|UFDC Home||myUFDC Home | Help|
Lifecycleassessmentofactiveandpassivegroundwater remediationtechnologiesPeterBayer ,MichaelFinkelCenterforAppliedGeoscience,UniversityofTuebingen,Sigwartstrasse10,72076Tuebingen,Germany Received17June2005;receivedinrevisedform28October2005;accepted10November2005 Availableonline27December2005Abstract Groundwaterremediationtechnologies,suchaspump-and-treat(PTS)andfunnel-and-gatesystems (FGS),aimatreducinglocallyappearingcontaminations.Therefore,thesemethodologiesarebasically evaluatedwithrespecttotheircapabilitytoyieldlocalimprovementsofanenvironmentalsituation, commonlyneglectingthattheirapplicationisalsoassociatedwithsecondaryimpacts.Lifecycleassessment (LCA)representsawidelyacceptedmethodofassessingtheenvironmentalaspectsandpotentialimpacts relatedtoaproduct,processorservice.Thisstudypresentstheset-upofaLCAframeworkinorderto comparethesecondaryimpactscausedbytwoconceptuallydifferenttechnologiesatthesiteofaformer manufacturedgasplantinthecityofKarlsruhe,Germany.AsaFGSisalreadyoperatingatthissite,a hypotheticalPTSofthesamefunctionalityisadopted.DuringtheLCA,theremediationsystemsare evaluatedbyfocusingonthemaintechnicalelementsandtheirsignificancewithrespecttoresource depletionandpotentialadverseeffectsonecologicalquality,aswellasonhumanhealth.Sevenimpact categoriesaredistinguishedtoaddressabroadspectrumofpossibleenvironmentalloads.Amainpointof discussionisthereliabilityoftechnicalassumptionsandperformancepredictionsforthefuture.Itis obviousthatahighuncertaintyexistswhenestimatingimpactspecificindicatorvaluesoveroperationtimes ofdecades.Anuncertaintyanalysisisconductedtoincludetheimprecisionoftheunderlyingemissionand consumptiondataandascenarioanalysisisutilisedtocontrastvariouspossibletechnologicalvariants. Thoughtheresultsofthestudyarehighlysite-specific,ageneralisedrelativeevaluationofpotential impactsandtheirmainsourcesistheprincipleobjectiveratherthanadiscussionofthecalculatedabsolute impacts.Acrucialfindingthatcanbeappliedtoanyothersiteisthecentralroleofsteel,whichparticularly derogatesthevaluationofFGSduetotheassociatedemissionsthatareharmfultohumanhealth.Inviewof that,environmentalcreditscanbeachievedbyselectingamineral-basedwallinsteadofsheetpilesforthe funnelconstructionandbyminimisingthesteelconsumptionforthegateconstruction.Granularactivated carbon(GAC)isexclusivelyconsideredasthetreatmentmaterial,bothin-situandon-site.Hereitis0169-7722/$-seefrontmatter D 2005ElsevierB.V.Allrightsreserved. doi:10.1016/j.jconhyd.2005.11.005 *Correspondingauthor.Tel.:+4970712973178;fax:+4970715059. E-mailaddress: firstname.lastname@example.org(P.Bayer). JournalofContaminantHydrology83(2006)171199 www.elsevier.com/locate/jconhyd
identifiedasanadditionalmaindeterminantoftherelativeassessmentofthetechnologiessinceitis continuouslyconsumed. D 2005ElsevierB.V.Allrightsreserved.Keywords: Lifecycleassessment;Funnel-and-gate;Pump-and-treat;Remediation;Groundwater;Uncertainty;Monte Carlo;Weightingtriangle1.Introduction Groundwaterisoneofthemostvaluablenaturalresources.Besideitscentralpositionfor terrestrialandaquaticecosystems,groundwaterisessentialfordrinkingandindustrialwater abstraction.Itrepresentsbyfarthelargestreservoiroffreshwater,withstoredvolumesoverfifty timesgreaterthantheamountofsurfacewater.Growingpopulations,continualurbanisationand industrialisationareincreasingtheriskofgroundwatercontamination,whilesimultaneously,the livingstandardsanddemandsonournaturalenvironmentarerising.Hazardouswastedisposal facilities,oilrefineriesandchemicalplantsaretypicalsourcesofwastestreams,whichcanreach theaquiferandproducelong-termcontaminationsinthesubsurface.Mostcontaminantsitesin industrialisedcountriesarecharacterisedbytheoccurrenceofhazardousorganicchemicalsin theunderground.Theycanformseparate,nonaqueousphasesinthesubsurface,which continuouslyfeedthepassinggroundwaterwithcontaminants.Downgradientofthesesource zones,plumesdevelopwithsignificantconcentrationsofthecontaminantsdissolvedinthe groundwater.Contaminantscanalsoslowlydiffuseintothelowpermeableaquifermatrixand, becauseofthesmallmasstransferratesintogroundwater,staythereaslong-termsecondary sources( Grathwohl,1998;Fetter,1999 ).Theseaspectssubstantiatetheimportanceofthe protectionofgroundwaterfrompollutionandthecontrolofalreadycontaminatedaquiferzones. However,theconsiderablylowmobilitytogetherwithatypicallyhighpersistencyoforganic contaminantsisthecruxforthedevelopmentofsuitablegroundwaterremediationtechnologies. Astillcommonandconventional b cleanup Q practiceis pump-and-treat ,meaningthepumping ofcontaminatedgroundwaterfollowedbyon-sitetreatment.Pumpingwellsareinstalledinorin thevicinityofthesourceorplumeareainordertohydraulicallycapturethecriticalaquifer zones.Experienceshowsthatduetothelimitedmasstransferrates,aquiferrestorationby conventionalpump-and-treatsystemsis,forthemostpart,notachievablewithinreasonable timeframes( EberhardtandGrathwohl,2002;Strooetal.,2003 ).Thoughtheuseofpump-andtreatsystems(PTS)isnottheonlytechnologicaloption,itstillseemstobethemostfavourable becauseoftheexperienceinappropriatehydraulicdesign,aswellasitsflexibilityandsimplicity ( USEPA,1996;Bayer,2004 ). Meanwhile,permeablewalls(PRBs)areawidelyacceptedalternativeforlong-termplume management.Continuouslyorlocallyreactiveverticalwallsareinstalledin-situfordowngradientcaptureandtreatmentofthecontaminatedplume( USEPA,2002 ). Funnel-and-gate systems (FGS StarrandCherry,1994 )areavariantbasedonthecombinationofimpermeable walls(funnels)andreactivezones(gates).Adjustedtotheregionalgroundwaterflowregime,the funnelsdirectthecontaminatedwaterthroughthein-situtreatmentfacilitieswithinthegates. ComparedtoPTS,FGSandcontinuouspermeablewallsaredenotedas b passive Q systemssince, afterproperinstallationofthetechnologicaldevices,noactivework,suchaspumping,is needed. Inthepresentedstudy,thefocusissetonthecomparisonofPTSandFGS,astypicalactive andpassivegroundwaterremediationoptions,intermsoftheirenvironmentalimpacts.SimilarP.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 172
towastewatertreatmentorrecyclingtechniques,theachievementofalocalenvironmental benefit,i.e.thecontrolandremovalofcontaminants,iscompromisedbycertainenvironmental burdenscausedduringtheconstruction,installationandmaintenanceofthetechnologies.These so-calledsecondaryimpacts( Volkweinetal.,1999 )canreachsignificantlevelsdependingon thetypeofmaterialsandenergyused,thesizeofthetreatmentsystemanditstimeofoperation. Alifecycleassessment(LCA)frameworkisusedtocontrastthepotentialimpactsindifferent categoriesreflectingtheinherentadverseeffectsonecology,humanhealthaswellasthe depletionofenergyresources.LCAconsidersthepotentialimpactsassociatedwiththesupply chainsthroughoutaproducts(i.e.here,technologys)lifecyclefromrawmaterialacquisition throughproduction,useanddisposal( ISO,14040,1997 ). 2.Previouswork ThemostcomprehensiveapproachesforusingLCAtoenvironmentallyrateimpactsfrom remediationofcontaminatedsiteshavebeenpresentedby Beinatetal.(1997) Volkweinetal. (1999) and Diamondetal.(1999) Sue `retal.(2004) giveabriefliteraturereviewofthefew availablestudiesinthisarea.TheRECmethodpresentedby Beinatetal.(1997) isastreamlined decisionsupportsystemforestimatingtherelevanceofenvironmentalimpactsoftechnological alternativeswithinariskassessmentframework. Volkweinetal.(1999) emphasizeadifference intheirUVAapproachtoREC:WhiletheRECtoolcanbeappliedtoderivesuitableclean-up levels,thelevelshavetobeassignedaprioriwhenusingtheUVAmethod.Asidefromthis, UVAisamoredetailedandelementaryapproach,whichoperatesongenericdatasetsprocessed inover50modules( LfU,1998 ). Volkweinetal.(1999) showtheapplicationoftheirmethodin accessingtheimpactscausedbypartialsoilexcavation,on-siteensuringandsurfacesealing. Benderetal.(1998) alsodiscussthesuitabilityoftheUVAtoolforanalysingtheimpactscaused byanumberofdifferentgroundwaterremediationtechnologies,suchaslong-termextractionof groundwater,in-situbioremediationandsoilvaporextraction.Thoughtheircasestudydoesnot exhibitadetailedsiteandinventorydatadescriptionandnoinformationisgivenonhowthesite dataisprocessedwithintheLCAframework,generalconclusionsarederivedwhencomparing thedifferentremediationtechnologies.Forexample,theyidentifiedenergyconsumptionasthe majorcauseofenvironmentalimpactwhenlong-termgroundwaterextractionisconsidered.This issupportedbythefindingsof Vignes(2001) Thelifecyclemanagementapproachof Diamondetal.(1999) isintendedasasystematic conceptforderivingqualitativecomparisonsofselectedsoilandgroundwaterremediation optionswithinadecisionmakingprocess.ItiscomplementedbyaLCAstepforadetailed investigationoftheenvironmentalburdensassociatedwiththetechnologies.Thegeneric remediationoptionshighlightedwereexcavationandoff-sitedisposal,in-situbioremediation, soilwashing,vaporextractionandnoaction.Theutilisationoftheconceptualoutlinepresented by Diamondetal.(1999) ,foraquantitativeinspectionofassociatedenvironmentalburdens,is showninacasestudyonsoilremediationofaPAHcontaminatedsiteinOntario,Canada( Page etal.,1999 ).Duringtheexaminedtimehorizonof75weeks,contaminatedsoilandsludgewas discardedatnearbydisposalfacilities.Sincenolongtermmeasurementssuchaspost-site processingandmaintenanceofwastelandfillwereinvestigated,themainemissionswere controlledbymaterialtransport.Thecreationofsolidwasteandlandusewasaccountedforin separatelifecycleimpactcategories.SoilremediationisalsosubjecttotheLCApresentedby Blancetal.(2004) ,wherefourtechnologiesarecomparedintermsoftheirenvironmental impactsandtherelatedenergyconsumptiontoclean-upasulfurcontaminatedsiteinFrance.P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 173
Thefourremedialactionsworkwithsoilexcavation,followedbyon-sitetreatment,backfilland/ ordisposal. 3.LCAframework 3.1.Overview ThestructuralandproceduralcomponentsofLCAaredeterminedbytheinternational standardseries ISO1404043(1997,1998,2000a,b) .Mandatoryinitialstepsincludethe definitionofgoalandscope,functionalunitaswellassystemboundaries.Theoverallgoalfor thisstudyisthecomparisonoftheenvironmentalperformanceoftwodifferentlongterm groundwaterremediationtechnologies.Apremiseforavalidcomparisonistopresumeanequal functionalityofbothapproaches,i.e.todefinethefunctionalunitsothatitprovidesareference towhichtheinputsandoutputscanberelated( Rebitzeretal.,2004 ).Thisisaddressedby analysinganexistingcase,theformermanufacturedgasplantsiteofthecityofKarlsruhein southernGermany(sitedescriptiongivenin Table1 ).In2000,aFGS( Fig.1 )wasinstalled ( Schadetal.,2000 ),whichwillsubsequentlybejudgedagainstahypotheticalPTS.Thelatteris Table1 TechnicalspecificationsofKarlsruhesiteandkeyelementsofFGSandPTSsubjecttoLCA SiteparameterformergasplantKarlsruhe Widthofcontamination210m Hydraulicconductivitylogmean3.9E 03m/s Hydraulicconductivitylogvariance1.31 Regionalgradient0.070.135%gradientdirection (25%seasonalvariation)NW/WNW ContaminantconcentrationsAcenaphthene400600ug/l AveragedtotalflowratethroughFGS10l/s FGS Funneldepth17m Funnellength240m Gatesdepth17m Gateboringsdiameter2.50m Gatecasingsdiameter1.80m Steel/gate8 12t Gravel/gate(50m3/gate;1.6t/m3)8 80t Clay/gate(8m3/gate;here2t/m3)8 16t GAC/gate(25m3/gate;0.6t/m3)8 15t Lifetime(regenerationinterval)5years PTS Numberofwells10 Installationdepth17m Pumpingrate18l/s PVCconduits100m HDPEcontainer(5cmwall)max.20m3each GACrefillinterval1year GACvolumeperfill2443.2t/fill Lifetimeofconduits,pumps,vessels10years P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 174
configuredassumingequivalentremediationgoals,i.e.thelongtermhydrauliccontainmentof thecontaminatedplumeemanatingfromtheupgradientsourcezone.Basedonthesite-specific resultsoftheLCA,thegoalistoidentifycrucialandenvironmentallysensitivetechnicaldesign elements,aswellastoderivesuggestionsfortechnologicalenhancementsfortheparticularsite andingeneral.Theexactdefinitionoffunctionalunitusedforthisstudyisexplainedindetailin thefollowingsection. ThesystemboundariesfortheLCAaresetbasedonthecharacteristicsofthesite,focusing ontheenvironmentalburdenrelatedtomajortechnologicalelements,whichwillbedelineatedin detailafterthesubsequentdescriptionofthesite.Thepresentedstudyisnotlimitedtothe analysisoftheinstalledFGSandahypotheticalPTS.Severalmodificationsofthereference scenarioareformulatedduringtheevaluationoftheresultsinordertoinvestigatetheroleof assumptionsmadeforthetechnologicalelementsandtodevelopsystemalternativeswith decreasedenvironmentalimpact.Anoverviewisin Table2 .Thescopeofthesescenariosis illustratedin Fig.2 AftertheinitialphaseoftheLCA,whichiscrucialtoestablishthecontextinwhichthe evaluationistobemade,theensuingstepsareinventoryanalysis,impactassessmentand interpretation.Theinventoryanalysis(LCI)examinesandcompilesallrelevantenergyand materialinputsandoutputsofprocessesduringthelifecycleofaproductoraservice.Theimpact Table2 VariationsbetweenscenariosdevelopedduringtheLCA ScenarioTime(years)FGSPTS ReferencescenariosheetpilesGAC72t/yr Scenarioswithdifferentwalltechnologies30diaphragmwall,slurrywall Scenarioswithdifferentwalltechnologies andvariableoperationtime 1050sheetpiles,diaphragmwall, slurrywall Scenarioswithdiff.walltechnologies, variableoperationtimeandvariable GACconsumptionforPTS 1070sheetpiles,diaphragmwall, slurrywall GACvar. Fig.1.Schemeoffunnel-and-gatesystemasinstalledattheformergasplantKarlsruhe-Ost.Verticalsheetpilesactas funnelsthatdirectthecontaminatedgroundwaterinto(intotal8)gates.Inthesegates,perforatedsteelcylindersarefilled withGACthatisregeneratedevery5years. P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 175
assessment(LCIA)iscarriedouttotranslatethecollectedemissionsandconsumptionsinto environmentaland/orhealtheffectsandiscommonlyexpressedbyanumberofrepresentative indicators.Finally,withintheinterpretationphase,theresultsofinventoryanalysisandimpact assessmentarediscussedtoextractthemainsourcesofenvironmentalburdenandtoderive recommendationsforthepreferableproduct.Inthenextsections,itisdescribedindetailhoweach crucialLCAelementhasbeenadoptedforthecomparisonofbothremediationtechnologies. 4.Functionalunit Thefunctionalunitdemarcatesthebasisuponwhichservicesarecomparedwithinthe predefinedsystemboundaries.Inthecontextofgroundwaterandsoilremediationprojectsthe desiredservice(s)maylargelydifferfromcasetocasedependingonthespecificobjectivesof theplannedremedialaction.AttheKarlsruhesite,sinceacompleteaquiferrestorationwasnot possible,neithertechnicallynoreconomically,themajorgoalwastopreventanyfurther increaseofriskcausedbyanongoingexpansionoftheexistingcontaminantplumeinthe groundwater.Bothtechnologicalapproachesthatareinthefocusofthisstudydoachievethis goalbyhydrauliccontainmentandareductionofcontaminantconcentrationstogiven standardlevelsbyon-siteorin-situwatertreatment.Thismeansthattheparticular technologies,evenif,inthenarrowersense,theyarecausingdifferenteffects,doprovide thesameservicewithrespecttothespecificremedialgoalofmanaging,i.e.controllingthe contaminantplume. Consequently,thefunctionalunitisdefinedhereasthecontrolofacertaincontaminated aquiferzone,whichcorrespondstothedefinitionproposedby Volkweinetal.(1999) .The followingargumentsshallexplainthereasonsfortheselectedformulationofthefunctionalunit, inparticularwithrespecttothepracticeinotherLCAapplications: (1)Thechoiceofthistypeoffunctionalunitfollowsthesuggestionsby Shakweerand Nathanail(2003) ,i.e.referringtothecontaminatedvolumeoftheaquifertoberestoredor tobecontrolled.Theperformancecriterion,i.e.hydrauliccontainment,isattainedwhen concentrationtargetsareensureddowngradientofthetreatmentsystem.Modelscanbe appliedtoconfigureequivalenttechnologies,ratedwithrespecttotheremediationgoals whilereflectingthepresentknowledge(e.g., Godinetal.,2004 ).Inthepresentstudywe employananalyticalmodeltodescribethepump-and-treatsystem,whereastheresultsof numericalmodellingareutilisedforthefunnel-and-gatesystem. Fig.2.KeyelementsofFGS(left)andPTS(right)forthescenariosconsideredfortheLCA. P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 176
(2)Ifremediationisdefinedinabroadersense,suchas b improvement Q ofsitesituation,amore exactfunctionalcomparisonofcompetingactionscouldbetocomputethetrade-offs betweeninputsandoutputswithinariskassessmentframeworksuchasaccomplishedby REC( Beinatetal.,1997 ).Inthiscase,allassociatedenvironmentalbenefitsandburdensare balancedwitheachothertoderivetheidealtechnologicalalternative.Wedonotgosofarin ourstudy,butkeeptheclassicdefinitionofaservice-relatedfunctionalunit,whichhasa prioritobesetandisnotadecisionvariable.Ourobjectionsaretotheimpracticablyhigh spatialandtemporalresolutionthatwouldbeneededforcalculatingrisk-benefittrade-offs. (3)AnotherlimitationoftheREC-approachisthatitsapplicabilityandexpressivenessis restrainedbythehighdegreeofparameterandprocessuncertainty,sincethelong-term effectandefficiencyinreachingoverallaquiferclean-upcanhardlybeestimatedprecisely. Thereareseveralreasonsforthis:Firstofall,afundamentalproblemisthatthephysical andchemicaldescriptionofthesubsurfaceisbasicallyinexact.Accordingly,thereis substantialuncertaintyinestimatingormodellingpresentandfuturestatusofthe remediationprocess.Furthermore,andevenworse,atnumeroussitestheinitialexactmass distributionofcontaminantsinthesubsurfaceisnotknown.Principally,descriptiveand predictiveuncertaintyiscrucial,independentofthetypeoffunctionalunitchosen. However,theselectionofaconcentrationthresholdrelatedfunctionalunitislessproneto theexactcontaminantmassdistribution. (4)Systemdesignbyreferringtoregulatorystandardsrepresentsthecommonpracticeandis compatibletoeconomicallybaseddesignconsiderations(e.g., RussellandRabideau, 2000;Bayeretal.,2005a ). Foracomparativeevaluationofremedialoptions,thequalityoftechnicaldesignand performanceparametersfundamentallydependsontheexperienceintechnologyandknowledge ofthesite.Ifatechnologyisimplementedatasite,itsperformancecharacteristicsareknown morepreciselythanforanother,hypotheticalalternative.Also,forarealisedmeasure,the associatedconsumptioninmaterialandenergyisapparent,whereaspotentialalternativescanbe describedonlytheoretically.Inthesubsequentcasestudy,theFGSisalreadyinstalledandthe taskistosetupanimaginaryPTSonthesamefunctionalunitbasis.Aspointedoutby NRC (1997) intheexampleofaneconomiccomparisonofexistingandhypotheticalremedialoptions, onesbiasmustbetakenintocarefulconsiderationduringanevaluation.Sincethereisextensive experienceintheapplicationofPTSandsincethistechnologyisbasedonarathersimple concept,configuringitappropriatelydoesnotrepresentamajordifficulty.Theremaining uncertaintyindesignandlongtermperformanceisaddressedbyconsideringpossiblePTS designalternativesinascenarioanalysis. Especiallythelongtermtrendinapproachingsiteclean-upcanhardlybepredicted.Thisplays acrucialrolewhencomparingsuchconceptuallydifferenttechnologies.DuringtheuseofPTS, (secondary)emissionsarereleasedandenergyisconsumedcontinuously,whereasforFGS,the majorimpactarisesduringconstructionandinstallationofthesystem.Lifecycleassessment commonlyattemptstomeasurethetotalenvironmentaleffectsofaproductortechnology b from cradletograve Q ( BaumanandTillman,2004 ).Obviouslythiscanhardlybeachievedwhenthe b grave Q ,i.e.thetimeoftermination,isnotknown.Toovercomethis,asisalsocommonfor economicanalyses( NRC,1997 ),afixedoperationperiodcanbepredefined,sothatthetemporal boundaryofthefunctionalunitisexplicitlyset.Forexample, Benderetal.(1998) suggestan operationperiodof50yearsforcomparinglong-termin-situgroundwaterremediation technologies,whereasthetimeframe Pageetal.(1999) assumeforlongtermmonitoringisP.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 177
onlyhalfofthat.Itisobviousthat,duetothelackoflong-termexperiencewithmostremediation technologiesandthegeneraluniquenessofsiteapplications,thesetimelimitsarechosenquite arbitrarily.However,timeframesshallreflectplanningperiodsandthusalsoseemtobeavalid inputfortheLCA.Inthesubsequentanalysis,weusetimeasafixed,aswellasvariable, parameterinordertorevealitsinfluenceonthecomparativeratingoftheselectedtechnologies. 5.Inventoryset-upandimpactassessment Afterdefiningthekeyelementsoftheremediationtechnologies(cp. Fig.2 ),aninventorywas setupinordertocollectassociatedmaterialandenergytransfersthatcrosstheboundary betweenservicesystemorproductandenvironment.Theconfigurationofkeyelementswillbe describedbelowwhenthecasestudyisintroduced.Groupingandassigningmaterialandenergy transfersthatexhibitenvironmentalburdensand/orinterventionstopre-specifiedcategories,i.e. characterisationandclassification,isthetaskoflifecycleimpactassessment(LCIA; ISO14042, 2000a ).Thesecategoriesrepresentthreatenedenvironmentalcompartments,statesorrelations, suchastheconsumptionofenergyresources,globalwarmingorhumantoxicity( UdodeHaeset al.,1999;Penningtonetal.,2004 ).Forexample,burningcrudeoilmeansanexploitationof availableenergyresources.IfthisprocessisconsideredwithinLCA,theamountofconsumed oilisassignedtotherespectiveimpactcategorysuchas b depletionofenergyresources Q .Burning crudeoilreducesavailableresources,butalsoreleasesemissionssuchasCO2orSO2,whichare assignedtoothercategoriessuchas b globalwarmingpotential Q and b acidificationpotential Q Aquantitativeevaluationisonlypossibleifthedifferentcontributionstoonecategory, originallyexpressedindifferentunits,canbemerged.Thisimpliesthenecessityofa characterisationstep,whichunifiesallcontributionsandindicators.Calculationmethodsto determinecharacterisationfactorsaremanifold.Theydependontheunderlyingspecificmodels andonthepurposetheyareusedfor.Thustherearenouniversallyvalidcharacterisationfactors. Especiallyforthehumantoxicitypotentialthereisnoconsensusabouttheidealcharacterisation Table3 Characterisationfactorsforselectedinventorydata Impactcategory UnitofindicatorUnitchar. factor Characterisationfactors Cifor inventorydata Depletionofenergyresources(DER)crudeoileq. (COE) kgCOE/kgbrowncoal0.0409,hardcoal0.1836, naturalgas0.5212 Globalwarmingpotential(GWP)CO2eq.kgCO2/kgCH423,N2O296,CF45700 Acidificationpotential(AP)SO2eq.kgSO2/kgNOx0.70,HCl0.88,HF1.60, H2S1.88,NH31.88 Terrestrialeutrophicationpotential(TEP)PO4 3 eq.kgPO4 3 /kgNH30.35,NOxasNO20.13 Aquaticeutrophicationpotential(AEP)PO4 3 eq.kgPO4 3 /kgNh4+0.327,NO3 0.095,N-comp0.42, P-comp.3.06,COD0.022 Photochemicalozonecreation potential(POCP) C2H4eq.kgC2H4/kgC2H60.189,formaldehyde0.421, CH40.007,NMVOC0.416,VOC0.377 Humantoxicitypotential(HTP)Aseq.kgAs/kgbenzopyrene20.9,C2H60.0019, dioxine10500,PCB0.28,Cd0.42, Cr(VI)0.28,Ni0.06Indicatorsarecalculatedas Indicator iMiCi,where Mi(kg)istheinventorydataand Ciisthecharacterisation factor.IndicatorofPOCPisexpressednitrogencorrectedas Indicator MNOxiMiCi ,where MNOXcompounds(UBA 2000). P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 178
model,whichcanbebasedonabroadrangeofdifferenthealthriskassessmentandhuman toxicologycalculationmethodologies.Inthepresentedapproach,thecharacterisationmethodis adoptedfrom UBA(2000) ,withthemaincharacterisationfactorslistedin Table3 .Arecognised setofenvironmentalimpactsisselected( BaumanandTillman,2004 ).Thesearedepletionof energyresources(DER,incrudeoilequivalents),globalwarmingpotential(GWP,inCO2equivalents),acidificationpotential(AP,inSO2equivalents),terrestrialaswellasaquatic eutrophicationpotential(TEP,AEP,inPO4 3 equivalents),photochemicalozonecreation potential(POCP,inC2H4equivalents)andhumantoxicitypotential(HTP,inAsequivalents). Theallocationofpotentialenvironmentalimpactsisnotdiscussedwithrespecttospatialor temporalvariance.Inthisregard,wefollowtheunderlyingstreamliningconceptof Volkweinet al.(1999) ,though,especiallyforconglomerate,long-termoperationssuchasthoseinspected here,amoredetailedanalysisseemsdesirable.Distinguishingbetweenthespatiallydifferent effectsofemissionsforexampleisdiscussedby Potting(2000) and RossandEvans(2002) ,who recognisedtheinaccuracyofgeographicallyaveragingLCIAindicatorsbyneglectingthe regionallydiverseeffectsofemissions.Acommonproblemisthatappropriatedatasourcesare fragmentaryorcannotbedirectlytransferredfromthecontexttheyaregatheredfor. Accordingly,nospatialdiscretisationoftheimpactsisattainedhere,andtherefore,noanswer willbegiventothequestioniftheenvironmentalbenefitgainedfromlocalgroundwatercleanupcanbebalancedwiththeinherentglobal(secondary)impacts.Adetailedtemporalresolution oftheenvironmentalimpactwasnotaddressedsinceestimationsabouttechnologicalevolution, thepredictionofmodificationsintheproductionofmaterialandenergy,aswellasthe proceedingchangesintheevaluationandallocationofenvironmentalimpactsisbeyondthe scopeofthispaper. Inviewofthevarietyofapproachesthatarecompetingforanappositerealisationofthe obligationsaccordingtotheISOstandards,thereisnouniquewayofconductingaLCAfor remediationtechnologies.Moreover,availablesecondarydatasourcesandprocessspecific libraries,whichhavebeenpreferredforthepresentedstudy,canexhibitsignificant discrepanciesdependingonsub-systemboundaries,desiredtemporalandspatialresolution, specificcalculationmodelsusedandactuality.Thisinherentlymeansthattheexpressiveness oftheresultscanonlybejudgedbyreflectingthevalidityoftheinputdata.TheLCA frameworkdevelopedinthisstudyisorientedatandappliedtoanexamplecaseinGermany, soalsothefocuswassetonexploiting(secondary)databaseswithcloserelation(partlyonly availableinGerman),andadoptingthestandardsrecommendedforLCAinotherdisciplines bytheGermanenvironmentalagency(e.g., UBA,1998,2000 ).Duetothefactthatnotsolely case-specificbutpartlyalsopre-existingapproachesareusedforthedescriptionof interconnectedsub-processesinvolved(e.g.,productionofelectricalenergy,oil,steel),the accuracyofthecalculationsforthespecificcasestudyislimited.However,thispresented streamlinedconceptisnotaimingatanexactquantificationofthepotentialenvironmental burdenofthetechnologiesconsidered.Insteaditshallconstitutesimilarboundaryconditions forthedescriptionofthetechnologiesandthiswayshallenableavalidcomparative assessment.Inordertoincorporatetheexpectedimprecisenessofinventorydataintothe presentedLCA,rangesinsteadofdeterministicvaluesareconsidered,andtheseareprocessed withinasubsequentuncertaintyanalysis. AnoptionalstepinLCAisacumulativeinvestigationofallcategoriestogether.Itcanbe usedforamorecondensedpresentationoftheresults.Ingeneral,aggregationofimpacts shouldbedonecarefullyinordertoavoidalossintransparencyconcerningtheroleof individualimpacts.AcriticaleffectisthatitcanleadtoseverecompensationbetweenimpactP.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 179
results(agoodimpactresultscancompensateaverybadone),andsoitisrecommendedto discusstheresultsforindividualimpactsbeforetheaggregation.Subsequently,thetwo remediationtechnologiesarecontrastedbymeansofascenarioanalysisanditseemed desirabletopartlyaggregatecategoriesforthepresentationoftheresults.Wegroupthe categoriesintothreetypes:thosedenotingadverseeffectsonenergyresources(DER), ecosystemqualityEQ(GWP,AP,TEP,AEP)andhumanhealthHH(POCP,HTP).Thisstep enablesabasicexaminationoftheroleofcrucialparametersandvisualisationbymixing trianglespresentedby Hofstetteretal.(2000) ,whichalsosuggestedthedistinctionofthese threesafeguardsubjectsaccordingtotheEcoIndicator99methodology.Theaggregationinto EQandHHclassesiscarriedoutbyaveragingthecategory-specificindicatorvalues.First,in ordertoachieveacomparabilitybetweenthedifferentkindsofimpacts,theindicatorvalues havebeennormalisedwithrespectto(annual)inhabitantequivalents( Table4 ).By normalisation,theindicatorsrepresentthecategoryspecificmultipleoftheprevailingamount Table4 ImpactassessmentforkeyelementsofFGSandPTS Inhabitantvalue/yearUnitsDER (kgCOE) GWP (kgCO2) POCP (kgC2H4) AP (kgSO2) TEP (kgPO4 3 ) AEP (kgPO4 3 ) HTP (kgAs) 244713167565.77.8140.006 GACton904.810922.214.171.1240.521.1E 034.2E 05 GACrec.ton328.31166.60.541.760.3003.1E 05 GACtransportton34.8126.96.36.1990.1301.4E 05 Facilities97.9305.52.520.250.551.4E 074.2E 05 FGSWALL Sheetpilem244.5363.31.020.200.242.8E 035.7E 03 Sheetpile+rec.m219.5164.90.5188.8.131.52E 041.4E 03 Facilities562.81799.514.41.433.0704E 05 Diaphragmwallm27.25184.108.40.206E 026.0E 021.1E 101.7E 06 Facilities765.32336.7220.127.116.1102.9E 04 Slurrywallm218.6139.90.516.8E 020.121.4E 107.4E 06 Facilities1757.75320.468.75.7013.005.0E 04 FGSGATE Excavationm323.0156.80.420.330.0500 Facilities470.31400.620.60.841.9200 Rebuildm34.1418.104.22.168E 023.2E 025.9E 091.2E 06 Facilities89.5222.214.171.124.491.3E 073.7E 05 Steelton284.825126.96.36.199.232.1E 024.2E 02 Steel+rec.ton124.61178.32.990.620.705.6E 031.1E 02 Gravelton6.56188.8.131.52E 023.0E 0203.1E 06 Clayton27.977.90.390.069.4E 0201.1E 05 PTS Installationwell(17m)146.3351.13.710.370.581.3E 032.2E 05 Facility12.5642.460.233.2E 023.8E 0204.6E 06 Pumping10E+6L6.6896.77.3E 021.7E 022.4E 021.5E 042.4E 06 Pump2.3523.14.9E 021.2E 021.1E 022.1E 041.2E 04 Conduits100m2.527.954.0E 026.4E 036.9E 032.2E 041.6E 08 GACcontainervessel(20m3)1340.02243.220.62.522.192.2E 024.8E 06 AverageinhabitantemissionsandconsumptionsforGermanyaretakenfrom IFU(2001) P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 180
ofemissionsorconsumptionrates.Hence,basedonthesameunits,theyreflecttherelative additionalburdencausedbythetechnologiesunderconsideration.However,theydonot reflectthelevelofseriousnessadherentintheindividualcategories. TheLCAframework,includingmaterialflowandinterrelationsofindividualprocesssteps describingthelifecycleoftheremediationtechnologies,hasbeendevelopedwithinthe modellingenvironmentUMBERTO,acommercialLCAsoftware( IFU,2001 ).Itrepresentsa widelyusedmodellingframework,whichenablesmaterialflowanalysesbasedoncomprehensivestate-of-the-artlibraries(materialgroups,processdata)thatcanbeexpandeddependingon thespecificproblemrequirements.Afterintroducingthesiteinthefollowingsection, assumptionsonprocessesinvolvedandondescriptiveparametersforthecasestudyare exposed.Adetailedlistoftheinventorydataisgivenin ZAG(2003) 6.Interpretation AnelementaryandnotanexclusivelyfinalstepofLCAistheinterpretationphase.As demonstratedhere,itshallaccompanythecourseofLCA.Itspurposeistoanalyseresults,reach conclusions,explainlimitationsandproviderecommendationsbasedontheresultsofthe precedingsteps(ISO2000).Inparticular,interpretationshallidentifyanddiscussenvironmentallymostrelevantelementsoftheremediationtechnologies,whilereflectinguncertaintyin inventorydataandtechnicalassumptions.Ideally,itispossibletoderivesuggestionsfor technologicalimprovementwithrespecttothedifferentLCAimpactcategories. 7.Casestudyformergasplantsite,KarlsruheOst Intheyear2000,aFGSwasinstalleddowngradientofaPAHdominatedcirca200mthick plumeemanatingfromtheformermanufacturedgasplantsiteofthecityofKarlsruheinthe Rhinevalley( Fig.1 ).Thesitecoversanareaofabout100,000m2( Schadetal.,2000 )with severalsupposedsourcezonesoftaroil.Theslightlyconfinedaquiferis12-mthick,underlain byaclaylayeratadepthof16mbelowthesurface.Duetothehighconductivitiesofthe alluvialgraveldominateddepositsoftheRhine,thecontaminatedgroundwaterflowrate emanatingfromthesiteisestimatedtoreach12l/sundernaturalconditions.Thehighest concentrationsofthecontaminantsintheplumearemeasuredforAcenaphthene(600 A g/l),with benzeneandvinylchlorideassecondarypollutants.Thelatterissuspectedtooccurintheaquifer upgradientoftheformergasplantandthusoriginatesfromanothersource. TheFGSconsistsof240mfunnelandeightgates,whichentirelycapturetheplume.The funnelswereconstructedofsheetpiles,whichareinstalledatadepthof17m.Forthegates,the excavationofboreholecasingswithdiametersof2.5mwereneededthroughouttheentire aquiferdepth.Thencylindricalgatesegments,eachbeing3minlengthand1.80mindiameter, wereinstalled.Thesearepre-fabricatedoutofsteelthatisperforatedatthein-andoutletandis connectedtotheaquiferbygravelfillings.Ateachgate,claysealingsconnectthesheetpileand steelcylindersegments( Schadetal.,2000;ZAG,2003 ).Aftertheconstructionofthepermanent facilities,atotalmassof120t(200m3)granularactivatedcarbon(GAC)hasbeenfilledintothe steelcylinders.Withareducedflowrateof10l/s,theGAClifetimeisestimatedat5years.This periodisalsosetasareferencevaluefortheFGSsGACexchangeandregenerationcyclesfor theLCA(cp. Tables1and2 ). ThedimensioningofanequivalenthypotheticPTSdependsonthegroundwaterflowrate,the areatobecaptured,thetypeandconcentrationofcontaminants,aswellasregulatoryrestrictions.P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 181
Hereweassumethattherequired(total)pumpingratecanbedeterminedbyreferringtothe flowrateofcontaminatedwater.Weusearoughestimationaccordingto JavandelandTsang (1986) ,whichpresenttypecurvesforsteady-statewellcapturezonesinconfined,homogeneous aquifers.Followingtheirapproach,thepumpingratetocaptureacertainareaiscalculatedasa multipleoftheuniformflowrate,whichis12l/sattheKarlsruhesiteandservesasalowerlimit.In ourstudy,weset18l/sasareference.Thisvalueisselectedwiththeassumptionthatdowngradient pumpinghastobemoderatelyhigher,butnotmorethandouble,theuniformratetoassure completehydrauliccapture( Bayeretal.,2003 ).Theidealnumberofpumpingwellstobeinstalled isuncertain,butduetothehighextractionrate,itisassumedthatintotal,tenfullypenetratingwells havetobeinstalled.Accordingtothein-situtreatmentinthegatesoftheFGS,GACisalsochosen asareactivemediaforPTSon-site.Thecleaned-upwaterisdisposedofinlocalsewersordirected totheRhineriver.Thoughthereisachancethatlocalregulationsimposefurtherprocessingofeven decontaminatedwaterinlocalwastewaterfacilities,thisisbeyondthescopeofthispaperandisnot consideredwithintheboundariesoftheLCA. 8.ClassificationoftechnologicalelementsconsideredwithinLCA 8.1.LCAtechnicalboundaries Thefocusofthiscomparativestudyissetonthefacilitiesinstalledandmaintainedforeach particularremediationtechnology.TheFGSandPTSfacilitycomponentschosenfortheLCA areexpectedtorepresentthecoreelements,intermsoftheirenvironmentalrelevance.Itis assumedthatadditionalservicesnotmentionedin Table2 ,suchaslongtermmonitoringand sampling,representonlyalowerontributiontotheLCAresults(cp. Pageetal.,1999 )and thereforecanbeneglectedforbothtechnologies.Itisobviousthatseveralfurtherprocessesand elementscouldbeconsidered,whichtogethermayhaveanoticeableinfluenceonthefinalLCA basedcomparison.However,mostofthesearehighlysite-specific(personalactivities,local powercontrolfacilities,etc.),veryvariableand/orcanonlybeassessedthrougha disproportionalefforttogatherreliabledata.Therefore,andinordertodrawgeneralconclusions fromthepresentedassessment,nofurtherworkhasbeendoneinthisdirection.Moresignificant impactscanbeexpectedfromrequiredtechnicalmodificationsbecauseofunsatisfactorysystem performance.Forexample,theFGSinKarlsruhewasslightlyrevisedduringthefirstyearssince itsimplementationandcappingelementswereinstalledatthetopofthegatestoimproveits hydraulicperformance.However,consideringthesemodificationsseemedunreasonable,andit isnotpossibletodevelopasimilarscenarioforthePTSalternative.Despitethis,itshouldbe emphasisedthattheassumptionofnomodificationsandnotechnicalfailurecanbeprincipally realistic,butrealisticallyisaveryoptimisticassumption. 9.Upstreamprocesses Theprocessflowdiagramsofthemaintechnologicalelementsaredepictedin Fig.3 .There areanumberofbasicseparableprocessesthatdeliverinputsatseveralpositionsandare distinguishedasupstreamprocesseswithinthemodellingframework.Theseareinparticularthe supplywithdieselandelectricalenergy.Forthese,theavailableUMBERTOlibrariesareused, whicharebasedontheinventorydataprovidedby Frischknechtetal.(1996) .Electricalcurrent isconsumedbyprocessessuchassteelconstructionandslurrymixing.Thepercentage contributionamongdifferentgenerationsourcesissetaccordingto IFU(2001) ,with32.5%P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 182
nuclearpower,27.0%hardcoal,27.3%browncoal,7.6%natural,3.9%waterpowergas,0.5% fueloil,0.3%windand0.9%waste.Thequalityoftheinventorydataondiesel(fossilfuel refinement,production)andelectricalcurrentcanbejudgedassufficientforthepurposeofthis studysince,despiteitsage,thedatabaseiswidelyusedand27comprehensive. Onefurthercommonprocessistheusageofmachinery,suchasforsitepreparation,drilling and(de-)installationoftechnicaldevicesandtransportactivities.Pleasenotethatthefabrication ofthismachinerywasnotconsidered,onlytheenergyconsumptionsandemissionsduring operation.Thedieselconsumptionandcombustionemissionsarecalculatedthesameforall devicesaccordingto BUWALetal.(1994) and Borkenetal.(1999) .Utilisingtheempirical approximationsprovidedbyBUWALformachinesofcapacitiesabove150kW,weestimateda Fig.3.FlowsheetofthescopeoftheLCAforthekeyelementsoftheremediationtechnologies.Circulararrowsindicate temporalreplacementofelementsduringoperation. P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 183
demandfordieselof242g/kWh,andemissionsof14.3g/kWhNOx,0.96g/kWhVOC,2.4g/ kWhCOand1.0g/kWhdust.Thesevaluesareutilisedformachinesusedfortheinstallationof walls(silentpiler,clamshell)andreactorexcavation.Theemissionsfromtheusageofdiesel drivendeviceswithcapacitiesbetween25kWand150kWarealsoapproximatedbythe empiricalfunctionsrecommendedby BUWALetal.(1994) Anotherubiquitousprocessisthetransportofmachinestothesiteandback,thetransportof mineralsubstances(e.g.,gravel,sand,clayexcavatedsoil),aswellasoftechnology-specific equipmentandmaterial(e.g.,sheetpiles).Fortheallocationofthedieselconsumptionand emissionscausedfromtransportactivities,theinternalUMBERTOmodulewasused,whichis basedontheTREMOD(trafficemissionestimationmodel)by Borkenetal.(1999) Theunderlyingdataqualityonlogisticprocessesisnotverygoodand,especiallyduetoits deficientup-to-dateness,thequantitativedescriptionhastobeusedwithcaution.Apreceding sensitivityanalysis( ZAG,2003 ),however,indicatedonlyaminorrelevanceoftransportactivities ontheentireenvironmentalburdenfortheselectedremediationtechnologies.Inparticular,a scenarioanalysiswithreasonablemin/maxassumptionsfortransportdistancesandinventorydata revealedthatvariationsleadtochangesoftheimpactcategoryspecificindicatorsbelow10%. 10.Verticalbarriers ThekeyelementsoftheLCAforFGSarelistedin Table1 :funnel,gateandGAC.Forthe funnel,sheetpilesareimplementedoveratotalareaof240m 17m=4080m2.Theinventory wassetuputilisingsecondarydatabasesandaccordingtotheinformationofthemanufacturers ( ZAG,2003 ).Forthefunnelconstruction,anumberofdifferentwalltypesareavailable,suchas slurrywalls,thindiaphragmwallsorsheetpiles.Thechoiceusuallydependsonseveralfactors, includingdepth,length,stability,longevityandeconomicconsiderations.Afirstinspectionof thecategoryrelatedtoemissionsandconsumptions,aslistedin Table2 ,revealsthatsignificant benefitsperm2wallcanbeachievedbyinstallinganalternativebarriersuchasa(thin) diaphragmwalloracementbentoniteslurrywall.Thisobservationmotivatedacomparisonof theexistingsheetpiletypetothesealternatives,whicharepotentiallyassociatedwithalower environmentalburden.Pleasenotethatanumberofothertypesofwallsexist,suchassheetpiles ofdifferentmaterial(aluminium,precastconcrete),soil-bentonitslurrywalls,borepileor compositecutoffwalls( MeggyesandSimon,2000;Careyetal.,2002 ). ThefunnelattheKarlsruhesitewasconstructedoutofpre-fabricatedsteelstrips,whichwere successivelypressedintothesoil,keyedintheunderlyingaquitardandjoinedtoforma continuoussubsurfacebarrier.Dieselconsumingpress-inmachines(here:silentpiler,provided byGIKEN)wereusedtopushthewallsegmentsintotheground.Thesilentpilerspecific nominalcapacityissetto650kWat25m2/hnetprogresswheninstallingthepiles.Asa preparatorymeasure,ahydraulicdriller(200kWat30m2/h)wasutilisedtoloosentheground andenhancethesubsequentpilecountersink.Thetransportdistancesofthesheetpileswere assumedtobe300kmat140kg/m2. Formanufacturingthesteelstrips,theprocesschainofsteelproduction,inclusiveraw materialmining,ironcastingandformingisincluded.Theindividualprocessesweredescribed usingtheUMBERTOlibraries.AccordingtocommonpracticeinGermany,thesteelisassumed tobeproducedinpartsinablastfurnace(75%)andaselectricsteel(25%),thelatterbeing environmentallypreferableduetotheuseofdiscardedmetal.Whensteelrecyclingissuggested aftercompletionoftheremediation,thesteelisassumedtobeprocessedinanelectricarc furnaceatavolumelossof10%.TheemissionsandrawmaterialconsumptionforprovidingtheP.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 184
steelstripsarecalculatedbasedonthedatasetsprovidedby CorradiniandKo ¨hler(1999) .The dataqualityisestimatedassatisfactoryforthepurposeofthisstudy. Theproductionofthetwootherwalltypesissignificantlydifferent,sincetheyare constructedentirelyin-situ.Forexample,thindiaphragmwallsareemplacedbyusinghighpressurejettingoftheslurryintotheground,asadrillstemisraisedupthroughtheground.A continuousbarrierisproducedbyconstructingarowofoverlappingpanels( USEPA,1998; Pearlman,1999 ).Asaresult,slurryandenergy(diesel,electricpower)isconsumed,and respectivesitefacilities(drillingdevices,slurrymixingplant)havetobemaintained.Transport distancesforthetechnologicaldevicesandmachineryareassumedtobe150km.Mineral compoundsareestimatedtobetransportedoveradistanceof20km. Similartoathindiaphragmwall,acementbentoniteslurrywallisfabricatedinthesubsurface byinjectingbentonite-basedsolidifyingfluids.However,forthiswalltype,atrenchhastobe excavatedinthetargetzonebeforethebackfillisinjected.Accordingly,atrench-cuttingmachine (clamshell)hastobesupplied.Theexcavatednativesoilisnotassumedtobecontaminatedin thisstudyandisdisposedofwithinthevicinityatadistanceof15km.Thetransportdistancefor themillingcuttermachineryisheldfixedat150km.Again,transportactivities,energyandwall materialdemandareconsideredfortheallocationofthecategoryspecificinputs. Bentoniteisaclaymineralwhichiscommonlyobtainedfromnaturaldeposits.Here65%ofthe bentoniteisestimatedtobeimportedfromotherEuropeancountries(ZAG,2002).Fortheslurry wall,thewallthicknessissetto80cm,withameancontentof14%(byweight)bentonite,16% blastfurnacecementand70%water( LfU,1995 ).Accordingto LfU(1995) ,45%lossofthewall cubaturehastobeexpectedduetosoliddisplacementinthesubsurfacecontrolledbyin-situ sedimentation,penetrationandfiltrationprocesses.Thediaphragmwallisconstructedoutof8% bentonite,9%blastfurnacecement,44%waterandadditionalnativefinesandfillers,witha demandof0.14tperm2wall.Thecementisexpectedtobeamixtureof70:30ofmetallurgical/ Portlandcement,whereasthishighportionofmetallurgicalcementshallguaranteethedesired sulfateresistanceofthewall.Furtherinformationontheinventoryset-upforthematerialusedis givenin Frischknechtetal.(1996) Kohler(1999) and ZAG(2003) ,whichevaluatethedataquality assatisfactory. Withinthisstudy,theselectedwalltechnologiesareexpectedtobeequalintheir performance,i.e.allexhibitthedesiredlowpermeabilitywithoutfailureorleakageoverthe wholeplanninghorizon.Itisobviousthatthisisaveryidealisedassumptionbecauseeachwall typeinvolvesindividualstrengthsandweaknesses.Sheetpiles,forexample,areespecially favourableinviewoftheirlong-termstability,chemicalresistanceandtherelatedquasiimpermeability,thoughleakagecanoccurwhentheseparatesteelstripsarenotappropriately connected( USEPA,1998;Careyetal.,2002 ).TheexperienceoftheKarlsruhesiteindicates thatsuchconstructionaldeficienciesarenotpresentandthuswillnotbeconsidered. Accordingly,potentialdefectsofthealternativehypotheticalwalltypesareneglected.However, amatterofdiscussionmightbethelongevityofthebentonite-basedbarriers.Thoughnot discussedfurtherwithintheLCAframework,shrinkagecracksmightoccurlocally,increasing thebarrierspermeability.Furthermore,contaminantsdissolvedinthegroundwatercancausea gradualdegradationofthewall( Pearlman,1999 ).Incontrasttosheetpiles,bentonite-based wallscannotberecycledanditisassumedthatnorebuildingorrestorationactivitiesareplanned aftertheremediationisfinished. TheimpactsaggregatedduringtheLCIAareexpressedasalinearfunctionofthewallarea (indicatorunitsperm2),whereasmachinerytransportisapproximatedbyafixedvalueforeach category.Indicatorvaluesarelistedin Table4 .ThemostnotabledifferencebetweensheetpilesP.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 185
andtheothertwovariantsisthehugedemandforsteel,resultinginrelativelyhigh-energy consumptionandadistinctivelystrongerratingofitsHTP( Table4 ).Itisquestionableifsheet pilewallscanbepulledoutofthegroundafterthelongplanningperiodsofgroundwater remediation,especiallyifcorrosionprocessesmergethesteelwiththesurroundingsoilmatrix.If excavatingthewallisapracticableoption,thentherawmaterialgainedbyrecyclingyields savingsof50%forthedifferentcategories( Table4 ).Theimpactonaquaticeutrophicationand humantoxicitypotentialevendecreasesby75%.Itisapparentthattheunitimpactsforthewalls aredecidedbythematerialtheyareconstructedof,thoughtheyarealsoaffectedbythework involvedin(de-)installation. 11.FGSGate Themainprocessstepsinvolvinggateconstruction,maintenanceandrebuildingare: Soilandsedimentexcavation:eightgatesareinstalled,eachatadepthof17m.Pergate,a volumeof84m3soil,sandandgravelwasexcavated.Transportoftheexcavatedmaterialfor disposalisassumedtobe15km,asdescribedintheprevioussection.Thenominalcapacity ofthedredgerissetto70kWat5m3/hnetprogress. Cylindricalsteelcasings( Schadetal.,2000 )arepre-fabricated,perforatedbytwothirdsinthe inflowandoutflowfaceatthetubesides.Asanapproximationofsteelproductionand processing,thesamedatasetsareusedandequivalentassumptionsaremadeasinthecaseofthe steelstripsusedforthesheetpileinstallationmentionedearlier.Transportdistancesfrom manufacturingtoinstallationsitesaresetto200km.Afterthetubesarecarriedtothesite,a craneliftsthesteelcasingsegmentsintothegates.Approximately12tsteelwasspentforeach gateyieldingatotalof96tsteelfortheentireFGS.Forthecrane,atransportdistanceof100km isconsidered. Ineachgate,8m3claywasusedtosealtheopeningbetweenfunnelandgatecasing. Furthermore,50tgravelpergatewasneededtofilltheremainingvoidsandfocusthe groundwaterflowtothesteeltubes,whichweresubsequentlyfilledwithgranularactivated carbon(GAC).Thetreatmentmaterialwillbethesubjectofanextrachapter. DuringtheFGSinoperation,nofurtheractivities(exceptofGACexchange)areconsideredfor thegate.Thisisaveryconservativeassumption,sincemonitoringmaycauselowbutlong-term emissions(wellinstallation,periodicalpumping,sampling,transportation,labtests,etc.). However,especiallyduetotherelativelysmallexpectedoverallcontributionandtheverysitespecificcharacterofmonitoringactivities,theyaresetbeyondtheLCAboundaries. Itisanticipatedthatthegateconstructionwillbeexcavatedbycranesafterapre-specified planningperiod.Thesteelwillberecycledandtheremainingholebackfilledatdoublenet progresscomparedtoexcavation.Steelrecyclingshallfollowthesameprocedureas describedaboveforrecyclingthesteelofsheetpiles.Backfillissuppliedfromnearbygravel pitsatadistanceof20km. 12.Pump-and-treatsystem ThecoreelementsofthehypotheticalPTSarewellswithpumpingdevices,conduitsandan on-sitetreatmentfacility.Thetenwellsassumedforthiscasestudyfullypenetratetheaquiferto adepthof17m.Allwells(F300mm)aredrilledinadvanceoftheremedialoperation.Dieselis consumedfortransport(50km)andapplicationofthedrillingapparatus.Forthelatter,aP.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 186
respectivenominalcapacityof80kWat2.5m/hnetprogressisassumedandtheassociated emissionsareapproximatedaccordingto BUWALetal.(1994) .Plungedwellcasings(F100 mm)stabilisetheboreholewhileallowinggroundwatertoenterthewell.PVCischosenasthe materialforthepipes(50kgeach)usedforthecasings.Theemissionsarecalculatedaccording tothecomprehensiveprocesschainsanddatasuppliedby Boustead(1999) .Thecasings(density 1.5g/cm3)aresurroundedbyafilter,forwhichgravelisusedthatistransportedoveradistance of20kmfromnearbygravelpits. Theemissionscausedbymanufacturingthepumpshavenotbeenanalysedindetailduetoa presumablylowcontributiontotheLCAcategories.Instead,theemissionsareapproximatedby themassofsteelspentontheirfabrication,whichisveryconservativelysetto50kgperpump. Afterusage,thesteelisexpectedtoberecycled.Theelectricityneededforoperatingthepumps isapproximatedusingthemethodof Bayeretal.(2005a) ,assumingadegreeofefficiencyof 60%forthepumps(inclusivemotor),andanaverageheadlossof20m.Wecalculatedademand of0.15kWh/m3groundwaterprocessed. Thetotallengthoftheconduitsconnectingthewellswiththeon-sitetreatmentfacilityischosen as100m.Forthefabricationoftheconduits(F100mm),atotalmassof300kgPVCisestimated. Theconduitsdirecttheextractedgroundwaterintoon-sitevesselsforwhichonlythematerial (HDPE)consumedisanalysed.ForthecontainersfilledwithGAC,amaximumvolumeof20m3eachandawallthicknessof5cmisassumed,duetoroadtransportrestrictions.Afixedvalueof 100kmisselectedasthetransportdistancebetweenthesiteandmanufacturinglocationof conduitsandvessels. Thelongevityofconduits,vesselsandpumpsisassumedtobelimitedto10years,sothat theyhavetobereplacedeverydecade.Also,theproductivityofpumpingwellshastobe maintained,whichcommonlyaffordsregenerativeactionsandmaterialreplacement.Asarough approximation,toincludethisaspectintheLCAframework,10%oftheenergyandmaterials computedfortheinitialwellconstructionareaddedinat5-yearintervals. 13.Granularactivatedcarbon Theuseofgranularactivatedcarbon(GAC)isverycommonforgroundwaterclean-up, thoughonlyafewrecentapplicationsemployitin-situ( USEPA,1998;KraftandGrathwohl, 2003;Susaetaetal.,2005 ).Itisbasedonastraightforwardprinciple:thecommonly hydrophobicorganiccontaminantsingroundwatersorbonthesurfaceofthiscarbon-rich, highlyporousmaterialwhenwaterflowsthroughtherespectiveGACfilters.Afteracertain operationtime,whenthesorptioncapacityisexploited,theGACfillinghastobeexchanged. Forlong-termremediation,wheretreatmentfiltersarecontinuouslyfedwithcontaminated groundwater,thismeansaperiodicalexchangeofusedwithnon-usedmaterial.Ideally,used GACcanberecycledbythermaltreatmentandthenre-filled.Thereareseveralsourcesthat serveasrawmaterialfortheproductionofGAC,suchascoal,woodoranyorganicwaste. Accordingly,thefabricationstepsdiffer,forexamplehowtherawmaterialisproduced,then destructedandagglomerated,alsoaffectingthefinalsorptioncapacityduringwatertreatment. TheassumptionsmadefortheGACinthisworkareadoptedfromapreviousstudy( Bayeretal., 2005b )discussingtheproductionofGACoutofhardcoal.Afterpulverisationoftheraw material,abinderisadded,followedbyagglomerationintobriquettes,dryingandthermal activation.Thereafter,theproductissieved,packagedandtransported.Intotal,producing1ton GACaffords3tonsofhardcoal,consumes1600kWhpowerand12tonsofwatervapourthatis heatedbyburning330m3naturalgas.RegeneratinginsteadoffabricatingvirginGACisP.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 187
commonlyassociatedwithlessenvironmentalimpacts.Here,usedGACisheatedandthen treatedathightemperatureswithvapourtogasifyanddegradethecontaminants.This reactivationprocedureusuallycausesalossinmaterialweightofabout10%. ThedataqualityanddeepnessinassessingGAClifecycleprocessesadoptedfrom Bayeret al.(2005b) isjudgedasfair,especiallyduetothehypotheticalassumptionsmadeinregardto typeandprocessesinvolvedinthesupplyoftherawmaterial.Sincethereisnouniqueraw materialfortheproductionofGAC,itcanbeexpectedthat,forexample,inthecaseofusing regenerativeeductsinsteadofhardcoal,particularlytheinfluenceofGACconsumptiononthe depletionofenergyrecourses(DER)willbelower.However,thenumberofregeneration intervals,whichriseinthecourseofthelong-termremediation,reducetherelativeamountof virginGAC.Increasingly,regeneratedGACisapplied,whichisassumedtobeproduced throughthesameprocess,independentlyfromtheoriginalrawmaterial. 14.Comparisonofscenarios 14.1.Scenariosset-up Theset-upofdifferentscenariosisnecessarytocomparethetechnologiesdependingon variableanduncertainparametersthatcontrolsystemdesignandperformance.Uncertainties arisebothwithrespecttotheaccuracyofthebasicinventorydataaswellastothedescriptionof performanceanddurationoftheremediationtechnologies.Additionally,forthehypothetical PTS,therequiredtechnicaladaptationtothesitecannotexactlybespecified.Toaddressthis, severalscenariosaredesignedfordelineatingthespaceofpossibletechnologicalvariants.The focusissetondiscussingthemostenvironmentallyrelevantelementsandfactorsinorderto elaborateimprovementstrategiesforthetechnicaldesign.AmainfactorfortheLCAoutcomeis obviouslytheplanningperiod,whichisthereforevariedamongthescenariosinorderto examineitsinfluence.Furtherscenariosaredefinedafteraprecedingexaminationoftherelative impactofkeytechnicalelements,themostsignificantofwhichturnedouttobeGAC consumptionandtheamountofsteelused.Particularlytheirrolewillbequantitativelyinspected. Asidefromthis b scenariouncertainty Q ( Huijbregts,1998 ),thereisaconsiderableparameter uncertaintyintheunderlyinginventorydata,i.e.theestimatedemissionsandconsumptions, reflectingqualityandsubjectiveinterpretationofthedatasourceschosen.Simultaneously,the descriptionofthetechnicalelements,theiradoption,performanceandrelatedenergyconsumption issubjecttoacertaindegreeofuncertainty,especiallybecauseastaticrepresentationofthe prevailingconditionsattheKarlsruhesiteisusedhere.Thisassumptioniscontrarytothefactthat crucialnaturalprocessesarecommonlyhighlydynamic(e.g.,groundwaterflowandlongterm contaminantmovementintheaquifer).Amoredynamicmodellingandassessmentmethod, however,isbeyondofthescopeofthispaper.Thepresentedevaluationissolelybasedon temporallyaveragingfunctionsandthereforeusesnotime-dependentestimationsofenvironmentalimpacts.Thisisforthepurposeofsimplificationandtoavoidthehighlevelofeffortthat wouldberequiredtoappropriatelydelineatetheimperfectknowledgeoffutureprocesses,suchas throughtheuseofnumericalgroundwaterflowandtransportmodels. TheuncertaintyanalysishasbeencarriedoutusingtheCrystalBallsoftware(Decisioneering, Inc.).Followingtheapproachof Canteretal.(2002) ,auniformBeta-distribution(shape parameters a = b =2)wasassumedtocharacterisethe(potential)variabilityofparameters( Fig. 4 ).ThemeanoftheBeta-distributionissettotheexpectedvalueofeachparameter;therangeis scaledbythedegreeofuncertaintyoftherespectiveparameter( F 25%or F 50%).TheP.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 188
uncertaintyassociatedwiththeunderlyinginventorydataisassumedtobe F 50%,reflectinga moderateimpreciseness.Thisaveragevalueisanticipated,sinceexactuncertaintyestimatesor discussionsofexpecteddatavariabilityarescarce.Itissetslightlyhigherthandenotedby Volkweinetal.(1999) inordertoincludeestimationerrorswhentransferringconsumptionand emissionvaluestotheKarlsruhesite.Inaccuracyfromtheimprecisedeterminationofmaterial andenergyconsumptionisassumedtobe F 25%forfunnelandgateconstruction.Allfurther keyelementscategorisedabovearesubjecttoahigheruncertainty( F 50%),becausetheyevolve (partly)inthefuture(e.g.ongoingsupplyofGAC)orarehypotheticalestimates(PTSelements). TheresultsoftheLCAcarriedoutforthevariousscenariosarediscussedincomparisontoa referencescenario,whichshallserveasabasisforouranalysis( Table2 ).Startingfromthis scenario,furtherscenariosaredevelopedtoscrutinisetheinfluenceofthesetentative assumptionsandtohighlighttherelevanceofparticulartechnicalelements(e.g.,typeof verticalwall)aswellasservices(e.g.,sheetpilerecycling). 15.Referencescenario ThereferencescenariocomparesaFGSwithsheetpiles,asinstalledatthesite,toa hypotheticalPTS,configuredasdiscussedabovewithtenwellspumpingatatotalrateof18l/s. Fig.4.Schemeofbetadistributionfunctionwithshapefactors a =2and b =2,scaledwithinuncertaintyintervalaround expectedvalue E ( X ). Fig.5.Referencescenariowithcategory-specificindicators:depletionofenergyresources(DER),globalwarming potential(GWP),acidificationpotential(AP),terrestrialaswellasaquaticeutrophicationpotential(TEP,AEP), photochemicalozonecreationpotential(POCP)andhumantoxicitypotential(HTP).Indicatorshavebeennormalized againstinhabitantemissionsorconsumptionsandareexpressedasinhabitantequivalents(Ieq.). P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 189
TheGACconsumptionisassumedtoincreaselinearlywiththevolumeofwatertreatedand thereforeamountsto40tonsperyearfortheFGSand72tonsforthePTS.Theoperationtime, i.e.planningperiod,issetat30years.Toaccountfortheparameteruncertainty,aMonteCarlo analysiswascarriedoutbasedon10,000samples. Fig.5 depictstheresultscomputedforthe individualimpactcategories.Allpotentialimpactsareexpressedasinhabitantequivalents(Ieq.), i.e.impactsarenormalisedtothecategory-specificexpectedannualemissionsperinhabitantin Germany,whichareextrapolatedovertheoperationperiodof30years( Table4 UBA,2000 ). TheMonteCarloanalysisdeliversnotonlymeanvaluesoftheseindicators,butalsotheir probabilitydistributionsaccordingtothevariabilityassignedtoinventoryandtechnical parameters.Fromtheprobabilitydistributions,the5%and95%percentileswereselectedto delineatetheexpectedranges.Theseareshownascumulativeerrorbarsfortheindividual categories( Figs.5and6 ). ForthisscenarioandtheFGS,theimpactscausedbyfabricatingandinstallingthefunnel (sheetpiles)dominateinnearlyallcategories.Thepotentialconsumptionofenergyresources andtheemissionscausingglobalwarmingareespeciallyinfluencedbyGACduetotheusageof coalanditshightemperatureactivation.Themostsignificantimpactofthefunnelinstallation canbeobservedforthehumantoxicitypotential(HTP).Theinhabitantequivalentsindicate relativeemissionswhichareordersofmagnitudehigherthanthosecalculatedfortheother categories.Thisisparticularlyeffectedbythehighemissionrateofheavymetalsandfurther carcinogenby-productsduringtheproductionofsteel.Consequently,comparedtothePTS,the usageofsteelturnsouttobeamajordisadvantagefortheFGS.TheDERandGWPindicators determinedforPTSandFGSarenearlyequalforbothtechnologies,sincetheimpactsfrom higherconsumptionofGACassumedforPTSoutweighthosecalculatedforfunnelinstallation. 16.Alternativewalltechnologies Withinthisstudy,fortheKarlsruhesite,twoalternativeverticalwalltechnologiesare suggested.Thisiseitheraslurryora(thin)diaphragmwall.Thequestioniswhethertheimpacts associatedwithfunnelconstructioncanbeloweredifadifferentwalltypeisused.Theresults areshownin Fig.6 .Comparedwiththehugeenvironmentalburdenfromusageofsteelfora sheetpile( Fig.5 ),thepredominantlymineralmaterialsrequiredfortheconstructionofthetwo otherwalltypesyieldlowerimpactsinallcategories.Theindicatorscalculatedforthe Fig.6.ImpactscalculatedforthereferencescenarioassumingtwoalternativewalltypesfortheFGS,athindiaphragm wallandacementbentoniteslurrywall. P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 190
diaphragmandslurrywallareinthesamerangewithslightbenefitsfortheformer,whichcanbe primarilyattributedtolowerrawmaterialconsumption.Themostimportanteffect,when selectingamineral-basedwalltype,isthereductionofHTPtolessthanonetenthofthat calculatedforthesheetpile.Thesteelrequiredforcylindricalcasingsinstalledintheeightgates becomesadominatingfactorleadingtoanindicatorvalueofabout6inhabitantequivalents. Pleasenotethatitisassumedthatthesteelisrecycledaftertheoperationisceased. Weightingtriangles( Figs.7and8 ),orientedontheworkof Hofstetteretal.(2000) ,represent acondensedviewoftheresultsfromLCIA.Here,onlythepotentialdepletionofresources (DER),theexpectedencroachmentofecologicalquality(EQ)andthehazardouseffectson humanhealth(HH)aredistinguishedbyaveragingthecorrespondingindividualcategories.Each pointwithinthetriangledenotesaparticularrelativeweightingofthesethreeaspects.The maximumweightof100%issignifiedbyoneofthecornersanditgraduallydecreasesto0%for thespecificaspecttowardstheoppositesideofthetriangle.Dependingontherelativeweighting ofthethreeaspects,onetechnologyispreferableduetoalower(normalisedandaggregated) overallimpact,yieldingareasofprosforPTSorFGSintheweightingtriangles.These preferenceareaswerederivedfromtheMonteCarloanalysesinordertoincorporatethe prevailingdatauncertainty.Basedontheseanalyses,onlyrelativeweightings,whichfavourone remediationtechnologybymorethan75%ofthesamples,areassignedtothisalternative.Asa Fig.7.Weightingtrianglesbalancingpotentialdepletionofresources(DER),theexpectedencroachmentofecological quality(EQ)andthehazardouseffectsonhumanhealth(HH).Figs.6acdepictthedecreasingimpactsbychangingthe walltypefortheFGSfromsheetpiles(withrecyclingafter30years)toslurrywallsandthindiaphragmwalls.Between thepreferenceareasforPTSandFGS,awhiteareaofindifferencedelineateswherenoclearbenefitsforeither technologycanbeascertained. Fig.8.WeightingtrianglesforcomparisonbetweenPTSandFGSatdifferentoperationtimeswithtechnology-specific preferenceareas.ThewalltypeforthefunneloftheFGSconsideredisthethindiaphragmwall. P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 191
result,anareaofindifferencewithnoclearadvantageforonetechnologyexists.Thispresented conceptofincorporatinguncertaintybyMonteCarloAnalysesintoweightingtrianglesis,toour knowledge,newandcouldbedevelopedfurtherby3Ddiagramsorbyusingfuzzymembership functionstorepresenttheresults. Whenusingweightingtriangles,itispossiblethatonealternativeispreferablewithrespectto allcategories,atrivialcasewhichrenderstheoveralltriangleasapreferencearea.Suchatrivial case,forexample,isthereferencescenariowhereasheetpileisinstalledandnorecyclingofthe steelisassumed.Ifthesheetpileisde-installedandrecycledaftertheoperationperiodof30 years,theFGSispreferableonlyifthedepletionofenergyresourcesissetasthecriterion.The correspondingweightingtriangleisdepictedin Fig.7 a.BothPTSandFGSareindifferentwith respecttotheassociatedecologicalimpacts,butthesignificantdiscrepancyinadverseeffectson humanhealthfavorsPTSevenifonlyasmallweightisassignedtothisaspect.Asisshownin Fig.7 bandc,thepreferenceareasforFGSextendifaslurrywallisinstalledinsteadofasheet pile.ThediaphragmwallrevealsanevenbroaderpreferenceareafortheFGSthanforthePTS, withbenefitsdueespeciallytoaminorecologicalburden. 17.Influenceofoperationtimeandpredictionoflong-termperformance OneofthemostimportantfactorsfortheoutcomeoftheLCAistheplanningperiodover whichemissionsoccurandoverwhichresourcesareconsumed.Thereisacrucialdifferencein theeffectoftimeontheevaluationofactiveandpassiveremediationtechnologies.Active technologiessuchasPTScauserelativelysmallinitialimpacts(forconstructionand installation),butentailcontinuousimpactsovertheentireplanningperiod,whilepassive methods,suchasFGS,arecharacterisedbyarelativelyhighratioofinitialimpacts.Thisisthe reasonwhyFGSarecommonlyecologicallyaswellaseconomicallyinferiortoPTSwhenonly consideringtheinitialoperationphase.Thisimbalanceisexpectedtodisappearinthelong-term withbenefitsevolvingfortheFGS.TheroleoftimefortheLCAoftheremediationtechnologies inquestionisfurtherscrutinisedbytheweightingtrianglesshownin Fig.8 ac.Thewalltype withthebestLCAresults,thethindiaphragmwall,isselectedfortheFGS.Furthertechnical assumptionsareforthesameasthoseusedinthereferencescenario. Fig.8 ademonstratesthat forashorttimeperiodof10years,expectedlong-termadvantagesfortheFGSarenotyet accentuated.Extendingtheplanningphasebeyond10years,however,increasinglyfavorsthe FGSwithrespecttothedepletionofenergyresources.Thisisespeciallycausedbythehigher GACconsumptionforPTSandthepermanentdrainforpumpinggroundwater.Asidefromthis, thecalculationofinhabitantequivalents(Ieq.)heremeansanaveragingofallimpactsoverthe entireplanningperiod.Thelongerthetimeframeassumed,thelesstheannuallyaveragedimpact ofthedominatinginitiallycausedenvironmentalburdens.After20years,theoverallpotentialof theFGStodeferecologicaldamagesislowerthanthatoftherespectivePTS( Fig.8 b).Thetotal humanhealthimpactsforthePTSaccumulateovertime,and,ascanbeseenintheresultsfor50 years( Fig.8 c),approachtheimpactsoftheFGSwhicharepredominantlyassociatedwiththe constructionofthesteelcasings. Apossiblepointofdiscussionistheassumptionofequalclean-up(oroperation)timeforthe twocompletelydifferentremediationtechnologies.Infact,thetimerequiredtofinallyachieve theremedialobjectives,suchassufficientdecontamination,mayvaryamongthedifferent technologies.Forexample,itcouldbeanticipatedthatthehighergroundwaterflowrate associatedwiththeapplicationofaPTS,willspeeduptheremediation,throughahigher contaminantextractionrate.However,sincetodaysknowledgeissimplyinsufficienttoguessP.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 192
theexactremediationprogressofsuchapplicationsinthefuture,forecastinglong-termchanges ishighlyspeculative.Therefore,adetailedquantitativeexaminationoftheinfluenceof contaminantconcentrationvariationswillnotbeaddressedhere.TheconceptofLCAbasedon b planningperiods Q insteadofpotentialremediationtimesparticularlymakessense,because,after decadesofoperation,thereisahighprobabilitythatinnovativeapproacheswillsupersede todaysstandardtechnologies.Consequently,notremediationtimebuttechnicalevolutionwill bethedeterminingfactorforthe b grave Q oftheproductsandservicesconsideredwithinthe LCA,particularlybearinginmindthelongtimeframeinvolved. 18.InfluenceofGACconsumption ThereisahighuncertaintyinthepredictionofGACconsumptionovertime.Ifitissuspected thattheremediationoftheaquifercanbeachievedwithintheplanningperiod,e.g.within30 yearsforthereferencescenario,thenassumingaconstantcontaminantconcentrationinthe treatedgroundwaterovertimedoesnotseemrealistic.Thiswouldmeanthatthetransferof contaminantsintothegroundwaterstopsabruptlyafter30years.However,acommon observationisthattheconcentrationinthetreatedgroundwatergraduallydecreases.If concentrationsbecomelowerasaconsequence,thedemandonGACwillsimultaneously decline.Inviewofthis,theassumptionofaconstantGACconsumptionrateduringthe remediationappearsratherconservative. SincetheestimateddemandonGACisassumedtobehigherforthePTSthanfortheFGS,a declineoftheGACmassperm3groundwatertreatedwouldespeciallyfavourPTS.Theessential roleofGAC,especiallyforthecategoriesDER,GWPandPOCP,isdepictedin Fig.5 aandb andtheinfluenceofanyGACcutbackcanbeappraised.Anotherissueisthat,asGAClifetime isnotoftendeterminedexclusivelybyitssorptioncapacityandtheprevailingcontaminant concentrations,thismattercouldbeadisadvantagefortheFGS.Long-termapplicationsofGAC beartheriskofbio-foulingormineralclogging,causingareducedperformanceofthereactors andthusnecessitatinganearlyexchangeoftreatmentmaterial(KraftandGrathwohl2002). Since,duetothehighervolumeofGACperfill,thelifetimeofin-situ-reactorssignificantly exceedstheregenerationintervaltimeframeofanon-sitereactor(here5yearsfortheFGS,1 yearforthePTS),performancelossappearstobemorelikelyfortheFGS.FurthermorethePTS isalessstaticimplementation.ContrarytotheFGSwithfixedgateconfigurations,GACvessels canbedynamicallyreplacedbyothersofadifferentsizeinordertocontrolthelengthofthe regenerationperiod. Thereisasignificantinfluenceofthetechnology-specificGAContheimpactindicatorvalues, whichcannotbepreciselyestimatedandcouldbefurtheraddressedbymorecandidatescenarios. However,thenumberofpossiblealternativesisendless.Instead,weaskwhatthehypothetical PTSsGACdemandisnecessarytobalancetheoverallimpactsofbothtechnologies?TheFGSis configuredaccordingtotheKarlsruhesiteand,asdescribedforthereferencescenario,withGAC exchangesevery5yearsataconstantconsumptionrate.Additionally,thealternativeFGSusing differentwalltypesareinspected.TheconfigurationofthePTSagreeswiththereferencecase althoughtheGACvolumeissetasavariabletobeadaptedfordifferentplanningperiods.Please notethatnouncertaintyininputparametersisconsideredhere. Fig.9 acshowthelinesofindifference,indicatingtherelativeGACvolumeperyear requiredforthePTSandresultinginthesamepotentialimpactswithrespecttodepletionof energyresources(DER),ecologicalquality(EQ)andhumanhealth(HH).Abovethese b break eventrade-offs Q ,thehigherGACdemandsofPTSdelayabetterratingofFGSandviceversa.P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 193
DuetothetemporallyincreasingGACdemandandtheattenuatedrelativeimportanceofFGS installationimpacts,theselineshavehyperbolictrends.Regardlessofthewalltypeinstalledfor thefunnel,ifextremelylargeGACdemandsarepresumedforthePTS,longplanningperiods willbenecessarytoobtainthesameimpactsonHHasfortheFGS.EventheFGSwithathin diaphragmwall,whichinterferestheleastwithhumanhealthaspects,canonlycompetewiththe PTSatanoperationtimeof70yearsandonlyiftheannualrelativeGACdemandisassumedto beabout250%oftheFGSsdemand. Withrespecttootherclasses,theGACthresholdratiosarecomputedsignificantlylowerthan thosedenotingpotentialadverseeffectsonhumanhealth.Thisisbecauseofthehigherrelevance ofGACcomparedtoothertechnologicalelementswhenfocusingonDERandEQ,whichis especiallyinfluencedbythehighresourceandenergydemandforproduction,supplyand regenerationofGAC.Therefore,withrespecttoDER,alowerconsumptionofGACfortheFGS iscomputedthatleadstothesameoverallemissionsasforPTS.Weobserveasteeperdeclineof thetrade-offswithrespecttoEQversusDERsothatintersectionsoccurafter40yearsormore. Fig.9 acdepicttheGACPTS/GACFGSratiosassumedforthereferencecase(180%)andthe equalityline,atGACPTS/GACFGS=100%,asdashedlines.Anticipatingthattheextractionrate forthePTS(18l/s,FGS:10l/s)isnotunderestimated,an80%higherannualGACdemandfor thePTSmayserveasanupperlimit.Thisisbecausethecontaminantconcentrationscanbe expectedtobeequaltoor,duetodilution,lowerthanthosegivenfortheFGS.Further argumentsforthispostulationaretheabove-mentionedadaptabilityandthebettercontrolofthe on-sitetreatmentunit.Moreover,itisdoubtfulif,inallgates,themaximumsorptioncapacityof theGACcanbereachedanotherpotentialdisadvantageforFGS. Ascanbeseenforallthreewalltypes,humanhealthrelatedemissionsoftheFGSalways surmountthoseforPTSunderrealisticassumptionsfortheplanningperiodandPTSsGAC demand.Thisrelationshipisespeciallycontrolledbytheusageofsteel.Pleasenotethatforthe scenariosdiscussedhere,creditsfromsteelrecyclingarealreadyconsidered.Withrespectto DER,the180%thresholdcanalreadybereachedafter15yearsofoperationforallwalltypes.In contrast,thebreak-evenpointfortheimpactscomputedfortheecologicalqualityis1020years later.Asalreadydiscussedabove,funnelsmadeoutofthindiaphragmwallsarepreferableover thoseconstructedoutofsoilcementbentoniteslurrywallsandsteelsheetpiles,thusofferingan earlierbreak-evenpoint.Forexample,assumingan80%higherannualGACdemandforthePTS, Fig.9.ComparisonofPTSandFGSwithrespecttopotentialdepletionofresources(DER),theexpectedencroachment ofecologicalquality(EQ)andthehazardouseffectsonhumanhealth(HH),assumingthattheGACdemandforPTS (GACPTS)isnotknown.Trade-offcurvesdepicttherelativeconsumptionrateGACPTS/GACFGSwherethetotalclassspecificimpactsareequal,givenafixedestimatedconsumptionofGACFGS=40t/yearfortheFGS.Abovethecurves, FGSarebeneficial,below,PTS. P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 194
thedepletionofenergyresourcesisalreadyundercutbytheFGSafteronedecade.Iftherelative annualGACdemandislowerthan180%,thentheindifferencetrade-offsincreaseexponentially. IfthesameannualGACamountisconsumedforbothtechnologies(GACPTS/GACFGS=100%), thenperiodsofover40yearsarerequiredtoachievelowerimpactsusingtheFGS. 19.Conclusions Therearetwomaincontributionsofthisstudy:theset-upandtheapplicationofaLCA frameworkforcommonlong-termgroundwaterremediationtechnologies.Eachstepofthe LCAthedefinitionoffunctionalunits,systemboundariesandobjectives,therealisationofthe impactassessmentandthecomparisonoftechnology-specificresultsnecessitatedapurposebuiltapproachtofitthetechnologicalkeyelementsintotheLCAframeworkandtoobtainvalid results.Theapplicationsite,theformergasmanufacturingplantofthecityofKarlsruhe,enabled quantificationofrealisticestimatesoftheenvironmentalburdencausedbysecondaryimpactsof applyingremediationtechnologies. TheidealfunctionalunittobeadoptedforaLCAframeworkisshowntobethecontroland/ orremediationofasite.Thisisdefinedbycomplyingwithconcentrationlevels,sincethis reflectstheperceptionofequalfunctionalityaccordingtoriskassessmentobjectiveswithout examiningtheexactsystemperformance.Forthedefinitionofsystemboundariesaswellasthe spatialandtemporalresolutionoftheenvironmentallyrelevantprocesses,acompromisehadto befoundthatmostaccuratelyoutlinestheadverseeffectsonenvironmentandhumanhealth,but focusesonthemostcrucialprocessesandeffectsofthemanifoldtypesinvolved.Toachievethis, keytechnicalelementsaredistinguished,suchasfunnelinstallation,gateconstruction,pumping devicesandgranularactivatedcarbon(GAC)asthetreatmentmaterial.Theimpactassessment carriedoutthendeliversthespecificimpactswithrespecttodepletionofenergyresourcesand burdensonecologicalqualityorhumanhealth.Duetothedissimilitudeofthevariousprocesses involved,theunderlyinginventorydataforestimatingproduct-relatedmaterialdemand,energy consumptionandoccurringemissionsisbasedonvariousliteraturesources,databasesand manufacturersinformation.Asaconsequence,itishardlypossibletoguaranteeasimilarquality foralldatautilisedandfunctionalrelationshipsassumed.Additionally,availabledatabasesare commonlynotsufficientforobtaininganexactquantitativedescriptionofallproductsand serviceswithinthesystemboundariesfromcradle-to-graveasrecommendedforanapposite LCA.Inviewofthelimiteddataavailabilityandcomparability,acertaindegreeof b streamlining Q wasnecessary,whichheremeanttoforegoadistinctionofspatialandtemporal characteristicsoftheprocessesinvolved. Deficienciesindataqualityandtechnicaldescriptionweretackledbyanuncertaintyanalysis whichyieldedconfidenceintervalsoftheimpactcategoryspecificindicators.Althoughbroad rangesofuncertaintyofupto50%wereassumedfortheunderlyinginventorydata,cleartrends couldbeobservedandrecommendationsderivedforimprovingthetechnologieswithrespectto theassociatedenvironmentalburden.Aparamountfinding,asexemplifiedbytheKarlsruhesite andingeneral,isthatamaindeterminantfortheoutcomeofanLCAistheamountofsteelused forthetechnology.Evenifthelifecycleincludesrecyclingofmetals,significantemissions occuranditisdesirabletominimisethesteelconsumptionparticularlyduetotheassociated adverseeffectsonhumanhealth.ThisismuchmoreessentialfortheFGS,particularlyifsteel sheetpilesareconsideredforthefunnels.Butthecasingsofthegatesforthetreatmentmaterial attheKarlsruhesitearealsofabricatedoutofsteelinordertosecurethelong-termstabilityof thein-situconstructions.FordifferentsiteswithFGS,otherconstructions,suchascaissons,mayP.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 195
beconsideredforthegate,whichprobablydevourasmalleramountofsteelbutcanstillbe expectedtocontributetoapoorevaluationoftheFGSwithrespecttohumanhealthissues. Comparedtoexistingstudiesonthesecondaryimpactsofgroundwaterremediation technologies,differentcriticalelementsofthetechnologieshavebeenidentifiedas environmentallyrelevant.Thisisbecauseofthediscrepancyintypeandtimeframesofthe technologiesinspected,whichwascommonintherangeofyears,contrarytotheexpectation ofdecadesinthisstudy.Therefore,transportprocessesfortechnicalelementsordisposalof excavatedsedimentsplayonlyaminorrolehere,notwithstandingtheirsignificantinfluencefor short-termapproachessuchas b diganddispose Q applications.Long-termoperationsmeana continuoususeofenergyandtreatmentmaterial(i.e.,GAC),whicharethemainsourcesof secondaryimpactsforthePTS.FortheKarlsruhesite,iftheassumptionsongroundwater extractionrateandGACconsumptionforthehypotheticalPTSregardingtheresource depletioncategoryareveryconservative,itwilltake1530yearsuntilweobserveabreakevenpointforbothtechnologies(dependingonthewalltypeoftheFGS).Withrespecttothe ecologicaldamage,atimeframeofatleast10yearsisneededuntiltheimpactsfromFGS installationbalancewiththecontinuousimpactsforthePTScausedbypumpingandtreatment material. Asaconclusivestatementconcerningthecomparisonofsecondaryimpacts,itseemstobea disadvantageoftheFGSthatitrequiresahigherlevelofstability,andthereforedifferent,more resistantmaterialsforthein-situconstruction.Fromtheenvironmentalpointofview,itisthe pricewhichhastobepaidtoguaranteesteadinessoverthelongoperationperiodfortheFGS, whereasthePTSismuchmoreaccessible,flexibleandlessdemandingontechnological elements.WhenwetrytogeneralisetheresultsoftheKarlsruhecasetoothersites,ratherhigh pumpingratesarerequiredforthePTStoobtainsimilarorevenbetterLCAresultsthanforthe FGS.PleasenotethattheaquiferattheKarlsruhesitehasahighconductivity,sothatevenhere enormousextractionrateswouldbenecessarytopumpthegroundwater.Despitethis disadvantageforthePTSandtheexpectationthatGACconsumptionratesriselinearlywith theflowrate,thisdoesnotinevitablyleadtoenvironmentalcreditsfortheFGS.Itcannotbe ascertainedifothercommontreatmentmaterials,suchaszero-valentiron(ZVI),canavailthe FGSorifcontinuouswallsarepreferableduetopotentiallylowerimpacts.However,itcanbe anticipatedthattheproductionofZVI,eitherasaprimaryorsecondarywasteproduct,andits longevitywheninstalledin-situ,willplayacentralrole. Amainquestionisiftheassumptionofastaticsystemisappropriateandifitisadequateto extrapolatetheenvironmentalburdenfromtodayspointofviewtothefutureoverdecades.On theonehand,theprogressofthegroundwaterremediationapplicationmaychangethe technologicalrequirementsinthefuture,suchastheGACdemand,whichiscontrolledbythe floworextractionrateofgroundwaterandthecontaminantconcentrations.Ontheotherhand, technicalevolutionwillinfluence,ideallyreduce,theemissionsreleasedduringthemanufacture oftechnicaldevices,productionofGACandenergyproduction.Finally,inviewofthelongtime framesofupto70yearsexaminedhere,theLCAassumptionsandthevaluationofimpactsmay change.However,itseemedimpracticabletoassesstheremediationtechnologiesfroma perspectivedifferentthantodays,andtherefore,representsourbestguessforthefuture. Definingastaticperformancefortheremediationtechnologiesisarigoroussimplification, whichismeantasastartingpointforfurtherworkwithamorerealisticconsiderationofdynamic processes.However,thisstudyhasalreadybroughtlighttothemajorsourcesofpotential environmentaldamages,whichisessentialtodeducerecommendationsfora b greendesign Q of theselong-termoperatingremediationtechnologies.P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 196
Acknowledgements FundingforthisresearchisprovidedbytheGermanDepartmentofEducation,Science, ResearchandTechnology(BMBF),contractNo.02WT0019,andtheGermanResearch Foundation(DeutscheForschungsgemeinschaftDFG),grantBA2850/1.Theauthors acknowledgethecontributionsmadebyEddaHeuerandUteKarlfromtheFrenchGerman InstituteforEnvironmentalResearch(Karlsruhe)forthesetupoftheLCAframeworkand inventorydatabase.Wehighlyappreciatethesuggestionsoftwoanonymousreviewerswho helpedtoimprovethepaper,especiallywithrespecttothepresentationoftheLCAframework. SpecialthanksareduetoClaireAanes,UrsulaMcKnightandFernandoDAffonsecafortheir assistanceinpreparingthemanuscript. ReferencesBauman,H.,Tillman,A.-M.,2004.TheHitchHikersGuidetoLCA:AnOrientationinLifeCycleAssessment MethodologyandApplication,FirstEdition.543pp. Bayer,P.2004.Modelling,economicassessmentandoptimisationofin-situgroundwaterremediationsystems.PhD thesis,CentreforAppliedGeosciences,UniversityofTuebingen,Tu ¨bingerGeowissenschaftlicheArbeiten,ReiheC, Nr.70,78pp. Bayer,P.,Finkel,M.,Teutsch,G.,2003.Bridgingthegapbetweenconventionalpump-and-treatsystemsandpermeable walls:Hydraulicandeconomicaspectsofbarrier-supportedpump-and-treatsystems.Proc.of8thInternationalFZK/ TNOConferenceonContaminatedSoil(CONSOIL),vol.1216.Gent,Belgium,May,pp.15921600. Bayer,P.,Finkel,M.,Teutsch,G.,2005a.Cost-optimalContaminantPlumeManagementwithaCombinationofPumpand-treatandPhysicalBarrierSystems,GroundWaterMonitoringandRemediation25(2),pp.96108. doi:10.1111/ j.1745-6592.2005.0022 Bayer,P.,Heuer,E.,Karl,U.,Finkel,M.,2005b.Economicalandecologicalcomparisonofgranularactivatedcarbon (GAC)adsorberrefillstrategies.WaterResearch39,17191728. Beinat,E.,vanDrunen,M.A.,Janssen,R.,Nijboer,M.H.,Koolenbrander,J.G.M.,Okx,J.P.,Schutte,A.R.,1997.The RECDecisionSupportSystemforComparingSoilRemediationOptionsAmethodologyBasedonRiskReduction, EnvironmentalMeritandCostsCUR/NOBIS,Gouda,(September). Bender,A.,Volkwein,S.,Battermann,G.,Hurtig,H.-W.,Klo ¨pffer,W.,Kohler,W.,1998.In:Telford,T.(Ed.),Life CycleAssessmentforRemedialActionTechniques:MethodologyandApplication,ContaminatedSoil,vol.98, pp.367376. Blanc,A.,Me tivier-Pignon,H.,Gourdon,R.,Rousseaux,P.,2004.Lifecycleassessmentasatoolforcontrollingthe developmentoftechnicalactivities:applicationtotheremediationofasitecontaminatedbysulfur.Advancesin EnvironmentalResearch8,613627. Borken,J.,Patyk,A.,Reinhardt,G.A.,1999.Basisdataforenvironmentalbalancingcommercialvehiclesapplicationin transport,agricultureandmining.ViewegVerlag.(InGerman:Basisdatenfu ¨ro ¨kologischeBilanzierungen-Einsatz vonNutzfahrzeugeninTransportLandwirtschaftundBergbau). Boustead,I.,1999.Eco-ProfilesofPlasticsandRelatedIntermediates.PreparedfortheEuropeanCentreforPlasticsin theEnvironmentoftheAssociationofPlasticsManufacturersinEurope(APME).Bru ¨ssel. BUWAL,Umwelt,Bundesamtfu ¨r,derSchweiz,WalsundLandschaft(Eds.),PollutantEmissionsandFuelConsumption ofBuilding/Construction/EdificialMachines,UmweltmaterialienNr.23,Bern(inGerman:Schadstoffemissionenund TreibstoffverbrauchvonBaumaschinenSynthesebericht). Canter,K.G.,Kennedy,D.J.,Montgomery,D.C.,Keats,J.B.,Carlyle,W.M.,2002.Screeningstochasticlifecycle assessmentinventorymodels.InternationalJournalofLifeCycleAssessment7(1),1826. Carey,M.A.,Fretwell,B.A.,Mosley,N.G.,Smith,J.W.N.,2002.Guidanceontheuseofpermeablereactivebarriersfor remediatingcontaminatedgroundwater.NGWCLCreportNC/01/51,Solihull. Corradini,R.,Ko ¨hler,D.,1999.HolisticBasicMaterialandSemi-FinishedProductsAccounting:PartIII.Metals(in German:GanzheitlicheBilanzierungvonGrundstoffenundHalbzeugen-TeilIIIMetalle).Forschungsstellefu ¨r Energiewirtschaft(FfE),Mu ¨nchen,EuropeanCentreforPlasticsintheEnvironmentoftheAssociationofPlastics ManufacturersinEurope(APME).Eco-ProfilesofPlasticsandrelatedIntermediates.Bru ¨ssel.127pp. Diamond,M.L.,Page,C.A.,Campbell,M.,McKenna,S.,Lall,R.,1999.Life-cycleframeworkforassessmentofsite remediationoptions:methodandgenericsurvey.EnvironmentalToxicologyandChemistry18(4),788800. P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 197
Eberhardt,C.,Grathwohl,P.,2002.Timescalesofpollutantsdissolutionfromcomplexorganicmixtures:blobsand pools.JournalofContaminantHydrology59(12),4566. FetterJr.,C.W.,1999.ContaminantHydrogeology,SecondEdition.Prentice-HallPublishingCompany,UpperSaddle River,NJ.500pp. Frischknecht,R.,Bollens,U.,Bosshart,S.,Ciot,M.,Ciseri,L.,Doka,G.,Hischier,R.,Martin,A.,Dones,R.,Gantner, U.,1996.EcologicalInventoriesforEnergySystemsBasicPrinciplesfortheEnergySystemsEcological Comparison/EquationandforEnergySystemsInclusioninEcologicalFinancialStatementsforSwitzerland.Zu ¨rich (inGerman:O ¨ koinventarefu ¨rEnergiesysteme-Grundlagenfu ¨rdeno ¨kologischenVergleichvonEnergiesystemenund denEinbezugvonEnergiesystemeninO ¨ kobilanzenfu ¨rdieSchweiz). Godin,J.,Me nard,J.-F.,Hains,S.,Desche nes,L.,Samson,R.,2004.Combineduseoflifecycleassessmentand groundwatertransportmodelingtosupportcontaminatedsitemanagement.HumanandEcologicalRiskAssessment 10(6),10991116. Grathwohl,P.,1998.DiffusioninNaturalPorousMedia:ContaminantTransport,Sorption/DesorptionandDissolution Kinetics.KluwerAcademicPublishers,Boston.207pp. Hofstetter,P.,Braunschweig,A.,Mettier,Th.,Mu ¨ller-Wenk,R.,Tietje,O.,2000.Themixingtriangle:correlationand graphicaldecisionsupportforLCA-basedcomparisons.JournalofIndustrialEcology3(4),97115. Huijbregts,M.A.J.,1998.ApplicationofuncertaintyandvariabilityinLCA:PartI.Ageneralframeworkfortheanalysis ofuncertaintyandvariabilityinlifecycleassessment.InternationalJournalofLifeCycleAssessment3(5),273280. IFU,Institutfu ¨rUmweltinformatik,2001.Umberto3.5SoftwareforEnvironmentalManagement,Hamburg(in German:Softwarefu ¨rdasbetrieblicheStoffstrommanagement). ISO,InternationalOrganizationforStandardization,1997.EnvironmentalManagement-LifeCycleAssessment Principlesandframework.EuropeanstandardENISO14040.Geneva.11pp. ISO,InternationalOrganizationforStandardization,1998.EnvironmentalManagement-LifeCycleAssessmentGoal andScopeDefinitionandLifeCycleInventoryAnalysis.EuropeanstandardENISO14041.Geneva.16pp. ISO,InternationalOrganizationforStandardization,2000a.EnvironmentalManagement-LifeCycleAssessmentLife CycleImpactAssessment.EuropeanstandardENISO14042.Geneva.11pp. ISO,InternationalOrganizationforStandardization,2000b.EnvironmentalManagement-LifeCycleAssessmentLife CycleInterpretation.EuropeanstandardENISO14042.Geneva.18pp. Javandel,I.,Tsang,C.F.,1986.Capture-zonetypecurves:atoolforaquifercleanup.GroundWater24,616625. Kohler,N.,1999.MaterialFlowandCostsinHabitationandConstructionSectors.SpringerVerlag(inGerman: Stoffstro ¨meundKostenindenBereichenWohnenundBauen). Kraft,S.,Grathwohl,P.,2003.Investigationoflong-termin-situremovaloforganiccontaminantsfromgroundwater usingactivatedcarbon.Grundwasser8(1),2331(inGerman:UntersuchungenzumLangzeiteinsatzderIn-situAktivkohlefiltrationzurEntfernungvonorganischenSchadstoffenausdemGrundwasser). LfU,Landesanstaltfu ¨rUmweltschutzBaden-Wu ¨rttemberg(Eds.),1995.Controlofgroundwatercontaminationsbythe useofslurryandthinwall.HandbuchAltlastenundGrundwasserschadensfa ¨lle,Karlsruhe(inGerman:Sicherungvon AltlastenmitSchlitz-oderSchmalwa ¨nden). LfU,Landesanstaltfu ¨rUmweltschutzBaden-Wu ¨rttemberg(Eds.),1998.Softwareforenvironmentalbalancingof contaminatedsites(inGerman:SoftwarezurUmweltbilanzierungvonAltlastensanierungsverfahren). Meggyes,T.,Simon,F.-G.,2000.Removaloforganicandinorganicpollutantsfromgroundwaterusingpermeable reactivebarriers:Part2.Engineeringofpermeablereactivebarriers.LandContaminationandReclamation8(3), 175187. NRC,NationalResearchCouncil,1997.InnovationsinGroundWaterandSoilCleanup.FromConceptto Commercialization.NationalAcademyPress,WashingtonD.C.292pp. Page,C.A.,Diamond,M.L.,Campbell,M.,McKenna,S.,1999.Life-cycleframeworkforassessmentofsiteremediation options:casestudy.EnvironmentalToxicologyandChemistry18(4),801810. Pearlman,L.,1999.SubsurfaceContainmentsandMonitoringSystems:BarriersandBeyond(OverviewReport).U.S. EnvironmentalProtectionAgency,TechnologyInnovationOffice, www.epa.gov/tio/products/intern/pearlman/ finalpap.htm ,61pp. Pennington,D.W.,Potting,J.,Finnveden,G.,Lindeijer,E.,Jolliet,O.,Rydberg,T.,Rebitzer,G.,2004.Lifecycle assessment:Part2.Currentimpactassessmentpractice.EnvironmentInternational30(5),721739. Potting,J.,2000.Spatialdifferentiationinlifecycleimpactassessment.InternationalJournalofLifeCycleAssessment5 (2),77. Rebitzer,G.,Ekvall,T.,Fischknecht,R.,Hunkeler,D.,Norris,G.,Rydbverg,T.,Schmidt,W.-P.,Suh,S.,Weidema,B.P., Pennington,D.W.,2004.LifeCycleAssessment:Part1.Framework,goalandscopedefinition,inventoryanalysis, andapplications.EnvironmentInternational30(5),701720. P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 198
Ross,S.,Evans,D.,2002.Excludingsite-specificdatafromtheLCAinventory:howthisaffectslifecycleimpact assessment.InternationalJournalofLifeCycleAssessment7(3),141150. Russell,K.T.,Rabideau,A.J.,2000.Decisionanalysisforpump-and-treatdesign.GroundWaterMonitoringand Remediation20(3),159168. Schad,H.,Haist-Gulde,B.,Klein,R.,Maier,D.,Maier,M.,Schulze,B.,2000.Funnel-and-gateattheformer manufacturedgasplantsiteinkarlsruhe:sorptiontestresults,hydraulicandtechnicaldesign,construction. Contaminatedsoil2000.ProceedingsoftheSeventhInternationalFZK/TNOConferenceonContaminatedSoil, Leipzig,Germany,1822,pp.951959.September. Shakweer,A.F.,Nathanail,P.C.,2003.LCAMethodologyforRemediationStrategySelection.ConSoil20038th InternationalFZK/TNOConferenceonContaminatedSoil.Gent,Belgium,pp.1216.May. Starr,R.C.,Cherry,J.A.,1994.In-situremediationofcontaminatedgroundwater:thefunnel-and-gatesystem.Ground Water32(3),465476. Stroo,H.F.,Unger,M.,Ward,C.H.,Kavanaugh,M.C.,Vogel,C.,Leeson,A.,Marqusee,J.,Smith,B.P.,2003. Remediatingchlorinatedsolventsourcezones.EnvironmentalScienceandTechnology37(11),224A230A. Sue `r,P.,Nilsson-Pa ledal,S.,Norrman,J.,2004.LCAforsiteremediation:aliteraturereview.SoilandSediment Contamination13(4),415425. Susaeta,I.,Aguirre,I.,Cagigal,E.,Bonilla,A.,2005.GACadsorptionbarriersfortheremovaloforganicsfrom groundwaters.In:Gavaskar,A.R.,Chen,A.C.C.(Eds.),RemediationofChlorinatedandRecalcitrantCompounds. UBAUmweltbundesamt,1998.EnvironmentalBalancingofGraphicPapers,FinalReport.Heidelberg(inGerman: O ¨ kobilanzgraphischerPapiere,Endbericht). UBAUmweltbundesamt,2000.EnvironmentalBalancingofDrinksPackingIIMainPart.UBA-Texte37/00.Berlin(in German:O ¨ kobilanzfu ¨rGetra ¨nkeverpackungenIIHauptteil). UdodeHaes,H.A.,Jolliet,O.,Finnveden,G.,Hauschild,M.,Krewitt,W.,Mu ¨ller-Wenk,R.,1999.Bestavailable practiceregardingimpactcategoriesandcategoryindicatorsinlifecycleimpactassessment.InternationalJournal LCA4(2),6674. USEPA,1996.Pump-and-treatGroundWaterRemediationAGuideForDecisionMakersandPractioners.United StatesEnvironmentalProtectionAgency,EPA625/R-95/005.WashingtonD.C. USEPA,1998.EvaluationofSubsurfaceEngineeredBarriersatWasteSites.EPA542/R-98/005.WashingtonD.C. USEPA,2002.Fieldapplicationsofin-situremediationtechnologies:permeablereactivebarriers.WashingtonD.C. Vignes,R.,2001.Uselimitedlife-cycleanalysisforenvironmentaldecisionmaking.ChemicalEngineering,4054. Volkwein,S.,Hurtig,H.-W.,Klo ¨pffer,W.,1999.Lifecycleassessmentofcontaminatedsitesremediation.International JournalLCA4,263274. ZAG,CenterforAppliedGeoscience,2003.Technical,economicalandecologicalassessmentofreactivewalls:Final report(inGerman:Technisch-umwelto ¨konomischeBewertungReaktiverWa ¨nde.Projektabschlu h bericht),LAG03 01/0346,Tu ¨bingen,http://safira.ufz.de:7777/frames eng3.html P.Bayer,M.Finkel/JournalofContaminantHydrology83(2006)171199 199
1 L IFE CYCLE ANALYSIS OF F OUR GROUNDWATER REMEDIATION TECHNOLOGIES FOR IRON BACTERIA SMOTHERING IN WETLANDS ADJACENT TO AN OPEN DUMP LANDFILL IN NORTHWEST FLORIDA By SANDRA K. GAYNOR A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR T HE DEGREE OF MASTER OF ENGINEERING UNIVERSITY OF FLORIDA 2013
2 2013 Sandra K. Gaynor
3 To m y m om w ho b elieves in me
4 ACKNOWLEDGMENTS I am forever eternally grateful to the Lord Jesus for His great love that is freely bestowed on me every day and for providing me with strength and amazing grace for each new day. I want to thank my mother for the hours she spent teaching me how to read, w rite, spell, and do math. I am thankful for her tireless devotion to providing me the loving home environment whereby which I could learn easily and grow in the knowledge of our environment I want to thank my many Christian friends who have prayed me thr ough the graduate classes, s pecifically Ms. Lori Higgs, each Saturday praying for my personal needs and the needs of my children; Ms Jan Peters for her selfless giving and hospitality, sharing food and a bed each time I pass through Tallahassee, FL gathe ring information for graduate work. I also want to thank m y aunt, Ms. Chan West for her boundless energy and love for preserving the environment for the next generation.
5 TABLE OF CONTENTS Page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 7 LIST OF FIGURES .......................................................................................................... 9 LIST OF OBJECTS ....................................................................................................... 12 LIST OF ACRONYMS ................................................................................................... 13 ABSTRACT ................................................................................................................... 16 CHAPTER 1 INTRODUCTION .................................................................................................... 18 1.1 Problem ............................................................................................................ 18 1.2 Goals and Objective.......................................................................................... 19 1.3 General ............................................................................................................. 20 1.4 Regulatory Drivers ............................................................................................ 21 1.4.1 Federal Regulatory Background .............................................................. 21 1.4.2 Federal Directives .................................................................................... 21 1.4.3 Water Quality Standards ......................................................................... 22 2 SITE DESCRIPTION .............................................................................................. 26 2.1 General ............................................................................................................. 26 2.2 Site Assessment Background ........................................................................... 26 3 LITERATURE REVIEW .......................................................................................... 41 3.1 Life Cycle Analysis Literature ............................................................................ 41 3.1.1 General Discussion ................................................................................. 41 3.1.2 LCA System Boundaries ......................................................................... 45 3.1. 3 Remedial Action Periods ......................................................................... 45 3.1.4 Groundwater Fate and Transport of Contaminants ................................. 46 3.1.5 Long Term Monitoring (LTM) Period ........................................................ 47 3.1.6 Capital Equipment and Infrastructure Exclusions .................................... 47 3.1.7 Greenhouse Gases ................................................................................. 48 3.1.8 Energy Consumption ............................................................................... 48 3.2 Green Sustainability Remediation (GSR) .......................................................... 49 3.3 Summary .......................................................................................................... 50 4 NORTHWEST FLORIDA SITE CONDITIONS ........................................................ 51
6 4.1 Unlined Landfills ............................................................................................... 51 4.2 Biological Assessments in Streams .................................................................. 52 4.3 Florida Soils ...................................................................................................... 53 4.4 Iron Redox and Mobilization in GW ................................................................... 53 4.5 Microbial Activity in Aquifers ............................................................................. 55 5 METHODOLOGY ................................................................................................... 68 5.1 SiteWise LCA/GSR ........................................................................................... 68 5.1.1 Remediation Action Investigation ............................................................ 69 5.1.2 Remedial Action Construction ................................................................. 69 5.1.3 Remedial Action Operation ...................................................................... 70 5.1.4 Long Term Monitoring ............................................................................. 70 5.2 Da ta Sources .................................................................................................... 70 5.4 Remediation Technology Selection ................................................................... 71 5.4.1 Limestone Interceptor Trench System (LITS) .......................................... 72 5.4.2 Groundwater Pump and Treat ................................................................. 73 5.4.3 Air Sparging ............................................................................................. 73 5.4.4 Monitored Natural Attenuation ................................................................. 74 6 RESULTS AND DISCUSSION ............................................................................... 85 6.1 General ............................................................................................................. 85 6.2 Results .............................................................................................................. 85 6.2.1 Limestone Trench Scenarios ................................................................... 85 184.108.40.206 Heavy equipment ........................................................................... 87 220.127.116.11 Active vs passive trench system .................................................... 87 6.2.2 Groundwater Pump and Treat ................................................................. 87 6.2. 3 Air Sparging ............................................................................................. 88 6.2.4 Monitored Natural Attenuation (MNA) ...................................................... 88 6.3 Discussion ........................................................................................................ 89 6.3.1 General .................................................................................................... 89 6.3.2 LCAs with MNA vs. Groundwater Pump and Treat ................................. 91 6.3.3 Impacts from Remedial Investigation and LTM Phase ............................ 92 6.3.4 Transportation ......................................................................................... 92 6.3.5 Long T erm Monitoring Periods and Point of Compliance ........................ 9 3 6.3.6 Limestone and Crushed Concrete Permeable Barrier Media .................. 95 6.3.7 Unconventional Approaches to Remediation ........................................... 97 7 SUMMARY AND CONCLUSION .......................................................................... 117 APPENDIX: ADDITIONAL DATA ................................................................................ 123 REFERE NCES ............................................................................................................ 124 BIOGRAPHICAL SKETCH .......................................................................................... 129
7 LIST OF TABLES Table Pa ge 1 1 SiteWise modeled scenario remediation technologies for OU1 landfill NAS Pensacola ........................................................................................................... 23 1 2 Florida drinking water standards for OU1 landfill at NAS Pensacola, FL site specific standards (FAC 62550) ........................................................................ 24 1 3 Florida surface water standards for OU1 landfill at NAS Pensacola, FL site specific standards (FAC 62302) ........................................................................ 24 1 4 Best available technology (BAT) discharge limits for acid mine wastewater (Mohana, 2006; 40 CFR Part 434) ..................................................................... 25 2 1 Site specific surface water quality results for iron wetland 3, wetland 4D and OU1 landfill NAS Pensacola, FL [CAS 743989 6] (Navy NIRIS, 2013) ............. 33 2 2 Site specific groundwater analysis iron in shallow background wells OU1 landfill NAS Pensacola, FL [CAS 743989 6] (Navy NIRIS, 2013) ...................... 34 2 3 Site specific groundwater analysis iron in shallow wells OU1 landfill NAS Pensacola, FL [CAS 743989 6] (Navy NIRIS, 2013) ......................................... 34 2 4 Site specific groundwater analysis iron in piezometer well at limestone trench OU1 landfill NAS Pensacola, FL [CAS 7439896] (Navy NIRIS, 201 3) .. 35 2 5 Site specific groundwater analysis iron in intermediate wells OU1 landfill NAS Pensacola, FL [CAS 743989 6] (Navy NIRIS 2013) ................................. 37 2 6 Site specific analysis of iron in soils & sediments at OU1 landfill & wetland 3 [CAS 7439 89 6] (Navy NIRIS, 2013) ................................................................. 39 4 1 Iron content in soil samples at northwest Florida county landfills (Subramaniam, 2007) ......................................................................................... 61 4 2 Biological assessments in streams adjacent to unlined landfills in northwest Florida (Ray, 2013) ............................................................................................. 62 4 3 Chemolithoautotrophs: energy source and waste products (Dyer, 2003) ........... 65 4 4 Use of redox sensitive compounds for assignment of redox conditions in groundwater aquifers (Christensen, 2000) ......................................................... 66 4 5 Criteria and threshold concentrations for id entifying redox processes in groundwater (McMahon and Chapelle, 2008) .................................................... 67 5 1 Modeled remediation technology summary input parameter s ............................ 83
8 5 2 Modeled scenario remediation technology implementability, effectiveness and cost summary .............................................................................................. 84 6 1 SiteWise sustainability matrix groups ............................................................... 100 6 2 SiteWise sustainability metric quantified for each alternative global warming & air .................................................................................................................. 101 6 3 SiteWise sustainability metrics quantified for each alternative other .............. 101 6 4 SiteWise relative impact ................................................................................... 102 6 5 SiteWise remedial investigation phase priority air pollutants environmental impact (%) ........................................................................................................ 102 6 6 Remedial investigation phase greenhouse gas emission environmental Impact (%) ........................................................................................................ 103 6 7 Remedial investigation phase energy consumption environmental impact (%) 103
9 LIST OF FIGURES Figure P age 2 1 NAS Pensacola, FL OU1 landfill general site map. Figure courtesy of US Navy, 2008. ........................................................................................................ 28 2 2 NAS Pensacola, FL OU1 landfill historical groundwater wells and sample locations. Figure courtesy of US Navy, 2008. ..................................................... 29 2 3 NAS Pensacola, FL OU1 landfill current groundwater monitoring wells. Figure courtesy of US Navy, 2008. ..................................................................... 30 2 4 Sampling location SW 01 groundwater seep and bacteria growth wetlands 3 OU1 landfill NAS Pensacola, FL. Photo courtesy of Ensafe, 2012. ................. 31 2 5 Sampling location SW 01 groundwater seep and bacteria growth wetlands 3 OU1 landfill NAS Pensacola, FL. Photo courtesy of Ensafe, 2012. .................... 31 2 6 Surface water sampling location SW 02 bacteria growth OU1 landfill NAS Pensacola wetlands 3, FL. A) close up view at SW03; B) view at SW03 pipe crossing roadway. Photo courtesy of Kathy Gaynor, 2012. ................................ 32 4 1 Spatial distribution of total iron concentrations based on soil suborders (Valcare and Townsend 2011 as cited in Chen et al., 1999) .............................. 58 4 2 Redox zones vs depth (Mill ero, F.J., 1996). ....................................................... 59 4 3 Reducing conditions in groundwater aquifer underlying unlined landfill (Christensen, T.,et al 2000). .............................................................................. 60 5 1 Aerial View of Surface Water Monitoring Points in at OU1 Landfill, NAS Pensacola, FL. Photo courtesy of US Navy, 2012. ............................................. 76 5 2 Groundwater and surface water sampling locations at OU1 landfill, NAS Pensacola, FL. Figure courtesy of US Navy, 2010. ............................................ 77 5 3 Limestone interceptor system trenching heavy equipment 2,000 hp. A) barrier wall trenching equipment with boom exposed, B) barrier wall trenching equipment with boom fully deployed, C) caricature of barrier wall trenching equipment in cross sectional view, D) HDPE pipe with sock being installed in trencher at some time as backfill material. Photos courtesy of Dewind Dewatering 2013. ................................................................................. 78 5 6 Limestone trench construction 675 LF utilized in 1999 at OU1 landfill NAS Pensacola, FL. Photo courtesy of US Navy, 1999. ............................................. 79
10 5 5 Limestone trench construction 675 LF in 1999 at OU1 landfill NAS Pensacola, FL. Photo courtesy of US Navy, 1999. ............................................. 79 5 7 Limestone Trench 675 LF 1999 as built construction details OU1 landfill NAS Pensacola, FL. A) cross sectional view of trench at 20 LF depth, B) riser pipe details, C) cross sectional detail with HDPE pipe in bottom for active pump & treat of groundwater. Figures courtesy of US Navy, 2009. ......... 80 5 8 Modeled scenario 675 LF limestone trench footprint. Photo courtesy of US Navy, 2012. ........................................................................................................ 81 5 9 Modeled scenario 1300 LF limestone trench footprint. Photo courtesy of US Navy, 2012. ........................................................................................................ 81 5 10 Modeled scenario groundwater pump and treat system footprint. Photo courtesy of US Navy, 2012. ................................................................................ 82 5 11 Modeled scenario air sparging system footprint. Photo courtesy of US Navy, 2012. .................................................................................................................. 82 6 1 Greenhouse gas emissions from all sources during the modeled SiteWise period. .............................................................................................................. 104 6 2 Energy used from all sources during the modeled SiteWise period. ................. 105 6 3 Water consumed at site & not returned to groundwater system on site impacts from all sources during the modeled SiteWise period. ........................ 106 6 4 NOx emissions from all sources during the modeled SiteWise period. ............. 107 6 5 SOx impacts from all sources during the modeled SiteWise period. ................ 108 6 6 PM10 emissions from all sources during the modeled SiteWise period. ............ 109 6 7 Accident risk fatality to all workers during the modeled SiteWise period. ......... 110 6 8 Accident risk injury to all workers during the modeled SiteWise period. ........... 111 6 9 Non hazardous waste disposed from all sources during the modeled SiteWise period. ............................................................................................... 112 6 10 Hazardous waste disposed from all sources during the modeled SiteWise period. .............................................................................................................. 113 6 11 Costing for remediation technologies from all sources during the modeled SiteWise period. ............................................................................................... 114
11 6 12 Lost hours injury from all sources during the modeled SiteWise peri od. .......... 115 6 13 Final cost with footprint reduction from all sources during the modeled SiteWise period. ............................................................................................... 116
12 LIST OF OBJECTS Object P age A 1 SiteWise Remediation Technology Input Data ..................................................... 123 A 2 Cost Estimate Supporting Data for Remediation Technologies ............................ 123
13 LIST OF A CRONYMS AMD Acid M ine D rainage ARAR Applicable or Relevant and Appropriate Requirement BAT Best Available Technology BMP B est Management P ractice BOD B iochemical O xygen D emand C&D Construction Demolition Debris COD C hemical O xygen D emand CERCLA Comprehensive Environmental Response, Compensation, and Liability Act of 1980, as amended CH4 Methane gas CO2 Carbon Dioxide gas COC Contaminants of Concern DO Dissolved Oxygen DoD U.S. Department of Defense EPA U.S. Environmental Protection Agency FDEP Florida Department of Environmental Protection FY Fiscal Year GHG Greenhouse Gas GSR Green and Sustainable Remediation GW Groundwater GWM Groundwater Monitoring GW P&T Groundwater Pump & Treat System HSWA Hazardous & Solid Waste Amendments hp Horsepower ICE Internal Combustion Engine
14 IPCC Intergovernmental Panel on Climate Change United Nations ITRC Interstate Technology and Regulatory Council kW kilowatt kWh kilowatt hour LCA Life Cycle Analysis LFG Landfill Gas LTM L ong T erm M onitoring MNA Monitored Natural Attenuation MSWLF Municipal Solid Waste Landfill MT Metric Ton MMBTu Million British Thermal Units MW Megawatt N2O N itrous O xide NA Natural Attenuation NAS Naval Air Station NAVFAC Naval Facilities Engineer Command NIRIS Navy Installation Restoration Information System NOx Nitrogen Oxide gas NPL National Priorities List NREL National Renewable Energy Laboratory N WF Northwest Florida OU1 Operable Unit 1 (i.e. OU1 unlined landfill at installation) O&M Operation and Maintenance P&T Pump & Treat PM Particulate Matter
15 PM10 Particulate Matter 10 microns in diameter or less PVC P olyvinyl C hloride RAC Remedial Action Construction RAO Remedial Action Operation RCRA Resource Conservation and Recovery Act of 1976, as amended RI Remedial Investigation ROD Record of Decision (CERCLA) RSE Remedial System Evaluation SARA Superfund Amendments and Reauthorization Act SOx Sulfur Oxide gas USACE U.S. Army Corps of Engin eer VC Vinyl Chloride VOCs V olatile Organic C ompounds
16 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Engineering LIFE CYCLE ANALYSIS OF FIVE GROUNDWATER REMEDIATION TECHNOLOG IES FOR IRON BACTERIA SMOTHERING IN WETLANDS ADJACENT TO AN OPEN DUMP LANDFILL IN NORTHWEST FLORIDA By Sandra K. Gaynor May 2013 Chair: Tim othy Townsend Major: Environmental Engineering Sciences A life cycle analysis was performed to evaluate secondary environmental impacts using a software tool known as SiteWise on 4 groundwater (GW) remediation technologies Technologies were selected that were viable for remediation of an iron/sulfur enriched GW plume emanating from an open dump landfill discharging into a wetland adjacent to the landfill. The analysis consist ed of 8 modeled scenarios quantifying impacts of remediation for the remedial investigation (RI), remedial construction (RC), remedial ac tion construction and operations (RAC and RAO), and long term monitoring (LTM ) phases of activity at the site. The SiteWise analysis revealed that conventional aggressive and energy intensive technologies have the greatest environmental footprint, confirmi ng findings by other researchers. Active remediation systems, verses passive, had higher GHG emissions and energy usage for secondary environmental impact metrics. An active pump and treat system (GW P&T) in a sand gravel aquifer with 17 recovery wells, at a depth of 40 ft, will emit 46,693 MTs emissions vs 1,247 MTs for monitored natural attenuation (MNA). A passive 1300LF limestone trench system (LITS), 3 ft wide and 40 ft depth emits of 6,178 MTs of GHG emissions, while a similar active LITS will emit
17 28,523 MTs. An air sparging system (AS) emits 1,285 MT of GHGs. The energy footprint in MMTBU is: a) 15,600 for MNA, b) 21,300 for GW P&T, c) 54,900 for active LITS, d) 51,200 for passive LITS, d) 15,900 for AS. The priority air pollutant in MTs is: a) 6. 01 for MNA, b) 15.28 for GW P&T, c) 7.99 for active LITS, d) 6.75 for passive LITS, d) 6.15 AS. The SiteWise model proved to be very effective as a tool to aid decision makers in remediation technology feasibility phase of assessing effects to human health and environmental for secondary impacts of the use of a remediation technology at a site.
18 CHAPTER 1 INTRODUCTION 1. 1 P roblem Landfills are a permanent feature of our landscapes in this generation, prior generations, and the next generation. Many of the prior generations landfills open dumps prior to 1980s, are unlined Municipal Solid Waste Landfills (MSWLF) disposal areas with contaminated groundwater plumes discharging to surface water bodies. In the state of Florida, MSWLFs old or new, have become preservationist s of large sections of natural wetlands, the most productive ecosystems in our nation. In N orthwest Florida (NWF), these landfills are surrounded by large plots of undisturbed wetlands classified as Type 1 and 2 headwater stream s. These areas are ecologically sensitive areas as nursery of biota. Questions have arisen regarding the environmental impacts from the discharge of leachate from unlined MSWLF s on headwaters streams in NWF B iologists with Florida Department of Environm ental Protection (FDEP) performed biological assessments in streams adjacent to 16 unlined landfills in NWF. These biological assessments revealed total degradation of water quality and habitat for aquatic benthic macroinvertebrate in these streams using biological monitoring techniques qualitatively and quantitatively, as ecological health indicators for water quality The se assessment further proved that there was a MSWLF leachate plume, enriched with nutrients discharging into waters of the state, caus ing massive growth of iron/sulfur bacterial i n the headwaters streams. F DEP b iologist s a massive reduction in benthic macroinvertebrate organism taxa, sensitive species : Ephemeroptera, Plecoptera, and Trichoptera (EPT), with an
19 increase in filamentous iron/sulfur oxidizing bacteria forming large mats, in the streams where leachate wa s being released from the unlined landfills. In many of these streams the benthic communities are in competition with the microbial mats f or the limited resources of dissolved oxygen in the stream. In these cases bacteria became the dominant coverage, smothering habitat for EPT, in the stream for large areas down gradient of the unlined MSWLF This competition limit s the development of any benthic communities. The benthic communities changed from communities of 50 + taxa aerobic organisms to monocultures of anaerobic organisms (tubificidae) [Ray 2003] 1.2 Goals and Objective The goal of this study is to perform a Life Cycle Analysis (LCA), otherwise known as a Green Sustainable Remediation (GSR), of various groundwater remediation technologies for iron/ sulfur enriched groundwater (GW) plumes emanating from unlined landfills. The objective of the LCA/GSR analysis is to assess a matrix of environmental impacts associated wit h all the stages of the life of remediation activities from the remedial investigation (RI), remedial action construction (RAC), remedial action operation (RAO), and long term monitoring (LTM). The analysis will incorporate five remediation technologies th at are implementable and effective for use in sand/gravel aquifers at unlined MSWLF sites in NWF, with iron/sulfur bacteria impacts in surface waters adjacent to the site. The software that will be utilized is SiteWise which contains sustainability matrice s for analysis of energy consumption, Greenhouse Gases (GHG) emissions, air pollutants, water resource consumption, worker accident risk, footprint reduction through energy conversation and renewable energy sourcing.
20 1.3 General An unlined MSWLF known as O perable Unit 1 (OU1) at Naval Air Station (NAS) Pensacola, FL, has a groundwater plume that discharges into a surface water body wetland adjacent to the landfill. The wetland adjacent to OU1 known as Wetland 3, have iron/sulfur filamentous bacteria mats similar to the 16 sites that were evaluated by FDEP biologists. An LCA/GSR evaluation of GW remediation technologies will be modeled for seven remediation technologies that would be considered to be implementable and effective in reducing contaminants reac hing impacting Wetlands 3 give at this site for removal of impacts of contamination in the wetlands adjacent to assist decision makers in quantifying, not only biological impacts in streams, but other environmental impacts during the remedial investigati on, remediation implementation, and long term monitoring of wetlands impacted by landfill leachate plumes in GW that impact the wetlands and streams. Th is study will e valuate the environmental footprint with a LCA/ GSR model known as SiteWise. Site Wise was developed by US Army/US Air Force/U S Navy and Battelle for US Department of Defense ( DoD ) installation remediation managers (IR) as an environmental decision making process tool Th is LCA/ GSR evaluation will aid decision makers in minimizing the environme ntal footprint for remediation of the GW plume and maximiz ing the overall benefit of cleanup actions which include direct and indirect consequences of performing a particular remediation action at a site. The remediation technologies evaluated for this sit e are technologies that would be consider ed viable for remediation of iron and sulfur bacteria in the wetlands. The technologies evaluated are noted in Table 1 1
21 1. 4 Regulatory Drivers 1. 4 .1 Federal Regulatory Background The D o D estimates that investigations and/or cleanups are planned or are underway at nearly 8,000 areas. These areas are located on hundreds of active and inactive installations and formerly used defense sites. DoD is mandated to investigate and remediate environmental contamination at installations under federal regulations Resource Conservation and Recovery Act (RCRA ) of 1976, as amended by the Hazardous and Solid Waste Amendments (HSWA) of 1984; Comprehensive Environmental Response, Compensation, and Liability Act of 1980 (CERCLA), as amended by the Superfund Amendments and Reauthorization Act (SARA) of 1986. Many states have parallel regulations which mirror these federal laws with regulatory oversight and responsibility for cleanup and reme diation of these DoD sites [EPA 2008] 1. 4 .2 Federal Directives DoD sites are subject to additional environmental energy reduction goals for compliance with Executive Orders (EO) 13423 and 13514 to reduce their environmental footprint by reducing energy and water consumpt ion. A summary of the orders is noted below: EO 13423 requires Federal agencies to reduce energy intensity by 3% each year, leading to 30% by the end of fiscal year (FY) 2015 compared to an FY 2003 baseline. This goal was given the weight of law when ratified by EISA 2007. EO 13423 mandates that Federal agencies reduce water intensity (gallons per square foot) by 2% each year through FY 2015 for a total of 16% based on water consumption in FY 2007 EO 13514 Federal Leadership in Environmental, Energy, and Economic Performance, signed on October 5, 2009, establishes an integrated strategy towards sustainability in the Federal Government and makes
22 reduction of greenhouse gas (GHG) emissions a priority for Federal agencies. 1. 4 .3 Water Quality Standards The FDEP regulates groundwater, drinking water, and surface water quality standards. DoD sites are required to remediate contamination of groundwater and surface water to these standards under CERCLA based on the applicable or relevant and appropriate req uirements (ARARS) under the CERCLA Record of Decisions (ROD) for the site State drinking water standards for groundwater (GW) that have been established for the site are noted in Table 1 2 Table 1 3 list surface water criteria that have been established for the site under the ROD The biological criteria that is driving site cleanup in wetlands is: FAC 62 302.530 Table: Surface Water Quality Criteria (10) Biological Integrity The Index for benthic macroinvertebrates shall not be reduced to less than 75% of established background levels as measured using organisms retained by a U. S. Standard No. 30 sieve and collected and composited from a minimum of three Hester Dendy type artificial substrate samplers of 0.10 to 0.15 m2 area each, incubat ed for a period of four weeks. The metal ore mining industry has similar contaminants of concerns (COC) resulting from stormwater runoff in contact with abandoned mines and mining waste piles. C ontamination from mining areas, known as acid mine drainage ( AMD), typically covers very large sections of water sheds and headwaters streams, The applicable best available technology (BA T) water quality standards for AMD utilizing remediation techniques of constructed wetlands, are listed in Table 14
23 Table 1 1 SiteWise m odeled scenario remediation t echnologies for OU1 l andfill NAS Pensacola Technology Activity Comments Natural Attenuation (NA) Passive Simulated LCA Scenario Groundwater P ump & T reat Active pumping Simulated LCA Scenario Limestone Interceptor Trench System = 675 ft, 18 ft deep (LITS 675ft 18D Active) Active pumping Constructed in 1999 Limestone Interceptor Trench System = 675 ft, 40 feet deep (LITS 675ft 40D Passive) Passive Simulated LCA Scenario Limestone Interceptor Trench System = 1300 Ft, 40 Feet deep (LITS 1300ft 40D Active) Active Pumping Simulated LCA Scenario Similar to LITS constructed in 1999 but 40 ft deep Limestone Interceptor Trench System = 1300 Ft, 40 Feet deep (LITS 1300ft 40D Passive) Passive Simulated LCA Scenario Air Sparging Active Air Injection Simulated LCA Scenario
24 Table 12 Florida d rinking water s tandards for OU1 l andfill at NAS Pensacola, FL site specific standards ( FAC 62 550 ) COC's Florida DW Stds (ug/l) Other Reference Document Organics Benzene 1 ROD Performance Criteria Chlorobenzene 100 ROD Performance Criteria Naphthelene 14 Proposed May 2012 1,1,2,2 PCA 0.02 Proposed May 2012 Total Xylenes 20 Proposed May 2012 Vinyl Chloride 1 ROD Performance Criteria Inorganics Aluminum 200 Proposed May 2012 Cadimum 5 Proposed May 2012 Chromium 100 Proposed May 2012 Iron 300 ROD 2009 Manganese 50 Proposed May 2012 Nickel 100 ROD Performance Criteria Table 13 Florida s urface water s tandards for OU1 l andfill at NAS Pensacola, FL s ite specific standards ( FAC 62 302 ) COC's Florida SW Stds (ug/l) Class III Fresh Water Organics Benzene < 71.28 annual avg. Chlorobenzene NA Naphthelene NA 1,1,2,2 PCA < 10.8 annual avg Total Xylenes NA Vinyl Chloride NA Inorganics Aluminum 1,500 Cadimum Cd
25 Table 14 Best a vailable t echnology (BAT) d ischarge l imits for a cid m ine w astewater ( Mohana 2006; 40 CFR P art 434) Pollutant Parameters Discharge Limitations (30 day average) (mg/l) Daily maximum (mg/l) Iron 3.5 7.0 Manganese 2.0 4.0 pH 6.0 9.0 Alkalinity Alkalinity > acidity at all times
26 CHAPTER 2 SITE DESCRIPTION 2.1 General Solid w aste generated at NAS Pensacola was disposed in Operable Unit 1 (OU1), an 80 acre unlined MSWLF, from the mid 1950s until 1976. The landfill is approximately 20 feet above mean sea level (msl) and is heavily wooded with trees and natural shrub vegetation. OU1 is bordered to the north by Bayou Grande, and to the east by the Nav ys A.C. Reed Golf Course, to the west by Navy property covered with vegetation, and to the south by the Barrancas Cemetery. Figure 2 1 to 2 3 are general layouts of the area surrounding OU1 and Wetlands 3 and 4 [Navy, 1996] S urface water runoff from OU 1 drains to the Bayou Grande t hrough two wetlands, Wetland 3 a nd Wetland 4D Wetland 3 is a freshwater wetland that receives shallow groundwater and surface water runoff from OU1 seeping from beneath the unit at the north end of the wetland ( Figure 2 4 through 2 6 ) Sediment in W etland 3 is highly organic, with total organic carbon (TOC) detected up to 24 percent. The lower section of this W etland 3 flows into a drainage culvert that discharges into Wetland 4D. Wetland 4D is an estuarine open water body fed by Wetland 3 from the west and by Wetland 4C from the south. Wetland 4D discharges north into Bayou Grande through a culvert beneath an service road. Wetland 4D i s surrounded by the golf course [Navy, 2012] 2.2 Site Assessment Background Based on GWM analytical results from 1994 to 2012, the greatest impact to site shallow and intermediate groundwater quality with respect to inorganics appears to be limited to the center of the site, along the landfills eastern, western, and northwestern boundaries. Except for aluminum, iron, and manganese (indicated by background data
27 to naturally occur at elevated concentrations), inorganic concentrations exceeding EPAs CERCLA Record of Decision (ROD) a pplicable or relevant and appropriate r equirement ( ARARS ) are generally limited to areas within and around the landfills perimeter. Organic constituents have consistently been detected at relatively low concentrations in surficial groundwater ( Navy RI, 1996 ; Navy GWM 2 0 10, 2012) Analysis of waste left in place was performed at 13 locations across the unlined landf ill OU1 Landfill c apping /cover material was encountered at depths of 1 to 3 foot below land surface; M aterial consisted of sandy soil cover material. W astes below a depth of 3 feet consisted of heter ogeneous deposits of burned and unburned domestic refuse; C&D rubble; industrial refuse including plastic, glass, metallic refuse, and crushed drums; clayey silty sludge; and tar/ sludge ( Navy RI, 1996) Table 2 1 to 2 6 contain a summ ary of historical GMW data for i ron from 1993 to 2012 that is publically available. Table 1 10 contains analysis for soils and sediments (Navy NIRIS, 2012 ) It is noted that soils underlying waste and surrounding OU1 have not been assessed for iron content.
28 Figure 2 1 NAS Pensacola, FL OU1 l andfill g eneral site m ap Figure courtesy of US Navy 2008.
29 Figure 22 NAS Pensacola, FL OU1 l andfill h istorical g roundwater wells and sample l ocations Figure courtesy of US Navy 2008.
30 Figure 23 NAS Pensacola, FL OU1 landfill c urrent g roundwater monitoring w ells Figure courtesy of US Navy 2008.
31 Figure 24 Sampling l ocation SW 01 g roundwater seep and b acteria g rowth w etlands 3 OU1 l andfill NAS Pensacola, FL Photo courtesy of Ens afe 2012 Figure 25 Sampling l ocation SW 01 groundwater s eep and b acteria g rowth w etlands 3 OU1 landfill NAS Pensacola, FL Photo c ourtesy of Ensafe, 2012.
32 A B Figure 26 Surface w ater sampling l ocation SW 02 b acteria g rowth OU1 landfill NAS Pensacola w etlands 3, FL A) close up view at SW03; B) view at SW03 pipe crossing roadway. Photo courtesy of Kathy Gaynor 2012.
33 Table 2 1 Site specific surface w ater q uality results for i ron wetland 3, wetland 4D and OU1 landfill NAS Pensacola, FL [CAS 743989 6 ] (Navy NIRIS 2013) Location ID Sample Date (YYYYMMDD Duplicate Matrix Result (ug/l) Qualifier Exceeds ARAR 01SW01 20110831 SW 265 01SW01 20110831 01SW01 0811 SW 291 01SW01 20100513 SW 1,690 J 01SW01 20100513 01SW01 0510 SW 4,850 J X 01SW02 20120117 01SW02 0112 SW 5,840 X 01SW02 20120117 SW 6,500 X 01SW02 20100513 SW 6,890 J X 01SW02 20110831 SW 12,900 X 01SW03 20100513 SW 345 J 01SW03 20120117 SW 347 01W001 20060208 SW 10,800 X 01W001 20060208 SW 11,000 X 01W002 20060208 SW 7,390 X 01W002 20060208 SW 19,300 X 01W01 20011113 SW 6,710 X 01W01 20080330 SW 6,810 J X 01W01 20020501 SW 7,930 X 01W01 20010523 SW 9,190 X 01W01 20000320 SW 28,000 X 01W01 20000818 SW 1,790,000 X 01W02 20080330 SW 3,260 J 01W02 20000320 SW 19,500 X 01W02 20011113 SW 36,800 X 01W02 20010523 SW 57,300 X 01W02 20000818 SW 703,000 X 01W02 20020430 SW 816,000 X 01W03 20080330 SW 4,390 J *Note: Exceeds Applicable or Relevant and Appropriate Requirement (ARAR) for site of 720 ug/l iron for fresh water S W = surface water J = Indicates an e stimated value
34 Table 2 2 Site s pecific g roundwater a nalysis i ron in shallow b ackground w ells OU1 l andfill NAS Pensacola, FL [ CAS 743989 6 ] ( Navy NIRIS, 2013) Location ID Sample Date (YYYYMMDD Duplicate Matrix Result (ug/l) Qualifier 01UPGW1 20050602 GW 8,480 01UPGW1 20031117 GW 5,810 01UPGW1 20100511 GW 2,140 01UPGW1 20051215 GW 1,400 J 01UPGW1 20021009 GW 144 01UPGW1 20011113 GW 91 J 01UPGW1 20030611 GW 81 01UPGW1 20020502 GW 59 U GW = groundwater J = Indicates an estimated value U = Indicates compound was analyzed for, but not detected Table 2 3 Site specific groundwater a nalysis i ron in shallow w ells OU1 landfill NAS Pensacola, FL [CAS 743989 6 ] (Navy NIRIS 2013) Location ID Sample Date (YYYYMMDD Matrix Result (ug/l) 01GS64 19930722 GW 103,000 01GS64 20031119 GW 98,500 01GS64 20051216 GW 86,400 01GS64 20050602 GW 77,400 01GS64 20101109 GW 68,700 01GS64 20110829 GW 66,500 01GS64 20100510 GW 64,400 01GS64 20030611 GW 62,800 01GS64 20120116 GW 61,500 01GS64 20011113 GW 46,800 01GS64 20011113 GW 46,750 01GS64 20011113 GW 46,700 01GS64 20010524 GW 39,100 01GS64 20020501 GW 34,100 01GS64 20021009 GW 24,700 01GS64 19940630 GW 13,900 GW = groundwater
3 5 Table 2 4 Site specific groundwater analysis i ron in p iezometer w ell at limestone t rench OU1 l andfill NAS Pensacola, FL [CAS 7439896 ] (Navy NIRIS 2013) Location ID Well Type Sample Date (YYYYMMDD Matrix Result (ug/l) 01PZ01 PZ 20100511 GW 46,300 01PZ01 PZ 20100511 GW 46,200 01PZ01 PZ 20100511 GW 46,100 01PZ01 PZ 20101107 GW 36,900 01PZ01 PZ 20101107 GW 36,200 01PZ01 PZ 20101107 GW 35,500 01PZ01 PZ 20110829 GW 25,000 01PZ01 PZ 20110829 GW 24,750 01PZ01 PZ 20110829 GW 24,500 01PZ01 PZ 20120117 GW 21,400 01PZ01 PZ 20120117 GW 21,000 01PZ01 PZ 20120117 GW 20,600 01PZ06 PZ 20031118 GW 36,200 01PZ06 PZ 20050601 GW 35,000 01PZ06 PZ 20100511 GW 28,600 01PZ06 PZ 20101107 GW 19,600 01PZ06 PZ 20030609 GW 19,200 01PZ06 PZ 20020501 GW 18,300 01PZ06 PZ 20051216 GW 16,800 01PZ06 PZ 20120117 GW 16,000 01PZ06 PZ 20011113 GW 14,600 01PZ06 PZ 20110829 GW 7,530 01PZ06 PZ 20021009 GW 5,680 01PZ06 PZ 20031118 GW 2,540 01PZ06 PZ 20101107 GW 1,350 01PZ06 PZ 20110829 GW 1,000 01PZ06 PZ 20051216 GW 952 01PZ06 PZ 20120117 GW 901 01PZ06 PZ 20100511 GW 820 01PZ07 PZ 20051216 GW 78,100 01PZ07 PZ 20031118 GW 61,600 01PZ07 PZ 20051216 GW 46,800 01PZ07 PZ 20050601 GW 45,600 01PZ07 PZ 20110829 GW 27,600 01PZ07 PZ 20110829 GW 25,400 01PZ07 PZ 20100510 GW 17,900 01PZ07 PZ 20030609 GW 17,500 01PZ07 PZ 20101107 GW 16,800 01PZ07 PZ 20120116 GW 14,600 Note: GW = groundwater, PZ = piezometer well; 01PZ1, 01PZ 0 6, 01PZ 0 7, 01PZ 0 8 are down gradient of the limestone interceptor trench system ( Figure 2 3 )
36 Table 2 4 C ontinued Location ID Well Type Sample Date Matrix Result (ug/l) Qualifier 01PZ07 PZ 20120116 GW 12,800 01PZ07 PZ 20100510 GW 11,200 01PZ07 PZ 20101107 GW 9,830 01PZ07 PZ 20020501 GW 8,120 01PZ07 PZ 20031118 GW 5,200 01PZ07 PZ 20031118 GW 3,660 01PZ07 PZ 20011113 GW 1,690 01PZ07 PZ 20100510 GW 1,090 01PZ07 PZ 20101107 GW 1,070 01PZ07 PZ 20110829 GW 1,040 01PZ07 PZ 20120116 GW 896 01PZ08 PZ 20031118 GW 4,410 01PZ08 PZ 20050601 GW 3,460 01PZ08 PZ 20100510 GW 2,940 01PZ08 PZ 20030609 GW 2,620 01PZ08 PZ 20051216 GW 1,870 01PZ08 PZ 20110829 GW 1,560 01PZ08 PZ 20101107 GW 1,170 01PZ08 PZ 20120116 GW 846 01PZ10 PZ 20011114 GW 5,630 01PZ10 PZ 20021009 GW 3,260 01PZ10 PZ 20031118 GW 2,810 01PZ10 PZ 20051220 GW 2,670 01PZ10 PZ 20050601 GW 1,920 01PZ10 PZ 20030609 GW 1,900 01PZ10 PZ 20020501 GW 1,800 01PZ10 PZ 20100510 GW 1,740 01PZ10 PZ 20101107 GW 945 01PZ10 PZ 20120116 GW 846 01PZ10 PZ 20110829 GW 671 01PZ101 PZ 20080330 GW 1,220 01PZ102 PZ 20080330 GW 14,700 01PZ102 PZ 20080330 GW 14,600 01PZ102 PZ 20080330 GW 14,500 01PZ103 PZ 20080330 GW 312 01PZ104 PZ 20080330 GW 32,200 01PZ105 PZ 20080330 GW 3,780 01PZ106 PZ 20080330 GW 505 01PZ11 PZ 20011114 GW 16,600 Note: GW = groundwater, PZ = piezometer well ; 01PZ1, 01PZ 0 6, 01PZ 0 7, 01PZ 0 8 are down gradient of the limestone interceptor trench system (Figure23).
37 Table 2 5 Site specific groundwater a nalysis i ron in intermediate w el l s OU1 landfill NAS Pensacola, FL [CAS 743989 6 ] (Navy NIRIS 2013) Location ID Sample Date (YYYYMMDD) Duplicate Result (ug/l) Qualifier 01GI36 19930715 11,100 J 01GI36 19940629 761 01GI36 20000315 829 01GI36 20000816 2,210 01GI36 20010521 1,540 01GI36 20011113 1,550 01GI36 20020430 1,150 01GI36 20021008 777 01GI36 20030611 452 01GI36 20030611 417 01GI36 20030611 01GI3600 061103 382 01GI36 20031119 947 01GI36 20050602 1,450 01GI36 20051215 1,840 J 01GI36 20100511 234 01GI36 20101109 827 01GI36 20110830 232 01GI36 20120117 163 01GI41 19930721 78,100 01GI41 19940706 52,500 01GI41 20000317 42,800 01GI41 20000817 01GI41 AUG00 38,500 01GI41 20000817 37,950 01GI41 20000817 37,400 01GI41 20010523 34,900 01GI41 20011112 39,600 01GI41 20020430 21,300 01GI41 20021009 21,300 01GI41 20030610 29,600 01GI41 20031119 10,200 01GI41 20050602 39,000 01GI41 20051215 39,300 J 01GI41 20100511 01GW01GI41 0510 39,000 01GI41 20100511 37,400 01GI41 20100511 35,800 01GI41 20101110 44,900 01GI41 20101110 43,600 01GI41 20101110 01GW01GI41 1110 42,300 01GI41 20110830 36,400 01GI41 20110830 36,250 01GI41 20110830 01GW01GI41 0811 36,100 01GI41 20120117 10,300 01GI41 20120117 9,665 01GI41 20120117 01GW01GI41 0112 9,030 01GI43 19930721 31,600 01GI43 20000817 15,300 01GI43 20010523 15,800 01GI43 20011113 17,700 01GI43 20020501 01GI43 20020501 16,800 01GI43 20020501 15,950 01GI43 20020501 15,100 01GI43 20021008 14,700 01GI43 20030610 16,400 01GI43 20031119 14,300 01GI43 20050601 14,200 01GI43 20051216 12,800 01GI43 20100512 8,130 01GI43 20101109 8,110 01GI43 20110830 6,710 01GI43 20120117 7,150 J = Indicates an estimated value
38 Table 2 5 Continued Location ID Sample Date (YYYYMMDD Duplicate Result (ug/l) Qualifier 01GI44 19930726 179,000 01GI44 20000317 58,200 01GI44 20000817 104,000 01GI44 20010522 53,700 01GI44 20011112 55,900 01GI44 20020430 41,100 01GI44 20021008 34,200 01GI44 20030610 36,600 01GI44 20031119 57,300 01GI44 20050601 33,600 01GI44 20051216 28,700 01GI44 20100511 25,300 01GI44 20120119 15,900 01GI46 19930722 49,300 01GI46 19940628 42,700 01GI46 20000317 30,100 01GI46 20000817 37,500 01GI46 20010522 43,000 01GI46 20011112 49,200 01GI46 20020501 41,000 01GI46 20021008 44,600 01GI46 20030610 43,100 01GI46 20031119 40,200 01GI46 20050602 PEN OU1 GI46 07 73,300 01GI46 20050602 72,900 01GI46 20050602 72,500 01GI46 20051215 36,400 J 01GI46 20051215 36,250 01GI46 20051215 PEN OU1 GI46 08 36,100 J 01GI46 20100512 17,500 01GI46 20101110 19,500 01GI46 20110830 21,300 01GI46 20120117 12,800 01GI65 19930722 55,400 01GI65 19940630 19,900 01GI65 20000315 34,500 01GI65 20000817 30,700 01GI65 20010523 33,700 J 01GI65 20011113 30,500 01GI65 20020501 29,100 01GI65 20021009 29,400 01GI65 20030611 30,300 01GI65 20031119 22,500 01GI65 20050602 22,800 01GI65 20050602 22,500 01GI65 20050602 PEN OU1 GI65 07 22,200 01GI65 20051216 PEN OU1 GI65 08 24,200 01GI65 20051216 23,350 01GI65 20051216 22,500 01GI65 20100510 18,400 01GI65 20101108 18,600 01GI65 20110829 27,900 01GI65 20120116 29,100 J = Indicates an estimated value Range = 163 to 179,900 ug/l Average = 32,719 ug/l Geomean = 27,499 ug/l Standard Deviation = 20,874 ug/l
39 Table 2 6 Site specific a nalysis of i ron in soils & sediments at OU1 landfill & w etland 3 [CAS 7439 89 6] (Navy NIRIS 2013) Location ID Sample Date (YYYYMMDD Duplicate Matrix Result (ug/l) Qualifier Sediments 001M0001 19940629 SD 1,490 J 001M000301 19940628 SD 13,200 J 001M000302 19940628 SD 15,800 J 001M000303 19940628 SD 1,940 J 001M0016 19940629 SD 4,680 J 001M0018 19940629 SD 15,000 J Soils 001S0001 19930618 SO 1,820 001S0039 19930618 SO 365 001S0042 19930527 SO 958 001S0062 19930618 SO 856 001S0064 19930617 SO 556 001S0067 19930602 SO 9,180 001S0067 19930602 SO 1,450 001S0067 19930602 SO 4,790 001S0067 19930602 SO 1,140 001S0067 19930602 SO 866 001S0067 19930602 SO 482 001S0067 19930602 SO 483 001S0067 19930602 SO 449 001S0067 19930602 SO 237 001S0067 19930602 SO 225 001S0069 19930601 SO 911 001S0069 19930601 SO 1,010 001S0069 19930601 SO 963 001S0069 19930601 SO 697 001S0069 19930601 SO 595 001S0069 19930601 SO 205 001S0069 19930601 SO 497 001S0071 19930826 SO 850 001S0071 19930826 SO 989 001S0071 19930826 SO 963 001S0072 19930824 SO 1,690 001S0072 19930824 SO 2,140 001S0072 19930824 SO 111 001S0073 19930820 SO 2,300 001S0073 19930820 SO 2,810 001S0073 19930820 SO 310 SO = Soils, SD = Sediments, J = Estimated Value
40 Table 2 6 C ontinued Location ID Sample Date (YYYYMMDD Duplicate Matrix Result (ug/l) Qualifier 001S0074 19930830 SO 2,970 001S0074 19930830 SO 760 001S0075 19930823 SO 1,300 001S0075 19930823 SO 1,170 001S0076 19930825 SO 1,020 001S0076 19930825 SO 481 001S0077 19930825 SO 685 001S0077 19930825 SO 2,520 001S0077 19930825 SO 4,700 001S0078 19930826 SO 1,230 001S0078 19930826 SO 1,440 001S0078 19930826 SO 91 001S0079 19930718 SO 653 001S0080 19930819 SO 3,580 001S0081 19930820 SO 1,030 001S0081 19930820 SO 4,000 001S0081 19930820 SO 102 001S0082 19930819 SO 42,300 001S0082 19930819 SO 15,900 001S0082 19930819 SO 1,830 001SI060 19930617 SO 944 01GI28 19930617 SO 9,760 01GI30 19930617 SO 333 01GI32 19930721 SO 1,670 01GI35 19930617 SO 1,260 01GI36 19930617 SO 1,380 01GI43 19930618 SO 1,200 01GI44 19930607 SO 46 01GI46 19930617 SO 1,320 01GI48 19930617 SO 1,230 01GI59 19930618 SO 1,010 01GI59 19930625 SO 16,600 SO = Soils, SD = Sediments For Soils: Range = 46 to 42,000 ug/l Average = 2,593 ug/l Geomean = 1,063 ug/l Standard Deviation = 5,937 ug/l
41 CHAPTER 3 LITERATURE REVIEW 3.1 Life Cycle Analysis Literature A literature review of published journal articles was performed for LCA studys pertaining to iron/sulfur bacteria in wetland ecosystems impacted by landfill leachate. There were no LCA studies found pertaining to iron/sulfur bacteria plumes and unlined landfills. There was a large volume of LCAs performed for manufacturing of various products and disposal of waste products, furthermore there was a limited volume of studies pertaining to groundwater remediation technologies associated with remediation of groundwater plumes emanating from MSW landfills and/or contaminated site remediation 3.1.1 General Discussion S olid waste recycling, reuse, and disposal were the focus of worldwide attention in the late 1980s. National and international laws were developing for solid waste landfill liners, waste disposal criteria, recycling, and beneficial use of m unicipal and industrial waste. During this time period, life cycle assessment (LCA) framework reemerged as a tool for anal ysis of environmental impacts from the manufacturing of products and disposal of waste In subsequent years, well defined methodologies and modeling tools for LCA were refined to better assess and quantify the cradleto grave environmental footprint o f industrial systems (EPA, 2006; Bonano et al., 2000). Today there are complex computer modeling systems, specific to various industrial processes available to assess potential environmental risk associated with the development of a product or technology utilized to manufacture that product. The objectives and me thodologies are spelled out in a variety of regulatory frameworks by US
42 E nvironmental Protection Agency (E PA) International Standards of Operation (ISO) 1404043, European Union (EU) directives and other national laws, regulations, standards or guidance documents These frameworks provided uniform methods for conducting assessments and marked a consensus of knowledge for assisting industry to go beyond compliance to pollution prevention strategies thereby improving environmental performance (EPA 2006; Morais and DelerueMatos 201 0 ; US Office of White House EO 13514 & EO 13423). The scope of this literature focuses on review of published journal articles or LCA studies in which a computer modeling approach is utilized or reviewed for groundwater remediation and disposal of solid waste. The site conditions modeled in the studies varied widely in site contaminants, geology, aquifer recharge, groundwater flow rate, extent of contamination, regulatory point of compliance for regulatory, and remediation techn ology. The wide variety of sites conditions resulted in many different metrics causing difficulty in presenting comparison of LCAs studies one to another. Additionally, the differences in environmental impact metrics measured with different computer models utilized added another layer of complexity for comparison. Only the Navy 2009 studies utilized the SiteWise LCA/GSR tool. In all the studies that were reviewed, q uantitative and qualitative metr ics were used in the analysis. LCA computer simulation modeling was proven by several researchers as an effective tool for assessing the environmental impacts of waste management techniques groundwater remediation technologies ( Lemming et al. 2010; Boldin et al. 201 1; Zhao et al., 2012).
43 In LCA literature r elated to contaminated site remediation, the impacts related to a sites physical state were labeled as primary impacts. I mpacts associated with the remediation service are labeled secondary impacts, and the environmental impacts associated with the ef fects of the postrehabilitation fate of the site were label ed tertiary impacts (Lesage et al. 2007; Morais and DelerueMatos et al. 2010) The application of LCA modeling facilitated rapid assessment of primary and secondary environmental impacts on a wide variety of remediation technologies and integrated systems for ex situ and insitu treatment of soils and groundwater. The remediation technologies modeled in this literature review were : a) liming of contaminated soils (Blanc et al. 2004; Lemming e t al. 2012), b) monitored natural attenuation (Lemming et al., 2012 ; Navy, 2009), c) in situ chemical oxidation (ISCO) (Lemming et al. 2012; Navy, 2009), d) activated carbon treatment (GAC) (Lemming et al. 2012 ; Bayer and Finkel 2006), e) pump and treat (Lemming et al. 2012; Bayer and Finkel 2006 ; Cadotte et al. 2007 : Navy, 2009), f) soil grouting (Gallagher et al. 2012) f) in situ air sparging (IAS) (Navy 2009), g) soil vapor extraction (SVE) (Navy, 2009), h) enhanced bioremediation (Navy, 200 9). A number of LCA tools were reviewed and/or utilized in these studies : US EPA TRACI2, EASEWASTE, EDIP97 /2003, USES LCA and SiteWise These computer models analyzed processes and the environmental impact/burden of remediation technologies, compare d trea tment alternatives, and provided guidance for decision making (Lemming et al. 2012; Zhao et al., 2012). A brief summary of each model and the environmental impact metrics for each is noted below: EASEWASTE Developed by Technical University of Denmark. Typically utilized to perform an LCA on solid waste materials or processes ( Boldrin et al. 2011 ;
44 Zhao et al., 2012). Quantifies secondary environmental impact metrics for on global warming, acidification, nutrient, enrichment, stratospheric ozone depletion, photochemical ozone formation, ecotoxicity, human toxicity, stored toxicity, spoiled groundwater US EPA TRACI Developed by USEPA. Typically utilized in industrial manufacturing processes impact assessment and pollution prevention. Tool for the Reduct ion and Assessment of Chemical and Other Environmental Impacts (TRACI). Quantifies s econdary impacts for ozone depletion, global warming, acidification, human health risk for cancer/non cancer, eutrophication, smog formation, ecotoxicity, fossil fuel use, land use, and water use. EDIP97/2003 D eveloped by the Institute for Product Development (IPU) at the Technical University of Denmark Environmental Development of Industrial Products (EDIP). Quantifies secondary impacts for photochemical ozone formation, acidification, eutrophication, ecotoxicity, human toxicity, hazardous waste, bulk waste, radioactive waste, and noise SiteWise D eveloped by US Navy, US Army, & Batelle for use at DoD sites. Designed for use at contaminated sites and remediation technologies. Measures secondary impacts for greenhouse gases (GHGs) reported as carbon dioxide equivalents (CO2e) energy usage, nitrogen (NOx), sulfur oxides (SOx), particulate matter (PM10) human health injury and risk of fatality risk ) CO2e is measured as summation of car bon dioxide (CO2), methane (CH4), and nitrous oxide (N2O) global warming potential (GWP). The computer simulated LCA models provided a measure of consistency in quantify ing the primary and secondary impacts within the LCA framework They differed in the types of remediation technology and the secondary impacts (i.e. energy consumption, GHGs, acidification, ecotoxicity ). Some studies regarded the total contaminant mass as an emission to either soil or surface water; others used generic ste ady state multimedia models to forecast the long term redistribution of the contaminant in the environment. Several studies reviewed coupled a groundwater fate and transport model with the LCA to evaluat e the primary impact on wat er resources (Godin et al. 2004, Cadotte et al. 2007; Lemming et al. 2012). Lemming et al., i n 2010 performed a literature review of the body of published LCAs on remediation technologies for groundwater and soil over the time period of
45 1998 to 2010. This review revealed that earlier LCA studies utilized simplified impact assessment models, the more recent studies based their impact assessment on established methodologies covering the conventional set of environmental impact categories and resource consumption utilizing computer simulation models LCAs varied across the board due to the diversity of contaminants at a site, remedial technologies that could be effectively used at a site, regulatory drivers, economics driving a site, human health risk associated with contaminants, and LCA approaches utilized. A discussion of some of the major parameters is presented below. 3.1.2 LCA System Boundaries System boundaries for an LCA site remediation are set based on characteristics of the site, focusing on the greater environmental bur den Some researchers only review the installation and operation of remediation theologies verses the systems boundary being widened to include landfill operation time period, remedial site investigation, remediation action construction, operation and long term monitoring (LTM) ( Lemming et al. 2012, Navy, 2009) Others include energy consumption associated with the processes of mining and transport ation of the materials ( Blanc et al. 2004, Navia et al. 2006, Bayer and Finkel 2006, Lemming et al. 2010, Gallagher et al. 2013). 3.1.3 Remedial Action Periods For the studies that evaluated remedial pump and treat systems, the recovery well pumping rate selected by researchers proved to be a critical driver for energy consumption and potential for complete r emediation of aquifer (Bayer and Finkel 2006) High groundwater pumping rates were selected for total extraction of contaminants verses long term hydraulic containment of contaminated plume. It was proven in these
46 studies that the design parameters of tec hnology, or addition of enhancements to a selected remedy, were driven by the site specific conditions. Additionally, other drivers for design were the regulatory water quality criteria that must be met at a designated point of compliance (POC). 3.1.4 Groundwater Fate and Transport of Contaminants The effectiveness of remediation technology was evaluated in all studies presented. The studies differed in the approach taken to the respective remediation time period for remediation and regulatory POC. Several authors incorporated groundwater/soil fate and transport models into the evaluation, coupled with a LCA models (Cadotte et al. 2007, Godin et al. 2007, Lemming et al. 2012). For example, a set timeframe was set for 99% mass removal in the source area an d a downstream monitoring point at the system boundary. The results of the fate and transport models revealed that if remediation is not conducted at the site, it would take 700 + years before contamina nts in the source area leached. Remediation with in si tu chemical oxidation ( ISCO ) reduces this timeframe to 80 years. The regulatory point of compliance would was set 100 m downstream fr om the source (Lemming et al. 2012). The utilization of a groundwater transport model coupled with an LCA proved to be ver y effective in determining system boundaries, rate of mass removal at source, POC systems boundaries and LTM time period. Groundwater fate and t ransport modeling assisted in predicting contaminant concentrations at a set time period and whether the water quality regulatory criteria could be m et in a reasonable time frame. If it could not be met, then the system boundaries were changed or remediation technology revised to meet that criterion.
47 3.1.5 Long Term Monitoring (LTM) Period US EPA defines long term monitoring of groundwater at a site as verification that contaminants pose no risk to human health or the environment and that natural processes are reducing contaminant levels such that there is no significant risk to human healt h or the environment. I n the United States for MSWLFs or RCRA sites this time period is between 10 30 years. The time period is established based on site hydrology, soils type, contaminate type, toxicity of contaminant and degradation products, and aquifer recharge, allowing decision makers to observe fluctuations in the systems and make conservative judgments as to what will occur if the site was not being monitored. Cost estimating for providing financial assurance for a site is calculated using a minimum of 30 years LTM. In the literature reviewed, there was no consensus as to a specified LTM period. Some researchers set it as a regulatory compliance period or when a water quality criterion was achievable with cost effective technology (Lemming et al. 2012 ; Bayer and Finkel 2006). A fundamental problem is that the physical and chemical description of the subsurface is basically inexact. There is uncertainty in estimating or modeling present and future status of the remediation process and at numerous sites the initial exact mass distribution of contaminants in the subsurface is not known ( Bayer and Finkel 2006) 3.1.6 Capital Equipment and I nfrastructure E xclusions C apital equipment buildings and machinery that are needed to produce the remediation equipment, were not inclu ded in LCAs I t is assumed that capital equipment had a long lifespan that its contribution to the LCA would be insignificant
48 after being apportioned by its years of use or by the amount of work a piece of equipment produced (Morais and DelerueMatos 2012). 3.1.7 Greenhouse Gases In all studies reviewed, the GHG emissions increased based on the type of manufacturing associated with the remediation media material selected and the volume of material consumed (Blanc et al. 2004, ) In the LCAs performed for Almeda and Parris Island sites, ISCO proved to be highest source of GHGs emissions due to the use of a chemical oxidant media (Navy 2009). T he use of activated carbon in a remediation system generated high GHG emissions ( Bayer and Finkel 20 06, Lemming et al. 201 2 ) Transportation of media materials to a site, from the gate of the manufacturer (i.e. gate to gate), was a large contributor to additional GHGs. One author analyzed distribution of bottled water utilizing ocean freig ht transportation verse tractor trailer transport and the resulting GHG emissions. The analysis revealed that there was a decrease in GHG emissions as a result of the efficiency of ocean transport despite the drastic increase in distribution distance from the national distribution center (Dettore, 2009). 3.1.8 Energy Consumption In all studies reviewed, active remediation technologies had the highest energy consumption and passive technologies the lowest. The use of electrical resistive heating (ERH) technology at the Almeda site (Navy 2009) and pump and treat technology proved to be the largest consumers of energy. These technologies also resulted in having the highest GHGs footprint (Navy, 2009; Bayer and Finkel 2006; Lemming et al. 20 10; Lemming et al 2012).
49 3.2 Green Sustainability Remediation (GSR) In 2008 EPA publish ed the G reen Sustainable Primer (GSR) Th e primer outlined to the regulated public what needed to be reviewed in LCAs for remediation of contaminated sites The core elements of the GSR process of analysis are to: a chieve remedial action goals support use and reuse of remediated parcels i ncrease operational efficiencies r educe total pollutant and waste burdens on the environment m inimize degradation or enhance ecology of the site and other affected areas r educe air emissions and greenhouse gas production m inimize impacts to water quality and water cycles conserve natural resources a chieve greater long term financial return from investments, and increase sustainability In 2009, the Department of Defense (DoD) developed software known as SiteWise for evaluation of remediation technologies using the principles laid out in the EPA Primer. This software is currently being used by the US Navy and Army installation restoration managers throu ghout the United States in evaluating the environmental footprint of contaminated site remediation technologies. In the GSR decision making process, decisions makers consider the following factors: economic natural resources ecological systems natural hydrologic cycle human health risk and safety quality of life The goal s of the LCA/GSR are to use natural resources and energy efficiently, reduce negative foot print impacts on the environment, minimize and eliminate pollution sources for protection of the com munity in an economical fashion, maximizing the overall benefit of cleanup actions. B est management practices (BMPs) that are included
50 in the overall consideration of remediation activities are water conservation measures, storm water runoff controls, and recycling of treatment process water. In 2011, the Interstate Technology and Regulatory Council (ITRC) published the Green Sustainability Remediation (GSR) Framework. This document formalized the basic framework for the GSR process for state regulatory agencies. 3.3 Summary LCAs have been proven to be effective screening tool s to help identify sig nificant environmental issues. LCAs that were developed to predict primary environmental burdens were typically defined by site specific acute and chronic water ecotoxicity categories as being the dominant impact categories of the environmental pro file and consequently (Godin et al. 2007). LCAs that were developed to predict secondary impacts were driven by environmental burdens for energy consumption, air pollutants, and global warming potential (Blanc et al. 2004; Bayer and Finkel 2006; Lemming et al., 2010; Lemming et al., 2012 ; Navy, 2009).
51 CHAPTER 4 N ORTHWEST FLORIDA SITE CONDITIONS 4.1 Unlined Landfills Waste that is landfilled, and the resulting leachate generated, has a life cycle that advances in fiv e sequential and distinct phases as the waste or leachate matures to stabilization The characteristics of leachate and gas produced from the waste vary from one phase to another and reflect the microbial ly mediated processes taking place inside the landfi ll. The phases are: a) initial adjustment phase, b) transition phase, c) acid formation phase, d) methane fermentation phase, and e) maturation phase. When biological activity in the leachate shifts to relative dormancy the leachate has reached the maturation phase methane g as production drops and leachate strength stays at much lower concentrations ( Navy, 1996; Reinhart and Goosh, 1998 ) Organic constituents within leachate tend to decompose and stabilize with time, verse inorganic cons tituents like heavy metals, ammonia, chloride, and sulfide remain steady in solution long after waste has stabilized. Metals often precipitate within the landfill and are infrequently found in hig h concentrations in leachate, with the exception of iron ( Reinhart and Goosh, 1998) When an unlined landfill discharges leachate into a surface water body, selection of a remedial technology for remediation of the groundwater plume impacting the water body, the net environmental consequences of the remediation are not al ways positive. The cost to the environment and human health in the form of increased greenhouse gases emissions (GHG) particulate emissions (PM10), water resources, and energy consumption, may outweigh the gain obtained by remediation of the plume throug h active pump and treat (P&T) groundwater systems ( Suer and AnderssonSkold 2011)
52 The major contributing factors at the site that are influencing extensive growth of iron and sulfur bacteria in the Wetlands 3 adjacent to OU1 unlined landfill are: waste i n unlined landfill resulting groundwater plume as a carbon source for microbial growth anaerobic conditions in groundwater soil beneath landfill as well as those surrounding the footprint of the landfill aquifer geochemistry beneath the landfill with groundwater flow direction toward the east into the wetlands stormwater runoff interacting with soils surrounding the landfill wetlands organic soils acting as a sump for metals aerobic conditions in wetlands and anaerobic conditions in sediments 4 2 Biological Assessments in Streams FDEP staff biologist surveyed 28 headwater streams in NWF that were impact ed by landfill leachate Th e leachate plume was the catalyst for extensive iron and sulfur bacterial plumes in the surface waters down gradient of the landfill. Table 4 2 lists the results of biological assessment at the sites evaluated in NWF by FDEP staff biologist. The effect to the surface water bodies from the plumes were: failed tests for Biorecons and Stream Condition Index (S CI) biological assessment of habit and benthic community health in fresh water streams severe degradation of the aquatic macroinvertebrate communities i ron and sulfur bacteria became the dominant coverage, smothering benthic habitat for EPT biological indicators were evidence of c hemical contamination leaching from landfill waste into the SW and GW failure of acute toxicity test for fish and water fleas at using sites using with bioassays laboratory testing
53 i ron concentrations in streams w ere elevated above FDEP state s tandards in the landfill compliance monitoring wells pH of the western panhandles sand and gravel aquifer ranges from 4 to 6 standard units. This naturally low pH readily dissolves buried waste materials and results in greater mobilization metals in an acidic environmental r esults of bioassessments demonstrated that landfills without liners do not provide the necessary best management practices adjacent to streams to prevent impact to the biological communities in waters of the State 4 3 Fl orida Soils L eachate from wast e buried at OU1 is a carbon source for biological activity as it infiltrates into the underlying aquifer. Distinct sulfate and iron redox zones are formed within the GW plume down gradient of the landfill ( Wang et al., 2012 ) originating from sources of Fe(II) in the GW : a) Fe originating from waste left in place, 2) Fe originating from soils surrounding and aquifer beneath the waste, c) bacterial activity. S oils at OU1 analyzed for iron are noted i n Table 2 6 A study was performed by Florida State University (FSU) t o review the ir on content in soils at landfill s in N WF Table 4 1 summarizes the iron content in soils from 15 landfills site in NWF S tudies wer e performed at University of Florida (UF), assessing aquifers for F e(II) content in soils at landfills in Florida confirming that the aquifers are rich in Fe(II) resulting f rom both waste disposed of in the landfill and soils surrounding the landfill s ( Wang et al. 2012; Townsend et al., 2011) Additional studies were performed assessing iron content in special waste and select soil types in Florida noted in Figure 4 1 and Table 4 1 ( Chen et al. 1999; Valcarce and Townsend, 2011) 4 4 Iron Redox and Mobilization in GW The cycling of iron consists largely of oxidationreduction reactions that reduce Fe(III) to Fe(II) and oxidize Fe(II) to Fe(III)
54 Fe(II) and Fe(III) i ons have very different solubility properties. Fe(III) precipitates in alkaline environments as ferric hy droxide. Fe(III) may be reduced under anaerobic conditions. H owever in the presence of sufficient H2S it may precipitate ferrous sulfide. Flooding of soil, which creates anaerobic conditions, favors the accumulation of Fe(II) ( Atlas and Bartha, 1993 ; R ussell and Cohn, 2012) Under alkaline to neutral conditions, Fe(II) is inherently unstable in the presence of O2 and is oxidized spontaneously to Fe(III) In these conditions, ferrooxidans bacteria proliferates and gain energy they need to liv e and multiply by oxidizing dissolved Fe(II) (or less frequently available manganese) ( Atlas and Bartha, 1993; Dyer 2003) Equation 4 1 below shows the reaction as this bacteria proliferate: 2Fe2+ + 1/2O2 + 2H+ 2Fe3+ 6.5 kcal /mole) (4 1) G roundwater seeping through sand formations dissolves ferrous salts in anaerobic zones, prevent ing iron oxidation. When groundwater seeps to the surface, ferroxidans bacteria convert the Fe(II) ions to ferric Fe(III) precipitat ing as ferric h ydroxide. ( Atlas and Bartha, 1993). Nitrates inhibit Fe(III) reduction. Some iron reduction may occur non enzymatically when H2S react s chemically with Fe(III) ( Atlas and Bartha, 1993) If H2S is present in the aquifer bacteria convert it to sulfur I f conditions are acidic enough it is converted to a weak sulfuric acid ( Dyer 2003) Tables 4 4 and 4 5 denote various redox/microbial mechanisms at work in GW aquifers. I ron reduction produces 32 times more Fe(II) per mole of organic matter (CH20) oxidized than H2S produced by sulfate reduction. A bout 2% of the Fe(II) generated enters solution the other 98% is absorbed on aquifer material or forms iron bearing
55 minerals ( Lovley and Phillips 1988) I f conditions shift from iron to sulfate reduction along a given flow path, excess Fe(II) must first be consumed before the Fe(II) / H2S ratio will fall below 0.3.This means that while Fe(II) /H2S ratios of l ess than 0.3 are compelling evidence of dominant sulfate reduction, Fe(II) / H2S ratios greater than 10 are more e quivocal and encompass dominantly iron reducing conditions as well as the shift from iron to sulfate reducing conditions within complet e removal of Fe(II) ( Chapelle et al, 2009) Figure 42 depicts redox zones vs depths. Figure 43 depicts reduc ing conditions in a groundwater aquifer underlying an unlined landfill. US Geological Service (USGS) National Water Quality Assessment Program, developed the workbook entitled An Excel W orkbook for Identifying Redox Potential in Groundwater 2009 R e dox conditions within an GW aquifer, are subject to insitu concentrations of dissolved oxygen (O2), nitrate (NO3), manganese (Mn2+), iron ( Fe(II) ), sulfate (SO4 2 ), and sulfide (sum of dihydrogen sulfide [aqueous H2S], hydrogen sulfide [HS], and sulfide [S2 ]). The USGS workbook assigns predominate redox process analysis based on concentrations of these chemicals in the aquifer (Table 4 5) 4 5 Microbial Activity in Aquifers Iron bacteria are known to grow and proliferate in waters that contain as low as 0.1 mg/l of iron. However, at least 0.3 ppm of dissolved oxygen (DO) is needed to carry out oxidation ( Russell and Cohn, 2012; Atlas and Bartha, 1993) Iron bacteria colonize the transition zone where deoxygenated water from anaerobic environmental flows into and aerobic environment. Groundwater containing the dissolved organic material may be deoxygenated by microorganisms feeding on that dissolved org anic material. Where concentration of a n organic material exceed the
56 concentration of DO required for complete oxidation, microbial populations with specialized enzymes can reduce insoluble ferric oxide in the aquifer soils to solubilize ferrous hydroxide and use the oxygen released by that change to oxidize some of the remaining organic material ( Dyer 2003) The chemical reaction is noted in Equation 42 below. H2O + Fe2O3 2Fe(OH)2 + O2 (4 2 ) When deoxygenated water reaches a source of ox ygen, iron bacteria use that oxygen to convert the soluble ferrous iron back into an insoluble reddish precipitate of ferric iron The chemical reaction is noted in Equation 43 below. 2Fe(OH)2 + O2 H2O + Fe2O3 (4 3 ) Groundwater may naturall y be deoxygenated by decaying vegetation in swamps and useful mineral deposits of bog iron ore have formed where the groundwater has historically emerged to be exposed to atmospheric oxygen. Anthropogenic sources of landfill Leachate, septic drain fields or leakage from light petroleum fuels like gasoline are other possible sources of organic material allowing microbes to deoxygenate groundwater ( Dyer 2003) Bacteria known to feed on iron are Thiobacillus ferroxidans and Leptospirilum ferroxidans. Thiobacillus is a species of bacteria oxidize iron (or sulfide or copper) that use s the chemical energy to release to make sugar from carbon dioxide. Thiobacillus gains more energy for growth from sulfides compounds than from iron or copper. If sulfides ar e in the environment, these are the compounds of choice for Thiobacillus. H2S a waste product of sulfate reducing bacteria (SRB) is abundant in many anaerobic sediments subsurface aquifers and some areas where a landfill leachate plume is
57 discharging to surface water In many of these environments, Thiobacillus preferentially oxidizes H2S into sulfates ( Dyer 2003) With the exception of L. ferrooxidans, the same microorganisms ox idize and reduce sulfur compounds, leading to sulfur and iron bacterial pl umes in surface water bodies. These microbes, through enzymatic conversation of soil sulfates to volatile hydrogen sulfide, use sulfur compounds as an alternative source of oxygen in anaerobic environment ( Dyer 2003) Thiobacillus and its relatives are fo und in anaerobic environments that are sulfide and iron rich, in the interface between anaerobic and aerobic z ones. Thiobacillius can thrive in very acidic waters, unlike most ironoxi di zing bacteria described previously. However, metal oxidation is not an easy source of energy for Thiobacillus, so thes e bacteria are not often abundant enough to display field marks ( Dyer 2003)
58 Figure 41. Spatial distribution of t otal i ron concentrations b ased o n soil suborders (Valcare and Townsend 2011 as cited in Chen et al., 1999)
59 Figure 4 2 Redox z ones vs d epth (Millero, F.J. 1996)
60 Figure 43 Reducing conditions in g roundwater a quifer u nderlying unlined l andfill (Christensen, T.,et al 2000)
61 Table 4 1 Iron content in soil samples at n orthwest Florida county l andfills ( Subramaniam, 2007) County Landfill Name Iron Content (mg/ k g) Av erag e Iron Released (mg/l)* Soil Type ** Bay Steelfield 49, 6 00 80.23 Lakeland Sand  Calhoun Calhoun County 84, 1 00 Dorovan Pimlico Rutlege  Franklin Franklin County 39, 4 00 20.97 Resota Fina Sand  Gadsden Quincy Byrd 65, 8 00 20.94 Susquehann Sawyer Complex  Gulf Five Points 46, 4 00 Mandarin Fine Sand  Holmes Holmes 1 NA 40.36 Stilson Loamy Sand  Holmes Holmes 2 NA 24.12 Dothan Loamy Sand  Holmes Holmes County 91, 2 00 Orangeburg Fine Sandy Loam  Jackson Springhill 3 4 00 198.55 NA Leon US 27 South 43, 8 00 NA NA Liberty Liberty County 68, 8 00 NA NA Okaloosa Baker 119, 9 00 50.03 Dothan Loamy Sand  Santa Rosa Santa Rosa Holley 9 4 00 80.33 Troup Loamy Sand  Santa Rosa Leachate Pond NA 78.03 NA Santa Rosa Santa Rosa Central 83, 2 00 56.50 NA Wakulla Lower Bridge 67, 3 00 85.63 Udorthents & Quartzipsammements  Walton Walton County Central 9 0 00 256.39 Lakeland Sand  Washington Mudhill 84, 3 00 NA Lakeland Coarse Sand  Average iron released over 5 day period using simulated landfill leachate ** USDA Classification Systems
62 Table 4 2 Biological a ssessments in streams adjacent to unlined l andfill s in n orthwest Florida (Ray, 2013) Site EcoSubregion Sample Date Total Taxa Florida Index EPT Habitat Assessment Classification Comments BACKGROUND SITES Coffee C above Beulah Landfill 65f 3/4/1997 38 31 17 71% HEALTHY EMC basin survey, above Landfill clearing and covering operation impoundments/ new subdivisions upstream Pipeline Branch off Cox Road above gasoline 65f 3/21/2002 43 27 20 94% HEALTHY background for Camp 5 Landfill for Waste Mgt cleanup section Sourwood Branch at Mouth 75a 5/2/2006 42 28 21 92% HEALTHY Back ground 1st order stream for Coyote Landfill near Freeport Seminole Hills Branch 75a 10/11/2006 47 30 22 72% HEALTHY background for Steelfield Landfill at Sunbelt Sand Mine Point Baker Branch above 65f 1/30/2002 46 30 24 88% HEALTHY 1st order background site in Milton area CONTAMINANTED SITES Reststop Run Creek 65f 7/12/1995 29 24 13 91% SUSPECT Background for Perdido landfill and Champion Paper Coffee Creek 65f 7/13/1995 30 21 10 66% SUSPECT Monitoring site for upper Beulah landfill Superfund site & background for Champion Beaver Pond Ck Below I 10 65f 7/12/1995 11 4 1 54% IMPAIRED Monitoring site for Perdido landfill runoff Coffee Creek 65f 2/12/1996 21 21 9 42% IMPAIRED Declined after clear cutting Superfund Landfill site
63 T able 4 2 C ontinued Site EcoSubregion Sample Date Total Taxa Florida Index EPT Habitat Assessment Classification Comments Beaver Pond C reek below I 10 65f 2/14/1996 13 8 2 36% IMPAIRED Historically impacted Perdido Landfill drainage stable Reststop Run Creek 65f 2/14/1996 24 18 9 77% IMPAIRED Declined after new runoff from Perdido Landfill Coffee C below Beulah Landfill 65f 3/4/1997 22 18 7 38% IMPAIRED EMC basin survey, below clear cut superfund site at old Beulah Landfill Eleven Mile C. US 90 65f 3/6/1997 17 9 3 51% IMPAIRED Eleven Mile Creek basin survey, below Beulah Landfill & in paper mill waste assimilation zone "Central Landfill" Branch 65f 10/12/2000 4 4 1 38% IMPAIRED Walton county landfill enforcement Baggett Creek North Fork Baker Landfill 65f 3/20/2001 25 9 5 66% IMPAIRED Baker Landfill impacts from groundwater migration Turkey Creek tributary at Niceville Landfill 75a 3/22/2001 9 2 0 49% IMPAIRED Niceville Landfill impacts from groundwater migration Turkey Creek tributary at Niceville Landfill 75a 3/22/2001 9 2 0 IMPAIRED Compliance monitoring for Solid Waste Section "Harrison" Branch below Sterling Fiber Landfill 75a 6/12/2001 15 13 4 86% IMPAIRED Impacts from Sterling Fibers closed Landfill with RCRA staff St. Regis Branch below Sterling Fiber's SW3 75a 6/12/2001 10 0 1 68% IMPAIRED Impacts from Sterling Fibers upper landfill, sprayfield, & lagoons with RCRA staff Little Reedy Creek below Mudhill Landfill 65g 1/24/2002 12 2 0 43% IMPAIRED Washington County Landfill impacts with D. Koonce Sodom Branch east of Mudhill Landfill 65g 1/24/2002 26 14 10 83% IMPAIRED Washington County Landfill impacts with D. Koonce
64 T able 4 2. C ontinued Site EcoSubregion Sample Date Total Taxa Florida Index EPT* Habitat Assessment Classification Comments Air Products Stormwater Creek 75a 1/29/2002 9 5 0 64% IMPAIRED Industrial site impacts for water facilities & RCRA staff Dridglers Creek above Parkview Street 75a 1/29/2002 17 8 2 92% IMPAIRED Industrial site impacts for water facilities & RCRA staff Camp Five Branch below Landfill 65f 3/21/2002 16 5 1 56% IMPAIRED monitor Camp 5 Landfill for Waste Management cleanup section Goodwin Creek above Hwy 20 Freeport 75a 2/27/2003 10 3 2 31% IMPAIRED Monitoring Freeport C&D landfill Fairgrounds Branch below Autoshred Landfill 75a 5/20/2003 8 1 0 62% IMPAIRED Monitoring Auto shred landfill Turkey Creek tributary at Niceville Landfill 75a 7/9/2003 2 0 0 54% IMPAIRED For TMDL SCI verification with T. Thom USFWS Turkey Creek tributary at Niceville landfill SCI 75a 7/9/2003 5 0 0 54% Very Poor TMDL SCI score 13 of 33 "Central Landfill" Branch 65f 7/29/2003 2 0 1 36% IMPAIRED For TMDL SCI verification Swampy Creek below Coyote Landfill 75a 5/2/2006 20 3 2 49% IMPAIRED Monitoring adjacent Landfill Eleven Mile Creek above US90 75a 5/3/2006 19 6 3 51% IMPAIRED Record low flow, gross decomposition odor from paper mill waste feeding biomass background for Klondike Landfill Eleven Mile Creek below Klondike Landfill 75a 5/3/2006 20 9 3 60% IMPAIRED Downstream of Klondike Landfill overwhelmed by paper mill waste Otter Creek Steelfield Landfill SW3 75a 10/11/2006 16 5 1 49% IMPAIRED Monitoring Steelfield Landfill Bay Co Provided by Donald Ray FDEP staff Biologist NWD *EPT = Ephemeroptera, Plecoptera, & Trichoptera
65 Table 4 3 Chemolithoautotrophs: energy s ource and waste products (Dyer, 2003) Mineral/.Organic Compound used a source of energy Waste Product Bacteria H2 (hydrogen) CH4 (methane) Methanogens H2 H2O Hydrogen oxidizers (including alpha proteobacteria) H2 CH3 -----COOH (acetic aid) Some acetogenic bacteria (alpha proteobacteria ***) CO (Carbon monoxide) CO2 Carbon monoxide oxidizers (gamma proteobacteria) NH4 (ammonium) NO2 (nitrite) Ammonium oxidizers (beta proteobacteria) NO2 (nitrite) NO3 (nitrate) Nitrite oxidizers (beta proteobacteria) Fe2+ (iron) Fe3+ (rusted iron) Iron oxidizers (beta proteobacteria, ie Theobacillus) S or S2O3 2 ** (sulfur) SO4 2 (sulfate) Compound is oxidized with oxygen (O2) or some other compound. Source of carbon for all types of chemolithoautotrophs is carbon dioxide (CO2) ** good field mark *** autotrophic producers of acetic acid are different from those that produce vinegar
66 Table 4 4 Use of r edox sensitive com pounds for a ssignment of redox c onditions in groundwater aquifers ( Christense n 2000) Redox sensitive compound Indicator Transport and Geochemistry Sampling precautions Oxygen, O2 Electron acceptor: low concentrations (< 0.51 mg/l O2/l) indicate anaerobic conditions Field analysis by flow cell is recommended Nitrate, No 3 Electron acceptor: presence of nitrate indicates aerobic or nitrite reducing conditions Dinitrogen oxide, N2O, (dissolved gas) Intermediate in conversion of N compounds. May indicate nitrate reduction or nitrification Transport Volatile Nitrite, NO2 I ntermediate in conversion of N compounds. May indicate nitrate reduction Transport Instable Manganese, Mn2+ End product generated by manganese reduction. Presence of Mn2+ indicates anaerobic conditions and manganese reduction Transport, cation exchange, precipitation as carbonates, sulfides & oxidation by oxygen Filtration, elevated concentration of dissolved manganese mainly presents Mn(II) at pH 5 8 Iron, Fe2+ End product generated by iron reduction. Presence of Fe2+ indicates anaerobic cond itions and iron reduction Transport, cation exchange, precipitation as carbonates, sulfides & oxidation by oxygen Filtration, dissolved iron mainly present as Fe(II) at pH 59 Sulfate, S O4 2 (dissolved gas) Electron acceptor. Presence but decreased sulfate concentrations under anaerobic conditions indicate sulfate reduction Sulfite, S2 (dissolved gas) End product generated by sulfate reduction. Presence of sulfide indicates anaerobic conditions and sulfate reduction Transport precipitates with dissolved iron and manganese Volatile, field analysis Methane, CH4 (dissolved gas) Methane is created from reduction of carbon dioxide or degradation/fermentation of organic carbon. Presence of methane indicates anaerobic conditions and methanogenesis Transport Volatile
67 Table 4 5 Criteria and t hreshold concentrations for i dentifying r edox p rocesses in g roundwater (McMahon and Chapelle, 2008) Note: Table was modified from McMahon and Chapelle, 2008. Redox process: O2, oxygen reduction; NO3, nitrate reduction; Mn(IV), manganese reduction; Fe(III), iron reduction; SO4, sulfate reduction; CH4gen, methanogenesis. Chemical species: O2, dissolved oxy gen; NO3, dissolved nitrate; MnO2(s), manganese oxide with managanese in 4+ oxidation state; Fe(OH)3(s), iron hydroxide with iron in 3+ oxidation state; FeOOH(s), iron oxyhydroxide with iron in 3+ oxidation state; SO42 dissolved sulfate; CO2(g), carbon dioxide gas; CH4(g), methane gas. Abbreviations: mg/L, milligram per liter; greater than or equal to; < less than; >, greater than Redox category Redox process Electron acceptor (reduction) half reaction Criteria for inferring process from water quality data Dissolved oxygen (mg/L) Nitrate, as Nitrogen (mg/L) Manganese (mg/L) Iron (mg/L) Sulfate (mg/L) Iron/sulfide (mass ratio) Oxic O2 O2 + 4H + + 4e 2O 0.5 <0.05 <0.1 Suboxic Suboxic Low O2; additional data needed to define redox process <0.5 <0.5 <0.05 <0.1 Anoxic NO 3 2NO 3 + 12H + + 10e 2(g) + 6 H 2 O; NO 3 + 10H + + 8e 4 + + 3H 2 O <0.5 0.5 <0.05 <0.1 Anoxic Mn(IV) MnO2(s) + 4H + + 2e 2+ + 2H2O <0.5 <0.5 0.05 <0.1 Anoxic Fe(III)/SO4 Fe(III) and (or) SO4 2 reactions as described in individual element half reactions <0.5 <0.5 no data Anoxic Fe(III) Fe(OH)3(s) + H + + e 2+ + H2O; FeOOH(s) + 3H + + e 2+ + 2H2O <0.5 <0.5 >10 Mixed(anoxic) Fe(III) SO 4 Fe(III) and SO4 2 reactions as described in individual element half reactions <0.5 <0.5 10 Anoxic SO4 SO4 2 + 9H + + 8e + 4H2O <0.5 <0.5 <0.3 Anoxic CH4gen CO2(g) + 8H + + 8e 4(g) + 2H2O <0.5 <0.5 <0.5
68 CHAPTER 5 METHODOLOGY 5 .1 SiteWise LCA /GS R SiteWise TM Tool for Green and Sustainable Remediation (GSR) was developed jointly by US Navy, US Army Corps of Engineers (USACE), and Battelle. The SiteWiseTM GSR tool is an Excel based tool comprised of spreadsheets used to conduct a baseline assessment of GSR metrics. The quantitative metrics ca lculated by the tool include: (1) greenhouse gases (GHGs) reported as carbon dioxide equivalents (CO2e) and includes carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O); (2) energy usage (expressed as BTU and MWH); (3) air emissions of criteria pollutants including oxides of nitrogen (NOx), sulfur oxides (SOx), and particulate matter (PM10); and (4) accident ri sk (risk of injury and risk of fatality) (Navy, 2011) To estimate the environmental footprint for each remedial alternative, only the remedial activities that produce significant emissions were considered. These include the following: GW W ell installation : manufacturing of well materials; transportation of personnel; transportation of equipment; operation of equipment onsite; and management of residual waste System C onstruction and O peration: manufacturing of materials consumed ( i.e. chemical additives ) ; transportation of personnel; transportation of equipment and waste; and operation of equipment onsite ( i.e. drill rigs) M onitoring: transportation of personnel and equipment; operation of equipment onsite; and management of residual waste The SiteWise model is divided into 4 modules for input: a) Remedial Investigation (RI), b) Remedial A ction Construction (RAC) c) R emediation Action Operation (RAO), d) Long T erm M onitoring (LTM). The variables for these modules in each different scenario are defined below ( Navy 2011)
69 5 .1. 1 Remediation Action Investigation The RAI module is the module for input of values associated with cost and / or activity associated with the initial remediation action investigation. For each of the modeled scenarios in this study, the RAI module input is the same. The input parameters for this module were taken from the 1995 Remediation Investigation. The time period for this portion of the site investigation was 1990 to 1995. The major components of the investigation include: a) ins tallation of 33+ GW wells, b) collection and analysis of soils, GW, surface water (SW) and sediments c) engineering and scientific technical services, d) disposal of solid and hazardous waste, e) mobilization of construction equipment. Details of input parameters are noted in Appendix Additional Data 5 .1. 2 Remedial Action Construction The RAC module is the module for input of values associated with cost and/ or activity associated with the remediation action construction. For each of the modeled scena rios the RAC period and input parameters var ied based on remedial systems modeled. The input parameters for this module were developed from various remedial feasibility/optimization study documents for the site and conversations with experts in remedial technology design. The time period for each scenario will vary based on the system being installed. The construction of a system includes: a) installation of GW wells, b) collection and analysis of soils GW, SW and sediments, c) engi neering and scientific technical services, d) treatment of waste waters, e) disposal of solid and hazardous waste, f ) mobilization of construction equipment, g ) testing of system Details of input parameters for each scenario are noted in Appendix Addit ional Data
70 5 .1. 3 Remedial Action Operation The RAO module is the module for input of values associated with cost and/ or activity associated with the remediation action operation time period. For each of the modeled scenarios the RAO the time period is se t to 10 years. The input parameters for this module were obtained from various documents for the site and conversations with experts in remedial technology design. The operation of a system includes: a) operation and maintenance (O&M) of the system and GW wells b) monitoring of GW and SW b) collection and analysis of GW and SW samples c) engineering and scientific technical services, d) treatment of waste waters, and e) disposal of solid and hazardous waste. Details of input parameters for each scenario are noted in Appendix Additional Data 5 .1. 4 Long Term Monitoring The LTM module is the module for input of values associated with cost and/ or activity associated during the LTM period. For each of the modeled scenarios the LTM time period is set t o 20 years. The time period is set by EPA and state regulators based on regulatory stipulations for Financial Assurance. The LTM at the site is set as equal for all systems. The LTM of a system includes: a) O&M of GWM system b) collection and analysis o f GW and SW samples, c) engineering and scientific technical services, d) disposal of solid and hazardous waste. Details of input parameters for each scenario are noted in Appendix Additional Data 5 .2 Data Sources Data for this study was collected from the document repository known as Navy Installation Restoration Information System (NIRIS) for the site and through IR managers. Historical data was collected on GW analysis, remedial investigation,
71 interceptor trench design, optimization studies cost estimates, a nnual GWM reports feasibility studies for remediation t echnology, CERCLA Record of Decision (ROD) and Amendments documents, Navy GRX geospatial imaging web site and staff at NAVFAC southeast regional headquarters. The following CERLCLA reports f or the installation were reviewed in detail for assessing potential remediation technologies for the LCS/GSR simulation: Final Site Feasibility Study 1996` Final Remedial Investigation Report 1996 Limestone Interceptor T rench C onstruction R eport 1999 5 year O ptimization Study report for OU1 L andfill August 2004 and April 2008 G roundwater monitoring report f or 2010 to 2012 5 .4 Remediation Technology Selection Remediation technologies that were selected as viable scenarios to model in the SiteWise GSR/LCA were taken from these sources: a) historical knowledge of remediation technologies, b) consultation with FDEP Waste Cleanup staff, 3) consultation with remedi ation technology design engineers, geologist, and contractors 4) site specific feasibility studies Selected remediation t echnolog ies and the input parameters are described in the text below and summarized in Table 1 1 5 1 and 52 A detailed list of all model scenario input parameters is placed in Appendix Additional Data The four technologies 8 modeled scenarios, reviewed were evaluated based on a comparison of their i mplementability, e ffectiveness, and cost This analysis is summarized in Table 5 2 A general layout of the site is noted in Figures 5 1 and 5 2.
72 5 .4.1 Limestone Interceptor Trench System (LITS) A limestone interceptor trench system (LITS) was approved by EPA for the site as a viable remediation technology in 1997 (Navy 1998) The tr ench was constructed and began operation in 1999. This technology was selected as a final remedy based on its Implementability and its inherent geochemi cal properties to raise pH, buffering capacity, limit microbial growth, and potential to aid in precipit ation of F (III) in the upper portion of the aquifer prior to reaching the soil/surface water interface at Wetland 3. There are 4 model scenarios that utilize the LITS The length, width, and depth of the trench are varied in each modeled scenario in order to capture larger portions of the GW plume that would be discharging into Wetlands 3. T he LITS inc orporated an active GW pumping system which pumped GW from the lower portion of the trench to an onsite wastewater treatment plant (WWTP) for treat ment. The modeled scenarios vary active pumping of GW to a n onsite WWTP or GW passively flow ing through the trench (Permeable Reactive Barrier PRB) Variables that are modeled for LITS scenarios are noted in Table 5 2 Figures depicting construction an d equipment utilized in the trench modeled scenarios ar e noted in Figures 5 3 through 5 11 A barge was utilized for transportation of limestone material for the 1300 LF LITS modeled scenarios (i.e. large volume of material). The limestone was barged from 250 miles away to Port Operations dock at NAS Pensacola. Trucks were utilized on site from the dock to the construction site (i.e. 5 miles). The use of this means of transportation greatly reduced the GHG footprint associated with the modeled scenario. Fo r all LITS scenarios, the RAO period is set to 10 years T he LTM period is set for 20 years These time periods were selected in accordance with financial assurance guidelines set in state/federal CERCLA/RCRA administrative codes
73 5 .4.2 Groundwater Pump and Treat Groundwater pump and treat systems (GW P&T) are the conventional method for remediation of groundwater. GW P&T system s remov e of wastewaters in the aquifer, geo spatially in relationship to the location of the recovery well s (ie. cone of depression/ influence) to effectively remove contaminated waters exceeding cleanup criteria. The systems includes a series of recovery wells which pump contaminated groundwater out of the aquifer, and contaminated waters are treat onsite or sent to a WWTP. For the modeled scenario the recovery wells (RW) are spaced 50 ft on center to a depth of 3540 ft. The RW s are connected to a central pumping system and GW is pumped and treated at an onsite WWTP. A general layout of the GW P&T system is shown in Figu re 5 10. For the GW P&T scenario, the RAO period is set to 10 years. The LTM period is set for 20 years. These time periods were selected in accordance with financial assurance guidelines set in state/federal CERCLA/RCRA administrative codes. 5 .4.3 Air S parging Air S parging (AS) is a conventional method for remediation of Fe(II) in GW contamination AS is an in situ GW remediation technology that involves the injection of air/oxygen under pressure into a well installed into the saturated aquifer zone ( Miller 1996) When air is inject ed into an anaerobic saturated zone aquifer it is hypothesized that the Fe(III) will precipitate out within the aquifer prior to upper portion of the aquifer reaching the sur face water interface at Wetland 3. For the modeled scenario, the sys tems include a series of 113 air sparge points / wells (1 in), installed utilizing a truck mounted direct push drill rig system, connected to a central 6.5 hp air compressor inject ing ambient air into the system T he
74 sparging points are s paced <710 ft on center to a depth of 1 5 20 ft. The system will be solar powered and operated up to 10 hrs per day. The screened interval of the spar ge wells will be from 5 to15 ft ( Reiss 2013) The i mplementability of this technology is questionable, w ith dissolved Fe(II) in GW being greater than 20 mg/l. EPA Office if Underground Storage T anks recommendations for use of AS are: Fe(II) <10mg/l = Good <10 mg/l Fe(II) < 20 mg/l = wells require periodic testing and may need replacement Fe(II) > 20 mg/l air sparging not recommended! While it is not recommended by EPA to utilize AS at this site, in that the system will clog with bacteria, the system is modeled to provide a reference for this at other site where it may be implementable. For the GW P&T scenario, the RAO period is set to 10 years. The LTM period is set for 20 years. These time periods were selected in accordance with financial assurance guidelines set in state/federal CERCLA/RCRA administrative codes. 5 .4.4 Monitored Natural Attenua tion Monitored natural a ttenuation (NA) is the utilization of the natural ecosystem which functions as its own wastewater treatment system without human intervention. GW contaminant concentrations are expected to decrease through natural biotic and abiotic attenuation processes, thus rendering the waste less threatening to GW and improving GW quality to meet remedial goals with time. For t he M NA modeled scenario, data parameters for RAC and RA O is null The modeled scenario consists of data input for the RI and LTM modules. The LTM time period is set to reflect 30 years of
75 monitoring of GW. These time periods were selected in accordance with financial assurance guidelines set in state/federal CERCLA/RCRA administrative codes.
76 Figure 51 Aerial View of Surface Water Monitoring Points in at OU1 Landfill, NAS Pensacola, FL. Photo courtesy of US Navy 2012.
77 Figure 5 2 Groundwater and surface water s ampling l ocations at OU1 l andfill, NAS Pensacola, FL. Figure courtesy of US Navy 2010.
78 A B C D Figure 53 Limestone i nterceptor system t renching heavy e quipment 2,000 hp. A) b arrier wall trenching equipment with boom exposed, B) barrier wall trenching equipment with boom fully deployed, C) caricature of b arrier wall trenching equipment in cross sectional view, D) HDPE pipe with sock being installed in trencher at some time as backfill material. Photos c ourtesy of Dewind Dewatering 2013.
79 Figur e 56 Limestone t rench c onstruction 675 LF u tilized in 1999 at OU1 l andfill NAS Pensacola, FL. Photo courtesy of US Navy 1999. Figure 55 Limestone t rench c onstruction 675 LF in 1999 at OU1 l andfill NAS Pensacola, FL. Photo c ourtesy of US Navy 1999.
80 A B C Figure 5 7 Limestone Trench 675 LF 1999 asb uilt construction d etails OU1 landfill NAS Pensacola, FL. A) cross sectional view of trench at 20 LF de pth, B) riser pipe details, C) c ross sectional detail with HDPE pipe in bottom for active pump & treat of groundwater. Figures courtesy of US Navy 2009
81 Figure 58 Modeled scenario 675 LF limestone t rench f ootprint Photo courtesy of US Navy 2012. Figure 59 Modeled s cenario 1300 LF l imestone trench f ootprint Photo courtesy of US Navy 2012.
82 Figure 510. Modeled scenario g roundwater p ump and t reat system f ootprint Photo courtesy of US Navy 2012. Figure 511. Modeled scenario a ir sparging system f ootprint Photo courtesy of US Navy 2012.
83 Table 5 1 Modeled remediation t echnology summary i nput p arameters Technology Modeled Scenario Name Length Width Depth GW Treatment Construction Period Remediation Action Operation Long Term Monitoring Other Comment LF FT FT # Weeks # Years # Years Limestone Trench 1 LITS 675ft 18D Active 675 1.5 18 Active to onsite WWTP 8 10 20 Limestone Trench 2 LITS 675ft 40D Passive 675 3 40 Passive 8 10 20 Limestone Trench 3 LITS 1300ft 40D Active 1300 3 40 Active to onsite WWTP 8 10 20 Limestone Trench 4 LITS 1300ft 40D Passive 1300 3 40 Passive 8 10 20 GW Pump & Treat GW Pump Treat 1300 LF 1300 NA 40 Active to onsite WWTP 4 10 20 17 recovery Wells, 2in 50ft OC Monitored Natural Attenuation Monitored Natural Attenuation NA NA NA NA NA NA 30 Air Sparging Electric 1300 1520 NA 4 10 20 130 Sparge Points, 1 in wells, 7 10 ft on center Air Sparging Solar 1300 1520 NA 4 10 20 130 Sparge Points, 1 in wells, 7 10 ft on center GW = groundwater, LITS = limestone interceptor trench system, WWTP = wastewater treatment plant, NA=not applicable.
84 Table 5 2 Modeled s cenario r emediation t echnology i mplementability, effectiveness and c ost summary Technology Modeled Scenario Name Implementability Effectiveness Cost Limestone Trench 1 LITS 675ft 18D Active Yes Low High Limestone Trench 2 LITS 675ft 40D Passive Yes Low High Limestone Trench 3 LITS 1300ft 40D Active Yes Medium High Limestone Trench 4 LITS 1300ft 40D Passive Yes Medium High GW Pump & Treat GW Pump Treat 1300 LF Yes Low High Monitored Natural Attenuation M onitored Natural Attenuation Yes High Low Air Sparging Electric Air Sparging Electric Yes Low Low Air Sparging Solar Air Sparging Solar Yes Low Low GW = groundwater, LITS = limestone interceptor trench system.
85 C HAPTER 6 RESULTS AND DISCUSSION 6 .1 General The Site Wise model was run for the 4 technologies with 8 modeled scenarios, varying the input parameters based on technology. The RAI and LT M input parameters remained the same for all technologies. A discussion of the LCA/GSR output results for each technologies result is discussed in Groups as Table 61 A comparative analysis of the GSR evaluation for the OU1 site remedial scenarios is summarized in Table 6 2 to 6 4 Among the 8 scenarios M NA had the lowest environmental footprint overall. GW P&T however, on the other hand had the highest footprint. In general, the environmental footprint of the alternatives followed this order: M NA< Air Sparging < LITS 675ft 18D Passive < LITS 1300ft 40D Passive < LITS 675ft 40D Active < LITS 1300ft 40D Active < GW P&T. Table 6 2 and 6 4 show the quantitative environmental footprint metrics evaluated for each remedial alternative during the LCA/GSR review. Figure 6 4 to through 6 1 3 are graphical representation of the environmental footprint for the enhanced bioremediation alternative broken down by specific activities that make up the remedy. Table 6 2 and 63 are an overall summary and describes which metric has the greatest footprint for each alternative, and al so highlights which specific activities contribute to the footprint of the metric. The environmental footprint for each alternative is discussed below. 6 .2 Results 6 .2.1 Limestone Trench Scenarios A L ITS was installed at the site i n 1999. The trench operated from 1999 to 2009, with an active GW pumping system The pumping system discontinued in 2009 due to
86 ineffec t ive ness of the s ystem to capture the pl um e and remediate the COC iron. There were various reasons the system did not work properly. F actors which may have played a role in the failure of the sy stem are : a) reduced size of trench from original design (i.e. length, depth, and width), b) iron bacteria fouling, c) groundwater flow direction being radial vs. one direction, d) F e(I II) precipitate blocking preferential pathways, e ) dissolution of the lime media, f) natural organic material in the aquifer being high, g) high concentration of Fe(II) in aquifer, g) low pH of the aquifer, h) biogeochemistry and competition for other cati ons, i) other unknown factors. In the LITS s that w ere modeled in this study the length and depth of the trench was varied to represent partial to total capture of the GW plume. The trench was modeled using the parameters of the existing trench size but expanded from 675 1300 LF in length 20 40 ft in depth and 1.5 3 ft in width Group 1 m e trics for the LITSs were impacted significantly when a trench system was an active P&T system during the RAC and RAI time period. A major contributing factor during the trench construction period was the use of limestone in the trench system. The model incorporates the GHG emission for lime as if the limestone was pulverized l ime that is dr ied in a lime kiln T he lime CO2e coefficient utilized did reflect pulveriz at ion and drying in a kiln, an assumption w hich was supported by the I ntergovernmental Panel on Climate Change (I PCC) which suggest that all lime is eventually reduced in the natural ecosystem to CO2, the refore it all is emitted as CO2 (IPCC, 2006). The lime stone CO2e value was not revised to ref lect any non drying or pulverizing stage. The selected value of 0.847654955 for lime GHG CO2e proved to be a high value for the media compared to other media available for use. This coefficient
87 was supported in references by EPA and IPCC Tier 1 ( Navy, 2009 ; IPPC 2006) Other studies supported a 47% reduction for use as an agricultural amendment, similar to the use at the site ( W est 2005) T he value of 0 .8476 54955 was retained and considered to be the more conservat ive v alue for this study. For Group 1 metrics a ctivities with the greatest environmental footprint include: installation of wells, personnel and heavy trenching equipment transportation, and heavy equipment use during clearing/grubbing and trench installation. Trench installation is discussed further under heavy equipment usage below. Group 2 metrics for all modeled LITSs remained similar except for cost. The cost for installations varied based on length of trenching and lime required to backfill trench and/or number of wells required for recovery of GW. 6 .2. 1.1 Heavy equipment A 750 hp trencher was utilized for the 675 LF @ 20 ft depth LITS s. A 1200 hp trencher was utilized for all other scenarios. Both trenchers are able to perform at a rate of approximately 250 LF per day with a crew of 3 men working 10 hours per day. Trenching equipment installed the horizontal GW collection piping while digging trench system. 6 .2. 1.2 Active vs p assive t rench s ystem For LITSs with active pumping systems their energy consumption was higher. The operation and maintenance (O&M) of these systems was higher but less than the modeled GW P&T system. 6 .2. 2 Groundwater Pump and Treat GW P&T had the highest environmental footprint, an order of magnitude higher than the other remedial alternatives. Heavy equipment usage led to high emissions of
88 GHGs and air pollutant criteria. However, GW P&T had the lowest accident risk injury The accident risk leading to injury was due to the transportation activities undertaken to drill GW wells and personnel for monitoring. 6.2.3 Air Sparging Air sparging h ad the second lowest footprint among all of the alternatives considered. Activities with the greatest environmental footprint include: installation of wells, personnel and equipment transportation, and equipment use during clearing Accident risk leading to fatality for this alternative was due mostly to personnel transportation during long term groundwater monitoring. Accident risk leading to an injury was driven by both transportation and the heavy machinery use during the installation of wells. The energy consumed in this technology was the second lowest next to MNA There were two scenarios modeled, solar and electric The simulation reveale d that with respect to air sparging because the electric use w a s low, set at 30 minutes per day specifically for the precipitation of Fe(II) to Fe (III) The use of solar vs electric did not impact secondary impacts to the remedial technology If air sparging had been coupled with soil vapor extraction for removal of a more volatile contaminant like BTEX, the technology would have had a much higher environmental burden. Since the tow mod eled scenarios were used without soil vapor extraction there is no comparison to make to previous LCAs studies reviewed in literature. 6.2.4 Monitored Natural Attenuation ( M NA) M NA h as the lowest footprint among all of the alternatives considered. Activiti es with the greatest environmental footprint include: installation of wells, personnel and equipment transportation, and equipment use during clearing Accident risk leading to
89 fatality for this alternative was due mostly to personnel transportation during long term groundwater monitoring. Accident risk leading to an injury was driven by both transportation and the heavy machinery use during the installation of wells 6 .3 Discussion 6.3.1 General F our remediation technologies were evaluated, consisting of a total of 8 modeled scenarios. E ach technology was ranked as to whether the technology had a high, medium, or low total environmental footprint. Based on the remediation technology metric ranking, much like the new landfill siting proc ess, d ecision makers may utilize the analysis performed to determine which metric or combination of metrics will be the driver for the site. For some sites the GHG foot print may be the driver for decisions if they site is located in a n air pollution nonattainment area. At others sites, water consumption may be the driver for decision making if water resources are scarce or if pumping groundwater will dry up a streambed which is home to a n endangered species (i.e. Okaloosa darter). At most DoD sites the cost of construction and O&M for the system is the driver for technology selection. E PA will consider a LCA/GSR analysis in review ing feasibility studies for remediation at CERLCA sites, they do not always use the analysis to determine implementation of remediation technology. The US Navy requires that a LCA/GSR be performed when a contaminated site is in the final remedial technology selection at a CERCLA or state restoration site. For this study, the SiteWise LCA/GSR model demonstrated that an LCA is an effective screening tool and can be utilized in evaluating environmental impacts from the groundwater remedial technology selected. While SiteWise does not evaluate the
90 primary impacts, it complements the use of other models that do. For example if the LCA/GSR had been coupled with a fate and transport model, although they have different goals, it would have complimented and provided a global understanding of the total sit e impac ts (Godin et al, 2004). Other LCA models such as EASEWASTE, EPID2003 and a site specific Ecological Risk Assessment would have provided a better understanding of the true risk associated remediation of contaminants in Wetlands 3 and 4 of the wetlands system as a whole and the effectiveness of the system to remediate in a timely fashion with minimal cost to owner or secondary impacts from the construction and operation/maintenance of the selected remedy. The site conditions in other LCAs varied widely in sit e contaminants, geology, aquifer recharge, extent of contamination, regulatory point of compliance for regulatory, and remediation technology. The wide variety of sites conditions resulted in many different metrics causing difficulty in presenting comparis on of LCAs studies one to another. Additionally, the differences in environmental impact metrics measured with different computer models utilized added another layer of complexity for comparison. In all the studies that were reviewed, q uantitative and qual itative metrics were used in the analysis. SiteWise and the other studies were only similar in quantifying GHG and energy metrics, other metrics varied based on site and model utilized. This study proved that, Site Wise and LCA studies that computer simulation was utilized, is a very effective screening tool for assessing the secondary environmental impacts of groundwater remediation technologies ( Lemming et al., 2010; Boldin et al., 2011, Navy, 2009).
91 6.3.2 LCAs with MNA vs. Groundwater Pump and Treat In this study MNA proved to have the lowest secondary environmental impact overall and GW P&T had the highest or all metrics presented in this study The total GHG emission for MNA was 1,24 6 MT and 46,693 MT for GW P&T. The total energy consumed was 1.56E+04 M MBTU for MNA and 2.13E+04 MMBTU for GW P&T. Th e higher value of GHG emission and energy consumption metrics for similar use of these technologies was confirmed in previous studies by Lemming et al., 2012 and Navy, 2009. This study proved that active pump and treat remediation systems had higher environmental footprints and passive remediation has the lowest secondary environmental impact. While other studies reviewed utilize a variety of approaches to quantify GHG emission and energy consumption, our results aligned with previous studies. This study showed that activities associated with construction, operation and maintenance of a system and transportation to the site of equipment were major contributors to this burden. The location of the POC toxici ty of contaminant, regulatory water quality standards and location of receptors are factors a regulatory agency takes into consideration when to allow MNA at a site. The POC is not just moved down gradient from a contaminated area to allow for dilution as a means of remediation. Therefore the owner must demonstrate that the selection of the POC at a cer tain location is valid and location does not pose a threat to receptors, or active remediation must be implemented. MNA has the lowest cost to the owner and has been implemented at the site for the last 3 years. Site specific surface water quality standards for Fe(II) and Fe(III) total
92 iron concentrations within Wetlands 3 are not met at location SW1, but the site meets site specific water quality criteria at SW3, a POC further away of the source. The LCA/GSR does not address exceedances of regulatory standards at SW1, nor does the model provide additional considerations for this failure to meet these standards. If regulatory agencies required compliance at SW1, the model would need to be coupled with a fate and transport model to evaluate Fe(III) concentrations and potential LTM time periods greater than 10 years. Additionally, if the regulatory agencies required compliance at SW1, DoD may consider purchase of wetlands via mitigation banking equivalent to the total area of Wetlands 3 & 4 and utilize nonconventional remediation and convert the wetlands areas to wetlands treatment systems. 6.3. 3 Impacts from Remedial Investigation and LTM Phase In the literature reviewed, many of the LCAs did not include the RI phase in their system boundary. Further analysis in this study proved that this can contribute a significant amount of GHH and priority air pollutants. For GHG emissions, the RI phase can contribute for active technologies up to 11% and the LTM phase up to 63% of the burden ( Table 6 6 ). For priority air pollutants, the RI phase can contribute for active technologies up to 11% and LTM phase up to 65% ( Table 6 5 ). For energy consumption, the RI phase can c ontribute for active technologies up to 12% and LTM phase up to 64% ( Table 6 7 ) 6.3. 4 Transportation Transportation of large equipment to and from the site proved to be a source of both GHG and Air pollutants. T he 1300LF LITS scenario utilized ocean transportation, which proved to be an efficient means of moving the large volume of material to the site, conserving resources and reducing t he overall environmental burden, confirming other
93 studies (Dettore, 2009). The active technology air sparging scenario, utilizing no reactive media only oxidation, had the second lowest footprint and proved that there was a significant reduction in environmental impact and O&M cost. Selection of other media, with stronger reaction rates, would have potentially r educed the environmental impacts for transportation. 6.3.5 Long Term Monitoring Periods and Point of Compliance For all modeled scenarios, the LTM was set for 10 years and the POC was set at the SW3 sampling location. This LTM period and POC location wer e selected in order to quantify and analyze secondary environmental impacts for all scenarios equally and perform a rapid assessment of secondary environmental impacts. If LTM period or POC had been set based on fate and transport of the contaminant throug h system to the POC and another location closer to the MSWLFs, sampling station SW1, remediation cost associated with regulatory compliance would have been unreasonable for owner. This is similar to discussion presented in by Cadotte et al., 2007, Godin et al., 2007, Lemming et al. 2012. While a fate transport model was not presented in this study, the topic presented to build the case that to have a robust LCA once should consider inclusion of one in the analysis for primary impacts to a system. The LTM phase for the air sparging scenario was set to 10 years. It is reasonable to assume that this is sufficient time to determine whether the technology will be effective, it is not reasonable to assume that the technology will be totally remediating the groundwater plume over time. This active treatment system was designed to alter Fe (II) through an oxidation process as it passes through the air barrier system. If the active treatment system is cut off, it will no longer act at remediating what remains in the g roundwater. A longer period of O&M should be assumed to control contaminants until:
94 a) all organic matter and carbon source is consumed, b) stabilization of waste in the landfill is achieved, c) the system boundaries equalized that are controlling the plum e (i.e. recharge, tidal influence, etc.)., d) carbon from all sources in the groundwater rendered consumed, e) competition for ions is consumed (i.e. nutrients, sulfates, nitrates, TKN, Mn, etc.), f) other biogeochemical effects. It would be reasonable to assume that this time period could be another 50 to 75 years. For the active LITS remedial systems, once the pumping system is cut off, the material remains in place and continues to remediate iron as it passes through the barrier systems. While the reme diation time period for all scenarios was set for 10 years, this system continues to work, only the media may need to be replaced at the end of it useful life. Similar to the air sparging systems, the scenario modeled should reflect a longer per iod of O&M The O&M phase should reflect a reasonable time period based on monitoring well data for the contaminant of iron when it exceeds the established site specific groundwater quality criteria is exceeded consistently over 4 monitoring periods (i.e. 1520 year s) similar to what is presented in Lemming, 2010. In the literature reviewed, there was no consensus as to a specified LTM period. Some researchers set it as a regulatory compliance period or when a water quality criterion was achievable with cost effecti ve technology (Lemming et al. 2012 ; Bayer and Finkel 2006). Previous studies presented best professional judgments as to POC and LTM based on fate transport models, this study utilized actual LTM periods and POC established in the ROD signed by regulator y agencies with oversight of the CERCLA site.
95 6.3. 6 Limestone and Crushed Concrete Permeable Barrier Media L ime utilized in the LITS was a significant source of GHG emissions as a result of from the manufacturing of lime which includes the heating, drying and pulverizing of the stones. Additionally, t he large volume of lime, 150 m3 for the 1300LF LITS and 16 m3 for the 675LF LITS, contributed to the environmental burden. For GHG emissions the use of lime in this study contributed to th e overall GHG metric of each technology as 24% for the 1300LF LITS to and 68% for the 675LF LITS. This confirm ed what was presented in LCA studies on the use of lime in bio leaching of soils and passive treatment of acid mine drainage (Blanc et al., 2004, Ziemkiewicz et al, 2003). A conservative approach was taken into consideration in this study that all lime is eventually reduced in the natural ecosystem to CO2, therefore it all is emitted as CO2 (IPCC, 2006). If a larger stone, greater than 2, was used in this model and a different CO2e coefficient used to reflect no kiln drying or pulverizing of stone, it would have significantly reduced the GHG emissions impacts It is questionable as to whether a larger stone would be viable and /or effective at thi s site, based on the 10 year s of GW monitoring of the 675LF LITS at the site (S ection 6.2.1) and the failure of this system from 1999 to 2009. If crushed concrete was used as the media in the trench system in lieu of the lime rock, there would be a slight reduction in GHG emission burden. Navia et al. 2006 utilized contaminated soil in a roadway cement application, in which the reuse of material slightly lowering overall air pollution emissions. Lesage et all 2007 assumed that the recycling of concrete at a site displaced an equivalent amount of crushed gravel for use at the site.
96 For this LCA study, c rushed concrete w as not modeled as media in an interceptor trench system but the US EPA waste reduction model ( WARM ) was used and GHG emission credits for rec ycling of concrete are presented for the replacement of lime in the trench system For the 675 LF trench there would be a reduction in GHG emissions by 7MT CO2E and 97 MBTU for energy consumed. For the 1300LF LITS this would be a reduction in GHG emission s of 60 MT CO2E and 802 MBTU of energy consumed At this site, this is a potentially viable option since the installation has an abundance of recyclable concrete from demolition of structures. The overall GHG emission footprint for each modeled scenario w ould be lower from the use of recyclable materail and from transportation of material to this site since the material is located at the installation. Wang et a; 2012 d emonstrated in the laboratory that construction demolition debris waste, 40% concentration of crushed concrete in combination with organic matter, will reduce the concentration of iron in simulated leachate bench test. Wang et al. in 2013 studied iron removal efficiency of calcium carbonatebased material in groundwater with natural organic mat erial C rushed concrete media and limestone rock of various particle sizes was studied. The studies proved that crushed lime rock and concrete had a 99% Fe (II) efficiency over time. The study concluded that v arious environmental factor s, including pH, coexisting cations, and natural organic matter can change the water chemistry and impact Fe(II) removal for these materials L imestone in a l o w pH 4 environment, will significantly reduce the removal effectiveness of Fe(II). In a neut ral pH range, pH between 5 and 9, the removal effectiveness was improved slightly by increasing pH Sodium ( Na+ ) and calcium ( Ca 2+) exhibited similar effects
97 with a slight effect on the removal process due to the incr ease in solution ionic strength. L im estone is effective at removal of Manganese ( Mn 2+) a nd is a competitive cation for Fe(II), as they compete for sorption sites on the media. In addition, NOM decreases Fe(II) uptake by limestone and thus red uces the removal effectiveness. While these bench tests proved the plausibility of crushed concrete to work in the laboratory, it is hard to extrapolate what the effect would be in the field at NAS Pensacola. The LITS that was constructed in 1999, employ ing pure limestone rock it failed to provide an adequate Fe(II) removal efficiency at the sampling station SW1, adjacent to the constructed LITS The s ystem operated for 10 years and was declared ineffectual by regulatory agencies. Concrete is a mixture of materials, of which coal fl y ash i s commonly uti lized, Coal fly ash may or may not contain higher concentrations of arsenic than pure lime. At all times, the designer of the remediation system should carefully consider the selected media and the technology such that the system selected does not introduce new contaminants into the system. Chemical analysis should be performed on media to see if the media leaches other contaminants of concern that may pose a problem with regulatory compliance. Additionally, a thorough review of literature should be perfor med, and input from peers as to implementability of the proposed media should be discussed with peers. For new technologies utilizing media which has not been tested, pilot testing will be necessary to be better able to see what will actually occur in the field. 6.3.7 Unconventional Approaches to Remediation While there are only 4 active and passive remediation technologies presented in this study. T here are sever a l unconventional passive technologies and one active that is noteworthy for discussion that one would consider for this site: a) enhanced
98 phytoremediation, b) wetlands treatment in Wetlands 3 and air sparging c) mitigation banking and Wetlands 3. Enhanced phytoremediation is a potentially viable passive technology that has been proven by many researchers in the past at remediating many types of contaminants. For this site the owner would consider enhancing the existing wetlands with trees and plants that are effective at remediating iron and pumping groundwater awa y from the wetlands (i.e. tulip popular). For the uplands area surrounding Wetlands 3, there would be a phased harvesting of the short leaf pine and replanting with long leaf pine, tulip poplar and other native trees. Inside Wetlands 3, intermittently planting the areas with other native wetlands plants and future research on plants within the wetlands that uptake iron (i.e. hyper bioaccumulators). The environmental impacts from this technology would be similar to the MNA, except for a slightly higher envi ronmental burden from the heavy equipment used in pla n ting pine trees in the uplands The burden from the RAC would be minimal and no additional burden for the RAO phase of the study, therefore it is hypothesized that the secondary environmental impacts w ould be similar to the AS technology and the similarity of heavy equipment used in construction of this technology The state of Florida will permit wetlands to be utilized as a w etlands t reatment system with appropriate documentation (FAC 62 611) For thi s site, the use of Wetlands 3 as a permitted wastewater treatment system may be consider viable if the owner coupled enhanced the system with an air sparging curtain spaced intermittently throughout the wetlands. This technology, based on the iron content in the surface water would work similar to the air sparging curtain presented in the study except the
99 curtain is in surface water verse groundwater. This system would capture any Fe(II) that remains in the surface water and precipitate it as Fe(III) in v arious sump areas throughout Wetlands 3. The environmental impacts from this technology would be similar to the AS modeled. The burden from the RAC would be minimal and no additional burden for the RAO phase of the study, therefore it is hypothesized that the secondary environmental impacts would be equal to the AS technology. A pilot test using a solar powered air compressor at the site would be easy to implement near SW2 and be very low cost to the owner to evaluate the effectiveness of a system throughout the wetlands. The state of Florida has a wetlands mitigation banking system that allows owners to purchase wetlands in a banking system if they chose to destroy wetlands. The Wetlands 3 area, approximately 45 acres, could be purchased and the owners r ehabilitate the site The land could be back fill with concrete rubble and use d as golf green area, or forested with long leaf pine, or any other similar use Depending on the end use of the land, the secondary environmental impacts could be very low to medium depending the whether the land was raised to a higher elevation and heavy equipment utilized to backfill the wetlands area. This unconventional technology would be viable, effectively eliminate groundwater from discharging into surface water, in essence removing the problem at the source. The short term secondary environmental impact would be minimal depending heavy equipment use and the specific end us e of the land but over the long term the environmental impact would be similar to MNA if not eliminating the problem all together.
100 Table 61 SiteWise s ustainability m atrix g roups Grouping Metric Group 1 GHG Emission include carbon dioxide (CO2), nitrous oxide (N2O), methane (CH4), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and sulfur, hexafluoride (SF6) Total Energy Use energy from nonrenewable versus renewable sources Water Water consumption c an be evaluated both qualitatively and quantitatively NO x SO x PM 10 Emission Criteria air pollutants activities such as transportation, electrical usage, and heavy machinery and equipment use during remedy implementation Accident Risk Fatality and Injury Worker safety/accident risk is the risk of fatality or injury of carrying out a specific task of a remedial activity Group 2 Non Hazardous Waste Landfill Hazardous Waste Landfill Space Costing Lost Hours Injury Final Cost with Footprint Reduction Group 3 Community Impacts Resources Lost
101 Table 6 2 SiteWise s ustainability m etric q uantified for e ach a lternative global w arming & a ir Remedial Alternatives GHG Emissions Total energy Used Water Consumption NO x emissions SO x Emissions PM 10 Emissions metric ton MMBTU gallons metric ton metric ton metric ton Air Sparging 1300LF 20D Electric 1 285.13 1 5 9 00 1.04E+05 3.61 2.44 0.10 Air Sparging 1300LF 20D Solar 1 285.13 1 5 9 00 1.04E+05 3.61 2.44 0.10 GW Pump Treat 1300 LF 46 693.45 2 1 3 00 4.23E+08 9.53 4.66 1.09 Monitored Natural Attenuation 1 246.53 1 5 6 00 8.10E+04 3.56 2.36 0.09 Long Term Monitoring Only 20 YR 746.08 9, 24 0 5.00E+04 2.13 1.42 0.06 Long Term Monitoring Only 30 YR 1 119.12 1, 39 00 7.50E+04 3.19 2.12 0.08 Remedial Investigation Only 127.41 1, 71 0 6.00E+03 3.71 2.40 0.01 Trench 1300LF 40D 3W Active 28 523.24 5, 49 00 2.08E+08 4.93 2.43 0.63 Trench 1300LF 40D 3W Passive 6 177.73 5, 12 00 8.50E+04 4.14 2.48 0.1 3 Trench 675LF 18D 1W P assive 1 806.98 1, 91 00 8.35E+04 3.34 2.12 0.10 Trench 675LF 18D 1W Active 12 938.37 2, 04 00 1.04E+08 2.59 1.30E 0.3 4 Table 6 3 SiteWise s ustainability m etrics q uantified for e ach a lternative o ther Remedial Alternatives Non Hazardous Waste Landfill Space Hazardous Waste Landfill Space Costing Lost Hours Injury Final Cost with Footprint Reduction Accident Risk Fatality Accident Risk Injury tons tons $ $ Air Sparging 1300LF 20D Electric 9.50 1.00 1.83E+06 724 1.83E+06 5.84E 04 0.09 Air Sparging 1300LF 20D Solar 9.50 1.00 1.83E+06 724 1.83E+06 5.84E 04 0.0 905 GW Pump Treat 1300 LF 9.50 1.00 3.28E+06 770 3.28E+06 6.36E 04 0.09 62 Monitored Natural Attenuation 5.50 1.00 1.33E+06 568 1 1.33E+06 4.65E 04 0.07 10 Long Term Monitoring Only 20 YR 3.00 0.00 9.75E+05 0.290 9.75E+05 2.46E 04 0.03 62 Long Term Monitoring Only 30 YR 4.50 0.00 9.75E+05 0.4.35 9.75E+05 3.69E 04 0.05 44 Remedial Investigation Only 1.00 1.00 3.50E+05 0.1.33 3.50E+05 9.59E 05 0.01 66 Trench 1300LF 40D 3W Active 6.50 2.00 3.03E+06 1.52 3.03E+06 1.26E 03 0.190 Trench 1300LF 40D 3W Passive 6.50 2.00 2.28E+06 1.44 2.28E+06 1.12E 03 0. 1.80 Trench 675LF 18D 1W PASSIVE 7.50 3.00 1.88E+06 1.87 1.88E+06 1.39E 03 0.2 33 Trench 675LF 18D 1W Active 6.50 2.00 2.05E+06 1.24 2.05E+06 8.52E 04 0.156
102 Table 6 4 SiteWise relative impact Remedial Alternatives GHG Emissions Energy Usage Water Usage NOx E missions SOx Emissions PM10 Emissions *Accident Risk Fatality *Accident Risk Injury Air Sparging 1300LF 20D Electric Low Low Low Low Low Low Medium Medium Air Sparging 1300LF 20D Solar Low Low Low Low Low Low Medium Medium GW Pump Treat 1300 LF High Medium High High High High Medium Medium Monitored Natural Attenuation Low Low Low Low Low Low Medium Medium Long Term Monitoring 20 Y r Low Low Low Low Low Low Low Low Long Term Monitoring 30 Y r Low Low Low Low Low Low Low Low Remedial Investigation Only Low Low Low Low Low Low Low Low Trench 1300LF 40D 3W Active Medium High Medium Medium Medium Medium High High Trench 1300LF 40D 3W Passive Low High Low Low Low Low High High Trench 675LF 18D 1W P assive Low Medium Low Low Low Low High High Trench 675LF 18D 1W Active Low Medium Low Low Low Medium Medium Medium *Accident Risk is an estimate of how many accidents may occur. This risk is not the same as Cancer Risk, which is the probabi lity (for a single person) of getting cancer. Accident risk is not comparable to Cancer Risk due to inherent fundamental differenc es. Table 6 5. SiteWise r emedial i nvestigation p hase p riority air p ollutants environmental i mpact (%) Remedial Alternatives NOx E missions SOx Emissions PM10 Emissions Sum Remedial Investigation % LTM 20 YR % metric ton metric ton metric ton Air Sparging 1300LF 20D Electric 3.25 2.21 0.09 5.55 11 % 65 % Air Sparging 1300LF 20D Solar 3.25 2.21 0.09 5.55 11 % 65 % GW Pump Treat 1300 LF 95.34 46.55 1.09 142.99 0 % 3 % Monitored Natural Attenuation 3.56 2.36 0.09 6.02 10 % 60 % Trench 1300LF 40D 3W Active 49.34 24.35 0.63 74.31 1 % 5 % Trench 1300LF 40D 3W Passive 4.14 2.48 0.13 6.76 9 % 53 % Trench 675LF 18D 1W PASSIVE 3.34 2.12 0.10 5.56 11 % 65 % Trench 675LF 18D 1W Active 25.93 13.03 0.34 39.31 2 % 9 %
103 Table 6 6 Remedial investigation phase greenhouse gas e mission e nvironmental Impact (%) Table 6 7 Remedial i nvestigation phase energy consumption e nvironmental i mpact (%) Remedial Alternatives Total E nergy Used Remedial Investigation LTM 20 YR MMBTU % % Air Sparging 1300LF 20D Electric 14,399 12 % 64 % Air Sparging 1300LF 20D Solar 14,399 12 % 6 4% GW Pump Treat 1300 LF 21,260 8 % 43 % Monitored Natural Attenuation 15,573 11 % 59 % Trench 1300LF 40D 3W Active 54,893 3 % 17 % Trench 1300LF 40D 3W Passive 51,182 3 % 18 % Trench 675LF 18D 1W P assive 19,081 9 % 48 % Trench 675LF 18D 1W Active 20,393 8 % 45 % Remedial Alternatives GHG Emissions Remedial Investigation LTM 20 YR metric ton % % Air Sparging 1300LF 20D Electric 1,182 11 % 63 % Air Sparging 1300LF 20D Solar 1,182 11 % 63 % GW Pump Treat 1300 LF 46,693 0 % 2 % Monitored Natural Attenuation 1,247 10 % 60 % Trench 1300LF 40D 3W Active 28,523 0 % 3 % Trench 1300LF 40D 3W Passive 6,178 2 % 12 % Trench 675LF 18D 1W P assive 1,807 7 % 41 % Trench 675LF 18D 1W Active 12,938 1 % 6 %
104 Figure 61. G reenhouse gas e missions from a ll sources d uring the m odeled SiteWise p eriod. 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 45,000 50,000 Air Sparging 1300LF 20D Electric Air Sparging 1300LF 20D Solar GW Pump Treat 1300 LF Trench 1300LF 40D 3W Active Trench 1300LF 40D 3W Passive Trench 675LF 18D 1W PASSIVE Trench 675LF 18D 1W Active Metric Tons GHG Emissions
105 Figure 62. Energy u sed from a ll sources during the modeled SiteWise period 0 10,000 20,000 30,000 40,000 50,000 60,000 Air Sparging 1300LF 20D Electric Air Sparging 1300LF 20D Solar GW Pump Treat 1300 LF Trench 1300LF 40D 3W Active Trench 1300LF 40D 3W Passive Trench 675LF 18D 1W PASSIVE Trench 675LF 18D 1W Active MMBTU Total Energy Used
106 Figure 63. Water c onsumed at site & n ot r eturned to groundwater s ystem on site i mpacts from all s ources during the m odeled SiteWise p eriod. 0.00E+00 5.00E+07 1.00E+08 1.50E+08 2.00E+08 2.50E+08 3.00E+08 3.50E+08 4.00E+08 4.50E+08 Air Sparging 1300LF 20D Electric Air Sparging 1300LF 20D Solar GW Pump Treat 1300 LF Trench 1300LF 40D 3W Active Trench 1300LF 40D 3W Passive Trench 675LF 18D 1W PASSIVE Trench 675LF 18D 1W Active Gallons Water Impacts
107 Figure 64. NOx emissions from a ll sources d uring the m odeled SiteWise p eriod 0 20 40 60 80 100 120 Air Sparging 1300LF 20D Electric Air Sparging 1300LF 20D Solar GW Pump Treat 1300 LF Monitored Natural Attenuation Trench 1300LF 40D 3W Active Trench 1300LF 40D 3W Passive Trench 675LF 18D 1W PASSIVE Trench 675LF 18D 1W Active Metric Tons NOx Emissions
108 Figure 65. SOx i mpacts from a ll sources d uring the modeled SiteWise p eriod. 0 5 10 15 20 25 30 35 40 45 50 Air Sparging 1300LF 20D Electric Air Sparging 1300LF 20D Solar GW Pump Treat 1300 LF Monitored Natural Attenuation Trench 1300LF 40D 3W Active Trench 1300LF 40D 3W Passive Trench 675LF 18D 1W PASSIVE Trench 675LF 18D 1W Active Metric Tons SOx Emissions
109 Figure 66. PM10 e missions from all s ources d uring the m odeled SiteWise p eriod 0 0 0 1 1 1 1 Air Sparging 1300LF 20D Electric Air Sparging 1300LF 20D Solar GW Pump Treat 1300 LF Monitored Natural Attenuation Trench 1300LF 40D 3W Active Trench 1300LF 40D 3W Passive Trench 675LF 18D 1W PASSIVE Trench 675LF 18D 1W Active Metric Tons PM10 Emissions
110 Figure 67. Accident r isk f atality to a ll w orkers d uring the m odeled SiteWise p eriod 0.0000 0.0002 0.0004 0.0006 0.0008 0.0010 0.0012 0.0014 0.0016 Air Sparging 1300LF 20D Electric Air Sparging 1300LF 20D Solar GW Pump Treat 1300 LF Monitored Natural Attenuation Trench 1300LF 40D 3W Active Trench 1300LF 40D 3W Passive Trench 675LF 18D 1W PASSIVE Trench 675LF 18D 1W Active Risk of Fatality Accident Risk Fatality
111 Figure 68. Accident r isk i njury to all workers during the m odeled SiteWise p eriod 0.000 0.050 0.100 0.150 0.200 0.250 Air Sparging 1300LF 20D Electric Air Sparging 1300LF 20D Solar GW Pump Treat 1300 LF Monitored Natural Attenuation Trench 1300LF 40D 3W Active Trench 1300LF 40D 3W Passive Trench 675LF 18D 1W PASSIVE Trench 675LF 18D 1W Active Risk of Injury Accident Risk Injury
112 Figure 69. Non hazardous w aste d isposed from all s ources d uring the m odeled SiteWise p eriod. 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 Air Sparging 1300LF 20D Electric Air Sparging 1300LF 20D Solar GW Pump Treat 1300 LF Monitored Natural Attenuation Trench 1300LF 40D 3W Active Trench 1300LF 40D 3W Passive Trench 675LF 18D 1W PASSIVE Trench 675LF 18D 1W Active Tons Non Hazardous Waste Landfill Space
113 Figure 610. Hazardous w aste d isposed from all s ources d uring the modeled SiteWise p eriod. 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 Air Sparging 1300LF 20D Electric Air Sparging 1300LF 20D Solar GW Pump Treat 1300 LF Monitored Natural Attenuation Trench 1300LF 40D 3W Active Trench 1300LF 40D 3W Passive Trench 675LF 18D 1W PASSIVE Trench 675LF 18D 1W Active Tons Hazardous Waste Landfill Space
114 Figure 611. Costing f or remediation t echnologies from all s ources d uring the m odeled SiteWise p eriod. 0 500,000 1,000,000 1,500,000 2,000,000 2,500,000 3,000,000 3,500,000 Air Sparging 1300LF 20D Electric Air Sparging 1300LF 20D Solar GW Pump Treat 1300 LF Monitored Natural Attenuation Trench 1300LF 40D 3W Active Trench 1300LF 40D 3W Passive Trench 675LF 18D 1W PASSIVE Trench 675LF 18D 1W Active USD Dollars Costing
115 Figure 6 12. Lost hours i njury from a ll sources d uring the modeled SiteWise period. 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 Air Sparging 1300LF 20D Electric Air Sparging 1300LF 20D Solar GW Pump Treat 1300 LF Monitored Natural Attenuation Trench 1300LF 40D 3W Active Trench 1300LF 40D 3W Passive Trench 675LF 18D 1W PASSIVE Trench 675LF 18D 1W Active Hours Lost Hours Injury
116 Figure 613. Final cost with f ootprint r eduction from a ll sources d uring the modeled SiteWise period. 0 500,000 1,000,000 1,500,000 2,000,000 2,500,000 3,000,000 3,500,000 Air Sparging 1300LF 20D Electric Air Sparging 1300LF 20D Solar GW Pump Treat 1300 LF Monitored Natural Attenuation Trench 1300LF 40D 3W Active Trench 1300LF 40D 3W Passive Trench 675LF 18D 1W PASSIVE Trench 675LF 18D 1W Active USD Dollars Final Cost with Footprint Reduction
117 C HAPTER 7 SUMMARY AND CONCLUSION A life cycle analysis was performed to evaluate secondary environmental impacts using a software tool known as SiteWise on 4 groundwater (GW) remediation technologies Technologies were selected that were viable for remediation of an iron/sulfur enriched GW plume emanating from an open dump landfill discharging into a wetland adjacent to the landfill. The analysis consisted of 8 modeled scenarios quantifying impacts of remediation for the remedial investigation (RI), remedial construction (RC), remedial action construction and operations (RAC and RAO), and long term monitoring (LTM) phases of activity at the site. Each technology was ranked as to whether the technology had a high, medium, or low total environmental footprint. Based on the remediation technology met ric ranking, much like the new landfill siting process, decision makers may utilize the analysis performed to determine which metric or combination of metrics will be the driver for the site. For some sites the GHG foot print may be the driver for decisions if they site is located in an air pollution nonattainment area. At others sites, water consumption may be the driver for decision making if water resources are scarce or if pumping groundwater will dry up a streambed which is home to an endangered speci es (i.e. Okaloosa darter). At most DoD sites, the cost of construction and O&M for the system is the driver for technology selection. EPA consider s a LCA/GSR analysis in reviewing feasibility studies for remediation at CERLCA sites, they do not always use the analysis to determine implementation of remediation technology. The US Navy requires that a LCA/GSR be
118 performed when a contaminated site is in the final remedial technology selection at a CERCLA or state restoration site. The SiteWise analysis revealed that conventional aggressive and energy intensive technologies have the greatest environmental footprint, confirming findings by other researchers. Active remediation systems, verses passive, had higher GHG emissions and energy usage for secondary envir onmental impact metrics. An active pump and treat system (GW P&T) in a sand gravel aquifer with 17 recovery wells, at a depth of 40 ft, will emit 46,693 MTs emissions vs 1,247 MTs for monitored natural attenuation (MNA). A passive 1300LF limestone trench system (LITS), 3 ft wide and 40 ft depth emits of 6,178 MTs of GHG emissions, while a similar active LITS will emit 28,523 MTs. An air sparging system (AS) emits 1,285 MT of GHGs. The energy footprint in MMTBU is: a) 15,600 for MNA, b) 21,300 for GW P&T, c) 54,900 for active LITS, d) 51,200 for passive LITS, d) 15,900 for AS. The priority air pollutant in MTs is: a) 6.01 for MNA, b) 15.28 for GW P&T, c) 7.99 for active LITS, d) 6.75 for passive LITS, d) 6.15 AS. The SiteWise model proved to be very effecti ve as a tool to aid decision makers in remediation technology feasibility phase of assessing effects to human health and environmental for secondary impacts of the use of a remediation technology at a site. In all the LCA studies that were reviewed, q uanti tative and qualitative metrics were used in the analysis. SiteWise and the other studies were only similar in quantifying GHG and energy metrics, other metrics varied based on site and model utilized. This study proved that, Site Wise and LCA studies that computer simulation was utilized, is a very effective screening tool for assessing the secondary
119 environmental impacts of groundwater remediation technologies ( Lemming et al., 2010; Boldin et al., 2011 Navy, 2009). Additionally, this study confirmed what other researchers finding that GHG emission and energy consumption resulted from the use of media that was manufactured like pulverized lime or activated carbon ( Lemming et al. 2012 ; Bayer and Finkel, 2006, Blanc et al., 2004, Ziemkiewicz et al, 2003). While there are only 4 active and passive remediation technologies presented in this study. There are several unconventional passive technologies and one active that is noteworthy for discussion and potential for futur e research that one would consider for this site: a) enhanced phytoremediation, b) wetlands treatment in Wetlands 3 and air sparging, c) mitigation banking and Wetlands 3. Enhanced phytoremediation is a potentially viable passive technology that has been proven by many researchers in the past at remediating many types of contaminants. For this site the owner would consider enhancing the existing wetlands with trees and plants that are effective at remediating iron and pumping groundwater away from the wetlands (i.e. tulip popular). For the uplands area surrounding Wetlands 3, there would be a phased harvesting of the short leaf pine and replanting with long leaf pine, tulip poplar and other native trees. Inside Wetlands 3, intermittently planting certain are as with native wetlands plants and future research on plants within the wetlands that uptake iron (i.e. hyper bioaccumulators). The environmental impacts from this technology would be similar to the MNA, except for a slightly higher environmental burden from the heavy equipment used in planting pine trees in the uplands. The burden from the RAC would be minimal and no additional burden for the RAO phase of the study, therefore it is hypothesized that the secondary environmental impacts would
120 be similar t o the AS technology and the similarity of heavy equipment used in construction of this technology. Additional analysis should be done on quantifying the actual secondary environmental impacts from this type of remediation approach and the viability of reme diating iron with hyper bioaccumulator plants. It would be of interest to the research and regulated community to do an in depth study of the plants in Wetlands 3 to see which plants are up taking and bioaccumulating iron, and further analysis of the enzy me s or proteins that may be secreted through their roots systems. The state of Florida will permit wetlands to be utilized as a wetlands treatment system with appropriate documentation (FAC 62611). For this site, the use of Wetlands 3 as a permitted waste water treatment system may be consider viable if the owner coupled enhanced the system with an air sparging curtain spaced intermittently throughout the wetlands. This technology, based on the iron content in the surface water, would work similar to the air sparging curtain presented in the study except the curtain is in surface water verse groundwater. This system would capture any Fe(II) that remains in the surface water and precipitate it as Fe(III) in various sump areas throughout Wetlands 3. The envir onmental impacts from this technology would be similar to the AS modeled. The burden from the RAC would be minimal and no additional burden for the RAO phase of the study, therefore it is hypothesized that the secondary environmental impacts would be equal to the AS technology. A pilot test using a solar powered air compressor at the site would be easy to implement near SW2 and be very low cost to the owner to evaluate the effectiveness of a system throughout the wetlands. Additional analysis should be done on quantifying the actual secondary environmental impacts from this type of remediation approach and the viability of
121 remediating iron with an AS system insitu at SW02. It would be of interest to the research and regulated community to do an in depth st udy of various spacing of air curtains throughout the Wetlands 3 system, focusing on various sections of the wetlands in a phased approach and observe the impact to benthic macroinvertebrate communities (i.e. reemergence of various types of communities through the system and at what location in the system). The state of Florida has a wetlands mitigation banking system that allows owners to purchase wetlands in a banking system if t hey chose to destroy wetlands. The Wetlands 3 area, approximately 45 acres, could be purchased and the owners rehabilitate the site. The land could be back fill with concrete rubble and used as golf green area, or forested with long leaf pine, or any other similar use Depending on the end use of the land, the secondary environmental impacts could be very low to medium depending the whether the land was raised to a higher elevation and heavy equipment utilized to backfill the wetlands area. This unconventional technology would be viable, effectively eliminate groundwater from disc harging into surface water, in essence removing the problem at the source. The short term secondary environmental impact would be minimal depending heavy equipment use and the specific end use of the land, but over the long term the environmental impact would be similar to MNA if not eliminating the problem all together. One area of interest to the res earch and regulated community f r o m this type of unconventional approach, and of value in further research, is to observe groundwater fluctuations and groundwater flow if the wetlands were filled with concrete rubble. Questions to address in the research are: a) would the iron problem disappear?; b) what are the impacts from concrete rubble vs. filling with soils
122 from the site?; c) if the area was planted with native s pecies that are natural pumps, i.e. tulip popular, would this remediate iron? Would GW flow direction change?; d) what would be the actual secondary environmental impacts if a SiteWise model was employed with a fate and transport model? Focus on com parison and contrast of LCA s using EASEWASTE EPA TRACI, EDIP2003 and SiteWise
123 APPENDIX ADDITIONAL DATA This a ppendix contains documents used to support cost estimating and input assumptions in SiteWise tool for remediation technologies. The hyperlink for documents is located below: Object A 1 SiteWise Remediation Technology Input Data Object A 2 Cost Estimate Supporting Data for Remediation Technologies
124 REFERENCES Atlas, R. Bartha, R., Microbial Ecology Fundamentals and Applications, 3rd Edition, 1993, Benjamin Cummins Publishing Company, 390 Bridge Parkway, Redwood City CA. 94065, pp 334335. Bayer, P., Finkel, M., Lifecycle A ssessment of A ctive and P assive Groundwater R emediation T echnologies, Journal of Contaminant Hy drology 83 (2006) 171 199. Blanc, A., Metivier Pignon, H. Gourdon, R., Rousseaux, P., Life Cycle Assessment as a Tool for Controlling the Development of Technical Activities: Application to the Remediation of a Site Contaminated by Sulfur, Advances in Environmental Research 8 (2004) 613 627 Boldin, A. Neidel, T.L., Damgaard, A. Bhander, G. S. Moller, J. Christensen, T. Modeling of Environmental Impacts from Biological Treatment of Organic Municipal Waste in EASEWASTE, Waste management 31 (2011) pp. 619 630. Bonano, E., Apostolakis, G., Salter, P. Ghassemi, A., Jennings, S. Application of risk assessment and decision analysis to the evaluation, ranking and selection of environmental remediation alternatives. Journal of Hazardous Materials 71 (2000) 35 57. Cadotte, M Deschnes L Samson, R Selection of a Remediation Scenario for a Diesel, Contaminated Site Using LCA. International Journal of Life Cycle Assessment 12 (4), ( 2007) pp 239 251. Chapelle, F., Bradley, P. Thomas, M. ; McMahon, P. Distinguishing ironreducing from sulfate reducing conditions, ( 2009) Ground Water, 47: pp 300 305 Chen, M., Ma, L., Harris, W., Hornesby, A., Background Concentrations of Trace Metals in Florida Surface Soils: Taxonomic and Geographic Distribution of Total total and Total recoverable Concentrations of Selected Trace Metals, University of Florida, Florida Center for Solid and Hazardous Waste, Report #997, (1999), pp. 1 156. Christensen, T., Bjerg, P. Banwa rt, S. Jakobsen, R. Heron, G. Al brechsten, H. Characterization of Redox Conditions in Groundwater Contaminant plumes over time, Journal of Contaminant Hydrology 45 (2000) pp 1 8 5 Dettore, C., Comparative Life Cycle Assessment of Bottled vs. Tap Water Systems, Report No. CSS0911, Center for Sustainable Systems University of Michigan, ( 2009) pp 1109. Dyer, B., A Field Guide to Bacteria, Comstock Publishing, Cornel University, ( 2003) pp 114115, 140 1 57.
125 FDEP Florida Administrative Code (FAC), 62302.530 Criteria for Surface Water Qua lity Classifications, ( 2007) FDEP Florida Administrative Code (FAC), 62611 Wetlands Application (1989). Gallagher, P., Spatari, S. Cucura, J. Hybrid Life Cycle Assessment of Comparison of Colloidal Silica and Cement Grouted Soil Barrier Remediation Tech nologies, Journal of Hazardous Materials, 250251 (2013), pp. 421430. Godin, J; Menard, JF; Hains, S; et al., Combined Use of Life Cycle Assessment and Groundwater Transport Modeling to Support Contaminated site Management, Human and Ecological Risk Asses sment, ( 2004) Vol.10 6, pp 1099 1116. Intergovernmental Panel on Climate Change IPCC Guidelines for National Greenhouse Gas Inventories Chapter 11: N2O Emissions from Managed Soils, and CO2 Emissions from Lime and Urea Application, ( 2006) IPCC pp 1130. Jurgens, B.C., McMahon, P.B., Chapelle, F.H., and Eberts, S.M., 2009, An Excel workbook for identifying redox processes in ground water: U.S. Geological Surv ey OpenFile Report 2009 1004, p. 8 Krug, C Dewind Dewatering, phone Interview trenc hing Equipment, Dec. 27, 2012. Lemming, G., Hauschild, M., Bjerg, P., Life Cycle Assessment of Soil and Groundwater Remediation Technologies: Literature Review, International Journal of Life Cycle Assessment Remediation Technologies, ( 2010) 15:115 127. Lemming, G., Chambon, J., Binning, P., Bjerg, P., Is There an Environmental Benefit From Remediation of a Contaminated Site? Combined Assessments of the Risk Reduction and Life Cycle Impact of Remediation, Journal of Environmental Management, 112 (2012), p p 392403. Lesage, P., Ekvall, T., Deschnes, L., Samson, R., Environmental assessment of Methodological approach, International Journal of Life Cycle Assessment V ol. 12(6)(2 007) 391 398. Lovley, D. Phillips, E., Novel Mode of Microbila Metabolism Organic Carbon Oxidation Coupled to Dissimilatory Reduction of Iron or Manganese. Applied Mi crobiology 54(6): ( 1988) pp 1472. Miller, R. GroundWater Remediation Technologies Analysis Center, Air Sparging, Technology Overview Report, TO 96 04, GroundWater Remediation Technologies Analysis Center, October ( 1996 ) p i. Millero, F.J. Chemical Oceanography. CRC Press LLC, (1996) pp. 469.
126 Mohana, D., Chandera, S. Removal and R ecovery of M etal Ions from A cid Mine D rainage U sing L ignite A Low C ost S orbent, Journal of Hazardous Materials B137 (2006) pp. 1545 1553. Morais, S. A. Delerue Matos, C., A P erspective on LCA A pplication in S i te R emediation Services: Critical Review o f C hallenges, Journal of Hazardous Materials 175 (2010) 12 22. Navia, R., Rivela, B.,K., Lorber, Mendez, R., Recycling Contaminated Soil as A lternative Raw M aterial in Cement F acilities: Life C ycle A ssessment, Resources, Conservation and Recycling 48 (2006) 339 356. Ray, Donald FDEP phone interview Iron Smothering November 1, 2012. Ray, D onald Landfill Impacts on Stream Benthos in the Florida Panhandle, 2003 presentation. Reiss, Rich, US NAVY NAVFAC Southeast, Environmental Section, Remediation Design En gineer interview February 2, 2013. Reinhart, D Goosh, C., Analysis of Florida MSW Landfill Leachate Quality, Report #973, July 1998. p p 110. Russell, J,; Cohn, R.; Iron Bacteria, Bookivia Publishing Lennex Corp, Mitchell House Edinburgh, Scotland United Kingdom, www.pubmix.com ( 2012 ) pp 5 8 Srgio Alberto Morais Cristina Delerue Matos, A Perspective on LCA Application in Site Remediation S ervices: Critical Review of Challenges, Journal of Hazardous Materials 175 (2010) pp 12 22 Subramaniam, P. I mpact of Landfill Leachate on Iron Release From Northwest Florida Iron Rich Soils Electronic, Theses, Treatises and Dissertations. Paper 1519, ( 2007) pp 25. Suer, P. An dersson Skold, Y. Biofuel or E xcavat ion? Lifecycle Assessment (LCA) of S oil R emediation O ptions, Biomass and Bioenergy 35 (2011) 969981. Townsend, T, Presentation: Assessing the Potential for Mobilization of Naturally Occurring COCs from Soils and Aquifer Sediments as a Result of Landfill Construction and Operation, Dec 6, 2011. US EPA, Green Remediation: Incorporating Sustainable Environmental Practices into Re mediation of Contaminated Sites Office of Solid Waste and Emergency Response ( 2008) EPA 542 R 08002.
127 US EPA, L ife Cycle Assessment Principles and Practice, National R isk Management Research Laboratory, Systems Analysis Branch O ffice of Research and Development EPA/600/R 06/060, SAIC, Contract No. 68 C02 067, Work Assignment 315, ( 2006 ) pp 13. US EPA, Chapter VII Air Sparging, pp VII 15, web site: http://www.epa.gov/oust/pubs/tum_ch7.pdf US EPA Office of Underground Storage Tanks web site: http://water.epa.gov/scitech/wastetech/guide/coal/upload/2009_03_24_guide_coal_sup port_support1.pdf US EPA, Office of Climate Change and Waste, WARM model: http://www.epa.gov/c limatechange/waste/calculators/Warm_Form.html US Navy, Ensafe/A llen and Hoshall Final Focused Feasibility Study NAS Pensacola Site 1, Contract # N6246789 D0318 CTO 059, November 8, 1996. US Navy, Ensafe/Allen and Hoshall, Final Remedial Investigation, NA S Pensacola Site 1, Contract # N6246789D0318 CTO 059, January 5, 1996. US Navy, Ensafe, Final Remedial Investigation Report Site 41Operable Unit 16, Volume 1 of IXI, S e ctions 1 9, NAS Pensacola Contract # N6246789 D0318 CTO 036, August 17, 2005, Revis ed November 16, 2007. US Navy, Ensafe, Final Record of Decision OU1 NAS Pensacola, August 14, 1998. US Navy Optimization Study Report for OU1 NAS Pensacola, FL Contract N6246794D 0888 Task Order 0296, Tetra Tech NUS Inc., April 2008. US Navy Optimization Study Report for OU1 NAS Pensacola, FL Contract N6246794D 0888 Task Order 0296, Tetra Tech NUS Inc., August 2004. US Navy, SiteWise Case Study Alameda, CA ( 2009) pp 1 2 US Navy, SiteWise Case Study Parris Island, (2009) pp 12. US Navy, SiteWise V ersion 2 User Guide, NAVFAC Engineering Service Center, Battelle Memorial Institute, June 2011 US Office of the White House Executive Order EO 13423 Executive Order 13423 of, Strengthening Federal Environmental, Energy, and Transportation Management, President George Bush, January 24, 2007. US Office of the White House EO 13514 Executive Order 13514 of Federal Leadership in Environmental, Energy, and Economic Performance, US President Barack Obama, October 5, 2009.
128 Valcarce, C., Townsend, T. Factors Affecting Iron Mobilization in Surface and Groundwater Sites near Florida Landfills, unknown date, pp. 14. Wang, Y Sikora, S., Kim, H, Dubey, B, To wnsend, T., Mobilization of iron and arsenic from soil by construction and te, Waste Management 32 (2012) pp 925 932. Wang, Y ., Sikora, S., Kim, H., Boyer, T. Bonzongo, J., Townsend, T., Effects of S olution C hemistry on the Removal Reaction Between C alcium Carbonatebased M aterials and Fe(II), Science of the Total Environment, 443 (2013), pp 717724. Wang, Y ., Sikora, S., Kim, H., Boyer, T. Bonzongo, J., R h ue, D. Townsend, T., Evaluation of M ineral Substrates for I n situ I ron R emoval from G roundwater Environmental Earth Science, published online, University of Florida DOI 10.1007/s1266501220548 ( 16Feb2013) West T O ., McBride A.C., The Contribution of A gricultural L ime to Carbon D ioxide E missions in the United States Dissolution, T ransport, and N et E missions, Agriculture, Ecosystems and Environment 108 (2005) pp 145 1 54 Zhao, Y., Xing, W., Lu, W., Zhang, X., Christensen, T., Environmental I mpact A ssessment of the I ncineration of M unicipal Solid W aste with A uxiliary C oal in China, Waste Management 32 (2012) 1989 1998. Ziemkiewicz, P, Skousen, J., Simmons, J, Long Term Performance of Passive Acid Mine Drainage Treatment Systems, National Min e Land Reclamation Center, West Virgina University Morgantown VA. ( 2003) pp 112.
129 BIOGRAPHICAL SKETCH Ms Kathy Gaynor o btained her undergraduate degree in c ivil e ngineering fr o m the Univer sity of Alabama in 1983. S h e work e d for the Florida D epartment of Environm ental Protection (FEP) for 15+ years in Tallah a ssee F L as an environmental engineer in S olid and Hazar dous Waste (S&HW) Over 10 years of this time of se rvice were dedicated to site remediation at Resource Conversation and Recovery Act ( RCRA ) sites and management of hazardous waste. An additional 5 + years were occupied environmental management of solid waste facilities and landfills. While employed at FDEP she was called to provide technical assistance to US EPA and US Agency for International Development (USAI D) to provide training to small and medium size municipalities in Central America for improvement of their solid waste management systems (i.e. MSW landfills, recycling facilities and hazardous waste management). She has travelled to Central America 30 + times assisting communities and training officials in this capacity. Ms. Gaynor currently works for the Naval Facilities (NAVFAC) Engineering Command Southeast (SE) region in the office of S olid and H azardous W aste in Jacksonville Florida. At NAVFAC SE she is program manager for the Other Environment al Liability (OEL) program for N aval facilities throughout the SE region.