Journal of cave and karst studies

Citation
Journal of cave and karst studies

Material Information

Title:
Journal of cave and karst studies
Series Title:
Journal of Cave & Karst Studies
Alternate Title:
Continues NSS bulletin (OCLC: 2087737)
Creator:
National Speleological Society
Publisher:
National Speleological Society
Publication Date:
Language:
English

Subjects

Subjects / Keywords:
Geology ( local )
Genre:
serial ( sobekcm )
Location:
United States

Notes

General Note:
Articles: http://dx.doi.org/10.4311/2010ES0138R1 Electrical resistivity imaging of cave Divas¡ka jama, Slovenia / M. Andrej and S. Uros http://dx.doi.org/10.4311/2011JCKS0194 A survey of the algal flora of anthropogenic caves of Campi Flegrei (Naples, Italy) archeological district / P. Cennamo, C. Marzano, C. Ciniglia, G. Pinto, P. Cappelletti, P. Caputo, and A. Pollio http://dx.doi.org/10.4311/2010ES0188R Assessment of spatial properties of karst areas on a regional scale using GIS and statistics - the case of Slovenia / M. Komac and J. Urbanc http://dx.doi.org/10.4311/2011PA0239 Lizards and snakes (Lepidosauria, Squamata) from the late Quaternary of the state of Ceara´ in northeastern Brazil / A.S. Hsiou, P.V. de Oliveira, C.L. Ximenes, and M.S.S. Viana http://dx.doi.org/10.4311/2011JCKS0208 Diet of the newt, Triturus carnifex (Laurenti, 1768), in the flooded karst sinkhole Pozzo del Merro, central Italy / A. Romano, S. Salvidio, R. Palozzi, and V. Sbordoni http://dx.doi.org/10.4311/2011MB0227 Human urine in Lechuguilla Cave: the microbiological impact and potential for bioremediation / M.D. Johnston, B.A. Muench, E.D. Banks, and H.A. Barton http://dx.doi.org/10.4311/2011ES0247 A method to determine cover-collapse frequency in the Western Pennyroyal karst of Kentucky / J.C. Currens, R.L. Paylor, E.G. Beck, and B. Davidson
Restriction:
Open Access - Permission by Publisher
Original Version:
Vol. 74, no. 3 (2012)
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:
K26-02063 ( USFLDC DOI )
k26.2063 ( USFLDC Handle )
12024 ( karstportal - original NodeID )
0146-9517 ( ISSN )

USFLDC Membership

Aggregations:
Added automatically
Karst Information Portal

Postcard Information

Format:
serial

Downloads

This item has the following downloads:


Full Text

PAGE 1

rf r fntbnnntntntttnt rfrf ttnttnntt ntnn nftbftftntnnfnftnnnfnftffffnnfrfr tntntttbnttnttnn tnt ffbfrrf ntttntntttttttttnttn bnfnffntnfbfbnnrf nttnntntt fffbffnfffnfnfffrf tnnnttnnntttntnntn nffnfrf ntntt bttnnnfnf fntbb nnt b rtf ff r r r r rf

PAGE 2

r fnftntnnnnt ttttt fn nn nntntnnnn ttt tntntnnn tnntnt f fnft ntfntntttttn nntntttntt tttt f nnttnttn tfnnnntnt fnnttn tfnn fnntnn tntnntttt ntttnn ntttttn ffttntntnnt ttttttttn nttnnt btttnntnt ttttnttnttt ntntnt f fnnttt tn bffbffnn tttttt ttntntn tttntttnnnt tttntnnn ttntnnt tnnr ttntnntnt tttnt nttn nntntnttntb nnttntt ntttt ntnnttnntn tntntttn tnntntnnn tt fntttnt nnttttn nntn ntntttnnn tntntttnn ntnttn ttt tnnttnt nttttttt tntnttt bttntnnttn ntnttt ttttttn fntttnfnn nttttb tttttf btnttntnt ntnntntn tnnntt nttntt nntnt ttntntt ttnttnnnnt ttttttt nnfnnnfnntt nt fnnntttn tntntttttnntt tttnttnn nttnttttttt tnntttt nntnntn nn fntnnt nnntnnnt tntnnnn ntnnnnnt nnttntt fbfftnn tntnnnnt nnttntnt tnn bfntnttnn tntttttnn nnt nnnnn nttttn tntntt f ttt nntt bbbff ntnnnttnt tnt ttttnnntn ttttttnn bf ntnttt nbntnt tntnttn ntntbntnt tntntfntt nfnnnttnntn ntnttntn nftnn ntntt ffbffnnntn ntntnnnttfnftrrntnnnnttnnt ttntntntntttrrrtrr tntntt fnnnnttnttnnntttnt fbttfnfttntttr fnftnnnnntnftnn tntnttftn ntntntn ttfnttnttnntnnnt fffn tntfnt btt fntn tnt tn nt ttff r rtntn n rrr r r r n rr nntt ff nnt nntff nftn ttnn ttnttntt tnn ntt tb tnttntt tft tn tnnt tnt ftttt tnttnn tfff nt tn tt r n ntftn tnnt ttt

PAGE 3

ELECTRICALRESISTIVITYIMAGINGOFCAVEDIVAS KA JAMA,SLOVENIA M IHEVC A NDREJ 1 AND S TEPIS NIK U ROS 2 Abstract: Electricalresistivityimagingisawidelyusedtoolingeophysicalsurveysfor investigationofvarioussubsurfacestructures.Toassessitsapplicabilityforsubsurface karst,electricalresistivityimagingwasconductedinthesoutheasternpartofthekarst plateauaboveDivas kajamaanditssediment-filleddenudedcontinuationonthesurface. Cavepassagesthatarenotfilledwithsedimentwerenotdetectedwithelectrical resistivityimaging,becausetheelectricalresistivitydifferencebetweenvoidsandhighly resistivecarbonatebedrockissmall.Ontheotherhand,denudedcavesandcavesections thatarefilledwithloamymaterialcanbeclearlydistinguishedfromlessresistive carbonatebedrock. I NTRODUCTION ThestudyareaissituatedinasoutheasternpartoftheKras plateaucalledtheDivac akarstandonthenorthwesternsideof theDivac akarstabovethecavesDivas kajamaandTrhlovca andtheirdenudedcontinuati ontowardstheeast.This manuscriptdiscussest heapplicationofele ctricalresistivity imagingonthesurfaceaboveknownpassagesinDivas ka jama,itspresumedsubsurfacecontinuation,anditsdenuded continuationontherimandsl opeofcollapsedolineGorenjski Radvanj.Themainpurposeofthepaperistotestthe applicabilityofelectricalresisti vityimagingtotheinvestigation ofsubsurfacestructureswher etherearesmallresistivity differences. TheKrasisalimestoneplateausituatedabovetheTrieste BayinthenorthernAdriaticSea.StretchingintheDinaric (northwest-southeast)direction,itis40kmlong,14kmwide, andcoversabout440km 2 .Itismorphologicallyquitedistinct fromthesurroundingregions.L owerflyschregionsandthe AdriaticSeabounditonthesouthwestandthenortheast,and tothenorthwestitissurroundedbythefluvialsedimentsofthe RiverSoc a(Isonzo)plain.Towardsthesoutheast,theborder oftheKrasiswell-definedbythenon-carbonateflyschBrkini HillsandtheRiverRekavalley. TheDivac akarstissituatedinthesoutheasternpartof theKrasplateaubetweenthehinterlandoftheRiverReka ponorandthetownofDivac a(Fig.1).Thebedrockinthe areacomprisesthickly-beddedCretaceouslimestone,dippingapproximately20degreestowardsthesouth,andis boundedtothesouthandnorthbyPaleogenethin-bedded limestone.Ontheedgeofthearea,theRiverRekasinks intoS kocjanskejameattheelevationof317ma.s.l.The terminalsumpofthis5800mlongcaveisat190ma.s.l. Beyondabout900mofunexploredpassages,theundergroundriverflowsthrough12,750mlongKac najama. ThesurfaceoftheDivac akarst,atapproximately430m, islargelyflat,withnumeroussolutiondolines,collapse dolines,anddenudedcaves.Solutiondolinesare50to 100mindiameterandareabout10mdeep.Theirdensity canbehigherthantwohundreddolinesperkm 2 .The volumesvarybetweensomethousandstoseveraltensof thousandsofcubicmeters(Mihevc,1997).Onthesurface, therearealsotwenty-sevenlargecollapsedolineswitha totalvolumeofmorethan41 3 10 6 m 3 .Theirmeandepthis about45m,andtheirmeandiameteris135m.Onthe planatedsurface,itispossibletorecognizeseveraldenuded cavesthataremostlyunroofedsectionsofhorizontal orsub-horizontalepiphreaticcavepassages.Thelargest sectionisabout30mwideandcanberecognizedovera distanceofabout600m(Mihevc,1997). Thestudyarea(Fig.2)issituatedontheedgeofa dolineinthenorthwestpartoftheDivac akarst.The surfaceismostlyflatatanelevationofabout460m,andis interruptedbyseveraldolineswithdiametersupto100m andabout15mdeep.Theeasternpartofthesurface graduallydipsintotheelongateddepressionofadenuded cavethatinitseasternpartcontinuesintotheDivas ki Radvanjcollapsedoline.Twolargecavesareknowninthis area.ThebiggestisDivas kajama,whichrunsapproximatelysouthwest-northeastatanelevationbetween350 and410m.Theotherlargecave,Trhlovca,islocated southwestofDivas kajama. Divas kajamaisdevelopedinbeddedlimestoneof Senonianage(Jurkovs eketal.,1996).Limestonebedsin thecavediptowardthesouthwestinthenortheasternpartof thecaveandtowardthesouthinthesouthwesternpart (Gospodaric ,1985).Thecaveisaroughly700mlongrelictof anoriginallylargercavesystemofepiphreaticandpartially phreaticorigin.Themainpartofthecaveconsistsofalarge passageupto20mhighand15mwide.Thecaveisfilled withatleast30moflithologicallyvariedsedimentsand speleothemsofdifferentages.Themostextensivesedimentin thecaveisthick,laminatedfloodloam.Theloamfilledup mostofthecave,butwaslaterpartiallyerodedawayinlower partsbypercolatingwater.Bothendsofthecavearechoked withallogenicsedimentsandflowstone.Theonlyknown 1 KarstresearchinstituteZRCSAZU,Titovtrg2,SI-6230Postojna,Slovenia, mihevc@zrc-sazu.si 2 UniversityofLjubljana,DepartmentofGeography,As kerc eva2,SI-1000 Ljubljana,Slovenia,urosstepisnik@hotmail.com M.AndrejandS.Uros ElectricalresistivityimagingofcaveDivas kajama,Slovenia. JournalofCaveandKarstStudies, v.74,no.3, p.235242.DOI:10.4311/2010ES0138R1 JournalofCaveandKarstStudies, December2012 N 235

PAGE 4

subsurfacecontinuationoftheDivas kajamaisTrhlovca, althoughtheyarenotdirectlyconnected.Trhlovcais142m longand22mdeep.Theentrancetothiscaveisbelowvertical wallsatthesideofadoline.Thedolineprobablyrepresents theunroofedcontinuationofthecave,asthepassagethat connectsthemainpassagewiththesurfaceisaphreatic channelinterceptedbythesurface.Thesouthwesternendof Divas kajamaisabout40mbelowtheendofthisdoline. Trhlovcaisdevelopedinbedded,southerlydippinglimestone oftheSez anaformation(Jurkovs eketal.,1996).Themain partofthecaveisameanderingcanyonapproximately15m high,about3mwide,and60mlongrunningnorth-south atanelevationof404to419m.Scallopsandundulating notchesaredevelopedonwalls,indicatingevolutionin phreaticandpartiallyinparageneticconditions.Thispassage wascompletelyfilledwithclasticfluvialsediments.Thecave becameaccessibleafterthesedimentswerewashedout (ZupanHajnaetal.,2008). Intheeastofthestudyareaaretwocollapsedolines, Divas kiRadvanjandGorenjskiRadvanj,whichistheactual easternlimitofthestudyarea.TheslopesofGorenjski Radvanjaremostlybalanced.Lowerpartsoftheslopesare coveredwithloamymaterial.O nthewesternslopesthereare twoerosiongulliesfilledwithsed imentconsistingofclay,silt andsand,andflowstone. M ETHODS Althoughelectricalresistivityimaginghasbeensuccessfullyutilizedforcharacterizingthesubsurfaceformany years,ithascertainlimitations.Themethodislaborintensive,interpretationofthedataistimeconsuming,and theresultsarebasedonsubjectiveinterpretation(Roman, 1952;Zhouetal.,2002).Thedevelopmentofcomputer controlledmulti-electrodesystemsandresistivitymodeling softwarehaveallowedmorecost-effectiveresistivitysurveysandbetterinterpretationofthesubsurface(Lockeand Barker,1996).Thesesurveysareusuallyreferredtoas electricalresistivityimaging(ERI)orelectricalresistivity tomography(ERT)(Zhouetal.,2002).Thesemethods allowdatatobecollectedandprocessedquickly,sothat ERIsurveysbecomeavaluabletoolinsubsurface investigations(Zhouetal.,2000). ERIsurveysaretypicallyconductedtodeterminethe resistivityofsubsurfacefeaturesandcanbeusedto determinethelocationofvariousgeologicandsoilstrata, bedrockfractures,faults,andvoids.Fundamentaltoall resistivitymethodsistheconceptthatcurrentisinjectedinto theground,andthevoltagesinducedbythiscurrentcanbe measured.Thesepotentialsordifferencesofpotential,ratios ofpotentialdifferences,orsomeotherparameterthatis Figure1.Locationofthestudyarea. E LECTRICALRESISTIVITYIMAGINGOFCAVE D IVAS KAJAMA, S LOVENIA 236 N JournalofCaveandKarstStudies, December2012

PAGE 5

directlyrelatedtothesevariablesarethemostcommonly measuredeffectoftheinjectedcurrent.Theprincipal differencesamongvariousmethodsofelectricalresistivity lieinthenumberandspacingofthecurrentandpotential electrodes,thevariablecalculated,andthemannerof presentingtheresults(Zhouetal.,2000). Generally,carbonaterockhasasignificantlyhigher resistivitythanloamymaterial,becauseofitsconsiderably smallerprimaryporosityandfewerinterconnectedpore spaces.Itsresistivityvalueisabout1000ohm-m(Telford etal.,1990).Loamymaterialscanholdmoremoistureand havehigherconcentrationsofionstoconductelectricity; therefore,theirresistivityvaluesarebelow250ohm-m (Telfordetal.,1990).Thehighcontrastinresistivityvalues betweencarbonaterockandloamymaterialfavorstheuse ofelectricalresistivitytodeterminetheboundarybetween bedrockandoverburdenorloamysediment(Zhouetal., 2000). Afrequentlyoccurringproblemwithelectricalresistivityimagingisdecidingwhichelectrodeconfigurationwill respondbesttothematerialchangesinkarstfeatures.Each typeofarrayhasdistinctiveadvantagesanddisadvantages intermsofsensitivitytomaterialvariations,depthfrom whichinformationmaybeobtained,andsignalstrength. Themostcommonarraysarethedipole-dipolearray,the Wennerarray,andtheSchlumbergerarray.Thedipoledipolearraygivesgoodhorizontalresolution,whilethe WennerandSchlumbergerarraysaremoreintendedfor Figure2.Top:Detailedmapofthestudyareashowingtheelectricalresistivityimaginglinesoverthecave(13),itspresumed filledcontinuation(4),andtheunroofedcaveatthesinkhole(58).Bottom:ProfilesketchalongthecurveABinthetoppart, withdepthsoftheelectricalresistivityimagingprofilesthatwerecalculated. M.A NDREJAND S.U ROS JournalofCaveandKarstStudies, December2012 N 237

PAGE 6

verticalresolution.Intheapplicationtokarstsurveys,the dipole-dipolearrayhasprovidedhighestprecisionof groundchangessensitivityandhasthegreatestsensitivity toverticalresistivityboundaries(Zhouetal.,2002). Electricalresistivitydatawerecollectedalongeight differentlinesaboveDivas kajama,itspresumedcontinuation,andthedenudedsectionontheslopeofthecollapse doline(Fig.2).TheSuperStingR1/IPearthresistivity meterdevelopedbyAdvancedGeosciences,Inc.wasused fordatacollection.Thesurveywasconductedwitha dipole-dipolearraywith5melectrodespacing.Inmost cases,twentyelectrodeswereusedsimultaneously,with alternationoftwocurrentandtwopotentialelectrodes. Forlongerprofiles,aroll-alongsurveywasused.Thedata wereprocessedtogeneratetwo-dimensionalresistivity modelsusingEarthimager2Dresistivityinversionsoftware developedbyAdvancedGeosciences,Inc.Thiscombinationofequipmentandsoftwarehavebeenshowntobe appropriateforprovidingarobustvisualizationofthe epikarststructureandthesubsurfacestructureofcollapse dolines(Stepis nikandMihevc,2008;Stepis nik,2008).The root-mean-squareerrorquantifiesthedifferencebetween themeasuredresistivityvaluesandthosecalculatedfrom thetrueresistivitymodel.AsmallRMSvalueindicates smalldifferences.TheminimumRMSerrorinthesurvey was2.59%,andthemaximumerrorwas8.2%. Previousapplicationsofthismethodinvariouskarst featuresintheSloveniankarstrevealedthattheresistivity valueforcarbonaterockexceeds1000ohm-m.Forsoiland weatheredbedrock,theresistivityvaluesarebetween approximately200and1000ohm-m.Loamymaterialhas resistivityvalueslowerthan150ohm-m(Stepis nik,2007; Figure3.ERIprofiles14,scale0to1500ohm-m. E LECTRICALRESISTIVITYIMAGINGOFCAVE D IVAS KAJAMA, S LOVENIA 238 N JournalofCaveandKarstStudies, December2012

PAGE 7

Stepis nikandMihevc,2008;Stepis nik,2008).However, thresholdresistivityvaluesthatwoulddiscriminatebetweenvoidsandcarbonatebedrockhavenotyetbeen determined.Sincevoidsshouldhaveinfiniteresistivity,the analysiswasrepeatedatdifferentrangesofresistivityto checkifsubsurfaceopeningscanbedetected. TheERIprofilesacrossDivas kajama(profiles1,2and 3)exhibitrelativelyuniformsubsurfacestructure,whichis aresultofthehighelectricalresistivityoflimestone bedrock,aswellascavevoids(Fig.3).Line1wassituated onthesurfaceabovethesoutheasternendoftheDivas ka jama,oriented290 u .Eventhoughthesurfaceisgently inclinedtowardsthenorth,theinclinationisuniform,and soforthepurposeoftheanalysis,thetopographyofthe profileispresentedasflat(Fig.3).Inthisprofile,bedrock withresistivityvaluemorethan1000ohm-miscoveredby thinlayersoflessresistivesoil,mechanicallyweathered rock,orloamymaterialwithresistivityabout500ohm-m. Inthecentralpartoftheprofile,atthedepthofabout25m thereisaclearlydistinguishedareawithelectricalresistivity lowerthan500ohm-m,whichmightbeahigh-level extensionofDivas kajamatowardsTrhlovcathatis completelychokedwithloamysediment.Knownpassages ofDivas kajamaarepositionedabout50mbelowthe surfaceandwerenotdetectedintheERIprofile,asthe maximumdepthinthisprofilewas28m. Line2wassituatedabovethecentralpartofDivas kajama thatliesapproximately15mbelowthesurface.Thesurfaceis onthenorthwesternslopeofadolineinthedirectionof300 u Inthisprofile,bedrockwithresistivityvaluemorethan Figure4.ERIprofile2computedwithadditionalhigherscalesofresistivityvalues. M.A NDREJAND S.U ROS JournalofCaveandKarstStudies, December2012 N 239

PAGE 8

1000ohm-miscoveredbythinlayersoflessresistivesoil, mechanicallyweatheredrock,orloamymaterialwith resistivityabout500ohm-m.Inthecentralpartoftheprofile, atthedepthofabout15m,acavepassageinDivas kajamais present.Itshouldbeseenontheprofile,asresistivityvalues shouldbeextremelyhigh.However,resistivityvaluesof thewholesectionoftheprofileappearlikethoseofthe surroundingbedrock.Differentrangesofresistivityvalues wereusedinanattempttofindaresistivitythresholdbetween thecavepassageandthesurroundingbedrock(Fig.4).None oftheappliedranges,differingbyanorderofmagnitude, allowedusdetecttheactualcavechamber.Athigher resistivityvalues,someanomaliesweredetectedthatmight betensionalfracturesabovethecave(15000and20000ohmm),buttheyarenotatthedepthofthecave. Line3wassituatednearthenortheasternendof Divas kajama,whichhereliesapproximately22mbelow thesurface.Thelineranacrossaflatkarstsurfacecovered withgrikes,inthedirectionof305 u .Inthisprofile,bedrock withresistivityvaluemorethan1000ohm-miscoveredby thinlayersoflessresistivesoil,mechanicallyweathered rock,orloamymaterialwithresistivityabout500ohm-m. PassagesinDivas kajamalieinthecentralpartofthe profile,buttheywerenotdetectedbyuseofERIbecause theyhavethesameapparentresistivityvaluesasthe surroundingbedrock. Profile4issituatedonaflatkarstsurfacebeyondthe northeasternendofDivas kajama,overitspresumed subsurfacecontinuationtowarddenudedcavenexttothe GorenjskiRadvanjcollapsedoline(Fig.4).Inthisarea,no Figure5.ERIprofiles58,scale0to1500ohm-m. E LECTRICALRESISTIVITYIMAGINGOFCAVE D IVAS KAJAMA, S LOVENIA 240 N JournalofCaveandKarstStudies, December2012

PAGE 9

accessiblecaveisknown.Directionoftheprofileis300 u .In thisprofile,too,bedrockwithresistivityvaluemorethan 1000ohm-mispartiallycoveredbythinlayersofless resistivesoil,mechanicallyweatheredrock,orloamy materialwithresistivityabout500ohm-m.Inthecentral partoftheprofile,fromthesurfacetothedepthofabout 10m,thereisevidenceofasmallsubsurfacestructurewith resistivityvalueslessthan500ohm-m.Mostlikelythe structureispartofanepikarstvoidfilledwithloamy materialorsoil.Heretoo,distinctgrikesarepresentonthe surface.Inthesoutheasternsectionoftheprofileatdepths greaterthan17m,asubsurfacestructurewithadiameter ofabout30misevident.Withresistivityvalueslessthan 500ohm-m,thestructureisapparentlyacavepassage completelyfilledwithloamymaterial.Thispresumablyisa continuationofDivas kajama. Lines5through8ranacrosstheunroofedcavesection completelyfilledwithloamymaterialandflowstone (Fig.5).ResultsofERIexhibitacleardifferencebetween cavefillandbedrock.Line5,inthedirectionof290 u ,was placedoverthesouthwesternsectionofthedenudedcave trench.Itsprofileexhibitssomebedrockwithresistivity valuearound1000ohm-minthesoutheasternsection.Itis coveredwithathinlayerofelectricallylessresistivesoil, mechanicallyweatheredrock,orloamymaterialwithresistivityvalueabout500ohm-m.Allotherpartsofthe profileexhibitsomeresistivityvalueslowerthan150ohmmthatindicatesloamymaterialandflowstonefillmaterial inadenudedcave.Materialwithresistivityvaluesaround 500ohm-mlocatedinthecentralandnorthwesternsectionsoftheprofileatadepthbetween10and15mis probablyweatheredbedrockthataccumulatedtheredueto slopeprocessesinsidethedenudedcave. Line6wassituatedonthedenudedcavenortheast ofline5,ontherimofthecollapsedoline.Thelinewasrun at290 u ,perpendiculartothedirectionofthedenudedcave. Intheresultingprofile,bedrockwithresistivityvaluemore than1000ohm-mispresentonbothslopesofthetrench. Itispartiallycoveredwiththinlayersofelectrically lessresistivesoil,mechanicallyweatheredrock,orloamy materialwithresistivityvalueabout500ohm-m.The centralpartoftheprofileshowsover15mofloamy materialandflowstonefragmentswithresistivityvalues below150ohm-m. Line7,runat350 u ,wasplacedonthewesternslope ofthecollapseddolineclosetothedenudedcave.Bedrock, withresistivityvaluesmorethan1000ohm-m,ispresent alongthewholeprofile.Alow-resistivityarea(below 150ohm-m)inthecentralpartoftheprofile,wherethe surfaceiscoveredwithloamandflowstoneparticles,isup to10mthick. Line8wassituatedonthefloorofthewesternpartof thecollapsedolinejustundertheslopewherethedenuded caveisdisintegrating.Bothendsoftheprofileshowthe presenceofbedrock,withresistivityvalueshigherthan 1000ohm-m,ontheslopesofthecollapsedoline.Inthe upperpartofcentralsectionoftheprofile,materialwith resistivityvalueslowerthan150ohm-mappearsuptoa depthof5m.Thisismostlikelyloamyoutwashofthe denudedcavefillfromtheslope.Belowtheoutwash,at depthsbetween5to15m,theprofileshowsresistivity valuesfrom150to500ohm-mthatmostlikelyrepresent weatheredbedrockaccumulatedasscreeatthefootofthe slope.Below,thereisagainthematerialthatexhibits resistivityvalueslowerthan150ohm-m,suggestingloamy fillinthedoline(Stepis nik,2008). C ONCLUSIONS Electricalresistivityimagingdatawerecollectedforeight linesovercavesDivas kajamaandTrhlovcaandacrosstheir denudedcontinuationontheslopeofthecollapsedoline GorenjskiRadvanj. TheERIprofilesacrossDivas kajama(profiles1,2and3) exhibitrelativelyuniformsubsurfacestructurethatisaresult ofthehighelectricalresistivityoflimestonebedrockandcave voids.Althoughthecavepassagesarerelativelyclosetothe surface,theywerenotdetectedwiththeapplicationofERI, evenatthehighestresistivityvaluesthatshouldshowthe differencebetweenbedrockandvoid. Profile4,acrossthepresumedundergroundcontinuationofDivas kajamainthedirectionoftheunroofedcave showssomedifferencesinsubsurfaceelectricalresistivity thatmayindicatetheexistenceofcaveconduitscompletely filledwithlessresistiveloamymaterial.Theunroofed sectionofthecaveiscompletelyfilledwithloamymaterial andflowstone.ERIprofiles4,5,6,and7exhibitaclear differencebetweenallogeniccavefillandbedrock.Inthe uppersectionabovetheslopesofthecollapsedoline,where thedenudedcaveisupto20mwide,theloamyfillis15m thick.Ontheslopes,thethicknessofloamyfilldiminishes, probablybecauseithasbeenwashedintothedoline.The ERIprofileinthelowersectionoftheslopeexhibitsupto 25mofloamymaterialfill. ApplicationoftheERImethodhasprovedappropriate fordetailedinvestigationofsubsurfacestructureswithlarge differencesinelectricalresistivity.Partsofdenudedcaves andcavepassagesthatarefilledwithloamymaterialcanbe clearlydistinguishedfromlessresistivecarbonatebedrock. InthemeasuredERIprofiles,resistivityvaluesofsoil-and sediment-filledfeaturesarelowerthan150ohm-mand weatheredbedrockisaround500ohm-m,whilebedrock exhibitsvalueshigherthan1000ohm-m. Ontheotherhand,undergroundpartsofthecaveswith hugechamberswerenotdetectedinthissurveybyERI method,asresistivitydifferencesbetweenvoidsandthe highlyresistivecarbonatebedrockareinsignificant.In calculatedprofileswithhighmaximumresistivity(Fig.4), limestonebedrockexhibitsresistivityvaluesapproximately between5000and10000ohm-m.Previousapplicationsof ERIovercavepassagesgaveresistivityvaluesoflimestone bedrockupto5000ohm-m,whilevoidshavehighervalues M.A NDREJAND S.U ROS JournalofCaveandKarstStudies, December2012 N 241

PAGE 10

(e.g.,Barbadelloetal.,2002;Brownetal.,2011).Inthis case,theproblemofnotdetectingthevoidsseemstobea consequenceoftheveryhighelectricalresistivityofthis typeoflimestone. R EFERENCES Barbadello,L.,Bratus,A.,Yabar,D.N.,Paganini,P.,andPalmieri,F.,2002, Integratedgeophysicalmethodstodefi nehypogenouskarsticfeatures:Atti delMuseoCivicodiStoriaNatura lediTrieste,v.40,p.1521. Brown,W.A.,Stafford,K.W.,Shaw-Faulkner,M.,andGrubbs,A.,2011, AcomperativeintegratedgeophysicalstudyofHorseshoeChimney Cave,ColoradoBendStatePark,Texas:InternationalJournalof Speleology,v.40,no.1,p.916. Gospodaric ,R.,1985,OspeleogeneziDivas kejameinTrhlovce:Acta Carsologica,v.13,p.534. Jurkovs ek,B.,Toman,M.,Ogorelec,B.,S ribar,L.,Drobne,K.,Poljak, M.,andS ribar,L.,1996,Formacijskageolos kakartajuz negadela Trz as kokomenskeplanate1:50.000:kredneinpaleogenske karbonatnekamnine,Ljubljana,Ins titutzageologijo,geotehnikoin geofiziko,143p. Mihevc,A.,1997,Dolines,theirmorphologyandorigin,casestudy: dolinesfromtheKras,westSlovenia(theS kocjankarst), in James,J., andForti,P.,eds.,FourthInternationalConferenceonGeomorphology,KarstGeomorphology(GeografiaFisicaeDinamica Quaternaria,Supplement3,part3),p.6974. Roman,I.,1952,Resistivityreconnaissance, in SymposiumonSurfaceand SubsurfaceReconnaissance,Philadelphia,AmericanSocietyfor TestingMaterials,SpecialTechnicalPublication122,p.171226. Stepis nik,U.,2007,Loamysedimentfillsincollapsedolinesnearthe LjubljanicaRiversprings,DinaricKarst,Slovenia:CaveandKarst Science,v.33,p.105110. Stepis nik,U.,2008,Theapplicationofelectricalresistivityimagingin collapsedolinefloors:Divac akarst,Slovenia:StudiaGeomorphologicaCarpatho-Balcanica,v.42,p.4156. Stepis nik,U.,andMihevc,A.,2008,Investigationofstructureofvarious surfacekarstformationsinlimestoneanddolomitebedrockwith applicationoftheelectricalresistivityimaging:ActaCarsologica, v.37,no.1,p.133140. Telford,W.M.,Geldart,L.P.,andSheriff,R.E.,1990,AppliedGeophysics (2.edition):NewYork,CambridgeUniversityPress,790p. Zhou,W.,Beck,B.F.,andAdams,A.L.,2002,Effectiveelectrodearrayin mappingkarsthazardsinelectricalresistivitytomography:EnvironmentalGeology,v.42,p.922928.doi:10.1007/s00254-002-0594-z. Zhou,W.,Beck,B.F.,andStephenson,J.B.,2000,Reliabilityofdipoledipoleelectricalresistivitytomographyfordefiningdepthtobedrock incoveredkarstterrains:EnvironmentalGeology,v.39,p.760766. doi:10.1007/s002540050491. ZupanHajna,N.,Mihevc,A.,Pruner,P.,andBosa k,P.,2008, Paleomagnetismandmagnetostratigraphyofkarstsedimentsin Slovenia,Ljubljana,Zaloz baZRC,CarsologicaSeries,266p. E LECTRICALRESISTIVITYIMAGINGOFCAVE D IVAS KAJAMA, S LOVENIA 242 N JournalofCaveandKarstStudies, December2012

PAGE 11

ASURVEYOFTHEALGALFLORAOFANTHROPOGENIC CAVESOFCAMPIFLEGREI(NAPLES,ITALY) ARCHEOLOGICALDISTRICT P AOLA C ENNAMO 1 ,C HIARA M ARZANO 2 ,C LAUDIA C INIGLIA 3 ,G ABRIELE P INTO 2 ,P IERGIULIO C APPELLETTI 4 P AOLO C APUTO 2 AND A NTONINO P OLLIO 2 Abstract: CampiFlegreiisalargevolcanicareasituatednorthwestofNaples,Italy. Twoarcheologicalsites,theSybil’sCaveandthePiscinaMirabilis,areartificialcaves dugintheyellowtuffandusedduringantiquityforvariouspurposes.Thispaper describesforthefirsttimethealgalbiodiversityofthesecavesanddetermineswhether environmentalfactorssuchaslightintensityandhumidityareinfluentialinspecies distribution.Atotaloftwenty-twoalgalspecieswereidentifiedbymolecularmethods (18SrDNA);thelargestgroupwasCyanobacteria(elevenspecies),followedbyalgae Chlorophyta(six),Rhodophyta(two)andBacillariophyta(two).Clusteranalysisofalgal distributioninthecavesinrelationtolightandhumidityshowednorelevantdifferences inalgaldistributionbetweenthetwocaves.Threedifferentalgalgroupswereidentified. Thefirstoneincludesstrainsstrictlydependentonlowhumidity,asecondclusterwas mainlyassociatedwithsiteswherehumidityisnotasevereconstraint,andathirdgroup, mainlyrepresentedbyfilamentouscyanobacteria,isprobablydependentonhigh humidity,sinceitwasdetectedonlyatPiscinaMirabilis. I NTRODUCTION Thealgalfloraofcavesisdrawingincreasingattention duetothepeculiaritiesoftheirenvironment.Reducedlight intensity,lownutrientinput,andabsenceofseasonality (DaynerandJohansen,1991;Pedersen,2000)arethepredominantfeaturesthatinfluencedistributionandcompositionofalgalassemblagesinnaturalandartificialcaves,but temperature,humidity,andoccurrenceofflowingwateralso playaroleintheestablishmentofalgalsettlements(Mulec etal.,2007).Manycavehabitatscurrentlysuffertheimpactof increasingtouristfluxes;artificiallightsandhumantrafficare strongfactorsthatcandeeplymodifyequilibriawithinalgal populations(Cheliusetal.,2009).Anexampleisgivenbythe anthropogeniccavesoccurringinthearchaeologicaldistrict ofCampiFlegreiinItaly,anareathatisthesinglelargest featureofthePhlegraeanVolcanicDistrict,whichincludes theislandsofProcidaandIschia,aswellassubmarinevents inthenorthwesternbayofNaples(Orsietal.,1996).Campi Flegreihasbeeninhabitedsinceprehistorictimesandbecame animportantcenterofcivilizationduringantiquity.Different undergroundhabitatshavebeenduginthedistrict,andtwo ofthemhaveremarkabledimensions.Themostancientisthe so-calledSybil’sCave(fourthcenturyBCE),nearCuma, whichisinfactalonggallerycutintothesofttuff,lightedby multipleopeningstothesurface.Theotherlargecave(dating backtothefirstcenturyCE)isthePiscinaMirabilis,located inBaia,acisternalsoexcavatedintuff,usedtostoredrinking water.Thetwocavespresentsomecommonfeatures.They sharethesameexternalclimaticconditions,beinglocatedless than6kilometersfromeachother,andsimilarviablealgal propagulescanbedispersedintobothcavesbyaircurrents andtransportbyanimals(Dobat,1970).Inaddition,they havebeenduginthesamegeologicalmaterialthatoriginated fromtheeruptionoftheNeapolitanYellowTuff(datedat 15kaBP;Insingaetal.,2004),whichproducedthecaldera collapseofthebayofPozzuoli(Scarpatietal.,1993). Ontheotherhand,thetwocavesdifferinsomecharacteristics.TheSybil’sCaveis,inpart,directlylightenedby daylightandisvisitedbynumeroustouriststhroughoutthe year,whereasthePiscinaMirab ilisdoesnotreceivedirectlight andisrarelyopentothepublic.Moreover,ithasbeenpartially filledforcenturieswithfre shwaterwithahighcarbonate concentration,causingtheformationofcalcareousincrustationsalongthewallsofthecave,andgroundwaterinfiltrations arestillfrequent.TheSybil’sCaveisadryhabitat,withvery littlewaterinfiltration.Thesimilaritiesanddifferencespresent anopportunityforobservinghowtheyinfluencethe compositionofthealgalassemblagesinthetwocavesand couldhelptoclarifythecontributionofmicroalgaetodeteriorationprocesses. Biodeteriorationincavesoccursasaconsequenceof thepresenceofmicrobialcommunities,formedofalgae, bacteria,andfungi,thatdevelopthickbiofilmsonany rocksurface,leadingtodecay(Albertanoetal.,2003). Subterraneanalgalcommunitiesaregenerallydividedinto *Correspondingauthor:pcennamo@unina.it 1 Facolta `diLettere,Universita `degliStudi‘SuorOrsolaBenincasa’,ViaS.Caterina daSiena37,I-80135Napoli,Italy 2 DipartimentodelleScienzeBiologiche,Universita `degliStudidiNapoli‘Federico II’,OrtoBotanico,ViaForia223,NapoliI-80139,Italy 3 DipartimentodiScienzedellaVita,SecondaUniversita `diNapoli,ViaVivaldi43, CasertaI-81100,Italy 4 DipartimentodiScienzedellaTerra,Universita `degliStudidiNapoli‘FedericoII’, ViaMezzocannone8,NapoliI-80134,Italy P.Cennamo,C.Marzano,C.Ciniglia,G.Pinto,P.Cappelletti,P.Caputo,andA.Pollio–Asurveyofthealgalfloraofanthropogenic cavesofCampiFlegrei(Naples,Italy)archeologicaldistrict. JournalofCaveandKarstStudies, v.74,no.3,p.243–250.DOI:10.4311/ 2011JCKS0194 JournalofCaveandKarstStudies, December2012 N 243

PAGE 12

theso-calledlampenflora,formedbyspecializedorganisms thatthriveonlyinundergroundhabitatsinthepresenceof artificiallight,andareophilicflora,composedbyterrestrial microalgae,mainlybelongingtoChlorophytaanddiatoms thattoleratecaveenvironmentalconditions,especiallyat theopeningsofcaves(Mulecetal.,2008).Cyanobacteria arethekeyorganismsinthegenesisofbiofilms,being abletoproduceexopolymericsubstancesthatallowthe formationofthemicrobialcommunityanditsadhesion torocks(Stal,2000).Cyanobacteriaexposedtohighlight intensitiesmayhaveanendolithicphaseofgrowth, changingthemineralstructureoftherockanddetermining itsdecay(AsencioandAboal,2001). Inanattempttoascertainthepotentialdamagecausedby thealgalfloraontheundergroundmonumentsofCampi Flegrei,asystematicsurveyofthealgalflorafromthetwo anthropogeniccaveshasbeenattempted.Atthesametime, consideringthatthealgaelivinginthesehabitatscannotbe consideredsolelyasathreat,butalsoasanimportantsource ofbiodiversity,aneffortwasundertakentoisolateand maintainthestrainsinthealgalcollectionofUniversity FedericoIIofNaplestostudytheirecophysiologicalfeatures. E XPERIMENTAL M ETHODS TheSybilÂ’sCave,builtinthefourthcenturyBCE,isa trapezoidalpassageover131mlongrunningparalleltothe sideofthehillonwhichwasbuilttheAcropolisofCumaat CampiFlegrei(Fig.1B1,B2).Itwascutoutinthe volcaniclasticNeapolitanYellowTuff.Thegalleryleadsto amoreenclosedpolygonalhall;themainentranceandthe openingsalongthehallwayensuretheentranceoflightanda continuousfluxofair. ThePiscinaMirabilis,builtinthefirstcenturyCE,isa cisternduginthetuff,70mlong,25.50mwide,and15m deep(Fig.1C1,C2).Thecavityhasarectangularlayoutwith forty-eighttuffpillarscoveredin opusreticulatum ,aformof brickworkusedinancientRoma narchitectureconsistingof diamond-shapedbricksoftuffplacedaroundacoreofconcrete,andisdividedintofourcorridorsthatcomposetheinner partofthecistern.Theouterwalls,in opusreticulatum ,are coveredwithathicklayerof cocciopesto ,mortarwithpotsherds.ThePiscinaMirabiliss upplieddrinkingwaterforthe RomanfleetthatwasbasedinMisenum,nearBaiae(Fig.1A). Duetotheirorientationandposition,theentrancesof bothcavesaredirectlyilluminatedthroughouttheyear. Moreover,theopeningspresentintheSybilÂ’sCaveallow penetrationoflightduringthewholeday,whereasthePiscina Mirabilisisalwaysshady.Sixsamplingpointswereselected ineachcave,onthebasisoflight,temperature,andrelative humiditydata(Fig.1B2,C2).Theyshowedevidenceof biodeteriorationintheformofagreenandbrownpatina.Ten samplingswerecarriedoutwithina20 3 20cmsquareateach point. Biofilmsampleswerecollectedintheautumnof2009by scrapingthewallsofthecaveswithasterilescalpeland Figure1.A,mapofPhlegreanFields.B1,pictureoftheSybilÂ’sCave.B2,mapofandlocationsofthesamplingsitesinthe SybilÂ’sCave(Paganoetal.,1982).C1,pictureofthePiscinaMirabilis.C2,mapofandlocationsofthesamplingsitesinthe PiscinaMirabilis(Rajolaetal.,1978). A SURVEYOFTHEALGALFLORAOFANTHROPOGENICCAVESOF C AMPI F LEGREI (N APLES ,I TALY ) ARCHEOLOGICALDISTRICT 244 N JournalofCaveandKarstStudies, December2012

PAGE 13

depositingmaterialsintosterilevialscontainingeithera specificmediumforCyanobacteria(BG11;Castenholz, 1988)orBBM(BoldBasalMedium;BischoffandBold, 1963)forallothermicroalgae.Sampleswerecollectedin September,October,andNovember(Table1).Temperature,lightintensity,andrelativehumidityweremeasured ateachsamplingsitebyusingappropriateinstruments (TESTO174,TESTO545,AGGermany).Lightirradiance ateachsamplingpointwasmeasuredbyusingaLI-COR radiationsensor(BIOSCIENCES). Thealgalsampleswereinspectedusinganoptical microscope(NikonEclipseE800equippedwithNomarski interference,magnification 3 100).Smallfragmentsof biofilms,aftercritical-pointdrying,weregold-coatedin anEmitechk550SputterCoaterandobservedbyscanning electronmicroscopy(PhilipsEM208S). Quantitativemineralogicalanalysisandthechemical compositionofrockscollectedinthesamplingareaswere determinedbymeansofx-raypowderdiffractionand environmentalscanningelectronmicroscope.Thex-ray diffractionwasperformedwithaPanalyticalXÂ’PertPro modulardiffractometer,andquantitativeanalyseswere obtainedbytheRietveldmethod(BishandPost,1993). Chemicalelementsweremeasuredusingamicroanalysis system(ZAFPB). Quantitativeestimationsoftheidentifiedtaxawerealso made.Foralgalcounts,0.5goffreshbiofilmsamplewas suspendedandfixedin100mLofaqueoussolutionof formaldehyde(3%)andcountedwithahemocytometer within24hours.Toestablishspeciescompositions,partof eachsuspensionwasputonaslide,andtwentyrandomly chosenmicroscopefieldswerecountedusingagrid(LUCIA version4.60SystemforImageProcessingandAnalysis);at leastonehundredcellswerecountedperslide.Threesuch slideswerecountedforeachsample,andthemeanswere analyzedforsignificantvariance(StudentÂ’st-test).The relativestandarderrorwasneverhigherthan5%.Thecount offilamentousCyanobacteriawascarriedoutbycounting filamentsintransectsofachamberfilledwithafew millilitersforsedimentationovernight.Backcalculatingto amlofsamplewascarriedoutbyconsideringthevolumeof thecountingchamberandmeasuringtheareaofthe transectsandofthechamberbottom(Lawtonetal.,1999). Additionally,theidentificationofthemicrobialcomponentsofbiofilmswasaccomplishedbymoleculartechniques.DNAwasextractedusingaproceduredescribedby DoyleandDoyle1990.PCRamplificationwascarriedout onanestimated10ngofextractedDNA.ForCyanobacteria,theprimersetfor16SrDNA(D ezetal.,2001)was used;foralgae,theprimersetfor18SrDNAindicatedby Hussetal.(1999)wasused.PCRreactionswerecarriedout inafinalvolumeof50 m Lcontaining5 m Lof10 3 PCR buffer,100mMofdeoxynucleotidetriphosphate,2.5mMof magnesiumchloride,0.5mMofprimers,and1Uof Taq polymerase(Quiagen,Hilden,Germany).ThePCRprogramconsistedofaninitialdenaturationat95 u Cfor4min and30cyclesincluding1minofdenaturationat94 u C,45s ofannealingat56 u C,and2minextensionat72 u C.Afinal extensionof7minat72 u Cfollowedbycoolingat4 u C terminatedthePCRprogram. Theidentificationoftheentirecommunitywasobtained employingcloninglibraries.AnaliquotofpurifiedPCR productwasligatedintothepGEM-TeasyVectorsystem (Promega,Vienna,Austria),andcloneswerescreenedand sequencedwitha3130geneticanalyzer(AppliedBiosystems).ThesequencesobtainedwerecomparedwithavailablesequencesintheGeneBankdatabase.Clusteranalysis todetectpossiblegroupingofalgaltaxaintermsofenvironmentalpreferenceswascarriedoutbyusingtheSYNTAXvers.5.1computerprogramfordataanalysisin ecologyandsystematics(Podani,2001). R ESULTSAND D ISCUSSION Atallthesamplingpoints,meanmiddaytemperature was18 6 2 u CintheSybilÂ’sCave(SC)and16 6 1 u Cinthe PiscinaMirabilis(PM).ThemeanamountofphotosyntheticallyactiveradiationinSCrangedfrom0.0165to 3.31 m Es 2 1 m 2 2 ,whereasinPMvaluesatsamplingpoints1 to6rangedfrom0.049to2.61 m Es 2 1 m 2 2 (Table1). RelativehumidityinSCrangedfrom69to72%;onthe otherhand,therelativehumidityofPMwasintherange 75to86%(Table1). LightmicroscopyrevealedthepresenceoftaxabelongingtoCyanobacteria,Chlorophyta,Bacillariophyta,and Rhodophyta.Thevariationsinrelativeabundanceofthese taxaamongthevarioussamplingpointinthetwocavesare showninFigure2.Cyanophytawerethedominanttaxaat severalsamplingpoints,bothinSC(sites1to4)andinPM (sites1to4),whereCyanobacteriacellpercentagesranged from50to80%inSCandfrom70to85%inPM.Atthe samesites,Chlorophytarepresentedaminorpercentage Table1.Temperature,lightintensityandrelativehumidityof theSybilÂ’sCave(SC)andthePiscinaMirabilis(PM) samplesites. SampleSites Temperature, u C m E,m 2 2 s 2 1 Relative Humidity,% SC117 6 20.05 6 0.00269 6 1 SC218 6 20.02 6 0.00169 6 1 SC318 6 20.16 6 0.00769 6 2 SC419 6 20.4 6 0.0268 6 2 SC519 6 20.407 6 0.0269 6 2 SC619 6 23.31 6 0.1572 6 3 PM116 6 10.15 6 0.00685 6 1 PM216 6 10.15 6 0.00686 6 1 PM316 6 10.22 6 0.0184 6 1 PM416 6 10.46 6 0.0281 6 1 PM516 6 10.05 6 0.00280 6 2 PM616 6 12.61 6 0.1375 6 2 P.C ENNAMO ,C.M ARZANO ,C.C INIGLIA ,G.P INTO ,P.C APPELLETTI ,P.C APUTO AND A.P OLLIO JournalofCaveandKarstStudies, December2012 N 245

PAGE 14

(lessthan25%)ofthealgalpopulation,butatSC5,PM5, andPM6,wherecyanophyteswerescarcelyrepresented, chlorophytesprevailed,withtheirrelativeabundancebeing directlyproportionaltotheincreaseinlightintensity. Bacillariophyta(diatoms)normallydidnotexceed10%in SCandinPM;atSC4,SC5,andPM6,however,theywere 30and45%respectively.Rhodophytawereabundantonly atSC3,atrelativelylowlightintensity,whileinPMtheir percentagesneverexceeded10%. Microanalysesofthesubstrate,alwayscomposedof NeapolitanYellowTuff(NYT),revealedthepresenceof thesameelementalcompositioninbothsites(Al,Ba,Cd, Co,Cu,Fe,K,Mg,Mn,Na,Si,Sr,Ti,U,andZn),albeit elementswerepresentatdifferentconcentrations.Mineralogicalcompositionofinvestigatedsamplesconfirmedthe compositionofNYT,mainlyphillipsite,chabazite,and analcime,andhydrousaluminosilicatespertainingtothe zeolitegroup,alongwithfeldsparsandminoramountsof clayminerals(deÂ’Gennaroetal.,2000).Datafromthe literature(Morraetal.,2010)statehighporosityvaluesfor NYT,whichischaracterized,moreover,bythepresenceof microporescausingahighcapillaryabsorptionofwater thatallowsmicrobialgrowth.SEMobservationsshowed thatthemicrobialcommunityisembeddedinanexopolysaccharidicmatrixthatfacilitatestheestablishmentof strongbondsbetweenbiofilmandsubstrate(Fig.3).In bothsites,CyanobacteriaadheredstrictlytoNYT,offering anidealenvironmentforthegrowthofotherorganisms, givingenduranceandthicknesstothebiofilm. Identificationofmanyalgalstrainsisdifficultbecauseof thescarcityofusefulfeaturesandthehomoplasiousmorphologyofsomelineages.Tominimizetheriskoferroneous identifications,moleculartechniqueswerealsoused.Only themostsimilarsequencematches(90%identity)with GenBankhavebeenconsidered.Sequencingrevealedatotal ofseventeenphylotypesforSCandthirteenforPMthat wereassignedtoCyanophyta(eleventaxa),Rhodophyta (twotaxa),Chlorophyta(sixtaxa)andBacillariophyta(two taxa)(Table2). Ahierarchicalclusteringwas carriedoutonadissimilarity matrixbasedonthepresence/absenceofalgaltaxaindifferent samplingsitesatdifferentlightintensitiesandrelativehumidities(Jaccardcoefficient1 2 a /( a + b + c )).Twoclassesof humidityweredefinedfortheanalysis,lowhumidity(lessthan Figure2.MicrobialcompositionofbiofilmsintheSybilÂ’sCaveandthePiscinaMirabilis. A SURVEYOFTHEALGALFLORAOFANTHROPOGENICCAVESOF C AMPI F LEGREI (N APLES ,I TALY ) ARCHEOLOGICALDISTRICT 246 N JournalofCaveandKarstStudies, December2012

PAGE 15

75%)andhighhumidity(77to82%),aswellasthreeclassesof lightintensities,lowintensity(from0.017to0.198 m Es 2 1 m 2 2 ), intermediateintensity(from0.33to0.4125 m Es 2 1 m 2 2 ),and highintensity(above2.607 m Es 2 1 m 2 2 ).AsshowninFigure4, algalstrainsaregroupedintwodistinctclusters:groupA(eight strains),whichincludesstrai nsfoundonlyatlowhumidity; Aphanotecae sp., Chroococcuslithophilus,Gloeocapsa sp., Gloeocystis sp., andScenedesmusarcuatus sp.aregroupedin thesamesubgroup,sincetheyhavebeenfoundinsampling siteswithbothmediumandhighlightintensities. Pseudococcomyasimplex and Cyanobiumbacillare appeardistantly relatedtothissubgroup,asthesetaxacolonizelowintensity sitesandhighintensitysites,r espectively.Asecondcluster (groupB)includestwosubgroups .InsubgroupB1,allspecies arenotaffectedbyhumidity; Chroococcusvarius Navicula sp., Chlorella sp., Stichococcus sp.and Scotiellopsisterrestis are ubiquitousstrains,while Oscillatoria sp., Synechococcus sp., Cyanidium sp.,and Phragnonemasordidum havebeendetected onlyinsamplingsiteswithlowtomediumlightintensities.A thirdgrouprepresentingSubgroupB2includes Leptolyngbya fragilis L.foveolarum Lyngbya sp.,and Nostoccommune ; thesestrainsweredetectedonlyinPM,thussuggestingthat theyareprobablydependentonhighhumidity. TheSybilÂ’sCaveandthePiscinaMirabilissharesome importantphysico-chemicalcharacters:location,anthropogenicorigin,Mediterraneanclimate,andnatureofthe substrate.Microbialcolonizationofrocksubstratesishighly dependentontheirfeatures(Macedoetal.,2009),andinthe caseofSCandPM,surfaceroughness,porosity,andmineral compositionoftherockpromoteinthesamewaythe establishmentofmicrobialassemblages.Temperature,irradiation,andrelativehumidityarerelevantdifferences betweenSCandPM.Intheinnerpartoftheformer,light intensityisveryreduced( 0.01 m Es 2 1 m 2 2 ),buttemperature andrelativehumidityarecomparabletoopen-airvalues. SimilarvaluesofrelativehumiditywerereportedbyHoffman (1989)foranotheranthropogeniccave,theRomannecropolis inSeville,Spain.SCcannotbeconsideredatypicalcave environment,butratherapoorlyilluminatedaerialhabitat, wheretemperatureandrelativehumidityareheavilyinfluencedbytheexternalclimaticconditionsduetothepresence oflargeopeningintherockwallsofthecave.Ontheother hand,PMisadimenvironmentwithamildtemperatureand arelativehumiditycomparabletothatrecordedfrom numerousnaturalcaves,whosevaluesrangefrom76to 96%(Coute andYe pre mian,2002). InthedarkareaofSC,Cyanophytesarethedominant organisms,beingrepresentedby Synechococcus sp.and Cyanobiumbacillare and Oscillatoria sp. Oscillatoria has beenfrequentlyencounteredinEuropeancaves(Coute and Ye pre mian,2002).Itisapioneerorganismbecauseofits capacitytogrowdiazotrophically(Gallonetal.,1991). Moreover,asothermembersofOscillatoriales,itiswell adaptedtoextremelylowirradiancecomparedtoother filamentousCyanophyta(Albertanoetal.,2000).The presenceof Synechococcus andofthecloselyrelated Cyanobiumbacillare insamplingpoints1and2ofSC confirmsthattheaveragetemperatureofSCduringthe yearisnotparticularlylow,sincethisspeciesdoesnot thriveincoldenvironments(SakamotoandBryant,1999). Synechococcus isnotatypicalcaveorganism:itsoccurrencehasbeenreportedonbothmarbleandgranite monumentsindifferentMediterraneancountries(Crispim andGaylarde,2005),anditalsocangrowendolithically (Saiz-Jimenezetal.,1990). Chroococcuslithophilus and C. various ,togetherwith Gloeocapsa sp.,arethedominant CyanobacteriainthewholeSybilCave,wherelightintensityisgreaterthan0.01 m Es 2 1 m 2 2 .Theincreasein Chroococcaceanspeciesatlowlevelsofirradiationseems tobeastandardfeatureofcavealgalassemblages(Asencio andAboal,2001),eventhoughtheoccurrenceof Chroococcus and Gloeocapsa hasbeenreportednotonlyindim habitats,butalsoonmonumentsexposedtodaylight (Scheereretal.,2009),confirmingthetoleranceofthese organismstoawiderangeofenvironmentalconditions. ThealgalcommunitiesofPMcanbesplitintwo types:thefirst,foundinpoints1through3,ismainly composedbyfilament ousCyanobacteria. Nostoccommune isthedominantspecies,responsibleforthe formationofthickmatscontainingalso Leptolynbya fragilis L.foveolarum ,and Lyngbya sp.filaments,along withChroococcaceanspeciesas Synechococcus sp.and Cyanobiumbacillare .Thedominantpresenceoffilamentouscyanobacteriainstableconditionsoflowlight intensityandhighrelative humidityhasbeenreported fordifferentcaves(MartinezandAsencio,2010;Rolda `n andHerna ndez-Marine ,2009). Nostoc isacosmopolitan terrestrialgenusthatcane nduredesiccation,aswellas verylowtemperatures(Dodds etal.,1995).However,the persistenceof Nostoccommune seemstobedetermined Figure3.Scanningelectronmicrograph(SEM)showing microbialcommunityembeddedinanesopolysaccharidic matrix(arrow). P.C ENNAMO ,C.M ARZANO ,C.C INIGLIA ,G.P INTO ,P.C APPELLETTI ,P.C APUTO AND A.P OLLIO JournalofCaveandKarstStudies, December2012 N 247

PAGE 16

bytheoccurrenceofliqui dwater(Stal,2000).This featurecouldaccountforthesh arpdifferenceobserved incommunitycompositionbetweentheSybilÂ’sCaveand thePiscinaMirabilis;inthefirst,adryhabitat, Nostoc commune wasnotrecoveredatanylightintensity,whereasinPM,anintensegrowthofthisCyanophytawas observedwherelightintensitywaslow. Chlorophytaanddiatomsfoundinpoints4to6ofboth SCandPMaretypicalmembersofalgalaerialcommunities.Chlorophytagenerasuchas Chlorella Stichococcus and Scotiellopsis anddiatomgenerasuchas Navicula ,and Pinnularia areubiquitousandwell-adaptedtocolonizea widerangeofsubstrata,regardlessofmicroclimaticand environmentalconditions(Macedoetal.,2009). Pinnularia obscura hasbeendetectedinthermoacidicenvironmentsof CampiFlegrei(Cinigliaetal.,2007),anditisprobably widelydistributedinthewholedistrict. TheclusteranalysisofalgaldistributionintheSybilÂ’s CaveandthePiscinaMirabilisasrelatedtolightand humidityshowedthatthemostrelevantdifferencesinthe algaldistributionswasthedetectionoffilamentousCyanobacteriaexclusivelyinthePiscinaMirabilis.Thisgroupof organismsseemstopreferhighandconstantvaluesof relativehumiditytogetherwithlowlightintensities,which werenotrecordedintheSybilÂ’sCave. Alsonoteworthyisthepresenceinbothcavesoftwo rareunicellularredalgae, Cyanidium sp.and Phragnonemasordidum .Theoccurrenceof P.sordidum inSC wasreportedforthefirsttimebySieminska(1962),who recoveredthealgaonashadyanddampwallofthis Table2.ListofalgaltaxaatsamplingsitesintheSybilÂ’sCave(SC)andinthePiscinaMirabilis(PM). Species SamplingSites SSC2SC3SC4SC5SC6PM1PM2PM3PM4PM5PM6 PROKARYOTA CYANOPHYTA Aphanotecae sp. ????????? XXX ?????????????????? Chroococcuslithophilus Ercegovic ?????? XXXX ?????????????????? Chroococcusvarius A.BrauninRabehorst ?????? XXXX ????????? XXX Gloeocapsa sp. ?????? XXXX ?????????????????? Leptolyngbyafoveolarum (Rabenhorstex Gomont)Anagnostidisetkoma rek ?????????????????? XX ???????????? Leptolyngbyafragilis (Gomont) Anagnostidisetkoma rek ?????????????????? XX ???????????? Lyngbya sp. ?????????????????? XX ??? X ?????? Nostoccommune VaucherexBornet& Flahault ???????????????????????? X ????????? Oscillatoria sp.XX ???????????? X ??? X ????????? Synechococcus sp.XX ???????????? XXX ????????? SynechococcusbacillarisCyanobium bacillare (Butcher)J.Koma rek, J.Kopeck&V.Cepa `K XX ?????????????????????????????? EUKARYOTA BACILLARIOPHYTA Navicula sp. ????????? XXX ????????? XXX Pinnulariaobscura Krasske ????????? XX ?????????????????? ??? CHLOROPHYTA Chlorella sp.Beijerinck ???????????? XX ?????? X ??? XX Gloeocystis sp. ????????? XXX ?????????????????? Stichococcus sp. ????????? XXX ???????????? XX Scenedesmusarcuatus Lemmrmanm ???????????? XX ?????????????????? Scotiellopsisterrestris (H.Reislig) Pucocharova&Kalina ??????????????? XX ????????? X ??? Pseudococcomyasimplex (Mainx)Fott ??????????????? X ?????????????????? RHODOPHYTA Cyanidium sp.X ??? X ????????? XX ?????? X ??? Phragnonemasordidum ZopfX ??? X ????????? X ?????? X ?????? A SURVEYOFTHEALGALFLORAOFANTHROPOGENICCAVESOF C AMPI F LEGREI (N APLES ,I TALY ) ARCHEOLOGICALDISTRICT 248 N JournalofCaveandKarstStudies, December2012

PAGE 17

cave.InSC, P.sordidum isfrequentlyassociatedwith Synechococcus ,anditoccursinallthestagesofitslife cycle.Thepresenceof P.sordidum hasbeenrecordedin othercavesintheMediterraneanregion(Friedmann, 1956),aswellasonmonumentsintheDeuxSe `vres district,France,underlowlightintensity(LeClercet al.,1983)andinhypogaeanRomanarcheologicalsites (Albertano,1993). Thegenus Cyanidium encompassesunicellularalgaerecentlyplacedinthephylumCyanidiophyta(Saundersand Hommerstand,2004)among theRhodophyta.Thisgenus includesboththermoacidophilicorganisms,suchasthewell known C.caldarium ,ubiquitouslypresen tinthermoacidicenvironments(Pintoetal.,2007),aswellasmesophilicones,as reportedbySchwabe(1936,1944)onthewallsofChilean coastalcaveswithnowaterseepage.AccordingtoOttand Seckbach(1994),themesophilicstrainisclearlydifferentfrom C.caldarium ,onthebasisofitsmesophiliccharacterandits habitat,butitsgeographicdistr ibutionispresentlynotknown. SincethestudiesofSchwabe(1936,1944),cave Cyanidium taxa havebeenfoundinmanyothersi tes,fromtheNegevDesert, Israel(Friedmann,1956)toFranceandItaly(Hoffmann1986, Skuja,1970). C ONCLUSIONS TheCampiFlegreidistrictpresentsmanyother hypogeanmonumentsthathavenotyetbeenexplored fortheiralgalcommunities.Studiesareinprogressto extendourknowledgeofthealgalbiodiversityoccurringin thesepeculiarhabitats. R EFERENCES Albertano,P.,1993,Epilithicalgalcommunitiesinhypogeanenvironments:GiornaleBotanicoItaliano,v.127,no.3,p.386–392. Albertano,P.,Bruno,L.,D’Ottavi,D.,Moscone,D.,andPalleschi,G., 2000,EffectofphotosynthesisonpHvariationincyanobacterial biofilmsfromRomancatacombs:JournalofAppliedPhycology, v.12,p.279–384.doi:10.1023/A:1008149529914. Albertano,P.,2003,CATS—Cyanobacteriaattackrocks:controland preventivestrategiesforavoidin gdamagecausedbycyanobacteriaand associatedmicro-organismsin Romanhypogeanmonuments, in Kozlowski,R.,etal.,eds.,Proceedingsofthe5thECConferenceon CulturalHeritageResearch:aPanEuropeanChallenge,Cracow,EUICSC,394p. Albertano,P.,Moscone,D.,Palleschi,G.,Hermosin,B.,Saiz-Jimenez,C., Sanchez-Moral,S.,Hernandez-Marine,M.,Urzi,C.,Groth,I., Schroeckh,V.,Saarela,M.,Mattila-Sandholm,T.,Gallon,J.R., Graziottin,F.,Bisconti,F.,andGiuliani,R.,2003,Cyanobacteria attackrocks(CATS):Controlandpreventivestrategiestoavoid damagecausedbycyanobacteriaandassociatedmicroorganismsin Romanhypogeanmonuments, in Saiz-Jimenez,C.,ed.,Molecular BiologyandCulturalHeritage:Lisse,Netherlands,Zwets&Zeitlinger, p.151–162. Asencio,A.D.,andAboal,M.,2001,Biodeteriorationofwallpaintingsin cavesofMurcia(SESpain)byepilithicandchasmoendolithic microalgae:AlgologicalStudies/Archivfu ¨rHydrobiologie,SupplementVolumesNo.103,p.131–142. Bish,D.L.,andPost,J.E.,1993,Quantitativemineralogicalanalysisusing theRietveldfull-patternfittingmethod:AmericanMineralogist,v.78, p.932–940. Bischoff,H.W.,andBold,H.C.,1963,SomesoilalgaefromEnchanted Rockandrelatedalgalspecies:Austin,UniversityofTexasPublication no.6318,PhycologicalStudiesno.4,95p. Castenholz,W.,1988,Culturin gmethodsforCyanobacteria, in Parker,L.,and Glazer,A.N.,eds.,Cyanobacteria :Elsevier,MethodsinEnzymology, no.167,p.68–93.doi:10.1016/0076-6879(88)67006-6. Chelius,M.K.,Beresford,G.,Horton,H.,Quirk,M.,Selby,G.,Simpson, R.T.,Horrocks,R.,andMoore,J.C.,2009,Impactsofalterationsof organicinputsonthebacterialcommunitywithinthesedimentsof WindCave,SouthDakota,USA:InternationalJournalofSpeleology, v.38,no.1,p.1–10. Ciniglia,C.,Cennamo,P., DeStefano,M.,Pinto,G.,Caputo,P.,andPollio, A.,2007, Pinnulariaobscura Krasske(Bacillariophy ceae,Bacillariophyta) fromacidicenvironments:characterizationandcomparisonwithother acid-tolerant Pinnularia species:FundamentalandAppliedLimnology: Archivfu ¨rHydrobiologie,v.170,p.29–47.doi:10.1127/1863-9135/2007/ 0170-0029. Coute ,A.,andYe pre mian,C.,2002,L’hommeetlesalguesdescavernes, in M.F.Roquebert,M.F.,ed.,Lescontaminantsbiologiquesdesbiens culturels,Paris,Elsevier,CollectionPatrimoine,p.33–55. Crispim,C.A.,andGaylarde,C.C.,2005,Cyanobacteriaandbiodeteriorationofculturalheritage:areview.MicrobialEcology,v.49,p.1–9. doi:10.1007/s00248-003-1052-5. Dayner,D.M.,andJohansen,J.R.,1991,Observationsonthealgalflora ofSenecaCaverns,SenecaCounty,Ohio:OhioJournalofScience, v.91,p.118–121. de’Gennaro,M.,Cappelletti,P.,Langella,A.,Perrotta,A.,andScarpati, C.,2000,GenesisofzeolitesintheNeapolitanYellowTuff:geological, volcanologicalandmineralogicalevidence:ContributionstoMineralogyandPetrolology,v.139,p.17–35.doi:10.1007/s004100050571. D ez,B.,Pedro s-Alio ,C.,andMassana,R.,2001,Studyofgenetic diversityofeukaryoticpicoplanktonindifferentoceanicregionsby small-subunitrRNAgenecloningandsequencing:Appliedand EnvironmentalMicrobiology,v.67,p.2932–2941.doi:10.1128/ AEM.67.7.2932-2941.2001. Dobat,K.,1970,Conside rationssurlave ge tationcryptogamiquedesgrottes duJuraSouabe(sud-ouestdel’Allemagne):AnnalesdeSpe le ologie, v.25,no.4,p.872–907. Figure4.Hierarchicalclusteringofthealgalstrainsbased onlightandhumidity.Themajorclustersdescribedinthe textare,fromthebottomup,A,B1,andB2. P.C ENNAMO ,C.M ARZANO ,C.C INIGLIA ,G.P INTO ,P.C APPELLETTI ,P.C APUTO AND A.P OLLIO JournalofCaveandKarstStudies, December2012 N 249

PAGE 18

Dodds,W.K.,Gudder,D.A.,andMollenhauer,D.,1995,Theecologyof Nostoc :JournalofPhycology,v.31,p.2–18.doi:10.1111/j.00223646.1995.00002.x. Doyle,J.J.,andDoyle,J.L.,1990,IsolationofplantDNAfromfresh tissue:Focus,v.12,p.13–15. Friedmann,I.,1956,Beitra ¨gezuMorphologieundFormwechselder atmophytischenBangioidee Phragmonemasordidum :Zopf.O ¨ sterreichischebotanischeZeitschrift,v.103,p.613–633. Gallon,J.R.,Hashemm,M.A.,andChaplina,A.E.,1991,Nitrogenfixation by Oscillatoria spp.underautotrophicandphotoheterotrophicconditions:JournalofGeneralMicrobiology,v.137,p.31–39.doi:10.1099/ 00221287-137-1-31. Hoffmann,L.,1986,Cyanophyce esae riennesetsubae riennesduGrandDuche deLuxembourg:BulletinduJardinBotaniqueNationalde Belgique,v.56,p.77–127. Hoffmann,L.,1989,Algaeofterrestrialhabitats:BotanicalReview,v.55, p.77–105.doi:10.1007/BF02858529. Huss,V.A.R.,Frank,C.,Hartmann,E.C.,Hirmer,M.,Kloboucek,A., Seidel,B.M.,Wenzeler,P.,andKessler,E.,1999,Biochemicaltaxonomy andmolecularphylogenyofthegenus Chlorella sensulato(Chlorophyta):JournalofPhycology,v.35,p.587–598.doi:10.1046/j.15298817.1999.3530587.x. Insinga,D.,Calvert,A.,D’Argenio,B.,Fedele,L.,Lanphere,M.,Morra, V.,Perrotta,A.,Sacchi,M.,andScarpati,C.,2004, 40 Ar/ 39 Ardating oftheNeapolitanYellowTufferuption(CampiFlegrei,southern Italy):volcanologicalandchronostratigraphicimplications.European GeophysicalUnion1stGeneralAssembly,Nice,2004(abstract). Lawton,L.,Marsalek,B.,Padisa k,J.,andChorus,I.,1999,DeterminationofCyanobacteriainthelaboratory, in Chorus,I.,andBartram, J.,eds.,ToxicCyanobacteriainWater:AGuidetoTheirPublic HealthConsequences,MonitoringandManagement,London,E& FNSponPress,p.347–368. LeClerc,J.C.,Coute ,A.,andDupuy,P.,1983,Leclimatannueldedeux grottesetd’unee gliseduPoitou,ouviventdescoloniespuresd’algues sciaphiles:CryptogamieAlgologie,v.4,p.1–19. Macedo,M.F.,Miller,A.Z.,Dion sio,A.,andSaiz-Jimenez,C.,2009, Biodiversityofcyanobacteriaandg reenalgaeonmonumentsinthe MediterraneanBasin:anoverview:Microbiology,v.155,p.3476–3490. doi:10.1099/mic.0.032508-0. Mart nez,A.,andAsencio,A.D.,2010,Distributionofcyanobacteriaat theGeladaCave(Spain)byphysicalparameters:JournalofCaveand KarstStudies,v.72,no.1,p.11–20.doi:10.4311/jcks2009lsc0082. Morra,V.,Calcaterra,D.,Cappelletti,P.,Colella,A.,Fedele,L., de’Gennaro,R.,Langella,A.,Mercurio,M.,andde’Gennaro,M., 2010,Urbangeology:relationshipsbetweengeologicalsettingand architecturalheritageoftheNeapolitanarea, in Beltrando,M., Peccerillo,A.,Mattei,M.,Conticelli,S.,andDoglioni,C.,eds.,The GeologyofItaly:tectonicsandlifealongplatemargins,Journalofthe VirtualExplorer,v.36,paper27.doi:10.3809/jvirtex.2010.00261. Mulec,J.,Kosi,G.,andVrhovs ek,D.,2007,Algaepromotegrowthof stalagmitesandstalactitesinkarstcaves(S kocjanskejame,Slovenia): CarbonatesandEvaporites,v.22,no.1,p.6–10.doi:10.1007/ BF03175841. Mulec,J.,Kosi,G.,andVrhovs ek,D.,2008,Characterizationofcave aerophyticalgalcommunitiesandeffectsofirradiancelevelson productionofpigments:JournalofCaveandKarstStudies,v.70, no.1,p.3–12. Orsi,G.,DeVita,S.,anddiVito,M.,1996,Therestless,resurgentCampi Flegreinestedcaldera(Italy):constraintsonitsevolutionand configuration:JournalofVolcanologyandGeothermalResearch, v.74,p.179–214.doi:10.1016/S0377-0273(96)00063-7. Ott,F.D.,andSeckbach,J.,1994,Newclassificationforthegenus Cyanidium Geitler1933, in Seckbach,J.,ed.,EvolutionaryPathways andEnigmaticAlgae: Cyanidiumcaldarium (Rhodophyta)and RelatedCells,NewYork,Springer,DevelopmentsinHydrobiology, v.91,p.145–152. Pagano,M.,Redd,M.,andRoddaz,J.-M.,1982.CampiFlegreiMarsilio Editore. Pedersen,K.,2000,Explorationofdeepintraterrestrialmicrobiallife: currentperspective:FEMSMicrobiologyLetters,v.185,p.9–16. doi:10.1111/j.1574-6968.2000.tb09033.x. Pinto,G.,Ciniglia,C.,Cascone,C.,andPollio,A.,2007,Species compositionofCyanidialesassemblagesinPisciarelli(Campi Flegrei,Italy)anddescriptionof Galdieriaphlegrea sp.nov., in Seckbach,J.,ed.,AlgaeandCyanobacteriainExtremeEnvironments,Dortrecht,Netherlands,Springer,CellularOrigin,Lifein ExtremeHabitats,andAstrobiology,v.11,p.487–502.doi:10.1007/ 978-1-4020-6112-7_26. Podani,J.,2001,SYN-TAX2000ComputerProgramfordataanalysisin EcologyandSystematics:User’sManual,Budapest,Scientia,53p. Rajola,T.,andN.Fiorillo in Paoli,P.A.,1978,Table60Naples,Biblioteca NazionaleVittorioEmanueleIII,Roma. Rolda n,M.,andHerna ndez-Marine ,M.,2009,Exploringthesecretsof thethree-dimensionalarchitectureofphototrophicbiofilmsincaves. InternationalJournalofSpeleology,v.38,no.1,p.41–53. Saiz-Jimenez,C.,Garcia-Rowe,J.,GarciaDelCura,M.A.,Ortega-Calvo, J.J.,Roekens,E.,andVanGrieken,R.,1990,EndolithiccyanobacteriainMaastrichtlimestone:ScienceoftheTotalEnvironment,v.94, no.3,p.209–220.doi:10.1016/0048-9697(90)90171-P. Sakamoto,T.,andBryant,D.A.,1999,Nitratetransportandnot photoinhibitionlimitsgrowthofthefreshwaterCyanobacterium Synechococcus speciesPCC6301atlowtemperature:PlantPhysiology,v.119,p.785–794.doi:10.1104/pp.119.2.785. Saunders,G.W.,andHommersand,M.H.,2004,Assessingredalgal supraordinaldiversityandtaxonomyinthecontextofcontemporary systematicdata:AmericanJournalofBotany,v.91,no.10, p.1494–1507.doi:10.3732/ajb.91.10.1494. Scarpati,C.,Cole,P.,andPerrotta,A.,1993,TheNeapolitanYellow Tuff—AlargevolumemultiphaseeruptionfromCampiFlegrei, SouthernItaly:BulletinofVolcanology,v.55,p.343–356. doi:10.1007/BF00301145. Scheerer,S.,Ortega-Morales,O.,andGaylarde,C.,2009,Microbial deteriorationinstonemonuments—anupdatedoverview, in Laskin,A.,Gadd,G.,andSariaslani,S.,eds.,Advancesin AppliedMicrobiology,Volume 66:SanDiego,AcademicPress, p.97–130. Schwabe,G.H.,1936,U ¨ bereinigeBlaualgenausdemmittlerenund su ¨dlichenChile:VerhandlungendesDeutschenWissenschaftlichen VereinszuSantiagodeChile,v.3,p.113–174. Schwabe,G.H.,1944,UmraumfremdeQuellen,Shanghai,M.No ¨ssler, MitteilungenderDeutschenGesellschaftfu ¨rNatur-undVo ¨lkerkunde Ostasienssupplement21,300p. Sieminska,J.,1962,Theredalga Phragmonemasordidum intheSibylcave nearbyNaples:ActaHydrobiologie,v.4,no.2,p.225–227. Skuja,H.,1970,Alghecavernicolenellezoneilluminatedellegrottedi Castellana(MurgediBari):LeGrotted’Italia,Ser.4,v.2,p.193–202. Stal,L.J.,2000,Cyanobacterialmatsandstromatolites, in Whitton,B., andPotts,M.,eds.,EcologyoftheCyanobacteria:TheirDiversityin SpaceandTime,Dortrecht,Netherlands,Kluwer,p.61–120. A SURVEYOFTHEALGALFLORAOFANTHROPOGENICCAVESOF C AMPI F LEGREI (N APLES ,I TALY ) ARCHEOLOGICALDISTRICT 250 N JournalofCaveandKarstStudies, December2012

PAGE 19

ASSESSMENTOFSPATIALPROPERTIESOFKARST AREASONAREGIONALSCALEUSINGGISAND STATISTICS–THECASEOFSLOVENIA M ARKO K OMACAND J ANKO U RBANC GeologicalSurveyofSlovenia,Dimic evaul.14,SI1000Ljubljana,Slovenia,marko.komac@geo-zs.si Abstract: InSlovenia,43%oftheterritoryiskarst,including42%ofallprotectedwater sourcesand53%ofallwater-protectionareasinthecountry.Over95%ofdrinkingwateris obtainedfromgroundwater,soassessmentofkarstareasandtheirspatialdistributionis essentialtobetterunderstandthewaterinthelithosphereandfortheassessmentofthe hydrogeochemicalpropertiesofthegroundwaterinalargepartofSlovenia.These groundwaterresourcesaresusceptibletodegradationorpollution,andaregional karstification-intensitymapwasdevelopedtoassistinthemanagementofthesewater resourcesandtointerpretthechemicalcompositionofthegroundwater.Forthepurpose ofclassifyingstratigraphicunitsintokarstification-levelclasses,threeparameterswere analyzedintheoutcropsofunitswithcarbonatecontentusingGISandsimplespatial statistics:thepresencesofsinksandcaveentrancesandtheabsenceofasurficialdrainage network.Whereatleasttwoofthethreeparametersshowedapositiverelationwith karstification,theunitwasregardedasintenselykarstified,whiletherestwereregardedas lesskarstified.Theformerareascover24%andthelatter21%ofSlovenianterritory. I NTRODUCTION Duetoitsnaturalfeatures,Sloveniaisacradleofkarst research.AccordingtoGams(2003),43%ofSlovenian territoryiskarst.Figure1(generalizedafterGams,2003) showsthegeneraldistributionofkarsttypesinSlovenia. Karstisaterrain,generallyunderlainbylimestoneor dolomite,inwhichthetopographyischieflyformedbythe dissolutionofrockandthatmaybecharacterizedby sinkholes,sinkingstreams,closeddepressions,subterraneandrainage,andcaves(Monroe,1970). Karstareashaveveryspecifichydrogeologicalproperties, andhence,akarstmapofhighqualityisessentialforthe purposeofbetterunderstandingtheflowandchemical compositionofwaterinthealmosthalfofSlovenia’sterritory thatconsistsofcarbonaterocks.Forthepurposeofstudying thecharacteristicsofSloven ianaquifersinalaterstage,a betterandnewermapofkarstareasonageneralscaleis needed.Suchamapwillalsobeveryusefulforassessing groundwatervulnerabilityandsitingofecologicallyproblematicprojects,buttheintention ofthispaperisnottoaddress thevulnerabilityissue.Anassessmentoftheseisthenextstep, whichwillbedoneinthefuturewhenadditionalparameters willhavetobeconsidered,suchassoilcoverandthethickness ofthevadosezone.Theintentionofthispaperismerelyto assigndifferentintensitiesofkarstificationtoareasinSlovenia onthebasisofavailabledatabyusingGISmethods. Inthepast,therewereseveralattemptstomapkarst areasinSlovenia(amongothers,BlaeuandBlaeu,1645; Valvasor,1977;Gams,2003),butnonewerebasedupona statisticalapproachordirectlyonstratigraphiclevels.The recentlypublishedGeologicalmapofSloveniaatascaleof 1:250,000(Buser,2010)enabledanovelapproachtoassessing karst-proneareasforthewholeSlovenianterritoryatan acceptableresolution. Variousstudieshavedealtwiththeanalysisofkarst terrainsandkarstfeaturesusingGIS,remote-sensingtechniques,andfieldworktoderivelarge-areakarstmaps.Florea etal.(2002)usedsystematicmappingofkarstfeaturesand GIStoproduceakarstmapofanareaofKentucky.Veni (2002)basedhiskarst(andpseudokarst)mapoftheUnited Statesonlithology,ratherthanonkarstfeatures.Armstrong etal.(2003)analyzedtherelationbetweenhydraulicconductivityandkarstsurfacefeaturesandfoundagoodcorrelationbetweenthetwoinwesternFlorida.AlsoinFlorida, Denizman(2003)investigatedmorphometricparametersof karstsurfacefeaturesandconcludedthatGISprovidesgood toolsforkarststudies.Florea(2005)usedsinkholeinformationtoidentifyknownandunknownstructuralfeaturesin Kentucky.ForthesmallerareainKentucky,Tayloretal. (2005)derivedsinkholedatafromdigitalelevationmodels. Staffordetal.(2008)usedkarstfeaturesdatatodevelopa karst-potentialmapinevaporates.Theydiscoveredthatusing aGISapproachexclusivelyunderestimatestheactualextent anddevelopmentofkarstfeatures.GaoandZhou(2008) gaveathoroughoverviewofpastGISanddatabaseapplicationinkarstresearch. S TUDY A REA SlovenialiesincentralEurope,atthecontactofthe AlpinearchwiththeDinaricchainandPannonianbasin.Its surfacemeasuresslightlyover20,200km 2 (Fig.2).Elevations extendfromsealevelupto2864m,andthelandscapevaries fromPannonianplainstohillyslopesandAlpinemountains, M.KomacandJ.Urbanc–AssessmentofspatialpropertiesofkarstareasonaregionalscaleusingGISandstatistics–thecaseof Slovenia. JournalofCaveandKarstStudies, v.74,no.3,p.251–261.DOI:10.4311/2010ES0188R JournalofCaveandKarstStudies, December2012 N 251

PAGE 20

allwithareasofkarstmorphology.Duetoitsrichnessin carbonateformations,morethan40%ofSloveniaiskarstor semi-karst.AccordingtoKomac(2005)almost92%ofthe carbonatesareofMesozoicageinSlovenia,wherekarstis onlydevelopedoncarbonaterocks.Oftheterritoryof Slovenia,49.25%iscomposedofclasticrocks,carbonate rockscover39.31%,andamixtureofthetwocovers4.27%of thearea.Theleastareaisoccupiedbyigneousrocks(1.49%), followedbypyroclasticrocks(1.78%)andmetamorphic rocks(3.9%).InSlovenia,almostallkarstphenomenaarethe productofhypergenickarstprocesses,andthereisvirtually nohypogenickarst. D ATAAND M ETHODS U SED Theanalysispotentiallycoveredtheareaofthewhole country,andareasonablescaleforthedatawasselected.The digitalGeologicalMapofSloveniaatascaleof1:250,000 (Buser,2010)inESRIshapeformatwasusedasabasisfor stratigraphicinforma tion.Lithologiesofthestrataandthicknessesmentionedinthetextwe reobtainedfromthebooklets accompanyingthegeologicalmapsatascaleof1:100,000.A digitalelevationmodelwitha cellsizeof25meters(GURS, 2005)wasusedformorphologican alyses,particularlyforsink occurrence.ThechosenDEMisthebestavailablerawand unbiasedinformationonSlovenia’ssurface,and,duetothe scaleandcoarsepositionalprec isionofthelithologicdata,a 25mDEMwasdetailedenough.Weareawarethatmore preciseresultscouldbeachievedwithmoredetaileddata,but forthepurposeoftheproject,thechosenscaleandresolution weresatisfactory.Drainagenetworkdataatascaleat1:50,000 wereobtainedfromtheEnvironmentalAgency(ARSO,2005) andusedtoassesstheabundanc eofsurfacewaterflows.A highlydetailedcatalogofcav eentrancewascollectedbythe SpeleologicalAssociationofSlo venia(JZS,2010).Allanalyses wereconductedwithESRIArcGISsoftware. Theapproachwechosewasbasedonsurfacekarstfeatures.Becauseoftheresolutionandotherlimitationsinthe inputdatasets,acombinationofthreefactors,cave,sink, anddrainage-networkdensities,waschosenasameasureof karstificationintensity.Themethodpresentedinthispaper isobviouslyapplicableonlyfortheassessmentofhypergenic karst,whileforthepurposeofassessmentofhypogenickarst itsuseisverylimited. Beforetheanalyses,allareasselectedforthestudywere classifiedbyanexperiencedhydrogeologistwithlong-term experienceinkarsthydrologyasatleastpotentiallykarstified areas.Theanalysiscoveredtheareaofthewholecountry,and areasonablescaleforthedatawasselected. PriortoanalysesperformedinGISona25mcellbasis, datausedintheanalyseshadtobeselected.Giventhefact thatkarstcanonlydeveloponsolublerocks(Monroe, 1970),stratigraphicformationswereusedasthebasisforthe analysis.ForthispurposeweextractedfromtheGeological Figure1.KarstregionsinSlovenia,generalizedafterGams(2003). A SSESSMENTOFSPATIALPROPERTIESOFKARSTAREASONAREGIONALSCALEUSING GIS ANDSTATISTICS – THECASEOF S LOVENIA 252 N JournalofCaveandKarstStudies, December2012

PAGE 21

mapofSloveniaallstratigraphicformationscontainingthe wordslimestone,calcite,ordolomiteintheirlithologic descriptionstogetasubgroupofunits(andhenceoutcrop areas)thatcouldbepotentialcarbonate-karstareas.Outof 114units,47wereselected,representedby2048polygons covering44.7%ofSlovenia.Purelynon-carbonateunitswere ignoredinourstudy,thoughseveraloftheselectedunits containbothcarbonateandnon-carbonaterocks,mainly clasticrocks.Weexpectedthatthissimpleselectionwould includesomeformationsthatarelesssusceptibletokarstification,asnotallwerepurecarbonates.Infact,insome formations,carbonatesareaminorlithologyoronlyoccurin smallareasandlimitedthickness,andthosemaynothost frequentorwell-developedkarstfeatures.InSlovenia,dolomites,whicharebydefinitionpurecarbonates,aregenerally notkarstified,butratherformfracturedaquiferswithcompletelydifferenthydrogeologicalcharacteristicsthanlimestoneareas.Weacknowledgeth esepotentialsourcesoferror andalsopointoutthat,giventhecoarsespatialresolutionof theDEMandgeneralizationofthestratigraphicmap,some smallareasofcarbonatesthatdohavekarstdevelopmentmay notberepresentedinouranalysis.Karstfields(poljes)were notincludedinthisselectionsincetheyarestratigraphically classifiedasQuaternarysedime nts.Stratigraphicunitswere classifiedasoneoffourlithol ogicclasses,clasticrocks, dolomites,limestones,andmixedlimestonesanddolomites. ThechosenDEMisthebestavailablerawandunbiased informationonSlovenia’ssurface,and,duetothescale andcoarsepositionalprecisionofthelithologicdata,a 25mDEMwasdetailedenough.Weareawarethatmore preciseresultscouldbeachievedwithmoredetaileddata, butforthepurposeoftheproject,thechosenscaleand resolutionweresatisfactory. Theoccurrenceofsinksindicatesthepresenceofkarst processes.Numbersofsinkholeswerederivedforthewhole studyareafromthe25mDEMusingtheSinkfunctionin ArcMapsoftware.Obviouslynotallfeaturesderivedwere karstsinks,becauseaproportionofthem(andthemost obviousones)occurredinflatalluvialbasinsorinthe vicinityofrivers.Afterafield-checkatrepresentativelocations,wedecidedtoexclude,inadditiontonon-carbonate outcrops,areasonalluvialplainswithina75m(3cells) bufferzonenearrivers.Ourfieldworkexperiencehasshown thatinthevicinityofriversnosinksoccur,andthosederived therefromaDEMareartifacts,mostprobablyrelatedto smallrelicchannelsorditches.Thebufferzonewidthwas definedbytrial,becauseinmanyareas—mainlyvalleys— thealluvialsedimentsareshowndisproportionallynarrow Figure2.MapofSloveniawithoutcropsofselectedunitsshowningray.LocationofSloveniainEuropeisshowninthe lowerright. M.K OMACAND J.U RBANC JournalofCaveandKarstStudies, December2012 N 253

PAGE 22

duetothescaleofthegeologicalmapandtheirbeingofless importanceonthegeologicalmapthanbedrock,especially becausethereisnosurfacewaterflowinvalleysinkarst areas.Altogethertherewere129,485acceptedsinkoccurrencesintheanalyzedareas.Densityvariedfrom0to43.7 sinks/km 2 ,withameanof10.6sinks/km 2 andastandard deviationof11.2.Weacknowledgethatduetotheautomaticderivationofthesinkdatasomebiasintheformof spurioussinksstillexists,especiallyinflatterrainwhere alluvialsedimentspartiallycomposedofcarbonaterocks wereincludedintheanalysis.Hencethesink-indexvaluesof theselithologicunitsneedtobeinterpretedwithcaution. Theoccurrenceofcavesisobviouslythemostdefinitive proofthatanareacanbeclassifiedaskarst.Atthesame time,thecavedatamaybebiased,sincethedatawere collectedbyincidentaldiscoveriesandgenerallynotbya systematicsearchoftheterrain.Despiteitslimitations,the cavecatalogistheonlyavailableresourceofthistype coveringallofSloveniathatcouldbeusedinGISanalysis, soitwasusedasoneofthebasesfortheevaluationof karstificationintensity.Thecavecatalogincludesrecordsof entrancestovarioustypesandstagesofkarstcaves,fully developedhorizontalorsub-horizontalcaves,verticalcaves, inactivecaves,freezingoricecaves,androck-shelters.The informationregardingcaveoccurrencewastheonlydataset thatwasusedinarawformfortheanalyses.Therewere 7554caveentrancesmappedinthestudyarea.Cavedensity valuesvariedfrom0to2.1caves/km 2 ,withameanof 0.6cave/km 2 andastandarddeviationof0.5. Theabsenceofasurfacedrainagenetworkonagiven formationcanalsoindicatethepresenceofkarstareaswhere subsurfacedrainageispredominant.Toavoidlosingthe informationonthesurfacedrainagenetworkdensitywhen performingraster-basedanalyses,wecalculatedthedensity ofthedrainagenetworkbysimplydividingthetotallength ofthisnetworkbytheareaofagivenstratigraphicunit’s outcrop.Thetotallengthofsurfacedrainagenetworkwith flowingwaterintheanalysisareawas6082.3km.The surficialdrainagenetworkdensitywasexpressedinkm/km 2 andvaluesvariedfrom0to2.5km/km 2 ,withameanof 0.95km/km 2 andastandarddeviationof0.75.Ourapproachwasbasedontheassumptionthatlowerdrainage networkdensitiesimplyhigherkarstificationintensities. Table1showstheextentofthestudiedfeatureswithin eachstratigraphicoutcropthatwasincluded,aswellasits area.Theaveragedensitiesoffeaturesintheoverallarea was0.83cavesperkm 2 ,14.3sinksperkm 2 ,and0.67kmof surfacedrainagenetworkperkm 2 Foraquantitativeassessmentoftheimportanceofthe presenceofcavesorsinksandtheabsenceofsurfacedrainage ineachstratigraphicunit,anon-parametricchi-squared( X 2 ) wascomputedas( O 2 E ) 2 / E foreachoftheobservedquantities,where O wastheobservedquantityand E wasthe expectedvalueofthatquantityintheareaoftheparticular unitbasedontheaverageoverallunits.Fromthis,anindex wascomputedbynegatingthe X 2 valueif( O 2 E )was negativeinthecasesofnumbersofcavesandsinksandifit waspositiveinthecaseofdrainagenetworks.(Amore extensivedrainagenetworkimplieslesskarstification.)Hence inallcases,anegativeindexindicatesthattheunitappears lesskarstifiedthanaveragebasedonthatparticularquantity, andapositiveindexindicatesthattheunitappearsmore karstified.Theresultingcaveindexes(CI),sinkindexes(SI), andsurfacedrainagenetworkindexes(SDNI)arelistedin Table2anddisplayedinFigure3. Foreachfeatureindexineachstratigraphicunit,a normalizedvaluewascomputed.Forexample, CI norm 5 ( CI 2 CI min )/( CI max 2 CI min ),where CI max and CI min arethe maximumandminimumvaluesofCIobservedforany unit.NormalizedvaluesforSIandSDNIwerecomputed similarly,andthethreenormalizedvaluesforeachunit wereaveragedtogiveasingleaverageofnormalizedindexes(ANI).TheANIsarealsolistedinTable2. Allcorrelationtestswereper formedusingPearsonproductmomentcorrelationcoefficient( r )formula,andvaluesrepresentlinearcorrelationcoefficients. R ESULTSAND D ISCUSSION G ENERAL R ESULTS Chi-squareanalysesshowedthatthethreephenomena weexpectedtoberelatedtokarstareaswerethecorrect choice.Forallteststhedegreesoffreedom(df)was46.In thecaseofcaveoccurrence, X 2 hadthescoreof3975.7and the p -value 0.001;inthecaseofsinkoccurrence, X 2 had thescoreof74070.2andthe p -value 0.001;andinthe caseofdensityofsurficialdrainagenetwork, X 2 hadthe scoreof4795.9andthe p -value 0.001. Foreachstratigraphicunit,thenumberofindexesthat weregreaterthan0(IIC)isshowninTable2.Anindicationofkarstificationintensity,wepresumedthatunitswith IIC $ 2areintenselykarstified,andthatotherunitsare lesskarstified.Inmostcases,theintenselykarstifiedareas haveANIsthatexceedthoseoftheareaslesskarstified.A thresholdofANI $ 0.4344coincidesratherwellwiththe simplifiedthresholdofatleasttwopositiveindexes,except fortheunits13and48,whichhaveANIvaluesabovethat thresholdbuthaveonlyonepositiveindex.Thehighvalue ofANI(0.602)forunit13istheconsequenceofveryhigh SIvalue,whiletheothertwoindexesarenegative.Despite thefactthatintheareaofunit13thenumberofsinksis high,theothertwoindexesindicatethattheunitshouldbe classifiedamongthelesskarstifiedareas.Thereasonfora highSIvalueisprobablyacombinationoferrorsofthe automaticallyderivedsinkdataandofthefactthatpartsof theexposureofunit13withwidelydifferingintensitiesof karstificationwerenotseparatedonthegeologicalmap. Unit48isaveryheterogeneousformation,composedof limestones,marls,andcoalandbauxitebedsthathas,like unit13,someintenselykarstifiedareas,whileingeneral, thekarstificationisverylimitedduetoitscomposition.So A SSESSMENTOFSPATIALPROPERTIESOFKARSTAREASONAREGIONALSCALEUSING GIS ANDSTATISTICS – THECASEOF S LOVENIA 254 N JournalofCaveandKarstStudies, December2012

PAGE 23

despitetherelativelylargevaluesofANI,theIICtest placesbothunitsinthelesskarstifiedclass. Analysesofthecorrelationvalues( r )betweenthethree setsofindexesshowthattheconnectionbetweencaveoccurrenceandsinkoccurrenceisnotverystrong(0.26),whilethe correlationbetweentheabsenceofsurficialdrainagenetwork andcave(0.58)orsink(0.66)occurrenceisstronger.The weakercorrelationvalueofthecaveoccurrencecouldbe explainedwiththefactthatcaveswerenotsystematically mapped,butwereregisteredbypurechance,andthatmost probably,therearestillsomecavestobediscovered.Also,the cavedataonlyrecordentrancesanddonotreflectthesizeof thecaves.Correlationsbetweenthetypesofindexesmight implythattheyarenotindependent;thisisinfactexpected forhypergenickarst.Althoughtherelationsbetweenthe studiedfeaturesmaynotapplyoutsidethestudyarea,for Sloveniathecorrelationbetweentheabsenceofsurfacedrainagenetworkandsinkorcaveoccurrenceisafactthatwas alsocheckedandprovedinthefield. Cavesareabsent(density 5 0.0km 2 2 )inthreeunits(34, 82,96),sinksinone(102),andsurficialdrainagenetworkin two(55,58).Thehighestdensityofcaves(2.1km 2 2 )was foundinunit45,thehighestdensityofsinks(43.7km 2 2 )in unit13,andthehighestdensityofsurfacedrainagenetwork (2.45km/km 2 )inunit83.Resultsforthedifferenttypesof indexvaluescanbediscussedinmoredetailwithreferenceto Table2andFigure3andadditionalinformation,not shown,aboutthelithologyofthestratigraphicunits. Thecompositionofunitswithhighlypositivevalueofthe caveindex(54,57,59,61,71,and76)indicatethattheseunits areeithercomposedof200to1500mofrelativelypure CaCO 3 (units54,57,76),orarecomposedofCaCO 3 and limitedimpurities,mainlybituminousdolomite(units59,61, 71).Forunitswheretheothertwoindexes,SIandSDNI, werenegative,lowCIvaluescorrespondwelltothelow karstificationintensity.Whentheothertwoindexesare positive,negativeCIvalues,whichfromthecave-occurrence perspectiveindicatelesskarstifiedareas,areprobablyresult ofrandomandsporadiccavediscovery.Inthecaseofhomogeneouscarbonatelithologicunits,itwouldbeexpectedthat evenonecaveoccurrencewouldindicatewell-developed karst,sothattheunitshouldbeclassifiedasintensely karstified.Sincealmostalllithologicunitsusedinthisstudy werenotpurecarbonatesbutratherheterogeneousformationsofclasticrocksandcarbonates,theassumptionthatone caveinthestratigraphcunitprovesthatthewholeareaofthat outcropisintenselykarstifiedisoversimplifiedandcouldbe wrong.Eveninhypergenickarst,basingthedistinction betweenintenselyandlesskarstifiedareasstrictlyonthebasis ofCIvaluesalonecouldbemisleadingbecauseofoverlooked caves.Thisiswhysupplementaryindexesofsinksandsurface drainagehavebeenused. Sinkindexvaluesforstratigraphicunitsareingeneral agreementwiththeirclassificationaccordingtokarstification intensity,exceptinthecaseoftheTriassicDachsteinlimestone(76).ExtremelylowSIvalueintheDachsteinlimestone Table1.Amountsofinventoriedfeaturesineachstratigraphicunitandtheirareas.SDNislengthofsurface drainagenetwork. No. a CavesSinksSDN,kmArea,km 2 13246627162.3151.6 276450290.5206.9 2919912155.3171.0 3403722.39.9 421016660.230.4 454451.71.9 46181254472.8188.0 4892158528.3112.0 502222.68.9 522640566.770.4 532422581.881.6 54186179318.2104.6 5582080.09.4 56910084.658.7 5711161188666.9584.3 5842940.017.6 59146328334107.51097.5 615431061426.2334.5 63136762.834.2 64277718222.1277.4 6574406020.1123.5 66195809916.7281.4 67111434931.0177.6 68153416.623.6 7034118.39.6 71333820038.0333.0 72324775085.0390.3 7333496.441.4 7442922.815.1 754695.715.2 7615046617386.1914.3 77283111771133.21096.8 7820131239.0150.1 8032447291.5180.6 8113638.525.7 82017.76.6 83227102.941.9 843872247643.3641.1 85302854.345.5 8611479110.4112.7 873229030.748.8 9167505458.3325.1 9285996806.7445.9 93135087.744.7 96072.71.1 97112320.519.2 102305.75.1 S 755412 9485 6082.39066.7 a Lithologicaldescriptionofunits/formationsisgiveninthesecondcolumnin Table2. M.K OMACAND J.U RBANC JournalofCaveandKarstStudies, December2012 N 255

PAGE 24

Table2.Shortstatementofthelithologyofeachstratigraphicunit,andresultsofthestatisticalanalysisofthedatainTable1. CI,caveindex;SI,sinkindex;SDNI,surfacedrainagenetworkindex;ANI,averagenormalizedindex;IIC,numberofindexes thataregreaterthan0,indicatingpronenesstokarstification;LC,lithologicalclassoftheunit(C = clasticrocks;D = dolomite;L = limestone;LD = mixedlimestoneanddolomite).TableissortedfirstbyincreasingIICandthenby increasingANI. No.Lithologicalunit(formation)descriptionCISISDNIANIIICLC 92Dolomite,micaceoussiltstone,sandstone,claystone, ooliticlimestoneanddolomite,marlstone,marly limestone(LowerTriassic) 2 220.9 2 4531.4 2 861.60.07200C 77Thick-beddedMainDolomite(UpperTriassicNorian-Rhaetian) 2 435.4 2 1284.9 2 214.70.23900D 91Thick-beddedandmassivedolomite,subordinately limestone(MiddleTriassic-Anisian) 2 153.4 2 3687.6 2 264.60.25050D 84Massivecoarse-crystallinedolomiteandlimestone (UpperTriassic-Rhaetian) 2 40.5 2 5213.9 2 105.70.28560D 27Lithothamniumlimestone,marlylimestoneandmarl (MiddleMiocene-Badenian) 2 160.6 2 2123.0 2 165.80.30590C 80Marlylimestone,marlstone,dolomite,shale(Upper Triassic-Carnian) 2 93.3 2 1762.9 2 239.40.31410C 78PlatyBac aDolomitewithchert(UpperTriassicNorian-Rhaetian) 2 88.2 2 1889.2 2 190.10.32450D 83Alternationofclaystoneands andstone,platylimestoneinthe upperpart-Amphiclinabeds(UpperTriassic-Carnian) 2 31.1 2 546.2 2 198.70.36670C 73Platymicriticlimestoneandcalcarenitewithchert(Lias) 2 28.7 2 525.0 2 169.60.37470L 29Lithothamniumlimestone(MiddleMiocene-Badenian) 2 107.0 2 959.3 2 14.30.38150L 93Thick-beddeddolomite,subordinatelylimestone(Upper Permian) 2 15.8 2 542.8 2 110.90.39180D 86Massivedolomite,subordinatelylimestone(Middleand UpperTriassic-Anisian-Norian) 2 73.2 2 794.4 2 16.00.39370D 56Platylimestonewithchertinalternationwithred marlstone-Krs kobeds(UpperCretaceous-Upper Cenomanian-Turonian) 2 32.6 2 650.7 2 51.80.39910C 53PlatyVolc elimestonewithchert,redmarlylimestoneand marlstone(UpperCretaceous-Coniacian-Campanian) 2 28.5 2 759.4 2 13.30.40700C 63PlatyBianconelimestonewithchert(UpperJurassicLowerCretaceous-Tithonian-Berriasian) 2 8.4 2 362.9 2 69.50.40750L 42Limestone-dolomiteconglomerate-S kofjaLokaand OkoninaConglomerate(MiddleOligocene-Rupelian) 2 9.3 2 165.6 2 77.70.40960C 81Platylimestoneanddolomit ewithchert.marlstone.marly limestone-TamarFormation(UpperTriassic-Carnian) 2 19.4 2 298.0 2 26.30.41630C 85MassiveWettersteinLimest oneanddolomite.thick-bedded limestone(MiddleandUpperTriassic-Ladinian-Cordevol) 2 1.7 2 595.6 2 18.50.41640LD 52Coarse-grainedlimestonebrecciawithintercalationsof flysch(UpperCretaceous-Maastrichtian) 2 18.2 2 358.3 2 8.00.41970C 34Lithothamnium-Lepidocyclinalimestone,sand,siltand clay(LowerMiocene-Ottnangian-Eggenburgian) 2 8.2 2 76.4 2 37.00.42150C 70Massivecrinoidalandooliticlimestone(Lias.Dogger) 2 3.2 2 67.9 2 21.60.42670L 97Light-graytoredlimestone-Dovz anovasoteskaand TrogkofelFormations(LowerPermian) 2 1.5 2 229.6 2 4.50.42770L 68ReddishandgreyishnodularlimestoneofAmmonitico Rossotypelimestonebreccias,marlstoneand claystone(UpperandLowerJurassic) 2 1.1 2 273.10.00.42800C 82Chert,platylimestone,claystoneandsiltstone-Kobla Formation(UpperTriassic-Carnian) 2 5.5 2 92.9 2 2.40.43010C A SSESSMENTOFSPATIALPROPERTIESOFKARSTAREASONAREGIONALSCALEUSING GIS ANDSTATISTICS – THECASEOF S LOVENIA 256 N JournalofCaveandKarstStudies, December2012

PAGE 25

couldbeexplainedbythefactthattheaverageslopeangleof thisunitis26.4 u ,whileaverageslopeanglesofotherintensive karstareasarebelow13 u .Fortheintenselykarstifiedoutcropsanalyzedinthispaper,thecorrelationbetweenthe numberofsinksinagivenslopeangleclassandtheareaof thesameslopeangleclasseswas0.79.Thedistributionofthe slopeangleclassesÂ’areascorrelateswellwiththedistribution ofsinks;hencethisdependencecanbeusedtotestthe correlationbetweenslopeangleandsinkdistribution.Correlationtestofthelatershowedthatathigherslopesthe No.Lithologicalunit(formation)descriptionCISISDNIANIIICLC 102Thick-beddedlimestoneinthelowerpart,reeflimestone inthemiddlepart,thick-beddedmicriticlimestonein theupperpart(Devonian) 2 0.4 2 72.7 2 1.50.43210L 96Neoschwagerinareeflimestone,limestonebreccia (MiddlePermian) 2 0.9 2 4.8 2 5.20.43260L 87Platymicriticlimestonewithchertnodules-Pokljuka Formation(MiddleandUpperTriassic) 2 1.8 2 237.30.10.42861L 75Reeflimestonewithcorals(UpperTriassic-Rhaetian) 2 5.9 2 100.92.00.43091L 50Reddishandgraymarlylimestoneandmarlstone (Turonian-Campanian) 2 4.0 2 87.31.90.43171C 48Thick-beddedmicriticlimestone-VremeandKozina beds(LowerPaleocene-UpperCretaceous-DanianMaastrichtian) 0.0 2 0.129.20.44111L 13Coherentfluvialdeposits;terraces(limestone conglomeratewithgravelintercalations)(Quaternar) 2 82.99195.7 2 36.10.60201C 74Limestone,dolomiteandlime stone-dolomitebreccia(Upper TriassicandLowerJurassic(RhaetianandLias)) 2 5.926.95.30.43442LD 58Platylimestonewithchert-Komenbeds(UpperCretaceousUpperCenomanian-Turonian) 2 7.77.411.80.43512L 45Alveolina-nummuliteslimestone(MiddleEocene)3.812.4 2 0.20.43542L 46Alveolina-nummulitesandmili olidalimestone(LowerEocene)3.8 2 7.422.50.44042L 67Reeflimestonewithcorals,hydrozoansandsponges (LowerpartofUpperJurassic-LowerKimeridgianOxfordian) 2 9.31294.265.20.47522L 72Micriticandooliticlimestone,limestonebrecciaand bituminousdolomite(LowerandMiddleLias) 0.0849.8119.40.48102LD 65Ooliticandmicriticlimestone(UpperpartofUpper Jurassic-UpperKimeridgian-Tithonian) 2 8.12986.747.50.50782L 66Thick-beddedmicriticandool iticlimestone(Lowerpartof UpperJurassic-LowerKimeridgian-Oxfordian) 2 6.64142.5156.90.55922L 76Thick-beddedDachsteinLimestonewithtransitionsto dolomite(UpperTriassic-Norian-Rhaetian) 723.1 2 3177.284.20.57872L 55Platylimestonewithcher t-DutovljeFormation(Upper Cretaceous-Campanian) 0.040.86.30.43653L 54Rudistlimestoneandcalcare nite-LipicaFormation(Upper Cretaceous-Coniacian-Campanian) 112.260.138.50.47463L 64Alternationofdolomiteandlimestone(Upperpartof UpperJurassic-UpperKimeridgian-Tithonian) 9.12617.2144.50.52753LD 71Micriticandooliticlimest one,bituminousdolomite(Upper Lias-Dogger) 11.12495.1153.90.52773LD 61Alternationoflimestoneand dolomite(lowerpart).micritic limestone(upperpart)(LowerCretaceous-BerriasianBarremian) 250.77132.8175.00.69683LD 57Rudistandmicriticlimestone-Sez anaFormation(Upper Cretaceous-Turonian) 813.31503.3269.50.74803L 59Thick-beddedmicriticlimes toneandbituminousdolomite (LowerCretaceousandlower partofUpperCretaceous) 329.210226.7536.80.87083LD Table2.Continued. M.K OMACAND J.U RBANC JournalofCaveandKarstStudies, December2012 N 257

PAGE 26

occurrenceofsinksislowerandthecorrelationbetweensink occurrencedistributionandslopeangledistributionwas 2 0.797.ThisagreeswithGams(2003,p.170),whoclaims thattheoccurrenceofsinkshasanegativecorrelationwith slopeangle.Inadditiontoitsgreaterslopes,unit76islocated atrelativelyhighelevation,anaverageof1200m,whilethe averageelevationofalltheotherintenselykarstifiedunitsis 708m.Thesteeperslopesthatresultinhighersurfacerunoff, extremeconditionsofprecipitation,winds,andtemperature variations,andtheabsenceofvegetationcoverathighelevationscouldhinderthedevelopmentofsinks. Highsurfacedrainagenetworkvalues,correspondingto smallamountsofsurfacedrainage,agreewellwiththe selectionofhighlykarstifiedunits.Asexpectedfromfield observations,highSDNIvaluesarerelatedtotwolithologicalclasses,limestonesandmixedlimestoneanddolomite. Thereasonisrathertrivial;unitsclassifiedaslimestonesor limestonesanddolomiteshavelittleornosoilcoverthat would,tosomeextent,preventverticalpercolationofwater throughsmallcracksandfracturestothesubsurface.Where considerablesurfacedrainageflowispresent,thereareusuallyfluvialsediments,andonthegeologicalmapsuchareasare labeledasQuaternaryalluvialsediments. Table3showssomeofthedatainTables1and2averagedoverthestratigraphicunits,bothasdividedintoless karstifiedunitsandintenselykarstifiedunitsandasdivided intothelithologicalclassesclastic,dolomite,limestone,and mixedlimestoneanddolomite.Indexanddensityvalues clearlyshowthedistinctionbetweenclassesthathave higher(limestoneandmixedlimestoneanddolomites)and lower(clasticsanddolomites)karstificationintensities. Nevertheless,thecomparisonoflithologicclasswithkarstificationintensityshowsthatnotalllimestones,andtoa lesserextent,notallmixedlimestonesanddolomites,can besimplyclassifiedasintenselykarstifiedareas.The differentiationisclearlyvisibleinathree-dimensionalplot whereaxesareCI,SI,andSDNIvalues(Fig.4).Thelow averagenormalizedindexesofnineunitsclassifiedaslimestoneandoneunitclassifiedaslimestonesanddolomites thatareexceptions(seeTable2)couldbeduetothin(5to 10cm)beddingwithtotalthicknessesofseveralmetersup toonly30meters(unit63),tohighcontentofintercalationsinlimestonesuchasquartzitepebbles,marls,cherts, andsandstones(units29,73,96),siltstonesandsandstones (unit97),orsiltstonesandmarls(unit102),ortoverysteep averageslopesabove20 u (units70,75,85,87).Thelast possibilityissupportedbythefactthatthecorrelationof sinkdensitywithslopeangleforlimestoneandmixed limestoneanddolomite,amongunitswithatleastone positiveindex,is 2 0.797.Theeffectoftheslopesofthose limestoneandmixedcarbonateunitsinwhichtheactual sinkoccurrenceismuchlowerthanexpected(63,70,75,85, and87),isprobablyexaggeratedbythefactthattheseunits coversmallareaswherealownumberofcaveshavebeen discovered,andresultinlowANIvalues.Basedonresults ofsinkdistributionanalyseswithinsloperanges,critical Figure3.Cave,sink,andsurfacedrainagenetworkdensityindexvalues,computedasdescribedinthetext,foreachstratigraphic unitincludedintheanalyses.Unitsarearrangedfromleftaccordingtoincreasingaveragenormalizedindex(ANI;seetext).Units determinedtobeintenselykarstifiedaretotherightofconspicuousunit13. A SSESSMENTOFSPATIALPROPERTIESOFKARSTAREASONAREGIONALSCALEUSING GIS ANDSTATISTICS – THECASEOF S LOVENIA 258 N JournalofCaveandKarstStudies, December2012

PAGE 27

slopeanglesabovewhichsinksareunlikelytodevelopata 95%(99%)probabilitylevel,are17 u (21 u )forunit63,15 u (20 u )forunit70,20 u (22 u )forunit75,23 u (26 u )forunit85, and19 u (25 u )forunit87.Ineachcase,the95%slope-angle valuesareeithersomewhatlowerthantheaverageslope anglesforagivenunit(23.5 u forunit63,20.1 u forunit70, 28.8 u forunit85,20.3 u for87)ormuchlowerthanthe averageslopeangle(32.3 u forunit75).Thisremainstrue evenforthe99%threshold,exceptinunit87.Thedistributionofsinkoccurrenceonshallowerslopesinunits63, 70,75,85,and87thathaverelativelyhighaverageslope angles,indicatesthatthesparseoccurrenceofsinkswithin theseunitscouldberelatedtothosehighaverageslopes. Intheintenselykarstifiedarea s,theaveragedensityofcaves isabout25timesgreaterthaninle sskarstifiedareas(Table3). Forsinkdensity,theratiobetweenintenselyandlesskarstified areasisgreaterthan3.Surfacedrainagenetworkdensityin intenselykarstifiedareasis almost7timeslowerthaninless karstifiedareas. Resultsoftheanalysesarebestcomprehendedifdisplayedgraphically.Figure5showsmapsofthethreefeature indexes,forcaves,sinks,andsurfacedrainagenetwork,in partsA–C,andtheresultofcombiningthedataintoaverage normalizedindexesinpartD.Thedifferencebetweenthe sinkindexSIandtheothertwoindexesismostobviousin theAlpineregioninthenorthwestandthenorth-central Quaternaryterraces.ThesummaryFigure5D,showingthe ANIvaluesdividedintothetwokarstification-intensity classes,showsthat23.9%ofSlovenia’sarea(outcropsof sixteenstratigraphicunitsrepresentedby550polygonsthat cover4850.35km 2 )canbeclassifiedasintenselykarstified and20.8%(thirty-oneunitsrepresentedby1498polygons covering4216.36km 2 )canbeclassifiedaslesskarstified. C ONCLUSIONS TheresultsderivedbyGISforSloveniashowthatitis possible,usingacombinationofappropriateindicative features,toidentifyhypergenickarstareas,andtoacertain extent,alsoassessthedegreeoftheirkarstification.We’ve shownthatusefulparametersarelithology,sinkandcave density,andtheabsenceofasurficialdrainagenetwork. Amongsomeoftheseparameters,wenoticedsomecorrelation.We’vealsoshownageneralnegativecorrelation betweenslopeanglesandsinkdensitywithhighreliability. Averagedensitiesofcaves,sinks,andsurfacedrainage networksinintenselykarstifiedareasareseveraltimes higherforthefirsttwoandseveraltimeslowerforthe latterthaninlesskarstifiedareas.Theresultspresented showregionalkarstificationlevelscanvaryonalocalscale incomparisontothegeneralscale. ThemapinFigure5Dprovidesabasisforthegeneral assessmentofgroundwatervulnerability,whichisespecially highinthekarstifiedareas,andforwater-resourceprotection planning.Inaddition,themapcouldbeusedinpreliminarystudiesofland-planningissuessuchasthesitingof Figure4.3Dplotofaveragevaluesofthethreeindexesover theanalyzedunitssortedbylithology(C,clasticrocks;D, dolomites;L,limestones;LD,mixedlimestoneanddolomite) andbydegreeofkarstification(LKA,lesskarstifiedareas; IKA,intenselykarstifiedareas),asgiveninTable3. Table3.Averagevaluesofindexes(CI,caveindex;SI,sinkindex;SDNI,surfacedrainagenetworkindex),theaverage normalizedindex(ANI),andotherparametersoverthestratigraphicunitsbylithologicalclass(C = clasticrocks;D = dolomite;L = limestone;LD = mixedlimestoneanddolomite)andbydegreeofkarstification(LKA,lesskarstifiedarea;IKA, intenselykarstifiedarea). Rock TypeCISISDNIANI Elevation (ma.s.l.) Slopeangle (degrees) Cavedensity (km 2 2 ) Sinkdensity (km 2 2 ) SDNdensity (km/km 2 ) C 2 51.11 2 180.67 2 122.590.387688.616.90.187.131.52 D 2 134.44 2 2235.46 2 150.320.314743.619.950.336.151.13 L73.33215.1122.380.46477416.121.2515.010.36 LD84.663250.41159.490.565628.611.381.1925.030.13 LKA 2 54.25 2 616.08 2 93.280.533708.818.80.066.251.26 IKA138.671888.21114.840.382678.613.51.4720.930.19 M.K OMACAND J.U RBANC JournalofCaveandKarstStudies, December2012 N 259

PAGE 28

problematicinfrastructure(landfills,industrialplants,gas stations,etc.)orforassessmentofkarsthazardsin engineeringgeology.Butthelimitationsofthepresented approachneedtobetakenintoaccount,includingthecoarse scaleofthegeologicaldata,thelowdigital-elevation-model resolution,andproblemswiththeautomatedderivationof surfacekarstfromtheDEM.Theresultscanbeusedas guidanceataregionalscaletoprioritizeareasoffuturemore detailedassessmentoflocalkarstareas. A CKNOWLEDGEMENTS TheauthorswishtothanktheSlovenianResearchAgency forfundingtheproject‘‘Aspatialmodelofgroundwater hydrochemicalcompositioninSloveniainGISenvironment’’ withinwhichthisresearchwasperformed. R EFERENCES Armstrong,B.,Chan,D.,Collazos, A.,andMallams,J.L.,2003,Dolineand aquifercharacteristicswithinHernan do,Pasco,andnorthernHillsborough counties:GSAAnnualMeeting,Seattle,November2–5,2003,p.39–51. ARSO,2005,EuropeanEnvironmentalInformationandObservationNetwork(Evropskookoljskoinformacijskoinopazovalno omrez je),Ljubljana,EnvironmentalAgencyofRepublicofSlovenia: http://nfp-si.eionet.europa.eu/Dokumenti/GIS/(accessedApril10, 2010). Blaeu,W.,andBlaeu,J.,1645,Karstia,Carniola,HistriaetWindorum Marchia, in TheatrumOrbisTerrarum,siveAtlasNovusinquo TabuletDescriptionesOmniumRegionum,EditaGuiljelet IoanneBlaeu:http://www.library.ucla.edu/yrl/reference/maps/blaeu/ index.htm # description(accessedAugust21,2010) Buser,S.,2010,GeologicalMapofSloveniaatscale1:250,000(Geolos ka kartaSlovenije1:250.000):GeologicalSurveyofSlovenia,1sheet. Denizman,C.,2003,Morphometricandspatialdistributionparametersof karsticdepressions,lowerSuwanneeRiverbasin,Florida:Journalof CaveandKarstStudies,v.65,no.1,p.29–35. Florea,L.J.,2005,Usingstate-wideGISdatatoidentifythecoincidence betweensinkholesandgeologicstructure:JournalofCaveandKarst Studies,v.67,no.2,p.120–124. Florea,L.J.,Paylor,R.L.,Simpson,L. ,andGulley,J.,2002,KarstGISadvances inKentucky:JournalofCaveandKarstStudies,v.64,no.1,p.58–62. Gams,I.,2003,KrasvSlovenijivprostoruinc asu:Ljubljana,ZRC SAZU,487p. Gao,Y.,andZhou,W.,2008,AdvancesandchallengesofGISandDBMS applicationsinkarst:EnvironmentalGeology,v.54,p.901–904. doi:10.1007/s00254-007-0894-4. GURS,2005,Digitalelevationmodel–DMV25,1998–2005(DEMwith resolution25 3 25m),SurveyingandMappingAuthorityofthe RepublicofSlovenia. Figure5.Mapsoftheanalyzedareas,withcolorsbasedonA,caveindex(CI);B,sinkindex(SI);C,surfacedrainagenetwork densityindex(SDNI);andD,averagenormalizedindex(ANI).InD,thedivisionbetweenlessandintenselykarstifiedareasis showninthekey;units13and48areshownhatchedbecausetheyareclassifiedaslesskarstified,despitetheirANIvalues. A SSESSMENTOFSPATIALPROPERTIESOFKARSTAREASONAREGIONALSCALEUSING GIS ANDSTATISTICS – THECASEOF S LOVENIA 260 N JournalofCaveandKarstStudies, December2012

PAGE 29

JZS,2010,CatalogofcavesinSlovenia:Sp eleologicalAssociationofSlovenia. Komac,M.,2005,StatisticsoftheGeologicalMapofSloveniaatscale 1:250000:Geologija,v.48,no.1,p.117–126. Monroe,W.H.,1970,AGlossaryofKarstTerminology:U.S.Geological Survey,Water-SupplyPaper1899,26p. Stafford,K.W.,Rosales-Lagarde,L.,andBoston,P.J.,2008,Castile evaporitekarstpotentialmapoftheGypsumPlain,EddyCounty, NewMexicoandCulbersonCounty,Texas:AGISmethodological comparison:JournalofCaveandKarstStudies,v.70,no.1, p.35–46. Taylor,C.J.,Nelson,H.L.Jr.,Hileman,G.,andKaiser,W.P.,2005, Hydrogeologic-frameworkmappingofshallow,conduit-dominated karst—ComponentsofaregionalGIS-basedapproach, in Kuniansky, E.L.,ed.,U.S.GeologicalSurveyKarstInterestGroupProceedings. RapidCity,SouthDakota,September12–15,2005,U.S.Geological Survey,ScientificInvestigationsReport2005-5160,p.103–113. Valvasor,J.V.,1977,SlavavojvodineKranjske:Ljubljana,Mladinska knjiga,365p. Veni,G.,2002,RevisingthekarstmapoftheUnitedStates:Journalof CaveandKarstStudies,v.64,no.1,p.45–50. M.K OMACAND J.U RBANC JournalofCaveandKarstStudies, December2012 N 261

PAGE 30

LIZARDSANDSNAKES(LEPIDOSAURIA,SQUAMATA) FROMTHELATEQUATERNARYOFTHESTATEOF CEARA INNORTHEASTERNBRAZIL A NNIE S CHMALTZ H SIOU 1 ,P AULO V ICTORDE O LIVEIRA 2 ,C ELSO L IRA X IMENES 3 AND M ARIA S OMA LIA S ALES V IANA 4 Abstract: WepresentthefirstformalreportonthesquamateassemblagefromParque NacionaldeUbajara.Thisparkcontainsthemostimportantcavecomplexinthestateof Ceara innortheasternBrazil,calledProv nciaEspeleolo gicadeUbajara.Thematerial comesfromtheUrsoFo ssilcaveatPenduradoHill.Allpreviouslyreportedfossil remainsfoundinthiscavearetentativelyattributedtotheQuaternary(latePleistoceneearlyHolocene).Probablyonly Arctotheriumbrasiliense representsarelictualfossilbear fromthelatePleistocenemegafauna.Thetaxarecognizedinthispaperbelongto Tropidurus sp., Ameiva sp.,cf. Epicrates ,andcf. Crotalusdurissus ,addingtothe knowledgeoftheBrazilianQuaternarysquamatefaunaasawhole,andcontributetoa majortaxonomicrefinementofthesquamateassemblagesfromtheearlyHoloceneof northeasternBrazil. I NTRODUCTION TheBrazilianQuaternaryrecordofSquamata(i.e, lizards,amphisbaenians,andsnakes)hasbeendocumented mainlyinthesoutheastandnortheastregions,withseveral taxacorrelatedtothecurrentBrazilianherpetofauna (CamolezandZaher,2010;Hsiou,2010).Manyrecords, however,havenotbeenformallystudiedanddescribed (Lund,1840;Paula-Couto,1978;Linoetal.,1979;BarrosBarretoetal.,1982;Gue rin,1991;Gue rinetal.,1993; Faureetal.,1999)andtheirtaxonomicandsystematic statusremainsunclear. Someofthemostdiversesquamatefaunasfromthelate Pleistocene–HoloceneofBrazilwererecentlystudied.The fossilswerecollectedincavesandrocksheltersinthestates ofBahia(northeasternBrazil),Goia s,MatoGrosso (midwesternBrazil),MinasGerais,andSa oPaulo(southeasternBrazil;CamolezandZaher,2010).Alargenumber oflizards(Tropiduridae,Leiosauridae,Polychrotidae, Teiidae,andAnguidae),snakes(Boidae,Colubridae, Viperidae,andElapidae)andamphisbaenians(Amphisbaenidae)weredescribed,andthemajorityoffossilswere attributedtoextantneotropicalspecies(Camolezand Zaher,2010).Allfossilswereidentifiedbasedonosteologicalcomparisonwithextantspecies.However,thereis onlyasinglerecordofsnakes(Viperidae)fromthelate PleistoceneofsouthwesternBrazilianAmazonia(Hsiou andAlbino,2011).Beyondtheserecords,someextinct speciesincludingtwoamphisbaenians, Amphisbaenabraestrupi and A.laurenti, werereportedfromthelate Pleistocene–HoloceneofLagoaSantaregion,Minas GeraisState(GansandMontero,1998),aswastheextinct teiidlizard Tupinambisuruguaianensis ,fromthelate PleistoceneoftheTouroPassoFormation,RioGrande doSulState,southernBrazil(Hsiou,2007). RecentfieldworkwasundertakenatParqueNacional deUbajara,wherethemostimportantcavecomplexinthe stateofCeara innortheasternBrazilislocated,partofa notablekarsticsystem(Oliveira,2010).Smallmammals, suchasbats,rodents,andmarsupials(Ximenesand Machado,2004),wereamongthetaxarecorded.Other recordsincludeartiodactyls(deerandpeccaries),perissodactyls(tapirs),xenarthrans(armadillos),andfelids (XimenesandMachado,2004;Oliveira,2010),aswellas asinglememberofthelatePleistocenemegafauna,the fossilbear Arctotheriumbrasiliense (TrajanoandFerrarezzi,1995).Otherrecordsarefromdepositsin tanques (naturaldepressionsformedingraniticrocksthataccumulatesedimentsandfossils),suchasundetermined remainsoflizardsandsnakesreportedbyPaula-Couto (1980)asbeingrecordedfromPleistocenedepositsofthe Itapipocaregion. Recently,Hsiouetal.(2009)brieflyreportedonsome snakevertebraeofthefamilies‘Colubridae’andViperidae fromthelateQuaternaryinProv nciaEspeleolo gicade Ubajara.Theirreportdidnotcontainstratigraphicdata (seealsoTrajanoandFerrarezzi,1995;Ximenesand Machado,2004)orradiometric(geochronologic)control. HerewedescribenewmaterialfromtheearlyHolocenein thestateofCeara inBrazilbasedonthemostrecent 1 DepartamentodeBiologia,FFCLRP,UniversidadedeSa oPaulo;Av.Bandeirantes3900,Ribeira oPreto-SP,Brazil. anniehsiou@ffclrp.usp.br 2 ProgramadePo s-Graduac a oemGeocie ˆncias/CNPq,DepartamentodeGeologia, CentrodeTecnologiaeGeocie ˆnciasdaUniversidadeFederaldePernambuco (UFPE);Av.Acade ˆmicoHe lioRamos,s/n u -CidadeUniversita ria,CEP50740-530, Recife-PE,Brasil. victoroliveira.paleonto@gmail.com 3 MuseudePre -Histo riadeItapipoca(MUPHI);AvenidaAnasta cioBraga,349, 62500-000,Itapipoca-CE,Brasil. clx.ximenes@gmail.com 4 Laborato riodePaleontologia,MuseuDomJose ,UniversidadeEstadualValedo Acarau (UVA);Av.DomJose ,878,CEP62010-190,Sobral-CE,Brasil. somalia_viana@hotmail.com A.S.Hsiou,P.V.deOliveira,C.L.Ximenes,andM.S.S.Viana–Lizardsandsnakes(Lepidosauria,Squamata)fromthelateQuaternary ofthestateofCeara innortheasternBrazil. JournalofCaveandKarstStudies, v.74,no.3,p.262–270.DOI:10.4311/2011PA0239 262 N JournalofCaveandKarstStudies, December2012

PAGE 31

fieldworkdoneintheUbajararegion,duringJuly2009, withprecisestratigraphicandradiometriccontrol. L OCATION A ND G EOLOGICAL S ETTING TheParqueNacionaldeUbajaraislocatedinUbajara Municipality(Fig.1),inIbiapabaCuestainthenorthwesternportionofthestateofCeara (northeasternBrazil), whichpossessesanotablekarstsystem.Thespeleological provinceoftheUbajararegionconsistsofninelimestone hillswithfourteenknowncaves(IBAMA,2002).The limestonerockcroppingoutintheregioncorrespondsto theFrecheirinhaFormationoftheUbajaraGroup, NeoproterozoicoftheUbajaraGraben(Quadros,1996; CPRM,2003).TheUbajaraGrouphasanunconformable contactwiththerocksoftheSerraGrandeGroup, Silurian-DevonianoftheParna baBasin(Nascimento etal.,1981).Amongthelimestonehillsinthestudied area,thePenduradoHillincludestwoimportantfossiliferouscaves:UrsoFo ssilandPendurado. AllfossilremainsareattributedtotheQuaternary(late Pleistocene-earlyHolocene),however,thefossilbear Arctotheriumbrasiliense isprobablytheonlyrelictinthisfauna ofthePleistocenemegafauna(TrajanoandFerrarezzi,1995), foundatUrsoFo ssilcave(03 u 49 9 58 0 S,40 u 53 9 34.4 0 W).The materialstudiedwasfoundinoneroomofthiscave,called SaladeEntrada.Acontrolledstratigraphicexcavationinthis roomexposedthreelayersofsedimentarydeposits,andall fossilremainsbelongtotheearlyHolocene.Ageological sectionprovidedinformationaboutunconsolidatedaccumulationsofallochthonous(biogenicandsiliciclasticfrom outsidethecave)andautochthonous(generatedinsidethe cave)material.Thestratigraphiclayersincludesediments fromthebottomtotop(Fig.2): Layer1 hasathicknessof0.20mandiscomprisedof carbonaceoussilt-claysediments,containingsmallerautochthonousfragmentsoflimestoneandlarge,angularfragments ofspeleothems.Inthislayer,thereareshellsoffreshwater clamsandseveralcarbonizedbonefragments.Asinlayer2,a samplewascollectedforthermoluminescencedating;more detailscanbefoundinOliveiraetal.(inpress).Remainsof squamatereptilesandmammalssuchasDidelphimorphia, Xenarthra,Rodentia,andArtiodactylawerefound. Layer2 hasathicknessofabout0.35mandiscomposedof lightgrayclaycontainingautochthonousfragmentsof limestone( 2cm),somesmallgeodes,andfragmentsof stalactites(approximately10cmdiameter)andother speleothems,insomecases,showingconcentrationsofiron oxide.Thetopofthelayercontainsahighconcentrationof twotypesofundeterminedseeds.Atthebottomofthelayer, somecompleteshellsoffreshwaterclamsandseveral fragmentsofshellshadaccumulated.Therearesmallfeces coveredbypowderedcarbonate,butstillunconsolidated. Some20cmfromthetopofthislevel,sedimentsamples werecollectedinPVCpipeforthermoluminescencedating, andbelowthat,sampleswereobtainedforrecoveryof palynomorphs.Gastropodsandseveralfragmentsof Didelphimorphia,Xenarthra,andRodentia,werecollected. Layer3 ,witha0.15mthickness,iscomposedoflight yellowish,silty-claysedimentsofcarbonate-richcomposition,containingsmallautochthonousfragments(0.5–1cm) ofamorphousandangularlimestone,andagreatamountof recentseedsandfecalmatter. M ATERIALAND M ETHODS Thematerialstudiedincludesisolateddentariesand vertebraedepositedinthecollectionofMuseuDomJose Figure1.LocationmapoflateQuaternaryofProv nciaEspeleolo gicadeUbajara,Ceara State,Brazil. A.S.H SIOU ,P.V. DE O LIVEIRA ,C.L.X IMENES AND M.S.S.V IANA JournalofCaveandKarstStudies, December2012 N 263

PAGE 32

(MDJ),intheStateCeara ,Brazil.Thematerialconsistsof twolizarddentaries(MDJR-004and005)andfive vertebraeofsnakes(MDJR-006,020,024,026,and027). Theosteologicalnomenclatureandsystematicsfollow Estes(1983),Presch(1974),Estesetal.(1988),Frost etal.(2001a,b),NydamandCifelli(2002),andNydam etal.(2007)forlizards;andAuffenberg(1963),Hoffstetter andGasc(1969),Rage(1984),Zaher(1999),andHolman (2000)forsnakes. S YSTEMATIC P ALEONTOLOGY SQUAMATAOppel,1811 IGUANIACope,1864 TROPIDURIDAEFrostandEtheridge1989 sensu Frost, Janies,andTitus,2001a Tropidurus Wied,1824 Tropidurus sp. (Fig.3) Material: MDJR-005,completerightdentary. Stratigraphicprovenance: ParqueNacionaldeUbajara, PenduradoHill,UrsoFo ssilCave,SaladaEntrada,layer1 (TL8,200 6 980yearsBP),earlyHolocene. Description: MDJR-005isacompleteanddelicateright dentarywithfourmentalforaminainlabialview.The dentarybearsthirteenpleurodontteethpreservedin eighteentoothpositions.Themesialteethareanteriorly inclinedandunicuspid.Thedistalonesaretricuspid,with twoaccessorycusps,amesialandadistalone,smallerthan themaincentralcusp.Underthelasttwoposteriorteeth, thesubdentalshelfofthedentarypossessesanotchthat extendsobliquelyuntilthelastteeth(CamolezandZaher, 2010).Thesymphysisissmallandslightlydorsally oriented.MeckelÂ’sgrooveisextensivelyclosed,withan anterioropeningrestrictedtoanelongateforamenanda posteriornotchunderthelasttwodistaltoothpositions. Theposteriorprocessofthedentaryislongwhen comparedtothetotalsizeofthedentary.Inlabialview, thedorsalmarginoftheposteriorprocessofthedentary showsaflattenedsurface,forthecontactwiththeanterior processofthecoronoid. Discussion: Speciesof Tropidurus arewidelydistributedin openareasinthetropicalandsubtropicalregions,from southernVenezuelaeastthroughtheGuianastonortheasternBrazil,andfromtheresouthwestoftheAmazonian regiontoeasternBolivia,northernmostUruguay,and centralArgentina(Etheridge,1964;A vila-Pires,1995; Frostetal.,2001b).Therearefourspeciesgroupsformally diagnosedwithinthegenus: T.spinolusus group,the T. borgeti group,the T.semitaeniatus group,andthe T. Figure2.StratigraphicsectionshowingthelayersL1,L2andL3;andtheirassociatedfossils. L IZARDSANDSNAKES (L EPIDOSAURIA ,S QUAMATA ) FROMTHELATE Q UATERNARYOFTHESTATEOF C EARA INNORTHEASTERN B RAZIL 264 N JournalofCaveandKarstStudies, December2012

PAGE 33

torquatus group(Frostetal.,2001b).Attributiontothe specieslevelofMDJR-005wasnotpossible;butthe presenceofalongitudinalimpressionofthelabialsurface oftheposteriorpartofthedentary,whichrisesposteriorly (slight‘‘erosion’’ofthealveolarshelf)shouldbean apomorphysupportingassignmentofMDJR-005tothe ‘‘ Tropidurus group’’(Frost,1992;Frostetal.2001b). SCLEROGLOSSAEstes,deQueirozandGauthier, 1988 SCINCOMORPHACamp,1923 TEIIOIDEAEstes,deQueirozandGauthier,1988 TEIIDAEGray,1827 TEIINAEPresch,1974 Ameiva Meyer,1795 Ameiva sp. (Fig.4) Material: MDJR-004,incompleterightdentary. Stratigraphicprovenance: ParqueNacionaldeUbajara, PenduradoHill,UrsoFo ssilCave,SaladaEntrada,layer1 (TL8,200 6 980yearsBP),earlyHolocene. Description: MDJR-004isarobustbutincompleteright dentary.Itsanteriorportionisfragmentedatthesymphysis,andtheventral,lingualandlabialportionsalsoare broken.Therearethreementalforaminainlabialview. Thedentaryrisesposteriorlytowarditslabialandlingual articulationswiththecoronoid.Lingually,onlythe anteriorportionofthesplenialispreserved.Meckel’s grooveisrestrictedtotheanteriorregionofdentarybythe developmentofthesubdentalshelfshowingastraight groove(Brizuela,2010).Thereareelevensubpleurodont teethpreservedinsixteentoothpositions,withina sulcus dentalis andwithheavydepositsofcementumattooth bases(Estesetal.,1988;NydamandCifelli,2002).The thirdmesialtoothpreservedisconicalandapparently unicuspid,relativelysmallerthanthedistalones.The fourthtoothisdamaged,buttwoaccessoriescuspscanbe seen,amesialanddistalone,bothslightlyposteriorly oriented.Thefifth,sixth,andseventharereplacementteeth withinthereplacementpitandshowtwoaccessorycusps, bothcuspsbeingalmostverticalandalignedwithone another.Theeighthtoothisbrokenatthebase.Fromthe ninthtotheeleventhtooth,toothsizeandinterdental spacingincreases(‘‘enlargedposteriorteeth:agreater degreeofmolariformy’’,EstesandWilliams,1984).Two accessorycuspsarepresent,asonthefourthpreserved tooth. Discussion: Thegenus Ameiva displaysawidegeographicaldistribution,occurrin ginsouthernMexico,Central andSouthAmerica,andinmanyCaribbeanislands,with differentspecies(A vila-Pires,1995;PiankaandVitt, 2003).Thegenushasbeenconsideredaparaphyletic group(Presch,1974;Reedere tal.,2002;Giuglianoetal., 2006,2007),althoughsom eauthorsdefendedmonophyly (HowerandHedges,2003).Amongthe Ameiva species, themoststudiedis A.ameiva, commonlyfoundinopen habitats,coastalandforestsenvironments,andfrequentlyseeninperianthropicsituations(A vila-Pires, 1995). FollowingCamolezandZaher(2010),thelargestsize amongsmallTeiidae(suchas Cnemidophorus Kentropix and Crocodilurus )andthedentalmorphologywouldallow referralofMDJR-004tothespecies Ameivaameiva Nevertheless,noadditionalosteologicalmaterialshave beenfoundtosupportthisattribution.Currently,we cannotidentifyMDJR-004tospecies. SERPENTESLinnaeus,1758 ALETHINOPHIDIANopcsa,1923 MACROSTOMATAMu ¨ller,1831 BOOIDEAGray,1825 BOIDAEGray,1825 Figure3. Tropidurus sp.completerightdentary,MDJR-005:A,labialview;B,lingualview.Scalebar = 10mm. A.S.H SIOU ,P.V. DE O LIVEIRA ,C.L.X IMENES AND M.S.S.V IANA JournalofCaveandKarstStudies, December2012 N 265

PAGE 34

Epicrates Wagler,1830 cf. Epicrates sp. (Fig.5) Material: MDJR-020,incompletemidtrunkvertebrae. Stratigraphicprovenance: ParqueNacionaldeUbajara, PenduradoHill,UrsoFo ssilCave,SaladaEntrada,layer1 (TL8,200 6 980yearsBP),earlyHolocene. Description: MDJR-020lacksmostofthedorsalpartofthe neuralarch.Thespecimenissmall,relativelyrobustand high,withashortcentrum.Thevertebraprobablyisa midtrunkvertebra,giventhepresenceofawellmarkedand anteroposteriorlydevelopedhaemalkeel.Ingeneralview, theprezygapophysesareslightlyinclineddorsally,anterolaterallyoriented,anddisplayashortprezygapophyseal process.Thearticularfacetsoftheprezygapophysesare triangular.Therearesmalllateralforamina.Thecentrumis triangular,widenedanteriorly,andrathernarrow.Adeep paracotylardepressionispresenttoeithersideofthecotyle, butforaminaareabsentthere.Theparadiapophysesare broken,butwereapparentlyrobustandorienteddorsoventrally,surpassingtheventraledgeofthecotyle.Thehaemal keeliswell-markedandbeginsontheventraledgeofthe cotyle.Thesubcentralridgesareweaklymarked;the subcentralgroovesareshallow,butthisismoreevidentin themiddleportionofthecentrum,lateraltothehaemalkeel. Thereisonepairofsubcentralforamina. Discussion: TheendemicNeotropicalgenus Epicrates is currentlyrecognizedasaparaphyleticgroupinrelationto Eunectes ,owingtorecentstudiesthatfoundmainland Epicrates inasister-grouprelationshipwith Eunectes (Burbrink,2005;NoonanandChippindale,2006).The genuscontainstenspecies(Kluge,1989;McDiarmidetal., 1999)andcomprisestwomonophyleticgroups(Kluge, 1989;Passos,2003;Burbrink,2005;NoonanandChippindale,2006;PassosandFernandes,2008).Aninsulargroup distributedintheWestIndianislandscontainstwenty-one taxa(HendersonandPowell,2007),whereas Epicrates cenchria (Linnaeus)isacontinentalendemic(McDiarmid etal.,1999;PassosandFernandes,2008).Oftheprevious ninesubspeciesof E.cenchria ,fivearenowrecognizedas distinctspecies E.alvarezi,E.assisi,E.cenchria,E.crassus and E.maurus basedonstatisticallyrobustdelimitationof speciesboundaries(PassosandFernandes,2008).The taxonomicassignmentofthespecimendescribedaboveis basedonthefollowingcombinationofvertebralcharacters sharedwiththegeneraofextantneotropicalboines:robust, shortandwidevertebra,lowinclinationofthearticular facetoftheprezygapophysis(lessthan15 u );short prezygapophysealprocess,vertebralcentrumshort, markedprecondylarconstriction,haemalkeelwelldevelopedinthemidtrunkvertebrae,andpresenceofsubcentral andlateralforamina(Rage,2001;LeeandScanlon,2002; SzyndlarandRage,2003;AlbinoandCarlini,2008;Hsiou andAlbino,2009,2010).Withintheneotropicalboines,the trunkvertebraissimilartosamplesfromindividualsofthe genera Epicrates and Corallus ,differingfrom Eunectes and Figure4. Ameiva sp.,incompleterightdentary,MDJR-004:A,labialview;B,lingualview.Scalebar = 10mm. L IZARDSANDSNAKES (L EPIDOSAURIA ,S QUAMATA ) FROMTHELATE Q UATERNARYOFTHESTATEOF C EARA INNORTHEASTERN B RAZIL 266 N JournalofCaveandKarstStudies, December2012

PAGE 35

Boa initssmallersize(HsiouandAlbino,2009,2010). AccordingtoHsiouandAlbino(2010), Epicrates and Corallus couldbedistinguishedbythemorphologyofthe anteriorlobeontheanterioredgeofthezygosphene,but MDJR-020lacksmostoftheneuralarch.In Corallus the prezygapophysesaremostlyhorizontalinanteriorview, whereastheyarerelativelymoreinclinedabovethe horizontalplanein Epicrates (HsiouandAlbino,2010). Forthisreason,wetentativelyassignedthetrunkvertebrae MDJR-020tocf. Epicrates. CAENOPHIDIAHoffstetter,1939 COLUBROIDEAOppel,1811 VIPERIDAEOppel,1811 Crotalus Linnaeus,1758 Crotalusdurissus Linnaeus,1758 cf. Crotalusdurissus (Fig.6) Material: MDJR-006,024,026and027,incompletetrunk vertebrae. Stratigraphicprovenance: ParqueNacionaldeUbajara, PenduradoHill,UrsoFo ssilCave,SaladaEntrada,layer 2(MDJR-006,TL8,000 6 990yearsBP)andlayer1 (MDJR-024,026,027,TL8,200 6 980yearsBP),early Holocene. Description: Thepreservationofthevertebraevaries amongspecimens.MDJR-006and024lackmostpartof theneuralspine,hypapophysis,rightprezygapophysisand parapophysealprocess,andbothparadiapophyses;MDJR026lacksthehypapophysisandrightprezygapophysisand parapophysealprocess;MDJR-027lacksmostoftheneural arch,zygosphene,hypapophysis,andleftprezygapohysisand paradiapophysis.Thezygospheneisthinandshowsaconcave anteriormargin,withsmallan ddorsallyangledarticular facets.Theneuralarchiswiderthanlong,ismoderately depressed,andbearsadeepposterodorsalnotch.Theneural canalissubtriangular,lowandwide.Thearticularfacetsof theprezygapophysesareslende r,longerthanbroad,withthe mainaxisratherlaterallyorie nted.Asmallprezygapophyseal processprojectsslightlybey ondthearticularfacetsofthe prezygapophysis.Theparadia pophysesareclearlyoriented dorsoventrallyasawhole.Thed iapophysialandparapophysialsurfacesaredistinctfromeachother.Theparadiapophysesarewelldeveloped,withaprominentandspherical diapophysis,distinctfromala rgeandconcaveparapophysis (seeninMDJR-024).Awell-developedandstronglyinclined parapophysealprocessisspatula tedprojectinganteriorly,and extendsclearlybeyondtheventralrimofthecotyle.The postzygapophysesareelongateda ndinclineddorsolaterally. Thezygantraarelargeanddeep,withasmallforamenwithin eachsideofzygantrum.Theneuralspineisverywell developed,high,andconsiderablyelongatedanteroposteriorly,seeninMDJR-026.Theinter zygapophysealconstrictionis deepandcurved.Smalllateralforaminaareevidentonthe sidewallsoftheneuralarch,m oreorlesspositionedatthe diapophysiallevel.Thecotyleandcondylearenearlycircular, andonepairofsmallparacotylarforaminaisevident;one foramenislocatedoneachsideofthecotyle,placedina shallowdepression.Thecentrumistriangularandbearsavery prominenthypapophysis(brokeninMDJR-006,0026,and 027).Thecentrumisdelimitedbysubcentralridgesthatare welldefinedanteriorlybutvanishintheposteriorhalfofthe centrum. Discussion: The Crotalusdurissus complex(Neotropical rattlesnakes)occursindryareasfromMexicotonorthern Argentina,butisabsentfromCentralAmericanand Amazonianrainforests,resultinginahighlydisjunctive distribution(Wu ¨steretal.,2005).TheBrazilian Crotalus durissus complexisrepresentedbyasinglespecies, Crotalus durissus ,whichhasalargegeographicaldistribution amongthecentralregionofCerrado,semi-aridandarid environmentsofnorthernregion,savannasandopenareas Figure5.cf. Epicrates sp.,incompletemidtrunkvertebra,MDJR-020:A,dorsalview;B,ventralview.Scalebar = 5mm. A.S.H SIOU ,P.V. DE O LIVEIRA ,C.L.X IMENES AND M.S.S.V IANA JournalofCaveandKarstStudies, December2012 N 267

PAGE 36

ofsouthern,southeasternandnorthernregionsofBrazil (Melgarejo,2003).Thegreatdistributionof C.durissus in Brazilisrepresentedbythepresenceoffivegeographical forms, C.d.terrificus,C.d.cascavella,C.d.collineatus,C. d.ruruima, and C.d.marajoensis .Allvertebraedescribed heresharetheonlyvertebralsynapomorphyrecognizedfor theViperidaefamily:awell-developedandstrongly anteroventrallyorientedparapophysealprocess(Zaher, 1999).However,thedistinctionamongtheBrazilian speciesonosteologicalfeaturesstillrequiresfurther studies.Atpresent,thevertebraedescribedherearesimilar tothoseof Crotalusdurissus intheanterioredgeof zygospheneconcave,greatanteroposteriorextentofthe neuralspine,spatulateparapophysealprocess,andpresenceofsmallparacotylarforamina.Allofthesesubtle characterscandistinguish Crotalusdurissus fromspeciesof Bothrops ,anothergenuswithbroadgeographicaldistributionthatoccursinmostopenareasinbothnorthernand southernregionsofBrazil. C ONCLUDING R EMARKS ThenewsquamateassemblagefromthelateQuaternary Prov nciaEspeleolo gicadeUbajaradescribedherecomprisesthelizardfamiliesTropiduridae( Tropidurus sp.)and Teiidae( Ameiva sp.),andthesnakefamiliesBoidae(cf. Epicrates sp.)andViperidae(cf. Crotalusdurissus ),in additiontoundetermined‘‘colubrid’’snakesreportedby Hsiouetal.(2009).Unfortunately,allspecimensarevery fragmentary,andspecificassignmentisnotpossible. CamolezandZaher(2010)reportedsomesquamate assemblagesfromseveralregionsofBrazil,butthiswork constitutesthefirstformallydescribedrecordfromthe stateofCeara .Hence,thetaxareportedinthispaper contributetoabetterunderstandingoftheBrazilian Quaternarysquamatefaunaasawhole.Mostofthe previousrecordsfromnortheasternBrazilweremadebased onuninformativereports,lackingaformaldescription,and theirtaxonomicvalidityisstillunclear(Paula-Couto,1980; Gue rin,1991;Gue rinetal.,1993;Faureetal.,1999).For thisreason,thepresentpapercontributestoamajor taxonomicrefinementofthesquamatefaunasduring thelateQuaternaryofnortheasternBrazil.Allmaterial describedherecomesfromlevelswithdatingaround 8,000yearsBP,correspondingtotheearlyHolocene. AccordingtoOliveiraetal.(inpress)theageisconsistent withthevertebratefauna,becauserepresentativesofthe PleistoceneSouthAmericanmegafaunahavenotbeen foundintheselevels.Likethetayassuids,marsupials, xenarthrans,andcaviomorphsfromtheselevels,the squamatesdonotindicatefaunisticalterationduringthe earlyHoloceneincomparisonwiththecurrentfauna (Oliveira,2010;Oliveiraetal.,inpress).Thepaleoecologicaldataindicatedbythetaxareportedareinaccordance withthemosaiccompositionofthecurrentenvironments oftheUbajararegion,havinghumidforestinhigher altitudesandopenanddryerareasintheplains(Oliveiraet al.,inpress). A CKNOWLEDGEMENTS WethankConselhoNacionaldeDesenvolvimento Cient ficoeTecnolo gico(CNPq)forfinancialsupportof theproject‘‘EstudoPaleontolo gicodosMam ferosdas CavernasdoParqueNacionaldeUbajara,Ceara ’’ (Universal/n u 473952/2008-4)headedbyA.M.Ribeiro (MNC/FZBRS).ThanksalsotoA.M.Ribeiro(MCN/ FZBRS),G.Lessa(UFV),andS.Teixeira(MUPHI)for helpandcamaraderieduringfieldworkinJuly2009;to Fundac a oCearensedeApoioa `PesquisaeaoDesenvolvimentoCient ficoeTecnolo gico(FUNCAP,BPI0341-1.07/ 08),andespeciallytoMuseuDomJose (MDJ)fortheloan Figure6.cf. Crotalusdurissus ,midtrunkvertebrae,MDJR-006(A),MDJR-024(B),andMDJR–026(C),inanterior (A1–C1),posterior(A2–C2),lateral(A3–C3),dorsal(A4–C4),andventral(A5–C5)views.Scalebar:10mm. L IZARDSANDSNAKES (L EPIDOSAURIA ,S QUAMATA ) FROMTHELATE Q UATERNARYOFTHESTATEOF C EARA INNORTHEASTERN B RAZIL 268 N JournalofCaveandKarstStudies, December2012

PAGE 37

ofthespecimens.A.S.HsioualsothankstoCNPqinthe formofPostdoctoralFellowship(n u .150803/2010-9);and P.V.OliveirathanksCoordenac a odeAperfeic oamentode PessoaldeN velSuperior(CAPES)andCNPqforMaster andDoctoralFellowshipsdevelopedatProgramadePo sGraduac a oemGeocie ˆnciasatUFRGSandUFPE, respectively.Wealsothanktothethreeanonymous refereesfortheircarefulreviewsandhelpfulsuggestions. R EFERENCES Albino,A.M.,andCarlini,A.A.,2008,Firstrecordof Boaconstrictor (Serpentes,Boidae)intheQuaternaryofSouthAmerica:Journalof Herpetology,v.42,p.82–88. Auffenberg,W.,1963,ThefossilsnakesofFlorida:TulaneStudiesin ZoologyandBotany,v.10,no.3,p.131–216. A vila-Pires,T.C.S.,1995,LizardsofBrazilianAmazonia(Reptilia: Squamata):ZoologischeVerhandelingen,no.299,706p. Barros-Barreto,C.N.G.,DeBlasis,P.D.,DiasNeto,C.M.,Karmann,I., Lino,C.F.,andRobrahn,E.M.,1982,AbismoPontadeFlecha:um projetoarqueolo gico,paleontolo gicoegeolo giconome diocursode RibeiradeIguape,Sa oPaulo.RevistadePre -Histo ria,v.3, p.195–215. Brizuela,S.,2010,Loslagartoscontinentalesfo silesdelaArgentina (exceptoIguania)[Ph.D.thesis]:LaPlata,FaculdaddeCiencias NaturalesyMuseo,UniversidadNacionaldeLaPlata,442p. Burbrink,F.T.,2005,Inferringthephylogeneticpositionof Boa constrictor amongtheBoinae:MolecularPhylogeneticsandEvolution,v.34,p.167–180.doi:10.1016/j.ympev.2004.08.017. Camolez,T.,andZaher,H.,2010,Levantamento,identificac a oe descric a odafaunadeSquamatadoQuaterna rioBrasileiro(Lepidosauria):ArquivosdeZoologia,MuseudeZoologiadaUniversidade deSa oPaulo,v.41,p.1–96. CPRM(Servic oGeolo gicodoBrasil),2003,AtlasDigitaldeGeologiae RecursosMineraisdoCeara ,Bras lia,Ministe riodasMinase Energia,CD-ROM. Estes,R.,1983,Sauriaterrestria,Amphisbaenia,HandbuchderPala ¨oherpetologie(Teil10A):Stuttgart,GustavFisherVerlag,249p. Estes,R.,DeQueiroz,K.,andGauthier,A.,1988,Phylogenetic relationshipswithinSquamata, in Estes,R.,andPregill,G.,eds., PhylogeneticRelationshipsoftheLizardFamilies:EssaysCommemoratingCharlesL.Camp:California,StanfordUniversityPress, p.119–281. Estes,R.,andWilliams,E.E.,1984,Ontogeneticvariationinthe molariformteethoflizards:JournalofVertebratePaleontology, v.4,p.96–107.doi:10.1080/02724634.1984.10011989. Etheridge,R.,1964,Theskeletalmorphologyandsystematicrelationships ofsceloporinelizards:Copeia,v.1964,p.610–631. Faure,M.,Gue rin,C.,andParenti,F.,1999,De couverted’uneme gafaune Holoce `nea `laTocadoSerrotedoArtur(airearche ologiquedeSa o RaimundoNonato,Piau ,Bre sil):ComptesRendusdel’Acade miedes Sciences,Ser.II,SciencesdelaTerreetdesPlane `tes,v.329, p.443–448. Frost,D.R.,1992,Phylogeneticanalysisandtaxonomyofthe Tropidurus groupoflizards(Iguania:Tropiduridae):AmericanMuseumNovitates,no.3033,68p. Frost,D.R.,andEtheridge,R.,1989,APhylogeneticanalysisand taxonomyofIguanianLizards(Reptilia:Squamata):TheUniversity ofKansas,MuseumofNaturalHistory,MiscellaneousPublications, no.81,65p. Frost,D.R.,Etheridge,R.,Janies,D.,andTitus,T.A.,2001a,Total evidence,sequencealignment,evolutionofpolychrotidlizards, andareclassificationoftheIguania(Squamata:Iguania):American MuseumNovitates,no.3343,38p. Frost,D.R.,Rodrigues,M.T.,Grant,T.,andTitus,T.A.,2001b, Phylogeneticsofthelizardgenus Tropidurus (Squamata:Tropiduridae:Tropidurinae):directoptimization,descriptiveefficiency,and sensitivityanalysisofcongruencebetweenmoleculardataand morphology:MolecularPhylogeneticsandEvolution,v.21, p.352–371.doi:10.1006/mpev.2001.1015. Gans,C.,andMontero,R.,1998,Twonewfossilamphisbaenids(Reptilia: Squamata)fromthePleistoceneofLagoaSanta(MinasGerais, Brasil):Steenstrupia,v.24,p.9–22. Giugliano,L.G.,Contel,E.P.B.,andColli,G.R.,2006,Geneticvariability andphylogeneticrelationshipsof Cnemidophorusparecis (Squamata, Teiidae)fromCerradoisolatesinsouthwesternAmazonia:BiochemicalSystematicsandEcology,v.34,p.383–391.doi:10.1016/ j.bse.2005.12.007. Giugliano,L.G.,Collevatti,R.G.,andColli,G.R.,2007,Moleculardating andphylogeneticrelationshipsamongTeiidae(Squamata)inferredby molecularandmorphologicaldata:MolecularPhylogeneticsand Evolution,v.45,p.168–179.doi:10.1016/j.ympev.2007.05.017. Gue rin,C.,1991,Lafaunedeverte bre sduPle istoce `nesupe rieurdel’aire arche ologiquedeSa oRaimundoNonato(Piau ,Bre sil):Comptes Rendusdel’Acade miedesSciences,Ser.II,SciencesdelaTerreetdes Plane `tes,v.312,p.567–572. Gue rin,C.,Curvello,M.A.,Faure,M.,Hugueney,M.,andMourerChauvire ,C.,1993,Lafauneple istoce `neduPiau (Nordestedu Bre sil):implicationspale oe cologiquesetbiochronologiques:QuaternariaNova,v.3,p.303–341. Henderson,R.W.,andPowell,R.,2007,Thebiologyofboasandpythons: aretrospectivelooktothefuture, in Henderson,R.W.,andPowell, R.,eds.,BiologyoftheBoasandPythons:EagleMountain,Utah, EagleMountainPublishing,p.2–22. Hoffstetter,R.,andGasc,J.P.,1969,V ertebraeandribsofmodernreptiles, in Gans,C.,Bellairs,A.d’A.,andParsons,T.S.,eds.,BiologyoftheReptilia, Volume1,MorphologyA:Lond on,AcademicPress,p.201–310. Holman,J.A.,2000,FossilSnakesofNorthAmerica:Origin,Evolution, Distribution,Paleoecology:Bloomington,IndianaUniversityPress, 357p. Hower,L.M.,andHedges,S.B.,2003,Molecularphylogenyand biogeographyofWestIndianteiidlizardsofthegenus Ameiva : CaribbeanJournalofScience,v.39,p.298–306. Hsiou,A.S.,2007,AnewTeiidaespecies(Squamata,Scincomorpha)from thelatePleistoceneofRioGrandedoSulState,Brazil:Revista BrasileiradePaleontologia,v.10,p.181–194. Hsiou,A.S.,2010,Lagartoseserpentes(Lepidosauria,Squamata)do Miocenome dio-superiordonortedaRegia oNorteAme ricadoSul [Ph.D.thesis]:PortoAlegre,UniversidadeFederaldoRioGrandedo Sul,239p. Hsiou,A.S.,andAlbino,A.M.,2009,Presenceofthegenus Eunectes (Serpentes,Boidae)intheNeogeneofSouthwesternAmazonia, Brazil:JournalofHerpetology,v.43,p.612–619.doi:10.1670/08295.1. Hsiou,A.S.,andAlbino,A.M.,2010,Newsnakeremainsfromthe MioceneofnorthernSouthAmerica:HerpetologicalJournal,v.20, p.249–259. Hsiou,A.S.,andAlbino,A.M.,2011,FirstrecordofViperidaesnakes fromthePleistoceneofsouthwesternBrazilianAmazonia:Alcheringa, v.35,p.389–395.doi:10.1080/03115518.2011.519646. Hsiou,A.S.,deOliveira,P.V.,andXimenes,C.L.,2009,Presenc ade Colubroidea(Squamata,Serpentes)noQuaterna riodaProv ncia Espeleolo gicadeUbajara,EstadodoCeara , in LivrodeResumosdo CongressoBrasileirodePaleontologia,21th,Bele m,Universidade FederaldoPara /MuseuEm lioGoeldi,189p. IBAMA(InstitutoBrasileirodoMeioAmbienteedosRecursos Renova veis),2002,ParqueNacionaldeUbajara:PlanodeManejo, Bras lia,Ministe riodoMeioAmbiente,CD-ROM. Kluge,A.G.,1989,Aconcernforevidenceandaphylogenetichypothesis ofrelationshipsamong Epicrates (Boidae:Serpentes):Systematic Zoology,v.35,p.7–25.doi:10.1093/sysbio/38.1.7. Lee,M.S.Y.,andScanlon,J.D.,2002,Snakephylogenybasedon osteology,softanatomyandecology:BiologicalReviews,v.77, p.333–401.doi:10.1017/S1464793102005924. Lino,C.F.,Dias-Neto,C.M.,Trajano,E.,Gusso,G.L.N.,Karmann,I., andRodrigues,R.,1979,PaleontologiadascavernasdoValedo Ribeira,Explorac a oIAbismodoFo ssil(SP-145):Resultados parciais, in AtasdoSimpo sioRegionaldeGeologia,2th,RioClaro, SociedadeBrasileiradeGeologia,p.257–268. Lund,M.,1840,NouvellesrecherchessurlafaunefossileduBre sil: AnnalesdesSciencesNaturelles,Zoologie,ser.2,v.13,p.310–319. McDiarmid,R.W.,Campbell,J.A.,andToure ,T.A.,1999,SnakesSpecies oftheWorld:ATaxonomicandGeographicReference,Volume1: WashingtonD.C.,TheHerpetologist’sLeague,511p. A.S.H SIOU ,P.V. DE O LIVEIRA ,C.L.X IMENES AND M.S.S.V IANA JournalofCaveandKarstStudies, December2012 N 269

PAGE 38

Melgarejo,A.R.,2003,Serpentespec onhentasdoBrasil, in Cardoso,J.L.C., Franc a,F.O.S.,Wen,F.H.,Ma laque,C.M.S.,andHaddad,Jr.,V., eds.,AnimaisPec onhentosdoBrasil:Biologia,Cl nicaeTerape ˆutica dosAcidentes:Sa oPaulo,Sarvier,p.33–61. Nascimento,D.A.do.,Gava,A.,Pires,J.de.,andTeixeira,W.,1981, GeologiadafolhaSA.24–Fortaleza:ProjetoRadambrasil,DNPM, v.21,p.23–212. Noonan,B.P.,andChippindale,P.T.,2006,Dispersalandvicariance:the complexevolutionaryhistoryofboidsnakes:MolecularPhylogenetics andEvolution,v.40,p.347–358.doi:10.1016/j.ympev.2006.03.010. Nydam,R.L.,andCifelli,R.L.,2002,AnewteiidfromtheCedar MountainFormation(Albian-Cenomanianboundary)ofUtah: JournalofVertebratePaleontology,v.22,p.276–285.doi:10.1671/ 0272-4634(2002)022[0276:ANTLFT]2.0.CO;2. Nydam,R.L.,Eaton,J.G.,andSankey,J.,2007,Newtaxaoftransverselytoothedlizards(Squamata:Scincomorpha)andnewinformationon theevolutionaryhistoryof‘‘teiids’’:JournalofPaleontology,v.81, p.538–549.doi:10.1666/03097.1. Oliveira,P.V.de.,2010,Mam ferosdeNeopleistoceno–Holocenodo ParqueNacionaldeUbajara,Ceara [M.S.thesis]:PortoAlegre, UniversidadeFederaldoRioGrandedoSul,167p. Oliveira,P.V.de.,Ribeiro,A.M.,Kerber,L.,Lessa,G.,andViana, M.S.S.,(inpress)LateQuaternarycaviomorphrodents(Rodentia: Hystricognathi)fromCeara State,northeastBrazil:JournalofCave andKarstStudies. Passos,P.,2003,Sistema ticadocomplexo Epicratescenchria (Linnaeus, 1758),comaproximac o essobreafilogeniade Epicrates Wagler,1830 (Serpentes,Boidae)[M.S.thesis]:RiodeJaneiro,Universidade FederaldoRiodeJaneiro,125p. Passos,P.,andFernandes,R.,2008,Revisionofthe Epicratescenchria complex(Serpentes:Boidae):HerpetologicalMonographs,v.22, p.1–30.doi:10.1655/06-003.1. Paula-Couto,C.,1978,Mam ferosfo sseisdoPleistocenodoEsp rito Santo:AnaisdaAcademiaBrasileiradeCie ˆncias,v.50,p.365–379. Paula-Couto,C.,1980,FossilPleistocenetosub-recentmammalsfrom northeasternBrasil:I–Edentata,Megalonychidae.Anaisda AcademiaBrasileiradeCie ˆncias,v.52,p.144–151. Pianka,E.R.,andVitt,L.J.,2003,Lizards—WindowstotheEvolutionof Diversity:Berkeley,UniversityofCaliforniaPress,seriesOrganisms andEnvironments5,333p. Presch,W.,1974,Asurveyofthedentitionofthemacroteiidlizards (Teiidae:Lacertilia):Herpetologica,v.30,p.344–349. Quadros,M.L.E.S.,1996,Estudotectono-sedimentardaBaciade Jaibaras,naregia oentreascidadesdePacuja eJaibaras,noroeste doEstadodoCeara [M.S.thesis]:Bele m,UniversidadeFederaldo Para ,134p. Rage,J.-C.,1984,Serpentes,HandbuchderPala ¨oherpetologie,Part11: Stuttgart,GustavFisherVerlag,80p. Rage,J.-C.,2001,FossilsnakesfromthePaleoceneofSa oJose de Itabora ,Brazil.PartII.Boidae:Palaeovertebrata,v.30,p.111–150. Reeder,T.W.,Cole,C.J.,andDessauer,H.C.,2002,Phylogenetic relationshipsofwhiptaillizardsofthegenus Cnemidophorus (Squamata: Teiidae):atestofmonophyly,reevaluationofkaryotypicevolution,and reviewofhybridorigins:AmericanMuseumNovitates,no.3365,61p. Szyndlar,Z.,andRage,J.-C.,2003,Non-erycineBooideafromthe OligoceneandMioceneofEurope:Krako w,InstituteofSystematics andEvolutionofAnimals,PolishAcademyofSciences,109p. Trajano,E.,andFerrarezzi,H.,1995,Afossilbearfromnortheastern Brazil,withaphylogeneticanalysisoftheSouthAmericanextinct Tremarctinae(Ursidae):JournalofVertebratePaleontology,v.14, p.552–561.doi:10.1080/02724634.1995.10011577. Wu ¨ster,W.,Ferguson,J.E.,Quijada-Mascaren as,A.,Pook,C.E., Saloma o,M.C.,andThorpe,R.S.,2005,Tracinganinvasion: landbridges,refugia,andthephylogeographyoftheNeotropical rattlesnake(Serpentes:Viperidae: Crotalusdurissus ):Molecular Ecology,v.14,p.1095–1108.doi:10.1111/j.1365-294X.2005.02471.x. Ximenes,C.L.,andMachado,D.A.N.,2004,Diagno sticopaleontolo gico daProv nciaEspeleolo gicadeUbajara,EstadodoCeara , in Resumos doEncontroBrasileirodeEstudosdoCarste,1th,BeloHorizonte: RedespeleoBrasileAssociac a oBrasileiradeA guasSubterra ˆneas (ABAS),40p. Zaher,H.,1999,HemipenialmorphologyoftheSouthAmerican xenodontinesnakes,withaproposalforamonophyleticXenodontinaeandareappraisalofcolubroidhemipenes:Bulletinofthe AmericanMuseumofNaturalHistory,no.240,168p. L IZARDSANDSNAKES (L EPIDOSAURIA ,S QUAMATA ) FROMTHELATE Q UATERNARYOFTHESTATEOF C EARA INNORTHEASTERN B RAZIL 270 N JournalofCaveandKarstStudies, December2012

PAGE 39

DIETOFTHENEWT, TRITURUSCARNIFEX (LAURENTI, 1768),INTHEFLOODEDKARSTSINKHOLEPOZZODEL MERRO,CENTRALITALY A NTONIO R OMANO 1 ,S EBASTIANO S ALVIDIO 2 ,R OBERTO P ALOZZI 1,3 AND V ALERIO S BORDONI 1 Abstract: Karsthabitatshostahighnumberofspecializedorganismsthatcontributeto complexandpeculiarfoodwebs.Inundergroundaquatichabitats,vertebratesarethe toppredatorsthatstronglyinfluenceandregulatepreycommunities.Inthisstudy,the dietoftheItaliancrestednewt, Trituruscarnifex ,intheworld’sdeepestkarstphreatic sinkhole,thePozzodelMerroinLatium,centralItaly,wasanalyzed.Weobtainedboth stomachandfecalcontentsfromtwentyadultnewts(tenfemalesandtenmales)sampled insummer2010.Availabilityofpreyinthesinkholealsowasdetermined.Preyitems wereidentifiedandclassifiedintotenecologicalgroups.AtPozzodelMerro,duringthe summer,theaquaticstageof T.carnifex wasspecializedonthepre-imaginalstagesofthe smallChina-mark, Cataclystalemnata .Therecentlydescribedendemicstygobitic crustacean Nipharguscornicolanus wasnotfoundinstomachcontents.Finally,our resultsshowedthatanalysesofstomachandfecalcontentsmayprovidedifferent informationonthedietofnewtsintheiraquaticphase. I NTRODUCTION Floodedsinkholesaretypicalelementsofkarstand mayhostarelativelyhighle velofbiologicaldiversity (e.g.,SchmitterSotoetal.,2002).Amongvertebrates, fewamphibianspeciesarerestrictedtokarstenvironments(Weber,2004;Ko ¨hleretal.,2010;Sket,1997),but moretypically,theyutilizekarstasoneofavarietyof suitablehabitats.IntheMediterranean,karsthabitats aremainlyassociatedwithcarbonaterocks(Lewinand Woodward,2007).AlthoughMed iterraneanamphibians arewidespreadinsurfaceha bitatsinthekarst(RomanazziandBonato,2011;Romanoetal.,2010;SchmitterSotoetal.,2002),fewstudieshaveexaminedtheir ecologyinkarstecosystems(e.g.,Schabetsbergerand Jersabek,1995;Schabetsbergeretal.,1995).Scanty ecologicalinformationisavailableforamphibianpopulationslivinginfloodedsinkholes. InEurope,mostamphibiansarestrictlyprotectedby theEuropeandirective92/43/EEC(theso-calledhabitats directive),whichisaimedatthecreationofanecological networkinEuropetopreservebiodiversity.Amongnewts thatoccurinItaly,onlytheItaliancrestednewt, Triturus carnifex (Laurenti,1768),isinannexIIofthehabitats directivethatlistsanimalsofcommunityinterestwhose conservationrequiresthedesignationofspecialareasof conservation. Whenastrictlyprotectedamphibianspeciesandan endemicorrareinvertebratespeciescoexist,theirecologicalrelationshipsaspredatorandpreyareofimportance forplanningadequateconservationmeasures.Inthestudy site,adeep,floodedsinkholeincentralItaly(Fig.1), ecologicalstudieswereparticularlyrequiredbecause, beginningin2003,theinvasiveaquaticBraziliantropical fern Salviniamolesta D.S.Mitchellspreadandsoon coveredthesurfaceofthelakeinthebottomofthestudy site,replacingthepreviouslydominant Lemnaminor L. withanapproximately5cmthickvegetativelayerupon whichterrestrialvegetationhadbeguntogrow(Giardini, 2003,2004).Theecologicalconsequencesofthisrampant invasionwereunknown,andinMarch2009a Salvinia molesta eradicationprogramstarted,asrecommendedin Giardini(2003).Asaresult, L.minor hadlargelybeen reestablishedasthedominantwatersurfacevegetationby spring2009. WestudiedthedietoftheItaliancrestednewtto determinewhethertheendemiccrustaceanrecently discoveredinPozzodelMerrowaspreyeduponby newts,toexaminewhetherthenewtsshowselectivityin feeding,andtocomparetheinformationobtained analyzingbothstomachandfecalcontents.Wealso checkedtoseewhetherthesexofthenewtsbiasedtheir choiceofprey. S TUDY S ITE ThesinkholePozzodelMerroislocatedat140ma.s.l., onthesouthernslopesofCornicolaniMountains(Lat.N 42 u 02 9 21 0 ,Long.E12 u 40 9 55 0 ,Latium,centralItaly)andis includedinthenaturalreserveMacchiadiGattacecae MacchiadelBarco(Fig.1a).Thesinkholeisthedeepest *Correspondingauthor:antonioromano71@gmail.com 1 DipartimentodiBiologia,Universita `diRoma‘‘TorVergata’’,ViaDellaRicerca Scientifica,I-00133Roma,Italy 2 DIP.TE.RIS.,Universita `diGenova,CorsoEuropa,26,I-16132Genova,Italy 3 DAF,Dipartimentoditecnologie,ingegneriaescienzedell’Ambienteedelle Foreste,Universita `diViterbo‘‘Tuscia’’,ViaS.C.deLellis,01100,Viterbo,Italy A.Romano,S.Salvidio,R.Palozzi,andV.Sbordoni–Dietofthenewt, Trituruscarnifex (Laurenti,1768),inthefloodedkarstsinkhole PozzodelMerro,centralItaly. JournalofCaveandKarstStudies, v.74,no.3,p.271–277.DOI:10.4311/2011JCKS0208 JournalofCaveandKarstStudies, December2012 N 271

PAGE 40

floodedkarstsinkholeintheworld(about460m; Caramanna,2002;Garyetal.,2003);itsfloodedpart extendsatleast392mbelowthewatertable(Fig.1b). PozzodelMerroisfunnel-shaped,withadiameterof about160matgroundlevel,narrowingto25matthe watersurfaceatabout70mdepth(Fig.1b,d).Belowthis depththefloodedsinkholecontinuesdownasanearly verticalconduitwithseveralminorcavitiesalongthewalls (Fig.1b).Thewatertemperatureof15.9 u Cisconstant throughoutthesinkhole(Palozzi,etal.,2010).Thewater chemistryisbicarbonate-calcic,andsamplescollected downthewatercolumnshowedaprogressivereduction ofpHfromneutralityuptoaminimumvalueof6.57at about100m(Palozzietal.,2010). B IODIVERSITY A SSESSMENT In2005,anendemicspeciesofamphipodcrustacean wasdescribedfromPozzodelMerro, Nipharguscornicolanus Iannilli&VignaTaglianti,2005.Fouramphibians arefoundthere,theItaliancrestednewt Trituruscarnifex (Laurenti,1768),thesmoothnewt Lissotritonvulgaris (Linnaeus,1758),theApenninefrog Ranaitalica Dubois 1987,andthecommontoad Bufobufo (Linnaeus,1758). Howeverthetwoanuransarefoundonlysporadically, while T.carnifex isthedominantnewt(captureratio T.carnifex/L.vulgaris 5 30:1,A.Romano,unpublished data).TheItaliancrestednewtisalarge-bodiednewtthat occursinItaly,southernSwitzerland,Slovenia,Istria,and someregionsofAustria,theCzechRepublic,andHungary Figure1.(a)Locationofthestudysite,thefloodedsinkholePozzodelMerro,incentralItaly(whitesquare).(b)Vertical geologicsection.(c)Thesinkholeasseenfromabove.(d)Samplingofthenewt Trituruscarnifex withalong-handleddipnet maneuveredfromarubberboat. D IETOFTHENEWT T RITURUSCARNIFEX (L AURENTI ,1768), INTHEFLOODEDKARSTSINKHOLE P OZZODEL M ERRO CENTRAL I TALY 272 N JournalofCaveandKarstStudies, December2012

PAGE 41

(seeVannietal.,2007). Trituruscarnifex prefersdeepand permanentwaterbodies(AndreoneandMarconi,2006; Vannietal.,2007). M ETHODS S AMPLING M ETHOD Forthisstudy,onlyadultnewtsweresampled.TwentytwonewtswerecapturedonAugust12,2010,withalonghandleddipnet3.5minlength,maneuveredfromaboat (Fig.1d).Newtswerecapturedwhensurfacingtobreathe. Newtsweremeasuredbysnout-ventlength,weighed, anesthetizedintricainemethanesulphonate(MS-222; Novartis).Theirstomachswereflushedusinga20ml syringefilledwithwaterjoinedtoasiliconecatheter1.0mm indiameter.Theflushingwasrepeateduntilnofurther contentcameout(Joly,1987;Salvidio,1992;Vignolietal., 2007).Twonewtshademptystomachs,leavingtwenty subjecttofurtheranalysis.Fooditemsstillpresentinthe oralcavityafterflushingwerecarefullyremovedby entomologicalforceps.Newtswerehousedinanaquarium forapproximatelysixteenhoursafterflushingtoverify theirreturntonormalactivityandthenreleasedatthe capturesite.Fecalmatterproducedbythenewtsduring theirhousingwerealsocollectedtoobtaindigestedprey. Thusfeedingmaterialwasobtainedfromtwosources, stomachcontents(STO)andfeces(FEC).STOwere individuallyassigned,butsincenewtswerehousedinthe sameaquariumallFECwerepooled. Preyavailability(AVA)wassampledintheaquatic environmentofthesinkholebyadipnetmeasuring60cm by30cmwitha0.1mmmesh.Weattemptedtoobtain samplesthatwererepresentativeofthethreegroupsinto whichaquaticinvertebratesmaybedivided.Weusedthe dipnettosamplepleustonicinvertebratesnearthewater surface,nektonicinvertebrateswithinthetop3mofthe watercolumn,andbenthicinvertebratesfromthetop3m onthewallsofthesinkholebelowthewatersurface.All thosehabitatsweresampledtentimes.STO,FEC,and AVAsampleswerepreservedin90%ethanol,and identificationwasmadeusingastereomicroscope. Thenewtsexratiowasexpressedastheproportionof maturemales:males/(males + females)(WilsonandHardy, 2002).Thevacuityindexwascalculatedasthepercentage ofemptystomachsoutofthetotalexamined.Because varianceswerenothomogeneous,Welch’stestforunequal varianceswasusedtodeterminemales-femalessnout-vent lengthandweightdifferences.TheMann-WhitneyUtest wasusedtodisclosedifferencesinthenumberofprey swallowedbymalesandfemales.Thenon-parametric Spearman’srankcorrelationwasusedtodetectany differencesinpreybetweenthetwosources,stomachand feces,andbetweenprey(STO + FEC)andtrophic availability.Tocomparetheresultsobtainedusingtwo differentsources(STOandFEC)andtodetectsex-related feedingdifferences,threediversityindiceswereused,the numberoftaxa,theSimpsonindex,whichmeasures evennessofthecommunityfrom0to1andisthemostusedstatistictocomparesmallsamples(Magurran,2004), andtheanalysisofsimilarity(ANOSIM)ontheBray– Curtisdissimilaritymeasure,whichiswidelyusedto analyzeanimalcommunities,andinparticular,thoseof freshwaterinvertebrates,withBonferronicorrection.ANOSIMisanon-parametricsignificancetestbetweentwoor moregroups,basedonanydistancemeasure(Clarke, 1993).Costello’s(1990)graphicalrepresentation,modified byusingprey-specificabundance(PI)insteadofpercentage abundance(Amundsenetal.,1996),wasmadetoestimate theimportanceofparticularfooditemsinthedietofthe newtsandtoassesstheirforagingstrategy.Thismethod classifiespreyselectionbyplottingprey-specificabundance PIontheY-axisagainstfrequencyofoccurrenceontheXaxisinthepredatorstomachs(Fig.2).PIisdefinedasthe proportionofapreytypeamongofallpreyitemsinonly thoseindividualsinwhichthatpreytypeoccurs(Amundsenetal.,1996).Thisgraphicalapproachallowsfor determinationofpreyimportance,thefeedingstrategyof thepredator,andthetwocomponentsthatcontributeto thepopulation’stotalnichewidth,within-phenotypicand between-phenotypic.ThemodifiedCostello’sgraphprovidesinformationonfeedingpatternsthatmightnotbe inferredfromsingle-dietindexes. Figure2.Feedingstrategy,basedonstomachcontents,of thenewt Trituruscarnifex inthePozzodelMerrovisualized usingtheCostellographicasmodifiedbyAmundsenetal. (1996).Thesmallsquareshowstheinterpretationoftheplot, whereBPCisbetween-phenotypecomponentofthepopulation’stotalnichewidthandWPCisthewithin-phenotype component.Sometaxawithlowabundanceandlow frequencyofoccurrenceinthepredatedsamplesareomitted forclarity. A.R OMANO ,S.S ALVIDIO ,R.P ALOZZI AND V.S BORDONI JournalofCaveandKarstStudies, December2012 N 273

PAGE 42

R ESULTS Nonewtmortalitywasobservedduringorafter stomachflushing.Thesexratioof Trituruscarnifex was 0.5( n 5 22)andthestomachvacuitywas9%.Thus,we obtaineddataonthedietoftenmaleandtenfemale Italiancrestednewts.Sexesdidnotdiffersignificantlyin bodysizeandweight( n 5 20;snout-ventlength 5 6.37 6 0.44cmand6.20 6 0.25cmforfemalesandmales, respectively;weight 5 8.26 6 1.80gand8.33 6 1.20gfor femalesandmales,respectively; p 0.1forallcomparisons). Theresultsofsamplingavailablepreyarepresentedin Table1.Basedontheirtaxonomyandecology,terrestrial oraquatic,thepreydetectedinstomachcontentswere dividedintoninegroupsplusundeterminedinsects (Table2).Overall,lepidopterans( Cataclystalemnata )and dipteranswerepreyeduponbynewtsmostfrequently(95% and55%respectively;Table2).Invertebratesthatwould appearonthehorizontalaxisinFigure2arethosethatare availableintheenvironmentbutwerenoteatenbythe newts. Theaveragenumberofpreyperstomachwas7.55 6 11.66and5.22 6 8.52formalesandfemales,respectively, anddidnotdiffersignificantly(Mann-Whitney,U 5 38, p 5 0.82).Therewasnooveralldifferenceinthediet compositionbetweenthesexes(ANOSIM, n 5 20;global R 5 2 0.06, p 5 0.92).Table2showsthenumbersineach preygroupfoundinthestomachsofmalesandfemales. Table3showsthediversityindexesforstomachandfeces separately,withthesexespooled,andforstomachcontents forthesexesseparately.ThecorrelationbetweenSTOand FECwasnotsignificant(flyingdipterawaspooledwiththe dipterathatwerenotTipulidae,giving n 5 9; r 5 0.34, p 5 0.37),suggestingthatsomepreygroupsweredegradated duringdigestion. Theoverallrichnessofthetypesofinvertebrateslisted inTable1inthedietof Trituruscarnifex ,poolingthe stomachandfecalcontents,waslessthantrophicavailability( n 5 15; r 5 2 0.17, p 5 0.95).Thenumbersof groupspreyeduponwaseightandfive,respectively,for femalesandmales(Table2).InthemodifiedCostelloÂ’s graphicrepresentation(Fig.2),almostallpreyitemsarein thelowerleftcorner,while Cataclystalemnata isinthe upperrightquadrant. D ISCUSSION TheoccurrenceinthesinkholePozzodelMerroofthe endemicamphipod Nipharguscornicolanus andthenewt Trituruscarnifex wasthemainmotivationtostudythe trophicnicheoftheItaliancrestednewt.Ifapreypredatorrelationshipwasdemonstrated,conservation implicationsconcerningthesespecies,theformerstrictly endemicandthelatterprotectedbytheEuropean Directive92/43/EEC,wouldhavetobeconsidered. Niphargus specimensareoftenpreyeduponbynewts, andingeneral,byaquaticsalamanders,incaseswhere theysharethesamemicrohabitat(Raccaetal.,2002; SchabetsbergerandJersabek,1995).However,inthe PozzodelMerro, N.cornicolanus isfoundmainlybetween 10and20mofdepth,andoccasionallyupto74m (Palozzietal.,2010;R.Palozzi,unpublisheddata).Inthe studysite, T.carnifex hasbeenfound(byRP)onlyupto7 or8mdeep,butdefinitedataonthisissuearelacking. Sincemanynewtsfeedatmorethan8mofdepth (SchabetsbergerandJersabek,1995),andatleastupto 12m(Georgeetal.,1977),asmalloverlapinthedepth distributionsof T.carnifex and N.cornicolanus seems possible.However, Nipharguscornicolanus wasnotfound inthenewtdiet.Thepresenceof Niphargus inthenewt dietcannotbecompletelyrejected,duetoourlimited samplesize.However,evenifitispreyedupon,the amphipodappearstobearelativelyrarepreyitem,andits conservation,inrelationtothepotentialpredationofthis largeItaliannewt,shouldnotrequirespecificmeasuresor managementplans. Trituruscarnifex feedsprimarilyonaquaticanimals andonterrestrialarthropodsfallingonthewater surface,showingageneralistdietaryhabit(Fasolaand Canova,1992;AnconaandBolzern,1993;Vignolietal., 2009).However,thefeedi ngstrategyoftheItalian crestednewtinthePozzodelMerrorevealedthatthere isadominantfooditem,the larvaeofthelepidopteran Cataclystalemnata ,thepointforwhichisintheupper rightcornerofFigure2andwhichhashighspecific abundanceinthewaterandhighoccurrenceinsamples ofpreyremains.Thepopulationof T.carnifex inhabitingthesinkholePozzodelMerroisspecializinginthis onepreytype.Almostallindividualshadbeenfeeding onthedominantprey,butasignificantproportionof otheravailablepreywasincl udedoccasionallyinthediet ofsomeindividuals(Table1 ;Fig.2).Sincepreytypes rangefromtheupperrighttothelowerleftinFigure2, thenewtpopulationpossesse sarelativelybroadtrophic niche.Thedietofmalesandfemaleswassimilar,but thenumberoftaxainthefemalestomachswastwice thatofthemales,andtheconfidencelimitofthenumber oftaxadiversityindex(Table3b)arejustslightly overlapping,suggestingtha tanalysisofalargersample couldallowadifferencetobestatisticallyconfirmed. Thecomparisonofthedataobtainedfromfecaland stomachanalysesisalsointeresting.Indeed,thestudyof fecalsamplesinsalamande rsmayleadtoadrastic underestimateinthenumberanddiversityofpreytaxa (Table3a),becausesmallanddelicatepreytendtobe completelydigestedbysalamanders(Corvettoetal., 2012).Thisselectivediges tionmaybeaproblemwhen searchingforspecificprey-predatorrelationshipsina givencontext.Therefore,theuseofstomachflushing shouldbepreferredtofec alsampling,whichdoesnot assureacompleteevaluationofasalamanderÂ’sdiet. D IETOFTHENEWT T RITURUSCARNIFEX (L AURENTI ,1768), INTHEFLOODEDKARSTSINKHOLE P OZZODEL M ERRO CENTRAL I TALY 274 N JournalofCaveandKarstStudies, December2012

PAGE 43

Table1.Trophicavailability(AVA)inthePozzodelMerroanddietcompositionoftwenty Trituruscarnifex asdeterminedfromtheanalysesoftheirstomach contents(STO)andfeces(FEC).Alltaxaorlifestagesareaquaticexceptwhereindicatedbyanasterisk(*). Taxon LifeStage Total (prey) LarvaePupaeAdult Class/OrderFamilyAVASTOFECAVASTOFECAVASTOFEC I NSECTA Collembola 1872 LepidopteraPiralidae(1sp.)1139595522153138 DipteraTipulidae9251 26 DipteranotTipulidae441*4 Ephemeroptera11 1 ColeopteraElmintidae 17112 Coleoptera* 11 undetermined 145 C RUSTACEA IsopodaAsellidae 61235 Concostraca 1 A RACHNIDA Acarina 131 G ASTEROPODA PulmonataPlanorbidae 5 T URBELLARIA SeriataPlanariidae 15 N EMATODA undetermined 111 O LIGOCHAETA undetermined 8 A.R OMANO ,S.S ALVIDIO ,R.P ALOZZI AND V.S BORDONI JournalofCaveandKarstStudies, December2012 N 275

PAGE 44

C ONCLUSIONS Thispaperisthefirststudyonthefeedingbehaviorofa newtinadeep,floodedkarstsinkhole.Apreviousstudyon thedietofnewtsduringthe2003–2009invasionof Salvinia molesta wouldhavebeenusefultodeterminetheeffectof thisinvasiveplantonthenewtpopulation.Theimportance ofthenativeaquaticplant Lemnaminor insustainingthe populationof Trituruscarnifex wasstronglysuggestedby thepresentstudy.Alterationstothenaturalstatusofthis karstecosystem,suchasthepermanentreplacementof autochthonous Lemna with Salvinia ,couldlikelyleadto ecologicalconsequences,becausethemostimportance foodresourceusedbynewtsinthesummerwouldbe unavailable. OtherItaliancrestednewtpopulationslivingin differentecosystemssuchasshallowkarstpondsandother lakesseemtobecharacterizedbyanopportunisticfeeding behavior(Vannietal.,2007).Astudyofapopulationfrom afloodedtuffquarryabandonedincentralItalysuggested the Trituruscarnifex mayhaveamixedfeedingstrategy (dominant-generalized)andsomeindividualsmayfeedon preywitharelativelyhighfrequencyofoccurrence(Vignoli etal.,2009).ThepopulationlivinginthePozzodelMerro exhibited,atleastinthesummer,adifferentfeeding behavior,becauseallnewtswerespecializedpredatorsof aninvertebratewithhighfrequencyofoccurrence. Generalistfeedingbehaviorof T.carnifex inhabitatswith variableenvironmentalparametersmayvaryslightlyover seasonsandyears,dependingonresources(Anconaand Bolzern,1993).Althoughinthesinkholephysicaland chemicalparametersarerelativelystable(Palozzietal., 2010),furtherresearchshouldaimatestablishingthe feedingstrategyadoptedbytheItaliancrestednewts duringthefallandwinter,whentheplant Lemnaminor and theassociatedsmallChina-mark Cataclystalemnata are absent.Furthermore,inwintertime,pleustonicpreywould becomelessavailable,andthenewtsmightmovedeeperin thewatertoforageon Niphargus Finally,theanalysisofthedietofthesmallersmoothnewt Lissotritonvulgaris andof Trituruscarnifex larvae,whichmay showaspecializationtowardsaquaticcrustaceans(Vignoli etal.,2009;StochandDolce,1 984),shouldalsobeundertaken tocompletelyrevealthetrophicrelationshipsbetweenthe endemic Nipharguscornicolanus andthesympatricamphibians whichliveinthePozzodelMerro. Table2.SummaryofthenumberofspecimensofpreyfoundinstomachsonlyforgroupsofthetaxainTable1,sortedby female(FF)andmale(MM).Thepercentageofthetennewtsofeachspeciesthathadeateneachgroupisalsogiven. Organism FFMMTotal N,prey%ofnewtN,prey%ofnewtN,prey%ofnewt Lepidoptera( Cataclystalemnata )3790431008095 DipteraTipulidae(larvae)113020502140 DipteranonTipulidae(larvae)510320815 FlyingDiptera(terrestrialandnotTipulidae)1100015 Ephemeroptera(larvae)1100015 Coleoptera(aquatic);1100015 Coleoptera(terrestrial)1100015 Insectaundetermined1100015 Crustacea,Asellidae0021025 Nematoda0011015 Total5869117 Table3.Indexesofpreytaxadiversity,with95%confidencelimits,of Trituruscarnifex inthePozzodelMerro(CentralItaly). IndexType StomachandFecalContentsComparedfor PooledMaleandFemaleSpecimens StomachandFecalContentsComparedfor FemalesandMales STO(sexespooled)FEC(sexespooled) DiversitySTO FemalesDiversitySTOMales Boostrap (95%) Boostrap (95%) Boostrap (95%) Boostrap (95%) Min.Max.Min.Max.Min.Max.Min.Max. IndexofTaxaDiversity959535838434 SimpsonIndex0.4760.3710.5520.2450.1140.380.3920.2260.5390.5110.3940.590 D IETOFTHENEWT T RITURUSCARNIFEX (L AURENTI ,1768), INTHEFLOODEDKARSTSINKHOLE P OZZODEL M ERRO CENTRAL I TALY 276 N JournalofCaveandKarstStudies, December2012

PAGE 45

A CKNOWLEDGEMENTS Captureandstomachflushingwereperformedwiththe permitfromtheMinisterodell’AmbienteedellaTutela delTerritorioedelMare(permitreference:DPN/2010/ 0002708).TheProvinceofRomeprovidedtheresearch grantforthisstudy.WeareindebtedtoGiovanniForcina forhelpinfieldsampling.Wethanktwoanonymous refereesformanyusefulsuggestionsthatsurelyimproved ourmanuscript. R EFERENCES Amundsen,P.-A.,Gabler,H.-M.,andStaldvik,F.J.,1996,Anew approachtographicalanalysisoffeedingstrategyfromstomach contentsdata—modificationoftheCostello(1990)method:Journal ofFishBiology,v.48,no.4,p.607–614.doi:10.1111/j.10958649.1996.tb01455.x. Ancona,N.,andBolzern,A.M.,1993,Alimentazionedi Trituruscarnifex e T.vulgarismeridionalis (AmphibiaCaudata)induestagni dell’AppeninoLigure:RendicontiScienzeChimicheeFisiche,Geologiche,BiologicheeMediche,IstitutoLombardo,v.B126, p.123–137. Andreone,F.,andMarconi,M.,2006, Trituruscarnifex (Laurenti,1768), in Sindaco,R.,Doria,G.,Razzetti,E.,andBernini,F.,eds.,Atlante degliAnfibiedeiRettilid’Italia/AtlasofItalianAmphibiansand Reptiles:Firenze,SocietasHerpetologicaItalica,EdizioniPolistampa, p.220–225. Caramanna,G.,2002,Exploringoneoftheworld’sdeepestsinkholes:the PozzodelMerro(Italy),UnderwaterSpeleology,February,p.4–8. Clarke,K.R.,1993,Non-parametricmultivariateanalysisofchanges incommunitystructure:AustralianJournalofEcology,v.18, p.117–143.doi:10.1111/j.1442-9993.1993.tb00438.x. Costello,M.J.,1990,Predatorfeedingstrategyandpreyimportance:anew graphicalanalysis.JournalofFishBiology,v.36,no.2,p.261–263. doi:10.1111/j.1095-8649.1990.tb05601.x. Crovetto,F.,Romano,A.,andSalvidio,S.,2012,Acomparisonoftwo non-lethalmethodsfordictarystudiesinterrestrialsalamanders. WildlifeResearch,v.39,p.266–270. Fasola,M.,andCanova,L.,1992,Residenceinwaterbythenewts Triturusvulgaris T.cristatus and T.alpestris inapondinnorthern Italy:Amphibia-Reptilia,v.13,no.3,p.227–233.doi:10.1163/ 156853892X00436. Gary,M.O.,Sharp,J.M.Jr.,Caramanna,G.,andHavens,R.H.,2003, Volcanicallyinfluencedspeleogenesis:formingElSistemaZacato n, Mexico,andPozzodelMerro,Italy,thedeepestphreaticsinkholesin theworld:GeologicalSocietyofAmericaAbstractswithPrograms, vol.34,no.7,session19-4,52p. George,C.J.,Boylen,C.W.,andSheldon,R.B.,1977,Thepresenceofthe red-spottednewt, Notophthalmusviridescensrafinesque (Amphibia, Urodela,Salamandridae)inwatersexceeding12metersinLake George,NewYork:JournalofHerpetology,v.11,no.1,p.87–90. Giardini,M.,2003,Notesullabiologia,l’ecologiaelemodalita `di controllodi Salviniamolesta D.S.Mitchell(Salviniaceae),specie infestantenuovaperilLazio:RivistadiIdrobiologia,v.42,no.1–3, p.263–282. Giardini,M.,2004, Salviniamolesta D.S.Mitchell(Salviniaceae):seconda segnalazioneperl’Italia(Lazio)econsiderazionisulcontrollodi questaspecieinfestante:Webbia,v.59,no.2,p.457–467. Weber,A.,Amphibia, in Gunn,J.,ed.,EncyclopediaofCavesandKarst Science:NewYork,FitzroyDearborn,p.61–62. Iannilli,V.,andVignaTaglianti,A.,2005,Newdataongenus Niphargus (Amphipoda,Niphargidae)inItaly,withthedescriptionofanew speciesofthe Orcinus group:Crustaceana,v.77,no.10(2004), p.1253–1261.doi:10.1163/1568540043166029. Joly,P.,1987,Lere gimealimentairedesamphibians:me thodesd’e tude: Alytes,v.6,no.1–2,p.11–17. Ko ¨hler,J.,Vences,M.,D’Cruze,N.,andGlaw,F.,2010,Giantdwarfs: discoveryofaradiationoflarge-bodied‘stump-toedfrogs’fromkarstic caveenvironmentsofnorthernMadagascar:JournalofZoology, v.282,no.1,p.21–38.doi:10.1111/j.1469-7998.2010.00708.x. Lewin,J.,andWoodward,J.C.,2007,Karstgeomorphologyand environmentalchange, in Woodward,J.,ed.,ThePhysicalGeography oftheMediterranean:Oxford,OxfordUniversityPress,p.287–317. Magurran,A.E.,2004,MeasuringBiologicalDiversity:Malden,Massachusetts,BlackwellPublishing,256p. Palozzi,R.,Caramanna,G.,Albertano,P.,Congestri,R.,Bruno,L., Romano,A.,Giganti,M.G.,Zenobi,R.,Costanzo,C.,Valente,G., Polani,D.,Vecchio,M.,Vinci,M.,andSbordoni,V.,2010,The underwaterexplorationoftheMerrosinkholeandtheassociated divingphysiologicalandpsychologicaleffects:UnderwaterTechnology,v.29,no.3,p.125–134.doi:10.3723/ut.29.125. Racca,L.,Bressi,N.,Dolce,S.,andGiacoma,C.,2002,Studioetologico delcomportamentoalimentaredi Proteusanguinusanguinus (Amphibia,Urodela)incattivita `:AttiMuseoCivicodiStoriaNaturaledi Trieste,v.49,p.157–167. Romanazzi,E.,andBonato,L.,2011,AnfibisulMontello:Distribuzione deisitiriproduttiviinunterritoriocarsicoprealpino, in Bon,M., Mezzavilla,F.,andScarton,F.,eds.,Attidel6 u Convegnodei FaunistiVeneti:BollettinodelMuseocivicodiStoriaNaturale, Venezia,suppl.to,v.61,p.88–95. Romano,A.,Ventre,N.,DeRiso,L.,Pignataro,C.,andSpilinga,C., 2010,Amphibiansofthe‘‘CilentoeVallodiDiano’’NationalPark (Campania,SouthernItaly):updatedchecklist,distributionand conservationnotes.ActaHerpetologica,v.5,no.2,p.233–244. Salvidio,S.,1992,Dietandfoodutilizationinarockfacepopulationof Speleomantesambrosii (Amphibia,Caudata,Plethodontidae):Vieet Milieu,v.42,no.1,p.35–39. Schabetsberger,R.,andJersabek,C.D.,1995,Alpinenewts( Triturusalpestris ) astoppredatorsinahigh-altitudeka rstlake:dailyfoodconsumptionand impactonthecopepod Arctodiaptomusalpinus :FreshwaterBiology,v.33, p.47–61.doi:10.1111/j.1365-2427.1995.tb00385.x. Schabetsberger,R.,Jersabek,C.D.,andBrozek,S.,1995,Theimpactof Alpinenewts( Triturusalpestris )andminnows( Phoxinusphoxinus )on themicrocrustaceancommunitiesoftwohighaltitudekarstlakes: Alytes,v.12,no.4,p.183–189. Schmitter-Soto,J.J.,Comin,F.A.,Escoba r-Briones,E.,Herrera-Silveira,J., Alcocer,J.,Sua rez-Morales,E.,Elias-Gutie rrez,M.,D az-Arce,V.,Mar n, L.E.,andSteinich,B.,20 02,HydrogeochemicalandbiologicalcharacteristicsofcenotesintheYucatanPeni nsula(SEMexico):Hydrobiologia, v.467,no.1–3,p.215–228.doi:10.1023/A:1014923217206. Sket,B.,1997,Distributionof Proteus (Amphibia:Urodela:Proteidae) anditspossibleexplanation:JournalofBiogeography,v.24,no.3, p.263–280.doi:10.1046/j.1365-2699.1997.00103.x. Stoch,F.,andDolce,S.,1984,Alimentazioneerapportialimentaridi Triturusalpestrisalpestris (Laur.), Trituruscristatuscarnifex (Laur.)e Triturusvulgarismeridionalis (Boul.).(Osservazionisull’alimentazione degliAnfibi:III):QuaderniETP,RivistadiLimnologia,Udine,v.9, p.17–28. Vanni,S.,Andreone,F.,andTripepi,S.,2007, Trituruscarnifex (Laurenti, 1768), in Lanza,B.,Andreone,F.,Bologna,M.,Corti,C.,and Razzetti,E.,eds.,Faunad’ItaliaXLII:Amphibia,Bologna,Edizioni Calderini,p.265–272. Vignoli,L.,Bombi,P.,D’Amen,M.,andBologna,M.A.,2007,Seasonal variationinthetrophicnicheinaheterochronicpopulationof Mesotritonalpestrisapuanus (Amphibia,Salamandridae)fromthe south-westernAlps:HerpetologicalJournal,v.17,p.183–191. Vignoli,L.,Luiselli,L.,andBologna,M.A.,2009,Dietarypatternsand overlapinanamphibianassemblageatapondinMediterranean centralItaly:VieetMilieu/LifeandEnvironment,v.59,no.1, p.47–57. Wilson,K.,andHardy,I.C.W.,2002,Statisticalanalysisofsexratios:an introduction, in Hardy,I.C.W.,ed.,SexRatios–Conceptsand ResearchMethods:Cambridge,CambridgeUniversityPress, p.48–92. A.R OMANO ,S.S ALVIDIO ,R.P ALOZZI AND V.S BORDONI JournalofCaveandKarstStudies, December2012 N 277

PAGE 46

HUMANURINEINLECHUGUILLACAVE:THE MICROBIOLOGICALIMPACTANDPOTENTIAL FORBIOREMEDIATION M ICHAEL D.J OHNSTON ,B RITTANY A.M UENCH ,E RIC D.B ANKS AND H AZEL A.B ARTON DepartmentofBiology,UniversityofAkron,Akron,OH44325 Abstract: Duringextendedexplorationtripsincaves,itissometimesnotpossibleto removeallexcretedurineduetoitsvolumeandweight.Excessurinecanbeparticularly problematicindrycaveswhere,withoutdilution,urineintroducesasignificantsourceof nitrogenintotheseotherwisenitrogen-limitedenvironments.Itwastheaimofthisstudy todeterminetheimpactthathumanurinecouldhaveoncavemicrobiotaoveran extendedperiodoftime.Todothis,weexaminedthemicrobialcommunitystructureofa heavilyimpactedsiteinLechuguillaCave,USA.Usingamolecularphylogenetic approachwegenerateda136-member16SrRNAclonelibrarythatdemonstrated representativesofthe Alpha-Betaand Gammaproteobacteria ,the Bacteroidetes Firmicutes Actinobacteria and Deinococcus-Thermus groupatthissite.Thestructure ofthemicrobialcommunityattheimpactedsitesuggeststhatitiscolonizedbyendemic cavespeciesratherthanhumancommensalorganisms,whilemetabolicinference suggeststhattheseorganismsaretakingadvantageofboththenitrogenousandorganic compoundsinurineforgrowth.Theintrinsicnatureofsuchmetabolicactivityinthe caveenvironmentwasconfirmedbyexaminingnon-impactedsitesusingcultivation, whichdemonstratedthatendemicspeciesexpressboththecapacitytodegradeurineand toreduceureatonitrogengas.Ourresultsdifferfromthoseofpreviousstudiesby implyingamoreresilientnatureofthemicrobialecosystemincavestoinvasionby exogenous(commensal)species,whilesuggestingthatendemicmicrobialspeciesmaybe abletomitigatetheimpactofexcessnitrogeninthecavethroughbioremediation. I NTRODUCTION Caveenvironmentsbecomemoreenergy-andnutrientlimitedasyoutravelfartherfromsurfaceinputs(Barr, 1967;HardinandHassell,1970;RaeslyandGates,1987; CulverandSket,2000;Lavoieetal.,2007),andthe environmentbecomesincreasinglydifficultfortroglobitic speciestosubsist.Asaresult,thebiomebecomes increasinglydominatedbymicroorganisms(Barr,1967; CulverandSket,2000;BartonandJurado,2007).Yetin thesenutrient-limitedenvironments,evenmicroorganisms aredependentontheavailabilityofbothenergyand nutrientssuchasnitrogen,sulfur,andphosphorusfor growth.Thepresenceofthesenutrients,inbothendogenous(autochthonous)andexogenous(allochthonous) form,canhaveaprofoundimpactonmicrobialgrowth andcommunitystructure(Ikneretal.,2007;Summers Engeletal.,2010;Ikeretal.,2010).Dependingonlocal geochemistry,sulfurandphosphorusmaybepresent withinthemineralmatrixoftherock;however,thereis rarelyanyintrinsicsourceofnitrogenincaves(Klimchouk, 2000).Asaresult,theavailabilityofnitrogenbecomes criticalformicrobialsubsistence,asdemonstratedbythe abundanceofnitrogen-fixingbacterialspeciesincave environments(Northupetal.,2003;Bartonetal.,2007; Ikneretal.,2007;Spearetal.,2007).Theintroductionof allochthonousnitrogencanhaveadramaticimpactonthe microbialcommunitystructureinsuchecosystems,created byanexcessofthisnormallylimitednutrient(Ikneretal., 2007;Ikeretal.,2010). Inordertounderstandcaves,humansmustenterand explorethem(Kambesis,2007);however,thehuman explorationofcaveshasthepotentialtoimpactmicrobial ecosystemsthroughtheintroductionofnutrients(Northup etal.,1997;LavoieandNorthup,2005;Ikneretal.,2007). Tominimizesuchimpacts,speleologistsuseanumberof techniques,includingstayingondesignatedtrails,wearing non-markingfootwear,avoidingleavingfoodcrumbs,and removingallwaste(Northupetal.,1997;Elliott2006). Somecavesaresufficientlylargeanddeepthatexploration requiresextendedperiodsunderground,sometimesexceedingthirtydays(Reamesetal.,1999;Stoneetal.,2002; Tabor,2010),andtheremovalofhumanwastebecomes impracticalduetoweightconstraints.Ifremovingallliquid wasteismandated,thereisadangerthatspeleologistswill undergoself-induceddehydrationtominimizewasteand weight.Dehydration,compoundedbyexercise,canadverselyaffectjudgmentandmotorskills,increasingthe likelihoodofanaccident(Lieberman,2007).Toavoidthe large-scaleimpactthatarescuewouldhave,itistherefore *CorrespondingAuthor:bartonh@uakron.edu M.D.Johnston,B.A.Muench,E.D.Banks,andH.A.Barton–HumanurineinLechuguillaCave:themicrobiologicalimpactand potentialforbioremediation. JournalofCaveandKarstStudies, v.74,no.3,p.278–291.DOI:10.4311/2011MB0227 278 N JournalofCaveandKarstStudies, December2012

PAGE 47

sometimesmorereasonabletoleavehumanwasteinthe cave. Duetothehighproteincontentofthehumandiet, excessnitrogenisproducedintheformofammonia(NH 3 ) fromthecatabolismofaminoacids(Wright,1995).This ammoniaisexcretedasurea[(NH 2 ) 2 CO],themajor componentofurineinmammals.Intheenvironment, microbialhydrolysiscanconvertthisureabackinto ammonia(Fig.1A),whichcanbeassimilatedbyplants ormicroorganismsbackintoaminoacids.Ammoniaand othernitrogenousspecies(NO { 2 ,NO { 3 ,NO,N 2 O)contain nitrogeninanumberofvalencystates,allowingthese compoundstofunctioninbiologicalredoxreactions (Fig.1B).Microorganisms,therefore,havethecapacity touseammoniafordissimilatory(energygenerating) metabolicprocesses(Fig.1A),withtheresultingproducts (NO { 2 ,NO { 3 ,andN 2 O)servinginadditional(denitrifying) redoxreactions(Francisetal.,2007;Fig.1B).Excess ammoniainanitrogen-limitedenvironmentcanimpact bothmicrobialnitrogenassimilatory(nitrogenuptake)and dissimilatory(energygenerating)processes,withthe potentialtodramaticallychangebothnitrogenscavenging activitiesandecosystemenergetics(Ikeretal.,2010). LechuguillaCave,NewMexico,isanextensivecave systemover200kminlengthand500metersindepth(S. Allison,personalcommunication,2010).Explorationof thislargesystemhasrequiredextendedunderground campingtripslastinguptoeightdays(Reamesetal., 1999).Duetothelongdurationofthesetrips,urinehas beendepositedatcampsitesthroughoutthecave(Northup etal.,1997;Reamesetal.,1999).Becauseofthepotential impactofthisurineonmicrobial-ecosystemdynamics, weexaminedhowmicroorganismsinLechuguillaCave respondtothisnutrientatbothimpactedandnonimpactedsitesusingacombinationofcultivationand molecularphylogenetics.Ourresultsdemonstrateadramaticchangeinthemicrobialcommunityinresponseto urinedepositionandsuggestthepotentialforbioremediationstrategiestominimizefutureimpacts. M ETHODS C AVE D ESCRIPTIONAND G EOLOGY LechuguillaCaveislocatedinCarlsbadCaverns NationalPark,EddyCounty,NewMexico.Thecaveis locatedinadesertregionanddoesnotconnecttoany knownsurfacestreams,limitingallochthonousnutrient input(Davis,2000).Thecavewasformedprimarilyinthe CapitanFormationoftheDelawareBasinbyhypogenic sulfuricacidspeleogenesiswithapostulatedbiogenicorigin (Hill,2000;PalmerandPalmer,2000;Barton,2013).Due tothecomplexgeologyofthecave,withbothback-reef andfore-reeffacies,numeroussecondaryelementscanbe foundwithinthelimestone,includingiron,manganese, titanium,silica,andotherconstituentsofsedimentary minerals(Scholleetal.,1992;Northupetal.,2003). S AMPLE C OLLECTIONAND M OLECULAR T ECHNIQUES AttheBigSkycampsite,asmalldepressionoffthe maincorridorhasservedastheurinedepositionsitefor morethantwentyyears(Fig.2).Thegypsumwherethe urinewaspouredhasacquiredablackpatina,similarto thatseenatotherurinedepositsiteswithinthecave. Figure1.Thedegradationofureaandthenitrogencycle.A. Asimplificationofthebreakdownofureabyureaseinto ammonia.B.Thenitrogencyclenotonlyprovidesnitrogenin abioavailableformforplantsandbacteria(NO { 3 ),butthe valencyofthenitrogenspeciesalsoallowsthemtoserveas electrondonorsorelectronacceptorsinbacterialenergygeneratingprocesses.Ammoniaisoxidizedbynitrifying bacteriaintonitriteandnitrate,whilenitratecanbereduced bydenitrifyingbacteriatogeneratenitrousoxideand nitrogengas.Underanaerobicconditions,ammoniacanbe directlyoxidizedtonitrogengasbyanaerobicammoniaoxidizing(ANAMMOX)bacteria.Ineithercase,either nitrogengasfromthenitrogencycleordirectlyfromtheair canbemadebioavailablebynitrogen-fixingbacteriainthe formofammonia. M.D.J OHNSTON ,B.A.M UENCH ,E.D.B ANKS AND H.A.B ARTON JournalofCaveandKarstStudies, December2012 N 279

PAGE 48

Approximatelyonemonthfollowingthemostrecentuse ofthisurinesite ,a2cm 3 sampleoftheblackpatina (designatedWee)wasobtainedformolecularphylogenetic analysisbybreakingoffapieceofprotrudinggypsum usingasterilespatula.Asimilarly-sizedpieceofgypsum wascollectedapproximately40mawayatanon-impacted sitetoserveasacontrol(designatedCTL).Whencollected, thesampleswereplacedin70%ethanolfortransportout ofthecaveandstoredat 2 80 u Cpriortoanalysisinthe lab. AllDNAprotocolswerecarriedoutinalaminar-flow hoodusingaerosol-resistantpipettetipstoreducethe likelihoodofcontamination.Negativecontrolisolations werecarriedoutinparalleltomeasureanycontamination ofthefinalsamples.DNAwasextractedfromeither0.5mL ofliquidcultureor0.5gofsedimentusingourpreviously describedprotocol(Bartonetal.,2006).Community16S ribosomalRNAgenesequencelibrariesforBacteriawere preparedfromextractedDNAbyamplificationusingthe universalforwardprimer8F(5 9 -AGAGTTTGATCC Figure2.MapofLechuguillaCave,showingtherelativelocationsofthesamplesites.Thesampleformolecularphylogenetic analysiswascollectedattheBigSkyCampurinesite(Wee),whilesamplescollectedforcultivationwerecollectedatEC34B, EC27Z,andEC26IintheWesternBoreholeregion. H UMANURINEIN L ECHUGUILLA C AVE:THEMICROBIOLOGICALIMPACTANDPOTENTIALFORBIOREMEDIATION 280 N JournalofCaveandKarstStudies, December2012

PAGE 49

TGGCTCAG-3 9 )andbacterial-specificreverseprimer 1391R(5 9 -GACGGGCGGTGWGTRCA-3 9 )atan annealingtemperatureof58 u C.ForanArchaeallibrary, thearchaeal-specificprimers333Fa(5 9 –TCCAGGCCC TACGGG–3 9 )and1100Ar(5 9 –TGGGTCTCGCTC GTTG–3 9 )wereusedacrossarangeofannealing temperatures(55to65 u C)(Halesetal.,1996).Purified PCRproductswereclonedintoapTOPO-TACloning Vector(InvitrogenCorp.,Carlsbad,California)according tothemanufacturer’srecommendedprotocol.RepresentativeclonesweresequencedattheUniversityofKentucky AdvancedGeneticTechnologiesCenter(UK-AGTC; http://www.uky.edu/Centers/AGTC).Partialsequencesof each16SrRNAgenewerecompiledusingDNABaser SequenceAssembler(HeracleSoftwareCo.,Germany)and comparedtotheNCBIdatabasebyBLAST(http://www. ncbi.nlm.nih.gov)(Altschuletal.,1997).Thecompiled sequenceswerescreenedforremovalofpotentialchimeric sequencesusingBellerophonesoftware(http://comp-bio. anu.edu.au/bellerophon/bellerophon.pl)(Huberetal., 2004).Remainingsequencesweredepositedinthe NCBIGenBankdatabase(accessionnumbersJN032353– JN032396).DNAalignmentswerecarriedoutusinga NASTaligner(DeSantisetal.,2006)withmanual correctionsintheARBSoftwarePackage(http://www. arb-home.de),withadditionalsequencesfromtheRibosomalDatabaseProject(RDP;Coleetal.,2009)and Raineyetal.(2005).Forevolutionarydistancecalculations,amaximum-likelihoodalgorithmusingageneral time-reversal(GTR)substitutionmodelwascarriedoutin RAxML7.2.7with1000bootstrapreplicatestotestthe robustnessoftheinferredtopologies(Stamatakisetal., 2008).Inallcases, Bacilluspumulis wasusedasthe outgroup(AB195283). S AMPLE C OLLECTIONFOR C ULTIVATIONAND M ETABOLIC A CTIVITY Samplesforcultivationwerecollectedfromthree pristinelocationswithinLechuguillaCave,includingtwo 5mLwatersamplesfromshallowpoolsites(EC26Iand EC27Z)an d1gof floorsediment(EC34B)collectedwith sterilesyringesandspatulas,respectively(Fig.2).These sampleswereusedtoimmediatelyinoculate20mLoffiltersterilizedurine(fromahealthy24yroldmale,no medications)underoxicandanoxicconditionsinseptated vials(WheatonScientific,Millville,NewJersey).The anoxicconditionsweregeneratedbyremovingtheoxygen from250mLofairusingascorbicacid,withtheanaerobic natureoftheresultantgasconfirmedusingananaerobic teststrip.Thisdeoxygenatedairwasthenusedtodegas each20mLofurinebyreplacingtheheadspacegasthree timesoveratwenty-four-hourperiod.Halfoftheurine withinthevialswastreatedwith40 m Lofurease (1mgmL 2 1 ).Theresultwasfourcultureconditions(urine underoxicandanoxicconditions,andurineplusurease underoxicandanoxicconditions)foreachsamplesite (Fig.3).Anuninoculatedsamplewasusedasacontrolto verifythesterilityoftheurineundereachcondition.All cultureswereincubatedinthecave(20 u C)forforty-eight hoursandtransportedtothelabat4 u Ctopreventcell damagefromthehightemperaturevariationsfromshipping duringthesummer.Theprimaryculturesgeneratedinthe caveweresub-culturedtwoweekslaterinthelaboratoryat 20 u C.Sub-cultureswerepreparedfrom1mLsamplesofthe primarycultures,whichwereusedtoinoculate20mLof urinemediaunderthedescribedconditions.Todetermine whetheracarbonandenergysourcewasnecessaryfor growth,40mgmL 2 1 offilter-sterilizedglucosewasincluded inonesetofthesub-cultures,resultingineightvarietiesof sub-culture,oxicoranoxic,withorwithouturease,andwith orwithoutglucose(Fig.3).Threereplicatesamplesfrom eachoftheoxiccultureswereanalyzedforthepresenceof ammonia,nitrate,andnitriteviaspectroscopywithammonia(TNT831)andnitrate(TNT839)testkits(Hach, Loveland,Colorado).Verylittlevariationwasobservedin theanoxiccultures,andthesemeasurementswerenot repeated. Forcomparativenitratereductionassays, Pseudomonas strainspreviouslyisolatedfromLechuguillaCave(Johnstonetal.,2011)weregrowninsterilenitratebrothtotest fordenitrificationaccordingtothemanufacturer’sguidelines(Difco TM NitrateBroth,BectonDickinson,Franklin Lakes,NewJersey)andincubatedatroomtemperaturefor 48hrs.Thepresenceofgasbubbleswasrecordedas positivefornitrogenousgasproduction.Sampleswerethen evaluatedfornitrateandnitritereductionfollowingthe manufacturer’srecommendedprotocol. R ESULTS ToassesstheimpactofhumanurineonmicroorganismsoverextendedperiodsinLechuguillaCave,we conductedacomparativeanalysisofthemicrobialcommunityprofileattheBigSkyurinesite(Fig.2).This areaofLechuguillaCavewasdiscoveredin1989andhas servedasapermanentcampsite,averagingtwotothreesixpersonexpeditionsperyear.Attheendofeachexpedition, urineispouredintoashallowdepressionoffofthemain trail,flowingoverrockandgypsumdeposits.Overtime, thishascausedablackpatinatodevelop,asisseenatall theurinesitesthroughoutthecave.Thechemistryofthis patinaisunknown,buttheamorphousnatureofthe residuesuggeststhatitisorganicinnature;itclosely resemblesamberat.Asmallsample( 2cm 3 )ofpatinacoveredgypsumwascollected(designatedWee)and examinedusingmolecularphylogenetictechniquesto describethemicrobialspeciespresentfollowingextended exposuretourine.Asimilarpieceofgypsumwascollected fromanun-impactedareainthegeneralregionoftheBig Skysitetoserveasanegativecontrol(CTL). WeattemptedtoextractDNAfromboththeWeeand CTLgypsumsamples.WhileDNAwasreadilyobtained M.D.J OHNSTON ,B.A.M UENCH ,E.D.B ANKS AND H.A.B ARTON JournalofCaveandKarstStudies, December2012 N 281

PAGE 50

fromtheWeesamples,threeindependentattemptswere unabletoisolateDNAfromtheCTLsample.Thissuggests thatthetotalnumberofmicroorganismsintheCTL samplewasbelowthedetectionlimitofthisassay( 10 4 cellsg 2 1 ;Bartonetal.,2006).Aclonelibraryof136 bacterialphylotypeswasgeneratedfromtheWeesample DNAviaPCRamplification;however,noamplifiable PCRproductswereobtainedusingarchaeal-specific primersafternumerousattempts(datanotshown).A comparativeBLASTofthebacteriallibraryattheWeesite (Table1)demonstratedthatasignificantproportionofthe phylotypes(42%)hadalowlevelofsimilarity( # 95%)to previouslyidentifiedbacterialspecies.Thislowlevelof identitywasunexpectedduetotheamountofhumanurine thathadbeendepositedatthissiteandthepresumed presenceofhumancommensalspecies(Hunteretal.,2004; LavoieandNorthup2005).Asmallpercentage(9%)ofthe 136phylotypesdidhavedistantsimilaritytospeciesthat havebeenassociatedwithhumanpopulations(WeeA_H02, 92%identitytoapulmonaryinfection;WeeA_C08,97% identitytoa Bacillus isolatedfromapatientÂ’sblood),butthe vastmajorityofthephylotypesidentified(78%)wererelated tosoilspecies(Table1),demonstratingabroaddiversity withintheBacteria,includingthe Alpha-,Beta-and Gammaproteobacteria ,the Actinobacteria Firmicutes Deinococci-Thermus ,and Bacteriodetes. Ofinterestwasthe presenceofrepresentativephylotypesfromtheOrder Deinococcales, whicharenotnormallyassociatedwithcave environments(CoxandBattista,2005).The Deinococcales identifiedintheWeeclonelibrarydemonstrateda comparativelylow16Sidentitywithpreviouslyidentified representativesofthe Deinococci (92to95%)andthe cultivated Truepera (91%)(Table1).Todeterminewhether these Deinoccales phylotypesrepresentapreviouslyundescribedgroupwithinthisphyla,wegeneratedaphylogenetic treethatincludedtheclosestculturedandun-cultured representativesofthe Deinococcale fromtheNCBIand RDPdatabases(Raineyetal.,2005).Usingamaximum likelihoodtree-buildingalgorithm,thedendogramsuggests thatthesephylotypesrepresentanew,previouslyunrecognizedcladewithinthe Deinococci-Thermus (Fig.4).Itis interestingtonotethattheotherrepresentativephylotypes withinthisclade(HQ727579-81)werefoundinnitrate-and ammonia-richenvironments. Thelackofrecognizedhumancommensalspeciesat theWeesitesuggestedthatendemicmicrobialspecies, ratherthanexogenousspecies,mightbesubsistinglongtermtoutilizetheurineinsitu.Todetermineifendemic speciesarecapableofutilizinghumanurine,weattempted toculturebacteriafrompristinelocationsinLechuguilla Caveusinghumanurineasaculturemedium.Nocarbon orenergysourceswereaddedtothisurine,allowingusto selectforspeciescapableofutilizingorganicmolecules presentinurineasacarbonandenergysource(Kusanoet al.,2011).Thethreesamplesites,EC26I,EC27Z,and EC34B,werechosenduetotheirdistancefromimpacted urinesites(DeepSeas,RedSeas,Rusticles,BigSky,and Far-Eastcamps)andsitesoflimitedhumanactivity(away frommajorsurveyjunctionsormaintrails)(Fig.2). Giventhaturineisaliquid,poolsweresampledatEC26I andEC27Zforplanktonicspecies,whileEC34Bwasa representativesedimentsample.Thesampleswereinoculatedinto20mLofsterileurine(Fig.3)underoxic(20% O 2 )andanoxic(0%O 2 )conditions.Anoxicconditions wereincludedduetothedependenceonanaerobic conditionsforureaseproductionbyanumberofbacterial species(MobleyandHausinger,1989;McCartyand Bremmer,1991),whileotherspeciescanonlyuse ammoniaasanenergysourceunderanaerobicconditions (KowalchukandStephen,2001).Withoutknowing whetherendemiccavespeciespossesstheureaseactivity Figure3.Diagramofdescribedcultureconditions.Liquid orsedimentsampleswereinoculatedinthecaveunderfour growthconditionstogeneratetheprimarycultures:urine underoxicoranoxicconditions,andurine + ureaseunder oxicoranoxicconditions.Thesecondarycultureswere inoculatedinthelaboratorytocreateeightgrowthconditions:urineunderoxicoranoxicconditions,urine + glucose underoxicoranoxicconditions,urine + ureaseunderoxicor anoxicconditions,andurine + glucose + ureaseunderoxicor anoxicconditions. H UMANURINEIN L ECHUGUILLA C AVE:THEMICROBIOLOGICALIMPACTANDPOTENTIALFORBIOREMEDIATION 282 N JournalofCaveandKarstStudies, December2012

PAGE 51

Table1.Phylogeneticaffinitiesofphylotypesidentifiedattheweesamplesite. PhylumClone No. Clones a ClosestIdentifiedSequence b OriginofClosestIdentifiedSequence b 16SID c NCBI AccessionNo. d Actinobacteria WeeA_H025/6 Microbacteriumsenuense Symptomaticpulmonaryinfectionpatient92%DQ536408 WeeA_F101/6Uncultured Mycobacteriumsp. Microbialcompositionassociatedwith tremblingaspen 91%EF019585 Deinococcus-Thermus WeeA_E0218/31UnculturedDeinococci bacterium MicrobialcompositionfromfiveHawaiian lakes 87%AF513964 WeeA_B0810/31UnculturedbacteriumSoilcompositionfromCapeEvans,Mcmurdo DryValley,Antarctica. 90%AY676482 WeeA_G072/31Uncultured Deinococcussp .SoilsofMarblePointandWrightValley, VictoriaLand,Antarctica 88%DQ366016 WeeA_E011/31UnculturedDeinococci bacteriumBC_COM473 Bioremediationofpetroleum-contaminated soilwithcomposting 92%HQ727578 Firmicutes WeeA_A0718/56 Bacillussp. SoilsfromTheNetherlands,Bulgaria,Russia, Pakistan,andPortugal 98%AY289499 WeeA_B0425/56 Sporosarcinasoli Microbecompositionofuplandsoilsisolated fromKorea 98%DQ073394 WeeA_C087/56 Bacillushackensackii BacillusisolatedfromapatientÂ’sbloodculture97%AY148429 WeeA_F022/56 Bacillussp. IDA3504Soilisolatesfornovel Bacillus -relatedlineages98%AJ544784 WeeA_E031/56 Bacillusclausii KSM-K16Alkalineproteaseactivityof Bacillussp .KSMK16,isolatedfromsoil 92%AP006627 WeeA_E101/56 Sporosarcinaaquimarina BacteriumisolatedfromseawaterinKorea95%NR_025049 WeeA_D041/56 Sporosarcinaluteola Isolatedfromsoysauceproductionequipment fromJapan 94%AB473560 WeeA_D011/56Uncultured Bacillussp .Microbialcompositionassociatedwith tremblingaspen 91%EF019702 Alphaproteobacteria WeeA_H075/10 Sphingomonaskoreensis Yellow-pigmentedbacteriaisolatedfrom naturalmineralwater 94%NR_024998 WeeA_D082/10Uncultured Bradyrhizobiaceae bacterium Microbialcompositionassociatedwith tremblingaspen 90%EF018691 WeeA_E072/10Uncultured Rickettsiales bacterium Microbiotaassociatedwithphylogenetically ancientepithelia 90%EF667896 WeeA_A041/10 Alphaproteobacterium HeterotrophicN2-fixingbacterioplanktonin theBalticSea 90%AY972871 Betaproteobacteria WeeA_H061/2 Thiobacillusdenitrificans [ATCC H 25259] Thiobacillusdenitrificans ATCC2525989%CP000116 WeeA_D071/2Uncultured Betaproteobacterium Radionuclide-contaminatedsubsurface sediments 93%EU236233 Gammaproteobacteria WeeA_B0214/29 Rhodanobacterfulvus Koreansoilsamplemixedwithrottenricestraw97%AB100608 M.D.J OHNSTON ,B.A.M UENCH ,E.D.B ANKS AND H.A.B ARTON JournalofCaveandKarstStudies, December2012 N 283

PAGE 52

necessaryforthefirststepinurinebreakdown,wealso includedureasepre-treatedurineasamedium(Fig.3). Theculturesweregrownatcavetemperature(20 u C), andaftertwoweekssignificantgrowthwasseenunder bothoxicandanoxicconditions,withandwithoutthe presenceofurease.Nogrowthwasobservedinthe uninoculatedcontrol.Subculturesoftheseprimarycultures werethenestablishedinthelaboratorytodetermineif nitrogenredox-cyclingwasoccurringintermsofnitrifying (NH z 3 ? NO { 2 )ordenitrifying(NO { 2 ? N 2 O ? N 2 )activity (Fig.1B).Theonemodificationtothesesecondarycultures wastheadditionofaglucose(40mgL 2 1 )todetermine whethertheorganicmoleculespresentinurinewere sufficienttoserveasacarbonandenergysourcefor growthofheterotrophicspecies,orifadditionalsources werenecessary(Fig.3). Afterfivedaysofgrowthinthelaboratory,thesecondary culturesweretestedforammonia,nitrite,andnitrate production.Allsterile,uninoculatedurinecontrolswere belowthelevelofdetectionforeachofthesecompounds (datanotshown).Thedata(Fig.5)demonstratethat ammoniawasproducedinallculturesregardlessofurease addition,althoughammoniaconcentrationsweregenerally higherunderanoxicconditions.Inalltheurinecultures, nitritewasonlydetectedattracelevels( 0.1mgL 2 1 ;data notshown);thislowconcentrationispresumablyduetothe rapidconversionofnitritetonitratebythemicrobialspecies present(Romanetal.,1991).Theadditionofglucosedidnot appeartosignificantlyincreasetheproductionofammonia ornitrateundertheconditionstested.Interestingly,the additionofureaseappearedtogreatlyreducevariabilityin theamountofammoniaproducedinthesecultures.This suggeststhatthevariationobservedintheEC27Zplanktonic cultureunderaerobicconditionsmaybeduetothe productionofthisenzyme.Nonetheless,thelevelsof ammoniaonlyhadaminimalimpactontheresultantnitrate levelsinthecultures,andnocorrelationbetweenammonia andnitrateconcentrationwasobserved. Toaccountforthecomparativelylowlevelofnitrateto ammoniainthesamples,wewantedtodetermineifnitrate wasbeingremovedfromtheculturebyreductionto gaseousnitrogen(NO,N 2 O,orN 2 ).Totestthis,weadded 1mLofeachsecondaryculture(Fig.3)into3mLoffresh urinemediacontainingaDurhamtube.Afterforty-eight hoursofgrowth,nobubbleswereobservedintheDurham tubes,suggestingthatnogaswasproducedandhenceno nitratereductionwasoccurringduringgrowth(datanot shown).Nonetheless,itwaspossiblethattheabsenceofgas intheseassayswasduetolimitationsinthemediaused, ratherthantheabsenceofnitratereduction.Wetherefore inoculatedatraditionalnitratebroth(Difco)containing Durhamtubesfromthesecondarycultures.Again,no bubbleswereobservedafterforty-eighthoursofgrowth, andtestingofthenitratebrothforthepresenceofnitrite suggestedthatnitratereductionwasnotbeingcarriedout inthesecultures. PhylumClone No. Clones a ClosestIdentifiedSequence b OriginofClosestIdentifiedSequence b 16SID c NCBI AccessionNo. d WeeB_A089/29Uncultured Xanthomonadaceae bacterium Cloneisolatedfrombiofilterstreatingdimethyl sulphide 98%FJ536874 WeeA_H122/29 Stenotrophomonashumi Nitrate-reducingbacteriaisolatedfromsoil99%AM403587 WeeA_F122/29 Frateuriasp. EC-K130Rhizomicrofloralcomposersisolatedfromsoil94%AB264175 WeeA_D061/29 Rhodanobacterlindaniclasticus IsolatefromKoreangranulesludge90%AB244763 WeeA_E121/29 UnculturedXanthomonadaceae bacterium Biofilterstreatingdimethylsulphide:with/ withoutmethanol 95%FJ536874 Bacteroidetes WeeA_E042/2 Flavobacteriumsp. MH51Cultivablebacteriaassociatedwiththesoil fungistasis 98%EU182879 a Numberofphylotypesidentified/totalnumberofphylotypesinphylum. b ViaaBLASTsearchoftheNCBIdatabase(Altschuletal.1997). c TonearestsequenceintheNCBIdatabase. d NCBIaccessionnumberofsequencewithhighestidentityinNCBIdatabase. Table1.Continued. H UMANURINEIN L ECHUGUILLA C AVE:THEMICROBIOLOGICALIMPACTANDPOTENTIALFORBIOREMEDIATION 284 N JournalofCaveandKarstStudies, December2012

PAGE 53

Eventhoughcompletereductionofammoniato nitrogengaswasnotoccurringintheurinecultures,the presenceofatraceconcentrationofnitriteindicatesthat nitrifyinganddenitrifyingreactionsareoccurring.To identifythespeciesintheseculturesthatmaybeinvolvedin suchreactions,weexaminedtheoxicsecondaryculturesof EC26I,EC27ZandEC34B(withoutamendment)usinga molecularphylogeneticapproach.Clonelibrariescontainingtwenty-four,twenty-two,andtwenty-threebacterial phylotypeswereobtainedfortheEC26I,EC27Z,and EC34Boxicsamples,respectively(Table2).Noarchaea wereidentifiedintheseculturesusingaPCR-based approach.ThepoolsamplesEC26IandEC27Zshowed thelowestdiversity,whilethesedimentsampleEC34B showedthegreatestdiversity(Table2).Itisnotablethat thespeciesidentifiedusingthistechniquedemonstrateda highpercentagesimilarity(99to100%)topreviously culturedbacteria,includingfivewithpotentialsimilarityto humancommensalpopulations(EC26ID10,EC34BG06, EC34BH08,EC34BE02andEC34BD06;Table2),even thoughallsamplescamefrompristinelocationswithinthe cave.Onepotentialexplanationofthisfindingisthatfiltersterilized,urinewasusedasaculturemedium.Itis thereforepossiblethatthesegenerarepresentedcommensal contaminantsobtainedduringurinecollection.Nonetheless,ouruninoculatedcontrolsdidnotcontainmicrobial growth,andthesegeneraarenotknowntoproduceultrasmallcellscapableofpassingthrougha0.22mmfilter (BakkenandOlsen,1987;Rappe etal.,2002;Hahnetal., 2003;Godoyetal.,2005;MitevaandBrenchley,2005).A numberoftheidentifiedgeneraarealsoknowntoproduce ureaseortolerateurea,includingrepresentativesofthe genera Bacillus,Corynebacteria,Micrococcus,Pseudomonas, and Ochromobactrum (Brenneretal.,2005).Additionally,membersofthe Bacilli,Corynebacteriaceae, Pseudomonads, and Ochrobactraceae areknowntocarry outdenitrificationreactions(Brenneretal.,2005). Despitethepresenceofdenitrifyingspeciesinthe cultures,theabsenceofnitratereductiontogaseousnitrogen suggeststhatthisfinalstepinthenitrogencycleisnot occurring.Inaseparatestudy(Johnstonetal.,2011),we haveculturedanumberof Pseudomonas speciesfrom LechuguillaCave,agenusthatisknowntopossessthis denitrificationphenotype(Brenneretal.,2005).Todetermineif Pseudomonas strainsendemictothecavearecapable ofcompletenitratereduction(Fig.1),wecarriedouta nitratereductiontestontwenty-three Pseudomonas strains, representingfivedifferentspecies(Table3).Ofthese isolates,threespecies, Pseudomonasabientaniphila,Pseudomonasgraminis, and Pseudomonasresinovorans, wereableto reducenitratetonitrite,while Pseudomonasstutzeri wasable tocompletelyreducenitratetonitrogengas.Thisresult demonstratesthepotentialforendogenouscavespeciesto completelyoxidizeureatonitrogengas. D ISCUSSION Humanexplorationallowsthesystematicdocumentationofcavesandprovidesacriticalcomponentinthe scientificunderstandingofthesesystems(Kambesis,2007). Suchworkhasallowedustobuildapictureofthe geologicalandhydrologicalprocessesthatleadtothe formationofcavesandallowedsignificantadvancesinour understandingofspeleogenesis,secondarymineraldeposiFigure4.Phylogenetictreeofculturedandunculturedrepresentativesofthe Deinococci ,demonstratingtheuniquecladeof uncultured Deinococci inwhichtheWeesitephylotypesarefound.ThedendrogramwasconstructedusingaMaximumlikelihoodalgorithm.Therobustnessoftheinferredtopologieswastestedwith1,000bootstrapreplicates,withthelikelihoodof generatingtheinferredtopologyateachnodeshown.Theoutgroupusedwas Bacilluspumulis (AB195283). M.D.J OHNSTON ,B.A.M UENCH ,E.D.B ANKS AND H.A.B ARTON JournalofCaveandKarstStudies, December2012 N 285

PAGE 54

tion,andcave-ecosystemstructureanddynamics.While hydrologicallyactivesystemscanwithstandahigherlevel ofhumanimpact,theconsequencesofhumanexploration onenvironmentallysensitivesystemsarenotknownor fullyunderstood(Lavoie,1995;LavoieandNorthup, 2005).Nonetheless,thelimitedsurfacewaterinputinto LechuguillaCaveandthestaticnatureoftheenvironment suggeststhatthiscavemaybeparticularlysensitivetosuch impacts(Cunninghametal.,1995;Northupetal.,2003). Toexaminehumanimpactonthemicrobialfloraof LechuguillaCave,Hunteretal.(2004)andLavoieand Northup(2005)examinedthecavesystemforthepresence ofbacterialspeciesthatcouldserveasamarkerforhuman contamination.Theseinvestigatorsexaminedthecavefor thepresenceoffecal( Escherichiacoli ),skin( Staphylococcus aureus ),orsoil( Bacilli spp.)bacterialspeciesusing chromogenictests.Hunteretal.(2004)suggestedthatfecal E.coli couldbefoundinpoolsthroughoutthecave,while LavoieandNorthup(2005)didnotfind E.coli inpools,but inpristinesitesandactiveurinedumps.Thediscrepancyin thefindingsbetweenstudieswasattributedbytheauthorsto theshortresidencetimeof E.coli incavesedimentsandits rapidentryintoaviablebutnon-culturablestate,making detectiondifficultusingthecultivation-basedtechniques usedinthesestudies(LavoieandNorthup,2005).Inthis study,weuseaculture-independentapproachthatisnot subjecttothelimitationsofcultivation(Pace,1997). Itwasourinitialaimtocarryoutacomparative analysisbetweentheendemicmicrobialpopulationina geochemicallysimilar,unimpactedsiteinthecaveandthe urine-impactedsite;however,theunimpactedcontroldid notcontainasufficientmicrobialpopulationtoallow DNAextraction.Theextractiontechniquesusedhavea detectionlimitofabout10 4 cellsg 2 1 ofsediment, suggestingthatthemicrobialpopulationofthecontrol sitecontainedveryfewcells,likelyduetotheosmoticstress ofgrowthongypsum.TheeaseofobtainingDNAfrom theimpactedWeesitesuggestsamuchhigherbiomassat Figure5.Productionofammonia,nitrite,andnitratefromurineinculture.BacterialcultureswereinoculatedfromnonimpactedsitesinLechuguillaCave,EC26I(gray,water),EC27Z(white,water),andEC34B(black,sediment).Theaverageof ammoniaandnitrateproducedfromthebreakdownofureaafterfivedaysincultureareshownunderaerobicandanaerobic conditions.Fortheoxiccultures,theaveragesandstandarddeviations(errorbars)ofthreereplicatesamplesareshown.Nitrite wasonlydetectedattracelevelsandisnotshown. H UMANURINEIN L ECHUGUILLA C AVE:THEMICROBIOLOGICALIMPACTANDPOTENTIALFORBIOREMEDIATION 286 N JournalofCaveandKarstStudies, December2012

PAGE 55

thislocation,whichcorrespondstothefindingsofLavoie (1995),whodemonstratedthaturineamendmentof sedimentsfromLechuguillaCaveledtoadramatic increaseincellnumber,fromlessthan10 4 to10 7 cells g 2 1 .ThissuggeststhattheurinedepositionattheWeesite ledtoadramaticincreaseinpopulationsize.Thisincrease innumbersalsomatchesthedetectionlimitsofourassay (Bartonetal.,2006;Lavoie,1995). Usingamolecularphylogeneticanalysisofthis increasedpopulationattheBigSkyurinesite,wefound twelvephylotypesthatdemonstratedsomesimilarityto humanisolates,includingsimilaritytoabacteriumfrom humanpulmonary(WeeA_H02)andblood(WeeA_C08) infections,butthesephylotypesshowonlyaweak similarity(92and97%,respectively)tothesepathogens. Suchlow16Ssimilaritysuggeststhatthesephylotypesdo Table2.Identityofculturedbacteriagrownonurinemediafrompristinesites. SampleSiteCloneNo.Clones a ClosestIdentified Sequence b OriginofClosest IdentifiedSequence b 16SID c NCBI Accession No. d EC26I(water) Actinobacteria EC26IB1010/22 Micrococcusluteus Medievalwallpainting99%AJ409096 EC26IB047/22 Micrococcussp. JL-76Marineenvironment99%AY745846 EC26ID105/22 Cellulomonas parahominis Clinicalisolatesof Coryneform group 99%AY655729 Firmicutes EC26IA022/2 Paenibacillusborealis Humusbacteriaof Norwayspruce 90%NR_025299 EC27A(water) Firmicutes EC27ZB0112/12 Bacilluspumilus SAFR-032 Soilpollutedwith chromium 99%DQ416781 Alphaproteobacteria EC27ZA0510/10 Ochrobactrumsp. TK14 Soilandwheatroots samples 99%AJ550273 EC34B(sediment) Actinobacteria EC34BG061/3 Corynebacterium jeikeiumK411 Nosocomialpathogen99%CR931997 EC34BH031/3 Rhodococcus erythropolis Rocksofanancient goldmine 99%EF491951 EC34BH081/3 Propionibacterium acnesJCM6473 16SrRNAgenesequence ofJCMstrain 100%AB573714 Firmicutes EC34BB054/5Uncultured bacteriumKSC2-41 Spacecraftcleanroom99%DQ532287 EC34BE101/5 Bacillus weihenstephanensis KBAB4 Root-associated bacteria 99%CP000903 Alphaproteobacteria EC34BA045/5 Bradyrhizobium elkanii Commercialrhizobial strains 99%FJ025139 Betaproteobacteria EC34BC012/2 Ralstoniasp.C1 Biofilminspentnuclear fuelpool 99%AY479983 Gammaproteobacteria EC34BD052/6 Acinetobactersp. Bovineproductsandsoil99%Z93442 EC34BE022/6Clonenbt97q09 (Pseudomonas) Humanskinmicrobiota99%EU539061 EC34BA021/6 Pseudomonas fluorescens Pf0-1 Cornplant,greenhouse conditions 99%AY958233 EC34BG091/6 Serratiaproteamaculans LauBasinhydrothermal vents 99%CP000826 Bacteroidetes EC34BD062/2Unculturedbacterium rRNA082 Humanvaginal epithelium 99%AY958855 a Numberofphylotypesidentified/totalnumberofphylotypesinphylum. b ViaaBLASTsearchoftheNCBIdatabase(Altschuletal.1997). c TonearestsequenceintheNCBIdatabase. d NCBIaccessionnumberofsequencewithhighestidentityinNCBIdatabase. M.D.J OHNSTON ,B.A.M UENCH ,E.D.B ANKS AND H.A.B ARTON JournalofCaveandKarstStudies, December2012 N 287

PAGE 56

notbelongtothesamespecies(orinthecaseof92% identity,thesamegenera)asthesepotentialhuman pathogens(Geversetal.,2005).Further,thesephylotypes arealsomembersofgeneracommonlyfoundincave environments( Microbacterium and Bacillus; Johnston etal.,2011),supportinganendemicoriginforthese species.Inthisstudy,wedidnotfind E.coli inourWee samplesiteorculturesusingthemoresensitiveandspecific molecularapproach.Thecultivation-basedchromogenic techniquesofHunteretal.(2004)andLavoieandNorthup (2005)measureasinglephenotypiccharacteristicto identify E.coli ,inthiscaselactosefermentation.Yeta numberoftheculturedspeciesidentifiedinthisstudyalso fermentlactose,includingmembersofthe Cellulomonaceae,Micrococci,Paenibacilli, and Bacilli (Tables1and2). Thus,theidentificationofbacteriabasedexclusivelyon lactosefermentationwouldgiveafalse-positiveforfecal coliforms.Assuch,thepresenceof E.coli withinthecave requiressupportingevidence(BartonandPace,2005). Suchevidencecouldcomeintheformofconfirmatory tests,includingculture(mTEC),morphological(Gram stain),andmolecular(16Sribotyping)assays(Bartonand Pace,2005),butnosuchconfirmatorytestswerecarried outinthesepreviousstudies(Hunteretal.,2004;Lavoie andNorthup,2005). Ourphylogeneticanalysisdoesnotdemonstratethe presenceof E.coli orotherfecalbacterialspeciesatthe samplesite.Thisdoesnotindicatethattheyarenotbeing deposited,butthatthecommensalsmaynotbeableto competeorsubsistundertheconditionsofthecave.The lossofcommensalspeciesfromanenvironmentcontaminatedbyhumanwaste(Germantoilets)haspreviously beendescribed(Egertetal.,2010).Therethesourceofthe endemicspecieswasdeterminedtobefromtheenvironment(fromflushing)andthismaybethecasehere;the environmentalsourceisthecaveitself. Thepresenceofanumberofphylotypespotentially involvedinnitrogen-cyclingactivitiesattheWeesite (specificallydenitrification;Table1)suggeststhatthe bacterialspeciesattheurinesitemaybetakingadvantage ofnitrogenouscompoundsderivedfromtheurinefor energygeneration.Thepresenceofnumerousphylotypes withsimilaritytoheterotrophicsoilspeciesalsosuggests thatthespeciespresentcouldbetakingadvantageofthe organiccompoundsfoundinurineforgrowth(Kusano etal.,2011).Theidentificationofmembersofthe Deinococcales groupismuchhardertoreconcileatthis location.The Deinococci-Thermus grouprepresenta heterotrophic,deeplybranchinglineagewithintheBacteria (Alburquerqueetal.,2005;Dworkinetal.,2006)thatare knownfortheirhighlevelofresistancetoionizing(X-, a -, b -and c -rays)andnon-ionizing(UV)radiation(Coxand Battista,2005).ItisunlikelythatLechuguillaCave providesselectivepressurefromradiation,particularly fromUVradiation,butthemechanismsofDNArepair thatprovideradiationresistanceforthesespeciesalso allowresistancetodehydration(MattimoreandBattista, 1996). Trueperaradiovictrix ,whichshowscloserhomology tothephylotypesinourphylogeneticanalysisthanother culturedmembersofthe Deinococci, hasbeenshowntouse nitrateasanelectronacceptor(Alburquerqueetal.,2005). Thissuggeststhatnitratewithintheenvironmentmaybe providinganadditionalselectionpressure.Itisinteresting tonotethatthe Deinococcales arereadilyculturedon heterotrophicmedia,suchasthatusedinpreviousstudies (Hunteretal.,2004;LavoieandNorthup,2005),and displaylactosefermentationcapabilities(Raineyetal., 2005;Alburquerqueetal.,2005). Ourphylogeneticresultssuggestedthatthebreaking downofhumanurineattheWeesitemaybebeing facilitatedbyendemiccavespecies.Totestwhether endemicspeciesdemonstratethiscapacity,weestablished culturesfromcavesitesthathadnotbeenimpactedby humanurine(Fig.2).Eventhoughthesecultureswere establishedwithmicroorganismsfrompristinesites,the bacteriaidentifiedhadsimilar16SrDNAidentityto humancommensalspecies,includingmembersofthe Cellulomonas,Corynebacterium,Propionibacterium, and Pseudomonas genera(EC26ID10,EC34BG06,EC34BH08, EC34BE02,andEC34BD06).Ratherthanbeingindicative ofcontaminationatthesesites,itislikelythattheurinein ourculturemediumenrichesforthegrowthofbacterial speciesabletoutilizeurea,resistthehighlevelsofurea encountered,orcatabolizetheorganicmoleculesfound Table3.NitrogenreductioninLechuguillaCave Pseudomonad cultures. Strains MetabolicTest Bacteria%ID a No. b N 2 NO 2 2 NO 3 2 Pseudomonasstutzeri 9916 + 2 + Pseudomonasabietaniphila 991 2 + ND c Pseudomonasgraminis 981 2 + ND c Pseudomonasresinovorans 992 2 + ND c Pseudomonasgingeri 993 222 a Totypestrain( P.stutzeri ATCC17587, P.abietaniphila ATCC700689, P.graminis ATCC700544, P.resinovorans ATCC14235, P.gingeri NCPPB3146). b ThetotalnumberofstrainsintheLechuguillaCultureCollection. c ND 5 notdone(nitratepositivetestscannotbetestedfornitratereductionusingthisassay). H UMANURINEIN L ECHUGUILLA C AVE:THEMICROBIOLOGICALIMPACTANDPOTENTIALFORBIOREMEDIATION 288 N JournalofCaveandKarstStudies, December2012

PAGE 57

intheurine(Kusanoetal.,2011).Indeed,thegenera representedbytheseisolateshavebeenassociatedwith humanurinary-tractinfections,increasingthelikelihoodof theircultivationusingthisapproach(Sellinetal.,1975; CarpenterandDicks,1982;MobleyandHausinger,1989; Kunin,1994;Kesseru ¨etal.,2002;Walteretal.,2007;Zhao etal.,2010).Similarphylotypeshavealsobeenfoundin othercavesandsubsurfaceenvironments,includinggold minesandthesoil(Table2),suggestingthattheoriginof thesespecieswasthecaveenvironment. Givenanendemicoriginofthespeciesfoundassociated withurineinthisstudy,itisinterestingtonotethatamong thesamplesfrompristinesites,thewatersamplesEC26Iand EC27Zshowedthelowestspeciesdiversity(fourandtwo phylotypes,respectively),whilethesedimentsampleEC34D demonstratedthehighest(thirteenphylotypesinsix divisions).TheEC26Icultureisdominatedbyheterotrophic species(Table2),includingmembersofthe Micrococci Cellulomonaceae, and Paenibacilli .While Micrococci are potentialhumancommensals,wehavepreviouslyisolated membersofthisgenuswithincaves(Johnstonetal.,2011). Indeed,theisolateculturedinthisstudy(EC26IB10)shared thehighestidentitytoabacteriumisolatedfromthewallpaintingofamedievalchurch(Table2),arguingforan organismbetteradaptedtonutrientlimitationthanhuman skin(Wieseretal.,2002).ThepoolcultureEC27Zcontained near-equalphylotypesof Bacilluspumilus andarepresentativeofthe Ochrobactrum (Table2).Bothhavepreviously beenisolatedfromcaveenvironments(Johnstonetal.,2011) andarecapableofheterotrophicnitrificationandaerobic denitrification(Zhaoetal.,2010).Theurineculturefromthe sediment(EC34B)samplecontainedthegreatestdiversityof heterotrophicspeciesandreflectsamorecomplexecosystem involvedinnitrogencycling,includingrepresentativesofthe Rhodococci,Bradyrhizobia, and Ralstonia (Dworkinetal., 2006).Membersofthe Cellulomonas,Micrococci, and Paenibacilli arealsolactose-fermentingspecies,suggesting thatcultivationfrompristinesitescanalsogeneratefalse positivesforthepresenceof E.coli (Hunteretal.,2004; LavoieandNorthup,2005). Inalltheestablishedcultures,themicroorganisms presentseemcapableofbreakingdownurinewithoutthe additionofanexogenouscarbonsourceorurease.Under anaerobicconditions,theconversionofureatoammonia viaureaseappearedtooccurathighlevels(Fig.5),an observationconsistentwiththeincreasedureaseactivityof soilspeciesunderanaerobicconditions(McCartyand Bremmer,1991).Althoughnoneofourestablishedcultures demonstratedthecompletereductionofureatoN 2 Pseudomonas speciespreviouslyisolatedfromLechuguilla CavewereabletoreduceaccumulatedNO { 3 toN 2 Collectively,thissuggeststhatitmightbepossibleto engineeracollectionofendemiccavespeciesthatcould completelyoxidizeureatoN 2 gas,withoutsupplementation withadditionalnutrients(anenergysourceorurease).These speciescouldthentobeaddedtotheurinebeforeitis dumpedinthecave.Byreducingthetotalamountof nitrogenintheurine,itmightbepossibletolimitthe introductionofnitrogencompoundsinurineusingan ecosystem-neutralmethodofinsitubioremediation. Ourdatademonstratethatevenunderthepressureof allochthonousnitrogenandexogenousbacterialspecies, themicrobialcommunityattheBigSkyurinesitestill appearstobedominatedbyindigenouscavespecies.The datasuggestthatthenativemicrobialcommunityremains resistanttoinvasionfromhumancommensalspecies. Factorsthatcouldmakethemicrobialecosystemresistant tosuchxenobioticinvasionmaybetheuniquelyoligotrophicnatureofthecaveenvironmentthatlimitsthe establishmentofcopiotrophicspecies(BartonandJurado, 2007),theimpactofhost-rockgeochemistryongrowth (Bartonetal.,2007),oreventheselectionpressureofthe gypsumpresentatthissite.Highspeciesdiversitycanalso playaroleincommunityresistancetoenvironmental perturbations,allowingmetabolicflexibilityandcommunityadaptationunderever-changingconditions(Girvan etal.,2005).Todeterminewhatfactorscontrolthe diversityandspeciationofthemicrobialcommunityat theWeeandotherurinesiteswillrequireadditional experiments,includingadditionalcultivationandin-depth biogeochemistryoftheurinesiteswithinLechuguillaCave. Thedramaticchangeinthestructureofthemicrobial communityattheWeesitesupportstheestablishedcaving principles:itisimportanttolimithumanimpactincave environments,particularlyfromtheintroductionofwaste. WhilethemicrobialcommunityattheWeesiteremains dominatedbyendemicspecies,itisclearlydifferentfroma nearbyun-impactedsiteinthecave(CTL).Without knowingthelong-termimpactofsuchadramaticchange ontheentiremicrobialecosystem,suchimpactsmustbe limitedbyemployingsoundmanagementpracticesto balancethedelicatenatureoftheecosystemwithourneed tounderstandandprotectthesecriticalhabitats. A CKNOWLEDGEMENTS TheauthorswouldliketothankBradleyLubbers,David Bunnell,andElizabethRosseauforassistancewithsample collection,threeanonymousreviewersforcommentsthat improvedthemanuscript,andDr.KathleenLavoiefor criticalcommentsandvaluablereferences.Theauthors wouldalsoliketothankPaulBurgerandStanAllisonof CarlsbadCavernsNationalParkforaccessandin-cave assistance.FundingwasprovidedbytheNSFKY EPSCoRProgramtoHAB. R EFERENCES Albuquerque,L.,Simo es,C.,FernandoNobre,M.,Pino,M.N.,Battista, J.R.,Silva,M.T.,Rainey,F.A.,anddaCosta,M.S.,2005, Truepera radiovictrix gen.nov.,sp.nov.,anewradiationresistantspeciesand theproposalof Trueperaceae fam.nov.:FEMSMicrobiologyLetters, v.247,p.161–169.doi:10.1016/j.femsle.2005.05.003. M.D.J OHNSTON ,B.A.M UENCH ,E.D.B ANKS AND H.A.B ARTON JournalofCaveandKarstStudies, December2012 N 289

PAGE 58

Altschul,S.F.,Madden,T.L.,Scha ¨ffer,A.A.,Zhang,J.,Zhang,Z.,Miller, W.,andLipman,D.J.,1997,GappedBLASTandPSI-BLAST:anew generationofproteindatabasesearchprograms:NucleicAcids Research,v.25,p.3389–3402.doi:10.1093/nar/25.17.3389. Bakken,L.,andOlsen,R.,1987,Therelationshipbetweencellsizeand viabilityofsoilbacteria:MicrobialEcology,v.13,p.103–114. doi:10.1007/BF02011247. Barr,T.C.Jr.,1967,Observationsontheecologyofcaves:American Naturalist,v.101,p.475–491. Barton,H.A.,2013,Biospeleogenesis:BiogeneticProcessesandMicrobial ImpactonSpeleogenesis,Frumkin,A.,ed.,TreatiseonGeomorphology:Atlanta,GA,Elsevier,(inpress). Barton,H.A.,andJurado,V.,2007,What’supdownthere?Microbial diversityincaves:Microbe,v.2,p.132–138. Barton,H.A.,andPace,N.R.,2005,Discussion:Persistentcoliform contaminationinLechuguillaCavepools:JournalofCaveandKarst Studies,v.67,p.55–57. Barton,H.A.,Taylor,N.M.,Kreate,M.P.,Springer,A.C.,Oehrle,S.A., andBertog,J.L.,2007,Theimpactofhostrockgeochemisryon bacterialcommunitystructureinoligotrophiccaveenvironments: InternationalJournalofSpeleology,v.36,p.93–104. Barton,H.A.,Taylor,N.M.,Lubbers,B.R.,andPemberton,A.C.,2006, DNAextractionfromlow-biomasscarbonaterock:animproved methodwithreducedcontaminationandthelow-biomasscontaminantdatabase:JournalofMicrobiologyMethods,v.66,p.21–31. doi.org/10.1016/j.mimet.2005.10.005. Brenner,D.J.,Krieg,N.R.,andStaley,J.T.,2005,eds.,Bergey’sManual ofSystematicBacteriology,Volume2:TheProteobacteria,PartC: The Alpha-,Beta-,Delta -and Gammaproteobacteria ,secondedition: NewYork,Springer,1392p. Carpenter,E.M.,andDicks,D.,1982,Isolationof Pseudomonas fluorescens aftersuprapubiccatheterisation:JournalofClinical Pathology,v.35,581p. Cole,J.R.,Wang,Q.,Cardenas,E.,Fish,J.,Chai,B.,Farris,R.J.,KulamSyed-Mohideen,A.S.,McGarrlll,D.M.,Marsh,T.,Garrity,G.M., andTiedje,J.M.,2009,TheRibosomalDatabaseProject:improved alignmentsandnewtoolsforrRNAanalysis:NucleicAcidsResearch, v.37,supplement1,p.D141–145.doi:10.1093/nar/gkn879. Cox,M.M.,andBattista,J.R.,2005, Deinococcusradiodurans —The consummatesurvivor:NatureReviewsMicrobiology,v.3,p.882– 892.doi:10.1038/nrmicro1264. Culver,D.C.,andSket,B.,2000,Hotspotsofsubterraneanbiodiversityin cavesandwells:JournalofCaveandKarstStudies,v.62,p.11–17. Cunningham,K.I.,Northup,D.E.,Pollastro,R.M.,Wright,W.G.,and LaRock,E.J.,1995,Bacteria,fungiandbiokarstinLechuguillaCave, CarlsbadCavernsNationalPark,NewMexico:Environmental Geology,v.25,p.2–8.doi:10.1007/BF01061824. Davis,D.G.,2000,ExtraordinaryfeaturesofLechuguillaCave,GuadalupeMountains,NewMexico:JournalofCaveandKarstStudies, v.62,p.147–157. DeSantis,T.Z.Jr.,Hugenholtz,P.,Keller,K.,Brodie,E.L.,Larsen,N., Piceno,Y.M.,Phan,R.,andAndersen,G.L.,2006,NAST:amultiple sequencealignmentserverforcomparativeanalysisof16SrRNA genes:NucleicAcidsReseach,v.34,supplement2,p.W394–399. doi:10.1093/nar/gkl244. Dworkin,M.,Falkow,S.,Rosenberg,E.,Schleifer,K.H.,andStackebrandt,E.,2006,TheProkaryotes:AHandbookoftheBiologyof Bacteria,thirdedition:NewYork,Springer,7volumes. Egert,M.,Schmidt,I.,Bussey,K.,andBreves,R.,2010,Aglimpseunder therim—thecompositionofmicrobialbiofilmcommunitiesin domestictoilets:JournalofAppliedMicrobiology,v.108,p.1167– 1174.doi:10.1111/j.1365-2672.2009.04510.x. Elliott,W.R.,2006,Biologicaldosanddon’tsforcaveconservationand restoration, in Hildreth-Werker,V.,andWerker,J.C.,eds.,Cave ConservationandRestoration:Huntsville,NationalSpeleological Society,p.33–46. Francis,C.A.,Beman,M.J.,andKuypers,M.M.M.,2007,Newprocesses andplayersinthenitrogencycle:Themicrobialecologyofanaerobic andarchaealammoniaoxidation:TheISMEJournal,v.1,p.19–27. doi:10.1038/ismej.2007.8. Gevers,D.,Cohan,F.M.,Lawrence,J.G.,Spratt,B.G.,Coenye,T.,Feil, E.J.,Stackebrandt,E.,VandePeer,Y.,Vandamme,P.,Thompson, F.L.,andSwings,J.,2005,Re-evaluatingprokaryoticspecies:Nature ReviewsMicrobiology,v.3,p.733–739.doi:10.1038/nrmicro1236. Girvan,M.S.,Campbell,C.D.,Killham,K.,Prosser,J.I.,andGlover, L.A.,2005,Bacterialdiversitypromotescommunitystabilityand functionalresilienceafterperturbation:EnvironmentalMicrobiology, v.7,p.301–313.doi:10.1111/j.1462-2920.2005.00695.x. Godoy,F.,Vancanneyt,M.,Mart nez,M.,Steinbu ¨chel,A.,Swings,J., andRehm,B.H.,2003, Sphingopyxischilensis sp.nov.,achlorophenol-degradingbacteriumthataccumulatespolyhydroxyalkanoate, andtransferof Sphingomonasalaskensis to Sphingopyxisalaskensis comb.nov.:InternationalJournalofSystematicandEvolutionary Microbiology,v.53,p.473–477. Hahn,M.W.,Lunsdorf,H.,Wu,Q.,Schauer,M.,andHofle,M.G.,2003, IsolationofnovelultramicrobacteriaclassifiedasActinobacteriafrom fivefreshwaterhabitatsinEuropeandAsia:AppliedandEnvironmentalMicrobiology,v.69,p.1442–1451.doi:10.1128/AEM.69.3. 1442-1451.2003. Hales,B.A.,Edwards,C.,Ritchie,D.A.,Hall,G.,Pickup,R.W.,and Saunders,J.R.,1996,IsolationandidentificationofmethanogenspecificDNAfromblanketbogpeatbyPCRamplificationand sequenceanalysis:AppliedandEnvironmentalMicrobiology,v.62, p.668–675. Hardin,J.W.,andHassell,M.D.,1970,Observationonwakingperiods andmovementsof Myotissodalis duringhibernation:Journalof Mammalogy,v.51,p.829–831. Hill,C.A.,2000,Sulfuricacid,hypogenekarstintheGuadalupe MountainsofNewMexicoandWestTexas(U.S.A.), in Klimchouk, A.B.,Ford,D.C.,Palmer,A.N.,andDreybrodt,W.,eds.,Speleogenesis:EvolutionofKarstAquifers:Huntsville,NationalSpeleological Society,p.309–316. Huber,T.,Faulkner,G.,andHugenholtz,P.,2004,Bellerophon;a programtodetectchimericsequencesinmultiplesequencealignments: Bioinformatics,v.20,p.2317–2319.doi:10.1093/bioinformatics/ bth226. Hunter,A.J.,Northup,D.E.,Dahm,C.N.,andBoston,P.J.,2004, PersistentcoliformcontaminationinLechuguillacavepools:Journal ofCaveandKarstStudies,v.66,p.102–110. Iker,B.C.,Kambesis,P.,Oehrle,S.A.,Groves,C.,andBarton,H.A., 2010,Microbialatrazinebreakdowninakarstgroundwatersystem anditseffectonecosystemenergetics:JournalofEnvironmental Quality,v.39,p.509–518.doi:10.2134/jeq2009.0048. Ikner,L.A.,Toomey,R.S.,Nolan,G.,Neilson,J.W.,Pryor,B.M.,and Maier,R.M.,2007,Culturablemicrobialdiversityandtheimpactof tourisminKartchnerCaverns,Arizona:MicrobialEcology,v.53, p.30–42.doi:10.1007/s00248-006-9135-8. Johnston,M.,Millette,J.,Banks,E.D.,Jurado,V.,andBarton,H.A., 2011,Theculturablediversityofoligotrophicheterotrophs:The lechuguillacaveculturelibrary, in ProceedingsoftheAmerican SocietyforMicrobiologyGeneralMeeting.NewOrleans,LA: Washington,D.C.,AmericanSocietyforMicrobiology,N1795p. Kambesis,P.,2007,Theimportanceofcaveexplorationtoscientific research:JournalofCaveandKarstStudies,v.69,p.46–58. Kesseru ¨,P.,Kiss,I.,Bihari,Z.,andPolya k,B.,2002,TheeffectsofNaCl andsomeheavymetalsonthedenitrificationactivityof Ochrobactrum anthropi :JournalofBasicMicrobiology,v.42,p.268–276.doi:10.1002/ 1521-4028(200208)42:4 268::AID-JOBM268 3.0.CO;2-E. Klimchouk,A.B.,Ford,D.C.,Palmer,A.N.,andDreybrodt,W.,eds., 2000,Speleogenesis,EvolutionofKarstAquifers:Huntsville, NationalSpeleologicalSociety,527p. Kowalchuk,G.A.,andStephen,J.R.,2001,Ammonia-oxidizing bacteria:Amodelformolecularmicrobialecology:AnnualReview ofMicrobiology,v.55,p.485–529.doi:10.1146/annurev.micro.55. 1.485. Kunin,C.M.,1994,Urinarytractinfectionsinfemales:ClinicalInfectious Diseases,v.18,p.1–12.doi:doi:10.1093/clinids/18.1.1. Kusano,M.,Mendez,E.,andFurton,K.G.,2011,Developmentof headspaceSPMEmethodforanalysisofvolatileorganiccompounds presentinhumanbiologicalspecimens:AnalyticalandBioanalytical Chemistry,v.400,p.1817–1826.doi:10.1007/s00216-011-4950-2. Lavoie,K.H.,1995,Theeffectsofurinedepositiononmicrobesincave soils, in Proceedingsofthe1993NationalCaveManagement SymposiumProceedings,Carlsbad,NM,:Huntsville,AL,National SpeleologicalSociety,p.302–312. Lavoie,K.H.,Helf,K.L.,andPoulson,T.L.,2007,Thebiologyand ecologyofNorthAmericancavecrickets:JournalofCaveandKarst Studies,v.69,p.114–134. H UMANURINEIN L ECHUGUILLA C AVE:THEMICROBIOLOGICALIMPACTANDPOTENTIALFORBIOREMEDIATION 290 N JournalofCaveandKarstStudies, December2012

PAGE 59

Lavoie,K.H.,andNorthup,D.E.,2005,Bacteriaasindicatorsofhuman impactincaves, in Rea,G.T.,ed.,Proceedingsofthe17thNational CaveandKarstManagementSymposiumProceedings:October31– November4,2005,Albany,NewYork,p.40–47. Lieberman,H.R.,2007,Hydrationandcognition:Acriticalreviewand recommendationsforfutureresearch:JournaloftheAmerican CollegeofNutrition,v.26,supplement5,p.555S–561S. Mattimore,V.,andBattista,J.R.,1996,Radioresistanceof Deinococcus radiodurans: Functionsnecessarytosurviveionizingradiationarealso necessarytosurvivedesiccation:JournalofBacteriology,v.178, p.633–637. McCarty,G.W.,andBremmer,J.M.,1991,Productionofureaseby microbialactivityinsoilsunderaerobicandanaerobicconditions: BiologyofFertileSoils,v.11,p.228–230.doi:10.1007/BF00335772. Miteva,V.I.,andBrenchley,J.E.,2005,Detectionandisolationof ultrasmallmicroorganismsfroma120,000-year-oldGreenlandglacier icecore:AppliedandEnvironmentalMicrobiology,v.71, p.7806–7818.doi:10.1128/AEM.71.12.7806-7818.2005. Mobley,H.L.T.,andHausinger,R.P.,1989,Microbialureases:Significance,regulation,andmolecularcharacterization:Microbiology Reviews,v.53,p.85–108. Northup,D.E.,Barns,S.M.,Yu,L.E.,Spilde,M.N.,Schelble,R.T., Dano,K.E.,Crossey,L.J.,Connolly,C.A.,Boston,P.J.,Natvig, D.O.,andDahm,C.N.,2003,Diversemicrobialcommunities inhabitingferromanganesedepositsinLechuguillaandSpiderCaves: EnvironmentalMicrobiology,v.5,p.1071–1086.doi:10.1046/j.14622920.2003.00500.x. Northup,D.E.,Beck,K.M.,andMallory,L.M.,1997,Humanimpacton themicrobialcommunitiesofLechuguillaCave:Isprotectionpossible duringactiveexploration?[abstract]:JournalofCaveandKarst Studies,v.59,166p. Pace,N.R.,1997,Amolecularviewofmicrobialdiversityandthe biosphere:Science,v.276,p.734–740.doi:10.1046/j.1462-2920. 2003.00500.x. Palmer,A.N.,andPalmer,M.V.,2000,Hydrochemicalinterpretationof cavepatternsintheGuadalupeMountains,NewMexico: Journalof CaveandKarstStudies ,v.62,p.91–108. Raesly,R.L.,andGates,J.E.,1987,Winterhabitatselectionbynorth temperatecavebats:AmericanMidlandNaturalist,v.118,p.15–31. Rainey,F.A.,Ray,K.,Ferreira,M.,Gatz,B.Z.,FernandaNobre,M., Bagaley,D.,Rash,B.A.,Park,M.-J.,Earl,A.M.,Shank,N.C.,Small, A.M.,Henk,M.C.,Battista,J.R.,Ka ¨mpfer,P.,anddaCosta,M.S., 2005,Extensivediversityofionizing-radiation-resistantbacteria recoveredfromSonoranDesertsoilanddescriptionofninenew speciesofthegenus Deinococcus obtainedfromasinglesoilsample: AppliedandEnvironmentalMicrobiology,v.71,p.5225–5235. doi:10.1128/AEM.71.9.5225-5235.2005. Rappe ,M.S.,Connon,S.A.,Vergin,K.L.,andGiovannoni,S.J.,2002, CultivationoftheubiquitousSAR11marinebacterioplanctonclade: Nature,v.418,p.630–633.doi:10.1038/nature00917. Reames,S.,Fish,L.,Burger,P.,andKambesis,P.,1999,DeepSecrets: TheDiscoveryandExplorationofLechuguillaCave:Dayton,Cave Books,381p. Roman,M.,Dovi,R.,Yoder,R.,Dias,F.,andWarden,B.,1991, Determinationbyionchromatographyandspectrophotometryofthe effectsofpreservationonnitriteandnitrate:JournalofChromatographyA,v.546,p.341–346.10.1016/S0021-9673(01)93032-8. Scholle,P.A.,Ulmer,D.S.,andMelim,L.A.,1992,Late-stagecalcitesin thePermianCapitanFormationanditsequivalents,DelawareBasin margin,westTexasandNewMexico:evidenceforreplacementof precursorevaporites:Sedimentology,v.39,p.207–234.doi:10.1111/ j.1365-3091.1992.tb01035.x. Sellin,M.,Gillespie,W.A.,Cooke,D.I.,Sylvester,D.G.H.,andAnderson, J.D.,1975,Micrococcalurinary-tractinfectionsinyoungwomen:The Lancet,v.2,p.570–572.10.1016/S0140-6736(75)90166-X. Spear,J.R.,Barton,H.A.,Robertson,C.E.,Francis,C.A.,andPace, N.R.,2007,Microbialcommunitybiofabricsinageothermalmine adit:AppliedandEnvironmentalMicrobiology,v.73,p.6172–6180. doi:10.1128/AEM.00393-07. Stamatakis,A.,Hoover,P.,andRougemont,J.,2008,AfastbootstrappingalgorithmfortheRAxMLweb-servers:SystematicBiology, v.57,p.758–771.doi:10.1080/10635150802429642. Stone,W.,amEnde,B.,andPaulsen,M.,2002,BeyondtheDeep:The DeadlyDescentintotheWorld’sMostTreacherousCave:NewYork, TimeWarnerBooks,351p. SummersEngel,A.,Meisinger,D.B.,Porter,M.L.,Payn,R.A.,Schmid, M.,Stern,L.A.,Schleifer,K.H.,andLee,N.M.,2010,Linking phylogeneticandfunctionaldiversitytonutrientspiralinginmicrobial matsfromLowerKaneCave(USA):TheISMEJournal,v.4, p.98–110.doi:10.1038/ismej.2009.91. Tabor,J.M.,2010,BlindDescent:TheQuesttoDiscovertheDeepest PlaceonEarth:NewYork,RandomHouse,286p. Walter,B.,Ha ¨nssler,E.,Kalinowski,J.,andBurkovski,A.,2007, Nitrogenmetabolismandnitrogencontrolincorynebacteria:Variationsofacommontheme:JournalofMolecularandMicrobiology Biotechnology,v.12,p.131–138.doi:10.1159/000096468. Wieser,M.,Denner,E.B.,Ka ¨mpfer,P.,Schumann,P.,Tindall,B., Steiner,U.,Vybiral,D.,Lubitz,W.,Maszenan,A.M.,Patel,B.K., Seviour,R.J.,Radax,C.,andBusse,H.J.,2002,Emendeddescriptions ofthegenus Micrococcus,Micrococcusluteus (Cohn1872)and Micrococcuslylae (Kloosetal.1974):InternationalJournalof SystematicandEvolutionaryMicrobiology,v.52,p.629–637. Wright,P.A.,1995,Nitrogenexcretion:Threeendproducts,many physiologicalroles:JournalofExperimentalBiology,v.198, p.273–281. ZhaoBin,HeYiLiang,andZhangXiaoFan,2010,Nitrogenremoval capabilitythroughsimultaneousheterotrophicnitrificationand aerobicdenitrificationby Bacillus sp.LY:EnvironmentalTechnology, v.31,p.409–416.doi:10.1080/09593330903508922. M.D.J OHNSTON ,B.A.M UENCH ,E.D.B ANKS AND H.A.B ARTON JournalofCaveandKarstStudies, December2012 N 291

PAGE 60

AMETHODTODETERMINECOVER-COLLAPSE FREQUENCYINTHEWESTERNPENNYROYALKARST OFKENTUCKY J AMES C.C URRENS ,R ANDALL L.P AYLOR ,E.G LYNN B ECK AND B ART D AVIDSON KentuckyGeologicalSurvey,UniversityofKentucky,228MiningandMineralResourcesBuilding,Lexington,KY40506-0107 Abstract: Todeterminetherateofcover-collapsesinkholeformationinChristian County,Kentucky,weusedlargescaleaerialphotographstakennearlytwentyyears apart.Thenegativeswereenlargedandprintedto1:3,000scaleandexaminedfor collapses.Thephotographsconstrainedthetimeperiodwithinwhichthecollapsecould haveoccurred,andthelargescaleoftheprintsprovidedameanstoidentify,locate,and field-verifythecovercollapses.Allfeaturesnotedonthephotographswerecheckedin thefield.Sinkholesseenonthelaterphotographs,butnottheearlierones,were recorded.Therateofformationcalculatedwas0.2cover-collapsekm 2 2 yr 2 1 I NTRODUCTION Covercollapseisthephenomenaofapparentlysudden collapseofsoilorotherunconsolidatedcoveroverkarstic bedrock.InKentucky,covercollapsefrequentlydamages buildings,roads,utilitylines,andfarmequipment.Ithas killedlivestock,includingsomethoroughbredhorses,and hasinjuredpeople.TheKentuckyGeologicalSurvey estimatesatotaleconomiccostof$20millionannuallyin Kentuckyfromkarst-generatedgeologichazards(Dinger etal.,2007).Thesurveyrecordsanaverageoftwodozen covercollapsesperyearandhasdevelopedacasehistory filespanningsomethirtyyears.Inthispaper,wereporta site-specificstudyofcollapsefrequencyinasmallareaof theWesternPennyroyalsinkholeplaineastofHopkinsville inChristianCounty,Kentucky(Fig.1). C OVER -C OLLAPSE P ROCESS Thedevelopmentofvoidsinunconsolidatedcover overlyingkarsticbedrockhasbeenstudiedfordecades (Beck,1991;WhiteandWhite,1992).Smallvoidsinsoilat depthsofafewmetersarecomparativelystablebecauseof thelateraldistributionoftheoverburden-inducedstressby thearchedroofofthevoid.Thevoidsareenlargedbya lossofcohesionandloadingofthearch-formingmaterial causedbyeitherawettingfrontofsoilwaterfrom infiltratingprecipitationorbyrapiddrainingofan inundatedvoid.Thesaturatedporesintheunconsolidated covercannotdrainasquicklyastheconduit-connected void.Thewettingandincreasedporepressureresultinan incrementallossofstrengthoftheregolitharch(Tharp, 1999)andtheundersideofthearchsloughingintothesoil void.Ultimately,therepeatedsloughingfromwettingand dryingoftheunconsolidatedcoverpropagatesthearchedovervoidtonearthelandsurface(HyattandJacobs,1996; Walthametal.,2005).Thesuddenappearanceofacovercollapsesinkholeisinitiatedwhenthearchbecomestoo thintosupportitsownweightandshearstheremaining soilinanearlycircularpattern(Fig.2).Ifsufficientvolume isnotavailableintheunderlyingbedrockcavitytostore thecollapsedsoil,theloosematerialistransportedawayby groundwaterflowthroughthebedrockconduit.Although thegenesisofcovercollapseiswellunderstood,precisely predictingthetimeandplaceatwhichacollapsewilloccur isnotyetpossible(Hyattetal.,2001). S TUDY A REA Thestudyareais4.04km 2 ineast-centralChristian County,Kentucky(Fig.3).Thetopographywithinthe studyareaiskarstplainandasinglelowhill,giving23.5m oflocalrelief,formedbyresistancetodissolutionofthe basalpartoftheBethelSandstone.Landuseatthetimeof thestudy(2004)waslargelypastureandrow-croppedfields withscatteredfarmsteads,aretailagriculturesupplystore, acementplant,andarestaurant.Theboundarieswere definedbytheoverlappingareaofstereoaerialphotograph pairs.Thestudyareawasselectedwithoutanyprior knowledgeofexistingcovercollapsesinthearea. TheexposedMississippiansection,inascendingorder,is Ste.GenevieveLimestone,RenaultLimestone,andBethel Sandstone(Klemic,1967).Thebedrockatthebaseofthe stratigraphiccolumnispredominantlyoobiosparitesand micriticlimestones,medium-tothick-bedded,andisgreater than95percentcalciumcarbonate.Interbeddedthinshale andargillaceouscarbonatesareaminorinterruptiontothe otherwiseverypurecarbonatesection.Theresidual10mof Bethelisacalcite-cemented,argillaceousquartzarenitethat weathersintoafriable,porous,sandyresiduumthatreadily slumpsintounderlyingsinkholes(Klemic,1967).TheLost RiverChertisexposednearthebaseoftheSte.Genevievein localquarries,butisbelowthedepthofkarstdevelopment inthestudyarea.TheexposedSte.GenevieveLimestoneis *CorrespondingAuthor:currens@uky.edu J.C.Currens,R.L.Paylor,E.G.Beck,andB.Davidson–Amethodtodeterminecover-collapsefrequencyintheWesternPennyroyal karstofKentucky. JournalofCaveandKarstStudies, v.74,no.3,p.292–299.DOI:10.4311/2011ES0247 292 N JournalofCaveandKarstStudies, December2012

PAGE 61

over52mthick,whiletheRenaultLimestoneissome15to 29mthick.Theregionalgentledipof3mkm 2 1 tothenorth istheonlystructureintheotherwiseflat-lyingbedrockatthe studysite. Thecovercollapsesinventoriedweremostlyinthe outcropareaoftheRenault.Becauseofthepurityand thicknessofthecarbonates,thepresenceoftopographically mappeddolines,andthemoderatetotalreliefof60mwithin thestudyarea,weexpectedtherateofoccurrenceofcover collapsetobecomparativelyhigh.Theconditionsinthe studyareaarenearlyidealforcover-collapsedevelopment. M ETHODS Weusedasimpleandinexpensivemethodtolocate sinkholesandconstrainthetimeofcovercollapse.Because wedidnothaveaccesstoamagnifyingstereoscope,the KentuckyGeologicalSurveypurchasedprintsofblackand-white,low-altitude,large-scale,visible-light,aerial photographyatanimagedscaleof1:12,000(1cm 5 120m;1in. 5 1,000ft)fromtheTennesseeValley Authority.ThephotographsweretakenMarch9,1971, andJanuary31,1991.Althoughwealsoobtainedstereo setsofcontactprints,themostusefulimageswere enlargementsofthecentralimagefromthesets.The enlargementswereprintedatascaleof1:3,000(1cm 5 30m;1in 5 250ft),fourtimesthescaleofthenegative. Using2-powermagnifyingglasses,wevisuallyscannedthe enlargementssystematicallyforfeaturesappearingtobe sinkholes.Theemulsiongrainsontheprintweresufficientlysmallincomparisonwiththetypicalcover-collapse thatshadowscastontheinteriorofacollapselessthana meterindiametercouldbediscernedfromthosecastby smallcedartrees,forexample.Further,labelingdevices couldbeattachedtotheenlargementprinttopreservethe interpretivedata. Wealsosearchedforcover-collapsefeaturesonthe stereo-paircontactprintsandthedigitalimagesmadeat theKGSfromscansoftheenlargements.Mostofthe featuresidentifiedonthe1:3,000-scaleenlargementscould notberelocatedwithconfidenceonthe1:12,000-scale contactprints.Scannedenlargementsweresavedas16-bit gray-scaleTIFFfilesat1,200dpi,givingapixelsizeof roughly0.5matgroundlevel.Thefilesizewaslarge (428,853kb)andpreventedviewingthescanwithimage viewersonmostoftheavailablecomputers.Wedidview theimageswithGISsoftware,butcouldnotrelocateany cover-collapselocationsonthedigitalimagesdueto pixilation. Fiftypotentialcover-collapsesiteswereselectedforfield inspectionfromtheenlargedprints(Table1).KGSstaff field-checkedallofthesitesanddeterminediftheywere,in fact,covercollapses.Thosesitesthathaddevelopedwithin Figure2.Aclassicexampleofacovercollapseinthe studyarea. Figure1.TheblackpolygonisthestudyareainChristian County,eastofHopkinsville,WesternPennyroyalregionin Kentucky.Thegrayshadedareaisunderlainbykarstic carbonates. J.C.C URRENS ,R.L.P AYLOR ,E.G.B ECK AND B.D AVIDSON JournalofCaveandKarstStudies, December2012 N 293

PAGE 62

thetwenty-yearperiodbracketedbythephotographswere identified.Wealsofoundasmallnumberofcovercollapses thatdidnotappearuntilafterthe1991photography.These weretoorecenttoincludeinthecalculationoftherate,but weredocumentedforfuturereference.Somefeaturesthat werevisibleontheearlierphotographs,butnotonthelater ones,werealsonoted.Wealsoreceivedalimitednumberof reportsthatacollapsehadoccurredandhadbeenfilledand gradedduringtheperiodbetweentheaerialphotography. Suchcovercollapseswereincludedintheratecalculation onlyifwefoundfieldevidencethatthereportwascorrect. Fieldevidenceforafilledsinkholeincludedacircular variationintextureandcolorofvegetation,buriedtrash exposedatthesurface,subsidenceduetosoilcompaction,or Figure3.Thecover-collapseinventoryareaintheHopkinsville7.5-minutequadrangleisinsidetheblackline.Thedisrupted texturedareaenclosedbyagraylineisthepropertyofalimestonequarry,whichwasexcludedfromtheinventory.Solid asterisksarecover-collapsesinkholesidentifiedonthe1991aerialphotographandverifiedinthefield.Hollowasterisksare otherfeaturesfromthesameimagesdeterminednottobecover-collapse. A METHODTODETERMINECOVER-COLLAPSEFREQUENCYINTHE W ESTERN P ENNYROYALKARSTOF K ENTUCKY 294 N JournalofCaveandKarstStudies, December2012

PAGE 63

Table1.Featuresidentifiedonaerialphotographsandfieldverifiedaspossiblecovercollapsessince1971. Fieldand MapID LocationPresenton 1971Aerial Photograph Presenton 1991Aerial Photograph CorrectlyIdentified onPhotoasCover Collapse? Stratigraphic UnitCommentsandNotes Long., u WLat., u N 5 c 2 87.418936.82111NYYRenaultPA a 6 c 2 87.417836.81944NYYRenaultSH b –Suspect 7 c 2 87.424236.81861NYYRenaultSH/PAGroup 11 2 87.405636.81833NYNRenaultPA 14 2 87.405636.81833YYNRenaultPA/SH–Omittedfromrate computation.Onboth photographs. 15 c 2 87.425836.81806NYYRenaultSH-Obvious 19 c 2 87.404736.81528PossiblyYYRenaultSH 22 c 2 87.407236.81389NYYRenaultSH-Suspect;photographed 32 c 2 87.402236.81083NYYRenaultSH-NearexistingsinkonUS71 34 c 2 87.402536.81056NYYRenaultSH-NearexistingsinkonUS71 37 2 87.405636.80917NYYBethelSandstoneSH-Maybeothershere. 41 2 87.417836.80667NYNSte.GenevievePA 46 c 2 87.417536.80306NYYSte.GenevieveSH-photograph 47 c 2 87.411736.80306Ahint?YYSte.GenevievePA-anotherPAtotheSE 48 c 2 87.416436.80278YYYSte.GenevieveSH-Seemstohavemoved 49 c 2 87.408636.80111Ahint?YYRenaultPA-hasbeenfilledin 50 c 2 87.407936.81183NNNRenaultOwnersaysformedbetween1975 and1980 1 2 87.414436.8225NYNRenaultPA 2 2 87.414436.8225NYNRenaultPA 3 2 87.417536.82222NYNRenaultSH/PA 4 2 87.419436.82167Ahint?YNRenaultPA 8 2 87.424236.81833NYNRenaultSH/PA-Group 9 2 87.423936.81833NYNRenaultSH/PA-Group 10 2 87.423936.81806NYNRenaultSH/PA-Group 12 2 87.423936.81806NYNRenaultSH/PA-Group 13 2 87.423936.81806NYNRenaultSH/PA-Group 16 2 87.406136.8175NYNRenaultPA 17 2 87.404436.81722NYNRenaultPA-Suspect 18 c 2 87.408136.81556NYYRenaultSH 20 2 87.406436.815NYNRenaultPA? 21 2 87.408336.81472NYNRenaultSH-Suspect 24 c 2 87.406136.81389NYNRenaultSH 25 2 87.425336.81361NYNRenaultPA 26 c 2 87.406436.81389NYNRenaultSH-Hasbeenfilled. 27 2 87.411936.81222NYNRenaultSH-Suspect J.C.C URRENS ,R.L.P AYLOR ,E.G.B ECK AND B.D AVIDSON JournalofCaveandKarstStudies, December2012 N 295

PAGE 64

Fieldand MapID LocationPresenton 1971Aerial Photograph Presenton 1991Aerial Photograph CorrectlyIdentified onPhotoasCover Collapse? Stratigraphic UnitCommentsandNotes Long., u WLat., u N 28 2 87.429736.81167NYNSte.GenevieveSH-Suspect-possiblyanotherto southwest 29 c 2 87.410836.81194NYNRenaultSH 30 2 87.411436.81194NYNRenaultSH-Suspect 33 2 87.427236.81028NYNSte.GenevieveSH/PA-Possibleberm? 36 2 87.405836.80972NYNBethelSandstoneSH 38 2 87.408636.80889NYNRenaultPA-Inareanowoccupiedby quarry 39 2 87.405636.80889NNBethelSandstoneOldpondsite.Inareanow occupiedbyquarry. 40 2 87.410336.80778NYNRenaultSH–Suspectfeatureinareanow occupiedbyquarry 42 2 87.405836.80639NYNBethelSandstoneSH 43 2 87.421936.80583NYNSte.GenevievePA 44 2 87.405636.80583NYNBethelSandstoneSH 45 c 2 87.418636.80333NYNSte.GenevieveSH–Twootherspossibleafew feettosouthwest 23 2 87.413336.81389NY?RenaultSH-NearexistingUS71sink destroyedbyroadinfilling. 31 2 87.405336.81083NY?RenaultSH?35 2 87.413136.81028NY?RenaultSH(quarriedaway) a PA–Anobstacleinacropfieldthatwasdrivenaroundduringtillageoftheground. b SH–Afieldverifiedsinkhole. c Usedtocalculaterateofoccurrence. Table1.Continued. A METHODTODETERMINECOVER-COLLAPSEFREQUENCYINTHE W ESTERN P ENNYROYALKARSTOF K ENTUCKY 296 N JournalofCaveandKarstStudies, December2012

PAGE 65

alocalchangeinsoilcolorifthesoilwasbare.After comparingrecentsinkholestosinkholesthatareclearlyold, wethinkthatinsufficienttimehaspassedinthe35years betweentheearlierphotographyandourstudyforanyrecent covercollapsestohavebeennaturallyobscuredbyslumping anderosion.Finally,ourmethoddoesnotworkinwooded areas,althoughwoodlandscanstillbefield-checked,which wedid.Nocovercollapsewasfoundinthesmallwooded area. L IMITATIONSON D ATA A NALYSIS Theprimarygoalofthisstudywasthedemonstration ofthetechnique,andweanticipatedseverallimitationson Figure4.The1991aerialphotographofthestudyareaintheHopkinsvilleEast7.5-minutequadrangle,ChristianCounty, Ky.Thedarkareainthecenteristhenow-floodedlimestonequarrythatwasexcludedfromtheanalysis.Northisatthetopthe photograph.Thescaleofthenegativeis1:12,000. J.C.C URRENS ,R.L.P AYLOR ,E.G.B ECK AND B.D AVIDSON JournalofCaveandKarstStudies, December2012 N 297

PAGE 66

theaccuracyofthecover-collapserate.Theaerialcoverage obtainedfromTVAhasoneofthelargestscalesavailable, anditisonlyavailableinKentuckyfortheTennesseeRiver Valley.Thelimitedareaofphotographyrestrictedthe selectionofstudyareas.Althoughtheareawaswellsuited forcovercollapse,wehadsomeconcernthatnonewould beidentifiableontheimagesorthatnocollapseshad actuallyoccurredinthestudyarea.Thisprovedunfounded.Third,thesizeofthestudyareawaslimitedbythe budget,whichresultedinitbeingtoosmallfora statisticallysignificantsampling(Beck,1991)thatwould berepresentativeofthelargerregion.Ideally,anareaof10 ormoresquaremilesshouldbeused,orseveralsmaller samplingsitesscatteredovertheregion. Large-scalephotographicprintsarenotcommonlyused becauseoftheincreasedcostofprocessingtheprints necessarytocoverthesameareaonthegroundwhen comparedtocontactprints.Itmaybecomeincreasingly difficulttoobtaincontactprintsandenlargementsof traditionalanalogphotographybecauseoftheclosureof processingfacilities.AsofFebruary2010,theTennessee ValleyAuthoritystillofferedtheenlargementservicethat KGSutilized(https://maps.tva.com/Scripts/MetaWeb/map_ aerial.asp). Moreproblematicwasamisunderstandingbetweenthe KentuckyGeologicalSurveyandtheTennesseeValley Authorityabouttheareatobecovered.TheKGSintended thequarry(thelargedarkareainthecenterofFig.4)tobe areferencelandmarkoutsideofthephotographframe,not theobjectoftheexercise.Wedidnotrealizethe photographsencompassedthestonequarryuntilwe receivedthem.Itwasnotpossibletoreplacetheprints. Thestaffofthestonequarrywascontacted,andthey statedthatthequarryopenedshortlyafter1971and operatedlessthantenyears.Theywerenotawareofany historyofcover-collapsesinkholesinducedbydewatering thequarry.Thequarryoccupies0.88km 2 or18percentof thestudyarea.Onlyoneprobablecovercollapsewasfound withinthequarryareaonthe1971image,anditcouldnot beverifiedinthefieldbecauseitslocationhadbeenmined by1991.Wecouldnotfindanyobviouspatternthat suggestedinducedsinkholeclusteringnearthequarry, eitherinthefieldoronthephotographs.Wethinkthe effectofthequarryandthestudywasnegligiblebeyondthe obviouslossofstudyarea. Twoofthefeaturesofthearea,quarrydewateringand highlyfavorablegeology,potentiallyinflatethepopulation ofcover-collapsefeaturesandoneoftheothers,uncounted sinkholesthatoccurredandthenwerefilledbetween photographs,possiblydeflatesthecount.Further,ifcover collapsesoccurredduringtheperiodbetweenthetwo photosthatwedidnotfind,thesewouldleadusto underestimatetherateofformation.Becauseofthepure, thicklimestoneand60mofreliefwithinthestudyarea,we thinkourrateofsinkholeformationshouldbeconsidered amaximumforapplicationtootherareas. R ESULTS Weidentifiedfiftycover-collapselikefeaturesthat occurredbetween1971and1991fromtheaerialphotographandreportsfromlocalresidents.Thephotographic enlargementsmadepossibletheidentificationofthe featuresandtheirlocationinthefield.Thedetailofthe photographicenlargementwasanimportantfactorinthe successofthetechnique. Also,wephysicallyexaminedanestimated90percentof thestudyarea.Ofthefiftypossiblesitesidentifiedonthe photographs,thirteencouldnotbefoundinthefieldand therewasnoevidenceofcollapseinthevicinityofthe featureseenonthephotograph.Oftheremainingthirtysevensites,sixteen(43percent)werecorrectlyidentifiedon thephotographsastoorigin(whethercover-collapseor not)whenlocatedinthefield.Fifteenofthecover-collapse sinkholeswereaccuratelyidentifiedonthephotographs andverifiedinthefield,theremainingcollapseswere thoughtsomeotherfeatureuntilfieldchecked(Table1). Onecovercollapseidentifiedonthe1991enlargementwas alsoonthe1971imageandwasnotcounted(Table1). Ultimately,atotalofeighteensitesweredeterminedtobe covercollapsesthatoccurredwithinthetimeframeofthe twoimages. Theprimarybenefitofthephotographicenlargements wastheydirecteduswheretofocusourfieldwork.The totalcovercollapsescorrectlyidentifiedfromthephotographs(36percentofthetotalfeaturesand43percentof thefeaturesthatcouldbefoundinthefield)isauseful successrate,butcouldbeexpectedtoimprovewith experience. Calculationoftherateofformationofthecovercollapseissubjecttothelimitationscitedabove.We excludedtheareaofthequarryasdefinedbytheproperty line(Fig.3)fromtheratecalculation.Therateofcovercollapseeventsforthestudyareais0.2km 2 2 yr 2 1 Table2.Resultsofinventoryandfield-verifiedcover-collapseeventsbetweenaerialphotographsbyTVAin1971and1991 (19.9years). AreaifQuarry Included/Excluded Count,Cover-Collapse SitesInventoryArea,km 2 CoverCollapseper UnitArea,km 2 CoverCollapse/Unit Area/Year,km 2 2 yr 2 1 Included184.923.70.18 Excluded184.044.50.22 A METHODTODETERMINECOVER-COLLAPSEFREQUENCYINTHE W ESTERN P ENNYROYALKARSTOF K ENTUCKY 298 N JournalofCaveandKarstStudies, December2012

PAGE 67

(0.6mi 2 2 yr 2 1 ;Table2),whichisconsistentwithsome previousstudies.Hyattetal.(1996)reportedmorethan 312covercollapsesinthevicinityofAlbany,Georgia,that developedwithina25km 2 studyareawithinaweek followingatropicalstorminJuly1994.Thedensityofthe featuresrangedfrom0.09km 2 2 to5.3km 2 2 (0.21mi 2 2 to 13.7mi 2 2 ).Afollow-upstudybyHyattetal.(1999)is perhapsthemostcompletedescriptivestudyofcovercollapsesinkholesintheUnitedStatestodate.Weused theirdatatocalculatetheoccurrenceofcover-collapseat 12.5eventspersquarekilometerovertheoneweek observationperiod.Theminimumannualratebasedon the1994datawouldbe0.2km 2 2 yr 2 1 .Beck(1991)reported 0.11km 2 2 yr 2 1 (0.29mi 2 2 yr 2 1 )inFlorida,andKetelleetal. (1988)reportedarangeof0.04to0.64km 2 2 yr 2 1 (0.1to 1.69mi 2 2 yr 2 1 )ineasternTennessee.Wilsonetal.(1987) reported0.01km 2 2 yr 2 1 (0.04mi 2 2 yr 2 1 ),alsoinFlorida. C ONCLUSIONS Theuseofenlargementsofconventionalaerialphotographytobothlocateandconstrainthedateofformationof cover-collapsesinkholesproved practical,inexpensive,and reasonablyrobustforthes inkholeplainoftheWestern KentuckyPennyroyal.Althoug hgroundinspectionisstill needed,fieldwork,regardlessofhowthorough,canonly grosslyestimatewhenanunobse rvedcollapseoccurred.The photography-basedfieldinve ntorywaspossiblebecausethe landuseofthestudyareawasdominatedbypastureandcrop fields.Exceptforanyundiscove redcover-collapsesinkholes thatoccurredandwerefilledintheinterimbetween photographs,themethodcanbe categorizedasexhaustive. Inthestudyarea,covercollapseoccursat0.2km 2 2 yr 2 1 (0.58mi 2 2 yr 2 1 ),consistentwithmostpreviousstudies. Becauseofthefavorablegeologyofthick,purecarbonates, theratecalculatedhereshouldbeconsideredamaximumif appliedtootherkarstareasinKentucky. R EFERENCES Beck,B.F.,1991,Oncalculatingtheriskofsinkholecollapse, in Kasting, E.H.,andKasting,K.M.,eds.,AppalachianKarst,Proceedingsofthe AppalachianKarstSymposium,Radford,Virgina,March23–26, 1991:NationalSpeleologicalSociety,Huntsville,p.231–236. Dinger,J.S.,Zourarakis,D.P.,andCurrens,J.C.,2007,Spectral enhancementandautomatedextractionofpotentialsinkhole featuresfromNAIPimagery—initialinvestigations:Journalof EnvironmentalInformatics,v.10,no.1,p.22–29.doi:10.3808/ jei.200700096. Hyatt,J.A.,andJacobs,P.M.,1996,Distributionandmorphologyof sinkholestriggeredbyfloodingfollowingTropicalStormAlberto atAlbany,Georgia,USA:Geomorphology,v.17,p.305–316. doi:10.1016/0169-555X(96)00014-1. Hyatt,J.A.,Wilkes,H.P.,andJacobs,P.M.,1999,Spatialrelationships betweennewandoldsinkholesincoveredkarst,Albany,Georgia, USA, in Beck,B.F.,Pettit,A.J.,andHerring,J.G.,eds.,HydrogeologyandEngineeringGeologyofSinkholesandKarst:Rotterdam,Balkema,p.37–44. Hyatt,J.A.,Wilson,R.,Givens,J.S.,andJacobs,P.M.,2001, Topographic,geologic,andhydrogeologiccontrolsondimensions andlocationsofsinkholesinthickcoveredkarst,LowndesCounty, Georgia, in Beck,B.F.,andHerring,J.G.,eds.,Proceedingsof GeotechnicalandEnvironmentalApplicationsofKarstGeologyand Hydrology,Rotterdam,Balkema,p.37–45. Ketelle,R.H.,Newton,J.G.,andTanner,J.M.,1988,Karstsubsidencein EastTennessee, in Proceedings2 nd ConferenceonEnvironmental ProblemsinKarstTerranesandTheirSolutions:Dublin,Ohio, NationalWaterWellAssociation,p.51–65. Klemic,H.,1967,GeologicmapoftheHopkinsvillequadrangle,Christian County,Kentucky:U.S.GeologicalSurvey,GeologicQuadrangle MapGQ-651,scale1:24,000. Tharp,T.M.,1999,Mechanicsofupwardpropagationofcover-collapse sinkholes:EngineeringGeology,v.52,p.23–33.doi.org/10.1016/ S0013-7952(98)00051-9. Waltham,T.,Bell,F.,andCu lshaw,M.,2005,Sinkholesand Subsidence:KarstandCavernousRocksinEngineeringand Construction:NewYork,Springe r-PraxisBooksinGeophysical Sciences,382p. White,W.B.,andWhite,E.L.,1992,Sinkholesandsinkholecollapses in Majundar,S.K.,Forbes,G.S.,Miller,E.W.,andSchmalz,R.F.,eds., NaturalandTechnologicalDisasters:Causes,EffectsandPreventativeMeasures,PennsylvaniaAcademyofScience,p.280–293. Wilson,W.L.,McDonald,K.M.,Barfus,B.L.,andBeck,B.F.,1987. HydrogeologicFactorsAssociatedwithRecentSinkholeDevelopmentintheOrlandoArea,Flori da;FloridaSinkholeResearch Institute,UniversityofCentra lFlorida,ReportNo.87-88-4, 109p. J.C.C URRENS ,R.L.P AYLOR ,E.G.B ECK AND B.D AVIDSON JournalofCaveandKarstStudies, December2012 N 299


Description
Articles:
http://dx.doi.org/10.4311/2010ES0138R1 Electrical resistivity
imaging of cave Divas¡ka jama, Slovenia / M. Andrej and S. Uros
http://dx.doi.org/10.4311/2011JCKS0194 A survey of the algal
flora of anthropogenic caves of Campi Flegrei (Naples, Italy)
archeological district / P. Cennamo, C. Marzano, C. Ciniglia,
G. Pinto, P. Cappelletti, P. Caputo, and A. Pollio
http://dx.doi.org/10.4311/2010ES0188R Assessment of spatial
properties of karst areas on a regional scale using GIS and
statistics the case of Slovenia / M. Komac and J. Urbanc
http://dx.doi.org/10.4311/2011PA0239 Lizards and snakes
(Lepidosauria, Squamata) from the late Quaternary of the state
of Ceara in northeastern Brazil / A.S. Hsiou, P.V. de
Oliveira, C.L. Ximenes, and M.S.S. Viana
http://dx.doi.org/10.4311/2011JCKS0208 Diet of the newt,
Triturus carnifex (Laurenti, 1768), in the flooded karst
sinkhole Pozzo del Merro, central Italy / A. Romano, S.
Salvidio, R. Palozzi, and V. Sbordoni
http://dx.doi.org/10.4311/2011MB0227 Human urine in Lechuguilla
Cave: the microbiological impact and potential for
bioremediation / M.D. Johnston, B.A. Muench, E.D. Banks, and
H.A. Barton http://dx.doi.org/10.4311/2011ES0247 A method to
determine cover-collapse frequency in the Western Pennyroyal
karst of Kentucky / J.C. Currens, R.L. Paylor, E.G. Beck, and
B. Davidson


printinsert_linkshareget_appmore_horiz

Download Options

close
Choose Size
Choose file type
Cite this item close

APA

Cras ut cursus ante, a fringilla nunc. Mauris lorem nunc, cursus sit amet enim ac, vehicula vestibulum mi. Mauris viverra nisl vel enim faucibus porta. Praesent sit amet ornare diam, non finibus nulla.

MLA

Cras efficitur magna et sapien varius, luctus ullamcorper dolor convallis. Orci varius natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Fusce sit amet justo ut erat laoreet congue sed a ante.

CHICAGO

Phasellus ornare in augue eu imperdiet. Donec malesuada sapien ante, at vehicula orci tempor molestie. Proin vitae urna elit. Pellentesque vitae nisi et diam euismod malesuada aliquet non erat.

WIKIPEDIA

Nunc fringilla dolor ut dictum placerat. Proin ac neque rutrum, consectetur ligula id, laoreet ligula. Nulla lorem massa, consectetur vitae consequat in, lobortis at dolor. Nunc sed leo odio.