone of steep gradient migrates headward in front of the propagating conduit. Openings with flow > 1 ml/s and hydraulic gradient >0.05 exceed the threshold for initiation of piping in erodible sandstone. The conduit evolution is initiated by piping of narrow slabs of erodible sandstone between sets of parallel fractures in shallow phreatic zone. In the second phase, the conduits mainly propagate vertically upward via the vadose zone by mass wasting of undercut sandstone slabs. Mass wasting is responsible for >90% of mobilized sandstone. No traces of erosion are left on fracture planes that form the conduit sides. Clefts in natural sandstone landscape may thus be developed by water erosion even if there are no traces of such erosion left. Based on physical modeling, the fluidization may occur in vadose zone at seepage face under a steep hydraulic gradient (>0.7). Based on the observations in the quarry, the following origin of various phenomena in clay-bonded sandstones is proposed. A) Clefts and eroded zones separating pillars and other rock masses in rock cities were initiated by piping and enlarged by mass wasting along subvertical fractures. Overland flow and fluidization erosion formed the undulating sandstone surface and lenticular hollows at pillars and other sandstone exposures. Erosion was active during glacial periods or at glacial/interglacial transitions. Frost weathering disintegrated firmer parts of sandstone. Landform evolution might be very fast (centuries). B) The case hardening started to form at sandstone exposures during the Holocene climate conditions, which decreased the erodibility below the threshold for natural flow and fossilized the forms into present shape except honeycombs, which are recently evolving. The postulated timing of erosion and formation of case hardening is in good accordance with published dating studies. The study does not support the existence of deep weathering along fractures (cf. Clek 2010). Sandstone is weak in massive blocks (forming pillars) as well as in fractured zones (forming clefts). In our opinion it is the mechanical weakness of narrow slabs of sandstones what enables the piping, triggers mass wasting and finally forms clefts.AcknowledgmentsMany thanks to Petr Miku, Luk Hronec, Iva K rkov, Vojt ch Stejskal, Michal Filippi, Ond ej Jger and Stanislav lechta for assistance with field measurements in 2009 and to the management of St ele Quarry for facilitating the research, and Ji Adamovi and Tom Lnczos for very valuable comments. The research was supported by grant projects GAUK 380511, GA R 13-28040S research plan MSM0021620855 and research plan AV0Z30460519.ReferencesAdamovi J, Mikul R, 2011. Origin of some ellipsoidal cavities by carbonate cement dissolution in the Jizera Formation sandstones, Koko n area. Geoscience Research Reports for 2010, 9. Czech Geological Survey. Adamovi J, Mikul R, Schweigstillov J, Bhmov V, 2011. Porosity changes induced by salt weathering of sandstones, Bohemian Cretaceous Basin, Czech Republic. Acta Geodynamica et Geomaterialia 8, 1, 29. Antonellini MA, Aydin A, Pollard DD, 1994. Microstructure of deformation bands in porous sandstones at Arches National Park, Utah. Journal of Structural Geology 16, 7, 941. Amsterdam. Bruthans J, Schweigstillov J, Jen P, Churkov Z, Bezdi ka P, 2012a. 14C and U series dating of speleothems in the Bohemian Paradise (Czech Republic): Retreat rates of sandstone cave walls and implications for cave origin. Acta Geodynamica et Geomaterialia 9 (1), 93. Bruthans J, Svetlik D, Soukup J, Schweigstillova J, Valek J, Mayo AL, 2012b. Fast evolving conduits in clay-bonded sandstone: Characterization, erosion processes and significance for origin of sandstone landforms. Geomorphology 177, 178. Clek V, 2007. Climate, microclimate and paleoclimate of sandstone areas of Central and Northern Bohemia (Czech Republic). In: Hrtel H, Clek V, Jackson A, Williams R. (Eds.), 2007. Sandstone Landscapes. Academia, Praha, 97. Clek V, 2010. Saxon-Bohemian Switzerland: sandstone rock cities and fascination in a romantic landscape. P. Migon (ed.) Geomorphological Landscapes of the World, 201. Springer. Hanson GJ, Cook KR, 2004. Apparatus, test procedures and analytical methods to measure soil erodibility in situ. Applied Engineering in Agriculture 20 (4), 455. Hrtel H, Clek V, Jackson A, Williams R. (Eds.), 2007 Sandstone Landscapes. Academia, Praha. HOROSVAZ, 2011. Pravidla sportovnho lezen na pskovcovch skalch v echch. Czech Mountaineering association. http://www.horosvaz.cz/index.php?cmd=page&type=2& article=35 Kune P., Jankovsk V., 2000. Outline of Late Glacial and Holocene vegetation in a landscape with strong geomorphologic gradients. Geolines 11, 11214. Lin ML, Jeng FS, Tsai LS, Huang TH, 2005. Wetting weakening of Tertiary sandstones microscopic mechanism. Environmental Geology 48, 265. k K, Richter DK, Filippi M, ivor R, Deininger M, Mangini A, Scholz D, 2012. Cryogenic cave carbonate A new tool for estimation of the Last Glacial permafrost depth. Climate of the Past Discussions 8, 2145.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings251

PAGE 254

Figure 5 a) Conduit developed by flowing groundwater in St ele quarry propagating mainly upwards along parallel sub-vertical fractures. Photo M. Filippi; b) Cleft limited by subvertical fractures in Apolena area; c) Fracture-guided opening in Adrpach nature reserve. Note the highest width at base and multiple fractures dying out upward; d) Lenticular hollows in Adrpach-Teplice nature reserve; e) Lenticular hollows developed at the side of a quarry lake (< 6 years old); f) Freezing groundwater seeping out of the lenticular hollows ( Adrpach); g) Fracture-guided opening in Adrpach area. Note remaining block in the ceiling; h) Cross-section via fracture surface (quarry); i) Inner part of the sandstone block is much easily erodible by flowing water than fracture surface (quarry); j) Lichen hyphae in a case hardened surface (quarry). Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings252

PAGE 255

ORIGINS OF THE KARSTIC SEDIMENT FILLING OF THE NORTHWESTERN PARIS BASIN CAVESS. Chdeville1, J. Rodet2, B. Laignel1, D. Todisco3, E. Dupuis1, G. Girot1, G. Hanin1 1Universit de Rouen, UMR 6143 M2C CNRS, SFR SCALE, 76821 Mont-Saint-Aignan, France, chedeville.stephane@gmail.com2CRNS UMR 6143 M2C, Universit de Rouen, SFR SCALE, 76821 Mont-Saint-Aignan, France3Universit de Rouen, UMR 6266 IDEES CNRS, SFR SCALE, 76821 Mont-Saint-Aignan, France Petites Dales, Villequier II and Orival are three privileged caves to study the sedimentary fillings in the chalk karst of the Northwestern Paris Basin (France). The present study is based on sedimentological characterization of the sediments from these caves using stratigraphy, grain-size and geochemical analysis. Observations and analysis point out a sedimentary variability both within and between sites. This may be explained by the variation of two main parameters: (i) the origin of karst sediments and (ii), their hydrodynamic behaviour in the karst system. The methodology used to analyse karst fillings allows identifying the origin of the components: (i) fillings made of surface formations (loess, sand deposits, or a mixture of them) and (ii), insoluble residue of chalk (clay with flints).1. IntroductionKarstificationprocessesoccurindifferentparts of the world on differentcarbonatedor notbedrocks(Rodet 1991; Willems 2000). These processescreate voidson the surface (e.g., sinkhole) andwithin thebedrock(i.e. karstic conduit) bychemical weathering and/ormechanical erosion(Mangin 1975). Depending onthe hydrodynamicsandsediment sources, these voids may present three kinds ofkarstic filling: (i) allochthonous(supply of sediments by mechanical erosion), (ii) autochthonous(insoluble residues from chemical alterationof the bedrock) or (iii), a mixtureof both (Laignel et al. 2004; Rodet et al. 2006).Thesedimentary fillings ofkarstare excellentrecordingsofhydrodynamic changesin thekarst (Losson and Corbonnois 2006). Their studiesallow understanding of thehydrosedimentary processesat the originof sedimentation,andsubsequently todetermine past climatic conditions (Falgures et al. 2001; Quinif 2006; Rodet et al. 2006; Hanin et al. 2011). Karstification phenomena are well developed in the chalk of the Northwestern Europe (NWE), particularly in the Northwestern Paris Basin (NWPB) in France, North-East Wallonia in Belgium and South-East England. In these areas, karstic introductions, networks and sedimentary fillings are most often similar (Rodet 1991; Laignel et al. 2004; Quinif 2006; Van Den Eeckhaut et al. 2007; Willems et al. 2007a; Cooper et al. 2011; Farrant and Smart 2011; Hanin et al. 2011). In limestone bedrocks, the weathered surface and karstification are linked (Rodet 1991; Laignel 2003; Laignel et al. 2004). The karstic network could be considered as zones of alteration, storage volumes of eroded material and also of transit. Whereas a lot of studies on karstic fillings are focused on archeology (Madeyska 2002; Cyrek et al. 2010) or climate change (Quinif 2006), for example, very few deal with the characterization and comparison of several karstic fillings (Rodet et al. 2009). The objective of the present study is then to propose a comparison between several karst sedimentary fillings in the NWPB. Three sites with karstic filling have been studied in the NWPB (Seine-Maritime, France): Villequier II, Orival and Petites Dales Caves (Fig. 1). The karstic sediment fillings of thesesites have been characterized with stratigraphical and sedimentological analysis in order to determine the origin of the sediments. After defining the origin of sediment sources, a comparison is drawn between the obtained results with those from Belgium (Quinif 2006; Willems et al. 2007a) and England (Farrant and Smart 2011).2. Context and methodologyIn the NWPB, caves and karst conduits are found in Cenomanian to Campanian chalk formations. The Upper Cretaceous chalks are covered by surface formations including from bottom to top: (i) PlioceneQuaternary clay with flints (CWF), produced by atmospheric weathering of chalk (Quesnel et al. 2000; Laignel et al. 2002), (ii) Tertiary detrital deposits occurring as pockets at the top of CWF (Auffret et al. 1975), and (iii) Quaternary loess (Lautridou 1985). The three studied sites are located in the NWPB, in the North of the Haute-Normandie region (Fig. 1). The Upper Cretaceous chalk formations of this region are subject to current or fossil karstification and voids may be filled by sediments. The three selected sitescan be defined as undercovered karsts, covered with unkarstifiable/insoluble layers, which not prevent the limestone to be weathered Figure 1. Location map of caves.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings253

PAGE 256

(Salomon et al. 1995; Salomon 2006). According to classification of Nicod (1992), under-covered karsts may have alterite coverage. 2.1. Presentation of study sites The studied sites are karst systems without current flooded network. Petites Dales site (Fig. 2c) The Petites Dales Cave entrance is open in the Petites Dalles non drained valley, in the town of Saint-Martin-auxBuneaux. This karst is very close to the sea-coast, at 1,200m from theChannel border. The current access to the cavity is through the working face of an open chalk quarry into the NE valley slope (N49.8 E000.0; Z +30 m a.s.l). The cavity does not have perennial flow; only re-flooding in the lower parts was recorded in 1995 and 2000. The gallery develops in the coniacian chalk with flint (Rodet and Viard 1996; Laignel et al. 2004). The Petites Dales Cave, described by Rodet et al. (2007), consists in a main gallery, recognized over 710 m of development (Viard and Rodet 2011), with 10 m height (according to a core) and a width of 2 to 5 m (maximum width in the karst entrance). It is the most important cave of the Seine Maritime department. The main axis shows a succession of rectilinear and sinuous sections. In this axis, three secondary drains join the main one. The drains morphology gives evidence of an evolution from an anastomosed network which was reduced to a main collector by concentration of drainage. According to the spatial distribution of surperficial deposits in the Western Paris Basin (Laignel 1997, 2003), the area include SaintEustache and Lozere sands, but also CWF (6 m) and lss (>5 m). 2.2. Methodology The initial contact with the sedimentary fillings was made with section surveys representative of the sites. Both the color and texture of the sediments were described. The color was determined from the Munsell soil color chart. Units were defined integrating relatively homogeneous sedimentaryfacies in terms of color, components and grain size.Sampling was then performed for grain-size and geochemical analysis. Grain sizes were performed by using the laser particle size analyzer LS230 BECKMAN-COULTER in the Laboratory of Continental and Coastal Morphodynamics (M2C) of the University of Caen. The results are expressed as percentages of sand (>63 m), silt (63 m) and clay (<4 m). Geochemical analyses were performed by the SRAM laboratory at CRPG-CNRS (Center of Petrographic and Geochemical Research), at Nancy. Major elements were measured by emission spectrometry (ICP-AES) Jobin Yvon JY-70 and absorption spectrometry (ICP-MS) Perkin-Elmer ELAN 5000 and 6000. Three major elements were used for the present study (SiO2, Al2O3, Fe2O3. They are given in mass percent. Flow velocities were determined with Hjulstrm diagram using the 95 percentil of grain size samples.3. Lithological characterization of the karst sedimentary fillings3.1. Observations of the karstic sediment fillings In Villequier II Cave, a section of 67 cm was survey in the middle of the gallery. The filling of the section VIIA has a Figure 2. Map of the studied caves: Villequier II, Orival and Petites Dales.Villequier II site (Fig. 2a) The Villequier II gallery is located in Villequier (N49.2 E 000.9; Z +8 m a.s.l). This gallery of 34.5 m length, develops in the white chalk of Upper Coniacian at the base of a 30 m high cliff. According to the spatial distribution of superficial deposits in the Western Paris Basin (Laignel 1997, 2003), the area include SaintEustache and Lozere sands, but also CWF (2 m) and lss (<0.5 m). Orival RF8 site (Fig. 2b) The Roche Foulon karst system develops in senonian chalk cliffs on the left bank of the Seine River, in the town of Orival (N49.7 E001.0; Z +75 m a.s.l). These delimit the concave side of the Elbeuf-Oissel meander, downstream of the confluence of the Essarts/La Londe Valley (Rodet 1991). Walking by the GR2, regionalhiking trail, an abandoned quarry and rock-shelter houses are observed and alsomany karst conduits, some of which revealing a sedimentary filling. The site called RF8 has been the subject of a sedimentary study(Rodet 1991). The RF8 section was taken over 1.36 m. According to the spatial distribution of superficial deposits in the Western Paris Basin (Laignel, 1997, 2003), the area include the SaintEustache, Lozere and Ypresian sands but also CWF (2m) and lss (<0.5 m). Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings254

PAGE 257

vertical organization with alternating beds of variable thickness, in the order of 1 mm to 14 cm. The fill is dominated by pale beige sands 10YR/8/3 (VIIA2 and VIIA5) to beige 10YR/7/3, which can be a little more silty (VIIA3). Stratifications are observed that can be crossed (VIIA3) or slightly oblique to sigmoidal (VIIA6). Seven units were identified on the section VIIA. In these units, three main kinds of sediments are distinguished (Table 1). In the Petites Dales cave, three sections were surveyed in the middle of the gallery. The height of the A, B and C sections was measured respectively at 160 cm and 126.5cm and 217 cm. The sediments of the three sections are vertically organized in beds of different thicknesses (0.1 to 20 cm) with a rhythmicity in millimetre to centimetre scale. The A and B sections consist of 14 units each and 13 have been identified in the C section. A similar filling was observed between A and B sections and a stratigraphic continuity can be underlined with the C section thanks to a benchmark of calcite. On A, B and C sections, three main kinds of sediments are described (Table 1). In addition, a coarse sand unit is also only observed at the base of C section. In the Orival site, the RF8 section, of 138 cm high, was surveyed. The filling has a vertical organization of alternating beds with variable thickness: in the order of 1mm to 22 cm. Fifteen units were found in the very heterogeneous filling. The section includes silty beds (e.g., RF8-8, RF8-6) and heterogeneousmedium to coarsesand, more or less clayey, sometimes with beds ofroundedquartz (RF8-15). According to Laignel (2003), such a sediment is characteristic of Lozere sand facies. In some units, pieces of flint can be found in a silty matrix (RF8-10, RF8-8 in particular). Three kinds of sediment were identified in this section (Table 1) with the presence of reddish Lozere sands facies in the filling (RF8-13).Table 1. Lithological characterization of karstic fillings and origin of sediment sources. Main source of sediment in the filling in bold.3.2. Grain size distribution and geochemical characterization of the karstic sediment fillings In the Villequier II site, the main grain size of the filling is sandy to sandy silt (Table 1). We have a fairly significant heterogeneity in the filling: values are rather different in Table 1. The clay content could be described as background noise with a ratio between 0.64% to 6.42%. The particle size confirms the observations with the exception of two samples, VIIA3 is really sandy silt rather than silty sand and VIIA5 is silty sand instead of sandy. Geochemistry of the filling is very siliceous: the values of SiO2are above 94% for all samples. The contents of Fe2O3and Al2O3are very low and do not exceed 4% and 1.8% respectively. The values are very narrow, indicating a homogeneous geochemical composition (Table 1). The coarser the sediment, the higher percentage in SiO2. Samples from the A section of Petites Dales are silty with values greater than 75% silt, and more than 80% of silt to B section (Table 1). The grain-size distribution variations are weak between samples. Observations and grain-size analyze point out a homogeneity between the A and B sections. Geochemical characteristics of sediments from the B section reveal that these are highly siliceous with values above 79% for all samples. The Fe2O3contents begin to be Site VILLEQUIERII Pale beige sands Pale beige silty sands Brown sandy silts Pale beige silts Beige silts A1, A2, A5, A6, A8, A9, A11, B1, B4, B6, B7, B9, B11, B14 A4, A7, A10, B3, B5, B8, B13 RF8-15 RF8-13 RF8-12, RF8-7, RF8-5, RF8-3, RF8-2 RF8-14, RF8-11, RF8-10, RF8-9, RF8-8, RF8-6, RF8-4, RF8-1 A12, B10, B12 VIIA3, VIIA4 VIIA1, VIIA5 VIIA2, VIIA6, VIIA7 Brown clayey silts Lozere sands facies Sands to silty sands Silts PETITESDALESORIVALKind of sediment Samples Clay (%) (0 m) 0.6 to 0.91.9.23.1.40 0.40.0.71.0.31.0.54.6.9 Silt (%) (4 m) 6.9.426.2.250.7.476.5.181.1.886.5.712.5.911.8.849.5.2 Sand (%) (63 m) 86.8.565.8.633.5.96.9.50.1.90.31.786.0.249.7.31.7.5 SiO2(%)97.0.895.7.094.0.990.6.285.0.979.0.783.4.888.7.576.3.9 Fe2O3(%)1.4.92.8.03.6.26.5.07.3.513.0.99.24.3.97.4.7 Al2O3(%)0.8.11.2.31.6.82.3.42.7.45.3.17.0.42.9.23.3.2 Flow velocity sedimentation (cm/sec) Sediment origin Surrounding surface formations 1.7.50.9.50.71.0.60.9.10.4.59.15.2.30.3.2 SaintEustache sandsIndeterminate sands SaintEustache sand Indeterminate sand / lss mixtureLss, SaintEustache sands Lss Lss, weathered chalk Lozere SandsUndetermined sands lssLss undetermined sand, claywith-flints Saint-Eustache sand, lozere sand, lss, clay-with-flint, chalk Lss, lozere sand, Saint-Eustache sand, clay-with-flints, chalk Lozere sand, Saint-Eustache sand,Ypresian sand, clay-with-flints, lss, chalk Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings255

PAGE 258

significant for samples B13, B3 and B8 with percentages of 13%. For Al2O3, levels are low and do not exceed 6%. Such values also likely indicate a relative homogeneity in the filling. For each kind of sediment, a slight geochemistry difference appears. Once again, the coarser the sediment, the higher the percentage in SiO2. The Orival sedimentary filling is heterogeneous with grain size ranging from sandy to silty (Table 1). The clay percentage is higher than in the two other sites, even if clay is less than 10% for most samples. In addition, four samples have percentage between 10 and 25%. The grain size confirms the observations with the exception of two samples which are really silty clay instead of silt (RF8-4, RF8-8). The filling is very siliceous, with SiO2 values above 76% for all samples. The analysed section reveals arelative homogeneitybut values are a little morescatteredthan Petites Dales andVillequier II (Table 1). The contents of Fe2O3and Al2O3are most often below 10% (except for RF8-08, RF8-10, RF8-06 respectively 15.7%, 14.2% and 11.7% in Al2O3). For each kind of described sediment, a slight geochemistry difference is again observed. The coarser the sediment, the higher the percentage in SiO2, with the exception of the sediment with Lozere sand facies. The latter have high levels of Fe2O3and Al2O3(respectively 6.96% to 7.42% and 9.17% to 9.24%). 3.3. Lithological synthesis of the karstic sediment fillings All the sediments from the studied karst fillings are dominantly siliceous, fine, mostly silty sediments, but can also be sandy and poor in clay (Table 1). Each filling has a relatively characteristic grain size: sediments of Petites Dales are mainly silty whereas those of Villequier II are primarily sandy. The Orival site presents the most heterogeneous filling, with the most variable size including sandy to silty clay samples. For each site, three kind of sediments are found, each of them havingits own grain size and geochemistry properties. A similarity in the sedimentary organization is noticed between the three studied karst fillings, with variable thickness of horizontal beds as well as rhythmicity. Nevertheless, there is an inter-site variability based on the nature and origin of sedimentary material.4. Sources of the karst sedimentsAccording to the geological context of the three sites described above, several sediment sources are possible: (i) lss, CWF and residual Tertiary sandy deposits (i.e. Lozere, Saint Eustache and indeterminate sands) through mechanical erosion, and (ii) insoluble residue of chalk through chemical weathering of the bedrock. Once deposited in karst, the insoluble residue of chalk is considered autochtonous (endogenous) deposits, whereas lss deposits, CWF and Tertiary sandy deposits are allochtonous (exogenous). The terms endogenous and exogenous correspond to the Ford and Cullingford English classification (1976) in Ek and Quinif (1988). In order to determine the likely sources, of each kind of karstic sediments, geochemical data have been used and compared (karstic fillings, surface formations, chalk; Fig.5), in addition to field observations and grain size data. For each site, a summary of whatsurface deposits could be at the origin of the filling was made (Table 1). The results given by the geochemistry (Fig. 3) are in the penultimate line, the main source is in bold. It appearsthat loessis the main sourceoffillings, but in varying proportions, explaining a relative homogeneity offillings. Nevertheless, thecharacterizationshowsboth anintra and inter-site variability. The latter is partly explained bythe presenceof severalsediment sourcesin addition to loess,such as the Saint-Eustache,Lozere, and indeterminatesands, but also possible in situ alterationofchalk, even if invery small quantities(on a singlesite). The surface deposits are constitutive of the karst fillings, but in different proportion. For thePetitesDales, the filling mainly consists ofthreekinds ofsurface deposits whereas fiveare present in thewatershed.Loess is found in each site, but in varying proportions, in relation to its large coverage of the NWPB (Laignel 2003). By contrast, the Tertiarysandsareonly present aspockets, which are covered by loess. Saint-Eustache sandsare present inthe watersheds of thethreesites, however they are presentin only twofillings(VillequierIIand PetitesDales).The same observationis made forthe Lozere sands, present in all threewatersheds, but only foundinthe Orival filling. Two conditions may explain the presence of one type of Tertiarty sands, i.e. a pocket that is intersected by the karstic network, in addition to specifichydrodynamic conditionsallowing erosion,transport and depositionofTertiary sandsin karst conduit. CWF are made offlintpacked ina clayey silty sandmatrix, whose proportionsof the different constituentsvary from onesite to another, covering a very large part ofthe NWPB (Laignel 1997; Laignel et al. 1998, 2002; Laignel 2003). Such components are however absent in the fillings, even if solution pipes are observed, forming theinput karst which cut the output karst network. Because of its nature, CWF is very cohesive then difficult to be eroded (Laignel 1997; Quesnel et al. 2000). If CWF drown downward solution pipe into the karstic network, they tend to stay in situ. Only the most erodible particles, such as lss (Le Bissonnais 1988) and sands (if a pocket is connected with a solution pipe) will undergo transport and deposition in the endokarst. We agree with the hypothesis of Laignel et al. (2004) to explain the absence of CWF in significant proportions in Figure 3. Ternary plot of the geochemistry of the potential sedimentary sources and of Villequier II, Petites Dales and Orival karstic sedimentary fillings.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings256

PAGE 259

the fillings. Therefore, our results do not reach those of Sweeting (1973), suggesting that clay particles come from erosion of CWF. Our studies are consistent with those of Reams (1968) showing that clay have two possible origins: the in situ alteration of the chalk in the endokarst and/or arrivals of surface sediments by mechanical erosion. To summarize, the most erodible sediments are primarily deployed in the karst network. Different sediment origins partly explain the variability between each site. Within a single site, sediment origin is probably not the only cause of the variability of the filling. Another parameter influences intrakarstic sedimentation, especially hydrodynamic conditions.5. Discussions and conclusionsKarsts of NWPB present introduction forms as sinkholes (Laignel et al. 1997; Quesnel et al. 2000) and current or ancient restitution forms, as springs or caves (Rodet 1991; Laignel et al. 2004; Hanin et al. 2011). Ancient karst systems could be filled depending on the hydrodynamic conditions and sediment sources. Sedimentary fillings characterization is essential to understand sediment dynamics in endokarst (Coquerel et al. 1993; Laignel et al. 2004; Hanin et al. 2011) and also to determine the paleoenvironmental conditions in a given area. Karst fillings from NWPB are made of surrounding superficial formations, from clayey silt to coarse sandy material, according to the sediment sources and the hydrodynamics in karst (Laignel et al. 2004). Studies in the North-East Wallonia (Belgium) have shown the same kind of karst introduction (solution pipes) associated with filled caves (Van Den Eeckhaut et al. 2007; Willems et al. 2007a, 2007b; Rodet et al. 2009). Fillings mainly consist in silty to sandy particles as examplified by the Montagne Saint Pierre (Willems et al. 2007a). The latter study has been shown that the filling is primarily made of sands and loess (surrounding surface formations) with also slight proportion of residual material from chalk alteration. South-East of England has similar forms of introduction as NWPB and North-East Wallonia like solution pipes and cave fillings with similar size particles (Cooper et al. 2011). Farrant and Smart (2011) have also shown that filling is partly composed from surrounding surface formations with also residuals from chalk alteration. To conclude, a sedimentological characterization of karst fillings allows determinating sediment sources. By comparing the sediment fills and the surface formations several sources were identified in studied caves: surface formations (Tertiary sand, clay-with-flints, and lss) and insoluble of residual alteration of chalk bedrock. A similarity is apparent between karst of the NWPB and that of NWE when considering (i) karst morphology, (ii) sediment fillings, and also (iii) particle sizes of the allochtonous material. One of the future perspectives is notably to apply GIS to the study of karst, e.g., in integrating data concerning surrounding surface formations and to draw comparisons between caves from different regions of the NWE. Such study should also be associated with large sedimentological analyses and when possible, accurate dating of the endokarstic deposits.AcknowledgementWe wish thank the members of CNEK (Centre Normand dtude du Karst http://www.cnek.org/) for their outstanding work at the Petites Dales cave. More than twenty years of cave excavation until today for the access to the site. Thank for keeping the history of works in referenced pictures archives. Thanks to the SCALE and the Upper Normandy Council for the study grant allocation.ReferencesAuffret JP, Bignot G, Blondeau A, Cavelier C, Pomerol C, 1975. Geologie du Bassin Tertiaire de la Manche Orientale au Large du Pays de Caux [and Discussion], Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences, 279(1288), 169. Cooper AH, Farrant AR, Price SJ, 2011. The use of karst geomorphology for planning, hazard avoidance and development in Great Britain. Geomorphology, 144, 118. Coquerel G, Lefebvre D, Rodet J, Staigre J-C, 1993. La grotte du Funiculaire. Splogense et tude dun remplissage ferromanganique. Karstologia, 22(2), 35. Cyrek K, Socha P, Stefaniak K, Madeyska T, MiroslawGrabowska J, Sudol M, Czyzewski L, 2010. Palaeolithic of Bisnik Cave (Southern Poland) within the environmental background, Quaternary International, 220(1), 5. Ek C, Quinif Y, 1988. Les sdiments dtritiques des grottes: aperu et synthtique. Annales de la Socit Gologique de Belgique, 111, 1. Falgures C, Bahain J-J, Yokoyama Y, Bischoff JL, Arsuaga JL, De Castro JMB, Carbonell E, Dolo J-M, 2001. Datation par RPE et U-Th des sites plistocnes dAtapuerca: Sima de los Huesos, Trinchera Dolina et Trinchera Galeria. Bilan gochronologique. LAnthopologie, 105, 71. Farrant AR, Smart PL, 2011. Role of sediment in speleogenesis; sedimentation and paragenesis. Geomorphology, 134, 79. Ford TD, Cullingford CHD. 1976. The science of speleology. London Academic Press. Hanin G, Laignel B, Massei N, Hauchard E, Ladhui V, Chdeville S, 2011. Hydrological variations and sediment transfer in a karst system (Radicatel springs, Upper Normandy, France) controlled by climate fluctuations. Proceedings of Hydrology & Hydrogeology of the Karst, 9th Conference on Limestone Hydrogeology, Besanon (France), 221. Laignel B, 1997. Les altrites silex de louest du bassin de Paris: caractrisation lithologique, gense et utilisation potentielle comme granulat., Thse de Doctorat, Universit de Rouen, dition BRGM, Orlans, France. Laignel B, 2003. Caractrisation et dynamique rosive de systmes gomorphologiques continentaux sur substrat crayeux. Mmoire de HDR., Universit de Rouen. Laignel B, Dupuis E, Rodet J, Lacroix M, Massei N, 2004. An example of sedimentary filing in the chalk karst of the Western Paris Basin: characterization, origins and hydrosedimentary behaviour. Zeitschrift fr Geomorphologie, 48(2), 219. Laignel B, Quesnel F, Meyer R, 2002. Classification and origin of the clay with flints of the Western Paris Basin (France). Zeitschrift fr Geomorphologie, 46(1), 69.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings257

PAGE 260

Laignel B, Quesnel F, Meyer R, Macaire J-J, 1998. Relations quantitatives entre les craies silex et les formations rsiduelles silex de louest du Bassin de Paris, Geodinamica Acta, 11(4), 171. Laignel B, Quesnel F, Spencer C, 1997. Une nouvelle ressource en granulats: les formations rsiduelles silex, Gochronique, 63, 9. Lautridou J-P, 1985. Le cycle priglaciaire plistocne en Europe du Nord-Ouest et plus particulirement en Normandie. Thse de Doctorat dEtat, Universit de Caen. Le Bissonnais Y, 1988. Analyse des mcanismes de dsagrgation et de la mobilisation des particules de terre sous laction des pluies. Thse de Doctorat es Sciences, Ecole Doctorale I.N.R.A., Universit dOrlans. Losson B, Corbonnois J, 2006. Les modes de sdimentation dtritique: nouvelle mthode de dtermination applique des remplissages endokarstiques. Geologica Belgica, 9(3), 257. Madeyska T, 2002. Evidence of climatic variations in loess and cave palaeolithic sites of Southern Poland and Western Ukraine. Quaternary International, 91(1), 65. Mangin A, 1975. Contribution ltude hydrodynamique des aquifres karstiques. Thse de Docteur en Sciences, Universit de Dijon 1975; Annales de Spelologie, 1974, 29, 3, 283; 1974, 29, 4, 495; 1975, 30, 1, 21. Nicod J, 1992. Les karsts sous couverture (sableuse, argileuse et/ou dtritique) en France, daprs des travaux rcents. Cuadernos de Seccion, Historia, Donastia: Eusko Ikaskuntza, 20, 165. Quesnel F, Laignel B, Bourdillon C, Meyer R, 2000. Les altrites silex de Seine-Maritime (France): typologie, chronologie et godynamique. Bulletin dInformation des Gologues du Bassin de Paris, 37(1), 17. Quinif Y, 2006. Complex stratigraphic sequences in Belgian caves correlation with climatic changes during the middle, the upper Pleistocene and the Holocene. Geologica Belgica, 9(3), 231. Reams MW, 1968. Cave sediments and the geomorphic history of the Ozarks. Ph.D. thesis, Washington University, St Louis, Missouri. Rodet J, 1991. Les karsts de la craie: tude comparative. Thse de Doctorat dEtat, Universit de Paris IV-Sorbonne. Rodet J, Viard J-P (eds.), 1996. La grotte des Petites Dales, SploDrack. Rodet J, Laignel B, Brocard G, Dupuis E, Massei N, Viard J-P, 2006. Contribution of a sedimentary study to the concept of karstic evolution of a chalk cave in the werstern Paris Basin (Normandy, France), Geologica Belgica, 9(3), 287. Rodet J, Viard JP, Poudras J, 2007. la dcouverte de la grotte des PetitesDales (Saint Martin aux Buneaux, Seine Maritime, France), Splo-Tract. Rodet J, Willems L, Brown J, Ogier-Halim S, Bourdin M, Viard J-P, 2009. Morphodynamic incidences of the trepanning of the endokarst by solution pipes. Examples of chalk caves in Western Europe (France and Belgium). Proceedings of the 15th International Congress of Speleology, Kerrville (Texas, USA), 19-26 july 2009, UIS, vol. 3, contributed papers: 1657. Salomon JN, 2006.Prcis de karstologie, 2me ed., Presses universitaires de Bordeaux, Collection Scienteren. Salomon J-N, Pomel S, Nicod J, 1995. Lvolution des cryptokarsts: comparaison entre le Prigord-Quercy (France) et le Franken Alb (Allemagne). Zeitschrift fr Geomorphologie, 39(4), 381. Sweeting MM, 1973. Karst landforms. Mc Millan, London. Van Den Eeckhaut M, Poesen J, Dusar M, Martens V, Duchateau P, 2007. Sinkhole formation above underground limestone quarries: A case study in South Limburg (Belgium). Geomorphology, 91(1), 19. Viard JP, Rodet J, 2011. Seine-Maritime: Grotte des Petites Dales, Saint-Martin-aux-Buneaux. Spelunca, 122(5), 0242. Willems L, 2000. Phnomnes karstiques en roches silicates non carbonates. Cas des grs, des micaschistes, des gneiss et des granites en Afrique sahlienne et quatoriale. Thse de Doctorat, Universit de Lige (Belgique). Willems L, Rodet J, Ek C, Dusar M, Lagrou D, Fournier M, Laignel B, Pouclet A, 2007a. Karsts des craies et calcarnites de la Montagne Saint-Pierre (Basse Meuse ligeoise). Bulletin des Chercheurs de la Wallonie, 46. Willems L, Rodet J, Fournier M, Laignel B, Dusar M, Lagrou D, Pouclet A, Massei N, Dussart-Baptista L, Compre P, 2007b. Polyphase karst system in Cretaceous chalk and calcarenite of the Belgian-Dutch border. Zeitschrift fr Geomorphologie, 51(3), 361.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings258

PAGE 261

SANDSTONE CAVES IN THE SYDNEY BASIN: A REVIEW OF THEIR CULTURAL AND NATURAL HERITAGEJohn R. Dunkley 5 Coleman Street, Pearce ACT 2607, Australia, jrdunkley@gmail.com Small caves are found throughout the onshore Sydney Basin, but an inventory of over 1,100 has only recently been compiled for the database of the Australian Speleological Federation Inc. Aboriginal sites have been studied for a century, but only a few academic studies of the development of the caves and karst features have been highlighted in recent research findings. Most caves are clustered low in the landscape, close to watercourses and to urban areas, and exhibit a variety of distinct processes: marine action, piping, block-gliding and other forms of mass movement, waterfall retreat, lateral meandering and deep canyon incision. In total they attract large numbers of visitors. They are, nevertheless, largely underappreciated and underestimated in cave and karst literature and by heritage managers, as well as by recreational cavers. On balance, the caves are of importance mostly for their association with human activity. However some contain unusual features such as silicate and other deposits worthy of research and proper management.1. People and LandscapeCaves are scattered widely throughout the onshore 36,000km of the Sydney Basin in some of the planets most spectacular landscapes, including World Heritage Areas which, however, gained that status primarily for their natural biological values. Most are small but over 1,100 are recorded as greater than 3m in length in the database of the Australian Speleological Federation Inc., the longest being 263 m (Dunkley, in press). Their density varies considerably, from less than one per 100 km, up to 3 per km. Most are clustered low in the landscape, beside or on mid-slopes above watercourses, some in the middle of the Sydney metropolitan area. The higher, usually drier sandstone ridges have few caves. Sandstone covers over 30% of the Basin and most of that is in the quarter of the Basin that is almost uninhabited national park and declared wilderness. Yet about 800 caves lie within at most a 30-minute drive and 20 minutes walking distance, of the population (nearly 6 million people) in the Greater Sydney urban region. Indeed many are in valleys within the metropolitan area. Although mostly small, in total the caves may attract more visitors, albeit fleetingly and only occasionally as a specific destination, than all the well-known limestone caves in the state combined. Aboriginal sites have been studied for a century, and although a few academic studies of karst and caves have occurred (e.g., Johnson 1974; Wray 1993, 1995; Young et al. 2009), these features are consistently underappreciated and underestimated in cave and karst literature and by natural heritage managers, as well as by recreational cavers, the last perhaps because their speleological pedigree is not recognised. The inventory includes about 114 sea caves, 20+ block-gliding caves, 40+ in narrow, deep and often dark canyons, 30 with unusual iron and silica concretions, and 2 stream caves. In addition, at least 200 contain evidence of Aboriginal occupation; 45 have white historical associations; 25 were used for a period as homes, mainly by destitute people in the last century; 23 are camping caves used by bushwalkers; 20 are connected with bushranger legends; 20 are used for rock climbing and bouldering; 14 inspired poetry, music or art, and 10 or 20 contain glow worm populations. 1.1. Aboriginal significance of caves The oldest of about 200 Aboriginal occupation cave sites dates back 22,000 years, and a further 100+ have been identified which, by location, size and aspect are potential aboriginal sites. For a century after white settlement, passing scientists just observed and described Aboriginal people and their customs, taking little interest in their cave sites and less in prehistory. Nearly a century passed before serious studies were undertaken of cave art sites and properly documented and scientifically executed excavations were undertaken. Even then, much of the work was undertaken by professional polymaths from other fields such as geology (Edgeworth David), palaeontology (Robert Etheridge) and surveying (R.H. Matthews), the last of whom apparently studied 50 caves and 70 rock art sites. In recent years Australian archaeology has been revolutionised by many more qualified archaeologists, and intensive studies have been undertaken in restricted areas in military reserves and water catchment areas, aiding site conservation. Attenbrow (2002) provided a comprehensive review. Important caves continue to be identified in remote areas of Wollemi and other national parks and there are undoubtedly many more awaiting rediscovery.Figure 1. Characteristic landscape of upper Blue Mountains. Caves typically occur near breaks of slope, mainly near the edge of the plateau, often in canyons. Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings259

PAGE 262

1.2. European perspectives and perceptions First Fleet settlers recorded Aboriginal life in caves around Sydney Harbour from the day they arrived in January 1788. A few caves gained historical significance when recorded during the first attempts to cross the Blue Mountains, in one case settling a long debate about an explorers route. About 20 caves are named after known or alleged early outlaws (called bushrangers in Australian legend), or in connection with other illegal activity such as stills and stolen goods. Then for many years the sandstone around Sydney became country to be hurried across rather than settled, and for nearly a century caves were used to house military guards, road gangs, railway and other itinerant builders and workers. Partly as the result of gold rushes, by the 1880s Australia was a wealthy country and railways began to deliver tourists to these wild landscapes. Over the next 50 years wealthy landowners, hotel entrepreneurs and local councils constructed walking tracks totalling 300 km in length to pleasant sights (Smith 1999), including numerous small caves which in turn acquired romantic or fanciful names, and which appeared on postcards and souvenir books. The early reservation of sites such as Mermaids Cave at Blackheath in 1882 began a period during which, eventually, nearly all sandstone caves in the region would remain in public ownership. However, after 1935 many such caves were forgotten or neglected, but most trails have survived because the more rugged land became incorporated into public reserves and later national parks. Over 30 caves are known to have been used for bush camping by walkers and climbers, and most still are, especially in rugged country more than a days walk from roads. Some caves were appropriated as weekend and holiday retreats by walkers, fishermen and others, and partly as a result have inspired works by writers, poets, artists and musicians. A number were used as permanent homes by recluses and the indigent, especially in the early 20thcentury and during depression years. As recently as 2011 inhabitants of several caves in Royal National Park on the southern outskirts of Sydney were removed by authorities. In recent decades the sports of rock-climbing, bouldering and canyoning have flourished throughout the region, especially in the upper Blue Mountains. At least 20 caves are used by climbers, some with fixed climbing routes. Caves in the splendid deep and narrow canyons are less frequented and more robust because of regular flooding.2. Geology and LandscapeMost caves are developed in either the Middle Triassic Hawkesbury Sandstone, or the Early Triassic Narrabeen Group of massive sandstones interbedded with claystones and shale. Beyond the Cumberland Plain of Sydneys metropolitan area, outcrops of the former are dominant over most of the coastal zone and from the Southern Highlands district to ridges north of the Hawkesbury River, while the latter predominate in the upper Blue Mountains west of Woodford. Block-gliding caves occur in a smaller area of Permian Shoalhaven Group sandstones near Nowra. Current research suggests that uplift of all these rocks occurred no more recently than the early Palaeogene period,Figure 2. A former Aboriginal habitation site below houses in the Sydney metropolitan area, now used by walkers and bouldering enthusiasts. Figure 3. Sydney Basin showing location of Hawkesbury, Narrabeen and Shoalhaven groups of sandstone. Cross-section AB is from Lithgow to Mona Vale, near Sydney. at least 65 Ma, after which the area has been relatively stable. Rivers maintained their courses, eroding deep gorges in relatively level plateaux surrounding Sydney. Over 100 sea caves up to 100 m+ long are developed extensively in cliffs lining most of the coastal exposure of Sydney Basin rocks north and south of the city, some being essentially eroded dykes of mostly Neogene age. There are no known volcanic caves. Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings260

PAGE 263

In the Nowra Sandstone of the Shoalhaven Group there are at least 25 block-gliding caves up to 240 m long in riverside cliffs, probably primarily the result of rock deformation and sliding on a clay base, and a few other mass-movement caves have been identified from this study elsewhere in the region. In the Hawkesbury and Narrabeen sandstones most caves are similar in form and at least partly solutional in origin. In the Narrabeen Group caves are commonly associated with knickpoints, waterfalls and intercalated beds of claystone, and at least 40 or 50 are found in deep, narrow and often quite dark canyons, which in some cases are no more than a metre wide at base. The paucity and lenticular nature of such beds in the Hawkesbury Sandstone produces few canyons but simpler and still numerous caves. The growing evidence of solutional activity in non-carbonate rocks such as quartz sandstone is discussed in Young et al (2009), the authors of which carried out extensive work in the Sydney Basin. A variety of karst-like surface features exists. Minor solution forms such as grikes and runnels occur, and solution basins (gnammas) were important sources of water for Aboriginal groups, especially on the dry, higher ridges. Ruiniform landscapes resembling ruined cities and towns are developed extensively in the Narrabeen Group, particularly above Wolgan Valley and in Gardens of Stone National Park, both with small caves usually at the base of sandstone towers locally called pagodas, analogous to remnant towers in limestone country (Washington and Wray 2011). Also in the Narrabeen Group are most of the 40+ deep canyons, some over 100 m deep, often with waterfalls associated with knickpoints. Nearly all incorporate lateral incision caves, especially where the canyon intersects a claystone bed. About a dozen natural bridges, some across streams, are recorded. 2.1. Formation, enlargement and age of the caves There is a long-entrenched popular belief that these caves (other than the sea caves) are formed by wind, sandblasting or at least by enlargement upwards due to convectional air currents. There is no evidence of this. Their real origin, well established for many years, is through the essential trigger role that solution plays in initiating caves and enabling them to enlarge. The host rocks are commonly characterised by matrices of silt, clay and soluble cements such as iron carbonate (siderite). The process commences with naturally acidic water (percolating from swamps carrying organic acids in the upper Blue Mountain) interacting with and removing the matrix. Hydration then releases iron, silica and other elements. The rock becomes less coherent and more susceptible to physical disintegration, and the grains disaggregate, fall or are washed away. Enlargement then takes place due to the interrelationship between the high permeability and higher rates of weathering in the less coherent interior, which weathers chemically faster than the entrance and surrounding case-hardened surface rock (Johnson 1974). Salt is not as active a factor in the Sydney region as commonly believed, and its presence assists chemical rather than mechanical processes (Young 1987). Field studies show that this process is very slow. However sufficient time has been available in the Sydney Basin since the end of the Triassic. Indeed, it may be that these caves can form only BECAUSE the process is so slow. If water movement through the rock matrix were too rapid, widespread surface lowering would occur. Hughes and Sullivan (1983) suggested that natural rate of cavern growth is generally less than 2 to 5 mm/1,000 years. On this basis, it seems likely that some caves are of the order of at least a few million years old. 2.2. Types of caves In addition to sea caves and mass-movement caves, several overlapping types of solutional and non-solutional caves are apparent: undercuts, lateral meanders and incisions, piping caves, those associated with waterfalls, a small number of through stream caves, caves in deep narrow canyons, and other small caves. Undercuts, some quite long, are usually associated with relatively thin (2 m) beds or lenses of claystone or shale, sufficiently high above valleys that it is difficult to attribute them to old lateral stream incision. Lateral meanders and incisions are particularly common where streams have incised through one or more claystone beds, sometimes on the outside of tight, entrenched meanders. Only a few caves are located predominantly behind waterfalls, negating an origin due primarily to plunge pools. Much more often they form laterally downstream, possibly from spray and/or interstratal seepage, lengthening as the waterfall retreats. The longest cave, 263 m, is a horseshoe-shaped example of this. Small-scale piping is evident in many caves such as in the Budawang Mountains south of Nowra, but two in particular, at Kincumber and Leura, are clearly formed primarily by this process. There are two excellent examples of through stream caves similar in all regards to those in limestone. TheFigure 4. Typical sea cave at Avalon, near Mona Vale. Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings261

PAGE 264

Underground River at Blackheath resulted from a landslip filling a tight incised meander, forcing water through joints in the meander neck. Although a fossil dry valley exists nearby, Hill Top Cave at Mittagong appears to have formed by a combination of a relatively weak joint and upstream migration of a knickpoint. 2.3. Speleothems and Speleogens The most widespread and characteristic forms are speleogens, notably cavernous weathering forms such honeycombing. Protruding plates of eroded bedrock are prominent in several caves, usually at a shallow sub-parallel angle to the back wall, for distances as great as a metre and invariably associated with cross-bedding. Their aesthetic attraction is enhanced by characteristic bands and swirls of iron staining of the sandstone, so that colour can vary from dark red and brown through orange and yellow to white. deterioration of caves and art sites. Their location in the more rugged terrain along and on the steep slopes above water courses rather than on ridge tops also aided survival. Intensive research has concentrated on Aboriginal sites, such as that relating to survival of cave art sites (Hughes 1978). Many sensitive sites are protected only by lack of publicity and/or by their location in military and water catchment or remote and rugged areas. A few close to urban settlement have suffered inexcusable vandalism, but some protective measures are now in place. Caves with aesthetically pleasing, thin protruding bedrock plates are vulnerable, with no additional protective measures provided; the same applies to those with unusual mineral deposits. The greatest environmental pressure comes from urban development which has concentrated run-off and thus flooding, sedimentation, pollution, infiltration of garden and other chemicals, and has increased graffiti. Although improvements have occurred in recent years, geoheritage has not enjoyed high priority among management authorities, and important landscapes such as that of Gardens of Stone National Park have not been adequately studied. Further persuasion of responsible public authorities, both state and local, is called for. The caves are worth celebrating primarily for their association with human life and endeavour, in other words, for their intangible cultural and heritage value. Some contain unusual features and deposits worthy of further research and proper conservation measures.4. AcknowledgmentsField work was greatly assisted by members of Highland Caving Group, and advice on documentation procedures was provided by Peter Dykes. The paper benefitted from critical comments from Dr Susan White, Latrobe University, Melbourne, and by Cathy Brown, Geoscience Australia, Canberra. Photographs by author except for Fig.4 (J. Sydney).Figure 5. Protruding bedrock plates (speleogens) in a Hawkesbury Sandstone cave north of Sydney. Many are very vulnerable to damage and deserve closer management, informed by further research, documentation and discussions with local Councils and other management authorities. Both carbonate and silicate speleothems are found, noticeably smaller than in limestone caves but still fairly common, including flowstone, cave coral, stalactites, stalagmites and small columns. A few caves in deep valleys with sheltered aspects facing east to south exude sticky gelatinous nodules presumed to be of partly dissolved limonite or other minerals, along with unidentified forms of what appear to be thin, parallel bands of mud stalactites possibly indurated with calcite, limonite and/or vegetation matter. Depositional minerals include carbonates of both calcium and iron, silicates, iron oxides, generic alum, alunite (potassium aluminium sulphate), and manganese. Secondary growths of both silica and carbonates long ago raised questions about origins (Osborne 1948; Lovering 1951).3. Significance, Conservation and ManagementMost caves are quite robust and have survived well, some having been used by humans for thousands of years, although evidence is that Aboriginal occupation did hastenFigure 6. Gelatinous nodules containing suspected iron minerals in a cave near Blackheath. Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings262

PAGE 265

ReferencesAttenbrow V, 2002. Sydneys Aboriginal Past. Univ. NSW Press, Sydney, 225. Dunkley J, 2013 (in press). Caves, people and land: Sandstone caves of the Blue Mountains and Sydney Region. Australian Speleological Federation Inc., Sydney. Hughes PJ, 1978. Weathering in sandstone shelters in the Sydney Basin and the survival of rock art. Proc. International Workshop on the Conservation of Rock Art, Perth, Sept. 1977, 36. Hughes PJ, Sullivan ME, 1983. The geoarchaeology of the Sydney Basin sandstones. In Young, R.W. & Nanoson, G.C. eds. Aspects of Australian sandstone landscapes. Australian and New Zealand Geomorphological Group, Wollongong, 120. Johnson ARM, 1974. Cavernous Weathering at Berowra, N.S.W. Aust. Geographer XII, 6. 531. Lovering JF, 1951. The Mystery of the Limestone Caves in National Park. Aust. Museum Magazine, March 12, 1951. 155. Osborne GD 1948. A Review of Some Aspects of the Stratigraphy, Structure and Physiography of the Sydney Basin. J.Proc.Linn.Soc.NSW 83 pts 1, 4. Smith J, 1999. Blue Mountains District walking track heritage study: Historical report prepared for NSW National Parks and Wildlife Service. Heritage assessment & conservation guidelines prepared for NSW National Parks and Wildlife Service by MUSEcape Pty Ltd & David Beaver. Washington H, Wray RAL, 2011. The Geoheritage and Geomorphology of the Sandstone Pagodas of the Northwestern Blue Mountains Region NSW. Proc.Linn.Soc. NSW 132, 131. Wray, RAL 1993. Solutional Landforms on Silicates: largely ignored or simply unrecognised? In Proc. 19thBiennial Conf. Aust.Speleol.Federation, Launceston, 110. Wray, RAL, 1995. Solutional Landforms in Quartz Sandstones of the Sydney Basin. PhD thesis, Univ. Wollongong, 381. Young, ARM, 1987. Salt as an agent in the development of cavernous weathering. Geology 15, 962. Young R, Wray RAL, Young A, 2009. Sandstone Landforms. Cambridge UP, Cambridge, 304.Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings263

PAGE 266

KARST EVOLUTION IN SANDSTONE: THE CHAPADA DOS GUIMARES SITE, BRAZILRubens Hardt1, Jol Rodet2,Sergio dos Anjos Ferreira Pinto3 1Instituto do Carste, Rua Seis, 1043 apto 112, 13500-050, Rio Claro, SP, Brazil, rubens.hardt@gmail.com2UMR6143CNRS/Universit de Rouen, 76821, Mont Saint Aignan, France, joel.rodet@univ-rouen.fr3UNESP, Rua 24A, 1515, 13506-900, Rio Claro, SP, Brazil, sanjos@rc.unesp.br Karstification of sandstone was considered a controversial topic in the past, but it has become increasingly accepted by karst researchers in recent decades. A large area of Brazils territory has sandstone and metasandstone outcrops. The countrys tropical climate, abundant rainfall and vegetation, allied to the presence of organic matter, iron oxides, salts and long weathering processes, among other factors, accelerated the chemical weathering of quartz. This combination of factors has produced a wide variety of karst landforms as well as a spatial organization that allows them to be referred to as Sandstone Karst Systems. The area under study was the Chapada dos Guimares. This study allowed the determination of the karst topography by understanding the organization of these landscapes processes and their systemic integration. The knowledge thus gained, served to underpin the formulation of hypotheses about the processes involved in the structuring of these landforms and systems. Above all, it shows the importance of vegetation as an aid for quartz dissolution, on the process of rock weathering ( in situ alteration), and subsequent removal of modified material by complete dissolution or by mechanical means. That resulted in karst landforms and systems. The aforementioned hypotheses served as the basis for the proposal to amend the definition of karst with respect to the term chemical dissolution, by using a broader term chemical weathering. This will allow the inclusion of changes in rock resulting from chemical processes and mechanical removal as karst, since they produce the same type of system and function in the same way as systems typically developed by dissolution of carbonaceous rocks.1. IntroductionSandstone karst is still a controversial subject, even after several studies showing its reality (Mainguet 1972; Jennings 1983; Young 1988). Although the term is accepted by some karst researchers, several questions remain open. Therefore, others still disagree with the possibility that sandstone is a substrate for karstification, without studying or even visiting a sandstone site themselves. In tropical areas, like in Brazil, it is possible to find large areas of sandstone, most of them presenting several aspects of karst (Hardt 2011). These include caves, kamenitzas, closed depressions, several types of karren, underground rivers and even some speleothems, showing not only the dissolution, but also precipitation of minerals. Since hydraulic gradient is needed to provide some movement to groundwater, usually these karstification areas are associated with gentle slope reliefs (e.g., cuesta backslopes) where the altitude provides such gradient and large area of exposition to weathering processes, giving rise to a karst system. One of these areas, the Chapada dos Guimares site, in Mato Grosso State, Brazil (Fig. 1) has developed a complex, polyphase karst system. In this area, a cuesta relief can be found, where water is abundant, and the gradient on the cuesta backslope side is low, providing conditions to the development of karstic features. The proximity of the cuesta front also has interfered with the system, generating some captures that altered the gradient and influenced the evolution of the karst system in the late stage. The occurrence of caves is in the contact from the sandstones of Ordovician Alto Garas formation to the sandstones of Silurian Vila Maria formation. The contact between the formations shows discontinuity. According to Kppens system the contemporary climate is classified as Aw, with dry season in autumn and winter, and wet season in spring and summer.Figure 1. Location in Mato Grosso State, Brazil. 2. MethodsIn the field, most of the work consisted of exploration guided by the knowledge of the main caves locations, in addition to the previous analysis of maps and satellite images from ChinaBrazil Earth Resource Satellite (CBERS), morphometric analysis of landforms, topography of caves and depressions. Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings264

PAGE 267

Exploration of sites that may present elements not described or known by previous researchers, but with the potential to have something important complementing the knowledge of the site, allowed to build a database of forms, measurements, spatial location and distribution. The endokarst measurements provide the morphometric characteristics of the conduits allowing one to understand the evolution of the caves. A map of the area was created with the spatial distribution of the forms, their organizations and the connections between them. This helps to understand how water dynamics behaves in the system by generating forms, describing, this way, the systems morphodynamics. At the end, a conceptual model was proposed, describing the evolution of the system through time and its consequences on landforms.3. Data and observationsSeveral exokarstic features were found in the area, including varieties of karren, kamenitzas, arches, unroofed caves and closed depressions. Two of these closed depressions, referred from now on as poljes, are very important for the comprehension of the system. The first one, denominated the upper polje, has about 45,300 m2of area, ellipsoidal form, and feeds with water the Kiogo-Brado cave. The other depression, denominated the central polje, has an area of 150,000 m2, and resembles the shape of a banana, an elongated ellipse with a turn in its center. Its water feeds the Aro-Jari system, and Lagoa Azul cave. The endokarst is complex but organized, showing a mature system evolving to a senile system with abandoned galleries, residual pillars, dendritic drainage, dissolution cupolas and speleothems. The main cave system, Aro-Jari, is 1,500 m long, with galleries going from sub-metric to more than 15 m high, and forming saloons more than 50 m wide by 100 m long. The main drain is about 1 km across the massif, with an average height of 12 m and width of 10 m. The Kiogo-Brado cave is about 210 m long with an average height of 13 m and width of 7 m. Its height can reach 16 m and the main entrance is about 33 m wide by 15 m high, but over an accumulation of blocks the void of the cave itself probably goes down under 10 m more. Another important cave is the Lagoa Azul Cave, a large flooded saloon 50 m 30 m, of a trapezoidal shape, with several underwater conducts at the end of the saloon going into the massif, following the main joints. Elements of an ancient karst system can be found above the current system. There are arches, unroofed caves, and fragmented conduits in residual rock and ancient deposits above the present day level.4. Results and discussionThe existence of a residual karst above the present day one demonstrates that the evolution of the landscape has involved several phases, while the water had to find its way down. But the complexity of the endokarst is the most important information about this system, since it is quite identical to several others in limestone. In the backslopes of the cuesta relief, the water following the gradient generates depressions by dissolution, where it accumulates. The weathering of the rock generates clay, which is deposited at the bottom of the depression, making it waterproof. For being filled with water and finding a barrier of rock at the side, the water penetrates amidst the rocks following the joints of the bedding plane or the structure, trying to find a way to cross the massif. Several pipesstarted to develop in those joints, especially in the gradient direction by dissolution, creating essays to cross the massif. These pipes were abandoned when one of them succeeded and concentrated the flux. The walls of these tubes are covered by a crust, which demonstrates the mobility of silica (Fig. 2). This is common to depressions, the upper polje and the central polje.Figure 2. One of the tubes and SEM image, demonstrating the chemical mobility of the silica. Photos by Rubens Hardt. After some time these pipes were abandoned in favor of one of them where flow has been concentrated, since it reaches the other side of the massif. Karst and Caves in Other Rocks, Pseudokarst oral2013 ICS Proceedings265

PAGE 268

Now it is important to separate what happened in the connection between the upper polje and the central polje, in Kiogo-Brado cave, and the central polje and whatever could have been after Aro-Jari cave: a polje or a valley. One of the essays to cross the first massif had reached the other side but it found another great depression, filled with water, and the flow had been blocked. That has allowed the deposition of sediments, partially provided by the incomplete solution of the rock, and partially transported by allogenic fluxes. This allowed a paragenetic evolution inside the caves. It is important to understand that the incomplete dissolution of rock is here regarded as in carbonate rocks, where the process denominated ghost rock formation (Quinif 2010), previously described as in situ alteration (Rodet 1996), is the main process in the formation of caves and its galleries, since residual alterite (ghost rock) was found in several places inside the caves. This circumstance allows a paragenetic evolution, with the water flowing above sediments, enlarging the galleries. Sediments trapped above the contemporary level (Fig. 3) and roof channels are the evidence of this phase.Figure 3. Sediment trapped on the roof. Photo by Rubens Hardt. Figure 4. Double-roof channel. Photo by: J. A. Labegalini. At Aro-Jari much of the protogalleries was filled with residual rock, and the water was running above it, so this cave also had a paragenetic evolution in a certain period. After a continuity solution (a connection between two points by a conduit) has been established from the upper polje via Kiogo-Brado to the central polje, and from the central polje via Aro-Jari to a polje or valley beyond Aro-Jari, the paragenetic evolution ceased and, in the Aro-Jari Cave many parallel conduits have coalesced, giving a large central collector as attested by residual pillar in the central collector and the double-roof channel on affluent conduit (Fig. 4). This characterizes a karst with captures of surface flow to the underground system. Then, the evolution of the landscape brings the front of the cuesta nearby the area where the karst has evolved, inducing the capture of the central polje by a drainage running into the depression, resulting in two different reactions in both caves. In the Aro-Jary, the lack of water converted the cave in an almost fossil cave, with too little water coming in and without enough energy to transport the sediments and organic matter from one side to another. The large saloons inside have become sediment traps and only the water that rises from a diffuse spring in the center of the central polje is captured to the cave. In the Kiogo-Brado Cave the fast drain created by the opening of the polje, lowered the cave in a syngenetic evolution, attested by the residual deposits on the side walls. After that, the upper polje has been captured and reduced the amount of water which feeded the Kiogo-Brado Cave, what originates a much smaller incision at the basement of it, giving the key role shape in the cross-section of the conduit. This is the actual situation at Chapada dos Guimares site (Table 1).Table 1. Comparative schema between Kiogo-Brado subsystem and Aro-Jari subsystem.From Upper polje to center polje (Kiogo-Brado connection): From center polje to the valley or lower polje (Aro-Jari connection): I Progradation fronts, primokarst. I Progradation fronts, primokarst. II Hydrokarstic connections, paragenesis. II Hydrokarstic connections, paragenesis. Water concentration, fully developed main conduit. III Lowering of water table due to central polje opening. Syngenetic evolution. III Lowering of water table due to central polje opening. Systems fossilization, reduced drainage, sediment trap. IV Upper polje opening, systems fossilization. Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings266

PAGE 269

5. ConclusionsSuch a karstic system in sandstone as described here is quite rare, since sandstone is much more susceptible to other forms of erosion that competes with chemical erosion or alteration. Karst in sandstone is not usually a pure karst with elements of a fluvial system competing in the structure of the system. But even when the system is not a pure karst, it is possible to state that karst in sandstone is a reality, since the processes (dissolution/alteration), the landforms (closed depressions, underground drainage, several small scale surfaces and underground forms) and finally the system may develop in this rock. The complex evolution and the response of the system to external factors put the sandstone karst in a similar level to carbonate karst.ReferencesHardt R, 2011. Da carstificao em arenitos. Aproximao com o suporte de geotecnologias. propos de la karsification dans les grs. Traitement par les technologies SIG. (Ph. D. Thesis) Instituto de Geocincias e Cincias Exatas. Universidade Estadual Paulista Jlio de Mesquita Filho. Em co-tutela com Morphodynamique Continentale et Ctire. Laboratoire de Gologie. Universit de Rouen. 224. Jennings JN, 1983. Sandstone pseudokarst or karst? In: Young RW; Nanson GC, Aspects of Australian Sandstone Landscapes. Wollongong: Australian and New Zealand Geomorphology Group Special Publication no. 1, 21. Mainguet M, 1972. Le model des grs: Problmes Gnraux. Paris: Institut Gographique National, 228. Quinif Y, 2010. Fantmes de roche et fantmisation. Karstologia Mmoires 18, 184. Rodet J, 1996. Une nouvelle organisation gomtrique du drainage karstique des craies: le labyrinthe daltration, lexemple de la grotte de la Mansonnire (Bellou sur Huisne, Orne, France). Comptes Rendus de lAcadmie des Sciences de Paris, t. 322 (12), srie II a: 1039. Young RW, 1988. Quartz etching and sandstone karst: Examples from the East Kimberleys, northwestern Australia. Zeitschrift fur Geomorphologie 32, 409. Figure 5. Schematic view (see table 1 for phases description).Karst and Caves in Other Rocks, Pseudokarst oral 2013 ICS Proceedings267

PAGE 270

THREE TYPES OF CREVICE-TYPE CAVES IN THE AREA OF CZECH FLYSCH CARPATHIANSJan Lenart1,2, Tom Pnek1 1Department of Physical Geography and Geoecology, Faculty of Science, University of Ostrava, Chittussiho 10, 71000, Ostrava-Slezsk Ostrava, Czech republic, honza.lenart@seznam.cz2ZO SS 7.01 Orcus, slavsk 407, 73581, Bohumn, Czech republic, honza.lenart@seznam.cz The crevice-type caves of the Czech part of the Outern Western Carpathians are formed by the processes of the gravitationall slope movements: spreading (or incipient translational sliding), toppling and rotational sliding. The Outern Western Carpathians are built by the flych Mesozoic rocks: sandstones, conglomerates and shales. The aim of the research was to detect types of the process and to divide the caves into the cathegories.1. IntroductionCaves are among the most distinctive forms of pseudokarst phenomena (Halliday 2007). Most extensive and widespread pseudokarst caves are crevice-type caves formed during evolution of slope failures (Vtek 1978). Margielewski and Urban (2003) considered the crevices to be free spaces between two rock blocks, when at least one of them is affected by slope movement. Most commonly, they accompany evolution of deep-seated gravitational slope deformations (DSGSDs);i.e. types of large-scale and relatively short-displacement mass movements affecting large volumes of mountain ridges (N m ok 1972; Dramis and Sorriso-Valvo 1994) and reaching the depth at least 30 m (Hutchinson 1988). They are mostly slow or extremely slow mass movements with the average velocity less than 18 mm per year (Varnes 1978). However, besides DSGSDs, the crevice-type caves can be produced also within the shallow and medium-deep (<30 m according to Hutchinson, 1988) structurally-predisposed landslides (Margielewski 2009). The crevice-type caves of the same genesis may occur in distinct types of rocks. The crevices very often cut incoherent sedimentary flysch rocks: sandstones and conglomerates with intercalations of shales, but can occur in granites, quartzites, gneisses and other lithologies (Vtek 1978). Crevices originate also in the carbonate karst rocks (limestones and dolomites), but these are usually classified as a karst caves (Vtek 1978). Typically, crevices and crevice-type caves occur within the areas built by anisotropic rigid rocks underlain by the plastic shales (Pnek et al. 2009 and 2010). Crevice-type caves originate by the process of gravitational widening and movement of tectonically or lithologically predisposed rock massif. This type of caves is mostly regularly narrow-shaped (A and V or H letter in crosssection) with high walls and flat roofs or floors. The corridors may reach appreciable length (Vtek 1978 and 1981). Within the crevice-type cave development, other processes might play a role, such as water erosi