Submarine and intertidal groundwater discharge through a complex multi-level karst conduit aquifer


previous item | next item

Citation
Submarine and intertidal groundwater discharge through a complex multi-level karst conduit aquifer

Material Information

Title:
Submarine and intertidal groundwater discharge through a complex multi-level karst conduit aquifer
Series Title:
Hydrogeology Journal
Creator:
Schuler, Philip
Duran, L.
McCormack, T.
Gill, L.
Publication Date:
Language:
English

Subjects

Subjects / Keywords:
Karst ( local )
Coastal Aquifers ( local )
Tracer Tests ( local )
Numerical Modelling ( local )
Ireland ( local )
Genre:
serial ( sobekcm )

Notes

Abstract:
The quantification of submarine and intertidal groundwater discharge (SiGD) or purely submarine groundwater discharge (SGD) from coastal karst aquifers presents a major challenge, as neither is directly measurable. In addition, the expected heterogeneity and intrinsic structure of such karst aquifers must be considered when quantifying SGD or SiGD. This study applies a set of methods for the coastal karst aquifer of Bell Harbour in western Ireland, using long-term onshore and offshore time series from a high-resolution monitoring network, to link catchment groundwater flow dynamics to groundwater discharge as SiGD. The SiGD is estimated using the “pollution flushing model”, i.e. a mass-balance approach, while catchment dynamics are quantified using borehole hydrograph analysis, single-borehole dilution tests, a water balance calculation, and cross-correlation analysis. The results of these analyses are then synthesised, describing a multi-level conduit-dominated coastal aquifer with a highly fluctuating overflow regime draining as SiGD, which is in part highly correlated with the overall piezometric level in the aquifer. This concept was simulated using a hydraulic pipe network model built in InfoWorks ICM [Integrated Catchment Modeling]® version 7.0 software (Innovyze). The model is capable of representing the overall highly variable discharge dynamics, predicting SiGD from the catchment to range from almost 0 to 4.3 m3/s. The study emphasises the need for long-term monitoring as the basis for any discharge studies of coastal karst aquifers. It further highlights the fact that multiple discharge locations may drain the aquifer, and therefore must be taken into consideration in the assessment of coastal karst aquifers.
Original Version:
Hydrogeology Journal, Vol. 26 (2018-07-06).

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
This item is licensed with the Creative Commons Attribution License. This license lets others distribute, remix, tweak, and build upon this work, even commercially, as long as they credit the author for the original creation.
Resource Identifier:
K26-05117 ( USFLDC: LOCAL DOI )
k26.5117 ( USFLDC: LOCAL Handle )

USFLDC Membership

Aggregations:
University of South Florida
Karst Information Portal

Postcard Information

Format:
serial

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

PAPER Submarineandintertidalgroundwaterdischargethroughacomplex multi-levelkarstconduitaquiferPhilipSchuler1&L.Duran1&T.McCormack2&L.Gill1Received:10February2018/Accepted:17June2018/Publishedonline:6July2018 #Abstract Thequantificationofsubmarineandintertidalgroundwaterdischarge(SiGD)orpurelysubmarinegroundwaterdischarge(SGD) fromcoastalkarstaquiferspresentsamajorchallenge,asneitherisdirectlymeasurable.Inaddition,theexpectedheterogeneity andintrinsicstructureofsuchkarstaquifersmustbeconsideredwhenquantifyingSGDorSiGD.Thisstudyappliesasetof methodsforthecoastalkarstaquiferofBellHarbourinwesternIreland,usinglong-termonshoreandoffshoretimeseriesfroma high-resolutionmonitoringnetwork,tolinkcatchmentgroundwaterflowdynamicstogroundwaterdischargeasSiGD.The SiGDisestimatedusingthe B pollutionflushingmodel ^ ,i.e.amass-balanceapproach,whilecatchmentdynamicsarequantified usingboreholehydrographanalysis,single-boreholedilutiontests,awaterbalancecalculation,andcross-correlationanalysis. Theresultsoftheseanalysesarethensynthesised,describingamulti-levelconduit-dominatedcoastalaquiferwithahighly fluctuatingoverflowregimedrainingasSiGD,whichisinparthighlycorrelatedwiththeoverallpiezometriclevelintheaquifer. ThisconceptwassimulatedusingahydraulicpipenetworkmodelbuiltinInfoWorksICM[IntegratedCatchmentModeling]®version7.0software(Innovyze).Themodeliscapableofrepresentingtheoverallhighlyvariabledischargedynamics,predicting SiGDfromthecatchmenttorangefromalmost0to4.3m3/s.Thestudyemphasisestheneedforlong-termmonitoringasthe basisforanydischargestudiesofcoastalkarstaquifers.Itfurtherhighlightsthefactthatmultipledischargelocationsmaydrain theaquifer,andthereforemustbetakenintoconsiderationintheassessmentofcoastalkarstaquifers. Keywords Karst . Coastalaquifers . Tracertests . Numericalmodelling . IrelandIntroductionIrishkarstaquifersconsistingofCarboniferouslimestones, whicharecharacterisedbyalow-lyingtopographyandexpositiontothecoast,presentanumberofrelevantresearchchallenges,includinggroundwaterfloodingdynamicsandtheinteractionoftemporaryfloodlakes,i.e.turloughs(Naughtonet al. 2012 ;Gilletal. 2013 ),saltwaterintrusion(Perriquetetal. 2014 ),pipenetworkmodellinginconduit-dominatedcatchments(Gilletal. 2013 )andassociatednutrientinputintothe aquaticcoastalecosystems(McCormacketal. 2014 ),and submarineandintertidalgroundwaterdischarge(SiGD; CaveandHenry 2011 ). Forkarsticcoastalcatchments,strongheterogeneitymust beexpected(Burnettetal. 2006 ),whichcreateschallengesin theapplicationofmethodstoquantifydynamics.Ontheother hand,SiGDplaysanimportantroleincoastalecosystems (Burnettetal. 2006 ),andthereforeanunderstandingofits dischargedynamicsisimportant.Forexample,nutrientloadingviasubmarinegroundwaterdischarge(SGD)drivenby landusemayenhancethedevelopmentoftoxicalgaeblooms (Silkeetal. 2005 )orharmfulmicro-ormacroalgae(Greenet al. 2014 ;Lietal. 2017a ),orpromotetheover-developmentof certainfishspeciessuchasjellyfish(Dongetal. 2010 ). However,giventhenatureofthedischargelocations,direct physicalmeasurementofSiGDorpurelySGDisnotpossible withthetypesofgaugingmethodsusedatonshoresprings. Instead,studieshaveapplieddirectandindirectmethodsfor measuringthemasstransferofgroundwateracrossthesea floor(Zektseretal. 2007 )ortowardsasurfacewaterbody, including(1)measuringtheseepageflowrateusingseepage * PhilipSchuler schulerp@tcd.ie1DepartmentofCivil,StructuralandEnvironmentalEngineering, UniversityofDublinTrinityCollege,Dublin2,Ireland2GeologicalSurveyIreland,HaddingtonRd,Beggar ’ sBush,Dublin 2,Ireland HydrogeologyJournal (2018)26:2629 – 2647 https://doi.org/10.1007/s10040-018-1821-3 TheAuthor(s)2018

PAGE 2

meters(CarrandWinter 1980 ;Corbettetal. 2003 )ormultilevelonshorepiezometers(FreezeandCherry 1979; TaniguchiandFukuo 1996 ),(2)applyingmass-balanceapproachesusingnaturalgeochemicaltracerssuchaselectrical conductivity(EC),short-livedradiumandradonisotopes,i.e.222Rn,223Ra,224Ra,226Ra,oroxygen-18( 18O)anddeuterium( 2H)(Moore 2006 ;Petersonetal. 2008 ;Santosetal. 2008;CaveandHenry 2011 ;Leeetal. 2012;Nulletal. 2014 ;Kneeetal. 2016),(3)applyingwaterbalanceapproachesbasedonthecontributingcatchment(Sekulicand Vertacnik 1996 ;SmithandNield 2003 ),(4)usinghydrograph separationtechniquestoquantifythegroundwatercontributionofsurface-waterstreamsandextrapolatingthiscontributiontothecoastalshore(Zektseretal. 2007 ),(5)employing numericalmodelling(Thompsonetal. 2007 ;McCormacket al. 2014 ;Taniguchietal. 2015 ),and(6)usingthermalimaging fromremotesensing(Johnsonetal. 2008 ;WilsonandRocha 2012 ;Tamborskietal. 2015 ).Whilemethods1 – 5canbe consideredquantitative,method6byitselfwillonlyyielda purelyqualitativeresult. InIreland,SGDhasbeenknownfromancienttimes,mainlyassociatedwithkarsticlimestones(Zektseretal. 2007 ).Due tothegreattidalvariationinthecontextofthisstudy,SGDis definedasanyinputthatisbelowthelowestebbofspring tides(whichisthereforetypically>4mbelowsealevel [mbsl]),whereasintertidaldischargeoccursalongtheshore, influencedbytidaloscillation.Whilemanyspringlocations drainingasintertidaldischargeareknown,thedischargelocationsofpurelySGDofftheshore,presumablylinkedtolower sealevelsduringthePleistocene(Drew 1990 ),areknowntoa lesserextent,althoughtheyarereportedanecdotallybylocal fishermenandmaricultureworkers.SGDislinkedtodeeper flowpathsinkarstaquifersthaninthecaseofintertidaldischarge.Deepkarstificationwithactivegroundwaterflowis knowntoexistinIrelandrelativelyclosetotheshore,e.g.at 77mbelowgroundlevel[mbgl;approx.67mbsl(Mayol 2011 )]orbetween70and80mbgl(approx.60 – 70mbsl), closetoKinvaraonthewesterncoastofIreland,County (Co.)Galway(CaveandHenry 2011 ).Itmustbeassumed thattheselow-lyingconduitsdrainintothesea(Bunceand Drew 2017 ). However,untilrecently,thedischargeofcoastalkarstaquifersinIrelandhaslargelybeenconceptualisedasintertidal dischargeorshallowSGDclosetotheshore — forexample, theBellHarbourcatchmentintheBurrenplateaulocatedin thewestofIreland(Perriquet 2014 ;McCormacketal. 2017 ). GroundwaterdischargeintoBellHarbourbayisknownto occurassubmarinedischargeaswellasinthevicinityof intertidalsprings,henceSiGD. Thepurposeofthisstudyistoimprovetheunderstanding ofthehydrogeologyofacomplexcoastalkarstcatchmentin Irelandbycombiningdifferentmethodsaimingtoquantify SiGD,linkedtounderstandingonshoregroundwaterflow dynamicsbyapplying(1)amass-balanceapproachinthe formofatidalprismmodelusingspecificECrecordsfrom seawater,(2)anonshorewaterbalancefortheassumed groundwatercatchmentofBellHarbour,(3)single-borehole dilutiontests(SBDT;Maurice 2009 ),(4)time-seriesanalysis, and(5)anumericalpipenetworkmodeltosimulategroundwaterflowdynamicsandSiGDintoBellHarbourbay. ThissetofmethodswascombinedtoquantifySiGD,and furthertoexplainthefunctioningofacomplexcoastalkarst aquifer.MaterialsandmethodsStudyareaThegroundwatercatchmentofBellHarbour(Fig. 1 )formsthe north-easternpartoftheBurrenlimestoneplateaulocatedin thewestofIreland.ThelimestonemassifoftheBurren,includingtheuplandcatchmentofBellHarbour,isdescribedas atemperateglaciokarstlandscape,whichwassubjecttorepeatedglaciationduringthePleistocene,showingfeatures typicalofglaciationsuchasice-pluckedcrags,scouredrock surfaces,limestonepavementsanderraticboulders(Simms 2014 ).Today ’ smorphologyofthecatchmentandthebay maybepartlytheresultofglacialerosionduringthelastice advance,whosedirectionislargelyparalleltothebay(Drew 1990 ). Thecatchmentwasdelineatedbasedontracertestspreviouslyexecutedintheregion(Drew 2003 ).Whiletheeastern andwesternboundariesareconstitutedbysharpescarpments andthenorthernboundaryintersectswiththeshore,theextent ofthecatchmenttothesouthisquiteuncertain.Untilnow,no tracersinjectedinthesouthernpartofthecatchmenthave beenrecoveredinBellHarbourbay.Thecatchmentcovers atleast50km2basedonthetopographicalextent,butmay extendfurthertowardsthesouth(BunceandDrew 2017 ), potentiallyaddingupto13km2.Thecatchmentisintersected bymultiplevalleys,withelevationsrangingfromsealevelin thenorthto300mabovesealevel(masl)alongtheescarpments.Alongtheslopesthebareoutcropisuncovered,showinghighdegreesofkarstification.Inturn,inthevalleys,relativelydeepsoilcoveronoutcropsofthelimestoneunitis attributedtoHoloceneweathering,formingthepresent erosion-resistantclay-richsoil(MolesandMoles 2002 ). Duetothehighdegreeofkarstification,surface-waterfeaturesarelimitedtoshortreachesofephemeralstreams,drainagefromadjacentnon-carbonaterocks,andturloughs(Drew 1990 ).Turloughsaredescribed(EPA 2004 )astopographic depressionsinkarstwhichareintermittentlyfloodedonan annualcycleviagroundwatersourcesandhavesubstrate and/orecologicalcommunitie scharacteristicofwetlands. Withinthecatchmenttherearetwoturloughs(Fig. 1 ):Luirk 2630 HydrogeolJ(2018)26:2629 – 2647

PAGE 3

Bealaclugga (Bell Harbour) Ballyvaughan Kinvara Carran Killnaboy1501 0 02 501501 5 01 003 002 5 01 0 05 01501 0 00000505005025 02001502 5 02002 0 02501 5 01 5 01 5 01 0 01502002 0 02 5 02002 0 01 501 5020 05 01 001 001502 0 0502 0 02001 501 5 01 0 02001505 00020 02 0 02001501 5 015 01501 0 010 01 001 0 01001001 001 00505 05 05 05 0505 05 0505 01501501 5015 015 01 5 020 00000000002001 5 01 5 0015 010010 01001001 001 0 01 001001 0 01005 050501 501 0 01 0010010 05 01 5 010010 0 523000 526000 529000 532000 535000 A 'A Atlantic Ocean / Galway BayBell H ar b o u r Bay LegendSampling sites Climate Rainfall GWpiezometer GW spring GW turlough SW bay SW lake SW river GW catchment Bell Harbour Boundaries relatively certain Potential contribution Spring Fergus River Minor stream Cross-section Mineral vein Boundary sandstone-limestone GSI GW recovered tracer test 50 m contour 25 m contour Fault Bell Harbour Bay outlet Geological formations Alluvial Tubber Formation Lower Burren Formation Upper Burren Formation Slievenaglasha Formation Namurian sandstoneRepublic of IrelandUK (Northern Ireland)Atlantic OceanStudy area (main map) 692000 696000 700000 708000 712000 704000C1 (PTHRW) P1 (P) BHB1+2 (CTD) T1 (D) T2 (D) BH1 (CTD) BH2 (D) (D) (D) SiGD1 (CTD) (D) (D) (D) Fig.1 MinimumgroundwatercatchmentofBellHarbourandpotential additionalcontributioninthesouthintheBurrenlimestoneplateau: topography,geology,structure,tracertests,samplinglocations(C= conductivity,D=depth,T=temperature,P=precipitation,H=relative humidity,R=netradiation,W=windspeed/direction).Inthelegend, GWisgroundwaterandSWissurfacewater HydrogeolJ(2018)26:2629 – 2647 2631

PAGE 4

(T2)intheLowerBurrenformation,andGortboyheen(T1) withintheMaumcahaformation. Geologyandstructure TheentirecatchmentisunderlainbyLowerCarboniferous well-beddedandpurelimestonerangingbetweentheTubber formation(earlyViséan),theLowerBurrenandUpperBurren formation(mid-Viséan,Asbian),andtheSlievenaglashaformation(lateViséan,Brigantian),togetherreaching>510min thicknessabovetheGalwayGranite.Dolomiticlayersare knowntoexistatthetoptheFinavarramember(Prachtetal. 2004 )andbetweentheLowerBurrenformationfromthe UpperBurren(Gallagheretal. 2006 ).Palaeokarsthorizons occuratthetopoftheMaumcahamemberbelowtheoverlyingAillweemember,whichisdistinguishedbetweenaLower andUpperAillweemember(MacDermotetal. 2003 ),andat thetopoftheAillweemember.Inaddition,severalimpermeableclay B wayboards ^ arelocatedwithintheAillweemember,interpretedaspalaeosols,whichrestonirregularlimestonesurfacesconsideredtobepalaeokarsts(Prachtetal. 2015 ).Thedipofthestrataisuniformlytothesouth,inthe rangeof2 – 3°(Fig. 2 ). Thereisonefaultinthecatchment,i.e.MacDermot ’ sfault inthewesternpartofthestudyarea,strikingintotheNNE sub-paralleltotheFergusShearZone,withaslight(<200m) sinistraldisplacementofmembersoftheBurrenand Slievenaglashaformations(Prachtetal. 2004 ).Jointsstrike NW – SEandE – W,andtheageofjointformationispostVariscan.TheVariscancontractionaldeformationcausedthe formationofveins,whicharelaterallycontinuousandverticallyconsistentacrossbeddingplanesanddiscontinuities (Gillespieetal. 2001 ).Horizontalpersistencyofveinsalong strikesextendsover7km,interconnectingcaves(MacSharry 2006 ).Increasedveiningwasrecordedtodepthsof204 – 228 mbslinthedeepboreholedrilledbytheGeologicalSurvey Ireland(GSI)atlocationBH1(Fig. 1 ),asdocumentedinthe respectivedrillinglog. Hydrogeology Thehydrogeologyofthecatchmentislargelyinfluencedby erosionduringtherecentglacialadvancesaswellaslimestone dissolution.Thegroundwaterflowisconduit-dominatedfrom southtonorth,i.e.updip,asintheneighbouringcatchmentin thewest,andthemaingroundwaterdischargefromthecatchmentisbelievedtohaveoccurredviaSiGDintoBellHarbour bay(Perriquetetal. 2012;McCormacketal. 2017).Bell Harbourbayhasaveryshallowtopography,largelyinfluencedbytidaloscillation:inthelongterm,89%ofwaterin thebayisdrainedduringebbtide.ThemainlocationofintertidaldischargeisPouldoodyspring(SiGD1,Fig. 1 ),whichis characterisedbyahighdegreeoffluctuationinEC,ranging from<300 S/cmduringperiodsofhighflowinwinterto> 40,000 S/cmduringperiodsoflowflowbetweenspringand autumn.Theassumptionofanorth – southflowdirectionalong thedipcanberuledout,astherearenopotentialdischarge locationsforthecatchmentinthesouth.Thesouth – northdischargepatterntowardsthebayallowsforthepossibilitythat preferentialsolutionofthelimestoneswouldhavetakenplace inthedominantsouth – northjointsinthezonewherefreshand salinewatermixtoproducesubsaturatedwaters(Drew 1990 ). Minorandseasonalgroundwaterdischargeoccursalongthe escarpmentoftheAillweemember,whereclaywayboards preventverticalpercolationandinsteadfavourlateralflow above.However,claywayboardsarenotcontinuouslaterally, andcanbelocallyintersectedbyverticallycontinuousdissolvedmineralveins(MacSharry 2006 ),resultinginpossible deeppercolation.Oneexampleoftheimpactofveinsonthe hy drogeologyisthedevelopmentofPollGonzocavefroma dissolvedveinthatwasexploredbetweenitstopat116masl andthewatertableat31masl(Bunce 2010 ).Withinthecave, laterallyinflowinggroundwaterfromthesouthtravelsrapidly downwards,andpotentiallyfurthertothenorthintoGalway bay(BunceandDrew 2017 ). Thecombinationofthestructuralpatternandextensive erosionanddissolutioncausedtheformationofconduits;the limestonesoftheBurrenareclassifiedas B regionallyimportantkarstifiedaquifersofconduittype ^ (GSI 2015 ).Ashallowconduitfollowingsouth – northwithinthevalleywasdetectedbyMcCormacketal.( 2017 ).Thisconduitisassumed toconnectthetwoturloughsinthevalley,Gortboyheen(T1) andLuirk(T2),allowingtheturloughstorapidlyfillfollowing persistentperiodsofrain,andsimilarlytoalmostemptyagain. Zonesofdeepkarstificationandhighdensitiesofperpendicularfractureswerefoundinthedeepborehole(BH1)at depthsbetween93and110mbgl,correspondingtothesea levelofthemostrecenticeage(McCormacketal. 2017 ). Theonlymappedfaultinthecatchment,MacDermot ’ s fault,showsevidenceofbeinghydraulicallyactive,assuggestedbyO ’ Connelletal.( 2012 ).DataAmonitoringnetworkwasestablishedtocontinuouslymonitorclimate,surface-water,andgroundwaterparameters(Fig. 1 ). Groundwaterlevelwasmeasured(1)attwoopenuncased boreholes(BH1andBH2)usinganINWCT2Xconductivitytemperature-depth(CTD)diver(InstrumentationNorthwest, Inc.[INW],Kirkland,WA,USA),and(2)atthebottomof turloughGortboyheen(T1)usingaSchlumbergerMini-Diver (SchlumbergerWaterServices,BC,Canada).T1isseasonally floodedwhilethemeasuredwaterlevelrepresentstheheadin theunderlyingconduitsystem.SiGDwasobservedatthe intertidalPouldoodyspring(SiGD1)usinganINWCT2X 2632 HydrogeolJ(2018)26:2629 – 2647

PAGE 5

CTDdiverthatwasinstalledonaconcreteplatformplacedat theshore. Abuoysamplerconsistingofananchoringbottomweight, asolidrope,andafloatingbuoyonthetopwasinstalledinthe centreofthenarrowoutletofBellHarbourbay.Thesampler wasequippedwithtwosensors,BHB1andBHB2(INW CT2XCTD/VanEssenCTD-Diver[VanEssenInstruments, Delft,Netherlands]),at~0.2and1mbelowthewatersurface, respectively,continuouslysamplingspecificEC,temperature (T),anddepthat10-minintervals.Theinstrumentsaimedto collecttheabovementionedparametersduringtheincoming floodtideandoutgoingebbtide,ofwhichECwasthenused toquantifySiGDintothebay. Rainfallandallrelevantmeteorologicalparametersforestimatingthereferenceevapotranspirationaccordingtothe FAO-56Penman – Monteithmethod(Allenetal. 1998 )were sampledattheBurrenOutdoorandEducationCentre(C1)at ~38masl,accordingtoWorldMeteorologicalOrganization guidelines(WMO 2008).Asecondrainfallsampler(P1) installedat~203maslontopoftheBurrenplateaushowed theabsenceofasignificantorographicimpactontherainfall regime,resultinginspatialheterogeneity. Allwaterlevelswerederivedfrompressurereadingscompensatedtothebarometricpressuremeasuredcontinuouslyat C1usinganINWPT2Xpressuresensor. Secondaryspatialdatausedforthisstudyconsistofthe following:A5-mbathymetricgridprovidedbytheGSIwas usedtocalculatethevolumeofthebayatdifferentwater levels.ThebathymetricgridwascombinedwiththeShuttle RadarTopographyMission(SRTM)elevationgrid(USGS 2014 )andcomplementedwithdatafromextensivetopographicalfieldsurveysusingaTrimble4700(TrimbleNavigation, Sunnyvale,CA,USA)differentialglobalpositioningsystem (GPS)withhorizontal/verticalaccuracyof0.01m,toderive animproved5-melevationgridforthestudyarea.Thisgrid wasusedtocalculatethevolumeofturloughsusingArcGIS (Esri,Redlands,CA,USA),followedbytheestablishmentof astage – volumecurveforeachturlough.Cross-correlationanalysisInkarsthydrology,cross-correlation(CC)hasbeenusedby severalauthorstocharacterisemostlyinput – outputrelationshipsbetweendifferenttimeseries(PadillaandPulido-Bosch 1995 ;Angelini 1997 ;Larocqueetal. 1998 ;Labatetal. 2000 ; Mathevetetal. 2004 ;Masseietal. 2006 ).Forexample, Larocqueetal.( 1998 )demonstratedtheusefulnessofCC analysisonspatiallydistributedtimeseriestocharacterise anddelineateakarstaquiferaccordingtocommonunderlying dynamics. Inthepresentstudy,cross-correlationwascarriedoutbetweenrainfall,headlevelsintheborehole,turloughwater level,andSiGDontheassumptionthatthetimeseriescould beregardedasbivariatestochasticprocessesthatarestationary(BoxandJenkins 1976 ;Juki andDeni -Juki 2015 ). Cross-correlationanalysiswasappliedonhourlyordaily timeseriesofdifferentmonitoringsitestoidentifyandquantifylinearrelationshipsbetweenseriestoimprovetheunderstandingofcatchmentdynamics.Single-boreholedilutiontestsAmbientgroundwaterdynamics,includingwaterlevel,temperature,andconductivity,werecontinuouslymeasuredin BH2andBH1usingaCTD-Diver.Inaddition,singleboreholedilutiontests(SBDT;Maurice 2009 )wereusedto gaininformationonflowdirection,meanflowvelocities,and tracerrecovery,aswellastolocateinflowandoutflow. NNE SSW-500 -400 -300 -200 -100 0 100 200 300 400 05,00010,00015,00020,000 Upper ViseanLegendUpper Aillwee Slievenaglasha Lower Aillwee Maumcaha Fanore Black Head Tubber Lower Visean Upper Tournaisian Early Devonian Middle Visean Ballys teen Galway GraniteDistance [m] Fig.2 Geologicalcross-section(A – A ,Fig.1 )throughthecatchmentofBellHarbour HydrogeolJ(2018)26:2629 – 2647 2633

PAGE 6

SBDTwereconductedinBH1underambient,i.e.nonpumped,conditionsfollowingthemethodoutlinedby Mauriceetal.( 2011 ),applyingconcentratedemplacements oftracersolutions(125g/lNaClanddeionisedwater)through ahoseintotheborehole,whichwasmeasuredatthepointof injectionusingaSchlumbergerCTD-Diver(DI263),aswell asaboveandbelowusingasetofAquiStar®Aqua4PlusCT2X CTD(INW). Thetotaltracerrecovery Mt[mg]wasestimatedbyEq.( 1 ) (Field 2002 ): Mt¼ 0Ct ðÞ Qt ðÞ dt ð 1 Þ withthetracerconcentration C [mg/l]attime t [s]andthe discharge Q [l/s].Thebackgroundtotaldissolvedsolids (TDS)concentrationwasaccountedfor. Themeanverticaltraveltime v [m/s]wascalculatedbased onthemeanresidencetime t [s],whichisgivenbyEq.( 2 ) (Field 2002 ): t ¼ 0tCt ðÞ Qt ðÞ dt 0Ct ðÞ Qt ðÞ dt ð 2 Þ Zonesofinflowand/oroutflowofgroundwaterweredeterminedbyrepeatingconcentratedtracerinjectionsafterloweringorliftingtherecordingCTDdiver,accordingtothedirectionoftracerplumemigrationfromeachprevioustest.QuantificationofSiGDQuantificationofSiGDintoBellHarbourbaywasperformed usingcontinuouslymeasured specificECconcentrations withinthenarrowoutletofthebay(BHB1+2)andapplying atidalprismmodelbasedonmass-balanceprinciples(Kinget al. 2010 ). Forthisstudy,azero-dimensional(spatiallyuniform) modellingapproachbasedontherepeatedexchangeofthe intertidalvolumewasapplied,baseduponapollutionflushing model(Barber 2003 ;BarberandWearing 2004 ).Themodel relatesthewaterqualityresponseofthebaytotheexternal forcingeffectsofthetide,theinitialpollutantloading,andthe rateofgroundwaterdischargeasinflowintothebay,i.e. SiGD.Themodelestimatestheconcentrationofaconservativepollutantattheendofanebbtideandtheendofaflood tide.Duringthefloodcycle,waterfromoutsideentersthebay andmixeswiththeexistingwaterinthebay,includingSiGD. Duringthefollowingebbcycle,themixedbaywateris drainedoutsidethebay,andtheconcentrationofapollutant isestimatedbythefundamentalmodelequation,asEq.( 3 ): Ce n ðÞ¼ Cf n Š 1 ðÞexp Qf V2 mŠ V2 tq 8 > < > : 9 > = > ; ð 3 Þ where Ce( n )isthepollutantconcentrationattheendofebbtide n , Cf( n Š 1)isthepollutantconcentrationattheendofthepreviousfloodcycle, QfistheSiGDwithinthebay[m3], isthe tidalangularfrequencygivenby =2 / T ,where T isthe periodofthetide[h], Vmisthemeanvolumeofthebasin [m3],and Vtistheamplitudeoftheoscillatorycomponentof thetidalvolume[m3]. Inthisstudy,theconservativepollutantwasdefinedasthe salinityofwaterrepresented,expressedasspecificEC. Accordingly,thesalinityinthebayresultsfromincoming seawaterduringfloodtideanddilutionofseawaterbySiGD intothebay.GiventhatSiGDintothebaycouldonlybe estimatedduringeachebbtide,Eq.( 3 )wasrearrangedtosolve for QftoyieldtheSiGDatebbtide n ,asshowninEq.( 4 ): Qn ðÞ¼ Š 2 V2 mŠ V2 tq ln Ce n ðÞCf n Š 1 ðÞ Te n ðފ Te n Š 1 ðÞ ð 4 Þ withthecombinedSiGD Q( n )[m3]offloodtide n Š 1andebb tide n attheendofeachtidalcycle,thespecificconductivity Ce( n )[ S/cm]attheendofebbtide n ,thespecificconductivity Cf( n Š 1)[ S/cm]attheendofthepreviousfloodcycle,andthe time Te( n )attheendoftheebbtide n ,aswellas Te( n Š 1)atthe endofthepreviousfloodtide n Š 1. TheEClevelsattheoutletofthebayweremeasuredattwo differentlevelstoaccountforstratification,andcombinedto provideanaverage. Thecombinedtimeserieswasusedtoextracttheminimum EClevelsattheendofeachebbtide( Ce)andthemaximum EClevelsattheendofeachfloodtide( Cf). Vmwascalculated usinglong-termtidalfluctuationsmeasuredatGalwayPort (MarineInstitute)andcombinedwith5-mbathymetrydata (GSI)toyieldameanvolumeofBellHarbourbayas2.65 millionm3.Accordingly, Vtwascalculatedasthestandard deviationoflong-termfloodmaximaandebbminima,yielding2.37millionm3.CatchmentwaterbalanceThecatchmentwaterbalancewasappliedaccordingtoEq. ( 5 ): P ¼ GWR þ R þ ET þ S ð 5 Þ withtheprecipitation P ,groundwaterrechargeGWR,runoff R ,evapotranspirationET,andchangeinstorage S .Since surfacerunoffispracticallynon-existentinthestudyarea, R wasconsideredtobezero.Changesinstorage(notatedbythe S )wasaccountedforbythedynamicwaterlevels/volumes inthesouthernturlough(T1).Thenorthernturlough(T2)was excludedfromthecalculation,asitsabsolutestoragevolume isnegligibleinthiscontext. 2634 HydrogeolJ(2018)26:2629 – 2647

PAGE 7

PipenetworkmodelBasedonthegeometricarchitectureofthediscreteconduit method(DCN;Thrailkill 1974 ;Jeannin 2001 ;Kovácsand Sauter 2007 ),apipenetworkmodelwas developedtorepresent thehydrogeologyofthecatchment.WhiletheDCNapproach onlyaccountsforgroundwaterflowinconduitsorinconnectionsbetweenfractures,thesemi -distributedpipenetworkmodelusedherealsoincludesadditionaldiffusegroundwaterrechargeininterceptioncatchment s.Multiplegroundwaterflow dynamicsandSiGDweremodelledusingInfoWorksICM®version7.0commercialsoftware(Innovyze,TweedHeads, Australia),guidedbyprinciplesoutlinedinGilletal.(2013 ) andMcCormacketal.( 2017 ).ThepipemodelresemblesdiffusegroundwaterrechargeinpermeablepipesmodelledaslaminarflowfollowingDarcy ’ slaw,combinedwiththeoutlineof majorconduitflowhorizonsconnectedtotheturlough(T1)on thesurface,consideringopenchannelandpressurisedflowhydraulicconditions.ThegoverningmodelequationsforconcentratedflowintheconduitsaretheSaint-Venantequationsof conservationofmassandmomentum,whiletheconveyance functionwasbasedontheColebrook – Whiteequation. Thecatchmentwasdividedintosevensub-catchments: threehigh-elevationsub-catchmentswithbareoutcrop,three low-lyingcatchmentswithasoilcover,andthewatersurface ofthebayoutlet. Meanelevationsofeachsub-catchmentinthemodelwere extractedfromtheimproveddigitalelevationmodel.Bydoing so,thepipenetworkmodelrealisticallyaccountsforhead differencesbetweenthelow-lyingcentreandthesurrounding escarpments,whichisbelievedtoaddadrivingforcetothe groundwaterflowdynamicsinthecatchment. Asmentionedearlier(see B Data ^ section),uniformrainfall asmeasuredatthetwositeswasappliedonallsub-catchments.ClimatevariableswerederivedfromthemeteorologicalstationatC1.Sealeveltimeserieswereintegratedintothe model,basedonvaluescontinuouslymeasuredatGalway Port(14kmnorthofthecatchment)bytheMarineInstitute. Themodelwascalibratedagainstestimatedgroundwater rechargerates(HunterWilliamsetal. 2013 ),waterlevelfluctuationsoftheturloughGortboyheen(T1),whichisafunction oftheheadintheunderlyingconduit,andestimatesofSiGD intoBellHarbourbay.Themodellingperiodwasfrom24 February2016to15July2017.ResultsBorehole(BH1)hydrographanalysisFigure 3 showsanexampleoftheambientborehole hydrographatBH1exhibitingarapidresponsetorainfall andmultiplerecessions.Itshouldbenotedthatthereareno pumpsclosetotheboreholethatcouldhaveanimpactonthe hydrograph. Duringthefirstrecessionfrom22Mayto10June2016, almostnorainfalloccurred.Strikingly,therecessionconsists ofmultipleconvexandconcavesections,whereasthetotal hydrographissplitintotwomainrecessions:above20.2masl [see(i)inFig. 3 a],andbelow20.2masl[see(ii)inFig. 3 a]. Below16.5masl,therecessionisclearlyinfluencedbytidal oscillation.Thelaggedcorrelationbetweenthetidalamplitude andtheboreholehydrographinthebeginningofJune2016 (Fig. 3 )showsanoscillationpatternof~6h,withadelayof ±2hattheboreholehydrograph. ThehydrographofBH1suggests: & Waterlevelsarehighlyfluctuating,rangingfrom24to9 masl. & Thewaterlevelrespondsveryrapidlytorainfallevents, confirmingthattheboreholeishydraulicallywellconnectedtotheaquifer. & Atleasttwodistinctmajorrecessionswithchangesfrom concavetoconvexsectionsarepresent — (1)>20.2masl, and(2)<20.2masl — suggestingtheexistenceofmultiple groundwaterflowhorizons. & Thewaterleveloscillatesinafrequencyof11 – 12hbelow 16.5masl.LaggedcrosscorrelationbetweenthetidaloscillationandBH1yieldsacorrelationcoefficientof+0.08 and Š 0.10,foralagof+3.5hand Š 2.5h,respectively, provingthatBH1isaffectedbythetidalfluctuationandis thereforeconnectedtothesea. & Headfluctuationisgovernedbyrainfallaswellasbythe overallpiezometriclevelintheaquifer,includingthelevel intheturloughT1. HydrographanalysisshowsthatBH1appearstobevery wellconnectedhydraulicallytothemainkarstaquifernetwork.Therecessionisdependentonthepiezometriclevelin theaquiferandintheturlough,clearlyvisibleintheperiod fromFebruarytoMay2016.Single-boreholedilutiontestsTwenty-threeconcentratedSBDTswereconductedinBH1on sixdifferentdaysunderdifferenthydrologicalconditions usingNaCl(125/62.5g/l)anddeionisedwaterastracer:campaigns1(11Feb2017),2(26Feb2017),3(4Mar2017),4 (23Mar2017),5(14Apr2017),and6(13Sep2017). Deionisedwaterwasusedtoruleoutthepossibilitythat downwardmigrationofNaCltracerwasdensity-driven. Table 1 summarisestheresultsof19successfultracertests, includingmeanflowvelocitiesandtracerrecovery,consideringtheprevailinghydrologicalconditions.Single-tracer breakthroughcurvesarepresentedinFig. 4 .Theprevailing flowdirectionduringrecessionswasdownwards(Fig. 4 a – c), HydrogeolJ(2018)26:2629 – 2647 2635

PAGE 8

whilejustatthepeakofarisinglimbfollowingarainevent, flowdirectionwasupwards(Fig. 4 d). Duringtracertestscarriedouton4March2017,the groundwaterflowdirectionwasupwards.Thelowrecovery rateof0.1 – 0.3%isaclearindicationofoutflowinggroundwateralongtheboreholesectionandassociatedlossoftracer. Meangroundwaterflowvelocitieswererelativelyhigh, reaching263m/h,confirmingthattheboreholemustbewell connectedtohighlytransmissivefractures/conduits.Along theverticalboreholesection,tracerwasrecoveredbetween thetopat23.4maslandthebottomat176mbsl. Ingeneral,thegroundwaterflowdirectionwasdownwards,andtracerrecoveryandmeangroundwaterflowvelocitydecreasedwithincreasingdepth,suggestingcontinuous groundwateroutflowalongthedepthoftheboreholeupto atleast176mbsl(wheretracerwasstillrecovered). Theresultshighlightthepresenceofmultiplehighlyconductivegroundwaterflowhorizons,rangingfrom23masltobelow 176mbsl.Itisthereforehypothes isedthatthegroundwaterflow regimeinthecatchmentislargel yinfluencedbyadeepconduit, presumablydraininggroundwaterasSGD.Surchargeofthis deepchannel,drivenbyheadsfromthesurroundingescarpments,temporarilyincreasesth eheadinBH1,activatingmore shallowconduitsanddrainingasSiGDintothebay.QuantificationofSiGDTheboreholedilutiontestresults(see B Single-boreholedilutiontests ^ section)indicatedtheexistenceofamulti-level conduitsystemdischargingtheflowsfromthekarstaquifer tothesea,withmoreintermittentSiGDintoBellHarbourbay viaashallowconduitsystemactivatedduringperiodsofhigh recharge.Hence,inordertomodelthegroundwaterflowdynamicsofthismulti-levelsystem,thebalancebetweendeep SGDandshallowSiGDneededtobeestimatedovervarying hydrologicalconditions. SiGDestimationwascarriedoutaccordingtothemethodologyinthe B QuantificationofSiGD ^ section,usingcontinuousECandtidalfluctuationmeasurementsinthebay. Figure 5 ashowsaplotoftheresultingestimatedaveraged dailySiGD[m3/s]intoBellHarbour,indicatingthattherelationshipbetweenSiGDandrainfalliscomplexandnon-linear. Inspring2016,SiGDreachespeaksofupto3m3/s, discharging5.8millionm3between24Februaryand28 April2016.Duringsummerandautumn2016,thedischarge intothebayclearlydropsandlargelyvariesbetweencloseto0 and0.6m3/s,whereevensignificantrainfalleventsdonot appeartocauseanincreaseinSiGD.Inwinter2016,SiGD startstoincreaseagain.Duringtheperiodfrom31May2016 to15July2017,thetotaldischargeaccountsfor12.7million m3,withapeakaveragedailydischargeof4.3m3/s.Figure 5 b showstheminimumandmaximumECvaluesforallebband floodtidesalongwithdailyevapotranspirationatC1usingthe Penman – Monteithequation.ThehigherECamplitudescorrespondtochangesinECbetweenebbandfloodtide,which translatesintohigherSiGD(Fig. 5 a).ETshowsatypicalseasonaltrend — thereisnoobviousindicationthatETinfluences ECamplitudes.Rather,ECamplitudesshowrapidsurges, suchasinNovember2016,whichmaybelinkedtotheoverall piezometricheadintheaquifer(Fig. 5 c). 2017 FebMarAprMayJunJulAugSepOct 10 15 20 25 30 35 40Head [masl]0 10 20 30 40 50 60Rain [ mm/d ] -505Lag (0.5 h) -0.1 -0.05 0 0.05 0.1CCC tidal impact 4 1 2 3 5 6 i) ii)(b) (a) Fig.3 a AmbientboreholerecessionatBH1anddailyrainfallbetween FebruaryandOctober2017,showingrapidresponsetorainfalland multiplerecessionswhichindicatedifferentreservoirs.Numbersand arrowsrefertotracerinjections(single-boreholedilutiontests).Forthe periodfrom17Julyto30September2017,rainfalldatafromthe MetEireannstationatCarronwereusedduetoadatagapatC1.Areas (i)and(ii)areseparatedbythe20.2-masllevel,whichformsasignificant breakinthehydrographrecession. b Correlogrambetweenthetidaloscillation(masl)andgroundwaterlevel(masl)ofBH1for30-min(0.5-h) timeseriesatlowlevel.Althoughthecross-correlationcoefficient(CCC) isrelativelylow,itdoesshowadelayandimpactoftidaloscillationonthe boreholehydrograph.Thehydrographthatisaffectedbythetideis highlightedinthedashedbox 2636 HydrogeolJ(2018)26:2629 – 2647

PAGE 9

Rainfallremainedfairlyconsistentthroughouttheperiod, withmeandailyrainfallof3.0,3.4,3.3,and3.3mminspring, summer,autumn,andwinter2016/2017,respectively,and 3.1mminspring2017;yetclearlytherateofSiGDdecreased duringthesummerandmuchofautumn2016.Thisobservationisconsistentwiththecontinuoussalinitymeasurementsin thebay,whichindicateverylowvariationinECbetweenthe seawaterenteringduringfloodtideandwaterleavingduring ebbtideinsummercomparedtospringandwinter.Thelow ratesofSiGDduringsummerandautumn2016suggestsubstantiallyreduceddischargefromtheintertidalspringsinBell Harbourbay.ThedischargepatternofSiGDintoBellHarbour fluctuatesconsiderably,suggestingthefunctioningofanoverflowspring.Thisinturnimpliesthatasignificantquantityof groundwaterfromthecatchmentmustalsobedrainingtothe sea,butbypassingBellHarb ourbay.Theresultsinthe B Single-boreholedilutiontests ^ sectionsuggestthatsuchbypassflowsarecarriedinmuchdeeperconduits. Time [h:min] 15:1015:2015:30 0 1000 2000 3000 4000TDS [mg/l] Time [ h:min ] 20:0000:0004:00 200 250 300 350 400TDS [mg/l] Time [h:min] 19:0021:0023:00 200 250 300 350 400TDS [mg/l] Time [ h:min ] 16:0516:0716:09 330 335 340 345 350TDS [mg/l] (a) (b) (d) (c) Fig.4 Totaldissolvedsolids (TDS)versustime. a Injectionof 500mlNaCl(62.5g/l)at4.3 mbsl,tracerrecoveryat43.7 mbsl; b injectionof500mlNaCl (62.5g/l)at4.3masl,tracerrecoveryat130.0mbsl; c injection of500mlNaCl(62.5g/l)at4.3 masl,tracerrecoveryat175.0 mbsl; d injectionof500mlNaCl (125g/l)at4.8masl,tracerrecoveryat18.4masl Table1 SummarytracerbreakthroughcurvesofsuccessfulSBDT conductedin2017(seeFig. 4 ),includinggroundwaterlevel,preceding rainfall(NA=notavailable)tracerinjectiondepth,tracersamplingdepth, tracerrecovery,andmeangroundwater(GW)flowvelocityandflow direction(upwards ordownwards ) ParameterDate 11Feb26Feb4Mar23Mar14Apr13Sep Head(masl) 21.925.1 28.1 25.225.224.5 Numberofinjections (NaCl/deionised) 1/05/0 6/2 1/12/01/0 Precedingrainfall 12h 024 1.6 01.4NA 24h 024.4 37.8 01.41.9 36h 031 53 2.61.4NA 48h 031.8 60 4.21.424.7 Tracerinjectiondepth[masl] Š 7 Š 4.3 +4.8 +4.8+5.1+4.8 Tracersamplingdepth(range)[masl] Š 16.9 Š 21.9to Š 43.7+18.4to+23.4 Š 44.7 Š 80.4 Š 129.4to Š 176 Tracerrecovery(range)[%] 99.845.8to49.60.1to0.389.820.09.4to25.6 MeanGWflowvelocity(range)[m/h]185.0146.3to156.7198.0to262.954.789.043.3to97.0 Flowdirection[ / ] HydrogeolJ(2018)26:2629 – 2647 2637

PAGE 10

Cross-correlationanalysisWhilethepreviousthreesectionsexplainedthecomplex patternoftimeseriesfromasinglesamplinglocation, cross-correlationwasusedto linkindividualtimeseries inordertodefinecommonlinearrelationshipsaswellas timelagsbetweeninputtimeseries(e.g.rainfall)andoutputtimeseries(e.g.head).Cross-correlationcoefficients (CCC)werecalculatedbetweenhourlyanddailytime series,withtheresultsmappedinFig.6 .SinceSiGD wasestimatedpertidalcycle,respectivetimeseriescould onlybecross-correlatedformeandailyrecords.Allother variableswerecross-correlatedusinghourlytimeseries. TheresultsshowalargevarianceinCCCsandlags,rangingfrom Š 0.21to+0.87,and Š 54hto+5days, respectively. NegativelagsarecalculatedfortheCCCbetweenrainfall andBH2(EC)andrainfallandBH1(T).Becausetherespectiverelationshipsareone-directional,thenegativelagssuggest thatotherprocessesareinvolved(Juki andDeni -Juki 2015 ),wherenegativelagsfortheCCCbetweenrainfalland ECandTingroundwaterarenotuncommon[e.g.Lietal. ( 2017b )]. ThewaterlevelsinBH1andBH2respondtorainfallwitha positivelagof25and49h,respectively;thesignificantlower lagtimeofBH1suggeststhattheboreholeisbetterconnected totheaquiferthanBH2.BH1andBH2arehighlycorrelated, indicatedbyaCCCof+0.78andalagof Š 2h,underlyingthe factthatthelowlandBH1respondsbeforetheuplandBH2. WhileBH2showsonlyaminorcorrelationwithwaterlevels inT1,waterlevelsinBH1andT1arehighlycorrelated (CCC=+0.84)byalagof0h.Thisisnotsurprising,as BH1andT1are~900mapart,anditindicatesthat,indeed, BH1mustbeverywellconnectedtothekarstaquifer,andits conduitsystemisalsoconnectingtheturlough. InadditiontothestronglinearrelationshipbetweenBH1 andT1,highCCCvalueswerecalculatedbetweendailytime seriesforBH1andT1andestimatedSiGDintoBellHarbour bay,accountingforaCCCof+0.51withalagof0days,and even+0.87withalagof+1day,respectively.Thedifference inlagmaybeexplainedbytheinertiaofthesystem:whilethe headinBH1isadirectfunctionofheadwithintheaquifer,the waterlevelinT1doesnotrespondlinearlytothefluctuationin headwithintheaquifer.Infact,theturloughstoragehasmuch higherinertia,whichisrelatedtothedepth – volumerelationshipoftheturloughstorage,wherebyanincreasingwater 0 1 2 3 4 5 6GW discharge [m3/s]0 10 20 30 40 50 60Rain [ mm/d ] GW discharge Rain1 2 3 4 5 6EC [S/cm]1040 2 4 6 8ET [mm/d]ECETA p r 2016Jul 2016Oct 2016Jan 2017A p r 2017Jul 2017 10 20 30Head [masl](c) (b) (a) Fig.5 a EstimatedSiGD(m3/s) atBellHarbourbayandrainfall (mm/day).Duetoinstrument malfunction,adatagapoccurred fortheperiod28Aprilto31May 2016. b Minimumandmaximum electricalconductivityvalues (EC, S/cm)ofebbandfloodtide anddailyevapotranspiration(ET, mm)asmeasuredatC1. c Hourly piezometrichead(logscale)at BH1(masl) 2638 HydrogeolJ(2018)26:2629 – 2647

PAGE 11

depthpromotesanincreasingsurfacearealinkedtothetopography;thevolumeofwaterintheturloughisthereforenota linearfunctionofitswaterdepth.However,thisresultisparticularlyinteresting,asitsuggestsastronglinearrelationship betweentheheadinthekarstaquiferandtheregimeofSiGD intothebay. Interestingly,CCCishighbetweenBH2andBH1,and betweenBH1andT1,butnotbetweenBH2andT1.This behaviourisinterpretedasindicatingthatBH1integrates twodifferentdynamicsthatmaygoverntheheadintheaquifer:aslow-respondingdynamic(overallpiezometricstatein thelowland)andafast-respondingdynamic(rechargedynamicintheupland).Thisinterpretationwouldexplainthelower lagbetweenrainfallandBH1(25h)asopposedtorainfalland BH2(54h).WaterbalanceWaterbalanceshavebeenusedtomatchestimatedSiGDto groundwaterrechargeinthecatchmenttoquantifytheshareof groundwaterthatappearstobypassBellHarbourbayviaa deeperconduitsystembeforebeingdischargedasSGDfurtheroutintothesea. Waterbalancesforthecatchmentwereestablishedacross twoperiodsofdifferingrechargecharacteristics,i.e.winter (24Febto28Apr2016,and1Nov2016to31Mar2017, Fig. 7 a,b)andsummer(1Julto31Oct2016,and1Mayto16 Jul2017,Fig. 7 c,d),distinguishingbetweeninput(rainfall andturloughstorage)andoutput(evapotranspiration,SiGD intoBellHarbourbay,andunaccountedresources).During thewinterperiod,thecatchmentreceivesmorerainfalloverall, C1 BHB1+2 T1 BH1 BH2 SiGD-D +0.51 (0 d) SiGDP +0.19 ( 5 d ) D P +0.15 (+2 5 h) EC-P 0 .10 (+9 h) D-P + 0.21 (+49 h) EC-P -0. 2 1 ( 5 4 h) D D + 0.78 ( -2 h ) D-D +0.84 ( 0 h) DD 0 . 0 8 ( -9 h ) GW piezometer GW turlough SW bay DTM [masl] 343.3 SiGD-D +0.87 (1 d)ClimateLegend-1.4 01 , 5 0 03 , 0 0 0 750 m Fig.6 Cross-correlation coefficientswithlagtimes(h/day) inbracketsfortimeseriesof rainfall(C1),boreholedata(BH1, BH2),turloughlevel(T1),and estimatedSiGD(BHB);D= depth,EC=electricalconductivity,P=rainfall,SiGD=submarine intertidalgroundwaterdischarge, T=temperature.Arrowsindicate thedirectionfrominputtooutput connectingthetwosamplingsites thattheCCCandlagtimereferto. DTMisthedigitalterrainmodel (topography) HydrogeolJ(2018)26:2629 – 2647 2639

PAGE 12

whichpartiallyrechargesturloughT1.SiGDaccountsfor53 and29%oftotalrainfall,andaccordingly,therateofunaccountedresourcesmissinginthebudgetsumsto32and50% oftotalrainfall. Duringthesummerperiod,rainfalldeclinesslightly,but moststrikingly,therateofSiGDdropsto8%oftotalrainfall forbothsummerperiods.Accordingly,therateofunaccountedresourcesis68and69%oftotalrainfall. Theresultsunderscorethelargemismatchbetweentotal groundwaterrechargeandSiGDforthetwoseasons.The mostplausibleexplanationbasedonthisresearchinthe hydrogeologicalcontextisthatasignificantshareofgroundwatermustleavethecatchmentbypassingBellHarbourbay viaadeeperconduitflowsystem,drainingthecatchmentcontinuouslyasSGDintotheAtlanticOcean.Ahigherrateof groundwaterbypassesthebayinsummer(68 – 69%)thanin winter(32 – 50%). Itshouldbenotedthatthewaterbalanceestimationswere calculatedfortheminimumsizeofthemostprobablecatchmentboundariesof50km2.Anyincreaseinthecatchment boundaries,upto13km2larger,asdisplayedinFig. 1 ,will increasetotalrainfallandETinthecalculation;however, SiGDinthecalculationwouldremainconstant,resultingin anincreaseinthecontributionforunaccountedresources.PipenetworkmodelTheresultspresentedpreviouslyledtothehypothesisofthe functioningofbothcontinuousSGDviaadeepconduitinto theAtlanticOcean,andtemporalSiGDintoBellHarbourbay viaashallowoutflow.Toverifythepotentialfunctioningof thisflowregime,apipeflowmodelwasdevelopedforthe groundwatercatchmentbasedonthefollowingfindings(as outlinedinprevioussections): & Thereisnostronglinearrelationshipbetweenthemeasuredheadintheaquiferandrainfallinput,asindicated bythelowCCCbetweenrainfallandBH1andBH2. & TheCCCbetweenrainfallandBH1andBH2,respectively,andCCCbetweenBH2,BH1,andT1,respectively, suggestthatatleasttwodifferentdynamicsarepresent, drivingheaddifferencesbetweenthelowlandandthesurroundinghigherescarpments. & SiGDvariesthroughouttheyear,showingapartlylinear relationshipwiththewaterlevelmeasuredatBH1anda strongrelationshiptoT1,providingevidencethatthedischargepatternatleastpartlyfollowstheheadintheaquifer.Duringperiodsofhighoverallpiezometricstateofthe aquifer,SiGDisactivatedthroughanoverflow mechanism. & Thewaterbalanceforthegroundwatercatchmentsuggeststhat32 – 69%oftotalrainfallisunaccountedfor, includingseasonalvariation;itisbelievedthisshareof waterdrainsthecatchmentviaSGD,bypassingBell Harbourbayviaadeepconduit. & TheboreholehydrographrecessionofBH1indicatesthe existenceofmultiplereservoirsandtheimpactoftidal oscillationontheboreholehydrographatdepthsbelow 16.5masl.Verticalgroundwaterflowdirectionsaregenerallydownwards,reachingbelow176mbsl,whileflow directionchangestoupwardsduringheavyrainevents, drivenbyhigherheadsfromthesurroundingescarpments. Themulti-levelkarstgroundwaterflowsystemofBell Harbourwasthusmodelledbyatwo-levelpipenetworkmodelconsistingof(1)ashallownetworkthatrangesbetween0 and10masl,dischargingperiodicallyasoverflowSiGDinto BellHarbourbay,and(2)adeepconduitsystemat~65mbsl (~95mbgl)bypassingBellHarbourbayanddischargingas SGD.Intheabsenceoftransmissivitydataforthesingle groundwaterflowlevels,simplificationofthesystemtotwo levelsseemstoachievethemostparsimoniousmodel.Both levelsarehydraulicallyconnectedwiththeconceptualmodel oftheconduitnetworkoutlinedinFig. 8 . Thesouthern,upperpartofthecatchmentwascalibrated againsthourlywaterleveltimeseriesoftheturlough,T1 (Fig. 9 ).Duetorestrictedlandaccess,dataforT1wereonly availableuntilJuly2016. Themodelsimulationwasin itiatedduringpeakwater levelsinJanuary2016,afterthebeginningofavailablehourly rainfalldata;therefore,theperiodforwhichT1wascalibrated couldnotextendbeyondJanuarytoJuly2016.Nevertheless, theresultshowsaverygoodfitbetweenmodelledwaterlevels andobservedlevels,withaNash – Sutcliffe(NS)coefficientof +0.99. ThemodelleddischargeoftheshallowoutflowwascalibratedagainstpreviouslyestimatedSiGDtimeseries,which importantlyinvolvesdifferenttemporalresolution,i.e.(1)the model(1h),and(2)theestimationofSiGDaccordingtothe occurrenceofminimum/maximumECrecordsofthetidal cycle(11 – 14h).Thegraphicalanalysisshowsthatthemodel iscapableofmodellingtheoveralldischargepattern:higher dischargesinspring2016,lowdischargeduringthesummer andautumnof2016,increasingdischargeinwinter2016,and againlowerdischargeinsummer2017.Furthermore,thedeclineinspring2016correlateswiththedecreasingheadinT1. TheabsolutequantityofmodelledSiGDmatchestheabsoluteestimatedSiGD(Fig. 10 ):duringperiod1(24Febto28 Apr2016),estimatedSiGDwas5.9millionm3,whilesimulatedSiGDis6.0millionm3;inperiod2(31May2016to15 Jul2017),estimatedSiGDwas13.0millionm3,whichis matchedbyasimulatedSiGDof12.9millionm3.Thereisa clearmismatchinJanuary/February2017,whichcannotbe directlyexplained.However,theinstrumentationhadbecome encrustedbythattime,whichmayhavetrappedhigher-EC water,makingitpotentiallylessresponsive.Further,itcan 2640 HydrogeolJ(2018)26:2629 – 2647

PAGE 13

InputOutput 0 5 10 15 20 25 30 24 Feb 2016 to 28 Apr 2016million m3InputOutput 0 5 10 15 20 25 30 01 Nov 2016 to 31 Mar 2017million m3InputOutput 0 5 10 15 20 25 30 01 Jul 2016 to 31 Oct 2016million m3InputOutput 0 5 10 15 20 25 30 01 May 2017 to 16 Jul 2017million m3Winter Summerc) a) b) d)GWR Turlough Rainfall ET GW discharge into the bay Unaccounted resources GWR Turlough Rainfall ET GW discharge into the bay Unaccounted resources Rainfall ET GW discharge into the bay Unaccounted resources Rainfall ET GW discharge into the bay Unaccounted resources Fig.7 WaterbalancesofBell Harbourbayfor( a,b )winterand ( c,d )summerperiods(million m3) 118 masl 1 maslDeep conduit / SGD shallow SiGD53 masl 118 masl 43 masl 65 maslT1168 masl Legend Sub-catchment Permeable pipe (Darcian flow) Empty conduit (turbulent/ open channel flow) Reservoir (turlough) Outfall (spring) Poll Gonzo~1,000 m~100 m BH2 BH1T1 SiGD1 BHB 1 & 2 “Deep GW flow horizon” Hydrogeological concept domain Model concept domain N Fig.8 3Dconceptualmodelofthehydrogeology,includingconduit networksthatresemblethoseinthepipenetworkmodel:permeable pipes(DarcyFlow)modellingdiffuserechargeandemptypipes(open channelandturbulentflow)modellingconduitflow.Yellowboxes conceptuallyrepresentsub-catchmentswithmeanelevations HydrogeolJ(2018)26:2629 – 2647 2641

PAGE 14

behypothesisedthatabnormalcurrentsstronglyinfluenced thesalinityatthemonitoringsite. Comparisonoftimeseriesofdifferenttemporalresolutions yieldsalossinNScoefficientperformance,especiallywhen parametersderivedfromalongertimestepwereusedfor simulatingflowwithashortertimestep(Bastolaand Murphy 2013).Therefore,theoverallmodelperformance wasevaluatedusingaveragemonthlySiGDrates,which yieldsanNScoefficientof+0.74,whichcanbeconsidered B good ^ (Moriasietal. 2007 ). Further,thegoodnessoffitofthemodelresultswasevaluatedusinganautocorrelationfunction(Labatetal. 2000 ; Fig. 11 ).Overall,thelossinmemoryisverysimilarwhen comparingtheestimatedandmodelledSiGD,indicatingreasonablemodelperformance. Whilethemodelledturloughlevelachievesagoodfitwith theobservedwaterlevel,themodelperformanceisweaker betweenthemodelledSiGDandestimatedSiGDtimeseries. Thisisnotasurprise,sincethetidalprismmethodofestimatingSiGDissusceptibletoerrorsaccordingtosomeofthe assumptionsmade,aswellasthecalculationtimestepofeach B tidalcycle ^ of~12hactingtosmoothoutdischargepeaks occurringinbetween.However,mostimportantly,themodel simulatesthegeneraldischargepatternandsingledischarge peakswell,inadditiontoaccuratelypredictingtheestimated SiGDoverbothperiods.DiscussionThisstudypresentsasuiteofmethods,combinedtocreatea novelapproachforquantifyingSiGD,andfurtherfor explainingthefunctioningofacomplexcoastalkarstaquifer inIreland.Inthisway,thestudyprovidesclearevidencefor theexistenceofamulti-levelcoastalkarstaquiferwithvery complexflowdynamics. HydrographanalysisofBH1clearlysuggeststheexistence ofatleasttwosub-systemsandthehydraulicconnectionof BH1withinthelimestoneaquifer,i.e.above20.2masland below20.2masl.Thelattersub-systemisinfluencedbythe tidalfluctuations,provingconnectiontothesea.Thegradient betweentheonshoregroundwaterlevelatBH1andtheseais Sep 2015Nov 2015Jan 2016Mar 2016Ma y 2016Jul 2016 15 16 17 18 19 20 21 22 23 24 25 26Head [masl] Head (observed) Head (modelled) Fig.9 Modelledvs.observedturloughlevel(masl) 0 1 2 3 4 5 6GW discharge [m3/s] 0 10 20 30 40 50 60Rain [ mm/d ] GW discharge estimated GW discharge modelled Rain Apr 2016Jul 2016Oct 2016Jan 2017Apr 2017Jul 2017 0 5 10 15Cum. GW discharge [mill. m3] Cumulated GW discharge estimated Cumulated GW discharge modelled (b) (a) Fig.10 a ModelledSiGDvs. estimatedSiGDintoBellHarbour bay(m3/s),and b cumulative modelledSiGDvs.cumulative estimatedSiGDintoBellHarbour bay(millionm3)between16July 2016and15July2017 2642 HydrogeolJ(2018)26:2629 – 2647

PAGE 15

sufficienttoexplainSGD:thelowestpotentialinBH1reaches 9masl,whichmeansthat,accordingtotheGhyben – Herzberg relation,thefreshwatercolumnatBH1mayreachathickness of360mabovepotentialseawater(Fleury 2005 ).Thebase levelofthelowersub-systemisnotknown;however,SBDTs proveverticalgroundwaterflowbelow176mbsl.Basedon theSBDTs,groundwaterflowispredominantlydownwardin BH1,withthegradientreversingonlyduringfloodpeaks. Thisdynamicisbelievedtoresemblethemainprinciplesof theaquifer:itiscontinuouslydrainedviadeepoutlets dischargingasSGDintotheocean.Theoccurrenceofrain eventscausesarapidincreaseinthepotentialheadlevelsin theaquifer(Fig. 5 c),yet,accordingtotherelativelylowCCC, rainisnottheonlyvariabledrivingheadfluctuations.Thelow CCCbetweenthepiezometricheadinBH1andrainfallis believedtoresultfromtheinfluenceofdifferenthydraulic conductivities/transmissivities(heterogeneity)withinthe aquifer,aswellasofthetopography,i.e.thesurrounding escarpments.Groundwaterrechargeinareasofhigherelevationisthoughttobethedrivingforcefortheshort-term (~hours/days)increaseinheadsinBH1,resultingtemporarily inupwardgroundwaterflowinBH1andfloodingofT1.This behaviourwassimulatedusingapipenetworkmodelresemblingatwo-levelconduitsystem,withacontinuousoutflowas SGDandanintermediateoverflowasSiGD.Themodelis capableofsimulatingtheimpactofhigherheadsoriginating fromthesurroundingescarpmentstomodelgroundwater/ surface-waterinteractions(i.e.turloughlevelfluctuations) andtheseasonalpatternofSiGDintoBellHarbourbay.The patternofSiGDintothebayshowsalinearrelationshipwith theheadintheaquifer,asmeasuredatBH1andT1. Thefindingsregardingthefunctioningofthecoastal karstaquiferraisequestionsandmayberelevantforother studiesinIrelandandworldwide.Thedevelopmentof coastalkarstsystemsmayberelatedtoglaciationandglacialmaxima,sealevelchanges,andtectonics,suchasthe large-scaleeventoftheMessiniansalinitycrisiswhich loweredtheleveloftheMe diterraneantoasmuchas 1500mbelowthepresentlevel(Doerflingeretal. 2009). Aresultofthatmaybecoastalkarstaquiferswithmultiple outletlevels,suchasPort-MiouintheMediterranean (Fleuryetal. 2007),whichconsistsofadeepbrackishconduitreservoirupto223mbslandashallowfreshwaterreservoir(ArfibandCharlier 2016). Asregardsthelastglacialmaximum(LGM),26kaago,the relativeseawaterlevelattheIrishcoasthasbeenmodelledto 80 – 100mbsl(EdwardsandCraven 2017 ),whilelocally,at GalwayBay,arelativesealevelwasmodelledtohave droppedbetween60and68mbsl15kaago(O ’ Connelland Molloy 2017 ).Olderglaciationofcomparablemagnitudeto theLGMoccurredduringthelatePalaeozoiciceage,with onsetindicatedduringtheLateViséan(Barhametal. 2012 ). IncomparisontotheMediterranean,thesystemofBell HarbourintheBurrenseemstobespecial,indicatingmultilevelconduitsatdepthsgreaterthanpreviouslyreportedfor thearea(Smyth 2000),reachingbelow176mbsl. Interestingly,thisdepthexceedsthedeepestsealeveldrop sincethePliocene.Therefore,theremaybeotherfactorsthat haveinfluencedtheformationofdeepconduits,suchas palaeokarstfeatures,whichmaybelinkedtothePalaeozoic iceageandloweringofsealevels,orpotentiallydissolved mineralveinspresentinthecatchment,asalsosuggestedby BunceandDrew( 2017 ).PotentiallocationsofSGDmaybe foundintheAtlanticOceanoffshoreatBallyvaughan,asillustratedbyMullan( 2003 ). ThepresentstudyintroducesanoveltechniqueforestimatingSiGDintheformofapollutionflushingmodel(Barber 2003 ;BarberandWearing 2004 ),whichwassupportedbyan extensivemonitoringnetworkthatallowstimesseriesanalysis,waterbudgetestimations,andsingle-boreholedilution tests,tolinkonshorecatchmentdynamicstotheoffshoredischargepattern.EstimationofSiGDischallengingandisoften reducedtosinglespotmeasurementsorshortperiodsofcontinuousobservations(inthecaseofradon).However,this studyemphasisestheneedtoestablishlong-termtimeseries thatenableestimationofSiGD.Withoutsuchanapproach,a highlyfluctuatingdischargeregimecannotbeassessed. Therefore,thisstudyusesthenaturaltracersalinity,representedasEC,toquantifytheSiGDintoBellHarbour.Thiswas previouslyestimatedbyPerriquet( 2014 )usingatidalprism approachdevelopedbyCaveandHenry( 2009 ).Interestingly, thelowestestimateddischargeinthecurrentstudymatches thatinthepreviouswork,whichwasfoundtobecloseto 0m3/sinMay2011.However,maximumdischargewasestimatedat~23m3/spertidalcycle,whichismorethanfive timestheestimatedpeakdischargeinthepresentstudy.It hasbeensuggestedthattheapproachdevelopedbyCave andHenry( 2009 )yieldshighlyexaggeratedestimatesof 020406080100 0.00.20.40.60.81.0 ACFLa g ( 1 d ) Modelled Estimated Fig.11 Autocorrelationfunction(ACF)ofmodelledvs.estimatedSiGD HydrogeolJ(2018)26:2629 – 2647 2643

PAGE 16

dischargerates(McCormacketal. 2014 ),whichthepresent studywouldseemtocorroborate. Bydefinition,apipenetworkmodelisasimplificationof thereality,whileresultssuggestthatdynamicsofSiGDare drivenbyamulti-levelconduitsystemwithpresumablydifferenttransmissivities,representedbyatwo-layersemi-distributedconduitsystem.Thissimplificationmaybeonefactor explainingtheNScoefficientof+0.74.Anotherfactorrelated tomodelperformanceisthetimestepof1h.Themodelwas calibratedagainsthourlyturloughwaterlevelsaswellasthe SiGDestimatedpertidalcycle,representedbytheperiodbetweentheoccurrenceofminimumECduringebbtideand maximumECrecordsduringfloodtide,rangingfrom11.5 to14h.Therefore,themodelwascalibratedagainsttime seriesofdifferenttemporalresolutions,reducingthemodel performance(BastolaandMurphy 2013 ).Toaccountatleast inpartforthis,themodelwasfirstcalibratedtofitthe1-h turloughlevelfluctuationsandwasthenfurthermodifiedtofit theestimatedSiGDtimeseries. ThewaterbalanceapproachhasbeenusedfortheestimationofSGDinmanyplaces;yettheapproachistypically impreciseforSiGDestimations,becauseuncertaintiesassociatedwithvaluesusedinthecatchmentsareoftenofthesame magnitudeasthedischargebeingevaluated(Burnettetal. 2006 ).Forexample,thedelineationofthecatchmentboundariesisacriticalaspectinkarstifiedcatchments(Marganeet al. 2018 ).Additionaluncertaintyisaddedbyusingspatially uniforminputdata,suchasrainfall,evapotranspiration,or salinityinthebay.However,inthisstudy,thewaterbalance approachisincorporatednottoquantifytheSiGDintoBell Harbourbay,butrathertoargueinfavouroftheexistenceof SGDonabasinscaleandtoprovideanestimateofdeep groundwaterflow.ConclusionsPreviously,thenorth-easternpartofthelimestoneBurrenplateauwasconsideredtodrainalmostexclusivelyviaSiGDinto BellHarbourbay(Perriquet 2014 ;McCormacketal. 2017 ). Thecurrentstudysuggeststhat,infact,thedischargepattern intothebayisdrivenbyacomplexoverflowmechanism, dependingontheoverallpiezometricpotentialinthekarst aquifer.Ithasonlybeenpossibletoinvestigatethisoverflow mechanismusingasetofdifferentmethodsthathavenotbeen appliedbeforeinthiscontext,aswellaslong-termmeasurementsofadenseonshoreandoffshoremonitoringnetwork: & Theexistenceofmultiplegroundwaterflowhorizonswas indicatedbyasimpleboreholehydrographanalysisshowingdifferentindividualrecessions,thelowestofwhichis clearlyinfluencedbytidaloscillation,provingaconnectionbetweenBH1andthesea. & Theexistenceofdeepgroundwaterflowwasprovento existusingSBDTinBH1thathighlightedgroundwater outflowfromtheboreholebelow176mbsl,aswellas relativelyhighflowvelocitiesaccordingtothehydrologicalregime. & Cross-correlationanalysisestablishedCCCvaluesof +0.51and+0.87betweentheheadinBH1andSiGD, andtheheadinT1andSiGD,indicatingthepartially linearrelationshipbetweentheheadintheaquiferand theSiGDregime. & Waterbalanceswereestablishedbasedonveryconservativecatchmentboundaries,indicatingaclearseasonal trendintheSiGDregime:duringthetwowinterperiods, SiGDaccountedfor29and53%,whileduringthetwo summerperiods,SiGDaccountedfor8%only.Thisseasonalpatternreflectstheroleoftheoverallpotentiometric headoftheaquiferdrivingthedischargeregimeintoBell Harbourbay.Further,theresultsindicatethatalargeshare ofthegroundwatermustdrainthecatchmentviaunknown springlocations(32 – 69%oftotalrainfall). & SiGDwasestimatedpertidalcycleusingarelativelysimplemass-balancetidalprismapproachbasedoncontinuousmeasurementsofECforaperiodgreaterthan1year. TheresultingSiGDisadirectfunctionofdifferencesin ECbetweenthefloodandebbtide,showingfluctuations inSiGDintoBellHarbourbaythroughouttheyear. ThepipenetworkmodeliscapableofsimulatingtheperiodicregimeofSiGDcharacteristicofanoverflowspring,asa functionofthepotentiometricheadintheaquiferand matchingtheoverallestimatedSiGDforbothperiodsofavailabledata.Nevertheless,amismatchbetweenestimatedand modelledSiGDinJanuaryandFebruary2017raisesthequestionofpotentiallyunidentifiedflowpathsthatmustbe accountedforintheconceptualunderstandingofthe catchment. Thisstudyhighlightstheimportanceoflong-termmeasurementsfortheassessmentof(coastal)karstaquiferswithcomplexorperiodicdischargeregimes.Further,itisarguedhere thatinthecontextofcoastalaquifers,differentdischargelocationsmayneedtobeconsideredinhydrogeologicalassessmentsinordertoenablereliablequantificationofthetotal discharge,aswellotherparametersofinterest(suchasnutrients),intothecoastalzones.Acknowledgements ThisresearchwasconductedwithintheIrishCentre forResearchinAppliedGeosciences(ICRAG),supportedinpartbya researchgrantfromScienceFoundationIreland(SFI)underGrant Number13/RC/2092,andisco-fundedundertheEuropeanRegional DevelopmentFundandbyiCRAGindustrypartners.Furtherfinancial supportwasprovidedbyabursaryfromtheInternationalAssociationof Hydrogeologists(IAH,IrishGroup). SincerestthankstoColinBunceandtheteamoftheBurrenOutdoor andEducationCentre(BOEC)inTurlough,Co.Clare,forhostingthe weatherstation,sitework,anddatacollection.Further,theauthorsthank 2644 HydrogeolJ(2018)26:2629 – 2647

PAGE 17

thetechniciansPatrickVeale,Dr.KevinRyan,andDavidMcAulayfrom theDepartmentofCivil,StructuralandEnvironmentalEngineering (TrinityCollegeDublin)fortechnicalsupport,aswellasSaraMakdessi forillustrations. Theauthorsareverygratefulforthecommentsofthetwoanonymous reviewers.OpenAccessThisarticleisdistributedunderthetermsoftheCreative CommonsAttribution4.0InternationalLicense(http:// creativecommons.org/licenses/by/4.0/),whichpermitsunrestricteduse, distribution,andreproductioninanymedium,providedyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinktothe CreativeCommonslicense,andindicateifchangesweremade.ReferencesAllenRG,PereiraLS,RaesD,SmithM(1998)Cropevapotranspiration — Guidelinesforcomputingcropwaterrequirements.FAO Irrigationanddrainagepaper56,FoodandAgriculture OrganizationoftheUnitedNations,Rome AngeliniP(1997)Correlationandspectralanalysisoftwo hydrogeologicalsystemsinCentralItaly.HydrolSciJ42:425 – 438. https://doi.org/10.1080/02626669709492038 ArfibB,CharlierJ-B(2016)Insightsintosalineintrusionandfreshwater resourcesincoastalkarsticaquifersusingalumpedrainfall – discharge – salinitymodel(theport-Mioubrackishspring,SEFrance). JHydrol540:148 – 161. https://doi.org/10.1016/j.jhydrol.2016.06. 010 BarberRW(2003)Analyticalmodelingofpollutionflushinginwellmixedtidalembayments.In:LakhanVC(ed)Elsevier OceanographySeries.Elsevier,p467 – 489 BarberRW,WearingMJ(2004)Asimplifiedmodelforpredictingthe pollutionexchangecoefficientofsmalltidalembayments.WaterAir SoilPollutFocus4:87 – 100. https://doi.org/10.1023/B:WAFO. 0000044789.67230.3e BarhamM,MurrayJ,JoachimskiMM,WilliamsDM(2012)Theonsetof thePermo-Carboniferousglaciation:reconcilingglobalstratigraphic evidencewithbiogenicapatite 18OrecordsinthelateVisean.J GeolSoc169:119 BastolaS,MurphyC(2013)Sensitivityoftheperformanceofaconceptualrainfall – runoffmodeltothetemporalsamplingofcalibration data.HydrolRes44:484 BoxGEP,JenkinsGM(1976)Timeseriesanalysisforecastingandcontrol.Holden-Day,SanFrancisco BunceC(2010)PollGonzo.IrSpeleol19:16 – 21 BunceC,DrewD(2017)WatertracingintheCarrandepressionand RiverFergusvalley,Burren,Clare.IrishGroundwaterNewsletter. GeologicalSurveyIreland,Dublin,p13 – 18 BurnettWC,AggarwalPK,AureliA,BokuniewiczH,CableJE,Charette MA,KontarE,KrupaS,KulkarniKM,LovelessA,MooreWS, OberdorferJA,OliveiraJ,OzyurtN,PovinecP,PriviteraAMG, RajarR,RamessurRT,ScholtenJ,StieglitzT,TaniguchiM, TurnerJV(2006)Quantifyingsubmarinegroundwaterdischarge inthecoastalzoneviamultiplemethods.SciTotalEnviron367: 498 – 543. https://doi.org/10.1016/j.scitotenv.2006.05.009 CarrMR,WinterTC(1980)Anannotatedbibliographyofdevicesdevelopedfordirectmeasurementofseepage.U.S.GeologicalSurvey, Open-FileReport80-344,37pp CaveRR,HenryT(2009)Groundwaterinputstothecoastalzoneof SouthGalwayandClare,westcoastofIreland.Paperpresentedat theASLOAquaticSciencesMeeting2009,Nice,France CaveRR,HenryT(2011)IntertidalandsubmarinegroundwaterdischargeonthewestcoastofIreland.EstuarCoastShelfSci92: 415 – 423. https://doi.org/10.1016/j.ecss.2011.01.019 CorbettDR,CableJE,ShinnEA,ReichCD,HickeyTD(2003)Seepage metersandadvectivetransportincoastalenvironments:comments on B SeepageMetersandBernoulli ’ sRevenge ^ byE.A.Shinn,C.D. Reich,andT.D.Hickey.2002. B Estuaries ^ 25:126-132.Estuaries 26:1383-1387 DoerflingerN,FleuryP,BakalowiczM,EkmekciM(2009)Specificities ofcoastalkarstaquiferswiththehydrogeologicalcharacterisationof submarinesprings – overviewofvariousexamplesinthe Mediterraneanbasin.Sustainabilityofthekarstenvironment. PaperpresentedattheInternationalInterdisciplinaryScientific Conference,PlitviceLakes,Croatia,23 – 26September2009 DongZ,LiuD,KeesingJK(2010)JellyfishbloomsinChina:dominant species,causesandconsequences.MarPollutBull60:954 – 963. https://doi.org/10.1016/j.marpolbul.2010.04.022 DrewD(1990)ThehydrologyoftheBurren,countyClare.IrGeogr23: 69 – 89 DrewD(2003)ThehydrologyoftheBurrenandoftheClareandGalway Lowlands.In:MullanG(ed)CavesofCountyClareandSouth Galway.UniversityofBristolSpelæologicalSociety,Bristol,p 31 – 43 EdwardsR,CravenK(2017)Relativesea-levelchangearoundtheIrish coast.In:CoxonP,McCarronS,MitchellF(eds)AdvancesinIrish QuaternaryStudies.AtlantisPress,p181 – 215 EPA(2004)Guidanceontheassessmentofpressuresandimpactson groundwaterdependantterrestrialecosystems,RiskAssessment SheetGWDTERA2a – Risktoturloughsfromphosphate.Water FrameworkDirectivePressuresandImpactsAssessment Methodology,GuidanceDocumentGW9.WorkingGroupon Groundwater,SubcommitteeonTurloughs. FieldMS(2002)TheQtracer2programfortracer-breakthroughcurve analysisfortracertestsinKarsticaquifersandotherhydrologic systems.USEnvironmentalProtectionAgency,Washington,DC FleuryP(2005)Sourcessous-m arinesetaquifèrescôtiers méditerranéens.Fonctionnementetcaractérisation(Submarine springsandMediterraneancoastalaquifers.Functioningandcharacterisation).Doctoralthesis,UniversityofParis FleuryP,BakalowiczM,deMarsilyG(2007)Submarinespringsand coastalkarstaquifers:areview.JHydrol339:79 – 92. https://doi.org/ 10.1016/j.jhydrol.2007.03.009 FreezeRA,CherryJA(1979)Groundwater.Prentice-Hall,Englewood Cliffs,NJ GallagherSJ,MacDermotCV,SomervilleID,PrachtM,SleemanAG (2006)Biostratigraphy,microfaciesanddepositionalenvironments ofupperViséanlimestonesfromtheBurrenregion,CountyClare, Ireland.GeolJ41:61– 91. https://doi.org/10.1002/gj.1033 Gi llLW,NaughtonO,JohnstonPM(2013)Modelinganetworkof turloughsinlowlandkarst.WaterResourRes49:3487 – 3503. https://doi.org/10.1002/wrcr.20299 GillespiePA,WalshJJ,WattersonJ,BonsonCG,ManzocchiT(2001) ScalingrelationshipsofjointandveinarraysfromtheBurren,Co. Clare,Ireland.JStructGeol23:183 – 201. https://doi.org/10.1016/ S0191-8141(00)00090-0 GreenL,SutulaM,FongP(2014)Howmuchistoomuch?Identifying benchmarksofadverseeffectsofmacroalgaeonthemacrofaunain intertidalflats.EcolAppl24:300 – 314 GSI(2015)GSIBedrockAquifers2015-06-16T00:00:00-revision. GeologicalSurveyIreland,Dublin2,Ireland. https://www.gsi.ie/ Mapping.htm HunterWilliamsNH,MisstearBDR,DalyD,LeeM(2013) Developmentofanationalgroun dwaterrechargemapforthe RepublicofIreland.QJEngGeolHydrogeol46:493 – 506 HydrogeolJ(2018)26:2629 – 2647 2645

PAGE 18

JeanninP-Y(2001)ModelingflowinphreaticandepiphreatickarstconduitsintheHöllochcave(Muotatal,Switzerland).WaterResourRes 37:191 – 200. https://doi.org/10.1029/2000WR900257 JohnsonAG,GlennCR,BurnettWC,PetersonRN,LuceyPG(2008) Aerialinfraredimagingrevealslargenutrient-richgroundwaterinputstotheocean.GeophysResLett35:L15606. https://doi.org/10. 1029/2008GL034574 Juki D,Deni -Juki V(2015)Investigatingrelationshipsbetweenrainfallandkarst-springdischargebyhigher-orderpartialcorrelation functions.JHydrol530:24 – 36. https://doi.org/10.1016/j.jhydrol. 2015.09.045 KingJN,MehtaAJ,DeanRG(2010)Analyticalmodelsforthegroundwatertidalprismandassociatedbenthicwaterflux.HydrogeolJ18: 203 – 215. https://doi.org/10.1007/s10040-009-0519-y KneeKL,CrookED,HenchJL,LeichterJJ,PaytanA(2016)Assessment ofsubmarinegroundwaterdischarge(SGD)asasourceofdissolved radiumandnutrientstoMoorea(FrenchPolynesia)coastalwaters. EstuarCoasts39:1651 – 1668. https://doi.org/10.1007/s12237-0160108-y KovácsA,SauterM(2007)Modellingkarsthydrodynamics.In: GoldscheiderN,DrewD(eds)Methodsinkarsthydrogeology. InternationalContributionstoHydrogeology26,International AssociationofHydrogeologists,Taylor&Francis,London,264pp LabatD,AbabouR,ManginA(2000)Rainfall – runoffrelationsforkarsticsprings.PartI:convolutionandspectralanalyses.JHydrol238: 123 – 148. https://doi.org/10.1016/S0022-1694(00)00321-8 LarocqueM,ManginA,RazackM,BantonO(1998)Contributionof correlationandspectralanalysestotheregionalstudyofalargekarst aquifer(Charente,France).JHydrol205:217 – 231. https://doi.org/ 10.1016/S0022-1694(97)00155-8 LeeCM,JiaoJJ,LuoX,MooreWS(2012)Estimationofsubmarine groundwaterdischargeandassociatednutrientfluxesinToloharbour,HongKong.SciTotalEnviron433:427 – 433. https://doi.org/ 10.1016/j.scitotenv.2012.06.073 LiH,ZhangY,TangH,ShiX,RivkinRB,LegendreL(2017a) SpatiotemporalvariationsofinorganicnutrientsalongtheJiangsu coast,China,andtheoccurrenceofmacroalgalblooms(greentides) inthesouthernYellowSea.HarmfulAlgae63:164 – 172. https://doi. org/10.1016/j.hal.2017.02.006 LiZ,XuX,LiuM,LiX,ZhangR,WangK,XuC(2017b)State-space predictionofspringdischargeinakarstcatchmentinSouthwest China.JHydrol549:264 – 276. https://doi.org/10.1016/j.jhydrol. 2017.04.001 MacSharryB(2006)Theinfluenceoffracturesontopographyand groundwaterflowintheBurrenandGortlowlands,western Ireland.Doctoralthesis,TrinityCollege,Dublin,Ireland,pp259 MacDermotCV,McConnellB,PrachtM(2003)BedrockGeology1:100, 000SacleMapSeries,Sheet14,GalwayBayGeologicalSurveyof Ireland,Dublin MarganeA,MakkiI,RaadR,SchulerP,Abi-RiskJ,DoummarJ, KoenigerP,HahneK,StoecklL(2018)HydrogeologyoftheJeita GroundwaterCatchmentinLebanon.GeologischesJahrbuchReihe BBundesanstaltfuerGeowissenschaftenundRohstoffe(BGR), Hannover,192pp MasseiN,DupontJP,MahlerBJ,LaignelB,FournierM,ValdesD,Ogier S(2006)Investigatingtransportpropertiesandturbiditydynamicsof akarstaquiferusingcorrelation,spectral,andwaveletanalyses.J Hydrol329:244 – 257. https://doi.org/10.1016/j.jhydrol.2006.02.021 MathevetT,LepillerM,ManginA(2004)Applicationoftime-series analysestothehydrologicalfunctioningofanalpinekarsticsystem: thecaseofBange-L ’ Eau-Morte.HydrolEarthSystSci8:1051 – 1064. https://doi.org/10.5194/hess-8-1051-2004 MauriceL(2009)Investigationsofrapidgroundwaterflowandkarstin theChalk.Doctoralthesis,UniversityCollegeLondon MauriceL,BarkerJA,AtkinsonTC,WilliamsAT,SmartPL(2011)A tracermethodologyforidentifyingambientflowsinboreholes. Groundwater49:227 – 238. https://doi.org/10.1111/j.1745-6584. 2010.00708.x MayolP(2011)ReviewoftheSouthGalwayFloodStudyReportOffice ofPublicWorks,Headford,Co.Galway,p43 McCormackT,GillLW,NaughtonO,JohnstonPM(2014) Quantificationofsubmarine/intertidalgroundwaterdischargeand nutrientloadingfromalowlandkarstcatchment.JHydrol519, PartB:2318 – 2330. https://doi.org/10.1016/j.jhydrol.2014.09.086 McCormackT,O ’ ConnellY,DalyE,GillLW,HenryT,PerriquetM (2017)CharacterisationofkarsthydrogeologyinwesternIreland usinggeophysicalandhydraulicmodellingtechniques.Journalof Hydrology:RegionalStudies10:1 – 17 MolesNR,MolesRT(2002)Influenceofgeology,glacialprocessesand landuseonsoilcompositionandquaternarylandscapeevolutionin theBurrenNationalPark,Ireland.CATENA47:291 – 321. https:// doi.org/10.1016/S0341-8162(01)00190-4 MooreWS(2006)Radiumisotopesastracersofsubmarinegroundwater dischargeinSicily.ContShelfRes26:852 – 861. https://doi.org/10. 1016/j.csr.2005.12.004 MoriasiDN,ArnoldJG,VanLiewMW,BingnerRL,HarmelRD,Veith TL(2007)Modelevaluationguidelinesforsystematicquantification ofaccuracyinwatershedsimulations.TransASABE50:885 – 900 MullanG(2003)CavesofCountyClare&SouthGalway.Universityof BristolSpelologicalSociety,Bristol,259pp NaughtonO,JohnstonPM,GillLW(2012)Groundwaterfloodingin Irishkarst:thehydrologicalcharacterisationofephemerallakes (turloughs).JHydrol470 – 471:82 – 97. https://doi.org/10.1016/j. jhydrol.2012.08.012 NullKA,KneeKL,CrookED,deSieyesNR,Rebolledo-VieyraM, Hernández-TerronesL,PaytanA(2014)Compositionandfluxes ofsubmarinegroundwateral ongtheCaribbeancoastofthe Yucatanpeninsula.ContShelfRes77:38 – 50. https://doi.org/10. 1016/j.csr.2014.01.011 O ’ ConnellY,DalyE,DuffyGP,HenryT(2012)Investigationofsubmarinegroundwaterdischargeandpreferentialgroundwaterflow-paths inaCoastalKarstareausingtowedMarine&TerrestrialElectrical Resistivity.PaperpresentedattheAmericanGeophysicalUnion Conference,SanFrancisco,3 – 7December2012 O ’ ConnellM,MolloyK(2017)Mid-andlate-Holoceneenvironmental changeinwesternIreland:newevidencefromcoastalpeatsand fossiltimberswithparticularreferencetorelativesea-levelchange. Holocene27(12):1825 – 1845. https://doi.org/10.1177/ 0959683617708447 PadillaA,Pulido-BoschA(1995)Studyofhydrographsofkarsticaquifersbymeansofcorrelationandcross-spectralanalysis.JHydrol 168:73 – 89. https://doi.org/10.1016/0022-1694(94)02648-U PerriquetM(2014)Characterizationofthehydrodynamicsandsaltwater wedgevariationsinacoastalkarstaquiferinresponsetotideand precipitationevents(BellHarbourcatchment,Co.Clare,Ireland). Doctoralthesis,NationalUniversityofGalway,Ireland PerriquetM,HenryT,CaveR,LeonardiV,JourdeH(2012) Hydrodynamicsofacoastalkarstaquiferaffetcedbysaltwaterintrusionunderoceanicclimaticinfluence,Co.Clare,Ireland.Paper presentedatthe22ndSaltWaterIntrusionMeeting,Búzios,Brazil, 17 – 22June2012 PerriquetM,LeonardiV,HenryT,JourdeH(2014)Saltwaterwedge variationinanon-anthropogeniccoastalkarstaquiferinfluenced byastrongtidalrange(Burren,Ireland).JHydrol519:2350 – 2365. https://doi.org/10.1016/j.jhydrol.2014.10.006 PetersonRN,BurnettWC,TaniguchiM,ChenJ,SantosIR,IshitobiT (2008)RadonandradiumisotopeassessmentofsubmarinegroundwaterdischargeintheYellowRiverdelta,China.JGeophysRes Oceans113:C09021. https://doi.org/10.1029/2008JC004776 PrachtM,LeesA,LeakeB,FeelyM,LongB,MorrisJ,McConnellB (2004)GeologyofGalwayBay:ageologicaldescriptionto 2646 HydrogeolJ(2018)26:2629 – 2647

PAGE 19

accompanytheBedrockGeology1:100,000mapseries,Sheet14, GalwayBayGeologicalSurveyofIreland,Dublin PrachtM,SomervilleID,WangY(2015)ArevisedMississippianlithostratigraphyofcountyGalway(westernIreland)withananalysisof carbonatelithofacies,biostratigraphy,depositionalenvironments andpalaeogeographicreconstructionsutilisingnewboreholedata. JPalaeogeogr4:1– 26. https://doi.org/10.3724/SP.J.1261.2015. 00065 SantosIR,NiencheskiF,BurnettW,PetersonR,ChantonJ,Andrade CFF,MilaniIB,SchmidtA,KnoellerK(2008)Tracinganthropogenicallydrivengroundwaterdischargeintoacoastallagoonfrom southernBrazil.JHydrol353:275 – 293. https://doi.org/10.1016/j. jhydrol.2008.02.010 SekulicB,VertacnikA(1996)Balanceofaverageannualfreshwater inflowintotheAdriaticSea.WaterResDev12:89 – 97 SilkeJ,O ’ BeirnFX,CroninM(2005)Kareniamikimotoi:anexceptional dinoflagellatebloominWesternIrishWaters,Summer2005.Marine EnvironmentandHealthSeriesNo.21,MarineInstitute,Galway SimmsMJ(2014)KarstandPaleokarst .In:ReferenceModuleinEarth SystemsandEnvironmentalSciences.Elsevier,p1 – 10 SmithAJ,NieldSP(2003)Groundwaterdischargefromthesuperficial aquiferintoCockburnsoundWesternAustralia:estimationbyinshorewaterbalance.Biogeochemistry66:125 – 144. https://doi.org/ 10.1023/B:BIOG.0000006152.27470.a9 SmythT(2000)AcasestudyintheGort/ArdrahanareaofSouthGalway. In:GendreauN(ed)FLOODawareFinalReport.European Commission,p163-171 TamborskiJJ,RogersAD,BokuniewiczHJ,CochranJK,YoungCR (2015)Identificationandquantificationofdiffusefreshsubmarine groundwaterdischargeviaairbornethermalinfraredremotesensing. RemoteSensEnviron171:202 – 217. https://doi.org/10.1016/j.rse. 2015.10.010 TaniguchiM,FukuoY(1996)AneffectofseicheongroundwaterseepagerateintoLakeBiwa,Japan.WaterResourRes32:333 – 338. https://doi.org/10.1029/95WR03245 TaniguchiM,OnoM,TakahashiM(2015)Multi-scaleevaluationsof submarinegroundwaterdischarge.Proc.IAHS365:66 – 71. https:// doi.org/10.5194/piahs-365-66-2015 ThompsonC,SmithL,MajiR(2007)Hydrogeologicalmodelingof submarinegroundwaterdischargeonthecontinentalshelfof Louisiana.JGeophysResOceans112:C03014. https://doi.org/10. 1029/2006JC003557 ThrailkillJ(1974)PipeflowmodelsofaKentuckylimestoneaquifer. Groundwater12:202 – 205. https://doi.org/10.1111/j.1745-6584. 1974.tb03023.x USGS(2014)ShuttleRadarTopographyMission(SRTM)1Arc-Second Global.USGSEarthExplorer. https://earthexplorer.usgs.gov/ WilsonJ,RochaC(2012)Regionalscaleassessmentofsubmarine groundwaterdischargeinIrelandcombiningmediumresolutionsatelliteimageryandgeochemicaltracingtechniques.RemoteSens Environ119:21 – 34. https://doi.org/10.1016/j.rse.2011.11.018 WMO(2008)Guidetometeorologicalinstrumentsandmethodsofobservation,8thedition.WorldMeteorologicalOrganization,Geneva ZektserIS,DzhamalovRG,EverettLG(2007)Submarinegroundwater. CRCPress,BocaRaton,pp466 HydrogeolJ(2018)26:2629 – 2647 2647


printinsert_linkshareget_appmore_horiz

Download Options

close


  • info Info

    There are only PDFs associated with this resource.

  • link PDF(s)



Cite this item close

APA

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

MLA

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

CHICAGO

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

WIKIPEDIA

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