Theoretical and Applied Karstology

Material Information

Theoretical and Applied Karstology
Series Title:
Theoretical and Applied Karstology
Emil Racovita Institute of Speleology
Emil Racovita Institute of Speleology
Publication Date:


Subjects / Keywords:
Geology ( local )
serial ( sobekcm )
46 x 25


General Note:
Radulescu, C. Editorial. pp. 7-8. TAK Articles Onac, B., P. Lauritzen, S.-E. The climate of last 150,000 years recorded in speleothems: preliminary results from north-western Romania. pp. 9-22. Seventy-four 230Th/234U dates on speleothems enable preliminary estimates to be made of the palaeoclimatic evolution of the north-western part of Romania. The growth frequency of speleothems was examined using the cumulative distributed error frequency method (Gordon, 1989). The climatic oscillations that are suggested by our curve correlate tentatively with known climatic events provided by pollen analysis, coral growth frequency and oxygen isotope records from both deep sea sediments and ice cores during the same timespan (the last 150,000 years). Pourchet, M., Mélieres, M., A., Silvestru, E., Rajka, G., Candaudap, F. Carbonnel, J.-P. Radionuclides in a caves lake sediment core from Ghetarul de sub Zgurasti (Romania). pp. 23-28. Some 250 m from the entrance of the cave Ghetarul de sub Zgurasti, on the bottom of a lake exclusively charged per ascensum, at 12 m depth, a 41 cm core of fine sediment had been taken for radionuclide investigation. The following have been identified: 210Pb-a natural radionuclide originating from atmospheric Radon decay, 134Cs and 137Cs-artificial radionuclides derived from the thermonuclear tests in the early sixties and the Chernobyl accident in the mid-eighties. Deposition fluxes of these elements at the water/sediment boundary have been found to be of the same order of magnitude as the ones at the surface, which shows a very effective transmission of pollution from the surface to remote cave environment. This is, to our knowledge, the first time such a substantial contamination has been measured. Coherent results were yielded by the profiles, especially in terms of sedimentation processes, allowing estimates of sedimentation rates, diffusion coefficients and the recent history of the sediment. The latter proved to be in perfect agreement with the climate recorded through the last decades. Labau, V., Gaspar, E. Paunica, T. Speleothems dating using the thermoluminiscence method. pp. 29-34. The age of a stalagmite fragment collected from Pestera Vântului (Wind Cave), Romania, was estimated using the thermoluminescence method. The natural dose accumulated by the calcite during the geological age was established. Using the relationship between the calcite thermoluminescence sensitivity and the applied pre-dose, an age of 59,052 years has been established. Carlier, E. Effet d'échelle de la dispersion dans un conduit karstique. Hypothèse linéaire et hypothese fractale. pp. 35-44. Les variations du coefficient de dispersion sont étudiées dans un conduit rectiligne et dans un conduit présentant une tortuosité à géométrie fractale. Dans le cas d'un conduit rectiligne, nous montrons que le coefficient de dispersion est sujet à un effet d'échelle, jusqu'à une certaine distance au?delà de laquelle on retrouve le processus classique fickien de dispersion, caractérisé par un coefficient constant. En revanche, dans un conduit à tortuosité fractale, nous montrons qu'il n'existe pas de valeur asymptotique du coefficient de dispersion apparent. La détermination du coefficient de dispersion par la variance est discutée. Carlier, E. Influence de la structure sur la variabilité spatio-temporelle du coefficient de dispersion. Cas théorique d'un faisceau de conduits karstiques soumis à une injection continue en absence de diffusion moléculaire (écoulement rapide). pp. 45-50. De nombreux modèles de simulation du transfert de masse en karst et milieu fissuré reposent sur l'hypothèse d'un coefficient de dispersion invariant dans le temps et l'espace pour une vitesse moyenne d'écoulement constante. A partir d'un ensemble de conduits cylindriques horizontaux, on montre que le coefficient de dispersion est, généralement, une fonction du temps, de la distance et de la structure du milieu karstique ou fissuré. Lorsque la fréquence de répartition des sections obéit à une loi exponentielle particulière, le coefficient de dispersion n'est plus fonction que de la distance. Miatovic, B., F. Relation entre les caractères physiques et les champs de leur distribution dans les roches karstiques. pp. 51-62. Dans les roches karstiques la structure des champs de distribution spatiale des majeurs paramètres physiques, tels la perméabilité et la porosité peut être caractérisée par: (1) la distribution statistique dans les blocs-matrices, (2) la distribution ponctuelle dans le réseau d'écoulement, et (3) l'effet d'échelle sur ces champs. Ces considérations demandent à être bien expliquées, notamment dans le domaine de la fracturation des massifs calcaires, dont les caractères physiques et hydrauliques sont sous l'influence de l'effet d'échelle. On constate souvent en terrains karstiques une coexistence de deux types bien distincts d'écoulements: l'un, correspondant aux classiques rivières souterraines, lié aux collecteurs des grandes dimensions du réseau karstique, l'autre, conforme à une variété de circulation plus ou moins lente, dans les blocs-matrices volumineux mais relativement peu perméables. Il se pose alors la question de la répartition des valeurs de perméabilité et de porosité efficace dans un réseau de conduits connectés et des blocs-matrices lorsque l'échelle varie. Papadopoulou, K., A. Comparative observations in karst formations of Parnass and Arcadia Mountains (Greece). pp. 63-68. The paper deals with the study of two great karst areas in Greece. The first one belongs to the Parnass Mountain (Central Greece) while the other one to the Arcadia Mountain (South Greece). A large number of dolines (300) have been studied in order to identify their common characteristics, differences, and the rules that control their development. The final purpose of the paper is to emphasize the rules that control the genesis of those areas. Papadopoulou, K., A. Gournellos, T. A study of the subsurface karst of Kopais (Central Greece). pp. 69-76. In this article we attempt to perform a quantitative analysis of the swallets ("katavothres") of the former lake of Kopais (Central Greece). These are situated at different altitudes around the former lake. The principal outcrops are various limestones, flysch and schists. We investigate the relationships between various variables such as lithology, altitude, horizontal development and geographical position of these karst forms. We proceed to a statistical analysis of the above variables and we study the spatial distributions of the swallets. The final results are indicative of the development of these karstic forms and their dependence on the various parameters. Zlokolica, M., Mandic, M. Ljubojevic, V. Some significant caves at the western rim of the Miro? karst (Yugoslavia). pp. 77-88. The western rim of the Miro¹ karst oasis is composed of Lower Cretaceous sandstone, marl, mudstone and limestone (K1,2), and Upper Jurassic limestone (J33), and it abounds with rivers which sink into caves of considerable dimensions. The deepest of them is the pothole Jama u Laniëtu (-272 m), while the longest is the cave Buronov Ponor (1950 m). The caves abound with diverse forms of passages - ranging from almost unaltered fissures to well-rounded phreatic tubes. The phases of fluvial sedimentation, of rock collapses, of formation of travertine coating, and of hydrologic reactivation and emptying of the cavities, can be distinguished as well. Stevanovic, Z., Dragisic, V., Dokmanovic, P. Mandic, M. Hydrogeology of Miro? karst massif, eastern Serbia, Yugoslavia, pp. 89-96. The karst massif of Miro? mountain is one of the national parks of Yugoslavia, situated in the northern extremity of the Carpatho-Balkanides of eastern Serbia and extending to the Danube and the Romanian border. The area of the calcareous massif, composed mainly of Tithonic limestones, exceeds 120 km2. Four karst springs used to drain a karst aquifer located in the north and the west of the massif, along the Danube, before being submerged by the reservoir behind the dam at Portile de Fier (The Iron Gates). To the south, the massif is drained by Blederija spring (Qav. about 30 l/s). The massif has a multitude of active concentrated swallets. Several deep caves (e.g. Rakin Ponor, 278 m deep, the deepest in Serbia) were recently investigated by speleologists. Explorations have been undertaken in Miro? area for karst ground water intake aiming at the water supply of Donji Milanovac resort center. Samani, N. Ebrahimi, B. Analysis of spring hydrographs for hydrogeological evaluation of a karst aquifer system. pp. 97-112. Karst aquifers are, in general, extremely heterogeneous in character. As a result, quantitative data obtained from selected points in the system, either by pumping or by using marker dyes, can be rarely extrapolated to evaluate the average function of the system as a whole. In contrast, recession curves of spring hydrographs show the overall response of the aquifer to precipitation events and help evaluating aquifer storage and drainage potentials. In this study, three analytic equations are used for estimating hydrogeological characteristics of a karst aquifer: Mangin's equations, a newly proposed equation that assumes that the recession curve is composed of both quickflow and baseflow, and Coutagne's equation, which considers the recession curve to be the response of a single reservoir. The hydrographs of three springs draining Sabzpooshan and Beaza aquifers in Fars Province, Iran, are analyzed. From the analysis, it is apparent that the saturated zone accounts for 99%, 91% and 99%, respectively for Pirbanoo, Polberengi and Ghorehdan springs. Iurkiewicz, A., Voica, M. Bulgar, A. Indirect evaluation of the Izvarna karst system discharge trend (Romania). pp. 113-120. The importance of Izvarna karst system stems from the fact that its water is being tapped for the supply of the city of Craiova. The hydrodynamic behavior of the aquifer under natural conditions has been investigated based on records from 1957-1965. The development of the Cerna-Motru-Tismana hydropower system has induced changes in the surface streams discharge regime and hence possible consequences on the karst aquifer supply. Compared to an accurate water budget evaluation, the correlative and spectral analysis of the discharge is a cost?effective method for the identification of the subsurface flow evolution trend induced by human activity. The paper illustrates the ability of the method to provide a quantitative evaluation of the water resource reduction as well as the limits of the adopted approach. Iurkiewicz, A., Dragomir, G., Rotaru, A. Badescu, B. Karst systems in Banat Mountains (Resita-Nera zone). pp. 121-140. The Banat Mountains area, located in the south-west of Romania includes more than 800 sq. km of karst zones belonging to the Resita - Moldova Noua large synclinorium. Although scientific karst investigations were initiated at the beginning of the century, the previously collected hydrogeological data were rather local and contradictory. Taking into account groundwater resources that karst areas should normally contain, "Prospectiuni" company has recently started a complex research program, in order to evaluate the amount of groundwater resources of this particular area. The karst territory was divided as a result into three distinct zones, which correspond roughly to the Caras, Minis-Nera and Nera-Danube catchment areas. The hydrologic, hydrodynamic and hydrochemical characteristics of the karst systems included within the first two zones are discussed. The water budget completed based on 1992-1995 data provides information on underground hydrologic connections between small catchment basins, and eventually for each of the above mentioned areas as a whole. Ristic, V., Prohaska, S. Stevanovic, Z. Some examples of karst springflow regime simulation and prediction for water management balance analyses. pp. 141-148. Karst occurrences in stream basins have not been given adequate consideration in the applied hydrology. As a consequence, water table and discharge of karst springs are not measured at stations of the official hydrometeorological network. For a proper assessment of the karst water regime, under the given study conditions, indirect methods and computation procedures have to be used. One of these is MNC (Multiple Nonlinear Correlation) model, used for filling-in, extending and three-dimensional interpolation of the hydrometeorological data. Based on this model, in the general areas of the karst springs (Izvor-Sveta Petka, Mlava, Grza, Nemanja, Mrljis with Groznicevac) on Kucaj and Beljanica mountains, eastern Serbia, this paper presents filled-in (where data are lacking) and extended series, which are essential for the investigation of the karst springflow regime, or for a water budget analysis that extends over several years. Stevanovic, Z., Markovic, M., Pavlovic, R., Jemcov, I., Dragisisic, V. Cupkovic, T. Correlation between regional fault pattern and karst water flow directions based on examples from Eastern Serbia (Yugoslavia). pp. 149-154. The analysis of the fault pattern of eastern Serbia was made as a part of the basic and purpose detailed hydrogeological investigations in some localities. By correlating the observed regional features with the point elements of the principal drainage systems, a significant coincidence of positions has been noticed. The strikes of the main faults, longitudinal NW-SE and younger NE-SW (W-E), largely coincide with the principal, general directions of karst ground water flows. Subjects of particular analysis were the calcareous massifs corresponding to the anticlinoria of Ku?aj and Beljanica, and of Rtanj and Ozren in central Carpatho-Balkan mountain range of eastern Serbia. Filipovic, B., Lazic, M., Vidovic, M. Sekulovic, B. Tapping and protection of Maljen Spring, on Mount Durmitor, Yugoslavia. pp. 155-164. Karst springs Veliki and Mali Maljen emerge, 30-35 m apart at the foot of one of the slopes of the mountain Durmitor at the altitude of 1320 m, approximately 15 km south of ?abljak, one of the largest winter resorts in Montenegro and Yugoslavia. The springs drain a vast karst reservoir consisting of Triassic carbonate rocks: they are of ascending type with the outflow zones masked by moraine material. The spring Veliki Maljen is tapped, whereas Mali Maljen still runs through glacial material. The discharge of the spring varies between a maximum of 150-200 l/s and 35-40 l/s indicating a characteristic karst regime. According to their physical and chemical properties, the waters are of exquisite quality and therefore they absolutely meet any norms or regulations for bottling, that is foreseen in the future. In order to determine the conditions of exploitation and of preservation of the water qualities, a necessary hydrogeological research has been carried out. It has been performed according to the regulations for establishing the boundaries of the sanitary protection zones for maintaining the necessary continuity during the building and the operation of the plant for bottling the natural drinking waters that flow through Veliki Maljen spring. Raeisi, E., Jehbez, O. Moore, F. Hydrochemical behavior of karstic and evaporitic formations surrounding Sarvestan Plain, Iran. pp. 165-174. The Sarvestan Plain is located in the central-southern Iran. The plain is surrounded by extensive karst carbonate deposits with minor amounts of evaporitic-argillitic sediments. These Mesozoic and Cenozoic formations are intruded by two Precambrian salt domes to the northeast and southeast of the plain. Despite the presence of karstic formations in the drainage basin of the plain, the quality of the ground water is very poor. In this study the lithological characteristics and mineral constituents of the surrounding formations are investigated by optical and chemical methods. During the summer 1992, the electrical conductivities of 500 operational wells, 15 springs, and 4 qanats were measured. Furthermore, 128 representative samples were analyzed for major ions including calcium, magnesium, sodium, potassium, chloride, bicarbonate, and sulfate using standard wet chemical methods. Temperature and pH of all the sampling sites were also measured. Using the WATQEF computer program, the saturation indices in various minerals were calculated. Results indicate that although the drainage basin of the plain consists mostly of carbonate formations, except for a narrow band in the south and a small region in the northeast of the plain where the type of water is bicarbonate, the rest of the aquifer exhibits a chloride-sulfate type water. The main sources of the chlorine and sodium ions are shown to be the exposed salt domes, and the sources of the sulfate are the dissolution of gypsum and anhydrite along with the oxidation of pyrite in the argillaceous formations. The reason for varying ionic compositions in different parts of the aquifer is discussed. The main reason for the poor quality of groundwater is apparently the mixing of the karstic water with water expelled mainly from salt domes and to a lesser extent from evaporitic and argillitic units. The higher solubility of halite minerals compared with that of carbonates, contributes to the poor quality of the groundwater. Gaspar, R. Pulse temperature and conductivity analysis in pseudo-karst structure investigation. pp. 175-184. The permeable zones and associated leakage through the pseudokarstic structure of Racova reservoir were determined by using natural tracers: electrical conductivity, TDS and water temperature. The variation of these parameters was measured in piezometers, boreholes, subaerial and underwater springs. To check the flow paths determined by natural tracers, artificial labelings were performed. Oraseanu, I. Contributions to the hydrogeology of the karstic areas of the Bihor Vladeasa Mountains (Romania). pp. 185-214. The Bihor Vladeasa Mountains are ones of the most important karst regions in Romania. They have important groundwater resources, unexploited till now. On the hydrogeological map there are pointed the main karst springs with their magnitude of discharges and the directions of the groundwater flows. More than 45 tracers labellings were performed and an average of 45 m/hour of flow velocity was recorded in 42 of these. As a consequence of the great diversity of the geological constitution and intense fracturation of the rocks, the karst systems are of binary type, with a large variety in size and hydrological parameters distribution. Due to the consistent observations and the hydro-meteorological measurements, the groundwater resources were evaluated and the processing of the flow rates series outlined a great diversity of the karstic systems. The quality of the groundwater is very good as indicate the results of the chemical and bacteriological analysis and the potential sources of pollution are leaks. Diaconu, G. Points de vue concernant le clivage de la calcite. pp. 215-218. A l'aide des quelques images obtenues au microscope électronique en balayage, l'auteur exprime un point de vue original concernant le clivage de la calcite. Iurkiewicz, A., Badescu, B. Marinica, E. Intensity of karst processes as a function of the carbonate formations in the north Resita-Moldova Noua Synclinorium. pp. 219-222. The exploration of the karst cavities in the Resita-Anina area led to the identification of a significant amount of "karst entrances", consisting mainly of caves and potholes, swallets and springs. The study of the characteristics of the karst cavities, as well as of their distribution with respect to the different types of carbonate deposits, has resulted in a qualitative evaluation of the karstification processes that have occurred in this specific area.
Open Access - Permission by Publisher
Original Version:
Vol. 9 (1996)
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:
K26-04800 ( USFLDC DOI )
k26.4800 ( USFLDC Handle )
11627 ( karstportal - original NodeID )
1012-9308 ( ISSN )

USFLDC Membership

Karst Information Portal

Postcard Information



This item has the following downloads:

Full Text
Radulescu, C.
Editorial. pp. 7-8. TAK Articles Onac, B., P. & Lauritzen,
S.-E. The climate of last 150,000 years recorded in
speleothems: preliminary results from north-western Romania.
pp. 9-22. Seventy-four 230Th/234U dates on speleothems enable
preliminary estimates to be made of the palaeoclimatic
evolution of the north-western part of Romania. The growth
frequency of speleothems was examined using the cumulative
distributed error frequency method (Gordon, 1989). The climatic
oscillations that are suggested by our curve correlate
tentatively with known climatic events provided by pollen
analysis, coral growth frequency and oxygen isotope records
from both deep sea sediments and ice cores during the same
timespan (the last 150,000 years). Pourchet, M., Mlieres, M.,
A., Silvestru, E., Rajka, G., Candaudap, F. & Carbonnel,
J.-P. Radionuclides in a caves lake sediment core from Ghetarul
de sub Zgurasti (Romania). pp. 23-28. Some 250 m from the
entrance of the cave Ghetarul de sub Zgurasti, on the bottom of
a lake exclusively charged per ascensum, at 12 m depth, a 41 cm
core of fine sediment had been taken for radionuclide
investigation. The following have been identified: 210Pb-a
natural radionuclide originating from atmospheric Radon decay,
134Cs and 137Cs-artificial radionuclides derived from the
thermonuclear tests in the early sixties and the Chernobyl
accident in the mid-eighties. Deposition fluxes of these
elements at the water/sediment boundary have been found to be
of the same order of magnitude as the ones at the surface,
which shows a very effective transmission of pollution from the
surface to remote cave environment. This is, to our knowledge,
the first time such a substantial contamination has been
measured. Coherent results were yielded by the profiles,
especially in terms of sedimentation processes, allowing
estimates of sedimentation rates, diffusion coefficients and
the recent history of the sediment. The latter proved to be in
perfect agreement with the climate recorded through the last
decades. Labau, V., Gaspar, E. & Paunica, T. Speleothems
dating using the thermoluminiscence method. pp. 29-34. The age
of a stalagmite fragment collected from Pestera Vntului (Wind
Cave), Romania, was estimated using the thermoluminescence
method. The natural dose accumulated by the calcite during the
geological age was established. Using the relationship between
the calcite thermoluminescence sensitivity and the applied
pre-dose, an age of 59,052 years has been established. Carlier,
E. Effet d'chelle de la dispersion dans un conduit karstique.
Hypothse linaire et hypothese fractale. pp. 35-44. Les
variations du coefficient de dispersion sont tudies dans un
conduit rectiligne et dans un conduit prsentant une tortuosit
gomtrie fractale. Dans le cas d'un conduit rectiligne, nous
montrons que le coefficient de dispersion est sujet un effet
d'chelle, jusqu' une certaine distance au?del de laquelle on
retrouve le processus classique fickien de dispersion,
caractris par un coefficient constant. En revanche, dans un
conduit tortuosit fractale, nous montrons qu'il n'existe pas
de valeur asymptotique du coefficient de dispersion apparent.
La dtermination du coefficient de dispersion par la variance
est discute. Carlier, E. Influence de la structure sur la
variabilit spatio-temporelle du coefficient de dispersion. Cas
thorique d'un faisceau de conduits karstiques soumis une
injection continue en absence de diffusion molculaire
(coulement rapide). pp. 45-50. De nombreux modles de
simulation du transfert de masse en karst et milieu fissur
reposent sur l'hypothse d'un coefficient de dispersion
invariant dans le temps et l'espace pour une vitesse moyenne
d'coulement constante. A partir d'un ensemble de conduits
cylindriques horizontaux, on montre que le coefficient de
dispersion est, gnralement, une fonction du temps, de la
distance et de la structure du milieu karstique ou fissur.
Lorsque la frquence de rpartition des sections obit une
loi exponentielle particulire, le coefficient de dispersion
n'est plus fonction que de la distance. Miatovic, B., F.
Relation entre les caractres physiques et les champs de leur
distribution dans les roches karstiques. pp. 51-62. Dans les
roches karstiques la structure des champs de distribution
spatiale des majeurs paramtres physiques, tels la permabilit
et la porosit peut tre caractrise par: (1) la distribution
statistique dans les blocs-matrices, (2) la distribution
ponctuelle dans le rseau d'coulement, et (3) l'effet
d'chelle sur ces champs. Ces considrations demandent tre
bien expliques, notamment dans le domaine de la fracturation
des massifs calcaires, dont les caractres physiques et
hydrauliques sont sous l'influence de l'effet d'chelle. On
constate souvent en terrains karstiques une coexistence de deux
types bien distincts d'coulements: l'un, correspondant aux
classiques rivires souterraines, li aux collecteurs des
grandes dimensions du rseau karstique, l'autre, conforme une
varit de circulation plus ou moins lente, dans les
blocs-matrices volumineux mais relativement peu permables. Il
se pose alors la question de la rpartition des valeurs de
permabilit et de porosit efficace dans un rseau de conduits
connects et des blocs-matrices lorsque l'chelle varie.
Papadopoulou, K., A. Comparative observations in karst
formations of Parnass and Arcadia Mountains (Greece). pp.
63-68. The paper deals with the study of two great karst areas
in Greece. The first one belongs to the Parnass Mountain
(Central Greece) while the other one to the Arcadia Mountain
(South Greece). A large number of dolines (300) have been
studied in order to identify their common characteristics,
differences, and the rules that control their development. The
final purpose of the paper is to emphasize the rules that
control the genesis of those areas. Papadopoulou, K., A. &
Gournellos, T. A study of the subsurface karst of Kopais
(Central Greece). pp. 69-76. In this article we attempt to
perform a quantitative analysis of the swallets ("katavothres")
of the former lake of Kopais (Central Greece). These are
situated at different altitudes around the former lake. The
principal outcrops are various limestones, flysch and schists.
We investigate the relationships between various variables such
as lithology, altitude, horizontal development and geographical
position of these karst forms. We proceed to a statistical
analysis of the above variables and we study the spatial
distributions of the swallets. The final results are indicative
of the development of these karstic forms and their dependence
on the various parameters. Zlokolica, M., Mandic, M. &
Ljubojevic, V. Some significant caves at the western rim of the
Miro? karst (Yugoslavia). pp. 77-88. The western rim of the
Miro karst oasis is composed of Lower Cretaceous sandstone,
marl, mudstone and limestone (K1,2), and Upper Jurassic
limestone (J33), and it abounds with rivers which sink into
caves of considerable dimensions. The deepest of them is the
pothole Jama u Lanitu (-272 m), while the longest is the cave
Buronov Ponor (1950 m). The caves abound with diverse forms of
passages ranging from almost unaltered fissures to
well-rounded phreatic tubes. The phases of fluvial
sedimentation, of rock collapses, of formation of travertine
coating, and of hydrologic reactivation and emptying of the
cavities, can be distinguished as well. Stevanovic, Z.,
Dragisic, V., Dokmanovic, P. & Mandic, M. Hydrogeology of
Miro? karst massif, eastern Serbia, Yugoslavia, pp. 89-96. The
karst massif of Miro? mountain is one of the national parks of
Yugoslavia, situated in the northern extremity of the
Carpatho-Balkanides of eastern Serbia and extending to the
Danube and the Romanian border. The area of the calcareous
massif, composed mainly of Tithonic limestones, exceeds 120
km2. Four karst springs used to drain a karst aquifer located
in the north and the west of the massif, along the Danube,
before being submerged by the reservoir behind the dam at
Portile de Fier (The Iron Gates). To the south, the massif is
drained by Blederija spring (Qav. about 30 l/s). The massif has
a multitude of active concentrated swallets. Several deep caves
(e.g. Rakin Ponor, 278 m deep, the deepest in Serbia) were
recently investigated by speleologists. Explorations have been
undertaken in Miro? area for karst ground water intake aiming
at the water supply of Donji Milanovac resort center. Samani,
N. & Ebrahimi, B. Analysis of spring hydrographs for
hydrogeological evaluation of a karst aquifer system. pp.
97-112. Karst aquifers are, in general, extremely heterogeneous
in character. As a result, quantitative data obtained from
selected points in the system, either by pumping or by using
marker dyes, can be rarely extrapolated to evaluate the average
function of the system as a whole. In contrast, recession
curves of spring hydrographs show the overall response of the
aquifer to precipitation events and help evaluating aquifer
storage and drainage potentials. In this study, three analytic
equations are used for estimating hydrogeological
characteristics of a karst aquifer: Mangin's equations, a newly
proposed equation that assumes that the recession curve is
composed of both quickflow and baseflow, and Coutagne's
equation, which considers the recession curve to be the
response of a single reservoir. The hydrographs of three
springs draining Sabzpooshan and Beaza aquifers in Fars
Province, Iran, are analyzed. From the analysis, it is apparent
that the saturated zone accounts for 99%, 91% and 99%,
respectively for Pirbanoo, Polberengi and Ghorehdan springs.
Iurkiewicz, A., Voica, M. & Bulgar, A. Indirect evaluation
of the Izvarna karst system discharge trend (Romania). pp.
113-120. The importance of Izvarna karst system stems from the
fact that its water is being tapped for the supply of the city
of Craiova. The hydrodynamic behavior of the aquifer under
natural conditions has been investigated based on records from
1957-1965. The development of the Cerna-Motru-Tismana
hydropower system has induced changes in the surface streams
discharge regime and hence possible consequences on the karst
aquifer supply. Compared to an accurate water budget
evaluation, the correlative and spectral analysis of the
discharge is a cost?effective method for the identification of
the subsurface flow evolution trend induced by human activity.
The paper illustrates the ability of the method to provide a
quantitative evaluation of the water resource reduction as well
as the limits of the adopted approach. Iurkiewicz, A.,
Dragomir, G., Rotaru, A. & Badescu, B. Karst systems in
Banat Mountains (Resita-Nera zone). pp. 121-140. The Banat
Mountains area, located in the south-west of Romania includes
more than 800 sq. km of karst zones belonging to the Resita -
Moldova Noua large synclinorium. Although scientific karst
investigations were initiated at the beginning of the century,
the previously collected hydrogeological data were rather local
and contradictory. Taking into account groundwater resources
that karst areas should normally contain, "Prospectiuni"
company has recently started a complex research program, in
order to evaluate the amount of groundwater resources of this
particular area. The karst territory was divided as a result
into three distinct zones, which correspond roughly to the
Caras, Minis-Nera and Nera-Danube catchment areas. The
hydrologic, hydrodynamic and hydrochemical characteristics of
the karst systems included within the first two zones are
discussed. The water budget completed based on 1992-1995 data
provides information on underground hydrologic connections
between small catchment basins, and eventually for each of the
above mentioned areas as a whole. Ristic, V., Prohaska, S.
& Stevanovic, Z. Some examples of karst springflow regime
simulation and prediction for water management balance
analyses. pp. 141-148. Karst occurrences in stream basins have
not been given adequate consideration in the applied hydrology.
As a consequence, water table and discharge of karst springs
are not measured at stations of the official
hydrometeorological network. For a proper assessment of the
karst water regime, under the given study conditions, indirect
methods and computation procedures have to be used. One of
these is MNC (Multiple Nonlinear Correlation) model, used for
filling-in, extending and three-dimensional interpolation of
the hydrometeorological data. Based on this model, in the
general areas of the karst springs (Izvor-Sveta Petka, Mlava,
Grza, Nemanja, Mrljis with Groznicevac) on Kucaj and Beljanica
mountains, eastern Serbia, this paper presents filled-in (where
data are lacking) and extended series, which are essential for
the investigation of the karst springflow regime, or for a
water budget analysis that extends over several years.
Stevanovic, Z., Markovic, M., Pavlovic, R., Jemcov, I.,
Dragisisic, V. & Cupkovic, T. Correlation between regional
fault pattern and karst water flow directions based on examples
from Eastern Serbia (Yugoslavia). pp. 149-154. The analysis of
the fault pattern of eastern Serbia was made as a part of the
basic and purpose detailed hydrogeological investigations in
some localities. By correlating the observed regional features
with the point elements of the principal drainage systems, a
significant coincidence of positions has been noticed. The
strikes of the main faults, longitudinal NW-SE and younger
NE-SW (W-E), largely coincide with the principal, general
directions of karst ground water flows. Subjects of particular
analysis were the calcareous massifs corresponding to the
anticlinoria of Ku?aj and Beljanica, and of Rtanj and Ozren in
central Carpatho-Balkan mountain range of eastern Serbia.
Filipovic, B., Lazic, M., Vidovic, M. & Sekulovic, B.
Tapping and protection of Maljen Spring, on Mount Durmitor,
Yugoslavia. pp. 155-164. Karst springs Veliki and Mali Maljen
emerge, 30-35 m apart at the foot of one of the slopes of the
mountain Durmitor at the altitude of 1320 m, approximately 15
km south of ?abljak, one of the largest winter resorts in
Montenegro and Yugoslavia. The springs drain a vast karst
reservoir consisting of Triassic carbonate rocks: they are of
ascending type with the outflow zones masked by moraine
material. The spring Veliki Maljen is tapped, whereas Mali
Maljen still runs through glacial material. The discharge of
the spring varies between a maximum of 150-200 l/s and 35-40
l/s indicating a characteristic karst regime. According to
their physical and chemical properties, the waters are of
exquisite quality and therefore they absolutely meet any norms
or regulations for bottling, that is foreseen in the future. In
order to determine the conditions of exploitation and of
preservation of the water qualities, a necessary
hydrogeological research has been carried out. It has been
performed according to the regulations for establishing the
boundaries of the sanitary protection zones for maintaining the
necessary continuity during the building and the operation of
the plant for bottling the natural drinking waters that flow
through Veliki Maljen spring. Raeisi, E., Jehbez, O. &
Moore, F. Hydrochemical behavior of karstic and evaporitic
formations surrounding Sarvestan Plain, Iran. pp. 165-174. The
Sarvestan Plain is located in the central-southern Iran. The
plain is surrounded by extensive karst carbonate deposits with
minor amounts of evaporitic-argillitic sediments. These
Mesozoic and Cenozoic formations are intruded by two
Precambrian salt domes to the northeast and southeast of the
plain. Despite the presence of karstic formations in the
drainage basin of the plain, the quality of the ground water is
very poor. In this study the lithological characteristics and
mineral constituents of the surrounding formations are
investigated by optical and chemical methods. During the summer
1992, the electrical conductivities of 500 operational wells,
15 springs, and 4 qanats were measured. Furthermore, 128
representative samples were analyzed for major ions including
calcium, magnesium, sodium, potassium, chloride, bicarbonate,
and sulfate using standard wet chemical methods. Temperature
and pH of all the sampling sites were also measured. Using the
WATQEF computer program, the saturation indices in various
minerals were calculated. Results indicate that although the
drainage basin of the plain consists mostly of carbonate
formations, except for a narrow band in the south and a small
region in the northeast of the plain where the type of water is
bicarbonate, the rest of the aquifer exhibits a
chloride-sulfate type water. The main sources of the chlorine
and sodium ions are shown to be the exposed salt domes, and the
sources of the sulfate are the dissolution of gypsum and
anhydrite along with the oxidation of pyrite in the
argillaceous formations. The reason for varying ionic
compositions in different parts of the aquifer is discussed.
The main reason for the poor quality of groundwater is
apparently the mixing of the karstic water with water expelled
mainly from salt domes and to a lesser extent from evaporitic
and argillitic units. The higher solubility of halite minerals
compared with that of carbonates, contributes to the poor
quality of the groundwater. Gaspar, R. Pulse temperature and
conductivity analysis in pseudo-karst structure investigation.
pp. 175-184. The permeable zones and associated leakage through
the pseudokarstic structure of Racova reservoir were determined
by using natural tracers: electrical conductivity, TDS and
water temperature. The variation of these parameters was
measured in piezometers, boreholes, subaerial and underwater
springs. To check the flow paths determined by natural tracers,
artificial labelings were performed. Oraseanu, I. Contributions
to the hydrogeology of the karstic areas of the Bihor Vladeasa
Mountains (Romania). pp. 185-214. The Bihor Vladeasa Mountains
are ones of the most important karst regions in Romania. They
have important groundwater resources, unexploited till now. On
the hydrogeological map there are pointed the main karst
springs with their magnitude of discharges and the directions
of the groundwater flows. More than 45 tracers labellings were
performed and an average of 45 m/hour of flow velocity was
recorded in 42 of these. As a consequence of the great
diversity of the geological constitution and intense
fracturation of the rocks, the karst systems are of binary
type, with a large variety in size and hydrological parameters
distribution. Due to the consistent observations and the
hydro-meteorological measurements, the groundwater resources
were evaluated and the processing of the flow rates series
outlined a great diversity of the karstic systems. The quality
of the groundwater is very good as indicate the results of the
chemical and bacteriological analysis and the potential sources
of pollution are leaks. Diaconu, G. Points de vue concernant le
clivage de la calcite. pp. 215-218. A l'aide des quelques
images obtenues au microscope lectronique en balayage,
l'auteur exprime un point de vue original concernant le clivage
de la calcite. Iurkiewicz, A., Badescu, B. & Marinica, E.
Intensity of karst processes as a function of the carbonate
formations in the north Resita-Moldova Noua Synclinorium. pp.
219-222. The exploration of the karst cavities in the
Resita-Anina area led to the identification of a significant
amount of "karst entrances", consisting mainly of caves and
potholes, swallets and springs. The study of the
characteristics of the karst cavities, as well as of their
distribution with respect to the different types of carbonate
deposits, has resulted in a qualitative evaluation of the
karstification processes that have occurred in this specific


ISSN 1012-9308 ACADEMIA ROMANA InstitutJl de Speologie ,Emil Racovita"




THEORETICAL AND APPLIED KAHSTOLOGT vol. 9 .. 1996 ... Director: dr. Radulescu EDITORIAL BOARD: + in-Chief"_. loan POV + kfembers: A,(Jrl.-niUIU(IEWICZ vior llor{of . .. + Associated Ed.ilor: ADVISORS BOARD: uort& .MI'fRoFAN E.iul:SILVESJRlJ . : SllViuCO:NSTANTIN Nicolae ANASTASIU (professorpfGeolpgy, BJ4cJiarestUnivetsiiy, Romania) .. Marcian BLEAHU. (prafesior of Geology, iJuch4fest) . . .. .. : Gabriel DIACONU (geologist, he(;u[ofthe "Geospdeology"Section, ''Emil RacoviJa" Speleological Institute-Bucharest, Romania).. ... ... ... .. ... -.-.. _. ... _. . . . .. Claude DROGUE (professor of Hydrogeology, Univei:sityof Montpellier; France) _. .. .. Emilian (senior scientist,lrutituteforPhysics aruJNuclea,rEngineefing, l)ubharest, Romania) . .. . . < > .-.. .... __ .... . _ CristianGQRAN (geographer, head ofthe :'Regiona:lf .. . .. ... . . . Grigore POSEA (professorofGeomorphology, B(lc/iarestJ]niversity,Romania) Gheorghe RACOVITA (senior scientist, '1EmilRacovija" Speleologic(llfnstituie -. .. Cluj-Napoca b"ranch, Romani(/,) : .. . _... . ': ... ..: . Costin RADULESC1J( senior scintist, "EmilRacoviJa Speleolqgicat1nstitute . Romania, co:Presponding memberofthe RomanianAciidemy) -. .... _... Florian ZAMFIRESCU (professor of Hydrogeo/dgy, Bucharest University, Romania) ISSN 1012-9308 Orders from abroad for issue or subscription may be also sent to: RODIPET S.A. or to ORION PRESS. Orders from Romania may be also sent to RODIPET S.A., ORION PRESS or AMCO PRESS. RODIPET SA, Piala Presei Libere 1, sect. 1, P.O. Box 33-57, fax; +40-1-2226407, teL:+40-6185103;.+40-1-2224126, Romania. ORION PRESS INTERNATIONAL S.R.L., $OS. Olteni\ei 35, sect. 4, P.O, Box 61-170, fax:+40-1-3122425; +40-1, tel:+40-1-6346345, Romania. AMCO PRESS S.R.L., Bd. Nicolae Grigorescu 29A, ap.66, sect. 3, P.O. Box 57-88, fax: +40-1-3125190, te1:+40-1-6439390; +40-1-3125109, Romania. EDITURA ACADEMIEI ROMANE Galea 13 Septembrie 13, R-76117 Bucurel?ti, Romania,P.O.Box 5-42 tl: +40-1-410.32.00




Cover photographs: Front : The Danube "Iron Gates" Gorge at Dubova, a majestic karst landscape carved by the river into Southern Carpathians range. Back : Aragonite speleothems from a protected cave in Apuseni Mountains (approx. 40 em in length) (top). Corosionalfeatures in Martel Cave (Mehedinti Mountains) (bottom). (all photographs by Cristian LASCU) THEORETICAL AND APPLIED KARSTOLOGY is an annual bulletin of karst science edited by the "Emil RacoviJa" Speleological Institute and published by the Romanian Academy's Publishing House. The bulletin is concerned with the publication of original scientific papers which cover the whole spectrum karstological research viz: karst geology and mineralogy, karstochemistry, karst geomorphology, karst hydrology and hydrogeology, speleochronology, climate and subterranean environment, speleo-paleontology, environmental problems in karst a.s.o. The Editors welcome the submission contributions in thesefieldsfiom all over the world. The first 4 volumes were published as a series by the ,Emil RacoviJa" Speleological Institute and are available by the Editorial Board. For the foreign readers, TAK volumes may obtained by exchange with specialized publications. All the correspondence concerning the exchange proposals or the submission contributions should be addressed to: T.A.K. EDITORIAL BOARD INSTITUTUL DE SPEOLOGIE "Emil Racovita" str. Frumoasa 11, R-78114 12 ROMANIA W: +40-1-650.34.65; fax: 40-1-2113874 Printed by "ILEX"-Bucharest Tel./ Fax: 653.21.89 DTP Editor: Silviu Constantin


Theoretical and Applied Karstology, vol. 9/1996 ( CONTENTS I SOMMAIRE) EDITORIAL................................ ................................................................ 7 TAK ARTICLES Bogdan-Petroniu ONAC & Stein-Erik LAURITZEN -The climate of last 150,000 years recorded in speleothems: preliminary results from north-western Romania......... ........ ....... ........................... .................. 9 Michel POURCHET, Marie Antoinette MELIERES, Emil SILVESTRU, Geza RAJKA, F. CANDAUDAP & Jean-Pierre CARBONNEL Radionuclides in a caves lake sediment core from Ghetarul de sub (Romania) ...... .. ........ ..... .. ... ...................... ............................ 23 Vasile LABAU, Emilian GASPAR & Tatiana PAUNICA-Speleothems dating using the thermoluminiscence method ........................ .... ......... 29 Erik CARLIER Effet d'echelle de Ia dispersion dans un conduit karstique. Hypothese lineaire et hypothese fractale ............ ......... ...... 35 Erik CARLIER Influence de la structure sur Ia variabilite spatio temporelle du coefficient de dispersion. Cas theorique d'un faisceau de conduits karstiques soumis a une injection continue en absence de diffusion moleculaire (ecoulement rapide).......................................... 45 Borivoje F. MIATOVIC Relation entre les caracteres physiques et lcs champs de leur distribution dans les roches karstiques.. ........ ...... .. .. ... 51 Kyriaki A. PAPADOPOULOU Comparative observations in karst formations of Parnass and Arcadia Mountains (Greece) ..................... 63 Kyriaki A. PAPADOPOULOU & Theodore GOURNELLOS-A study of the subsurface karst of Kopais (Central Greece)....... ...................... 69 Milena ZLOKOLICA, Mihajlo MANDIC & Vladimir LJUBOJEVIC Some significant caves at the western rim of the karst (Yugoslavia) ........................................................................................ 77 Zoran STEVANOVIC, Veselin DRAGISIC, Petar DOKMANOVIC & Mihajlo MANDIC Hydrogeology of karst massif, eastern Serbia, Yugoslavia ............. 89 Nozar SAMANI & Babak EBRAHIM! Analysis of spring hydrographs for hydrogeological evaluation of a karst aquifer system ............... .... 97 Adrian JURKIEWICZ, Mariana VOICA & Alexandru BULGAR Indirect evaluation of the Izvarna karst system discharge trend (Romania)................... .............. ..... ............................................ .......... I I 3


6 Adrian IURKIEWICZ, Gigi DRAGOMIR, Aurel ROTARU & Bogdan BADESCU Karst systems in Banat Mountains zone) ................................................... : .................... Vesna RISTIC, Stevan PROHASKA & Zoran STEVANOViCSome examples of karst springflow regime simulation and prediction for water management balance analyses ................ .... .. Zoran STEVANOVIC, Miroslav MARKOVIC, Radmila PAVLOVIC, Igor JEMCOV, Veselin & Tomas CUPKOVIC Correlation between regional fault pattern and karst water flow directions based on examples from Eastern Serbia (Yugoslavia) ..... ........... : ............................ ............... ...... .. Budimir FILIPOVIC, Milojko LAZIC, Mladen VIDOVIC & Branislav SEKULOVIC Tapping and protection of Maljen on Mount Durmitor, Yugoslavia ... .... ; .......................... .. Ezzat RAEISI, Omld JEHBEZ & Farid MOORE Hydrochemical behavior of karstic and evaporitic formations surrounding Sarvestan Plain, Iran ................................ ............................... .... Radu GASPAR Pulse temperature and conductivity analysis in pseudo-karst structure investigation ............................................. lancu Contributions to the hydrogeology of the karslic areas of Lhe Bihor Vllideasa Mountains (Romania) .......... Gabriel DIACONU-Points de vue concernant le clivage de la calcite Adrian IURKIEWICZ, Bogdan BADESCU & Elisabeta MARINICA Intensity of karst processes as a function of the carbonate formati ons in the north Nouli Synclinorium ................................................................................ 121 141 149 155 165 175 185 215 219 Contents


Theoretical and Applied Karstology, vol. 91)996,. pp. Z-:-8 .. EDITORIAL As we all know, this year, I997, has a particular significa nce for the Romanian speleologists. One hundred years have already elapsed since Emil RacoviJ{i, our emblematic eminent scientist, was asked, on the proposal of his French mentor HenH de Lacaze-Duihiers, to take part, as a zoologist, in the well-known Belgian expedition to the South Pole (I 897-I 899 ). A.f;er this temerarious scientific expedition, having a great experience in the realm of Natural Sciences, Emil RacoviJa discovered the .fascinating subterranean world during his .faunal investigations in caves of the Balearic Island.'l. In his famous "Essai sur les problemes biospeologiques ", published ninety years ago ( 1907 ), he founded a new branch science, the Biospeleology At last, let us recall that in November I997, we shall commemorate .f{fty years since Emil RacoviJa (I 868-I947) passed away in Cluj, the city hosting the .first Speleological Institute in the world, ./(Junded by this great man. His influence on the Ronumian scientUic community was and his works continue to shape our to orient our idea to new achievements and direct our organizational structure and scientific philosophy. This issue consists of the Proceedings of the X/Vth Symposium on Theoretical and Applied Karstology held on May 26June I, I996, in Baile Herculane. Annual meetings have been taking place since /983 and have provided an important .fbrum for a regular exchange information and views on current scientific activities concerning Speleology and Karstology. This issue Theoretical and Applied KarstoloRY includes important data on past climatic changes, evolution karst Landforms and on a broad variety of natural and human induced phenomena Linked to kar.'lt hydrogeology At the same time, we have to propose the appropriate measures that can help better emergin11 needs and challenges to basic questions. I Afieldtrip guide to the Symposium is also worth to be m e ntioning. This valuabl e guidebook, authored by our colleagues A. Jurkiewicz, S. Constantin and I. Povara, was devoted to the karst ofsouthwestern Romania inc:Luding a survey ofBanat Mountains with a special view on the thermo-mineral sources in Baile Herc:ulane area.


8 Editorial A welcome counterpart entitled "Karst' of northeastern Serbia" was authored, with this occasion, by our colleagues from University of Belgrade. This Monograph-Guide for excursion of the X/Vth International Symposium held in Baile Herculane, published by the Faculty of Mining and Geology, Institute of Hydrogeology, Belgrade (editor Professor Zoran Stevanovic) offered the participants rich and very interesting information concerning both the geological outline and hydrogeology of northeastern Serbia and other spec(fic topics. We are glad to welcome here the presence at the XfVth Symposium, ides the participants from European countries, of scientists from Iran (Shiraz University) who exposed and discussed topics connected to karst aquifer systems, drainage basins and quality of groundwater It worth emphasizing that many important mutual interests were discovered. Profound and new insigh,ts into the scientific achievements and future potential 4 the Symposium's deliberation.r were shared by all participants interested in the vast domain Speleology and Karstology. It is beyond any doubt that scientists cannot be isolated from the social and economic processes that are taking place around them. They have to examine and sometimes to rearrange their priorities, especially during periods contraction the national funding. Problems groundwater pollution, for instance, are of great concern not only to hydrogeologists, but also to cavers, since groundwater pollution can affect caves and the animal communities who live there. It is interesting to underline that topics that were almost exclusively con:fined to scientific: circles are now being dealt with by many governmental and non-governmental, national and international organizations. It is obvious that purposes of science are constantly changing and that science itself remains an outstanding agent of change at a world-wide scale. In recent years there has been dn increasing recognition the importance of scientific research in the future protection and conservation of the environment, the karst regions in particular. A major goal of our Symposia has been to encourage and sustain research and creative ideas in this area We hope that the issue of Theoretical and Applied Karstology is an eloquent illustration love of things having to do with nature. Costin RADULESCU


Theoretical & Applied Karstology, vol. 9/1996, pp. 9-21 TAK ARTICLES The the Last 150,000 Years . Recorded in Speleothenls : Prelimilliry Results from North-Western Romania Bogdan Petroniu ONAC1 & Stein-Erik LAURITZEN2 ABSTRACT Seventy-four 231Th/234U dates on speleothems enable preliminary estimates to be mape of the palaeoclimatic evolution of the north-western part of Romania. The growth frequency of speleothems was examined using the cumulative distributed error frequency method (GORDON, 1989). The climatic oscillations that are suggested by our curve correlate tentatively with known climatic events provided by pollen analysis, coral growth frequency and oxygen isotope records from both deep sea sediments and ice cores during the same timespan (the last 150,000 years). Key words: Speleothems, UraniumThorium dating, Quaternary palaeoclimate, Romania. Le clima(des der'niere 150.000 annees reflete par lesspeleothemes: resultats preliminaires dans le nord-ouest de Ia Roumanie RESUME 74 datations 230Th I 2'"u ont permis des estimations preliminaires sur I' evolution du dans La partie nord-ouest de La Roumanie. !.Afrequence de croisscmce des speteothemes a ete examinee en utilisant La methode de La distribution cumulative de lafrequenc:e des erreurs (GORDON, 1989) Les oscillations climatiques suggerees par notre courhe peuvent etre correlees avec les evenements climatiques mis tm evidence par l'anal y se du pollen La frequence de croissan c e des coraux et par l'intermediaire des enregistrement des i s otopes de I' oxygene dans les se diments de hassin profond et dans des corps de glace durant lu meme periode ( les dentiers 150.000 annees) Mots-eMs: speleothemes, datation U-Th, pateoclimat du Quaternaire; Roumanie. I. INTRODUCTION Except for the measurements made over the last 200 years or so at various meteorological stations, our knowledge of past climates is based on records in ice, in marine and continental sediments and in caves. The Quaternary period is characterized by tremen dous fluctuations of the global climate that induced changes in biota, landforms, sedimentation patterns, and oceanic and atmospheric circulation. In the . . mid-and high-latitudes, ice sheets and valley gla ciers waxed and waned, and the areas affected by periglacial processes expanded and contracted Throughout the world, weathering rates and pe dogenic processes varied with changes in tem perature and precipitation, river regimes fluctuated marktdly, sea-levels rose and fell several times These processes have left a vast variety of geo logical and geomorphological evidence, from which it has been possible to reconstruct the pa laeoclimatic evolution of this period. 1 Institutul de Speologie "Emil Racovit!", Sectia Cluj, str. Clinicilor 5, R-3400 Cluj-Napoca, Romania. 2 Department of Geology, Bergen University, Allegt 41, N-5007 Bergen, Norway


10 .... ._. .. .. . _: ... .. .:_.:-. The aim of this paper is to provide the. first indicators of climatiC .its validity is U-Th dated time-scale of climatic change of the ascertained aridit can: then be tisedto eXtent the. last 150 Ka (1 kilo-annum = 1,000 yeats} for the.. climatic records. In this paper we summarize the north-western part of Romania and compare this yVi.del}ce for clirnatit change in marine and contirecord with other records. Ifthe speleothem record . nentat e0viicriments; speleothem data, and corroborates other well-documented and reliable correlate them with previously published records .... 13 .. . 15 . : : .. c Cllinatlcch8ngesdiuing the tate T(!ijiai-y arid as re-Age (Ma) : II. THE NATURE OF THE PALAEOCLIMATE RECORD A. MARINE RECORDS ... _: .. : The separation of ol{ygen isotopes plays an portant role in palaeoclimatology providing infor mation both about past ice volumes and about past temperatures. The tie between the isotopic compo sition of ocean water and ice volume stems from the fact that the snow which accumulates to make ice caps is depleted by several percents in heavy oxygen. Because of this, the residual water vapor in the air mass becomes progressively depleted in the heavy isotope These missing heavy molecules are stored in the sea, giving a slightly higher 180 to 160 ratio (DANSGAARD, 1954). Thus the greater the volume of ice on the continents, the higher the 180/160 ratio in sea water. . by varlmis marine r ecords (LAU. RITZEN, 1993). A: The variation of icerafted detritus .(IRD) during the past 6 .. tnlllion years (Ma) (from JANSEN & 1991); B: Oxygen isotope In ODP site (Mid-Atlanc (from A .... .. YMOet al., 1990); C: oXygen isotope yarlations for the last 600 )(8 depleting the familiar peaks and troughs of ihe last 17 oxygen isotope (from IMIJRIE 1984) . Les changements climatiques durant le Neozo'ique et le Plei$tocime, riveUs par ies differents enre.gistrements en milieu /narin (LAURITZEN; 1993). A: lA variiliion du detritus porte ppr Ia glace ( iRD) durant les derniers 6 millions d'annees (Ma) (du J4NSEN & SJ0 HOLM, 1991), 8: les variationS de l'isowpe d'oxygene dans le site ODP (ume du milieu de !'ocean 4tlantique) duram les dernier.e$ Ma (du RAYMO et (ll., 1990); C: varlcztions des isotopes de l'oxygene pour les derniers 600 Ka met tant en les maximums et mini mumsfamiliers des demiers 17 elopes pa,r les isotopes de l'oxygene (IMBRIE et al., .1984). The 180/160 variations in carbonate shells and foraminifera from sediments are a direct measure of the volume of water Stored on the continents as glacial ice, being a sensitive indicator of global climate change (SHACKLE TON & OPDYKE, 1973). An oxygen isotope trace through a core of deep ocean sediment therefore reveals a record of glacial/interglacial changes spanning, in many instances, the whole of the Quaternary (Fig. lB). Times of various glaciation (stages labeled by even numbers) are indicated by 180-enrichment in marine foniminifera: as a result of accumulation of 180-depJeted ice on the continents (SHACKLETON, 1969). Only for the last 600,000 years or so, the oxygen isotope studies have indicated 7 cold and warm periods (Fig. lC).


The climate of the last 150,000 years reflected in speleothems There is abundant morphological evidence from many parts of the world for variations in sea level during the Quaternary as a result of episodic conti nental ice formation. Glacial-eustatic changes dominate the sea-level history of the tectonically stable areas of the world, and in such regions, lengthy sequences of sea-level changes can be re constructed. dating of corals and speleo thems from fossil coral reefs and flooded caves in Bermuda revealed a chronology of eustatic sea-level fluctuations which spanned almost 200 Kyr (HARMON & al., 1978). The corals' growth was active at times of high sea stands. Thus, the coral growth frequency can be used as a palaeoclimatic indicator (SMART & RICHARDS, 1992). Evidence of several major climatic oscillations that have occurred during the past 6 million years was recently revealed by the occurrence of ice-rafted detritus in the Norwegian Sea (JANSEN & SJ0-HOLM, 1991) (Fig. 1A). The details of these dra matic climatic changes have been further confirmed by oxygen isotope and other deep-sea stratigraphies : B. CONTINENTAL RECORDS Records of climatic variation during the Quater nary are preserved in sediments that have been laid down in a variety of terrestrial settings: in lakes, glacial deposits, swamp and marsh deposits, soil, flora and fauna (archaeOlogical sites included) and cave deposits. Unfortunately, surface deposits are commonly altered by weathering and pedogenic processes and offer poor chances of preservation of evi dence. However, caves frequently provide suitable sites for the accumulation of chemical, detrital and organic deposits and in cave environments the effects of sub-aerial weathering are much reduced, and the chances of long term preservation of both sediments and fossils are greatly enhanced. Speleothem growth is a sensitive indicator of both temperature and aridity, and changes in either one or both of these can cause periods of enhanced or reduced growth (ATKINSON et al., 1978; LAU RITZEN & GASCOYNE, 1980; BAKER et al., 1993; GOEDE et al., 1990). Speleothems may be dated by uranium-series methods up to 350 Ka (HARMON et al., 1975; SCHWARCZ, 1989), and have proven to be important tools in terrestrial Quaternary studies. Thus, cave and cave deposits are a powerful tool for the study of past environ mental changes (LAURITZEN, 1993). 11 Ill. METHODS AND DATA ., Seventy-four ages were computed based on alpha particle spectroscopy at the U-series dat ing Laboratory at the Department of Geology, Bergen (Norway). Samples of carbonate speleothems (15 to 50 g) were dissolved in concentrated HN03 The spiked acid extracts were preconcentrated by scavenger precipitation; Fe(OH)3 then U and Th were sepa rated and pmjfied by ion exchange chromatogra phy and finally eleCtroplated on stainless steel disks and counted for alpha-particle activity with silicon barrier detectors in vacuo. A commercial spike was used as an internal standard, the activity ratio of which is calculated for each analysis. Raw counts were corrected for back ground and time elapsed between nuclide separa tion and counting and processed to fmal dates, using standard algorithms of IV ANOVICH & HARMON ( 1992), implemented in Turbo Pascal '9Qrle, by LAURITZEN (1993a) .. The errors of all included in this study are Jcr. Twelve preparations out of a total .of seventy-four, suffered from low U yields 10% ) probably due to the presence of humics that were insufficiently oxidized. These low chemical yields induced low radiometric yields, so all calculated ages for these samples gave larger error intervals (lcr) than the other dates. For the rest of the samples, chemical yields were moderate to high, and almost all spec tra were well resolved, yielding reliable dates. The results are listed in Table 1. Here we enclose all dates from our work, but only finite dates with finite uncertainties have been included when we discussed our results. Also, all dates provided by analysis for which the chemical yields are known to be less than 10% were excluded. Age corrections were performed on 11 samples that showed high detrital Th contamination ( < 20), assuming an initial activity ratio equal to 1.5. These reduced the corresponding ages by 9 to40% The distribution of the ages in time and with re spect to Pleistocene climate can be done by plot ting the dates in a histogram, or by using a prob af>ility density function approach (PDF) (GORDON et al., 1989). As radiometric ages are based on exponential relationship (i.e. first-order proc esses), the error intervals are unsymmetrical and increase (exponentially) with age. Plotting age distributions in histograms with constant class size would therefore be misleading; high ages will appear with too high precision, and younger


12 (precise) ages may become grouped into intervals .. that are too large to pay justice to the resolution. Therefore, ages should be transformed into a con tinuous PDF (GORDON et al., 1989), yielding a much more realistic picture of significance levels. The cumulative growth frequency record of a large sample collection can also be used to define the timing of warm and cold periods by taking the timing of the peaks and troughs as a simple binary signal (BAKER et al., 1993). Therefore, we will only present the PDF curve (Fig. 3C) and do a tentative interpretation. The significance of peaks and troughs in a PDF distribution is strongly dependent on the size of the underlying sample set. Probably, a sample of about 300 dates is required to this purpose so that the following discussions must be regarded as tentative while more dates are in progress. Twenty-one different speleothems from 11 caves located in Bihor and Plidurea Craiului Mountains were included for the present study (Fig. 2). Both caves and speleothems were carefully selected in order to provide an efficient database for climatic B. P. Onac & S.-E. Lauritzen and future geomorphologic interpretations. We sampled caves at different altitudes (between 200 and 1,200 m a.s.l.) in regions of well-documented morphotectonic evolution during the Tertiary and Quaternary periods. The sampled caves are: Lithophagus (Fig. 2a), Vintului (Fig. 2b), Vadu-Criului (Fig. 2c), (Fig. 2d), Vartop (Fig. 2e), Sdlrioara Glacier (Fig. 2f), Piatr.a Altarului (Fig. 2g), Hum pleu (Fig. 2h), Rotaridesz (Fig. 2i), Ponora (Fig. 2j), and Urilor (Fig. 2k). The first five of these caves and their speleothem ages were presented in a previous paper (LAURITZEN & ONAC, 1995). These dates will be added to the present data set without giving the description of the caves and speleothems. However, a brief presentation of the other five caves investigated is given below. Glacier Cave (Fig. 2t) is located in the central area of the Bihor Mountains at 1,156 m above sea level altitude. The cave shelters an ice block of 75,000 m3 having on both sides lateral galleries that contain both ice and calcite speleothems (RACOVJT.A, 1994). Two samples were collected


The climate Q/the last lSO,OOO years in sp(!leotlmns . 13. Cave Lab. Descriptio U (ppm) 2.l4'J'h/134U correet,ed age Obs. No. n (Kyr) -500 ,; . :: .' <:, .. --. 100 I top (>.126 0.734+/-0.034 1.83+/-0.084 25 125.3 flowstone (+I 0.48/-9.7) 1902 .. l\7:33 >;s. f :);-. 1 -= 1014 middle 0.09 flowstone I 034 middle 0.()36 flowstone : 1104 base 0 .092 s tala gmite 1212 .. LFG-2 0.044 I 118 5em 0.()78 from base li19 9em 0.()85 from base : I 123 7 em 0.084 from base nz2 11 em 0.089 from base 1113 13 em 0.066 from base 1112 15 em 0.072 from base 1120 base 0.072 (2nd lime) 1156 is em 0.055 1157 1159 from base 22em from base 26em from base 1158 29 em from base 0.062 0 .045 0 082 1121 37 em 0.08 from base I I 05 top !).()68 stalagmite 1116 base 0.085 stalagmite I 153 base 0 036 stalagmite 1155 47 em 0.037 from base i 154 74 em 0.046 fr o m bas e 1117 top 0.089 1003 1004 s talagmite top flowslone flow stone 0.082 0.118 0.793+/-0.05 I 876+/-0. I 11 ,.-: L5. ,;.: ,. 0 748+/ -0.08 I 0;967+1-0.05 1 681+/0 212 L35+/"0;07 > 1000 120 0.718+/-0.034 1.52+/ 0 062 > 1000 o :016 0 717+/ 0.035 1.66+/ 0.073 > JO(XJ >1000 :.;. 0.608+/-0 043 1.725+/> 1000 0 156 0.6+/-0.05 z :o25+t0.27 > 1 oob 0.754+/ 0.()48 1 768+/> 1000 0 084 0.927+/-0.057 1.422+/-IS 0 102 0 65+/-0 073 1.457+/ 4 0 .105 0.612+/ 0.03 1:773+/-0.()6 5 1.087+/-0 077 1.412+/-0.06 > 1000 0.448+/-0 041 2.837+1-0.2 > 1000 0.418+/-0 032 3 63+/ 0 27 > I O 1000 0.073 1 039+/ 0.03 2 1.447+/ 278 ().()46 1.074+/-0.044 1.383+/-0.05 > 1000 1.099+/ 0.0 3 1 1.399+/ 0.034 0.74+/-0.03 I 1.896+/0 073 0 377+/-0 .021 1.682+/ 0 .0 65 > 1000 > 1000 44 0.312+/-0.026 1 923+1 0 07 24 143.17 132.69 (+17.59/-15.5 _1) rejected . ,. ..-.-.. 131.32 ( +27 .85/-22.85) 253:04 . .c+ss:.w.3s:or 124.21 (+10.72/-9 87) 122.13 (+10.66/-9.83) 97. 1 ':: (9:7/-9.02) 93 73 (+10.42/9 63) 90.43 ( +11.37/10 .46) 131.98 (+15 .88/-14. 14) 216.55 208.02 (#0/-33.26) 106.{)) ( +20.66/-17 67) 94 .7) <+ > 350 60.33 (+7.14/-6.78) 54 98 ( +5.33/5.14) > 350 > 350 >350 > 350 126.63 I (+9.63t-8 .97> 49.72 (3.49/ 3 39) 39.36 (+3.94/ 3.83) 78 34 ( +24 79/-2 2 .03) 73 . 04 ( +8;69/8.34) I 005 middle 0 .081 0 6 04+/ 0 057 1 .453+ / > I()()() 94 69 under ._ _______ .__ ____ -------_______ -------------


14 1-0 ..... Cll 1-;:::, "' cpflow" stone 1021 base flo w -stone 1022 midle tlow -stone 1006 t o p flow stone 1093 base stalag mite 10 9 4 top sta1ag mite 1032 base_ stalag mite 1103 calcite cr stal 1035 top flow stone 105 3 t o p flow stone 1054 base flow" stone -----... ... .. . ;' -.. : . ; ().095 Q.188+/-0.!i21 1.652+/'-(l082 > I 000 :. 22.42 ; . .. (+2.8 .8/..:2.-81) --... : > _;; __ .. _ >;_ . ...... ... . _<--/.: :; __ .. -: . : :; .. ? 4 . 0.064 o.l63+t;o. o 11 1 3S4+t-o.oo3 38 .. 19.19 ..' Y.58s;tl6:()M: >.160o > ; : ... ::: :." ,., .. :' /':' c :.:,;;::..-.:., .;} __:\_r_ :,:9_ _.'6 : _.,_ .. )__ .. .. . 0.041 . : ;..' . .-:---. ... '' :.::>: T6f . 13.61 .. (+2.34/-2.30) 0.118+/-0.()19 1.256+/-0.(164 > 1000 .o: tl4+J.,o.ow 0 102+/-0.016 i .417+/-0.i 13 > JOof) : ... 36); ... 0 .032 11 ( +2.03/-L99) ();031 _; ___ . -. : / (+1;'671-1;65) 0.108 0 627+/-0.025 I j 19+/c0 .03 > 1000 0;()7 very low very low very low 0.070 .... ... +o:o1 0.073 0.259+/-0.02 2 66+/-0.2 0.065 o.o28+/0.007 2.087+/0 102 0;.401+1-0 ,()5 .-0.052 0.429+/0 051 2.489+/-0.156 9 0.045 . l.l(i4+/C{),26 1.5034-_/0.JOj 2 0.()35 0 .607+/0 .131 2 .069+/-0. 143 >1000 0.178f:/ O Of3 .. >.'30..:. 0 .<)98 .. 2.4J 0.071 0 .065+/-0.02 1 3 .321 +/-0.324 > 1000 0 .069 .2.56H0.23. >1000 0 296 0 .653+/0 .038 1 329+/-0.!i55 > 1000 O.D3'7 1 .82+/,0.375 > loOo 0 039 0 .65+/0 .049 I 952+1 0 149 2 0.037 0 ;726+/-0.0 W 2 : ----------_____ .;..... 57.35 ( +8.84/8 .30) re 50.32 (+9.81/-9.30) 46.92 (+19.51/ 1 9.11) 59;58 ......


. . The climate of the last 150,000years reflected in speleothems 15 . . .. . : . --- Rotaridesz P.Ne Van tulul Vadu Crl-Gala-Lilie ell or Li manu Herulan c 1051 base flow0 027 0.467+/-0.121 2.539+/-0.163 > 1000 63.8 stone (+23.0/-19 65) io52., tqp no\V" o :o#. ;71/: ; . :; :5o4i : .:. .. wm. e ; . ; ;.-:. i .. ._,: ;, .. __ . . . 1095 base stu0 039 0 129+/-0 .01 5 2 .4 04+/-0.13 > 1 ()QO 1.4.83 lag mite . . ( + 1.87/-1.85) .tQ96 tpp o;o:n : ; }Q'oo.\.' . .. .. rriite ... . :.- _.: (+:Z.I71.2:l3} 1219 base 0.035 0.214+/-0.021 1.758+/-0.152 22 25.77 (+2.93/-2.87) 1264. sc-3 top 1000 15.38 . lagmite . (+1.45/-1.43) 1I 1000 60.07 .: ,, .. ,_ 490 top 0:12 O.I8S4-/"o.m( > 1 6ot f .... n 4 '---. ; stone c> ;; .. :. . . : : . ,,_, .. '.'_, ( +4.2l/4.o6) 503 base flow0 02 0.9266 2.321+/-0 .17 2 2773 187 7 stone (+29.4/-24.2) 809 .. .. o :oi o .ss+t0.042 i .B63+t-ci. il o :7n4 '; _t6i.5. i flowstone" '; .-{+ I7.42,q 538 ) 810 0 .02 ::;.,. 118.7 .. .. .-'. :-. fractions 1231. YTP-8 .. L32+/ Q ; 365 .. 1 5 . :> 350 1233 VTP-8 0 024 2 639+/ 0 26 I 407+/-0. I 5 0 .74 . .. UlO base sta7 0.051 ;: 0.213+!-0 07 2.053+/-0,12 >1000;. ._.. .. .... :. .... : II II top stalag0 027 0 18+/-0 052 1 77+/ -0.131 7 mite 1213 fiowstone .... 0 035 . 0.838+1-0.063 95 inn '. 1214 0 045 I .058+/ 0.073 1 066+/ 0 068 > 5 out .491. baSe Stac 0 ;094 .. 0.i'l3+i-0.Q22 1.71+/-0 142 S ; :. lagmite 497 flow stone 0. 14 0 .281 +/-0 082 1.592+/ 0 027 > 1000 811. flo w stone 0;1 3 0 293+/ 0.017 1.871+/ 0 061 > 1000 498 flo wstonc 0.1 I 0.38+/-0 082 I 85+/-0.67 > I 000 489' stalactite 0;22 0.516+/-0 011 1 35+/ -0 .025 9 ... 1 211 crust 0 .605 0 959+ / 0 0 2 3 0 958+ / 0 0 1 9 83 I .. O.i.39. 1 .07-i+ / 0.05 i .428+/-0.06 1406 clusterite ? > 1000 .. . :--... .. > 350 25.51 . 21.32 16.81 (.+6. 88/6 53) ( +8. 23/ 7.92) 180 .21 . . > 350 33 8 ( +3 24/-3 16) 35 2 ( + 12.5/-1 1.4) 36.67 I <+2.5/2.45) 49.7 ( +13 .9/-12.6) 75.9 .. ( .. > 350 >350 24. 6 ( +4.07/ 4 0) ..


16 from "The Cathedral" Room (SC-1, a 15 em thick flowstone and SC-4, a 25-cm high stalagmite). The .... flowstone sample was rather porous and the resul tant dates of ca. 55 Ka show a large standard error. The stalagmite shows a pattern of growth layers consisting of thin gray opaque horiwns separated by thick white opaque layers. Both base and top of the stalagmite gave reliable ages of 14.8 Ka and 10.7 Ka, respectively. Piatra Altarului (Fig. 2g; 1120 m a.s.l.) is a well decorated cave that lies in the north-western part of Bihar Mountains in the upper part of SomeUl Cald catchment area. The samples were taken from a fossil phreatic gallery close to the cave gate and from the "Big Room". Both samples (PA-l and 2) consist of layered flowstone (3 and 6 em thick, respectively) with alternating compact and porous zones. The color varies from white to reddish brown. The results suggest that the two sub-samples cut from PA-2 were significantly contaminated. While the reliability of all three individual dates is low, the samples agree in giving ages within the last interstadial (isotope stage 3) Humpleu Cave (Fig. 2h) opens in the right side of Firei Valley at 920 m a.s.l. It has over 27 km in length and several huge rooms showing various types of carbonate speleothems. The following speleothems were collected from the "Metro Gal lery" and the "Pocket Room": two broken stalag mites (PH-1, 75 em long and PH-2, 22 em long) and a calcite crystal that was detached from the cave wall (PH-6) (10 to 15 em in length). Four samples out of the three speleothems have been dated and they yielded reliable ages ranging from 7 to 108 Ka (see Table 1). Rotaridesz Cave (Fig. 2i; 630 m a.s.l.) is located on the right side of Rea Valley, some 300 m up stream from its confluence with I ada Valley. Both phreatic and vadose morphologies are typical for this cave Important detritic deposits (even within the upper floors of the cave) suggest that the cave was flooded several times. Detritic flowstones were cemented in the outer part of some speleothems. One stalagmite (R-2, 32 em long) was sampled but the two sub-samples were quite different. The lower part was better crystallized, while the upper part was much more porous and contaminated. A tempo rarily submerged period is indicated by the layer of detrital clay separating the base of the stalagmite from the top part. Considering the detrital material cemented on the stalagmite surface, the assumption of a closed system is clearly invalid (at least for the period of the flooding) and the obtained ages (both with high errors) should be regarded with caution and treated as maximum ages. B. P. Ollllc & S.-E Lauritzen Cave (Fig. 2j) lies in the north-eastern ,'part of: Pldurea Craiului -Mounti!.ins 605 m a:s.l.) in the Dami-Ponora de-l pression The total length of its galleries is over ; 4.5 km (GYOROG, 'pers. comm ) The cave con: tains various types of carbonate ; speleothems as well as interesting gypsum crystals. PP-1, a 49. cm high stalagmite, was the main sample from Ponora cave. Unfortunately, patterns of internal coiTosion were found. For this reason ,' only 4 dates were taken from the base, 15, and 31 em from the base and the top. No detrital contamination was recorded for these four samples The basal sample consisted of clear, well-crystallized calcite and gave a reliable age of 16 Ka. The top of the sta lagmite was dated to 3 Ka, but the sub-sample extracted at 15 em from the base appears to be older than the base (32 ka). The corroded internal morphology indicates U leaching. All the other flowstone samples collected were more or less porous with interlayered compact calcite layers but of a dark brown color. PP-2a was the only date performed on a flowstone from Ponorll cave that gave a reliable age (50 Ka but with a large error). Urilor Cave (Fig. 2k; 489 m a.s.l.) is situated in the westernmost part of Bihar Mountains. It is the most important show cave in Romania. The two speleothems we analyzed were collected from the main fossil gallery (Candles' Gallery) PU-1, a fragment from a broken stalagmite, was dated bottom and top. The other set of dates presented here was performed on a 72.5 em long stalagmite (PU-2) that had already been removed when the cave was developed for tourism. The distinctive physical features of PU-2 stalag mite are the numerous clear transparent and white, opaque laminae and the evidence in some areas of internal corrosion. Samples from corroded regions were avoided. PU-2 covered a time span of ca. 20,000 years (Table 1). A section of this stalag mite, as well as the results of our UTh dating are now displayed in the Urilor Cave Museum. IV. PLEISTOCENE CHRONOLOGY IN NW ROMANIA No detailed data are available for Romanian Qua ternary climate and there is no general agreement on the number and timing of interstadial periods in the last 150 Ka However, an attempt to recon struct the main climatic events for this period has been performed by CARCIUMARU (1980). He drew a climatic curve based on pollen analysis,


The climate of the last 150,000 years reflected in speleothems 17 -2.5 -2.0 -1.5 -1.0 0 -0.5 0.0 r,o 0.5 1.0 1.5 2.0 2.5 5e (A) 1 ......---:; Growth discontinuities ... / : : SC:1H .. I I I I PA-2 PH-2 : ... +---4 'I PP-1 I I .. t-f.. I I PU-5 (B) PU-1 I 10 9 11 8 22 37 7 3 :;.... 6 C'J = 5 = 0" "" 4 3 2 1 0 0 LFG-4 LFG-2 .... 1----------1----------1 .. of I 51 50 I I LFG-1 1"11----------I I 76 (C) 100 125 100 150 200 Age (Kyr) ... A: Oxygen Isotope tnclml$ng the llltm1ll& wt. i _th_ active . . = 62). the '" isottJpes de intltthe


18 arid 14C oil curve does not correlate in any parts with our for Romania, nor with the orbitally tuned oxygen isotope chnmology. This may betheresult of the location of his sites in high altitude from Southern Carpathians and also ofa lack of reliabie : chionology. : '. : : .In. to previous dates by LAURI'fZEN & ONAC(l995) and by ONAC (199()),. we have added 24 new dates up to 1:1 total tif:62in the PDF distribution shown in Fig; 3,: arid we correlated chronostratigraphically with : pollen, mals, oxygen isotope .. stratigiaphy and c6ra1 growth .. frequency; The following observations emerge: ":" . . . .. . . .,. Between 15 arid 35 climatic sigrials proxy feeoras in the celitralpartofRolilariia (RAbfJLESCtJ .& SAMSON. 1992)' areiri close agreement with otir data. . In summary, the dates obtained cover a thne . high latitude data from Norway (LAlJRI'fZEN et span from late Riss/Wiirm (Eemian), through at., J990)-and also with .the contJ:overs1arDevil's Wiirm (Weichselian), up to late Holocene; and Hole vein .. calcite in USA (WINOGRAD ei a/,, compare well withsimilar records published by 1992). The periods ofnon-deposition GORDON et al. (1989) for UK and BAKEReial. tbems shown i.n agree well in' age with (1993) fot NW Europe. The higher thari 142.5Ka ... similar. determined ror age of the LPG sample is iri good agreement with Europe by different authors (ATKINSON it'al., ..


The climate of the last 150,000 years reflected in speleothems 1978; GORDON et al., 1989; LAURITZEN et al., 1990; BAKER et al., 1993). The growth ranges of individual stalagmites or flowstones are limited in time, but when seen collectively, speleothems growth is almost continuously through the investi gated timespan (150 Ka), (Fig. 3B). This appar ently continuous growth is unexpected and may also be an artifact due to the relatively imprecise alpha particle dates (BAKER et al., 1995). This will be further investigated by means of thermal ionization mass-spectrometry dating. V. CONCLUSIONS Six periods of enhanced speleothem growth were found over the last 150 Ka in north-western Ro mania by means of the speleothem growth fre quency curve presented here. The PDF curve of the whole collection of dates (Fig. 3C) displays peaks and troughs that correlate broadly with known climatic variations during the same period. However, the timing of some cli matic events did not entirely coincide probably due to the geographical position of Romania, which was far south of the maximum limits of the Pleistocene ice sheets. Due to the low number of samples (n = 62) none of the peaks are significant from a statistical point of view. This problem will be solved when more dates become available, and the present correlation must be therefore regarded as tentative. REFERENCES ATKINSON, T., C., BRIFFA, K., R. & COOPE, G., R. (1987) Seasonal temperatures in Britain during the past 22,000 years reconstructed using beetle remains. Nature, 325, pp. 587-592. ATKINSON, T ., C., HARMON, R., S ., SMART, P. L. & WALTHAM, A., C. (1978) Paleoclimatic and geomorphic implications of 1rht-34U dates on speleothems from Britain. Nature, 212, pp. 24-28. BAKER. A SMART, P., L. & FORD, D., C. (1993) Northwest European paleoclimate as indicated by growth frequency variations of secondary cal cite deposits Palaeogeography, Pala eocl imatol ogy, Palaeoecology, 100, pp. 291-301. BAKER, A. & SMART, P., L. (1995) Recent flowstone growth rates: Field measurements in comparison to theoretical predictions. Chemical Geology, 122, pp. 121-128. BIRKS H H., PAUS, A., SVENDSEN, J., L., ALM, T MANGERUD, J & LANDVIK, J. Y (1994) Late Weichselian environmental change in Norway, 19 The speleothem growth frequency may provide the. most complete record of interstadial and intergla.:. cial periods for the last 350 Ka in Roqtania, not only because speleothem growth js-a ; sensitive indicator of both temperature and humidity, but it; has the advantage of being calibrated by a reliable radiometric time scale (GORDON etal., 1989). '' The geochronology of the Romanian Quaternary has yet to be established and, in view of the fact that the proxy climatic record from Europe is un derpinned almost entirely by conventional or AMS 14C and U-Th da,tes, a more complete radiocarbon, and UTh time-scale for the Romanian Pleistocene is required. ACKNOWLEDGEMENTS We wish to thank M. Gligan, F. Papiu and V. Laszl6 for providing access and collecting permit into Piatra Altarului, Humpleu and caves We are grateful for the patience and cooperation of M. Vremir and 0. who provided in-cave assistance. Dr. Joyce Lundberg is thanked for critically reading an early draft of the manuscript and for improving the English language. This is the 2nd contribution to the Speleothem Pole-Equator-Pole Transect III (SPEP III) program in Romania and it was done while one of the authors (B. Onac) was staying at the Department of Geology, Bergen University in 1995 and 1996. including Svalbard. J Quaternary Sci., 9, pp. 133-146. BLEAHU, M., D. (1972) Karst of Romania In: Karst: Important Karst Regions of the Northern Hemi sphere. (HERAK, M. & STRINGFIELD V ., T Eds.), Elsevier, Amsterdam, pp. 141-151. BOHNCKE, S., J. P. (1993) Late glacial environmental changes in the Netherlands : spatial and temporal patterns. Quaternary Science Revi ews 12, pp. 707-718. BOND, G., BROECKER, W ., S., JOHNSEN S., J., McMANUS, S ., LABEYRE, J., JOUZEL, J. & BONANI, G (1993) Correlations between cli mate recoq:ls from north Atlantic sediments and Greenland ice. Natu re, 365 pp. 143 147. BORDEI, E., I. (1983) R olulla nJului alpino-carpatic In evoluJia cic:lon ilor mediteraneeni Ed Academiei R.S.Romania, Bucureti. BURGHELE, A., RADULESCO, C., SAMSON, P. M ., E. (1994) La premiere faune intergla ciaire (Eemian ? ) de micromarnmireres d'Olt e nie. Trav. Jnst. Sp e o l "Emile Ra c ovit z a" XXXIII, pp. 101 106.


20 M. (1980) Mediul geograjic In Pleis tocenul superior culturile paleolitice din Romfinia, Ed. Acad. RSR, Bucureti, pp. 12-60. CORDONEANU, E. (1995) Particularitl1Ji ale dinami cii aerului deasupra Romaniei utilizand un model numeric al atmosferei. Ph.D. Thesis, Uni versitatea Bucureti. (Abstract). DANSGAARD, W. 1954. Oxygen isotopes in fresh water. Geochimic:a et Cosmochimic:a Acta, 6, pp. 241-260. DANSGAARD, 1., J., JOHNESEN, B., J., CLAUSEN, H., B., DAHL-JENSEN, D., GUNDENSTRUP, N., S., HAMMER, C., U., HVIDBERG, C., S., STEfFENSEN, J., P., SVEINBORNSDOTIIR, A., E., JOUZEL, J. & BOND, G. (1993) Evidence for general instability of past climate from a 250Kyr ice core record. Nature, 364, pp. 218-220. GOEDE, A., HARMON, R., S., ATKINSON, T., C. & ROWE, P., J. (1990) Pleistocene climatic change in southern Australia and its effect on speleo them deposition in some Nullarbor caves. Journal of Quaternwy Science, 5 {1), pp. 29-38. GORDON, D., SMART, P., L., FORD, D., C., AN DREWS, J., N., ATKINSON, T., C., ROWE, P., J. & CHRISTOPHER, N., S., J. 1989. Dating of late Pleistocene interglacial and interstadial periods in the United Kingdom from speleothem growth fre quency. Quaternwy Research, 31, pp. 14-26. GUIOT, J., PONS, A., de BEAULIEU, J., L. & REILLE, M. (1989) A 140,000-year continental climate reconstruction from two European pollen records. Science, 338, pp. 309-313. HARMON, R., S., THOMPSON, P., SCHWARCZ, H., P. & FORD, 0., C. (1975) Uranium-Series Dat ing of Speleothems NSS .Bull. 37 (2), 21-33. HARMON, R., S., SCHWARCZ, H., P. & FORD, D., C. (1978) Late Pleistocene sea level history of Bermuda. Quaternary Research, 9, pp. 205-218. IMBRIE, J., HAYS, J., D., MARTINSEN, D., G MciNTYRE, A., MIX, A C., MORLEY, J., 1., PISIAS, N., G., PRELL, W., L. & SCHACKLE TON, N., J. (1984) The orbitally theory of Pleistocene climate: Support from a revised chronology of the marine o180 record In: Mi lankowich and Climate Part I. (L. BERGER et al. Eds.), Reid! Pub!., pp. 269-305. IVANOVICH, M. & HARMON, R., S. (1992) Ura nium-series disequilibrium. Applications to Earth Marine, and Environmental Sciences. Clarendon Press, Oxford. 910 p. JANSEN, E & SJ0HOLM, J. (1991) Reconstruction of glaciation over the past 6 Myr from ice-borne deposits in the Norwegian Sea. Nature, 349, pp. 600-603. B. P. Onac & S.-E.Lauritzen LAURITZEN, S.-E. (1993) Natural environmental change in karst: the Quaternary record Catena supplement, 25, pp. 21-40. LAURITZEN, S.-E. (1993a) "Age4U2U". Progtamfor reading ADCAM energy spectra, integration peak-correction and calculation of 230Thoriumi3"Uranium ages. Department of Ge ology, Bergen, 5000 lines. LAURITZEN, S.-E (1995). High-resolution paleotem perature proxy record for the last interglaciar based on Norwegian speleothems. Quaternary Research, 43, pp. 133-146. LAURITZEN, S.-E. & GASCOYNE, M. (1980) The first radiometric dating of Norwegian stalag mites-evidence of pre-Weichselian karst caves. Norsk Geogr Tidsskr., 34, pp. 77-82. LAURITZEN, S.-E., L0VLIE, R., MOE, D. & 0STBYE, E. (1990) Paleoclimate deduced from a multidisciplinary study of a half-million-year-old stalagmite from Rana, northern Norway. Quater nary Research, 34, pp. 306-316. LAURITZEN, S.-E. & ONAC, B. P (1995) Uranium series dating of some speleothems from Roma nia. Theor. Appl. Karstol., 8, pp. 25-36. MARTINSON, D., G., PISIAS, N G., HAYS, J., D., IMBRIE, J., MOORE, T., C. & SHACKLETON, N., J. (19870 Age dating and the orbital theory of the ice age: Development of a high-resolution 0 to 300,000-years chronostratigraphy. Quater nary Research, 27, pp. 1-29. ONAC, B., P. (1996) Mineralogia speleotemelor din unele ale Munfilor PCJdurea Craiului, cu referiri asupra semnificaJiei lor paleoclimatice. Ph. D. Thesis, Universitatea "Babe-Bolyai", Cluj, 200 p. PISIAS, N., G., MARTINSON, D., G., MOORE, T., C., SHACKLETON, N., J., PRELL, W., HAYS, J. & BODEN, G. (1984) High resolution stratigraphic correlation of benthonic oxygen isotopic records spanning the last 300,000 years. Mar Geol., 56, pp. 119 136. PONEL, P & COOPE, G R. (1990) Late glacial and early Flandrian Coleoptera from La Taphanel, Massif Central, France: climatic and ecological implications. Journal of Quaternary Science, 5, pp. 235-250. RACOVITA, G. ( 1994) Bilan climatique de la grotte glaciere de Sclrioara (Monts du Bihor, Rouma nie) dresse sur dix annees d'observations Trav. lnst. Speol. "Emile Ra c ovit z a ", XXXIU, pp. 107-158. RAYMO, M., E., RUDDIMAN, W., F., SCHACKLE TON, N., J. & OPPO, D., W. (1990) Evolution of Atlantic Pacific o1 3C gradients over the last 2 5 m.y. Earth and Planetary Sci e nce Letters, 97, pp. 353-368.


The climate of the last 150,000 years reflected in speleothems R.ADULESCU, C. & SAMSON, P., M (1992) Chro nologie et paleoclimatologie de trois grottes des Carpates Orientales (Roumanie) d'apres les mammiferes. 1. Micromamiiliteres. Trav Jnst. Speol. "Emile Racovitztl", XXXI, pp. 95-104. SCHWARCZ, H P. (1989) Uranium series dating of Quaternary deposits. Quaternary International, 1, pp. 7-17. SHACKLETON, N., J. (1969) The last interglacial in the marine and terrestrial records. Philos. Trans. R Soc., Ser. B, 174, pp. 135-154. SHACKLETON, N., J & OPDYKE, N., D (1973) Oxygen isotope and palaeomagnetic stratigraphy of Equatorial Pacific Core V28-238: oxygen isotope temperature and ice volumes on a lOS year and 106 year scale. Quaternary Research, 3, pp. 39-55. SMART, P. L. & RICHARDS, D ., A. (1992) Age esti mates for the late Quaternary high sea-stands. Quaternwy Science Reviews, 11 (6), pp 687-696 21 VALEN, V., MANGERUD, J., LARSEN, E & HUFfHAMMER, A K. (1996) Sedimentology and stratigraphy in the cave Hamnsundhelleren, western Norway Journal of Quatemary Science, 11 (3), pp. 185-201. WINOGRAD, I., J., COPLEN, T B., LANDWEHR, J. M., RIGGS, A., C., LUDWIG, K.., R., SZABO, B., J., KOLESAR, P T. & REVEZ, K. M. (1992) Continuous 500,000-year climate record from vein calcite in Devils Hole, Nevada Sci ence,258,pp.225-260. WOHLFARTH, B., GAILLARD, M -J., HAEBERLI, W & KELTS, K. (1994) Environment and cli mate in southwestern Switzerland during the Last Tennination 15-10 ka BP. Quaternaty Sc i ence Reviews 13, pp 361-394.


and Applied K:arstology, pp. 2}-28 Radionuclides in a Cave Lake Sediment Core from 6hetarul de sub Zguri,ti [Romania] I f. Michel POURCHET1 Marie Antoinette MELIERES\ Emil SILVESTRU2 Geza RAJKA2 F. CANDAUDAP1 & Jean-Pierre ABSTRACT Some 250 m from the entrance of the cave Ghetarul de sub on the bottom of a lake exclusively charged per ascensum, at 12m depth, a 41 em core of fme sediment had been taken for radionuclide investigation. The following have been identified: 210pb-a natural radionuclide originating from atmospheric Radon decay, 134Cs and 131Cs-artificial radionuclides derived from the tests in the early sixties and the Chemobyl accident in the mid-eighties. Deposition fluxes of these elements at the water/sediment boundary have been found to be of the same order of magnitude as the ones at the surface, which shows a very effective transmission of pollution from the surface to remote cave environment. This is, to our knowledge, the first time such a sub stantial contamination has been measured. Coherent results were yielded by the profiles, especially in terms of sedimentation processes, allowing estimates of sedimentation rates, diffusion coefficients and the recent history of the sediment. The latter proved to be in perfect agreement with the climate recorded through the last decades. Key words: radionuclides, cave sediment, pollution, paleo-pollution, Romania. Radionuclides dans le sediment d'un lac souterrain de Ia grotte Ghetarul de sub Zgur8$ti (Roumanie) RESUME A quelques 250m de ['entree de La grotte GheJarul de sub Zgur(4ti, dufond (situe a une projondeur de 12m) d'un lac alimente exclusivemimt per ascensum, on a preleve une carotte de 41 em de sediment fm pour ['investigation de radionuclides. Les especes suivantes ont ete identijiees: le 210Pb-un radionuclide nature/for me par La decomposition du Radon de /'atmosphere, les JJ"Cs et 137Cs-des radionuclides artificiels provenus des testes nucleaires des annees '60 et de /'accident de Chernobyl. Les flux de deposition de ces elements a l'intelface eaulsediment ont le meme ordre de grandeur que ceux de La surface, ce qui demontre que La trans mission de La pollution a ete tres efficace. C'est, seton notre connaissance, La premierefois qu'on a mesure une contamination de cette envergure. Les profiles ont fourni des resultats coherents, specialement en ce qui con cerne le processus de sedimentation, en permettant des estimations sur les vitesses de sedimentation, les coeffi cients de d(ff'usion et l'histoire recente des sediments. La derniere s'est avere c:oncordante avec: le c:limat des dernieres decennies. Mots-c:les: radionuclides, sediment de grotte, pollution, paleo-pollution, Roumanie. 1 CNRS-Laboratoire de Glaciologie et Goophysique de l'Environnement. BP 96, 38402 St. Martin d'Heres, France. 2 Institutul de Speologie "Emil SecpaCluj, CP 58, str.Clinicilor 5, 3400 Cluj-Napoca, Romania. 3 Universite P. et Marie Curie, Laboratoire de Goologie 75252 Paris Cedex 05, France


24 SITE OF SAMPLING The cave Ghetarul de sub Zgud1ti is located in the centre of the westernmost segment of. the Roma nian Carpathians (known as AjJUseni MountainS), some 90 km west of the city of Cluj. It has rather large passages with a length of 5,210 m, extending over 74.8 (-44,8; +30) m elevation range There are four lakes with identical elevation of the water table (Fig. 1), all being supplied from below, via flooded conduits. The karst plateau located 250m above the cave (elevation 1,100 m a.s.l) displays no hydrographic network and therefore no sinking stream can account for the water stored in this karst aquifer. Areal infiltration is the only supply mechanism (at least in the present stage of the cave). The only sediment in the cave is the omni present clay which covers in some areas even the ceiling, indicating periodic flooding, due to an important raise of the water table (up to 8 m). The 41 em core was taken by diving to the bottom of Lake 1 (at 250 m from the entrance) which is actually the sloping tloor of a flooded passage (stalactites are present below water surface). The average sloping angle is 25' the coring being per formed in the of the floor (Fig. 2). GEOLOGICAL SETTING The previously mentioned karst plateau lies in the heart of an important geological unit of the Apuseni, called The Bihor Autochthonous, con sisting of a crystalline basement of Upper Protero zoic age, covered by a quasi-continuous stack of sedimentary formations ranging from the Lower Permian to Lower Cretaceous. The Permian layers are continental reddish, detritic sediments ('Verrucano' type) with local accumulations of M. Pourchet et al. uranium. The Lower Triassic is detritic as well, somewhat to the Werfen Formation. The remainder is mostly limestones (up to 1 ,300 m thick); There are no depo.sits .over!yijlg the limeexcept for a rather thiri, poor sou: .. Inside .the cave, no coarser sediments have beep identified so far, a featUre that strongly advocates foi: non-existent external input of sediments and consequently the muddy deposits niay be assumed to be autigenic This is valid for the present stage of the cave's evolution (an older cavity invaded by a niisirig water table) DESCRIPTION OF CORE There is little to be said about the macroscopic look of the core since it is a quite homogenous sample of fine grained sediment. However, there are three thin lllyers of ciark (almost black) deposit at 0, 4 and 28 em. (Fig. 2a). The section below the first such layer is visibly coarser, even if the larger than 60/l class does not exceed 30.2%. The coarser grains are mostly tiny angular limestone chips; no other petrographic species have been identified so far. The remainder of the core is highly homoge nous -light brown clay/silt mixture (according to the Wentworth scale) with ertatic green spots (decaying organic matter?) in the lower sectiori. Granulometric analysis has revealed two different populations with the coarser (above 60!l) fraction clearly marked at 5-6 em of depth (Fig. 3). THE RADIONUCLIDES The unsupported 2H'Pb (half-life 22.3 qa.ys-) is a natural radionuclide generated in the atmosphere by filiation of 222Ra (half-life 3.8 days) and L-----'-------'------'------'----=-==-=---'-s_a __ m_p_l_l._ :-=-s_i_t_e __,_ 5 O 700 600 500 400 Distance from entrance (m) I> Q) r-i ril Survey IJy A. S. 'Focu/ Viu': C.S. "Crista!': A.S. 11 Slinx'', C.S. '2"', C.S.


Radionuclides in a cave lake sediment core % m 0 5 10 12 35.00 30.00 25 00 20.00 15.00 10.00 5 .00 25 0 30 40cm 0 4 8 m ---------------------------1 .00 3 .00 5 .00 7 .00 9 .00 11 .00 1300 15.00 1700 1900 ongmates from steady atmospheric deposition at earth surface. 1 3 4Cs (half-life i.07 years) and 137Cs (half life 30.2 years) are both artificial isotopes with two different origins: the thennonuclear atmospheric tests extending from 1953 to 1975, displaying a p e ak around 1963; the Chernobyl accident in 1986. The entire core has been investigated for the above mentioned radionuclid e s, both in a and y spec trom e try. Only the upper section (roughly 20 em) yield e d measurable activities which indicat e that the 100 year limit (of the 21'1>b dating method) lies in this section. The profile (Fig. 4) marks a clear event at 6 em of depth-most probably a slumping that had brought two different sequences in direct con tact. The regular decrease of unsupported 2J('Pb in th ese two sequences yielded a sedimentation rate of 0.5 mm/y for the upper sequ e nce and 2 rnm/y for Depth ( em) the lower one. The slumping zone at 6 em is con finned by the lower content in 40K (Fig. 5) and a larger (Fig. 3). Given the short half life of 134Cs, the one identified in the core can only originate from ChernobyL I n the lower sequence, the 137Cs profile (Fig. 6) is interpreted in tertns of slow diffusion by applying the SDM (Simplified Diffusion Mixing) method (MELIERES eta/., 1988, POURCHET et a/.,1989, MELIERES et a/.,1991) This method gives the reconstruction 137Cs profiles form e d as a cons e qu encte of thermonucl ear tests, for di fferent sce narios of diffusion and sedimentation processes. The reconstruction of the observed signal indicates a slow diffusion process characterised by the di mensionless number G (G = DILS, where Dis the diffusion coefficient L the mixing d e pth S th e sedimentation rat e ) sma ller than 0 03. This typ e of diffu s ion implies that the po s ition o f the ma x imum


26 .M. Pourcl,l,et {!t a/. 2500 0 2 00 4 .00 6 .00 8 .00 10.00 1200 14.00 16.00 18 .00 20.00 1 00 ---------------------o go I 0 8 0 0 7 0 010 Depth (em) 0 00 ==;::=;:::.:::;::.:::;::::::;::::.==;==;::=;:::.:::;:==;==;==;;:=;:---=:;:::...r-100 3 .00 500 700 9 .00 11.00 13.00 1 5 .00 17.0019.0 0 D e pth (em) 1 .00 300 5 0 0 7 .00 9 .00 11 .00 13 .00 15 .00 17.00 1900 Depth (em) of the s i g nal ha s not been displaced and corre s pond s to 1963. Th e SDM also allows t o es timate a se dime ntation rate close to 2 mm/y, confirming the 21''Ph es timates, and a mixing depth of 6 8 em. Consequently, the sedimentation rate and the po si tion of th e 1963 peak (Fig 6) allow an es timatio n of the date of s lumping--close to 1983. Thi s s lumpin g dat e coincides with an important local cli m a t e sh irt in Tra n s ylv an i a (CARBONNEL e t al., 199 4): beginning with 1983 a severe drou g ht affected a territo ry ext e ndin g from Transylvania to the Bulgarian Balkans; rainfall d ecreased with 2 0 to 25% (compared to preceding decades) and ac cordinglythe water table dropped significantly. If the datil}g o f the slumping i s correct, n o C hernobyl contamination s hould b e recorded in th e seq uen ce bel o w it and thi s is exac tly th e c a se, sin c e no 134Cs s i gna l has be e n identified. In th e u pp e r sequenc e


Radionuclides ;,. a cave lake sqdimetrt core 1r1 160 00 I 14000 1"\ -4.00 . 3.50 : I \ \ 1' --.......... ,...--, ,.. 120.00 ,.-I "-, '--\. ,---;::-3.00 \, \ 'I ,...., I ',. --.......__, '-\\ I I ---. \ /'1 \, ,--.... E 2.50 100.00 \\ IJ \ 1 \ I ,,., '-"' \\ ,y i !,! I ,_ 80.00. 60.00 40.00 20.00 + 2.00 .... ( i c Tl 1 50 1.1,.. 1 .., I 1.oo 8 t 0 50 I 0.00 .l---+----+--1---0.00 1 3 5 7 9 11 13 15 17 depth (m) ---Pb ------Cr --zn ---cd ---Fe 134Cs is detected only in the upper 2 em, confirm ing the Chernobyl origin of the isotope This profile is coherent with the slow mixing (here "mixing" meaning diffusional) processes and will be devel oped in a further article. The sedimentation rate estimated froin the 2 10Pb profile indiCates that the upper depositional sequence extends to the begin ning of the century at 6 em of depth The coher ence between the two profiles (and sequences) advocates for a minor disturbance by the slumping and allows estimates of deposition fluxes. The integrated unsupported 2wPb is indicative of the averaged annual deposition nux at the wa ter/sediment interface (POURCHET et a! ., 1994), with the assumption of a constant deposition flux, as usually accepted. In the upper sequence the mean annual flux is 150 Bq/m2 and in the lower sequence it has been estimated to 240 Bq/m2 in cluding the correction of the missing part of the sediment since 1983. The two fluxes are of the same order of magnitude and, since they do not represent similar depositional mechanisms (except for the last 12 years they shared different deposi tional env ironments), th e y are not expected to be identical. In any case, these values arc only in dicative of the order of magnitude. Even if the precision of the measurements is 5% (lcr) the slumping and the coring its elf preclude precise interpretations of these values. The Chemobyl mcs total deposition flux at the water/s e diment interface can be es tim ated from the upper sequence of the core where it has two origins: Chemobyl and the thermonuclear tests. The accepted ratio of 2 mcs/134Cs in <::hemobyl emission has been used to separately estimate Lhe 137Cs f1ux relative to Chemobyl (5.3 kBq/m2 cor rected at deposition time), from the part relative to the thermonuclear tests (1 kBq/m2 at deposition time). Thelower sequence gives a 137Cs flux from the thermonuclear tests of 1.1 kBq/m2 These fluxes can be compared with the closest measured deposition fluxes at earth surface like the ones measuredin soil in Bulgaria (110 Bq/m2y, unpublished results ) which at their turn agree with other values measured in eastern Europe (PREISS eta!. 1996). The 137Cs inventory relative to ther monuclear tests (estimated from the deposition of 90Sr) has been measured at Vienna yielding a value of 2.95 kBq/m2y (HASL, 1977). The Chemobyl 1.17Cs deposition in Romania has been estimate d at Cluj to be 4 kBq/m2 but shows a wide range of values throughout the country, from 0.2 to 8.5 kBq/m2 (EUR 12800, 1991). The striking conclusion which emerges when separately comparing each deposition flux at the earth surface with its correspondent at th e wa ter/ se diment int erfa ce, is that th e y arc of the s ame order of magnitude This indicates an cffccti v e transfer of these radionuclides from the surface to th e cave environment.


28 KARSTOLOGICAL The first spectacular feature that emerges at first sight is the presence of stalactites some 3 m below water table. This is the first time in Romania that a flooded cave at such an extent, is explored. In the mean time, flooding of caves in perfectly continen tal conditions (at more than 600 km from the near est sea) is by itself a rather peculiar feature, re questing a special hydrogeological setting (neo tectonics included). Yet, since the subject of this paper is radionuclides, we shall skip further analysis at this point (which calls for a separate paper). The deposition rates yielded by radionuclide pro files are the first to be properly estimated in the saturated zone of karst in Romania. Their order of magnitude are a strong argument for an autigenic source (most probably the weathering of the walls) given the predominance of silt and clay and espe cially the lack of other than limestone coarser grains. It is important to take into account that the sedimentation rate of the two sequences differs by almost one order of magnitude (0.5 versus 2 mrnly). The most probable explanation is that the upper sequence was deposited at the foot of the wall where the suspended load might have been less important (Fig. 2). This is consistent with its coarser character which according to FORD & WILLIAMS (1992), is a typical feature for silts and clays (which "grade laterally from centres of passages"). The slumping marks a rather dramatic episode which we presume to be an important drop of the water table (consecutive to significantly dryer conditions at the surface, as revealed by precipita tion and river discharge graphs). It is possible that for a short time, the area of deposition of the upper sequence was close to the water table (even above REFERENCES CARBONNEL, J.P et al. (1994) Etude statistique de quelques series pluviometriques et bulgares Incidences sur l' evoluuon climauque recente. Roman ian J. of Hydrogeol. & Water Res., 1. FORD, D., C. & WILLIAMS, P., W (1992) Karst Geomorphology and Hydrology Chapmann & Hall, London, New York, Tokyo, 601 p. MELIERES, M., A. et al. (1988) Chernobyl134Cs, 137Cs and 210pb in high mountain lake sediments: measurement and modelling of mixing process. J ofGeoph. Res., 93, pp. 7055-7061. MELIERES, M., A. et al. (1991) Comments on decon volution profiles in sediment cores. Water and Soil Pollution, 60, pp. 35-42. M. Poutchet et aL it), its return to initial conditions providing the conditions for triggering the slumping. Finally, all data yielded by the core suggest a recent change in this cave system's condition, mainly its flooding. If the deposition rate of the lower se quence is extrapolated for the rest of the core, the age at its bottom is 188 years. If, on the other hand, the deposit on the floor of the flooded passage is not thicker than 50-60 em (as suggested by the coring), one may fairly assume that the deposition began some 250 years ago. Accordingly, the flooding could not be much older. No human-induced change recorded so far can account for such a change within the aquifer, so the only mechanism we can imagine at the time being, is neotectonics-mainly due to the continuous lifting of the entire block in the centre of the Apuseni. ENVIRONMENTAL SIGNIFICANCE To the best of our knowledge, this is the first time paleo-pollution could be measured in a karst aqui. . th t 210nb 134c d 137cs fer. It 1s qUite surpnsmg a r s an are so effectively transferred to a karst aquifer ex clusively by seepage and through a soil and some 250 m thick limestone cover. Under such circum stances, one must assume that fme sediments in the saturated zone of karst terrains can be excellent pollution recorders, especially when they are authi genic and there is no stream input into Seepage water carries the abovementioned 1ons which are subsequently fixed by the fine residual, poorly layered clays at the bottom of flooded cham bers, resulting in excellent pollution files One may also assume that thinner the soil covering the limestones, the more polluted are deep aquifers within those limestones. The further consequences are already commonplace ... POURCHET, M. et al. (1989) 137Cs and 210pb in Alpine lake sediment: measurement and modelling of mixing process J. of Geoph. Res., 94, pp 12761-12770. POURCHET, M. et al. (1994) Sedimentation recente dans le lac Titicaca (Bolivie). C.R.A.S., 319, pp. 535-541. PREISS, N. et al. (1996) A compilation of data on l ead 210 concentration in surface air and fluxes at the air-surface and water-sediment interfaces. J. of Geoph. Res., accepted. EUR 12800 EN Rep', 1991 HASL 329 Rep', 1977.


Speleothems Dating Using the_ Thermoluminescence Method V as ile LABAU, Emlllan G ASP A R & T atl ana PAUNICA 1 A B STRACT The age of a stalagmite fragment collected from Petera Vantului (Wind Cave), Romania, was estimated using the thermoluminescence method. The natural dose accumulated by the calcite during the geological age was es tablished Using the relationship between the calcite thermoluminescence sensitivity and the applied pre dose an age of 59,052 years has been established. Key words: speleothems dating, thermoluminescence sensitivity, naturally accumulated dose, pre-dose geo logical age, thermoluminescence method. La datation des speleothemes par /'utilisation de Ia methode de Ia thermoluminescence RES U ME En utilisant La methode de La thermoluminescence on a determine l'tige d'unfragment de stalagmite preleve de La Pejtera Vantului (Grotte du Vent), Roumanie La dose naturelle accumulfe dans La calcite durant le temps geologique a ete etablie. En utilisant La relation entre La sensibilite La thermoluminescence de La calcite et La pre-dose appliquee il a resulte un tige de 59.052 ans. Mots-cles: datation des speleothemes, sensibilite de La thermoluminescence, dose naturelle accumulee, predose, age geologique, methode de La thermoluminescence. 1. IN T RODUCTION Speleothems (stalagmites, flowstone, etc ) are regular calcite depositions formed by chemical precipitation from vadose seepage waters that penetrated through the roofs or the walls of lime stone caves. As these deposits grow, small amounts of seepage water are trapped within. Speleothems consist of two components : authi genic calcite and detritus particles which w e re deposited on the surface of speleothems as the latter grew. The detritus normally consists of car bonate or silicate rock fragments, clay or organic matter Speleothems may be dated through various means, such as the 14C method or uranium series disequilibrium methods, analyzing both th e stabl e and radioactive isotopes in calcite and in the fluid inclusions (GASCOYNE & SCHW ARCZ, 1982 ; ATKINSON et al 1984; GEWELT, 1984; GEYH et al., 1985; GEWELT, 1985). The application of uranium series methods to spe leothems dating can provide chronological infor mation on landscape development and p a leo environments. Here is the principle underlying this method: seepage waters contain a small amount of dissolv b d uranitJm in the form o f complex salts When, for instance, a stalagmite is formed, ura nium precipitates together with CaC03 On the a dau g hter of234U, i s insoluble and therefore the calcite in speleothems did not include it at the moment of deposition (GEWELT, 1985) 1 Institutul de Fizicli i'Ingbterie Nuclearl. Romania.


30 Two equilibrations intervene, which may be used as geochronometers. On the one hang, the activity ratio of the two uranium isotopes 234 and 238 which was different from 1 at the initial moment, fp,ward unit. Therefore; : an assessmeQt: of toe ' age of speleothems a knowledge of the initial isotopic ratio. However, a iiumber cif difficulties are encountered in this case. The second method is based 'on the as tion, namely that (authigenic) calcite did not con tain thorium, but only uranium, at the moment of deposition. In the chain of the 238U and 234U iso topes, the next long-life element is 23'111. There fore, the increase of the 23"rh concentration in authigenic calcite provides a measure of its age. Inter alia, suchlike datings are used for the study of the kinetics of speleothem growth in caves. Thus, Gewelt finds a variation in the velocity of stalagmite development which ranges from 0.22 to 9 29 cm/100 years, in alternation with hiatus peri ods. Special pieces of information may be supplied by an analysis of the isotopic content of the water and gas contained in various concretions and speleo thems, and more particularly, by tritium analysis. However, the available amounts of water in spe leothems, as well as their poor tritium content preclude the application of the current measuring techniques. By using of a particle accelerator and application of mass spectrometry current measur ing limitations will be reduced. In Romania, a stalagmite fragment collected from the submerged Taaul Cave was dated by using the Uranium-series method. The stalagmite grew dur ing the period from 130 ka to 70 ka b.p. (LASCU & LAURITZEN, 1996). In this paper, we tried to determine the age of a stalagmite fragment from the Wind Cave, Roma nia, by using the thermoluminescence method. 2. THE THERMOLUMINESCENCE METHOD 2.1. THE PHYSICAL PHENOMENON The thermoluminescence method (TL) is based on the thermoluminescence phenomenon. This phe nomenon consists in the fact that certain crystal lized anorganic substances, when exposed to ion izing radiations and then heated up to a certain temperature, characteristic of each material, are emitting a visible radiation V. lAbau et al. The intensity of the emitted light is proportional to the radiation dose to which the substance has been exposed. The most important appUcations of the thermoluphenomenoq occur in the radiations the age in geology by the TL method is based on the fact that the rocks contain certain minerals which have thermolumines ceiit properties and may be used as integrator do simeter. Calcite, which is the main constituent of the stalactites and stalagmites, is one such mineral. The event which is dated by the TL method is the moment of crystallization of the deposited calcium carbonate. From that moment the material is ex posed to the radiations emitted by the natural ra dioactive elements (U, Th, K) which exist in the matrix. During one year, in the material is accu mulated the dose (d ) named the annual dose. During the geological time, T, the dose Dn =.d. x T is accumulated, which represents the accumuiated natural dose. J, From this relation of accumulation of the natural dose, one can determine the geological age (T) as the ratio: T= Dn d (1) 2.2. THE CALCITE THERMOLUMINESCENCE . In order to establish the geological age, it is neces sary to measure the natural thermolumine,scence (NTL). For this determination, it has been used the thermoluminescence (TL) curve. The TL represents the variations of the flux of visible radiation emitted by the matrix that is pro gressively heated from the environmental tem perature up to a maximum tem perature (-450 C) In Fig. 1 are presented the natural thermolumines cence (NTL) and the artificial thermolumines cence (ATL) curves obtained for the studied sta lagmite. The NTL curve is due to the calci te irra diation by the natural elements (U, Th, K) con tained in very small amounts within the stalagmite. The NTL curve displays a single peak occurring at a temperature o f about 320 C. The artificial thermoluminescence curve (ATL) may be obtained by artificial irradiation using a beta or gamma nidiatiori source The A TL curve displays distin c t pea!cs at the temperature of


Spelepthems dating using the iiie(hod. 31 ,; Fig; The artmclal. ; curves fofthe. Stildlect:;.':_ stalagmite .. ,,:. ; ...... .. ;. . , ... thermoluminesce11ce .. .. 11aturelle et ar#ficie/le pour l'echantiilon de .. stalagmrtll etudie. . E -ao >-1-Vi 70 z w tjso z w w z --30 ::::) <520 ffi10 I 1-D = 720Gy 0 120 180 230 320 120 C, 180 C and 320 C It also displays a less distinct peak at the temperature of 230 The peak at the temperature of 120 "C :is unstable and disappears comparati vdy soon. when the irradi ated sample is stored a long time at the room temperature. In geologiCal tirrie, because of the thermal agitation, the peaks in the median temperature range ( 180 C and 230 C) display a declining (fading) trend which may eveiih:ially result in their disappearance For these reasons, the low (120 C) and middle ( 180 and 230 C) temperature peaks do not appear in the natural thermolumines cence curve (NTL). Through irradiation with ionizing radiations, the therrrioluminescence sensitivity of the peaks ftom the low and middle temperatures increases. This is due to the supplementiuy axes caused by the irra diation of the calcite (ZELLER, 1968) 2.3. THE DETERMINATION OF THE NATURALLY ACCUMULATED DOSE In order to estimate the geological age it is neces sary to e valuate the natural dose (Dn), accumu lat ed during the geoiogical time, T. The Dn dose is evaluated by measuring th e natural th ermo lumi nescence and converting it into radiation dose values For this purpose it would be normal to use the peak at high temperature (320 C). But this maximun;1, although stable; cannot be used foi: the measurem e nt of the th errno luminescence in do sime tric purposes, because its inten si ty depends TEMPERATURE ( c] not only on the absorbed radiation dose but also on other uncontrollable factors . In order to estali>iish the Dn dose we have used a method based on the combined action of pre irradiation (pre-dose) and heating (at 450 .0C) on the thermoluminescence of the cillcite at low and 01edium temperatures. For the stalagmite we have studied, the effect con sisted of an increasing of the thermoluminescence sensitivity at low and middie temperatures. proportionaltothe dose (pre dose). For the practical application of this method we have proceeded in the following way: 1. a calcite sample has been heated up to 450 oc and, after the cooling, the thermoluminescence sensitivity (SN) has been evaluated by using a test dose (D, = lGy); 2. for another sample the pre dose (D) has b ee n applied, theh; after heating up to 450C and subsequent cooling, it has been tested if the sensitivity had the value (SN +D). In all cases the measurement of the thermolumines cence sensitivity was made for the temperature peak at 120C, which was pi:oven to be reproducible. I Based on theoretical reasons, corroborated by the analysis of the experimental data, it results that th e relation S = f(D) between the thermoluminesc e nce sensitivity {S) of the calcite and t!1e dose (pre dose) D, has, as a first order approximation, the form:


32 S=a+bD By extending the resulting straight line, toward the negative side of the abscissa, the accumulated natural dose (Dn) is obtained. 2.4. THE EVALUATION OF THE ANNUAL DOSE As it results from relation (1) in order to establish the age it is also necessary to estimate the yearly dose (d) due to the radiations emitted by the natu ral radioactive elements (U, Th, K) very small quantities of which exist in the speleothems. The annual dose of the stalagmite was indirectly evaluated from the natural radioactive elements content, by using a concentration-dose conversion table (AITKEN et al., 1968). The natural radioactive elements content was evaluated by using the neutron activation analysis method. The following radioactive elements con tent was found: U = 0.3 ppm; Th = 0.0 ppm and K = 0.01 %. From the conversion of this radioactive content, the yearly dose resulted to be: d = 94.72 X 10"5 Gy/y 2.5. THE EXPERIMENT For the dating experiments we have used a stalag mite sample. For the thermoluminescence meas urement the fragment was processed by crushing, grinding and sieving, a microcrystalline powder being obtained as a result. The diameter of the microcrystals used in this experiment varied between 80 and 160 J..Ull. For natural (NTL) and artificial (A TL) thermolu minescence measurements, the same quantity of microcrystalline powder was used. The dosimetric information was derived during the thermolumi nescence measurement by means of a conventional device for detector thermoluminescence measure ments, type Mark IV TLD Reader. The thermoluminescence curve was established by using an X-Y recorder. As a measure of the ther moluminescence intensity one can take the l20C temperature peak height. For the artificial V. Labau et aL irradiation of the calcite samples, a 90Sr -90y beta radiation source was 'used. The thermolumines cence curve was recorded after calcite samples irradiation, directly. For the evaluation of the accumulated natural dose (Dn) we have used the 'predose' described above. As a measure of tlle thermoluminescence sensitivity the of the 120 oc peak after calcite irradiation at a dose test of 1 Gy was used. The heating of the samples up to 450 oc was made with the same Mark IV device. The heating of the samples was repeated four times, in order to eliminate the residual thermoluminescence influ ence on calcite sensitivity. 2.6. EXPERIMENTAL RESULTS In order to establish the enhancement of the calcite thermoluminescence sensitivity to the dose (pre dose), each calcite sample was at in creasing doses. The used irradiation doses (D) and the resultant thermoluminescence sensitivity are indicated in Table 1. The irradiation dose (D) is expressed in Gy and the thermoluminescence sensitivity (S) in mV. Table 1 D(Gy) S(nA') 0 45 105 90 270 210 570 380 720 500 The thermoluminescence sensitivity versus the dose (pre-dose) D is presented in Fig. 2. From these data one can see that t)le relation S = f(D) between the calcite sensitivity and the dose (pre-dose) is linear. The equation of this linear correlation is given by the relation: S = 35.233 + 0.6299 D (3) The accumulated natural dose is obtained extend ing the straight line to the X axis, where S = 0: Dn= 55 934 Gy (4)


Speleothems dati11g usi11g the thermoluminescence methtJd '. 33 > r---------------------------->-I-> i= Vi z w (/) x EXPERIMENTAL VALUES o COMPUTED VALUES The Iumbtescence sensitivity' versus the dose La sensibilite de Ia. therlnoluminescence en fMctUm de Ia dose D; z w u Vl w z :::l ...J 200 a: w :r: I-----+--------<1--------+---Dn 0 200 400 The geol()gical age of the studied speleothem was evaluated using the naturally accumulated dose established as above and the yearly dose. From these data the age of the stalagmite results: 59,052 years. REFERENCES AITKEN, M., J., ZIMMERMAN, D., W. & FLEMING, S., J. (1968) Therffio1uminescent dating of An cient Pottery. Nature, 219, pp.442-445. ATKINSON, T., C . SMART, P., L. & ANDREWS, J., N. (1984) Uranium series dating of speleothems from Mendip Caves. 1 :Rhino Rift, Charter house -on Mendip, In Proc. Univ. Bristol Speleol. Soc ., 17, 1 p. 55 CHERDYNSEV, V V ., KAZACHEVSKIY, 1., V. & KUZMINA, Y . A. (1965) Dating of Pleistocene carbonate formations by the Thorium and Ura nium isotopes Geochemist1y International 2, pp. 794-801. GASCOYNE, M. & SCHW ARCZ, H P ( 1982) Ura nium series dating of speleothems. In Uranium series disequilibrium. Applications to environ mental problems. (M. IV ANOVICH & R. S. HARMON Eds), Clarendon Press, Oxford, pp. 270-287. 600 BOO EXPOSURE DOSE [Gy) ACKNOWLEDGMENTS The authors thank Mrs. M. SlUligean and Mrs A. Pantelicli for the excellent neutron activation analysis of the natural radioactive elements con tent of the studied stalagmite. GEWELT, M., GASCOYNE, M QUINIF, Y. & CATILLANA, R. (1984) Les datations radiome triques. Kiilner Geogr Arb., 45, pp. 95-104. GEWELT, M. (1983) Datations 14C des concretions de grottes belges: vitesses de croissance durant !'holocene et implications paleoclimat iques. In New Directions in Karst. Proc. 2nd Angl o French Karst Symposium, Oxford GEWELT, M. (1984) Les datations radiometriques 14C et 2J(7hf-34U In Livret-gu ide des exc ursi ons. Colloque International de Karstologie appliquee. Liege pp. 12-16 and 24-25. GEWELT, M. (1985) Cinetique du concretionnemenl dans quelques grottes belges : Apport d e s Data lion 14C et 230Jb I 234U. Annates de La Societ e deologique de Belgique. 108, pp. 267-273. GEYH. M. A., HENNIG, G . J & GRUN. R. (1985) Multiple dating of a long flowstone profile In The 12'h Int Radi o carb on Confer e nce, Tmnd heim,Nmway, p. 15.


34 KOMURA, K. & SAKANONE, M. (1967) Studies in the dating methods for Quaternary samples by natural alpha-radioactive nuclides Sei. Rep Ka nazawa Univ. 12 (1) pp. 21-26. LASCU, C. & LAURITZEN, S., E. (1996) Speleothem dating in the Tasaul Cave, evidence of low seas tands in the Black Sea (Abstract) In XN Theo retical and Applied Karstology Symposium, Baile Herculane, Romania, 1996 Abstracts Vol ume, pp. 34-35. LIRITZIS, Y. & GALLOWAY, R., B. (1982) The 23<7ht-34U disequilibrium dating of cave traver tines. Nuclear Instruments and Methods, 201, pp. 507-510. V. lAbau et al. ROSHOLT, J., N. & ANTAL, P ., S (1962) Evaluation of the 231Pa/U -2rh!U method for dating Pleis tocene carbonate rocks US Geol .Survey, Prof Paper 450-E, pp. E108-Elll. TENU, A. & DA VIDESCU, F., D. (1991) La datation par radiocarbone dans Ia recherche sl&>logi que. Theor Appl. Karstology, 4, pp. ZELLER, E., J. (1968) Geologic Age Determination by thermoluminescence In: Thermoluminescence of Geological Materials (D. J. McDougall Ed.), Academic Press, London & New York, p. 311.


Theoreticaland Applied Karsto!Ogy, voL 9/1996, pp. 35-44 Effet d'echelle de Ia dispersion dans un conduit karstique Hypothese lineaire et hypothese fractale Erik CARLIER1 RESUME Les variations du coefficient de dispersion sont etudiees dans un conduit rectiligne et dans un conduit presentant une tortuosite a geometrie fractale. Dans le cas d'un conduit rectiligne, nous montrons que le coefficient de dispersion est sujet a un effet d'echelle, jusqu'a une certaine distance au-dela de laquelle on retrouve le processus classique fickien de dispersion, caracterise par un coefficient constant. En revanche, dans un conduit a tortuosite fractale, nous montrons qu'il n'existe pas de valeur asymptotique du coefficient de dispersion apparent. La determination du coefficient de dispersion par Ia variance est discutee. Mots cles: milieu fissure, karst, fractal, coefficient de dispersion. Scale effect of the dispersion in a karst pipe Linear and fractal cases ABSTRACT The variability of the coefficient of dispersion is investigated in a linear and in a fractal pipe. For the linear case, it is shown that the coefficient of dispersion exhibits a scale effect over a certain distance, beyond which the process becomes fickan, with a constant coefficient of dispersion. In the fractal case, it is shown that there is no asymptotic value of the coefficient of dispersion. The evaluation of the coefficient of dispersion by means of the variance is discussed. Key words: fractured medium, karst, fractal, coefficient of dispersion. INTRODUCTION De nombreuses experimentations de terrain ont montre que le coefficient de dispersion obtenu etait beaucoup plus grand que celui determine sur colonne de laboratoire pour un meme materiau (ANDERSON, 1979; 1984). Des mesures de dis persivite pour une meme fonction geologique a des distances differentes ont montre que ce parametre etait fonction de l'echelle d'observation (PICKENS & GRISAK, 1981; SUDICKY & CHERRY, 1979). En utilisant !'analyse spectrale, GELHAR et al. (1979) montrent theoriquement que, s'il y a des echanges entre les differents stra tes geologiques, la dispersivite peut atteindre une valeur constante pour une grande distance de par cours. MATHERON & MARSILY (1980) con eluent que l' equation classique de convection dispersion n' est generalement pas applicable, meme pour une grande distance, dans un milieu stratifie. PEAUDECERF & SAUTY (1978) trou vent experimentalement, sur des sables alluvionnai res, un coefficient de dispersion qui double quand la distance augmente d'un facteur de 2 a 2,5. 1 Laboratoire de Geosciences environmentales, Faculte de Sciences Jean Perrin, Universite d' Artois, rue Jean Souvraz, SP 18, F-62307 Lens Cedex, France.


36 Q t : temps tc :temps t h :.temps -hann<)rtitiJJe u : vitesse (L't'1 ) u : vitess6 . .-'; ; u' v x : distan(i.{Lf: ;;-r (Lf_' z : rayoll cle I a a : {L) a1 : ' J.ln : ::: : .eri ,. a;: variance entemps J . -. E. Cartier ARYA et al. (1985), en utilisant Ia theorie du mouvement brownien .fractal de MANDELBROT & VAN NESS (1968), demontrent que la dispersi vite est proportionnelle au deplacement eleve a une puissance comprise entre zero et }'unite. ROSS (1986) indique que l'effet d'echelle de Ia dispersivite pourrait etre de nature frac:tale; WHEATCRAFf & TYLER ( 1988) donnent une explication de cet effet d'echelle en utilisant Ia geometric fractale. 1. DISPERSION DANS UN CONDUIT CYLINDRIQUE RECTILIGNE TAYLOR (1953) a etudie la dispersion d'un solute dans un tube eta montre qu'au-deHl d'une certaine distance L de le processus de dispersion etait fickien. Si: 2 2 L> ua 3 82 D (1) ou u" est Ia vitesse moyenne (L11), ale rayon du tube (L), D le coefficient de diffusion moleculaire (L211), alors que l'equation classique de diffusion est applicable: ou C (x, r, t) repn!sente la concentration (ML -3). En posant z =ria et x1 = x-u"t, et en considerant que le temps d'homogeneisation des concentra tions radiales par diffusion est faible devant le temps necessaire a la convection pour induire des differences notables de concentrations radiales, TAYLOR (1953) montre que les quantites ac;axl et ac fax. sont cgales ct obtient !'equa tion de dispersion longitudinale suivante: i D _au L-48D (3) (4)


Effet d'echelle de Ia dispersion dans un conduit karstique D,_ est le coefficient de dispersion longitudinale (e11 ) qui est constant pour une vitesse moyenne u constante. Pour L 2u* a2/(3,82D), TAYLOR (1953) a etabli des expressions analytiques de 2 a c'(x,t)=-2 Jc(x,t,r)dr a o (5) TOMPSON & GRAY (1986a; b) ont propose un modele de dispersion de second ordre qui s' ecrit, pour un tube en absence de diffusion moleculaire: frepresente le flux dispersif (ML-7"1 ) C=C' +C';C'' =0 u = u + u'; u'' = 0 f =(u'C')' (6) (7) La demonstration de !'equation (6) est donnee en annexe. KLOTZ (1989) a etabli des solutions analytiques de !'equation (6) et en a deduit des expressions du coefficient de dispersion D (x, t): D(x,t)= -oC ox (8) II conclut en precisant que cette quantite manifeste un effet d'echelle. Considerons le cas d'une injection de traceur de masse M pendant un intervalle de temps !lt sur une partie centree de Ia section d'un tube. L' ecoulement est suppose laminaire et station naire. Le rayon de la surface d'injection est r(L ), celui du tube est a (L ). Pour un fluide incompres sible, de densite constante, le profil des vitesses est parabolique: (9) oil zest Ia valeur d'un rayon comptee a partir du centre du tube. La vitesse maximale est u(O) = 2u 37 A !lt, toute Ia masse M est injectee et occupe le volume suivant: TJ J J21tzdzdx 9 y avec: e = 2u -r2 1 a2 )ll 1 13 = 2u' Ill y=O (, i/2 TJ = a x I 2u !l t} (10) (11) Ia concentration a I' interieur du nuage est done : C0=MIV (12) Selon les hypotMses faites precedemment et en absence de diffusion moieculaire, le nuage va se deformer tout en gardant un volume constant. Pour le couple de variables (x, t), il est possible de determiner les expressions analytiques de la concentration moyenne en residant c; du flux moyen total F et du coefficient de dispersion longitudinale D1 (x, t). Les notions de concentration en residant et de con centration en flux ont ete introduites par KREFf & ZUBER (1978). j c;(x,t)= Co2 J21tZdz 1ta ; c j F' (x,t)= -T J21t u(z)iz 1ta ; ( ) F'-u'c; D1 x,t = -oCR OX (13) (14) (15)


38 Les homes d'integration i et j indiquent les limites radiales du nuage et scront explicitees ulierieure ment. Le tlux moyen F. est cgal a: F (x,t )= u c; (x,t) (16) c;.. f o > representc la concentration rnoyenne en flux; dans un contexte de dispersion cinernatiquc, il n'y a pas egalite entre c;( et c; .. i::::: 0 (19a) 11 'f2 t) (19b) -xj(2ut)] (i9c) (19d) )] (l9e) D1 (x,t)= u" x t )} (19f) 2u(t -M)< x < -r2 / a2 ) i = 0 (21a) J=r (2lb) c;((x,t )=Co r2 I a'}. (2lc) F(x,t)= uco( 2r2la2 -r4la 4 ) (21d) c;..(x,t)= co( 2r'J.Ia2 r4la 4 ) {2le) oo (2lf) 2u (t-M)Q-r2/a2)) {29c) (29d) (29e)


Effet d'echelle de Ia dispersion dans u11 co1uluit karstique D,,(x,t)= ru))} 2u '2 )t2 I a2 X!,-r 2 I a 2) (29t) Les quantites I9f, 2lf, 23f, 27f, 29f indiquent clai rcment que le coefficient de dispersion longitudi nale manifeste un effet d'echelle. Ces resultats sont analogues a ceux obtenus par KLOTZ (1989). Les domaines de definition des forrnules precedentes sont indiques sur les Figures 1 et 2. de pollution e concentration Co Fig. 2. 2: -Le parabolo'ide terminala dep&sse Ia base du lnlthil (solutions 19, '17, 29). Case 2; The !eft paraboloid overtaken the ba$e of the right one (solutions i9). :. . .. Nous illustrons Jcs equations precedentes par une simulat ion realisee en considerant un rapport entre surface polluee et surface totale de 0,25, une vi lesse moyenne de I rnls, un temps d'injcction de 120 s. les concentrations sont calculees en fonction du temps el pour des distances de 1 00 m e t I 0000 m. ( Fig 3 et 4). 2. DISCUSSION Tant que le s vites ses des particules du nuage se ront correlees par Ia loi parabolique de distribution (9) l e coefficient d e dispersion restera un e fane lion spatio-temporelle. II est bien evident que 39 !'utilisation des formules classiques utilisant Ia variance pour determiner le coefficient de dispersion longitudinale sont fondamentalement inappli cables: (30) a; represente Ia variance en distance. L'utilisation de ce type de formule n'est compatible qu'avec C/CO 0 5 ...--------------.., 0 4 0 3 0.2 0,1 a CRICO CF/CO 0 0 -+------tlt-----r------..---+--1 0 100 temps (s} 200 Fig. x = too m. Conctnfriduin l'entls time .ot i./10 .. 0,05.----------------, C/CO 0 ,04 0.03 0,02 0 01 a CRICO CF/CO o .oo 4000 5000 6000 7000 Fig. 4. de Ia concentration ii X = loobo m. ConcentiY,dioia versus time at ioooo m. . . . unc distribution gaussienne des concentrations et un mode en residant d'injection et de mesure de Ia concentration. Neanmoins, il sera dans certains cas possible de relier le coeffi ci e nt d e dispe rsion el Ia variance Par exemple, si l'on considere Ia condi tion (18) avec r = a e t D.t oo el en utilis a nt I a mcthode1classique d e calcu l des moments : x,. lln = J x"Cu(x.t)dt (31) 0 avec x"' = 2ut.


40 La variance en distance est definie par: Comme C11 = C0(1-xlxm), on obtient: (33) remarquons que la quantite cr; / t2 est constante. L'expression du coefficient de dispersion en fonc tion de la variance est: D1 = 2,12 x{l-0,235x/cr x) (34) Si la longueur pour laquelle les vitesses des parti cules du nuage ne sont plus correlees est atteinte, 1' evolution de la concentration en residant est donnee par !'equation (3), avec un coefficient de dispersion longitudinale constant. Dans le cas d'une injection instantanee en mode residant et d'une mesure de la concentration dans le meme mode, la solution est (TAYLOR, 1953): C11(x t)= M (35) 2 I )112 4D t 1ta \41tDLt / Le coefficient D1 peut etre determine par la rela tion (30). En regie generale, lorsqu'il s'agit d'experimentations sur le terrain, i1 est tres diffi cile de determiner la variance en distance; on dis pose d'un ou de plusieurs points d'observation au niveau desquels on mesure la concentration en fonction de temps; de plus, le mode en residant est, dans la plupart des cas, remplace par un mode d'injection et de prelevement en flux pour lequel la reponse a une injection instantanee est donnee par la formule (ZUBER, 1974; SAUTY, 1977): (36) Q represente le debit (L311 ) et est egal a la quan tite na2u. Il convient done d'etudier les relations entre le coefficient de dispersion DL et la variance en temps des concentrations cr; En appliquant -la methode des moments a !'equation (35): E. Carlier ao lln = JrncR(x,t)dt 0 On obtient comme expression de la variance en temps: 2 ={.2)' _1.)2 8DZ') cr' \!. \!. + 2 U U X U X (38) Il convient de noter que le temps moyen t = Jl1 est different de xlu quantite appelee temps de transfert par convection pure que 1' on note t, Le calcul du moment d'ordre 1, Jl1 donne: (39) WANG ( 1987) montre que t,. est egal a la moyenne harmonique des temps: (40) A partir de (39) et (40), on peut exprimer le coef ficient de dispersion par: (41) Si l'on compare les expression (41) et (4), on re marque que: (42) Si la section du tube et le coefficient de diffusion moleculaire sont constants, la quantite t th est uninvariant. En appliquant la methode des moments a !'equation (36), WANG (1987) obtient les rela tions suivantes: t' ,;.xju' =tc (43) (44) 2 zj(. ) cr1 =2D1_t (45)


E/fet d'echelle de Ia dispersion dans un conduit karstique L'equation (45) peut se mettre sous Ia forme: 2 cr/ D -ux= L 2t.2 (46) ou cr; / t 2 est appelee variance reduite. La methode de determination du coefficient de dispersion par Ia variance reduite n'est fondamen talement applicable qu'au mode d'injection et de detection en flux. En comparant les expressions (46) et (4), on remarque que Ia quantite (47) est un invariant si a et D sont constants Si l'on introduit le nombre de Peclet defini par: (48) on remarque que dans !'expression (38), la quan tite 8/Pe2 peut etre negligee par rapport a 2/Pe pour une valeur importante du nombre de Peclet; de meme, dans !'expression (39), la quantite 2/Pe pourra etre negligee devant 1. On peut done en deduire que pour une distance importante, !'expression (46), valable pour le mode en flux, constitue une bonne approximation pour le mode en residant. 3. DISPERSION DANS UN CONDUIT A GEOMETRIE FRACTALE 3.1. LA MATHEMATIQUE FRACTALE Dans ce qui suit, nous reprenons sommairement 1' excellente presentation de la matMmatique frac tale de WHEATCRAFf & TYLER (1988). MANDELBROT (1983) a montre que les formes tres irregulieres, telles que les fractures, les rivie res, les c6tes littorales, pouvaient etre caracterisees et quantifiees par une nouvelle methode matbema tique qu'il a appelee fractale. En geometrie eucli dienne, on peut mesurer une ligne a partir d'une unite de longueur g: L=Ng (49) N est le nombre d'unite de longueur necessaire pour mesurer la longueur L. Si l'on diminue l'unite f:, i1 con vi endra d'augmenter N car L est constante; par exemple, si 1' on divise t: par trois, N devra 8tre multiplie par trois. MANDELBROT (1967) a que certains objets naturels ne presentaient pas un produit Ns constimt: une cote littorale devient de plus en plus grande si devient de plus en plus petit: limNs ex> (50) e-.0 MANDELBROT (1967) a montre qu'il existait une dimension dans laquelle Ia mesure de !'objet restait constante: F= Nsd = const. (51) F est independant de s, d est appele dimension fractale. L'expression (49) peut s'ecrire: (52) Considerons une ligne fractale de longueur L1 ; sa projection sur une droite aura pour longueur L. Si 1' on choisit comme pas de mesure cette longueur L, !'expression (51) devient F = Ld et la longueur fractale L1 s' exprime par: (53) WHEA TCRAFT & TYLER ( 1988) definissent une tortuosite fractale: (54) Comme d > 1, Ia tortuosite devient infinie si la distance rectiligne devient infinie. La tortuosite fractale manifeste done un effet d' echelle. 3.2. EFFET D'ECHELLE DU COEFFICIENT DE DISPERSION Si l'on considere un conduit a geometrie fractale au niveau duquel on effectu e une injection et une detection de traceur en flux, le coefficient de dis persion peut etre determine par la variance reduite selon la formule (46). Il faut noter que si l'on ef fectue mesure de concentration a une distance rectiligne x de 1' entree du traceur, la distance reelle parcourue par le traceur est: (55) L' expression ( 46) devient done :


42 (56) u1 est la vitesse moyenne fractale que 1' on peut relier a la vitesse moyenne rectiligne apparente: I 1-d d-1 U I = X I tc = E X U (57) Si le debit Q et la section S du conduit sont cons tants u; = Q IS est constante; dans ce cas, !'equation (57) montre que six augmente, u" diminue. La vitesse moyenne rectiligne apparente ma nifeste, elle aussi, un effet d'echelle. Si u; est inconnue et si l'on mesure u" =xI tc, on determine alors un coefficient de dispersion appa rent par !'expression (46); or le coefficient de dis persion reel est donne par la formule (56). En combinant les expressions (46), (56) et (57), on obtient une relation entre DL1 et DL: (58) Comme D1 1 est constant et d > 1, le coefficient de dispersion apparent DL diminue si la distance x augmente, ce qui est logique puisque les vitesses moyennes rectilignes apparentes mesurees a des distances de plus en plus grandes seront de plus en plus faibles. Si a l'entree du conduit, on mesure le debit Q et la surface S, alors u; est connue. Comme tc = xf I u;, !'equation (56) peut se met tre sous la forme: (59) En combinant (59) et (55), on peut exprimer le coefficient de dispersion apparent DL: D D 1-d d-1 t = l.le x (60) : E. Carlier Comme DL,est constant et d > 1, coefficient de dispersion apparent augmente avec la distance; il y a effet d'echelle. WHEATCRAFT & TYLER (1988) ont utilise la solution de !'equation monodimensionnelle d'advection-dispersion dans une co1onne infinie de milieu poreux sujette a une injection a concentra tion imposee (BEAR, .1979, p. 264). Ils ont etudie l'effet d'echelle sur la dispersivite a., caracteristi que du milieu poreux consideree comme inva riante. Si le milieu poreux est defini par une di mension fractale, ils montrent que la dispersivite a. mesuree sur le terrain est reliee a la dispersivite fractale a.1 invariante par la relation: (61) Cette relation est identique a !'expression (60) aveca.=DL/ u; eta.1=DL11 u; . 4. CONCLUSION Dans un conduit rectiligne ou a faible tortuosite, le coefficient de dispersion est une fonction de la distance parcourue et du temps tant que les vites ses des particules du traceur sont correlees par la loi parabolique de distribution. Lorsque la distance limite de correlation des vitesses est depassee, le coefficient de dispersion devient constant et egal au coefficient de dispersion de TAYLOR (1953). Le choix de la methode de determination du coef ficient de dispersion est fonction des conditions d'injection et de mesure; !'utilisation de la methode classique de la variance en temps reduite n' est rigoureusement applicable qu' en mode in jection-detection en flux; elle est, toutefois, une bonne approximation pour le mode residant si la distance de parcours est importante. Quai qu'il en soit, cette methode ne peut etre utilisee que pour une dispersion de type Taylor, done au-dela de la distance limite de correlation des vitesses. Dans le cas d'un conduit a tortuosite fractale, le coefficient de dispersion fractal D1 1 est regi par les memes lois qui decrivent !'evolution du coeffi cient de dispersion dans un conduit rectiligne; en revanche, le coefficient de dispersion apparent manifeste un effet d'echelle, sans limite asympto tique.


Effet d'echeUe de Ia dispersion dans un conduilluustique 43 .. ANNEXE: EQUATION STOCHASTIQUE DE LA DISPERSION DANS UN CONDUIT Posons: c(x,t)=c(x,t)+c'(x,t) Le flux moyen sur Ia section du tube s'exprime comme Ia somme d'un flux convectif Fe= uc et u(x,t)= u(x,t)+ u'(x,t) d'un flux dispersif f= (u'c')": C I (x, f) et U I (x, f) representent les ecarts a la con centration moyenne c (x, t) et a la vitesse moyenne u(x,t). L'esperance matMmatique dec' (x, t) et u' (x, t) est nulle: E (c)= c"; E (c') = c'" = 0 E (u) = u"; E (u1 ) = u = 0 Le flux total au travers Ia section du tube s'ecrit: F=u + u':X< + c') . . F = u c + u c' + u'c + u c' Prenons I' esperance matMmatique de (Al): . ( )" F = u c +\u'c' BIBLIOGRAPHIE (Al) (A2) ANDERSON, M., P. (1979) Using models to simulate the movement of contaminants through ground water flow systems. CRC Crit. Rev Environ. Control 9, pp. 97-156. ANDERSON, M., P. (1984) Movement of contaminants in groundwater: groundwater transport advection and dispersion. In Groundwater Contamination, National Academy Press, Washington, pp. 37-45. ARYA, A., HEWETI, T., A., LARSON, R. & LAKE, L., W. (1985) Dispersion and reservoir heteroge neity. paper SPE 14364, 60th Annual Technical Conference, Soc. Petrol. Eng., Las Vegas, pp. 22 25 BEAR, J. (1979) Hydraulics of Groundwater. McGrawHill, New York, 569 p. GELHAR, L., W GUTJAHR, A., L. & NAFF, R., L. ( 1979) Stochastic analysis of macrodispersion in a stratified aquifer. Water Resour. Res., 15, pp. 1387-1397. KLOTZ, K (1989) On the applicability of the second order dispersion model. Water Resour. Res., 12, pp. 2497-2500. KREFT, A. & ZUBER, A. (1978) On the physical meaning of the dispersion equation and its tions for different initial and boundary condi tions. Chem. eng. Sci., 33, pp. 1471-1480. (A3) L'equation de conservation s'ecrit: aFjfJx =-acjat =-ac fat-ac' jar (A4) Si l'on prend l'e'Sperance matMmatique de (A4), il vient: aFjax = -ac" jar (A5) En considerant la vitesse moyenne u constante, et en combinant (A3) et (A5), il vient: ac" jar+ u"ac /ax+ a! ;ax= o (A6) MANDELBROT, B., B. (1967) How long is the coast line of Great Britain ? statistical self-similarity and fractional dimensions. Science, 155, pp 636-638 .. MANDELBROT, B., B. (1983) The Fractal Geometry of Nature. W.H.Freeman, New York, 468 p. MANDELBROT, B., B. & VAN NESS, J., W. (1968) Fractional Brownian motions, fractionnal noises and applications. SIAM Rev 10, 422 p. MATHERON, G. & MARSILY, de G. (1980) Is trans port in porous media always diffusive ? A coun terexample. Water Resour Res., 16, pp. 901-917. NERETNIEKS, I. (1983) A note on fracture flow dis persion mechanisms in the groundwater. Water Resour. Res., 19, pp. 364-370. PEAUDECERF, P. & SAUTY, J., P. (1978) Applica tion of a mathematical model to the characteri z,tion of dispersion effects on ground water quality Prog Water Techno! 10, pp. 443-453. PICKENS, J., F. & GRISAK, G., E (1981) Scale dependent dispersion in a stratified granular aq uifer. Water Resour Res., 17, pp. 1191-1211. ROSS, B. (1986) Dispersion in fractured fractal net works. Water Resour. Res 22, pp. 823-827


44 SAUTY, J., P. (1977) Contribution a /'identification des parametres de dispersion dans les aquiftres par /'interpretation des experiences de These Dr. Ing., Univ. Grenoble, 157 p. SUDICKY, E., A. & CHERRY, J., A. (1979) Field observations of tracer dispersion under natural flow conditions in an unconfined sandy aquifer. Water Pollut. Res. Can., 14, pp. 1-7. TAYLOR, G., I. (1953) Dispersion of soluble matter in solvent flowing slowly through a tube, Proc. Roy. Soc .London,233,pp.446-468. TOMPSON, A., F., B. & GRAY, W., G. (1986a) A second order approach for the modeling of dis persion transport in porous media, 1, Theoretical development. Water Resour. Res., 22, pp. 591-599. E. Carlier TOMPSON, A., F., B. & GRAY, W., G (1986b) A second order approach for the modeling of' dis persion transport in porous media, 2, Application to solute motion in pipes and capillarity tubes. Water Resour. Res., 22, pp. 601-614. WANG, H., Q. (1987) Modelisation des transferts de masse en milieu sature a double porosite These Dr., Univ. Paris Sud, 271 p. WHEATCRAFT, S., W. & TYLER, S W (1988) ex planation of scale-dependent dispersivity in het erogenous aquifers using concepts of fractal ge ometry water Resour. Res., 24, pp. 566-578. ZUBER, A. (1974) Theoretical possibilities of two-well pulse method. lsotop. Techn. in Groundwater Hydrology, 2, IAEA, Vienna, pp. 277-294.


TheoreticalaTutkppliedKarstdlogy, vol. 911996, pp. 45-49 Influence de Ia structure sur Ia variabilite spatio-temporelle du coefficient de dispersion Cas theorique d'un faisceau de conduits karstiques soumis a une injection continue en absence de diffusion moleculaire lecoulement rapidel Erick CARLIER1 RESUME De nombreux modeles de simulation du transfert de masse en karst et milieu fissure reposent sur !'hypothese d'un coefficient de dispersion invariant dans le temps et l'espace pour une vitesse moyenne d'ecoulement cons tante. A partir d'un ensemble de conduits cylindriques horizontaux, on montre que le coefficient de dispersion est, generalement, une fonction du temps, de la distance et de la structure du milieu karstique ou fissure. Lorsque la frequence de repartition des sections obeit a une loi exponentielle particuliere, le coefficient de dispersion n'est plus fonction que de la distance. Mots milieu fissure, karst, transfert de masse, coefficient de dispersion. Influence of the structure medium on the variability of the coefficient of dispersion Theoretical case of pipes submitted to a continuous injection of tracer with no molecular diffusion (fast flow) ABSTRACT A lot of transport models applied to karst and fractured media assume that the coefficient of dispersion is con stant when the groundwater velocity is constant. Starting from horizontal pipes, one shows that the coefficient of dispersion is a function of time, distance and structure of the lnedium. When the distribution of pipe sections follows an exponential law, the coefficient of dispersion is a function of the distance only. Key words: fractured media, karst, mass transfer, coefficient of dispersion. 1. INTRODUCTION Les travaux remarquables de TAYLOR (1953) ont montre que, dans un conduit cylindrique rectiligne en absence de diffusion moleculaire, le coefficient de dispersion etait dependant de Ia distance; alors qu'au deUt d'une certaine distance de parcours, la diffusion moleculaire devenant preponderante, le coefficient de dispersion s'exprimait par une constante, fonction de la vitesse moyenne, du rayon du tube et du coefficient de diffusion moleculaire. 1l..aboratoire de G6()sciences e11virqnmenWes, Facult6 de Sciences Jean Perriri,Umversite d' Artois, rue Jean Souvraz, sp 18, f.-:62307 c&iex, france:


46 De nombreux travaux experimentaux ont montre unc liaison entre I' ordre de grandeur de Ia dis sion et l'echelle d'etude du milieu (RENAULT et al., 1975; FRIED & COMBARNOUS, 1971; LALLEMAND-BARRES & PEAUDECERF, 1978; ANDERSON, 1979; 1984). Des mesures sur le terrain du coefficient de dispersion ont revele une augmentation de celui-ci avec Ia distance entre point d'injection et point d'observation (PEAU DECERF & SAUTY, 1978; SUDICKY & CHER RY 1978; SUDICKY et al., 1983; SILLMAN & SIMPSON, 1987; PICKENS & GRISAK, 1981). La quantification analytique de l'effet d'echelle de la dispersion a ete I' objet de nombreuses etudes theoriques, notamment basees sur une approche stochastique du phenomene (GELHAR et al .. 1979; MATHERON & de MARSILY, 1980; NE RETNIEKS 1983; KLOTZ, 1989; DAGAN & NEUMAN, 1991) a laquelle peut etre joint le con cept d'une geometric fractale (MANDELBROT & VAN NESS, 1968; ARYA et al., 1985; ROSS, 1986; WHEATCRAFT & TYLER, 1988; BER KOWITZ & BRJESTER, 1991). DAHLBLOM & HJORTH (1986) ont montre que le coefficient de dispersion etait fonction uniquement de Ia distance et de Ia repartition des sections des conduits. Dans E. Carlur cette note, nous demontrons que, generalyment, s'il.ri'y.a: pas d'ecluinge entre les et-i;j Ja .. : moleculaire est : negligeable, Je ; coeffi-.deJJ.t de djspersion est d 0 2ut {inrdr ( x J C=C0 2 :;:C0 1--(2) 1tR 2ut t=O x>O C=O C0 represente Ia concentration initiale(ML -3), C Ia concentration a Ia distance X > 0 et a !'instant t (ML-3), r le rayon de Ia section contaminee a dis tarice X et a }'instant t dont }'expression est (TAYLOR, 1953): (3) La vitesse maxi male dans le conduit, scion 1' axe, est egale a 2u. Le flux (ML"2T) a travers la surface contaminee est: C !r ( x2 ) F 21trv(r)ir = C0 u 1--2 2 1t R 4u t (4) v(r) (L11 ) represente Ia vitesse auxpoints distants de r de I' axe du conduit. Selon 1a tMorie classique du transfert de masse, le flux F peut s'exprimer


Influence de Ia structure sur Ia variobilite spatio-temporelk par !'expression suivante, dans le cas d'un ecou lement monodimensionnel: 6C F==-D-+uC ox (5) ou D (L :vr-1 ) represente le coefficient de dispersion longitudinale. De cette expression, on peut ecrire: uC-F D== oC ox (6) L'expression du coefficient de dispersion dans un conduit cylindrique sans diffusion moleculaire peut etre exprime par: x2 D == ux--pour x < 2ut 2t (7) 2. EXPRESSION DU COEFFICIENT DE DISPERSION POUR UN MILIEU CONSTITUE DE CONDUITS CYLINDRIQUES HORIZONTAUX DONT LA REPARTITION DES SECTIONS s OBEIT A UNE LOI DE DISTRIBUTION f(s) La loi de distributionf(s) (nombre de conduits par m2 ) est telle que: "' fJ (s)is == 1 0 Dans un conduit, la vitesse moyenne u peut s'ecrire: u = a s ou .1 d p a = -8).1 d I avec u exprime en e111 (8) dp!dl est le gradient de pression par unite de longueur et )lla viscosite dynamique (PaT). La vi tesse moyenne dans 1' ensemble des conduits est egale au rapport entre le debit moyen et la surface moyenne: ao Jas2 f(s)ds u=..::.o ___ ao Js!(s)ds 0 L' expression de la concentration est: (1-sf (s)ds 2 I 2ast C =Co -=a:....____,..-----r sf(s)is 47 (9) (10) Le parametre (x/2at) represente la section du con duit au niveau duquelle polluant vient juste d' atteindre la distance x a 1' instant t aux conditions suivantes: C = 0 pour x s 2at C > 0 pour x < s 2at en utilisant le theoreme de derivation sous somme, on obtient comme expression pour oC/ox: 1: f (s)is oC C0 -:;:;;-; -=--------ox 2at (12) De meme,le flux F s'exprime par: [_(1-2 )as2f(s)is 201 4a s t F == C0 ---------fsJ(s)is (13) A partir de ces expresions, on obtient comme ex pression du coefficient de dispersion: x2 D=ux-+ 2t + ""2.a.i [a. j s2f(s)dsu jsf(s)ds ] fJ(s)ds 2 a t 2 a t X 2 at (14)


48 Cette expression montre, qu'en general, le coeffi cient de dispersion depend de Ia dista.Ilfe, du tenips et de Ia structure dans le cas etudie, par Ia fonctionf (s). Ce parameU:e est iitdependant du temps si l'egalite suivante est realiste: 21

Influence de Ia structure sur Ia variabilite spatio-temporelle BIBLIOGRAPHIE ANDERSON, M., P. (1979) Using models to simulate the movement of contaminants through ground water flow systems. CRC Crit. Rev. Environ. Control, 9, pp. 97-156. ANDERSON, M., P. (1984) Movement of contaminants in groundwater: groundwater transport advection and dispersion. In Groundwater contamination. National Academy Press, Washington, pp. 37-45. ARY A, A., HEWETT, T A., LARSON, R. & LAKE, L., W (1985) Dispersion and reservoir heterogeneity. Paper SPE 14364, 60th Ann. Technical Confer ence, Soc. of Pet. Eng., Las Vegas, pp. 22-25. BERKOWTIZ, B. & BRJESTER, C. (1991) Dispersion in sub-representative elementary volume fracture networks: Percolation theory and random walk approaches. Water Resourc. Res., 27, 12, pp. 3159-3164. DAGAN, G. & NEUMAN, S., P. (1991) Nonasymptotic behaviour of a common Eulerian approximation for transport in random velocity fields. Water Resourc. Res., 27, 12, pp. 3249-3256. DAHLBOHM, P. & HJORTH, P. (1986) Derivation stochastique du coefficient de dispersion. Hy drogeologie, 2, pp. 207-214. FRIED, J., J. & COMBARNOUS, M., A. (1971) Disper sion in porous media. Adv. Hydrosci., 7, pp. 169-282. GELHAR, L., W., GUTJAHR, A., L. & NAFF, R., L. ( 1979) Stochastic analysis of macrodispersion in a stratified aquifer. Water Resourc. Res., 15, pp. 1387-1397. KLOTZ, K. (1989) On the applicability of the second order dispersion model. Water Resour. Res . 12, pp. 2497-2500. LALLEMAND-BARRES, A. & PEAUDECERF, P. (1978) Recherche des relations entre la valeur de dispersivite macroscopique d'un milieu aquifere, ses autres caracteristiques et les conditions de mesure. Bull. BRGM (2) III, 4, pp. 277-284. MANDELBROT, B., J3. & VAN NESS, J., W. (1968) Fractional Brownian motions, fractional noises and applications. SIAM Rev., 10,422 p. 49 MATHERON, G. & de MARSILY, G. (1980) Is trans port in porous media always diffusive ? A coun terexample. Water Resour. Res., 16, pp. 901-917. NERETNIEKS, I. (1983) A note on fracture flow dis persion mechanisms in the groundwater. Water Resour. Res., 19, pp. 364-370. PEAUDECERF, P. & SAUTY, J., P. (1978) Applica tion of a mathematical model to the characteri zation of dispersion effects on groundwater quality. Prog. Water Technol., 10, pp. 443-453. PICKENS, J., F. & GRISAK, G., E. (1981) Scale dependent dispersion in a stratified granular aq uifer. Water Resour. Res.,11, pp. 1191-1211. RENAULT, D., MUNTZER, P. & ZILLIOX, L. (1975) Propagation d'un front de pollution dans un aquifere: contribution experimentale a I' etude de la dispersion en milieux poreux stratifies. TSM l'Eau, 70 (4 ), pp. 153-157. ROSS, B. (1986) Dispersion in fractured fractal net works. Water Resour. Res., 22, pp. 823-827. SILLIMAN, S., E. & SIMPSON, E., S. (1987) Labora tory evidence of the scale effect in dispersion of solutes in porous media. Water Resour. Res., 23, pp. 1667-1673. SUDICKY, E., A. & CHERRY, J., A. (1979) Fields observations of tracer dispersion under natural flow conditions in an unconfined sandy aquifer. Water pollut. Res. Can., 14, pp. 1-7. SUDICKY, E., A., CHERRY, J., A. & FRIND, E., 0. (1983) Migration of contaminants in groundwa ter at a landfill, 4. A natural-gradient dispersion test. J. Hydrol., 63, pp. 81-108. TAYLOR, G., I. (1953) Dispersion of soluble mater in solvent flowing slowly through a tube. Proc. Roy.Soc.London,233,pp.446-468. WHEATCRAFT, S., W. & TYLER, S., W. (1988) An explanation of scale-dependent dispersivity in heterogeneous aquifers using concepts of fractal geometry. Resour. Res 24, pp. 566-578.


Theoretical & Applied Karstology,yol. 911996, pp. 51-61 Relation entre les.caracteres physiques et les champs de leur distribution dans les roches karstiques Borivoje F. MIJATOVIC1 RESUME Dans les roches karstiques la structure des champs de distribution spatiale des majeurs parametres physiques, tels Ia permeabilite et la porosite peut etre caracterisee par: (1) la distribution statistique dans les blocs-matrices, (2) Ia distribution ponctuelle dans le reseau d'ecoulement, et (3) l'effet d'echelle sur ces champs. Ces considera tions demandent a etre bien expliquees, notamment dans le domaine de Ia fracturation des massifs calcaires dont les caracteres physiques et hydrauliques sont sous )'influence de l'effet d'echelle On constate souvent en ter rains karstiques une coexistence de deux types bien distincts d'ecoulements: l'un, correspondant aux classiques rivieres souterraines, lie aux collecteurs des grandes dimensions du reseau karstique, )'autre, conforme a une va riete de circulation plus ou moins lente, dans les blocs-matrices volumineux mais relativement peu permeables. II se pose alors Ia question de Ia repartition des valeurs de permeabilite et de porosite efficace dans un reseau de conduits connectes et des blocs-matrices lorsque l' echelle varie. Mots-des: champ de distribution, permeabilite, porosite, effet d'echelle, ecoulement souterrain, aquifere karstique, emmagasinement. Relationships between physical characteristics and the fields of their distribution in the karst rocks ABSTRACT In the karst rocks the structure of the spatial distribution fields of the main physical parameters i.e. permeability und porosity could he characterized by these items: ( l) statistical distribution in block matrices, (2) site-related distribution in drainage network, and ( 3) scale effect on these fields. These qu.estions are to be carefully ob served and explained, especially in the field of limestone rocks frac turing, whose physi ca l and hydraulic proper ties are influenced by the scale effect In karst terrains coexistence of two types of flows has often been found: one of them is related to classical underground flows connected with channels of huge dimensions in a karst network, and the second one which fits slowed (retarded) circulation in voluminous but relatively poorly perme able block-matrices. We are interested in the values of permeability and effective porosity in a network of con nected karst channels and block-matrices when the scale is changing. Key words: field of distribution, permeability, porosity, scale effect, groundwater flow, karst aquifers. storag e . Universite de Novi Sad, Place D Obradovic 8, YU-21000, Novi Sad, Yugoslavia.


52 AVANT-PROPOS Une approche methodologique dans l'etude des aquiferes karstiques necessite une adaptation du schema des systemes d'ecoulement, tout a fait different de celui reconnu en milieu meuble ho mogene. Pour clarifier Ia terminologie, rappelons qu' en roches carbonatees, 1' aquifere existe dans un milieu tres permeable, discontinu et beterogene, dans lequel 1' eau s. ecoule a travers un reseau de fissures ouvertes et diversement connectees entre les blocs-matrices tres peu permeables par rapport au reseau de fissures. Le reseau des conduits karstiques (ou reseau kars tique) designe !'ensemble des larges fissures, che naux et cavites naturelles connectes, constituant dans !'aquifere un systeme de collecteurs a circu lation d'eau permanente, temporaire ou abandon nee (BURGER, 1983). Ce modele de !'aquifere que nous observons parfois sur le terrain karstique, couvre !'ensemble des eaux comprises dans le reseau et dans les blocs-matrices interconnectes dans Ia zone saturee de 1' aquifere, ce qui permet !'utilisation de !'expression de nappe karstique, qualifiant pourtant un milieu essentiellement dis continu. Par consequent, l'ecoulement souterrain dans les roches karstiques peut etre considere comme dO a Ia reunion de deux types de circula tion: ( 1) une circulation macroscopique dans les collecteurs privilegies (larges fissures, chenaux, cavites), consistant dans une sorte de reseau d'ecoulement, et (2) une circulation microscopique dans des itineraires secondaires, done dans des blocs-matrices (ilots) ainsi delimites, qui assument Ia majeure partie de la fonction de stockage du massif karstique et alimentent les collecteurs pri vilegies. Dans l'optique de Ia presente communication, l'objectif majeur est Ia connaissance des systemes d'ecoulement souterrain, definis par les champs de potentiel hydraulique ( et de vitesse des ecoule ments ( q ). Pour y parvenir, on aura besoin de disposer des caracteres physiques du milieu aqui fere, c'est-a-dire, les champs de permeabilite et d'emmagasinement, qui fournissent les parametres des calculs necessaires dans l'etude et les investi gations hydrogeologiques complexes et de simuler leurs effets. L' estimation des champs des caracteres physiques, tels que permeabilite et porosite, doit permettre done de rendre compte des multiples comporte ments des nappes dans les aquiferes karstiques. B. Mijatovic LA PERMEABILITE ET LE DEVELOPPEMENTDELA KARSTIFICATION DANS LES ROCHES CARBONA TEES .,. : : LE ROLE DE L'EvOLUTION .STRUCTURALE Chaque analyse exhaustive des donnees sur les proprietes de permeabilite et d'emmagasinement en regions calcaires met en evidence certaines regles generales de leur evolution, communes a toutes les roches carbonatees, a savoir: (a) }'existence d'une permeabilite de fissures, condi tion essentielle du debut de toute erosion karsti que. Selon la disposition et !'orientation des fissu res, cette erosion entraine une karstification d'extension regionale et locale, en faveur des zo nes tectoniques priyilegiees; (b) Ia diminution de 1' erosion karstique en fonction de la profondeur, resultant des trois principaux facteurs: permeabi lite des fissures, vitesse d'ecoulement et action corrosive des eaux souterraines; (c) l'accrois sement de !'erosion karstique au voisinage des zones privilegiees de drainage, la permeabilite des fissures, et la vitesse d'ecoulement etant en fort accroissement; (d) le developpement d'une zona lite hydrodynamique verticale ou se localisent plus ou moins quatre zones d'ecoulements a caracteres propres, relatives aux deux zones principales de }'aquifere: denoyee et noyee (MIJATOVIC & BAKIC, 1967) (Fig. 1). Une approche de I' analyse systemique sur les champs de permeabilite et d' emmagasinement met en relation reciproque ces regles generales avec les trois classes de facteurs: le contexte geologique, notamment tectonique, les variations climatiques et paleoclimatiques, ainsi que les facteurs geomor phologiques et paleogeographiques. Ceci permet de rendre compte des multiples comportements des ecoulements karstiques, reposant sur deux criteres d'essence, respectivement geologique (organisation geometrique du reseau des ecoule ments) et hydrodynamique (aptitude a emmagasi ner et a conduire les eaux souterraines). Le carac tere discontinu, beterogene et anisotrope du milieu karstique, depourvu a I' origine de proprietes re servoirs significatives, represente un obstacle majeur pour comprendre, ensuite traduire en terme quantifiable les indices typiques du reservoir. 11 importe de rappeler, a ce propos, que !'acquisition des proprietes des reservoirs karstiques est un pro cessus complexe, evolutif a l'echelle des temps geologiques et gouverne par un facteur determinant-la fracturation des roches carbonatees sous l'effet de contraintes tectoniques.


Relationelltre les caracteresphysiques et les.chQnips deieuriistdbution 53 ,: .. . : .. .. ; G, ... G, .. ... :.: tension distension -G3 Fig. 1. Relation entre les diagramme, B: . et eilip8e clsaiUemerlt (:dextres et senes tres); 2: de teltsi'on;'3: .. 1988) > . .. .: ... Relation be(Ween principM t!!citiiiic_; of fJSstires: A: Blofdiagm't(l, B: : Ellipsoid and ellipse of CoiJiufdted 2; TerisiC,n 3iCompresswn or re/axationfissures:(MIJATOVU5, 1988); . .. . . . . A cet egard .ta ctmnaissance : d es voirs en empleyailt que et lnicrc>tectoniqoe L'etude de hi fracturation oul'on consid.ei'e quement. toutes. les dlscoiitiriuites _-Ocs airisi que ies niiGrofissures). en 1 ronction de i'echene, revet rigine .tectonique .. dont I'identi.fication de precHie. peut .'etre etablie .A partir d'une classification :dnematique; compte t enu des .. de lion, cRAZA<;K, 1984; Ain$i,j)oui ini domie de conh;Unte trlaXiale, ies resliiWttesso.nt, definies._en.:foilctiqn: de trois p#ncipales: direction d'al (XO)..,ditection )nteririediaire (YJ, di.rect.lon de. (Z). Ces trois direc tions : llli ellipsorde 4e contrainte avec trOis contrairites .o- O'i; u]appetees contiairite principale; coiitrrunte i.nter.ffiediaiie et contJ:ainte Jnirilmuni (Fig. 1). Cet etat de con. b:ainte, en fonctiori des vhleurs relatives de dif motitte bien l'existerite .trois <>nentaiions princip .ales du plan, o\ll' 6tat de coiitiainte egalemenl se par une . 9e cisaillemerit coordonnees tect<,>niques, riotees pat COtiverifi(}il a; b, et C, Ce : sorit(Pig: 'f): i le plan be. sourriis a une tension. (u1 cr2>crl>O)(t:>ci;.> 0); )es plans de cis.ru_ llement corijogues ( el senestres) d.'otieritati o n hOt en coupes circuiaires .de l'ellipsO.i:de de contrairites e t de de-formations.


54 L'approche structurale; tectonique et microtecto nique, permet ams1 de distinguer et d'individualiser a l'echelle du massif et de l' affleurement, trois principaux groupes de. fissures ou de ruptures: ( l) fissures de tension, (2) fissures de compression ou de reHichemerit, et (3) fissures de cisaillement conjuguees (dextres et senestres). Les nombreuses mesures effectuees sur les affleu rements de roches carbonatees apportent la a--+---L orientation des contraintes principales, celles de 0"), 0"2 et O"J par rapport a des cteformatiqns, C()ndi tionne son jeu par le developpemerit de ces tiois types de fractures, soit en ouverture, soit en Jermeture et)nflue par ce biais surhl permeabilit6 du milieu fissure. Dans les conditions d'un etat de contrairite favorables a une ouverture sei:ont celles proches d'une posHion orthogonale a cr3 (contrainte minimale) alors que les fractures or thogonales a cr1 (contrainte maximale) auront tendance a se fermer. Quant aux fractures de ci saillement conjuguees dles traduisent un type de discontinuite marque par des accidents de decro chement, tres favorables pour un developpement notable des ecoulements souterrain s (Fig. 3); Du point de vue hydrogeologique on peut remarquer que toutes les directions et. plans de fracturation ne presentent pas un interet equivalent silr le develoP"' pement du processus de karstification: L' ecoulement souterrain, 6troitement lie ala peimeabilite etau gra dient hydraliliqu e es t une res\lltarit e d'imeinte ractiori B. MijatoYid. confirmation de ce modele theorique, pouvant expliquer l' apparition des fissures, anglilaii : es et leUr r61e lors qes deformations . Ainsi on constate souvent dans la tr3Jlche sliperieure massifs rocheux que les relachements de. con traintes ant pour cause une action gi:avitatioilnelle ou .la decharge lithostatique due a l' erosion, (DROGUE, 1980). lG' ... G, z a beaucoup plus complexe des compbsantes iectopi ques. st:rUcturahi s et du champ vedeurs vitess e des ecoulernents piutotqu'une fonticin simple de Ia densitC de ftactUration. La mise en evidence de la conn

Relation entre lt!s caracteres physiqu(!,s et ks champs de leur 55 Fig. 3. Bloc-diagramme de5 failles observees dans des terrains calcaires: les trois jeux succe$Sifs sont numerotes: 1 (normal), 2 (inverse), 3 (normal); les failles inverses 2a. et 2b sont corouguees, /. . e CMUATOVIC, 1988). . Blockdiagrain of faults obsen.ed in ter-. rains: the three successive displacements are Indicated by: 1 (normal faults), 2 (inverse faults), 3 (normal faults); the inverse faults 2a iirid 2b are . conjugated, (MIJATOVIC, 1988) . QUELQUES RESULTATS DES MESURES DE CARACTERES PHYSIQUES DANS LES ROCHES KARSTIQUES EFFET D'ECHELLE SUR LES PARAMETRES Plusieurs auteurs se sont penches sur le probleme de Ia representativite des mesures ponctuelles de la permeabilite exptimee par le coefficient K (L!'}, dcduites des essais d'injection d'eau de type Lu geon ou des essais de pompage. Cependant il faut remarquer qu'au sens strict du terme, la permeabi litc caracterise les milieux homogenes et isotropes et s'exprime par un tenseur de permeabilite qui Ia decrit dans trois directions principales de l'espace. En milieu Mterogene il convierit d' appliquer la notion de conductivite hydraulique, qui est !'aptitude d'un conduit nature/ ou d'un ensemble de conduits, dans un milieu aquifere discontinu, a permettre le mouvement de l'eau sous l''et d'un de clulrge hydraulique donne, dont La di rection en general de celle de l 'ecoulement (CASTANY & MARGAT, 19'Z7). La il est question iei doit eritendue dans ce sens J?ans trois regions clU ](arst Dinariqoe (BUSko blato, Penica et barrage$ de reienue), ime anatyse statistique permeabilitt : (par essai$ d'injection) a 'ete faite par BOREU..f& )? A yuN (1967) sur !'hyPothese de Ia correlation entre les caracteres physiques des roches ef l' apparitiori possi ble des cavernes et conduits karstiq(les. La distribu tion des mesures de permeabilite est presentee de exemplaiie par des courbes .de trecjuence cu mulative. De so a.% de valeurs prtserttees se si tuententre lxlo-o et Sxlo-5 rnls;Ja distribution de ces valeurs est log-normai en approxiniation et la queue de distribution est dans la direction des grandes penneabilites, ce qui indique que la probabi lite d'une zone tres permeable est. assez reduite (KIRALY, 1975). Dans les forages decrits, la permeabilite diminue nettement avec la profondeur, de sorte que les grandes valeurs de perineabilite se trou vent dans les premiers 60 m des forages (Fig. 4). tH lfl3 E-14 P-14 P-16 lP-n] tJ E-16 P...-16 P-18 P-8 :i5 700 0 P-14 P-3 P-18 E .... f-18 P-14 A-81 c. fl14 P-10 A-& In ,C) .... -E-14 P-3 A-6) -> c 0 B-3c N 650 E 'k2 c ( ... i "0 c 0 0 -c. 600-0 sur terrain D forage! a 0 forages a 550-BuJko )Jbito: La frequence des intervalles permeabillte elevee en fonctlon de profondeur (BORELLI & PA VLIN, 1967). BuSk{J Bl!Uo:. The frequency of increased permeability sectio.ns as a function of depth (BORELLI & PAVLIN, 1967): ,.


56 Les auteurs admettent une exponen tielle de Ia permeabilite avec Ia profondeur, mais i1 y a de fortes chances pour que Ia tendance mec ne soit plus demontrable, a partird'une cer taine profondeur oii les tests statistiques s ont tres approximatifs, a cause de Ia dispersion de valeurs mesurees. Les distributions citees caracttrisent vraisemblablement les blocs-matrices a faible permeabilite (BORELLI & PA VLIN, 1967). La Figure 5 montre Ia distribution de 399 mesures de permeabilite par essais d'injection dans trois synclinaux du Jura suisse (KIRALY, 1975; SIMEONI, 1976). Environ 80% de toutes les .20 -;'! 10 B.-Mijatovic valeurs sont inferieures a 2x 10'6 rnls et les plus grandes valeurs ne' depassent pas 4x 10'6 m/s. La permeabilite diminue avec Ia profondeur d'une significative dans les p:remiers 80 a 90 m environ, selon une loi exponentielle du type K = K0 exp(-ah) ou, a: pente de Ia droite; K0 : conductivite hydrau lique a Ia zone epictermique du terrain; h: profon deur depuis Ia surface (SIMEON!, 1976). 30 40 (x to7 > permeobilitb.. m Fig . s. Distribution de permeabiMte du .Jura ;sQ.isse.' tie Ia Brevine; Blanc: synclinaux du Locle et -., -. .. . -. Di$tribution of permeability in the (hr'e J'ura. Dotted:_ syncliJJ-e, of Brevine; White. synclines of ; -.--.' .' -_ -, '''/ .-_' ; La loi exponentielle s'observe jusqu'a -140 m dans deux forages isoles. La variabilite spatiale de Ia pernieabilite dans le sens lateral etant aussi un caractere structural important du champ K, les auteurs ont presente I' analyse de variance par mo dele emboite de 399 mesures de permeabilite. fectuees dans 8 .forages, eux-memes repartis dans 3 synclinaux. Pour savoir a quel niveau se situe Ia plus grande heterogeneite (entre Ies valeui:s des forages a I' interieur des synclinaux ou entre les valeurs individuelles a l'interieur des forages) un Lableau d'analyse de variance calculee par Simeoni donne une n!ponse claire a ces questions (voir Tableau 1): Ia variabilite est maximci.Ie entre les synclinaux ( = 75,8 xl0-14), ce qui semble a demontrer le caractere regional du developpement de Ia perm6abilite. II est a remarq1ier que Ia variabilite des mesures dans Ies forages ( = 29,32 X w-14) est plus importante que Ia variabilite des moyennes de forage a l'interieur des synclinaux ( =10,32xl0-14). A ce propos, KIRALY (1975) admet que Ia moyenne des blocs matrices peu permeables semble .etre distribuee d'une relativement homogeiie A des trois bassins hydrogeologiques. Dans une region karstique aux environs de Bel grade, sous I' effet de pompage intense, Ia nappe karstique captive est forter:nent deprimee et sa remise en pression, en vue de restaurer les niveaux piezometriques par 1' alimentation. aitificielle, s'avere indispensable. II s'agit alors d'un reservoir karstique du Trias superieur sous-jacent au depot neogene dont Ia puissance peut atteindre plus de 250 m environ. La nappe est captive ou ijbre sui vant Ies secteurs de pompage. Les forages d'exploitation actuels (au nombre de 5) se trouvent dans Ia zone ciii Ia nappe est encore en charge. Le Tableau 2 fournit des donnees hydrogeologi ques sur les parametres K et T dans les forages d'cxploitation, deduites des essais de pompage (MIJATOVIC, 1982).


Relation entre les caracteres physiques et les champs de leur distribution 57 Tableau 1. Source de variation Somrne des carres Degrede SC/DL Nombre effectif Variances liberte Entre synclinaux 18079,5xi0-14 2 9039,7x10-14 2,5 =75,8x 10""14 Entre forages 2474,7xl0-14 5 494,9x10-14 41,5 0'2 =10 32'10-14 8 Entre mesures 11441,7xl0-14 391 29,2xl0-14 0' =29,26x 10""14 To tal 31995,9x w-14 398 Tablea u 2 Forage Profondeur Epaisseur Debit to-5 K w-3 r Aquifere (m) (m) (1/s) (mls) La) 250 181 30 0,7 La4 375 275 40 1,0 L

58 grande par rapport aux resultats obtenus dans d'autres regions karstiques (KIRALY, 1975). Les rares essais de pompage effectues directement dans les parties des reseaux karstiques tres per meables (cavites, zones de sources) nous rensei gnent sur la permcabihte a grande echelle, done sa distribution locale, car )'influence d'un tel essai s'etend loin, mais bien limite au du rese&u karstique: sur !'ensemble de l'aquifere cette influence ne touche que le reseau karstique. Apres avoir effectue un essai de pompage sur la source de Verdier en France, PALOC (1964) Lrouve une valeur de permeabilite de 9x w-2 m/s a JQ-1 rn/s, et un coefficient d'emmagasinement de 0,038. MIJATOVIC (1970) calcule Ia transmissi vite les courbes de tarissement de deux sources karstiques au Liban (El Safa et Barouk) et trouve une valeur T = 9xl0-1 m2/s; la source Jadro sur Ia cote dalmate de l 'Adri&tique Ia trans missivite s'eleve a T = 4,7 rf.//s. Meme si J'epaisseur de )'aquifere est de l'ordre de 100 a 200 m, Ia permeabilite moyenne est autour de JQ -2 rn/s. MOULLARD et al. (1967) deterrninent Ia transmissivite par essa i de pompage dans une crevasse quaternaire effondree sur une zone de dolines karstiques dans les calcaires du Cenoma noTuronien recouverts par des mames du Seno nien. Les calcaires sont en general compacts (T = 9,8 x m2/s dans certains forages) mais ils peuvent presenter des zones karstiques tres impor tantes (T=l,3 m2/s dans crevasse) avec une per meabilite entre w z et w -3 rn/s. Les essaisde porn page dans une galerie de captage avec un piezo metre, sur Ia cote dalmate merite d'etre mention nes. Les coefficients T et S sont determines sur la base de deux essais de pompage en regim e non permanent (avec deux paliers a part, 15 lis et 40 1/s), a savoir: Ia transmissivite etant de 1,4x X } o l m2/S a } ,24x 1 o l m2 / S et }e Coefficient B. Mijatovic d; enunagas1nement de 0,0709 a 0,0976 (MIJATOVIC, 1980; 1984). Une comparaison de differentes valeurs de per meabilite de transrnissivite et de porosite effieace, obtenues dans les forages et en laboratoire, dans les galeries drainantes et dans les zones de sour ces, montre clairement que l'effet sur ces parametres ne fait aucun doute. Ce est dO a Ia coexistence du resea u des conduits, de densite et de volume faible, mais de permeabilite grande, et des blocs matrices volumi neux mais peu permeables. C' est pourquoi 1' effet d'echelle joue le role preponderal1t dans la deter mination de Ia limite de validite oti du rayon d' action des essais hydrauliques dims les aquiferes k&TStiques. REMARQUES GENERALES Les caracteres physiques les plus importants des roches karstiques sorit les champs 4e permeabilite et de porosite efficace. La connaissance regionale de leur distribution avec des limites spatiales per m e t Ia reconstitution de ces champs en fonction de l, effet d, echelle. Les mesures directes de Ia permeabilite et de Ia porosite montrent que leurs champs sont caracteri ses, generalement, par une tres grande htteroge neite et par un effet d'echelle tres marque. C 'est pour cela que, lors de Ia comparaison des valeurs mesurees, on doit tenir compte deJ'ordre de gran deur de l'echantiilori (KIRALY, 19 75). Ces COf!Siderations demandent a etre bien distin guees. En general Ia fracturation des massifs cal caires est a l'origine de l'effet d'echelle, interve riant pour la plupart des caracteres physiques et hydrauliques. De nombreuses propnet6s intrinse qu es varient quand le volum e considere change.


Relation entre les caracteres physiques et les clulmps de leu.r distributwn 59 Ce phenomene est particulierement marque dans Ies roches fissurees Dans une maniere generaie, on peut dire que la permeabilite d'un fissure decroit notablement des que le volume cotisidere augmcnte; sa direction peut egalernent varier (LOUIS, 1974) : Ceperidant, : l'emploi du modele pour Ia determination de Ia permeabilite moyenne equivalente a l'echelle d'uri bassin versant karstique a donne des resultats qui montrent que Ia per meabilite moyennc a l'echelle du bassin est K Cm/s) 10-J ... 10 .. ---.... ----L'ecoulement des eaux soutertaines etant tribu taire a }'hypothese de l'homogeneite, on C0090ll que dans les aquiferes karstiques cdte hypothese n' a pas de signification reelle. II en est ainsi, en particulier des singularites ponctuelles que l'on trouv e dans ces aquiferes. Quelques reflexions sur l'effet d'echelle dans le milieu karstique montrent qu'il est etroitement lie a l'homogeneite du volume d'aquifere coilsidere. Ainsi, un echantillon de calcaire dont les plus gran des fissures ont un diametre de I' ordre du centimetre ne pourra certainement pas etre consi d ere comme hom ogene a I' echelle du decimetre cube. De meme, un massif calcaire ne pourra etre suppose hoinogene si le plus petit echantillon du volume coilsidere n est pas au moins 50 fois plus grand qu'un volume cont e nant les blocs-matrices avec les conduits connex es. C'est a ce concept de l'effe t d'ech e lle qu'il faut rattacher la critique contre !'opinion generalement admise pour les valeurs moyennes de pernieabi lite et transmissivite, ainsi que du coefficient d' emmagasirtement, proclamees tou jours comme Ies valeurs reduites (MIJATOVIC, 1980). Nous voulons so uligner que ces Valeurs moyen nes, ainsi forini.Jlees, n'ont a ucun sens et se nettement plus que celle obtenue par les essitis de laboratoite et dans les .forageS" (Fig; 7) (iGRALY,l975). . -Face a ces premisses contradictoires, l'effet d' echelle demande etre etUdie sur uri plan fon dainental Quel est, lorsque I' echelle varie, la rela. tion et !'influence des principaux parairte"tres hy drogeologiques sur le comportement multiple des karstiques ? Effet du n!,eou trouvent rnises en defaut par <;ies experiences ac quises darts les investigation hydrogeologiques des aquiferes karstiques. Ainsi, les valeurs assez ele vees de transmissivite et du coeffiCient d'eirunaga s1nement, obtenues sur Ia base des essais de debit dans les reseaux karstiques ou dans les zones de graJides sources, sont representatives a l' echelle d'un systeme des ecoulemerits privi16gies ou bien d'un captage d'eau (grande echelle) et non pas a l'echelle de l'ensemble d'aquifere (petite echelle). Cependant, ces resultats de mesures ponctuelles, qui en cas d' extrapolation pourraient condtiire a de graves erreurs, ont toutefois grande importance dans !'orientation et Ia direction de Ia recherche des eaux souterraines karstiques. Cette conclusion n'est evidemment valable que pour des comparti ments bien delirnites, dont les valeurs d e transrnis sivite et de porosite efficace ne potirront etre con siderees des valeur s moyennes des echantillons d'ordre d e grandeur differents, mais des valeurs de distrlbution Iocale s, qui nous renseignent a la fois sur Ia structure des champs de perrrieabilite, de potentiel hydraulique et de vitesse des ecoule ments dans Ia zone saturee de I' aquifere karstique (Fig. 8)


60 : ... btoc-matdce .. ... :. 0 ..... : ____ .-.r ... : 1': r r : ..... C'est pourquoi le s lois hydrodynamiques des ccoulem e nts dans les aquiferes kar s tiques, relati ves A un niveau d'echelle generale restent inope rantes et sont sou vent inutilisables . Passer a un niveau d observation locale pour regler certains problemes preci s en des point'i directel11ent accessiblc s s'avere plus utile BIBLIOGRAPHIE BORELLI, M & PAVLI N B (1967) Approac h t he problem of undergro un d water leakage fr om th e storage i n karst regions. Actes du colloque d e Dubrovnik, Oct. 1965, A I H S UNESC O pub!., 73, I, p p.l20-138. BURGER, A (1983) P ro spection et captage des eaux souterraines des roches carbonatees du J u ra, So ciete Suisse de l'indu.-itrie du gaz et des eaux, 1037,pp.533-583. CASTANY, G. & MARGAT, J (1977) Dic t ionnaire Bull du BRGM, 249 p DROGUE, C (1980) Essai d 'identification d'un type de structure de magasi n s carbonates, fissures. Ap plication a !'interpretation de certains aspects de fonctionnement hydrogeologique Mem. Soc. Geol. France, 11, pp. 101-108. KIRALY, L. (1975) Rappo r t s ur l'etat actuel d es con naissances dans le domaine des caracteres physi ques des roches karstiques. Hydrogeologie des terrains karstiques, pub/. de l A I H et lUGS, se rie B, 3, pp. 53 67. LOUIS, C., L. ( 197 4) Introduction a 1' hydraulique des roches. Bull. du BRGM, section Ill, 2e serie, 4 pp 283-356. MIJATOVIC, B. & BAKIC, M (1967) Le kars t du Liban. Etude de son evolution d'apres l es recher ches hydrogeologiques. Chronique d'Hydrogeo 1 0 pp 95-107. B.,Mijato.viti .. ..... ,.': :_: __ .:. : .... : ,. _: .. .;.; . . ... :.': Ainsi, on revient problem e d'echelle de re cherche dans les tertains katstiqu e s, pour pouvoir decider A quel niveau d'echantilloim;1ge les rela tions entre les caracteres physique s i es champs de leur distribution soht niieux ptononc e e s et ex ptimees. MIJA T OVI C, B (19 70) Met h od of studying hydrody n amic regime of karst aquife r s by ana)ysis of the discharge c u rve and level fluctuations during re cession Bull. lnst. Geol. and ser. B. 8 pp 41-74. MIJAT OVIC B (1980a) Representativitc des parame tres hydrauiiques obtenus sur la base des essais de pompage dans les karsliques (en serbo-ctoate) 61!. Symposium Yougoslave sur !'Hydrogeologie et G(!ologie d'ingenieur, 1, pp. 2 4 1-253. MIJ A T OVIC, B. (l980b) Effet d'echelle sur les para metres hydrau l iques dans l e karst. Congres Geplogique International Paris, Sec. 15, W 0167, 3 p. MIJATOVIC B. (1982) Etude hydrogeologique de La nappe karstique dans La region de Nepricava pres de Belgrade. Evaluation des reserves dispo nibles des eaux souterraines et Ia progrwse du regime d'exploitation dans les conditi ons natu relles Rapport incdit de I'Instittit GEOZAVOD, pp. 1-75, Belgrade. MIJATOVIC, B. (1984) Captage par galeri e dans un aquifere karstique de 1a cote Dalmatc : Rimski Bunar, Trogir (Yougoslavie). Hydrogeol o gie d e s terrains karstiques-Des hisioires des cas. AIH, lUGS, U N ESCO, 1, pp. 152..:.155, Heise, Hanno ver. MIJATOVIC, B. ( 1988) Rapport sur l e theme 2: Fissu ration et karstification Quatrieme Colloque d'Hydrogco)og ie en pays calcaire .et en milieu fissure, 29 Sept 1 Oct, AniuJ!es de


Relation entreles caracteres phy#ques et les dikut disiribution 61 l'Universite de BesQ.nron, Memoire Hors S6rie, 6, 2; pp. 432-438. MIJATOVIC, B., F. (1994) Possibilite de la regulation d'une nappe karstique dans Ia region de priava pres de Lajkovac (en serbo-croate): jt Kecueil des iapports du Comite pour le Karst et La Speteologie, 69, pp. 81-'-97, Academic des Sciences et des Arts, Belgrade : . MULLARD, L. eta/. (1967) Exploitation d'unenappe karstique captive a exutoires sous-marins. Actes du Colloque de Dubrovnik, Octobre 1965, AIHSc UNESCO, puhl. 73, /, pp. 237-250. NEWBERRY, B. (1968) The perched water table in the Upper limestones aquifer of Malta. J. Water Engineers, 22/87 pp.551-570. FALOC. H. (1964) Caracteristiques hydrogeologiques des dolomites de la region languedocienne. Memoires du Centre d'Etudes et de Recherches Hy-drogeologiques; 1, pp; 123-127. RAZACK,. M. ( 1984) Application de methodes numeri ques et sUitis!iques a !'identification des reser voirs {T.Ssures carbonates:. en hydrogeologie. These de DOCtoral; Montpellier; pp ; 12-24 SlMEONI, G . P. (1976)Etude de la des formations calcaires du Jura neuchatelois Bull du centre d'Hydrogeologie a l'Universite de Neuchiitel, 1, pp. 9-18.


. Theoretical and Applied KtustQlogy, vol. 9/1996, pp. Comparative Observations in Karst Formations of Parnass and Arcadia Mountains [Greece] Kyriaki A. PAPADOPOULOU1 ABSTRACT The paper deals with the study of two great karst areas in Greece. The first one belongs to the Pamass Mountain (Central Greece) while the other one to the Arcadia Mountain (South Greece). A large number of dolines (300) have been studied in order to identify their common characteristics, differences, and the rules that control their development. The final purpose of the paper is to emphasize the rules that control the genesis of those areas. Key Words: karst, doline, ponor, bauxite, corrosion, past weathered relief. Observations comparatives dans /es formations karstiques des Montagnes Parnass et Arcadie (Grece) RESUME I.e travail concerne l'etude des deux grandes zones karstiques de La Grece. La premiere est situee dans les Montagnes de Parnass (Grece Centrale), tandis que La deuxieme se trouve dans les Montagnes d'Arcadie (Grece du Sud). Un grand nombre de dolines (300) a ete etudie en vue d'identifier leur caracteristiques communes, les differences et les lois qui controlent leur developpement. I.e but final du travail est La mise en evidence des regles qui controlent La genese de ces zones. Mots cles: karst, doline, ponor, bauxite, corrosion, paleoreliefs. INTRODUCTION By the study of 300 dolines, a karst analysis of Pamass Mountain in central Greece (21 0 km NW of Athens) and Arcadia Mountain in central Pe loponnissos (275 km SW of Athens, 338 km NW of Tripolis) has been carried out (Fig. 1). The se lected sites cover a total area of 110 km2 at an al titude of 1200-2280 m and 175 km2 at an altitude of 880-1720 m, respectively. Several studies have been carried out in the past, concerning the areas of Parnass and Arcadia mountain, with special emphasis on their geology: PAPASTAMATIOU, 1960 & 1962; MONOPOLIS, 1971, KALPAK.IS, 1979; KAROTSIERIS, 1978 & 1981; RENZ, 1930 & 1940; PHILIPPSON, 1892 & 1930; CELET, 1962. GENERALITIES GEOLOGICAL AND TECTONIC ASPECTS Dolines of central Parnass Mountain are mainly developed on the neritic, medium bedded lime stona formations of the Parnass-Gionas Unit of the Late Cretaceous, where the sedimentation was three times interrupted, leading to the development of three bauxite horizons. The basement consists of pre-Carboniferous formations, the domes of which may act as a base level for the karst. 1 National University of.Athens, Dept. of Geography and Climatology; Panepistimioupolis, 15784 Athens, Greece.


64 0 0 0 C" A . ""--cJT . .s' .. CRt'TAN """ 6 q The northern and eastern part of Patnass forms an E-W striking anticline. Each one consists of smaller, NNW-SSE striking anticlines. T h e folds and the up faulting front movement follow the same direction. The faults strike NNW-SSE, E-W, and NW -SE, as shown on the relevant giaph (Fig. 2) which was made using the directions of 73 faults. At the south Parnass there are only a few dolines developed on the mixed thin bedded lime stone formations which belong to the transition zone of the Parnass-Giona Unit to the Olonos Pindos Unit, and of the Parnass-Gionas unit to the eastern Greece unit. Dolincs found on the mountain of Menalo (NW of Tripolis) are developed on neritic, pure, unbedded to heavy bedded limestone formations of Late Tria s sic and Middle-Late Eocene of the Tripolis subunit. The sedimentation of this unit was inter rupted during the Eocene period and followed by a fault tectonism, karstification and bauxitogenesis (RICHTER & MARIOLAKOS, 1972). On Menalo mountain there is a great symmetric anticline structure striking NNW SSE Minor fold s strike NW SE or NE-SW. Studying 143 . K. a . s ... .. .61. different faults, one may conclude that the major directions of the faults are N-S, NE-SW, NNWS SE, NW-SE (Fig. 3). Few dolines are developed on the mixed limestone of the Olonos-Pindos unit, and o n the dolomite and dolomitic limestone for mations of both units THE CLIMATE The climate at Parnass and Arcadia area during the winter time is a mild one {lowest temperature of less than 10 "C, humid, with high air humidity and low evaporation). Snowfalls are frequent and intense: the frost lasts almost all.the year at altitudes of more than 1800 m. In summer time, the eijmate is a desolate one (high air temperature, sun beaniing, evaponiUon), and what more, the thermic storms are quite frequent between June and August. One can observe three main paleoclimatic periods, on what concern the Pamass Mountain; i the first one was recognized on the third baux ite layer of the late Cretaceous,. characterized by warm and humid climate like the present tropical and subtropical climate.


Comparative observations in karst formations of Parnass and Arcadia . .. 65 ; .. .. .. . w E w E s s N .. .. w w E s ... M!s (h). ii. the second one is characterized by the tropical climate of the Middle-Late Miocene to Early Pliocene which lead to the genesis of the super ficial cone karst, as seen on the surface of Ka lyvia at 10001200m. iii. the third one is characterized by th e cold and dry climat e of the Wtirrn period. Two glacial trails may be observed at 2000-2500m. In Arcadia area one may observe only one paleo climatic period belonging to the Late Eocene charac terized by tropical climate. As an example th e area between Vytina and Kamentsa, next to Nymfasia, is ; . ; . : . .. referred, where fossilized cone karst is developed on limestone formations of the Olonos Pindos area covered by flysch and bauxite is cited on Tripolis limestones of the Eocene period, 500 km west of Vytina (KAROTSIERIS, 1978). I THE VEGETATION In the Pamass area there exists woody vegetation mainly of up to an altitude of 2000 m. In the Arcadia area woody vegetation (fir trees) ex pands to the point wher e a vegetation full of trees begins.


66 DO LINES Central Parnass karst is characterized by dolines of dissolution and collapse, the dimensions of which range from several meters to 1 km with a depth up to 100 m. Their drai.I)age system " uses swallets and is directed towards the river Viotikos.:.Kifissos. They are found on three corrosion levels: (a1) at 1()()(}...1200 m, (ai) at 1240-1340 m, (b) at 15001800 m and (c) at 2080--2280 m The (a2) level (located 200m above (a1), and dipping south ward) is considered to be a pru.:t of the a1 Taking into consideration the high altitude on which these forms are developed, one may de scribe them as "high karst" fomiations. As "high karst" are also designated the karst forinations de veloped at an altitude of more than.1000 m, ob served at the area of Kalyvia (SWEETING, 1961). The largest amount of dolines is developed on the first and second paleo-surfaces, which ha:ve the largest extension The main direction of their long axes is parallel to the main direction of the area structure which is NNW-SSE, NW-SE (Fig. 3). Quite often dolines turn to mivillas, with their main ax. es striking N-S. On the third paleo-surface there are fewer dolines yet of larger size, (more deep and more elongated), which is attributed to the corrosion and the frost phenomena occurring at such high altitudes. The karst formations observed NW of Tripolis are characterized by collapse dolines which are drained through ponors to Argolikos bay. The ponors found in the unit ofT. ripolis are mainly diffuse as compared to those found in the Pindos unit which are "gate" shaped. They are mainly found on four corrosion levels; (a) 880-1060 m, (b) 1140-1340 m, (c) 1400-1560 and (d) 16001720 m Levels (b), (c), and (d) are observed in the northern part' of the area with a SE dip. The differ ence occurring in dip values are due to neotec tonism. As shown in the graph (Fig 4) most of the dolines are developed within th_e lower levels (a) and (b), which are the most extended. Fewer dolines are developed on the less extended third layer while on the highest, fourth layer where the influences of snow and frost are significant, the development of. dolines is the most intense. The higher the altitude, the smaller the do lines. The direction of the elongated dollne axis is N-S, NNW -SSE, and NE-SW, parallel to the strike of the main area faults. One may conclude that karst formation of both areas is tectonically controlled. Concerning the first level, it may be charactenzed as karst mar ginal plain, formed by chemical dissoiution during K. Paptidpoulou the Early Pliocene period, under humid and tropi cal Climate conditioJ:ls, arid disse(;ting _.a paleo Tertiary karst partially covered by f;Iyscb (RIEDL. 1977). Apart from the young, tectonically controlled karst, one may observe a tect<;>riidllly : coJ:ltrolled corte karst of Late Cretaceous in dmtrai Parnass, where the third layer is devetopect (ARONIS, 1955). A cone karsi 'otLate Early Pliocene age was also at an altitUde of 1000-1200 rri. Despite the fact that the dissolu tiori do lines found between the cone hills give the impression of a young karst, they must be a ldnd of renewed superficial karst, like tlie one observed ori the Elikonas tnotuitain, SE of Parnass tain, at altitude (PAPADOPOULOU VRYNIOTI, 1990). A cone paleokarst fTom the late Eocene period is observed iri Arcadia area, which was uncovered after the removal of the overlying la:yeti puring the main tectoniC phase of the Miocene Bauxites were found in a tectonically controlled cockpit doline from limestone formations of the Eocene, 500 km west of Vytina As a general conclusion, bauxite horizons in both studied areas define a cone type paleokarst. In what concerns the Parnass Mountain area, paleo karst constitutes the first observed karst level of the Greek Mesozoic periO

ComparatiYe obseTYations in karst formations of Parnass and Arcadia 67 Numbl2r of forms in Parness 2 8 10 12 1240 ------------1320 1360 1400 1440 1480 1520 1560 1600 .... 1640 E ....., til 1680 'C ::J 1720 ... ... 1760 1800 1840 1880 1t20 ,,60 2000 2040 2080 tectonically controlled karst. Despite the fact that most of the dolines are found at lower altitudes, those observed at higher altitudes are of larger size. The observed cone paleokarst and the superficial one in central Parnass have developed during the Late Cretac eous and the Late Miocene Early Plio cene period, respectively. Karst formations of 1<4 NumbC2r of forms in Arkodia ..... E ....., Tripolis area were formed during Late Eocene and were either covered by flysch sediments or devel oped surfaces suitable for bauxitogenesis. In all above mentioned cases paleokarst is tectonically controlled. Consequently, in both areas paleokarst and young karst as well are controlled in each case by the; corresponding structure


68 REFERENCES ARONIS, G. (1955) Geographical distribution, geological placing and point of views of the Hellenic bauxi togenesis. Bull. Hell. G .Soc., ll, J, pp. 55-79. CELET, P. (1962) Contribution a !'etude geologique du Pamass Giona et d'une partie des regions meri dionales de Ia Grece continentale. These d'Etat (Lille), An. Geol. Pay. Hell., Xlll, pp. 1-358. K.ALPAKIS, G. (1979) Contribution to the sedimentary study of biomicrite stratified rocks, hard ground horizons of Parnass-GiontlS unit, at the Late Cretaceous-Paleogene limit. PhD Thesis, Uni versity of Athens. K.AROTSIERIS, Z. (1978) A bauxite outcrop of the Eocene period at Vytina's area in Tripolis. Bul. Hell. G. Soc., 13, 2, pp. 153-161. K.AROTSIERIS, Z. ( 1981) Geological investigation around Vytina 's area (central Peloponnisos ). PhD Thesis, University of Athens, 202 p. MONOPOLIS, D. (1971) Hydrometric research of the karst area situated on the Pamass mountain. IGSR., Hydrological and Hydrogeological Re searches, 4, Athens PAPADOPOULOU-VRYNIOTI, K. (1990) Geomor phological investigation of Kopais area (Viotia). PhD Thesis, University of Alhens. PAPASTAMATIOU, I. (1960) La geologie de la region montagneuse du Pamass-Giona-Oete. Bull. Soc. Geol. France, 1, 2, pp. 398-409. Papadpoulo_u_ PAPASTAMJ\TIOU, I. et al. (1962) Geological map of Amfiklia. Geographic Military Service (GMS), lost. of Geol. & Subsurface Research (IGSR), Athens. PAPASTAMATIOU, I. et al. 1960. Geological map of Amjissa. Geographic Military Service (GMS), lost. of Geol. & Subsurface Research (IGSR), Athens. PHILIPPSON, A. (1892) Der Peloponnes. Friedlander Ver., Berlin, 642 p. PHILIPPSON, A. ) Beitrage zur Morphologie Griechenlands. Geogr. Abh., 343, pp. 55-60. RENZ, C. (1930) Geologische Reisen in Griechischen Pindosgebirge. Eclogae Geol. Hel., 23, pp. 301-377. RENZ, C. (1940) Die Tektonik der Griechischen Gebir ge. Mem. de l'Acad. d'Athenes, 8, pp. 1-171. RICHTER, D. & MARIOLAKOS, I. (1972) Palaomor phologie und Eozane Verkastung der Gavrovo Tripolis Zone aus der Peloponnes (Griechen land). Hel. Geol. Soc., IX, 2. RIEDL, H. (1977) Die Formenelemente in Bereiche des Arkadischen Zentralzuges und des Westarkadi schen Gebirger auf der Peloponnes (Griechen land). Ann. Geol. Pay. Hell., 29, pp. 209-225. RIEDL, H. (1993) Die Karstreliefgenerationen Grie chenlands und das Problem anthropogeographi scher Steuerung. Die Hohle, 42, pp. 57-70. SWEETING, M. (1961) Report on the geomorphology of the Pamass-Giona Region on the Greece. JGSR Geol.Reconn., 41, p. 24.


Theoretical and Applied Karstology, vol. 9/1996, pp. 69-76 A Study of the Subsurface Karst of the Kopais (Central Greece] Kyriaki PAPADOPOULOU & Theodore GOURNELLOS1 ABSTRACT In this article we attempt to quantitative anatysis of the swallets ("katavothres") of. the former lake of Kopais (Central Greece). These are situated at different altitudes around the former lake. 'The principal outcrops are various limestones, flysch and schists We investigate the rehitionships between various variables such as lithology, altitude, horizontal development and geographical position of these karst forms We proceed to a statistical analysis of the above variables and we study the spatial distributions of the swallets The final results are indicative of the development of these karstic forms and their dependence on the various parameters. ; Key words: karst, swallets, Kopais, Greece. Etude des formes karstiques souterraines de Kopais (Grece centrale) RESUME L 'article propose une analyse quantitative des ponors ( "katavothres") de /'ancien lac de Kopais (Grece cen trale). Ceux La sont situes a des altitudes differentes autour de l'ancien lac Les pdncipaux afjleurements in eluent des d!!Jerents calcaires, du flysch et des schistes. On analyse La relation entre les differents variables comme L a lithologie, ['altitude, le developpement horizontal et La position geographique de ces formes karsti ques On ejfectue une analyse statistique de ces variables et on etudie La distribution spatiale des ponors Les re sultats jinaux indiquent le developpement de ce.f formes karstiques et leur dependance des differents parametres. Mots cles: karst, ponor, Kopais, Grece. INTRODUCTION The 220 km2 Kopais region is situated in Central Greece, 110 km NW of Athens at an average alti tude of 92-100 m (Fig. 1 ) It basically consists of limestones and flysch. Scientists of different spe cialties, like : STRABON (50 BC), PAUSANIAS (2"d century AD), PHILIPPSON, (1894; 1930), KRAUS (1892 ), MAUL, (1921), RENZ (1913), MAHERAS (1972), MALERDOJS, (1981), KNAUST, (1986), BERTOLANNI & ROSSI, (1983), ALLEN, (1986), PAPADOPOULOU (1987; 1990). During the historic times, a lake was formed peri odically, that was drained in the sub-surface through swallets. During the 14'h century BC it was originally drained by the Minies and as the result the Orchomenians got prosperous. After the destruction of what the Minies had accomplished, the region regained its old lake marshy form and was finally drained in 1931. The dr1inage was acomplished by building a peripheral and a central channel which collect the large flow to reach, finally, Iliki lake though the Karditsa canal ap.d eventually into the sea. 1 University of Athens, Dept. ofGeography & Climatology, Panepistimioupolis 15784 Ana Ilisia, Athens, Greece.


70 ; 0 \ 1-.... 'Y ..,. J' I' .. 0 0 0 (' .. 1-t:? i I' "(] Q C R ETAN GEOLOGICAL SETTING AND KARST EVOLUTION The margins of Kopais a re made up of limestones of Mesozoic age, and of flysch of Upper JurassiC Lower Cretaceous and Paleocene age (Fig. 2). The preNeogene basement of this region belongs mostly to the geological unity of the Eastern Greece. Only part Of it, to the west, belongs to the Pamassos-Giqnaunit and to the Viotiki uni t in the sw. The Neogene formations that outcrop in a wide area in the NW and the NE are Pliocene d eposits consisting of alternate layers of conglomerates, marls, marly limestones and lignite intercahttions. These are the residual deposits of t h e Pliocene lakes or of a wide lake which was formed in this particular area. during the Pliocene. These Pliocene deposits have been eroded during the Quaternary aild they either have been carrie d to lower positions, as in the case of Kopais, or they were redeposited in the region of Thives. K Papadopoulou & T, Goumellos a SEA The creati o n of Kopais was due to fault tectonism which started in the Early Pliocene and createdthe big Phokic-Niotic tectoniC basin. The P liocene deposits that filled these depressions soon exceeded 800 m in thiclaiess. 1'he successive subsidence took place during Upper Pliocene-Lower Pleistocene period resulted in separating off the Kopais segment from the entire region The surrounding inflow was collected in this area and thus started the formation of an independent shallow lake. Under a clirriate th at was warmer and more huinid than today, the .of this lake; started to undergo karstification. A polja-type soil was created and the lake started to b e drained in the subsurface through . a number of swallets, mainly towards the North Evoikos g\Jlf. THE SVBSURFACE KARST FORMs The swallets of the Kopais regionhave developed \Vithin theN, NE, E and SW limestone margins of the foliner lake at different altitudes.


GEOLOGICAL MAP OF KOPAIS by K Papadopoulou (1990) D r-::::J I!'!'TI l'.zJ fgJ rml .s.V LEGEND Alluvial dClpOsits Old and naw talus conas and alluvial conas Pliocana dClPOSits Flysch (Lowar MiocanaPolaocana l LimClstonas !Uppar Jurrasic-Uppor Trlosic) Schist and charts formation (Jurrasic -Uppor Crotacaous) Dolomitas !Trios) Poridotitas of schist and charts formation. Upthrusts Ovarthrusts Inn I Faults Faults probablas Zonas of high dansity of Kotavothras SCALE o 300m ... \ I ......... : ... .....,_. ... \ .. -.......... ......__/ -' .. ,;l' ;''! ... .'P" N (_..._:) C'"-........... r' .: ") --... Fig. 2 "' "' "' ;:: $:1 "' t" i-! .sa. So "' >:: $:1 t; ...... .....


72. Altitud a5 of Kotovothras 400 e 3001--23 Longitudf25 fig. 4. swailets 1tin8ItU.4es vefsil$ .. des erifonction' E ()I '0 :J -... 400 22 24 28 30 32 34 38 Gaogrophical Latituda Apart Jrom a few "Schluckatavothren" .the major : icy o( swallets are "Torkatavothren" . .They have formed along .different lithologic arid tectonic dis'--toi:itinuities. Most of them are parallel to the general direction 'of the Jayers arid occur }nregions of steep erit, \here they bave become coastaJ sub horizontal caves; Certain swalletsin the NE occasionally furictiori as springs : e.g. "The Cave ofihe Spring"; Few lets are still active : ' ... : The drainage of the lake; through a nmltitucte pf swallets', took place : :practically tow aids .'fou'r' di rections: In. the North, the waters t1owed towards Atahintt in the NE mainly to the gulf of Larymna, but also to Skoponeria; where submarine springs .. stilfemergeand in tqe SE towards lake Iliki. The swallets of the eastern margin ate being de strayed : as a result of the recenl works for road building. QUANTITATIVE ANALYSIS . .. An aiuilysis of all the ve and quantitative !'actors of the Kopais swallets has been made, based on our field observations 'ruici on data. of BERTOLANNI& ROSSI (1983) (Table The histogram (Fig. 3) of the altitudes shows the medium height of the lake. The di;tribution of the subsurface karst features was subsequently As shown in Fig. 2, thelocation of the swallets is noL randomly distribuLed around the lake. Their occurrences are concentrated in several zones : This fact proves thaL their developme nt was not uniform aiong the mer lake shoreline, but that they were subject Lo a preferential development in certain weak zones. The diagrams of the latiLude and longitude versus elevation (Fig. 4, 5) show thaL the karst forms arc rriorc or less concentrated in a narrow zone as far as longitude is concerned. On the contrary. their location is widely in terms of latitude. It was the horizontal development of the karsl forms that was subsequently examined. Their de velopment is controlled by the degree of Lheir kar stification. The horizontal developmenl : of. Lhe swallets as a function of their altitude is shown in Fig. 6. We can therefore distinguish three zones: .. the shoreiine-tectonic zone; . the shoreline zone; the "katavolhres" zone.


The subsurface karst forms of the Kopais Table 1. No. Alt (m) Longitude Latitude Length Orient. 1 110 2 3 105 115 4 140 5 140 6 130 7 120 8 110 9 150 23 39 38 57 44 23 6 38 14 180 23 29 38 14 157 23 52 38 11 25 23 46 38 13 29 23 38 31 23 9 38 15 30 23 58 38 15 14 23 7 38 15 26 23 2 38 14 16 23 54 38 15 17 23 47 38 10 20 23 45 38 7 12 70 177 80 1 40 10 170 10 110 170 170 1 90 10 115 11 115 12 105 13 103 14 100 15 130 16 115 17 115 18 200 19 274 20 265 21 490 22 112 1.370 100 23 8 38 55 16 23 34 38 39 19 23 27 38 36 12 23 18 23 33 23 4 23 2 23 26 38 18 133 38 22 212 38 35 35 47 21 23 115 23 27 38 32 19 24 110 23 26 38 24 16 25 110 23 28 38 18 10 26 110 23 38 25 27 108 23 35 38 7 30 28 107 23 28 38 4 24 29 107 30 125 31 108 32 115 33 120 34 115 23 25 23 23 23 24 23 27 23 56 23 8 38 59 38 41 38 5 38 0 38 45 38 46 15 20 18 13 11 22 65 140 50 170 155 120 1 55 50 90 65 13 60 85 70 80 150 97 1 73 No. Alt (m) Longitude Latitude Length Orient. 35 115 36 110 37 110 8 31 38 110 23 56 38 20 19 39 110 23 54 38 23 62 40 115 23 13 38 57 96 41 105 42 105 43 98 44 195 45 178 46 125 47 110 48 105 49 110 so 120 51 110 52 95 53. 96 54 115 55 112 56 98 23 8 38 57 477 23 33 38 37 591 23 0 38 5 81 23 25 38 54 57 23 25 38 52 20 23 8 38 34 17 23 9 38 18 145 23 38 40 23 19 23 27 23 29 23 37 23 14 23 24 23 6 23 4 38 59 14 38 55 36 38 43 14 38 15 13 38 38 11 38 11 24 38 36 31 38 55 10 57 98 23 37 38 38 20 58 115 23 15 38 41 16 59 110 23 0 38 37 14 60 100 23 57 38 31 16 61 110 23 1 38 29 16 62 140 23 39 38 48 52 63 100 64 115 65 105 66 125 67 118 68 120 23 24 38 45 20 23 34 38 34 380 23 56 38 15 49 23 31 38 53 45 13 32 90 40 55 47 165 120 80 130 8 35 140 168 125 78 95 50 175 45 1 80 145 144 175 40 90 25 25 130 15 65 90 95


74 ..., E -10 ' ' ' . ' 0 ' \ \ 0 '. \ \ B \ 50 Horizontal 0 0 F n o Kotavothro 0 100 di2VI2Iopm12nt Orillntation Oricmtation 1(. PapadopoU(Cill & T Goiltf&ellos According to our observations it appears that these thr(1e zories extend.. within both : most important lithological types of limestones. Besides the hori zontal extenflength of the swallets, it is also iriter esting to study their orientation ; Two direction histograms haye been constructed In Figure 7 we can see the main directions of the swallets while in Figure 8 their both main and secondary !lirections, as well as their length have been considered. We notice that the main directions are N...:S, NE-SW, E-W and NW-SE. Some of them correspond to the main directions of the faults and joints of the region, as we have found out during our investigation. In Figure 9 we can see the relationship between the altitude of the swallets and their directions . It is obvious that there is a uniforin distribution of the directions of these karst forms at all altitudes around the former lake. In Figure 10 we can see the relationship between die shape of the subsurface karst forms

The karst forms of the Kopais 4001-JOOf-,': e -' 200 1)1 '0 ... ... 100 .. I ,I I I .: I a- .. : I ! ;{ . 0 I I I 0 30 t.o 90 120 150 180" Oriczntation 150 120 c: gor-.2 ... 601-c: 1)1 0 JOI-. I j I I 1 Coa$tal TClctonic Katavothra to. .. -: .. La rellltWn eNr;e -karstiques ttt leur:s _-. -'.;.: : ... c: Cl E 0 12 g g 9 0 .. c: 2 3 ;;: 0 I 75 Fig. 12. relatiomltip l;Jetween the geographical coordinaus and tlle'borizontal development. . . . ui rell.ztion entre les coordonnees geographiques: et le deve/Oppeine'ntlioriio"tal. :. > _: : ... c 2 ... a ... c: '-0 Gczogr: Longituaqs 1)1 50 l:l ::;1 ..-AO en c: 30 9 &.: 20 en 0 10 38'


76 3. "Katavothres" which practically drain the area. Their horizontal extension is increasing from zone 1 to zone 3. The distribution of the subsurface karst forms in the area shows that their occurrence was not acci dental, being related to certain zones of weakness within the limestone formations. That is why we observe concentration of forms in some areas. The maximum frequency of subsurface karst forms occurs at altitudes ranging between 105 and 115 m. Similarly to previous studies, we believe that this was the level that the lake reached during BIBLIOGRAPHY ALLEN, H., D. (1986) Late Quaternary of the Kopais basin, Greece. Sedimentary and Environmental History. Disert. Citron College, 282 p. BERTOLANNI, M. & ROSSI, A. (1983) La zona Spe leologica dellago Copaide (Beozia, Grecia Cen trale). Le Grotte d'ltalia, 4, XI, pp. 205-248. KNAUS, J. (1986) Die Melioration des Kopais Beckens durch die Minier im 2. Jt. V. Chr. Kopais 2-Wasserbau und Siedlungsbedingungen in Alter tum. Antike Welt, 11, 2, pp. 15-30. KRAUS, F. (1892) Sump/ und Seebildungen in Grie chenland mit hesondere Beriick.fichtigung der Karst Erscheinungen und besonders der Kata vothrenseen. pp. 5-49. MAHERAS, A (1972) Study of Hydrological balance, with hydrogeological survey of Biotikos Kifisos valley and fields of Kopaida and Thiba. Ministry of Public Constructions, Section 2, Athens. MALERDOS, G. (1981) Contribution a /'etude des reactions hydrologiques saisonnieres dans la plaine de Kopais, par nivellements de precision 1(, dl: T. the period of its largest expansion, where it re mained, with slight fluctuations, for a long time. The distribution of the karst features directions seems to be controlled by the structural pattern of the area, as far as the faults an d the joints are con cerned. The distribution is uniform and all kinds of subsurface karst forms are expanded equally in all directions. It is therefore obvious that a significant part for their formation is due to the tectonics and the in tense karstification realized achieved as a result of suitable conditions. et teledetection Beotie, Grece continentale. These Univ. Bordeaux I., 100 p. MAUL, 0. (1921) Beitrlige zur Morphologic des Pe loponesse und des sudlichen Mittelgriechisches Lands. A Pencks geog. Mitteil 3, pp. 179-340. PAPADOPOULOU, K. (1987) The blind valley of the Black river (Biotia). Greek Geographical Society, V-G, pp. 20-24, Athens. PAPADOPOULOU, K. (1990) Geomorphological study of Kopaida (Biotia). Phd Thesis, Athens Univer sity, 145 p. PAUSANIAS, (lind Century AD) Tour of Greece. Ath ens, reprinted 1967. PHILIPPSON, A. (1894) Der Kopaissee in Griechenland und seine Umgebung. Z S. G. Erdk., 29, pp. 1-90. PHILIPPSON, A. (1930) Beitrlige zur Morphologic Griechenlands. Geogr. Ahh., 3, H3, pp 55-60. RENZ, V (1913) Der geologische Aufbau der Gebirge und des Kopais Beckens. Z D.G.G., MB, 607-619. STRA VON (50 BC), Geographicals, Papiros publ., Athens, reprinted pp. 1169-1913.


Some Significant Caves at the Western-Rim of the Miroc Karst [Yugoslavia] Milena ZLOKOLICA 1 Mihajlo MANDIC1 & Vladimir LJUBOJEVIC2 ABSTRACT The western rim of the karst oasis is composed of Lower Cretaceous sandstone, marl, mudstone and limestone (K1 2), and Upper Jurassic limestone (13 3), and it abounds with rivers which sink into caves of con siderable dimensions. The deepest of them is the pothole Jama u (-272 m). while the longest is the cave Buronov Ponor (1950 m). The caves abound with diverse forms of passages-ranging from almost un altered fissures to well-rounded phreatic tubes. The phases of fluvial sedimentation, of rock collapses, of for mation of travertine coating, and of hydrologic reactivation and emptying of the cavities, can be distinguished as well. Key words: karst, cave, speleology, Miroc, Serbia, Yugoslavia. Quelques cavites importantes sur Ia bordure ouest du karst de Mirot (Yougoslavie) RESUME lA bordure ouest de /'oasis karstique de Miroi!, constituee de gres, mames, argiles et calcaires du Cretace in jerieur et calcaires du Jurassique superieur est riche en penes des rivieres temporaires qui disparaissent dons des gouffres et des grottes de dimensions imponantes. Le plus profond, est le gouffre Jama u wni!tu ( -272m) tant que Ia plus longue est Ia grotte Buronov Ponor ( 1950 m). Les cavites sont riches en differentes Jonnes de galeries, allant des fiSsures /egerement modifiees jusqu'a des formes bien arrondies. On peut remarquer aussi les phases de sedimentation des dept1ts alluviaux, d'ecoulement, des depots de travenin et de reactivation et vidonge hydrologique des cavites Mots cles: karst, cavite, spe/eologie, Miroi!, Serbie, Yougoslavie INTRODUCTION The Miroc karst oasis is a phenomenon of karst morphology the speleological research of which has been unjustifiably neglected Although nume rous researchers, among whom C VUIC (1921), MILIC (1965). ZEREMSKI (1988) and others performed diverse investigations, their interest was focused mainly on the surface morphology (the evolution of the planes and the Danube terrac es) and on the geological situation. The only author who gave some amount of data concerning the characteristics of the cavities was J. Petrovic in the monograph "Jame i peCine Srbije" (''Potholes and caves of Serbia"). In 1990, the Student Speleologic and Alpinistic Club (Akademski Speleolosko-Alpinisticki Klub -ASAK) from Belgrade began a systematic re search of this terrain, with a special focus on the areas of the western and eastern contact of the


78 0 (5>. 0 0 0 0 0 "' a" J) 13 C) c;:; c> 0 M. Zlokoli(J(I : et_(ll.!f& 1':. Fig. 1. Map of the rim of karst. a) GeOIOgiC!Bi riu!p: cheJ1y,'i(lne&tone; K,a niarl; argillaceous Ume8tone; limestone; K?.4-Umeston(!,: tnari, Sandstone; "tW.. ffiles .... goologlcallimlts; thick Hoes-faults (after BOGDANOVIC; 197$)... .. . : . ;-,::<. ... i -. Carte de lD bordure 'ouestdu karst de Mirof." a) Ci:zrte geologiqile:il..; calcaii'e a .. gres, ca.l-' caires argileux, calcaires; K/'4 -ca.lcaires, marnes, { drgikux, ires; fines -limites ,' gnesepaisses --faiUes' (d'apres BOGDtVVlr:, ''; __ ... .. ., -::. ,. : ':


Significant caves at the western rim of the .. . ' . .. ... karst oasis with the s1,nrounding, non-karst terrain. more than 30 cavities have been ex plored. The concentration of cavitiis of notable dimensions on the western rim (between Kopana Glavica and Bele Vode) is sigmficant. GEOMORPHOLOGICAL AND GEOLOGICAL CHARACTERISTICS OFTHE. TERRAIN . ... The terrain whose characteristics are presented, includes a narrow band of a karstified plateau lying by the western contact of the Miroc karst oasis, between Bele Vode (Dobre Vode) on the north, and the village of Kopana Glavica to the south. The terrain can be hypsometrically divided into two areas: one along the Danube shore, with an aJtltude of roughly 70 m a.s.l., and one which inCludes the karst plateau, with altitudes raJ1ging from 300 m to 512 m a. s. I. (at Kalailfirov Cukar) Large viuia tions of altitude over a short distance are due to the deep carving of the gorge of Danube, as well as to the existence of numerous short streams flowing from the non-karst terrain toward the contact aiea, ending in blind valleys. Among them are the streams of Rakin Ponor, Suvi Ponor, Ibrin Pol1or and Buronov Ponor, which sink in the caves at the contact with the limestone. The limestone is inten sively karstified and known in the literature as the dry valley ofBele Vode (MILIC, 1965). The geological structure of the terrain is relatively simple. Non-karst rocks, underlying the carbonate sediments, do not outcrop in the western contact area. The youngest rocks are Cretaceous conglom erates, sandstone, mudstones, and arenaceous lime stone (K1 2 ) which lie above Barremian-Aptian limestones, sandstones and mudstones (K1 2 1). The Upper Jurassic sediments have the widest extent on the Miroc karst oasis. They are represe:nted by lime stone with cherts, massive limestone and, to a smaller extent, by marly limestones (J/). Jurassic sandstones, mudstones and arenaceous limestones Cave Entrance altitude Bottom altitude Buronov Ponor 272 85 Suvi Ponor 370 237 Ibrin Ponor 350 232 Jama u Langtu 440 168 Rakin Ponor 380 124 -....... found in a narrow area in the center of Miroc on Strbac, in wlUr the magmatites of Velild form of the Miroc anticline. On the availaJ'-le geological maps, the tectonic setting of the :western-rim is nOt. distmct, except for sporadic of NE-sW :onentatioll. Having in mind the anticline structure Miroc, Jhe existence of numerous .tension roptures . of &-w. orientation (favorable for water Circulation) as well as the existence of ruptures of N-S orientation, .parallel ing the axis is expected. A system of well developed ruptures and ruptute pianes of NE-SW orientation been detect(:d in tlie cavities (Fig. 1). The li(Jlological characteristics of the terrain, in conjunction with the tectonic structUre, .induced a very deep and intensive development of karstification, mainly on the edges of the Mitoc karst oasis. No significant cavities liave been found in the central part of the Miroc karst oasis. The analysis of the characteristics of the cavities has not shown the expected correlation with the altitudes and the development of the terraces of the Danube, rior the confirmation of successive devel opment of the cave passages and their concordance with the local erosion base level the kaistified valleys of Bele Vode and This is shown best in the Rakin Ponor pothole, with the lowest explored passage at 124 m above sea ievel. Examining the surface only, the logic continuation for the stream of Rakin Ponor wOuld be the valley of Cvetanovac, with its swallet and the springs of Sokolovica (230 m a.s.l.) and Blederija (188 m a.s.l.). On the hand, the horizontal passages on the altitudes of 220-230 m a.s.l. in Buronov Ponor, Suvi Ponor and lbrin Ponor, do not have their equivalent in the formation of the terraces of the Danube,this being a zone of intensive carving of the Danube Gorge. I Total length Total depth Length/Depth J ratio 1950 187 10.42 930 133 6 .99 228 118 1.93 710 272 2 .61 634 256 2 .47


80 -256 .Q end i ng II siphon ? I' ? I' 0 / ... .. ;;.. . :. RAKIN PONOR PROJECTED ELEVATION 0180 0' 100


Significant caves at the western rim(}fthe Mirl!ck4rst : . . . . All the presented cavjties are of casc:ading type, the deepesfpits having so .in (in Rakin Portor :and Ibrin Ponoi'). Looking .from north to the length of the caves is decreasing, while their depth is increasing, having the length/depth ratio 10: l in Biuono.v Ponor, artd in Rakin Pon6r. The prevailing type of passages is .the c:ilriyon. (nft) shaped, except for the Buronov Poncir, where the existenc:e of rounded passages with quantities of accumulated material is typical . The clay sediments in the caves originate from fissures . a very enlarged ruptUre, which ends witliboulders :toffiiing a terrace. . The southeastern part of the : tjre(J.t Crevice inter cepts a system of fractures \Vhh a NESW orientation; called the Frssure Passage. This passage, 6.5 io 1 m wide and up to 20 m .high, ends with im unperietiable fissure. Along the opposite part of the Crevice, a passage descends steeply towards west arid inter cepts the lower level of the cave. Thi!i part consisL<; of an enlargedfissure,) in which ends in sand, gravel and rock deposits. A lateral passage, named the Jewel Passage, opens towards east. Completely covered with travertine coating flow stone and curtains, it is the only part in the cave where chemical sediments are found. It is ascend ing towards east, until it meets the fissure which, probably, originated the formation of the Fissure Passage. Afterthat, the passage turns towards SW, ascending .further and ending in a travertine pool, which is the highest part of this passage and, pos sibly, the place of water in11ow. In the next section of the Entrance Passage, the ascending Guano Passage is encountered; it repre sents the final part of the Entrance Passage. Its dimensions remain unaltered, while it ends in an unpenetrable fissure. The floor is covered with guano, which is encountered only in this part of the cave, along with sections of trav erti ne coating on the walls The Entrance Passage ends in an expanded joint between beds, dipping to the west, up to 3 m high, extremely eroded with 1 m deep erosion gu tt e r s both on the floor and the ceiling This part of th e cave differ s morpholo g ically from the previous part. Its development was dominated hy a perpendicular rupture with E-W orientation. The bottom of this canyon-shaped passag e is rarely visible, c:overed with boulders The width varies betwe e n 1.5 and 3 5 m, while th e height goes from 2.5 to 10 m. The walls are eroded hy waterflow, and travertin e coatings on th em arc rare The only lateral passag e in this part of the cave is Daca 's Passage, which is a narrow tube (0 .5 to 1 m in diameter) ascending in th e b eg in ning, th e n steeply descending to meet again th e Boulder Passage, just b efore th e b eginning of th e Gre at Pit. T h e ending part of the Boulder passage has different characteristics. There are n o h o uld e r


82 masking the bottom of the passage; this one is descending through cascades, and its dimensions rise up to 10m in width and over 12m in height. The end of this passage lies on a terrace formed on wedged boulders. The third, final part of the cave, is the one with the simplest morphology. It consists of the Great Pit, whose height exceeds 70 m, and the diameter is over 5 m. It ends in an enlarged rupture, descend ing towards NE. The passage has a complex cross section, with the width of about 5 m, and the height varying from 1 m to more than 20m. Ex cept for the bottom of the pit, no detritic sediments are encountered, while chemical deposits appear sporadically, mainly on cascades and at the inter sections with lateral fissures. The erosion was the dominating factor, leading to the formation of well polished and rounded walls and ceiling, and an gutter on the floor. The width of the gutter 1s 0.3 to 1 m, and its depth varies between 0.3 to 1.5 m. The passage reaches a sump of 3 min di ameter; through its clear water one may see that further on it preserves the same dimensions and inclination. The hydrologic characteristics of Rakin Ponor, as suggested by the morphology analysis are rela tively clear. The entrant part, before the Great Crevice, is hydrologically inactive, except after heavy rains or snow melting, when dripping wafrom the ceiling. Above the pit reach mg mto the Great Crevice, traces of significant and water inflow are detected. During certam penods, a waterfall is formed down the pit. The northern part of the Great crevice is, hydro logically speaking, the most active part of the cave. Fragments of wood, leaves and other alloch t?nous debris are clearly visible in the upper sec tions of the passage, while the great amount of gravel, sand and pebble confirms that assumption. It is clearly visible that a water inflow used to exist at the end of Jewel Passage. During repeated ex plorations, no traces of current waterflow were found, leading to the conclusion that the Jewel passage is hydrologically fossil. The same can be stated for the Guano Passage. The rest of the main passages, all the way to the final sump, gives clear indications of existence of a periodic flow with considerable amount of water and great energy. It is confirmed by the existence of deep erosion gutters at the entrance of Boulder Passage, and the total absence of smaller boulder in it. All the smaller fragments, carried in from the Entrance Passage, have been removed to the end ing part of the cave. The absence of travertine flowstone, and the polished and rounded walls also M. ZlokoUca et aL confirm the existence of a voluminous and strong flow. The end of the Boulder Passage, in the wid ened passage, also contains erosion gutters, scal lops, and no detritic sediments except for the great amount of leaves and wood in the high sections of the passage. The Great Pit, whose walls are eroded by a waterfall, ends in deep and enlarged fissures, enlarged by a great quantity of water falling from above. In the parts of this passage without a direct hydrological function, accumulations of pebble, sand and gravel mixed with leaves occur, while the active parts are lacking detritic accumulations. The ending part of the passage, just before the sump, indicates the flow of great quantities of water, which almost completely fill the passage, despite its significant dimensions. The sides are well pol ished and rounded, without sharp edges, and a gutter occurs both on the ceiling and on the floor. During repeated explorations, the gutter on the floor conducted a stream of at least 2 Usee. The ending sump displays large variations in water level, which oscillated 6 m between two explora tions. It must be stated that the Derdap lake accu mulation probably had a certain influence on the water level oscillations Based on the presented facts, it can be concluded that the Rakin ponor stream had the initial influ ence on the genesis of the cave. As the process of karstification was developing, the waters of the stream descended to lower levels, slowing down the development of the upper parts of the cave. The passages ahead of the pit entering the Great Crevice represent a part of a local collector, which is responsible for the lack of detrital sediments and the domination of erosive work of the water. The accumulation of debris at the bottom of the Great Crevice, and the shape of the passage indicate that waters circulating in this part do not have sig mficant energy. The situation in the Boulder Pas sage is completely different: where mechanical sediments are absent, and the erosive factor is dominating, even the travertine deposits are eroded by waterflow. analysis of the passages' morphology. par ticularly of their cross-sections, does not indicate any cyclicity and alternations in the flow regime. Except the passage leading to the final sump, there are no indicating the flow under pressure, but the existence of a free-level flow. Indices of filling of passages with sediments and later reactivation have not been found yet. It can that this is a classic type of dra1mng the surrounding terrain, and its most Important property is the fact that it is directly connected with the aquifer level.


Significant caves at the western rim ofthe Mirot JAMA U LANISTU The Jama u Lanistu lies in the central part of the western rim of the Miroc karst oasis, not far from the Rakin Ponor cave. It is located in the locality of Laniste, on a very karstified plain. Being 272 m deep, it is the second deepest known cave in Serbia. The total length of the cave is 710 m (Fig. 3). The cave has two entrances, located in a sinkhole at the end of a short temporary stream. The top entrance is a short pit. The cascading, entrant passage is leading from the bottom entrance. The passage width in the entni.nce part is between 1.5 and 4 m, while its height ranges from 6 to 18 m. Over a series of small boulder slopes, an entrance to the Main Passage is accessed. It is very dis tinctly canyon-shaped, with the width varying from 0.5 to 4 m, and the height between 1 and 30m, exceeding 20m in the most zones The ori entation of the passage is varying several times, starting from N in the entrance part, successively turning to W, SE, and NW and finally stabilizing towards NE. The cave is composed of two pas sages: the main Meandering Passage and the lat eral A./fluent Passage. 83 Along its entire length, the passage is gently me andering, while the flooi: and the walls are eroded and polished by the watet flow with pecasional scallops. The only sigriificant enlargement of the passage exists at the place ofifs intersection with a system of fissures, where a chimney of undeter mined height occurs, as well as several terraces at different heights. This part of the passage abounds with collapsed rocks, varying in size from 0.1 m up to more than 1 m. Atthe end of the main Meandering Passage, the only lateral passage (the Affluent Passage) is encountered Its dimensions are slightly .bigger than those of the Meandering Passage, ranging from 2 5 to 3 m in width, with the height over 20 m. At the intersection of the passages, the elliptical Final Chamberis formed, 20 m long and 15 m wide, with a height exceeding 30m. Like the rest of the Meandering Passage, the Final Chamber represents also an enlarged fissure, with parallel walls, inclined towards SE The sides and the floor of the hall are completely covered with clay coating, which was brought in through the Affluent passage, whose floor and sides are cov ered with thick layers of wet clay ....... :TENDEOELVATION l__ ------._ N ... .,,I>. .... . \ \ -272_ J


84 The access to the upper portions of the Affluent Passage is not possible. In the floor of the Final Chamber, a river bed of a temporary stream is carved (1 m wide, and 0.5 m deep). Limestone pebble, along with shale and sandstone fragments, is encountered on the floor and the river bed. The exact place where the stream sinks was not located, because the water infiltrates through boulders and clay and pebble deposits. Analyzing the plane and the profile of the cave, it can be presumed with great probability that the tectonic setting had a significant influence on the orientation and the morphometry of the passages. Two fissure systems can be noticed, with NW-SE and NE-SW orientations The dimensions of the passages lead to the conclusion that the initial fissures allowed rapid carving of the passage and lowering of its bottom, which led to the fonnation of extremely high and narrow passages The de velopment of short sections of the passage along fissures belonging to different fissure systems, led to frequent changes in passage orientation, as well as the appearance of wider portions of passages, which form the shapes of meandering passages. The water course is encountered in several loca tions in the cave. The waters of the temporary stream sinking into the cave disappear after some 30 m, reappearing later and flowing down to the cave's end The Affluent Passage is the only place of concentrated water inflow. It should be men tioned that, although 272 m deep, no river flow was registered in the cave, which leads to the con clusion that the cave is not a collector of an exten sive area, but rather has only local significance. The available data lead to the conclusion that the cave had a relatively simple genesis, in which the tectonics had the dominant influence both in early modeling and in the present development of the cave. The structural characteristics of the cave, as well as its hypsometric location are of secondary importance. The hydrological characteristics of the cave, and most of all the periodical appearance of a stream with turbulent flow, led to reduced deposi tion of detritic sediments in the upper parts of the cave and alternatively considerable deposition in the lower parts. The abundance of clay sediments in the Affluent Passage is probably caused by the in flow of mat erial through narrow fissures and th e existence of soil layers over the surface limestones. SUVI PONOR Suvi Ponor is locat ed 2 km NE from Kopana Glavica, at the end of a blind valley occupied by a M. Zlokolica et aL periodic stream coming from the south-west. It flows through impervious deposits, and sinks at their contact with limestone rocks. The contact follows the eastern side of the valley, and a small periodical swallet occurs 0.5 km upstream from the cave entrance. The total length of the cave is 930 m, with the depth of 133 m (Fig. 4). The entrance of the cave is 15m high, lying at the bottom of a 30-m high cliff. The cave passages are oriented along two systems of interchanging rup tures, with the orientations of 60 and .150. The passages with the orientation of 60 are narrow and high, except at the intersection areas. The fault planes in them are clearly visible in the deeper parts of the cave, where a great amount of fallen rock is encountered too. At the places of intersec tions, the passages are wider and form halls, espe cially in the entrance part. In the first 200 m of the cave, the Main Passage is relatively spacious (the width goes up to 5 m, the height up to 15 m, and the diameter of the cham bers of m). The floor is covered with lime stone debris and fallen rocks, as well as pebble carried in by the stream. Few pools filled with water, with the diameter varying from 3 to 5 m, and up to 2 m deep, are encountered as well. After the second chamber, a 17-m deep pit is en countered, and most of the debris and pebble car ried by the stream are deposited at its bottom. The passage leading further is almost closed by traver tine deposits, and allochtonous debris. At a place of fracture intersection', a small stream (0.5 l/sec in August 1990) is flowing from the SW. Further on, the Main Passage descends through several slopes, changing the direction from NE, to SE and SW, and back to SE. The Main Passage ends with three successive pits, covered with travertine deposits, with a total depth of 35 m, which intercept a spa cious passage ("the Wide Passage") occupied by a stream flowing from SW to NE. Downstream from the intersection point, the Wide Passage is up to 10 m large. Its floor is covered with a thick layer of clay, as well as small pebble, huge fallen rocks and travertine flowstone, both on the sides and on the center of the passage. The stream (approx. 1 l/sec in August 1990 and July 1991) p erio dically sinks while flowing through the deposits, forming several accumulations. The height the passage' is exceeding 15 m at the intersection point, then constantly decreases while going downstream. 120m further, a muddy sump stops the progress, at the depth of 133m. Upstream from the intersection point, a 15m high clay -covere d flowstone is encountered on the SE


Significant CDVf!S at the western rim of the Mirockarst .. .. .. ... : ' . .. . .. 8 5 EXTENDED ELEVATION '-: --......."'...t....... -/'-'\) . ;s't : ...... . --... . -133 N /sJpl>on -133 (237m a.s.l) I __ ,_.w._._.;. __ n_ __ o_,.r_, .-_ ..-.:..:_ ... ._ __ : __ __ _._._._:: __ . -.,'.-_. ., . .-:,.'.:-.'. __ .:.:_ . ., . : __ .. .:. . ., . -, :, -:: . ...... rv_ . . -. . --_ : ;::. .. .;. .. :-::. wall of the Wide Passage .At the top of the flow stone, an inaccessible entrance to a 2 m high fossil passage is visible. This fossil passage probably conducted the waters which, in conjunction with those from the Main Passage, formed the Wide Passage. At the bottom of the NW wall of the Wide Passage, the Affluent Passage is encoun tered, ascending gently towards SW. Compared with those of the Wide Passage, its dimensions are relatively small, until 70 m upstream, where an other intersection with the rupture of 60 is en countered. At the intersection, the Affluent Pas sa ge is alinost completely filled with loose fallen roc ks. A steep lateral passage leads towards SW, ascending almost 30 m until is closed by debris, pebble and clay. From the intersection, the Affluent Passage continues 30 rri further; being finally obstructed by pebble In its entrance part, Suvi Ponor is hydrologically activ e only periodically e arly in the spring and after h e avy rains. The first active affluent found in the cave is encountered in the passage leading from the bottom of the 17-m deep pit after the second hall. The stream flowing through the Wide Pass a g e is supplied by th e tributari e s from the Main Passage, a s well as from the Affluent Pas s a ge The Afflu e nt Pass ag e togeth e r with its lat eral passages, represents the genetically youngest part of the cave, draining the waters collected by the sumps upstream in the valley. In its final part, the stream flowing through the Wide Passage sinks through the debris and the muddy sump which closes the passage. IBRIN PONOR lbrin Pmwr is located 1 km north of Suvi ponor, a t a place where a periodic stream coming from the SW meets a limestone cliff with a SW-NE orien tation . Two swallets are encountered at the bottom of the 20-m high cliff: a bigger one which is the entrance tolbrin Ponor, and a smaller one (1 m in diameter, filled with debris) some 30m upstream. lbrin Ponor (Fig. 5) is dev e loped along two s ys tems of fissures : the fissur e s of 80 and 300 ar e visibl e to the largest ex t e nt, and they controlled the excavation of all the pits, whil e the corr e sponditlg -sec ondary fissur es of 30 and 170 ar e I

86 -118-(232m a. s .1) entnsnce PROJECTED E L E VA T ION 2 7 0 "-90" .. length of the explored c(\ve passages is 228 m, while the depth is ...: 118 m. Some 50 m from the entrance, at the depth of only 7 m, the passage changes its directi o n and bec o mes guided by the 80 fissure . On. its "\vesierri edge art ascending laterlil .passage filled w i th d ebris is in-. countered, leading probably to the abov6-ineritionecf upstream swallet towardsthe east; tlie rna:in passage descends through oeep pit and leaves the fissure sw' while af the eastern edge ofthe fissure a laterai leadingto the base of a shaft," were no signs of active "affluent have been found. Changing direction several times, the main sage encounters the 300 fissure; entering a series of pits and slopes .with tota,ldepth of 85 m. the bottom :of the last pit, a n accumulation of pebble M_. Zlokolica et aL an d water partially closes the pas s age; and makes further progress impossible without diggi ng Ibriri Ponoi: .is hydrologicail y act.ive only periodi : cally, d uring snow meltin g per io d s and after heavy rains . Altho ugh the entrance part :Is dry," in the deeper sections of the cave (starting m>. a very weak stream appears ( < 0.5 llsec), AtOthe end of. the passage explored so far, the stream sinks through pebble, flowing ovef. rai).s;, which . can be behind the impe)}etraple part o"fthe passage MAMUTOLO .. str:ellll1s are .sec.. tions of a fossil tiver floor which leadS froi:ri Ibrin ponoi. towards rioith. Two of ib()se join,


Significant caves a t t h e westem rim of t h e Miroc karst 87 upper entrance 0 PROJECTED ELEVATION 1ao-o western lake PLAN N 50 I .... . .. ... . -= .... .. and 30 meters further, contact rocks, they sink in the cave of Mamutolo. Although its entrance is impressive (over 20 m high)., the cave floor is filled with a thickiayer of debris, and the passage closes completely after only 40 m. BURONOV PONOR The valley of Buronov Ponor is the northernmost valley of the rim of the Miroc karst It contains a temporary stream, 2 km long, which sinks into Buronov Durin g the period of snow melting, the water .fills the entrant passages, making the c!lve inaccessible except through the dry, upper entrance, tliiough a 50 m long dry pas sage. During the dry part of the year, the entire cave is accessible over its 1950 m of known length. Buronov Ponor (Fig 6) is a cascading cave, with the total depth of 187 rn, with 8 pits, the deepest of which is a 30-m deep pit in the Rio Bravo lateral .. passage.:. The r:rtain passage of the cave strikes northward and northeastward. The cave is poorly ramified, with two basic types of passages: the oval-shaped passages, and the rift passages (canyon.;.shaped). Significant deviations from these shapes are noted in the entrance parts of the cave, as well as in the chambers, the collapsing process affects the shape of the passages. The canyon-shaped passages have a N-S orienta tion, and are notably enlarged by water erosion (width varying from 3 to 7 m), with polished walls and great heights (ranging from I 0 to over 30 m). The rounded shapes of passage walls are dominant in the Northeast Passage arid in the upper levels of the Pool Passage. Their shapes are semi-ellipses, 0.5 to 11.5 m high and 2 to 7 m wid e. The passages are floored with roughly rounded fragments of non carbonate rocks and with almost with out the clay component. A 0.1 to 0 5 m deep river bed i s carved in the sediments. Scallops are carved in the walls of the passages, and a gutter is en countered on th e ceiling of the passage, both in the


88 Northeast Passage, the upper level of the Pool Passage and the ceiling of the Collapsed Chamber. In the Northeast Passage and the upper level of the Pool Passage, a layer of clay has been deposited over alluvial sediments, being later covered by travertine in form of flowstone. Fragments of this sedimentation series are preserved above the pres ent-day level of river-bed erosion, at the places of sharp passage bends. In the lower part of the Pool Passage, rocks of metric dimensions have col lapsed over travertine deposits. The upper level is mostly formed above these rocks. The lowest part of the cave, Rio Negro, is com posed of two spacious passages with lakes-sumps (?), which require cave diving being at the abso lute height of 85 m a.s.l. This part of the cave also contains collapsed rocks, and, unlike the rest of the cave, deposits of pure, fine sand. The floor and the walls of the passage beyond the Collapsed Chamber are covered with a clay coat ing which is over 1 m thick. REFERENCES BOGDANOVIC P (1978) Tumac z a osn o vnu geolosku kartu list D. Milanovac 1 : 100.000. Savezni gezavod, Beograd CVIJIC, J (1921) Djerdapske terase. Beograd. itf. i The system of the passages with is particu. larly interesting. The Eastern 1s .a accumulation formed by a traverune parner; 1ts level is 5 m above the stream in the passage beyond. A maze of passages with lakes and sumps, yet' explored, is developed towards N and NE. Ne1ther the passages beyond the Western l-ake have been completely explored Beyond the Western Lake is encountered an absorbent sump, probably con nected with the passages beyond tbe Eastem Lake. An unexplored ascending passage, filled with clay covered boulders has also been eqcountered in that part of the c&ve. Rupture planes of NE-SW orientation have been detected in the cave. In the zones where they oc cur, collapsing processes (with boulder size from 0.1 m to over 10 m) as well as deposi tion of clay are active and the passages have a cascading form This is most obvious in the Pool Passage and in the Collapse Chamber MILIC, C (1965) Morjologija kraske oaze Miroc Zbomik Radova Geografskog Instituta J Cvijil! SANU pp. 15-54. ZEREMSKI, M. (1988) Geomorjoloska karta kraske oaze Miroea Zbomik Radova Odbora za Kras i Speleolog iju, SANU, Beograd


Theoreiical and Applied Karstol(Jgy, vol. 9/1996, pp. 89-95 Hydrogeology of Miroc Karst Massif, Eastern Serbia, Yugoslavia ABSTRACT Zoran STEVANOVIC, Veselin DRAGISIC, Petar DOKMANOVIC1 & Mihajlo MANDIC2 The karst massif of mountain is one of the national parks of Yugoslavia, situated in the northern extremity of the Carpatho-Balkanides of eastern Serbia and extending to the Danube and the Romanian border. The area of the calcareous massif, composed mainly of Tithonic limestones, exceeds 120 km2 Four karst springs used to drain a karst aquifer located in the north and the west of the massif, along the Danube, before being submerged by the reservoir behind the dam at Portile deFier (The Iron Gates). To the south, the massif is drained by Ble derija spring (Q . about 30 1/s). The massif has a multitude of active concentrated swallets. Several deep caves (e.g. Rakin Ponor, 278 m deep, the deepest in Serbia) were recently investigated by speleologists Explorations have been undertaken in area for karst ground water intake aiming at the water supply of Donji Milanovac resort center. Key words : Upper Jurassic limestones, karst aquifer, water balance, ground water reserve. Caracteristiques hydrogeologiques d u Massif karstique de Miroc dans Ia Serbie orientale, Yougoslavie R E SUM E Le massif karstique de Miroc, l'un des pares nationaux de La Yougoslavie, est situe a l'extremite de La partie septentrionale des Carpato Balkans de La Serbie Orientale et s'etendjusqu'au bord du Danube eta Lafrontiere de La Roumanie. La suiface du massif karstique, constitue de calcaires d'age Tithonien couvre au total plus de 120 km2 Dans La partie Nord et Ouest, au bord du Danube, ont ete situes quatre sources karstiques representant de principaux elements de drainage de l'aquifire karstique avant leur submersion par le Lac d'accumulation du barrage de PorJile de Fier (Partes de Fer). Le drainage de La partie Sud se fait par La source Blederije ( Qm"> 30 lis environ) Sur Le massif calcaire existent de nombreuses pertes concentree:f actives; au c ours des recher c hes speleologiques intensives effectuees recemment ont ere etudiees certaines pertes tres profondes (par ex. : Rakin p o nor, 278m de profondeur, qui est La plus profonde perte de La Serbie). Les recherches des eaux de sources karstiques sont en cours et on envisage de faire le captage des eaux des sources karstiques de La region de Miroc pour alimenter en eau I' important centre touristique de Donji Milanovac. Mots-cles : Calcaires du Jurassique superieur, aquifire karstique, bilan des eaux, reserves d'eaux souterrain. INTR ODUCTION The karst massif of Miroc, of meridian trend, is one of the largest in the Carpatho-Balkanides mountain range. It is situated in north-eastern Serbia, on the right side of the Danube valley. Its area is about h8 km2 (30 km l e ngth, 8 km max. width) and the average altitude is about 450 m. Prominent peaks of the massif are Veliki Strbac (768 m) and Mali Strbac (626 m). Mount Miroc is part of the Djerdap National Park which is under governmental protection. 1 University of Belgrade, Faculty of Mining and Geology,? Djusina str., Belgrade, Yugoslavia. 2 Geozavod HIG, ul. Karadordeva br,. 48, 11000 Beograd, Yugoslavia


90 The climate of the Miroc general area is moderate continental, characterized by cold winters and hot summers; and a mean annual temperature of about 11. C. Mean annual the massif is 790-860 mm, the larg(lst amount (some .. 30%) occurring from April to June, and the smallest(abotit 19%) in . A characteristic of the Miroc karst hydrography is the lack of perennial streams except during heavy rainfall and snow melting .periods. Intermittent Catchment area Submerged springs (Djerdap accumulation) Blederia and Sokolovica Karst Spring streams of northern and central ma.ssif: Suva Reka, Bela Voda, dolubinjska Reka, Ra}cina Rel

Hydrogeology of MirocKarst Massif Sch&f. ,..,, '"'' . interred"> lio .Qsi;Jawts;; ;_.:. < .... o !:',-, .. .. ' "" ' ', ;,, ''. , figiOiud-desfrtnires''il4ttil#s < nales (Jir,()..,_rlts ,t. mpp'Q$!-es)f. 2. ; : .. . .. .. : ' ,: .. 1\. .: I .. 5j{,.I;,R :. -_: v>r;{ .. 0 '' ... .. 91 ,..nhcm hJ'IIkl.:!iJ \ ....__.___.._ !!plftl . th;\tn.'\f('C!II dnn&e k:\rst .. m nnnk:t.rslified / LLLJ rncb dro'-'nC"d optin& 10 II llo


92 Lithostratigraphic members included in the mountain and its immediate surroundings are the following: Early Proterozoic schists: amphibolite, rnig matite, gneiss, etc. in the Getic Nappe; Cambrian dominantly chlorite-sericite schists, outcropping along the northwestern margin of the karst (in Miroc Anticline); Hercynian (Carboniferous) granites exposed by erosion in the Miroc Anticline; Middle Jurassic rocks unconformably dis posed over the older formations; basal rocks of this unit are conglomerates and sandstones, overlain by coal sandstones and shales with a thin bed of hard coal, passing upward into sandy limestones, sandstones, and conglomer ates. The thickness of the Middle Jurassic unit is about 100 m; Karstified reef, marly, red limestones of the Upper Jurassic, up to 300 m thick, form the Miroc 'karst oasis ; Valanginian/Hauterivian (Neocomian) rocks are exposed in the anticlinorium and Krajina syncline. They lie conformably over Upper Jurassic and are represented by reef limestones, calcareous sandstones, grey wackes, siltstones, and marlstones, in an aver age thickness of about 200 m; the Barremian/Aptian formations outcrop in limbs of the Miroe anticline, where they lie con formably over Neocornian rocks. The unit is composed of limestones, marls, shales, and sand stones with a total thickness of about 400 m; the Albian unit overlies Aptian rocks, and is composed of shales, marls, and marly limestones. the unseparated Albian/Cenomanian unit lies exposed in a narrow belt, along the eastern limit of the Miroc anticlinorium, where it unconformably lies over Upper Jurassic karsti fied limestones. The unit is composed of sand stones, conglomerates and shales, and has a thickness of up to 150m; the Turonian/Senonian formation lies con formably over Albian/Cenomanian rocks. It is uncovered in the eastern limb of anti cline where it is composed of silts, marls and sandy limestones, and has a thickness of about 300m. A series of soft sandstones and sand-marly clays, exposed on Miroc eastern slopes, lies Z. Stevanovic et al. '_ unconformably over Mesozoic or older rocks, and it was assigned to the Neogene age. the Quaternary unit, which occurs on the north-western and northern margins of the karst massifs, represented by several genetic types : river terrace sediments (Pleistocene) -gravel, sands, clays, sandstones, and lime stones; deluvial deposits (Pleistocene), formed dominantly in river-terrace backland; debris (Holocene); and alluvial deposits of the Da nube (Holocene). HYDROGEOLOGY AND KARST GROUND WATER BALANCE Rocks classified according to their hydrogeologi cal function include: the karst aquifer in the Upper Jurassic karsti fied limestones, which contains the highest water reserve; all other rocks, marginal on karst, which are less permeable than the karstified lime stones and form hydrogeological barriers to ground water flow. The total area of the exposed karst aquifer is about 128 km2 and that of non-karst deposits that pro vide additional supplies to the aquifer is about 34 kn12 The area of submerged springs (the north ern catchment) has about 104 km2 of exposed karst and 20 km2 on bordering non-karst terrains, whereas the Blederija and Sokolovica springs (southern catchment area) include 24 km2 of open karst and 14 km2 of bordering non-karst terrains (Table 1). Upper Jurassic limestones are the thick est (300 m) in the central massif, thinning out to the south (DRAGISIC et al., 1988; 1992). The main modes of ground water recharge are rainfall infiltration and percolation from surface streams which are intermittent over the year The karst aquifer is also supplied from numerous sinkholes, uvalas and ponors, which collect rain fall or sinking streams (MILIC, 1965; ZEREM SKI, 1988). Explorers of the Academic Speleo logical Club discovered, since the early nineties, more than twenty cavities on Miroc, including the deepest cave of Serbia ( Ponor: -280 m), and explored a length of about 8000 m of their pas sages Most of these caves are active swallets. Nemacki Ponor and Buronov Ponor are the caves of more 1000 metres of explored passages (MANDIC & LJUBOJEVIC, in press). The northern catchment of the karst aquifer is drained through four large and several smaller


Hydrogeology of Miroc l(arst Massif ., springs, now submerged by the Danube in Djerdap gorge. The southern catchment is drained through Blcderijaand Sokolovica springs. Part of the karst water is probably drained underground into the marginal, less permeable, rocks. The main characteristics of the springflow regime for Bele Vode, Hajducko Vrelo, Pena, and Pestera springs (Fig 1 ), which drain the northern catch ment of the karst aquifer, are given in Table 2; schematic sections of Bele Vode and Pena springs, before their flooding by the Djerdap storage reser voir, are shown in Fig 3 As shown in Table 2, the cumulative minimum flow rate of the submerged springs is about 65lls; and inthe period ofhydro:. : logic maximum itis about : 6 5 m3/s, or / 100 times .. higher (PETROVIC, 1968; STEV ANOVJC, 1991) . Large springs emerge at the cqntact of the lim.e stones with the alluvial deposits at the Danube, with the exception of l-Iajducko Vrelo (the highest with respect to the old. Danube level) .. which emerges at the coritact of the limestones with the Carboniferous granites. Blederija the south eastern margin (see Fig 1) i s a seepage spring of ascending type, composed of four emergence (springs), controlled by the contact of the limestones with the nian impervious rocks Three of the four springs yield a total of 30 lis ofcold (9.5 oq water. The fourth that discharges subtherrnal ( 17.5 C) wa:teris associated \vith deep flooded circulation; its o\itfiow is acco :mpanied : by inteimihent gas pulsation. the lowest discharge of the sub thermal spring is 10 lis. The of its watei: expressed by Kurlov equation: h} llanuhr .. --====. The water balance equation for Miroc karst massif can be written as: (1) where, P = rainfall; Qgw = ground water runoff; E = evapotranspiration, because surface runoff is nonexistent. Based on earlier research data, corre sponding values On mm of watercolumn per year) have been adopted for balance components P and E I> for each segment of the Miroc karst aquifer catchment area (surface areas are given in Table 1 ): a) Noithem : c atcluilent, ctta.hted l:)y. springs in Djerdap gorge: Karst : : P :::: 859 mm, E = 370 nlln.; Nonkarst: P :;;; 862 mm, E:::: 645 inrit; b) Southern catchment, drained by Blederija and Sokolovica .. springs: Karst: . P = 792 mm, E = 370 min; Nonkarsi: P 821 mm, =587 nun. The average underground runoffs from drainage areas of the Miroc karst aquifei:, calculated using equation (1), are: (a) Qgw = 1.75 m3/s and (b)Qgw := 0.43 m3/s, ot, the total average underground runoff or dynainic ground water reserve, is: . . 1 E for non -karst terrain are taken from the Water Mailagemerif Plari of Serbia; for karst areas this balance coinponent was c\llculated by the same method as that used for and Beljanica karst of eastern Serbia (RISTIC. : 1995) .


94 Based on the analogy with the and Beljanica karst springflow regime (RISTIC, 1995; STEVA NOVIC, 1994) the mean monthly minimum water yielding capacities of the karst aquifer at 95% confidence level are : (a) Q1w = 0.287 m3/s and (b) Q1w = 0 .071 m3/s, or for the whole drainage area of karst aqui-. Q1w = 0.358 m3/s or Q1w = 11.29 x 106m3/year Only a small part of the karst ground water reser voir is currently used, mostly at a number of spring intakes. The significant reserve of ground water in the catchment area related directly to the Danube can be used for supplying the resort center and small town of Donji Milanovac; hydrological, geophysical and hydrogeological surveys for this purpose are under way The principal problems arise from the small free areas available for building the intake structures in the steep-walled gorge, and from the possible influence of the Da nube water on the underground karst) CONCLUSION The karst massif of karst oasis'), in north-eastern Serbia, has an area of about 128 km2 It is composed of karstified Upper Jurassic lime stones about 300 m thick, which form an aquifer with a large ground water reserve. On the northern karst plain margin, the aquifer is largely drained through springs which have been submerged by the Danube water since the Iron Gates 1 (Djerdap) dam was built. The calculated average long-term REFERENCES BOUE A (1840) Esquisse geologique de Turquie eu ropeenne Vienne. CVInC, J (1903) Geomorfologija Spec. ed., Beograd. CVInC, J (1921) Djerdapske terase, Glas. Srpske Kraljevske Akademije, CI, Beograd V., FIUPOVIC, B. & DIMITRIJEVIC, N. (1988) Pojave termalnih voda na padi nama Zap i snici SGD za 1985-1986, pp 225 229 V., STEVANOVIC, Z. & FILIPOVIC B (1992) The occurrences of deep siphonal circula tion of karst aquifer of the mountain (NE Serbia), Theor Appl Karst S, pp 115-120. -ground water run-off is 2.18 m3/s, some 80% of which flows by gravity to the submerged large : springs. All the investigation data suggest that significant1 amounts of ground water are stored in the karst. The utility of this water, primarily for water supply, has not been considered mainly due to the inadequate hydrogeological exploration of the area and to the difficulties in tapping ground water. The principal tasks are: To define the extent of the karst aquifer catchment sub-areas and the main drains. To assess the amount and the regime of ground water flow at the submerged springs under the new hydrogeological (hydraulic) conditions created by the Djerdap reservoir. To establish the static ground water reserve, which is reasonable to expect to have in creased significantly (compared to the natural situation) under the backwater effect. To analyze the quality of the ground water, especially in the northern of the karst aquifer, considering the possible mixing with the polluted Danube river water. To protect the karst aquifer -a task partly facilitated by its location in the national park as an area of particular environmeQ.tal concern. Preservation of water quality in tfie northern catchment sub-area, where mixing with the Danube river water may occur, js an open problem for possible artificial extraction of ground water. MANDIC M. & UUBOJEVIC, V (in press) Prelimi narni prikaz rezultata dosaddnjih istra!ivanja krdke oaze. XI Kongres speleologa Jugoslavije, Valjevo MILIC t. (1965) Morfologija krille oaze Zbornik radova Geografskog instituta "J_ovan Cvijic", 20 MILOV ANOVIC B. (1953) Stratigrafija i tektonika i Velikog Grebema u Srbiji. Vesnik zavoda za geoloska i geofizitka is-traiivanja, Beograd PETKOVIC, V (1935) Geologjia i s tocne Srbije spec ed SKA, Beograd. PETROVIC, J. (1968) O_polot.aju krdkih vrela i peCina u Djerdapskoj klisuri Zhornik za prirodne nauke Matice Srpske, 35, Novi Sad


Hydrogeology of Mirof Karst RISTIC, V (1995) Metodologija analize i piognoze izdasnosti vrela Kucajsko-Beljanickog mas iva, Magisiar ski Rad, Rudarsko-GeoloSki Fakultet, Beograd. STEVANOVIC, Z. (1991) Hidrogeologia karsta Kar jJato-balkanida istocne Srbije i mogucrwsti vOdosnahdevanja Monography Spec. ed. Rudarsko geoloski Facultet; Beograd, 245 p. 95 STEVANOVIC, Z. (1994) Ground Waters in Carf?onate Rocks of the CarpatMan-Balkan Mountain .Range. Monograph. Spec. ed CBGA Alston Holding Co. Jersey, 237 p zEREMSKi, M. (1988). karta .kraSke oaze Zbomik radova fidbOra ia kras i SANU; DLXXXIX. Odeijen]e Pri63, Be6grad.


ABSTRACT Theoretical & Applied Karstology, vol. 9/1996, pp. 97-112 Analysis of Spring Hydrographs for Hydrogeological Evaluation of a Karst Aquifer System Nozar SAMANI & Babak EBRAHIMI1 Karst aquifers are, in general, extremely heterogeneous in character. As a result, quantitative data obtained from selected points in the system, either by pumping or by using marker dyes, can be rarely extrapolated to evaluate the average function of the system as a whole. In contrast, recession curves of spring hydrographs show the overall response of the aquifer to precipitation events and help evaluating aquifer storage and drainage poten tials. In this study, three analytic equations are used for estimating hydrogeological characteristics of a karst aq uifer: Mangin's equations, a newly proposed equation that assumes that the recession curve is composed of both quickflow and baseflow, and Coutagne's equation, which considers the recession curve to be the response of a single reservoir. The hydrographs of three springs draining Sabzpooshan and Beaza aquifers in Fars Province, Iran, are analyzed. From the analysis, it is apparent that the saturated zone accounts for 99%, 91% and 99%, re spectively for Pirbanoo, Polberengi and Ghorehdan springs. . Key words: karst aquifer, evaluation, spring hydrograph, analysis. RESUME Analyse des hydrographes des sources en vue d'une evaluation hydrogeologique d'un systeme aquifere karstique Les aquiftres karstiques ont, en general, un caractere tres heterogene. Par consequent, les donnees quantitati ves obtenues dans certains points du systeme, soit par pompage, soit par ne peuvent pas etre extra polees que rarement pour evaluer La fonction moyenne du systeme dans son ensemble. Par contre, des courhes de recession des hydrographes des sources presentent La reponse globale de l'aquiftre aux episodes pluvieux et aident a l'evaluation du potentiel de stockage et de drainage de l'aquiftre. Dans cette etude on utilise trois equations anaLytiques pour estimer Les caracteristiques hydrogeoLogiques d'un aquifore karstique, Les equations de Mangin et une equation proposee par les auteurs, en supposant que La courbe de recession comprend aussi un ecouLement rapide, qu 'un ecoulement de base, ainsi que /'equation de Coutagne, qui traite La courhe de re cession comme La reponse d'un seul reservoir. Sont analyses Les hydrographes des trois sources qui drainent les aquiflres de Sabzpooshan et de Beaza. dans La province de Fars, en Iran. L'analyse montre que La zone saturee represente 99%, 91% et 99% respectivement pour les sources tie Pirhanoo, Polberengi et Ghorehdan. I Mots ctes: aquifere karstique, evaluation, hydrographes des sources, analyse. 1 Department of Geology, College of Sciences, Shiraz University, Shiraz 71454, Iran.


98 INTRODUCTION The shape of the outflow hydrograph recorded at a spring is a unique reflection of the response of the aquifer to recharge. The shape and rate of reces sion, in particular, provide significant information on the storage and structural characteristics of the aquifer system supplying the spring. For these reasons, the analysis of spring hydrographs offers considerable potential insight into the nature and the operation of a karst drainage system. Due to the inherent characteristics of karst, there are two types of flow (ATKINSON, 1977). Slow flow, i.e. the so-called diffuse flow, occurs through karst fissures of small dimensions, generally in a laminar regime. Turbulent fast flow, or conduit flow, occurs in larger fissures, through irregular karst conduits, with dimensions varying from I 0 mm to one meter or more. A study of the spring hydrograph may help to dis tinguish the degree of structuration of the flow, the influence of any wide conduits, the controlling effect of the stored reserves, etc. (PADILLA et al., 1994). Quantitative analysis of hydrograph recession derived particularly from the work of MAILLET ( 1905), and the first attempts to apply a simple theoretical solution for karst regions were made by BONACCI (1987). The influence of the form of the drainage network on the hydrograph of karst systems has been studied by numerous authors; some have attempted to simulate real conditions by using laboratory scale models (ROBINSON & GALE, 1990), some via flow-simulation mathe matical models (KIRALY, 1975; PADILLA, 1990), and others with conceptual models based on regional studies, such as those of DROGUE (1972), MANGIN (1975), ATKINSON (1977), WILLIAMS (1983), PULIDO-BOSCH & CASTI'LLO (1984), BONACCI (1987), SMART (1988), and CHRISTOPHER ( 1992). SOULIOS (1991) has classified karst aquifers on the basis of recession of major karst springs in Greece. The study of the recession curve has been ap proached from two points of view: models which take the discharge of an aquifer via a spring to be the algebraic sum of two or more functions repre senting different flows within the system (SCHOELLER, 1965; MILANOVIC, 1976; MANGIN, 1975), and models which consider the discharge as being produced by a single system (MAILLET, 1905; COUTAGNE, 1968; SINGH, 1969; ATKINSON, 1977). N. Samani & B. Ebrahimi In this paper, we apply Mangin's equations, Cout agne's equation (PADILLA et al., 1994) and a newly proposed equation that assumes the reces sion curve to be composed of both quickflow and baseflow, to the hydrograph of three springs that drain Sabzpooshan and Beaza karst aquifers in Fars Province, Iran. This resulted in a better understanding of the importance of various compo nents of the flow, storage and transfer properties of the aquifers. GEOLOGY AND HYDROLOGY OF THE STUDY AREA The 55 km Sabzpooshan anticline in the Zagros Mountain Range (ZMR) is located southwest of Shiraz (29'-30' N and 52'-53' E). The anticline falls in zone three (simple folded belt) of Zagros Orogeny, with a maximum height of 1200 m from Shiraz plain. The exposed geological formations, in decreasing order of age, consist of Pabdeh shales (Late Cretaceous), Asmari-Jahrom limestone and dolomite (Eocene-Oligocene), Razak evaporites (Oligocene-Miocene), Aghajari sandstone (Miocene-Pliocene), and Bakhtiari con glomerates (Pliocene). The detailed lithology of these formations is described by STOCKLIN & SETUDEHNIA (1977). The well karstified Asmari-Jahrom formation has provided suitable conditions for reservoirs of karst water in the Sabzpooshan anticline. This karst aquifer is drained by two major springs. These are Pirbanoo spring and Polberengi spring, with elec trical conductivities of 708 and 1437 J.l!l!cm, respectively. The average annual rainfall measured in this area is 326 mm. SAHRAEI (1995) has showed that there is no hydrogeological connection between Polberengi and Pirbanoo springs; however, there is a hydro geological correlation between Polberengi spring discharge and the fluctuation of the water level in an observation well, in other words, Polberengi drains east Sabzpooshan and Pirbanoo drains west Sabzpooshan aquifer (see Fig. 1). Buozangan anticline in the ZMR is located at the Beaza region, 40 km northwest of Shiraz (30'30015' Nand 52'-52' E). The anticline falls in zone one (Zagros thrust zone) of Zagros Orogeny, with a maximum height of 2888 m from sea The exposed geological formations, in decreasmg order of age, consist of Pabdeh shales, Asmari-Jahrom limestone and dolomite, and Razak evaporites.

PAGE 100

Analysis of spring hydrograplts for hydrogeological evaluation 99 1 S71ririg Obsen;alinn uxJ1 Fcn.dt ihdidiTUII aii.s As-Ju Limestr:mr. Pirbanoo spring \ Y' Beaza Syslm Polberengi spring SabzJXXJshnn S rin Max. altitude (m) Spring altitude (m) Q,,,. (Vs) Q,mn (1/s) Qmcan (1/s) Standard deviation TDS (g/1) Ec. A uifer A vcrage annual rainfall (mm) Stud Pirbanoo 2446 1484.184 311 107 216 39.6 0.500 708 326 1987-1994 The well karstified Asmari-Jahrom formation has produced reservoirs of karst water in the Buozan gah anticline (in the Beaza region), too. This karst aquifer is drained by Ghorehdan spring, with an electrical conductivity of 392 J!nlcm. Ghorehdan spring .. a .Tehran { l () 10 Km 0 6 Krrl __. ;:;p u.: :: Beaza Ghorehdan 2446 2888 1474.636 2060 282 1381 75 810 144.5 1033 44.9 136.8 1.000 0.194 1437 392.3 326 510 J 1987-1994 1995 The average annual rainfall measured in this area is 510 mm. Table 1 summarizes the main charac teristics of the two karst aquifers.

PAGE 101

100 THEORIES MANGIN'S EQUATION Mang i n considers that two basic hydrological entities are distinguishable in the interior of a karst dra i nage system, one being the unsaturated zone with a nonlinear flood recession, which he repre sents with the function 91 and the other being the saturated zone with a linear baseflow recession represented by the function
PAGE 102

Analysis of spring hydro graphs for hydrogeological evaluation 101 PROPOSED EQUATION PADILLA et al. (1994) show that Coutagne's equation cannot describe the entire recession curve, thus they introduced Equation 9, the Q,. of which has no real physical meaning. On the other hand, other authors (MANGIN; ATKINSON) show that only the late section of the recession curve may be described by Coutagne's equation (with n = 1), while the early section is a nonlinear recession; therefore, MANGIN (1975) simulated the first part of recession curve by an empirical function (Equation 3). We assumed that this part of the recession curve may be described by a function of a form similar to Coutagne's equation and that quickflow dis charge has a nonlinear relationship with the stored volume available for drainage via the spring in the unsaturated zone: Q. = cvn (11) The discharge at any time ( Q;) is : dv Q, =-dt (12) and thus dV +cvn=o for n> 1 dt (13) where V is the stored volume available for draining in quickflow period. Same as Mangin, we have considered that the dis charge from the saturated zone (during baseflow period) h as a linear relationship with the stored volume available for draining via the spring (MILANO VIC, 1981, shows that this equation refers to laminar flow in an aquifer): b dV +cvb=O, dt (14) where V b is the stored volume available for draining in baseflow period. Therefore, in this method we have considered that during recession in a karst spring, the discharge at timet is expressed by the sum of two components -e,, that is the s olution of differential Equation 13, and"' that is the solution of Equation 14: 'l'z Q, =
PAGE 103

102 N. Samani & B. Ebrahimi Discharge (lis) :mo ,-------------------------------.----------------------------------:-l I Polberengi spring . 2::10 t-1 I I ISO I I 00 50 .. 0 L ___ j ___ .. .J-,-....... L--:---L.:... ___ L __ ____ __ __ _j 0 15 30 45 60 75 90 105 120 135 15 0 .165 180 195 4 F ebruary 1989 Time ( day)_ September 1989 Discharge (lis) . . 300 -----. -. 275 I I I 250 r 2251' 200 1751 1 1 50 1 ........... ..... J_:__. ___ -.1__ .... L. .. -, .. L -.. .. .. .Lc .... __J ___ _-_J _________ _L _______ _j 0 15 30 45 60. 75 90 105 120 135 150 165 180 195 210 4 F cbrt.1ary 1989 Time (day) September 1989 Discharge (lis) 1 500 ------------. 1350 -1 1200 -1050 -900 /'50 6oo -------.L,..:. ___,__ _____ c _ _ __ __j ____ () l s 30 H April 1995 45 60 'i'5 90 Time (day) 105 120 Augl)sl 1995

PAGE 104

Analysis of spring hydri).gttiphs for hydrogeplogiC"al evaluiziion . .... .. :. ,. : .. J?ischarie '(Its) 300 . . . -.: . . Adapted functions / I : . : .' baseflow : j 250 . -:-.;. quicklow ..-:: 1 ..... . --= ... ...,., l 200 150 .. 100 1 ... ;. ... 50 I . I 0 __ ___ _i__ __ _L_ __ ,_j _:_t __: __ j_: __ () 1 5 30 45 .60 75 90 105 120 135 150 165 180 195 4 F ebruary 1989. Tin1e(day) Seplerriber 1989 300 J ... t:. ,_I,) I 1'5 -Discharge (Vs) . .. Pirbanoo ... . 150 L..,_ _j___:.._ __ o 2o 40 .' :_6 o : -80 < ... 120 4 F'ebru a r y .1989 . <:lis); .. "1500 .. >. Ghorehdan 1350: 1200 .. 1050 i ---: l ........ :: baseflow ..., .. quiddow .. I 140 160 180 2QO ... ,1989 Adapted functions I .. baseflow J -quickJow l I I I 7501 L i 600 __ ,____!_ ___ ..J..._ ___ ___.L _______ l_ _________ j 0 20 40 60 80 100 120 140 8 April 1995 -Tirnti (day) August 1995 103

PAGE 105

104 This function does not depend on the input peak, thus it is useful for comparing the quickflow of different springs. Because Function 16 is always positive, we consider the time (t98 1.), when 98% of the volume of the quickflow reservoir is drained; after this time we consider the quickflow to be nearly zero. Thus the volume drained by the spring during the quickflow is: .. . Parameters Polberimgi spriiig .: ... Q0 (lis) 271.2 (lis) 107.05 a (day-') 0.001825 (lis) Hi4.i5 ti (day) 120 f..l (day-1 ) 0,008333 .. E (day-1 ) 0.028 v; /V0% 91.65 vc; /V0 % 8.35 0.998 r N. Samani & B.-Ebrahimf APPLICATION AND DISCUSSiON been applied to the hydro graphs of. Pirbanoo, Polberengi arid .:Goorehdan springs. The graphiea(representatior{ of the e, is shown in 3. .. for Mangin's formula have been calculated and are shown in Table 2. In Fig. 4 we show the values of the "functi()n. This function displays a raster decline tor noo and Ghorehdan springs for Poiberei1gi spring, w hile the decline rate for Gliorehdan spring .. . :.: .280 1350 .. .. 247.84 851,37 0.000644 0.000366 52.16 498.63 90 76 0.0111 i 1 0.013158 0.054 0.0369 99.70 99.50 0.30 0.50 0 .997 0.993

PAGE 106

Analysis of hydrogi'aphs for hydrogeolOgical eviiiUatJo.n lOS -: ::' , : . , ... a I Polbcreilgispiing Ghorehdan spring I : I I i I I . . . ... j 0 _ ____ _. ___;.. ___!,___, '...l.. ___ J. _______ ..... _......L.. ____ _ :..,...___! 0 1 5 30 45 .60 75 9 0 . 105 120 Time (day). Volume drained( %) ;.._ '. : .. . : (Ji .. ;_ : draiiie; par port a celui de' 1 'eau ."i : . .. : l 'ecoulfcmeni rapide ... (a), et pour ia. quantile .. (b) sysie.mes > 100 --. -.. . -.. -,..-----. ----. -.. .. . IJO I . ,. : . -. Polb e rengi spring (l() ... .. ....... ---:_... : .. 40 I -.... .-.. ...,' 2 0 ___ .;.-_ ----.: :::.. . . .. ... ;' l< ;. ::.._:;: .. '. ----0 :...:-: =-= ... = = == -= = = ; = -======== = ==....., ' .. 0 30 6 0 90 120 150 180 :::!10 240 270 :300 330 360 is the same as that of PirbanoO spririg, thus :we can conclude that the unsaturated zone of the system drained by Polberengi sptjng has a greaterrole _in the response of the discharge than in the case of Ghoretidan arid Pirbarioo springs. This difference becomes dear in Fig. 5a, which shows the percentage of the ratio between the volume drained in Hme t and ttie volume stored at time t = o, for the quickflow ( v, fvo % ). Over 91.65% of the tot(ll water supplied to Polber engi spring is drained during periods of baseflow and 8.35% is drained during qtiickflow On the other hand, 99.7% and 99.5% of the total water supplied to Pirba11oo and Ghorebdan respectively, are drained during periods of baseflow, and the quickflow has no important role in these springs. Fig. 5bshows the percentage of the drained vol. ume as compared to the total initial volume for_ each of the springs. Polberengi .spring display s a fasier increment rate of the .drained volume a5 compared to Pirbarioo and Ghorehdim springs Tim e (day) .In general, on the basis or this inethod we can conclude that the unsaturated zoiie .l:n the case of the aquifer drained by Polberengi spring : has a greater role during quickflow period than it does at Pirbanoo and Ghorehdan springs. We can also conClude that water infiltratkm from the unsatu rated toward the sarurated zone in i>itbanoo and Ghorehdan springs is faster than at Polberengi spring. This is probably due to the fact 'that the network of: channels in the \msatunited zone in Polbererigi is poorly developed, as compared to Pitbanoo and Ghorehdan springs. Oii the othe r hand, rapid discharge from Polberengi spring a s compared to other springs indicates that trans missivity in this system is befter thari in the case of Ghorc;hdan and Pirbcmoo springs . hi a general view, baseflow has a fairly important role in the Ghorehdan and Pirbanoo springs and controls over 90% of total flow in Polberengi spring. We applied Coutagne's equation for all three hydrogn1phs. The results are shown in Fig 6. The parameters of the functions were calculated and are shown in Table 3.

PAGE 107

106 Discharge (lis) 300 r .. 250 -.. -.. 200 150 -100 -50 r 0 [_ ____ j_ ___ .J_ ____ _j_ ____ L,:_. __ _ . ___ i.:...:_. 0 1 5 :30 45 60 75 90 105 120 135 150 165 160 i95 4 F'e brua r y 1989 Tirne(day) September 1989 Discharge (lis) :300 1 2751 250. 22fJ r--. . . j 150 i__ ____ L _____ J_ __ _!. ___ _ __,.: __ L___.__L_ __ c .L ____ _;__.i_ ___ _j____ ____ j __ 175 0 20 40 60 80 100 120 140 160 Li:lO 200 .:1 v 1989 Time (day) Septembe r 1989 Discharge (lis) 1 .. -: -----. ..... ----.------_--I :J50 1:WO 10 5 0 -1 _. 900 spiing :.-; : : . Adapted ful').ction . ' __ .-_. _.-_ .. _:. :' _.--.... .. : .. ... ,. : 750 -. . . : .. GOO. __ :_..:... ___ __ _.-0 2 0 W 60 80 too 120 n April 1995 Time (day ) August 1995 ,, .. ---:.: ... : :: .. Fig. 6. curve&. to equation .f(lr the three karstsprlngs . _.:,_ .. .. .. :._-:;::c.:-:-.<<.::<:_-_.._::. .-:. :<-'::, /. ,>.;_ _de recession _des propose par teqU!lt(Qn karstiquei -. : .--<->::: ,,:<'-':;; .. /;.;; :._-,.;.__ .. -__

PAGE 108

Analysis of spring hydrograp}IS for hydroge()logiCQt e_vtiluittU,n Parameters Q0 (Vs) (lis) a (day-1 ) (lis) t911% (day) Vt /V0 % V0 /Vo % (day) d ramed sbieitbelle / :> d,_I,J,i4 .. 271.2 : 92.50 0.001076 178.7 180 92.69 7.31 0.025800 1.1 0.995 Volume drained(%) .. Pirbatzoo spring 2so 227.84 0.000644 52.16 105 99.69 0.31 0.045999 1.1 0.996 1350 851.37 0.000366 498.63 111 99.48 0.52 0.039999 LI 0.991 1 00 I -------;,... I ..... .... = . I I .,....,.,., 'II : -----Pirbanoo spring I !' 60 f .. /f/. ---.. PQiberengisprirtg r ;:/ i .. 1 40. f f. t/ 20 I : I 0 _____ .J -_: . :..J_ L--J,.... .. __ _L . Ghorchdan spring I I .,, __ j_ ___ ;___ __ !. 0 30 60 90. 120 150 180 210 ::!40 270 :300 330 360 Time (day) Alpha ( 1/day) 0.04 r ... ____ :...._ ---.. ----.--------------, I 1 0031 __ ..:.-:.-.:. ---------.,.,.----_.-:--. -. ---0 0 .02 Pirbanoo spring ........... --: ........ .,,.., """. 0.011 ;..::;. ........ I ....-:. Polberengi spring f'' . Ghorehdan spring .. () .. __ _;_--I...,:.;.. 0 200 400 600 800 1000 1200 Discharge (1/s) 1400

PAGE 109

108 !1/ .Sa. mani & B. Ebrahimi . j .. . . paseflow . . quick! ow 150 100 50 Q ---L---' _ L .. ___ ]__, ___ 0 1.5 30 45 60 75 90 105 l20 I35 150 165 ttio 195 4 Februai y 1989 Tirne(cia.y) S eptember 1989 Discharge (Us) 300. _______ ..____,. _ -............ .... ... .... --, Adapted functions baseflow -quick low Pirbanoo spring 275 250 -. I 225 !' 200 ::: _ --"---.. -__ _L, ___ L.._._ o zo 4o eo .eo.. too 120 t4o. teo 1a? .. .. 4 : l' .Discharge. (lis) . .''1500 ..--....,.-' ..:.. _.....,. _, ..:..' . . Ghorelidan.$ring A4apted 1350 '1.200 'Ef . .. baseflow -.. quicklbw 6oo L --_ j ___ ...,._:;_ _ ,L _ .:__ ___ .. __ _:_j ____ o zo 4o 6o a o 8 April J995 Time (day ) 100. 120 .. (40 August i 995

PAGE 110

AmilysiS ofspm.zg hydrogrdphsjor hydrogeological 109 values for n suggest that the relationship be tween the flow of the spring and .th(!' hydraulic heud(h) for .all hydrogn\phs is nonlinear. Flow of. the Pirbartoo, Polberengi arid .dhorehdan springs is proportionalto h1 8 hl.9 hl.l, respectively. . . Q,. values for Polberengi;: Pirbanoo and Ghorehdan are '51, 198 and 799 1/s, respectively. . . Fig. 7 shows the volume percentage drained via the springs compared to the initial quantity'.of stored water. If we accept PADILLA (1994) cons'ideration, that the recession cu rve is caused by the discharge from the saturated zone of the aqui fer (after elimination of Qc values from original recession curve), we may conclude that the Ghorehdan aquifer has a karstified saturation zone with rapid discharge and little storage capacity, as compared to Pirbanoo imd Polberengi springs. Fig. 8 shows the variation in a., as a function of spring discharge calculated from Equation 8. The values of 0., are larger for Ghorehdan than for Pir banoo imd Polberengi springs, respectively. That is to say, transmissivity in the Beaza aquifer is larger than in aquifer that is dfained by Pirbanoo and Ghorehdan springs . , Param e t e rs Qn (1/s)
PAGE 111

110 8o .-. .,.., 60 ---.... Pirbanoo spiing 40 .. -. . spring I ___ j Ghorehdan spring L.._.. ____ __.___ ___ __L___ ____ ...,.:...t.. .:... _.........:_ 0 30 60 90 120 150' 180 Tirrie (day) Volume drained(%) 100 r-80 r .-.-.. Polberengi spring .. 60 -Ghordidan spring 40 20 -. ------. --: ---------------__ -------. I -.., 1 _ ____j_. ___t_ ___ . . ..J--0 < --------.. ---:-:-0 30. 60 90 120 150 180. 210 240. 27.0 300 330. 360 0.02 (l.() 1 Tirtte( day) Pol})efengi Spring PirbanOo Spring. --. Gb.orehdfih Sprjng 0 ______ ..J_ ___J..__ __.J__ __ .. ..:_L_ _____l ___ ,,.......J__; __ ....:.L__.J_ 0 f>O 100 150 200 300 350 400 450 !JOO 550 600 Discharge (Vs) .. N; & B. Ebrahimi ..-:r ., .. .... -()f' tb.e: .ajt-eft'h;i(!nt with dis;_cbarge i'or thrthree '; ppur lesttois etudi4es. .. . c Fig. 11 shows the variation of the a.; coefficient as a function of spring quickflow discharge calculated from Equation 18. The values of a.; for Pirbanoo arc larger than for Ghorehdan and Polberengi s pring s respectively, thus we can conclude that in Polberengi spring the flow rate in the unsaturated zone is lower than in Ghorehdan and Pirb
PAGE 112

. I)/spring hydrographs for hydrogeological ,evi,lluiiiion 111 The rapid change a.; at Pirbanoo spring .1ndi..: cates that. di.fferent .. sections, different sections; exist in .zone. In other woids, the:. geologica) characteristics in the unsaturated zone of Pirbanoo aquifer are heterogeneous as. compared to Ghorehdan arid Polberengi spririgs. In Fig. 12b we the percentage of the volume drained as .compared to the total initiat volume, for each of the studied The increment rate of. the drained volume in Polberengi spring : is faster than in the case of Pirbanoo and Ghorehdan springs, in other words, saturation zone in Ghorehclan .. aquifer is more progressive than in Pii:banoo and Polberengi springs. In general, in this method we have applied an analytic equation (Eql)ation. 16) in contrast with the empirical homographiC equation that is applied in Mangin's formula (Equation 3). The application of this equation confirms the results of Mangin's models. In addition, we have used Equation 18 to define the relative importance of various perme abilities in the unsaturated zone of the studied aquifers. CONCLUSION In order to define the relative importance of base now and quickflow from hydrographs of karst springs, Mangin's model and Coutagne's model are reviewed and a new model is proposed. The three models are applied to the recession hydrographs of Ghorehdan spring (draining Beaza aquifer), Pirbanoo spring (draining west Sabzpooshan aquifer) and Polberengi spring (draining east Sabzpooshan aquifer). Using both Mangin's and the newly proposed models, we conclude that over 99% of the water supplied to Ghorehdan arid Pirbanoo spri.flgs is drained from the saturated zone of their re.spective aquifers, and that 9% of the water discharged by Polberengi spring is drained from the unsaturated zone of the east Sabzpooshan aquifer. According to the proposed model, and on the basis of the variation of parameter a.;, the unsaturated zone in the east Sabzpooshan aquifer appears to be more homogeneous than in the case of the Beaza and west Sabzpooshan aquifers. Therefore, it can be said that the transmissivity of the unsat\.Ji:ated zone Of the West Sabzpooshan and Beaza aquifers is higher than that of East Sabzpooshan aquifer. In west Sabzpooshan, karstification in the unsatu rated zone is higher as compared to Beaza aquifer :"":.>. = lilitiar b{the .... c ,_ Ba&eflow recessiOn coefficient [T'1.]; ...... r e ces s i o n coe ffici e nt at .time ... -J.1. = lriv.erse of i; [tJ);. . I , . . . .... ,;; A and, for this reason, we have a rapid discharge with a small storage capacity in the case of the unsaturated zone of west Sabzpooshan. On the other hand, Beaza has a progressive satu rated zone with high storage capacity, as compared to the west Sabzpooshan and east Sabzpooshan aquifer.

PAGE 113

112 Based on Coutagne's formula, it appears that the saturation zone at Beaza aquifer is more karstified as compared to Sabzpooshan aquifer. This conclusion does not confirm the conclusion of Mangin's and proposed models. It is probably due to the fact that this method does not consider the flow com ponent Qc. On the other hand the conclusion of this method rather confirms the conclusion of Mangin's and the proposed models for the unsatuREFERENCES ATKINSON, T., C. (1977) Diffuse flow and conduit flow in limestone terrain in the Mendip hills, Somerset (Great Britain). J. Hydrol., 35, pp. 93-llO. BONACCI, 0. (1987) Karst Hydrology. Ed. Springer, Berlin, 179 p. CHRISTOPHER, G (1992) Geochemical and kinetic evolution of a karst flow system: Laurel Creek, West Virginia. Ground Water, 30, 2, pp. 187-198. COUTAGNE, A. (1968) Les variations de debit en periode non influencee par les precipitations. LA Houille Blanche, Sept-Oct., pp. 416-436. DROGUE, C (1972) Analyse statistique des hydro grammes de decrue des sources karstiques. J. Hydrol., 15, pp. 49-68. KIRALY, L. (1975) Rapport sur l'etat actuel des con naissances dans le domaine des caracteres physi ques des roches karstiques. In: Hydrogeology of karst te rr ains (A BURGER & L. DUBERTRET Eds), Ed. Intl Union Geol. Sci., B, 3, Paris, pp. 53-67. MAILLET, E (1905) Essais d'Hydraulique Souterraine et Fluviale. Eyroles, Paris 218 p MANGIN A (1975) Contribution a !'etude hydrody namique des aquiferes karstiques. These Doct. Univ Dijon. In : Ann. Speleol., 29, 3, pp. 283 -332;29,4,pp.495-601;30,J,pp.21-124. MILANOVIC, P., T (1976) Water regi me in deep karst. Case study of the Ombla spring drainage area. In: Karst Hydrology and Water Resour ces, l, Karst Hydrology, (YEVJEVICH Ed.), pp. 165-191. Colorado Water Resources Publications. MILANOVIC, P., T. (1981) Karst Hydrogeology. Wa ter Resources Publications, Littleton, Colorado. N. Samani & B. Ebrahimi rated zones. So we can say that this method can be applied to analyze the unsaturated zone of the aq uifers ACKNOWLEDGMENTS This research was financially supported by Shiraz University, Iran. PADILLA, A. (1990) Los modelos matematicos aplica dos al analisis de los acuiferos karsticos. Tes Doct. Univ. de Granada. 262 p. PADILLA, A., PULIDO-BOSCH, A. & MANGIN, A. (1994) Relative importance of baseflow and quickflow from hydrographs of karst spring Ground Water, 32, 2, pp. 267-277. PULIDO-BOSCH, A. & CASTILLO, E. (1984) Quel ques considerations sur Ia structure des aquiferes carbonates du Levant espagnol, d'apres les don nees de captages d'eau. Karstologia, 4, pp. 38-44. ROBINSON, J., W & GALE, J. E. (1990) A laboratory and numerical investigation of solute transport in discontinuous fracture systems Ground Water 28, 1, pp. 25-36. SAHRAEI, P., H. (1995) Response of karstic aqu(f'ers to hydrological factors : A stochastic approach. M Sc Thes Shiraz University 180 p. SCHOELLER, H (1965) Hydrodynamique dans le karst (ecou1ement et emmagasinement) in Hydrogeo logie des roches fissurees, 1, IAHS CoiL Hydro!. des Roches Fissurees, Dubrovink (Yugoslavia), pp. 3-20. SINGH, K., P. (1969). Theoretical baseflow curv e s J Hydrol ., Div. ASCE. 95, HY6, pp. 2029-2048. SMART, C., C. (1998) Artificial tracer techniques for the determination of the structure of conduit aq uifers. Ground Water, 26, 4, pp. 445-453. SOULIOS, G. (1991) Contribution a !'etude des c our bes de rec ess i o n des sourc es karstiques, Ex e mpl e du pays H e ll e nique. J Hydrol., 127, pp 29-42. STOCKLIN, J & SETUDEHNIA, A (1977) Stratigraphic Lexicon o.f Iran Geological Survey of Iran, Report no. 18. WILLIAMS, P., W (1983) The role of subcutaneous zone in karst hydrology. J. Hydrol., 61, pp. 45-67

PAGE 114

Indirect Evaluation of the lzvarna Karst System Discharge Trend [Romania] Adrian IURKIEWICZ1 Mariana VOICA2 & Alexandru BULGAR3 ABSTRACT The importance of Izvarna karst system stems from the fact that its water is being tapped for the supply of the city of Craiova. The hydrodynamic behavior of the aquifer under natural conditions has been investigated based on records from 1957-1965. The development of the Cema-Motru-Tismana hydropower system has induced changes in the surface streams discharge regime and hence possible consequences on the karst aquifer supply. Compared to an accurate water budget evaluation, the correlative and spectral analysis of the discharge is a cost-effective method for the identification of the subsurface flow evolution trend induced by human activity. The paper illustrates the ability of the method to provide a quantitative evaluation of the water resource reduc tion as well as the limits of the adopted approach. Key words: correlative and spectral analysis, water resources decrease, Izvarna, Romania. Evaluation indirecte des tendances du systeme karstique de /zvarna (Roumanie) RESUME L'importance du systeme karstique de Izvarna resulte de /'utilisation d'une partie de son debit pour ['alimentation en eau potable de La ville de Craiova. Le fonctionnement du systeme karstique dans un regime non-influence a ete etudie sur La base de donnees de La periode de 1957 a 1965. w mise en place du systeme hydroenergetique de Cerna-Motru-Tismana a introduit des modifications dans le regime d'ecoulement des eaux de surface avec des consequences sur ['alimentation du systeme karstique. En general, pour l'identijication de ces consequences on utilise La methode du bilan. Par rapport a cette methode, les analyses correlatoire et spec trale du debit des sources representent une alternative peu couteuse capable de mettre en evidence une possible tendance des ecoulements souterrains resultant d'une activite humaine. L'application de cette derniere methode dans le cas du systeme karstique de Izvarna revele La diminution reelle des ressources dynamiques ainsi que les problemes qui en decoulent. Mots-cles: analyses correlatoire et spectrale, diminution des ressources, Izvarna, Roumanie. Long term planning of groundwater use is closely related to the recharge mechanisms and to the hydrodynamic behaviour of the aquifers. For karst systems, the recharge--discharge relationship un dergoes significant modifications as a result of the influence of some specific periods, i.e. dry or rainy years. The sensitivity of this input....,()utput relation ship increases with the degree of karstification. The situation becomes even more complex if an already intricate natural system is subject to changes induced by human activities. In this respect, it is important to point out that our possibility to con trol or to optimise the specific yield of the aquifers (in this case, for water supply purposes) depends on our capability to identify each component of the system. In the case of a binary system, part of 1 S.C. "Prospe(:tiulli" S.A, str. Ca$Sebe l,R..:.78344;.Bucpre,ti,Romania .. ... 2 UlliverSitatC8: Bucureti, Facultatea de Oe,ologie, str. TraianYuia 6, Bucu!'eti; Romania. 3 Institutul National de Meteorologic iliidrologie;Sos; Bucureti-Ploieti 97, R-71552, Bucuresti, Romania.

PAGE 115

114 which is supplied by sinking streams originating in nonkarstic areas, the regulation of the inflows could be an efficient solution if we try to maintain a reasonably constant rate of the output. On the contrary, the existence of uncontrolled but perma nent intakes from the inflows may contribute to a constant diniinution 'of the discharge, which in fact occurs as a long-term trend. An usual approach to the evaluation of such a trend is to estimate the recharge-discharge balance (hydrogeologic budget). Yet the accuracy of the results obtained by this approach is poor for large and complex systems. Also, the importance of costs is not negligible. Another procedure, based on the study of the pie zometric level evolution and/or of the permanent discharge time series, can lead to the identification of a possible over-exploitation or of a degradation of the resources (MANGIN & PULIDO-BOSCH 1991). To avoid a misinterpretation of the results due to the interdependency between the recharge of the system and the dynamic stored water vol ume, it is important to analyse the evolution of the climate of the considered area during the consid ered period. METHODOLOGY The method is similar to the time series analysis consisting of a descriptive investigation of dis charge values (MANGIN, 1994). The detection of the trend results from the analysis of the different components included within the series. The spectral analysis decomposes the series in the frequency domain, which often results in a more accurate identification of the components. If a trend exists, it will be detected when the spectra presents a higher value for low frequencies. The "observation window" chosen for a long-term analysis was 10 to 1250 days. The next step is to eliminate the components (random or periodic) in order to preserve the trend. To this purpose, it suffices to perform a linear transformation of the series with the filter of equally balanced mobile averages (BARBUT & FOURGEAUD, 1971). Finally, the decreasing evolution of the dynamic stored water volume results from the recession curve analysis, as indicated by MANGIN (1975). GENERAL FRAMEWORK The studied karst system is located in Valcan Mountains, that are included in the group of the western ranges of the Southern Carpathians. The A. lurkiewicz et al. whole group has a similar tectonic pattern, domi nated by the overthrust of the Getic Domain over . ; the crystalline and sedimentary formations of the Danubian Domain (POP, 1973). The limestones, the thickness of which may reach 1000 m, are of Dogger-Aptian age and are layered (the lower sequence) or massive (the upper se quence). Part of the limestones occupy a normal position between the .Liassic sandstones and the Upper Cretaceous flysch, while other part are thrusted over the flysch and occur as areally ex tended, yet relatively thin ( < 200 m) bodies (see Fig. 1). The surface stream-network supplies the carbonate reservoir through diffuse or concentrated swallets, which may be either partial or total losses. The inflows vary between 5-300 1/s, which accounts for 50-100% of the total flow rate of a stream. The mean runoff displays significant variations, be tween 12.5 and 45 l/slkm2 according to the season and to the considered elevation range. Another component of the aquifer recharge includes the rainfall on the limestone outcrops, which varies between 700 rnrnlyear (at 250-300 m elevation) and 1100 mrnlyear (at 1500 m elevation). Izvarna karst system ranges among the most constant large discharge springs in Romania. De tailed studies of the karst aquifer systems in Val can mountains (IURKIEWICZ, 1994) indicated that Izvama karst system exhibits a complex or ganisation, consisting of sub-systems with various degrees of interconnectiveness. Systemic analysis identified the relationships between sub-systems and provided an overall characterisation of each system behaviour. The system is developed within a low and elon gated limestone bar, dissected by many surface streams. There is morphologic and hydrologic evidence that an upper underground karst drainage level is overlying the main deep drainage. It has been also inferred that the karst system receives an additional underground supply provided by a fis sured aquifer occupying a granite body located to the north (RADULESCU 1985). The recession curves display a very steep initial, post rainfall decrease, which features the response of the shallow component of the discharge, i.e. the fast depletion of the unsaturated zone. Long de pletion periods follow, when the discharge, pro vided mainly by the deep component of the sys tem, slowly decreases. a, vdyn i and k parameters interpretation (MANGIN, 1975) is awkward, since depletion occurs in a rather non-karstic flow re gime: the computed a values are abnormally low

PAGE 116

lndirec. t evt#uation ()j the lzvarna karst system discharge trend us 'a.[J b.(J c:--d -e .c=J g.[J h.Q i.-+) V. Mare lake (0 0003 day-:1), while Vdyn is correspondingly large. Anyway, according to the i and k parameters the s ystem behaviour is resembling mor e that of a por ous aquifer. All the above information is based o n the flow rates and rainfall time series recorded between 1957-1965. After this period moSt of the dis charge was tapped for the water supply of the city of Craiova; so that at the beginning of 1970 the overall tapped yi eld amounted to 770-800 Vs. The works associated to the development of the Cema MotruTismana hydropower system have induced significantchanges in the regime ofboth surface stteairls and groundwater. RADULESCU et al. ( 1987) . had already pointed out .. that thes e works might affeCt th e dischar ge of the karst springs ; arid as a result the p ro j ec t of in cr easing f. 2:23 0 2 4 the yield tapped by the lzvama intake was recon sidered. (However, starting from 1992-1993 the building of the supplementary pipelin e has be e n initiated, without a prior updat e d evaluation of the karst system behaviour having been undertaken). DATA ACQUISITION After being tapp e d, most of th e disch arge of th e s yst e m could not be measured anymor e As a con sequence, the tlow rate time series recorded at Celei gauging station, on Orlea stream, has been used as a primary variable. The catchment area of the latter amounts to 61,8 km2 and includes two additional small basins (Sohodol..,-a dry valley that carries water only during rainy periods, and Pocruia). Besides th e surfac e discharg e th e flow

PAGE 117

116 rate gauged at Celei station (Q0 ) also includes the overflow from the main source Izvama, as well as the flow rate of other karst sources in its neigh bourhood: Qo = Qsc + Qp + Qs where: Q0 -Orlea valley flow rate; Qsc -karst springs flow rates; Q" -Pocruia valley flow rate; Qs -Sohodol valley flow rate. Because the average Qp is no more than 20-25% of the total discharge recorded at Celei and since the tapped yield is more or less constant, we can assume that the general evolution trend of Q0 parallels that of the karst springs flow rates On the other hand it has been shown that the rain fall input is much delayed by the system, which makes the evaluation of the trend due to human activity more difficult. Therefore we consider that it is more appropriate to use as a second variable, instead of the rainfall, the streamflow records in a nearby catchment basin, which benefits of similar geomorphologic and climatic features, without yet being influenced by human activity Verde valley gauged at Vaideei). The main fea tures of Orlea basin versus Verde basin are presented in Table I Table 1. Catchment basin Orlea !!!!! Verde Mean elevation (m) 573 1029 Area (km2 ) 61.8 79 ArealiiiiCitonc (km2 ) 19. 5 9.6 (km2) 396 Jim .. tone (km2 ) 27.5 Qmin (m3/s) 0.280 0.130 Qmax (m3/s) 21.6 22.7 Qmax/Qmin 77.142 174 Qmcan (m3/s) 1.408 1.888 Qrnean min. (m3 /S) 0.630 ('93) 0.370 ('93) Qmeon mu (m3/S) 2.24 ('79) 2.474 ('79) A. Jurkiewicz et al. For both variables (Q0 and the overall re cording period covers 17 years, namely 1978-1994. Since the flow rate tapped for the supply of the city of Craiova froin the main spring remained virtually constant (750-800 lis), it has not been considered for the data processing. RESULTS AND DISCUSSIO N S The first hint about the existence of a trend is given by the comparison of the hydraulicity (H), the ratio between the annual average discharge and the multiannual average discharge of each catch ment area for the two basins The diagram representing the ratio (Fig. 2) indi cates a three-steps evolution: during an initial pe riod (1978-1983) the Hsv values are smaller than H0 over the middle period ( 1983-1988) the two parameters are virtually identical, while more re cently (1988-1994) the H0 values become smaller than Hsv It is well known that the average values are influenced by momentary characteristics of the flow (flood or drought periods) and as a conse quence, the real evolution is frequently obscured. Nevertheless, the previously discussed observa tions also suggest to continue the analysis with more detailed methods. The spectra of two flows (Fig. 3 and 4) show the existence of an annual cycle with a peak in the frequency 0.02 (which corresponds to 357 days). Furthermore, it is important to notice the differ ence between the two maxima, i.e. S(f) = 27 for Verde Valley and S(f) = 14.7 for Orlea Valley respectively, due to the regularisation ef fect induced by Izvarna karst system. A second peak, present only in Fig. 4 and which is two times smaller than the first one, indicates an additional seasonal periodicity (179 days) displayed by Verde valley. For the scope of this paper, the most relevant are the spectra values in the origin. Thus, while for Verde valley S(O) is relatively small (5), the spectrum of Orlea valley begins with a large value (21 ,5 ), which according to MANGIN (1994) is indicative of the existence of a trend (Fig 4). The next step is to separate the trend out of th e entire time seri es, by using th e filter of equally balanced mobile averages with a 365 day ampli tude. According to the properties of this filter, the 365 days components and their submultiples ( 179 days) should fade out and the random component should dampen in the 1/365 ratio (Fig. 5 and 6).

PAGE 118

H ; ill : ze :;:: .. a. .g;. 25 24 23 22 .. 2D 18 17 18 15 14 13 12 II 10 8 'Hydra,ulicity variation 1.8 1,4 1.2 0.8 o.e 0.4 0.2 o ;anu l 1979 1981 S(f l. SUSiTA -VERDE .. 1983 -a-Orlea _,nu. 1987 Time (years) JJ o .2.9 2.7 2.5 2.3 -2.1 J : 9 i.} ; } 5 IJ. u .. .. . 9 117 1989 1991 1993 VERDE ; ... 0.7 f
PAGE 119

118 Qr(m3/d*10i 80 120 70 100 60 50 80 40 80 30 40 20 20 10 1983 1985 1987 1989 1991 1993 Time (years) Because of the high variability of the original se ries, the resultarit series are not completely line arised. Still some aspects can be mentioned: the presence on both charts of a newly out lined 5 years periodicity, due to weather fea tures. the decreasing trend of both series in the 19781983 range; after 1983, a decreasing trend associated only to the "Orlea" series; during the saine period the series maint:alns a cyclic evolu tion, withoutdroppii1g belOW a Virtually con stant value (1,1 m3/s); hence 1983 can be con sidered as the start of the degradation of the Orlea system dynamic resources. a steeper decline of the "Orlea" series after 1992: it might reflect the Izvama system "inertia" subsequently to the drought of that year and/or the filling of the Vija reservoir; the latter is part of the Cema-MotruTis mana hy dro-power system, to which it provides im portant flow rates, being on the other hand situated (Fig. 1) upstream with respect to the main swallet which supplies the karst aquifer ( Qins. = 300-400 l/s ). The analysis can be pushed even further, to pro vide a suggestion about the moment when the flow A. Jurkiewicz et aL rate tapped for the city of Craiova will be affected. A quantitative approximation cif the decay is given by the evolution of the dynarriic stored water vol ume (Vd) which according to MANGIN (1975) is the mobile water volume whose flowregulates the discharge of the different springs. The recession curve analysis performed starting from 1983 indicates a dramatic reduction of the dynamic stored water volume (Vd), from 1236m3 in 1984 to 9.166m3 in 1994 (Fig. 7). Although during 1993-1994 there is an apparent stability of the considered parameters, and even an increase of the flow rate recorded at the end of the recession period (Q,), the analysis of the cyclic (five years periodicity) behavior indicates that at the end of the 1993-1998 cycle (Fig. 6) the flow rates of the karst springs might drop below the yieldrequired for the supply of the city of Craiova (800 1/s). CONCLUSIONS Despite the complexity of the lzvama karst sys tem, the adopted approach outlined a definitely decreasing trend of the overall discharge, by using already existing, inexpensive data sets. The evaluation of the dynamic stored water volume vd has indicated a. dramatic reduction (about I 0 times) of the water reserves and a possible re equUibration trend, which however might be fol lowed by a decline of the discharge below the requirements of the water intake. In order to identify the future management issues related to discharge optimization, it is urgent to initiate anew research project for the investig!ltion of Izvama karst system. The necessary primary data have to be provided by an adequate network, to include flow rate gauging stations at the springs and the swallets, as well as meteorological sta tions. The objective of this new project has to be unambiguously stated, that is, to devise a model that will explicitly consider the changes that hu man activities have induced in the behavior of the aquifer. This model would ultimately lead to the identification of the optimal solutions, able to rec oncile both the interests of the proj ect and those of the city of Craiova water supply system.

PAGE 120

Indirect evalUIItion of the Izvarna karst system discharge trend 119 REFERENCES BARBUT, M. & FOURGEAUD, C. (1971) Elements d'analyse mathematique des chroniques. Ha chette Universite, Paris, 207 p. JURKIEWICZ, A., (1994) Etude hydrodynamique du karst du versant sud des monts Vdlcan. Memoire DEA, Universitatea TehnicA de Construc\ii, Bu89 p. MANGIN, A. (1975) Contribution a l'etude hydrody namique des aquiferes karstiques. These Doct. Sci. Nat. Dijon, in Ann. Speteol., 29, 3, pp. 283-332,29,4,pp.495-601,30, J,pp.21-124. MANGIN, A. (1994) The karstic milieu, the karstic aquifer, classification of karst. In Groundwater Ecology (1. GIBERT, D. DANIELOPOL & J. STANFORD Eds.), Academic Press, Orlando Florida. MANGIN, A. & PULIDO-BOSCH, A. (1991) Identifi cation of the overexploitation of karstic aquifers using the statistical treatment of time series (theory and applications). XXIIII.itH. Congress "Aquifer overexploitation", Canary Islands, Spain, pp. 65-72. POP G. (1973) Depozitele mezozoice din MunJii Viican. Ed. Academiei, 155 p. RADULESCU, D (1985) Studiu hidrogeologic privind injluenta amenajarilor hidrotehnice din bazine lor rti.urilor Cerna-Motru-Tismana si Jiu asupra izvoarelor Izvarna dezvoltarea alimentarii c:u apa a municipiului Craiova Arhiva ISLGC, BuR.ADULESCU, D., STANESCU, I GASPAR, E & BULGAR, A. (1987) Aquiferous interconnex ions in the Motru-Izvama-Tismana-Bislrita karst area. Theor Appl. Karstology, 3, pp.199-214.

PAGE 122

ABSTRACT Theoreticaland Applied Karstology, vol. 911996, pp. 121-140 Karst Systems in Banat Mountains [Re,ila-Nera zone, Romania] Adrian IUnKIEWICZ, Gigi DRAGOMIR, Au rei ROT ARU1 & Bogdan BADESCU2 The Banat Mountains area, located in the south-west of Romania includes more than 800 km2 of karst zones be longing to the Reita Moldova Noull. large synclinorium. Although scientific karst investigations were initiated at the beginning of the century, the previously collected hydrogeological data were rather local and contradic tory Taking into account groundwater resources that karst areas should normally contain, "Prospectiuni" com pany has recently started a complex research program, in order to evaluate the amount of groundwater resources of this particular area The karst territory was divided as a result into three distinct zones, which correspond roughly to the Cara, Mini-Nera and Nera-Danube catchment areas. The hydrologic, hydrodynamic and hydrochemical characteris tics of the karst systems included within the .first two zones are discussed. The water budget completed based on 1992-1995 data provides information on underground hydrologic connections between small catchment basins, and eventually for each of the above mentioned areas as a whole. Key words: karst systems, regional karst studies, Banat, Romania. Systemes karstiques dans les Monts du Banat (zone de Re$ita-Nera, Romania) RESUME La region des monts du Banat, situe au sud-ouest de Ia Roumanie comprend une zone calcaire qui 'etend sur presque 800 km2 Bien que cette zone soit Ia plus large de Roumanie, les informations sur Ia hydrogeologie de La zone ont eu un caractere local et paifois contradictoire. C'est pour cela qu'en avec 1992 La societe "Pro:rpecfiunr a demarre un programme complexe de recherche en vue d'etablir les reserves en eaux potables de cette zone. Par consequent, La suJface du territoire karstique a ete partagee en trois zones. qui correspondent plus ou moins aux bassins des rivieres du Caraf, de Minif-Nera et de Nera-Danube. Pour les systemes karstiques ident(fies dans les deux premieres zones, on presente les carac:teristiques hydrologiques hydrodynamiques et hydmc:himiques. La encore, grace au hilan hydrologique qui a ete dresse sur Ia base des donnees de Ia periode de 1992 u 1995, ont ete etablies les relations souterraines entre les petits, etfinalement entre les grands hydrologi ques. Mots cles: systemes karstiques, etudes regionales, Banat, Roumanie. 1. s.c. "Prospeepuni" s.A.,str. t,R-78344, Buclifeti, Romania. : 2. A. S. "Exploratorii" str. Lalelelor 7, R,..;t700 Reita, Romania.

PAGE 123

122 A. et al. 1. INTRODUCTION southem part,withNei:a valley as the bounlt:is worth Major contributions to the investigation of the . 110ttcing, for ii' proper oataric computation; that karst geomorphology in this area are due to the weather chaiacteristics of'the eastern slope 'of SENCU (1970; 1978; 1986). His activity covered, ihe ai:e StronglY. different as : compared to with small gaps, more than 30 years. In order to those from the western side, due to the Mediterraidentify the reciprocal relationships that exist neiui influences. . . tween karst aquifers and the coal deposits from Anina area, many dye tracing experiments have ; 'GEOLOGY AND STRUCTURE been carried out by SENCU (1973) and SENCU & ClOACA (1980). For the evaluation of the karst water potential of Lhe Moldova Nouli syncli.Oorium starting with 1992 "Prospecpuni" $.A. has beguri a .;oln plex program that includes dye tracing drometric records, h.ydrocheinical.: analyses, etc. ,. (IURKIEWICZ & DRAGOMIR, IURKIE WICZ & 1996); the main results being summarized in the following. 2. NOUA SYNCLINORIUM --GENERAL PRESENTATION The large Noul!. synclinorium extends over two physiographic units, Aninei Mountains, that occupy its northen1 and central section, and Locvei Mountains, lying .in its As al!eady mentioned, this : zone belongs to the structural unitofthe Getic Nappe. The deposits are folded as a large synclinorium With southward plunge. The first teetonic synthesis on the ReHa-Moldova Nolll __ Zone been completed :by Schreter in 1912;' its main ideasate still valid imd they have been additionally developed by the researchers which have been studying this area (RAILEANU et al., 1957, MUTIHAC 1959, NASTASEANU, 1964, STiLLA er a/., 1972 etc.) .. In fact, ihe major structural elements established by Schieter ..arc figured on the maps at the 1:50 000 scale com pleted during the last decades by the Geological Institute of Romania. The pQSition of this unit .within the general struc tUJ:e of Southern Carpathians has been extensively discussed {BUCUR, 1991). A relatiVely

PAGE 124

Karst system s in Banat Mountains correlation of the alpine structures on both sides of the Danube (Nfi..STASEANU & MAKSIMOVIC 1983) led to the separation of the following (from east to west): the Getic Unit, the Sasca Gornjak Nappe, the Refifa Nappe, the Dognecea Luznica Nappe and the Nappe (Fig. 1 ). The present-day structural constitution is the result of the Hercynic and the Alpine tectoniC cycles. The beginning of the Alpine movements is related to. the Liassic transgression, while evidence con cerning the various phases of this cycle of move ments is provided by the Early Jurassic and the Albian-Cenomanian transgressions. As a result of these movements, the sedimentary deposits were imposed a very complicated structure, characteris tic of which are the dislocation lines with thrust planes and the inclined folds, or those displaying significant attenuation of one limb. In the synthesis on the geology of the Moldova Noua area, NASTASEANU (1964) out lines, from west towards east, the following folds displaying a jurassian style: the Syncline, the Natra Anticline, the Valea Do manuluiValea Jitinului Syncline, the Cornetul Mare-Ilidia (Valea Aninei) Syncline, the Polom Anticline, the Bradet Syncline, the Anina Anti cline, the Central Syncline ( Colonovi1I). the Beu Sec Anticline, the Pitulaii Syncline, the Anticline and the Eastern Syncline. Among them, the largest extension have the Bradet arid Central synclines, the Anina Anticline and the Eastern Syncline, whose central infilling consists of Al bian sandstones. Many longitudinal and transverse faults, associ ated in some cases to important strike-slip faults, dissect this extremely folded geologic complex The main regional longitudinal faults are Polom, Terezia, Valea Aninei, Prolaz, Sodol, Ranchina, Beu Sec and Geologic investigations performed between 19701985, mainly by S. Nastaseanu, assisted by several other scientists associated to the Geological Institute of Romania, resulted into several sheets of the geologic map of Romania, at the 1 :50 000 scale, which cover this area. The sedimentary succession covers the entire Late interval, with still several hiatuses being recorded . Within the lower half of the interval, mostly non carbonate deposits occur, including conglomer ates and sandstones of various fadeses; In addi tion, clays, s}lales, coal sheds and coal layers are assodat. ed to this Carboniferous Permian and Liassic_..;Earl y Callovian age series. 123 The transition fades occurs in the form of an es sentially marly formation, which includes marly sandstones, compaCt marls, mariy lim estones and carbonate sandstones. A carbonate series of Late Jurassic-Cretaceous age follows, which includes, chronologically, Gumpina limestones, marls, Valea Aninei limestones, Bradei limestones, Marila limestones and Crivina marls, (lower and uppev). This carbonate series displays a typical pelagic facies. It is followed by middle shelf and reef limestones which corresponds to tile major transgression beginning with the Barre mian: Plopa and Valea Miniu{ui limestones The carbonate sequence lasts Uiltiltbe end of Bedoul ian (Lower Aptian) when conditions are recorded. During the Albian sedimentation

PAGE 125

124 resumes the deposition of a detriuc formation (Golumbu Sandstone) . mebtioned carbonate series, dis tinct lithologies (micritic and biomicritic, organic or reef limestones), and accordingly distinct ages can be separated. Although up to 1200 m thick, the carbonate stack is relatively homogeneous in what concerns its water transfer and storage properties. The indicated tectonic and geologic factors con tributed also to the current configuration of the water accumulations in the carbonate deposits. The complex role of the main fractures and of the ac companying second-order faults is outlined by their coincidence with the lineaments of swallets and springs through which the karst aquifer dis charges. Besides improving drainage characteris tics, the fractures also separate distinct aquifer compartments, and as a consequence they act as boundaries between the main karst areas and sys tems (Fig. 2). 2.2. PREMISES OF KARSTIFICATION PERMEABLE ROCKS The pervious rocks complex begins with Middle Jurassic deposits Gumpina Limestones. These are followed by the complex of the so-called Tlimaa Marls which are, in fact, marls and marly limestones, which, although less-permeable, do not form an impervious screen and may be consid ered as 'pervious'. These deposits support the limestones of Valea Aninei (Upper Oxfordian Lower Kimmeridgian), followed by the limestones of Brlidet (Upper Kimmeridgian-Lower Tithonian) and Marila (Upper Tithonian Berriasian). After a marly sequence belonging to the Valanginian, the carbonate series continues with middle shelf and reef limestones of the Bar remian Lower Aptian interval (Plopa and Valea Miniului limestones). Among them, the Plopa Limestones are the most pure (94.08-99.07% cal cite and 3.86-4.92% dolomite). In the same category might be included the ther mal and chemical contact rocks (skarns and horn fels) formed at the contact with 'banatites' intru sions. The original permeable properties of these deposits have been improved on account of the intrusions, that have induced a secondary perme ability, finally resulting a karst-type water circula tion. This type of circulation has been recorded at the -50 m level of the Ciclova Montanli mine. Thermal waters with a temperature of 29.3C and pressure up to 3.5 bar have been intercepted in several sites at this horizon, either in boreholes or in the galleries 1972). A. Jurkiewicz et al. DEPOSITS"WITH AMBIGUOUS CHARACTER (PERVIOUs/IMPERVIOUS) . . . It seems that the two previously mentioned marly limestone formations (Tmt"laa and Crivina marls) may restrict or not the hydrogeologic continuity, depending on the component which prevails (i.e. marl or limestone) and/or on the deposit thickness. The Crivina Marls formation (Valanginian) for example has a prevalently marly character only in the stratotype zone, on the left side of Mini val ley, downstream of Crivina forestry hut. Another formation whose hydrogeological char acter is a function of its thickness is the Golumbu Sandstone. Thus, in some zones of Poiana Rochii, where the thickness of this deposit does not exceed 5-10 m, numerous sinkholes, swallets and even small potholes have been identified. They are ini tially developed in sandstone, subsequently con tinuing into the limestones beneath. The process seems to be, however, a pure mechanical one. On the other hand, within the same area, on Golumbu Valley (stratotype), where the formation thickness ranges between 50 and 100 m, the sandstones have an obviously impermeable character, carrying sub aerial streamcourses. STRUCTURAL-TECTONIC CONDITIONS The faults and fissures whose main direction is NNE-SSW impose the main drainage direction. The latter is also controlled by the analogous ori entation of the folds (synclines or anticlines). Be sides improving the drainage properties, the frac tures separate the aquifers, acting as boundaries between the main karst systems and/or zones. The role played by the main fractures and the associated fissure systems is outlined by their coin cidence with the springs lineaments by which the karst aquifer The support fissures of the regional faults may be involved in either the recharge or the discharge of the karst aquifer sys tems, hence water circulation may be directed in one sense or another. 2.3. KARST MORPHOLOGY The karst morphology displays the entire range of surface and underground features, for which dif ferent evolution stages are recorded. In this respect, it -is worth mentioning the abundance of the sinkhole valleys, as well as that of the sinkhole plateaus. Thus, the typical evolution: active valley-swallets-dry valley-sinkhole valley is complete

PAGE 126

Karst systems in Ba1141 Mountains 125 2 L..: : l 3 5 ,.....-6 7 c::> 8 9 to 11 12 n 13 v 1 15 15 @ ------------'. carbonate stream; to. swai-15. spring 'rii'Cftl TO permanelit; 9. d'ciyep; 14. 'iaUe-

PAGE 127

126 The most spectacular exokarst feafure of Banat Mountains are the gorges. These are usually de veloped transversally with respect to the general structure, with steep slopes, 150-200 m high. The most important gorges are Caraului ( 19 km), Garlitei (9 km), Miniului (14 km) and Nerei (22 km). The local base-level evolution is outlined by the several cave systems floors (elevation range: 50-I 00 m) within the steep slopes of the gorges. Finally, a special mention should be made for the associated continental carbonate deposits i.e. cal careous tufa and travertine. Although there are currently no extensive studies concerning these formations, there are wide occurrences within the entire Banat karst, which suggest that the associ ated springs are over-saturated with respect to calcite. CAVES AND POTHOLES Although the discussed karst area is the largest of the country, the currently known caves are of moderate size (ORGHIDAN et al., 1984). An ap proximate inventory of the caves explored and surveyed by the speleologists associated with the Speleological Institute "Emil Racovitli" and the clubs in Reita1 Anina, Ora vita, Timioara and Bucharest displays the following situation : 12 caves with lengths over 1000 m, among which 3 longer than 5000 m; 11 caves with lengths between 500 and 1000m; 10 caves with lengths between 300 and 500 m, and hundreds of caves under 300m in length. The usual depth of the numerous potholes which may be encountered on the karst plateaus ranges between 30-50 m and only rarely exceeds-100m. Major caves are Buhui (7282 m in length) and Comarnic (6203 m) -both through-trips and the pothole Poiana Gropii (-236 m in depth, 1029 min length). When considering this overall context, areas like the one situated at the confluence of the streams Comarnic and Cara (Comarnic Cave: 6203 m; Exploratorii Cave: 5172 m) or that of Valea Mare catchment basin (Poiana Gropii pothole -236 m; Cioaca Mare pothole -137m) might be rather 1 Most of the data concerning the caves from the northern and central parts, including the precise geographic location (using theodolite) of the main caves from basin area, are due to the "Exploratorii" Caving Club in A. Jurkiewicz et al. considered as exceptions. They are, in fact, the results of the simultaneous occurrence of karst-favoring factors (lithological, tectonical, hydrological). The genetic types of cavities which may be distin guished in Banat frequently depends on the posi tion of the limestone with respect to the impervi ous basement. In this respect, Valea Aninei (J3ox2 -J3km1) and Plopa (K1br) limestones are strongly favoured as they correspond to important trans gressive episodes and usually extend more than any other carbonate formation and lie directly over the impervious basin margins. The cavities result ing from such an allochtonous water supply are typically sub-horizontal stream caves, often offer ing complete traversesalong the main stream, with large-size passages along the stream-way and fos sil, abundantly decorated, upper floors ( Comarnic cave excavated in the Valea Aninei limestones, Buhui cave-carved within the Plopa limestones). Cavities associated to an autochthonous (i.e. karst) water supply may be recorded either within the recharge areas where the flow is predomi nantly vertical (potholes, steep caves) or within the discharge (active or inactive) zones, where quasi horizontal caves may be found. Among the major caves belonging to the first type it is worth men tioning the pothole Ponorul de la Stana lui Jvacu, with a depth of -101 m for a horizontal extension of 300 m only, and Avenul Raurilor Suspendate (-70 m deep), both from Poiana Rochii area. For the second type, the typical example is pro vided by Petera de la Captare (near the Valea Miniului holiday-camp), with a total length of 961 m. Its 50 m depth extends between the en trance and the stream passage. This is a good ex ample to illustrate the changes induced by the fluctuations of the local base level. During a first stage, the outflow occurred through the present day entrance (located 60 m above the present-day spring). Following the deepening of the local base level, the drainage ran along a lower-positioned floor, so that the outflow currently uses a secon dary fault, forming a Vaucluse-type spring. 2.4. HYDROGEOLOGY AND KARST SYSTEMS CATCHMENT AREAS The recharge of karst springs (or karst systems) originates either directly in rainfall or in swallets or water-loss zones. The last two situations often occur at the contact impervious/pervious formations, or within the still-active sections of the valleys cross-

PAGE 128

Karst systems in Banat Mountains ing carbonate rocks. Classical concentrated swallets, such as that of the Golumbu sub-aerial stream, allow the transfer of some 50-100 Vs, that may increase to 200-250 Vs under flood conditions. The swallets associated with cave traverses (upstream cave entrances) receive average flow rates of 50-60 Vs, while the rest of the swallets rarely exceed 8-10 Vs, most of them receiving only 1-3 Vs. As we have already mentioned, an outstanding role in the constant recharge of the karst springs have the surfaces covered by imper vious rocks (crystalline schists and igneous rocks). Most of the karst networks or important springs are related to such recharge surfaces, the associ ated karst systems being of a binary-type (e.g. the karst system Certeje-Buhui, that of the Cara Spring, of Comamic cave, etc.). The Golumbu Sandstone also favours the devel opment of some perennial streams which recharge many karst aquifers, such as those discharging through the springs Ducinu, Moceri, Lllpunic and even one of the major karst systems of Anina Mountains: Liciovacea-Bigh. Furthermore, the water courses formed into the Anina anticline crest continuously recharge both the Ponor-Plopa Sys tem and the Irma Spring. In many cases, large areas within the karst pla teaus (Brlldet) are covered by Pleistocene residual deposits (usually terra-rosa-like clays with blocks), which persistently supply a low, yet constant rate for the springs recharge. One may conclude as a consequence that the ma jority of large karst systems (including or not ac cessible caves) are directly conditioned by a sup ply originating in non-karst surfaces, one of the noticeable exceptions being Tolosu cave. On the other hand, it is also true that the evolution of the karst systems supplied by perennial streams is extremely rapid, transforming them into real drains with continuously decreasing reserves. SUBTERRANEAN DRAINAGES The 43 tracing experiments (35 of them presented in Table 1), performed mainly (25) by V. Sencu, outlined karst systems which are concentrated in three areas, belonging to the catchment basin of and to the Mini-Nera and Nera-Danube watersheds. Most of the subterranean connections between the swallets and the discharge areas (karst springs) have been proven by tracer tests (fluorescein). Sometimes, due to the lack of water courses SENCU (1973, 1980) used as dye-solvent water transported in tanks (10-20 m3). In other cases 127 (Anina zone), the connection has been proved by the coincidence between drilling fluid losses and a delayed pollution recorded at a neighb.oring spring. One may notice 'that the fictitious transit velocities of the water strongly vary between 3-10 mlh for the zones of and Liciovacea-Bigar, and 150-180 mlh in the case of the streamways run ning through major caves. Besides the karst systems whose configuration has been proved by tracer tests there are still many other systems whose recharge area is only in ferred. CATCHMENT BASIN ZONE Within this zone, six areas with well organized karst structures were identified, namely: Caraova, Sodol-Baciului, Certeje-Buhui, Izbucul Caraului, Brlldet Plateau, Ponicova-Comamic (Fig. 3). The underground drainage paths from the recharge toward the discharge areas are either divergent or convergent, and the fictitious tracer transit veloci ties range from 3.3 to 180 mlh. The location of the springs is controlled by three factors, which acted either independently, or inter connected with each other: a tectonic fracture; the carbonate-non-carbonate boundary; the local erosional base. The most important springs in this area are Izbucul CaraUlui (Cara Spring) and the springs in Gorge (between Comamic and Prolaz). The flow rates of the latter ar estimated based on the significant difference recorded between the average discharge values recorded at two Carq stream gauging stations, situated upstream of Co marnic and at Prolaz respectively (Table 2). A similar situation is that of Garlite stream, where the total springs inflow was estimated based on the difference between the values recorded upstream and respectively downstream of the gorge section The fact that the flow rates of these springs have been measured in different periods induces a degree of uncertainty for a comparative analysis. On the other hand, the intense tectonic and karst procunderwent by the reservoir rocks resulted in high variability indexes of the karst springs. Therefore, the characteristic flow rates and the variability indexes for the considered time inter. vals have been synthesized, as a first estimate, in Table 2. The minimal flow rates generally range, for all springs, between 0.001-0.030 m3/s_," while the river sections Garlite (downstream) and Cara (at all measurements points) exhibit minimal flow

PAGE 129

'fl!_ -----, . s.QI...-.BAC.IULUI .... 0 i "4' 0 g 1' f qo J D .. 0 o PONICOVACOMARNIC 0 500 1000m CARASOVA-PROLAZ o soo mo .. 3 -----L 5 C) 6 7 6 9-10 11 n 12 v 13 1L 15 16 17 CERTEJE-BUHUI IZ.BlXll CARASULUI .. A. f.urkiewicz et aL .. ;;_, ... .. :-.\" ... ;_: ; 1-,

PAGE 130

',' : .:.:.-. 1 .. PP .. . . ac: a. .. : c ... : . :: .. .. ;: .... t9:07: 1992 : L5 .. : captil(. : ... :r 2.5 2 i1 . ll.os;t99i oi:. ... : t.5 3 :-: .:,>:::; 2L07.1992 0.1 ... ,... . : 3 4 PetComarQic .:: 9 9 .-: ::;:.. 1 : . :.: .> i 6 Pat!tiCom.arilic ., ; -80 40 7 : 21 .... i .. : .. :.:. 34 8 .. .: .... 18.08.1967 .: : .'. : 30 9 18.1 i.I958 25 25 10 Parauizvarnita 1 1 Pariu cer.teJe: ::. 22 07.1977 I 06.04.1968 8 Buhui .. ': ... 8 CerieJe :: :.; ; :_. . . 12 lzvor Schlucht .: ,o 5 Izvor treiia: ... .... .. ,:. 12 : .. 250 -.... -:: 12 SondaJ Brlidet ; ' ; : 13 Soil(la .:; : ... 14 (:an toil Camealli '; -. 18 V alea Morii . . : . 19 Ponor Uteri . :. . .. 28.07 1976 20 06.1968 09.04.1958 15 30 7 .... .. Izv. ... . 250 Iiv : b1!;iuiui. 2so > : . ,: J4 Ffutr cti. Fra5inF. ... ; 5 Izbucul rrnla 10 . . . ' u,.; : 20 Pil.rauPonor >'. .. -.:::: '' ;_(>;: 15.08.1967 10 Petera Plopa 10 21 Sorida DLCovliciei 22 1 03.08.1975 23 r:ioi. Livada Mare. 10.07 1979 24 Dol. Poiana 16.07 1979 Paraui Alb 18.07.1978 . 26 Ponor LiCiovacea 09 07 .1978 27 :, .. 04.08.1975 28 P Scocu 12.06.1976 29 Cuce :: 3 1.01.197 6 30 La Albii 18. 07.1978 31 Pori or v. .Oolu,Jilbu 32 Og; .. din P Scocl,liui. . 10.10.1994 1965 09.06.95 .. .. : =: 7 . Iz\ioru Mare Izvoru Mare. 0 5 Izbucul Big11r. 0.5 Izbuchl Bjgh. 2.5 3.0 Izbucul Bigh 4.0 lzbucul '. 0.5 Izv. Upunic 65 lzbucul Biglir .' 2.0 IzvorLiipunic 1 . . 3 .. 300 300 300 300 300 300 300 50 50 290 129 DiSt. T (h) Vel. (kt:Jl) (m/h) 145 1.8 70 0.87 75 0 5 20 1.8 95 0.7 26 0.6 36 0.7 50 1.5 10 2.5 40 15 180 30 107 107 132 120 30 0 .8 0.55 1.7 0.7 0.8 0 6 1.3 1.3 3 100 1.2 30 0.7 50 0.8 10 0 2 90 2 2 80 2.4 310 4.0 300 3 5 340 5.0 350 6 0 390 7.0 90 0.4 90 76 405 3.6 1.85 2.25 192 9.4 264 3.3 144 3 5 12 150 27 28 20 30 17 42 241 6 18 139 10 3 62 17 72 44 12 51 16 2 5 50 58 205 175 370 154 432 16 384 280 26 80 183 12 35 18 29 40 23 43 80 40 41 20 20 14 39 16 25 93 6 5 86 5 33 . 34 Av: din l'oiana.Rochil 35 : 19.06.1995 0.5 Izvor Lapunic l 270 405 2 91 21 1965 Izvor Poniasca -335 2 2 105 21 water brought in tanks.

PAGE 131

130 Hydrometric Station Period lis Q lis . min a: vol.. . ME (xHl5Jn3 ) TF A: et al. RT M Lazarovlit .. 1992-93 72.5 Q4 3 6 .. 1 20.13 19 0.01 64 i635 0.011 0.713 32 29 4 27 29 10 12 28 12 29 29 -29 0.136 .. .. o.io8 28 B . 19 : B Jitin 1 Jitin 2 1976 1976 1320 22 162 Cara spring 1992-93 599 Cara spring _. 1976 1170 Certeje 1976 100 6 13 30 15 60 46 .39 6.6 64.6 nr 29.7 0.001 0.017 0.008 0.008 21 Los 1.62 3.26 0.288 5 16 0.124 29 0.144 8 0.168 6 A B B A A 475 5 95 66.4 ... 0.018 L77 .. 22 B 0.160 8 A 0.12 23 Schlucht 1992-93 398 6 66.3 45.02 0 0055 1.76 Cara (Jerviuti) 1992-93 3266 59 55.3 404.5 13.9 Cara (Comarnic) 1992-93 4620 70 Cara (Pro1az} 7546 165 (Caraova) (Sch1ucht) 1992-93 16300 155 1992-93 910 120 66 577 ,. : .. 45;7 1121 105 1621 7.6 < 270. 1992-93 5970 362 16. 5 1041 0.013 13.1 50.9 28.8 0,0048 114 -26 26 0.12 27 0.12 23 0.192 25 0 .152 18 B B (F) ME = memory effect; TF = truncation frequency; RT =regulation time; M = model. rates exceeding A cai:etui analysis of fast of resources. The other catethe minimal fl<)w rates, characteriStic for. the gory;, including mainly the discharge from the drought period, highlights -for the gorge sections, displays low u values, ranging the conti'ibutiori of the springs from the section .between 0.002 and O.OOS day1 A special deserves (95 lis )as well as the lack the spring Jitin 1, for which the depletion coefTiadditional contribution along the section Prolaz:.... cient is 0.001, i ,e. indicative of a non:kaist beCaraoVa . At the same time, itls wo.rth haviOur or at Jeasi for the existence of a compoing the diffuse recheen continuously monitored over one year, indicates two major classes of springs (Table 2, Fig. 4).

PAGE 132

Karst in Banat Mountains 131 0,6 0,6 0,6 Of> 0,4 CD 0.4 0 o. 0 0,4 O,L. 0,2 0,2 0,2 0.2 o l 25 50 0 24 0,6 0,6 0,6 0,6 0.6 0,4 (7) Of.t @ OA 0,2 0,2 0.2 F ig. 4. Uidt step response funt:tiob of Jw:st systemS in tile Cara baslil zone. Rejumsf/.impulsio,nelle .. > .. :_ ': . . . :. -. ''--:: : .; . 1. 2. Jltin 1; 4; J ltlil2; 6. CarJi (1976); Certeje sprblgs; 8. Comarmc cave; 9. spring; 10 Carll (Jervant) : streain#)ow; 14. GArlite (Schluchh ; The first class includes systems more or less similar to the Aliou type, namely systems with small reserves, which only slightly modulate the rainfall information and have poor regu lating capacity. The typical example could be considered the rapid component of the discharge of Jilin l spring which conceals the real behaviour of the system (the spectral analysis reveals that the first truncation frequency could indicate the existence of a slower com ponent of the discharge). In the same class are also included the systems Izvorul (during the 1976 period}, Certeje, Schlucht (at least the fast component), especially if the regulating period is considered, because on the other hand, their memory effect is large, as a consequence of the supply from the surface streams. It has to be however pointed out that for short time series ( < 1 y ear), the regulating period appears to be a much more stable criterion as compared to the truncation frequency. The second class includes systems the hydrody namic behaviour of which is similar to the Baget type, namely systems with average size re serves and with enhanced regulating capacity spring and Jitin 2 springs for in stance). The analysis of the 1992-1993 behavior of Izvorul Caraului might suggest that the latter also he longs to this second class; yet, according to the above mentioned criteria (short time series record relatively long freezing period}, we think that such an ascription would b e erroneous. In what concerns the spring Jilin 1 its first truncation fre quency is characteristic for an inertial system Nonetheless, there are several karst systems that display a rather small first truncation fr e quency close to that identified for the Fontestorbes type Besides the previously mentioned springs, two stream sections, (dowstream) and (at Jervani}, display characteristics similar to those of the Bagel type. The memory effect of all the systems included within thi s class is e ven larger than that displayed hy the Baget type, nam ely 24-29 days, although their regulating periods are sometimes shorter (16-23 days). The systems inducted in the first class display a more or less sharp unit step response function. whil e those in the second class ex.hihit a signiti cant spread out (Fig 4).

PAGE 133

132 The analysis of all the considered parameters leads to the following general conclusions concerning this area as a whole: the fault systems striking NE-SW and the associated parallel, deep valleys resulted in a tectonic and topographic dissection of the car bonate aquifers, and as a result the installation of major karst systems was prevented; in addition, most of the surface streams network is totally dis-organized, the small brooks formed on the impervious substratum sinking just at the contact with the limestones; the degree of organization of the karst cavities decreases with the depth, leading to a local lack of homogeneity of the storage and trans missive properties (the karst cavities reached a highly organized stage, which resulted in very good transmissive properties, but poor storage properties); for the unitary (only carbonate rocks included) karst, the lack of distinct sinking points results in systems with large regulation capacity and constant transmissive and storage properties. In the case of the Bradet plateau, the presence of the Pleistocene residual clays appears to have a major influence on the performance of an aquifer with non-karst behavior, that se cures constant flow rates over long drought periods, both to the Jitin 1 spring and to the valley; the binary karst systems (that include surface supply from impervious areas too), are rela tively well organized, with good transmissivity and low active storage capacity. The water balance performed over the 1992-1993 period, first for small catchment areas (tributaries and stream sections), then for the entire northern section, resulted from a co-operation with INMH In order to compute the input, rainfall records from the stations and Anina have been used, which allowed a subsequent evaluation of the rainfall distribution according to elevation ranges and catchment areas surfaces. The runoff regime has been derived from stream level records (performed continuously or twice a day) at 18 gauging stations Evapotranspiration, computed as a first estimate (A) according to the Turc formula, amounts to a total value of 475 mrnlyear (13.5%), which results in a coefficient of 15.9 l/slkm2 and 1.1 m3/s water budget deficit. A A. Jurkiewicz. et al. second estimate of the water balance (B) considers equal inputs and outputs (zero infiltration) for Jervani catchment area (VII), resulting an 18.781/slkrn2 evapotranspiration coefficient and a deficit of only 281.5 1/s (3.38%). Both water balance estimates (Table 3) suggests similar directions of the water exchanges between catchment sub-areas Hydrometric mesurements performed during drought periods also substantiate this flow pattern, which can be summarized as follows: groundwater transit from Nermed catchment basin (II) toward Doman (I) and (12) catchment basins; the hydrogeologic catchment basin which cor responds to Valea Mare river is smaller than the surface catchment basin of the same val ley; hence it appears reasonable to suppose a subterranean flow from the Sohodol area to the X catchment basin gorges between Comarnic and Prolaz); downstream of Prolaz, the surface watershed is situated much closer to the valley of than to that of important diffuse and concentrated discharges in the median catchment basin of valley (IV) suggest that the hydrogeologic boundaries of the latter are nearer to Buhui, and Jitin valleys than its surface water sheds. Still taking into account the measurement and esti mation errors of the considered water budget ele ments, the resultant total deficit value according to the A estimate (1.1 m3/s, corresponding to 13.5%) presumably indicates a water transfer, although not very significant, from the northern toward the southern zone, occurring mainly at the southern boundary of the first. (Note: the surface watershed between and catchment basins is conventionally outlined across the karst plateau of Anina). MINI-NERA ZONE The investigations performed in this zone did not indicate significant differences in what concerns the setting and organization of the karst systems Based on hydrometric measurements performed over different hydrologic cycles, the hydrody namic parameters of nine such karst systems were obtained.

PAGE 134

!((Irs( systeiii;S in BaiUlt .Mountains . . , ... . ,. .. Doman II Nermed III Ciudanovila IV v Garlite-Schlucht VI VII VIII Top lila IX Comarnic cave X Cara gorges XI Valea Mare 439.86 .. 452.657 532.606 1336.92 282.391 747.313 933.878 387.927 196.808 2127.716 874.703 201 57.836 140 768 273 119.27 404.5 92.339 66.431 130 Cara ova (12) VI+VII+VIII+IX+X 4393.642 1621 The analyses comp leted in .the sanie way as for the previous zone. For most .. springs, the mini mal flow rate values gerieralJy tange between 0.0035 and 0,040 m3/s. The exceptions are serited by 1zvorul de lei. valea caidlifil arid the spting OchJu Beului, with minimal flow rates of 0.064 and 0.156 m3/s, respectivdy. The variability indexes itre included into a ally single group (Q-100), which encompasses all the springs. Table 4 that r:nost systems display a karst-type<;>r, with a exceeding o.oi (rapid drainage of a part of the water resel'Ve).Jn this group inay be inchided the springs. Bigar (Fig. 5), (F:ig. 6), Moceri and Bei4riiJa . Another grqup, with .lower thari q;b1 tetation of the results of the correlative and -.: : ){1/s). .. r } I(%) 13.73 165.93 146.32 -79.3 -119. 9 286.22 104.04 114.46 38.31 322.43 678.48 3.12 36.65 27.47 -5.93 -42.48 38 .30 11.14 29 5 19.46 36.89 15.44 ;'13,56 -41.17 110 86.129 -237.71 -151.6 202.69 0 -6 -5.163 219.23 166.73 281.5 133 -9.359 24.3 16. 7 -17.7g -53. 68 27.122 0 -1. 54 -2.6 25.06 3 79 3.38 spectral analysis indicates two major classes of springs. The first category includes the systems more or less similar to the Aliou type', with low reserve s. which only slightly modulate the rainfall, having a low regulation capacity as well. The typical exam ple is provided by the Ponor-Plopa system. To the .same class seem to belong also the springs from the Vicinic stream, analyzed in 1974. In their case the value of the parameters is, however, affected by the reduced recording period (8 month no winter season values). Within the second category are included the sys tems disphiying a hydrodynamic. behavior of 'Baget type', with medium sized. reserves and large regulating capacity : This class includes, by and large, aU .the other springs. Two sub-classes may be still deiineated The first one displays regulating period and the first tnlncation frequency values similar to the 'Bagettype' (the springs from yalea Bigl\r, Ochiu Beului and Beunita). For ail these systems the memory effect is even larger than that of Bagei, 23-30 days respectively In what concerns the second sub-class, their parameters a transition behavior between th e 'Baget type' and the 'Fontestorbes type', espe cially In what concerns the regulating period and th e second truncation (Moceri, and Vicinic) lri fact, several karst

PAGE 135

134 0 0 Oo 0 0 '': . . 0 q 0 0 o A. Jurkiewicz et al. !----'---'--'"'-1. '. .. . : : .. : '.'. :: .: : :.:o, .[]3 --. 4 0 5 Oo @ ;;-----f () 0 ": ()0 0 0 G 0 o @ 6 0 () 7 n 10 Fig. 5. Liciovacea-Bigar karst l; rocks; 2; imperVious rockS; 3 Golulnbu SandStones; 4. pron;n underground dr&inage(S. spring Q > 6. spring Q < 25 J/s; 7; sJnldioJe; s. swalleti 9: cave; IQ. cave entrance. Le systeme kars#.qu.e 1. roches cizibonat'ies/ 2. roches. j . Greide" Golumbu; 4. Drainage$ .. Q>2$ ils; Q<-251/s;,?,_ doline; .B. p .onor; 9 grotte; 10. entree de grotte.!' > .. /,, ,. ., ;.,_-: :' .. : . : ,: . ., ... : :. .,, .. : ,., ,, systems .)iave .. a low. 'fir;t' fte:. quency, close to that identified for the 'Fontestorbes type'. The shape of the unit step response Junction is very sharp, without a significant s pn:;ad out in the case of the first class, and composite, indicating the presence of two distinct discharge components in the of the second (Fig. 7). Even though the anillysis is incomplete, only tially reflecting the hydrodynamic behavior :of some spi"ings, it calls for the same general remarks : as i n the case of the previous zoile; it can be addi tionally stated that: .. .."The presence of. some .impervious .. deposits areas withiri the carbonate surface llllows the formation" of "some. perennial streams dolumbti) which supply most im(>ortant karst systen:1.s iii Aninei Mountains cochiuBeului and Biglii karst systems) . The rest of ihe hydrographic network is iotally dis organized, the small brooks formed on the pervious basement sinkii1g just. at the contact with the lime stoiles;

PAGE 136

l(.a.rst sys.tems. in Banat Mountains 0 -, -, 0 0,5 I I I I I I I ........... .... __ \.,.---.... ---......... 11. n 12. v + + + + + + + 135 ,' I ,' 1 Krn '',,, ---====lilll.-=:::::::::::1 __ -Fig. 6. Liipqlilcul Mare k&rst system.. L carbonate rokS; 3. impervious rocks; 4. igneous 6. proven undergt()und 'drainage;7. swallet; 'spiiJ.Ig: (Q < 51/s). 9. sprbig (Q> 51/s).; H). sinkhole; 11. cave entrance; 12. entrance. : '< '' Systeme de UJpufnicu Mare. I. carbonatee's; 2. Ores de (;olu!"??.u; 3. roches impermeables; 4. roches ehQulisj 6: soutertain deniotttre; pon;,r; fcjU'f,ce (Q <. 51/s);-9. source (Q >5 lis); 10: dolirie; 11. errtree de. iroite; 12. entree d'aven. > . :. < . . .. THE WATER BUDGET The resu.lts of .the water budget completed for the 1994-1995 period (Table 5) supports the conclu sions obtained for the previous zone: for the Mini river catchment basin a supply from the plateaus Anina, Bradet and Buhui has been accounting for 6-7% of the calculated balance; the Nera valley catchment basin (tributaries included) displays over the limestone area a deficit of 1 L3%, which might indicate a water lfansfer toward the Neogene deposits existing at the western limit of the synclinorium. The current data do not substantiate an older hy pothesis, which assumed the existence of a groundwater flow from the central Aninei Mountains towards the Danube, passing be neath the Nera stream bed.

PAGE 137

136 . . . Q,;.;,:: Q min .ME TF id.. IJ;-: QJ11in. .vs: .. lopa Cave.. 3i30 31 101 180.6 3 0.268 .. ; __ ; 112<>, Jd. : : "f4:'t .. . 1 976 482o 14o. :34;42 s11 a.oo9 . .. 24.1' .. :23 o.os6 __ :hoo 42:-.. :4 83-:;0JJI" ; ... : .. .:.6.216 2620 'is .. 276. o.'oi31 .... 3.52 . 43 0,144 37: icinic (Ilidia) Ochiu Beu . Simion 1 Simicin 2 Simion 3 .: : mrf(.; :_:;.:; :i ;j{ . 1994-95 17-. 52.47 ,118 O.DIOS ... L92. At .. 0..140 ... 1 r2 3.5 3 2 .. -_,if > .. 2a::.:o ::1s4 ,' .. 1973-74 1300 32 40.6 2oi: o.o3ss 2:2 n o.J67 199+-9s.'Io2o. 18:::i7.5 :o .oi21 :31 .. o.l83 946o 1s 6 6o.6 112 ... o .o221 .. l3, t 24 0 .168 J .) 22oo : 35: i60 : 0.0153 .. ;; : . ;''" ; 26:. .... ... 0,168 1964-65 189' 1Q.4 '18.17 43; 8 ;. .,.ht6. ::;3o'.\ :7:3;.i: 1964-65 146 9 7 15.05 26.7 ME = effect; TF =truncation RT = regtdating M =model. : .. ,' . . .. .. .. ': M:iri! :.-:< .. : 67o : o:; 948.s:< ,.:165.3 .:.:..s .?, Tiidia ... .-. 369'.{ 17.5.8::: 6o. f .... /: ... 16. 3 -.. . .. J6.3 -i..liiJu$ni2 ... : . : -;is6.8 / .. : :.:.:46.0 .. .-.. :. -t4s.r .-. :.:.92,5 .:..14s:t : ;.i 92. 5 Moceri$ .. ... .. 6-/ _rsio ..... .:_ ;..::39:4.-:--si.oi .:.;tos,8:: .. :.:8uii bucin ... -2S7. s.-:; -:i4.i-:-: ii6:t(:. ::: _:_ .. s.3 ... .: : s,3 : J i .. Nera 387.5 .'11.3 Poneasca ().:: .. o Total basins 7269. 7 2697.6 _: .1390 :'i9J -1310.8. -18 :4. . 39:9:_: 0 : 5 ): :: 10 .: ;:;: : .:: a :sj,c.ifi/ cat<;hment basins + 1390 lis (19.l %}; .: : ....... :' ... where they lia:ve been ciilculated: :-: : : 2; .: . :the : :: :. :: 3; ; 1S58 vincm2 s and witness basin Porieasca. (Very dose l!J the :_ ... SE area: b e 10ltsJkm2 .. .. : 1993 for C2:1ra-Jervl:lni catchmeQ.tbasin .;:. : .. .. 18.7811sllcin2); now rrom otnei: eatc:iuiient 6a.sin:(.: .< ... Miiu catchritent. => other 1310. 81/s(--:-iSA%) ; .. .. A possibly ritoie realistic. .(3). ..... )'oial budget takes irito iccount different evapotranspi-: catclunenfbasins 39.9 Vs .. : ' . . . . : ... ..

PAGE 138

in Banat 0,6 r 0,1. 02 ANALYSES ChemiciU analyses pe:rfom;1ed for 40springs indiGate The chemical qiJalHy is genetan.y .. Total mineralization (IDS)" values betWen 300 arid :600 ingll, except for the water from the Ponor Pl()pa cave system, which eXltibits a higher .TDS value The. e.xcessive minera1j.zation of this water is ptainly due to the pollution ; hlduced by the .mining works of Criviria. According to their SO/content, distinc t groups of springs have been sepru;ated :within the central section of Banat mouritains. The first group includes the karst waters with less than l nigli. sulphate, discharged by spririgs such as Ochiu Beului, Bigar, or LapUnic (Fig. 8). The second group of springs (Simi on 1, 2. Vicinic, Plopa) inc}Qde Crivina marls with coal interbd dirigs within their areas, which results in im increased content (i.ip to mg/1, Fig. 9). The last group of springs (Fig . 9) consists of those whl(;h are $trorigly i nfluenced by the coal miffing activity at Aillii!l {Pon
PAGE 139

138 .. .. : .. 10 200 100 200 ,: ; 1: \ \ :1 , 101 10 'I \I I \ 0,1 I 0,01 0 1 0,03 Ca .. .:tQ: 0.01 I i \ : I \ I i I \ I I 0.1 I O.DJ I 1 I ....... 0,1 : i 9 .0 2 Mg ',:I 0,1 0,03 ,,_ '-------2 300 100 0,1 0.04 Cl 400 100 10 I I I I .I I I I I I ;, I ;. I ....... ...... 3 0,1 0,1 600 1Q: :.: : -100 I I I I I I 10 I I 0,1 I I Hco/'EPM ---. -4 ..... 100 0,1 0,01 -0,1 0,001 EPM Co 0 04 0 05 q001 Mg Cl Sel[: HCDJ EPM ............ -.. -.. 5 --2 ------3 ----4 ----6 ---7 -----8----9 . ... A. Jurkiewicz f!t. (II.

PAGE 140

Karst systems in Banal Mountains swallets catchment area, which provides a virtually constant, 75-80 Vs supply. Not far from this system, at the margin of the Bd.det plateau, more than ten karst springs totalling 180-200 Vs are also used for the water supply of Anina and of the Liava colony (mining activities included). The Simion springs, located 300 m upstream of the Ciclova Montanii village, have been successively used as a supply for the brewery of Oravita (founded in 1865), then for the tap water supply, and currently again for the brewery (rebuilt in 1975). The main springs from the central part of Aninei Mountains, namely Bigiir and Ochiu Beu lui, are used only for fish hatchery. CONCLUSIONS The tectonic and geologic factors contributed to the current configuration of the water accumula tions in the carbonate deposits of Anina Moun tains. The complex role of the main fractures and of the accompanying second-order faults is out lined by their coincidence with the lineaments of swallets and springs through which the karst aqui fer discharges. The karst morphology, a direct result of the complex physico-chemical processes, displays the whole range of surface and under ground features, for which different evolution stages are recorded. The computed hydrodynamic parameters (depletion coefficient, dynamic stored water vol ume) and the results of the correlative and spectral REFERENCES L., NEGREA, & NEGREA, A. (1967) Grottes du Banat explorees de 1960 a 1962. In Recherches sur les grottes du Banat et d'Oltenie (Roumanie, 1959-1962), Ed. CNRS, Paris. L. & NEGREA, (1976) Drumefind prin Muntii Banatului. Ed Sport Turism, Bucureti, 234 p. BUCUR, I. (1991) Studiul Jurasicului # Cretacicului din unele perimetre de perspectiva pentru car bunii liasici lntre va:ea # valea Nerei (compartimentul central al zonei Banat), Tezli de Doctorat, Universitatea "Babe-Bo1yai", Cluj Napoca, 190 p. CINETI, A. (1990) Resursele de ape subterane ale Ronu2niei. Ed. Tehn., Bucureti, 295 p. IURKIEWICZ, A. & DRAGOMIR, G. (1994) Studii hidrogeologice pentru evaluarea resurselor de 139 analysis, performed on flow rate histories of the springs which have been continuously monitored for one year, indicate mature karst aquifers and lead to the following general conclusions con cerning this area as a whole: the tectonic and topographic dissection gener aly prevents the installation of major karst systems; the karst cavities reached a highly organized stage, which results in very good transmissive properties, but poor storage properties; the presence of some impervious deposits areas within the carbonate surface allows the formation of some perennial streams which supply the most important karst systems in Aninei Mountains. The water budget performed over the 1992-1993 period for the Cara-basin zone indicates a water transfer, although not very significant, across its southern boundary, toward Mini basin. The results of the water budget of Mini-Nera zone completed for the 1994-1995 period also confirms the previous assertion (i.e. a supply from the Anina, Bddet and Buhui plateaus toward the Mini river basin). Furthermore, the Nera valley catch ment basin (tributaries included) displays over the limestone area a deficit of 11.3%m, which might indicate a water transfer towards the Neogene deposits existing at the western limit of the syncli norium. ape subterane din depozitele carbonatice mezo zoice ale sinclinalului Noud, (etapa 1-a. 1992-1993; perimetrul Anina). Arh. S.C.Prospecpuni S.A., Bucureti (unpublished). IURKIEWICZ, A. & MANGIN, A. (1994) Utilisation de ]'analyse systemique dans !'etude des aquife res karstiques des Monts VaJ.can. Theor. Appl. Karst., 7, pp. 9-96. IURKIEWICZ, A. & ROTARU, A. (1996) Studii hidro geologice pentru evaluarea resurselor de ape subterane din depozitele carbonatice mezozoice ale sinclinalului Moldova Nouli, (etapa 11-a, 1994 1995; perimetrul Anina Sasca Montana. Arh. S.C.Prospecpuni S.A., Bucureti (unpublished). JEANNEL, R. & RACOVITZA, E., G. (1929) Enume ration des grottes visitees 1918-1927 (septieme sene). In Biospeologica, LIX, Arch. Zool. Exper. et Gen., 68, 2, pp. 293-603

PAGE 141

140 MANGIN. A. (1975) Contribution a retude hydrodynainiqtie partie, . Constitution ei des aquiferes karstiques, Ann. Sp'lliiol., 3Q pp 21-124. . MANGIN A. (1984) Pour une meilleure connaissance 9es systemes a partir des analyses correlatoire et spectral e. J. Hydrology; 67; pp; 25-4J. . MUTIHAC, V ; (1959) Studii geologice In partea me diana a zonei Noui:i (Banal). Ed. Acad. R.P.R., 106 p. NASTASEANU, S . (1964) Prezentarea hhtii geologice a zonei Noull ,sc. 1:10 .000 An. Com Geo/ ., XXXID, pp. 291-342. NASTASEANU, S & MAKSIMOVIC, B (1983) La correlation des unites structurales alpines de Ia partie interne des Carpates Meridionales de Roumanie et de Yugoslavie. Ann.1GG, LX. I. & N. (1972) ProspecJiun i hidrogeologice hidrochimic:e pentru fosfaJi f n zo na jud. -Severin. Arb S C Prospectiurii S.A., Bucureti (U11puhlished). ORGHIDAN, T NEGREA .. RACOVITA. G. & LASCU, <::;. (1984) din Romania. Ghid Turistic.< Ed. Sport-Turism, Bucureti, 454 p PASCU, M R. (1983) Apele suhterane din Romania. Ed Tehn .. Bucureti, 411 p. A,. Jurkiewicz et al. RAILEANU, G., NASTASEANU, S. & MUTJHAC. V . (I 957) Cercetlri geologice fri regiunea : dintre Doman (zona Noull) R.P.R. Ser, Geql. Geogr., 2. SENcu: v. & ClOACA. A. (1980) Cercetari pentru ident;fic:area golurilor carstice din zmia Anina Raport de sinteili, Universitatea Bucureti; lnst. Geagr: Arhiva C.M.Anina, 133 p (uizpuhli.5hiul) SENCiJ; V. (1970) Vrule de doline din mun\ilor .. Banatuhti. Si. Cere. Geol.. Geojiz : Geogr;, Seria Geografie, XVII; 2, pp. 231:.:..237. SENCU, V (1973) Les eauxkarstiques de Ia zone mi niere d'Anina, (Banat). In Livre du Ciliq!4lnte naire de l'lnst. Speol. "Emil Rac oviJi:i", Ed. Academiei SENCU, V. (1978) MunJii Aninei . Ghid turistic Ed.Sporl -Turism, 86 p SENCU, V (1986) Field measurements relating to the subterraneim runoff of karst waters in ihe Anina Mountains Z Geommph: N.F pp.59-68 STILLA, A., RADU, A . IA VORSCHI, M., DRAGO MIR, B . VISARION, A., IVAN, M., GHIU T. & GHEORGHIAN, D. ( 1 972) Sinteza Lui:: ri:irilor de c:ercetare geologica pentru stabilirec/pe r.rpec.: ti velot de ci:irbuni superiori in : z( jna. Moldova Nou/1, perimetrulc ; uprins intre paralelu Anina s i Dunare judeJu( Arh S.C. Prospectiuni S.A. (unpuhli rhed).

PAGE 142

Theoretical and Applied Kar$t0logy, vol. 911996, pp.l41-147 Examples of Karst Springflow Regime Simulation and Prediction for Water Management Balance Analyses Vesna RISTIC, Stevan PROHASKA & Zoran STEVANOVIC1 ABSTRACT Karst occurrences in stream basins have not been given adequate consideration in the applied hydrology. As a consequence, water table and discharge of karst springs are not measured at stations of the official hydro meteorological network. For a proper assessment of the karst water regime, under the given study conditions, indirect methods and computation procedures have to be used. One of these is MNC (Multiple Nonlinear Corre lation) model, used for filling-in, extending and three-dimensional interpolation of the hydrometeorological data. Based on this model, in the general areas of the karst springs (Izvor-Sveta Petka, Mlava, Grza, Nemanja, Mrljis with Groznicevac) on Kucaj and Beljanica mountains, eastern Serbia, this paper presents filled-in (where data are lacking) and extended series, which are essential for the investigation of the karst springflow regime, or for a water budget analysis that extends over several years. Key words: karst springs, stochastic prediction methods, water management balance. Quelques exemples de simulation et de prevision du debit des sources karstiques pour /'analyse du bilan hydrologique RESUME Au cours des recherches hydrologiques sur le regime des eaux effectuees jusqu'ii nos jours en Yougoslavie, on n 'a pas donne /'attention necessaire aux occurrences de karst dans le bassin d'une riviere. Par consequent, le reseau d'observation du service meteorologique ofjiciel ne prevoit pas de mesures du niveau et du dehit des sources karstiques. Pour determiner precisement le regime des eaux karstiques dans les conditions hydrologi ques donnees, il est necessaire d'appliquer des methodes indirectes et des procedures de calcul. L'une de cel les-ci est le modele CMN (Correlation Multiple Nonlim!aire) utilise pour completer, extrapoler et interpoler dans l' espace des donnees hydrometeorologiques. Le present travail illustre le fait que, sur La base d'un tel mo dele, on obtient des series completes (La ou les donnees manquent) et extrapolees qui sont essentielles pour {'observation du regime des eaux, c'est-ii-dire pour ['analyse du bilan des eaux pendant une periode de plu sieurs annees dans La zone plus large des sources karstiques (La source /zvor-Sainte Petka, La source des rivie res: Mlava, Grza, Nemanja, Mrljis avec Groznicevac), puis les sources des ITJOntagnes Kucaj et Beljanica dans La Serbie Orientale. Mots-cles: sources karstiques, methodes stochastiques et prognostiques, bilan de La gestion de l'eau. 1 University of Belgrade, Faculty of Mining and Geology, 7 Djusina str., Belgrade, Yugoslavia.

PAGE 143

142 INTRODUCTION V . Ristit et al. of water supply to major towns and industrial centers; ,but. their degree of utiJi:zation Can be inGround water reservoirs are sources C:rea:sed: Large water reserves of good quality piyjo individtial con sun)ers; but orilfa nririor part reliltively easy jq .. proteCt against poiiution are of this resouree is ti,tilized . Karsf aquifers in the Ori the other haild, the degree of investi-of eastern Serbia and. in the giition' and availabhrinvestmen.ts not adeDimirldes of westettt" ate "the main ,: .. quate tcnnilk:e an :effiCient use of this resource: Flg.l. Hydrographlc map of Kucaj-B eljanica p18$Slisiio:Wbig watersheds. i: Karst aquifer; 3. Perennlal stream; 4. Intei:DJittent Uke.y dire(:tiori. ofgrqundwater flow; 6. ApproXimate 1. rrt3isi s.Sprlng 1 m3/s; 9. Swallet; tO. Inferred cif if. Proved t ,1t\;. tantes. I. limite dtl, cot!fP.lexe 2 }. . superficielperiodif[Ue;"S. t!ts eti_ll:X liniite _. (. approximative ' bassin$; 7. source au debit de. plus (]e._ J lftYs; /j;: source, au.'cfe_bltmoyen 9. perte; I 0. (lirection d' ecolflement/!es eatlf source de Ne11WIIja, Ill source de Mr(i4. . . ,:

PAGE 144

of karst springflow regime simulation The available = hydr
PAGE 145

.... ... .. : .:= ;-. : .. v: Ristic et"til. .... .. .. : WATER BALANCE FOR KUCAJ AND BELJANICA KARST . SPRINGS Basic quantitative parameteisof the water for Kucaj an.d .Beljanica karst springs, .for provF .. sional use, cair be deduced from th e mol)thly and annual average karst .springflows given iri Tabl e L It follows from the above Ulblethat 3) Cn3/S; <)r 116:74 X j06 .mj :Of good viater are diS-. charged annuallyon average by the mentioned ::. karst Springflows .are the highest, abot}t 45.6% of the total aririualamount, from Marchto May, and the. lowest; ,onlylO%, during the dry season (Augustto October). .. Table numerical.values of t h e main element$ of the water equation for the study : area. It .also ghi es numerical .value!!. of domestic stream :flo ws, taken front (h Water Management Plan .of. Serb i a (PROHASKA et a/.;. 1993) . The quantitieS given in rows 2 and 3 refer io multiannual...: .: . averages ..... derived . for : .. the ColUmns-in: Table 2 refe,tto: catchment atea; F (ktn2); average l ong t e rm dis charge, Q ofo\itflow, W(l06m3), specific rUnoff, q {l/s/ktn2); rul)offlayer, h (mm); 1:\ .armual precipitations, P (tn,m); average aniniaJ E (inm); and average long ... term runoff cO:efficient, q>. The behavior of other water balMce components Tile above elements of the water balance equation should b e amuyzed by considering the hydrogeologilead tO the conclusion that the considered area is cai setting and their characteristic iJ}.terreJ.ationS for .. very hel<:I'Ogeneous with respect to the genesis of the drainage areas of f{:ucaf.. and Beljanica kafst . the resource. Some of the water bahince Mlave Grzc 0;751 0.415 0.136 0.137 < Sv. P etke 0.418 0.593 0.742 0.504 0.311 0.224 0.17 0 0,186 Nemanje II 0.01 1 0.018 0.023 0.012 0.064 0.037 0.022 0.033 General Mrl is. zon e 0.4 87 1.172 2.311 0.642 00382 .. :' 3.1 80 s ois 7.624 7J9i6 s;(:>36 2.429 To t al . Table +. ': ,.,,., .: .. ,. Representation l!ti .... .. No Catchmen t area F Q w q h km2 m3/s io6m1 Vs/km2 mm ..... Kucaj and Mlavc 179.2 2.039 6423 1L38 358 Beljanica Grzc 36.6 0.376 11.84 10.27 323 I karst Sv. Pctke 27.2 0.432 13.61 15.88 500 springs Mr1jis general zone 99.5 0.848 26.71 8.5 7 270 Total 342 3.706 116.74 10.84 3 41 2 Rivers Basin Kuca i and Belian i ca 181 7 1 9 : 599 617.37-10.78 340 3 Dom esti c w a t er 88 361 508.8 160 27 5.76 181 1.685 2.039 0.211 0.285 0.375 0 .2 72 0.356 0.432 0.062 0.084 0 .011 0.372 0.683 0.848 2.222 3.093 3: 7 0 6 p E cp mm mm 601 243 0.59 658 335 0.49 635 135 0.79 732 462 0.37 629 288 0.54 710 370 0.48 7 3 4 553 0.2 5

PAGE 146

Some e,xampks of springjlqw regime simukltion oquatiop meJl1bers vary as much as more .than 100%, the evapotranspiration alpne by over 200%. Karst is fro(Jl the hydrological setting, as an factor of drastic influence on the drainage p _attem in the area. Therefore, any re gional telii.tion deduced .by using one of the con -hydrological qtethods without tiling .karst into has not a practical appliUrifortunately, karst is given adequate consideraU _on .in applied hydrology in this CQUntry, A consequence -is the wrong estimate of the effective water t1ow for the design of con structed water structures That is the i"easori why sortie reserVoirs cannot. be filled for years or some .water supply $ystems are affected. by shortages during the dry season, when the water den1and is. the highest. At the same time, good healthy waier from Kucaj and Beljanica karst springs flows out uncontrolled. . THE SIGNIFICANCE OF KUCAJ AND BELJANICA KARST MASSIF FOR THE\tVATER MANAGEMENT PLAN -- 4 ... management, activity includ:ing lhe protection from of water : utiHiat.lon water and their adequate contr<>l; on the kriowldge of all the available water resources. One type of besides surface and ground water bodies, is karst aquifers.: A karst spripg is .impor tant, because it gives good water alinos t for dfrect use . Where used for hulnan consumption, one of the priority uses in the water plartni11g, water from ;k8rst requires little purification treatment, which makes it least cost atti:ac;tive. I ; Water resources are the main consideration in any water management branch and any d,eveioptnent scheme. ::and so is consequently the k arst spring flow regime. The purpose of many water Mlava 0.379 0.328 145 agement branches is to balance water budget in amount, quality, .and in relation ':to the-' environ ment: to average flows with the demands of the users, to contiolflood flows, io supplem ent lpw flows, to improve quality, structures must be avllilable (storage lakes, barrage pools. water treatment and purification fadlit1es, etc ) to this purpose. In this respect, the use q f karst water has many aclvantages Each karst spring : is supplied from a karst aquifer or resei'Voir, _and the redistri bution of ihe total amount of waier i s natural over one or several years Thus, the desired . effect is accomplished: balance of average flows, reduction of ntaltiinum. and increase of triiriimum flows As to the water quali t y. it is superior to that o f surface streams or other ground water bodies, becaus e karst water is well protected against external influ ences From the man-agement point of view particularly f considering wa:ter aviilability . two re quirements regimes are prinipa.I: :. the rPiniQl.Un:t and the .. water eGOilQrnY m:inim9m. The former 1s. an lowest stream fiow that be provided from a storage reservoir and a pool. < as, .Overflow fr9m an intake and the like, to preserve the biocen0sis in the stream as a biotope The other nurtiinum is a category-the lowest strea.m flow that must be deti vered by upstream users so as to satisfy the needs of all downstreain coJiSumers The minimum J:rtonthly average discharge of 95% occurie.nce-ba,sed dependability at the intake point is taken for tiie warranted minimum. In the ligbt of the a,bove stated, all relevant water regime parameters will be given fm the considered karst springs on Kucaj arid Beljanica, taJQng into considera,tion the mentioned requirements and limitations The pqrpose of this analysis is .to assign the karst springs o f Kucaj Beljanica the proper phtce Cllld significance J.n the .water re sources development schemes of Serbia. . '"... ....... .'" .:. 9.8. '.'. 'Q9 . 0. 294 0 .270 Grze o 029 o 026 o.o2s o 024 0. 263 0.024 0 096 0 001 0 246 0 023 0 001 0 040 1.. Sv. ,Petke 0.103 0.099 0 097 0 097 Nemanje II.1 i 0 .001 0 001 0.001 0 001 Jieneral zone o.o95 o o 7 5 o .o6t o o 5 I 0 048 .::_ :: : .. ... .. . .. _: _:.::

PAGE 147

146 The most interesting quantity in this respect is the minimum monthly springflow of high frequency and the 95% frequency flow, as a regulatory pa rameter for the definition of warranted minimum. Table 3 gives occurrence-based dependability of minimum mean monthly values for the considered springs. The above quantities show that the total minimum mean monthly springflow of Kucaj and Beljanica large springs is 0.478 m3/s. This quantity of water is regulated by the Water Law of the Republic of Serbia as the (warranted) minimum flow to be provided for the downstream users and for the preservation of the biocenosis in surface streams. It means that the difference: Qsr-Qmin ,95 = 3.7-0.478 = 3.222 m3/s can be used to meet the demands, primarily for domestic water supply. The considered springs of Kucaj and Beljanica, in terms of water budget, provide control over 3 706 = 0.189 = 18.99% 19.599 of all water resources in the general area of Kucaj and Beljanica According to the ratio of catchment areas 342 = 18.8% 1817 the water-yielding capacity of the karst catchment areas is almost the same as that of the total Kucaj and Beljanica general catchment area, and 95% higher than the average long-tern value of domes tic streams in the Republic of Serbia (see Table 2). The most interesting from the standpoint of water management is the comparison of the minimum monthly average flows of 95% occurrence-based dependability for the study area. The ratio of the REFERENCES PROHASKA S., PETKOVIC, T. & SIMONOVIC S (1979) Matematicki model za prostomo preno senje i interpolaciju hidrometeoroloskih podatalCa. Saopstenje /nstituta "J. Cemi", 64, Beograd. \. 't : : discharges from the direct catchment areas of the karst springs .and the catchment areas o( surface streams in the general Kucaj and Beljanica area is: . 0 478 = 0.394 = 39. 4% .' 1214 Considering the size of the two catchment areas, water yield from the karst spring drainage area is twice as high compared to the yield of the surface streams in the general Kucaj and Belj8nica area. Compared with domestic streams of the Republic of Serbia, the area drained by Kucaj and Beljanica karst springs yields more water than the rest of Serbian territory which is planned to be 70% con trolled by 52 surface reservoirs (Water Management Plan of Serbia, 1993). CONCLUSION The karst springs on Kucaj and Beljanica moun tains (lzvor, Mlava, Nemanja, Grza, and Mrljis with Groznicevac) discharge on the average more than 3.7 m3/s, or 11. 6 74 x 106 m3/year of good water. The cumulated minimum monthly average flow of Kucaj and Beljanica karst springs amounts to 0.478 m3/s. This amount of W!lter is regulated by the Water Law of the Republic of Serbia as the lowest permitted discharge for downstream users and maintenance of biocenosis in surface streams. It means, that the difference Q s r -Qmi n 95 = 3 .7-0.478 = 3222 m3/s is a quantity potentially usable to meet the de mands of various users, primarily for domestic water supply. A general conclusion from all the above consid erations is that the discussed karst springs are comparatively richer in water than average do mestic streams in Serbia, and that they constitute a significant resource. PROHASKA, S. et al. (1992) Vodoprivredna osnova Srbije-meteoroloske podloge, Dok Instituta "J. Cerhi" i Rudarsko-Geoloskog Fakulteta, Beograd PROHASKA, S STEVANOVIC Z RISTIC V & MARCETIC I. (1994) Analiza osnovnih kompo nenti bilansa raspolo 'ivih voda u visegodisnjem periodu u siroj zoni karstnog sliva Mrljis. XX Savetovm:tje hidrogeologa, Kik:inda.

PAGE 148

Some examples of korst spring/low regime simulation PROHASKA, S. & STEVANOVIC, Z. (in press) Raz voj autokrosgresivnog modela za simuliranje izdasnosti karstnih vrela, ref. simp. "Covjek i krs", casopis "Nas krs", Sarajevo. RISTIC, V. (1995) Metodologija analize i prognoze izdasnosti vrela kucajsko-beljanickog masiva, Mag. rad, Rudarsko-geoloski fakultet, Beograd. STEVANOVIC, Z (1986) Neki aspekti odnosa povrsinskih i izdanskih voda u karstu. Geoloski glasnik, XI, pp. 103-111. 147 STEVANOVIC, Z (1988) Rezultati simultane hidrometrije na nekirp vodotocima u karstnim terenima. Zapisnici Srpskog geoloskog drustva za 1985-1986, pp. 325-331. STEVANOVIC, Z. (1991) Hi4rogeologija karsta Kar pato-balkanida istocne Srbije i mogucnosti vodos nabdevanja, Monografija, izd. RGF, Beograd

PAGE 150

Theoretical vol. 9/1996, pp. 149-154 Correlation Regional Fault Pattern and Karst Water Flow Directions Based On Examples from Eastern Serbia (Yugoslavia] Zoran STEVANOVIC, Miroslav MARKOVIC, Radmila PAVLOVIC, Igor JEMCOV, Veselin DRAGISIC & Tomas CUPKOVIC1 ABSTRACT The analysis of the fault pattern of eastern Serbia was made as a part of the basic and purpose detailed hydro geological investigations in some localities. By correlating the observed regional features with the point ele ments of the principal drainage systems, a significant coincidence of positions has been noticed. The strikes of the main faults, longitudinal NW-SE and younger NE-SW (W-E), largely coincide with the principal, general directions of karst ground water flows. Subjects of particular analysis were the calcareous massifs corresponding to the anticlinoria of Kufaj and Beljanica, and of Rtanj and Ozren in central Carpatho-Balkan mountain range of eastern Serbia. Key words: satellite image, regional fractures, karst sources. Correlation entre /'ensemble regional des fractures et les directions d'ecoulement des eaux des sources karstiques selon des exemples de Ia Serbie orientale (Yougoslavie) RESUME L'analyse de I' ensemble regional des fractures sur des terrains de La Serbie Orientale a ete faite au cours des recherches hydrogeologiques normales et de detail dans certaines localites. Par la correlation entre les fractu res regionales observees et les elements ponctuels des principaux systemes de drainage ---: de fortes sources karstiques, on a constate une coincidence significative. Les directions principales des fractures, NW-SE (lon gitudinales) et de NE-SW (W-E) (transversales, plus recentes) correspondent en majorite aux directions gene rales principales d'ecoulements des eaux des sources karstiques. On a fait particulierement des analyses des massifs calcaires, qui correspondent aux anticlinaux de Kucaj et Beljanica, puis de Rtanj et Ozren, tous situes dans la partie centrale des chaines Carpato-Balkaniques de La Serbie Orientale. Mots-cles: image satellite, fractures regionales, sources karstiques. INTRODUCTION This analysis of regional hydrogeology and fault pattern covers the northern and central parts of the Carpatho-Balkan karst massif of eastern Serbia. The region is bounded by Homolje Mountains. to the north, the Velika Morava depression to the west, the Beli Timok valley to east, and the Nisava valley to the south. It is significant for the extent of karst topography and the large reserves of ground water used in municipal water supply systems. A correlation between the fault pattern and the major karst drainage zones revealed a sig nificant control of karstification by the position and the function of the faults.

PAGE 151

150 . METHODOLOGY OF THE REGIONAL FAULT PATTERN -STUDY ' .: .. :: ..... ,.'' ... .' (faults) were primarily identifie'd on expressions in the land configuration, and chissi fled by the plausibility of identification into orpartly traced, inferred faults. .. ' .. Linear structures; which form the regional fault The degree offauli expression in the land configupattern of eastern Serbia, have : been 'eX:plored by ration depeildS .On the tiine of its activity; Neotecusing the remote detection method. Considering tonic the Neogene, probably the size of the exploration area and the small scale from the Quaternary, are putlined by the. rectilinear of representation, the most suitable method of data pfwater-:c::ou.rses, steep scaq)s, and linear collecting was the satellite images (LANDSAT breaks in slope gradient, Le. elongation of large missions 2 and 3) on scale 1:500.000, analyzed by karst features. Faults ofpre.,.Neogene origin, wbich means of the logical comparative technique. were repeateclly activated during the rieotedonic, White-black contrast images Were used, taken in the youngest, perl.od of the geological history," . are green, red and related infrared spectral areas. The similarly outlined (KRESIC & PAVLOVIC, 1990; results obtained were correlated with the analytical PAVLOVIC & KRESIC, 1990). results of color-composite images at the same Faults slightly marked in the topography; only scale (Fig. 1 ). The transposition of data from itnpartly traced artd classified as inferred, are always ages into topographic maps was visual, by recog-..... old, pre-Neogene in age. Their activity through the nition of topographic features. : Neogene alld the Quaternary was single, or the Fig. I. SatelUte n easterri ; lliwge-satelJite .. 4e chee de /ij Strl,i(i movements were feeble, without notable influence on landform modification. It erisues that the classi. fication byplitusibility of the regior1al fault pattern data has additional chronological implications . . CHARACTER OF THE REGIONAL FAULT PATTERN The fault .pattern of the eastern Serbia study area is shown on the map (Fig. 2). Faults, classifiedby the significance in the structural fab ric, are divided. into two ranks. One includes hirgc features o( regional importance, Which can be traced in the length rarige of hundreds of meters. Their traces riever extend beyond the perimeter of a study terrain. The other rank includes faults o( local significance for the regional pattern. Their lengths are of tens of kilometers, within the limits of the study area. . . Large regional faults are compiex structures. Broad fault sets of related,paral]el faults, echelon faults and large single faults alternate along the stdke, with more .or Jess htrge breaks. : Their geometry, kinematics and time of activity differ from one "pliwe to another. The convenient name for suc}l a structure would he "preferential fault line". . . Faults of local significance in the reglonal of the discussed part of eastern Serbia can also be compleX: structures, . such as actual : fault zcmes, fault sets 'etc., but more often they are single faults of different kjnerriatics, gravity faults, fronts of nappes or slices, faults with horizontal dis placements, etc; Assessments of fault kiriemati.cs by remote dete&tion alone are not reliable alldare omitt(:d

PAGE 152

Correlation between regional fault pattern and karst water flow directions 151 The main regional faults (Fig. 2) strike NW-SE. Lithostratigraphic units in the region have the same trend. Faults of NW -SE strike may be there fore taken for longitudinal structures. Faults of perpendicular strike of E-W trend are less fre quent in the region; they are transverse faults. Faults of local significance are mostly satellite faults of the large regional ones. Longitudinal, NW-SE oriented, and transverse structures of NESW to E-W strikes are equally present. Deviations from the previously mentioned general trends are much more frequent in the case of the local faults. Longitudinal structures vary in strike from NWSE to NNE-SSW. The transverse faults include some of ESE-WNW trends. Regional, large faults vary within the magnitude range of tens of kilometers. Small faults of local significance extend within the kilometric range. This makes the structural pattern of the eastern Serbia study region extremely well developed. Most of the identified faults are neotectonically active, displaying significant vertical movements. These features must have affected the hydrogeol ogy, particularly the movement of karst ground water. Faults are potential paths water circulation, or barriers that divert the flow Differential move ments of the faults separated blocks have a definite influence on the water table. HYDROGEOLOGY OF KARST IN THE ANALYZED REGION Karst terrains of the discussed region are quite extensive, occupying a surface area of about 1800 km2 or 76% of the entire region. They are represented by a complex of carbonate rocks of Mesozoic age, dissected by regional faults and separated by noncarbonate rock complexes. Most of the karst terrains are single antifonns of non carbonate rocks and cores of older sediments. Karst areas are characterized by a high level of karstification, equivalent to holokarst, and large reserves of ground water. For all major towns and industrial centers in the region, aquifers are the main sources of water supply (STEVANOVIC, 1991; STEVANOVIC & FILIPOVIC, 1994). Nine regional karst aquifers have been identified in the reported region, viz.: Gornjak zone, Beljanica massif, Ravanica limestones, Rtanj, Krs, Tupiznica and Vrska Cuka, Ozren and Devica, Kalafat, and SvrljiSke Planine. The largest is the Kucaj-Beljanica massif of more than 1100 km2 Three of the aquifers have prevailingly karst catch ment areas. The covered marginal parts of the karst aquifers have a total area of more than 800 krn2 Karst ground water circulates through the developed systems of fissures, fractures, channels, caverns in which turbulent flow prevails: fast propagation high filtration rates. STEV ANOVJC & DRAGISIC (1992) give a synthetic presentation of flow-tracing results for the CarpathoBalkanides of Serbia and indicate relatively high fictitious rates of ground water flow, ranges from 0.0045 to 0.222 m/s, resulting an average rate of 0.063 mls. They also indi cate a multidirectional circulation in the hydrauli cally complex karst systems, and the occurrence of water piracy between adjoining catchment areas Ground water in karst areas is prevailingly dis charged by large springs, in addition to under ground runoff and evapotranspiration. Large karst springs usually discharges at the margins of the karst massifs (close to the regional erosional base level), at contacts between carbonate and noncar bonate rocks. Their location is often controlled by fault structures or fracture systems. Unlike this, many springs emerge in limestone massifs (in zones of the local erosional base level), have smaller catchment areas and, consequently, lower yields. Their location is also controlled by local barriers (impermeable rocks or fault structures) More than 40 karst springs are documented in the study region, all of them are significant sources of regional water supply systems (the lower limit of the minimum yields is 10 lis in an average hydro logical year). There are several temporary springs, which in the flowing periods may discharge hun dreds of lis. Ground water movements in the emergence zone are either gravitational or ascending, depending on the karstification level and the position of the karst channels. The discharge of karst springs, in par ticular that of gravitational springs, is often non uniform, with large amplitudes and fast propaga tion after rainfalls. The sum of the mean springflows, corresponding to the dynamic reserve, after STEVANOVIC ( 1991; 1995), varies from only 0.1 m3/s (Kalafat) to 8.67 m3/s (Kucaj-Beljanica massif), or expressed in per ceo{ of the rainfall over the catchment area, it may amount up to 23% (Svrljiske Planine)

PAGE 153

152 LEGEND karst sprm!IS Qmm:> mtarmrtent karst sprmg Qmax> 100 1/s "'( thermommeral spring pradrctacJ direction or karst ground water now smkholo karst spnng conection delmmined by tracer lost potential underground outflow .,_ --venf. ] TP.giooaJ fauns supp. vera.] supp. local fauns -karst aquifer [ Fig. 2. Map Qfthedistrlbiltlonof Carte de Ia. repamtion.des .. Z. Stevan01ic et al.

PAGE 154

Correlation between regional fault paltern and karst water flow directions 153 RELATION OF FAULTS TO THE GENERAL KARST GROUND WATER FLOW The recent fault pattern is the result of repeated strong tectonic movements, and many of the identified faults are still active. The privileged directions of karst ground water flow are mainly inferred based on fault structures and on the rela tionship between the karst aquifers and other less permeable rocks. Locations of main karst springs are controlled by the regional faults; there are a few springs outside the fault structures (Fig. 2). Thus, many large karst springs: Krupaja (1), Grza (2), Krivi Vir (3), Pop sicko spring (4), Zlot (5), and springs in Veliki Krs and Mali Krs zones emerge along longitudinal faults Karst springs also appear along local, trans verse faults in zones of intersection with the re gional, longitudinal, faults. Karst springs that emerge from regional transverse faults are more abundant in water than those along the longitudinal faults, e.g. the source springs of Mlava, Quv = 1.73 m3/s (6); Mrljis, 0.6 m3/s (7); Ozren and Dcvica, and karst springs of Svrljiske Planine. Flow directions of ground water are controlled by the geological fabric, karstification level, and local hydrogeological situation. They are mostly geneti cally determined by the karstification progress. Tectonic features and limestone structure arc the principal factors of karstification and ground water circulation. The first stage, which has the strongest influence on the subsequent ground water circulation, is certainly the uplifting and the formation of lime stone massifs. Longitudinal faults of NW-SE trend (e.g. Senje-Krepoljin and Zlot faults) determined the general directions of ground water movement toward the regional erosional base level th e Tertiary basins of Mlava in the north and Soko Banja in the south. In the following stage of karsti fication, periclinal inclination of the terrain, dispo sition of impermeable rocks, and chiefly the trans verse r egio nal fault s of E W (NE-SW) trend (e.g. the faults of Ml a v a, Lukovo, and Beljanica) con trolled the paleo-hydrographic pattern (radial type) and defined the position of the regional erosional base level. In the subsequent stages of evolution karstification became a dominant process as com pared to fluvial erosi on, which re s ulted in complete degradation of the river systems and lowering of the water table to a greater depth, but still pre serving the same flow direction as the pa leostreams controlled by transverse -fractures. The recent stage of karstification processes (including the present time) is characterized by the redistri bu tio n of th e drainage between some of the ad joining areas, as a result of gravity flows toward the regional erosional base level (controlled by longitudinal faults), abandon of local base levels and disappearance of minor catchments (including frequent ground water piracy). In this respect, we point out that the tracer, injected during high water flow in the ponors of Velika Brezovica karst area of the Kucaj-Beljanica massif, which is drained by the Grza spring (as the local erosional base level controlled by a transverse fault), appeared at Ra dovansko spring (regional base level), only in very low concentration at Grza source spring; during low water flow, the tracer appeared only at the Radovansko spring. Like karstification, the redistribution of ground water and circulation toward the regional erosional base level are neither simultaneous nor uniform over the entire massif This assertion is illustrated by Veliko Vrelo and Malo Vrelo in the Resava valley on Beljanica mountain, related to the ground water piracy along the transverse Beljanica fault, which is now partly included in the Mlava drainage area (recent regional erosional base level) The increasing natural rerouting of the drainage to the north and the south, at the pres ent stage of karst processes results in the largely increased. relatively uniform outflows, recorded in the areas of the regional erosional base levels (e .g. the Mlava, Mrljis). CONCLUSION Relying on an analysis of the regional fault pattern and its correlation with the main drainage zones il has been inferred that the areal distribution and interrelation of longitudinal and transverse faults hav e had a significant control over th e preferential directions of karst ground water circulation at differ&nt stages of karstification. The general direc tions of ground water circulation toward the ero sional regional base lev el have been controlled mainly by longitudinal fault structures

PAGE 155

154 REFERENCES KRESIC, N. & PAVLOVIC, R. (1990) Daljinska detek cija kao metod odredjivanja pravaca kretanja podzemnih voda u karstu. Nas Krs, 28-29, pp. 21-34. . PAVLOVIC, R (1990) Kompleksna analiza reljefa kao metod geoloskog istrazivanja Doct. dissert., F aculty of Min & Geol., Belgrade PAVLOVIC, R. & KRESIC, N. (1990) Remote sensing in hydrogeological research of eastern Serbia karst. Ref Intern Symposium "Remote sensing and water resources ", Enshede, The Nether lands STE VANOVIC, Z (1991 ) Hy dro geology karst aq u ( fer of Carpatho Balkanides of easten Serbia and water supply possi bilities. Monograph. sp. ed of Fac of Mining and Geology Inst. of Hy drogeology Belgrade 245 p. Z. Stevanovic et al. STEVANOVIC, Z. (1995) Karst ground waters of bia-Present and potential use in regional water supply. In: Lithospheric Water Mineral Re sources of Serbia Sp. ed. Fac. of Mining and Ge ology Inst. of Hydrogeology, Belgrade pp 77-119 STEVANOVIC, Z. & DRAGISIC, V. (1992) Directions of circulation of karst ground waters in the CarSerbia. In: Tracer Hydrology (H. HOTZL & A. WERNER Eds.), Balkema Rotterdam pp. 291-295. STEVANOVIC, Z. & FILIPOVIC, B. (1994) Ground waters in carbonate rocks of the Carpathian Balkan mountain r ange. Sp. ed CBGA Alston Holding Co. Jersey 237 p.

PAGE 156

Theoretical and Applied Karstology, vol. 911996, pp. 155-163 Tapping and Protection of Maljen on Mount Durmitor, Yuu.oslavia Budimir FILIPOVIC, Milojko LAZIC, Mladen VIDOVIC & Branislav SEKULOVIC1 Karst springs Veliki and Mali Maljen emerge, 30-35m apart at the foot of one of the slopes of the mountain Durmitor al the altitude of 1320 m, approximately 15 km south of Zabljak, one of the largest winter resorts in Montenegro and Yugoslavia. The springs drain a vast karst reservoir consisting of Triassic carbonate rocks: they are of ascending type with the outflow zones masked by moraine material. The spring Veliki Maljen is tapped, whereas Mali Maljen still runs through glacial material. The discharge of the spring varies between a maximum of 150-200 Vs and 35-40 Vs indicating a characteristic karst regime. According to their physical and chemical properties, the waters are of exquisite quality and therefore they absolutely meet any norms or regulations for bottling, that is foreseen in the future. In order to determine the conditions of exploitation and of preservation of the water qualities, a necessary hydrogeological research has been carried out. It has been performed according to the regulations for establishing the boundaries of the sanitary protection zones for maintaining the necessary continuity during the building and the operation of the plant for bottling the natural drinking waters that flow through Veliki Maljen spring. Key words: karst, exploitation, protection, karst spring. Conditions de captage et de protection de Ia source Veliki Maljen a Durmitor, Yougosla vie RESUME Au pied de l'un des nombreux versants de La montagne de Durmitor, a !'altitude d'environ 1320 m, jaillissent sources karstiques Veliki et Mali Maljen, eloignees l'une de ['autre d'environ 30-35 m. Les sources se trouvent a environ 15 km sud de Zabljak, l 'un des plus grands centres touristiques du Montenegro et de La Yougoslavie. Les sources drainent des eaux karstiques provenant d'un grand collecteur compose des calcaires triasiques. L'ecoulement de l'eau des sources est ascendant, La place d'apparition de l'eau etant masquee par des mate riaux morainiques. La source Veliki Maljen est deja captee, tandis que Mali Maljen s'ecoule toujours a travers le materiel glacial. Le debit de La source varie de 150 a 200 1/s en maximum a 35 a 40 1/s en minimum, ce qui correspond a un regime typiquement karstique. Compte tenu des nombreuses analyses chimiques effectuees jus qu 'a present, on peut dire que les qualites physiques et chimiques des eaux, sont exceptionnelles et satisfont toutes les normes qui conditionnent leur embouteillagt, ce qui est prevu en perspective. Pour determiner les conditions dans lesquelles it faut exploiter et maintenir [a qualite des eaux, ont ete effectuees des recherches hy drogeologiques d'apres les normes pour La definition des limites, des zones de securite sanitaire, a.fin de reali ser une continuite necessaire pour achever La construction de l'usine d'embouteillage des eaux naturelles et potables qui s'ecoulent par La source Veliki Maljen. Mots-cles: karst, captage, securite, source. 1 Faculty of Mining and Geology, Institute of Hydrogeology, Djusina 7, 11000 Belgrade, Yugoslavia.

PAGE 157

156 INTRODUCTION About 80% of the Montenegrin territory is karstland. Many karst springs discharge ground water of good quality for multi-purpose use, about 80% of which is at present supplied to various types of users Many of these springs discharging healthy water, often far from urban centers and impossible to be used in municipal water supply, have remained uncontaminated and preserved for bottling. The scarcity of hydrogeological data on remote karst areas is a consequence of their re duced or inadequate study. This is particularly true for the mountain massifs of Durmitor and Sinja jevina. Durmitor is a mountain of the High Dinarides, a climatic boundary between the Mediterranean and the inland zones. Some 15 km south of Zabljak, one of the major winter-sport centers of Montene gro and Yugoslavia (Fig. 1), or about 5 km beyond the southern perimeter of the Durmitor National Park, Veliki Maljen and Mali Maljen karst springs emerge at the foot of one of many slopes of Mount Maljen at the altitude of 1320 m. The environment of the resurgences is unspoilt and healthy, in an area abounding in small and big streams and springs In order to define the conditions of the spring in take and of preservation of the quality of the water for bottling, extensive hydrogeological exploration was carried out and the zones of sanitary protec tion outlined GENERAL CHARACTER OF THE VELIKI AND MALl MALJEN SPRING AREA The larg e extent and marked vertical differentia tion o f the Durmitor mountain massif is responsi ble for it s notable climatic variation Climate char a ct e ristic s howev e r, ar e not well studi e d due to the lack of corresponding measurements and rec ords. The climate is that of a mountain region: short, cool summers and long, cold winters; high, non uni form annual precipitation ; abundant pre cip i t a tion s and low air t e mperatures in a utumn and wint er months, and muc h l e ss abund a nt r ainfall in the short cool summ e rs. Gen e rally the annual rainfall at high altitudes (ove r 1400 m) of Durmitor a mounts 2000 mm; the mean monthly maximum m a y exceed 517 mm in the a utu m n At low e r altitude s 1300-150 0 m, the long t erm a v era g e is w ithin th e rang e 1400-B. Filipovic et al. 1800 mm. The highest amounts of rainfall, about 2/3 of the total, fall in the autumn and winter sea sons, while the smallest quantity (about 1/3 of the total annual amount) is recorded during summer. A large part of the precipitations is represented by snow, which makes Durmitor one of the highest and longest snow-covered mountains of Yugosla via. Mean monthly air temperatures vary from 5 4 octo 14 C. The climatic characteristics of Durmitor have a very important hydrogeological impact, because high precipitations are one of the main water budget elements. Geomorphologically speaking, the area of Veliki and Mali Maljen springs is a part of the Durmitor mountain massif. The landforms belong to karst, fluvioglacial and fluvial types, resulting from the corresponding processes. The terrain is dominantly composed of Triassic lime stones, with typical karst features, among which frequent sinkholes and uvalas may be noticed, particularly upstream of Maljen springs, in Buk ovicka Gora area Karstification processes are enhanced by the almost complete absence of vegetation in the region. Besides the karst forms in the general Gornja Bukovica area, large deposits of fluvioglacial ma terial lie along the Bukovica river. This river has carved its valley in the fluvioglacial material of heterogeneous composition (very large boulders to fine-grained limestone arenites). Bukovica valley is the lowest-positioned (extending from 1372 m in its upper reaches to about 1200 m at Donja Bukovica). Mountains on eith e r side of the valley rise to ov e r 1600 metres. T he surface configura tion is rugged, intersected by many steep-sided gullies of the mountain torrents that flow after heavy rainfall or rapid snow m e lting. Geologically (see Fig. 2), the oldest rocks in the study area are of Permi a n age, and con s ist of fly s ch s chists, micaceous sand s tones, con g lomer ates, and limestones. The deposits overlying the Permian series are the Lower Triassic micaceous sandstones, schists, marls, limestones, and marly dolomites with a total thickness of about 2 00 m. Lower Triassic rocks are directly overlain by a s e quence o f Middl e Tri a ssi c lim e ston e s a nd dolo mitic limeston es about 350 m t hick The deposition of Middle Triassic limestones was followed by rhyolite intrusion along the Bukovica valley, and the s ub s equent partial metamorp hi sm o f th e Triassic rocks. Th e rhyolit e intrusion wa s p ara ll ell e d to th e d e p os ition o f th e Mi d dl e T ri ass i c tu ffs and ch e rt s follow e d by lim e ston e s with

PAGE 158

Tapping and protection of Maljen Spring Flg .l. J of Durritltor,;; ; /:-. Positi'bn de La montagi;(/de '' chert and th e massive reef limeston es of the Upper Triassic. The thickne ss of the entire Middle and U pper Triassic carbonate series is about I 000 m ; it is intensely folded and faulted. Upper Juras sic limestones and Creta ce ous Pal e o gene 11ysch deposits in the explored area arc p artly overthrust by the Triass ic series. The youngest rocks are glacial and diluvial depos its in the Bukovica vall e y Gl ac ial morainal mate rial is very heterogeneous and loc a lly it has a thickness of several tens of meters. T h e rocks in 157 the area of Gornja Bukovica have been strongly crushed during vigorous tectonic events, which included thrusting of Triassic series. first over the Upper Jurassic, then over the younger Cretac e ous Paleogene unit s, and heavy faulting, particularly in Donja Bukovica zone The hydrogeolo g y of th e re gidn is much inl1ucnced by the t ec tonic pattern. Preferential flow paths o f gro und water arc along faults and fractures; this, and other hydro geo logic factors r es ult e d in the occurrence of many karst springs down th e Bukovica valley.

PAGE 159

2(XX) 1000 A 0-l Alluvial d..:posit __ ____ Mominal deposit -basal breccia __ _I and konglomcrutc -strati ficd limcston..:s -Thick and bedded limestones L ... T1 2 200) Fig11re 2: Geological section trough Veliki Maljen spring (1 :25 000) Figure 2: Co11pe geologique au-tle.\.o;us tie la source Veliki Malje11 (1:25 000) -13cddec.l brown, purpk or reed -Stratifi..:s r massiv..: gray lim..:ston..:s limcston..:, dolomitic limeston..:, I and lh>lumile J -Qumtz-micaccous sandston..:, ai;J --Siratificd mnrly limestone, 3 doll>mitil: limesh>ne and dnlomitc Tt schist, siltstone, marlstone, t-_:-;-----limestone, and marly dolomite lhiThJ [--? i/T-J -Rhyolite (hydrogeological -massive limestone barrier) -Limestone and cheltllt>dules, j :1;;;1:; Purple phyllite schist, quartzchert. and r..:eflimcslt>nc mi CiiCC<>Us sandstone, -. --I 2: conglnmcratc-and limestnne Vnlcanic--sedimenlary 1\>rnmli>n: IU ff. luffac..:ous, sandsh>ne. d1e1t, I. ..... .1L bentonite. and limesl<>nc 500 0 B otlter symbols : -11<>rmul boundary eros/teet. boundary -filUit -nappe front general ground water flll'IIV direction -til QO .g: a ..: Q :""'

PAGE 160

Tapping and protection of Maljen Spring HYDROGEOLOGY OF THE VELIKI MALJEN SPRING AREA In the general area of Gornja Bukovica, there are two types of reservoirs and almost impervious rocks (Fig. 3), namely: the karst aquifer of the Triassic and Upper Jurassic limestones; the aquifer of the alluvial and moraine depos its in the Bukovica river valley; virtually impervious flysch deposits of Cretaceous Paleogene age. The subject of our study is the ground water origi nating from Triassic and Upper Jurassic rocks, or the karst aquifer. The other two hydrogeologic media will not be discussed in this paper. KARST GROUND WATER IN TRIASS I C AND UPPER JURASSIC LIMESTONES A series of Triassic and Jurassic carbonate rocks outcrop over a large part of the study area It plunges locally under till deposits, in the Bukovica river valley to the north and south-east. The thick (about 1000 m) Triassic deposits and the 450 m thick Upper Jurassic limestones form a significant reservoir of ground water. The high degree of kar stification and fracturing of the limestones and the previously mentioned geometry add to the perme ability of rocks (ALIPOVIC & LAZIC, 1995) The large aquifer is recharged mostly by infiltra tion of precipitations, which are most abundant in the autumn and in the late spring, subsequent to the melting of the thick snow cover. Rainfall is far less abundant in the summer, but then the effective percolation into lower aquifer levels is the highest. Ground water is drained by a multitude of karst s pring s, to mention the largest ones: Glava Buk ovice, Veliki and Mali Maljen, Krvavac, Dzedzekovac, and many smaller sources The dis charge of the largest one, Glava Bukovice, varies from th e maximum o f over one thousand to the minimum of some hundr e d 1/s. The second largest in size and discharge is Veliki and Mali Maljen of 150-200 1/s maximum and only 35-40 1/s mini mum outflows, typical for karst springs Unlike the left side of the Bukovica vall e y, the right side abounds in natural resurg e nces, which indicat es the probabl e direction of w a ter flow, from south west to north east. Thi s hypothesis is also s ustained by the presumed hydrol ogic barrier of Cretaceous-Paleogene flysch, which generally 159 directs ground water flow toward the Bukovica valley and further to the Tara river, the regional erosional base level. The ground water outflow on the margin of the glacial deposits in the Bukovica valley is primarily controlled by the Lower Triassic basal sandstones, thick heterogeneous deposits of till, and probably also by the buried rhyolite and dacito -and esite intrusions which are characteristic in areas of ex posed Lower Triassic rocks. Future explorations aiming at the verification of the above-mentioned suppositions, definition of the aquifer extent, the rates of ground water flow and the possible deviations at different water ta bles, and a more accurate definition of the drain age area, will certainly include water flow tracing tests and other exploratory activities VELIKI AND MAL l MALJEN SPRING DISCHARGE AND DE VE L OPM E NT PROSPECTS The karst springs of Veliki Maljen and Mali Maljen drain an abundant ground water reservoir in karstified Triassic limestones of Bukovicka Gora (Fig 4). Topographically, the drainage area is reduced ( -10 km2). The flow rates of both spri ngs, however, suggest a larger drainage area, probably due to ground water divides, but this has to be verified by tracing tests Water emerges at both springs from glacial material; the resurgences are masked by till. The springs are located 30-35 m apart, the Mali Maljen some 2-3m lower than th e other Available records and a large number of complete physical and chemical analyses of water from both springs indicate its very good quality. Water temperature is quite stable over the year, 5-5.5 C, equivalent to the mean annual air tem perature. Total dissolved solids cont e nt is low, as a result of high water circulation in the karst. The water type is carbohydrate-potassium with 150-250 mg/1 dry residue. The contents of other ground water components are, both in species and amounts, lower than the permissible concen tra tions. Veliki Maljen is tapped and used in a new wat erbottling plant. Mali Maljen is still naturally flowing, probably planned for the expansio n of the new plant. The exca vation for building th e V e liki Maljen intake and stripping of the thick till deposit revealed that the several apparent outflows were actually one spring which spread throu g h the de posit. The situation at Mali Maljen is probably similar. V e liki Maljen is an ascending spring. Ground w a ter ascends through a system of fractures and

PAGE 161

160 B. filipoviC et al Figure 3: Hydrogeological map of Veliki Maljen spring general drainage area and protection zone (1 :25 000) Carte hydrogeologique presentee des environs plus larges de Ia region de Ia source V.Maljen avec /es zones de securite sanitaire (1 :25 000) [ 0 0 0 l al I -Alluvial deposit lr I Hfr I -mass ive limestone +f 2,3 symbols: -normal bounJary -Morainal deposit Fly5ch : breccia and conglomerate -Thick and bedded limestone s I I Zone or immediate prote ctio n -and chert nodules, chert, and reef limestone Volcanic -sedimentary formation : tuff, tuffaceous, sHndstone, chert, bentonite. and limestone massive gray limestone, dolomitic limestone, and dolomite -Quartz-micaceous sandstone, schist, siltstone, marlstonc:, lime!ltone. and morly dolomite PROTEcnON ZONES eros/teet. boundary -rault ........_...... -nappe fiont Key of !lydrogeological symbals: -spring of I 002 00 1/s V J\.faljen, spnngs of I 0 -100 1/5 general ground water now A ___ B hydrogeol.ogic .ul cross-sectu.m I me: II Protection z one proper Ul Pro tecti on zone gcncrn l NOTE: Dnrnfur period jirmr a /99.1

PAGE 162

Tapping (Jnd_prQtection of Maljf!n !$pring WSW fissures, sl.lggesting a deep flooded circulation. An impermeable barrier, probably concealed by gla cial material, is preventing drainage to lower levels, as the spring emerges some 30-40 m above the erosional base level of the Bukova rivet bed. The hypothesis of an impermeable bariier is based on the presence of a series of Lower Tiiassic rocks with its basal part composed maiirly of low permeable sandstones, schists and marlstones, and quite likely. along the Bukovica valley, of rhyolite or which are responsible for the metamorphism underwent by the Lower Triassic limestones in .this zone. Marble is not so easily karstified as limestone, and it forms a barrier to percolation within the non-uniforinly kiirstified limestones. The inflow water cannot easily and rapidly drain into deeper levels; therefore it is diverted and overflows the barrier, to finally as cend to the surface. The intake, constructed at the spring itself, pro tects the latter from direct contamination and fits well in the natUral envhonment. H is provided with two outlets corresponding to the future plant A chamiel (about 35 mlong) leads from one of the outlets to the plant site.-The other outlet will dis charge surplus water in periods of high water ta ble, or will carry an additional amount of water to a small fish-pond considered for being built in the future 161 nu.knv. ic:t PROTECTION OF VELIKI MALJEN SPRING Where ground water domestic use is planned, par ticular consideration is given to the local hydro geological situation for the definition of the pro tection zone. The principal interventions at Veliki Maljen spring are prevention works against the possible poll)ltion of ground water at the intake and in its recharge zones, under both natural and artificial conditions (FILIPOVIC & VUJASI NOVIC, 1982). Generally speaking, a karst aquifer is difficult to protect, because karst channels, caverns and fissures, as the privileged paths of prevailingly tur bulent flow, are numerous and large. ProteCtion of karst springs and of their intakes in the specific hydrogeologic conditions is a complicated prob lem which calls for an equally serious approach The first step in the case of Maljen spring was the delineation of its drainage area, of its recharge zone, and zone of preferential flows in the area This was necessary, because the fast ground water flows, typical for karst, may rapidly carry pollut ants from remote parts of the catchment to the resurgence zone. For the identification of the sanitary protection zone, the regulatory statutes were observed, and in

PAGE 163

162 this respect consideration was given, at the present level of our knowledge, not only to the geological and hydrogeological situation, but also to other relevant factors, such as woods, size and type of vegetation cover, records of human activities and possible sources of pollution. The available infor mation was used only in a preliminary design of the sanitary protection zones II and III (see Fig. 3), before a detailed hydrogeologic study being made. SANITARY PROTECTION ZONE I OR IMMEDIATE PROTECTION ZONE The zone of direct or strict protection is designed to cover the intake, its immediate surrounding, and the supply line to the plant over a distance of 35 m. This zone is intended to prevent uninten tional or intentional pollution through the intake structure. Its size and shape are therefore small related to the hydrogeological situation, but rather to the types and layout of water intake structures. The intake is well designed, considering both the architectural and functional criteria. It is built over the spring itself, protecting it from direct surface pollution and isolating it from the subsurface envi ronment. The water intake-bottling plant system will also be closed and physically protected. The intake site and the area overlooking it is amphi theatred and forested with beech trees. This zone includes the immediate intake area of 30m on either side and up to 50 m uphill. Its shape is ar with a total area of 1000 m2 It will also include a 2.5 m-wide belt on either side of the supply line from the intake to the plant. The immediate protection zone will be sub ject to statutory and sanitary regulations which will not permit any activity that may affect the quality of the bottled water, including trespassing or prolonged presence in the intake area. This area will be fenced, and the service premises issued as provided by the sanitary regulations. SANITARY PROTECTION ZONE II OR ACTUAL PROTECTION ZONE The actual sanitary protection zone, which extends from zone I upstream of the Veliki Maljen spring, must be defined as accurately as possible in rela tion to the local hydrology. Boundaries of such a zone in karst areas are di rectly dependent on the resurgence mechanism. Because this is an ascending spring, suggesting deep flooded circulation, the limits of the proper protection zone will be reliable only if they en compass terrains that are more elevated hypso-B. Filipovit et al. metrically than the outflow, up to 1000 m in the spring background. The proposed actual protection zone seems sufficiently safe for the spring itself, since it is uninhabited and forested. Naturally, regulatory statutes on the behavior of persons vis iting or performing in this zone any activity that may threaten the water quality will have to be strictly observed. This primarily refers to the in terdiction or the control of the construction, cattle or sheep breeding or similar purposes facilities. The zone must be properly marked, but not neces sarily fenced. If any structure of some potential pollution hazard exists, it will be dismantled and the possible contamination sources will be removed. PROTECTION ZONE Ill OR GENERAL VELIKI MALJEN SPRING PROTECTION ZONE In the projects of exploration or design of tapping natural karst ground water as springflow or by means of artificial wells, the extent of zone III, or the zone of general sanitary protection has been usually selected to coincide with the real drainage area. Topographic and hydrogeological water di vides are often non-coincident in karst regions. The available data for the Veliki Maljen spring indicate a far greater hydrogeological catchment area as compared with the topographic one; this is indicated by the ascending type of the spring and its relatively high discharge, compared to the topo graphic drainage area. Similarly, it is not always possible in karstland to delimit the actual and the general zones, therefore only zones I and II are designed. At any rate, the zones that should be identified and protected are sink-areas or zones of increased underground percolation (uvalas, sink holes etc.), regardless of their proximity to the spring, because they are directly communicating with the karst aquifer. Any direct or indirect source of pollution in such zones is a direct threat to the ground water. The present situation, as far as protection of karst aquifer in the spring drainage area is concerned, is good, and the presence of woods and the absence of settlements are additional favorable factors. In order to preserve the present situation, each major activity in the area should be well planned and its effect on the discharging water properly monitored. At any rate, development of industrial activities, animal farms, or similar projects should not be allowed, particularly in the zones of infiltration and surface percolation of water.

PAGE 164

Tapping and protection of MaljenSpring SUMMARY This work presents general and hydrogeological chara cteristics of the Veliki Maljen spring area and its surroundings. This spring water is ph1.nned to be bottled for commercial use. The natural and artificial intervention conditions are explained. Ground water is ascending from the Lower Triassic limestones and is mainly accumu lated in the main reservoir located within the Mid dle Triassic limestone series. REFERENCES FILIPOVIC, B. & VUJASINOVIC, S. (1982) Zastita podzemnih voda Fatcultet, Beograd. 163 Field prospectin g and compiled information were used to establish the zones of sanitary protection at the preliminary engineering report level. Three zones are defined in conformity with the regula tory statutes, and activities and works are proposed for the protection of the Veliki Maljeil spring and of the quality of its water. Finally, detail hydrogeological investigations are expected to provide data for a more reliable de lineation of the proposed sanitary protection zones. FILIPOVIC, B. & LAZIC, M. (1995) Odredivanje zona .wnitarne zastite oko iZ\'Ora v. Maljen u selu Gornja Bukovica na Durmitoru Savnik). za potrebe flasiranja prirodne pijace vnde. Projekat, Fakultel, Institut za Hidrogeologiju, Beograd

PAGE 166

ABSTRACT Theoretical & Applied Karstology, vol. 9/1996, pp. 165-173 Hydrochemical Behavior of Karstic and Evaporitic Formations Surrounding Sarvestan Plain, Iran Ezzat RAEISI, Omid JE H BEZ, & Farid M OOR E1 The Sarvestan Plain is located in the central-southern Iran The plain is surrounded by extensive karst carbonate deposits with minor amounts of evaporitic-argillitic sediments. These Mesozoic and Cenozoic formations are intruded by two Precambrian salt domes to the northeast and southeast of the plain. Despite the presence of kar stic formations in the drainage basin of the plain, the quality of the ground water is very poor In this study the lithological characteristics and mineral constituents of the surrounding formations are investi gated by optical and chemical methods. During the summer 1992, the electrical conductivities of 500 operational wells, 15 springs, and 4 qanats were measured Furthermore, 128 representative samples were analyzed formajor ions including calcium, magnesium, sodium, potassium, chloride, bicarbonate, and sulfate using standard wet chemical methods. Temperature and pH of all the sampling sites were also measured. Using the WATQEF com puter program, the saturation indices in various minerals were calculated. Results indicate that although the drainage basin of the plain consists mostly of carbonate formations, except for a narrow band in the south and a small region in the northeast of the plain where the type of water is bicarbon ate, the rest of the aquifer exhibits a chloride-sulfate type water The main sources of the chlorine and sodium ions are shown to be the exposed salt domes, and the sources of the sulfate are the dissolution of gypsum and anhydrite along with the oxidation of pyrite in the argillaceous formations. The reason for varying ionic compo sitions in different parts of the aquifer is discussed. The main reason for the poor quality of groundwater is ap parently the mixing of the karstic water with water expelled mainly from salt domes and to a lesser extent from evaporitic and argillitic units. The higher solubility of halite minerals compared with that of carbonates, contrib utes to the poor quality of the groundwater. Key words: karst aquifer, water quality. Le comportement hydrochimique de Ia plaine de Sarvestan entouree par des formations karstique e t evaporitiques RESUME La Plaine de Sarvestan est situee au sud-ouest de La province Fars en Iran. Lil plaine est entouree par les dep ots karstiques-carbonates ave c une quantite mineure des sediments eva poritiques -argileux. Au nord et sud-es t de La plaine, ces formations Mesozoi'ques et Cenoz.oiques sont traversees par deux domes de sel precambriens. Malgr e La presen ce des formations karstiques1dans le bassin de drainage de La plaine, La qua lite des eaux souterraines est tres pauvre Dans cette etude Les caracteristiques Lithologiques et les composants mine raux des formations sont etudies par l'intermediaire des methodes optiques et chimiques. Durant l'ete /992, La conductivite electri que de 500 puits operationnels. /5 sources, et 4 qanats a ete mesuree De plus, 128 echantillons representatifs ont ete analyses pour des ions majeurs. respectivement Ca. Mg. Na, K. Cl HC03 et SO" En utilisant des methodes chimiqu e s humides standard, on a determine aussi La temperature et le pH de taus les echantillons des sites L 'indice de saturation des differents mineraux a ete cal cule par /'utilisation du logiciel WATQEF. Bien que La majeur partie 1 Shiraz University, College of Sciences, Department of Geology, Shiraz, Iran.

PAGE 167

166 E. Raeisi et al. du bassin de drainage delaplaine consiste en formations carbonatiques, a ['exception d'un etroit terrain ou le type de l'eau carbonate, le reste de l'aquifere contient de l'eau de type chlorure-sulfate. Jl a ete montre que les sources des wns de chlorure et de sodium sont des domes sates, et que la source du sulfate est La dissolution du et et aussi l'oxydation de la pyrite dans des formations argileuses. La variation des comP.o:Ltzons wmques dans des parties de ['aquifere a ete discutee. La raison principale pour La P_auvre des eaux souterrames est apparemment le melange des eaux karstiques avec des eaux eliminees les domes sates et dans une certaine mesure par des unites evaporitiques et argile uses. La solu btltte plus elevee mineraux de chlorure de sodium par rapport au carbonate contribue a La qualite pauvre des eaux souterrames. Mots cles: aquifore karstique, qualite des eaux. INTRODUCTION Even though several factors such as topography, climate, vegetation and time, control the hydro chemistry of groundwater (DREYER, 1988), the main reason behind the observed differences in groundwater chemistry is rock lithology (GARRELS, 1967; GARRELS & MACKENZIE, 1967; FLINT, 1971; PALMER & CHERRY, 1985; JOHNSON, 1985; ROGERS, 1989 and SIGURDSSON, 1993). The chemical composition of groundwater can indicate its origin and the his tory of the underground material the water has been in contact with. Water draining out of lime stone and dolomite often has calcium, magnesium and bicarbonate as the only significant solutes, although sulfate is commonly present from associ ated gypsum or pyrite oxidation. The amount of the total dissolved solids is limited by the solubil ity of the carbonate minerals. Evaporites weather by simple dissolution and can give rise to water of very high salinity. Typically, the main anions are sulfate. and/or chloride, although this may vary, dependmg on the nature of the evaporite. The geo chemistry of groundwater is largely a function of the mineral composition of the aquifer, but the quality of groundwater can differ significantly over a relatively short distance, for example at the boundary between different formations. Karst waters in the Zagrosides of Iran which are not in direct contact with salt domes or gypsum evaporite formations have good quality. In general the electrical conductivities are less than 500 110/cm (RAEISI & MOORE, 1993a). The analysis of 119 karst springs in the Zagrosides of Iran indi cates that 85.7% of the total dissolved ions consist of calcium, magnesium and bicarbonate, and the rest includes sulfate, chloride, sodium and potasin that order. There are about 200 salt plugs m the southern part of the Zagros Mountain Ranges in Iran and the Persian Gulf (KENT, 1970). These salt plugs have risen from a depth of more than 4,000 m. On some occasions they protrude through karst aquifers and deteriorate the good quality of karst water (SHARAFI et al., 1996). Sarvestan Plain is surrounded by karst for mations, but the quality of groundwater is poor. The objectives of this study are to establish: first the water quality of karst and non-karst forma tions, second, the main source responsible for dete riorating karst water quality and third, the method ology of preventing contamination of karst water. GEOLOGICAL SETTING The studied area falls within the simply folded belt of the Zagros Orogeny. The stratigraphy and structural characteristics of the Zagros sedimen tary sequence have been described by JAMES & WYND (1965) and FALCON (1974). The geo logical map of the study area is presented in Fig ure 1. The Sarvestan syncline basin which trends NW-SW, is bordered by the Maharlu Salt Lake and three anticlines, namely Gar, Ahamadi and Panel (Fig. 1). The exposed formations are domi nantly made of Gurpi (Santonian-Maestrichtian), Tarbur (Late Campanian-Maestrichtian), Sachun (Late Maestrichtian to Early Eocene), AsmariJahrum (Paleocene to Early Miocene), Razak (Early Miocene), Aghajari (Late Miocene-Plio cene) and Bakhtiari (Late Pliocene) formations. Two prominent salt domes of infra-Cambrian age (Hormuz formation) protrude in the Ahmadi and Panel Anticlines. Four major faults occur in the region. The one in the east has resulted in the outcropping of the Sa chun, Tarbur and Gurpi formations in the core of the Panel Anticline. The Asmari-Jahrum in the western flank has almost disappeared. The two salt domes protruded along this fault (Fig. 1). The Mount Ahmadi Faults have played a significant role in exposing the Sachun formation especially in the southern, and the Tarbur formation in the east ern parts of the anticline. The Gar Fault exposed the Gurpi formation in the core of the anticline,

PAGE 168

D r::::::l F.3 F-?3 e;j 8 [I] ._ C:J. 0 l( Alluvium Bakh\Yari Aghajari Razak Asmari-Jahru m Gur p 1 Sac hun hrbur Bang rstan Salt dome Fault Anti cline Syncline Spring HC03 so4 Cl Wate r type Bound<&ry -s--<1 -i> E! 2 -;:---= c; ., ., :: 15.. --= {; Q ., 'Q> ::::. -s ;:: :.: --.l

PAGE 169

168 and in some parts of the northern flank, the As mari-Jahrum formation is so shattered that it is either completely removed or visible as small blocks. Ghale Gorikhte syncline is composed of Bakhtiari Conglomerate, left by differential ero sion in the north-west of the plain. Several alluvial fans are located at the foot of the three mentioned anticlines, which are recharged by run-off from the high altitude area. These alluvial fans were found to consist of coarse grained ero sion products of Asmari-Jahrum and/or Tarbur formations, along with some fine-grained material of Sachun or Gurpi formations. The Sarvestan aquifer is made of Quaternary alluvium that con sists of medium to fine-grained clastic material brought about by stream and flood wash of the surrounding geological units. A decrease in grain size is observed from the edges to the center of the aquifer. The topographic elevations range from about 1460 m above mean sea level (m.s.l.) in the basin near Maharlu Lake to 2470 m m.s.l. at the crest of the Ahmadi Anticline. The annual average rainfall and temperature are 412 mm and 18.2 oc respec tively. METHOD OF STUDY In this study the lithological characteristics and mineral constituents of the surrounding formations are investigated by optical (79 rock samples) and chemical (48 rock samples) methods. The cal conductivity of 500 operational wells, 15 springs and 4 qanats were measured at the sites. Furthermore, 128 representative samples were analyzed for major ions including calcium, mag nesium, sodium, potassium, bicarbonate, sulfate and chloride using standard wet chemical methods. Temperature and pH of all representative samples were measured at the sites. Saturation indices of major minerals were calculated using the WATQEF model (PLUMMER et al., 1984). HYDROCHEMICAL CHARACTERISTICS OF SARVESTAN PLAIN RAEISI & MOORE (1993b) studied the hydro logical behavior of Sarvestan Plain. The isodepth map reveals that the depth of water table varies from zero at the boundary of Maharlu Lake to 36 m near the plain boundary The isopotential map clearly indicates that the general direction of flow is down dip from south-east to north-west E. Raeisi et al. (sloping down from the foot of the mountains to the plain and from the plain toward the Maharlu Lake). The overall direction of flow seems to break into two separate components near Maharlu Lake, moving to the south and north of the main flow. A more elaborate map with contour intervals of 2 m showed that the individual components of flow from the plain, mountain slopes and Maharlu Lake are directed towards the area of maximum concentration of production wells. RAEISI & MOORE (1993b) suggested that the magnesium content of groundwater can be used as an indicator for the intrusion of Maharlu Lake water into the aquifer. ZAK & GAT (1975) used isotopic C80 and 2H) techniques to determine the origin of wa ter in the Sarvestan area. They concluded that the origin of groundwater is locally recharged water. The isoelectrical conductivity map of the Sarvestan Aquifer was prepared by measuring the electrical conductivity (EC) of 500 operating wells (Fig. 2). The results of 128 chemical analyses of groundwater resources were plotted on a trilinear diagram and the types of water were determined (Fig. 1). The Sarvestan Plain contains a combina tion of bicarbonate, sulfate and chloride water. HYDROGEOLOGICAL CHARACTERISTICS OF THE MAIN FORMATIONS GURPI FORMATION The Gurpi formation is 320 m thick and consists mainly of soft, bluish-grey marl and shale with subordinate argillaceous limestone bands at the type locality. In the study area, the Gurpi forma tion overlies the Bangestan Groups and underlies the Tarbur formation. Optical studies show that it mainly consists of clay minerals and calcite, with small amounts of iron oxides and dolomite. The calcium and silica (Si02 ) constitute on the average, 24 and 14 percent of the rock samples respectively (Table 1). The texture is biomicrite with inter granular porosity. The Gurpi formation contains an abundance of index fossils such as Globotruncana bulloides, G. falsostuarti, Rugoglobigerina, Radiolaria and pelagic fauna. It is mostly exposed in the core of the Gar Anticline. The run-off from this formation infiltrates into the alluvial fans at the foot of the northern flank of Gar Anticline. The Gurpi formation with its marly and shaley lithol ogy is actually an aquitard, therefore it does not seem to affect the hydrochemistry of the scattered Asmari-Jahrum Aquifer at the northern flank of

PAGE 170

D El . t;ZT*l r-==-=3 Ed [J] Alluvium Bilkh\YlH l Aghajari Asmui-Jahru m G
PAGE 171

170 E. Ra.eisi i!.t al. .. Formation Minerals by optical method Average cheniicai anal uses by weight percent ... Ca! Ml!:!+ .. Na+ 1(+ .. .. AI]+. ... SiO MnO Fe,01 : Sarvak calcite with small amounts of 27.06 0.18 . 0.02 0.94 0.0 0.20 dolomite and clay minerals clay minerals and calcite with : Gurpi small amounts of iron oxides 24.16 0:30 1.65 Q.l7 0.01 13. 76 0.0 1..52 and dolomite Tarbur calcite with small amounts of 19.62 10.14 1.70 0.02 0.0 -0.034 0.20 clay minerals and iron oxides gypsum, anhydrite and clay Sachun minerals with small amounts 25.50 4.0 2.1 0.04 O.Q2 1.58 0.018 -0.68 of calcite & dolomite Asmaricalcite, dolomite 28.22 0.22 1.97 0.02 0:0 0.019 0.14 Jahrum .. -Razak clay minerals, quartz, calcite, 13.83 1.59 2.19 1.02 0.1 40.23 0.13 4 .83 feldspar, and iron oxide sandstone, quartz, chert, calcite, with small amounts of 1.70 0.01 43.37 0.059 2.69 Aghajari clay minerals, iron oxides, and 20.12 0.61 0.22 gypsum Table 2. Major electrial,ting from ; "L _; . ; . .. :- ... ;:.;.. Ions majeurs, conduetivite type t!el'ealf. pto ,velia7JI des formationsconsi
PAGE 172

Hydrochemical behavior of karstic and evaporitic formations the Gar Anticline. As there are no springs or oper ating wells within the Gurpi formation, its hydro chemical behavior is studied via the operating wells at the alluvial fan on the northern flank of the Gar Anticline (bicarbonate B region in Fig. 1). The average chemical analyses of this region are presented in Table 2. The electrical conductivity is less than 1000 ).1.0/cm and the water type is cal cium bicarbonate. This implies that the Gurpi for mation does not seriously affect the quality of the adjacent Asmari-Jahrum formation. SACHUN FORMATION The Sachun formation is 1400 m thick at the type locality. The lower 387 m consists of marls, marl stones and silty, white limestones. These are fol lowed by 91 m of gypsum and dolomite which are overlain by 112 m of madstone and dolomite. The succession is continued with 82 m of gypsum with bands of dolomite. Following this, are 335 m of chalky marlstones, dolomites and limestones. The upper 405 m is composed of massive gypsum, marl and ribs of dolomite. The lower contact with the underlying Tarbur formation and overlying Jahrum formation is conformable. The Sachun formation outcropping in the study area consists mainly of gypsum, anhydrite and clay minerals with small amounts of calcite, dolomite and iron nodules. The texture is argillitic micrite. No index fossils are observed in the thin sections. The Sa chun formation with gypsum and marl lithology could probably be considered as a weak aquitard. The rather high solubility of gypsum enhances the capacity of this formation to reduce the quality of groundwater. There are no springs or wells which directly discharge the Sachun groundwater. There fore, the hydrochemistry of this formation is stud ied by its effect on the neighboring formations such as the Asmari-Jahrum and Tarbur in the Panel Anticline. Springs 4, 5 and 6 which emerge from karst formations contain sulfate water instead of the expected carbonate water (Table 2 and Fig. 1). The seepage of Sachun water into the adjacent karst aquifer reduces the quality of karst water such that the electrical conductivity of springs range from 1030 to 1230 ).1.0/cm. Sulfate regions A, B, C, and D mostly originate from neighboring Sachun formations (Fig. 1). Due to the low solu bility of gypsum and anhydrite, the electrical con ductivity of sulfate regions is generally less than 4000 ).1.0/cm. 171 RAZAK FORMATION The Razak formation with a thickness of 805 m consists of red, gray and green silty inarls inter bedded with subordinate silty limestone and sand stone ribs at the type section. The lower contact of this formation with the underlying Asmari forma tion is gradual and conformable. In the Gar and Ahmadi Anticlines, exposure of the Razak forma tion is limited to a narrow area between the As mari-formation and Sarvestan Plain, and in most parts, Razak formation lays beneath the alluvial fans. The major minerals of Razak formation near the contact of Asmari are quartz, feldspar, calcite, clay minerals and iron oxides. The chemical analyses show that the exposed Razak formation has 40% silica, 14% calcium, and 5% ferric oxide (Table 1). In the absence of wells or springs in this forma tion, its hydrochemical behavior is studied by its effect on neighboring aquifers. Razak formation could be considered as an aquiclude or a poor aquitard. The carbonate water with low electrical conductivity in the alluvial fan (carbonate region A in Fig. I) reveals that the Razak formation be neath the alluvium does not have any considerable effects on reducing the quality of karst water. THE HORMUZ SALT FORMATION The Hormuz Salt formation outcrops mainly in the form of piercing salt domes. Many geologists studying the salt dome in the region have consid ered the buoyancy of the salt, the vertical pressure of several thousand meters of sedimentary cover, and the lateral pressure resulting from the opening of the Red Sea as the reasons for the salt diapric intrusion. The upward movement of the salt has been facilitated by faults. NABA VI & SABZEI (1987) (as quoted by HERA VI et al., 1990) be lieve that 70% of the salt domes in the south of Iran are located on several parallel faults. The exposed parts of the two salt domes in the Sarvestan Plain are predominantly made of halite with some minqr quantities of gypsum and other evaporite minerals. A small spring emerges from the Gar Anticline Salt Dome (number 3 on Fig. 1). As shown in Table 2, the EC of this spring is about 359000 ).1.0/cm and the concentration of sodium and chloride is about 3100 equivalent per million (epm). It is almost saturated with calcite, dolomite, gypsum and halite. The high values of dissolved solids indicate that the small amount of water out flowing from the salt domes has a great effect on decreasing the quality of large quantities of karst water. The effect of the salt domes on the quality

PAGE 173

172 of the water in the Sarvestan Plain can be observed by two chloride regions, cl(A) and cl(B), with high electrical conductivity at the entrance of the plain (Fig 1 and 2). ASMARI-JAHRUM FORMATION The Asmari formation is composed of feature forming well-jointed limestone at the type local ity. Its thickness ranges form a few meters to more than 518 m. The Jahrum formation consists of 467 m of dolomite and dolomitic limestone at the type section. The difficulty in differentiating between Asmari and Jahrum has led to the combined name of Asmari-Jahrum formation. The most exposed formations in the anticline surrounding the Sarvestan Plain are the Asmari-Jahrum. The opti cal method shows that the Asmari-Jahrum forma tion is composed of calcite, dolomite and dolomite limestone, while the percentage of calcium and magnesium are 28 and 0.2 respectively (Table 1). The abundance of calcite in the chemical analyses i s in contradiction with the lithology of the type section. The sampling sites are limited to the ex posed thickness of the Asmari-Jahrum formation, ther e fore they do not represent the whole contents o f this formation. Asmari-Jahrum is a highly kar s tified limestone. The Asmari-Jahrum aquifer karst water f1ows into adjacent alluvial fans which we dge out towards the center of the Sarvestan ba si n and qanats and springs at the foot of the anticline ranges. Springs probably represent a11 overflow of the aquifer. Results of chemical analy ses of water emerging from the Asmari aquifer are pr e sented in Figure 1 and Table 2 by numbers 1, 2, 7, 8 9 and 10 The electrical conductivity ranges from 350 to 640 J.L.Oicm. As expected the water is mostly of the magnesium calcium bicarbonate type and is saturated with calcite and dolomite In s p i te of the extensive exposure of Asmari-Jahrum formation around the Sarvestan Plain, bicarbonate wa ter is limited to a narrow area near this formatio n (Fig. 1 ). The sulfate water area is adjacent to the bicarbonate water region and a sharp increase of electrical conductivity is observed upon reach ing the sulfate water region These sulfate waters originate from two source s. First, by contamina t i on of karst waters by the Sachun formation ex po s ed in the core of the anticline that changes the l y pe of water to sulfate at the entrance of flow into the Sarvestan Plain and second by the washed off S a chun formation deposited in the alluvial fan and S a rvcstan Plain. E. Raeisi et al. TARBUR FORMATION The Tarbur formation is characterized by 527 m of resistant, mainly massive, cliff-forming, partly anhydritic limestone at the type locality. It inter fingers with marls and shales of the Gurpi forma tion at the study area The optical method show that this formation consists mainly of calcite and dolomite, with small amounts of clay minerals and iron oxides. The chemical analyses show that the main anionic constituents are calcium and magne sium (Table 1). The texture is biomicrite, biomi crosparite and biosparite The microfossils are mainly Loftusia minor, Sidrolites calcitrapoides Orbitopsella perocourson The Tarbur formation is exposed as two high altitude mountains in the Panal and Ahmadi Anticlines. The type of Tarbur karst water is carbonate and its quality is good in those places where it does not underlie the Sachun formation in the region (RASHED!, 1994). Spring 6 (Fig. 1 and Table 2) emerging from the Panal Anticline Tarbur formation is sulfate in type and has an electrical conductivity of about 1030 J.L.Oicm. It should be concluded that the Sa chun formation is responsible for the decrease in quality of the Tarbur karst water. In the Ahmadi Anticline, the Tarbur formation i s in contact with the salt dome on its southern flank. The salt dome penetrates the Tarbur formation at the contact zone, therefore Tarbur water is con taminated by salt. The Tarbur formation and salt dome are both adjacent to an alluvium fan which in tum is connected to the Sarvestan Plain. Un contaminated Tarbur water from the region not in direct contact with the salt dome, flows into the alluvial fan, mixing with salt dome water on its way to the entrance of the plain, resulting in chlo ride water with an EC range of 3000 to 10000 f.!O/cm at the plain entrance (chloride re gion Bin Figures 1 and 2). CONCLUSIONS Although the Sarvestan Plain is mainly surrounded by karst formations, bicarbonate water with low electrical conductivity is limited to a narrow strip near the karst formations, and its groundwaters are mainly of the sulfate and chloride types. The sul fate and chloride regions are mostly affected by formations made of gypsum and halite respec tively. These formations do not deteriorate the quality of water to the same extent. Although evaporite formations such as the Sachun reduce the quality of water, they do not increase the EC to more than 4000 f.!O/cm and at present agricultural

PAGE 174

Hydrochemical behavior of karstic and evaporitic formations crops are being cultivated with this water quality. But salt domes reduce the quality of water signifi cantly such that the contaminated water is of no use for agriculture (EC up to 15000 J.l!llcm). Thus for proper management, preventing measures must be taken regarding the salt dome water. In this respect, the following are suggested: 1. Water flowing from karst formations should be pumped out before reaching salt domes. For this purpose the water budget of karst forma tions and the direction of flow should be de termined. 2. In situations where salt dome discharging water mixes with fresh water, a qanat or pumping well may be constructed for removing saline water before mixing. A qanat is a slightly dip ping underground gallery which drains water REFERENCES DREYER, J. I. (1988) The geochemistry of natural water. Prentice Hall, New Jersey, 437 p. FALCON, N., L. (1974) Southern Iran: Zagros Moun tains in Mesozoic-Cenozoic orogenic belts. Geological Society. London. Special Publication, 4, pp. 199-211. FLINT, R., F. (1971) Glacial and Quaternary geology. John Wiley, New York, 892 p. GARRELS, R., M. (1967) Genesis of some ground waters from igneous rocks. Researches in Geo chemistry, 2, pp. 405-420. GARRELS, R., M. & MACKENZIE, F., T. (1967) Ori gin of the chemical compositions of some springs and lakes, equilibrium concepts in natu ral water system. Am. Chern. Soc. Adv., 67 pp. 222-242. HERA VI A., M., HOUSHMANDZADEH, A. & NABA VI, M., H. (1990) New concepts of Hor muz formation stratigraphy and the problem of the salt diaprism in south of Iran. Proceeding of Symposium on Diaprism with Special References to Iran, 1 pp. 1-20. JAMES, G., A. & WYND, Y., G. (1965) Stratigraphic nomenclature of Iranian Oil Consortium Agreement Area. Bull. AAPG, 49. JOHNSON, K., S. (1985) Hydrogeology and recharge of gypsum-dolomite karst aquifer in south west ern Oklahoma, USA. Karst Water Res. IAHS Pub!., 161, pp 343-357 KENT, P., E. (1970) The saiL plugs of the Persian Gulf region. Soc. Trans., 64, pp. 56-88. 173 from either side. As compared to a pumping well, a qanat requires suitable topography, but it is advantageous in that it does not require any operating expenses and intensive maintenance. In the Sarvestan Plain, saline water from qanats may be directed to the Maharlu Lake or evapo rated in basins. 3. Salt domes may be mulched to prevent the in filtration of rain water. ACKNOWLEDGEMENT The authors would like to thank the Research Council of Shiraz University for financial support of this research (grant No. 71-Sc-732-397). The authors are grateful to M. H. Nejati and H. Ja hangiri for their valuable assistance in this work. PALMER, C.D. & CHERRY, J., A. (1985) Geochemi cal evaluation of groundwater in sequence of sedimentary rocks. J. Hydro/., 75, pp. 27-65. PLUMMER. L., N., JONES, 8., F. & TRUESDELL, A., H. (1976) A computer program for calcu lating chemical equilibrium in natural water. US Geol. Water Resour. Invest., 76, pp. 13-61. RAEISI, E. & MOORE, F. (1993a). Hydrochemistry of karst springs from two carbonatic units in Za grosides oflran. J. Sci. Iranian. Rep., 4, pp. 302-307. RAEISI, E. & MOORE, F. (1993b). The effect of evaporitic formation on the quality of karst wa ter. Iranian J. of Sci. and Tech., 17, pp. 91-103. RASHEDI, M. (1994) The effect of climatological and geological factors on karst processes. M.Sc. Dis sertation, Shiraz University, Shiraz, Iran, 227 p. ROGERS, R., J. (1989) Geochemical comparison of groundwater in areas of New England, New York, and Pennsylvania. Groundwater, 27 pp. 690-710. SHARAFI, A., RAEISI, E. & FARHOUDI, G. (1996) Contamination of the Konar Siah karst spring by salt dome. lntl. J. Engineering, 9, pp. 37-44. SIGURDSSON, F. (1993). Groundwater chemistry and aquifer classification in Iceland. Memories (){ the XXIVth Congress qf IAHSAS Oslo, pp. 507-518. ZAK, I & GAT, J. R. (1975) Saline waters and residual brines in the Shiraz Sarvestan Basin, Iran. Chemical Geology, 16, pp. 179-188.

PAGE 176

Theoretical & Applied Karstology, vol. 9/1996, pp. 175-183 Pulse Temperature and Conductivity Analysis in Pseudo-karst Structure Investigation Radu GASPAR1 ABSTRACT The permeable zones and associated leakage through the pseudokarstic structure of Racova reservoir were determined by using natural tracers: electrical conductivity, TDS and water temperature The variation of these parameters was measured in piezometers, boreholes, subaerial and underwater springs. To check the flow paths determined by natural tracers, artificiallabelings were performed. Key words: pseudokarst, artificial reservoir, electrical conductivity, TDS, temperature, artificial tracers. Analyse de Ia pulsation de temperature et de Ia conductivite dans /'investigation d'une structure pseudo-karstique On a determine les zones permeables et l'ecoulement dans La structure de type pseudo-karstique du reser v oir artificiel de Racova en utilisant des traceurs naturels: la conductivite electrique, la mineralisation totale et l a temperature de l'eau. La variation de ces parametres a ete mesuree dans des piezometres, des forages et dans des sources suhaeriennes et suhmergees. Des artificiels ont ete effectues p o ur verUier les resultats mi s en evidence par les traceurs Mots ctes: pseudo-karst, reservoir artijiciel, conductivite electrique, mineralisation t o t ale, temperature. traceurs artijiciels. 1. INTRODUCTION The artificial Racova reservoir, on Bistrita river, Romania, stores 5.5xl06 m3 of water for power production. The foundation of the dyke from the right bank of the lake consists of fissured bed rock having salt intrusions. During the years, the water reservoir partly dissolved the salts and created a network of holes In the latest time, some of these holes have structured in drains and many emergences ap peared downstream. Some of these springs are submerged by the Bistrita river. This case is simi lar to a pseudokarstic phenomenon. In order to investigate the behavior in time of the dyke, many piezometers and boreholes ranging between 30-55 meters depth, have been drilled. A number of 40 springs, boreholes and piezometers have been investigated by means of environmental tracers: temperature, conductivity and the total dissolved salts (TDS) content. The analysis of the results indicates flow direc tions and possible transit flow paths for leaks from the reservoir. For checking these results and for the determination of the flow velocities and their spatial variation, multi-tracers artificial tracings and multi samplings were performed. 2. CONDUCTIVITY AS A TRACER One of the most characteristic natural tracer s o f karst waters is electrical conductivity. BAKALOWICZ (1974) studied the relationship between the 1 Institutul de Fizicli i lnginerie Nuclearli, P.O. Box MG-6, Bucureti-Mligurele, Romania

PAGE 177

176 electrical conductivity of a given water and its mineralization and concluded that a linear correla tion exists between them. Owing to the fact that each ion has an influence on the electrical conduc tivity, which depends on both the nature and con centration of the ion, the slope coefficient of the linear regression is characteristic for the associa. tion of ions in the water of an emergence and pro vides for a geochemical definition of the aquifer under study. MULLER & PLANCHEREL (1982) studied the distribution of the values of electrical conductivity of waters from various regions and noted that a relationship existed between water mineralization and the altitude of the respective regions: miner alization increases when the altitude of the supply area decreases. Comparisons between karst waters in various re gions show that, whereas the absolute values of water mineralization are influenced by the altitude of the supply area, the variations around the aver age are, on the contrary due to flow conditions determined by the structure of the reservoirs. Con sequently, water mineralization depends not only on altitude but also on permeability conditions and water residence time in the respective karst for mation. Groundwater conductivity greatly depends on the large variety of rock-water interactions, which eventually can be reflected in equilibrium models. The process of dissolution is determined by a se ries of criteria which must be considered as a whole. One is the time of contact between the moving water and the rock and, furthermore, the contact surface of the solid or unconsolidated rock. The rock-water interactions are also influenced by the dissolution capacity of the water and vice versa by the ability of the rock to release particles, i.e. to be dissolved. In carbonate rocks, C02 and its par tial pressure are important factors controlling the chemical reactions (MOSER eta/., 1986). In this way, dissolved solids can be used as natural tracers, due to the rock geochemistry in the re charge areas of springs and wells. So parameters like calcium, magnesium sulfate and other ions, as well as the varying relationships between them can indicate the location of the recharge areas and can help to devise hydrodynamic models (ERIKSSON, 1981 ). The chemical composition of the water always gives valuable information on the origin of the emerging water and on the groundwater flow around the man made reservoirs. Most frequently, the reservoir water has a low salt content that R. Gaspar changes along its pathway through the rocks. This change is usually more pronounced when the leak age takes place through granular media than in the case of consolidated rocks. When several springs occur, the conductivity or the total dissolved salts (TDS) of the water provide qualitative information on the relative transit times of the water from the reservoir to the emergence point. Usually, higher conductivities correspond to longer transit times. But, in these cases, it is neces sary to make sure that all the emerging water is coming from the reservoir because the change in conductivity can be also a result of the mixing of reservoir water with groundwater derived from local rainfall (PLATA BEDMAR, 1991). When many springs exist, the groundwater contri bution to the leakage can be calculated for each individual spring and for the total leakage using the simple equation for two-component mixing. Obviously, when the borehole intercepts the groundwater flow coming from the reservoir to ward the spring, an intermediate value of the con ductivity will be observed (supposing that this parameter changes along the pathway). If this is the case, the conductivity profile along the bore hole gives information on the localization of the permeable sections of the layers. A borehole conductivity value higher than the spring conductivity indicates, in most cases, that the borehole has not intercepted the permeable formation responsible for the leakage; it is how ever also possible that this borehole water under goes subsequent mixing with other lower conduc tivity water before emerging. On the contrary, a deep section of the water column having lower conductivity than the upper section, usually indi cates low conductivity leakage from the reservoir. Otherwise the water column would not be stable due to the density effect. However, for a thorough evaluation of the stability of the water column, the temperature profile has to be also taken into ac count; density differences due to salt concentration may be compensated by the temperature effect (PLATA BEDMAR, 1989). 3. THE PULSE TEMPERATURE ANALYSIS TECHNIQUE To verify a number of hypothetical connections, the pulse train analysis technique is ever more often resorted to. It relies on the principle that a known input pulse (e.g. of discharge, temperature, or water quality) will be transmitted to a resurgence where its arrival may be monitored. In certain

PAGE 178

temperature and conductivity analysis situations, natural or artificial temperature anoma lies may be used as tracers over short distances and for rapid transiti vi ties. In this case, a highly sensitive device is required to measure small tem perature contrasts. It is well known that the water temperature is an important geophysical parameter in the hydrologi cal investigation of karst systems BLA VOUX et ul. ( 1979) found a correlation be tween the altitude of the catchment area and the temperature of the karst springs from Jura Moun tains. POVARA & MARIN (1984), in a complex study about the evolution of the major ions in the water of the geothermal spring Hercules depend ing on the flow rates and temperatures, found a meteoric component in the genesis of this spring. W atcr temperature is one of the best tracers for the investigation of the leaks from reservoirs and lakes also. Most frequently, these water bodies display transient thermal stratification, not only in areas with temperate and cold climates, but even in in ner or high zones of tropical areas (PLAT A BEDMAR, 1991). The climatic and meteorological contrasts induce, in general, a density and temperature stratification of the water in the reservoir. So, in spring time, under the influence of the wind and of the solar radiation, a steep thermal gradient occurs at a Wind direction <>. 177 certain depth and divides the water body into dis tinct zones separated by the 'thermocline'. This vertical structure has a fundamental importance for ecosystems, because it is the vector of internal movements of the water bodies (MOLINARI 1976). As it is well known, in the case of stratified lakes, vertical diffusion coefficients arc very low com pared with horizontal planes and bottom infiltra tion areas are also fed, almost exclusively, with bottom water. Thermal stratification of the reservoir can also he responsible for many failures in tracer experiments performed in Jakes (Fig. 1 ) The tracer injected above a zone qf preferential infiltration can mi grate horizontally instead of sinking underground (CALMELS & SALENCON, 1991). The correlation between the water temperature in the reservoir at the springs and, respectively, in the boreholes, gives in many cases preliminary infor mation on the depth at which the infiltration from the reservoir takes place. Temperature measure ments should be initiated at an early stage of the investigation because the results that arc ohtaincu may serve for the definition of other future activi ties. For a complete study, a long record of tem perature data (e g. one year) is necessary (PLATA BEDMAR, 1989) I> )_ () Thermocline Epilimnion ... . :0.... I J

PAGE 179

178 4. A CASE STUDY: RACOVA DAM The Racova dam is a typical case where the geo physical and geochemical parameters of the water, the temperature and the electrical conductivity provided valuable information on its behavior and stability. During the cold and warm seasons of 1995, in situ measurements of temperature, con ductivity and TDS have been systematically per formed. These parameters have been measured in 40 points, in Racova reservoir, in piezometers and boreholes drilled in the right bank dyke and the dam structure and in the emergences downstream. To assess the changes in temperature and conduc tivity, a well has to cross the water-bearing layer down to the impermeable layer. Some of the boreholes were screened to allow the water now. When the well <.hilled in the bank or in the dam structure intercepts the water flow in coming from the reservoir towards the spring, an intermediate value of the conductivity will be ob served, provided that this parameter changes along the pathway. In our case, in some of the investi gated boreholes, high conductivities have been noticed. These large conductivity and TDS values arc encountered only when there are salt inclusions in the bedrock. The water has dissolved the salt R. Gaspar and created big holes filled with brine, like in pseudokarst structures. If the conductivity and TDS measured in the water of a spring situated downstream had low values, this probably indicates that the underground drains were not yet organized. The seepage risk exists. The vulnerability and stability of the bank and of the dam structure depend on the origin of the springs. In the Bistrita river some of them are sub-aquatic. Four longitudinal measurement profiles of TDS, conductivity and temperature were performed in order to determine the origin of the submersible springs downstream. The measurement of the geophysical and geo chemical parameters in boreholes, piezometres and springs led to the identification of the origin of springs and of permeable zones in the right bank. Multitracing experiments were necessary for as certaining the flow path, i.e. the horizontal and vertical spectrum of infiltrations established as a result. Simultaneously, labelings using activable tracers and fluorescent dyes: In-EDT A, Br-, Uraninc and Rhodamine B have been performed within two series of experiments. CONDUCTIVITY [ m Sjcm] 0.5 1.0 1.5 2.0 ... .. t -----. t ---------------------; BOREHOLE FS2 \ \ \ + ' 15.02. 95 Fig. 2. Conductivity and temperature profiles the Jake and tlie }i\S2 .. Profils de conductivite et de temperature dans I.e lac :et dans le forage FS2. "'' ..

PAGE 180

Pulse temperature and conductivity analysis 5. EXPERIMENTAL DATA Investigations performed in a number of 40 pie zometers, boreholes and springs resulted in a picture of the general pattern of the subsurface flow. In Fig. 2 is presented a comparative picture of the variation of the temperature and conductivity in the lake and the FS2 borehole, recorded in the winter season The vertical distributions of the temperature and electrical conductivity in the Ra cova lake and the FS2 borehole, recorded in the right bank, arc similar. This similarity suggests the existence of a horizontal current through the per meable layer bet ween the water reservoir and the borehole. The downwa rd increase of the conduc tivity values in FS2 sugge s ts that the permeable layer is situated bet ween 10 to 15 m depth. Figure 3 presents a typical c a s e without circulation The PH-IOC bore hole i s drilled in the same bank but the downward increasing conductivity i s due only to the di ssolution o f th e salt inclu s ion s A very slow horizontal infiltr a tion may e xist down to 30 m of depth, but bet ween 30 and 50 m, the quasi constant conductivity and TDS values indicate the existence of a s tagn a nt water z o ne The o rigin o f some s ubm e r ge d sprin gs along t he course o f th e river has b ee n identified by monitorin g 179 the temperature, conductivity and TDS param e t e r s in longitudinal profil e s downstream from the dam Such a longitudinal downstream profile is pre sented in Fig. 4 The minimum temperatur e conductivity and TDS values correspond to t wo su b merged springs Their origin is local rainfall. The subaquatic s prings in the median se cti o n o f the l o ngitudinal profil e are s uppli e d by the rese r voir water. To confirm these suppositions and to obtain in for mation on the local groundwater flow and o n the horizontal and vertical di s tribution of p er meabili tie s tracer tests have been carried o ut. Thus multi tracer and multisamplin g ex p e rim ents were per formed in two stages using In-EDTA, Bras ac ti vable trac e r and Uranine and Rhodamine B fluorescent dye tracer s S ome important r es ult s a rc present e d in Figur e s 5 6 an d 7 All o f th e m hav<.: s hown th e s ame fl o w p a tt e rn : h orizont a l flow with the water inl e t locat e d at th e upp e r semi permeabl e l a y e rs o r karstified m a t e ri a l s a nd the outlet di s tributed in dra ins and sprin g s The r a n ge of t1ow ve locities v aried het w een 0. 18 m/h ami 5 m/h as mean value s a nd 2 .96 m/h t o 10. 50 m / h as maximum trans it ve l oci t i e s (see T a bl e I ). PH -10 C Flg.3. 22t-2 y---i X t20 E TDS : Variations of conduc tivity, temperature and TDS as a func tion of depth in the PHlOC borebole. Graphe de variation de Ia conductivite, Ia temperature, et le TDS enjonc tion de Ia profondeur dans i e forage 2 1 22 E / l1s= l /1COND UCTIVI T Y I 20, ,21 ; 162 TEMPE RAT U R E t : : z X I 8 19. 18 {10 I I 18 s J ? 1 r ----------+------r--------8 5 1 6 32 41 5 5 OEP TH[m]

PAGE 181

180 TRACE R Indium-EDT A FSIO-MII 86 Uranine FS30-PH40 35 6 7 5.80 FS30-MII 66 8 24 8.25 FS30-M10 42 1 0.50 Rhodamine B FS40-FH 18 6 100 3 .0 0 FS40-PH40C 30 4 24 7 .50 FS40-PH30C 30 4 65 7.50 17 DOWNSTREAM PROFILE u .__. w w _J 0:: ffiw =)1 VlU 0::0:: 0:: W:::J LJJ 2::0 0 .. 0){/) 2:: :::J w {/) 1--J 15 Con ductivity 13 LENGTH[m]-----100 200 3 0 0 400 600 7 0 800 Fig. 4. Temperature, conductivity and TDS a La variation de la temp6rature, de la conductivite et du TDS daM k projilwngitudinpl:aval. R. Gaspar 1.23 5 2.75 1.45 0.18 1.2 5 0.46 E l/1 E ........ >-....... I-CJ) > E I-{/) u 0 ::) I-0 z 0 u 0,5 Q2 0,4 0,3 0.1 0,2 0,1

PAGE 182

Pulse temperature and conductivity analysis Rhodamine 8 E q>'cr. 1 ,21 1 z Q 0,8l r-. I
PAGE 183

182 RACOVA LAKE CJ Indium Uranine !!!!!8 Rhodamine B I The horizontal spectrum of infiltrations based on the hydraulic interconnections established by us ing artificial tracer investigations is illustrated in Fig. 8. 7. CONCLUSIONS Investigation of water losses from the Racova reservoir by means of natural tracers, such as tem peratur e electrical conductivity and TDS allowed to perform preliminary tests concerning the struc ture vulnerability and dam stability. The boreholes drilled in the bedrock, at depths under the bottom of the lake, intersected large holes The TDS and conductivity measurements showed the existence of brines in these holes. The REFERENCES BAKALOWICZ, M. (1974) Geochimic des eaux d' a quifcrcs kar s tiqu es Relation entre minerali s ation cl c onductivite. Ann. Sp e / eol. 29, 2 pp. 167 173. R. (Jaspar . ,: . ..... \ ". ' extension of these cavities arid drains network are a real danger for dam stability and particl.llarly for the right flank The temperature, conductivity and TDS variations analysis in longitudinal profiles downstream the reservoir Jed to the identification of the origin of some submerged springs. Some of them arc supplied by local precipita tion. The use of acti vable and fluorescent dye tracers in two series of field ex periments has confirmed the existence of perme able zones in the dam structure and under the right bank. This has demonstrated th<.it in the case of artificial reservoirs leak detection and water loss investigations, the water temperature, electrical conductivity and TDS are excellent environmental tracers. The changes in the chemical composition of water can be quickly determined by measuring the elec trical conductivity. There is a linear dependence between TDS and conductivity. The applied methodology can be successfully used for leak detection and water loss investigation in the case of dams from karst zones. BLAVOUX, B., BURGER, A., CHAUVE, P., & MUDRY, 1. ( 1979) Utilisation des isotopes du milieu a Ia prospection hydrogeologique de Ia chaine karstiquc du Jura. Revue de G e ol o gie Dynamiqu e e t d e Geo gruphi e Physique. 21, 4 pp 2 95 -306.

PAGE 184

Pulse temperature and conductivity analysis CALMELS, P. & SALENCON, M., J. (1991) Techni que de marquage pour l'etude des circulations des masses d'eau dans les lacs. Application a la rctenue de Parelup. International Symposium on the Use of Isotope Techniques in Water ResoUIces Developement, IAEA-SM-319/11, Vienna. ERIKSSON, E. (1981) Hydrochemistry chemical process in the water cycle. Technical Documents in UNESCO, Paris. MOSER, H., RAUERT, W, MORGENSCHWEIS, G. & ZOJER, H. (1986) Study of groundwater and soil moisture movement hy applying nuclear, physical and chemical methods., UNESCO, Paris, 104 p. MULLER, I. & PLANCHEREL, R. (1982) Contribu tion a I' etude de !'hydrologic karstique du massif du Vanil Noir et de Ia chaine des Gostlosen. Bull. Soc. Frih. Sc. Nat., 71, (112), pp. 102-132. 183 MOLINARI, J. (1976) Recherche et localisation de fuites sur retenues naturclles ou artificielles par techniques de traceurs. La Houille Blanche 314, pp. 301-312. PLAT A BED MAR, A. ( 1989) Leakage study at lake Laja. Final Report TC Project CHI/8/016, IAEA, Vienna. PLAT A BED MAR, A. ( 1991) Detection of leaks from reservoirs and lakes. Proc. Adv. Group Meel. Use of Artificial Tracers in Hydrology. IAEATECDOC 601, pp. 71-129. POVARA, I. & MARIN, C. { 1984) Hercule thermomin eral spring. Hydrogeological and hydrochemical considerations. Theor. Appl. Karst., I, pp. 183194.

PAGE 185

Theoretical and Applied Karstology, val. 9/1996, pp. 185-214 Contributions to the hydrogeology of the karst areas of the Bihor-VIideasa Mountains Romania] lancu ABSTRACT The Bihor Vladeasa Mountains are ones of the most important karst regions in Romania. They have important groundwater resources, unexploited till now. On the hydrogeological map there are pointed the main karst springs with their magnitude of discharges and the directions of the groundwater flows. More than 45 tracers la bellings were performed and an average of 45 m!hour of flow velocity was recorded in 42 of these. As a consequence of the great diversity of the geological constitution and intense fracturation of the rocks, the karst systems are of binary type, with a large variety in size and hydrological parameters distribution. Due to the consistent observations and the hydro-meteorological measurements, the groundwater resources were evaluated and the processing of the flow rates series outlined a great diversity of the karstic systems. The quality of the groundwater is very good as indicate the results of the chemical and bacteriological analysis and the potential sources of pollution are leaks. Key words: hydrogeology, karst, analysis of the flow rates series, groundwater quality, Bihor Vladeasa Moun tains, Romania. RESUME Contributions a /'hydrogeologie des regions karstiques des Monts de Bihar-Vladeasa Les Monts Bihar-Vladeasa occupent l'une des premieres places parmi les regions karstiques de Roumanie. En meme temps, ils disposent d'importantes ressources d'eaux souterraines non exploitees jusqu 'a present. Sur fa carte hydrogeologique ci-jointe, sont indiquees les principales sources karstiques, l'ampleur de leur debit et les directions de deplacement des eaux souterraines mises en evidence par plus de 45 marquages aux traceurs. La vitesse moyenne enregistree dans 42 marquages atteint 45 mlheur.On a localise aussi les plus grands avens et grottes cartes. Suite a Ia constitution geologique variee et d La fracturation accentuee des formations geologiques, les systi!lnes karstiques sont, pour La plupart, de type binaire, de dimensions tres d(fferentes et a compartaments hydrogeolo giques distincts. Un programme rigoureux d'observations et de mesurements hydro-meteorologiques a mene a !'evaluation du potential global d'eaux souterraines et ['etude des donnees acquises par d!ffbents (!'etude des cowbes de recession, ['analyse correlative et spectrale des series de debits, etc.) a pennis de mettre en evidence Ia grande diversite de functionnement des systemes karstiques. Les analyses variees chimiques et hacteriologiques ant evalue La potabilite des eaux karstiques de toute La region etudiee, en etablissant La presence d'une eau de bonne qualite et !'absence totale des sources de pollution. qui pourraient conduire a Ia degradation de La qualite de l'eau. Mots cLes: hydrogeologie, karst, analyse des series d'ecoulements, qualites des eaux souterraines, monts Bilwr Vladeasa, Rownanie. 1 S.C. Prospectiuni S.A. str. Caransebe 11, R-78344 Bucureti, Romania.

PAGE 186

186 INTRODUCTION The Bihar mountains occupy a central position within the Apuseni mountains range, extending in a longitudinal direction between the basin to the west and the GiU\u mountains to the east, and transversely between the Repede valley to the north and the Mic valley to the south. They consist of three distinct transverse compart ments, that are well defined in what concerns both their topography and their geologic framework: Vladeasa, Bihor and Biharia massifs (IANOVICI eta/., 1976). The Vllideasa massif consists mainly of intrusive and of igneous ophiolitic formations, which induce an overall heavy-looking topography. The south ern half of the intrusive body is surrounded by sedimentary formations, within which carbonate formations occupy a pre-eminent position: the karst area Meziad-Ferice-Valea Rea to the west and south-west, and the graben of Cald to the east and south-east. The central compartment, for which the "Bihar mountains" denomination should be preserved, due to the fact that here the karstic topography, characteristic for these mountains, is widely de veloped, is separated from the Vllldeasa mountains by Cald and Pietras stream courses To the south, the Mare and CriUl BiHta streams delimit this compartment with re spect to the Biharia massif, made up of crystalline schists. The present paper is an introduction to the hydro geology of the carbonate deposits occurring within the central compartment the Bihor mountains, as well as within the southern half of the Vllideasa mountains, areas that for the sake of the current analysis are reunited under the denomination of Bihor Vladeasa mountains. The central compart ment is simply referred to as "Bihar mountains". I. OROHYDROGRAPHV OF 81-MOUNTAINS The complex geological constitution of the Bihar Vladeasa mountains, that includes a puzzle of rocks, with limestones and dolomites prevailing, followed by sandstones, conglomerates and igne ous rocks, results in a multitude of topography types, among which the most outstanding is clearly the karstic type, which considering its ex tent variety and amplitude of the karstic land forms, ranks this specific area in the top position among all Romania's karstic territories. L Orilream Due to the presence of many ridges and isolatec massifs, and to the absence in the case of th( large karst platforms -of major topographic leading lines, it is difficult to perform a systematic description of the orohydrography of this area anc as a result the presentation will follow the rive1 catchment areas. This choice is also supported the occurrence in the Bihar mountains of the most important water divide from all Apuseni moun tains, wherefrom the rivers CriUl Negru, SomeUl Cald and Arieul Mare originate. The origins of those three major catchment basins are separated by two mountain ridges: one, strik ing north-south, marked by the summits Bruta Custurilor-Poienii-Bohodei-Fantana Rece-Mligu ra tra Grrutoare, borders Criul Negru catchment basin to the west; the other one, striking west-east, branches perpendicularly to the previous one at Biserica Motului summit, to continue westward along summit-Clujului Summit-Ursoaia saddle. The latter separates the catchment basin of Someul Cald, situated to the north, from that of Arie, situated to the south. At the junction of those three major catchment basins is situated the closed catchment area PadiPonorului, sutounded by a belt of ridges which preclude it from being included in any of the three previously mentioned basins. Westward from Rece summit branches the main ridge of Vllldeasa mountains, that along the section Cornul-Miclau-Muncelu Mare forms the divide between the catchment basins of Criul Repede, situated to the north, and that of Someul Cald, situated to the south. CRI!;lUL NEGRU CATCHMENT BASIN Between the valley of Meziad to the north, and that of Criul Bruta to the south, Criul Negru re ceives from the Bihor Vllldeasa mountains a series of major tributaries: Beiuele, Valea Mare, Ferice, Criul Pietras, Craiasa and Sighitel. Those streams, together with the main tributaries of Criul Pietras (Aleu, Bulz, Galbena), isolate a series of west-east striking ridges, that branch from the main ridge of Bihor mountains, to smoothly connect in the end to the hilly topogra phy of the eastern part of the Beiu Basin. North of Criul Pietras, the topography is domi nated by the imposing Dealul Mare, built up of ba natites and carrying the road that leads to de Vale, and by the solitary summit of Magura Ferice, as well as by the narrow and steep valleys of Zapo die, Cohuri, Aleu and Sebiel, the fountain-heads of

PAGE 187

The hydrogeology of the karst areas of the Bihor Vl4deasa Mountains 187 which reach up to 1600 m. Among the caves exist ing in that area noteworthy are the ones at Cresuia (Cornilor cave) and Ferice. Bulz valley, forming the upper reaches of Criul Pietras, collects its water from beneath the main ridge of Bihor, between the summits Carligatele and via a series of steep tributaries, some of which display inaccessible waterfalls (Boga, Oelu, Bulbuci), while others benefit of less rugged courses (Valea Rea, Valea Plaiului) The topography of Bulz valley is dominated by the majestic Boga escarpment, with vertical, over 300 m high walls in its upper part, climbing just beneath Piatra Boghii peale. The landscape also includes many abundant springs (Boga, Oelu, Bulbuci), peaks and bluffs, which make this area the wildest of the entire Bihar massif The median section of the Criul Negru catchment basin situated within the Bihor mountains, that extends between Criul Pietras to the north, and the ridge Vartopul-Tapul-Prislop to the south, is broken by the intermediate ridges Tapuland Tapul-Dosurile, into three subordi nate catchment basins, namely Galbena, Crruasa and Sighitel. The first of those two ridges dis plays characteristic steep slopes in its median sec tion and a small karstic plateau around peak, strewn with sinkholes and having a pothole entrance just next to the summit. Galbena valley originates between the peaks Bortigul and in terrains consisting of sandstones and conglomerates of the Arieeni unit. When entering a limestone substratum, north of Vartopul summit, the valley, that from here down stream is called Lunqoara, flows between the steep walls of a canyon, eventually to sink into the streambed fissures, most of the time of the year completely, some 2 km downstream of the junc tion with the Crianului valley. Two km downstream of this swallet, Luncoara receives a powerful right hand tributary that origi nates in the Galbenei Spring, one of the major outflows of the closed catchment area Ponorului. From here downstream the valley is called Galbena and assumes a perennial character, while receiving only left hand tributar ies (Valea Budeasa). Before the junction with the valley, the flow rate of Galbena doubles, as a result of the inflow of spring. Along its entire course, from be neath V artop peak and down to the site called intre Ape, the junction with the Bulz stream, Galbena v alle y displays a rectilinear course, tectonically controlled by the major Galbena fault. Valea Seacii, the ongm of which is located in Groapa Ruginoasli, beneath Tapul peak, has the most extensive catchment basin of all Galbena valley tributaries. In spite of this, due to the mul titude of karstic stream piracy processes that occur both in its own catchment basin, as well as in that of its main tributary, Tiganului valley, Valea carries water only during heavy rainfall periods, and, as a consequence, its direct contribu tion to the Galbena valley flow rate is small. Groapa Ruginoasii is a huge, still active ravine, 125 m deep and almost 700 m in diameter, open ing toward Valea Sead, excavated in the Permian sandstones, shales and conglomerates of the Arieeni nappe thrust outlier located on Tapul peak. During rainy periods, a yellow-reddish mud stream flows out of Groapa The asso ciated sand suspensions are deposited downstream, on the Galbena beaches, and further on, on those of Criul Pietras, almost down to the junction with Criul Negru. Among the remarkable karstic features existing in the catchment basin of Galbena valley, we men tion the 288 m deep pothole in Hoanca Urzicaru lui, one of the deepest in Romania, and the virtu ally inaccessible Jgheabului gorge, excavated by Galbena stream in the massive limestones next to its junction with the Bulz stream. CrlHasa valley catchment basin extends west of that of Galbena and collects its water from beneath the ridge, via the streams Fagului, Sibioara and Pietrele Roii. In this catchment basin are situated three of the most beautifully decorated caves of Bihor mountains: the commercial Pestera Urilor at Fagului cave, intercepted by a mining gallery excavated on the left side of Fagului valley, and Micula cave, whose underground stream emerges in the Giuleti spring. The Sighitel catchment basin extends mostly on limestone deposits where the stream has excavated a deep valley, which in its upper reaches has a canyon appearance. The summits of the ridges that surround it are covered by quartzite sandstones ascribed to the Arieeni nappe. Those deposits favor the organi zation of a scanty runoff, that when reaching limestone t errains sinks under ground via a multitude of swallets, to supply a well developed karstic aquifer. The intense karst processes of this area resulted in the development of a large number of caves (about 70), out of which Mligura, Coliboaia, Piolca and the pothole in are worth mentioning.

PAGE 188

188 X Bruta catchment area, extending west of Piatra Grrutoare peak and south of the Tapul Prislop ridge, displays a high energy environment and steep slopes, across which the streams -most of them temporary (Hoanca Motului, Corlatul Couri, Hoanca Codreanului), have in cised deep, canyon-like valleys, broken by many waterfalls which make progression extremely dif ficult. Outstanding karst features of this catchment basin are the outflow cave Izvorul Criului, as well as the temporary stream cave Poarta Bihorului, the latter especially due to the size of its entrance porch. The flow regime of both surface streams and groundwater in the karst area of the upper Criul Baita catchment basin is dramatically influenced by the existing mining activities. PONORULUI CLOSED C AT CHMEN T ARE A Padi-CeUitile Ponorului closed catchment area extends over a surface of 37.2 km2 and is sur rounded by a belt of ridges that prevent surface flow connections with any of the adjoining catch ment basins. However, tracer tests have proven that its groundwater flow discharges into the Criul Negru catchment basin The origin of the closed basin is intimately related to the geological constitution of the area, specifi cally to the alternating karstic and non-kartsic substratums. The sandstones and shales on allow a scanty organization of the rainfall water into perennial streams, which sink when entering a limestone substratum, with a re sulting dissection of the terrain, leading to the existence of 9 subordinate closed basins, within the overall Ponorului closed catch ment area: Padi, Groapa de la Barsa, Valea Cetlipi, Barsa Cohanului, Paragina, Lumea and Poiana Ponor. CALD CA T C HM E NT BASIN Characteristic of SomeUl Cald upper reaches are the outstanding morphology and karst topography of Radesei and the associated spectacular canyon When reaching out of the canyon, the river re ceives four main, left hand tributaries: Alunul Mare, Alunul Mic, Ponorul and Valea Firii, all of which cut across the prevalently carbonate depos its of the SomeUl Cald graben, displaying a char acteristic karst topography, with karst plateaus I. Orllfeanu (Piatra Altarului, Humpleu, Onceasa, etc.), pot holes, springs and swallets. From the Bihor mountains karst area, Someul Cald receives two main, right hand tributaries: and Beli. The first one originates in the junction of the streams Izbuc and the flows of which are collected from the karstic pla teaus Blitrana and Clilineasa, while the second one, Beli, has its fountain-head east of Clilineasa pla teau and receives as its main tributary Apa the source of which is located beneath Ursoaia Saddle (the latter stream is delimiting Bihor from mountains). MARE CATC H M E NT AREA A significant part of the carbonate terrains in Bi hor mountains occurs in the Arieul Mare catch ment area, more specifically on the left side of the river, between its source area and the junction with Albae stream. From the fountain-head situated beneath the Vartop pass and down to the village of Arieeni, where it receives the Cobli tributary, the river is called Raul Alb. Along this section it crosses exclusively terrains consisting of sand stones, conglomerates and shales, ascribed to the Arieeni unit and displaying a characteristic topog raphy of rounded ridges and steep mountain slopes. The most important tributary of ArieUl Mare in the Bihor mountains karst area is Garda Sead. The latter is almost 20 km long, having its foun tain head beneath Garzii, close to Padi, while its first significant inflow is provided by the spring at Gura Apei. After a rectilinear course along a narrow valley, where the flow rate doubles via the Coliba Ghiobului and Apa din springs inflow, the entire flow of the stream-that in this section is called Gardioara, sinks into the cave Coiba Further on, from Casa de hamlet downstream the valley is called Garda It enters a narrow gorge section and re ceives the left hand tributary Vulturul, then next to Fileti, via Tliuz spring, the valley recovers the flow sunk in Coiba When leaving that gorge section, the flow rate of Garda increases on account of the discharge provided by the Coroaba spring, whichafter follows a long course acro ss Permian Werfenian sandstones and conglo merates, interrupted by the Ladinian limestones at Cotetul Dobretilor hamlet, where the homonym outflow cave is located. Before reaching the course of ArieUl Mare, in the center of the villa ge Garda de Sus, Garda Seac!i valley r ecei ves from the left side its most important

PAGE 189

. .. The hydrogeology of the karst areas of the Bihor VliidetlSa Motintai'ns 189 tributary, the course of which is parallel to, yet shorter than that of Garda Seadi. BetWeeri the valleys of Garda and is perched the second closed catchinent area in the Bibor mountains, flows, on its first 4 km upstream of the confluence with Garda Seaca, through a narrow canyon, With up to 200 rn .high walls, cut into .. the limestone s ubstnttum : Within the canyon, stream receives its most important supply, the discharge of the lui Ioanel. The closed catchment area situated at 1100-1300 m altitude, is traversed in its upstream section by Ocoale brook, that when Pothole ofPoienita. (146)-Cave of J-h1mpleu hill, (150) 2 Cilve of Ptl!aul Hodobanei stream (I 03) 3 Cave of Zapodie (8iJ)-Neagra cave (81) 4 Valea Rea system (Adrian pothole, 54) 5 Corniior cave, (I) 6 Coiba Mare cnve, (97) 7 Darninii cave, ( 122 ) 8 cave, (I 14) 9 Cerbului cave, ( 137) Avenul cu Vaca pothole 10 Pothole of (I 07) II Cave of Fantiina (59) 12 Coltului cave, ( 138) 13 Lume.rPierduta system, (70-71) 14 Ceti\t:ilc Pohorului cave, (75) 15 Ponorul din Cuciulata cave, (13 1 ) 1 6 Ghetanil de Ia Bursa cave, (79) 17 (Micula) cave, (25) 18 de dupl\ Delut cave, (99) 19 Cave of S ecaturii hill,{36) 20 V5pothole(Fata Munchllui). (55) 21 Pothole of Hoanca Urzicarului, (89) 22 Pothole of Cuciulata, ( 1 3 2 ) 23 Z llp o diei ponor 24 Sohodo1 2 pothole 25 ''Gaura care Sufl!l" pothole, (Petit Tibi, 24) .. *Data in colaboraiim1 with H. MITRO FAN from quartzite sandstones on a limestone substratum gradually sinks, eventually to disap pear completely ; The valley downstream assumes the appearance of a wide depression, with its bot tom strewn with sinkholes, where also the enirancc ofthe pothole is located. On the divide between the closed catchment basi n a nd t he actwd Garda Seaca stream basin opens the wide entrance of the shaft that leads to the SdlriOara Glacier, while westward, benea th the divide, is located one of the most beautifully deco rated caves in Romania, Pojarul Politci. 22. 142 181 (-211; +60) 1982 ) 12.<148 178 (-162;+16) 11.718 -264 DAMM 10.140 +I 12 BRIJAN, 1987 . 5 .680 121 (-76;+45) 1978 .. 5.645 -112 SILVESTRU 5 .210 -75 CIUBOT ARESCU 5 .094 -125 SILESTRU et al. l995 4.010 240 (-220;+20) 3.550 129 {-40;+89) V 1977" 1978 3.526 167 (-86;+81) SIVESTRU et al. i995 3 3Z2 c..J37 3 .214 -117 BRIJAN, 1978 3.140 85 (-75;+10} 1978 3.010 112 Ji:n?-1978 3 .000 (?) 1.480 -142 V 976 J .45(} -230 HALAS! 1.446 273 DAMM,I994 1.125 288 (-286;+2) V etaL 1982 925 186 v i 978 705 122 C:112 ;+ 10) .1977 1978 507 -193 et al., 1982 241 161 F-160;+1) KOPACZ, LAZAR.I996

PAGE 190

190 REPEDE CATCHMENT BASIN Valea Seadi (Paraul Stanciului) karst area and the carbonate deposits at Stana de Vale area occur respectively in the catchment basins of Hent and lad streams, tributaries of Repede. Paraul Stanciului collects its water from the east ern slopes of the VHideasa massif. In its fountain head area, in a high energy environment, the stream crosses a succession of compartments risen and sunken along fault lines, with Triassic and Jurassic limestones and dolomites outcrops in the elevated blocks and Senonian deposits in the downthrown blocks. From a hydrologic point of view it is worth mentioning the presence of Varspring, of the temporary dry section of the Valea Seadi stream between the swallet "de Ia Tau" and Nimoioasa springs, and the karstic stream piracy features along Valea Podurilor. Sev eral caves and potholes have been explored, among which the outstanding 2250 m long Varcave (KOMIVES & NAGY, 1976). The presence close to Stana de Vale of a little compartment consisting of limestones and dolo mites resulted in shaping a karst topography of Fig. 1. Distribution of discharge gauging stations in the Bihor Hideasa Mountains: 1. discharge gauging sta tions (DGS) in the INMH national net work; 2. suppressed DGS; 3. manmade r e scrvoir; 4. study area. La distributi01/cks Sla tions dans les monts Bilwr VUideasa: 1. stations llydrometriques (SH) dans lit reseau nationale du INMH; 2. SH suprimees; 3. lac arti ficiel; 4. zone etudiee. :. . .... 1 I. Oriieanu modest dimensions, yet including varied land forms (sinkholes, dry valleys, swallets, caves and springs). Outstanding among them is lzvorul Min unilor, a source of still water of excellent quality, that emerges from a small cave excavated in Ani sian dolomites. The large number of karst cavities, their dimen sions and their impressive beauty place the Bihar VIMeasa mountains in the top position among Romania's karst areas. In Table I are indicated the largest caves and potholes surveyed in the area. II. GENERAL HYDRO-METEORO LOGICAL DATA Rainfall across the Bihor VHideasa mountains area has an uneven distribution. Records performed during hydrologic year October 1984-September 1985, as well as multiannual average values dis play an increase of the annual amounts from the basin (Budureasa-941.3 mm, Pietroasa-948.6 mm, Baita-884.2 mm) eastward, up to the Stana de Vale-Piatra GdHtoare ridge area (Slana de Vale-1608.5 mm), while further east a de crease intervenes (Vll!.deasa-943 mm, Smida-s / F; I .'0 j:)!.(l ( r ;r:., t_)l :: }, 11 i ; ";'-'. ___ / :.l I ...... CAMPEN I ..

PAGE 191

The hydrogeology of the karst areas of the Bihor Vllideasa Mountains 191 952.3 mm, Poiana Horea-714.5 mm, Casa de Piatd1-836.5 mm, mm). Within Bihar VHi.deasa mountains are included a series of stations of the national stream gauging network, under the authority of INMH (National Institute of Meteorology and Hydrology) (Fig. 1 and Table 2). They are generally located at the border of the mountain massifs and gauge runoff originating in catchment areas of varied lithologic constitution (limestones, dolomites, crystalline schists, igneous rocks, Permo Werfenian molasse deposits, etc.). The distribution of the multiannual specific discharge of those streams mirrors the rainfall distribution, ranging from 31.4 Vsfkm2 in the case of Dragan river, at the gauging station Padiul Crucii upstream, with a high altitude catchment basin, to 19.4 l/s/km2 in the case of Cald river, at the gauging station in the eastern part of the mountains. The memory effect (MANGIN, l98la, 198lb, 1982, 1984) of th e catchment basins (Table 2, Fig. 2), reflecting the groundwater reserves that sustain the runoff, has large values (47-123 days) in the case of basins extending on igneous rocks and/or crystalline schists, and low values (15 days) in the case of basins extending prevalently on limestones or PermoWerfenian molasse deposits. ... .-.-staUon Ri'Ver. F km1 .. H.m 1 2 3 4 Cald 320 1247 Cald Smida 110 1293 119 1249 Poiana Horea 83 1259 Dragan P. Crucii am. 119 1228 P.Crucii 39,4 1172 lad 101 979 lad Stiina d e Vale 27 1210 Pietras Pietroasa 123 956 Baita Baita 36 892 Scarisoara 200 1099 .. The large memory effect values in the case of the igneous and crystalline rocks are due to the ad vanced development of the wheathering layer, that acts as a storage reservoir which slowly delivers the rainfall derived water. A quite distinct category form the VlMeasa ignimbritic rhyolites, that occur under a variety of facieses and form a major reser voir, most of which is drained by the Dragan river and by its tributary, the stream. The knowledge of the runoff characteristics has an outstanding importance in the case of the binary karst systems, since together with rainfall this is the input function to the system. Interpreting the system output function (the springs flow rates hydrograph), without considering the parameters of the non karstic catchment area (provided by correlative and spectral analysis) may result in erroneously ascribing filtering properti es to the karst aquifer. Since the springs hydrographs of th e binary karst systems preserve in their memory the hydrologic characteristics of the non karstic catchment basins, especially for small and medium size systems, usc of graph k/i (MANGIN, 1975) in establishing the degree and the places of the karstification must b e performed with caution. . Q .;.3is .. 0 q 1fslkm1 Br .. EM days TRdays TF 5 6 7 8 9 lO 6.22 19.4 0 24 48 37.1 0.196 2 .92 26.5 0 25 2 32 19.5 0.268 4 7 34.2 0.208 1.80 21.2 0.28 3.74 31.4 0.25 1.19 30.2 0.25 123 81 0 .0 92 2 83 28 0 0.205 1.21 44 .8 0 .2 5 24 24.7 0. 1 92 4 .15 33.7 0.215 15 21.4 6.208 0 .94 26.5 0 0.19 5.45 2 7. 25 0 27 35 28,9 0.232 Notes: Data in columns 5, 6 and 7, after Dia c onu eta!., I 97 I; Discharges for 1950-1967, time interval ; F, surface .of catchment area; H, mean altitude (as1); Q, m ean annual discharges; q, specific dischar ges; Bf. base mdex; EM, memory effect; TR, regulation time; TF, truncation frequency Data m 8, 9 and lO for I971-I975 time interval, exceptfor the Dragan and Scbiel rivers (1970-1 973) and and Cald (g.h rivers (1971....;1 974).

PAGE 192

192 ;. Qf tb dlo\V rates:> );_;:. : roasa). s imple d e d e b its (f1Z=12S dayS, k=l). :": }-:'. .'. : Ul. HYSTORICAL REVIEW ON THE BIHOR MOUNTAINS KARST HYDROLOGY INVESTIGATION . . The investigations on the Bihar Vllldeasa moun tains karst have been inaugurated in 1863, by the printing of Adolf Schmidt's monograph "Das Bi har Gebirge", the first extensive geographic study on the karstology and speleology of a specific area of our country The work includes the geographic results of a geological-geographical scientific ex pedition performed in the Codru Morna, Bihar and Metaliferi mountains during 1858-1862. Detailed descriptions of a series of caves, including the Glacier are provided, as well as a table where the caves are listed into two categories, breakage" (swallet) caves and "eruption" (outflow) caves, which witnesses an accurate knowledge of the water movement through karst massif s (p 33). 4-5 -"M contrary, the upper areas of the lbnestone territ'ory are ch(J,racterized by water shortage, there the springs are scarce and weak, yet at the feet of the Limestone formations vigorous springs burst out, exceeding the flow abundance oft hose in crystal line areas, frequently forming streams that might even drive water-mills"(p. 34). In this chapter the provides two tables including the cold and the warm springs in the mountain domains of the Cris-es, Cald and catchment basins; indicating for each of them the geologic substra tum of the emergence, the tempeni.ture, the date and the author of the observations. A brief presen!ation of this monograph has been published by in 1980-1981. . . An important contribution to the. hydrologic in vestigation of the karst in Bihar VHldeasa moun tains is provided by the scientists of the Institute of Speleology in Cluj Napoca, D. Coman, M. and I. Viehmann, together with the geologist M Bleahu, team that will be subsequently joiried by In the introduction to the chapter concerning the T. RUSU and Gh. Racovi{il, arid which during hydrography; Schmidl provides a vivid characteri1946-1956 conducted explorations that led to the zation of the Bihar mountains: "The crystalline discovery of major caves (Pojarul Politei cave, the igneous rocks the main ridge are those where pothole in Petera Neagril de Ia Barsa, .fi"mn the considerable amount of springs spout, Cilput cave, the cave networks in Lumea Pierduta even the weak (mes. The water springs out, partly ar1d Cetiltile Ponorului) Still during that period, _li-om beneath the stdne, mH unfrequentlyfrom the the first fluorescein tracing experiments have been actual with cracks, from the mountain ridge, performed by et al. (1957); RUSU et al. straight from beneath the meadow, from the . (1970) and VIEHMANN (1966), the flow connec-smooth streambeds of the ridge, from bineaih tions between Ocoale closed catchment area and small marshes, as well as from other Sp(J/s.On the the springs at Cote{Ul Dobretiloi:, respectively the existence flows along the

PAGE 193

The hydrogeology of the karst areas of the Bihor Vliideasa Mountains 193 -Poiana Ponorului-Galbenei Spring lineament being outlined as a result. In 1957 BLEAHU publishes the paper 'The kar stic stream piracy and its importance for the mor phologic evolution of the karstic regions", which exposes the methodological approach for the mor phologic and hydrologic investigation of the karst areas and proposes a systematization of the corre sponding terminology. The entire approach of the author is supported with examples taken from the Bihar mountains karst. During 1976-1985, alone or in co operation, publishes in a series of papers the re sults of speleological investigations, which as sumed a definite hydrologic character too, con ducted in the karst of Bihar VlMeasa mountains and which have brought important contributions in this domain, as an outcome of the exploration of Groapa de la Barsa cave system (1977-1978), of Coiba Mid-Coiba Mare cave system, of the cave in Paraul Hodobanei ( 1982), of the karst at Casa de (1976), in the upper reaches of Someul Cald (1978), Lumea (1982) and in other areas. Many of the published papers are dedicated to karst areas or objectives that are representative for the karst in Bihar VlMeasa mountains, papers where the authors bring informations concerning the present day or the original, (i.e. the landform engendering) hydrologic context: VIEHMANN et al. (1980) Ponorului), RUSU (1981) (Petera Urilor de la Chidu), COCEAN (1988) (Gorges and defiles), BRIJAN (1978; 1982; 1987) (the pothole in Hoanca Urzicarului, Valea Bulzu lui-Fanate area, Cresuia area). The hydrogeologic investigations in the Bihar VlMeasa mountains have been initiated in 1983 through the activities conducted by I. and N. During 1983-1985 they prepare the first hydrogeologic map of the karst areas, as well as the groundwater reserves evaluation, and perform some 36 new tracer The hydro-meteorological data acqmsttion has been performed in co-operation with G. and P. Hotoleanu from INMH, while E. Gaspar and I. from IFIN (Institute of Physics and Nu clear Engineering) and I. Pop from the University in Baia Mare have taken part in the completion of the tracer tests, the results of which have been published in 1991. In 1992, I. and N. performed a hydrogeologic study the evaluation of the still water potential m the Apuseni mountains; the existence of still water sources in Bihar VlMeasa mountains is outlined as a result, and a detailed investigation of the identi fied sources and of Izvorul Minunilor at Stana de Vale is performed during 1995-1996. In 1995 SIL VESTRU et al. published the prelimi nary results of hydrogeologic studies performed in the upper of Someul Cald. IV. GEOLOGIC-STRUCTURAL FRAMEWORK OF THE BIHOR MOUNTAINS Within the overall structural setting of Bihar and Padurea Craiului mountains (see insert within the enclosed map), the lowest position is occupied by the Bihar Unit, that is usually called the "Bihar Autochthonous". It includes metamorphic forma tions, ascribed to the Arada and Some series, and a sedimentary stack consisting of Mesozoic pre Senonian formations, locally with detritic Permian deposits at their bottom. Out of the Codru nappes system, in this specific area have been outlined the following units (starting from the lower one): the the Garda nappe, the Ferice nappe and the Seblel and wedges, the Batranescu nappe, the UrmlH nappe, the Vetre nappe and the Arieeni nappe. Out of the second nappes system in the Bihar mountains, i.e. the Biharia nappes system, at the southern boundary of the considered map, south of the valley of ArieUl Mare, crystalline schists and gneiss belonging to the Muncel-Lupa nappe occur, while in the area of Piatra summit, albitic gneiss ascribed to the B1har1a nappe outcrop. The legend of the hydrogeologic map indicates the stratigraphic correlation of the Bihar Unit forma tions and of the Codru nappes, according to the works of BLEAHU et a!. (1981) BORDEA & BORDEA (1973) and to the sheets Pietroasa (BLEAHU et al., 1985), (MANTEA et a/., 1987) and Biharia (BORDEA et al., 1988) of the geologic map of Romania, scale 1:50,000. THE BIHOR UNIT The sedimentary formations of the Bihar Unit include at their bottom Permian deposits, that con sist of breccia with crystalline schists fragments, with a red matrix, and of ignimbritic rhyolites. The Permian deposits, or directly the crystalline schists, are transgressibely overlain by a thick stack of Mesozoic deposits, that display two de tritic within the Werfenian (the

PAGE 194

194 Werfen Formation) and one within the Early Jurassic (the Gresten Formation), a stratigraphic hiatus in the Late Triassic and another one at the beginning of the Cretaceous, and the occurrence of three major carbonate depositions, in the Early Middle Triassic, in the KimmerirdgianTithonic and in the Barremian-Aptian respectively. The first carbonate series includes either grey and white dolomites, locally with black limestone in terbeddings, or black limestones (Gutenstein and Vida limestones), with grey-yellowish shales (Pestis shales) at their top. This series, ascribed to the Anisian, is overlain by Late Anisian-Early Carnian deposits, consisting of white reef lime stones (Wetterstein limestones) and white, tiled limestones (Padi limestones), locally with poly genic breccia and red shales, interbedded with white limestones or quartzite sandstones (the Zugai formation) at the lower part. In the south-eastern part of the Bihor mountains, in the median section of the deposits that are as cribed to the Late Anisian-Early Carnian interval, the Ordancua formation occurs, consisting of white, tiled limestones and red shales at its bottom, and of sandstones and violaceous shales at its up per part. In the Triassic limestones and dolomites are in cised the large karstic platforms Padi, as well as the closed basin Ocoale Scarioara. The Late Triassic has a strictly local occurrence west of Padi, where it consists of the deposits of the formation, that includes fine limestones and red shales, with sandstone-clayey cement. The Early Jurassic formations of the Bihor Auto chthonous transgressively overlie the Triassic de posits, occurring in a typical Gresten facies which includes quartzite sandstones and conglomerates, with Hettangian-Early Sinnemurian shales inter beddings, of 200-300 m overall thickness. The upper part of the Early Jurassic (Late Sinne murianToarcian) includes at its bottom encrinitic reddish and greyish limestones and pink quartz marly sandstones (Late Sinnemurian-Carixian), overlain by marly limestone and grey marls with cherts (Domerian) and black marls and limestones, with or without phosphatic nodules (Toarcian), the thickness of the entire stack ranging between 6 and BOrn. The Early Jurassic of the Bihor Unit includes lo cally, at its bottom, either a conglomerate with fragments of Middle Triassic limestones and with I. Or4feanu a carbonate matrix with oolitic iron (Sdrita), or a carbonate welded mega-breccia (east of Garda). The Middle Jurassic includes a succession of red oolitic iron limestones, yellow spotted limestones and encrinitic limestones, of about 10 m maximum thickness. The next major stack of carbonate deposits has been deposited during the Late Jurassic, which includes exclusively a carbonate facies, consisting at its bottom of white, reef, partly Stramberg type limestones (Farcu limestones) of Oxfordian-Early Tithonic age, and of layered limestones with blackish onchoids .(Albioara limestones) of Tithonic-Berriassian age at the top. The continental regime which occurred during the Early Cretaceous resulted in a paleo-karstic topog raphy and in the accumulation of bauxite pockets, that trace this hiatus. During the following stage, in the Early Cretaceous, a submerged limestone platform provided the environment for the deposi tion of the last major carbonate series, which in cluded fenestral lamination limestones, limestones with miliolids and limestones with orbitolins (Barremian-Early Aptian). Along the Horea-Ocoale lineament, the entire sedimentary series of the Bihor Unit occurs as a homocline, with an overall NW-SE strike, between Poiana Horea and Ocoale. In its northern half the structure dips south westward, while in its southern part a westward dip is recorded. As a general rule, there are neither recurrences of the succession due to reverse strike slip faults, nor folds. These structures have been built during the Turonian or the Middle Creta ceous, before the nappe of the Codru nappes. The Someul Cald graben is the northernmost compartment of the Bihor mountains. It is bor dered to the south by crystalline formations, which are juxtaposed via the Someul Cald fault, and to the north by the igneous formations of the VlMeasa mountains, the latter contact being sometimes overlain by Late Cretaceous forma tions. The Someul Cald fault is one of the major faults of the Bihor mountains, prolonged westward beyond the main ridge, up to the Bulz fault. The graben is filled mainly with Jurassic and Early Cretaceous deposits, and its structure includes a succession of transverse faults, that delimit a se ries of compartments with overall westward dip (MANTEA, 1986).

PAGE 195

The hydrogeology of the karst areas of the Bihor Vladeasa Mountains 195 THE CODRU NAPPES SYSTEM In the constitution of the Codru nappes system, carbonate deposHs display different weights and lithologic compositions (see insert within the hy drogeological map). Within the Vlilani nappe, carbonate deposits closely parallel those of the Bihor Unit, with the occurrence of the same three distinct limestone and dolomite series. The Ferice and Batranescu nappes, consisting of Permian and Triassic depos its, include at their bottom (Anisian) dolomites with sandstones and dolomitic schists, overlainwithin the Ferice nappe-by black limestones with cherts (the Roia Formation, Ladinian-Early Carnian), while in the case of the Batranescu nappe the upper term consists of black limestones with cherts (Ladinian) and white limestones (Carnian-Norian). Within the Urmlit nappe, a limestone stack occurs within the Late Jurassic, while within the Vetre nappe the carbonate deposits consist of the Frlisi nel dolomites (Norian) and the Bliita marble (Early Rhaetian). The Arieeni nappe includes black dolomites (Anisian) and limestones and dolomitic limestones (Ladinian-Carnian). POST-TECTONIC COVER At the end of the Cretaceous, three major geologic events have taken place: a) the overthrusting of the Codru nappes, during the Turonian; b) the forma tion of fracture systems, along which the subsi dence of sedimentary basins of epi-continental facies took place, with associated accumulation of Gosau type Senonian formations; c) intense subse quent volcanic activity. The Senonian deposits featuring a Gosau facies form the Late Cretaceous post-tectonic cover of the Bihor Unit and of the Codru nappes system. Such deposits outcrop over relatively restricted areas in the Someul Cald graben, as well as on the terrains covered by the Vllideasa igneous forma tions. The succession of the Senonian deposits in the SomeUl Cald graben begins with conglomerates with arenitic matrix and well rolled embedded gravel, that includes crystalline schists, limestones and sandstones. Dark grey-reddish clayey marls and micaferous yellow-green sandstones follow. The -reef facies of the Senonian consists of lime stones with many corals, while the volcanic sedimentary formation includes alternating ashes, tuffs, tuffites, sandstones, micro-conglomerates, breccia and conglomerates with terrigenous volcanic matrix (MANTEA, 1985). ISTRA TE (1978), in a paper dedicated to the pe trographic study of the Vllideasa mountains (the western part), distinguishes within the Senonian series a lower, sedimentary complex and a vol canic-sedimentary formation. The lower sedimen tary complex (the Gosau formation) includes a succession consisting of three sections: a bottom, conglomeratic one, a median, marly, sandy, mi caferous one, and an upper, micro-conglomeratic one. ALPINE SUBSEQUENT IGNEOUS ROCKS AND ASSOCIATED PRODUCTS The alpine subsequent (banatitic) magmatic activ ity is documented in the western part of northern Bihor mountains and along the northern border of this unit, by means of a large variety of rocks. Within the Pietroasa-Aleului valley area, and further north, up to Budureasa, granodiorites outcrop. They are part of a single batholitic body, that within Bihor mountains extends, both at the sur face and in the underground, up to the Galbena fault. An exception is recorded in the Bulz valley area, where an igneous body having penetrated along the previously mentioned fault outcrops. A multitude of veins of andesitic or basaltic compo sition, that have been identified especially in the upper reaches of Criul Bliita, along the valleys Hoanca Motului, Corlatul and Flecuta. as well as in the V alea Seacli catchment area, are of hypo abissal origin, being associated to the indicated banatitic intrusion. Vllideasa mountains are built up of rhyolitic rocks of different facieses, ranging from massive to vitrophyres, as a function of the place where the rhyolitic magma solidification has occurred (i.e. under the Senonian sedimentary cover or at the surface). In the evolution of the magmatic activity of this area there have been two outstanding events, namely the setting of the ignimbritic rhy olites formations and the setting of the intrusive bodies. The intrusion of the banatites has resulted in con tact processes that concerned the sedimentary de posits being traversed. At the contact of the banatites with the limestones, marbles and various types of calcic skarns have been formed, while at the contact with the detritic and pelitic rocks, hornfels, garnet skarns, etc. are met.

PAGE 196

196 NEOGENE FORMATIONS On the western rim of the Bihor mountains, Pan nonian (Malvensian) deposits consisting of clays with coal interbeddings, sands and gravel from the Neogene Basin filling OUtcrop. In the close neighbourhood of the mountains border coarse deposits prevail, that are however rapidly substi tuted by a pelitic facies, of wide occurrence across the entire Basin. Quaternary formations consist of sands, gravel, boulders and, subordinately, clays. They occur in the terraces of Pietros and of the other streams that originate on the western slopes of the Bihor mountains, in the present day streams allu via, in the ancient and the recent deluvial and col luvial deposits. A noteworthy extent have the de posits on the elevated karstic platforms Cetatile Ponorului, Batrana and Apa divide. They consist prevalently of sands with quartzite sandstone fragments, which have been carried away by runoff originating on the nearby slopes, then left in place, once the surface stream lets had taken an underground course through the carbonate substratum. V. HYDROGEOLOGICAL FEA TURES OF THE BIHORVLADEASA MOUNTAINS The deposits included in the geological framework of the Bihor mountains display a wide variety of lithologies and different intensities of tectonic dislocation. As a result, distinct hydrogeologic features occurred, which allowed the hydro geological separation of five types of deposits, with specific groundwater recharge, flow and dis charge characteristics (enclosed hydrogeological map): 1. Carbonate Mesozoic series (limestones, dolo mites), highly fractured and karstified, charac terised by very high effective groundwater flow. Numerous karstic systems of various size and prevelent by of binary type Spring flow rates up to 550 Vs. Important water resources in large karstic systems; 2. Sub s equent alpine magmatites (banatites) and metamorphites with permeability of fisures with discontinuous distribution and intensity. The weathering zones are well developed and provide a continous and important supply to the rivers (memory effect is 55 120 days for 0.2) and to the binary karst systems; I. Ortleanu 3. Prevalent by detritic Permo-Mesozoic deposits (sandstones and conglomerates with argi laceous shales and rhyolites) with different permeabilities. The groundwater flow is mostly confined to the fissured areas. They act as an caprock impervious barrier for karst water res ervoirs and frequently form bedrock and/or of the caprock of the latter; 4. Senonian postectonic deposits (sandstones, con glomerates and less frequently argilaceous shales) with local extension in the northern part of the map area. Senonian reservoirs supply springs with discharge up to 3 Vs, and also subjacent karstic reservoirs from the north and north-eastern part of the map area; 5. Marly and argillaceous deposits, devoid of groundwater flow, and flysch-like series, in cluding rock-complexes of variable permeabil ity (marls, argillaceous shales sandstones, lime stones), hosting occasionally discontinuous aq uifer accumulations occuring in the more per meable terms. VI. HYDROGEOLOGY OF CARBONATE TERRAINS Karst systems in the Bihor Vladeasa mountains are generally of binary type, those extending exclu sively over carbonate terrains being less frequent, with their occurrence mainly restricted only to the Apa Calda-Hoanca Seaca area. They display a wide variety of dimensions, lithologic constitu tions and performance mechanisms, that are mir rored by the physical, chemical and hydrogeologi cal characteristics of the springs. Table 3 indicates the major springs in the Bihor Vladeasa mountains, as well as the formation from which they emerge, their in situ measured tem perature and pH, the saturation index with respect to calcite and dolomite, the TDS content, the Mg/Ca ratio and th e measured or estimated annual average (over October 1984-September 1985) discharge The karst springs are situated at different eleva tions, as a result of the pronounced dissection of the carbonate deposits and of the rugged topogra phy. At the scale of th e entire karst region a gen eral base level cannot be outlined, each sp e cific karst area having its own base level. The karst springs flow rates extend over a very wide range, with a 550 Vs maximum multiannual average value recorded in the case of Galbenei Sprin g over th e hydrologic y ea r Octob e r 1984-S e pt e mber 1985.

PAGE 197

The hydrogeology of the karst areas of the Bihor Vllideasa Mountains 197 TRACER TESTS In spite of a large amount of investigations com pleted and of papers published prior to 1984 on the Bihar Vllideasa mountains karst morphology and hydrology, only a small number of tracer tests (6) have been performed in order to outline the groundwater flow paths Out of those tests, two have been performed by VIEHMANN et at. (1958, 1961) and the other et at. (1957), RUSU et at. (1970), (1.974), HALAS! & PONTA (1984). Starting with 1983, 36 new tracer tests have been performed by in co-operation with POP and TANASE, with rhodamine B, fluorescein, stralex (optical brightener agent pre pared in Romania), radioactive tracers (I-131, Br82) and activable tracers (ln-EDTA, Dy-EDTA), within the framework of a complex investigation program concerning the hydrogeology of the Bihar Vllideasa mountains karst, initiated by the com pany "Prospectiuni S.A.". The results of the tracer experiments have been published by et at. in 1991. Considering the tracer tests performed prior to 1991 as a whole, an average groundwater flow velocity of 45 mlhour has resulted, the longest identified underground flow path, 4500 m, being that between the pothole in Hoanca Urzicarului and Pliuleasa spring, while the largest elevation range covered, 665 m, was that between Muncelu cave and Blidaru spring. HYDROGEOLOGIC WATER BUDGET In order to evaluate the budget of surface water and groundwater in the Bihar VHideasa mountains, during the hydrologic year October 1984-September 1985 the national hydro meteorological observations network has been filled in with dis charge gauging sections, set up on the main streams, at sites where they left the carbonate ter rains, and at the main karst springs. For filling in the meteorological observations network, rainfall gauging devices have been installed at Runcu Ars and Vartop. Additionally, a meteoro logical platform, provided with equipments for ga u gi ng rainfall, evaporation next to wat e r surface and evapotranspiration by means of lysimeters has been built at Casa de Piatrli. The discharge gauging stations located at the out skirts of the karst areas in Bihar mountains have provided control over a 527 km2 surface area. Those s t ations are: valley d.g.s.), Crliias a valley (downstream spring d.g.s.), Galbena valley (lntre Ape d.g.s.), Bulz valley (Boga forestry hut d.g.s.), Cald river (Smida d.g.s.), river (Poiana Horea d.g.s.) and river d.g.s.) For the considered area, the rainfall recorded over the hydrologic year October 1984-September 1985 has amounted to 1220 mm. This value was computed by taking into account the data provided by rainfall gauging stations of the INMH network (Bliita, Pietroasa, Stana de Vale, Vllideasa, Smida, Poiana Horea, the values transmitted by gauging station correlated with Stana de Vale gauging and the data provided by the previously mentioned rainfall gauging stations During the same period, the evapotranspiration value obtained by processing the lysimetric data provided by the temporary meteorological station at Casa de Piatrli amounted to 374.6 mm. The available water amount (845 6 mm), distributed between runoff and seepage, has been recovered at the discharge gauging sections at the outskirts of the karst areas, which indicates that within the range of error of the primary data, there are no significat water transfers from or toward ajoining structural units. GROUNDWATER QUALITY Observations, measurements and analyses per formed at the main springs of the Bihar Vllideasa mountains result in the following considerations concerning the groundwater quality: A. The temperature of the kar st springs ranges between 5.4 and 10C, directly related to the elevation of the supplying karst system. Some springs discharge higher temperature flows, as a result of deep e r underground circulations along nappe or fault planes Thes e specific springs are: the warm spring at Valea Neagrli Izvoar e no 29) 17.2"C, the warm spring at Cotetul (Feredeu, no 112) 16.2"C and Izbucul Mic at Garda de Sus (no. 142)-14.4"C ; B. The pH of the discharged water is slightly alka lin e, ranging b etwee n 7.15 and 7 86; C. Five karst springs di s play gas outflows: the warm spring at Val e a Neagd Izvoare, no. 29), the warm spring at Cotetul (Feredeu, no. 112), the spring at the confluence of Paraul Sec with Cald (no. 1 40), the warm spring in Alunul Mic valley (no. 142) and Izbucul Mic at Garda de Sus (no 116), th e chemical compos ition of th e outflowing gas b e ing indicated, for the firs t 4 springs, in Table 4.

PAGE 198

Table 3; : Charaderi5Pesot the nirun springs -ill . .. .. Les sources de Bihor...;VJaauisa, \. : l : ',: No Source l 2 Hydrographic basin of Ncgru ril'cr I spring (2) 2 Cuciului spri n g (5) 3 Comilor spring ( 8 ) 4 Cerbasca spring (10) 5 spring (14) 6 Berbece s pring (IS) 7 Spr.of confi.(l6) 8 Aleu spring (I B) 9 Spr. of Valea Popii stream (I 9) 10 Ulmului spring (20) II spring (2 5) 12 WamJSpring Valea Neagr:i (29) 13 Cold spring Valea Neagrl (29) 14 Hidrei spri ng (30) 15 Blidaru spring (32) 16 cave (33) 17 Coliboaia spring (35) 18 Spring of F a na(e ( 43) 19 Springs ofTopli(a valley (45) 20 spring (50) 21 Bog a spring (56) 22 srring (57) 23 Bulbuci spring ( SX) 24 !zvorul Rcce sprini; 25 l7. bucul Ursului s pring (69) 26 Galbenei spring (X3) 27 spring CX4) H . -. L (a.m.s.l.) Lithology. 3 .. 4 1000 granodiorites 830 lms ( n o+rh) 425 lm s (J3 ) 570 dol. (an) 675 lm s. (rh) 485 Lms. (ld-no ) 435 granodi ori tes 950 Mg-lms (rh1 ) 1300 rhiolite s 575 ss (w) 505 dol. (an) 375 d ol. (an) 376 dol. (an) 390 hns (ox-v ) 435 c (ox -v) 500 c (ox-v) 515 1ms. (ox-v) 450 dol.(cr.3-no2 } 525 dol. (an) 700 dol.(cr1n o2 ) 675 dol. (an) 910 dol. (an) 900 (an2-cr,) 1075 lms (br-ap1 ) 1095 lms (an2-cr1 ) H60 lms (br-ap ) 570 lms. (br-ap1 ) pa !base' 5 : 6 1: 7 .. 1<.5 6 39 7 5 8 8 lU 8 .5 7.9-10 7 6 10. 2 7 87 .. 6 .0-6. 2 7 55 6 6 7.6 7 8 7 77 R.5 1 7.2 7 3 + 10. 6 7.6 8.7 !1. 6 9.4 9 2 1<.9 8 7 7 0 6 8 7 0 6 5 5 4 5 6 7 .15 6.3 5 6.6 s.r .. ,t.n.S 8 .,. _.9';:. : -2 .43 7 63 156 5 ; 172 0 330 0 -0 303 -1.417 225.. 2 0 282 -0.307 317.1 0. 710 1.25.9-349 3 (4) -1.498 4 : 038 62. 5 -0 616 -1.779 121.5 381. 7 465 0 (5) O.i45 -0 72 0 415.4 0.112 ..{).773 3 20 .7 271.1 2 69 7 453 .0 302 3 208 7 272.1 (4) 258 .0 396 5 29 4 8 -0. 537 -2.853 247.7 306 5 (4) 2X6.8 ; 275 0 255.4 ; 262 5 -:'1\):::: 0 0 0.201 0.126 0.05 9 b 0.08 0 .236 0.45 7 0.07 4 0.05 5 0 1.22 9 0 0 0 .440 0 11 8 0 163 0 23 7 0.0 3 6 .iJ.38 0 O.J l lfJ __ :: .. 2 10 1 0 35 4 4 3 9 2 3 79 2 3 55 70 25 2 0 15 20 215 500 50 100 4 0 20 550 462 -\0 00 !'-< ..., ::: ;::

PAGE 199

Hydrographic basin of Ariel} river 28 Gura Apei spring (90) 1125 dol. (an) 6.0 372.2442.6 .(3) 0.439 . 61. i9 Apa din Piatla spring (93) 1100 dol. (an) 6:2 302:5 : 362.4 0,098 : 23 .. 30 Coliba Chiobului spring (94) 1100 dol. (an} 6.2 332:.8 0 30 31 Vulturului spring (101) 1040. lms. ( 6.8 326.7 ; 390.8 o 75, ... 32 Tlluz spring (I 04) 850 lms. (ox-be) 7.5 255.4373.0 (4) 0.056 530 33 Corobana spring (105) MOO lms. (an2-cr1 ) 7.1 .286.1 55 34 Cotetul spr. (Ill) 770 dol. (an) 7.2-7.5 7.5 -0.009 -0.381 3442 "470.5 (6) 0.314 280 .. 35 Warm spr Cotetul (112) 757 dol. (an) 16.2 7.54 + -0.095 -0.089 296.7 o.i34 5 36 Morii spring (113) 760 dol. (an) 7.2-7.5 386.0;447,7 o:332 .. 20 37 lui loariel cave (114) 810 dol. (an) 7.7 ; 7.51 0.069 .. -0.300 274.6 436.0 (5) o:t6s .... 90 ; ...... 38 lzbucul Mic spring (116) 730 6.8 7:74 + -0.092 0.056 285.9 464.9 0.892 to 39' Izbucul Mare spring (117) 725 lms. 6.2 7.38. -0.044 348.1 463.8 0.083 ... .. ... 45 Hydrographic basin of Someljul Cald river 40 Apa Calda spring (123) 1120 dol. (an) 8.0 444.3 ().149 40 41 .Spring of Hoanca Seadi (124) ',1150 dol. (an) 6.3 529.7 : 0.270 ; .. 42 Olineasa.(Rece) spring (127) 1170 dol .. (im) 6.5 ... 461.2 o.337 25 '43 Izbucul Mic spring (129) 1255 dol. (an) 6.5 363.4 0.363 : 40' 44 Alunul Mare spring (136) 1180 lms. (ox-be) 5.6 . 7.72 .,0,322 i77.3 ; 218.5 0.013 no 45 Pepii cave (139) 1125 lm$. (ox-be) 5.6 7.86 0.159 -1.667 : 264.9 ; 320.1 0 20 (140) 1095 aluvia (Q) i!.8 7.8 + -0:223 -1.574 171.4; 345.5 0.054 30 :. Alunu) Mic spring (141) IHlO (ox-be) 9.0 7.64 -0.390 -2.245 167.6; J22.1: o.o11 180 Wilrm spring Alunul Mic (142) 1065 lms. (f3). 14.4 7 .77 + -0.035 -0.832 206:2 ; 247.0 0:074 .. : 6 of din Firea, 149 1070 lms (br-ap1 ) 7.5 372.4 o 100 -------------Hydrographic basin of Negro river 50 .. spring (156) 1175 dol. (an) 6 0 7.47 166.5 0 033 105 51. Izvorul Minunilor spring ( 157) 1150 dol. (an) 5.8-6.2 7.34 -0.820 -2.323 '139.3 0.190 15.6 52 Ranipei spring (159) .1145 dot.'(an) 5.4 5.6 7.37 -0.744 -2.509 187:4 0.087 15 53 : lzvorul Plistriivliriei spring (160) 1100. dol. (an) 5.4" 6.0 7.53 -0.555 180.5 0:. 50 54 Pavelspring (161) 101!5 dol.(an) 6.1. 7.44 -0.597 -1.915 211.3 0:158 7 55 spring (162) 930 sl;,p(sn) 7.8 (7} 7.65 c0.900 99.8160.2 (6) 0:008. 12' 56 Raduspring (163) tOss doL(an) 6.4 7.24 -0.613 278.0 o.sns 2 57 Ariei spring (164) i325 rhyolites 5.8 7.63. 4 472 -9.437 65.0 0 .21!5. .. I Note: In brackets: C;Jiumn 2 number of the spring im the map; Culunui 4 Column 5-nu;nber of measurements; Column 9-number of Abbreviations: S.l. saturtion index; Q-mrn annual discharge (X.l9l:!4-IX.I985); M-measured E estimated discharge; lms. limestones; dol. ss.sandstones; p-rhyolites. ::tO II> .;:to 't .,. <::1 I)Q . <::1. & s.. II> ir a. C ., .. ... II> c .... <::1 ._,. ;;:. bo ., :;. <::1 ., . <:<:. .. B< ... c . <::1 .. ;::, :: s : s ... .... 'C 'C

PAGE 200

200 The gas outfows have a composition siniilar to that of atmospheric gas, hence they consist of the air dissolved in the water of the springs, thatleaves the solution ensuing to the increase of the water temperature. D. The computed saturation indexes (Table 3) in dicate that the water of most karst springs in Bihar VH'ideasa mountains is undersaturated, to a larger or smaller extent, with respect to both calcite and dolomite. The warm and cold springs at Valea Neagra (no. 29), the stream emerging from Pepii cave (no. 139) and the spring Poarta lui Ioand are supersaturated with respectto calcite, the latter spring having large associated travertine de posits. Izbucul MiC at Garda de Sus has tion indexes very close to equilibrium, its water being slightly supersaturated with respect to dolomite. The water of the springs in the catchment areas of the binary karst systems is strongly under saturated with respect to calcite and dolomite, inducing as a result of its aggressivity an intense dissolution of the carbonate deposits. Quite typical in this respect is the water of the springs originating in igneous formations (for instance the spring of Valea Popii, no. 19 and Ariei spring, no. 164 ), which explains the in tense development of the karst within the car bonate deposits in the Cald graben induced hy runoff water originating on the southern and south-eastern slopes of the Cor nul-MiclauVladeasa ridge. The water of the karst systems is of calcium bi carbonate, magnesium bicarbonate and magnesium-calcium bicarbonate type, depend ing on the chemical composition of the trav ersed formations (limestones and/or dolomites), with TDS values ranging between 125 529.7 mg/1. Stiff chemical composition diagrams for water (Fig. 3) indicates a larger TDS content for the springs that either have their supply derived exclusively from large extent carbonate forma tions, or include a small nonkarstic catchment basin that supplies part of a large karst system (for example the karst systems in OcoaleGarda de Sus and Apa Calda-Hoanca Seaca ar eas). The karst systems of small dimensions and with a larger contribution of the nonkarstic catchment basins have lower TDS content (for example the karst systems on the western slopes of the VHideasa E. The major elements contents of the karst springs water in the Bihar VlMeasa mountains range below the maximum accepted concentra tions stipulated by the ST AS 1342-91 regula tion for drinking water. For some of these springs however, the bacteriological and the toxic elements content does not comply with the requirements of the regulation. F. In order to evaluate the bacteriological content of the karst springs, specific analyses have been performed at the Preventive MediCal Center in under the leadership of dr. Mocuta. Table 4. La compo$itiOtl 'de$,, ,:;; . -. . ' .. ' . ... . No. .. ::.:-,>"' Warm spring of Valca Neagra (29) 2 Spring of conf1encc Cald -Piiriiul Sec (140) 3 Warm spring of Alunul Mic stream (142) 4 Warm spring of Cotetul Do( 112) 1.5 17.2 415.4 15.0 8.8 171.4 5.0 14.4 206.2 2.5 16.2 252.3 *In brackets number of spring on hydrogeological map . 6.90 19.42 72.80 1.27 20.20 77.59 0.31 20.26 78 .15 0.54 17.83 80.81 Other compounds for which the gases were analysed, c;Hz, C3H, C4Hw, He and H2 are lacking. Ar 0.86 0.89 0.91 0.79

PAGE 201

The hydrogeology of the karst areas ofthe Bihor Vllideasa Moutttaiizs 201

PAGE 202

202 2500 T M t E 2000 + u I (f) r co t c 15001 Q) t5 co ..c -1000 0 L.. Q) r ..c E 500 :::1 z r 0 .Cb.,.frl-t '! 1 1 [ D N.t.b ./cm3 Iii N b.c./cm3 I. OrtieatfU ' <0 N Number of water sources Fig. 4. Displays of the total number of bacteria (N.t;b.) and_ ofthe p r obable number of coliform bacteria (N.b.c.) in spring . .. .. - . -Le contenu ba c t eriol o g i q i t e total e t le numero le plus probable d esbactedes c ollifo r m es dan s les sources kars tiques. For urinking water provided by local sources (wells, springs, etc.), the STAS 1342-91 regula tion stipulates the following maximum accepted bacteria contents: the total number of bacteria, (N.t.b.), devel oping at 370C/cm' should be Jess than 300; the probable total number of coliform bacteria/urn\ (N.d.), should be Jess than 100; the probable number of coliform fecal bacteria/dm', (N.c.f.), should he Jess than 20. Figure 4 displays the total number of bacteria and total coliform bacteria contents of some karst springs, illustrating a wide range of contents, with the karst systems supplied by swallets (for exam ple the springs Cornilor, Hidrei, Camenita) ex hibiting a higher bacteriological content, as op posed to those systems the recharge of which is diffuse and/or is taking place through an aquifer confined beneath an aquiclude (the springs Aleu, lzvorul Minunilor, etc.), that have a smaller bacteriological load (see 1994) Along one year time span, the bacteriological content of the karst springs water is subject to important lluctuations, relatively high contents being associated to the rainy or snowmelt periods, as compared to drought periods. G. Toxic elements content. The STAS 1342-91 regulation stipulates that in drinking water maximum allowed concentrations are 50 pph manganese, copper, chromium and lead, I 00 ppb nickel, and 5000 ppb zinc. Exceptionally allowed concentrations are 300 ppb manga nese, 100 ppb copper and 7000 ppb zinc.Thc analyses of 26 water samples, performed hy means of the ICP equipment of the "Prospeqiuni S.A." laboratories, indicate Mn, Cr, Ni and Zn contents below the maximum allowed concentrations. On the contrary, in the case of many springs the maximum allowed lead concentrations arc exceeded up to two times, while those of copper are exceeded up to six times (Fig. 5 and 6). The springs Aleu ( 18), Izvorul Minunilor ( 157), Izvorul Rece (68) and Izvorul Radu (162) display extremely low contents of the considered micro elements. Relatively high Cu, Pb and Zn con centrations occur in the springs (2) and Ariei ( 164), that emerge from banatites, and in the springs emerging from the dolomites that outcrop in the southern end of the study area: Cotetul ( 111 ), the warm spring at Cotetul Dobretilor ( 112), Poarta lui Ioanel ( 114), Izbucul Mic (116) and Izbucul Marc (117).

PAGE 203

The hydrogeology of the karst a r e a s of t h e B ihor Vli1deasa Mow,tains 203 BOO 700 600 .0 500. Q. Q. : ... :. vi c 400-QJ c 0 300 . u 200-'\ ., 1 0 0 . 0 ,...( ll)
PAGE 204

204 I. abl.e S. tJ!espr}rigs .. s debits : : Source J'auz ... J:'iiuleasa t < Alunul : Iz:vottd ., Giulett ._livorul Parameter ,': ... . ... Mlc :'crij\tl.rl . : --. Mtnunilor .. Q n1eon.l/s 529 477 306 274.7 217.9 77.7 19.2 Q 1/s 68 180 2 0 58 3 1 5.6 Q m:nimunl' ll s 4640 1920 3160 2120 826 57 1 30.0 11_. (Qma/Qmin) 62.2 10.7 1580 14.2 190.3 1.9 ME, days 18 31 16 31 37 7 44 RT. days 24.6 42 23.8 34.5 42.7 12.3 46.4 TF 0.208 0.092 0.112 0.160 0.172 0.420 0.096 n,, index of variability; ME, memory effect; RT, regulation time; TF, truncation frequency able 6. Main parameters > .. s qui If! . Source Tauz Pauleasa Alunul Parameter .. Mic Period of reces-26.06.85 24.01.85 22.06.85 sion 30.09.85 04.03.85 06.08.85 Qo 1490 670 560 4 h () 290 316 283 a 0.0116 0.0093 0.0231 4n 1200 354 227 l, 32 13 1 5 Y] 0.0313 0.0769 0.0667 E 0.114 0.339 0.046 Y} 2. 1 6 2.94 1.06 y() 0.87 0.97 0. 1 5 Yo 3.03 3.91 1.21 Y0 I Y0 err, 7 1 75 88 Yo"/ Yo% 29 25 12 Cotetul IZvorul Giuleti Izvorul Dobretilor Critdui Minunilor 06.10.84 09.08.85 20.11.84 31.12.95 1 9.11.84 30.09.85 19.12.84 04.06.96 368 349 124 29.7 106 100 73.8 21.6 0.0495 0.0078 0.0549 0.0034 262 248 50.2 8.2 14 17 9 32 0.0714 0.0588 0. 1111 0.0313 0.446 0.394 0.025 0.042 0.185 1.290 0.116 0.556 0.066 0.073 0.018 0.008 0.251 1.363 0.134 0.564 74 95 87 99 26 5 13 1 Q0 total discharge at the beginning of the recession, (1/s); q0 discharge at the beginning of the recession for the base flow. (Vs); a (day "1), baseflow coefficient; q0'. discharge at the beginning of the recession for the quickflow, (1/s); t,, duration of the quickflow, (days); YJ and E, parameters adopted for the curves of the quickflow, (day1),; Y0, initial volume which will be drained during the quickflow m3); Y0 initial total volume stored in the aquifer, (10" m1); V.,", initial volume that will be drained during the basellow (106m'). ln the correlative and spectral analysis of the flow rates, we used the methodology proposed by MANGIN (198 1 a, 198 lb, 1982, 1984) and in the study of the recession, the papers published by MANGIN (1975) and PADILLA et a/. (1994) (Fig. 9). TATAROAIA KARST AREA Tataroaia karst area is developed in Anisian dolo mites and Ladinian limestones that occur as a strip extending between Craiasa and Galbena valleys. Most of this karst area overlies the karst system of spring, tharis supplied almost exclusively by rainfall and does not include a non-karst catchment basin.

PAGE 205

The hydrogeology of the karst areas of the Bihor Vliideasa Mountains 205 The relatively scarce exokarst landforms are re stricted to the sinkholes in the Varcioroagele plateau; in contrast, there are several significant ea vi tics, among which it should be mentioned the pot hole "Gaura care sufH\" and two major stream caves, Micula and Fagului, the latter being discovered ensuing to the excavation of a geological exploration mining gallery. The stream in Fagului cave emerges in spring, as indicated by the rhodamine tracing experiment performed on 7 October 1984. Figure 7 illustrates the simple correlograms and Figure 8 the variance density spectrum of the flow rates series. Some additional supply to the system might also originate in the sinking stream of Valea During heavy rainfall periods the water transfer capacity of the cracks and channels network of spring is exceeded, so that part of the flow is discharged through the entrance of Micula cave, that behaves as an overflow to the system. Besides Giuleti spring, the aquifer located in the carbonate deposits of the Tataroaia wedge discharges also through the springs at the fountain head 6f Fagului valley, of lO 1/s average cummulated flow rate, and through the !PEG mining gallery excavated eastward from Prelucilor valley. along 3 km, that discharges on the average some 15 1/s of 15 "C water. To the north, this aquifer supplies the springs at the fountain-head of Pauleasa and Buteasa brooks (Bonciului and La Scoici springs) ; the latter area being designated by the locals by the name "Ape Caldc" (Warm Waters). The global interpretation of the data concerning spring contained in Tables 3, 4 and 5 indicates that its discharge is derived from a karst system of very poor inertia. subject to intense karst development, prevalently conductive and much less in what concerns its storage capaci ty The reserves of the system are small, the weight of the fraction discharged by the fast now amounting to 13 % There is no signi11cant filtering of the rain fall induced information, so that heavy rainfall is immediately followed by intense Hoods The discharge of the springs quickly decline s after th e rains stop, and as a consequence the prolonged draught periods result in a drastic reduction of the flow rate, occasionally till the discharge ceases completely. Fig. 7. Analysis of dis. charge time series of seven karstic systems:J. Izvorul Minunilori 2: lzvorul Criului; 3. Piiuleasa; 4 Cotetui Dobretilor; 5. Alunull\lliC, 6 ; Tauz; 7. Simple correla tion of the flow rates {m=125 days, k =l): Analyse des chroniques de debits de sept systemes kaistiques: J.Iz-..orul Mimmilor; 2. lzvorul. lui; 3 : Piiulea!ia; 4; Coteful Dobre!jtilor; 5 . Alunul Mic; Ttluz; 7. Giuleti. Corridogramnte simple de de. (m=l25jours, k::l).

PAGE 206

206 .. Qo .. b q .. .0 .. 0 tT TIME Fig. .diagrJil,i" hqwing basenow. r: :al. Dfiig { am'nii sants : d' IIi jQrmbnellt a !le > MAN..fJ(, i .. ( PADILLA e(al. 1994). .:-. :' :.:.: ; 1 .,-. KARST AREA I. Oraeanu This karst area, that extends between Craiasa and Criul Baita valleys and is traversed along its main axis by the Sighitel valley, is dominated by the central position of the Arieeni nappe thrust outlier located on Prislop peak, marked by a multitude of swallets through which the runoff water sinks in the underground at its entrance on carbonate do mains. When additionally considering the aquifer concentrated supplies originating on the western slopes of the Tarlau peak (Secatura valley, Cheia Rea valley, Muncelului valley) and those in the area of Dosurile peak, a realistic image is obtained on the multitude of the impact points between surface water and limestones, that over the ages resulted in the excavation of an impressive number of caves. Ensuing to the intense tectonic dislocation of the area, one karst aquifer occurs, the resources of which are subject to competition between many large flow rates springs. In the valley upper reaches, on its right hand side, arc located the caves Coliboaia and Piolca. Access in the first

PAGE 207

The hydrogeology of the karst areas of the Bihor Vliideasa Mountains 207 of those caves does not follow the stream way, which in turn emerges via a group of 20 lis aver age flow rate springs, at the bottom of the scree slope beneath the cave entrance. The 25 1/s aver age flow rate stream in cave is probably supplied through the karst area next to Grohotilor peak and through the diffuse sinking of Pietrele valley. The most important karst spring along valley is Blidaru. It has 70 1/s average flow rate and during flood periods the cave Rasuflatoarea Blidarului, located 5 m upslope, acts as an over flow to the karst system. Tracer tests have indicated that Blidaru spring discharges the entire amount of water derived from the north-western slopes of Prislop peak and from the western slopes of Tarau-Pietrele Negre ridge. The tracer has been injected at the diffuse sinking points along Preluca Nasului valley (fluorescein), Secatura valley (rhodamine) and in the surface stream that enters Muncelului cave. The tracking of the tracer has been performed by means of active coal filters immersed in the main springs on the valley (Coliboaia, Blidaru, Hidrei), in the spring of Baita and in the underground mining works of the Baita Molibden mine. While the fluorescein and the I-131 have been carried only toward Blidaru spring, rhodamine injected in the sinking section of Seditura valley has been carried, due to an under ground diffluence, both to Blidaru spring (most of the tracer amount) and to Coliboaia cave. We pre sume that originally Seditura swallet had dis charged via Coliboaia cave, the pothole in Sed(tura (Arago) and the above mentioned cave forming a unitary karst system. Coliboaia cave currently acts as an overflow to Blidaru spring. Some 20 m downstream Blidaru spring there is another spring of 10 1/s average flow rate. The fact that small tracer amounts have been recovered from this spring too indicates that it belongs to the same aquifer system. Hidrci spring is the second largest outlet (35 lis average flowrate). It is impenetrable and occurs at the bottom of a 15 m high Tithonic limestones cliff, discharging mainly the water originating on the western slopes of Prislopul peak, as indicated by the fluorescein tracing experiment completed on 28 September 1984 at the diffuse sinking point along Sodolul valley. Considered as a whole, the karst springs in catchment basin are supplied by a single karst aquifer that is subject to an intense contem porary evolution of the karst processes. The large number of completed tracer tests 1991) indicate that groundwater flow currently concentrates toward Blidaru spring, with an obvi ous tendency to abandon outlets situated upstream (Coliboaia and and to augment of the flow rates discharged by Hidrei spring, located at the bottom of the erosional level of the considered karst area. During the previously indicated observation pe riod, stream had an average flow rate of 456.5 1/s, with the extreme values ranging between 2130 and 110 1/s. The most outstanding underground feature on the western water divide between Craiasa and valleys is cave at with a small stream running along its bottom floor. In the same area, at the fountain-head of Izbucului val ley, in a small grassy depression on its left hand side, at the site called by locals izvoare", occurs a group of springs, the one situated down stream displaying gas outflows, a temperature of 17 oc and a flow rate of 4 1/s. In its close neigh bourhood, immediately upstream, two other springs of 8 oc and 5 1/s cumulated flow rate emerge from the scree. KARST AREA The Werfenian deposits on the south-western slopes of Magura Vanata mountain, the Hettan gian-Sinemurian ones occurring north of Iezere and in Groapa de Ia Barsa, and the Permian ones in the area of the Glavoiu and Bortig peaks, favour the organization of a well-defined surface drain age, among which worth mentioning are, due to their large flow rates and their perennial character, the valleys Cutilor, Renghii, Arsurii, and Ursului. When entering carbonate terrains, the surface f1ow is abandoned for alternative underground paths, directed toward the springs in the catchment ba sins of Bulz and Galbena valleys. The tracer tests performed in order to identify the destination of the streams sinking through the multitude of swallets in Ponorului closed catchment area, have outlined underground flows from subordinate closed basin and from the northen half of subordinate closed basin, toward Boga spring. The surface streams in the southern half of subordinate closed basin, Arsura, and Garjoaba, enter first an underground course down to the spring in Poiana Ponor, wherefrom they take another, about 500 m long, surface course, eventually

PAGE 208

208 to sink again into an impenetrable swallet and to emerge shortly afterwards in the outlet next to the entrance in Cetatile Ponorului cave, where from they continue their underground course down to Gal benei spring. To the supply of this latter spring also contribute, directly or via the underground cm,rse in Cetlitile Ponorului, the water collected by the subordinate closed basins Paragina, Barsa Cohanului, Cetlitilor valley and Lumea Pierdutli, the latter having a significant impact due to the contribution of Paraul Sec and of Izvorul Ursului brooks, that sink in Caput cave. The inclusion of Balileasa subordinate closed basin is open to de bate, since there is no perennial stream course to be traced. The karst system of Galbenei spring and that of Tauz are the systems with the largest extent in Bihar mountains. They are complex karst systems, that include in their constitution several karst drainage units (i.e. simple karst systems). The average annual flow rate of Galbenei spring during the considered hydrologic year has been 550 Vs. The left side of Galbena valley, upstream of the junction with Pliuleasa stream, displays advanced karstic stream piracy phenomena, especially in the catchment areas of Valea Sead and Crianul. Subsequently to diffuse sinking in a swallet lo cated downstream of the boundary with the Per mian sandstones of the overthrust, the water of Tiganului valley, a tributary of Valea Seacli, emerges in Pliu1easa spring, as indicated by an In-EDT A tracing experiment. Pliuleasa spring is also supplied by stream, that sinks un derground before the junction with Galbenei spring Valea Seadi has its fountain-head located in Groapa Ruginoasli and displays a temporary flow regime, both upstream and downstream of the banatite body that outcrops in its median section. In August 1984 the water sinking in the upper section of the valley has been traced with In EDTA For identifying the subsequent ground water flow direction, the springs Pliuleasa, GiIzvorul Criului, all from southward located CriUl Baita catchment area, as well as the water discharged by the 3 km long Valea Sead direc tional gallery, that runs from Bliita Molibden minin g area southward, down b e neath Tapul peak, have been monitored by means of adequate meth ods The tracer has been identified at the latter two monitoring stations, in an advanced degree of di lution, over the time interval 2 25 October 1984 (at Izvorul and over the time interval 24 A u g u s t 10 S epte mber 1984 (in Valea Seacli gal l e ry). The tracing e xperiment has proven th e I. Oriieanu continuity of the Tithonic limestones of the Bihar Autochtonous beneath the Permian quartzite sandstones, substantiating by means of a hydro geologic investigation method the napping posi tion of the quartzite sandstones. At the same time, it has proven that the radius of hydrodynamic in fluence of the minig works in the upper catchment basin of Bliita trangresses the surface water divide of this stream. Pliuleasa spring has a 477 Vs average flow rate, with a 25% contribution of the fast flow to the volume discharged during recession period. The relatively long duration of the rain unitary impulse influence (42 days regulating time) and the significant memory effect (31 days), indicate Pliuleasa system to have important groundwater reserves, while the small values of the cut fre quency (0.092) mirror the influence of the surface runoff in the spring supply. The flow rate gauging performed along Galbena stream has identified underground sinking, that occurred between the junction with Lunqoara valley and Pliuleasa spring. For instance, on 12 July 1985 the flow rate between the two gauging sections diminished from 426 to 370 Vs, while on 9 October 1985 it decreased from 130 to 90 Vs. We presume that water sinking is related to the position of Galbena fault, that in this section acts as a drainage path. The hypothesis that those stream losses might be recovered at Pliuleasa spring has not been confirmed by the tracer tests. Between Pauleasa spring and the junction with Bulz stream the flow rates of Galbena have a nor mal evolution, no significant inflows or outflows being recorded within the error range of the gaug ing methods. Upstream of the junction with Valea Rea stream, Boga stream has an annual average flow rate of 700 Vs, provided by Boga spring (500 Vs), spring (50 Vs), Bulbuci spring (100 Vs), and by the other springs, of s e condary interest, existing in th e catchment basin (50 Vs). Along a section where no tributaries exist, be tween the junction with Valea Rea and the junc tion with Plaiului valley, Boga valley displays significant sinking into the streambed amounting to about 20 % of the entire flow rate. The topogra phy of thi s spe cific zone app e ars as a wid e v a ll e y section, abundantly covered with alluvial deposits where Boga cottages settlement is located. The abundance of the alluvia is the result of the d e po sition of part of the suspended solids carried by the stream, which occurred wh e n the latter diminish e d its flow ensuing to under g round s inking. T h e sinking i s r e lated to the pr e sence of th e ma jor

PAGE 209

The hydrogeology of the karst areas of the Bihor Vliideasa Mountains 209 draining fault of Bulz, along which the valley is incised. Except for this specific section, the hy drogeological role of the fault is not known; it can be only stated that gauging performed down to the site called "Intre Ape" has indicated that the flow which sinks upstream_ is not recovered along this section. TAUZ KARST AREA The geologic structure in Garda Seaca upper reaches, that includes extensive carbonate terrains occurring as a monoclinal structure, subject to intense tectonic dislocation and covered by the detritic deposits of the Arieeni nappe, favoured the occurrence of several large flow rate karst systems (Gura Apei, Apa din Coliba Ghiobului, etc.). The discharge of those systems supplies the most upstream section of Garda Seadi stream, that sinks in the cave Coiba to eventually emerge in spring. Downstream the cave Coiba Mid, starting from Casa de hamlet, Garda Sead valley carries water a new and additionally receives a strong left hand tributary, Vulturului valley, supplied by the spring with the same name, of 75 lis average flow rate. The complex Tauz karst system has an annual average flow rate of 529 lis and a 68.2 ratio be tween the extreme daily average flow rates re corded over the observation period (October 1984 -September 1985). When all described springs are consid ere d, it can be observed that the fast flow of Tauz has th e highest weight (29%) with respect to the water volume discharged by the spring during the considered period of flow rates recession, which is normal if the prevalent supply of the aq uifer via Gardioara stream is taken into account, and which is also corroborated by the large cut frequency (0.208), characteristic to systems that are highly inertial and that hav e und ergone int ense k a rst development. The relatively small value of the memory effect (18 days) indicates relatively s mall groundwater reserves compared to the very large surface area of the system. The rain has a sma ller infl u ence period (26.4 days) than in the case of the other sprin gs (Table 5) KARST AREA The water divide between Garda Seaca and O r dancua v alleys, dominated by th e Ocoaleclosed catchment basin, wa s th e object of many s peleological investigations, stimulated by the existence of the glacier, the larg est cave glacier in Romania Those investigations have been paralleled by observations concerning the groundwater flow directions, the hydro geological connection between the pothole in and Pojarul Politei spring, as well as that between the Ocoale valley stream losses and Cotetul and Morii springs being out lined by means of fluorescein tracer tests. Cotetul Dobrqtilor spring is the main outlet of the Ocoale-Cotetul Dobrqtilor karst aquifer. The average flow rate recorded over the previously indicated hydrologi<: year has been 280 lis while the maximum monthly flow rate has been 1.06 m3/s. During draught periods the spring flow rate declines progressively to complete dry out the outlet being actually an overflow of the system. The perennial outlet is Morii spring, located some 100 m downstream, and the springs that occur along the left side of Garda Seaca valley, over the specified 100m distance, at stream level or below. The cumulated flow rate of those springs and of Morii spring has been occasionally gauged during periods when Dobretilor spring had dried out, resulting a value of 85 1/s. The parameters provided by the correlati vc and spectral analysis indicate the karst system Dobretilor to have relatively important ground water reserves. The variance density spectrum displays amplified cycles (due to the rain) for cut frequencie s lower than 0 16 amplification th at is due to the regulating capacity of the system. On the right side of Garda valley, opposite the junc tion with Cotetul Dobretilor sp ring a s ub th erma l spring (Feredeu), of 15.8-16.2 oc and with intense gas outflows emerges from the alluvia of the flood plain next to the stream channel. Its occurrenc e is related to the deep flow of th e karst water along the nappe plane of th e Arieeni napp e. The southern part of the Ocoale-S dlrioara closed catchment basin discharge s via Poarta lui Ioanel spring. Over the previously mentioned hydrolo g ic year, the average flow rate value has been 90 1/s, and that of the variability index has been 17. 2. PRELUCA KARST AREA This name has been used to d esig nate the lime stone water divide betw ee n and Mare valleys, the topogr ap hy of which is mark ed by the sinkhole plain in the area of Preluca hamlet and by the gorg e of Ordancua. Th e dominating hydrogeolo gica l feat ur e of this area is th e massiv e diffuse, t em pora'ry complete sinking of stream along its last 2 krn section before the

PAGE 210

210 junction with Garda Seaca stream. The water is recovered in Izbucul Mic and Izbucul Mare springs, on the left side of Arieul Mare. Izbucul Mare emerges at the contact of the Wetter stein limestones with the Permian deposits of the Moma nappe, and discharges an average flow rate of 45 lis, derived from seepage across Preluca sinkhole plateau and from diffuse sinking, that occurs along the median course of Preluca stream at crossing the above mentioned limestone area. CALDA KARST AREA The easternmost occurrences of carbonate deposits in Bihar mountains are located in Beli-Apa Calda area. They consist of Triassic dolomites and limestones, that occupy a synclinal structure with quartzite sandstones at the bottom. The supply of the structure is derived from rainfall and its dis charge is directed eastward, mainly toward Apa Calda, a 50 lis and 7"C spring. In the catchment basin of Bcli stream, the only significant karst spring is that in Hoanca Seaca. The latter dis charges a 10 lis average flow rate, derived from water accumulations in the northern part of Clu jului peak. The upper section of Beli streamcourse displays a temporary flow regime, while its left side tribu taries, running on non karstic terrains, arc sinking in the streambed when penetrating in the karst domains. CALD GRABEN KARST AREA Runoff originating on terrains consisting of igne ous rocks of VUldeasa massif and of Werfenian and Senonian deposits existing in the northern part nf Somqul Cald graben, supplies a widely dcveloped karst aquifer that discharges through impor tant flow rate springs. An outstanding position among them occupy, as a result of their significant flow rates, the springs Alunul Mic, Alunul Mare and the springs in the area din Firca. Alunul Mic spring has a karstic system that ex tends northward to the upper reaches of Ponorului valley, an area that for most of the time of the year supplies the system via the swallet of Ponorului valley. The tracer tests performed by injecting In EDTA in the streamway of Diaclaza cave, fluo rescein in that in Lucii pothole and rhodamine in the swallet of Ponorului valley have indicated an active karst flow. Alunul Mic spring has a flow rate that fluctuates over a very wide range (2-3160 1/s), with an average of 180 lis. The base flow is prevalent (88%) in the water volume discharged by the spring during recession periods. The system has undergone intense karst development, it has a very poor inertia and small groundwater reserves. Between Ponorului valley to the west and Firii valley to the east stretches Humpleu karst plateau. The plateau is built up in Barremian-Aptian lime stones and has as an outstanding speleological mark the presence of the cave with the same name, one of the largest in Romania (ONAC, 1995). The cave extends from Firii valley up to the environs of the swallct in Ponorului valley, and provides a major drainage path for the water accumulations in the plateau. The tracer experiments we performed have indicated that the surface streams running down the southern slopes of Miclau peak, which sink through the swallets in the area VartoapelcPonorul cu Pod, also belong to this system. The system discharges through the springs and the cave at lui Firea, of about 80 lis average flow rate. Fig. 10. Hydrogeological map of Stana de Vale area (geological data after BORDEA et al., 1984,arid ISTRATE, 1978): 1. Quaternary: a-alluvium; b-diluvium; 2. Laramian intrusives: a-microgranite rhyolite; b-tonalite; c:quartz-diorite; 3. Early laramian volcanics: a-ignhnbrite rhyolite formation; dacite; 4. Senonian (Gosau facies: congh>merates, siltstones, marlS,limestones); S. Anisian (dolomites); 6. Werfenian ( quartzitic sandstones, argilites); Perennialstream; 8,; Temporary stream; 9. Ge ologfcal limit between sedi mentary deposits; .10. Magmatites limit; 11, limit; 12. Faults; 13. Diffuse in stream bed; 14. Ponor; 15. Watershed; 16. Direction of groundwater flow; 17. Spring; 18. Outflow cave; 19. Chalet; 20. Direction of cross-section. ... Carte hydrogeologique de la zone karstique Stiina de Vale (les domzees geologiques d'apresBORDEA et al., 1984, et /STRATE, 1978): 1. Quaternaire; a-alluvions; b-diluviurn; 2. Roches intrusives laramiques: a-rhyolite micro granitique; b-tonalite; quartzifire; 3. Roches.volctmiques (debutdu,cycle larami(jue): a..:.jormation de rhyolite ignimbritique; b-dacite avec biotite et homblende; 4, Senonnien (jades de GOS(lll: conglomerats, siltites, mames, calcaires); 5. Anisien (dolomies); 6. Werfenien (gfes quartzitiques, argilites); 7. Ruisseau permanent; 8. Ruisseau temporaire; 9.limitegeologique entre les depots sedimentaires; IO.limite de magmatites; 11. limite de transgression; 12. failles; 13. pertes diffusesdans le thalweg de 14. ponor; 15. ligne departage des eaux; 16. direction d' ecoulement souterrain; 17,. source karstique; 18. grotte emergente; 19. hutte; 20. direction de section transversale. . . . . . . Numbers key: 1. Ariei spring; 2. Brebu ponor; 3. Rampeispring; 4; Piistraviiriei spring; 5. Meteorologist's spring; 6. Pavel spring; 7. Radu sprin.g;8.

PAGE 211

The hydrogeology of the karst areas of the Bihor Vlddeasa Mountains J Q11 11 II b I ;c c ..... f"! l.. ; 1 i ! 0 10. 11. .......... :2.----1 ---) 1 6 ------.. ,., 13. n 19. 10 f-----4 Yf M UloCPIAfU 1L15.Q 1 2Km __ l-_____ _J 1 NW 500 211 SE

PAGE 212

212 Two springs displaying .gas outflows occur in Cald graben: in the catchment basin of Alunul Mic stream, next to the major fracture that delimits the graben deposits with to the crystalline schists, emerges a 14.5 oc sub-thermal spring, that has a flow rate of about 5 lis and gas out flows; on the left bank of Cald stream, op posite to the junction with Paraul Sec, from the alluvia of the flood plain next to the stream channel emerges a spring of about 50 lis flow rate and 9.8 oc temperature, that displays vio lent gas outflows. IZVORUL KARST SYSTEM Izvorul karst system is located in the up per reaches of Biiita stream, in an extremely rough topography area, that had been also subject to intense tectonic dislocation and where carbonate deposits outcrop in the streambed and on the right side of the previously mentioned streamcourse, being nappeed in their northern, southern and east ern parts by the Permian deposits of the nappe. The outcrop area of the carbonate deposits does not have a permanent surface runoff: the multitude of streams running down the adjoining non karstic mountain slopes (Corlatul, Corliitelul, etc) sink diffusely, at their entrance on carbonate terrains. Before the excavation of the underground mining works, those groundwater flows ran to Izvorul spring, but currently a significant part of the flow is drained by the galleries of Biiita Molibden mine. The tracer tests have indicated that the radius of influence of those mining works has extended over the entire Biiita upper catchment basin, as well as in the upper reaches of Valea Seacii, a tributary of Galbena stream. The spring ensuingly underwent a drastic decline of its discharge, to such an extent that during draught periods it is not able to meet the drinking water supply demand of Nucet town, located down stream. Izvorul karst system has a strong inertia, a significant regulating capacity and important water reserves, 95% of the goundwater flow during re cession periods being provided by the base flow. STANA DE VALE KARST AREA In the area of Stana de Vale climateric resort, car bonate terrains outcrop over a small area (about 2.5 km2). The area (Fig. 10) includes several karst I. springs, supplied mainly by diffuse seepage of runoff originating in surrounding non karstic ter rains. Springs have relatively large flow rates, supported by the abundant rainfall, the most im portant in such terms being Plistrliviiriei spring, followed by Minunilor spring. Minunilor spring emerges from a small cave, ex cavated in a little outcrop of Anisian dolomites that protrude from beneath Senonian deposits and ignimbritic rhyolites, at the contact with a quartz rhyolites body penetrated along a fracture. Minunilor spring has a 15.6 lis average flow rate and very small annual fluctuations of the daily average flow rate (nv = 1.9), due to the almost exclusive contribution of the base flow component (99%) to the discharge recorded during recession periods. The strong memory effect ( 44 days, Fig. 7) suggests relatively important reserves. Minunilor spring karst system displays a typical "pass below filter" behaviour (MANGIN, 1982), completely suppressing the high frequency fluc tuations (Fig. 8). The long period of influence of the rainfall phenomenon (46.4 days) accounts for the very small flow rate fluctuations. Strong rain fall does not result in a major increase of the spring discharge, increasing in turn the stored re serves. The system is inertial, i.e. very capacitive and only slightly transmissive. Leakage from water accumulations existing in the Senonian deposits that transgressively cover the dolomites, and from ignimbritic rhyolites that cover part of the Senonian deposits (Fig. 1 0) pro vides the main supply to the karst aquifer. Subor dinately, the karst aquifer is also supplied by the surface stream course that sinks into the dolomite substratum through Brebu swallet and by the seep age that occcurs across the dolomites outcrop area. The Brebu's ponor was labelled with fluoresceine at 17.10 199 5. The tracer was detected in large amount in Pastriiviiriei spring with the maximum intensity after 90 hours. In the same labelling test, in Izvorul Minunilor Spring, the level of fluores cence was situated near that of the blank (0.2 ppb) which do not prove the arrival of the tracer at the source (observation time: 5 days). The outstanding constancy of the physico chemical parameters of the water of the spring and the continuous lack of any bacteriologic content, qualities that rank Minunilor spring among Roma nia's best still water sources, are the result of outstandingly favourable hydrogeological circum stances in terms of supply, flow and discharge of the karst system.

PAGE 213

The hydrogeology of the karst area s of the Bihor Vliideasa Mountains 213 ACKNOWLEDGEMENTS The author is using this opportunity to express his deep gratitude to the geologists Josefina Bordea, S. Bordea and Gh. Mantea for their kind assistance in providing the geological maps that had been used for the prepara tion of the geological background of the hydrogeologi-REFERENCES BLEAHU, M. (1957) Captarea carstidi i importanta ei pentru evolutia morfologicll. a regiunilor carstice. Probleme de geografie, V, pp. 55-99. BLEAHU, M DUMITRESCU, R., BORDEA, S., BORDEA, J. & MANTEA, G. (1980) Poiana Horea. Harta geologicll. a RSR, Scara 1:50.000, Ed. IGG, Bucureti. BLEAHU, M. & BORDEA, S (1981) MunJii Bihar Vli'ideasa. Ed. Sport-Turism, Bucureti, 496 p. BLEAHU, M ., LUPU, M., PATRULIUS, D., BORDEA, S., A. & PANIN, (1981) The structure of the Apuseni Mountains. Guide to ex cursion B3. XII Congres of Carpato-Balkan geological association, IGG, Bucharest, 107 p. BLEAHU, M., BORDEA, S ., BORDEA, J., MANTEA, G., POPESCU, A., MARINESCU, F., CIOFLICA, F. & A. (1985) Pietroasa Harta geologicll. a RSR, Scara 1:50 .000, Ed. IGG, B u cureti. BORDEA, S., BORDEA, J, MANTEA, G., MARINESCU. F ., M IONESCU, G & POPESCU, A. (1992) Meziad Harta geologici!. a Romaniei, Scara 1:50 000, Ed. IGG, Bucureti. BORDEA, S., BORDEA, J. STEFAN, A ., BLEAHU, M., MANTEA, G. & G. (1984) Harta litologica Stlna de Vale, Scara 1:25.000. Ed. IGG, Bucureti BORDEA, S., DUMITRESCU, R MANTEA, G A., BORDEA 1., BLEAHU, M & COSTEA, C. (1988) Biharia. Harta geologidi a RSR, sca ra 1:50.000 Ed. IGG, Bucureti. BRIJAN, P. (1978) Avenul Independenta din Hoanca Urz icarului. Buletin informativ CCSS, 2, pp. 20-23 BRIJAN, P. (1982) Fenomene endocarstice in zona valea Bulzului-Fanate. Carst, 2, pp. 54-60. BRIJAN, P (1987) Peterile din zona Cresuia. But. Speol. FRTA CCSS 11. COCEAN, P. (1988) Chei defilee In MunJii Apuse ni. Editura Academiei, Bucureti, 166 p DAMM, P & BOTEZ, M (1994) E xplorll.ri speologice In Bazinul lnchis Vll.rll.oaia. Ardealul Speologic, 4, pp. 5-8. DIACONU, C. (1971) Raurile Romaniei. IMH, Bu cureti, 500 p. cal map, as well as for their suggestions on the inter pretation of the geologic and structural framework of certain springs. I am also grateful to the hydrologists Paraschiva and Gh. Hotoleanu from INMH, for their permanent support in collecting and processing the hydrologic data DUMITRESCU, R., BLEAHU, M. & LUPU, M. (1977) Avram Iancu. Harta geologidi a RSR, Scara I :50.000, Ed. IGG, Bucureti E. & I. (1987) Natural and artificial tracers in the study of the hydrody namics of karst. Theor. Appl. Karst 3, pp. 31-107. HALAS!, G. & PONTA, GH. (1984) Subterranean drainage in the upper part of the Sigh itel valley (Monts Apuseni) Theor Appl. Karst. I, pp. 239-242. IANOVICI, V ., M., BLEAHU, M., PATRULIUS, D LUPU, M., DUMITRESCU, R .. & SAVU, H. (1976) Geologia Munfilor Apuseni. Ed. Academiei Bucureti 631 p. ISTRA TE, G (1978) Studiul petrografic al masivului Vlll.deasa (partea de vest). An. /GG, LIII, pp.177-298. KOMIVES, E. & NAGY, I. (1976) Carstul din bazinul Vll.ii Seci (Masivul Vlll.deasa) Bul.CSER, pp 101-128. MANGIN, A. (1974) Contribution a !'etude hydrody namique des aquiferes karstiques Ann. Speleol Paris,29,3, pp. 283-332;29,4, 495-601;30, 1, pp. 21-124. MANGIN, A. (1981a) Apports des a naly se s correlatoirc et spec tral e croisees dans la connaissance des systemes hydrologiques. C. R .. Acad Sc Paris, 293, II, pp. 1011-1014. MANGIN, A. (1981 b) Utilisation des analyses correla toire et spectrale dans r approche des systemes hydrlogiques. C. R Acad. Sc. Paris, 293, II, pp 401-404. MANGIN, A. (1982) Mise en evidence de l'originalite et de la diversite des aq uiferes karsti ques. Je Call. Hydro/. Pays calcaires, pp. 159 172 MANGIN, A (1984) Pour une meilleure connaissance des systemes hydrologiques a partir des anal ys es correlatoire et spectrale J ournal of Hydrology, v. 67, pp. 25 -43. MANTEA. G. (1985) Geological studies in the upper basin of the Someul Cald Valley and the V alea Seacll. Valley region (Bihor Vllideasa Mountains) An. IGG, 66, pp. 5-59. MANTEA, G H., STEFAN, A., RUSU, A. & DUMlTRESCU, R. (1987) RtlchiJele. Harta geologica a RSR, Scara 1:50 000, Ed. IGG, Bucureti

PAGE 214

214 I. (1994) Considerations regarding the karstic aquifers pollution in Romania. Especially Apuseni Mountains reference. Proceed. Inti. Symp. Impact of Industrial Activities on Groundwater", Constanfa, 23-26.05.94, pp. 420-429. 1., E., POP, I. & TANASE, T. (1991) Tracers experiments in the karst area of Bihar Mountains (Romania). Theor. Appl. Karst., 4, pp. 159-172. ONAC, P., B. (1995) Mare din Dealul Humpleu, Theor. Appl. Karst. Abstracts XII/th Symp., 26-30. 05.95, pp. 52-54. PADILLA, A., PULIDO-BOSCH, A. & MANGIN, A. (1994) Relative importance of baseflow and quickflow from hydrographs of karst springs. Ground Water, 32, 2, pp. 267-277. RUSU, T (1981) La Grotte des Ours de (Monts Apuseni). Rev. Roum Geol., Geoph. et Geogr., 25, 2, pp. 193-204. RUSU, T., RACOVITA. GH. & COMAN. D (1970) Contribution a !'etude du complexe karstique de Ann. Speleo., 25, 2, pp. 383-404. SCHMIDL, A., A (1863) Das Bihar-Gebirge an der Grenze von Ungarn und Siebenburgen. Verlag von Forster & Bartelmus, Vien, 442 p SERBAN, M., COMAN, D & VIEHMANN, I. (1957) Recherches speologiques dans les Monts Apuseni (Roumanie) Zvlastni otisk z ca.wpisu Ceskoslovensky Kras, 10, 1, pp. 11-25. SILVESTRU, E T. & FRATILA, G (1995) Preliminary data on the hydrogeology of karst I. terrains around the springs of Cald river (BiharVllideasa Mountains, Romania). Theor Appl. Karst 8, pp. 81-89. A. (1980) Studiul petrografic al de est a masivului eruptiv Vllideasa, An. /GG, LV, pp. 207-325. V L. (1976a) Carstul de Ia Casa de Piatrli (Muntii Bihar). Bul. CSER, 4, pp. 150-169. V L. (1976b) Privire de ansamblu asupra carstului din Muntii Bihorului. Nymphaea, pp. 21-58. V L., (1977-1978) Explorarea retelei subterane din Groapa de Ia Barsa (M.Bihor): Bul CSER, 5, pp. 170-211. VALENAS, L. (1980-1981) Cons iderati_i asupra in formati_ilor documentare despre carstul Munti_lor Apuseni in lucrarea "Das Bihar-Gebirge" (1863) de A. Schmid!. Nymphaea, VIII-IX, pp. 549-560. V ALENAS, L. (1984) Studiul complex a! carstului din zona Izvorul Ursului-Paraul Sec (Muntii Bihar) Crisia, XIV, pp. 559-580. VIEHMANN, I. (1966) Colorllrile cu fluoresceinli in hidrografiei carstului Hidrotehnic:a, Gospodarirea ape/or Meteorologia, 1111, 11/2, pp. 37-42,92-96. VIEHMANN, I., CRISTEA, E., SERBAN, M ., CUC, 0. & GHITEA, S. (1980) La morphologie du complexe karstique Cetlllile Ponorului" (Les Monts Apuseni, Roumanie) Trav /nst. Spe ol. "Emile Rac:ovitza", XIX, pp. 261-274.

PAGE 215

Theoretical and Applied Karstology, vol. 9/1996, pp. 215-218 Points de vue concernant le clivage de Ia calcite Gabriel DIACONU1 RESUME A l' aide des quelques images obtenues au microscope electronique en balayage, I' auteur ex prime un point de vue original concernant le clivage de Ia calcite. Mots cles: cristallographie, clivage, calcite. Some considerations concerning the calcite cleavage By commenting some images ohtained at ESM the author expresses an original point of view concerning the cleavage of the calcite. Key words: c1ystallography, cleavage, calcite. Il est a presumer qu' en dehors de Ia couleur et de !'eclat, le clivage est une des proprietes physiques les plus anciennes connues chez Jes mineraux. Apprecie comme une deformation irreversible, Je clivage est defini comme la qualite des mineraux de se detacher d' a pres des faces planes, toujours les memes, s'ils sont soumis a une force mecani que orientee dont la grandeur est suffisamment elevce pour depasser leur capacite de cohesion. Le clivage n'est pas une deformation au long d'unc discontinuite qui se trouve dans un mineral au hasard. II se produira seu!ement aux especes dont Ia symetrie d' agencement des a tomes en re seau est compatible avec un tel effet de reponse a un effort mccanique. En fonction de la facilite de detachement et de !'aspect de la face de plan resulte, le clivage se caracterise par deux notions: facilite de clivage et perfection. Conformement au loi de Bravais les plans de cli vage sont toujours des plans reticulaires de la plus grande densite du reseau. Entre l'annee 1915, quand STARK a donne une explication concernant !a realisation du clivage et 1' annee 1973, quand LIPPMANN pub lie un des plus completes etudes sur les carbonates des ro ches sedimentaires, a ce qu'il para1t il n'est pas intervenu rien essentiel. C'est le motif pour lequel nous preferons de citer directement cet auteur, ses explications etant edificatrices pour le cas en discussion-le clivage de la calcite: ... The realistic results, obtained from the electro static structure model, may be taken as evidence for the generally accepted view that the structure of calcite is indeed held together by the electro static forces between the component ions cl+ and co/-. This view is further substantiated by the perfect cleavage of calcite along the faces the morpho logical rhombohedron. This property may be understood according to STARK (1915) by consid ering that the atomic planes parallel to the cleavage faces are occupied by oppositely charged ions.... Adjacent atomic planes parallel to the cleavage rhomb are held together by juxtaposition I Institutul de Speologie "Emil Racovitli", str. Frumoasli n, R-78114 Bucureti, Romania.

PAGE 216

216 G. Diaconu c ------6 0 1 2 3 4 s .. c:==;;:' =""":!==.J A o-ca o-C03 ;; .. Fig. 1. La projection des pllUis te!i' pillS leur prolongement sur les faces rholttboedre divage{Ct(1c A projections sont conteriues Ia projectign tipliee a un prisnie hexagonal complet .. Spatial projection of the most dense planes de, tei-minet!ld tile c(Jo,:dinpiion ((:aOJ (D);. wi(ht}zef f pro longations on the "p" faces of the morphologicalcleav.age rhotnbohedro n . c{t'e : spatial correlation projection between the hexagolral ceu(6CaClN a tlie s}zarp rhombohedric cell (2CaC0 ,1 ) (H) a11d the cleavage morphoifJgic rhombohedron ( C) (wii!z .six Ca ele,me,11ts1 IHitn"-: bered fronz 1 to 6, .presurna.biy Outsi4e o/. th __ --. <:::,;: :. : '; '

PAGE 217

P))oto l, surface chJ.P:face of a C.aleite.mic_ro cry'$tdfi()(.l(JOQ)::;;;:_ > llhoto 3. Detail du Photo 1. ( xSOOO). ... ,_ :, .o 217

PAGE 218

218 rl oppositely charged ions. By way of some severe deformation, two of these planes are thought to be displaced with respect to each other to such an extent that equally charged ions come into juxta position. The resulting electrostatic repulsion then destroys the coherence of the parallel. to these planes .... (LIPPMANN, 1973). Done, nous pouvons considerer que, de point de vue cristallochimique, le reseau de Ia calcite est ... anisodesmique, etant caracterise de liaisons ioniques heteropolaire etabli entre les ions simples de calcium {Ca++) et les ions complexes de carbon ates Les derniers (COn, etant constitues comine des molecules mesomeriques planes (hybrides de resonance), agissant unitaire en ayant deux electrons disponibles compenses dans reseau, par liaisons ioniques, des charges electriques posi tives du calcium (Ca++). Par Ie dessin de Fig. 1, nous avons essaye une visualisation du parallelisme d'entre celles trois plans de Ia plus grande densite determines de l'octaedre de coordination (C06 ) et les faces "p" du rhomboedre morphologique de clivage, les deux clements (l'octaedre et le rhomboedre) etant en cadres dans Ia correlation graphique d' entre Ia symetrie rhomboedrique et celle hexagonal. Mais, ces relations de symetrie, elles ne font pas autre chose que confirmer (d'ailleurs ir-BIBLIOGRAPHIE DIACONU, G. (1990) Cave. Minerulogical and genetic study of carbonates and clays. Misc. Speol. Rom., 2, 135 p. G. Diaconu reprochablement) les acceptations theori ques.Pratiquement, un cristal de calcite qui est soumis a un effort mecanique au-dela de sa limite de cohesion, presentera toujours un premier plan de clivage, facile et parfait, parallele a l'une des faces du rhomboedre morphologique de clivage ( p }. Les autres deux directions des plans de clivage seront realisees, aussi faciles et parfaits, mais dans un certain etat de subordination par rapport au premier plan de eli vage. La justification de cette subordination a ete sur prise, sans doute, a l'aide d'une image sur un mi crocristal de calcite, agrandie successivement au microscope electronique en balayage (Photo 1, 2, 3 et 4). Regardant ces photographies on peut distinguer que le plan de clivage qui sera le plus facile cor respond a celui qui se trouve dans une position perpendiculaire a Ia direction de croissance de Ia face p, done parallele a Ia Stratification. Les autres deux directions de clivage (correspon dantes aux autres faces p) seront depourvues d'une facilite de detachement similaire etant donne Ia maniere d'imbrication des elements constitutifs du crista] pour Ia zone de croissance donnee. LIPPMANN, F. (1973) Sedimentary carbof)ate mi nerals. Springer Verlag, Berlin Heidelberg New York, 228 p.

PAGE 219

Theoretical and Applied:Karstology, val. 911996, pp. 219-226 Intensity of karst processes as a function of the carbonate formations in the north Re,ita -Moldova Noua Synclinorium Adrian IURKIEWICZ1 Bogdan BADESCU2 & Elisabeta MARINICA3 ABSTRACT The exploration of the karst cavities in the area led to the identification of a significant amount of "karst entrances", consisting mainly of caves and potholes, swallets and springs. The study of the characteristics of the karst cavities, as well as of their distribution with respect to the different types of carbonate deposits, has resulted in a qualitative evaluation of the karstification processes that have occurred in this specific area. Key words: karst entrances, karstification, statistical analysis. lntensite des processus de karstification en function du substrat carbonate dans Ia partie nord du Synclinal Re$i[a Moldova Noua Les recherches speleoloKiques qui se sont deroulees sur plus d'une trentaine d annees dans La partie nord des monts de Anina (zone de ant mis en evidence un nomhre important de phenomenes karstiques (grottes, avens, ponors, sources, etc) enregistres comme des "entrees karstiques". L'etude statistique de certai nes caructeristiques des ces entrees, ainsi que leur repartition dans les d(fferents types des deptm carbonates ont pennis une evaluation qualitative de La karstification de cette zone. Mots cles: entrees karstiques, karstification, etude statistique The present paper intends to delineate the different ways in which the karst processes have operated within the carbonate deposits in the Moldova Noua area and to outline a statistically based hierarchy of their intensity. The ResitaAnina zone occupies the northern part of the major sedimentation area of the Getic Unit, the so called Noua Synclinorium" which extends over 800 km2 The Getic Unit includes the crystalline basement, which occupies the eastern zone, pierced by granite intrusions and covered by a sedimentary stack belonging to Permian -Lower Cretaceous. (for more detailed information about the geological framework of the area see IURKIEWICZ et al, this issue) From the entire unit, only the Jurassic-Lower Cretaceous cycle (out of which the lower series of the Liassic should be also excluded) exclusively includes limestone deposits. From a speleological point of view, the northern part of Anina Mountains is the most thoroughly investigated of all the structure, ensuing to the widespread speleologic investigations performed by the Speleological Association "Exploratorii" from (709 cavities recorded on 31 December 1995). 1 S.C. Prospectiuni S.A., str. Caransebe nr.l, Sect. 1, Bucureti, Romania. 2 Asociatia Speologicli Exploratorii Reita. str. Lalelelornr. 2, Reita. R.:.01700, Romania 3 R.A. "Apele Romane" Filiala str: ClUri.inelol', nr. 9, Reita, R-1700, Romania

PAGE 220

220 The impact of the overall underground karst processes is mostly visible between 200 and 600. m altitude. Major caves arc Buhui (6547 m long) and Comarnlc (6203 m long) both through trips and the pothole in Poiana Gropii (-236 rri deep, 1029 m long). Table 3 further indicates main karst features. The adopted approach is to consider the Lion of the "karst entrances" with respect to fcrent criteria (stratigraphic, lithologic, hydro. logic, tectonic). In this way the karst entrances (GORAN 1982, POVARA et al., 1990) are con" -.. .. Age . .. Name .. .: ' Aalenian-Lower Callovian V alea Morii laye.-s (aa+cl ) Middle Callovian (c l ,) Gurripina lms. Upper Callovian-Lower Oxfordian Tamasa marls (cl1+ox ) Upper Oxfordian-Lower KimmeValea Aninei lms. ridgian (ox,+km ) Upper Kimmeridgian-Lower Brlidet lms. Tithonic (km,+th ) Upper Tithunic-Beriassian Marila Ims. (lh,+bc) Valanginian (v) Crivina marls Hautl!rvian (h) Lower Plopa lms : BaiTI.!mian-Lower Aptian (br) Upper Plopa lms Uppl!r Ap tian Valea Ims. Rock geochemistry. Relying on the analysis of 20 rock samples from Anina area, LAZAR ( 1963) shows that the Upper Plopa limestones have the largest carbonate contents (94.08-94.97% calcite, 4.92-3 .86% dolomite, 0.64-0.37% impurities), followed by Valea Aninei limestones, Marila limestones, Gumpina limestones, V. limestones. and Bradet limestones (Table 2, clear l ines). A. Jurkiewicz et al. ventionatly defined as entrances of caves and potholes, swallets and springs whiCh are situated within the area (which corresponds to catchment basins to 2248 from the Karst Survey Register-GORAN, 1982). Data supplied by this statistical analysis may result in some prediCtive information about karstification and endokaist features from this region . . . . . The stratigraphic criterion is the most important for delimiting carbonate deposits in the studied area, according to the following .. separations CNASTASEANU, 1964): . . ... -, .. x;'',' }.:; Thick. (m) Characterisation .. 100 marly lins. with ellipsoidalconcretions lms. with nodular silica concretions . 100 limestones and 200-300 biomicrite with silica tnterbeddings 100 biomicrite nodular lms. with silica .. concretions 200 micrite, lms. & ellipsoidal silica nod-ules 100 marls and marly lms. mady lms and lms. 350-400 massive reef lms. 400-500 organogenous lim es tones As a result of performing 80 chemical analyses and 37 spectrographic analyses (BUCUR & POMARJANSCHI 1986) of the carbonate rocks collected from 3 drill holes completed by IPEG south of Doman and from natural and artifiCial galleries, the percentile contents of <;alcite, dolo mite, clay, dendrites and silica was established (Table 2, shaded lines).

PAGE 221

Intensity of karst processes Age (name). CALCITE(%) 9i.z9/94;ss: '-86.02-93.1 94.08 94.97 : 90A6 91.35 -. - .. 88.2 lt can be noliced that while the calcite content is roughly similar in all formations where from samples have been taken and analysed, the dolomite rises the carbonates content of the Valea Aninei and Upper Plopa limestones to almost 100%, hence setting a clear difference with respect to the Bradet and Valea limestones, the carbonates content of which does not exceed 95%, because of impurities occurrences. The spectrographic analyses show small quantiLies of micro-elements i.e. 30-1000 ppm Sr, 3-400 ppm Ba, 3-2900 ppm Mn, 3-1000 ppm Ti, 0-25 ppm Ni, 0-30 ppm Cr, 100 ppm Cu, and exceptionally 260 ppm Pb. 100 C+P 50 16;7 100 C +P 100 C+P 91,11 50 ::._: {;; 0 .. ... P. AV. 1,86 2?>7 o.LJ._J:::::;r.::::::J U D 0 L F S A .. DOLOMITE(%) 4.60-16.06 8.62 . 3.86-4.92 Ll8 -0.22 3.94 IMPURITIES (%) 2:()8 ---1.01 1.84 -:);[;\. . 8.71 . t.9t .--.--: 0.34 7.32 5.03 221 The vanatton range is large and approximately the same for all categories of rocks, except for the Valea Minisului limestones, which sometimes contain large quantities of Ti, Ni, Cr, V and minor clay elements. Data acquisition. Within the considered area. II speleological basins included in 5 stream catchment areas have been separated (GORAN, 1 982) The investigations resulted in the identification of 1001 "karst entrances", 709 of which are caves, the characteristics of which have been plotted in the area speleogram, 135 are springs and 59 arc swallets. The percentages that resulted from the comparative analysis of the different sections that 100 c .. ;::> 50 73,37 n I I 15,65 0 .L-...l.-.._i_.....J Pl. Ve. Va. compose the speleogram are in a direct relationship with the surfaces of the stratigraphically delimited carbonate rocks (Table 1 ). as follows: Primary statistic analyses and discussions. 83.351/r of the cavities identified within the studied area arc caves and 16.78% arc potholes (Fig. 1, a). The cumulated length of the caYcs is 56.77 km and their cumulated depth is 5. 79km. Among the horizontal Fig.1: The freqljency of'ciitj'erent types of caves: a; Cilves (C) and (f); b; shape ofnetwork: uDlque gallery (U), dendritic {D); labyrinthlc {L), orthogonal (0); c; bydrologlc character: fossil (F); active(A); d. localisation: cavities, the majority (50.08%) display a dendritic pattern, while the minority ( l. 86o/c) is formed by those that arc orthogonally developed (Fig. I b). plateau (l.>l),_ sJope (Vc);vaUey (Va). "' . -. .. , La de dijjh"enfs tjpes des c;Qvitis: 4<. grottes ( C) etavens (P); kz phologie de-resea'u;: galerie tmique ( U),. (D), ltibirintigue orthoglf11af#. ((J); c. le (F),"subfossile (SF), actije (A); i : .,.

PAGE 222

222 From the hydrologic poiRt of view, the majority (91.11%) are fossil cavities, 6.20% are temporary stream cavities and only 2.67% are perennial stream cavities (Fig. 1, c), while among the tem porary stream cavities, about 1.50 % include short sections of stream passages This observation might suggest a non-functional contemporary karst. which has been abandoned perched above the drainage levels It has to be mentioned (in the case of the large cavities -Table 3) that the large temporary stream cavities are dominant (45%), followed by the perennial stream cavities (35%) . : ... 100 C+P 100 Swaltets, . 57,10 50 50 30,52 By an accurate location of the cavities on the maps, their percentile distribution with respect to the geologic formations has been identified It has resu lted that most cavities are developed in Upper Plopa limestones (br+ap, 57.10%) and in Yalca Aninei lime s tones (ox2+km,, 30;52%) (Fig. 2, a). The cavities identified in any other. formation amount to 3 4.5%, except for the Lower Plopa (h) and Bradet (km2+th1 ) limestones. In th e case of the latter formations the very low percentages, o: 14% and even 0%, can be ascribed, at least partially, to their restricted outcrop areas It results that the two formations wh ere 87.62% of the investigated karst cavities occur are the most propitious for karst development, which may provide a guideline for further speleologic prospecting. A Jur/ciewici eta/. and the, fossil cavities {20% ), an distri!Ju: .. tlon which in part does not support the previous supposition . in fact the statistic results tonfirin iri both cases an advanced degree of karst deyelop ment, yet also an advanced orgarusatiori of the cavities iil active karst systems. . The position of the networks within the physiog raphic unit is the following: 79.97% ori niountairi slopes, 15.66% on the plateaus and 4.37%' on beds, the latter being perenniaL at tempo-ra:i-y karstic stream piracy cavities (Fig : l, d). _. .. . . . 100 Springs :. : .. ' .... 35,59 If besides cavities, sinking points and emergences are considered, the re..: suiting image is virtually the same: Upper .limestones 53.48%, Valea Aninei limestones 30.34%, with a slight increase for the lime..: stones, namely 5.53%. The distribution with i :espect to geologic formations ofthe swalletS and springs considered separately (Fig. 2, b and re spectively c) also falls within the previously indicated rimges. . Since the outcrop areas percentages ofe
PAGE 223

Intensity ofkarst processes (Fig. 4); then another comparison was made by. means oftheir ratios As a function of the latter parameter (derisity of entrances D.E., expressed as number of entrances/ km2 ) 3 main groups have resullcd (Fig. 5), namely: 100 Surface so 40,80 r--27,60 .---DD. 15,0 ,--14 12 10 8 6 4 2 0 tufa 223 1 6,80 ' II 2,56 III r:-:::::-1,19 h-' I I l'h Fig, . J'he of the d.ifferent .. te outcrop area$ . -. . . .. : . Les pourcerltages df!s diffeiel!ts affkuretiJent's Fig. 5. The karst entrance density (DD) in different car b(mate outc.-op areas ; .:, .'' ::.. . .. t..a dei,s i t{d'entreesk arstiques (DD) sur les bonates. . .. '" . .. types d'affle.'Ureinents carbo11ate .. Krn2 6 0 7,14 500 50 D DO 400 40 2 I / DC DC 300 30 G' nr. LJ 200 20 3 100 10 D 4 0 0 o,... -? 0 + 0-+: <',< <'_.;: QQ tiv .. "<> '1 "'"1:: 2 '<'..-: 1< :.0 0 0+. ? 7 ? 7 A T he kars(entrance the carbonate outcrop areas: a. the karst en trance den s it y; b. the nuinl)er ofthe karst entrnces'; c. the sui-face oftl:te outcrop carbon,a t e formations; d; calcarolismarls. . I.e sw (es.(J_ifjirellis. types d'aj./leur ements carbonates: adade..nsite kar sti ques; lJ; [e. d 'entrees karstique.s; c. w surf des affl.eurisements carbolaatees; d. mames calcaires.

PAGE 224

224 6,80 D.E. Mini limestones, Upper Plopa limestones, Valea Aninei It en compasses 64,45% of the considered statistic units (entrances); according to the resulting value, V. Mini limestones are as propitious to karst development as the other t'WO forma tions previously discussed. 2,52 D.E. Marila limestone, Gumpina lime stones, Lower Plopa limestones; they amount to 24,26%. a 100 C +P 82,79 so 0 sup.merl.mar. b 100 C +P 60,93 Fig. 6. Frequency of the horizontal; b. vertical . .. ... . .. . < La frequence des classes tes: a. lwrizonttiks; b; verticales. ,. ,:. .: mm/mil. k.mlkmz 100 1 80 08 60 06 OK 40 Ott 20 02 Fig. 7. Comparison between gloi;Jal karsLdenuda tion velocity and .. . . ; . ... Comparaison eiJire la denudation karstiqilf! global!J etle UND. . . . .. A. et al. 1,19 D.E. Valea Morii layers, lime stones, Tlimaa .. marls. They amount to 11,27%. Although the density of the karst entrances in travertine (15/ km2 ) exceeds that in limestones, their importance from a speleological point of view is minor, due to the improbability of a large spatial development (small rock volumes). Out of the entire amount of caves arid potholes, those with different degrees of horizontal devel opment are distributed as follows (Fig. 6, a): small caves (between 5-50 m) 82,79%; medium developed networks (between 50-500 m) 14,80%; caves (over 500 m)-2,39%. An analogous analysis, considering this time cavities with different degrees of vertical devel opment indicates (Fig. 6, b): virtually horizontal nernorks (between 0-5 m) 60,93%; networks .small vertical extent (between 5-50m) 36,81 %; netwm"ks of medium vertical extent (between 50I 00 rn) -1,55%; networks rllar!{e 1ertical extent (between 100-500 m) -(0,70%); The previously mentioned percentages seem to indicate a relatively restricted penetration, both on the horizontal and on the vertical, suggesting a relatively young karst. Yet if the weight of the cumulated lengths for each of the three (horizontal) groups of caves with respect to the total cumulated length is considered instead, it can be noticed that the distribution is reversed (18,68%, 28,17% and respectively 53,13%), prevalently organised in only a few karst systems ofrelatively large development. Comparatively analysing the karst denudation velocity mm/mil. (SENCU, 1982) and the under ground network density (UNO) = the ratio be tween total length of horizontal projection of so Fig. 8:The of the tectonic influence to tlui . . : .' deJ!eloppe!luntt tie :

PAGE 225

Intensity of karst processes the passages and the total surface of the lithological unit km!kin2 (Fig. 7) in the stream catchment areas, it can be observed that for catchment areas with a similar denudation veloc ity value (Doman, Rau Mare, Nermet and Garlite) different values of the underground net work density have resulted. This result leads to the hypothesis of a stronger underground denuda tion in the Rau Mare catchment area as compared to the Nermet valley catchment area. In order to outline how the overall tectonic setting of the area has influenced the karst processes, a 225 frequency histogram of the mairy bearings of the passages of 20 main cavities (Table 3), distributed across the entire study area has been prepared. The resulting image (a prevalence of the N30W and N30E bearings, Fig. 8) confirms the strong influence that the structural elements (anticline and syncline structures and/or NNE-SSW striking longitudinal faults, as well as their associated faults) had in imposing the development direc tions of the passages in the case of the considered cavities. Table 3. Malnkatst . from >-SoQl)l;Deptb >lOOm),.. . Prine. ip.aux phenombles dans lazone Re#d-Ailina (LOngeur >SOOm; Profondeur >lOOm) -_-' .. . -. .. . . l.CaveNal1le . BASIN. 2236 P. cu Apa din Chcile 910 +5 320 30 1 A ox 2 +km1 s D Vidra 521 -25 375 1,5 1 SF br v D BUHUI LUCA BASIN 2237 Buhui 6547 -50 650 10 5 A br+th 2 v D P. din Valea Seaca 537 16 640 60 1 SF br s D (the left slope) 2238 P. de dupa Cirsa 1300 54 243 35 1 A lh2 s D P. 2Mai 700 16 483 108 1 F br s D GORGES (the right slope) 2240 Popov111 1121 15 417 37 1 SF br s D Tolosu 1847 87 350 50 1 SF br v D P. de sub Cetale 2 576 14 252 7 1 SF br s D Liliecilor 640 34 232 15 1 SF br s D Rasullatoarei 740 35 471 141 1 F br s D Ramonci 534 27 515 140 1 F br s L DOMAN BASIN 2242 lzvor 530 -10 380 10 1 A ap2 s 0 Stirnic 700 -11 370 11 1 F ap2 s L TOPLITA COMARNIC BASIN 2243 Comarnic 6203 -100 512 44 9 A br. v D OX2 +km1 Exploralorii 5172 48 387 I 0 1 A OX2 +km1 v D RAU MARE BASIN 2246 A v. din Poi ana Gropii 1050 -236 750 0 1 A ox 2 +km1 v v A v. din Cioaca Marc 759 -137 883 100 1 SF ox 2 +km1 s v P. Gaura Turcului 1603 77 454 35 I SF OX2+km1 s D Av din Dolina 217 -129 818 1 F OX2 +km1 p v Av. Bizonului 444 -101 549 6 1 SF br s v .. p Poslllon: V valley, S slope (valley s1de), P-plateau, Hydro type H.T .. A acuve cave, SF subfossii. F tossii. Network shape -N.S.: D-dendritic, L-labyrinthic; V-vertical; 0 orthogonal.

PAGE 226

226 CONCLUSIONS The underground investigation of karst cavities resulted in the identification of a hierarchy in the liability to karst development of the carbonate deposits in the northern. part of the Moldova synclinorium. In this respect, out of the ten considered formations the most favour able to karst development appear to be the Upper Plopa limestones (50%), to which Valea Aninei limestones (30%) should be added. REFERENCES BUCUR, I. & POMARJANSCHI, D (1986) Microfa cies, diagenesis and geochemistry of the car bonate rocks in the surroundings of Reita. D.S.lnst Geol.Geofiz., vol.70-71, 4, pp. 63-77. GORAN, C. (1982) Catal ogul sistematic : al din R omania. Ed.CNEFS, Bucureti, 496 p IURKIEWICZ, A DRAGOMIR, G., ROTARU, A & BADESCU, B (1997) Karst systems in Nera zo ne (Banat Mountain s) (this issue). LAZAR, L. ( 1963) Cercetari geologice-tehnice asupra carstului din regiunea Anina (Banal), Com. A. Jurkiewicz et al. This result is in accordance with the extent of the outcrop areas of those formations Moreover, the karst entrances frequency weighted against the extent of the outcrop areas ranges in the same group of highly prone to karst development for mations Valea limestones too. The other formations are less favourable to karst de velopment, or they even prevent the development of penetrable underground cavities. Geol., Studii. Tehnice Economice, Ser. E, 6, pp. 217-240. NASTASEANU, S. (1964) Prezentarea o1ogice a zonei 1:100.000 An. Com. Geol., pp. 291-342. hartii ge Noua, sc XXXIII, POVARA, I., GORAN, C. & GUTT, W (1990) Sp eo l ogie, Ghid practic. Ed Sport-Turism Bu236 p. SENCU, V (1982) Denudarea carstici1 in muntii Aninei, But. So c G eogr., VI,

PAGE 227

INFORMATION FOR CON T R I BUTORS Papers may be submitted for publication as Articles (up to 15 pages) or as Short Notes (up to 5 pages). Larger articles, i.e. Synthesis or Reviews may also be published but the subject and contents of these should be discussed first with the Editorial Board. In any case, priority will be given to the papers presented at the previous annual Symposium of Theoretical and Applied Karstology. Manuscripts must be submitted in English or French, if possible on a floppy disk (IBM-compatible format) accompanied by two paper-copies. Typed manuscripts will be double-space typed on one side of the paper, with no more than 2000 characters/page (31 lines/page; 67 characters/line) Papers should be headed by the title and the author(s) full name(s) The complete address of the author(s) will be on a separate sheet. If more than one author, please underline the name and address of the person to whom the correspondence and proof should be sent. An abstract in paper's language should be placed directly after the title and author(s)' name(s), followed by the key-words. These will be followed by the translation of the titlf., abstract and key-words in the other language of the volume (English if the paper 's language is French and French if the paper's language is English). Acknowledgements should be placed at the end of the text, after the heading Acknowledgements, before the references. Mathematical equations should be numbered in consecutive order throughout the text, using Arabic numbers. The footnotes are to be consecutively numbered with exponential numbers and double-space typed on a separate sheet. References cited in the text (and only these) should be listed, in alphabetical order, on (a) separate sheet(s), after the heading REFERENCES/BIBLIOGRAPHIE taking for example this issue of TAK References should be cited in the text in parentheses, in capital letters, by the name(s) of author(s) followed by the ye a r of publication e.g. "(AHRENS, 1962)" except when the author's name is part of the sentence, e g "AHRENS (1962) has shown that . When reference is made more than once to same author and year a, b, c etc should be a dded to date in the text and referen ce list. If there are more than two authors it is in order to put "et al." behind the first name, e.g. "(KLEIN et al., 1988)" Tables are to be numbered in Arabic numerals in order of their appearance within the text, e.g. "Table I". Each table must be submitted on a separate sheet(s). If the table has a title and/or a legend these must be translated in both volume's languages Figures (maps or line drawings in black Indian ink on transparent paper) should be prepared in anticipation of reduction; small sym bol s or fine details may reproduce poorly. The authors are requested to limit the size of their illust rati ons as much as possible. The original figures must not exceed 31 x 20 em. The number of folded figures being limited, only ONE figure exceeding this size will be as an exception, accepted. Plates a nd text-figures should be numbered consecutively, with Arabic numerals, in order of appearance. The top of the figure should be indicated whenever exists any possibility of ambiguity. Please indicate in the manuscript where the caption-figure or group of figures is to be inserted The captions and explanations of the figures will be doubled-spaced typed on a separate sheet, in both volume's langu ages Authors bear sole responsibility for the content of their contribution The Editorial Board reserves the right to refuse any manuscript submitted if the subject is out of th e volume's topics and to make suggestions a nd modifications before publication The autho rs are requested to pay special attention l o the accuretcy of the figures or photogr aphs The manuscripts of the refused papers will NOT be returned. Papers accepted by the Editorial Board will become the propert y of the Publis her No article or a ny part thereof may be reproduced in wh a t e ver form without the written permission of the P ublisher 1 A total of 30 reprints of each paper will be supplied to the authors free of charge.

PAGE 228

TH'i:ORETICAL AND APPLIED KARSTOLOGY volume 9, 1996 Legend or the numbers on the map In . dtllldc uf tht: potnto; m nru:n; as I Hydrographtc of Negru river 1 0'11-iltktC.;we\:ns m);u'od ,pnng(370 m} 2 (II XXI m) 3 Spnng ot Re.t ( 4 2 5 m) 4 C>cm bnJge {7 1()m) j "LJtretll\l(!J"C 0J...1Ulult.pnng (515 m) 36 Sec1illlr.l pothole (930 m) and of Scditur, 'tream (925 m) 37 P othole nt Dn s u l Mum.:c:lulu1 (1100 m) I 38 l_.,:,sse, n l SOOolul Pictrc1 stream (550 m) 39 ot Sodulul TorrqU!or stream (550 m) 40 of Cf.lcJUne strea m (880 rn) 41 nf Prcluca NC!jUllll (81 5 m) 42 U.>!i'.!'; 1>l &>dcllul Ghct<'VUIUI stream (740 m) I 43 Civei500m)es u strewn (850 m) 4t! Lo."o;cs ol (750 m) 49 Elena poth,Jie (!\45 m) 50 (7C() m) 51 ot llilatul stream (1040 m) 52 o f0lfhi\elul stream ( 1150 m) 53 n t stream (II SO m) 54 Adnan potho!e(Valea Rea, 1275 m) 55 Pnn()r of ( 129:1 m) and V5 (Fa!& MunceluluJ) p:Jth()1e (1400 m) 56 Iklga (675 m) "i7 CJiclu (91fl m) Sli. HlllbtKJ spnng (9(XJ m ) 59 hwtana spnng(ln5 m ) andcave (Y-\5 m1 W (lf Plmtlllll stream (lOCi) m) f. I Ponor ol Cu\tlur stu:am {l260 m) ul {1235 m) Pnnor nl ,\rsum stream ( 1245 m) Ponr." nt Tr.ln.;hejll >lrethole ( 121Xl m) 7 1 AvenuiNl'gntpotiJulet1220m) 72 of! Set: ( 1205 ITI) 73 Ponors of !el.Cfe ( 1260 m} 74 of B"!'Sil Cohitnulm stream ( 1090 m) 75 Ccta[1le Porortor(ll25 m) 7M Groapa de Ia Barsa ( 11 IX)...! 15() m I 7!) Gbejarul de !11 Bar.;a cave ( 1125 m ) 1 80 GtveofZAPixhc(IIOOm) 81 Pq1era Neugri\ cav e { 1100 m) 82 Ghetarul Ftx.:u l Vm (1170 m) 83. G.dbene1 sprmg (815 m) 84 Fauleasa spnng (570 m) r 85 of v ganulUJ stream (8211 m l 116 of V alea Seacli stream { 1100 m) 11!7 88 Losseo: of Cnj:anut sm::am f975 m l I 89 Pothole ol H!)'I Or. tJ.t c ;.re:.tnl r I i 70 m) 1!17 ll!!' G hC\te1u1 D. bre,tl nr {7]ll m) 112 V.arm t C >lC\1'1 ( Feredeu 757 ml II:\ M,msprm b,7r:olm) 114 rothok (KIl rn and fu<. Jut !oand ti!IO ml 115 L>s' c.' ol ,tcrarn (7 45 m) 110 llhmuiMKspnngC>\.;mJ I P Mar, m) Pom>r nt (}!7) !': II Y lI[UIUlt:aVC(IU!ilm) IJY Pqtcraf'cpuuve(l125m) 140 Spflng ol Ca1d--Pirilul Sa: cunllueocc ( !C!95 m} 141 142 W,lffi1 spnng of Alunul Mtt: stream (ICXiS m) 143 P1>nur uf Ponor ,tream( 1130 m) 144 Diada7.A cave (1235 m) 145 Lltw pulhnle ( 1140 m) 146 Pment\11 potho le ( 1395 m) 147 Punur J\11<11\a Virtopulut {1322 m) 14S l4Y Jm Firca { ltl70 m ) ISO Hump1eu ( 1175 m) Hydrographic basin ol Rcpede river 151 L.espcz1 spnng(950m) 152{ele waterfall (Hnl m) 153. ( 1 HXJ m} 154 t:U Sao;e lnlliit cave(ll40 m) !55 Ptll'lnr and ofTriU (I ISO m} 1!56 V!trfur.r!jul and O!Ve (1175 m) 157 lzvnrul Mmumlnr m) 1St! BrebU ponor(I21KJm) ISIJ Rampe1 spnng (1145 m) 160. Pi'itdiviittCI spnng ( 11(.10 m) 161. P.dvtl { Hltl5 m) 162. Radu "Pf!Og {10115 m) 163 Murpi!jul spnng (9)() m) 164 Anc spnns 0325 m) I I 5 ' ; --/ / / 3 l I 9 I 7 I otn I I "-1 ---------1--__ ...J_---------r ---' 2 : 8 / / ' I I I 1 --------------___ ,.. __ ----------4 + + Geological data : Cascade ladohna A A A + Dt.Pia1ului ' + + 1103 + + / A / TatOroaia V 23 1289,3 -----. P. p / .. '/ :. / ... . I . -. o"' .: Plai 'I I. Ble ahu M. Dmmtrescu R .. Bordea, S., Bordea J osetina. Mantea Gh (1980) sc 1 50.000 . ' ,. I 1: I II I d ,I I, vf Munceilor I 1411 I sn I I + ,vartop :--If' 1294.8 2 Bleahu M Bordea S, B o rdea J osefina, Mantea Gh, Popescu Agapia, M arinescu FL, Ciuflidi G $tefan A (1985) sc. I 50 000 3 Bord ea S . Bordea Josefina, A .. Bleahu M .. Man tea Gh . Gh. ( 1984 ) sc. I :25.000. manuscnpt 4 Bordea S Dumotrescu R. Mantea Gh A Bordea J osetina. Bleahu M. Cos tea C (1988). sc.l :50 000 5. Bordea S Bordea Josefioa, Mantea Gh., Marinescu Fl M Ion escu G, P opescu A ( 1992), sc 1:50 000 6 Dum!lrescu R Bleahu M . Lupu M. (1970), sc. 1 : 50.000 7 Istrat e G (1978). sc. 1.50 000 8 Man tea Gh ( !986 ), sc 1 000 9. Mantea Gh . A Rusu A i))ummescu R (1987 ) sc. 1:50.000 10 A (1980). sc. 1 :50.000 ----L A A A + + + + ' p A A I ( ' ........ . A \ 1426,3 \ _,. \ -. i-p ,, -. . / --/. w 'i ---.: . /o : _,. .. ._:L/':: Y.> I ' I I \ \ i I w r -- I \ -,, -. \. ' ' Ariel?eni I ' ' \/... w -.,/-. "' umpelu -132l5 .... ----..___ "' "' ... w \ Ticta'u N w 6 i -:-'"1-'-;::s-'--t:Y::: "' IANCU ORASEANU > HIDROGEOLOGICAL MAP OF THE KARST AREAS \..1 -FROM THE BIHOR VLADEASA MOUNTAINS "' StanCIU/ .., .... Doda Pilii .... --------' / 1 /-W r : -;---= ' i . / . : .. ;--/ ..... ; ' -w .. \ 0 1 2 3km LEGEND HYDROGEOLOGICAL CHARACTERISTICS OF THE MAIN FORMATIONS EJ' . [[][]]]] Carbonate MesozOic senes (limestones, dolomnes), highly fractured and karsttfied, charactensed by very htgh etfecuv e mfil tra u o n and mtenstve groundwater flow Numerous karst systems with various SIZe and prevelant binary type Spnng flow rate up to 550 U s Important water resources in large karst systems. Laramian magmarites (banatltes. a-mtrustves, b-volcanics) and metamorphttes (c) wtth permeability of fissures wtth dtscontmu ous dtstnbutton and mtensity The weathenng zone is well developed and provide a continuous and Important supply of nVefs (memory effect is 55-120 days for 0 2) and of bmary karst systems Prevalent detrittc Penn o-Mesozotc deposits (sandstones and conglomerates with argillaceous shales and rhy o lite s) w1th dtfferent penneabthty. The groundwater flow lS mostly confined to the fissured areas They act as an impervious hamer for kirSfWat er reserv01rs and frequently form bedrock and/or the caprock for these. Sen onian posHectomc deposits (sandstones, conglomerates and less frequently argillaceous shales) with local extension m northern part of map area Senoman reservo1rs s upply spnngs with discharge up to 3 Vs, and also in north-east subJacent karst reserv01rs. Marly and argillaceous depOSitS, devoid of groundwater flow, and flysch-like series, includmg rockRcomplexes of variabl e perrneab1hty ( marls, argillaceous shales, sands tones hmes_tones). hosting occas10nally discontmuous aqmfer accumulat10n occumng, in the more permeabl e terms. Pann o nian-Quat ernary deposits (marls, argtllaceous shales, sands, gravels) hosting discontinuous water accumulations in the more permeable tenns I CORRELATION OF THE BIHOR UNIT AND THE UNITS OF THE CODRU NAPPES SYSTEM c R E T A c. J u R A s s I c T R A Tithnnir Rhaetian Norian S Ladinian 8 l H 0 R UNIT Urgoruan reef limestones, marls, sandstones OF THE BIHOR VLADEASA Urgoman limestones limestones BLEAHUet co oR u black lms with cherts N A P P E S limestones SYSTEM gray ltmestones sandstones (Unn!t Fonn) (1!+12) dolonutes gray lms & s dolomttJc I C Ams1an PERMIAN Carpathian Keuper. argtllaceous shales and marls wnh gray dolomites, Codru Formation arglilaceous shales, breccias, limestones Beds ortoquanznes, sandstones and argillaceous shales m pyrometamorph!C fac1es, Ecleja Formation marls, sandslones, limestones Gresten }.'ormation quartzit1c sandstones and conglomerates, argillaceous shales limestones KOssen Formation -hmes10nes, black marly shales,' -Fom1at1on, Sc:irita Formation limestones, argillaceous shales, breccms, Valea Frunzii Formation -magnes1an limestones sandstones, conglomerates black marly shales, Zugi'il Formation -breccl