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

Restriction:
Open Access - Permission by Publisher
Original Version:
Vol. 73, no. 1 (2011)

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:
K26-02058 ( USFLDC DOI )
k26.2058 ( USFLDC Handle )
11260 ( karstportal - original NodeID )
0146-9517 ( ISSN )

USFLDC Membership

Aggregations:
Karst Information Portal

Postcard Information

Format:
Serial

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

DETERMININGGEOPHYSICALPROPERTIESOF ANEAR-SURFACECAVETHROUGHINTEGRATED MICROGRAVITYVERTICALGRADIENTANDELECTRICAL RESISTIVITYTOMOGRAPHYMEASUREMENTS M ARCO G AMBETTA 1 *,E GIDIO A RMADILLO 2 ,C OSMO C ARMISCIANO 1 ,P AOLO S TEFANELLI 1 ,L UCA C OCCHI 1 AND F ABIO C ARATORI T ONTINI 3 Abstract: Vertical-gradientmicrogravityandelectrical-resistivitytomograph y geophysicalsurveyswereperformedoverashallowcaveintheItalianArme tta Mountainkarstarea,closetotheLiguria-Piedmontwatershed.Theaimoft hisstudy wastotestthegeophysicalresponseofaknownshallowcave.Theshallowes tportionof thecaveexhibitsnarrowpassagesand,atabout30metersbelowtheentranc e,afossil meanderlinkingtwolargechambers,thetargetofthegeophysicalsurvey. Theintegrated resultsofthetwosurveysshowacleargeophysicalresponsetothecave.Th esurveys exhibitedhighresistivityvaluesandanegativegravityanomalyoverthe largecave passages.Thisworkconfirmstheabilityofthesegeophysicaltechniques togivetheprecise locationofthevoids,evenincomplexenvironments.Theapplicationofth esetechniques canbesuccessfulforsitesurveyingwherethepresenceofhollowsmaybeex pected. I NTRODUCTION Geophysicaltechniquesprovidequick,low-costimaging ofvoids,andcanplayakeyroleinunderstandingepikarst. Thisunderstandingiscrucialbecausekarstconduitsmay conductpollutantstowardgroundwaterreservoirsand contaminatevulnerablewaterresources(Field,1993). Moreover,collapseofundergroundvoidsmayresultin extensivedamagetopropertyanddangertopeople (Waltham,2005).Inthispaperweshowtheresultsofa geophysicalsurveyconductedoverashallowcavedeveloped undertheCarteidaCollaplainintheMountArmettakarst, whichislocatedneartheLigurian-Piedmontwatershedin northwesternItaly.Theaimofthisstudyistocontributeto thebodyofvoid-detectiontechniques.Thisgeophysical surveyconsistedofamulti-electrodeelectrical-resistivity tomographicsurveycombinedwithavertical-gradient gravitysurvey.Thegeophysicalimagingofunderground voidswasverifiedbycomparisonwiththeauthors’ measurementsofthecavedimensions.Amongvarious geophysicalmethods,electrical-resistivitytomography (ERT)andmicrogravityvertical-gradient(MVG)were chosenbecauseoftheirrecognizedabilitytodetectshallow featuressuchashollowsandcavevoids.Inparticular,the MVGtechniqueemphasizesshallowgravityanomalieswith astrongreductioningeologicnoisetoallowdetermination ofthehorizontalpositionofvoids.Thehighresolution ERT,conversely,outlinestheshapeandverticaldistribution ofthehollows,whichappearashigh-resistivityvolumes. A REA G EOLOGY ThegeophysicalsurveywascarriedoutinCarteida Collaplain(Figs.1and2),locatedat1400mabovesea levelandencompassingmorethan9km 2 intheMount ArmettakarstcomplexinValPennavair,Italy,closetothe Liguria-Piedmontwatershed.Thekarstaquiferfeedsa springusedbylocalmunicipalitiesfordrinkingwater supplies. ThekarstterranebelongstotheCapraunaArmetta tectonicunit,whichisacovernappeemplacedduringthe Brianc onnaisorogenyontheexternalmarginofthe LigurianBrianc onnaisterrane.Theunitfeaturesfour superimposeddeformationsthatproducedlargerecumbent isoclinalfolds(phase1foldsinFig.2)associatedwitha strongaxial-planecleavageandasouthwest-trending lineation.Thesefoldscanberelatedtoasouthwestdirectedoverthrustshear.Thesecondphaseofdeformation(phase2foldsinFig.2)yieldedopentomoderately tightfoldswithsubverticalaxialplanesthatareoverturned towardsthenortheast.Thelaterdeformationsarelongwavelengthopenfoldsaffectingonlythelarge-scalesetting ofthenappe(Menardi-Noguera,1988).Foldsfromthetwo phasesconstrainedthecave’sevolution.Thelargeroom locatedattheintersectionofthephase1and2anticlines (Fig.2)hasbeeninterpretedasthetopmostpartofapit, nowfilledwithcollapsedebris.Thislargepitappearsto haveevolvedoveradome-likestructurefeaturinga pervasiveaxial-planecleavage.Thecave,droppingfrom theentranceat1395mabovesealeveltoapproximately 1200m,alsoseemstobeguidedbothbytherocks’brittle deformationsandthestratigraphy.Thedeepeningconduits followboththestratadipdirectionof20 u andasetof *Correspondingauthor,marco.gambetta@ingv.it 1 INGV—ViaPezzinoBasso2FezzanodiPortovenere(SP)—Italy 2 UniversityofGenova,DIP.TE.RIS,V.leBenedettoXV,Genova—Italy 3 GNSScience,1FairwayDr,Avalon,LowerHutt,NewZealand M.Gambetta,E.Armadillo,C.Carmisciano,P.Stefanelli,L.Cocchi,andF .C.Tontini–Determininggeophysicalpropertiesofanearsurfacecavethroughintegratedmicrogravityverticalgradientandelec tricalresistivitytomographymeasurements. JournalofCaveand KarstStudies, v.73,no.1,p.11–15.DOI:10.4311/jcks2009ex0091 JournalofCaveandKarstStudies, April2011 N 11

PAGE 2

fracturestrending95 u ,withaprevailingdrainagealongthe former.Thedeepestcaveregions,conversely,exhibitan evolutionalongafracturetrending135 u thatcannotbe recognizedbysurfaceobservations. G EOPHYSICAL M ETHODS Fivetwo-dimensional,direct-currentelectrical-resistivitytraverseswerecarriedout.TheERTprospectingwas performedusingaSyscalR1(IrisInstruments)multielectrodesystemwithasetof48electrodesevenlyspaced every5m.AWenner-Schlumbergerarrayconfiguration wasusedwithageometricalfactor, k p n 1 a ,where a isthepotential-electrodespacingand n isaninteger.The Wenner-SchlumbergerarraycombinesthestandardindividualWennerandSchlumbergerarrays,butisadaptedfor Figure1.Thesurveyarea:(A)landscapeoftheCarteidaCollaplain;(B)pi ctureofthelargestroom,whichwasthetargetof thegeophysicalexperiment;partialplanandcross-sectionofthecavesy stem.Cavesurveyperformedwithcompassandtape. r Figure2.Topographyofthesurveyarea,withstructural featuresandtheoutlineofthecave. D ETERMININGGEOPHYSICALPROPERTIESOFANEAR SURFACECAVETHROUGHINTEGRATEDMICROGRAVITYVERTICALGRADIENTANDELEC TRICAL RESISTIVITYTOMOGRAPHYMEASUREMENTS 12 N JournalofCaveandKarstStudies, April2011

PAGE 3

usewithalineofelectrodeswithconstantspacing,as normallyusedin2-Dimaging.Besidesbetterhorizontal coverage,themaximumdepthofpenetrationofthisarray isabout1.5timesgreaterthantheWennerarray(Lokeand Barker,1996).Themeasuredresistancevaluesarefirst reducedtoapparentresistivityvaluesandtheninverted (LokeandBarker,1996)toimagetheresistivitydistributionbelowthetraversepath(Fig.3). TheMVGsurveywasperformedusingaLaCosteRombergmodelDgravitymeter,equippedwithdigital dataacquisitionthroughtheAliodfeedbacksystem,GPS tracking,andautomatictidecorrection.Themeterhasa nominalresolutionofonemicrogal(10nms 1 ).The geographicalpositionsofthe53gravitystations(Fig.2) weredeterminedbyusingadifferentialGPS.Thevertical gradientofthegravityfieldwasmeasuredastheratio betweenthedifferenceoftworecordingsatdifferent heightsandtheheightdifference.Measurementswere recordedforfiveminutesatthetopandbottomofa1.80metertower.Therecordingprocedureincludedatide correction.Therecordingscanbestatisticallyaveragedto giveabestestimateofthegravity,reducingthenoise inherentwithinasinglemeasurementresultingfrom instrumentalerrors,vibrationsofthesupport,andother relatedsourcesofuncertainty.Thegravitydatawere processedtoremovetheinstrumentdrift,determinedby periodicallyreoccupyingthegravitybasestationlocated nearthesurveyarea.Theobserveddrifterrorwasabout 25 gald 1 .TheoverallerroroftheMVGdata,including effectsofnearbytopography,tide,anddriftwasapproximately4to5 galm 1 (Stefanellietal.,2008). R ESULTSAND D ATA I NTERPRETATION Figure3showstheresultsoftheinversion(Lokeand Barker,1996)ofapparentresistivitydatacollectedalong profiles1,2,and,3thatcrossthemainroom.Thecrosssectionoftheroomissuperimposedontheplots.The resultsfromtraversesnumber4and5appeartobemuch noisierthantheothers,eventhoughthecaveoutlineisstill recognizable.Becauseofthehighnoise,theseresultsare notshown. Theresistivityvaluesassociatedwiththeroomrange from6,000to40,000ohm-m(Fig.3,line1).Thehigh variabilityofthevoidÂ’selectricalsignaturemaybeduetoa micrometeorologicaleffectinsidethecave.Aseasonalcave Figure3.Electrical-resistivitytomographyresultforlines1,2,and3w iththeoutlineofthemainroom(seeFig.2). M.G AMBETTA ,E.A RMADILLO ,C.C ARMISCIANO ,P.S TEFANELLI ,L.C OCCHI AND F.C.T ONTINI JournalofCaveandKarstStudies, April2011 N 13

PAGE 4

wind(Fig.4)flowsfromthedeepestregionstowardsthe entrance,yieldingasignificantdecreaseincavemoisture. Thisairflowdriesthecavewalls,thusincreasingtheir electricalresistivity;conversely,thepartsofthecavefar fromthewindpatharemuchmoredamp,andthus theresistivitydrops.Thiseffectisclearlyshowninthe sectionsforlines1and3.Line2,conversely,showslower resistivityvaluesthanline3.Thismaybeduetothefact thatthepassagesbetweentheareaofline3andthelarge roomprobedbyline1arefilledwithlargedebrisand blocks. TheERTsurveyshowsanunpredictedoval-shaped maximumofresistivity(labeled2inFig.3)andis interpretedasevidenceofanunknowncave.Thewelldefinedlow-resistivitybody(labeled3inFig.3)is interpretedasasinkholefilledwithresidualclays. ThedistributionofMVGdatawiththecave-plan outlineareshownintheupperpanelofFigure4.Thelower panelofFigure4showsaprofileacrossthemainhollow. Thereisaverygoodcorrelationbetweenthehorizontal geometryofthecaveandtheminimaofthegradientmap, asitisexpectedfromthelargenegativedensitycontrast betweenvoidsandrocks( 2400kgm 3 ).Thisshowsthat effectsoflocaltopographyhavebeeneffectivelyfiltered outinthegradientmeasurements.Thegradientsurveyhad ameanvalueof0.268mGalm 1 ,incontrastwiththe standardverticalgradientvalueofapproximately 0.3mGalm 1 .Thismayresultfromthesuperpositionof thelocalizednegativeanomalythatiswell-correlatedtothe cavegeometryandamoregeneralregionalfeaturecoming fromotherlarge-scalegeologicaleffects.Theamplitudeof thegradientvariesacrosstheareabyapproximately 0.130mGalm 1 becauseofthelargeundergroundvoids. Thegradientprofileexhibitsanegativeanomalyapproximatelycoincidentwiththeknownvoidgeometry.The minimumabovethelargestroom(Figure4,labeled1)is consistentwiththepresenceofalarge(about20 3 20 3 80m)collapsedpitpartiallyfilledbycollapsedebris.The minimumlabeled2inFigure4isattributedtothetwo underlyinglevelsofcavepassage. C ONCLUSIONS Thecombinedapplicationofmicrogravityverticalgradientsoundingandelectricalresistivitytomography yieldsaccurateimagingofundergroundhollows.The MVGshowsminimacoincidentwiththeknownvoid, showingacapabilityforvoiddetection.TheERTresults, combinedwithofthecavelocationandvolumesuppliedby MVG,cangivehigh-resolutionimagesofunderground voids.Thelowcostandthemoderatelogisticsneedsof bothmethodsallowrapidnoninvasivesurveys,providing reliableresults. A CKNOWLEDGMENTS TheauthorswishalsotothankR.CastelloandF.Poggi (EnvironmentDepartment,RegioneLiguria)fortheir confidenceintheSpeleologyandNatureinPennavaire HighValley(SNAP)ResearchProject,whichhasbeen fundedbythedepartmentactn.154302/12/2005,granted totheAlassioCavingClub.Thanksarealsoduetothe municipalauthoritiesofAlassioandAltoforlogistical supportandpermissiontoaccessthecaves.Theauthors alsoacknowledgetheassistanceprovidedbytheAlassio CavingClubandE.MassaandA.Maifrediinthe undergroundsurvey. Figure4.Themicrogravityvertical-gradientsurvey.Top panel,spatialdistributionofthegradient;bottompanel, profilewiththecavecross-section. D ETERMININGGEOPHYSICALPROPERTIESOFANEAR SURFACECAVETHROUGHINTEGRATEDMICROGRAVITYVERTICALGRADIENTANDELEC TRICAL RESISTIVITYTOMOGRAPHYMEASUREMENTS 14 N JournalofCaveandKarstStudies, April2011

PAGE 5

R EFERENCES Field,M.,1993,Karsthydrologyandchemicalcontamination:Journalof EnvironmentalSystems,v.22,no.1,p.1–26. Loke,M.H.,andBarker,B.R.,1996,Rapidleastsquaresinversionof apparentresistivitypseudosectionsbyaquasi-Newtonmethod: GeophysicalProspecting,v.44,p.131–152. Menardi-Noguera,A.,1988,Structuralevolutionofabrianc onnais covernappe,thecaprauna-armettaunit(LigurianAlps, Italy):JournalofStructuralGeology,v.10,no.6,p.625– 637. Stefanelli,P.,Carmisciano,C.,CaratoriTontini,F.,Cocchi,L.,Bever ini, N.,Fidecaro,F.,andEmbriaco,D.,2008,Microgravityvertical gradientmeasurementinthesiteofVirgointerferometricantenna (PisaPlain,Italy):AnnalsofGeophysics,v.51,no.5–6,p.877–886. Waltham,T.,Bell,F.G.,andCulshaw,M.G.,2005,Sinkholesand subsidence:Karstandcavernousrocksinengineeringandconstruction,Chichester,U.K.,Springer,382p. M.G AMBETTA ,E.A RMADILLO ,C.C ARMISCIANO ,P.S TEFANELLI ,L.C OCCHI AND F.C.T ONTINI JournalofCaveandKarstStudies, April2011 N 15



PAGE 1

INVERSIONFORTHEINPUTHISTORYOFADYE TRACINGEXPERIMENT M ALCOLM S.F IELD 1 AND G UANGQUAN L I 2 Abstract: Theadvection-dispersionmodel(ADM)isagoodtoolforsimulating transportofdyeorsolutesinasolutionconduit.Becausethegeneralprob lemof transportcanbedecomposedintotwoproblems,aboundary-valueproblema ndan initial-valueproblem,thecompletesolutionisasuperpositionoftheso lutionsforthese twoproblems.Inthispaper,thesolutionforthegeneralproblemisexplai ned.Adirect applicationofthesolutionfortheboundary-valueproblemisdye-tracin gexperiments. ThepurposeisinclusionoftheinputhistoryofasolutedyeintotheADM.Th e measuredbreakthroughcurveofadye-tracingexperimentisusedtoinvert fortherelease historyofthedyeattheinputpointthroughtheADM.Itismathematicallys hownthat thebreakthroughcurvecannotbedirectlyusedtoinvertfortheboundaryc onditionata tracerreleasepoint.Therefore,aconductance-fittingmethodisemploy edtoobtainthe inputhistory.Theinvertedhistoryforasimpleexampleisthenshowntobe astep functionwithamplitudeof420 g/Landadurationof10minutes.Simulationsillustrate thatthebreakthroughcurvesatdownstreamspringsprovideameansforund erstanding themigrationofdye.Adiscussionoftheimplicationofthesolutionforan initial-value problem(e.g.,simulatingtransportofpreexistingsolutessuchasdisso lvedcalcium carbonateinsolutionconduits)isalsoincluded. I NTRODUCTION Solute-transportmodeling,aspartofaquantitative tracer-testconductedinkarsticaquifers,iswellestablished tobecriticaltodevelopinganunderstandingofthenature ofsolutemigrationinsolutionconduits(e.g.,Fieldand Pinsky2000;Birketal.2005).Theessentialsolutetransportparametersofvelocity,dispersion,andretardation,tonamejustafew,aregenerallydeterminedfrom groundwatertracingandsolute-transportmodelingprocesses.Unfortunately,thegeneralcomplexityofatracerbreakthroughcurve(BTC)oftenleadstothedevelopment oruseofmodelsofeverincreasingintricaciesinattempts toobtainimprovedmodelfitstomeasureddata.Although theimprovedmodelfitsareusuallypreferredandvaluable, thereremainsomeconcernsastotheoverallapplicability ofmodelswithverylargenumbersofparameters, especiallythosewhoseparametersmaynotbetransferrable tootherexamples. Theadvection-dispersionmodel(ADM),alsoknownas theequilibriummodel,mayberegardedasthesimplestof thevariousmathematicalmodelsusedtodescribesolute transportinsolutionconduits.AlthoughtheADMis theoreticallyreasonable,itsapplicationtomeasuredBTCs isoftendisappointingbecauseofexcessivelyskewedBTC tailsinthemeasureddatathatcannotbematchedusingthe ADM.Theskewnessisoftenattributedtostrong exchangesbetweenmobile-andimmobile-flowregions (Torideetal.1993),solutereactionswithaquifermaterials (SvenssonandDreybrodt1992),ormultipleflowpaths. Currently,theADMandthetwo-regionnon-equilibriummodel(2RNE)aretwoofthemostpopularmodels usedforsimulatingsolutetransportinsolutionconduits (Torideetal.,1993;FieldandPinsky,2000;Birketal., 2005;Go ¨ppertandGoldscheider,2008;Goldscheider, 2008).The2RNEisadvantageousinthatexcessiveBTC skewnesscanbewellsimulated,buthasthedisadvantage ofrequiringadditionalparametersthatleadtomathematicalcomplicationsandpossibleerrorsinparameterization iflocalminima(asopposedtoaglobalminimum)createa conditionofnon-uniquenessduringinverseanalysis(see, forexample,More andWright,1993).Thelikelihoodofa localminimumbeingencounteredincreasesasthenumber ofmodelparametersincreases,especiallyiftheinitialvalue foroneormoreofthevaryingparametersisfarfromthe realvalue. ThecurrentlyfavoredADMisactuallythesolutionfor theboundary-valueprobleminwhichazeroinitialconditionisassumedsuchthatthesolutesourceisaselected boundarycondition.Inotherwords,itistheboundary conditionthatdrivestheBTCs,andthatismostapplicable totracingexperiments.Inthispaper,wefirstdescribea completesolutionthatconsidersthecontributionofthe initialcondition(similartothatofTorideetal.,1995, pp.4–6).Second,weuseaconductance-fittingmethodto obtaintheinputhistoryofadyetracingexperiment. *CorrespondingAuthor 1 U.S.EnvironmentalProtectionAgency,NationalCenterforEnvironmenta l Assessment(8623P),1200Pennsylvania,Ave.,N.W.,Washington,D.C.204 60, USA,field.malcolm@epa.gov 2 DepartmentofGeophysics,YunnanUniversity,2NorthGreenLakeRd., Kunming,Yunnan650091,P.R.China,guangquanli74@gmail.com Disclaimer :Theviewsexpressedinthispaperaresolelythoseoftheauthorsanddo notnecessarilyreflecttheviewsorpoliciesoftheU.S.EnvironmentalPr otection Agency. M.S.FieldandG.Li–Inversionfortheinputhistoryofadyetracingexperi ment. JournalofCaveandKarstStudies, v.73,no.1,p.16–20. DOI:10.4311/jcks2010es0143 16 N JournalofCaveandKarstStudies, April2011

PAGE 2

Finally,wediscussthedifferenceoftheADMandthe 2RNEinmodelingtransportinsolutionconduits. A DVECTION -D ISPERSION M ODELINGIN S OLUTION C ONDUITS P ROBLEM F ORMULATION Theessentialfeaturesofsolutetransportinasolution conduitmaybedescribedusingaone-dimensional advection-dispersionequation(Taylor,1954),whicharises frommassconservationofthesolute,butwaterexchange betweenthesolutionconduitandthesurroundingrock matrixisneglected.Thegoverningequationofthegeneral ADMisgivenas L C L t W L C L z D L 2 C L z 2 1 Soluteconcentrationisdenotedas C .Thedispersion coefficient D canbeparameterizedwiththeconduitradius, a ,andsolutevelocity W asfollows(Lietal.,2008), D b aW 2 wherethedimensionlessdispersioncoefficient b isusedto quantifythestrengthofdispersion.Wenotethat b a is oftenreferredtoasdispersivity. Becausethelengthoftheconduitisinvariablyfinite, thereisaninitialconditionwithinthesolutionconduit (0 z L ,where L isthedownstreampositionofthe spring), C ( z ,0) C I ( z ), 3 andaboundaryconditionatthesinkhole( z 0) C (0, t ) C B ( t ) 4 ThegeneralproblemconsistsofEquation(1)subjectto conditionsshowninEquations(3)and(4).Becausethe mathematicalproblemislinearwithrespecttoconcentration,wecandecomposethisgeneralproblemoftheADM intotwoproblems;aboundary-valueproblem(BVP)andan initial-valueproblem(IVP)(Fig.1).Thesolutionisthe superpositionofthetwosolutionsforthesetwoproblems. S OLUTIONFORTHE B OUNDARY -V ALUE P ROBLEM Theboundary-valueproblemconsistsofEquations(1) and(4),withazeroconcentrationinitialcondition, C I ( z ) 0.TheGreenÂ’sfunctionforthisproblemcanbe obtainedbytheLaplacetransform,andthesolutionis C BVP ( z t ) t 0 C B ( t ) z 4 p D ( t t ) 3 exp z W ( t t ) 2 4 D ( t t ) d t 5 whichisequivalenttothesolutionlistedinTable2ofKreft andZuber(1978)andTable2.2ofTorideetal.(1995). Here, t representsthepasttime,i.e.,thetimeearlierthan t S OLUTIONFORTHE I NITIAL -V ALUE P ROBLEM Thesolutionfortheinitial-valueproblemisaconvolutionoftheinitialconditionwiththeassociatedGreenÂ’s function: C IVP ( z t ) L 0 C I ( f ) 1 4 p Dt exp ( z Wt f ) 2 4 Dt t 0 1 4 p D t exp ( W t f ) 2 4 D t z 4 p D ( t t ) 3 exp z W ( t t ) 2 4 D ( t t ) d t d f 6 NotethattheterminsidethebigbracketistheGreenÂ’s functionfortheinitial-valueproblem.Here, f isavariable denotingthespatialcoordinateofthesolute-concentration distributionat t 0. TheGreenÂ’sfunctionfortheinitial-valueproblemin infinitespacecanbeobtainedfromtheFouriertransform, whichisthefirstterminsidethebigbracket.However,this termcausesapositivevalueattheboundary z 0.The initial-valueproblemrequiresazerovalueatthislocation. Figure1.Theschematicdecompositionofthegeneral problemoftheADM.Thecompletesolutionisasuperpositionofthesolutionsforeachsubproblem. M.S.F IELDAND G.L I JournalofCaveandKarstStudies, April2011 N 17

PAGE 3

Forthisreason,anegativevaluemustbeprescribedatthis boundaryandthecontributionofthisnegativeboundary valuemustbeincluded,whichisthesecondterminsidethe bigbracket.ThisGreenÂ’sfunction(i.e.,theterminsidethe bigbracket)isequivalenttothatlistedinTable2.2of Torideetal.(1995). T HE C OMPLETE S OLUTION ThecompletesolutionofEquation(1),subjecttothe initialandboundaryconditionsimposedbyEquations(3) and(4),respectivelyisasuperpositionofsolutions(5)and (6)thatisgivenas C ( z t ) C BVP ( z t ) C IVP ( z t ) 7 A PPLICATION ThissectionappliesEquation(5)fortheboundaryvalueproblemtoatracerexperimentconductedattheLost RiverCaveSystemlocatedinBowlingGreen,Kentucky. FormedintheSte.GenevieveandSt.LouisLimestonesof MiddletoLateMississippianageanddevelopedontopof theLostRiverChert,theLostRiverCaveSystembegins withtheLostRiverflowingdirectlyintothecaveentrance. LostRiverthentraversesthecavetoeventuallyreemerge 8kmdownstream. Flowalongthelengthofthecaveisrelativelyuniform alongmuchofitslength,butolder,higherpassagescanbe inundatedbyflood-flowconditions(Crawford,1986,p.7). BlocksoftheLostRiverChertroutinelydetachfromcave wallsandlinethecavefloorasthelimestonedissolves away.Thedetachedchertblockscreateminordetention backwaterscontainingimmobile-flowregions.Undercut benchesandrecirculationwithscallopedwallsalsoserveas minorimmobileflowregions. Thetracerinputismodeledusingatypicalstep-like functionwithtime.ByfittingthetheoreticalBTCagainst themeasuredBTC,theparametersofthesolutionconduit andsolutetransport,aswellastheinputhistoryoftracer, areinverted(Table1).Inourinversionprocess,thetotal massofdyeprovidesaconstraintasfollows: T ST M QC 0 8 where M isthetotalmassofdye(476gram), T ST isthe durationofdyeinputatthesinkhole, C 0 istheamplitude ofdyeconcentrationatthetracerreleasepoint,and Q p a 2 W isthespringdischarge(1.78m 3 /s).Forthe ADM,theboundaryconditionattheinputpointis assumedtobeastepfunctionwithtime(constrainedby Equation(8)). The2RNEisalsousedtosimulatethemeasuredBTC. Bothmodels(ADMand2RNE)achieveagoodfitagainst themeasuredBTC,asdepictedinFig.2.Thepeakofthe BTCmodeledfromADMhasanerrorof2hours(Fig.2) becausetheADMcannotadequatelysimulatetheskewness ofthemeasuredBTC.The2RNEvisuallyappearstoabetter fitthantheADMinsimulatingthefallingskewedlimb. Often,thedyereleaseisassumedtobeinstantaneous(a Diracd source).Thisassumptionisreasonableasafirstorderapproximation.However,thisassumptionmathematicallyimpliesaninfinitesolute-concentrationatthe injectionmoment.Toovercomethisphysicalproblem,we includetheinputhistoryintotheADM,andobtaina reasonablehistoryofabout10minutes.Thisresultshould beregardedasthesecond-orderofapproximation. Actually,becausethisdurationismuchshorterthanthe widthofthemeasuredBTCandtheamplitudeofsolute concentrationatthetracerreleasepoint C 0 (420 gL 1 )is muchlargerthanthemaximumamplitudeofthemeasured BTCatthespring(18.5 gL 1 ),thecontinuousinput couldberegardedasaDiracd source.Forthisreasona Diracd sourceassumptionoftenworkswell.Nevertheless, Table1.Theparametersofthesimulationexample. ParameterValueUnits Conduitradius, a 2.44m Conduitlength, L 8.0km Flowvelocity, W 0.095ms 1 Dimensionlessdispersion coeff., b 7.0 Totalmassofdye, M 476gram Dyeconcentrationat sinkhole, C 0 420 m gL 1 Inputdurationatsinkhole, T ST 10min Timesamplinginterval1min Figure2.Thetheoreticalbreakthroughcurvesversusthe measuredcurve.Theinitialconditioniszero,andforthe ADM,theboundaryconditionatthesinkholeisassumedto beastep-likefunctionwithtime. I NVERSIONFORTHEINPUTHISTORYOFADYETRACINGEXPERIMENT 18 N JournalofCaveandKarstStudies, April2011

PAGE 4

thecontinuousinputisphysicallymorereasonable, becauseconcentrationcanneverbeinfinitelylarge. D ISCUSSION Fortheinversionprocedure,weuseconductance-fitting, whichisanindirectinversionmethod.Directinversionmay bethoughtofasusinganegativevelocityandanegative dispersioncoefficienttoinfertheinputhistorydirectlyfrom themeasuredBTC.However,theGreen’sfunctionfora Diracd inputatthespringdoesnotexist.Inaphysical sense,apointsourceatthespringwouldberestoredtoan inputhistorythatneverexists.Mathematically,thiscanbe shownbyrewritingEquation(5)to C BVP ( z t ) C B ( t ) G B ( z t ), 9 where*representsconvolution.PerformingtheLaplace transformwithrespecttotimeonEquation(9)yields C B ( p ) C BVP ( z p ) G B ( z p ) 10 andapplyinganinverseLaplaciantransformandconvolutiononEquation(10)yields C B ( t ) C BVP ( z t ) LT 1 1 G B ( z p ) 11 where LT 1 denotesinverseLaplaciantransform.The inverseLaplaciantransformofthereciprocalof G B ( z p ) doesn’texist.Therefore,wecannotinvertfortheinput history C B ( t )directlyfromthespringBTC C BVP ( z t ),and theconductance-fittingmustbeused. TheADMisoftenappliedtotracerexperimentsin karsticaquifersforwhichtheinitialconditioniszero,butthe boundarycondition(historyofthetracersolute)attheinput pointisnotzero.Besidestracingexperiments,theADMis alsoapplicabletothesimulationoftransportinwhichthe boundaryconditioniszero,buttheinitialconditionisnot zero.Suchasituationmaybeencounteredwhenwater enteringasinkholeorsimilarinputpointistracerfree,but waterrichindissolvedsolutes(e.g.,calciumcarbonate) preexistsinsidetheconduit.Becausetheinitialconditionis oftenunknownandnotmeasurable,thedistributionof solutesinthesolutionconduitmustbeassumed apriori Torideetal.(1993)proposeda2RNEmodel,to describenonequilibriumsolutetransportwithfirst-order decayandzero-orderproduction.TheADMmodelcould beseenasasimplificationoftheirmodel.TheTorideetal. modeldealtwiththecaseinwhichthereisastrongsolute interactionbetweenmobile-andimmobile-flowregionsor aquiferkineticreactions,whileourmodelfocusesonthe caseinwhichaquiferkineticreactionsandsoluteinteractionbetweenmobile-andimmobile-flowregionsare negligible.OursolutiontotheADMmayberegardedas morerobust,becauseitrequiresfewerparameters,whichis similartothefindingsofGo ¨ppertandGoldscheider(2008). Inthissense,ourmodelmaybemoresuitabletothe relativelyrapidtransportconditionsthattypicallyoccurin solutionconduits,becausewhensolutesarerapidlyflushed andentrainedthroughthesolutionconduit,onlyasmall portionofthesolutewilllikelybesequesteredinan immobile-flowregion(Lietal.,2008)andprobablyonly foraveryshorttime.Therefore,thesoluteinteraction betweenthemobile-andimmobile-flowregionsshouldbe verylimited,andourmodelshouldbehaveverywellasa first-orderapproximation.Nevertheless,the2RNEmodel providesamoreaccuratetool,becauseitconsiders temporarysolutedetentioninimmobile-flowregions.For thisreason,the2RNEisbetterforsimulatingBTCsthat exhibitsignificantskewnessandlongtailingphenomena (seeBirketal.,2005),whiletheADMissuitablefor simulatingtheprimarylargesignalsthatarriveearlier. S UMMARYAND C ONCLUSION Transportinasolutionconduitisoftendescribedbythe advection-dispersionmodelthatissubjectedtovarious initialandboundaryconditions.Thegeneralproblemisa superpositionoftwosolutions:oneisfortheproblem consistingofaboundaryconditionandazeroinitial condition(Equation(5)),andtheotherisfortheproblem consistingofaninitialconditionandazeroboundary condition(Equation(6)). WeusedEquation(5)andthebreakthroughcurve measuredfromadyetracingexperimenttoinvertforthe historyofdyeinjectionattheinputpoint.Theinverted parametersarereasonable,whichillustratesthattheBTCs canbeusedtoobtainthehistoryofsoluteinjectionatinput points.FromFigure2,thebreakthroughcurveofthe boundary-valueproblemstillexhibitssomeskewness.Inthis sense,the2RNEisabettermodelthatbetterreplicatesthe typicallystrongskewnessandtailingobservedinmeasured BTCs.ItisourcontentionthattheADMwillhaveimportant andprofoundapplicationsinmodelingtransportofsolutes thatpreexistinsolutionconduits(e.g.,dissolvedcarbonates). A CKNOWLEDGEMENTS Thisresearchwasfinanciallysupportedinpartbythe UniversityFundofYunnanUniversityundercontract KL090020.Theauthorsaredeeplygratefultothe AssociateEditor,Prof.GregoryS.Springerforhis insightfulcommentsandconstructivesuggestions. M.S.F IELDAND G.L I JournalofCaveandKarstStudies, April2011 N 19

PAGE 5

N OTATION a conduitradius(m) C C ( z t )soluteconcentrationinconduit(gL 1 ) C 0 concentrationofdyeatsinkhole(gL 1 ) C B ( t )boundaryconditionatsinkhole(gL 1 ) C BVP ( z t )soluteconcentrationfortheboundary-value problem(gL 1 ) C I ( t )initialconditionwithinconduit(gL 1 ) C IVP ( z t )soluteconcentrationfortheinitial-valueproblem(gL 1 ) D dispersioncoefficient(m 2 s 1 ) G B ( z t )Green’sfunctionfortheboundary-valueproblemwithsourcefixedat t 0(s 1 ) L conduitlength(m) M totalmassofdye(gram) p Laplaciantransformvariable(s 1 ) Q waterdischarge(m 3 s 1 ) t time(s) T ST inputdurationofdyeatsinkhole, W flowvelocityinconduit(ms 1 ) z conduitdownstreamlocationstartingfrom sinkhole(m) b dimensionlessdispersioncoefficient R EFERENCES Birk,S.,Geyer,T.,Liedl,R.,andSauter,M.,2005,Process-based interpretationoftracertestsincarbonateaquifers:GroundWater, v.43,no.3,p.381–388,doi:10.1111/j.1745-6584.2005.0033.x. Crawford,N.C.,1986,Karsthydrologicproblemsassociatedwithurban development:groundwatercontamination,hazardousfumes,sinkholeflooding,andsinkholecollapseintheBowlingGreenarea, Kentucky,GuideBooktoFieldTripB:EnvironmentalProblemsin KarstTerranesandTheirSolutionsConference(BowlingGreen): NationalWaterWellAssociation,Dublin,Ohio,85p. Field,M.S.,andPinsky,P.F.,2000,Atwo-regionnonequilibriummodel forsolutetransportinsolutionconduitsinkarsticaquifers:Journalof ContaminantHydrology,v.44,no.3–4,p.329–351. Goldscheider,N.,2008,Anewquantitativeinterpretationofthelong-ta il andplateau-likebreakthroughcurvesfromtracertestsintheartesian karstaquiferofStuttgart,Germany:HydrogeologyJournal,v.16, p.1311–1317,doi:10.1007/s10040-008-0307-0. Go ¨ppert,N.,andGoldscheider,N.,2008,SoluteandColloidTransportin KarstConduitsunderLow-andHigh-FlowConditions:Ground Water,v.46,no.1,p.61–68,doi:10.1111/j.1745-6584.2007.00373.x. Kreft,A.,andZuber,A.,1978,Onthephysicalmeaningofdispersion equationanditssolutionfordifferentinitialandboundaryconditions: ChemicalEngineeringScience,v.33,no.11,p.1471–1480. Li,G.,Loper,D.E.,andKung,R.,2008,Contaminantsequestrationin karsticaquifers:Experimentsandquantification:WaterResources Research,v.44,W02429,doi:10.1029/2006WR005797. More ,J.J.,andWright,S.J.,1993,Optimizationsoftwareguide: Philadelphia,SocietyforIndustrialandAppliedMathematics,154p. Svensson,U.,andDreybrodt,W.,1992,Dissolutionkineticsofnatural calcitemineralsinCO 2 -watersystemsapproachingcalciteequilibrium:ChemicalGeology,v.100,no.1–2,p.129–145. Taylor,G.I.,1954,Thedispersionofmatterinturbulentflowthrougha pipe:ProceedingsoftheRoyalSocietyLondon,Ser.A,v.223, p.446–468,doi:10.1098/rspa.1954.0130. Toride,N.,Leij,F.J.,andvanGenuchten,M.T.,1993,Acomprehensive setofanalyticalsolutionsfornonequilibriumsolutetransportwith first-orderdecayandzero-orderproduction:WaterResources Research,v.29,no.7,p.2167–2182,doi:10.1029/93WR00496. Toride,N.,Leij,F.J.,andvanGenuchten,M.T.,1995,TheCXTFITcode forestimatingtransportparametersfromthelaboratoryorfieldtracer experiments;version2.0:USSalinityLaboratoryResearchReport 137,121p. I NVERSIONFORTHEINPUTHISTORYOFADYETRACINGEXPERIMENT 20 N JournalofCaveandKarstStudies, April2011



PAGE 1

FIRSTRECORDSOFPOLYCHAETOUSANNELIDSFROM CENOTEAEROLITO(SINKHOLEANDANCHIALINECAVE) INCOZUMELISLAND,MEXICO S ARITA C.F RONTANA -U RIBEAND V IVIANNE S OLI S -W EISS Lab.deEcolog ayBiodiversidaddeInvertebradosMarinos,InstitutodeCienciasdelMa ryLimnolog a,UniversidadNacionalAuto nomadeMe xico,Me xico, solisw@cmarl.unam.mx Abstract: Inthisstudy,polychaetousannelidsarerecordedforthefirsttimeinMex ican cenotesandanchialinecaves.TheseorganismswerecollectedintheCenot eAerolito (CozumelIsland,ontheCaribbeancoastofQuintanaRoo)duringthreesamp lingevents fromFebruary2006toApril2008,amongalgae,rootsofmangroves,andinka rst sediments.Atotalof1518specimensbelongingtofivefamilies(Paraonid ae,Capitellidae, Nereididae,Dorvilleidae,andSyllidae),tengenera,andelevenspecies werecollected.In thecavesystem,twospecimensoftheamphinomid Hermodicecarunculata werefound. Thiscenoteanditsbiotaarenowindangerofdisappearingbecauseofamari na constructionprojectinitswesternshore. I NTRODUCTION Polychaetesconstitutethelargestclassofthephylum Annelida,withabout12,000speciesdescribedtodatein morethan80recognizedfamilies(RouseandPleijel,2006). Theyareubiquitousinmarinehabitatsfromintertidalto abyssaldepths,wheretheyareoftenthemostdiverseor dominantgroup.However,theyarealsofoundinsome freshwaterhabitatsandinkarstsystemslikecenotes(or sinkholes)andevenintheassociatedcaves.Caverecordsare scarce,butthemostimportantcanbefoundinAugener, 1932;Remy,1937;Hartmann-Schro ¨der,1977and1986; SketandIliffe,1980;CulverandSket,2000;Wilsonand Humphreys,2001;andMart nez-Garc aetal.,2009. In2002,Schmitter-Sotoetal.madeanexhaustivelistof organismscollectedinYucata ncenotes,frombacteriaand protozoatoalgaeandsmallinvertebrates,suchas copepods,amphipods,andisopods,buttheseauthorsdid notreportpolychaetes;largeranimals,suchasfishes, amphibians,iguanasorcrocodilesarebetterknownin thesehabitats.Previously,Sua rez-MoralesandRiveraArriaga(1998)mentionedthatnopublishedreferences existaboutfreelivingnematodes,annelids,ormollusksin theYucata ncenotes.Mej a-Ort zetal.(2007b)studied someofthecenotes’macrofaunaofCozumelIslandand mentionedthepresenceofwormsinthecavesystemof CenoteAerolito,buttheydidnotspecifyiftheywere indeedpolychaetes.Inthisstudy,wefocusedsolelyonthe polychaetousannelidsinthiscenoteanditscavesystem. Cenotesaresinkholesfoundalmosteverywhereinthe Yucata nPeninsula,whichcomprisestheMexicanstatesof Yucata n,Campeche,andQuintanaRoo,aswellas northernBelizeandtheGuatemalandepartmentofEl Pete n.Theyareaproductofthegeologicalcharacteristics ofthatregion:inthewholepeninsula,noriversarefound becauseitscalcareouslandscapeischaracterized,among otherfactors,byitshighpermeability.Thus,theabundant rainfall(average1300mmperyear)sinksdowntothe phreaticlevel,whereacomplexwebofsubterraneanrivers, mostlyunknown,areformedandflowtowardstheseaat differentlevels(Schmitter-Sotoetal.,2002).Inplaces wheretheundergroundflowhasproducedcaves,occasionallytheirceilingcollapses,uncoveringthesubterranean waters,andacenoteisborn(Aguilar,2003). CozumelIslandisformedfromreefsedimentswithan averagethicknessof100mormorefromtheOligoceneto theQuaternary;intheselimestonebeds,akarstaquiferis presentand18cenotesareknown(WurlandGiese,2005). S ITE D ESCRIPTION Withdimensionsof52kmby14kmandasurfacearea ofabout650km 2 ,CozumelisthelargestislandinMexico. ItislocatedinthenortheasternregionoftheYucata n Peninsula,inthenorthernMexicanCaribbean,about 18kmfromthemainland(PachecoandVega,2008).Its soilismainlykarstic,composedoflimestone(Wurland Giese,2005),withfourkindsofcenotesrepresented:those withsurfaceconnectionnarrowerthanthediameterofthe waterbody( cenotesca ntaro ),thosewithverticalwalls ( cenotescil ndricos ),thosedegradedtoshallowwater basins( cenotesaguadas ),andthosewithhorizontal entrancestodrysections( grutas ).Theisland’sfreshwater supplycomesmainlyfromthecenotesandassociated subterraneanwatersystems(Mej a-Ort zetal.,2007b).It hasapproximately70,000inhabitantsandistotally devotedtotourism,whichnowadaysisexpandingata rateofmorethan100%peryear(Sol s-Weissetal.,2007). CenoteAerolito,oneofthe18cenotesknownin CozumelIsland,islocatedclosetothewesterncoastof Cozumel,at240mfromshore(20 u 28 00 Nand86 u 58 45 W).Itisapproximately68mlong,25mwide,and8m deep.Itisconnectedtotheseathroughanunderwatercave system(Mej a-Ort zetal.,2007a).Atitsnorthwesternend, S.C.Frontana-Uribe,andV.SolI s-Weiss–FirstrecordsofpolychaetousannelidsfromCenoteAerolito(si nkholeandanchialinecave) inCozumelIsland,Mexico. JournalofCaveandKarstStudies, v.73,no.1,p.1–10.DOI:10.4311/jcks2009lsc0107 JournalofCaveandKarstStudies, April2011 N 1

PAGE 2

arelictofmangrovevegetationispresent( Rhizophora mangle Linnaeus,1753)(Fig.1),andlargeaggregationsof algaearefoundallarounditsedges. M ETHODS SampleswerecollectedbyhandinFebruary2006as partoftheprojectEchinodermsfromCozumelAnchialine Caves.Eachsampleconsistedof200mlofalgaeorkarst sedimenttakenamongmangroverootsinanareacloseto themaincaveentranceoftheCenoteAerolito(station106).InJune2007,aspartoftheprojectBenthicFaunaof theMexicanCaribbeanShores,furthersamplingwas carriedoutatfourstations(1-07,2-07,3-07,and4-07); approximately1000mlofalgaewerecollected,aswellas karstsedimentalongtheeasternedgeofthecenote.Athird samplingoccurredinApril2008atfivestations(1-08,2-08, 3-08,4-08,and5-08),andanadditionalstation(6-08)was sampledusingscubadivinginitscavesystem(Fig.1, Table1).Physicalandchemicalparametersweremeasured duringeachvisitwithaHydrolabDataSonde(HDS3) multiparameterprobe. Figure1.StudyArea.CenoteAerolitoonCozumelIsland. F IRSTRECORDSOFPOLYCHAETOUSANNELIDSFROM C ENOTE A EROLITO ( SINKHOLEANDANCHIALINECAVE ) IN C OZUMEL I SLAND ,M EXICO 2 N JournalofCaveandKarstStudies, April2011

PAGE 3

Allthebiologicalmaterialwasfixedin7%formalinin thefield,laterrinsedwithwater,sievedthrougha0.5mm sieve,preservedin70%alcohol,andidentifiedtospecies level.ThetaxonomicarrangementfollowsRouseand Fauchald(1997).Inthefaunallist,thesyllidslistedassp. 1areincompletespecimens,sotheiridentificationcould onlybecarriedouttothegenuslevel.Thedorvilleids identifiedas Ophryotrocha sp.Acouldnotbereferredtoa knownspeciesandarenowunderstudyaspotentiallynew toscience.AllthespecimensaredepositedintheNational PolychaeteCollectionoftheLaboratoriodeEcolog ay BiodiversidaddeInvertebradosMarinos,Institutode CienciasdelMaryLimnolog a,UNAM(CPICMLUNAM,DFE.IN.061.0598). ThesedimentwasclassifiedfollowingFolk’s(1974) methodforsizeclasses,whilemineralogicalandorganic compositionwasdonewiththemethodofGaudetteetal. (1974). R ESULTS Atotalof1518specimensbelongingtofivefamilies (Paraonidae,Capitellidae,Nereididae,Dorvilleidae,and Syllidae),tengenera,andelevenspeciesarereportedfor CenoteAerolito,whileinitscavesystemtwospecimensof Hermodicecarunculata werefound.Allthepolychaete speciescollectedduringthisstudyrepresentfirstrecords forthelocationinwhichtheywerefound. S PECIES A CCOUNT CapitellidaeGrube,1862 Capitella cf. capitata (Fabricius,1780)speciescomplex Materialexamined.CenoteAerolito,CozumelIsland, QuintanaRoo,February15,2006(25specimens),July5, 2007(66specimens),April19,2008(21specimens)insoft bottomsassociatedwithmangroveroots,algae,andkarst sediments(CPICML-UNAM-PO-17-015). Description.Largestspecimencompletewith36chaetigers,11-mmlongand0.9-mmwideincludingparapodia, smallerspecimencompletewith12chaetigers,5-mmlong and0.5-mmwideincludingparapodia.Prostomiumbroad andtriangular;eyesabsent.Bodyelongate,thoracicregion broadest,partiallyinflated,withnarrowsegments;thoracic chaetigerswithreducedpodiallobesandcapillarychaetae inbothramifromchaetigers1–6;chaetiger7with capillariesormixedfasciclesofcapillaryandhooded hooks;parapodiaofchaetigers8–9withonlyhooksand,in mostofthespecimens,withenlargedopposingnotopodial genitalspines.Abdominalregionnarrowerandsegments longerwithtoriofhoodedhooks.Pygidiumwithout appendixes. Habitat.Inintertidalandsubtidalmudandsand, especiallyinorganicallyenrichedsediments(Blake,2000), deadcoral(Ochoa-Riveraetal.,2000). Distributionremarks.Thespecies C.capitata hasbeen consideredcosmopolitanbutiscomposedbyseveralsibling speciesfoundtobegeneticallydistinct,butmorphologicallysimilar(GrassleandGrassle,1976;Grassle,1980). FollowingtherecentpublicationbyBlake(2009)onthis subject,webelieve,ashedoes,thatthetrue C.capitata will ultimatelybeconfinedtoarcticandsub-arcticenvironmentsandthatallotherrecordsofthespecieswill ultimatelyberecognizedasdifferentspecies,someofthem newtoscience.Hisimportantstudyofthetaxonomic problemssurroundingtheso-called C.capitata complex willresultinfurtherstudiesofthesespecies,sothat,inthis case,thespeciespreviouslyreportedfromCozumelIsland byOchoa-Riveraetal.(2000)as C.capitata willprobably havetobecorrected,andthatrecord,aswellasthepresent onefromCenoteAerolito,willhavetobere-evaluated. Capitellaaciculatus (Hartman,1959) Materialexamined.CenoteAerolito,CozumelIsland, QuintanaRoo,April19,2008(2specimens)inkarst sediments(CPICML-UNAM-PO-17-034). Table1.SamplingstationsatCenoteAerolito. Sample No.Latitude(N)Longitude(W) Depth (m) Salinity (ppt) Water Temp.( u C)pHSubstrate 1-06 a 20 u 27 58.64 86 u 58 41.43 0.518.125.087.25algaeandsediment 1-07 b 20 u 27 56.76 86 u 58 42.21 0.315.127.027.30algaeandsediment 2-07 b 20 u 27 57.74 86 u 58 41.55 0.515.827.087.42algae 3-07 b 20 u 27 58.48 86 u 58 41.27 0.515.827.047.37algae 4-07 b 20 u 27 58.54 86 u 58 41.54 0.515.826.894.44algae 1-08 c 20 u 27 58.70 86 u 58 41.20 1.019.6824.677.35algae 2-08 c 20 u 27 58.11 86 u 58 41.49 0.320.3324.617.34algaeandsediment 3-08 c 20 u 27 57.40 86 u 58 41.70 1.020.4124.887.40algae 4-08 c 20 u 27 56.80 86 u 58 42.30 0.520.3025.077.45algaeandsediment 5-08 c 20 u 27 58.55 86 u 58 41.51 0.319.6824.677.37algae 6-08 c Anchialinecave8.034.7126.037.74sediment a CollectedFebruary15,2006. b CollectedJuly5,2007. c CollectedApril19,2008. S.C.F RONTANA -U RIBE AND V.S OLI S -W EISS JournalofCaveandKarstStudies, April2011 N 3

PAGE 4

Description.Bothspecimensincomplete,males,largest with20chaetigers,11-mmlongand1-mmwideincluding parapodia,smallerspecimenwith17chaetigers,9-mmlong and1.2-mmwideincludingparapodia.Bodyelongate, thoracicregionbarrel-shaped,partiallyinflated,with narrowsegments,abdominalregionnarrower,segments longer.Prostomiumbroadlytriangular;eyesabsent. Thoraxwith9chaetigers,thefirstandsecondareunique inbearingheavyacicularspines,2or3inafascicle,they areknowntooccurinbothnotopodiaandneuropodiain themales,3–7withcapillarychaetae.Onchaetigers8–9 (sometimes6–8),enlargedopposingnotopodialgenital spinesarepresent.Abdominalsegmentswithtoriof hoodedhooksinslightlyelevatedridgeslocatedcloserto theposteriorendofthesegment;singleseriesoffewuncini numbering5to10inarow.Hoodedhookswithlongshafts anddistallyendinrecurvedfangssurmountedbyfive smallerteethintworows. Habitat.Subtidalmud(Tagatzetal.,1982) Distribution.AtlanticOceaninFlorida(Hartman, 1959),CostaRica(Dean,2004).Thisspeciesisnewly recordedforMexicanwaters. Heteromastusfiliformis (Clapare `de,1864) Materialexamined.CenoteAerolito,CozumelIsland, QuintanaRoo,April19,2008(2specimens)inkarst sediment(CPICML-UNAM-PO-17-029). Description.Bothspecimensincomplete,largestwith30 chaetigers,7.5-mmlongand0.25-mmwideincluding parapodia,smallerspecimenwith20chaetigers,4.1-mm longand0.25-mmwideincludingparapodia.Prostomium thinandconical;eyesabsent.Eversibleproboscisinflated, withpapillae,peristomiumasingleachaetousring.Thorax with12segments,firstachaetous.Chaetigers1–5with capillarychaetae,chaetigers6–11withhoodedhooks. Thoracic-abdominaljunctionindistinct,butanteriorabdominalsegmentslargerincross-section,especiallydorsally.Posteriorchaetigerscampanulate,trapezoidin section,widestventrally.Parapodialdorsalbranchiae (locatedposteriorly)andpygidiumcouldnotbeobserved. Habitat.Intertidalmuds,anaerobicandestuarine habitats(Blake,2000).Muddysandsandsands(Dean, 2004). Distribution.MediterraneanSea(Hutchingsand Rainer,1981)AtlanticandPacificOceans,Australia (Blake,2000),SouthernCalifornia(CadienandLovell, 2008). ParaonidaeCerruti,1909 Paradoneislyra (Southern,1914) Materialexamined.CenoteAerolito,CozumelIsland, QuintanaRoo,February15,2006(154specimens)insoft bottomsassociatedwithmangroveroots(CPICMLUNAM-PO-02-004). Description.Largestspecimencompletewith105 chaetigers,18.5-mmlongand0.2-mmwide;smaller specimencompletewith51chaetigers,6.8-mmlongand 0.2-mmwide.Bodyelongate,prostomiumshort,bluntly conical,fusedtoachaetigerousperistomium.Antennaeand subdermaleyesabsent.Prebranchialchaetigersfrom1–3, branchiaefromsegments4to12to14,long,slenderand blunt,firstfewpairsgenerallyshorter.Notochaetaelyrate, slender,fromchaetigers2–3withtwounequaltineswith innerrowofspines,acicularneurochaetaeabsent.Pygidiumrounded,analcirrishort. Habitat.Sandymuds,sands,gravels(Mackie,1991). Distribution.NortheasternAtlantic(Mackie,1991), Indo-Pacific,Mediterranean,Panama(AguadoandLo pez, 2003),SouthernCalifornia(CadienandLovell,2008). AmphinomidaeLamarck,1818 Hermodicecarunculata (Pallas,1766) Materialexamined.AnchialinecaveoftheCenote Aerolito,CozumelIsland,QuintanaRoo,April19,2008 (2specimens)(CPICML-UNAM-PO-49-008). Description.Specimenscompletewith45to48chaetigers,45-to65-mmlongand8-to10-mmwideincluding parapodia.Bodyelongate,intenselyredincolor,the largestspecimenwithintersegmentaltransverseblacklines fromchaetigers5–6.Prostomiumcoveredbyanelaborate carunclecoveringthefirst3chaetigersandformedbytwo seriesof7or8transversefolds;twopairsofeyesand3 antennae.Doubledendriticbranchiaepresentalongwhole body.Parapodiawithabundantnotochaetaeandneurochaetae.Notochaetaeverylongcapillariesandharpoon chaetae(stoutpointedchaetaewithrecurvedbarbsnearthe tip)ofsmallersize;threetypesofacicularneurochaetae: subdistallyflat,distallydenticulate(3to5teeth),subdistallywithatoothanddenticulateuntiltheapex,withupto 15smallteeth.Notopodialcirribiarticulatedandlarger thanneuropodialcirri.Lastsegmentencirclingtheanus. Habitat.Associatedtosessileorganismsinrocky substrate(Salazar-Vallejo,1996–1997);indeadcoral (Ochoa-Riveraetal.,2000). Distribution.TransatlanticandMediterraneanintropicalandsubtropicalwaters(Salazar-Vallejo,1996–1997);in theMexicanCaribbeanpreviouslyreportedfromCozumel IslandbyOchoa-Riveraetal.(2000). DorvilleidaeChamberlin,1919 Ophryotrocha sp.A. Materialexamined.CenoteAerolito,CozumelIsland, QuintanaRoo,February15,2006(10specimens),July5, 2007(758specimens),April19,2008(227specimens)in softbottomsassociatedwithmangroveroots,algae,and karstsediments(CPICML-UNAM-PO-53-012). Description.Mostofthespecimenscompletewith12to 15chaetigers,0.5-to0.8-mmlongand0.4-to0.5-mmwide includingparapodia.Bodyshort,cylindrical,compressed dorsoventrally,taperingtowardspygidium.Prostomium bluntlytriangularwithtwolateralantennaeandciliary aggregationsbothinthefrontalregionandatthetop. Peristomiumwithtwoapodousrings.Parapodiaunirramous,withventralretractilelobeandsimplechaetae. F IRSTRECORDSOFPOLYCHAETOUSANNELIDSFROM C ENOTE A EROLITO ( SINKHOLEANDANCHIALINECAVE ) IN C OZUMEL I SLAND ,M EXICO 4 N JournalofCaveandKarstStudies, April2011

PAGE 5

Supra-acicularfasciclewith2to3simplechaetae,infraacicularfasciclewith4to5heterogomphfalcigersandone inferior-mostsimpleChaeta.Bothtypesofchaetaefinely serratedsubdistally,taperingtosmalldistaltooth. Pygidiumwithonepairofminuteovatepygidialcirri. Jawapparatuswithelongatedistallybifidmandibleswith serratedcuttingedge.MaxillaryapparatusofPorKtype; PtypepresentinsmallestspecimensandKtypeinlarger specimens.K-forcepsdistallyfalcate.D1–D7denticles attachedbyligamenttoforceps.Withbluemethyldyea reducedciliarydistributiononeachsegmentandpygidium isevident Remarks.ThemaxillaryapparatusknownasP-type mayhavefalcateorbidentateforcepsandthenumberof freedenticlesrangingfromfivetosevenpairs,whilethe maxillaryapparatusknownasK-typemayhavetwo unidentate,twobidentate,oroneunidentateandone bidentateprongandthenumberoffreedenticlesranges fromsixtoeightpairs;theremaybeone,twoorthree differentkindsofdenticlesinonejawapparatus.The remarkablefeaturehereisthatintheseorganisms,both typesarefound,dependingonthesizeofthespecimens, raisingthequestionoftheimportanceofthischaracterto differentiatebetweenspeciesinthisgenus. Otherspeciesof Ophryotrocha areopportunistic(Desbruye `resetal.,2006)andthereforefoundinmanydifferent habitats,whichcouldexplainwhythisspeciesisso abundanthere.Itsstatusasanewspeciesispresently understudy. NereididaeJohnston,1865 Stenoninereismartini Wesenberg-Lund,1959 Materialexamined.CenoteAerolito,CozumelIsland, QuintanaRoo,February15,2006(14specimens),July5, 2007(203specimens),April19,2008(9specimens)insoft bottomsassociatedwithmangroveroots. Comparativematerialexamined.SanJulia n,Lagunade Te rminos,Campeche,Me xico1March1984(18specimens)(CPICML-UNAM-PO-39-039). Description.Largestspecimencompletewith33chaetigers,10.5-mmlongand1.5-mmwideincludingparapodia,smallerspecimencompletewith21chaetigers,2.5-mm longand0.6-mmwide.Prostomiumpentagonal,slightly notchedfrontally.Twopairsofeyes.Frontalantennae cirriformandnotlongerthandistalpalps.Palpsmarginal, globular,biarticulatewithelongateconicalpalpostyles. Peristomiumslenderwithfourpairsoftentacularcirri, anteriordorsalpairreachingposteriortochaetiger6. Pharynxwithpairedjawsarmedwith10to12teeth,no paragnathsorpapillae.Firsttwoparapodiasubbiramous withnotopodiareducedtosmallnotoacicula;following parapodiabiramous;anterioroneswithlongdorsalcirri consistingofelongatebasalcirrophoreandshortpyriform cirrostyle,becominglongertowardsendofthebody. Notopodiatrilobate,superiorlobeshortanddigitiform, decreasinginsizeinposteriorchaetigers;inferiorlobes subulate,withsmallpresetallobeatbaseofupperlobe. Neuropodiawithbluntlyconicalacicularlobeinanterior region,becomingmoreelongateinmiddlechaetigersand shorter,morepointedinposteriorregion.Supra-acicular notochaetaesesquigomphspinigers,infra-acicularnotochaetaehomogomphspinigers;bothwithslenderappendix, serratedontheinneredge.Supracicularneurochaetae heterogomphandsesquigomphspinigers;oneheterogomphspinigerwithastronglyserratedshortbladeonat leastthreequartersofitslengthonitsinnermargin; heterogomphfalcigerswithspinulosedistallyhookedlong blade.Pygidiumwithapairoflateralflattened,widelobes, andapairoflonganalcirri. Habitat.Softbottomsassociatedwithmangroveroots (deLeo n-Gonza lezandSol s-Weiss,1997). Distribution.LagunadeTe rminos,Campeche,Mexico (deLeo n-Gonza lezandSol s-Weiss,1997),greaterCaribbeanregion,SanMart nIsland,SarasotaFlorida,western GulfofMexico(Texas),Cuba,NorthCarolina(Wesenberg-Lund,1958;Pettibone,1971;Hartman-Schro ¨eder, 1977). SyllidaeGrube,1850 Erinaceusylliscentroamericana (Hartmann-Schro ¨der, 1959) Materialexamined.CenoteAerolito,CozumelIsland, QuintanaRoo,February15,2006(1specimen),July5, 2007(4specimens),April19,2008(1specimen)(CPICMLUNAM-POH-076). Description.Onespecimencompletewith21chaetigers, 1.5-mmlongand0.2-mmwideincludingparapodia.Body small,coveredwithsmall,scatteredpapillae.Prostomium ovalwith4smalleyesinrectangulararrangementand2 anterioreyespots;antennaepyriform,medianandlateral antennaesimilarinsize.Palpsshort,fusedalongtheirlength. Peristomiumlong,tentacularcirrisimilartoantennae. Dorsalcirrisimilartoantennae,withbulbousbasesand shorttips,absentonchaetiger2.Ventralcirridigitiform. Compoundchaetaeheterogomph,similarthroughoutbody; bladesslender,elongate,unidentate,distallyslightlyhooked, providedwithproportionallylongmarginalspinesonbases oflongerblades;parapodiaeachwithonelongbladed compoundchaetaand6falcigers.Fromchaetiger1,simple unidentatechaetaewithshortmarginalspinesdorsally. Simplechaetaeventrally.Aciculaeacuminate,oneper parapodiumthroughoutthebody.Pharynxextendingfrom chaetigers1to3oreven4;pharyngealtoothsmalland enlargedonanteriormargin.Proventriclebarrel-shaped, extendingfromchaetigers3–6;withabout13to16rowsof transversemusclebands.Pygidiumsmall,withtwoanalcirri. Habitat.Insand,algaeandmangroves(SanMart n, 2005). Distribution.Circumtropical:ElSalvador,Gala pagos Islands,CaribbeanSea,Hawaii,Samoa,Angola,Mozambique,Tanzania,Australia(SanMart n,2005).Newrecord forMexicanCaribbean. S.C.F RONTANA -U RIBE AND V.S OLI S -W EISS JournalofCaveandKarstStudies, April2011 N 5

PAGE 6

Salvatoria sp.1 Materialexamined.CenoteAerolito,CozumelIsland, QuintanaRoo,July5,2007(4specimens),insoftbottoms associatedwithmangroveroots,algae,andkarstsediments (CPICML-UNAM-PO-37-077). Description.Allspecimensincomplete,largestwith21 chaetigers,1.8-mmlongand0.25-mmwideincluding parapodia.Bodysmall.Prostomiumwith3antennae,4 eyesand,usually,without2eyespots.Palpswelldeveloped, joinedalongtheirlengthbyadorsalmembrane.Twopairs oftentacularcirri.Antennaetentacular,dorsalcirrilong andslender,slightlybulbousattheirbaseandwithan elongate,acutetip;dorsalcirripresentonallsegments. Ventralcirridigitiform,shorterthanparapodiallobes. Compoundchaetaeheterogomphwithsmallindistinct subdistaltooth;dorsalsimplechaetaebidentate,withshort subdistalmarginalspines,ventralsimplechaetaenot observed.Aciculadistallyrounded,oneperparapodium throughoutthebody.Pharynxextendingthroughchaetigers1–3,surroundedbyacrownof12softpharyngeal papillaeandasmalltoothonanteriormargin;proventricle ofthesamelengththroughchaetigers4–6,withabout19 rowsoftransversemusclebands.Pygidiumunknown. Exogone(Paraexogone) sp.1. Materialexamined.CenoteAerolito,CozumelIsland, QuintanaRoo,July5,2007(1specimen),insoftbottoms associatedwithmangroverootsandalgae(CPICMLUNAM-PO-37-077). Description.Specimenincompletewith15chaetigers, 0.5-mmlongand0.2-mmwide.Bodysmallandslender. Prostomiumwithoutantennae,foureyeswith2eyespots. Palpsslender,welldeveloped,completelyfusedtoeach other.Tentacularcirrifragmented.Dorsalcirrismall, ovoid,presentonallsegments.Compoundchaetae falcigers,incompleteandnospinigersvisible,dorsalbifid simplechaetaepresentfromfirstchaetiger.Aciculadistally truncate.Pharynxsinuous,extendingthroughchaetigers1 and2,pharyngealtoothsmallonanteriormargin. Proventriclethroughchaetigers2–4,withabout22rows oftransversemusclebands.Pygidiumunknown. Syllisprolifera Krohn,1852 Materialexamined.CenoteAerolito,CozumelIsland, QuintanaRoo,July5,2007(15specimens),inalgaeand karstsediments(CPICML-UNAM-PO-37-007). Description.Largestspecimencompletewith48chaetigers,9-mmlongand0.25-mmwide.Bodythick. Prostomiumovalwith4smalleyesintrapezoidal arrangement;medianantennaelongerthanprostomium andpalpstogether,withabout24articles;lateralantennae withabout29or30articles.Palpstriangular,robust,larger thanprostomium.Dorsaltentacularcirrilong,withabout 37to45articles;ventraltentacularcirriwithabout33to38 articles.Dorsalcirrilongandslender,withabout45to52 articles.Ventralcirridigitiform.Compoundchaetae falcigers,stronglybidentatewithteethofsimilarsize. Aciculaeroundedwithhollowtips;theirnumberchanges dependingonbodyregion,fourinanteriorandmedian regionandonlyoneinposteriorregion;dorsalbifidsimple chaetaepresentfromchaetiger18andventralsimple chaetaeonlypresentinposteriorchaetigers.Pharynx extendingthroughchaetigers1–7;pharyngealtoothsmall onmiddledorsalposition.Proventriclethroughchaetigers 7–10,withabout25rowsoftransversemusclebands. Pygidiumsmall,withtwoanalcirriwith28to30articles. Habitat.On Rhizophoramangle roots,corallinerocks (SanMart n,1992).Abundantinalgae,sandyandhard substrate(SanMart n,2003). Distribution.Acosmopolitanspeciesintropicaland temperateseas(SanMart n,2003).Theclosestrecordto thestudyareaisCuba(SanMart n,1992). Syllismaryae SanMart n,1992 Materialexamined.CenoteAerolito,CozumelIsland, QuintanaRoo,April19,2008(1specimen)(CPICMLUNAM-POH-37-045). Description.Specimencompletewith61chaetigers,8.5mmlongand0.5-mmwide.Bodythick.Prostomiumoval with4smalleyesintrapezoidalarrangement;median antennaelongerthanprostomiumandpalpstogether,with about25articles;lateralantennae,withabout12to14 articles.Palpstriangular,similarinlengthtoprostomium. Dorsaltentacularcirrilong,withabout13to15articles; ventraltentacularcirriwith10to12articles.Dorsalcirri slender,alternatinglong(23to26articles)andshort(10to 16articles).Ventralcirridigitiform.Dorsalglandson segments14–16.Compoundchaetaeincluding1or2 bidentatepseudospinigerswithteethofsimilarsizeand about6bidentatefalcigerswithproximaltoothshorter thandistalone,andwithshortspinesoncuttingmargin withdorsoventralgradation.Aciculaedistallytruncate, formingarightangle,theirnumberchangingalongthe body,twoinanteriorandmedianregionandonlyonein posteriorregion.Pharynxextendingthroughchaetigers1– 7;pharyngealtoothsmallonanteriormargin.Proventricle throughchaetigers7–10,withabout29rowsoftransverse musclebands.Pygidiumsmall,withtwoanalcirriwith16 to18articles. Habitat.Shells(SanMart n,1992),deadcoral(Granados-Barbaetal.,2003). Distribution.NorthCarolina,Cuba(SanMart n, 1992),GulfofMexico(Granados-Barbaetal.,2003). S EASONAL S PECIES O CCURRENCE Inthisstudy,thefamilyDorvilleidaeisdominant,with 995specimensor65.5%ofthepolychaetescollected, althoughitisrepresentedbyonlyonespecies, Ophryotrocha sp.A.Somedifferencesinseasonaldistributionwere noted.InJuly2007andApril2008, Paradoneislyra was absent,whileinFebruary2006itwasbyfarthemost abundantspecies(75%ofthetotal). Syllisprolifera was presentonlyinJuly2007, Capitellaaciculatus and F IRSTRECORDSOFPOLYCHAETOUSANNELIDSFROM C ENOTE A EROLITO ( SINKHOLEANDANCHIALINECAVE ) IN C OZUMEL I SLAND ,M EXICO 6 N JournalofCaveandKarstStudies, April2011

PAGE 7

Heteromastusfiliformis wherepresentonlyinApril2008, andthespecimensof Salvatoria sp.1and Exogone sp.1were presentonlyinJuly2007.Inallsamplesandatalmostall stations,exceptinthecavesystem,(February2006,July 2007,andApril2008)the Capitellacapitata species complexand Ophryotrocha sp.A,werepresent(Table2). Thedifferencesobservedinspeciescompositionand distributionarerecordedhereaspartofthisfirststudy onthepolychaetefaunaofMexicancenotes,butcannotbe fullyinterpretedatthispointuntilmoresamplingiscarried out.Nevertheless,thisstudydocumentsboththefactthat polychaetesarepresentanddiversifiedinthishabitat,and thatseasonalchangesorspeciesreplacementsmighttake place.Itisinterestingtonotethat,inthesecondsampling period,epitokeswerecollectedfromthefamiliesNereididaeandSyllidae,whilenonewerepresentinthefirstand thirdsamplingperiods. Bottom-watersalinityandtemperatureweredifferentin eachsamplingperiod.InFebruary2006,thesalinitywas 18.1pptwithatemperatureof25 u C,inJuly2007the salinitywasbetween15and15.8pptandtemperature26.89 to27.08 u CwhereasinApril2008thesalinitywasbetween 19.68and20.33pptandtemperature24.61to24.88 u C.In thecavesystem,thesalinitywas34.71pptandtemperature 26.03 u C.Thesedimentinthecaveconsistsofsandof biogenicorigin(veryfinecoraldebris). D ISCUSSION Sincethisisthefirststudyaboutpolychaetesincenotes inMexico,allrecordsarenewforCenoteAerolitoitself anditscavesystem,butsofaronlythedominantspecies Ophryotrocha sp.Aispotentiallynewtoscience. Thepolychaetespeciesalreadyrecordedforthe Cozumelareainclude Hermodicecarunculata andthesocalledcomplexofspeciesunderthename Capitella capitata ,bothrecordedfromdeadcoral(Ochoa-Riveraet al.,2000).Inthiscase,thepresenceofthisspeciescomplex couldbeattributedtotheabundanceoforganicmatter (mangroveroots,algae)inwhichthespeciescomplexis knowntothrive. Attheregionalscale, Stenoninereismartini isknown fromFlorida,downtotheGulfofMexico(Te rminos lagoon)andthegreaterCaribbean(deLeo n-Gonza lezand Sol s-Weiss,1997),andthecapitellids Capitellaaciculatus and Heteromastusfiliformis havealsobeenrecorded previouslyforthegreaterCaribbean.Althoughsome speciesofExogoninaeoftheCaribbeanhavebeenreported byRu z-Ram rezandSalazar-Vallejo(2000),thespecies Erinaceusylliscentroamericana isanewrecordforthe MexicanCaribbean.Thisspecieshadbeenpreviously recordedfromtheWestIndies(SanMart n,2005). Syllis prolifera isconsideredcosmopolitanbut,like Syllismaryae theclosestrecordstothestudyareaarefromCubaandthe GulfofMexico(SanMart n1992;2003). C ONCLUSIONS ThisfirststudyonpolychaetousannelidsinMexican cenotesrevealedarelativelydiversified(10generaand11 species)andinterestingpolychaetefaunainthebest-known andmostpopularcenoteonCozumelIsland.Webelieve thattheunderwaterconnectiontothesea,inadditionto thenearbymangrovehabitat,createpropitiousconditions forPolychaeta,andmanyotherspeciescanbeexpectedin theseuniquehabitats. Eventhoughthereareimportantdifferencesbetween thecenotehabitatandatrulymarineone,nomorphologicaldifferencescouldbeobservedinanyofthespecies recorded,withthepossibleexceptionofthe Ophriotrocha sp.1. Althoughthebest-representedmacrofaunalinvertebrateanimalgroupincavesandalsocenotesismacrofaunalcrustaceans(Schmitter-Sotoetal.,2002;Mej a-Ort zet al.,2007b),morerecentstudiesarerevealingnewtaxa. Mej a-Ort zetal.(2007a)andSol s-Mar nandLaguardaFigueras(2008),forexample,foundspeciesofallthe echinodermclassesexcepttheCrinoideaforthesystemof cavesconnectedtoCenoteAerolito. Inaddition,thisstudyconstitutesabaselinestudythat willbeespeciallyimportantbecausethecenoteisonthe vergeofbeingseverelytransformedbytheconstructionof amarinaatitswesternend,whichwillincludethe destructionofthecaveincreatingadirectconnection betweenthemarinaareaofthecenotetothesea.Thiswill certainlyaffectthelocalfaunaasweknowitnow,notonly becauseofthechangeinsalinity,whichwillbecomelike thesurroundingsea,butbecauseoftheconstructionitself andsubsequentmaritimetraffic.Allthisimpactwill changetheenvironmentalconditionsandnotonlythe polychaetecompositionbutalsotheotherfaunagroups alreadyrecorded. Thereisconcerninthescientificcommunity,because legislationconcerningthecenotesinMexicoisnotyetwell defined.Theyarenotdifferentiatedfromotherfreshwater bodies,sotheycannotbeproperlyprotected.TheMexican governmentagencyCONAGUA(Comisio nNacionaldel Agua)hasonlybegunin2009toworkonalegal frameworktopromotethesustainableuseandconservationoftheseuniquewaterbodies. A CKNOWLEDGEMENTS WearegratefultoAlfredoLaguardaandFranciscoSol s, fromtheInstitutodeCienciasdelMaryLimnolog a, UNAM,aswellastodiversGerma nYa nez,AdrianMedina, andEmmanuelTeysier,fortheirhelpinthefield,andto DiegoTrujilloandEfrainCha vezfortheirhelpincollecting thecavesamples.VictorOchoa,Mar aElenaGarc a,and PabloHerna ndezarealsothankedfortheirhelpinsome identifications.AntonioMa rquez,LaboratoriodeSedimenS.C.F RONTANA -U RIBE AND V.S OLI S -W EISS JournalofCaveandKarstStudies, April2011 N 7

PAGE 8

Table2.AbundanceofpolychaetescollectedinCenoteAerolitoanditsanc hialinecavein2006–2008. Taxa Stations Total Specimens 1-061-072-073-074-071-082-083-084-085-086-08 FamilyCapitellidaeGrube,1862 Capitellacapitata (Fabricius,1780) sensulato 259164132268112 Capitellaaciculatus (Hartman,1959) 22 Heteromastusfiliformis (Clapare `de,1864)22 FamilyParaonidaeCerruti,1909 Paradoneislyra (Southern,1914)154 154 FamilyAmphinomidaeLamarck,1818 Hermodicecarunculata (Pallas,1766) 22 FamilyDorvilleidaeChamberlin,1919 Ophryotroca sp.A1014434239341848104121995 FamilyNereididaeJohnston,1845 Stenoninereismartini Wesenberg-Lund,1959149111218226 FamilySyllidaeGrube,1850 SubfamilyExogoninaeLangerhans,1879 Erinaceusylliscentroamericana (HartmannSchro ¨der,1959)121116 Salvatoria sp.1314 Exogone sp.1 11 SubfamilySyllinaeGrube,1850 Syllisprolifera Krohn,1852113115 Syllismaryae SanMart n,1992 11 F IRSTRECORDSOFPOLYCHAETOUSANNELIDSFROM C ENOTE A EROLITO ( SINKHOLEANDANCHIALINECAVE ) IN C OZUMEL I SLAND ,M EXICO 8 N JournalofCaveandKarstStudies, April2011

PAGE 9

tolog afromUnidaddeOceanograf a,ANIDE,Mexico,is deeplyacknowledgedforhisanalysisofthesediment. R EFERENCES Aguado,M.T.,andLo pez,E.,2003,Paraonidae(Annelida:Polychaeta) delParqueNacionaldeCoiba(Pac fico,Panama ),conladescripcio n deunanuevaespeciede Aricidea Webster,1879:RevistaChilenade HistoriaNatural,v.76,p.363–370. Aguilar,V.,2003,Aguascontinentalesydiversidadbiolo gicadeMe xico: Unrecuentoactual:Biodiversitas,v.8,no.48,p.1–14. Augener,H.,1932,DiePolychaetenundHirudineendesTimavo-gebietes inderAdriatischenKarstregion:AbdruckausZoologischeJahrbucher,AbteilungfurSystematik,OkologieundGeographiederTier, v.63,no.5–6,p.657–680. Blake,J.A.,2000,FamilyCapitellidaeGrube,1862, in Blake,J.A.,Hilbig, B.,andScott,P.H.,eds.,TaxonomicAtlasoftheBenthicFaunaofthe SantaMariaBasinandWesternSantaBarbaraChannel,Volume7. TheAnnelida,Part4.California,SantaBarbaraMuseumofNatural History,p.47–96. Blake,J.A.,2009,Redescriptionof Capitellacapitata (Fabricius)from WestGreenlandanddesignationofaneotype(Polychaeta,Capitellidae):Zoosymposia,v.2,p.55–80. Cadien,D.B.,andLovell,L.L.,eds.,2008,ATaxonomicListingofSoft BottomMacro-andMegainvertebratesfromInfaunal&Epibenthic MonitoringProgramsintheSouthernCaliforniaBight,FifthEdition: TheSouthernCaliforniaAssociationofMarineInvertebrateTaxonomists,210p. Chamberlin,R.V.,1919,AlbatrossPolychaeta(Rep.Sci.Res.Exp. Albatross):MemoirsoftheMuseumofComparativeZoology, HarvardUniversity,v.48,514p. Cerruti,A.,1909,Contributoall’anatomia,biologiaesistematicadell e Paraonidae(Levinsenidae)conparticolareriguardoallespeciedel golfodiNapoli:MitteilungenausderZoologischenStationzuNeapel, Berlin,v.19,p.459–512. Clapare `de,E.,1864,GlanureszootomiquesparmilesAnne lidesdePortVendres(Pyre ne esOrientales):Me moiresdelaSocie te dePhysiqueet d’HistoireNaturelledeGene `ve,v.17,no.2,p.463–600. Culver,D.C.,andSket,B.,2000,Hotspotsofsubterraneanbiodiversityi n cavesandwells:JournalofCaveandKarstStudies,v.62,no.1, p.11–17. Dean,H.K.,2004,MarinebiodiversityofCostaRica:ClassPolychaeta (Annelida):RevistadeBiolog aTropical,v.52,supl.2,p.131–181. deLeo n-Gonza lez,J.A.,andSol s-Weiss,V.,1997,Anewspeciesof Stenoninereis (Polychaeta:Nereididae)fromtheGulfofMexico: ProceedingsoftheBiologicalSocietyofWashington,v.110,no.2, p.198–202. Desbruye `res,D.,Segonzac,M.,andBright,M.,2006,Handbookofdeepseahydrotermalventfauna.2 nd edition,Denisia18,Linz-Dornach, Austria,Biologiezentrum,544p. Fabricius,O.,1780.FaunaGroenlandica.Systematicesistens,Animalia Groenlandiaeoccidentalishactenusindagata,quadnomenspecificum, triviale,vernaculumquesynonymaauctorumplurium,descriptionem, locum,victum,generationem,mores,usum,capturamquesinguli; proutdetegendioccasiofuit,maximaquepartisecundumpropias observationes,HafniaeetLipsiae,J.G.Rothe,xvi,452p. Folk,R.L.,1974,PetrologyofSedimentaryRocks:Austin,Texas, Hemphill,170p. Gaudette,H.E.,Flight,W.R.,Toner,L.,andFolger,D.W.,1974,An inexpensivetitrationmethodfordeterminationoforganiccarbonin recentsediments:JournalofSedimentologyandPetrology,v.44, no.1,p.249–253. Grassle,J.P.,andGrassle,J.F.,1976,Siblingspeciesinthemarine pollutionindicator Capitella (Polychaeta):Science,v.192,p.567–569. Grassle,J.P.,1980,Polychaetesiblingspecies, in Brinkhurst,R.O.,and Cook,D.G.,eds.,AquaticOligochaeteBiology:NewYork,Plenum, p.25–32. Granados-Barba,A.,Sol s-Weiss,V.,Tovar-Herna ndez,M.A.,and Ochoa-Rivera,V.,2003,DistributionanddiversityoftheSyllidae (Annelida:Polychaeta)fromtheMexicanGulfofMexicoand Caribbean, in Sigvaldado ttir,E.,Mackie,A.S.Y.,Helgason,G.V., Reish,D.J.,Svavarsson,J.,Steingr msson,S.A.,andGuomundsson, G.,eds.,AdvancesinPolychaeteResearch,Hydrobiologia,v.496, no.1–3,p.337–345. Grube,A.E.,1850,DieFamilienderAnneliden:Archivfu ¨rNaturgeschichte,v.16,p.249–364. Grube,A.E.,1862,NocheinWortueberdieCapitellenundihreStellung imSystemederAnneliden:Bonn,CarlGeorgi,p.65–378. Hartman,O.,1959,CapitellidaeandNereidae(MarineAnnelids)fromthe GulfsideofFlorida,withareviewoffreshwaterNereidae:Bulletinof MarineScienceoftheGulfandCaribbean,v.9,p.153–168. Hartmann-Schro ¨der,G.,1977,DiePolychaetenderKubanisch-RumanischenBiospeologischenExpeditionnachKuba1973, in Re sultats desExpe ditionsBiospe ologiquesCubano-Roumainesa `Cuba, Volume2:Bucharest,AcademieiRebublichSocialisteRoma ˆnia2, p.51–63. Hartmann-Schro ¨der,G.,1986,Polychaeta(Incl.Archiannelida), in Botosaneanu,L.,ed.,StygofaunaMundi:AFaunistic,Distributional, andEcologicalSynthesisoftheWorldFaunainhabitingSubterraneanWaters(includingtheMarineInterstitial:Leiden,E.J.Brill, p.210–233. Hutchings,P.A.,andRainer,S.F.,1981,Designationofaneotypeof CapitellafiliformisClaparede,1864,typespeciesofthegenus Heteromastus (Polychaeta:Capitellidae):RecordsoftheAustralian Museum,v.35,no.4,p.373–381. Johnston,G.,1865,ACatalogueoftheBritishNon-parasiticalWormsin theCollectionoftheBritishMuseum:London,BritishMuseum (NaturalHistory),366p. Krohn,A.,1852,UeberdieErscheinungenbeiderFortpflanzungvon Syllisprolifera und Autolytusprolifer :Archivfu ¨rNaturgeschichte, Berlin,v.18,no.1,p.66–76. Lamarck,J.B.P.,1818,HistoireNaturelledesAnimauxsansVerte `bres, Volume5:Paris,DetervilleandVerdiere,612p. Linnaeus,C.,1753,SpeciesPlantarum,ExhibentesPlantasRiteCognitas adGeneralRelatas,cumDifferentiisSpecificis,NominibusTrivialibus,SynonymisSelectis,LocisNatalibus,SecundumSystemaSexuale Digestas:Holmi,Sweden,ImpensisL.Salvii,twovolumes1200p. Mackie,A.S.Y.,1991, Paradoneiseliasoni sp.nov.(Polychaeta:Paraonidae)fromnorthernEuropeanwaters,witharedescriptionof Paradoneislyra (Southern,1914), in Petersen,M.E.,andKirkegaard, J.B.,eds.,Systematics,BiologyandMorphologyofWorldPolchaeta: Copenhagen,OpheliaSupplement5,p.147–155. Mart nez-Garc a,A.,Palmero,A.M.,Brito,M.C.,Nu n ez,J.,and Worsaae,K.,2009,AnchialinefaunaoftheCoronalavatube (Lanzarote,CanaryIslands):diversity,endemismanddistribution: MarineBiodiversity,v.39,p.169–182. Mej a-Ort z,L.M.,Ya nez,G.,andLo pez-Mej a,M.,2007a,Echinoderms inananchialinecaveinMexico:MarineEcology,v.28,no.1,p.31–34. Mej a-Ort z,L.M.,Ya nez,G.,Lo pez-Mej a,M.,andZarza-Gonza lez,E., 2007b,Cenotes(anchialinecaves)onCozumelIsland,QuintanaRoo, Me xico.JournalofCaveandKarstStudies,v.69,no.2,p.250–255. Ochoa-Rivera,V.,Granados-Barba,A.,andSol s-Weiss,V.,2000,The PolychaetecryptofaunafromCozumelIsland,MexicanCaribbean: BulletinofMarineScience,v.67,no.1,p.137–146. Pacheco,M.A.,andVega,F.J.,2008,Resen aGeolo gica, in Mej a-Ort z, L.M.,ed.,BiodiversidadAcua ticadelaIsladeCozumel:Universidad deQuintanaRoo,p.33–42. Pallas,P.S.,1766,MiscellaneaZoologica:Quibusnovaeimprimisatque obsuraeanimaliumspeciesdescribunteretobservantionibusicomibusqueillustrantur:TheHague.Comitum:Hague,v.12,224p. Pettibone,M.H.,1971,RevisionofSomeSpeciesReferredto Leptonereis Nicon and Laeonereis (Polychaeta:Nereididae):Washington,D.C., SmithsonianInstitution,ContributionstoZoologyNo.104,53p. Remy,P.,1937,Sur Marifugiacavatica AbsalometHrabe,serpulidedes eauxdoucessousterraneesduKarstadriatique:BulletinduMuseum d’HistoireNaturelle,Paris,v.9,p.66–72. Rouse,G.W.,andFauchald,K.,1997,Cladisticsandpolychaetes: ZoologicaScripta,v.26,p.139–204. Rouse,G.W.,andPleijel,F.,2006,Annelidphylogenyandsystematics, in Rouse,G.W.,andPleijel,F.,eds.,ReproductiveBiologyand PhylogenyofAnnelida:Enfield,NewHampshire,SciencePublishers, p.3–21. Ru z-Ram rez,J.D.,andSalazar-Vallejo,S.I.,2000,Exogoninae(Polychaeta:Syllidae)delCaribemexicanoconunaclaveparalasespecies delGranCaribe:RevistadeBiolog aTropical,v.49,no.1, p.117–140. S.C.F RONTANA -U RIBE AND V.S OLI S -W EISS JournalofCaveandKarstStudies, April2011 N 9

PAGE 10

Salazar-Vallejo,S.I.,1996–1997Anfino midosyeufros nidos(Polychaeta) delCaribeMexicanoconclavesparalasespeciesreconocidasdelGran Caribe:RevistadeBiolog aTropical,v.44–45,no.3-1,p.379–390. SanMart n,G.,1992, Syllis SavignyinLamarck,1818(Polychaeta:Syllidae:Syllinae)fromCuba,theGulfofMexico,FloridaandNorth Carolina,witharevisionofseveralspeciesdescribedbyVerrill: BulletinofMarineScience,v.51,no.2,p.167–196. SanMart n,G.,2003,Annelida,PolychaetaII:Syllidae, in Ramos,M.A., ed.,FaunaIbe rica,Volume21:Madrid,MuseoNacionaldeCiencias Naturales,p.336–447. SanMart n,G.,2005,Exogoninae(Polychaeta:Syllidae)fromAustralia withdescriptionsofanewgenusandtwenty-twonewspecies:Records oftheAustralianMuseum,v.57,p.39–159. Schmitter-Soto,J.J.,Escobar,E.,Alcocer,J.,Sua rez-Morales,E.,EliasGutierrez,M.,Marin,L.E.,andSteinich,B.,2002,Hydrobiologyof cenotesoftheYucata nPeninsula:Hydrobiologia,v.467,p.215–228. Sket,B.,andIliffe,T.,1980,CavefaunaofBermuda:Internationale RevuederGesamtenHydrobiologie,v.65,p.871–882. Sol s-Mar n,F.A.,andLaguarda-Figueras,A.,2008,Equinodermos, in Mej a-Ort z,L.M.,ed.,BiodiversidadAcua ticadelaIsladeCozumel: UniversidaddeQuintanaRoo,p.187–214. Sol s-Weiss,V.,Granados-Barba,A.,andMalpica-Mart nez,J.,2007, EnvironmentalevaluationofCozumelIslandMexico, in Ozhah, E.,ed.,LandCoastalInteractions:ManagingCoastalEcosystems, ProceedingsofthejointconferenceVIIIMEDCOAST,Volume2: p.775–786. Southern,R.,1914,ClareIslandSurvey:Part47,Archiannelidaand Polychaeta:ProceedingsoftheRoyalIrishAcademy,v.31,p.1–160. Sua rez-Morales,E.,andRivera-Arriaga,E.,1998,Hidrolog ayfauna acua ticadelosCenotesdelaPen nsuladeYucata n:Revistadela SociedadMexicanadeHistoriaNatural,v.48,p.37–47. Tagatz,M.E.,Ivey,J.M.,Dalbo,C.E.,andOglesby,J.L.,1982,Responses ofdevelopingestuarinemacrobenthiccommunitiestodrillingmuds: Estuaries,v.5,no.2,p.131–137. Wesenberg-Lund,E.,1958,LesserAntilleanpolychaetes,chieflyfrom brackishwater,withasurveyandabibliographyoffreshand brackish-waterpolychaetes:StudiesontheFaunaofCurac aoand OtherCaribbeanIslands,v.30,p.1–41. Wilson,R.S.,andHumphreys,W.F.,2001, Prionospiothalanji sp.nov. (Polychaeta:Spionidae)fromananchialinecave,CapeRange,northwestWesternAustralia:RecordsoftheWesternAustralianMuseum Supplement,v.64,p.105–113. Wurl,J.,andGiese,S.,2005,GroundwaterqualityresearchonCozumel Island,StateofQuintanaRoo,Mexico, in FraustoMart nez,O.,ed., DesarrolloSustentable:Turismo,CostasyEducacio n:Universidadde QuintanaRoo,p.171–176. F IRSTRECORDSOFPOLYCHAETOUSANNELIDSFROM C ENOTE A EROLITO ( SINKHOLEANDANCHIALINECAVE ) IN C OZUMEL I SLAND ,M EXICO 10 N JournalofCaveandKarstStudies, April2011



PAGE 1

THEROLEOFSMALLCAVESASBATHIBERNACULA INIOWA J OSEPH W.D IXON IowaGrottooftheNationalSpeleologicalSociety,P.O.Box228,IowaCity ,IA52244,joedixon@mchsi.com Abstract: Smallcavesprovidehabitatforavarietyofspecies,includingbats.Past researchoncavebatsinIowahasfocusedonafewlargecaves.Largecavesar e uncommonandrepresentonlyaportionoftheknowncavesinthestate.Since few hibernaculaareprotectedinIowaandnoassessmentofsmallcaveshasbeen done,bat censusdatawerecomparedtocavemorphologytodeterminethesignificanc eofsmall cavesashibernacula.Twelveyearsofcensusdata(1998–2009)werereview edforsmall caves( 50.0minlength)wherehibernatingbatshadbeendocumented.Four morphologicalfeatureswerecomparedagainstthedata:entranceaspect, entrancesize, cavelength,andinternalsurfacearea.Student’st-testandSpearmanran kcorrelation wereusedtotestforrelationshipsbetweenthepresenceandabundanceofe achspecies andeachofthefourmorphologicalfeatures.Theeasternpipistrelleoccu rredin68%of thecaves,andthelittlebrownbatin24%.Student’st-testshowedasignif icant correlationwithcavelengthforeasternpipistrelles.Spearmanrankcor relationshoweda significantnegativecorrelationwithentranceaspectandsignificantp ositivecorrelations forcavelengthandinternalsurfaceareaforeasternpipistrelles.There sultsaredifferent frompreviousstudiesonlargerIowacaves,whichshowedbigbrownbatsand little brownbatsasthemostabundantspecies.Easternpipistrellespreferredl argercaveswith verticalentrances.However,largeisasubjectiveterm,andtheresultsi ndicatethatsmall cavesareanimportantsourceofhibernaculafortheeasternpipistrelle. I NTRODUCTION Smallcavesoftenreceivelittlenoticefromresearchers, cavers,andthegeneralpublic.Largerandmorecomplex cavesreceivemoreattentionintheformofexploration, study,orrecreationalcaving(Kastning,2006).Thisbiasis oftentheresultofanthropocentriccriteria,suchascave length,whichisusuallyarbitrarilydetermined(Curl,1966). Forexample,amongstatecavesurveysintheUntied States,aminimumlengthisoftenestablishedtofilterout thosecavesdeemedlessimportant;cavesthatfallbelow thisdiscretionarylengtharesimplynotincludedinthe survey(Mylroie,2007).Suchdiscriminatorypractices towardssmallcavesareunfortunate,sincesmallcaves canoccasionallybegeologically,historically,orarchaeologicallysignificant(Kastning,2006).Inaddition,small cavesmayalsobebiologicallysignificant,sincetheyhave beenknowntoprovidehabitatforavarietyoftemperate vertebratetrogloxenicspecies,suchassnakes(Drda,1968), salamanders(BrigglerandPrather,2006;Campand Jensen,2007),andfrogs(Blair,1951;PratherandBriggler, 2001). Thebiologicalsignificanceofsmallcavesmaybe applicabletobatsaswell.Thefactthatlargercaves supportbothgreaternumbers(RaesleyandGates,1987; BrigglerandPrather,2003)andgreaterdiversityofbats (Arita,1996;Fuszaraetal.,1996;BrunetandMedell n, 2001;Nuietal.,2007)hasbeenwelldocumented,butsmall caveshavealsobeenobservedasasourceofhibernacula forsomespecies.Forexample,Ozarkbig-earedbats ( Corynorhinustownsendiiingens )(PratherandBriggler, 2002)andeasternpipistrelles 1 ( Perimyotissubflavus ) (BrigglerandPrather,2003)bothutilizesmallcavesas hibernacula.PratherandBriggler(2002)observedthatthe endangeredOzarkbig-earedbatusedsmallcavestosuch anextantthattheyrecommendedthoseintheirstudyarea beprotectedandfurthersurveysforOzarkbig-earedbats beconductedinadditionalsmallcaves. Thelackofinterestinsmallcavescouldpotentially resultintheunintentionalexclusionofpertinentdata. Cavesareknowntoprovidesomeofthemostimportant hibernaculasitesforbats(Pierson,1998),andbat communitiesinsmallhibernaculacandifferdramatically fromthoseutilizinglargehibernacula(Lesin skietal., 2004).Numerousspeciesofcave-hibernatingbatsexhibit considerablefidelitytohibernacula(Harvey,1992).For manyspeciesoftemperatebats,appropriatehibernacula areessentialfortheirsurvival,andagreaterunderstanding ofhibernaculaisnecessarytomakeappropriateconservationandmanagementdecisions(Brack,2007). ThestateofIowahasnotbeenanexceptiontothis pattern.PreviousresearchonbatsinIowacaveshasbeen irregular(Bowlesetal.,1998)andfocusedonlargercaves, averagingseveralhundredmeters,andoccasionallyovera thousandmeters,inlength(seeMuirandPolder,1960; KunzandSchlitter,1968;PruszkoandBowles,1986;and 1 Editor’sNote:Theeasternpipistrelleisnowmoreproperlycalledthetri -colored bat.(http://www.batcon.org/news2/scripts/article.asp?articleID 128). J.W.Dixon–TheroleofsmallcavesasbathibernaculainIowa. JournalofCaveandKarstStudies, v.73,no.1,p.21–27.DOI:10.4311/ jcks2010lsc0145 JournalofCaveandKarstStudies, April2011 N 21

PAGE 2

Clarketal.,1987,fornamesofcavessurveyedandLace andOhms,1992;Lace,1997;LaceandKlausner,2001;and LaceandKlausner,2004,forcavelengths).Cavesofthis sizeareuncommonandrepresentonlyafractionofthe cavesfoundwithinthestate.Oftheapproximately1,300 cavesdocumentedbytheIowaGrottooftheNational SpeleologicalSociety(KambesisandLace,2009),itis estimatedthatasmanyas90%are50morlessinlength (Dixon,2009).Althoughpreviousresearchprovided valuableinformationonthestatusanddistributionofthe state’sbatpopulations,thesedatamaynotrepresenta completelyaccurateassessmentbecausetheydidnot includethetypicalsmallcavesthatdominateIowa’skarst landscape. OftheelevenspeciesofbatsrecordedinIowa(Laubach etal.,1994;Bowlesetal.,1998),fivehavebeendocumented roostingincaveswithinthestate:thebigbrownbat ( Eptesicusfuscus ),theeasternpipistrelle,theIndianabat ( Myotissodalis ),thelittlebrownbat( Myotislucifugus ), andthenorthernmyotis( Myotisseptentrionalis )(Muirand Polder,1960;KunzandSchlitter,1968;Pruszkoand Bowles,1986;Clarketal.,1987;IowaDepartmentof NaturalResources,unpublisheddata;IowaGrotto, unpublisheddata).Allfiveareconsideredregularcaveusingspecies(Harvey,1992),andofthefive,allbutthe IndianabatregularlyhibernateincavesineasternIowa (Laubachetal.,1994).Twooftheremainingfourspecies, thelittlebrownbatandtheeasternpipistrelle,mayeven hibernateexclusivelyincavesinsomeregions(Tuttle, 2003). SincethemajorityofthecavesinIowaareconsiderably smallerthanthosethathavebeensurveyedforbatsinthe past,andfewmajorhibernaculaareprotectedinthestate (Bowlesetal.,1998),datacollectedbytheIowaGrotto wereexaminedandcomparedwithattributesofcave morphologyusingpublishedcavemapstodeterminewhat, ifany,rolesmallcavesplayedinprovidingbathibernacula inIowaandwhichofthefourcommonlyoccurringcave speciesutilizedthesehibernaculaonaregularbasis.Cave morphologywasselectedasthelimitingfactortobetested, becauseexteriorhabitatappearstohavelittletono relationshipwiththeselectionofacaveasahibernaculum (RaesleyandGates,1987;BrigglerandPrather,2003), whiletheinteriorclimatethatresultsfrommorphological featureshasbeenknowntoinfluenceroostselection(Arita andVargas,1995;Rodriquez-Duran,1998). M ETHODS In1987,theIowaGrottobegantheIowaSmallCaves SurveyProject(ISCSP)withthegoaltosystematically surveyallthecavesinthestatewithoutbias.In1998,bat censuseswereaddedasacomponentoftheISCSP.Surveys wereconductedforbothsummerroostingandhibernating batsusingeitherthedirectcountorsurfaceareaestimate methodsasdescribedbyThomasandLaVal(1988).For directcounts,eachindividualbatwascountedand identifiedtospecies.Forsurfaceareaestimates,species wereidentified,butnumbersofbatswereestimated.In situationswherespeciesidentificationwasnotpossible (e.g.,ahighroostingposition),onlynumbersofbatswere recorded.Allbatcensuseswereconductedusingvisual observationsonly.Nobatswerehandledormolested duringsamplingprocedures. TwelveyearsofbatcensusdatacollectedbytheIowa Grottowerereviewedforcavesthatmetfourpreestablishedcriteria:(1)thetotalsurveyedcavelengthwas 50morless,(2)thecensusdatewasduringhibernation,(3) ifbatswerepresent,thedirectcountmethodwasused,and (4)anyobservedbatswereidentifiedtospecies.A maximumlengthof50mwaschosenasthelimitabove whichacavewouldbeconsideredtoolargeforthisstudy because,asstatedpreviously,manyofthehistoricalbat surveyswereconductedincaveslargerthanthisandcaves longerthan50minlengthareuncommoninIowa.Any censusconductedduringSeptemberthroughmid-Maywas consideredanobservationofhibernatingindividuals.Of thefivespeciesofbatsthatroostinIowacaves,theeastern pipistrelleistypicallythefirsttoenterhibernaculaandthe lasttoleave,usuallyenteringinmid-Octoberanddeparting inmid-April(SchwartzandSchwartz,2001,p.80–83). However,theyhavebeenknowntoentercavesfor hibernationasearlyasSeptemberandremainaslateas mid-MaytotheendofMay(WhitakerandRissler,1992; VincentandWhitaker,2007).Therefore,thehibernation periodforthisstudywasexpandedtoSeptemberthrough mid-Mayinordertoensurethatearly-arrivingandlatedepartingbatswereincludedinthesample. Fourmorphologicalfeatureswereselectedforcomparisonagainstthebatcensusdata:entranceaspect(compass direction),entrancesize,cavelength,andinternalsurface area.Publishedcavemaps(Lace,1997;LaceandKlausner, 2001;LaceandKlausner,2004)wereusedtoquantify morphologicalfeatures.Allcaveshadbeensurveyedand mappedusingstandardcartographictechniquesdescribed byDasher(1994).Entranceaspectsweremeasuredin degreesofazimuthusingaSuuntoA1000compass.The compasswasorientedtomagneticnorthoneachmapand azimuthbearingsweretakenonthecenterlineofthecave entrancethatwasperpendiculartotheplaneofthe entrance.Forcavesthathadaverticalentrance(i.e.,a sinkhole)anaspectof 90degreeswasarbitrarilyassigned torepresentthenadirsinceanazimuthbearingdidnot exist. Entrancesizewasdeterminedbymeasuringthecave entranceatthewidest(plan)andhighest(profile)points usingaStaedtlerEngineer’sScalecalibratedtothemap scaleinmeters.Thesevalueswerethenusedtocalculatethe approximatesizeoftheentranceinsquaremeters.Cave lengthsweretakendirectlyfrommapsasthisfeatureisa standardcomponentofcavemapspublishedbytheIowa Grotto.Althoughcavelengthisastandardformeasuring T HEROLEOFSMALLCAVESASBATHIBERNACULAIN I OWA 22 N JournalofCaveandKarstStudies, April2011

PAGE 3

cavesize,ithasbeencriticizedforbeingpotentially inaccurate(Mylroie,2007).Tocompensateforthis, internalsurfaceareawasselectedasanalternativemeans ofmeasuringcavesizeandwasdefinedasthesurfacearea availableforroostinginsideeachcaveandincludedthe wallsandceilingsofeverysurveyedpassage.Caveswere dividedintosectionsbasedonprofileviews.Theheight, width,andlengthofeachsectionwasthenmeasuredusing thesameStaedtlerEngineerÂ’sScalecalibratedtothemap scale.Thesevalueswerethenusedtocalculatethesurface areainsquaremetersandtheresultsofeachsection summedtodeterminethetotalinternalsurfaceareafor eachcave. Relationshipstocavemorphologyweredeterminedby comparingthepresenceandabundanceofeachspeciesof battoeachofthefourmorphologicalfeaturesforthecaves inwhichtheyweredocumented.Studentst-testswereused totestfordifferencesbetweencavesthathadaspecies presentversusthosethatdidnotforeachofaspect, entrancesize,length,andinternalsurfacearea.Spearman rankcorrelation( r s )wasusedtodetermineiftherewere anycorrelationsbetweeneachmorphologicalfeatureand thenumberofbatsfromeachspecies.Allstatisticaltests weredoneat95%significance( P 0.05). R ESULTS Atotaloftwenty-fivecavesmetthesamplingcriteria; allcavesweresolutionalcaveswithasingleentrance.Caves werelocatedinsevendifferentcountiesinnortheastern Iowa(Fig.1).Thenumberofcavespercountyranged from1to10withanaverageof3.6( 6 3.4SD).The averageaspectforallcavescombinedwas97.4degrees ( 6 169.8degreesSD).However,32%(8/25)ofthecaves hadaverticalentranceandwerethereforeassignedavalue of 90degreessincetheydidnothaveageographicaspect. Theremaining17caveshadhorizontalentrances.These caveshadaspectsrangingfrom0to352degreeswitha meanof185.5degrees( 6 132.1degreesSD).Cave entrancesrangedfrom0.2to27.1m 2 andhadameanof 4.8m 2 ( 6 6.4m 2 SD).Cavelengthrangedfrom6.4to 34.9m,withanaveragelengthof18.7m( 6 9.9mSD).The internalsurfaceareahadarangeof15.5to383.8m 2 ,the meaninternalsurfaceareawas125.3m 2 ( 6 103.7m 2 SD). Threespeciesofbatsweredocumentedfrom23ofthe 25caves.Theeasternpipistrellewasthemostcommon speciesandwasfoundin68%(17/25)ofthecaves.The numberofeasternpipistrellesrangedfrom0to7withan averageof1.20( 6 1.47SD)batspercave.Thelittlebrown batwasthesecondmostcommonspecies,thoughit occurredmuchlessfrequentlythantheeasternpipistrelle. Littlebrownbatswerefoundinonly24%(6/25)ofthe cavessampled.Thenumberoflittlebrownbatspercave rangedfrom0to6withanaverageof0.76( 6 1.67SD). Easternpipistrellesandlittlebrownbatsdidnotoccur togetherinanyofthecavesinthesample.Lastly,asingle bigbrownbatwasfoundinonecavealongwithseveral littlebrownbats. TheStudentÂ’st-testindicatedamarginallysignificant differenceforcavelengthbetweencavesoccupiedby easternpipistrellesandthosethatwereunoccupied( t 2.09, df 23, P 0.048).TheStudentÂ’st-testsdidnot indicateasignificantdifferencefortheremainingmorphologicalfeatures,nordiditindicateanysignificant differencesforanyofthecavesselectedbylittlebrown bats(all P 0.237).Spearmanrankcorrelationindicated thatcavesselectedashibernaculabyeasternpipistrelles hadasignificant,moderatelynegativecorrelationwith entranceaspect( r s 0.408, P 0.043).Therewasno discerniblerelationshipbetweenentrancesizeandhibernaculaselection( P 0.289).Therewashoweveravery significant,moderatepositivecorrelationforcavelength( r s 0.506, P 0.010)andasignificant,moderatepositive correlationforinternalsurfacearea( r s 0.455, P 0.022) forcavesselectedbyeasternpipistrellesashibernacula.No significantcorrelationsforlittlebrownbatsweredetected foranyofthefourmorphologicalfeatures(all P 0.388). Nostatisticalanalysiswascompletedforbigbrownbats sincethesingleindividualdidnotrepresentasufficient sample. D ISCUSSION Withonly25cavesmeetingthepre-establishedcriteria, itappearsthatthecriteriawereperhapstoorestrictive. However,sincethefocusofthisstudywasonsmallcaves andtheiruseashibernacula,modifyingthecriteriato includelargercavesorthosethatcontainedobservations outsideofthehibernationperiodwouldhaveinvalidated anyresults.Also,thesmallpercentageofcaveswithout batsisprobablytheresultofsamplingerror.Someofthe surveyedcaveswerenotrecordedifbatswerenotpresent, accordingtoEdKlausner(personalcommunication).An additionallimitationwouldbethefailuretoincludecavelikefeaturesorexceptionallysmallcaves.Likeotherstate cavesurveys,Iowautilizesaminimumcavelength,inthis Figure1.MapofIowadepictingcounties,withnumbersof cavesincludedinthisstudy. J.W.D IXON JournalofCaveandKarstStudies, April2011 N 23

PAGE 4

case15ft(about4.5m),andnobatsurveyshavebeen undertakenbelowthislimit.Thewiderangeinsampling dates(1998to2009)mayalsointroducetemporalvariation indistributionandabundancethatcouldpotentiallybias theseresults.Duetotheselimitations,itmaybe inappropriatetoinferthattheseresultsarebroadly applicable.However,severalsignificantrelationshipswere presentandmeritdiscussion. Theresultscontradictprevioussurveysthatfoundthe bigbrownbat(MuirandPolder,1960;Pruszkoand Bowles,1986;Clarketal.1987)andlittlebrownbat(Iowa DepartmentofNaturalResources,unpublisheddata)tobe themostabundantspeciesinIowacaves.Thisislikelydue tothehistoryoffocusingonlargecaveswithinthestate. Bestetal.(1992)observedthattheeasternpipistrellewas themostcommonspecieshibernatinginsmallcaves( 65mlong)surveyedinsouthernAlabama.Priortotheir study,thespecieswasthoughttobeuncommoninthat partofthestate.Theeasternpipistrelleisnotconsidered uncommonineasternIowa(Laubachetal.,1994;Bowles etal.,1998),andtheseresultsaresupportedbyBracketal. (2004),whoobservedthateasternpipistrellesareusually theonlyspeciesfoundinsmallcaves. Onlyasinglecaveharboredmorethanonespecies (littlebrownbatsandthesinglebigbrownbat).Although cavesareknowntoprovideshelterformultiplespecies (Kunz,1982),smallercavesandsmallerhibernacula,in general,usuallyhavelessdiversebatpopulations(Arita, 1996;Fuszaraetal.,1996;BrunetandMedell n,2001; Lesin skietal.,2004;Nuietal.,2007).Theresultsofthis studyfitthatpattern.Theobservedsegregationoflittle brownbatsandeasternpipistrelleswasunexpected, however,sincebothspeciesareknowntohibernatein thesamelargercaveswithinIowa(Laubachetal.,1994). Littlebrownbatsareusuallymoreabundantinlarge, complexcaves(Gatesetal.,1984),sothefewindividuals encounteredinthesesmallcaveswasnotsurprising. However,thereappearstobewidearrayofmorphological featuresforthesurveyedcavesthatwereselectedas hibernaculabylittlebrownbats. Indeed,thecavesutilizedbylittlebrownbatsfellwithin therangeofcharacteristicsofthoseutilizedbyeastern pipistrellesand,withtheexceptionofcavelength,there wasnotasignificantdifferencebetweenthecavesselected byeachspecies.Duetothesmallsamplesizeforlittle brownbats,thissegregationmaysimplybetheresultof chance.However,itcouldbetheresultofpreviously undocumentedinterspecificcompetition,sinceasmaller cavewouldbeamorelimitedresource(i.e.,lessspace availableforroosting).Previoussurveysinthestatewould nothaveobservedthiscompetitionbecausetheywere conductedinlargercaves.Futureresearcheffortsinsmall cavesshouldincludelittlebrownbatstofurtherclarifythis behavior. Thelownumbersofeasternpipistrellesmatched previouslydocumentedbehaviorforthisspecies.Eastern pipistrelleshibernateinsmallnumbersassolitaryindividuals(FujitaandKunz,1984;VincentandWhitaker,2007), andtheirnumbersincavescanvaryconsiderablyduring hibernation(Bracketal.,2003).Thelackofarelationship withcaveentrancesizematchedexpectedbehavioraswell. Aswiththisstudy,BrigglerandPrather(2003)foundno significantrelationshipwithentrancesizeofcavesusedby easternpipistrelles.BrigglerandPrather(2003)also observedthatlongercavessupportedgreaternumbers,as indicatedinthisstudybyboththeStudentÂ’st-testandthe Spearmanrankcorrelation.Thisrelationshipwithcave lengthismostcertainlytheresultofthepipistrellesÂ’ preferenceforhibernatinginthermallystableenvironments (Rabinowitz,1981;FujitaandKunz,1984;Brigglerand Prather,2003;Brack,2007;VincentandWhitaker,2007), sincethelongerthecave,thegreaterthethermalstability (TuttleandStevenson,1978;Ferna ndez-Corte zetal., 2006).Futurestudiesinsmallcavesshouldincorporate climatemonitoring. Internalsurfaceareawasselectedasoneofthe morphologicalfeaturestobetestednotonlyasan alternativemeansofmeasuringcavesize,butalsobecause ithasalsobeenpositivelycorrelatedwithbat-species richness(BrunetandMedell n,2001).Therewasno evidencefromthisstudytosupportthisrelationshipin smallcaves.Apositivecorrelationwitheasternpipistrelles andinternalsurfaceareaissupportedbyRaesleyand Gates(1987),whofoundthatlargercavessupported greaternumbersofhibernatingbatsingeneral.Thiswould appeartobesimilartocavelength,sincecavevolumeis alsoknowntoinfluencecavetemperature(Tuttleand Stevenson,1978).However,anextremelysignificant, strongcorrelationwasnoticedbetweencavelengthand internalsurfacearea( r s 0.877, P 0.001).Thatis, longercavesinthisstudyhadcorrespondinglylarger internalsurfaceareas.Therefore,theseresultscannotsay withanycertaintyifthepreferenceexhibitedbyeastern pipistrelleswastheresultofcavelength,internalsurface area,oracombinationofbothinterrelatedspatial parameters. Thenegativecorrelationbetweenentranceaspectand easternpipistrellehibernaculawouldseemtoindicatea preferenceforsinkholeentrancecavessinceapproximately one-thirdofthecavesinthesample(32%)werecaveswith thearbitrarilyassigned 90degreeaspect.Greater numbersofeasternpipistrelleswerefoundincaveswith verticalentrances.Sinkhole-entrancecaveshadanaverage of2.13( 6 2.23SD)easternpipistrellespercave,while horizontal-entrancecavesaveraged0.76( 6 0.66SD)per cave.BrigglerandPrather(2003)foundlargercaveswith east-facingaspectswerepreferredbyeasternpipistrelles overothermorphologies.Theynotedhoweverthatthe majorityoftheeast-facingcavestheyexaminedwere sinkhole-entrancecavesoncomparativelyflatground.In thisstudyaswell,sinkhole-entrancecavesweretypically longerandlarger,havinggreaterinternalsurfacearea. T HEROLEOFSMALLCAVESASBATHIBERNACULAIN I OWA 24 N JournalofCaveandKarstStudies, April2011

PAGE 5

Theaveragelengthandinternalsurfaceareaof horizontalentrancecavesinthisstudywas15.6m( 6 8.8mSD)and82.3m 2 ( 6 53.1m 2 SD),respectively.The averagelengthandinternalsurfaceareaofvertical entrancecaveswas25.2m( 6 9.3mSD)and216.5m 2 ( 6 128.2m 2 SD).BrigglerandPrather(2003)hypothesized thatthepreferencetheyobservedforsinkholeentrance caveswasduetothesizeofthecavesandnottheresultof entranceaspect.However,consideringthepreviously mentionedpreferenceforthermallystablehibernacula, thismaybeevidenceofapreferredmorphologicalfeature whenselectingahibernaculum,sinceverticalentrance cavesaremorethermallystablethanthosewithhorizontal entrances(DaanandWichers,1968;TuttleandStevenson, 1978).Thisbehaviorhasbeenobservedinotherspeciesas well,suchastheOzarkbig-earedbat,whichalsohasa preferenceforhibernatingincaveswithsinkholeentrances (Clarketal.,1996;PratherandBriggler,2002).Likecave lengthandinternalsurfacearea,theseresultsarestill uncertainduetotheassociationbetweenentranceaspect andcavesize.Basedonthesedata,thereisnowayto determineifthiscorrelationisaresultofentranceaspect alone. Despitethepreviouslylistedlimitations,theresultsof thisstudyandthecitedworksofothersdemonstratethat easternpipistrellesselectlargercavesashibernacula.This issupportednotonlybythepositiveSpearmanrank correlations,butalsobytheStudent’st-testthatfounda significantdifferenceforcavesselectedbyeasternpipistrellesbasedoncavelength.Thispreferenceismostlikely duetomorestabletemperaturespresentinlargercaves. However,itisapparentfromthesedatathat‘‘large’’isan extremelysubjectivetermsincethelargestcavesinthis studywereonlyaround34minlengthwithapproximately 300m 2 ofinternalsurfacearea.Basedontheseresults,I canonlyconcludethateventhoughtheyweresmall,the cavesinthissamplewerestillofsufficientsizetoexhibit sufficientstabilityoftemperaturethateasternpipistrelles tolerate.Althoughthesamplesizewasmodestandthe samplingdatesspreadoutoveratwelve-yearperiod,these resultsshouldnotbediscounted.Trombulaketal.(2001) usedsimilarirregularcensusdataspanningseveraldecades fromonlytwenty-threehibernaculacavestodocument populationtrendsinVermont. Thisstudyalsodemonstratesthatlocalgrottosofthe NationalSpeleologicalSocietycanbeavaluablesourceof dataandplayakeyroleinmonitoringandstudyinglocal batpopulations.Knowledgeofhibernacularoostinghabits isanindispensabletoolinunderstandingthebiogenicand anthropogenicimpactsonpopulations(O’Sheaetal.,2003; Tuttle,2003).Accurateconservationdecisionscanonlybe madewithafullunderstandingofhibernacularequirements(Brack,2007),andgiventhethreatthatmanyspecies facetoday(e.g.,white-nosesyndrome,destructionof habitat,etc.),thisknowledgeisperhapsmoreimportant nowthanever.Moredetailedandorganizedsurveyefforts bytheIowaGrotto,andothergrottosaroundthecountry, isbothencouragedandrecommended. C ONCLUSIONS Gatesetal.(1984)statedthateasternpipistrelleswere generalistswhenitcametocaveselection.Thisis supportedbyHarvey(1992)andBracketal.(2003),who notedthatintheeasternUnitedStatestheyareknownto utilizemorecavesashibernaculathananyotherspecies. However,easternpipistrelleshavealsobeenknownto expressapronouncedfidelitytoparticularhibernacula (FujitaandKunz,1984).Intheirstudyofunderground cellarsinPoland,Lesin skietal.(2004)statedthatsmall, well-distributedhibernaculaaremostlikelytobeusedby localbatcommunities,asthiswouldincreasetheirchance ofsurvivalasopposedtoalongmigrationtoalarger hibernaculasite.Sincehibernaculaforeasternpipistrelles arepresumedtobewithin100kmofsummerroostingsites (VincentandWhitaker,2007)andcomparativelypermanentroostssuchascavesareabletosupportstablebat populationsandannualusepatterns(Agostaetal.,2005), itisapparentthatsmallcavesareanimportantsourceof hibernaculafortheeasternpipistrelle. Furtherresearchontheroleofsmallcavesas hibernaculaisneeded.First,thepreferredmorphologyof sinkholeentrancesbyeasternpipistrellesevidentinthis studyandnotedbyBrigglerandPrather(2003)needstobe validatedorrefuted.Atthepresenttime,thereisnowayof discerningwhetherthisisanactualpreferredfeatureor whetherthiswasmerelyacoincidentaloccurrencewith largercaves.Second,additionaldataareneededinorderto determinewhetherthepresenceofsomespecieshibernating insmallcaves,suchasthelittlebrownbatsobservedinthis study,areanomaliesorwhetherthesespeciesalsoutilize smallcavesonaregularbasis.Specificknowledgeof hibernaculaisnecessarytoadequatelymeetconservation needs(O’Sheaetal.,2003;Tuttle,2003;Brack,2007),even forcommonspeciesthatareconsideredabundant(Agosta, 2002).Amorecompleteunderstandingofroostinghabitats isessentialsincethedestructionofmoresignificant hibernaculamayresultinlessimportantsites(suchas smallcaves)emergingasmajorhibernacula(Gatesatal., 1984). A CKNOWLEDGMENTS IwouldliketothankEdKlausnerandMikeLacefor reviewingthismanuscript.Iwouldalsoliketothankthe followingfellowmembersoftheIowaGrottofortheir diligentworkinrecordingbatobservationsinIowacaves; withoutthemthispaperwouldnothavebeenpossible:Ed Klausner,ElizabethMiller,MikeLace,ChrisBeck,Gary Engh,JimRoberts,RichFeltes,PhilLaRue,andRay Finn. J.W.D IXON JournalofCaveandKarstStudies, April2011 N 25

PAGE 6

R EFERENCES Agosta,S.J.,2002,Habitatuse,dietandroostselectionbythebigbrown bat( Eptesicusfuscus )inNorthAmerica:Acaseforconservingan abundantspecies:MammalReview,v.32,p.79–198,doi:10.1046/ j.1365-2907.2002.00103.x. Agosta,S.J.,Morton,D.,Marsh,B.D.,andKuhn,K.M.,2005,Nightly, seasonal,andyearlypatternsofbatactivityatnightroostsinthe centralAppalachians:JournalofMammalogy,v.86,p.1210–1219, doi:10.1644/05-MAMM-A-012R1.1. Arita,H.T.,1996,Theconservationofcave-roostingbatsinYucatan, Mexico:BiologicalConservation,v.76,p.117–185. Arita,H.T.,andVargas,J.A.,1995,Naturalhistory,interspecific association,andincidenceofthecavebatsofYucatan,Mexico: SouthwesternNaturalist,v.40,p.29–37. Best,T.L.,Carey,S.D.,Caesar,K.G.,andHenry,T.H.,1992, Distributionandabundanceofbats(Mammalia:Chiroptera)in coastalplaincavesofsouthernAlabama:NSSBulletin,v.54, p.61–65. Blair,A.P.,1951,WinteractivityinOklahomafrogs:Copeia,v.2,178p. Bowles,J.B.,Howell,D.L.,Lampe,R.P.,andWhidden,H.P.,1998, MammalsofIowa:Holocenetotheendofthe20thcentury:Journal oftheIowaAcademyofScience,v.105,p.123–132. Brack,V.Jr.,Johnson,S.A.,andDunlap,R.K.,2003,Wintering populationsofbatsinIndiana,withemphasisontheendangered Indianamyotis, Myotissodalis :ProceedingsoftheIndianaAcademy ofScience,v.112,p.61–74,(www.indianaacademyofscience.org/ media/attachments/IAS_v112_n1_p61-74.pdf). Brack,V.Jr.,2007,Temperaturesandlocationsusedbyhibernating bats,including Myotissodalis (Indianabat),inalimestonemine: Implicationsforconservationandmanagement:Environmental Management,v.40,p.739–746,doi:10.1007/s00267-006-0274-y. Brack,V.Jr.,Whitaker,J.O.Jr.,andPruitt,S.E.,2004,BatsofHoosier NationalForest:ProceedingsoftheIndianaAcademyofScience, v.113,p.76–86,(www.indianaacademyofscience.org/media/attachment s/ PIAS_v113_n1_p76-86.pdf). Briggler,J.T.,andPrather,J.W.,2003,Seasonaluseandselectionofcav es bytheeasternpipistrellebat( Pipistrellussubflavus ):American MidlandNaturalist,v.149,p.406–412,doi:10.1674/0003-0031(2003) 149[0406:SUASOC]2.0.CO;2. Briggler,J.T.,andPrather,J.W.,2006,Seasonaluseandselectionofcav es byplethodontidsalamandersinakarstareaofArkansas:American MidlandNaturalist,v.155,p.136–148,doi:10.1674/0003-0031(2006) 155[0136:SUASOC]2.0.CO;2. Brunet,A.K.,andMedell n,R.A.,2001,Thespecies-arearelationship inbatassemblagesoftropicalcaves:JournalofMammalogy, v.82,p.1114–1122,doi:10.1644/1545-1542(2001)082 1114:TSARIB 2.0.CO;2. Camp,C.D.,andJensen,J.B.,2007,Useoftwilightzonesofcavesby plethodontidsalamanders:Copeia,v.3,p.594–604,doi:10.1643/ 0045-8511(2007)2007[594:UOTZOC]2.0.CO;2. Clark,B.S.,Bowles,J.B.,andClark,B.K.,1987,Summeroccurrencesof theIndianabat,Keen’smyotis,eveningbat,silver-hairedbat,and easternpipistrelleinIowa:ProceedingsoftheIowaAcademyof Science,v.94,p.89–93. Clark,B.K.,Clark,B.S.,Leslie,D.M.Jr.,andGregory,M.S.,1996, CharacteristicsofcavesusedbytheendangeredOzarkbig-earedbat: WildlifeSocietyBulletin,v.24,p.8–14. Curl,R.L.,1966,Cavesasameasureofkarst:JournalofGeology,v.74, p.798–830. Daan,S.,andWichers,H.J.,1968,Habitatselectionofbatshibernatingi n alimestonecave:ZeitschriftfurSaugetierkunde,v.33,p.262–287. Dasher,G.R.,1994,OnStation:ACompleteHandbookforSurveying andMappingCaves:Huntsville,Alabama,NationalSpeleological Society,242p. Dixon,J.W.,2009,CavesizeinIowa:HowlongistheaverageIowacave?: Intercom,publishedsemi-spasmodicallybytheIowaGrottoofthe NationalSpeleologicalSociety,v.45,p.17–22. Drda,W.J.,1968,Astudyofsnakeswinteringinasmallcave:Journalof Herpetology,v.1,p.64–70. Ferna ndez-Corte z,A.,Calaforra,J.M.,Jime nez-Espinosa,R.,and Sa nchez-Martos,F.,2006,Geostatisticalspatiotemporalanalysisof airtemperatureasanaidtodelineatingthermalstabilityzonesina potentialshowcave:Implicationsforenvironmentalmanagement: JournalofEnvironmentalManagement,v.81,p.371–383,doi: 10.1016/j.jenvman.2005.11.011. Fujita,M.S.,andKunz,T.H.,1984, Pipistrellussubflavus :Mammalian Species,v.228,p.1–6. Fuszara,E.,Kowalski,M.,Lesin ski,G.,andCygan,J.P.,1996, Hibernationofbatsinundergroundsheltersofcentraland northeasternPoland:BonnerzoologischeBeitrage,v.46,p.349–358. Gates,J.E.,Feldhamer,G.A.,Griffith,L.A.,andRaesly,R.L.,1984, Statusofcave-dwellingbatsinMaryland:Importanceofmarginal habitats:WildlifeSocietyBulletin,v.12,p.162–169. Harvey,M.J.,1992,BatsoftheEasternUnitedStates:LittleRock, Arkansas,ArkansasGameandFishCommission,46p. Kambesis,P.,andLace,M.,2009,Iowa, in Palmer,A.N.,andPalmer, M.V.,eds.,CavesandKarstoftheUSA:Huntsville,Alabama, NationalSpeleologicalSociety,p.151–154. Kastning,E.H.,2006,Verysmalleclecticcaves:conservationand managementissues, in Rea,G.T.,ed.,Proceedingsofthe17 th NationalCaveandKarstManagementSymposium:Albany,New York,NCKMSSteeringCommittee,p.92–101. Kunz,T.H.,1982,Roostingecologyofbats, in Kunz,T.H.,ed.,Ecology ofBats:NewYork,NewYork,PlenumPress,p.1–55. Kunz,T.H.,andSchlitter,D.A.,1968,Anannotatedchecklistofbats fromIowa:TransactionsoftheKansasAcademyofScience,v.71, p.166–175,(www.bu.edu/cecb/files/2009/08/iowachecklist.pdf). Lace,M.J.,ed.,1997,TheIowaCaveMapBook,VolumeII:IowaCity, Iowa,IowaGrottooftheNationalSpeleologicalSociety,210p. Lace,M.J.,andKlausner,E.,eds.,2001,TheIowaCaveMapBook, VolumeIII:IowaCity,Iowa,IowaGrottooftheNational SpeleologicalSociety,223p. Lace,M.J.,andKlausner,E.,eds.,2004,TheIowaCaveMapBook, VolumeIV:IowaCity,Iowa,IowaGrottooftheNational SpeleologicalSociety,183p. Lace,M.J.,andOhms,M.,eds.,1992,TheIowaCaveMapBook, VolumeI:IowaCity,Iowa,IowaGrottooftheNationalSpeleological Society,151p. Laubach,C.M.,Bowles,J.B.,andLaubach,R.,1994,AGuidetotheBats ofIowa:DesMoines,Iowa,IowaDepartmentofNaturalResources, NongameTechnicalSeriesNo.2,18p. Lesin ski,G.,Kowalski,M.,Doman ski,J.,Dzie cioowski,R.,LaskowskaDzie cioowska,K.,andDzie gielewska,M.,2004,Theimportanceof smallcellarstobathibernationinPoland:Mammalia,v.68, p.345–352. Muir,T.J.,andPolder,E.,1960,NotesonhibernatingbatsinDubuque Countycaves:ProceedingsoftheIowaAcademyofScience,v.67, p.602–606. Mylroie,J.,2007,Cavesurveys,cavesize,andflankmargincaves: Compass&Tape,v.17,no.4,p.8–16,(www.caves.org/section/sacs/ rr/cavelenth.pdf). Nui,H.,Wang,N.,Zhao,L.,andLui,J.,2007,Distributionand undergroundhabitatsofcave-dwellingbatsinChina:Animal Conservation,v.10,p.470–477,doi:10.1111/j.1469-1795.2007. 00136.x. O’Shea,T.J.,Bogan,M.A.,andEllison,L.E.,2003,Monitoringtrendsin batpopulationsoftheUnitedStatesandterritories:Statusofthe scienceandrecommendationsforthefuture:WildlifeSocietyBulletin, v.31,p.16–29,(www.fort.usgs.gov/Products/Publications/21045/ 21045.pdf). Pierson,E.D.,1998,Talltrees,deepholes,andscarredlandscapes: ConservationbiologyofNorthAmericanbats, in Kunz,T.H.,and Racey,P.A.,eds.,BatBiologyandConservation:Washington,D.C., SmithsonianInstitutionPress,p.309–325. Prather,J.W.,andBriggler,J.T.,2001,Useofsmallcavesbyanurans duringadroughtperiodintheArkansasOzarks:Journalof Herpetology,v.35,p.675–678. Prather,J.W.,andBriggler,J.T.,2002,UseofsmallcavesbyOzarkbigearedbats( Corynorhinustownsendiiingens )inArkansas:American MidlandNaturalist,v.148,p.193–197,doi:10.1674/0003-0031(2002) 148[0193:UOSCBO]2.0.CO;2. Pruszko,R.,andBowles,J.B.,1986,SurveyofsomeeasternIowacaves forwinteringbats:ProceedingsoftheIowaAcademyofScience,v.93, p.41–43,(137.227.227.130/BPD/BPD_bib_viewObs.asp?BibID 1300). Rabinowitz,A.,1981,Thermalpreferenceoftheeasternpipistrellebat ( Pipistrellussubflavus )duringhibernation:JournaloftheTennessee AcademyofScience,v.56,p.113–114. T HEROLEOFSMALLCAVESASBATHIBERNACULAIN I OWA 26 N JournalofCaveandKarstStudies, April2011

PAGE 7

Raesly,R.L.,andGates,J.E.,1987,Winterhabitatselectionbynorth temperatecavebats:AmericanMidlandNaturalist,v.118,p. 15–31. Rodr guez-Dura n,A.,1998,Nonrandomaggregationsanddistributions ofcave-dwellingbatsinPuertoRico:JournalofMammalogy,v.79, p.141–146. Schwartz,C.W.,andSchwartz,E.R.,2001,TheWildMammalsof Missouri:Columbia,Missouri,UniversityofMissouriPress,392p. Thomas,D.W.,andLaVal,R.K.,1988,SurveyandCensusMethods, in Kunz,T.H.,ed.,EcologicalandBehavioralMethodsfortheStudyof Bats:Washington,D.C.,SmithsonianInstitutionPress,p.77–89. Trombulak,S.C.,Higuera,P.E.,andDesMeules,M.,2001,Population trendsofwinteringbatsinVermont:NortheasternNaturalist,v.8, p.51–62,doi:10.1656/1092-6194(2001)008[0051:PTOWBI]2.0.CO;2. Tuttle,M.D.,2003,Estimatingpopulationsizesofhibernatingbatsin cavesandmines, in O’Shea,T.J.,andBogan,M.A.,eds.,Monitoring TrendsinBatPopulationsoftheUnitedStatesandTerritories: ProblemsandProspects:U.S.GeologicalSurvey,BiologicalResourcesDiscipline,InformationandTechnologyReportUSGS/BRD/ITR2003-0003.p.31–39. Tuttle,M.D.,andStevenson,D.E.,1978,Variationinthecave environmentanditsbiologicalimplications, in Zuber,R.,Chester, J.,Gilbert,S.,andRhodes,D.,eds.,Proceedingsofthe1977National CaveManagementSymposium:Albuquerque,NewMexico,Adobe Press,p.108–121. Vincent,E.A.,andWhitaker,J.O.Jr.,2007,Hibernationoftheeastern pipistrelle, Perimyotissubflavus ,inanabandonedmineinVermillion County,Indiana,withsomeinformationon Myotislucifugus : ProceedingsoftheIndianaAcademyofScience,v.116,p.58–65, (www.indianaacademyofscience.org/media/attachments/Proc_v116_ 1_2007_pp58-65.pdf). Whitaker,J.O.Jr.,andRissler,L.J.,1992,Seasonalactivityofbatsat CopperheadCave:ProceedingstheIndianaAcademyofScience, v.101,p.127–134. J.W.D IXON JournalofCaveandKarstStudies, April2011 N 27



PAGE 1

ANEWSPECIESOFTHEGENUS PLUTOMURUS YOSII,1956(COLLEMBOLA,TOMOCERIDAE)FROM GEORGIANCAVES R EVAZ D JANASHVILIAND S HALVA B ARJADZE EntomologyandBiocontrolResearchCentre,IliaStateUniversity,Chavc havadzeav.31,0179,Tbilisi,Georgia,shalva1980@yahoo.com Abstract: Anewspecies, Plutomurusbirsteini sp.n.,fromGeorgiancavesisdescribed andillustrated.Itissimilarto Plutomurusbaschkiricus (Skorikow,1899).Differences betweenthespeciesarediscussed.Akeytothegenus Plutomurus speciesfoundinthe Caucasusisprovided. I NTRODUCTION Thegenus Plutomurus Yosii,1956isrepresentedby twenty-threespeciesworldwide(KnissandThibaud,1999; Bellingeretal.,1996–2010).Thespeciesbelongingtothis genusaredistributedinNorthAmerica,Europe,andAsia andliveincavesorsoils(Yosii,1956,1966,1967,1991; Chrisitiansen,1964,1980;Gruia,1965;Martynova,1969, 1977;RusekandWeiner,1977;KnissandThibaud,1999). Beforenow,fourspecies, Plutomurusabchasicus Martynova,1969(Georgia,insoils), P.jeleznovodski Knissand Thibaud,1999(northCaucasus,incave), P.kelasuricus Martynova,1969(northandsouthCaucasus,incaves)and P.sorosi KnissandThibaud,1999(northCaucasus,incave) havebeenrecordedintheCaucasus(Martynova,1969; KnissandThibaud,1999;BarjadzeandDjanashvili,2008). Re-examinationofthespringtailscollectedbyDr.R. DjanashviliinGeorgiancavesduring1966–1970ledusto describehereanewspecies. D IAGNOSISOFTHE G ENUS P LUTOMURUS Y OSSI ,1956 Eyesfrom6 6to0 0.Prelabialsetaefrom2 2to4 4.Thisgenusiseasilydistinguishedbythepresenceof largespine-likesetaeattheexternalbasesofthedens combinedwithawelldevelopedmultisetaceoustrochateral organsonthetrochanterandonthebaseofthefemur. Therearefourorfewerintermediateteethonthemucro, andthereisonlyasinglesmallmucronallamella,usually onthebasaltooth.Mosthavesomepigment,buteyesare oftenreduced.Sexualdimorphismabsent.Thegenusis Holarcticbutdoesnotoccurinarcticregions.Itoftenis foundinalpineareas.WithintheHolarcticregion,theyare limitedtoeasternAsia,westernEurope,theCaucasian region,andwesternNorthAmerica.Although60%are foundincavesandmanyofthesehavenotbeenfound elsewhere,theyshowverylittletroglomorphy. M ATERIAL E XAMINED Holotype = fromSakishoreCave,06.VII.1969,leg.R. Djanashviliand21paratypesareontheslides.Specimens fromthesamelocalityaremountedonthesameslide. Paratypes:threespecimens,ChuneshiCave,21.IX.1968, leg.R.Djanashvili;onespecimen,SatapliaIVCave, 10.X.1966,leg.R.Djanashvili;onespecimen,TvishiCave, 12.IX.1968,leg.R.Djanashvili;onespecimen,Magara Cave,18.VIII.1967,leg.R.Djanashvili;threespecimens, TsakhiCave,31.VIII.1966,leg.R.Djanashvili;one specimen,SharaulaIICave,20.VII.1970,leg.R.Djanashvili;fourspecimens,NikortsmindaCave,03.VII.1969,leg. R.Djanashvili;onespecimen,SakishoreCave, 06.VII.1969,leg.R.Djanashvili;twospecimens,Bnelaklde Cave,18.IX.1967,leg.R.Djanashvili;onespecimen, TarokldeCave,15.IX.1967,leg.R.Djanashvili;three specimens,KhvedelidzeCave,29.VI.1969.,leg.R.Djanashvili.Holotypeandparatypesaredepositedinthe EntomologyandBiocontrolResearchCentreofIliaState University,Tbilisi,Georgia. D ESCRIPTIONOF P LUTOMURUSBIRSTEINI D JANASHVILI AND B ARJADZESP N Bodylength1.0to3.3mm.Bodylightyellowish-gray; anteriorpartofheadandIIIandIVantennalsegmentsa littledarker.Antennac.1.5timesshorterthanbody.Eye patchsmall,blackwith6eyes(Fig.1A).Prelabialsetae 2 2,labrumwith5,5,4setaeand4curvedsetaeonthe distalpartoflabrum.Trochanteralorganwelldeveloped ontrochanterandfemur,composedoflargenumberof shortsetae(morethan40)andseverallongsetae.Allparts ofventraltubewithmanysetae.Spine-likesetaeon tibiotarsusI,II,andIII:0,0,1.Tenenthairsweaklyto moderatelyclavate.Ratioofunguis,unguiculus,and tenenthairinholotype16:9:11.Unguiswithwelldeveloped pseudonychia,0.38to0.56timesaslongasinnerpartof unguis.Unguisofprolegswith2to4innerteeth.Unguisof mesoandmetalegswith2to3innerteeth(Fig.1B). Unguiculusofalllegsalwayswithoutinnerteeth. Tenaculumwith4 4teethand1heavysetaoncorpus (Fig.1C).Ratioofmanubrium/dens/mucroinholotype 6.3:9.8:1.Ventralsideofmanubriumwithnumerous relativelythickandlargesetae.Outermarginsofdens with4thickandlargespine-likesetae.Mucrowith2basal R.DjanashviliandS.Barjadze–Anewspeciesofthegenus Plutomurus Yosii,1956(Collembola,Tomoceridae)fromGeorgiancaves. JournalofCaveandKarstStudies, v.73,no.1,p.28–30.DOI:10.4311/jcks2010lsc0147 28 N JournalofCaveandKarstStudies, April2011

PAGE 2

denticles,withoutsubapicaldenticlesand1setaprojecting prominentlybeyondapex(Fig.1D).Dentalformula variable:5-15 1-2 /2-9 1 0-3 0-1 0-3 0-1 0-3 0-1 0-2 0-1 02 0-1 (Fig.1E). E TYMOLOGY ThespecificnameischoseninhonoroftheRussian biospeleologistDr.J.A.Birstein,whowasasupervisorof thefirstauthor. D ISCUSSION Thelackoftroglomorphicmodificationsmakesitlikely thatthisisatroglophilespeciesandwillbefoundoutside ofcaves.Thisspeciesisaverysimilarto Plutomurus baschkiricus ,whichdiffersfrom P.birsteini by: 1. Larger numberofsetaeonthetrochanteralorgan(morethan40 shortsetae)andseverallongsetaein P.birsteini ,while thereare18to40shortsetaeand2to4largeonesin P. baschkiricus ; 2. Spine-likesetaeontibiotarsusI,II,andIII are0,0,1in P.birsteini ,while0,0,2in P.baschkiricus ; 3. Unguiculusofalllegsalwayswithoutminuteinnerteethin P.birsteini ,while2to3innerteethareoneachunguiculus of P.baschkiricus ; 4. Dentalformula:5-15 1-2 /2-9 1 0-3 01 0-3 0-1 0-3 0-1 0-2 0-1 0-2 0-1 in P.birsteini ,while6-8 1 / 9-13 1 in P.baschkiricus .Akeytothegenus Plutomurus Yosii,1956speciesdistributedintheCaucasusisprovided inTable1. A CKNOWLEDGEMENTS WewouldliketothanktoDr.KennethA.Christiansen, GrinnellCollege,USA,forthevaluablecommentsonthe speciesdescription. Figure1. Plutomurusbirsteini sp.n.:(A)eyepatch;(B) clawIII;(C)tenaculum;(D)mucro;(E)dens. Table1.Akeytothegenus Plutomurus Yosii,1956speciesdistributedintheCaucasus. 1.Tenenthairsweaklytomoderatelyclavate. .................................................2 Tenenthairsacuminate... .............................................................4 2.Eyepatchalwayswith6eyes.Mucrowithoutsubapicalsmalldenticles.. ............................3 Eyepatchwith4or5eyes.Mucrowith3subapicaldenticles... .......... P.sorosi KnissandThibaud,1999 3.Trochanteralorganwithmorethan40setae.Unguiculusofalllegsalway swithoutminuteinnerteeth.Dental formula:5-151-2/2-910-30-10-30-10-30-10-20-10-20-1 ....................... P.birsteini sp.n. Trochanteralorganwithlessthan40setae.Unguiculusofalllegsalwaysw ithafewminuteinnerteeth.Dental formula:4-51/6131. ................................ P.jeleznovozski KnissandThibaud,1999 4.Spine-likesetaeontibiotarsusI,II,andIII0,0,1.Prelabialsetae3 3........ P.kelasuricus Martynova,1969 Spine-likesetaeontibiotarsusI,II,andIII0,0,2.Prelabialsetae2 2......... P.abchasicus Martynova,1969 R.D JANASHVILIAND S.B ARJADZE JournalofCaveandKarstStudies, April2011 N 29

PAGE 3

R EFERENCES Barjadze,Sh.,andDjanashvili,R.,2008,Checklistofthespringtails (Colembola)ofGeorgia:CaucasianEntomologicalBulletin,v.4, no.2,p.187–193. Bellinger,P.F.,Christiansen,K.A.,andJanssens,F.,1996–2010,Check list oftheCollembolaoftheWorld.http://collembola.org[accessedApril 27,2010] Christiansen,K.,1964,Arevisionofthenearcticmembersofthegenus Tomocerus :Revued’E cologieetdeBiologieduSol,v.1,p.639– 678. Christiansen,K.,1980,Anewnearcticspeciesofthegenus Tomocerus (Collembola:Entomobryidae):ProceedingsoftheIowaAcademyof Sciences,v.87,p.121–123. Gruia,M.,1965,ContributiilastudiulcollembolelordinRoma ˆnia: LucrarileInstitutuluideSpeologie‘‘EmilRakovita’’,v.4,p.191–202. Kniss,V.,andThibaud,J.,1999,LegenrePlutomurusenRussieeten Georgie(Collembola,Tomoceridae):Revuefranc aised’Entomologie (N.S.),v.21,no.2,57_64p. Martynova,E.,1969,SpringtailsofthefamilyTomoceridae(Collembola) fromthefaunaoftheUSSR:EntomologicalReview,v.43, p.299–314(inRussian). Martynova,E.,1977,SpringtailsofthefamilyTomoceridae(Collembola) fromthefaunaoftheFarEast:InsectFaunaoftheFarEast,v.46, no.149,p.3–16(inRussian). Rusek,J.,andWeiner,W.M.,1977, Plutomuruscarpaticus sp.n. (Collembola:Tomoceridae)fromtheCarpathianMountains:Bulletin del’Acade miePolonaisedesSciences,v.25,p.741–747. Yosii,R.,1956,MonographiezurHo ¨hlencollembolenJapans:ContributionfromtheBiologicalLaboratoryKyotoUniversity3,109p. Yosii,R.,1966,ResultsofthespeleologicalsurveyofSouthKorea1966, IVcaveCollembola:BulletinoftheNationalScienceMuseumTokyo, v.9,p.541–561. Yosii,R.,1967,StudiesonthecollembolanfamilyTomoceridaewith specialreferencetoJapaneseforms:ContributionfromtheBiological LaboratoryKyotoUniversity20,54p. Yosii,R.,1991,AnewspeciesofTomoceridCollembolafromthecaveof thePref.Iwata:AnnalsoftheSpeleologicalResearchInstituteof Japan,v.9,p.1–2. A NEWSPECIESOFTHEGENUS P LUTOMURUS Y OSII ,1956(C OLLEMBOLA ,T OMOCERIDAE ) FROM G EORGIANCAVES 30 N JournalofCaveandKarstStudies, April2011



PAGE 1

SUBGLACIALMAZEORIGININLOW-DIPMARBLESTRIPE KARST:EXAMPLESFROMNORWAY R ANNVEIG VREVIK S KOGLUND 1 AND S TEIN -E RIK L AURITZEN 1,2 Abstract: Mazecavesornetworkcavesareenigmaticintheirevolution,astheyform flownetsratherthanmoreefficient,directpoint-to-pointflowroutes. Networkcavesare oftencharacterizedbyuniformpassagedimensionsinseveraldirections ,whichindicates simultaneousdissolutionofmostavailablefractures.Nonshauggrottai nGildeska l, northernNorway,isformedinlow-dipmarblestrataandsituatedasarelic tina topographicalandhydrologicalhangingposition,thuslackingamodernd rainagearea. Thecavedisplaysareticulatenetworkgeometrydictatedbytwoorthogona lfracturesets. Passagemorphologyandpaleocurrentmarksinthecavewalls(scallops)de monstrate thatthecaveevolvedunderwater-filledconditions(phreatic)andthatt herelativelyslow flowwasdirecteduphilltowardstheconfiningaquicludeandtheclifffac e.Inthatsense, ithassomeresemblancetohypogenecaves.However,weproposethatthecav eisaresult ofice-contactspeleogenesis,asitdevelopedintheleesideoftheNonsha ugenridgeunder topographicallydirectedglacierflowandseemsindependentoftheother wisevariable flowregimescharacteristicoftheglacialenvironment. I NTRODUCTION Karstcavesinlimestoneareformedbydissolution whenwateraggressivewithrespecttocalciumcarbonate flowsthroughintegratedopeningsintherock.This fundamentalprocessresultsinawidevarietyofcave geometries,controlledbytheoriginalporosityintherock andhydraulicboundaryconditions(Palmer,1991).Maze cavesaremoreenigmaticintheiroriginthanmeteoricor ‘‘common’’caves(FordandWilliams,2007),becausemaze cavesconsistofflownets,incontrasttoefficientpoint-topointflowroutes,asinlinearandbranchworkcaves. Mazesevolvebysimultaneousdissolutionofallavailable fracturesatsimilarrates,soconduitsinthesecavestendto beofequaldimensions.Inaccordancewiththisobservation,twoalternativemechanismsofmazeoriginare proposed(White,1969;Palmer,1975,1991;Fordand Williams,2007).Mazescanoriginatethroughdiffuse,slow inter-formationalrecharge(transversespeleogenesis) (Klimchouk,2009)orbyrechargethroughrepeated floodinginavariablefluvialregime.Diffuserecharge suppliesallavailablefractureswithsolutionallyaggressive waterfromanoverlyingorunderlyingunit,generallya sandstone,sothatallfracturesaredissolvedandenlarged simultaneouslyandatsimilarrates.Asimilareffectoccurs ifthereareshortflowpathsfromwherewaterentersthe rock,causingallfracturesexceptforthetightesttoenlarge simultaneouslyatsimilarrates,asinepikarst(Palmer, 2002). Floodingcausestemporalvariationindischargeand steepeningofthehydraulicgradient,forcingaggressive waterintoavailablefractures,whichaccordinglydissolve atsimilarrates(Palmer,2001).Highdischargeandhigh hydraulicgradientmayoccurinhydraulicregimeswith greatvariability,asintheepiphreaticorflood-waterzone ofrivercaves.Aparticularlyefficientsituationoccurs whenatrunkpassageischokedduetocollapse.Here,the dammingeffectoftheconstrictionincreasesthehydraulic headandamplifiestheeffectoffloods,whichwillforce waterintoallavailablefractures(Palmer,2002). Mazecavesarelociofextremekarstporositythat,like largechambers,arepossibleprecursorstobrecciazonesin paleokarst(seeforinstancemechanismssuggestedby Loucks,1999).Areasofmoderatetohighkarstporosity areofmajorimportanceforthecapacityandyieldof groundwateraquifers,formineralization,andforpetroleummigrationincarbonatereservoirs.Thisispartofthe motivationforunderstandingthespeleogeneticmechanismsformazecaves. M AZE C AVESIN S TRIPE K ARST Mazecavesarequitecommoninthestripekarstof centralScandinavia.Stripekarstoccursinthinlayersof marbleinterbeddedwithlayersofimpermeablebedrock, mainlymicaschist(Horn,1937;Lauritzen,2001).Stripe karstoftenhasgreatlateralextentandintersectstheland surfaceatanangle.Theinsolubleaquicludesprovideboth geometricandhydrologicconstrainttocavedevelopment. Accordingly,thecavesinstripekarstareessentiallytwodimensional.Metamorphicalterationhasremovedall originalporosityintherock,sothatallpre-karstwater circulation,andthusspeleogeneticinception,isrestricted tofractures,faults,andlithologicalcontactsdeveloped 1 DepartmentofEarthScience,UniversityofBergen,Alle gaten41,N-5007Bergen, Norway.rannveig.ovrevik@geo.uib.no 2 DepartmentofPlantandEnvironmentalSciences,TheNorwegianUniversit yof LifeSciences,1532A s,Norway.Stein.Lauritzen@geo.uib.no R..SkoglundandS.-E.Lauritzen–Subglacialmazeorigininlow-dipmarb lestripekarst:examplesfromNorway. JournalofCaveand KarstStudies, v.73,no.1,p.31–43.DOI:10.4311/jcks2009ES0108 JournalofCaveandKarstStudies, April2011 N 31

PAGE 2

underpost-metamorphic,brittleregimes.Thisgivespredictableboundaryconditionsandmakescavesinstripe karstfavorableforanalysisandmodeling. DuringtheQuaternary,repeatedglaciationsofvariable durationandextension(e.g.,Sejrupetal.,2000)shapedthe landscapeofwesternScandinavia,andthus,thestripe karst.Thekarsthydrologicalregimehasvariedaccording totheclimaticconditionsbetweentheextremesoffluvial, duringinterglacialssimilartothepresent,andsubglacial, duringglaciationswiththickcontinentalice-sheets. Theorientationsanddipsoffracturesandlithological contactsandtheconnectivitybetweenfracturesand interfacesarethemostimportantpassivefactorsfor speleogenesisinagivenrockmass.Theirintersections withthesurfacedefineinputandoutputboundariesforthe karst.Waterrechargeandwaterchemistryareactive factors.Lauritzen(2001,p.75)examinedhowpassive factorsconstraincavepatternsinstripekarstandfound that‘‘thevariousmorphotypesshowsystematicdependenceonstrataldipoftheallogenicstripecontacts,ofthe typeofcontacts,andtoalesserextent,thefracture patterns,’’wheretypeofcontactmeansaquifertype(i.e., confined,unconfined,orperched).However,thereis considerableoverlapbetweentheoccurrenceofmazecaves andothercavepatternswithrespecttotheseparameters.In otherwords,astripekarstsettingdoesnotgiveacomplete explanationofmazeevolution.Ofthepresentlyaccepted mechanismsofmazeorigin,diffuserechargeseemsquite improbableinthemetamorphicstripekarst,withits impermeablewallrocks.Sincemanysuchmazesare situatedinhangingpositionsinglacialvalleysorat summits,isolatedfromanypresent-dayfluvialdrainage, theflood-watermodelisalsonotasatisfactoryexplanation ofmazemorphologyinglacialstripekarst.Ourhypothesis isthat,inthisgeologicalsetting,aspectsofthesubglacialor ice-contactregimemayhavebeenresponsibleformazecaveevolution. Paleocurrentsinandspeleogenesisofatieredmazecave instratadippingsteeplymorethan60 u (Lauritzen,2001), Pikha ggrotteneinRana,northernNorway,werepreviouslydescribedandanalyzedbyLauritzen(1982).Inthe presentwork,wehaveexaminedagroupofmazecavesin stripekarstwithlowerdipsoflessthan30 u (Lauritzen, 2001),NonshauggrottaandadjacentcavesinGildeska l, Nordland.Thecavesarerelictandhaveahanging topographicposition,displayingaquiteuniformpassage geometryandmorphology.Thisinturnsuggeststhatthe cavepassagesmayhaveacorrespondinglyuniformhistory. Throughdetailedstudyofcave’sgeometry,passage morphology,flowfunction,andstructuraltemplate,we aimatacloserunderstandingofbothpassiveand,more important,theactivefactorsofsubglacialmazeorigin. TheseissuesarefurtherexploredbysurveyandexaminationofLnngangen,asmallcaveafewkilometersaway, andtwoothersmallcavesinNonshaugenridge. B ACKGROUND G EOMORPHOLOGICALAND G EOLOGICAL S ETTING Nonshauggrotta(66 u 57 N13 u 58 E)islocatednearthe coastnorthwestofSvartisenglacier,northernNorway (Fig.1).Theareaischaracterizedbyglaciallyincised fjordsandvalleys,surroundedbymountainswithalpine peaksreachingelevationsof800to1100m.NonshauggrottaissituatedonthenorthernsideofNonshaugen,a small,300mridgeattheendofSrfjordenfjord. Nonshaugenhastheshapeofaglacialwhaleback,witha steepleesidetothenorthandamoregentlydippingstoss sidetothesouth. Thecavernouscarbonateessentiallyconsistsofcalcite marblethatwasfoldedinseveralphasesduringthe Caledonianorogeny(Gustavson,1985;Stephensetal., 1985;GustavsonandSolli,1989).TherocksofNonshaugencontainarecumbentfold,theNonshaugenfold, thatisoverturnedtothenorthwithapredominantlyeastwesttrend(WellsandBradshaw,1970).Duringthelast phase,therocksequencewasfoldedtoformanopen antiformtrendingnorth-south,theSrfjordenfold(Wells andBradshaw,1970),sothattheNonshaugenfold plungestotheeastandwestfromthetopofthehill (Fig.2).Themarble-schistinterfaceinNonshauggrotta dipsgentlysouthward(082 u /25 u ).Althoughinclinedstripe karstofthistypemaybeconsideredcoveredkarst,the exposureintheNonshaugenwallisneverthelessastripe karst. Figure1.TheareanorthofSvartisenglacierwithlocation ofNonshauggrottaandLnngangen.Whitearrowsshowthe directionoficeflowduringmaximumglaciation.Redarrows showthedirectionofglaciermovementduringthelast deglaciation.GlacierflowsfromRasmussen(1981). S UBGLACIALMAZEORIGININLOW DIPMARBLESTRIPEKARST : EXAMPLESFROM N ORWAY 32 N JournalofCaveandKarstStudies, April2011

PAGE 3

G LACIAL H ISTORY Incoastalareas,glaciationwasnotacontinuousperiod ofglacialcover,butratherasequenceofglacialadvances ofvariousdurationsandextents(e.g.,Sejrupetal.,2000; Mangerud,2004).AccordingtoOlsen(1997),theperiods oficegrowthandicerecessionhaddurationsofonlyafew thousandyearsduringtheperiod15–40kaBP.During maximumglaciation,iceflowinGildeska lwaswestward (Rasmussen,1981)(Fig.1).Theicesheetwasthick,sothat flowwasessentiallyindependentofbedtopography. Duringdeglaciation,theicethicknesswasreducedand iceflowwasdirectedbythehighestpeaksandpartly divertedintothemainfjordsandvalleys.Duringthelatest stagesofglaciation,valleyandfjordglaciersflowedfrom localicemassesatSvartisenandGlombreen(Rasmussen, 1981).Observationsofglacialstriationsonthetopand easternsideofNonshaugenindicateaniceflowtowardthe northwest(302to316 u ),correspondingtoayoung, topographicallydirectediceflow.Therapidiceretreat afterthemaximumiceextentatabout18kaBPprobably occurredpartlythroughiceshelfbreakupandmassive icebergformationafterarapidglobalsea-levelrisestarting atabout15kaBP(Lingeetal.,2007).Srfjorden,andthus Nonshauggrotta,havebeenice-freesinceEarlyAllerd time(13.7kaBP)(Rasmussen,1981). M ETHODS NonshauggrottawasfirstsurveyedbyCorbel(1957) andlaterbyHolbyeandTrones(unpublishedplanmap), andLnngangenwassurveyedbyEikeland(1986). However,ourstudyrequiredmorecompleteandaccurate three-dimensionalcavemapswithdetailedgeological information.Bothcaveswerethereforeresurveyedto BCRAgrade5C(Day,2002),usingpassage-centerline polygons,wherewalls,ceiling,andflooraredefinedby theirdistancefromeachsurveystation.Fromsurveydata, 3-Dmodelsofthecaveswereobtainedbyusingthe Grottolfcavesurveyprogram(Lauritzen,2004).During thesurveyprocess,thepassagemorphologywasthoroughlyexamined,sketched,andphotographed.Thedistribution andapproximategrainsizeofsurfacesedimentsalongthe passagefloorwerealsorecorded.Guidingfractures(i.e., thosefracturesthathadbeenwidenedintocaveconduits bydissolutionalenlargement)werelogged,inadditionto themarble-schistinterface,foliations,andsurfacefractures.Someofthesurfacefracturewereobservedinthe overlyingmica-schistandinastratigraphicallyhigher marblelayer. Passagemorphologyreflectsthehydrologicalconditionswhentheyformed(e.g.,LauritzenandLundberg, 2000).Underphreatic(water-filled)conditions,conduits formbycorrosioninalldirectionsoutwardfromthe passagecenter.Thisresultsincircular,elliptical,or lenticularpassagecross-sections.Geologicalstructuresact aspassiveconstraintsonthepassageshape.Underopenchannel(vadose)conditions,corrosionanderosiononly attackthefloor,forminganincisedcanyon.Atransition fromphreatictovadoseconditionsnormallyresultsinthe keyholeshaped,two-stagecross-section. Scallopsarecorrosionalflowmarksinthecavewalls (Curl,1974).Theyareimprintsofthelastactivestagethat lastedlongenoughforwatertoeraseanypreviousscallop patterns.Scallopanalyseswereperformedinaccordance withtheprotocoldescribedinLauritzen(1982)andin LauritzenandLundberg(2000).Fromscallopsymmetry andwavelength,paleocurrentdirectionwasdefined.Fluid velocityand,inturn,dischargecanthenbecalculatedfrom anestimateofwatertemperature( 1 u C)andpassage dimensions.Forstatisticalconfidence,populationsofat least30well-shapedscallopsfromthesameareamustbe measuredfordischargedetermination.However,flow directioncanbededucedfromasinglewell-developed scalloporasmallgroup. Variationinmarblecarbonatecontentwasexaminedby loss-on-ignition(LOI)andacid-insoluble-residuetestsof rocksamplesthroughoutthemarblelayer,andLOIwas measuredontwosamplesofmicaschistfromtheconfining beds.DuringtheLOItestsat800 u C,CO 2 isexpelledfrom calcite(CaCO 3 )anddolomite(CaMg(CO 3 ) 2 )intherock, formingthecorrespondingoxides.Theoreticalweightloss inpurecalciteis44%andinpuredolomite48%. Carbonatesdissolvecompletelyin1Mhydrochloricacid, andtheresidualisameasureofinsolubleimpuritiespresent inthemarble. S PELEOMETRIC C HARACTERIZATION ThecavesurveyprogramGrottolfprovidesseveral geometricalparametersofthevoidthatcanbeusedforits characterization.Cavelengthisthetotallengthincluding allsurveyedpassagesegments.Cavedepthisthevertical differencebetweenthehighestandlowestelevatedpassage. Figure2.Low-dipstripekarst.AlongthenorthernclifffaceofNonshauge nthelow-dipmarble-schistsequencecropsoutand makestheopenSrfjordenantiformvisible. R..S KOGLUNDAND S.-E.L AURITZEN JournalofCaveandKarstStudies, April2011 N 33

PAGE 4

Caveareaisthetotalhorizontalplanareaofthepassage outline.Cavevolumeisthetotalvolumeofallcave passagesbasedonellipticalcross-sections.Therockarea thatenclosesthecaveisestimatedasitsconvexhull,a polygonofonlyconvexanglesenclosingallcavepassages inplanview.Therockvolumethatholdsallcaveconduits isherecalculatedastheconvexhullareamultipliedbythe maximumpassageheight.Fromthesedatathefollowing speleometricparameterscanbecalculated.Meanpassage cross-sectionalarea(cavevolumedividedbycavelength). Passagedensity(cavelengthdividedbyconvexhullarea). Arealcoverage(caveareadividedbytheconvexhullarea expressedaspercent).Caveporosity(cavevolumedivided byrockvolumeexpressedaspercent). Asameasureofhowcompletelythecavefillsaplaneof projection,weusethefractaldimensionorboxdimension, determinedbystandardbox-counting(Feder,1988).The boxdimension, D ,ofthecavecanbecalculatedas (SimancaandSutherland,2002) D lim e ? ? log N e ( S ) log e 1 where N e ( S )istheminimumnumberoftwo-dimensional squaresofside-length e neededtocoverthefilledoutlineof thecave( S ).Accordingly, D istheslopeofthelog-logplot of N e ( S )versus e .Thisapproachdiffersfromthe volumetric,modularmethodofCurl(1986),butserves ourpurposebest,becausethecaveisessentiallytwodimensionalduetothestripegeometry,anditpermits comparisonwithothercaveswherevolumetricpassage detailsarenotavailable. R ESULTS C AVE D ESCRIPTIONOF N ONSHAUGGROTTA Nonshauggrottawassurveyedtoatotalpassagelength of1.5km,atelevationsbetween231and260m(Fig.3). Thehighestpassagesarelocatedinthenortheastandalong thecliffface.Thepassagedensityisquitehigh,148km/ km 2 ,asistheplane-fillingfractaldimension, D 1.5 (Table1).Explorablecaveconduitspenetratelessthan 50mdown-dipfromthecliffface,butextendnearly300m alongit.Thesurveyedpassagescomposeonlyaminimum estimateoftheactualcaveextent.Almostallsouthtrendingconduitsterminateinsandchokes;thosethatare opencontinuebeyondexplorabledimensions.Essentially allnorth-trendingconduitsendasopeningsinthecliffface, andwereapparentlytruncatedeitherbyglacierpluckingor bygravitationalretreatofthecliffface.Therefore,the absoluteboundaryofcavepassagescannotbeestablished unambiguously,althoughourobservationssuggestthat manypassagestaperouttosmalldimensionsdown-dipto Figure3.PlanmapofNonshauggrottawithpassagecross-sections(5 3 exaggeration).A-A :Longitudinal(vertical)profileof aN-Spassage(2 3 exaggeration). 3 :Circularshaft.Siteofscallopsanalysisreferstocross-sectionnumbe r.v:calculated paleocurrentvelocity.Q:calculatedpaleodischarge. S UBGLACIALMAZEORIGININLOW DIPMARBLESTRIPEKARST : EXAMPLESFROM N ORWAY 34 N JournalofCaveandKarstStudies, April2011

PAGE 5

thesouthwhenfolloweddeeperintotherockmass.Sothe densityofcavevoidsappearstodecreaseinthedown-dip direction. StructuralSpeleology Thecavesystemdisplaysareticulatednetworkarchitecture(Fig.3).Caveconduitsareorientedparallelwith thestrikeanddipofthemarble-schistinterface(082 u /25 u ), thusintersectingatnearlyrightangles.Thepredominate setofguidingfracturesissteeplydipping,strikingsouthsouthwest(190 u /79)(Fig.4).Theothersetofguiding fractureshasamoderatedipandstrikeswest-northwest (282 u /49 u ).Onlythepredominatefracturesetcouldbe detectedonthelandsurfaceabovethecave(Fig.4).We havenotbeenabletodetectanydisplacementalong fracturesorfoundotherbrittlekinematicindicators.At severallocations,thewest-northwestfracturesterminateat thesouth-southwestfractures,thoughthisrelationshipwas notestablishedunambiguously.However,ifso,itimplies thatthesouth-southwest-trendingfracturesareolderthan thewest-northwest-trendingfractures.Ifnot,thetwo fracturesetsmighthavebeenformedsimultaneously. Morphology Thepassagecross-sectionsarehighandnarrow,with ellipticalorirregularriftshapesreflectingtheguiding fractures(Fig.3).Thepassagecross-sectionsarerelatedto theslopeoftheguidingfractures;southtrendingpassages haveverticalcross-sections,whereaswest-trendingpassageshaveslantingcross-sections(Fig.5).Meanpassage cross-sectionalarea,approximatedasellipses,isestimated asabout1.5m 2 .Thelargestmeasuredcross-sectionalarea isabout16m 2 andisaresultofbreakdownmodification. Deadend,tightfissuresarecommon,whilenarrowvadose channelincisionsarerareand,evenwhenapparent, ambiguous.Wherepassagesterminateinthecliffface andtheirentirecross-sectionisvisible,theylackvadose incisions.Insum,thecaveappearsessentiallyphreatic, withnegligible,ifany,vadosemodification.Oneofthe southernconduitsendsinacircularshaft2mdeepplugged withsandandmicaceoussilt(Fig.3). Mica-SchistCapRock Themica-schistceilingformsahydrologicalconstraint thatrendersthecavesystemessentiallytwo-dimensional. Passagesweredevelopedalongandjustbelowtheupper marble-schistcontact(i.e.,inhydraulicallyconfined settings)(Fig.3). Corrosionaldrip-pitsinthemarbleunderlyingthe schistceilingdemonstratethatseepagefromtheschistis occasionallyacidic.Substantialefflorescenceofgypsumon cavewallsandrustyweatheringofthemicaschistreveal thatironoxidesandsulfidesarepresentinthemicaschist, thesulfideshavingproducedsulphuricacidbyoxidation. Lithology Theresultsfromtheloss-on-ignitionandacid-insoluble-residuetestsareinverselyproportional(r 0.99), andconsistent.Theinsolubleresidualswereintherange1 to21%(Fig.6),whichindicatethatthepurityofthe marbleisvariable,withbetween79and99%CaCO 3 and CaMg(CO 3 ) 2 .However,thecompositionoftheupper, Table1.Speleometricdata. ParameterNonshauggrottaLnngangenUpperNonshauggrotta Length,10 3 m1.50.300.23 Depth,m291236 Cavearea,10 3 m 2 1.60.40.3 Cavevolume,10 3 m 3 2.30.80.4 Convexhull,10 3 m 2 10.24.20.9 Rockvolume,10 3 m 3 92256 Meanpassagecross-sectionalarea,m 2 1.52.81.9 Passagedensity,km/km 2 14870256 Arealcoverage,%16936 Caveporosity,%2.53.27.0 Fractaldimension1.51.21.1 Figure4.Left:Stereographicprojection(equalarea,lower hemisphere,magneticnorth)ofpolestofoliationand fracturesatNonshaugen.Right:Rosediagrams(equalarea, magneticnorth)oftrendofsurfacefracturesandguiding fractures.Circle:10% % R..S KOGLUNDAND S.-E.L AURITZEN JournalofCaveandKarstStudies, April2011 N 35

PAGE 6

cavernouspartofthemarblelayer,wherecaveconduits occur,doesnotdifferfromtherestofthemarblelayer eitherinloss-on-ignition( t -test, t 17 0.67, p 0.51)or acidinsolubleresidual( t -test, t 17 0.55, p 0.59). Consequently,variationinmarblepurityfailstoexplain thestratigraphicpositionofthecaveconduitsadjacentto theconfiningstratainthesequence.Theabsolutecontent ofinsolubleresidueisquitehigh,buthasnotchokedoff speleogenesis,probablybecausemicaandquartzdonot swellincontactwithwater. Sediments Thepassagefloorconsistsessentiallyoffinetocoarse sand(Fig.3).Micaisabundantandseemstodominatethe coarsersandfractions.Someofthenorthernpassages containpebblesandcobbles.Angularbreakdownoccursin afewofthelargestpassages.Someofthenorth-trending passagesarepluggedbybreakdownandscreematerialthat penetratedfromthecliffface.Novarved,clayeysediments haveyetbeenfoundinthecave,andthelayersofsandand gravelareprobablyafluviallag. Paleohydrology Nonshauggrottaisrelictanddryandlacksamodern drainageareaduetoitstopographicpositioninthe northernclifffaceofNonshaugen,intheleesideofthe glacialwhaleback.Scallopsinconduitwallsandceiling displayaconsistentflowpattern,eastwardineast-west trendingpassagesandnorthwardinnorth-southtrending passages(Fig.3).Accordingly,waterflowwasupward, artesian,andthehydraulicgradientwasdirectedtoward thenortheast.Hencethenetworkhadaneffluentflow underphreaticconditions;itrepresentedthedischarge boundaryofthekarstaquifer.TheSautermeanofscallop length, L 32 ,wasdeducedfromscalloppopulationsin threedifferentsites(Fig.3),32cmatonesiteand20cm attwoothersites.Calculatedpaleoflowvelocitieswerein therangeof7to16cms 1 ,whilepaleodischargeswerein therangeof0.02to0.5m 3 s 1 .Althoughonlyafewsites weresuitableforstatisticalanalysisofscallops,numerous otherobservationsindicatethatscalloplengthsthroughoutthenetworkarewithinthesamesizerange.Wewere unabletodetectanyvariationinmeanscalloplengthwith heightabovethepassagefloor. Figure5.PassagesofNonshauggrottawithdistinctguidingfractures.Le ft:south-trendingpassagewithsubverticalguiding fracture.Right:west-trendingpassagewithmoderatelydippingguiding fracture,micaschistceiling,efflorescencesofgypsum andhydromagnesite,rustycoloring,andfinesedimentsonlowerwallandf loor. S UBGLACIALMAZEORIGININLOW DIPMARBLESTRIPEKARST : EXAMPLESFROM N ORWAY 36 N JournalofCaveandKarstStudies, April2011

PAGE 7

O THER C AVESAND K ARST F EATURESIN N ONSHAUGEN UpperNonshauggrottaissituatedabout100msouthof Nonshauggrottainastratigraphicallyhighermarblelayer atanelevationofabout285m,butinasimilarposition belowanuppermarble-schistinterface(Fig.7).Thecave wassurveyedtoalengthof175mandadepthof36m.The lowerpartofthecaveisashaftslantingdownwardat about36 u andtrendingsouth,endinginabreakdown chokeatthebottomofa3mverticaldrop(Fig.7).Inits upperpart,thecaveconsistsoftwostoriesandsome branchesandloops.Morphologyandscallopsindicatethat paleoflowwasslowanduphill. Asmallcaveissituatedatthesouthernsideof Nonshaugen,atanelevationofapproximately140m,here termedSmallCave(Fig.7).Thecaveisasinglepassage trendingnorthandpluggedbysedimentsafterabout30m. Thispassageismainlyinschistwithanangularand irregularcross-section,indicatingthatithasmigrated upwardbybreakdownfromthedissolutionalconduitin theunderlyingmarbleunit. Fourdolineswerenotedintheuppermarblelayerat thetopofthewesternsideofNonshaugenridge(Fig.7). Threesmalldolines(2to5m 2 incross-sectionalareaand 0.6to1.2mdeep)aresituatedinaroworientedeast-west. Southofthesedolinesliesasinglelargerdoline,about8m indiameterand3.7mdeep. L NNGANGEN Lnngangen(67 u 00 N13 u 57 E)issituatedonthe peninsulawestofSrfjorden(Fig.1).Thecaveislocated onthewesternlimboftheSrfjordenfold.Accordingly, themarble-schistinterfacedipsgentlytowardthewestsouthwest(154 u /11 u ). Lnngangenwassurveyedtoalengthofabout300m. Thecavehasfouropeningsinaclifffacelessthan10m high,atelevationsbetween60and70m.Inplanview,the cavehasanangularpattern(Fig.8).Comparedto Nonshauggrotta,thecavehasalowpassagedensityof 70km/km 2 andalowfractaldimensionof1.2(Table1). Thepassagemorphologyandgeometryaredictatedbytwo orthogonalsetsofsteeplydippingguidingfractures.The guidingfracturesstrikesouthwest(216 u /75 u )andnorthwest (315 u /74 u )(Fig.9).Passagesareessentiallyhighand narrowriftswithameancross-sectionalareaof2.8m 2 in Figure6.Loss-on-ignitionandacid-insoluble-residuefrom 19marblesamplesand2mica-schistsamples(onlyLOI) fromthemarblelayerinwhichNonshauggrottaissituated (cavepositionmarkedinblack). Figure7.Upper:AerialphotographofNonshaugenwith caves.A-A :Lineofcross-section.Lower:VerticalcrosssectionofNonshaugen.Marble-schistcontactsinthemiddle oftheridgearehypothesized.Redarrows:paleoflow directionfromscallops.Greyarrows:Topographically directediceflow(inaccordancewithRasmussen1981). R..S KOGLUNDAND S.-E.L AURITZEN JournalofCaveandKarstStudies, April2011 N 37

PAGE 8

ellipticalapproximation(Fig.8).Afewsmall,sub-circular sideconduitsintersectthemainpassageandformclosed loops.Negligiblevadosemodificationisseen,andmarble breakdownoccursinonlyafewplaces,mostprominently bytheeasternentrance.Sandysedimentscomposethe passagefloor,andshellsandplugstheinnermostconduit. Sub-roundedgraveloccursinafewplaces(Fig.8).The mica-schistcaprockfrequentlyformstheceilingofthe passages(Fig.8).Corrosionaldrip-pitsarefoundon marblebouldersbelowmica-schistexposures. Thepaleoflowdirectiondeducedfromscallopsisfrom easttonorthwest(Fig.8).Theeasternentrancesconveyed waterintotherock,whilewateremergedfromthe northwesternentrances(Fig.10).Accordingly,thecave containsacompleteflowroutebetweeninputandoutput boundariesatthetopographicsurface. D ISCUSSION S TRUCTURAL G UIDING NonshauggrottaandLnngangenareeachguidedby orthogonalfracturesets.However,thefracturesetsofthe twocaveshavedifferentorientations. InthemapoftectoniclineamentsofNorway(Gabrielsenetal.,2002),theregionaltrendsofNordlandand Tromsarenortheast-southwestandnorthwest-southeast. Onthemap,lineamentstrendingnorth-northeast–southsouthwestandwest-northwest–east-southeastareabundantintheGildeska larea,whereaslineamentstrending north-south,east-west,andnortheast-southwestarepresentbutsparse.Northtonorth-northeastisalsothe regionalCaledonianstrikedirectionandtheaxialtrend oftheSrfjordenantiform. ThetwosetsofguidingfracturesofNonshauggrottaare paralleltothepronouncedtrendsoftectoniclineamentsin theGildeska larea.InNonshaugen,thesouth-tosouthsouthwest-strikingfracturesetisdominant.Thisfracture setguidesthepassagetrendsintheNonshaugencavesand isthesinglepronouncedfracturesetonthesurfaceabove thecave.Thewest-northwest-strikingfractureswereonly detectedwithinNonshauggrotta.Weassumethatby loggingsurfacefracturesinthevicinityofthecave,the truefracturedistributionisidentified.However,herethe attitudeofthewest-northwest-strikingfracturesetmay havemadeitlesspronouncedinsurfaceoutcrops. Thesouthwest-andnorthwest-strikingguidingfracturesinLnngangenareconsistentwithsurfacefractures (Fig.9),locallineamentsdeterminedfromaerialphotos (Fig.10),andtheregionallineamenttrendsinNordland andTroms(Gabrielsenetal.,2002).Ontheotherhand, theydifferfromthesouth-southwestandwest-northwest Figure8.PlanmapofLnngangenwithpassagecross-sections(3 3 exaggeration). Figure9.Left:Stereographicprojection(equalarea,lower hemisphere,magneticnorth)ofpolestofoliationand fracturesfromLnngangen.Right:Rosediagrams(equal area,magneticnorth)oftrendofsurfacefracturesand guidingfractures.Circle:20% % S UBGLACIALMAZEORIGININLOW DIPMARBLESTRIPEKARST : EXAMPLESFROM N ORWAY 38 N JournalofCaveandKarstStudies, April2011

PAGE 9

strikeofguidingfracturesfoundinNonshauggrotta.This pronounceddifferenceinfracturesetsmayberelatedtothe locationofLnngangenatthelimbofthefoldofthe Srfjordenantiform,whereasNonshaugenissituatedin thehingeofthefold. Thefracturespredatethespeleogenesisinthemetacarbonates.Otherwise,thereisnounequivocalevidenceof chronologicalrelationshipsbetweenthefracturesets.Since theDevonian,theregionaltectonicstressfieldhasvaried significantly(Berghetal.,2007),anditdoesnotseem prudent,onthebasisofthislocality,toassociatethelocal fractureswithspecificphasesoftheregionalgeological history. S TRATIGRAPHIC C AVE P OSITION Thelackofcorrelationbetweenmarblepurityand concentrationofcavevoidsinNonshauggrottasuggests thatthestratigraphicpositionofthecaveiscontrolledby hydrologicalfactorsratherthanthedistributionof solubility.Itisconceivablethatthepositionofthecaveis aresultofartesianflowtowardsalow-dipconfining contact. Occurrenceofironoxidesandpyriteinthemicaschist suggeststhatcaveinceptionmayhavebeenaidedby sulfuric-acidspeleogenesisatthemarble-schistinterface. P ALEOHYDRAULIC F UNCTION FlowDirectionandFlowComponents Allinterpretationsofpaleocurrentsarebasedonscallop morphometryandsurfacesedimentsatthepassagefloor, allofwhichareremnantsofthelastactivestage.Our interpretations,therefore,primarilyconcernthelaststages ofevolution. InNonshauggrotta,thepassagecross-sectionsindicate symmetricaldissolutionoutwardfromtheguidingfracturesunderphreaticconditions.Artesianwaterflow, deducedfromscallops,isconsistentwithevolutionunder confinedsettings.Inaccordancewiththepaleocurrent direction,thesouthernverticalconduit(Fig.3)isinterpretedasaphreaticartesianfeeder.Sinceseveralpassages seemtotaperoutsouthwardsintotherockmass,we suggestthatonlyafewrisertubesfedthesystemfrom below.UpperNonshauggrottadisplaysthesameflow functionandconduitpatternasNonshauggrotta,feeder tubesanddistributarynetworks,althoughmorefragmentaryduetoitssmallersize(Fig.8).Thissupportsthe interpretationofartesianwaterflowfromthesouth throughaneffluentflownetwork.Thelow-dip,confining aquicludemayhavecausedaspreadingandslight retardationoftheflow,andthus,enhancedmaze formation.TherudimentarynetworkevolutioninUpper Nonshauggrottamaybepartlyattributedtothesteeperdip ofthemarble-schistinterfaceexceptintheupper, outermostpart. ThestratigraphicpositionofSmallCaveatthe southernsideofNonshaugenmayalignwitheither NonshauggrottaorUpperNonshauggrotta.Itmight thereforebeanexampleofaninfluxconduitfeedingthe cavenetworkfromthesouthern,stosssideoftheglacial whaleback. Thededucedhydraulicgradientdirectiontowardthe northeastandtheconsistentflowdirectiontowardstheeast inthewesternpartofthecaveandineast-westpassages togetherimplythattheremighthavebeenadditional rechargefromthewesternorsouthwesternsideofthe ridge,givingaflowcomponentsub-parallelwiththecliff face.Thismaybeconsideredasasortofflankmargin subglacialpatternofgenesisorperhapsflankmargin adaptationofacoveredorconfinednetworkofconduits. However,thereisnosignofsolutionsculpturebyflowinto oroutofthecaveinthenortherncliff-face. SlowPaleocurrent AsscalloplengthsseemtobequiteconsistentthroughoutNonshauggrotta,themeanpaleoflowvelocityinthe cavesystemasawholeissimilartothespotobservationsof calculatedvelocities(afewtensofcms 1 ).Slowflow conditionswerealsoobservedinUpperNonshauggrotta. Lowflowvelocitiesarealsoconsistentwiththewidespread depositsoffine-grainedsediments.Accordingtothe Hjulstrmdiagram(Sundborg,1956),finesandisdepositedatvelocitiesoflessthan20cms 1 ,whilecoarsesandis Figure10.Upper:Aerialphotographoftheareaaround Lnngangen.A-B:Lineofcross-section.Lower:Vertical cross-sectionofLnngangenwithtopography.Black:cave passages.Redarrows:paleowaterflowdirectionfrom scallops.Greyarrows:Topographicallydirectediceflow (inaccordancewithRasmussen,1981). R..S KOGLUNDAND S.-E.L AURITZEN JournalofCaveandKarstStudies, April2011 N 39

PAGE 10

depositedatvelocitieslessthan40cmsec 1 .Discharge determinedfromscallopshasbeenshowntorepresent dischargeswithinthehighest2to15%oftheannualflows (Lauritzen,1989;LauritzenandLundberg,2000).The conspicuouslackofcoarsefluviallagmaterialfurther supportstheimpressionthatlowvelocitieshavedominated thenetworkflow. S PELEOGENETIC S ETTING Basedonthecavegeometry,passagemorphology, hydraulicfunction,andtopographicallyandhydrologically hangingposition,severaldifferentoriginsofNonshauggrottamaybeconsidered:(1)hypogeneorigin,wherethe aquiferisfedfrombelowandwithoutrelationtosurface drainage,(2)post-glacialorlateinterglacialorigin,ina topographicalpositionsimilartothepresent,(3)pre-glacial orearlyinterglacialorigin,asaflood-watermazeinthe bottomofafluvialvalley;and(4)subglacialorigin,ina topographicalpositionsimilartothepresentandincontact withaglacialaquiferaspartofasubglacialdrainage network. Thenetworkgeometryandtheeffluentflowfunction aresuggestiveofcavesofhypogeneorigin(Klimchouk, 2003,2009).However,thissortofhypogenespeleogenesis ischaracterizedbywaterinjectionfromaporousaquifer beloworabovethesolublestrata,andthepresenceof impermeablestratainstripekarstmakesahypogeneorigin implausible.Incontrasttoisolatedmazecaves,whichare typicallyformedinconfinedaquifersbyalocallygenerated boostofthegroundwateraggressivity(Frumkinand Fischhendler,2005),thenetworkofNonshauggrottais anintegratedcavesystemthatwasformedbywaterthat wasalreadyaggressivewheninjectedintothemarble. Pyriteorsulfideoxidationcouldhavebeenasourceof locallygeneratedaggressivenesswhenascendinggroundwatermixedwithoxygenatedsurfacewater.Forpyrite,the totalreactionis 4FeS 2 16CaCO 3 15O 2 14H 2 O 16Ca 2 16HCO 3 8SO 2 4 4FeOH 3 2 Oneformulaweightunitofpyritecandissolvefour formulaweightsofcalcite,orbyvolume1mm 3 pyrite woulddissolve6.4mm 3 marbleandleaveaconsiderable amountofinsolublehydratedironoxide.Theiron hydroxidewillactasasurfaceinhibitor.Ifspeleogenesis wasbasedonpyriteoxidationalone,thedissolutionofthe marblewouldhaverequired16%ofitsvolumeinpyrite. Therefore,itisevidentthatthemodestsulfideconcentration(nopyritegrainshavebeendirectlyobserved)inthe uppermica-schistlayerisnotenoughtoaccountforthe totalvolumeofthecave,sothatthesulfuric-acid mechanismcouldonlyhavebeenimportantatthecaveinceptionstage.Itis,however,quiteconceivablethat oxygenatedwater,eitherdirectlyfromthesurfaceorfrom theglacierenvironment,combinedwithsomesulfide, mighthavebeencrucialinestablishingintegratedflowat theverycommencementofspeleogenesis. PhreaticformationofthecavesinNonshaugenrequires thatthewatertablewasabovethetopofthecaves.Two differentsettingshaveprovidedhighwatertablesinthe past:bedrock,beforethevalleywasentrenched,and glaciers,whentheyfilledthevalleyandraisedthewater tableintheadjacentrocks.ThecavesofNonshaugenand Lnngangenlackedasurfacedrainageareaduringpast andpresentinterglacialsduetotheirhangingpositionin thecliffface.Consequently,post-glacialorlateinterglacial originofthecavesisrejected. Inpreviousworks,valleyerosionratesbyglaciershave beenestimatedtorangefrom15to55mper100ka (Lauritzen,1990,andreferencestherein;Nesjeand Sulebak,1994).Thisimpliesaminimumof500kasince Nonshauggrottamayhavebeensituatedbelowthevalley floor.Smallpassagecross-sectionalareasindicateashort evolutiontimebecause‘‘afterapassagereachesthe maximumannualenlargementrate,itssizedependsmainly ofthelengthoftimeitcontainsflowingwater’’(Palmer, 1991,p.10).Undersubglacialsettingsthecaveconduits tendtobewater-filled,andactivewatercirculation,and thusdissolutionalwidening,occursatleastoccasionally (Ford,1977;Lauritzen,2006).Smallpassagedimensions areinconsistentwithevolutioninavalleybottomduring fluvialregimesandrepeatedsubglacialconditions,andpreglacialorearlyinterglacialevolutionofNonshauggrottaas afloodwatermazeinthevalleybottomisrather improbable. Themostprobablesettingforartesianconditionsinthis topographicandgeologicalenvironmentissubglacial. Subglacialandenglacialwaterdrainageiscontrolledby thesurfaceslopeoftheglacierandisessentiallyindependentofbedtopography.Waterflowtowardthenortheast inNonshauggrottaandtowardthenorthinUpper Nonshauggrottaisconsistentwithobservedglacialstriationstowardthenorthwestontheeasternsideoftheridge andwithyounger,topographicallydirectedglacierflow (Rasmussen,1981)(Fig.7).Whendirectedbythetopography,theglacierwouldflowoverthesummitandaround Nonshaugen,sothatboththesouthernandwesternsideof theridgehavebeensubjecttohighpressure.Effluentflow isconsistentwiththenetworkoccurringattheleesideof theridge.Theconsistentlyslowflowratesobservedinthe Nonshaugencavesmaybedueeithertogreatglacier thicknesswithagentlesurfaceslopeorsomethingelsethat dampedthestrongfluctuationsofglacialflow. Theobservedflowdirectioninthecavesisperpendiculartothedirectionofglaciermovementduringfull glaciation.Agentleglaciersurfaceslopeparallelwiththe observeddirectionofwaterflowisthereforemorelikelyto correspondtoearlyperiodsofdeglaciation,whenthe mountainsintheweststartedtocontroltheglacierflow andturneditnorthward.Duringdeglaciationitcanbe assumedthathighflowratesoccurredduringthemelting S UBGLACIALMAZEORIGININLOW DIPMARBLESTRIPEKARST : EXAMPLESFROM N ORWAY 40 N JournalofCaveandKarstStudies, April2011

PAGE 11

season.Thelackofsignsorremnantsofhighflowratesor flushing,suchascoarse-grainedlag,suggestthatthecave wasnotexposedtothenaturalfluctuationsofmeltwater discharge.Accordingly,wesuggestthatsomecontrolofthe flowoccurredinthekarstsystemorinthecontactbetween thekarstsystemandtheglacier.Aglacierisadynamic aquiferwithanephemeraldrainagepatternanddischarge, andthemazewasestablishedintheleesideofthehill. Accordingly,wesuggestthatrestrictedoutflowtowardsthe icecontacthascausednetworkevolutionintheoutflow zone.ModelingexperimentsbySkoglundetal.(2010) supportthisinterpretation. S PELEOGENESISOF L NNGANGEN TheunderlyingcavepatternofLnngangencannotbe decidedunequivocallyduetoitssmallsizeandfragmentary form.Wesuggestthatitisarudimentaryordegenerated networkcave;iteitherneverdevelopedintoaproper networkorhasbeentruncatedandpartlydestroyedby glaciers.WaterflowinLnngangenwasparalleltothecliff faceandthecap-rockcontact,nottowardthem(Fig.10). ThiscontrastswiththesettingofNonshaugen,andthelack ofoutflowrestrictionisapossibleexplanationofthe rudimentarynetworkpattern. ThepassagemorphologyinLnngangenalsoindicates symmetricaldissolutionaroundsteeplydippingguiding fracturesunderphreaticconditions.Lnngangenwas situatedbelowsealevelforabout4kaafterthelast deglaciation(ML89ma.s.l.,Rasmussen,1981),sincesea leveldidnotdropbelowthecaveleveluntiltheBoreal. However,thephreaticpassagemorphology,thesmall passagecross-sections,andthecaveÂ’slocationintheseafacingcliffindicatethatthecaveisquiteyoungand,in accordancewiththepreviousdiscussionofNonshauggrotta,hasprobablydevelopedundersubglacialconditions.Thepaleocurrentdirectiondeducedfromscallopsin Lnngangenisnotcompatiblewithsubmarineorshoreline erosion.Paleocurrentdirectionisconsistentwithglacier flowduringbothfullglaciationanddeglaciation(Figs.1 and10).However,itisconceivablethattherewasless wateravailablebelowtheglacierduringfullglaciation,and accordingly,thatwatercirculationanddissolutionwere moreefficientduringglacialretreatandadvance. C ONDITIONSOF S UBGLACIAL S PELEOGENESIS Thehomogeneouscharacterandlackofmodifications ofthepassagemorphologyinNonshauggrottasuggestthat evolutioneitherhasbeencontinuousorthecaveevolved throughaseriesofstageswithaquitesimilarhydrological regime.Glaciationsarecharacterizedbystrongfluctuationsinhydrology,bothonashorttimescale(annually) andonalongtimescale(glacial/interglacialcycles). Accordingly,activesubglacialsolutionalwideningofkarst aquifersismostlikelydisruptedbyperiodsofstagnation, possiblesiltingup,anddrainedperiodsduringinterstadials andinterglacials(Ford,1977). WemayestimatetheageofNonshauggrottaby consideringthetimerequiredtoformthepassagesunder differentwall-retreatrates.Inanallogenicstreaminacave southofSvartisenglacier,thedissolutionrateisestimated at0.2mm/ainanunpublishedstudybyLauritzen,soa scallopreliefofabout2cmwouldrequireabout100years underinterglacialconditionstobeestablishedorreplacea previouspattern.Fromcorrosionofaninterglacial speleothem,Lauritzen(1990)reportedatotalsubglacial solutionalwallretreatrateof5to10cm/100ka, correspondingtoanaveragerateofabout0.001mm/a. Atthisrateitwouldtake20katoformthescallop,which isagrossoverestimate,asthecorrosionwasmostlikely discontinuous. IfNonshauggrottawaswidenedfromtheproto-cave stage( 1cm,FordandWilliams,2007)tothepresent conduitsize(meanwidth 1.2m,maxdissolvedwidth 2.0m)duringthelastglaciation,thiswouldhaverequireda meansubglacialwall-retreatratebetween0.006and 0.01mm/a.Thisisabout10timeshigherthanthesubglacial ratemeasuredbydirectradiometricdating.Basedonthis estimate,nounequivocalstatementcanbemadeonwhether thecaveevolvedduringthelastorseveralofthelatest glaciations.Evolutionofthecaveduringtheabout4kaof thelastdeglaciation,whenlargeamountsofwaterwere available,wouldhaverequiredawall-retreatrateofabout 0.2mm/a.Thisrateisequaltothepresentwallretreatrate byanallogenicstreamand200timeshigherthanthe subglacialrateofLauritzen(1990).Thatwouldbean improbabledissolutionrateforsubglacialwater,whichis characterizedbylowaggressiveness(Tranteretal.,1993),in themoderatequantitiespermittedbytheslowflowvelocity observed.Accordingly,wemaystatethatthecaveis certainlyolderthanthelastdeglaciation. C ONCLUSIONS Nonshauggrottaisanetworkcavesituatedina topographicallyandhydrologicallyhangingposition.It wasdevelopedunderartesianwaterflowtowardsthe confiningmica-schistcaprock.Itsstratigraphicposition justbelowtheconfiningcap-rockwascontrolledbythe contact,asthereisnocorrelationbetweenpurityofthe marbleandthelocationofthepassages. NonshauggrottaandLnngangenarebothdeveloped alongorthogonalfracturesets.Wesuggestthatthe differentorientationsoftheguidingfracturesinthetwo cavesisaresultoflocalstructure;Nonshaugenislocatedin thehingeareaandLnngangeninthefoldlimbofthe Srfjordenantiform. ThenetworkofNonshaugenridgehadaneffluentflow function,andwesuggestthatthenetworkwasfedfromthe southbyafewfeedertubes.ThisissupportedbythecavesÂ’ architectureandthepaleoflowpatternofUpperNonshauggrotta.However,theconsistentflowdirection towardthenortheastinNonshauggrottaindicatesan R..S KOGLUNDAND S.-E.L AURITZEN JournalofCaveandKarstStudies, April2011 N 41

PAGE 12

additionalflowcomponentalongthecliffface.Accordingly,thenetworkwasprobablyfedfromthesouthwestern orwesternsideoftheridgeaswell. Lnngangenisashortcavewithanangularpassage patternandseemstohavedevelopedbyphreaticwater flowparalleltothecliffface.Thecavedoesnotdisplaya propermazepattern. TheflowfunctionsofUpperNonshauggrottaand Lnngangenmayservetodemonstratethetwosuggested rechargecomponentsofNonshauggrottaseparately.Upper Nonshauggrottadisplaysasinglenorth-trendingartesian feedertubewitharudimentarylabyrinthstructuretowards theclifffacethatwasdevelopedinaconfinedsetting. Lnngangendisplaysarudimentarynetworkdeveloped alonganescarpmentwithbothrechargeconduitsand outflowconduits,butfewclosedloops.Wemayspeculate thatacombinationoftwoflowcomponentsmighthave createdtheelaboratelabyrinthofNonshauggrotta. Allobservationssupportthehypothesisthatthemaze cavesweredevelopedundersubglacialconditionsrelatedto wet-based,topographicallydirectedglacierflow.The consistentlyslowwaterflowintheNonshaugencaves impliedbyscallopmorphometryandsedimentdistribution makesussuggestthattheice-surfaceslopewasverygentle, thattheicecontactrestrictedoutflow,orboth. A CKNOWLEDGEMENTS TheworkofthispaperwasfundedbytheResearch CouncilofNorway,grantNo.160232/V30‘‘Porosity developmentinmarblestripekarst.’’DavidSt.Pierre, HelgeSkoglund,TerjeSolbakk,andArnfinnJonsenare thankedfortheirassistanceduringcavesurveys.Elsa Norumisthankedforherhospitalityduringfieldwork. WalterWheelerandDerekFordarethankedforconstructivecommentsthatimprovedthemanuscript.Statens KartverkprovidedtheaerialphotosinFig.7and10. R EFERENCES Bergh,S.G.,Eig,K.,Klvjan,O.S.,Henningsen,T.,Olesen,O.,and Hansen,J.-A.,2007,TheLofoten-Vestera lencontinentalmargin:a multiphaseMesozoic-Palaeogeneriftedshelfasshownbyoffshoreonshorebrittlefault-fractureanalysis:NorwegianJournalofGeology, v.87,p.29–58. Corbel,J.,1957,Leskarstsdunord-ouestdel’Europeetdequelques re gionsdecomparaison:Institutdese tudesrhodaniennesdel’UniversitedeLyon,Me moiresetDocuments12,541p. Curl,R.L.,1974,Deducingflowvelocityincaveconduitsfromscallops: NSSBulletin,v.36,no.2,p.1–5. Curl,R.L.,1986,Fractaldimensionsandgeometriesofcaves:MathematicalGeology,v.18,p.765–783. Day,A.,2002,CaveSurveying:Buxton,BritishCaveResearch Association,CaveStudiesSeries11,40p. Eikeland,R.,1986,Lnngangen:NorskGrotteblad,v.16,p.25–26. Feder,J.,1988,Fractals:NewYork,PlenumPress,PhysicsofSolidsand Liquidsseries,283p. Ford,D.C.,1977,KarstandglaciationinCanada, in Proceedings,7 th InternationalSpeleologicalCongress:Sheffield,BritishCaveResearc h Association,p.188–189. Ford,D.C.,andWilliams,P.,2007,Karsthydrogeologyand geomorphology,2 nd edition:Chichester,JohnWileyandSonsLtd, 562p. Frumkin,A.,andFischhendler,I.,2005,Morphometryanddistribution ofisolatedcavesasaguideforphreaticandconfinedpaleohydrologicalconditions:Geomorphology,v.67,p.457–471. Gabrielsen,R.H.,Braathen,A.,Dehls,J.,andRoberts,D.,2002,Tectoni c lineamentsofNorway:NorwegianJournalofGeology,v.82,no.3, p.153–174. Gustavson,M.,1985,Glomfjord,forelpigberggrunnsgeologiskkart: NorgesGeologiskeUnderskelse,1928I,1:50000. Gustavson,M.,andSolli,A.,1989,BerggrunnskartGildeska l:Norges GeologiskeUnderskelse,1929II,1:50000. Horn,G.,1937,U ¨ bereinigeKarstho ¨hleninNorwegen:Mitteilungenfu ¨r Ho ¨hlenundKarstforschung,p.1–15. Klimchouk,A.B.,2003,Conceptualisationofspeleogenesisinmulti-sto rey artesiansystems:amodeloftransversespeleogenesis:Speleogenesis andEvolutionofKarstAquifers,v.1,no.2,18p.[Availableathttp:// speleogenesis.info/pdf/SG2/SG2_artId24.pdf.] Klimchouk,A.B.,2009,Principalfeaturesofhypogenespeleogenesis, in Klimchouk,A.B.,andFord,D.C.,eds.,HypogeneSpeleogenesisand KarstHydrogeologyofArtesianBasins:Simferopol,Ukrainian InstituteofSpeleologyandKarstologySpecialPaper1,p.7–16. Lauritzen,S.E.,1982,ThepaleocurrentsandmorphologyofPikha ggrottene,Svartisen,NorthNorway:NorskGeografiskTidsskrift, v.36,p.183–209. Lauritzen,S.E.,1989,Scallopdominantdischarge, in Proceedings,10 th InternationalSpeleologicalCongress,v.1:Budapest,Hungarian SpeleologicalSociety,p.123–124. Lauritzen,S.E.,1990,TertiarycavesinNorway:amatterofreliefandsiz e: CaveScience,v.17,no.1,p.31–37. Lauritzen,S.E.,2001,MarblestripekarstoftheScandinavianCaledonides:Anend-memberinthecontactkarstspectrum:ActaCarsologica,v.30,no.2,p.47–79. Lauritzen,S.E.,2004,GrottolfProgramforprocessing,plottingand analysisofcavesurveydata,UniversityofBergen. Lauritzen,S.E.,2006,CavesandspeleogenesisatBlomstrandsya, Kongsfjord,W.Spitsbergen:InternationalJournalofSpeleology, v.35,no.1,p.37–58. Lauritzen,S.E.,andLundberg,J.,2000,Solutionalanderosional morphologyofcaves, in Klimchouk,A.B.,Ford,D.C.,Palmer, A.N.,andDreybrodt,W.,eds.,Speleogenesis:EvolutionofKarst Aquifers:Huntsville,Alabama,NationalSpeleologicalSociety, p.406–426. Linge,H.,Olsen,L.,Brook,E.J.,Darter,J.R.,Mickelson,D.M., Raisbeck,G.M.,andYiou,F.,2007,Cosmogenicnuclidesurface exposureagesfromNordland,northernNorway:implicationsfor deglaciationinacoasttoinlandtransect:NorwegianJournalof Geology,v.87,p.269–280. Loucks,R.G.,1999,Paleocavecarbonatereservoirs:Origins,burial-de pth modifications,spatialcomplexity,andreservoirimplications:AmericanAssociationofPetroleumGeologistsBulletin,v.83,p.1795– 1834. Mangerud,J.,2004,IcesheetlimitsonNorwayandtheNorwegian continentalshelf, in Ehlers,J.,andGibbard,P.,eds.,Quaternary Glaciations-ExtentandChronology:Part1Europe:Amsterdam, Elsevier,DevelopmentsinQuaternaryScienceseries2,p.271–294. Nesje,A.,andSulebak,J.R.,1994,QuantificationoflateCenozoic erosionanddenudationintheSognefjorddrainagebasin,western Norway:NorskGeografiskTidsskrift,v.48,p.85–92. Olsen,L.,1997,Rapidshiftsinglacialextensioncharacterizeanew conceptualmodelforglacialvariationsduringtheMidandLate WeichelianinNorway:NorgesGeologiskeUnderskelseBulletin, no.433,p.54–55. Palmer,A.N.,1975,Theoriginofmazecaves:NationalSpeleological SocietyBulletin,v.37,no.3,p.56–76. Palmer,A.N.,1991,Originandmorphologyoflimestonecaves: GeologicalSocietyofAmericanBulletin,v.103,p.1–21. Palmer,A.N.,2001,Dynamicsofcavedevelopmentbyallogenicwater: ActaCarsologica,v.30,no.2,p.13–32. Palmer,A.N.,2002,Speleogenesisincarbonaterocks, in Gabrovs ek,F., ed.,EvolutionofKarst:FromPrekarsttoCessation:Ljubljana, Ins titutzaraziskovanjekrasa,ZRCSAZU,p.43–59. S UBGLACIALMAZEORIGININLOW DIPMARBLESTRIPEKARST : EXAMPLESFROM N ORWAY 42 N JournalofCaveandKarstStudies, April2011

PAGE 13

Rasmussen,A.,1981,ThedeglaciationofthecoastalareaNWof Svartisen,NorthernNorway:NorgesGeologiskeUnderskelse Bulletin,no.369,p.1–31. Sejrup,H.P.,Larsen,E.,Landvik,J.,King,E.L.,Haflidason,H.,and Nesje,A.,2000,QuaternaryglaciationsinsouthernFennoscandia: evidencefromsouthwesternNorwayandthenorthernNorthSea region:QuaternaryScienceReviews,v.19,p.667–685. Simanca,S.R.,andSutherland,S.,2002,FractalDimension, in MathematicalProblemSolvingwithComputers,coarsenotesfor MAT331atUniversityofStonyBrook,http://www.math.sunysb.edu/ scott/Book331/Fractal_Dimension.html,[accessedJuly5,2009]. Skoglund,R..,Lauritzen,S.E.,andGabrovs ek,F.,2010,Theimpactof glacierice-contactandsubglacialhydrochemistryonevolutionof mazecaves:amodelingapproach.JournalofHydrology,(inpress; postedonline2010). Stephens,M.B.,Gustavson,M.,Ramberg,I.B.,andZachrisson,E.,1985, TheCaledonidesofcentral-northScandinavia-atectonostratigraphic overview, in Gee,D.G.,andSturt,B.A.,eds.,TheCaledonideOrogen -ScandinaviaandRelatedAreas,Part1:Chichester,U.K.,John WileyandSonsLtd.,p.135–162. Sundborg,A .,1956,TheriverKlara ¨lven,astudyoffluvialprocesses: GeografiskaAnnaler,v.38,p.125–237. Tranter,M.,Brown,G.,Raiswell,R.,Sharp,M.,andGurnell,A.,1993,A conceptualmodelofsoluteacquisitionbyAlpineglacialmeltwaters: JournalofGlaciology,v.39,p.573–580. Wells,M.K.,andBradshaw,R.,1970,MultiplefoldingintheSrfinnset areaofNorthernNordland:NorgesGeologiskeUnderskelse Bulletin,no.262,89p. White,W.B.,1969,Conceptualmodelsofcarbonateaquifers:Groundwater,v.7,p.15–21. R..S KOGLUNDAND S.-E.L AURITZEN JournalofCaveandKarstStudies, April2011 N 43



PAGE 1

BOOKREVIEW GroundwaterHydrologyofSprings:Theory,Management, andSustainability NevenKresicandZoranStevanovich(eds.),2010.Burlington, Mass.,Butterworth-Heinemann(imprintofElsevier),573p.,7 3 9 inches,ISBN978-1-85617-502-9,hardbound,$130.00. Springflowandqualityareimportantgroundwaterissues, particularlyinkarst.Manytextbooksaddressgroundwater hydrology,buttomyknowledgethisisthefirstbookthat specificallyaddressesthehydrologyofsprings.Allkindsof springsandinallkindsofterranesareconsidered,aswellastheir relationtosurfacehydrology.Karstreceivesspecialattention, however,becausespring-flowissuchanimportantaspectofkarst aquifers,andkarstisthemainspecialtyoftheeditorsandthe majorityoftheauthors.Authorswereselectedbytheeditors basedonrecognizedexpertiseandgeographicdistribution.The editorsalsocontributedmaterialtothebook. Thebookincludesageneraldiscussionofsprings,technical aspectsandmethods,andspecificexamples.Thediscussion beginswithachapteronthesustainabilityandmanagementof springsthatincorporateslargeexcerptsfrompreviousreports.It isagoodgeneraldiscussionoftheuseandmanagementofsprings andcoversmuchmaterialonhydraulicsystems(man-made structures)atspringsintheU.S.andinforeigncountries.Chapter 2isadiscussionofspringtypesandtheirclassification,both karsticandnon-karstic(itevenbrieflymentionsgeysersand fumaroles).Manyphotosareprovidedforclarification. Chapter3beginsthetechnicalsectionofthebookwitha discussionofrechargetosprings.Basicprinciplesareintroduced, aswellasmethodsforquantifyingrecharge(e.g.,viaartificialand environmentaltracers).Mathematicalandchemicalprinciplesare mentioned,but,asisappropriateforabookofthistype,notin greatdetail,asthisisalreadyavailableintextbooks.Chapter4 providesafairlycomprehensivediscussionofspring-hydrograph analysis.Quantitativetechniquesarepresentedinastylethatis relativelyeasy,useful,andclear.Itispossibletoobtainmuchof thisinformationfrombooksonsurface-waterhydrology,but theserarelydiscussspringflow. Chapters5and6addresshydrologicmodelingandgeochemistry.Theyareclearandinformativesummariesofvasttopics, andthosewhowishfurtherinformationcanconsultspecialized textbooks.Itwouldhavebeenhandyifthesechaptersincluded referencesandlinkstorelevantsoftware. Chapter7concernswater-qualitytreatmentofspringsto acceptabledrinking-waterstandards.Itincludesseveralcase studiesthatfitbetterherethaninaseparatechapteroftheirown. Chapter8describesthedelineationofspring-rechargezonesand strategiesforprotection.Thisconcepthasrecentlyreceivedmuch attentioninEurope,butlesssointheU.S.Itisabriefdiscussion ofasubjectthatiscontinuallybeingrevisedwithinindividual Europeancountriesandincludesrecommendationsinthe researchofRavbar(2007).Manyusefulreferencesareincluded. Chapter9addressestheutilizationandregulationofspring flow.Thehistoricaldevelopmentofcapturingandprotecting springwatersisdescribed,andseveralcasestudiesareprovided. ThistopicisrarelyconsideredintheU.S.,butitprobablyshould be.Thischaptercouldprovideguidance. Chapter10isacollectionofcasestudies.Thisisthelongest chapterinthebook(176pages)andconsistsoftensections. Includedaredetaileddescriptionsofspringsusedforwatersupply insoutheasternEurope,Austria,Romania,Turkey,Iran,Texas (EdwardsAquifer),andChina.Topicsincludegeology,hydrogeology,waterquality,exploitation,protection,andregional distributionofsprings.Again,thischaptermaybeespecially usefulintheU.S.ifcommunitiesbegintoutilizespringwaters morethantheycurrentlydo. Thisbookisworthpurchasingbyanygroundwaterhydrologistwhoworkswithspringwaters.Itisnotcheap,butitslarge amountofmaterialandgood-qualityback-and-whitefigures makethepricereasonable.Thereareinevitablyafewtypographic andformattingerrors,buttheseareminor(exceptforsomefaulty references).Thetopicsflowsmoothlyfromchaptertochapter, reflectingthecareinpreparationandediting.Imighthave recommendedincludingachapteronspringbiota(e.g.,thatin Gibertetal.,1994),which,amongotherthings,canbeameasure ofthelong-termhealthofthespringanditswatersource.Iexpect tokeepthisbookcloseathandasaguidetomyownwork. R EFERENCES Gibert,J.,Danielopol,D.L.,andStanford,J.A.,eds.,1994,Groundwate r ecology:SanDiego,AcademicPress,571p. Ravbar,N.,2007,Theprotectionofkarstwaters:Acomprehensive Sloveneapproachtovulnerabilityandcontaminationriskmapping: Ljubljana,Slovenia,Zaloz baZRC(ZRCPublishing),256p. ReviewedbyMalcolmField,NationalCenterforEnvironmental Assessment(8623P),OfficeofResearchandDevelopment,U.S.EnvironmentalProtectionAgency,1200PennsylvaniaAve.,NW,Washington, DC20460-0001(field.malcolm@epa.gov). DOI:10.4311/jcks2010br0135 B OOK R EVIEW 44 N JournalofCaveandKarstStudies, April2011



PAGE 1

BOOKREVIEW AdvancesinHypogeneKarstStudies KevinW.Stafford,LewisLand,andGeorgeVeni(eds.), 2009.Carlsbad,NewMexico,NationalCaveandKarst ResearchInstituteSymposium1,182p.,8.5 3 11inches,$50 plusshippingfromsales@nckri.org.ISBN978-0-9795422-4-4 HypogeneSpeleogenesisandKarstHydrogeologyof ArtesianBasins AlexanderKlimchoukandDerekFord(eds.),2009.Simferopol,Ukraine,UkrainianInstituteofSpeleologyandKarstologySpecialPaper1,292p.,8.2 3 11.7inches,$50plusshipping fromnssbookstore.org.ISBN978-966-2178-38-8 Formostofthepastseveralcenturies,studentsofcaves andkarstweresatisfiedwithasingle,underlyingconceptualmodel.Meteoricwaterinteractedwithsoluble bedrock,usuallylimestoneordolomite,sometimesgypsum,andoccasionallysalt.Differentialdissolutionon exposedlandsurfacesproducedthecharacteristickarst landforms:closeddepressions,sculpturedbedrock,and residualhillsofunusualshape.Infiltrationofmeteoric waterthroughcloseddepressionsandsinkingstreams developedcaveswithgreatvarietiesoflengthsand patterns.Theinletpointsandtheirsources,combinedwith theoutletpointsatsprings,allowedtheidentificationof karstdrainagebasinsandtheunderlyingkarstaquifers. Karstsystemstookonatremendousvarietyofdetail, dependingonlocalclimateandgeologicsetting,butthe underlyingprocesseswerethesame—asingleconcept. Thisconceptwaschallengedinthelaterdecadesofthe 20 th century.Studiesofcavessuchasthoseinthe GuadalupeMountainsofNewMexicoandtheBlackHills ofSouthDakota,aswellasthegiantmazecavesof Ukraine,haverevealedotherprocessesinvolvingupward migratingfluids,someofthemathightemperaturesor carryingsulfuricacidinadditiontocarbonicacid.The conceptofdeep-seatedorhypogeneticspeleogenesishad beenborn.TheseideasweresummarizedinAlexander Klimchouk’s HypogeneSpeleogenesis (reviewedbyJohn Mylroieinthe JournalofCaveandKarstStudies 70,129– 131,2008).Thehypogeneconcepthasnowreceivedenough attentiontowarranttwosymposia,theproceedingsof whicharethevolumesreviewedhere. TheStaffordetal.bookconsistsoffourteenpapersthat describetheinfluenceofdeepgroundwaterflowonthe originofcavesandrelatedfeatures.Itisbasedon presentationsataspecialsessionatthe2008national meetingoftheGeologicalSocietyofAmerica.Thisvolume representsaNorthAmericanpointofview,astwelveofthe fourteenpapersarebyUSandCanadianauthors.The bookisverynicelyproduced,withcolorphotographsand illustrationsthroughout. Inthefirstchapter,AlexanderKlimchoukdescribes hypogenecaveorigin,withspecialattentiontotheriseof wateracrossstratalboundariesinthedistalportionsof regionalandintermediate-scalegroundwatersystems.A suiteofcharacteristiccavefeaturesisdescribed,including floorslots,wallgrooves,ceilingchannels,andcupolas,all ofwhichheattributestorisinggroundwater.Inthenext chapter,however,JohnandJoanMylroieurgecautionin ascribingthesefeaturessolelytorisinggroundwater.They citeexamplesofsea-coastcavesinpoorlylithified limestonesthatcontainmostorallofthefeaturesdescribed inthepreviouschapter,butwhichhaveneverbeenin confinedsettingsorexposedtorisinggroundwater.Shorttermfloodingbyepigenicprocessesalsoproducesmany suchfeatures. MarcusGaryandJohnSharpdocumentseveraldeep Mexicanspringsthatarefedbyrisinggroundwater,with CO 2 ,H 2 S,andmildheatsuppliedbyvolcanicsources. Theyintroducetheterm volcanogenickarst forfeatures originatinginthisway.TheirtypeexampleisElZacato n,a 319mwater-filledshaftthatisoneofthedeepestknownin theworld.PhilippeAudraandcoauthorsdescribecave folia,whichconsistofarraysofsub-horizontalfungus-like calcitegrowths.Theyattributethesedepositstohypogene degassingofCO 2 onthebasisofmorphologicalfeatures suchasbubble-trailsinscribedinthecavewallsbythe escapinggas. Acollectionoffieldexamplesfollows,eachdocumentingaspecificaspectofhypogeneprocesses.KeltonBarr andCalvinAlexanderdescribedepressionsinMinnesota wherewaterrisesintovalleysburiedbyglacialdepositsand causescollapseattheoutlets,atopicofgrowingconcernto engineersandland-usemanagers.PaulBurgerexamines structuralandfaciescontrolofcavesintheGuadalupe MountainsofNewMexico.Thesearesulfuricacidcaves DOI:10.4311/jcks2010br0155 B OOK R EVIEW JournalofCaveandKarstStudies, April2011 N 45

PAGE 2

formedbytheoxidationofrisingH 2 S.Hedemonstrates, forexample,thatmazepatternsinthecavesstrongly correlatewithpaleokarstandearlytectonicbreccias. GeorgeVeniandLynnHeizleruseRobberBaronCave, inCretaceouslimestoneofsouth-centralTexas,asan exampleofcaveoriginbyrisinggroundwater.Ceiling cupolasandresidualbedrockbridgesacrosspassagesare citedasrepresentativebyproducts.VanBrahanaetal. describeanunusualcaveinArkansasinwhichepigenic passagewayshaveintersectedmucholderroomslinedwith calcitesparcrystalsupto1.9mlong.Isotopicsignatures indicatethatthecalcitewasofhypogeneoriginwith temperaturesgreaterthan100 u C.Relationshipsto dolomitization,brecciation,andMississippi-Valley-type oredepositsarediscussed. ThreechaptersconcernhypogeneprocessesandfeaturesinPermiangypsuminsoutheasternNewMexicoand westernTexas.Thisareaisoneoftheleast-studiedkarst regionsoftheUSA.RayNanceandKevinStafford describecavesandsurfacekarstfeatureswithhypogenic characteristics,suchascalcitizationofevaporites,sulfur deposits,andsolutionbreccias.Theseareattributedtothe influenceofhydrocarbonscarriedbyrisingwater.Kevin Staffordetal.describeregionalflowpatternsrelatedtothe PecosRiver.Theriverhaslongservedasatargetfor groundwaterrisingundersemi-confinedconditions.Hypogenedissolutionhasproducedmulti-storycaves,oil-field porosity,andsurfacedepressions.LewisLanddiscussesthe impactofdeep-seatedprocessesonthewaterresourcesof thearea.Karsticartesianbasinssupplynearlyallofthe waternecessarytosustainalevelofpopulationgrowthand agriculturaldevelopmentthatwouldotherwisebeimpossibleinthissemi-aridregion. Threechaptersconcerntheapplicationofhypogene karstprocessestoeconomicgeologyandregionaltectonic history.HarveyDuCheneprovidesfieldevidencethatoil fieldsnorthandwestoftheGuadalupeMountains suppliedthehydrogensulfidethatformedthelocalcaves (e.g.,CarlsbadCavern).Herelatesthechangesinoil-field character,groundwaterpatterns,andcavedevelopmentto blockfaultingintheRioGrandeRiftzonetothewest. DerekFordinterpretstheparagenesisofcarbonate-hosted sulfideoresintheNanisivikminingareaofBaffinIsland, northernCanada.Heinterpretstheoreashavingformed alongtheinterfacebetweensalinewaterandgasoroil about1600mbelowthesurface,withsimultaneous carbonatedissolutionandoredeposition.Langhorne Smithdescribeshydrothermalpetroleumreservoirsin OrdovicianrocksofeasternNorthAmericathatformed indolomitizedzonesaroundbasement-rootedtrans-tensionalfaults.Dissolutionandmineralizationalongthe faultswasaccomplishedmainlybyrisingthermalfluids. Naturalgasisabundantinandaroundelongatefaultboundedstructurallows(interpretedtobenegativeflower structures).Thistimelychaptertranscendstheusual boundariesofkarststudiesbyconcentratingondeepseatedtectonicandgeochemicalprocesses,aswellastheir applicationtopetroleumgeology. TheKlimchoukandFordvolumecontainsthethirtyninepaperspresentedattheInternationalConferenceon HypogeneSpeleogenesisheldatChernivtsi,Ukraine,in May2009.Thirty-sixofthepapersareinEnglishandthree areinRussian.ContributionsarefromEurope,Russia, Australia,Brazil,theUS,andCanada.Onceidentified,it appearsthathypogenecavesareeverywhere. TheopeningpaperbyAlexanderKlimchoukandthe twofollowingpapersbyPhilippeAudraandhiscolleagues setouttoidentifythecharacteristicmorphologicalforms thatresultfromhypogenespeleogenesis:cavepatternswith elaboratethree-dimensionalstructure,cupolas,half-tubes, andtheplanatedsurfacesassociatedwithcondensationcorrosion.Morphologicalformscanbesubjecttomultiple interpretations.Moresolidevidenceisprovidedbymineral deposits,especiallytheisotopiccompositionofcoarsely crystallinecalcite.DublyanskyandSpo ¨tldescribeoxygen isotoperatiosinthecalcitecoatingofcavitiesinacavein theAustrianAlpsthatidentifythetemperatureofwaters thatmovedthroughthecave. Microbialprocessesareknowntobeimportant catalystsinthegeochemicalreactionsofhypogene speleogenesis.P.J.Bostonandhercolleaguesgivean overviewandanassessmentoftheirimportance. Modelingofhypogenesystemsisextremelydifficult becausetherearefewharddataonthesources,earlyflow paths,andchemistryofthefluids.Threeattemptsaremade toatleastdescribethemechanicsoffluidflow.Rehrl,Birk, andKlimchoukofferagenericmodelshowinghow fracturesystemsmightbeexpectedtoevolve.Dreybrodt, Romanov,andKaufmanndescribeaquantitativemodel formixing-corrosionthatcanbeappliedtocoastalkarst withfreshwater-saltwatermixingzones.Itshouldalsobe applicabletodeep-seatedupwellingfluids.Anotherapproachtomodelingofmixingzonekarstisdescribedby AntoineLafareandhiscolleaguesandappliedtothe Mediterraneankarst. Asmightbeexpected,mostofthepapersdescribe specifichypogenecavesthattheauthorsthinktheyhave identified,ortocaveswherehypogeneticprocessesare thoughttohaveplayedanimportantrole.Thereisan amazingdiversityofsites.TherearetheObruks,giant collapseshaftsinTurkey(Bayariandcolleagues).Thereis theendokarstofMallorca(Gine sandcolleagues).There arethehypogenecavesoftheItalianApennines(Sandro Galdenzi),whichincludetheimportantFrasassiCaves,in whichmuchmicrobiologyresearchisnowunderway. Beyondthese,thereareexamplesofhypogenecavesfrom Austria,Slovenia,Israel,theCrimea,Brazil,Poland, Romania,Greece,Norway,Russia,Jordan,andSaudi Arabia,tonameonlythemainlocalities. Inthiscollectionofpapersfrombothsymposia,thereis someimpressiveprogressandatleastoneseriousgap.As toprogress,cavesofhypogeneticoriginhavebeen B OOK R EVIEW 46 N JournalofCaveandKarstStudies, April2011

PAGE 3

identifiedfrommanyregionsandinmanyenvironments. Additionally,manycaves,clearlyremnantsofthedevelopmentofcontemporarywatersheds,alsohaveaninitial hypogeniccomponent.Thegapistheabsenceofany detailedgeochemicalmodelforadissolutionprocesswhere risinggroundwateraloneisresponsibleforcaveorigin. Fluidspercolatingupfromdeptharerarelyseenandare difficulttoanalyze. Thesebooksillustratethatkarstprocessescanextendto considerabledepthandinvolvechemicalprocessesthatare seldomobservedatthesurface.Readerswhowishtoapply themethodsdiscussedhereshouldkeepinmindthat interpretationsmustbecompatiblewithregionalgroundwaterflowrates,chemicalenvironments,andthegeologic timeframeandthatmanyrelictgeomorphicfeaturescan beattributedtomorethanasingleprocess.These importantbooksshowthewiderangeoftheseinterpretationsandapplications. ReviewedbyArthurN.Palmer,Dept.ofEarthSciences,StateUniversity ofNewYork,Oneonta,NY13820-4015(palmeran@oneonta.edu),and WilliamB.White,MaterialsResearchLaboratory,PennsylvaniaState University,UniversityPark,PA16802-4801(wbw2@psu.edu). B OOK R EVIEW JournalofCaveandKarstStudies, April2011 N 47



PAGE 1

rf rfntbftfbfbtfbt rfnrf ftftbttbbttnff ff tbrrbrrrbbbrf fft trbrrrf nfft nrf tffrftbtf rrrf tff rntb rtrntb b r trbfbfftbfbft fb r rf ff tr r r rt

PAGE 2

r t rfbft bbtbfft r ff fbftfttfttfb tffffbbf tttf ftfttf rff tfffffbbbff tftbft ftff rfttbffbtff tffbbff fbffbf fbff rtf tfbtfb ftbtb ffbbff ttf fbbbfffftb tbfttfffttbfb tf nftffffb ffftttfftff ftftb rtfft fffff nrntfffb fbbfbbb fftbt ftbftff ffbbtf bfttt fbfbfftfb fbbtftf ttbbtffb ff ffttfn fftbbbffttffb tfttffbf ttffbfff fftfff fbfttbffb b tftf fbtfbffff fbbffftftftf bbbbffff bffbffbt ft fbbbfff ftftt ftfbfff ftftftbbf nrbbft fttfbbbffb ffffb rbbbbb ftbbfnr fbftbbfrr nfffftffr fftfttbt ftbfbfttb ffbftffbbftbb ffbbftffbbf ffbftfbtf tffffttbftbf ttfffb tbbfbffb tb rffttb tfbbfb fbfbbf bbftfb tbbb tbftfffff fb rtf ftfffbffb bfbbftfffft tftfffff tf nfff ttfttf fbtfftb ffbf rnrfbfbr tffbffb rffbbftbb frffbtbt fbfffb tfbff tfrf bbf nnn ffffttt bfttfffbf fbffbfffff ftfbbbttftf n fbftfff fbnbftfb tfbffb fnfbftfb fftbtbfttt tbfbtffb ttttff frftff fb nrrfbf fffbtttrfbftbfbf ttfr fftfbbtf trffftftfffbtbtft nfbbbfftfrf rftbftfttbbbtf bffttbttf ttt rtffffbftbtfbbfff fff ffff tnftb tt f f tr btff r n fttttr r t rt n rttr ft tfbff b bffftf fbtf nbb ftftff tttf f ff fb nftft ffb ff f f fb ff t b rf tbbffff f f bffbntb r ttf f b