DESIGNINGANOPTIMIZEDCONSTRUCTEDWETLANDSYSTEMWITHCIRCULARSUSTAINABILITYFORWASTEWATERTREATMENTINMACAOADissertationPresentedtoTheAcademicFacultybyMicaelaLeiInPartialFulfilmentoftheRequirementsfortheDegreeofMasterofEnvironmentalScienceandManagementintheInstituteofScienceandEnvironmentUniversityofSaintJoseph,MacaoDecember/2020

3AcknowledgementsIwouldliketothankmysupervisor,ProfessorCristinaS.C.Calheiros,forprovidingguidanceandfeedbackwithendlesspatiencethroughoutthisproject,andProfessorRaquelVasconcelosforhersupportandencouragement.Finally,Iwouldliketoexpressmydeepgratitudetomyfamilyfortheirenduringsupportduringmydissertationjourney.

4AbstractinEnglishConstructedWetlands(CWs)areexamplesofnature-basedsolutionsthatmimicthebiogeochemicalprocessesthatoccurinnaturalwetlandsystems.Theyareoftenappliedtoremovepollutantsfromdifferenttypesofwastewater.Besideswatertreatment,CWscanalsoprovideotherecosystemserviceslikewaterreuseforirrigationpurposes,biomassforenergyproductionandenhancelocalbiodiversity.Basedonlifecycleassessmentsdata,CWshaveshowntobecost-effectiveandalowenvironmentalimpacttechnology.ThisstudyaimstoreviewandevaluateenvironmentalimpactofthemainmaterialsthatmakeupthecomponentsofCWs,aswellascombiningtheconceptofwatercircularityintourbanwastewatermanagement.Intensifiedstrategiestotacklerate-limitingfactorsforimprovingCWperformancehavealsobeenexplored.AfinaloptimizeddesignwithselectedsustainablematerialswasproposedforimplementationinMacaoSAR,acityinChinawiththesecondhighestpopulationdensityintheworld.Weattempttocalculatetheareaneededtotreatthepost-primarytreatedwastewaterfromtheMacaoPeninsulawastewatertreatmentplantandfurtherlocatesuitableareasforCWimplementation.Itwouldalsoprovideaninsightintocouplingnature-basedsolutionswithurbanwastewatertreatment,bioenergyproductionandatthesametime,achievecircularwatermanagement.Thepresentstudywillenabledecision-makerstolookatwastewatermanagementinadifferentperspectiveotherthantraditionalhard-engineeringprojects.AbstractinPortugueseAszonashúmidasconstruídas(ZHC)sãoexemplosdesoluçõesbaseadasnanaturezaquemimetizamosprocessosbiogeoquímicosqueocorremnaszonashúmidasnaturais.Elassãofrequentementeaplicadaspararemoverpoluentesdediferentestiposdeáguas

5residuais.Paraalémdotratamentodeágua,asZHCtambémpodemforneceroutrosserviçosecossistémicos,comoreutilizaçãodeáguaparafinsdeirrigação,biomassaparaproduçãodeenergiaepromoveroaumentodabiodiversidadelocal.Combasenosdadosdeanálisedeciclodevida,asZHCtêm-semostradoumatecnologiaeconómicaedebaixoimpactoambiental.EsteestudotemcomoobjetivorevereavaliaroimpactoambientaldosprincipaismateriaisqueintegramoscomponentesdasZHC,bemcomoaliaroconceitodecircularidadedaáguaàgestãodeefluentesurbanos.EstratégiasintensificadasparaabordarosfatoreslimitantesdeformaamelhorarodesempenhodasZHCtambémforamexploradas.UmdimensionamentootimizadocommateriaissustentáveisfoiselecionadoparaimplementaçãonaRAEdeMacau,umacidadedaChinacomasegundamaiordensidadepopulacionaldomundo.Tentou-secalcularaáreanecessáriaparatrataraságuasresiduais,apóstratamentoprimário,provenientesdaestaçãodetratamentodeáguasdaPenínsuladeMacau,eaindaidentificarasáreasadequadasparaaimplementaçãodeZHC.Tambémforneceráumavisãosobreacombinaçãodesoluçõesbaseadasnanaturezacomotratamentodeáguasresiduaisurbanas,aproduçãodebioenergiae,aomesmotempo,atingirumagestãocirculardaágua.Opresenteestudopermitiráqueosdecisoresolhemparaagestãodeáguasresiduaisdeumaperspetivadiferente,paraalémdosprojetostradicionaisdeengenhariaconvencional.AbstractinChinese

6Publicationsandawards●BlueThinkConference-AbstractandaPosterrelatedtothisthesiswerepresentedinaConferenceheldbyCIIMAR(CentroInterdisciplinardeInvestigaçãoMarinhaeAmbiental)on17thSeptember,2020:LeiM.,VasconcelosR.,CalheirosC.S.C.2020.Constructedwetlandsforwastewatertreatmentindenselyurbansettlements.InBookofAbstractsofBlueThinkConference:ShareScience,SpreadKnowledge(pp.110).Matosinhos:CIIMAR.ISBN:978-989-54965-0-1●4thMeetingofFunctionalBiologyandBiotechnologyofPlants-AbstractandaPosterwithtitle“DesigninganOptimizedConstructedWetlandSystemwithCircularSustainabilityforWastewaterTreatmentinMacao”werepresentedinaConferenceheldbytheUniversityofPortoon2ndDecember,2020●AScholarshipwasawardedtothisthesisproposalbyMacaoFoundationintheir“MacaoStudies”PostgraduateScholarshipsScheme(No.19023).

7ContentsENDORSEMENT........................................................................................................2ACKNOWLEDGEMENTS........................................................................................3ABSTRACTINENGLISH.........................................................................................4ABSTRACTINPORTUGUESE................................................................................4ABSTRACTINCHINESE.........................................................................................5PUBLICATIONSANDAWARDS.............................................................................6CONTENTS.................................................................................................................7LISTOFABBREVIATIONS...................................................................................111.INTRODUCTION.................................................................................................141.1WASTEWATERMANAGEMENTINCITIES...............................................................141.1.1Importanceofwastewatermanagement......................................................141.1.2Conventionalwastewatertreatmentplantinurbanareas............................151.1.3Advantagesandlimitationsofconventionalwastewatertreatment............191.1.4Futuretrendsforwatermanagement:CircularSustainability.....................191.2.CONSTRUCTEDWETLANDSFORWASTEWATERMANAGEMENT............................211.2.1.Constructedwetlandtechnology................................................................211.2.2Advantagesandlimitationsofconstructedwetlands..................................221.2.3.Componentsofconstructedwetlands.........................................................231.2.4.Functionandservices.................................................................................251.2.5.Classificationofconstructedwetlands.......................................................261.2.6.Mechanismsofpollutantremovalinconstructedwetlands........................281.2.7.Constructedwetlandimplementation.........................................................31

81.2.8.Designingaconstructedwetland................................................................331.3.AIMSANDOUTLINEOFTHISTHESIS....................................................................352.METHODS.............................................................................................................382.1LITERATUREREVIEW...........................................................................................382.2CONSTRUCTEDWETLANDDIMENSIONING............................................................403.RESULTSANDDISCUSSION............................................................................423.1.CONSTRUCTEDWETLANDS’MATERIALTOWARDSCIRCULARSUSTAINABILITY...423.1.1Substrateselection.......................................................................................423.1.1.1Filtermaterialsforphosphorusremoval...............................................433.1.1.1.1Soil,sandandgravel......................................................................433.1.1.1.2Dolomite.........................................................................................443.1.1.1.3Zeolite.............................................................................................453.1.1.1.4Lightweightaggregates..................................................................453.1.1.1.5Tabletporousmaterial....................................................................453.1.1.1.6Redmud.........................................................................................463.1.1.1.7Biochar...........................................................................................463.1.1.1.8Slag.................................................................................................463.1.1.1.9Oilshaleash...................................................................................473.1.1.1.10Crushedautoclavedaeratedconcrete...........................................473.1.1.1.11Alumsludge.................................................................................483.1.1.1.12Magnesiumslagparticle...............................................................493.1.1.1.13Fragmentedmoleanoslimestone..................................................493.1.1.2Phosphorusrecoveryandfilterrecyclabilityfromsaturatedsubstrate.503.1.2Vegetationselection....................................................................................56

93.1.2.1Vegetationasbioenergyfeedstock:First-generationbiofuels..............603.1.2.1.1Sugarcane.......................................................................................603.1.2.1.2Corn................................................................................................613.1.2.2Vegetationasbioenergyfeedstock:Second-generationbiofuels........623.1.2.2.1A.donax,P.australis,T.latifolia..................................................633.1.2.2.2C.indicaandI.pseudacorus..........................................................643.1.3Linerselection.............................................................................................643.2EVALUATIONOFCONSTRUCTEDWETLANDSYSTEMSWITHLIFE-CYCLEASSESSMENTSTUDIES................................................................................................663.2.1LCAscomparingconventionalWWTPandCWs.......................................663.2.2LCAofCWwithlightweightexpandedclayaggregateassubstrate..........683.2.3LCAofaFieldTidalFlowCWwithDewateredAlumsludgeassubstrate693.2.4LCAofbiofuelproductionthroughCW.....................................................713.3REDUCTIONINFOOTPRINT(LANDREQUIREMENT)OFCONSTRUCTEDWETLAND.763.3.1FactorsthatreduceCWefficiency:Landavailability.................................783.3.2FactorsthatreduceCWefficiency:Oxygenavailability.............................793.3.2.1LCAofMicrobialFuelCellcoupledCW.............................................823.3.3FactorsthatreduceCWefficiency:Temperature........................................833.3.4FactorsthatreduceCWefficiency:Carbondeficiencyindenitrification...843.3.5FactorsthatreduceCWefficiency:Re-releaseofnutrientsbydecomposingvegetation..............................................................................................................843.3.6FactorsthatreduceCWefficiency:Microbialdegradation.........................843.3.7FactorsthatreduceCWefficiency:Clogging.............................................853.4.CONSTRUCTEDWETLANDSINHIGHDENSITYURBANSETTLEMENTS:CASESTUDYOFMACAO................................................................................................................86

103.4.1BackgroundinformationofMacao..............................................................863.4.2CWdimensioningforMacao......................................................................913.4.3FinalconstructedwetlanddesignforMacao...............................................943.4.3.1CWdesignguidelinesfromLCAstudies.............................................953.4.3.2SubstrateselectionforpilotCWinMacao...........................................963.4.3.3VegetationSelectionforpilotCWinMacao......................................1013.4.3.4LinerSelectionforpilotCWinMacao...............................................1063.4.3.5FinalCWdesignforMacao................................................................1063.4.4LocationsforCWimplementationinMacao............................................1104.CONCLUSIONSANDFUTUREWORK.........................................................1154.1.GENERALCONCLUSIONS...................................................................................1154.2.FUTUREWORK..................................................................................................1175.REFERENCES.....................................................................................................119

11ListofAbbreviationsAAArtificialaerationAACAutoclavedaeratedconcreteADFAbioticdepletionoffossilfuelsASTSActivatedsludgewastewatertreatmentBODBiochemicaloxygendemandCAACCrushedautoclavedaeratedconcreteCIIMARCentroInterdisciplinardeInvestigaçãoMarinhaeAmbientalCODChemicaloxygendemandCWConstructedwetlandDASDewateredalumsludgeDODissolvedoxygenDSDissolvedsolidECEuropeanCommissionEPAEnvironmentalProtectionAgencyEPDMEthylenepropylenedienemonomerEREffluentrecirculationEUEuropeanUnionFGDFlue-gasdesulfurizationFMLFragmentedMoleanoslimestoneFTWFloatingtreatmentwetlandsFUFibrinunitFWSFreewatersurfaceGHGGreenhousegas

12GJGigajouleGWPGlobalWarmingPotentialHCWHybridconstructedwetlandHDPEHigh-densitypolyethyleneHFHorizontalflowHLRHydraulicloadingrateLCALifeCycleAssessmentLECALight-expandedclayaggregatesLIHDLow-inputhigh-diversityLWALightweightaggregatesMFCMicrobialfuelcellMJMegajouleNANotapplicableNBSNature-basedsolutionsNEBNetenergybalanceNYCNewYorkCityOCROxygenconsumptionrateOLROrganicloadingratePPhosphorousPSBPhosphate-solubilizingbacteriaPVCPolyvinylchlorideSARSpecialAdministrativeRegionSFSubsurfaceflowSSWStormsurgewarning

13TDSTotaldissolvedsolidsTFTidalflowTIPTransborderIndustrialParkTNTotalnitrogenTOCTotalorganiccarbonTPTotalphosphorusTSTotalsolidsTSSTotalsuspendedsolidsUNUnitedNationsUSUnitedStatesUSAUnitedStatesofAmericaUVUltravioletVFVerticalflowVOCVolatileorganiccompoundsWWTPWastewatertreatmentplantWWTSWastewatertreatmentstation

141.INTRODUCTION1.1Wastewatermanagementincities1.1.1ImportanceofwastewatermanagementWaterisoneofthemostimportantresourcesprovidedbyourplanet,itsusageisincorporatedintooureverydaylives,rangingfromhousehold,agricultural,industrial,recreational,transportationtoelectricitygenerationpurposes.Itisthereforecrucialandourutmostresponsibilityasmankindtoproperlymanagewastewatergeneratedfromouractivitiestopreventenvironmentalpollutionandpromotecircularityofresources.Accordingtothe2017UnitedNationsWorldWaterDevelopmentReport,itislikelythatover80%ofglobalmunicipalandindustrialwastewaterisreleasedtotheenvironmentwithoutadequatetreatment(WWAP,2017).Untreatedmunicipalwastewatercontainsunsafelevelsofinorganicandorganicmaterialsthatleadstounsuitableconditionsformarinelivesanddisturbstheglobalnetworkofecosystembalance.Pollutantscanbeoriginatedfromseveralsources,suchasfoodandhouseholdrelatedproducts,faeces,fuel,pesticidesandherbicidesfromparksandevenindustrialby-products.Emergingcontaminantssuchasantibiotics,pharmaceuticalandpersonalcareproducts,microscopicplasticfibersreleasedfromclotheswashing,werealsofoundtoenterintomarineecologicalsystems;thiswouldcausebioaccumulationandinturncouldenterintothehumansystemandcauselong-termhealthimpacts(Tranetal.,2018).Inordertosustaingoodwaterqualityforhumanhealth,socialandeconomicdevelopmentandcontinuetomaintaintheecosystems’integrity,thereisaneedtoensurethesustainableuseofwaterresourcesbytreatingthemproperlyaftertheir“usagelifetime”,effectivelychangingthelinearsystemintoacircularsystem.Instead

15ofhavinga“pipe-it-away”mindset,therecognitionofwastewaterasavaluableresourceshouldbeestablished.Safelymanagedwastewaterisanaffordableandsustainablesourceofwater,energyandnutrients(WWAP,2017).By2050,70%oftheglobalpopulationisestimatedtobelivingincitiesascomparedto50%in2014.Furthermore,itisprojectedthatthenumberofmegacities(morethan10millioninhabitants)willrisefrom33in2018to43in2030.Thepressureonwastewatermanagementwouldbeevenhigherbythen,asmostcitiesdonothaveadequateinfrastructuretotreatwatersustainably(UN,2018;UNDESA,2014).TheEuropeanCommissionhasrecentlyadoptedanewCircularEconomyActionPlanensuring,amongotherissues,thattheresourcesusedarekeptintheEUeconomyforaslongaspossible.Also,alignedwithanintegratednutrientmanagementplan,directivesonwastewatertreatmentandsewagesludgewillbereviewedandnaturalmeansofnutrientremovalwillbeassessed(EC,2020).WatermanagementrequiresaglobaleffortthatisimplicitintheresolutionadoptedbyUnitedNationsGeneralAssemblywhichconcernsthe“InternationalDecade(2018–2028)forAction–WaterforSustainableDevelopment”tohelpputagreaterfocusontheefficientwaterusageatalllevels,takingintoaccountthewater,food,energy,environmentnexus;andhighlightingtheimportanceoftheengagementofallrelevantstakeholders(UnitedNations,2016).Accordingtothe“DevelopmentPlanoftheGuangdong-HongKong-MacaoGreaterBayArea”,Macaoshouldalsoadoptaninnovative,greenandlow-carbondevelopmentmodelbyreducingenergyandmaterialconsumption,recyclingresourcesandimplementingwaterconservationactions.1.1.2ConventionalwastewatertreatmentplantinurbanareasInurbanareas,twotypesofsewersystemaremainlyadopted:separatesanitaryandstormwatersewersystemandcombinedsewersystem.Intheseparatesystem,

16stormwaterduringwetweatherwillbedrainedseparatelyfromthedomesticwastewaterusuallyintoawaterbody(e.g.sea,river);inthecombinedsystem,stormwaterwouldbedrainedintothesamesystemasdomesticwastewaterintoatreatmentplant.Traditionally,inacombinedsewersystem,stormwatermixedwithrawsewagewouldoverflowthesystemandwouldbedischargedintothenearestwaterbodywithoutpropertreatment(ManninaandViviani,2009).Untreateddomesticwastewaterisclassifiedintolow,mediumandhighstrengthcategories,accordingtothecontaminantconcentration.Contaminantsaremainlymeasuredintheformof:totalsolids(TS),totaldissolvedsolids(TDS),totalsuspendedsolids(TSS),biochemicaloxygendemand(BOD),totalorganiccarbon(TOC),chemicaloxygendemand(COD),totalnitrogen(TN),totalphosphorus(TP),chlorides,sulphates,oilandgrease,volatileorganiccompounds(VOCs)andbacterialcountsuchastotalcoliformandfecalcoliform.Thesemeasurementsarelikelytovaryaccordingtoseasons,infiltrationfromgroundwaterthatflowsintothedrainagesystemcoulddecreaseBODandTSSconcentrationofwastewater.Inacombinedsewersystem,constituentvaluescanbegreatlyinfluencedbyprecipitation;forexample,BODandfecalcoliformbacteriaconcentrationsaretypicallylowduringastormeventasrunoffflowsarehigh(HathawayandHunt,2011).Nowadays,mostcitiesadopttheconventionalwastewatertreatmentprocessforwastewatermanagement.Conventionalwastewatertreatmentplant(WWTP)consistsofacombinationofphysical,chemicalandbiologicalprocessestoremovesolids,organicmatterandsometimes,nutrientsfromwastewater(Vanraesetal.,2016).AgeneralflowwithincreasinglevelsofwastewatertreatmentisshowninFig.1.TraditionaltreatmentstartswithPreliminaryfollowingbyPrimary,Secondary,Tertiaryand/orAdvancedlevel.Conventionalwastewatertreatmentplant,fordomestic

17wastewater,isatypeofcentralizedmanagementwhichconsistsofacollectionsystemandtreatmentinanoff-sitelocation,finaldisposalofthetreatedeffluentwouldusuallybefarfromthepointoforigin(WildererandSchreff,2000).Forinstance,onlyoneoutofsixwastewatertreatmentplantsinBerlinislocatedinthecity,theotherfiveareinthecity’ssurroundingarea(BerlinerWasserbetriebe,n.d.).NewYorkCity(NYC)has14wastewatertreatmentplantsthattogethertreat1.3billiongallonsofwastewaterdaily.Thesludgefromthewastewaterplantsisusedasavaluableresource(e.g.asfertilizersorsoilconditioners)becauseofitshighnutrientandorganiccontents(NYCDEP,n.d.).In2008,areportstatedthat1.75-2.0billionkWhofelectricitywasconsumedbythewastewatertreatmentsysteminNYCperyear(MalcolmPirnieEngineers,Inc,2008).Itwasestimatedthataround70%oftheelectricityusedinconventionalplantsareforaerationand20%forpumping(Pakenas,1995).OthermajorcitiessuchasLondon,Berlin,HongKongandTokyoalsoemploytheconventionalapproachtotreattheirwastewater,withmostlysimilarmechanicalsystemsforPrimaryandSecondarytreatment(BerlinerWasserbetriebe,2020;DrainageServiceDepartment,2020;Thameswater,2020).Fig.1:Simplifiedschemeofaconventionalwastewatertreatmentplant(Source:adaptedfromVanraesetal.,2016)

18Preliminarytreatmentinvolvestheremovalofwastewaterconstituentsuchasrags,sticksandlargematerialsthroughascreentopreventthemfromcausingoperationalblockages.Primarylevelremovesaportionofthesettleableorganicandinorganicsolidsthroughsedimentation.Someorganicnitrogenandphosphorusthatareassociatedwithsolidswouldalsoberemoved.ThePrimarysludge,inconventionaltreatmentplants,ismostcommonlytreatedbiologicallyinananaerobicdigester.SecondaryTreatmentinvolvesremovingtheresidualorganicsandsuspendedsolidsthroughaerobicbiologicaltreatment.Microorganismsthatconsumeoxygenarepresentandwillmetabolizetheorganicmatterinthewastewater,multiplyandproduceinorganicendproducts.Dependingontheamountoforganicmatterinthewastewater,differentprocessesareadopted.Activatedsludgeprocessiscommonlyusedwherebythecontentscontainingmicroorganismsandwastewateraremixedvigorously,sometimescompressedairisinjectedtoincreasetherateofreaction.Aftertheprocess,microorganismsareseparatedoutfromtheliquidbysedimentation,asmallportionofthesemicroorganismswillberecirculatedbackintothesystemtomaintainthedigestionrate.TheSecondarysludgecouldbemixedwiththePrimaryonesforsubsequentanaerobicdigestion.Tertiarytreatmentremovesconstituentssuchasnitrogen,phosphorus,heavymetalsorotherwastesthatcannotberemovedintheSecondarytreatmentlevel.Sometreatmentplantsusespecialisedfilterssuchasmembranefilters,otherswouldcreatemultiplezones,bothaerobicandanoxicfordifferentbacteriatoreactwithphosphorusandnitrogen(usuallyinammoniaform)andinturnremovethesecontaminants(Vanraesetal.,2016).Disinfectionstepisoftenincludedintertiarytreatment,UVradiationisoftenusedtodisinfectthewater.Filtrationwouldalsotakeplacetoremoveanyremainingsuspendedsolids.Tertiarytreatmentvarieswithdifferentwastewatertreatmentplantsasitdependsonvariousfactorssuchasland

19availability,wastewatertype,cost,publicproximity,dischargepermits(Burtonetal.,2003).Sludgeproducedintheprocessisusuallydewatered,thentheleftoversludgecakecanbeincineratedordumpedinlandfill.Insomecases,itispossibleforvalorizationofthesludgethroughcomposting(Tarragoetal.,2017).Researchisbeingcarriedouttooptimizetheprocessestowardssustainablewastewatertreatment(e.g.Manninaetal.,2018).1.1.3AdvantagesandlimitationsofconventionalwastewatertreatmentIntermsofadvantagestheconventionalapproachgenerallycharacterizesbyhavinghightreatmentefficiencieswithrelativelylowlandrequirements.Asitreliesheavilyonmachinery,thereishighflexibilityinoperatingconditionsandprocessesarehighlycontrolled.Effluentqualityisthereforemoreconsistent.Mainlimitationsattributedtotheconventionalapproacharetherequirementofmultiplemechanicalparts,whichconsumelargeamountsofenergyannuallyandthereforearesusceptibletopowercuts.Skilledlaborforcehastobepresenteddailyforoperationandhighmaintenanceisrequired.Therefore,thistreatmentplantapproachisnotthemostadequatemethodfordevelopingcountriesassomeruralareasonlyhavelimitedelectricitysupply,limitedmechanicalpartsforreplacement,andtherequirementofextensivesewerpipingsystemsmightnotbesupported.Thisaddstotheinitialconstructioncost.Conventionalapproachoftenhasalargeimpactontheenvironment(Garfíetal.,2017)whichwillbediscussedintheLifecycleassessmentsinsection3.2.1.1.4Futuretrendsforwatermanagement:CircularSustainabilityTheInternationalWaterAssociationhaslaunchedthePrinciplesofWaterWiseCitiesin2016toencouragetheshiftincurrentwastewatermanagementparadigmtomakecitiesmoreresilientandwaterusagemoresustainable.Insteadofusingascattered

20urbanwatermanagementapproach,thePrinciplesprovideguidelinesfordecisionmakerstomanagewatercollectivelyfromdesigntowaterregeneration(IWA,2016).Thereisanurgentneedtoincreaseefficiencyinurbanwatermanagementtofollowacirculareconomictrend.Intoday’sestablishedlineareconomicsystems,atake-make-consume-throwawaypatternisadopted.Theend“waste”isoftennotrecoveredastheprocessofdesign,productionanddisposalisnotdesignedforwastestobeconvertedbackasmaterials.Forthepurposeofthepresentthesis,“CircularSustainability"referstoasustainablecirculareconomy,inwhichresourcesareextractedandreusedinacircularmannerthatensuressustainability.Watershouldbemanagedinacircularsustainabilitywaywherebysustainableresourcesareconsumedandlimitedsecondarywasteisproduced.Stefanakis(2019)hashighlighted5Rsruleswhichencompassi)waterlossreductionandpromotionofwaterefficiency,ii)waterreuse,iii)recyclingwaterresourcesandwastewater,iv)restorationofwatertospecificqualitytowhereitwastakenfrom,andv)recoveringresourcesoutofwastewater.TheinclusionofNature-BasedSolutions(NBS)offermultiplewater-relatedbenefitsandoftenhelptoaddresswaterquantity,qualityandriskssimultaneously.Thecombinationofgreenandgreyinfrastructurescanleadtocostsavingsandoptimizationofthewatertreatmentsystems(WWAP,2018).NBSare“inspiredandsupportedbynatureandmimicnaturalprocessestocontributetotheimprovedmanagementofwater.AnNBScaninvolveconservingorrehabilitatingnaturalecosystemsand/ortheenhancementorcreationofnaturalprocessesinmodifiedorartificialecosystems”(WWAP,2018).Intheurbancontext,constructedwetlandsareaninterestingNBStosupportastrategyofresource-orientedwatermanagement(Masietal.,2018;Stefanakis,2019).

211.2.Constructedwetlandsforwastewatermanagement1.2.1.ConstructedwetlandtechnologyNaturalwaterdepurationoccursaswaterflowsthroughrivers,lakesandwetlands.Wateralsoinfiltratesintotheground,chemicalsareabsorbedandadsorbedbysoilparticlesandorganicmatter;livingmicroorganismswillalsouptakesomeofthenutrientsanddecomposetheminsoil.Thisisnature'swayofpurifyingwater,usingcomponentswithintheecosystemtopurifywaterandultimatelycirculateresourcessustainably(Balkeetal.,2008).However,duetoincreaseinurbanizationandconsequentimpermeabilization,watercycledynamicsisoftencompromised(McGrane,2016).Decentralisedwastewatertreatmentsystemsthatmimicnaturalpurificationprocesseswouldbeaninterestingoptiontoconsiderastheyoftenhavecomparativelylessimpactontheenvironmentthanconventionalprocesses(Garfíetal.,2017;Rouxetal.,2010).Somedecentralisedexamplesincludeoxidationponds,anaerobicponds,facultativepondsandconstructedwetlands(Mahmoodetal.,2013).Naturalwetlandsystemisoneofnature’swaystotransformandremovewaterpollutantsthroughaseriesofphysical,chemicalandbiologicalprocesses.Wetlandshavesurfaceornear-surfacewaterwhichcouldfluctuateseasonally.Waterentersthewetlandandissloweddownbyexistingvegetationandbythewiderwaterbodyvolume.Sedimentationofsolidsoccursduetoslowwaterflow,chemicalandbiologicalreactionsoccurnaturallybythediversecommunityofmicroorganisms,therhizosphereandwithinthesoilaswaterseepsintotheground.Wecanthereforeexploitthesenaturalideasandmimicsimilarprocessesbydesigninghuman-madewetlandsystems,oftenknownasConstructedWetlands(CW).CWsystemscanbedesigned,andfunctionscanbeenhancedtotreatdifferent

22wastewatersuchasstormwaterrunoff,domesticwastewater,agriculturalwastewater,leachates,industrialwastewaterandminedrainage(KadlecandWallace,2009).Dependingonthewastewatercharacteristicsanddischargepermits,theycanbeastand-alonetreatmentsystem,ortheycanbeacomponentinatreatmentsequence.Forexample,anumberofdecentralisedCWshavelongbeenestablishedinEuropeandNorthAmericafortreatingsmallvolumesofdomesticoragriculturalwastewater(Vymazal,2011a).InNimir,Oman,thereisaCWwhichisthelargestindustrialCWsystemfortreatingindustrialeffluentfromitsoilexplorationactivities(Stefanaki,2018).AnetworkofurbanwetlandswasalsobuiltinBeijing,Chinaformultiplepurposessuchaswaterregenerationandfloodcontrol(Jiaetal.,2011).Recently,CW’sadoptioninsuburbanareashasincreasedgloballyduetoclimatechangeandtherealizationoftheirmultipleecosystemservices(Masietal.,2018).However,CWisstillrarelyintegratedintocitycentresofurbanareasaslargesurfaceareaisneededforgoodtreatmentefficiency.1.2.2AdvantagesandlimitationsofconstructedwetlandsCWsarerelativelyeasytobuild,lessexpensivethanconventionalwastewatertreatmentwithlowoperationandmaintenancerequired.CWcanbebuilttofitharmoniouslytothesurroundingenvironmentwithhigheraestheticvaluethantheconventionalapproach(Stefanakis,2019).However,asCWsareopenair-exposed,theirperformancemaybeaffectedbychangingclimate(e.g.extremerainfall,typhoons)andthereforemaybelessconsistentthanconventionaltreatment.ConditionsinCWarelesscontrolled,thebiologicalcomponentsaresensitivetotoxicchemicalsandtroubleshootingwillbelongerthanautomatedconventionalapproach.OneofthebiggestlimitationsofCWisthatthesystemgenerallyrequireslargerlandareascomparedtoconventionaltechnology,areaswithlandscarcityandhighlandeconomic

23valuewillfacethebiggestchallengeinimplementation.Nevertheless,recentstudies(IlyasandMasih,2017)haveshownthatitispossibletoreduceCWlandrequirement(hereinreferredtoasfootprint)whilemaintainingitspollutantremovalefficiencybyemployinginnovativedesignsthatwillbeexploredinlaterchaptersofthisthesis.TosupporttheCWimplementationandtoevaluateitspositiveandnegativeimpacts,aLifeCycleAssessment(LCA)canbecarriedout.AsreferredbyGkikaetal.(2015),thismethodologyallowsustoassesstheenvironmentalimpactsassociatedwithallstagesofaproduct’sorstructure’slife;theauthorshaveassessedtheenvironmentalfootprintofaCWthroughLCAwhichcomprisesthequantificationofboththeconstructionandoperationphasesoftheCW(Gkikaetal,2015).1.2.3.ComponentsofconstructedwetlandsAconstructedwetlandsystemconsistsofadesignedbasinthatholds,ingeneral,water,asubstrateandmostcommonly,vascularplants.Thesecomponentscanbemanipulatedaccordingtodesiredwaterqualityandlandscapeprofile,whileothercomponentssuchasmicroorganismsandaquaticinvertebratesdevelopnaturally(KadlecandWallace,2009).Water:Aswaterseepsintotheground,animpermeablelinerhastobelaiddownatthebottomofthebasin.ShapeofthelandsurfaceshouldbeutilizedasgravitationalforceandcanbeexploitedforwatertoflowacrosstheCW.HydrologyisthemostimportantdesignfactorinCWasitcontrolshydraulicretentiontime,andinturn,timeavailableforwaterpollutantstoberemoved.Hence,thesizeofCWhastobewellthoughtoutasitdeterminesthevolumeofwaterthattheCWcanhold;thehigherthesurfacearea,thelargertheinteractionisbetweenthewatersurfaceandtheatmospherethroughrainfallandevapotranspiration(Davis,1995).

24Substrate:Abovetheimpermeablelineristhesubstratelayer.Commonlyusedsubstratesincludeexpandedclay,sandandgravel(Davis,1995).Substrateshavetobechosencarefullyastheirsizeandporosityaffectswaterpermeabilityandsupportadiversecommunityofmicroorganisms,allowchemicalandbiologicaltransformationstotakeplaceandhenceadsorbresultingcontaminants.Litterandsedimentfromsedimentationwillformontopofthesubstrates,theyactasasourceofcarbonwhichdrivesimportantbiologicalprocesses.Vegetation:VegetationinCWhastobetoleranttosaturatedconditions.Macrophytesareaquaticplantsthatgrowinornearwater,theycouldbefree-floatingplants,floating-leavedplants,emergentandsubmergentplants.Non-vascularplantssuchasalgaewillphotosynthesizeandincreasethedissolvedoxygencontent,whilevascularplantswillstabilizethesubstrate,slowwatervelocities,takeupnutrientsandtheirstemandrootsprovidesitesformicrobialattachment.Whentheseplantsdecay,litterwillbecreatedandcontributetothecarbonsourcewithintheCW.EmergentvegetationisusuallyplantedintheCW.SomecommonemergentvegetationusedinCWincludebulrushes,cattailsandreeds(Davis,1995).Microorganisms:SomeoftheorganismsinhabitingCWarebacteria,yeasts,fungi,protozoaandrindalgae.Theyarethemaincomponentthatdrivesredoxreactionswithinthesubstrate,andsometimesproduceinsolublesubstancesfromcontaminantreactions.Bothaerobicandanaerobicreactionstakeplace,dependingonthetypeofCWconsidered,andconsequentlyavailabledissolvedoxygen(Davis,1995).Animals:CWsmostlysupportinvertebratessuchasinsectsandworms.Theyfragmentdetritusandconsumeorganicmatter.Aquaticlarvaeofmanyinsectsconsumeorganicmatterduringthelarvalstageaswell.AsCWmatures,othervertebratessuchasmuskratsarelikelytobeattracted(Davis,1995).

251.2.4.FunctionandservicesCWs,besideswatertreatment,providemultipleecosystemservices.Ecosystemservicesarebothdirectandindirectcontributionsfromtheecosystemstohumanwell-being,thoseprovidedbyCWsinclude:waterreuseforirrigation,biomassforenergyproduction,airpurification,floodresilience,biodiversityenhancement,ornamentalplantsforaestheticvaluesandanurbaneducationalsite(Calheirosetal.,2018;MEA,2005).CWsmaybeconsideredforapplicationinthreemainareas(Stefanakis,2019):Wastewatertreatment:Thisisthemostusedapplicationworldwide.Chemicalandbiologicalreactionswilltakeplaceonthesurfacesofplants,substrateandlitter,withmicroorganismstransformingordecomposingthepollutants.Vascularplantswillalsouptakeacomparativelysmallportionofthenutrients.HabitatCreation:Thepresenceofwaterandvegetationcouldprovidesuitablehabitatforwildlifespeciesespeciallyforbirdsandinsects.Theycouldpossiblycreateanewecosystemtopreviouslydamagedonebyprovidingasourceoffoodandfiber.ThishabitatcreationcouldalsobeusedaspublicrecreationalareasifCWsareproperlydesignedtoallowpublicaccess.Multiplefloweringvegetationcanalsobeplantedtoenhancelandscapeaesthetics.FloodControl:Duringstormseason,CWcanactasstormwaterstoragetoreceiverunoffandinturntreatpollutedstormwater.Theysignificantlyreducethewetdrainageflowespeciallyintheurbancombinedsewagesystem.

261.2.5.ClassificationofconstructedwetlandsAccordingtoStefanakis(2019)therearemainlythreetypesofconstructedwetlandbasedontheflowpath(Fig.2):FreeWaterSurface(FWS),SubsurfaceFlow(SF)wetlandsandFloatingTreatmentWetlands(FTW).SubsurfaceFlowisfurtherdividedintoHorizontal(HFCW)andVerticalFlowConstructedWetland(VFCW).Acombinationofdifferentwetlandsisknownashybridsystems(HCW)(Fig.2).FWSsystemscontainawatercolumnof10-50cmaboveathinsubstratelayer;vegetationcaneitherberootedemergentorsubmergentplants,ortheycanbefreefloatingplants.Thissystemlookslikenaturalmarshesandcancreatewildlifehabitat.Thewaterlayernearthesurfaceisusuallyaerobicwhiledissolvedoxygendecreasesindeeperlayers,beingmoreanaerobicjustabovethesubstrate.FWSsystemisrelativelyeasytomanageandrequireslowcapitaltosetup.However,ithasarelativelylargefootprintcomparedtoothersystems.Usuallyitisnotadequatetotreatheavilyloadedcontaminantwatersduetothesurfaceexposuretoopenairandpossibleinteractionwithhumansandwildlife(Stefanakis,2019).SFsystemscontainporoussubstrateandwaterlevelismaintainedjustbelowthetopofthesubstrate.Rootsofvegetationpenetratetothebottomofthebed.Poroussubstrateprovidessurfaceareaforwatercontactandmicrobialattachment.SFsystemsarealsobetteratcoldtolerance;aswaterflowsbeneaththesubstrate,pestandodorproblemsarealsogreatlyminimized.However,itismoreexpensivetoconstructwithhighermaintenanceandrepaircoststhanFWS.Problemslikeclogging(duetoimproperdischargeofsolids)andunintendedsurfaceflow(duetohighornonuniformflow)couldalsoariseduetothesystem’snature(Stefanakis,2019).

27Wastewaterpretreatmentmaybenecessaryiftherearehighsolidsloadingthatleadstocloggingproblems.Pretreatmentmainlyinvolvesasedimentationtankand,insomecases,followedbyanaerationtanktoimprovenitrification(Bosaketal.,2016).Ifthewastewatercompositionishighlyheterogeneous,wastewaterequalizationphasemightbenecessary.Equalizationtankscouldbesetuptomitigatechangesinflowrate,itwouldstoreexceedingdailyflowanddivertitifinflowislessthantheaveragedailyflow,thusallowingconstantinfluenttoflowintotheCWs(Manderso,2018).SFsystemscanbedividedintohorizontalflow(HFCW)andverticalflowCW(VFCW).Althoughtheflowisdifferent,thebasicmechanismsarethesame.HFCWiscontinuouslyfedfromonesideofthewetlandandthedepuratedwateriscollectedontheotherside.Substrateiswatersaturatedandthereforewaterandaircontacttimearelimitedwithsloweroxygentransfer.Hence,anoxicreactionsdominateasmostofthesubstratedoesnotcontainhighdissolvedoxygencontent;nitrificationprocessisgenerallyslow,renderingtheperformanceofoverallnitrogenremovaltobeslowtoodespitefastdenitrificationprocess.Ontheotherhand,VFCWissequentiallyfedthroughouttheentiresurfaceofthewetland,byintermittentflow,andthedepuratedwateriscollectedatthebottomoftheCW.Partsofthesubstratewillbesaturatedandpartsofitwillbeunsaturated,thismechanismincreasesairandwatercontactandinturnoxygentransfer.Nitrificationprocessisfasterwhiledenitrificationisrelativelyslow.Phosphorusremovalinbothsystemsissimilar(Menaetal.,2008).AHybridsystemofVFCWandHFCWtakesadvantageofbothsystemsandperformsbetterinnitrogenremoval(IlyasandMash,2017).Inaddition,floatingtreatmentwetlands(FTW)arefloatingplatformsplantedwithemergentvegetation.Theseartificialplatformsfloatandallowtherootsofthevegetationtopenetratetheplatformandspreadintothewater.Thesedenseroots

28providelargesurfaceareaforbiofilmtobeformed.Theyfunctionasasmallportablewetlandtotreatpollutedponds,lagoonsandevenoilspillssites(IISD,2020).Fig.2Classificationofthevarioustypesofconstructedwetlands(Source:Stefanakis,019).1.2.6.MechanismsofpollutantremovalinconstructedwetlandsInordertochoosetheoptimalcomponentswhendesigningaCW,wehavetofirstunderstandhowpollutantsareremoved.Assuggestedinsection1.2.5,SFsystemismoresuitablethanFWSwhenimplementedinurbanareasduetolimitedexposureofwastewatertopublic.Hence,onlypollutantremovalmechanismsofSFsystemwillbeintroducedhere.Suspendedsolidsremoval:ForSFwetlands,gravitationalsettlingandadsorptionofsuspendedsolidsontothelargesurfaceareaprovidedbythesubstratehelptoremovethemfromthewastewater.However,cloggingisamajorproblem,itreduceshydraulicconductivityofthemediaandreducesoverallpollutantefficiencyoftheCW(Norton,2014).Therefore,substratemediawiththeappropriateparticlesizeshouldbechosencarefully.

29Organicmatterremoval:Organicmattercontainshighlevelsofcarbonthatmicroorganismscanuseasenergysources.Aerobicmicroorganismsconsumeoxygenwhenbreakingdownorganicmatterforenergywhileanaerobicmicroorganismcanbreakdownorganicmatterandproducemethane.Oxygencanthereforebetherate-limitingfactorinorganicremovalefficiency.Biologicaloxygendemand(BOD)isusedtomeasuretheamountofoxygenrequiredformicroorganismstobreakdownorganics,whichinturnsuggesttheamountoforganicsthatareleftpresentinthewastewater.Chemicaloxygendemand(COD)isanotherindicativemeasureoftheamountofoxygenrequiredtooxidizeorganicmatterthroughchemicalreaction;oxidizablecompoundsthataretoxictomicroorganismswouldalsobedetectedbyCOD.Itisamorerapidtestandhenceoftenusedasameasureoforganicmatterwithinwastewatersamples.Inaddition,plantscanalsouptakeorganiccarbonfornutrientsandbiomass,buttheirroleinorganicmatterremovalissmallcomparedtomicroorganisms’break-down(Norton,2014).Nitrogenremoval:Nitrogeninwastewatereffluentisamajorproblemthatcontributestoeutrophicationasitpromotesexcessiveplantgrowthandconsequentlydepletionofoxygeninthewater,makingitanoxicforotherorganismstosurvive(Norton,2014).NitrogenismainlyremovedbyplantsuptakeorbrokendownbymicroorganismsinCW.Aswastewateroftenlacksoxygen,nitrogenismostlyavailableasammonium.MicroorganismssuchasNitrosospiraandNitrosomonascanfirstconvertammoniumintonitriteunderaerobicconditions,Nitrobacteriawillthenoxidizenitriteintonitrate.Whenoxygenisdepleted,denitrifyingbacteriacanconvertnitratetonitrogengasunderanoxicenvironment,completingtheremovalofnitrogenfromtheCWsystem.Acombinationofnitrificationanddenitrificationprocessesisthemajormechanismfornitrogenremoval.Asammoniaconcentrationishighandnitrate

30concentrationindomesticwastewaterisusuallylow,nitrificationprocessisoftentherate-limitingstepandoxygencouldbetherate-limitingfactorinthereaction(Norton,2014).Phosphorusremoval:Phosphorus(P)canalsocontributetoeutrophicationifpresentinhighconcentrations.RemovalofphosphorusappearstobemuchmorechallengingthannitrogeninCWasthemajormechanismdoesnotinvolvedirectmetabolicpathways(Norton,2014).ThepredominantformofPisorthophosphatewhichcanbeusedbyalgaeandmacrophytes,howeverPwillbereleasedbacktothewastewaterasvegetationdecomposes,hencenotofferingalong-timestorageofP.Adsorptionbysubstratemediaplaysamoremajorrolecomparedtovegetation.Theefficiencyinadsorptionhighlydependsondifferentsubstratematerials.Nevertheless,thesubstratemayhaveacertaincapacityofholdingP,oncethelimitisreached,efficiencyofP-removalwilldecline(Norton,2014).P-retentionisthereforethemaincriteriawhenselectingasubstrateinthisthesis.Pathogenremoval:Wastewatermaycontainvariousformsofbacteria,viruses,fungiandprotozoans.Numerousfactorssuchasnaturaldie-off,predation,inactivationcouldleadtodecreaseinpathogen;studiesthatweredonemainlyfocusedonfaecalindicatororganisms,littlestudiesweredoneonspecificmicroorganisms.Specificremovalmechanismthereforeremainsunclear(Alufasietal.,2017;WeberandLegge,2008).However,itwasfoundthathigherhydraulicretentiontimecorrelateswithhigherpathogenremovalefficiency(Karimetal.,2004).Metalsremoval:Metalssuchascadmium,mercuryandleadcanbefoundinindustrialwaterwhichcanaffecthumanhealthevenatlowconcentrationiftheygetintooursystemthroughtheecosystemnetwork.CWcanremovemetalsviaplantuptake,substrateadsorptionandprecipitation.Somemetalscouldformstrongcovalent

31bondswiththesurfaceofthesubstrateandcouldberelativelystableandlong-term,thismechanismissimilartoP-removal.Otherssuchasnickel,zinc,leadcaninteractwithsulfidesunderanaerobicconditionandprecipitateasinsolublecompounds(Norton,2014).1.2.7.ConstructedwetlandimplementationConstructedwetlandwasfirstimplementedtotreatwastewaterfromnearbypoint-sourceinareaswithlandavailability(Stefanakis,2019).Theirapplicationiswidelyacceptedasamoresuitablewayforwastewatertreatmentinruralareasandinlow-incomeregions(IlyasandMasih,2017;Wuetal.,2015a).Therealizationofthenegativeaspectsofconventionaltreatment,coupledwithclimatechangeandtheincreaseinglobalsustainabilityawareness,hasledtothegradualrecognitionofCWssustainabilityandtheirincreaseinadoptioninternationally(Masietal.,2018).However,CWtechnologyisnotreallydeeplyintegratedintotheurbanareasduetoonemainissue:landavailability.Whenlandisahigheconomiccostandexistingtechnologyisalreadylongestablishedtocurrentlycopewithwastewatertreatment,thereislikelytobelessincentivetointroduceCWinurbanareas.Nevertheless,theglobalneedforsustainabilitydevelopmenthaspushedrapidengineeringdevelopmentinthelast10-15years.Thisgapbetweenlandavailabilityandenergyconsumptionhasgraduallybeenpulledcloserbyintensifiedandaeratedwetlands(Stefanakis,2019)(Fig.3).

32Fig.3.Qualitativecomparisonofconstructedwetlandandconventionaltreatmentsystemintermsofarearequirementandenergyneeds.O&M:OperationandMaintenance.(Source:Stefanakis,2019).Itishoweverdifficulttocompletelyreplaceconventionaltreatment.TheimplementationofCWinurbanareaswillminimizetheincreasingpressureonexistingconventionalwastewaterplants.HowevercentralisedCWscouldnotcompletelyreplaceconventionalwastewaterplantswiththehighpopulationdensityinthecity.Therefore,aCWdecentralisedsystemcouldbeconsideredinstead,thissystemcouldtreatwastewaternearpointsourceandtreatedwatercanbeusedinnearbyfacilitiesorwithinthesamebuilding.Overallurbanpipingsystemscanbereduced.DecentralisedsubsurfaceflowCWswithinthecities,couldbeaestheticallypleasingbyincreasinggreenpatcheswithinthecity.DecentralisedCWinurbanareascouldreducerun-offandserveasurbanfloodcontrol,aconditionwhichisincreasinglyfrequentandseriousduetoclimatechange.Insteadofbuildingconcretefloodcontrolwhichpropagatestheproblemitself,decentralisedCWsystemisaveryattractivealternativethatprovidesseveralecosystemservices.AninterconnectedhubusingNature-basedsolutionsuchasoneillustratedinFigure4wouldimproveoverallwaterresiliencewithinacity(Masietal.,2018).

33Fig.4.Asustainablewatermanagementschemeinanurbansettlementwherenature-basedsolutionssuchasdecentralisedconstructedwetlands(CWs)areintegrated(Source:Masietal.,2018).1.2.8.DesigningaconstructedwetlandDuringthedesignstageoftheCWsystem,itisimportanttofollowsomebasicguidelines(Davis,1995;Dotroetal.,2017):●Designsshouldbekeptassimpleaspossible,avoidover-engineeringwithminimalrequiredmaintenance.●Designsshouldexploitnaturalenergiesandshouldbeasintegratedintotheenvironmentallandscapeaspossible.●Designsshouldconsiderextremeweatherandclimate.●Localmaterialsandvegetationshouldbeused.●Dischargestandardsshouldbedetermined.●Designsshouldconsidercircularsustainabilityandhowwatercouldbemanaged,integratedandreusedinanurbansettlement.

34ForCWimplementationinurbanareas,FWSwetlandsystemisnormallynotpreferredduetohighlandrequirement.SFCWsshouldbeconsideredinstead.OutofthetwoSFsubtypes(horizontalandvertical),VFCWrequireslesslandasitmainlyusesgravitytodrivethehydraulicflow.Thetreatmentefficiencyisdirectlyrelatedtothefiltermaterialused(Dotroetal.,2017).Finematerialsallowlongerretentionofwastewaterandinturn,higherpollutantremovalefficiencies,butthepotentialofcloggingincreases.Coarsematerialsallowhigherhydraulicloadingratebutarecompromisedbylessefficientpollutantremovalefficiency.SomecountriessuchastheU.S.,GermanyandDenmarkhavepublisheddimensioningguidelinesfordesigningVFCWthatmeetthenationaleffluentdischargelimit(Dotroetal.,2017).Ingeneral,guidelinesaddresstheneedofadrainagelayerofgravelatthebottomofthebedandatransitionlayertopreventthefiltrationlayer(mainsubstratemedia)frommigratingtothecoarselayerbelow.Itissuggestedtoincludeatoplayerofgraveltopreventerosionduringintermittentloadingandtopreventfreewateronthesurface.WhiletherearedifferentCWdimensioningguidelinesavailablefromdifferentcountries,westillneedtoemployacomprehensivetooltoestimatetheoverallpotentialenvironmentalimpactsexertedbydifferentCWs.ALifecycleassessment(LCA)canbeusedduringthedesignphasetochoosebetweendifferentwastewatertechnologies;itcanalsobeusedtoidentifywhichlife-cyclestageexertsthemostsignificantenvironmentalimpactsforaparticulardesignsothattailoredimprovementsandalterationscanbemade.LCAscanhelptodecidewhichtypeofCWhaslessimpactsontheenvironment.Forexample,Fuchsetal.(2011)haveperformedanLCAtocompareVFCWwithHFCWwastewatertreatment.TheauthorshavefoundthatVFCWwillcauselessenvironmentalimpacttreatingthesameamountofwastewater

35throughoutitslifecycleduetohighertreatmentefficiency,lowerGHGemissionsandsmallerfootprint.Therefore,thereisaneedtoextensivelyevaluatematerialsusedforsettingupaparticularCWdesigninagivenarea.MoreLCAsshouldbeperformedtocomparedifferentmaterialsandevendifferentCWdesignstoselectthemostsustainablewayofbuildingaCWandtoensureitssustainableoperation.ThepresentthesiswilladdressthisissuebyevaluatingdifferentexistingCWmaterialsanddesigns.ExistingLCAstudieswouldbereviewedinordertoselectmoresustainablematerialsforCWconstructionandtoselectaCWdesignwiththeleastenvironmentalimpactforimplementation.1.3.AimsandoutlineofthisthesisIntheprevioussections,CWswereintroducedanddiscussedasanalternativewastewatertreatmentsystem.MostoftheWWTPsintheworldaredesignedtomeettherequirementsoftheeffluentqualitywithoutconsideringenergydemands.Electricityinputinwastewatertreatmentplantsactuallyaccountsfor3%oftheworld'selectricityconsumption(Lietal.,2015).AlthoughCWprovidesmultipleecosystemservices,oneofthemajorlimitationsofCWisthattheytraditionallyrequirelargefootprint.Nevertheless,multiplecitiessuchasSanFranciscoandCopenhagenhavealreadyimplementedCWintheirsuburbanarea.CWs'relativelylargefootprinthaslimiteditsimplementationincitycentres(Smith,2009).Inthepresentthesis,anoptimizedCWsystemwillbeproposedtoimplementinMacaoSpecialAdministrativeRegion(SAR),China.MacaoSAR(hereinreferred

36toasMacao),with21,420populationperkm2,isthesecondmostdenselypopulatedarea/regionintheworld(UnitedNations,2019),CWimplementationwouldthereforebeinterestingandnovelinsuchadensearea.Firstly,mainmaterialsforsettinguptheCWsystemwillbereviewed,namelysubstrate,vegetationandlinerwhichareselectedtoensuresustainablewastewaterpollutantremovalandtoachievecircularsustainabilitywithinacity.Also,LCAstudieswouldbeextensivelyreviewedinordertocomparedifferentmaterials,CWsystemsandtheirvaryingenvironmentalimpacts,sothatmoreinformeddecisioncouldbemadewhendesigningforaCW.Asetofidealmaterialswillthenbeproposed.Inaddition,themainrate-limitingfactorsofaCWsystemanddifferentimprovementstrategieswillbeassessedtoreduceCWlandrequirement.Finally,aCWdesignwithrecoverablematerials,lowfootprint,goodtreatmentefficiencyandwiththeleastexpectedenvironmentalimpactswouldbeproposedforimplementationinMacao.ThenecessaryCWareawouldbecalculatedassecondarytreatment,andlocation(s)wouldbebrieflyidentifiedforimplementingasetofnovelconstructedwetland(s)systeminMacao.Theaimsofthisthesisaresummedupasfollows:Aim1.Reviewtheexistingmainmaterialsforconstructedwetlandsetup:Substrate,vegetationandliner.•Substratewillbeselectedfortheir:(1)P-removalcapacity,(2)P-recoveryandfilterrecyclabilityaftersubstratesaturation•Vegetationwillbemainlyselectedfortheir:(1)CODtoleranceinwastewater,(2)floodtolerance,(3)NitrogenPhyto-uptakeand(4)potentialasbioenergyfeedstockbasedonthecirculareconomyconcept.•Linerwillbeselectedfortheirimpermeabilitywithleastenvironmentalimpact.

37Aim2.ReviewexistingLCAstudiesforvariousenvironmentalimpactsofdifferentmaterialsandCWsystem.Aim3.Reviewrate-limitingfactorsinCWsystemandtheirimprovementstrategiesforfootprintreduction.Aim4.UsingMacaoSARasacasestudy,calculateanecessaryarearequiredtotreatthepost-primarytreatedwastewaterfromtheMacaoPeninsulaWWTP.Aim5.BasedonfindingsfromAim1to4,proposeamodelofCWfordomesticwastewatertreatmentinMacaowithselectedlowenvironmentalimpactmaterialsandintensifiedstrategiestocomplementexistingWWTP.Aim6.Identifylocation(s)forimplementingtheproposedCWmodelfromAim5inMacaoPeninsula.

382.METHODS2.1LiteraturereviewToachieveAim1and2statedinsection1.3,researchpapershavebeencollectedandreviewedtogiveanoverviewofCWconstructionmaterialsintermsoftheirpollutantremovalefficiency,theirlifecycleandtheabilitytoachievecircularsustainability.Forsubstrate,specialfocuswouldbegiventophosphorussincepreviousstudies(Haynes,2015;Vohlaetal.,2011)haveshownthatitisoneofthetoughestpollutantstoberemovedanditsremovalhighlydependsonadsorptiononsubstrateandlessonmicroorganismscomparedtonitrogenandorganicmatter(Norton,2014).Phosphorusremovalefficiencyshouldthereforebeoneofthemostimportantcriteriawhenchoosingasubstrate;theirrecoveryandsaturatedsubstraterecyclabilitywerealsoexplored.AsthefinalaimistoproposematerialforMacaoCW,filtermaterialsthatwouldbeavailableinurbanareaswerealsoexplored.ISIWebofScienceandScopusdatabasewereusedforthisreview.Themaintopicsusedfordatabasesearchwereasfollows:“phosphorusremoval”,“phosphorusretention”,“phosphorusadsorption”.Allofthesetermsweresearchedaddingthefollowingterms:“Constructedwetland(s)”,“filtermaterial(s)”.Bothterms“phosphorusremoval”and“Constructedwetland(s)”werealsosearchedaddingthefollowingterms:“filterrecyclability”,“phosphorusrecovery”,“circulareconomy”,“lifecycleassessment”and“urbanmaterial”.Ananalogoussearchwascarriedoutusingtheterms“phosphorous”and“phosphate(s)”insteadof“phosphorus”inthesewordcombinations.Forvegetationandliner,theterms:“vegetation”,“liner”wereeachsearchedwiththeterm“constructedwetland(s)”.Bothterms“vegetation”and“constructed

39wetland(s)”werealsosearchedaddingthefollowingterms:"biofuel",“bioenergyfeedstock”,“circulareconomy”,"Nphytouptake",“CODtolerance”,“floodtolerance”,“lifecycleassessment”.ForAim2,LCAstudieswerereviewedtogaininsightsintheenvironmentalimpactsofdifferentmaterials,componentsordifferentCWsystems.AsidefromtheabovesearchwiththeCWmaterials,bothterms“lifecycleassessment”and“constructedwetland(s)”weresearchedaddingtheterm:“biofuel”,“constructionphase“,“operationphase”and“intensification”.Atotalof93paperswereusedinourdiscussionsection,ofwhich54paperswereusedinourmaterialevaluation:25usedforfiltersubstrate,24forvegetationand5forliners.Papersexploringtheuseoflinermaterialsarelimitedandthereforeonly5wereused.Ofthe21LCApapersfoundwithinthe93papers,mostLCApapersfoundwererelatedtothecomparisonofWWTPandCW;inwhichonly8LCApaperswerefoundtocomparedifferentfiltermaterialsand2werefoundrelatedtocombinedCW-biofuelsystems.ToachieveAim3,existingtechnologieswereexploredforCWfootprintreductionforurbanCWapplication.Itisalsoimportanttomaintainahighlevelofpollutantremovalefficiencywiththefootprintreductionstrategies.Theterms“constructedwetland(s)”and“footprintreduction”weresearchedwiththefollowingterms:“intensification”,“greeninfrastructure”,“efficiency”,“performance”.ForCWintensification,atotalof18paperswereincludedinthediscussionandevaluation.However,theseintensificationstrategyandtechniqueswerenotcomparedagainsteachotherquantitatively,thereforeonlyqualitativediscussionwasdone.

402.2ConstructedwetlanddimensioningAsVFCWrequiresasmallerfootprintthanHFCWtotreatthesamevolumeofwastewater(Fuchsetal.,2011),intheory,VFCWwouldbemoresuitedinplaceswherelandisexpensiveandscarce.Therefore,toachieveAim4,onlythemethodologyofVFCWdimensioningwillbepresentedhere.ThetreatmentefficiencyofaVFCWpartiallydependsonthefiltermaterialused;mostoften,finermaterialwouldincreasetheretentiontimeofthewastewater,whichoftenincreasesremovalefficiency.Coarsematerialallowshigherhydraulicloadingrate(HLR)andlesscloggingbutwouldhavelowerremovalefficiencies;thiscouldpotentiallybeovercomebyincreasingthedepthofthemainlayer.AccordingtoGermanguideline,DWA(2017),themaximumHLRshouldnotexceed80lm-²·d(0.08m3m-2·d).LargerVFCW(>100P.E.)commonlyhasdividedsurfaceareaintosmallerones(cells)sothattheycanbeindependentlyloaded.TheGermanguideline(DWA,2017)recommendsVFCWtobedesignedwithaquarterofthetotalsurfaceinrestingphase(non-loadingphase).ForVFCWsinwarmerclimates,studieshaveshownthatlowersurfaceareaisrequiredforthesametreatmentefficiency(Langergraberetal.,2007),andthatHLRcanbeupto200mmd-1(Stefanakisetal.,2014).Intemperateclimates,amaximumorganicloadingrateof20gCODm²·dcanbeapplied.TheOLRsintertiarytreatmentcanbehigherandthusthesurfacearearequirementcanbereduced(DWA,2017).Tocalculatethenecessaryarearequiredtotreatthepost-primarytreatedwastewaterfromtheMacaoPeninsulaWWTP,asimpleVFCWdimensioningisshowninthefollowingstepsaccordingtoDotroetal.(2017):Step1:Defineinfluentflow(Qi),effluentconcentrationemissionstandard(dependsoncountry)andrawpollutantconcentrations.

41Step2:Designthethree-chamberseptictank.Asthedomesticwastewaterisprimary-treated.Thereisnoneedforseptictanksinthiscase.Step3:DesigntheVFCWDependingontheuseofsubstrate,differentsurfaceareacanbeestablishedby:1)Settingamaximumarealorganicloadingrate(gCODm-2d-1)(Dependsonlocalclimate)2)Calculatingorganicload(gCODd-1)3)Calculatingrequiredsurfacearea(m2)4)DecidingVFCWcellconfigurationbasedonsurfacearea(m)ForAim5and6,qualitativeanalysisandevaluationwillbediscussed.MapofMacaoPeninsulawithexistingopenspaceswillbeusedforidentifyinglocationsforCWimplementation.

423.RESULTSANDDISCUSSION3.1.Constructedwetlands’materialtowardscircularsustainabilityAsustainableCWhastohavelong-termtreatmentperformanceanditscontinualoperationshouldnotbeachievedattheexpenseoftheenvironment.Therefore,theconceptofcircularsustainabilityisexploredduringthesubstrate,vegetationandlinerselectionprocesswhendesigningaCW.3.1.1SubstrateselectionSubstrateisacrucialcomponentinalltypesofCWsasitaffectswaterpermeability,providessitesforbiochemicalandchemicalreactions,adsorbremovedpollutants,supportplantgrowthandbiofilmestablishment.Theselectionofoptimalsubstratesismainlydeterminedbytheirhydraulicpermeabilityandcapacityforadsorbingpollutants.PoorhydraulicconductivitywouldleadtocloggingandinturnreducetheperformanceofCW;lowadsorptionofsubstratescouldleadtonegativelong-termimpactinCW’sperformance(GorgoglioneandTorretta,2018).Itisrecommendedtouselocalmaterialsforsubstratestoreducelong-distancetransportation.Substratematerialalsoaccountsforhighproportionintheconstructioncost,thereforeselectionofappropriatesubstrate,botheconomicallyandfunctionally,isanimportantstepduringthedesigningstageofaCW(Wuetal.,2015b).Traditionally,commonlyusedsubstratesincludeexpandedclay,sand,gravel,rockandorganicmaterials(Davis,1995).Despitetheimportanceofsubstrateselection,studiesoffull-scaleCWsystemwerenotextensivelydonetoevaluatetheperformanceofdifferentsubstrates.Duringourliteratureresearch,batchandcolumntestswerealsoincludeddespitenotbeingafull-scaleCWsystemduetothelackoffull-scaleexperiments.

433.1.1.1Filtermaterialsforphosphorusremoval3.1.1.1.1Soil,sandandgravelTraditionally,localsoilwasoftenusedassubstratesinCW.Soilwasselectedfortheirorganiccontenttofuelplantgrowthandmicrobialactivity.However,soilgenerallyhaslowhydraulicconductivityandthereforecouldcauseoverlandflowandshort-circuitingofwastewaterbetweeninletandoutlet(PucherandLangergraber,2019).Densesoilsuchasclayandshales,shouldthereforebeavoidedasitlimitsrootpenetrationandhaslowhydraulicconductivity.Coarse-texturedsoilwasusedtoallowrootgrowthbuthaslowpollutantretentioncapacity.Loamysoilisabetteroptionasitissoftandfriable,allowingrootpenetrationwhilehavinghighpollutantretention.Sandandgravelsubstrateweregraduallyadoptedduetotheirseeminglymoresuitablehydraulicconductivitythansoil.AnexperimentwasconductedinLangenreichenbach,GermanybyNivalaetal.(2010)tocomparetheimpactofsubstratesizedifferenceontheoxygenconsumptionrate(OCR)ofaCW;theyhavecomparedOCRofasand-based(1-3mmsand)andgravel-based(4-8mmgravel)VFCWsystem.Sand-basedsystemwasfoundtodisplayhigherOCRsthangravel-basedsystemswhichindicatesthatthesizeofthemediaaffectsthechemicalreactionsinthetreatmentprocess.Butthegavel-basedverticalsystemhashigherNRemovalduetoalowOxygenenvironmentthatfavorsdenitrificationprocess(Nivalaetal.,2013a).Pwasfoundtobeoneofthehardestpollutantstoremovefromdomesticwastewaterandmainlyreliesonsubstrateadsorption.NumerousresearcheswereconductedtofindthebestfiltermaterialforP-removalfromwastewatertopreventfurthereutrophicationofwaterbodies(Hylanderetal.,2006;Lietal.,2006a;Vohlaetal.,2011).

44SandandgravelhavenotdisplacedaconsistentP-removalcapacityandgenerallyarepoorcandidatesintermsofsustainableP-removal(Wuetal.,2015c).P-sorptioncapacityofsomesandswasdepletedonlyafewmonthsinabatch-scalesystem(Ariasetal.,2001),whereasotherscouldlastuptoafewyears(Brixetal.,2001).Insands,Vohlaetal.(2011)havefoundthatcalcium(Ca)contentiscrucialintheirP-removalcapacity.InpH>6,precipitationoccursascalciumphosphates,butpHlowerthanthatmainlydependsonFeandAltoprecipitateasironandaluminumphosphates.Hence,withacidicwastewater,sandsubstrateswithhighFeandAlcontentarepreferred.Ina8-yearfull-scalesystemexperimentwithsandsubstrate,Vohlaetal.(2007)hasobtained,0.117gPkg-1retention(72%P-sorption).Forafull-scalegravel-basedCWexperiment,Korkusuzetal.(2005)hasonlyobtained4.33%ofP-removal.3.1.1.1.2DolomiteDolomiteisacommonsedimentaryrock-formingmineralwithachemicalcompositionofCalciummagnesiumcarbonate.Theycanbefoundthroughouttheworldinsedimentaryrockknownasdolostone,theycanbehundredstothousandsoffeetthickinmassivebeds.Karacaetal.(2004)haveuseddolomiterockwhichisabundantinTurkeyinabatchstudytotesttheirP-removalcapacity,theywerefoundtoobtainP-sorptionof9.7to52.9gPkg-1withanincreaseintheinitialphosphateconcentrationfrom10to60mg/Lfor20◦C.TheamountofphosphateadsorbedalsoincreasedwithincreasingpHfrom1to11.DolomiteadditionwascommoninanattempttoincreaseP-sorptionofsand(Pantetal.,2001;ProchaskaandZouboulis,2006)andotherman-madematerialssuchasinlightweightaggregates(LWA)toguaranteecertainlevelofP-sorption(JenssenandKrogstad,2003).

453.1.1.1.3ZeoliteNaturalzeolitesformwhenvolcanicrocksandashlayersreactwithalkalinegroundwater.Asnaturallyfoundzeolitecontainvaryingdegreesofminerals,thereforeitsapplicationassubstratewasreportedwithvaryingP-sorptioncapacity.Chenetal.(2006)haveuseddifferentzeoliteswithvaryingpHandobtainedonly0.01to0.05gPkg-1,whereasSakadevanandBavor(1998)haveobtainedP-retentionof2.15gPkg-1withinitialPconcentrationrangingfrom0to600mg/Lfor21◦C.3.1.1.1.4LightweightaggregatesLightweightaggregates(LWA)orlight-expandedclayaggregates(LECA®)werealsousedforP-removalsubstrateinCW.TheyhaveshowngoodwaterpermeabilityandP-sorptioncapacity(JenssenandKrogstad,2003;Vohlaetal.,2011).Inabatchscalestudy,JenssenandKrogstad(2003)hasobtainedP-retentionof12gPkg-1.Theyweremanufacturedbyputtingpalletizedclayaggregatesinarotarykilnat1200oC.Johansson(1997)hasconductedcolumnexperimentswithLECA®andfoundthatinitspureformisconsideredtobechemicallynon-reactive,butlimeadditionhasincreaseditsP-sorptioncapacity.EnhancementofP-sorptionofLECA®withaluminiumandironoxidecoatingswasdemonstratedbyYaghiandHartikainen(2013),desorptionofsorbedPwasalsoreducedbytheoxidecoatings.3.1.1.1.5TabletporousmaterialAnoveltabletporousmaterial(TPM)wasdevelopedwithKanumaclay,CaOandcornstarchwherebyporeswereformedundercalcinationat600oC.TPMhasdisplayedmaximumPadsorptionof4.39gPkg-1(96%)inabatchstudyperformedbyYangetal.(2013).P-recoveryofTPMwillbeexploredinthenextsection.

463.1.1.1.6RedmudRedmudisaby-productofaluminaproductionusingtheBayerprocess,aprocessthatisdevelopedbyBayertoproducealuminabyrefiningbauxite.Itwasshownfromabatch-scaleexperimentthatextraheatedredmudhasahighmaximalsorptioncapacityof345.5gPkg-1.UnheatedrawredmudwascalculatedtohaveaP-sorptioncapacityof113.9gPkg-1(Lietal.,2006b).Duetoitshighalkalinity,untreatedredmudcancauseseriousenvironmentalimpacts;therefore,usingthewidelyavailableredmudassubstratecouldbeagoodsolutionforbothproblems.However,apilotstudywouldhavetobeperformedtofurtherexploreitsfeasibility.3.1.1.1.7BiocharBiocharismadebythermalheatingoforganicmaterialinlimitedO2toproduceaC-basedresidue;enrichedbiocharisproducedbypreorpost-pyrolysisusingmineralssuchashematite(Fe2O3)whichallowsthemtobondwithorganicphases,forminghighconcentrationofexchangeablecationsforPsorptioninCW.A7-monthstudyhasshownthattheenrichedbiocharcanreducePtobelow2mgL-1withaverageinletPconcentrationof15mgL-1.ThePremovalefficiencywasabove94%(0.208gPkg-1)forthestudyperiod;Pinthebiocharhasincreasedby77%andNhasincreasedby36%(Boltonetal.,2019).IncreaseinbiocharsurfaceareaalsofacilitatesP-sorption,astudyhasfoundthatincreaseinpyrolysistemperaturehasthemostsignificanteffectinincreasingsurfaceareaofbiochar(Luaetal.,2004).3.1.1.1.8SlagSlagisaporousnon-metallicco-productproducedintheironandsteelindustry.P-sorptionrangeswerereportedrangingfrom0.1upto420gPkg-1,thelargescatterofP-sorptioncapacityismainlyduetoundefinedchemical,mineralogicalandphase

47compositionsoftheseslags(Vohlaetal.,2011).AsslaggenerallyhashighalkalinityandPisknowntobindtoCawithhighpH.Blastfurnaceslag’smechanismforP-removalwasinvestigatedbyJohanssonandGustafsoon(2000),CaconcentrationwasfoundtodecreaseasafunctionofP,forminghydroxyapatite.ThissuggeststhatCa-PprecipitationisthemajormechanismforP-removal.Korkusuzetal.(2005)havecomparedagravel-filledVFCWwithablastfurnacegranulatedironslag-filledVFCW;itwasreportedtheslag-basedVFCWhashigherremovalefficienciesforallthepollutants:TSS,COD,NH4+-N,TNandPO43--PandTP.Inparticular,PO43--PandTPremovalwas44timesand11timeshigherthanthegravel-basedVFCW.Nonetheless,Cameronetal.(2003)hasreportedthatresearchusingslagcolumnfiltersforphosphateremovalfromwastewaterhascontributedtohigheffluentpHof11.5.Therefore,pHofeffluentinfuturepilot-scaleslag-basedCWstudyhastobemeasuredandcorrectedifnecessary.3.1.1.1.9OilshaleashCombustionofoilshaleinthermalpowerplantsproducesashrichinCaOandCaSO4,ashismixedwithwaterinaratioof1:20duringitstransportationtowasteheaps.Duringthisprocess,oilshaleashishydratedtoformreactiveCa-minerals.Kaasiketal.(2008)havereportedaP-retentionof65gPkg-1(85%)withhydratedoilshaleashsediment,whileKoivetal.(2009)havereporteda71%P-removalfromdomesticwastewaterusinghydratedoilshaleashinahydraulicallysaturatedexperimentalfilterstudy.3.1.1.1.10CrushedautoclavedaeratedconcreteInurbanareaswhereindustrialactivitiesarenotpresent,anothermaterial,crushedautoclavedaeratedconcrete(CAAC)wassuggestedasapotentialfilter

48mediumforPremovalbyCa-Pprecipitation.CAAC,inwhich70%iscomposedofCaOandSiO2,isaby-productfromautoclavedaeratedconcreteproductionandfromthedemolitionofhousesbuiltwithautoclavedaeratedconcrete(AAC)(Castellaretal.,2019).AACisahighlyavailablebuildingmaterialusedworldwide.IthastobecrushedandsievedtoachievesuitableparticlesizedistributionforCWsubstrate;CAACwith2-4mmwasshowntoremovePwithanefficiencyof93-99%(57gPkg-1)inafieldpilot-scaleexperiment(RenmanandRenman,2012).Reusingwastematerialsorby-productssuchasCAACwouldintegratewastewatermanagementwithcirculareconomy.Mkumboetal.(2019)haveshownthatAACcanbeaddedtosoilsforphytoremediation;plantssurvivalrateinheavymetal-contaminatedsoilincreasedwithincreaseofAAC.Therefore,P-saturatedCAACislikelytofurtherincreaseplantgrowthduetotheadditionof“Pfertilizer”,PintheformofCa-Pwasshowntobeslow-releasingforplantuptake(Koivetal,2012).3.1.1.1.11AlumsludgeAnotherpotentialurbanmaterialforfiltermediaisalumsludge,aby-productfromdewateringofalumsludgecakeinconventionalwastewatertreatmentplantswherealuminiumsaltsareusedduringthecoagulationprocess.Asurbandensityishighandlandisexpensive,conventionalwastewatertreatmentplantsarenotlikelytobecompletelyreplacedbyCW,butinsteadcouldcomplementeachother.Alumsludgeisoftendisposedofinlandfillandincreaseslandpollution,usingitassubstratewouldthereforereducesolidwaste.Amulti-stageCWusingalumsludgefromlocalWWTPassubstratewasshowntoenhancesimultaneousremovalofPandorganicmatterfromwastewater.Thesystemcouldremove90.6%BOD5,71.8%COD,93%reactivePand97%solublereactiveP(Babatundeetal.,2010).

493.1.1.1.12MagnesiumslagparticleMagnesiumslag,aby-productgeneratedduringmagnesiumproduction,providesagoodsourceofmagnesium(Mg)whichcanreactwithNH4+andPO43-toformlowsolublestruvitecrystalline(MgNH4PO4.6H2O)andsimultaneouslyremovebothpollutantsfromdomesticwastewater.Magnesitebyproduct(MgCO3)wasshowntoremove90%ofP(Etteretal.,2011)and65-80%ofammoniainhumanurine(Lindetal.,2000)inacolumnexperiment.3.1.1.1.13FragmentedmoleanoslimestoneVohlaetal.(2011)hasshownthatthechemicalformofCamattersinP-removalcapacity.StudieswithlimestonewhichmainlyconsistsofCaCO3intheformofcalciteonlyachieved20%P-removalinthelimestonewetlandsystemasdemonstratedbyStrangandWarehamin2006.Opoka(bedrockmaterialfromSEPoland)mainlycontainsCaCO3(50%)andlowretentionof20%ofPwasdetected.ByheatingOpoka,CaCO3istransformedtoCaOandhasincreaseditsPretentiontoaverageP-removalof96-99%.AnotherP-sorptionstudyhasalsoshownthatPolonite®,acalcinatedformofopokawhichcontainshighcontentofCaO,retainsPten-foldmorethanopoka(Hylanderetal.,2006).FragmentedMoleanoslimestone(FML)isaconstructionby-productrockfromcivilconstructionactivities.AstudyconductedbyMateusetal.(2012)hasusedFMLfromPortugalasasubstrateinCWandwasabletoobtain61%ofP-removal;perhapsheatingitwouldfurtherincreaseitsP-removalcapacity,butheatingmightresultinapowderwithverylowhydraulicconductivity.Heatingwouldaddtotheenergyconsumptionwhichisnotrecommended.InadditiontoselectingmaterialswithgoodP-removalefficiency,wehavetoconsidertheproblemwithclogging.CloggingisamajorissueinSFwetlandsystem.A

50coupleofstrategieswillbementionedinsection3.3tohelpwithcloggingonthetoplayer.Forlowerlayers,asthemainmechanismofPsorptioninvolvesCaOdissolution,thenPprecipitationandcrystallization;thesmallerthematerialsize,thegreaterthesurfaceareaforCaOdissolution,leadingtoincreaseinpHandincreaseinCaCO3precipitation,crystallisationandfinallycementationofthesystem(Vohlaetal.,2011).Usingpre-filtertoremovesolubleCO2andorganiccarbonandusinglargersizemediacouldhelprelieveclogging.Pretentioncapacity,however,formostfiltermaterialreviewedbyVohlaetal.(2011),significantlydecreasesaftera5-yearperiodofapplication.PerhapsaseparatefilterunitcontainingreplaceablematerialwithhighPbindingcapacityshouldbeconsideredforlongerlifetimebydecreasingtheriskofcloggingandforquickermaintenance.This,however,dependsonanumberoffactorssuchaslandavailability.3.1.1.2PhosphorusrecoveryandfilterrecyclabilityfromsaturatedsubstrateAsfinitephosphorusrocksusedforPfertilizers’productionarenon-renewableandPisestimatedtobedepletedin50-100years(NesetandCordell,2012),Padsorbedinthefiltermaterialsshouldberecycledratherbeingdisposedintolandfill,otherwisepollutionproblemswouldbepostponedinsteadofbeingeliminatedandwewouldfaceachallengeinglobalfoodproductionifPisnotusedsustainably.PhosphorusenrichedfiltermaterialscanberecycledbackasafertilizerprovidedthatPisavailabletoplantsandtheycantoleratethepathogenandtoxiccontentswithinthefilter.Biotoxicityoftheplantswouldhavetobetestediftheyarefoodcrops.Ifsaturatedsubstratecanbereusedwithoutreprocessing,overallenergyinputofthecircularsystemwillbereduced.Forsaturatedsubstratefromvariousmaterialstobeeffectivefertilizers,bioavailabilityofbothinorganicandorganicphosphateforplantuptakehastobeunderstood.Theorthophosphates,H2PO4-andHPO42-,aretheprimaryformsof

51phosphorustakenupbyplants;withH2PO4-asthepredominantformifpHislessthan7.0.However,certainorganicPcanalsobeavailablefordirectplantuptake,thoughthisislesscommon.InorganicPformscanbeclassifiedintothreecategories:(1)Plant-availablePthatisdissolvedinsoilreadilyforuptake,(2)SorbedPthatisattachedtoclay,Fe,Al,CAoxidesinsoilthatwouldbeslowlyreleasedforuptake,(3)MineralPsuchasstruviteisevenmoreslow-releasing(Prasad,n.d.).Phosphateadsorbedinsubstrateismostlyincategorytwoandthreeandthereforewouldbeslowreleasingifusedasfertilizers.Slow-releasingPfertilizers,however,havethepotentialtocontributetomoresustainablecropproductionsystems.TheyreducetheneedforcropstoallocatephotosynthatetosoilPacquisitionstrategiesandreducethefertilizerPrunoffinstormswhichcauseseutrophication.AlthoughhighlysolublePfertilizersleadtoearlycropgrowth,thePhosphorusquicklybecomesadsorbedandimmobilizedontosoilsurfaces;Pdemandinlatercropstagesisthennotsupported(Prasad,n.d.).P-recoverystudieswerenotfoundforsoil,sandandgravelinthisreview,thismightbeduetotheirpoorP-retentioncapacity.Inaddition,noresearchcanbefoundforLWA,FMLandzeoliteinthisliteraturereview.Forothermaterialssuchasredmudanddolomite,theyweretestedforPremoval,buttheseadsorbentscannotbewidelyappliedinpracticeduetodifficultiesinseparationfromaqueousphaseandsomecouldonlybeachievedathighcost(Yangetal.,2013).Theirdirectapplicationtosoilisyettobetested.ForTPM,itwasshownthat70.29%ofadsorbedPcanbetotallyrecoveredfor5cycleswhen0.2NHClwasusedaseluent.HighHClconcentrationhowevercouldreducethenumberofPadsorptionandrecoverycycles;0.5NHClwasshowntoreducePsorptioncapacityrapidlyinthe2ndcyclewhile0.2NHClonlydisplayedadecrease

52inthe4thcycle(Yangetal.,2013).AlthoughTPMdisplaysbothgoodadsorptionandPrecoveryquality,theenergyinputinitsproductionishighandKanumaclaymightnotbeavailablelocallywhichwouldincreasethetransportationcostandenergyinput.Theirdirectapplicationtosoilisyettobetested.TheaforementionedrecoverymethodwouldstillproducesolidwasteasTPMsubstratecouldonlyberecycleduptoaround4cyclesbeforeadsorptioncapacitystartstodroprapidly.Ideally,thesubstrateshouldbecheap,portablewithlittleornoenergyinputforitsproductionandproducenosecondarypollution.Usingindustrialby-productassubstratesthereforeallowsreuseofindustrialsolidwasteanddoesnotrequireadditionalenergyinput.Magnesiumslag-packedfilterispotentiallyagoodoptionthatfulfillstheabovecriteriainchoosingasubstrate.Struvitecrystalline(MgNH4PO4.6H2O)formedduringPandNremovalcouldpotentiallybeanexcellentslow-releasefertilizer;Ryuetal.(2012)hasevaluatedtheenvironmentalsafetyandreliabilityofusingstruviteprecipitationasdirectfertilizerandresultshaveshownthatstruvitefromsemiconductorwastewaterpromotedcabbagegrowthmorefavorablythancommercialPfertilizer,nohazardelementssuchasCu,Cd,As,PbandNiweredetectedinthecabbagebiomass.AnotherstudyfromFanetal.(2018)showedthatMgslag-basedfertilizerpromotedgrowthofmaizeplants,whileheavymetalconcentrationinmaizeislowerthanthelimitofestimateddailyintake,indicatingthatconsumptionofmaizegrainislowriskandwouldnotcausenon-carcinogenicrisks.However,struvite’slowsolubilitymeansthatearlycropgrowthisnotsupported.Talboysetal.(2016)hasshownthatonly9%ofstruvitehaddissolvedat36daysinapotexperiment,butat90days,26%haddissolvedandspringwheatwasgrowntograinharvest.Amixtureofslow-releasingPsubstratesuchasstruviteandreadilysolubleP

53seemstobepromisingandmoresustainable,theysupportearlyandlaterstageofcropdevelopment;residualPfromtheslow-releasingfertilizerwouldalsobeavailableforsubsequentcrops.Nonetheless,thelowsolubilityofP-enrichedsubstrateintheformofstruviteandhydroxyapatitecouldbeenhancedbymixingwithphosphate-solubilizingbacteria(PSB)andphosphate-solubilizingfungi(PSF);thesemicrobesproduceorganicacidsthatcandissolvehydroxyapatiteandstruvite,makingPsolubleforplantuptake(De-BashanandBashan,2004).Arangeoffertilizersfromslowtofast-releasingcouldbedevelopedbymixingsubstrateswithdifferentchemicalcompositionandwithPSBandPSFthatcouldspeedupPsolubilization,dependingonthestageofcropdevelopment.AnovelMg-packedwetlandfilterwasproposedbyTangetal.(2017)toeffectivelyremovestruvite(Fig.5);theaccumulatedstruviteprecipitationcanbeseparatedthroughsieveandoscillationbasethatareinstalledatthebottomoftheslaglayer.Separatedstruvitewouldbecollectedatthebottombygravity.Preliminarybatchesshowedthatapproximately2mmMgslagparticlescanrecover43.20–72.39%ofPandthepresenceof5–50mol/LNH4+contributedto11.71–29.11%enhancementofPrecoverymainlyduetostruviteprecipitation.ThisapproachiseasytooperateandreducestransportationcostfordownstreamPrecovery.UsingMgslag-packedfilterinCWthereforepotentiallyreducesindustrialsolidwasteandusingthesaturatedfilterasfertilizercausesminimaldischargeofsolidwaste.

54Fig.5NovelapproachforstruviteprecipitationseparationinConstructedwetlandwithfilterpackedwithmagnesiumslagparticle,where(1)-influentzone,(2)-precipitationreactionandretentionzone,(3)-precipitationseparationzone,(4)-precipitationsedimentationzone,(5)-effluentzone,(6)-influentpipe,(7)-electromagneticflowmeter,(8)-demountablepipe,(9)-sieve,(10)-magnesiumslagparticle,(11)-perforatedwaterdistributionpipe,(12)-oscillationbase,(13)-electricalmudvalve,(14)-watercollectionweir,(15)-effluentpipe,(16)-phosphorusonlinemonitoringinstrument(Source:AdaptedfromTangetal.,2017).Forsaturatedbiocharsubstrate,itcanalsoberemovedforsoilamendment;nutrientcapturedfromeffluentusingbiocharhavebeenfoundtobeinabioavailableformforplantuptake(Streubel,2011;Taghizadeh-Toosietal.,2012).SeveralstudiesalsosuggestedthatthepercentageofPdesorptionfrombiocharincreaseswithincreasinginitialadsorbedPconcentrationonbiochar(Trazzietal.,2016).Therefore,saturatedbiocharislikelytohaveahighPdesorptionrateforplantuptake.Also,biocharactsasacarbonsinkandcansequestercarboninsoilforthousandsofyearslikecoalwhichcouldhelpmitigateglobalwarming.Byapplyingrestingordryingperiods,biocharcouldpotentiallyberegeneratedasstudyhasshownthatfourweeksofrestingsteelslagoxidisesFe(II)toFe(III)oxides,increasingP-sorptioncapacity(Drizoetal.,2002).Therefore,aftersaturation,nutrientscouldbedesorbedandremovedfrombiocharforfertilizersandbiocharcouldpotentiallybereusedassubstrateinCWorsimplyactasacarbonsink.Although,furtherstudiesshouldbeundertakentotestthispossibility.

55Blastfurnaceslags,anindustrialby-product,wasproventoadsorbPefficientlybyforminghydroxyapatite.Thesaturatedblastfurnaceslagscouldpotentiallybeappliedasfertilizerdirectly.However,hydroxyapatitehaslow-solubilityandcannotbeeasilytakenupbyplants(JohanssonandGustafsoon,2000).Nevertheless,inapotexperimentconductedbyHylanderetal.(2006)wherebarleywascultivated;amorphousandcrystallineblastfurnaceslagswereshowntohavethebestfertilizereffectivenesswhencomparedtofourotherP-enrichedfiltermaterials(sand,opoka,limestone,Polonite®).Barleygrowninthetwotypesofslag,withfinegrainsizeof0.25-4mm,yieldedthehighestdrymatter,withrelativeeffectivenessofaround74%whencomparedtothepotfertilizedwithstandardfertilizer.ThisshowsthatPretainedbyslagscouldbeusedbyplantstosomeextent.Mixtureofhydratedcalcium-richoilshaleashandwell-mineralisedpeatwasshowntoalsohavegoodfertilizingefficiencyinpotexperimentswithsilverbirchseedlings(Koivetal.,2012).Pinpeatismainlyinphysicallyadsorbedplant-availableformwhereasPisimmobilisedasastableCa-Pforminoil-shaleash,slowlyreleasingPtotheenvironment.Also,PboundtoCawasshowntobemoreavailabletoplantsthanPsorbedwithAlorFeinastudyconductedwithcrystallineblastfurnaceslagbyHylanderandSiman(2001);PsorbedtothecrystallineslagwasdeliveredtothebarleyplantsmoreefficientlythanPaddedinK2HPO4fertiliser.ForCAAC,aby-productconstructionmaterialthatislikelytofoundinurbanareas,Mkumboetal.(2019)haveshownthatplantsurvivalrateinheavymetalcontaminatedsoilincreasedwithincreasedadditionofAAC.AACamendmentsprovideextrasorptionsitesandthereforereducemetalbioavailability,itraisessoilpHandfavorsmetalionprecipitation,reducingitsmetaltoxicitytoplants.Therefore,thisstudysuggeststhatAACcouldbeusedasphytoremediationofheavymetal

56contaminatedsoilandenhanceplantgrowth.Inaddition,asthedominantmineralinAACistobermorite,astudyhasshownthatbyusingtobermorite-richwastecompoundstorecoverPfromwater,theproductgeneratedhasmettherequirementsoftheP-industryasaphosphaterocksubstitute(Bergetal.,2006).Therefore,saturatedCAACwithadsorbedPislikelytofurtherincreaseplantsurvivalastheycouldactasfertilizer.However,furtherstudieswouldhavetobeperformedtoconfirmthistheory.Alumsludgementionedpreviously,thoughhasgoodP-sorptionandislikelytobeavailablewithinurbanareas,wasshowntohavelimitedabilitytoprovidesufficientPforplantproductionwithin2yearsafteritsapplication.SoilanalysisindicatedthatformsofPinalumsludge-amendedsoilmightnotbeavailableforcropuptake(Rigbyetal.,2013).Zhaoetal.(2009)hastesteddifferentacidsandbasesforPdesorptionfromP-saturatedalumsludgefromCWandfoundthatH2SO4seemstobeanefficientandcost-effectivereagentforPrecovering;itiscapableofachievingmaximum98.2%Precovery.However,furtherexperimentswouldhavetobeconductedtoexploreitsapplicabilityasPfertilizersforcrops.ConcernswerealsoraisedfortheapplicationofalumsludgeasaluminiumsulfatethatismostoftenusedinWWTPcouldincreasetoxicionsconcentrationintheenvironment(Kluczkaetal.,2017).3.1.2VegetationselectionVegetationinCWhastobetoleranttoaquaticandsaturatedconditions.Macrophytesarethereforecommoninwetlands,rootedvegetationshouldbeabletoadapttogrowinsubstratesthataremoderatelyorconstantlysaturated.Selectionofvegetationdependshighlyontheclimateofthelocation,thepollutantcontentofthewastewatertobetreatedandthusitstolerancetoit,thegrowthpotentialofroots,diseaseresistance,theexpectedaestheticvaluesoftheCWandmanagementrequirements.

57Mostcommonlyselectedmacrophytesisemergentspecies;examplesare:Typhaspp.(Typhaceae),Phragmitesspp.(Poaceae),Irisspp.(Iridaceae),Scirpusspp.(Cyperaceae),Juncusspp.(Juncaceae),andEleocharisspp.(Spikerush)(GorgoglioneandTorretta,2018).Nativespeciesarerecommendedduringselectiontoavoidintroducinginvasivespeciestothesurroundingecosystems.StudieshaveshownthatinSFwetlands,themostfrequentlyusedplantaroundtheglobeisP.australis,secondmostcommonisTyphaspp.(Vymazal,2011b).Intropicandsubtropiccountries,ornamentalspeciesarebettersupportedsuchasIrispseudacorus(YellowIris)(YanandXu,2014).However,plantsthataremostfrequentlyusedinCWisnotnecessarilysuitableforallwastewatertreatment.Wastewaterwithvarioustypesandlevelsofcontaminantsrequireplantsthatcantoleratethespecificrangeofconcentrations;P.australis,forexample,couldtoleratewastewaterwithCODconcentrationofupto2010mg/l(Calheirosetal.,2007).ForSFCWintropicalcountries,locallyavailablespeciesofPhragmites,Cyperus,BulrushandTyphahavebeenthemostcommonchoicetodate(Ranietal.,2011).Intropicandsubtropiccountries,ornamentalspeciesarebettersupported(YanandXu,2014).Forexample,inThailand,toincreaselocalpeople’sawarenessofwastewatertreatmentandtoincreasetheecstaticvalueofwetlands,Konnerupetal.(2009)havesuccessfullyusedfloweringplantsHelicorniaPsittacorumandCannaGeneralisinaHFCW.Threeornamentalplants(Cannaindica,CyperuspapyrusandHedychiumcoronarium)werecomparedintermsoftheirpollutantremovalcapabilityfordomesticwastewaterafter180days;allthreeplantswereabletoremovemorecontaminantsthantheunplantedsystem.However,C.indicahasbetterammoniaandTSSremovalsthanotherspeciesandalsoproducesthehighestnumberofflowersandshoots.Noneofthem

58exhibitwiltingorhavepestproblems(Zamoraetal.,2019).Astropicalcountriescouldbettersupportthegrowthofornamentalspecies,thesefloweringplantscouldperhapsbeusedforsettingupfloralgardensintropicalurbanareasandtoprovideaestheticvaluestoconcretejungles.Thecapacityofplantsinpollutantremovaldependsonthesystemconfiguration,hydraulicretentiontimes,loadingrates,wastewatertypesandclimaticconditions(Wuetal.,2015c).Itisthereforedifficulttoconcludefromstudieswhichplanthasthebestpollutantremovalefficiencyduetonumerousvariables.SomestudieshaveshownawiderangeofNandPremovalbyplants,15-80%forNand24-80%forP(GreenwayandWoolley,2001),whileothershavefoundplantstoremoveamuchlowerrange,14-52%forNand10-34%forP(Wuetal.,2013).PlantedandunplantedwetlandswerecomparedintermsofoxygentransferinCWsystem(Nivalaetal.,2013b).Plantsinsand-basedVFCWwereshowntoincreaseoxygenconsumptionrate(OCR)slightlyinCWcomparedtounplantedsystems,howeverthiscouldbeduetoincreaseinammoniumremovalandnotnitrificationprocess.Ontheotherhand,aplantedgravel-basedsystemdidnotseemtoaffectOCRasitshowsnosignificantdifferencethananunplantedgravel-basedsystem(Nivalaetal.,2013b).Althoughtheimportanceofvegetationinpollutant-removalprocessesisnotconsensual,itisagreedthatplantsdocontributetowastewaterpurificationeitherbydirectabsorptionorbyprovidingfavorableenvironmentformicroorganismsintheirwebroot(Gottschalletal.,2007).Asidefromselectingplantsbyitstoleranceandspecificpollutant-removalcapabilities,weshouldalsoexamineits“after-lifeusage”.Dependingonthemaintenanceschemeadopted,plantsmaybeperiodicallyharvestedtoavoidre-releaseofnutrientsbackintotheCWsystem,andmaintainingthewetlandvegetationinalog

59(growth)phaseofhighphysiologicalactivitytoenhanceremoval(Davis,1995).However,severalstudieshavesuggestedthatharvestingofvegetationinSFCWisnotnecessaryasplantuptakeofpollutantsrepresentarelativelyminorpathwaycomparedtosubstrate(Crites,1988;EPA,2000).However,ifbiofuelcropsareplantedasvegetationinCWs,harvestingcouldbeconsideredtosupportacirculareconomybygenerationofrenewablefuel.AttentionisrisinginrelationtothecontributionofbiomassofCWplantsforenergeticpurposes,throughdirectcombustion,andbiogasandbioethanolproduction.Whenbiomassvalorizationisconsidered,certainkeydesignshouldbeaddressed,suchas:i)thenutritionalcontentofthewastewatermustbeadequatefortheneedsofthetargetcrops;ii)theoptimizationofCWsystemflow;iii)theselectionoffastgrowingplantspecieswithhighyieldsofabovegroundmaterialandwithhighenergeticvalueandiv)planningthefrequencyofharvestinordertoenhancebiomassproduction(Langergraberetal.,2019).Biofuelisafuelthatderivesfrombiomass,insteadofthosetraditionallyderivedfromfiniteresourcessuchasoil.Primarybiofuelreferstothosethatareproducedwithnaturalrawmaterials,whereasSecondarybiofuelaremadefromprocessedfeedstocks(Dragoneetal.,2010).Also,dependingonthefeedstockusedandtheprocessingtechnology,biofuelscanbefurtherclassifiedasfirst,secondorthird-generationbiofuel.BioethanolorBiobutanolarefirst-generationbiofuelsthataremadebyfermentationofcarbohydrateswithinfoodcropssuchassugarcaneandcorn;biodieselisproducedbytransesterificationofoilcropssuchassunflowerandsoybean.Second-generationbiofuel’sproductionreliesonthehydrolysisoflignocellulosicmaterialsofnon-foodcrops,followedbysugarfermentationintobioethanol.Third-generationbiofuelsarederivedfrommarineresourcessuchascyanobacteria(Dragoneetal.,2010).

603.1.2.1Vegetationasbioenergyfeedstock:First-generationbiofuels3.1.2.1.1SugarcaneMacrophytesgrowninCWcanpotentiallybeusedasbioenergyfeedstock.CarbonintheformofCO2iscapturedandconvertedtocarbohydrates,complexoilandfibercompoundsinplants.Thesebiologicalmaterialscanbeusedasfeedstocktoproducebioenergybyharvestingandburningtheplant,essentiallyreleasingthecapturedcarbonbackintotheearth’satmosphere.Intheory,thisisacarbon-neutralfuelthatcouldhelptomitigateglobalwarming.Sugarcaneandcorn,forexample,arecommonlyusedfeedstocksfortheirhighsugarandstarchcontentforfermentationintoalcohol,whichinturnisusedtomakebiodieselandethanolfuels(KindbergandEnergy,2010).Mateusetal.(2014)haveinvestigatedtheuseofsugarcaneasCWvegetationinlaboratory-scaleinfourdifferentfilters(Filtralite®,Moleanoslimestone,basaltgravelandclaybrickfragments).Saccharumofficinarum(sugarcane)wassubjectedtolow-strengthwastewater;stemheight,diameterandleafareaallincreasedregularlyinthefourpotswherebyplantgrowthwasgreatestonlimestonefillingandlowestonthebasalt.IntermsofP-removal,theuptakeofPbythesugarcanefromallpotsincreasesfrom15%to30%,exceptforbasaltmediumwhichwasshowntobesaturatedwithinthefirst100days.Overall,estimatedPaccumulationbysugarcaneinthereportwasbetween15and30kgPha-1year,whereastraditionalsugarcaneproduction’svaluesarearound30to40kgPha-1year.Also,theP-removalvaluesobtainedinthisstudywerelowerthancommonCWemergentmacrophytes(30-150kgPha-1year)(Brix,1997).SugarcanewasshowntobeabletoproducebiomassregularlyinlowstrengthwastewaterandsimultaneouslyremoveP.AlthoughsugarcanewasshowntobelesseffectivethanothermacrophytesinitsP-removalcapability,PismainlyremovedbysubstrateandthereforeshouldnotgreatlyaffectoverallP-removalefficiency.Another

61studybyetDiresetal.(2019)hasshownthatsugarcanecantolerateBOD5concentrationof221mg/l,CODconcentrationof713mg/l;withremovalrateabove96.5%inagravel-bedwetland;italsoremoved97%ofTSSwith335mg/l.SugarcanethatgrowsinCWdoesnotoccupyarablelandandrequirednutrientsareacquiredfromwastewater,sugarcanewasalsoreportedtoberesistanttofloodconditions(Viatoretal.,2012)andisgenerallynotaggressive(Cheavegatti-Gianottoetal.,2011).Therefore,selectingsugarcaneasCWvegetationpresentsagreatopportunitytocouplewastewatertreatmentwithnetenergyproduction.3.1.2.1.2CornCornisalsowidelyusedforbioethanolproduction.Over90%ofbioethanolisstillmadefromcornintheUnitedStates(KindbergandEnergy,2010).Astudywasconductedtoinvestigatetheuseofcornasvegetationinagravel-basedrecirculatingVFCW(García-Pérezetal.,2011).Pollutantremovalefficiencieswerehighforfecalcoliformbacteria(99%),BOD(98.4%),TSS(95.6%),NH4+-N(95.3%),TN(79.5%)andTP(75.5%).P-removalefficiencyreportedinthisstudywasbetterthanothergravel-basedrecirculatingVFCWusingtypicalwetlandplantswherebysomeonlyreached33%ofP-removal(García-Pérezetal.,2009).Regulargravel-basedCWsnormallyexhibitpoorP-removalcapabilities(Wuetal.,2015c)andcornproduceisspecificallyknownforitslargeusageofPduringgrowth.García-Pérezetal.(2011)havealsoobtainedgoodcornyield(onlyKernel)equivalentto10,026kg/ha.AsidefrompollutantremovalanalysisandcorngrowthoftheCW,theauthorshavealsofurtheranalyzedthemicrobialandchemicalcontentofthecorn;nomajorchangesinnutrientvalueswerefoundwhencomparedtotheexpectedstandardsandaerobicplatecountvalueswerefoundtobebelowacceptablelimits(García-Pérezetal.,2013).Corngrown

62inthisstudywasalsonaturallyfield-driedtobelow14%moisturewhichcouldavoidfieldfungiandyeastcontaminationiftheharvestedcornisstoredandhandledproperly.CornwasalsotestedtotolerateCODconcentrationupto1450mg/l(JosephSahayarayanetal.,2019),generallynotaggressive(OECD,2003)andisfloodtolerant(Zaidietal.,2004).Hence,recirculatingVFCWthatusescornasvegetationwasshowntotreatwastewaterwhilesimultaneouslyrecyclingnutrientstogrowcorn,withnoadverseeffectsonmicrobiologicalandnutrientcontent,topotentiallyproducebiofuelorenergybiomass.Consideringthat40%ofthemaizeproducedin2010intheU.S.wasusedforethanolproductionandthatthishascausedwaterirrigationtoincreasefrom3%of6%from2007to2010(Faeth,2012),resultspresentedinthisrecirculatingVFCWstudy(García-Pérezetal.,2013)havemadepositiveimplicationstowardstheenvironmentalandeconomicviabilityofusingcornasvegetationinVFCW.3.1.2.2Vegetationasbioenergyfeedstock:Second-generationbiofuelsSugarcaneandcornareregardedasfirst-generationbiofuels.Second-generationbiofuelsrefertobiofuelsthatareproducedbyconversionofplantcellulose,hemicelluloseandlignintosugar,whichisthenfermentedtoethanol.Extractionofsugarfromtheseplantmatter,however,ismorecomplexandmuchlessstraightforwardthanfirst-generationbiofuelswhichaddstotheenergyinputintheprocess.Nevertheless,theoptiontochooseothervegetationasidefromsugar-richcropspresentsasanattractivealternativeaswecanchoosevegetationthatismoresuitedtothelocalclimateandenvironment.

633.1.2.2.1A.donax,P.australis,T.latifoliaAstudyhascomparedsixperennialwetlandplantsfortheirbiomasspotentialandbioenergyuse.Phragmitesaustralis(Commonreed)andArundodonax(Giantreed)wereshowntoobtainthehighestbiomass,calorificvalueandhighestNandPuptake(Zhaoetal.,2014).P.australisandA.donaxwerereportedtotolerateCODconcentrationof2010mg/l(Calheirosetal.,2007)and425mg/l,respectively(Calheirosetal.,2012).Althoughtheyarefloodtolerant(Mannetal.,2013;Pagteretal.,2005),theywerebothhighlyaggressive(Davis,1995;Lambertetal.,2010)andcaneasilybecomeinvasive.Someregionsorcountrieshavebannedtheirplantation(Davis,1995).Linetal.(2019)havealsostudiedfiveplantspeciesfromtwoChineseCWfortheirpotentialtoproducebioethanolusingsimultaneoussaccharificationandfermentation.P.australisdisplayedthehighestcontestsofholocelluloseof54.66%andalsohadthehighestconvertedglucoseconcentrationof36.44g/L.EthanolconcentrationsobtainedfromP.australiswasalsothehighest(34.89g/L),withethanolconversionof21.81%ofdrymass;thisiscomparablewithcornstoverwhichhasethanolconversionrateofaround19-22%(ZhaoandChristensen,2018).Therefore,withP.australisasthemostfrequentlyusedvegetationinSFwetlandsaroundtheglobe(Vymazal,2011b),resultsinthisstudywashighlyencouragingtofurtherconverttheharvestedP.australisforbioenergyproduction.However,moreresearchisneededtosupportthescaleupofbiomassproductioninCWforenergeticvalorizationpurposes.Typhalatifolia(Bulrush)isanothercommonemergentthatcouldbeconvertedtosecond-generationbiofuel.Itisalsofloodtolerant(Gucker,2008)andwasreportedtotolerateCODconcentrationof2010mg/l(Calheirosetal.,2007).However,sameasP.

64australisandA.donax,itishighlyaggressive(Davis,1995)butwithlowerNphyto-uptakecapabilitythanthetwospecies(Rebaqueetal.,2017).3.1.2.2.2C.indicaandI.pseudacorusForornamentalplants,Cannaindica(Cannalily)andIrispseudacorus(YellowIris)werereportedtotolerateCODconcentrationof1100mg/l(Haritashetal.,2015)and1156mg/l(Tangahuetal.,2019),respectively,thereforetheywouldbeabletogrowindomesticwastewater.Althoughfloodtolerant(SchlüterandCrawford,2001;Zamoraetal.,2019),theyarestillaggressivebutnotashighlyaggressiveasthethreespeciesmentionedabove(section3.1.2.2.1).Theycouldalsobeconvertedtobioenergyfeedstockbuttheirreportedproductivity(GJha-1yr-1)isnotashighasP.australisandA.donax(Liuetal.,2012).3.1.3LinerselectionTopreventcontaminationofundergroundwaterfromCWwastewater,alowpermeabilitylinerisoftenusedtolinethebottomofCWs.Themostcommonlyusedlinersareclays,claybentonitemixturesandsomesyntheticmaterialssuchaspolyvinylchloride(PVC)andhigh-densitypolyethylene(HDPE)(Ahnetal.,2001).AnLCAwasperformedtocomparethreescenarioswithdifferentwaterprooflinermaterialsofahybridCWfacilityinGreece(Gkikaetal.,2015):(1)originaldesignwithanaerobictankmadeofreinforcedconcrete,earth-coveredwith1m-thickHDPEgeomembranefortheCWbeds;(2)anaerobictanksandCWsareallearthenwith1mm-thickHDPEgeomembrane;(3)anaerobictanksandCWsareallearthenwith30cmcompactedclaylayer.Scenario3wasfoundtobethemostenvironmentallyfriendlyinallcategoriesasreinforcedconcretewasreplacedbytheclaylayer.Scenario2performsworstintheozonelayercategoryduetotheuseofHDPEgeomembrane.Thisstudy

65highlightstheimprovementsachievedifalternativeenvironmentallyfriendlyconstructionmaterialisused.Furthermore,theclay-layeredscenarioisalsothecheapestintermsofconstructioncostsamongthethreesystems,whereasscenariooneisabout22.1%moreexpensivethanscenariothree.SyntheticlinersarerelativelyexpensiveandnaturalclaysarenotalwaysavailablewhereCWisimplemented.By-productssuchasthosefromFlue-gasdesulfurization(FGD)werestudiedforthepurposeoflinerselection.FGDmaterialhasverylowpermeabilitywhenproperlyappliedandcheaptoobtain.However,FGDby-productscontainhighboroncontentwhichcouldbeleachedtotheenvironment.Ahnetal.(2001)havefoundthatFGD-linedmesocosmsretainmoretotalandsolublePandthereisnosignificantdifferenceintotalbiomassproductionofwetlandplantsforbothFGD-linedandunlinedmesocosms.However,potentiallyphytotoxicBoroncontentoftheplanttissuesofbelowgroundbiomasswassignificantlyhigherinFGD-linedmesocosms.Therefore,althoughusingbyproductsaslinerdoseemtobesustainableandcheap,furtherpilotandfull-scaleexperimentshavetobeconductedtomeasurethepositiveandnegativeeffectsofbyproducts.Garcia-Perezetal.(2011)havedesignedarecirculatingVFCWwithtwoliners.ThecellwasbuiltwithPVClineratthebottom,fillingthebottomcompartmentwithrivergravels(13-25mm);asecondPVClinerwasusedtoseparatethetwocompartments,extendingovermostofthetopareaofthebottomgravellayerwhileleaving25%oftheareanearestthewetlandinletuncovered.Topcompartmentwasthenfilledwith4mmofpeagravel.Thiscouldcontributetothegoodpollutantremovalefficiencyasthetoplinerkeepstheuppercompartmentaerobicandthebottomcompartmentactingasananaerobicenvironmentfordenitrification.Nevertheless,the

66constructioncostandenvironmentalimpactcouldincreaseduetotheuseofmoreconstructionmaterials.3.2EvaluationofconstructedwetlandsystemswithLife-CycleassessmentstudiesLifecycleassessment(LCA)isacomprehensivetooltoestimatetheoverallpotentialenvironmentalimpactsexertedbyaprocessorproductoveritslifecycle.Itcanbeusedduringthedesignphasetochoosebetweendifferentwastewatertechnologies;itcanalsobeusedtoidentifywhichlifecyclestageexertsthemostsignificantenvironmentalimpactsforaparticulardesignsothattailoredimprovementsandalterationscanbemade(Chenetal.,2011;Ledonetal.,2017).However,LCAstudieswerenotextensivelyperformedforCWsystems.SomeoftheLCAresearchfoundinthisthesiswerediscussedbelowtoprovidefurtherinsightsintodifferentenvironmentalimpactsofvariouscomponentsofCWs.3.2.1LCAscomparingconventionalWWTPandCWsByperformingLCAonaCWsystem,givenasetofboundariesandassumptions,wecancompareitagainstconventionalwastewatertreatmentsystemstodecidewhichtechnologyisbestsuitedtoaspecificcase.Forexample,Rouxetal.(2010)haveconductedanLCAtocomparetwowastewatertreatmentsystems:aVFCWandanactivatedsludge.TheVFCWwasshowntohavesmallerinfluenceonallenvironmentalimpactcategories,exceptforeutrophicationduetotheincompleteremovaloftotalNitrogenandtotalPhosphorus.ImproveddesignscanthenbemadefocusingonTNandTPremovaloptimization,forinstance,differentsubstratescanbeusedtoincreaseP-sorptioncapacity.Sensitivityanalysiscanthenbeperformedtoevaluatethealtered

67impactsbyreplacingdifferentsubstratesintheLCAprogram.Garfíetal.(2017)havealsoconductedanLCAtocompareanactivatedsludgesystemandtwonature-basedtechnologies(aHybridCWandhighratealgalpondsystems).Thenature-basedsystemswereshowntobethemostenvironmentallyfriendlywithimpact2to5timessmallerthantheconventionalwastewatertreatmentplant.However,differentsubstrate,liner,vegetation,designofconstructedwetlandwouldaffectitsenvironmentalfootprintandcost(Garfíetal.,2017).Insteadofaccountingthewholelifecycle,someLCAsonlyfocusontheCWcollectionsystem,treatmentsystemtosludgedewateringandsludgetransportationfromseptictank;constructionphaseoftheCWwasnotincludedasthestudywouldliketofocusontheimpactofmacrophytesandthereforeconstructionphasecontributesequallyregardlessofthespecies(Ledonetal.,2017).ExcludingtheconstructionphaseintheHFCWsystemdescribedinLedonetal.(2017),98%ofCH4wasemittedfromCWoperation,specialattentionshouldthereforebepaidtoreducingitsemission.Therefore,dependingonthefocusofthestudy,differentboundariesandassumptionscouldbeset.Nevertheless,ifanoverallenvironmentalimpactofaCWistobeevaluated,theconstructionphasemustbeincluded.Constructionphasewasknowntoaccountfor58.77%oftheoverallGHGemissionsfromaVFCWsystem(Chenetal.,2011),otherstudieshavefoundthattheconstructionphaseconstitutesupto82.31%oftheCW’stotalGHGemissions(Gaoetal.,2012).FromvariousLCAstudies,itwasgenerallyfoundthattheoperationalphasehasgreaterimpactsinconventionalWWTPwhileconstructionphasehasgreaterimpactsinCWs(Chenetal.,2011;Dixonetal.,2003;Gaoetal.,2012).LCAscanhelpustodecidewhichtypeofCWhaslessimpactsontheenvironment.Fuchsetal.(2011)haveperformedanLCAtocompareVFCWwith

68HFCWwastewatertreatment.TheauthorshavefoundthatVFCWwillcauselessenvironmentalimpacttreatingthesameamountofwastewaterthroughoutitslifecycleduetohighertreatmentefficiency,lowerGHGemissionsandsmallerlandarea.TheCH4emissionswerefoundtobesignificantlylowerinVFCW(Median3.0mgm-2h-1)thanHFCWsystems(Median6.4mgm-2h-1)(Manderetal.,2014).VFCWsystemalsorequiresfourtimeslessareathanHFCWtotreatthesamevolumeofwastewater(Fuchsetal.,2011).Thissmallfootprintsuggestedthatlessmaterialsandenergyarerequiredduringtheconstructionphase.3.2.2LCAofCWwithlightweightexpandedclayaggregateassubstrateAssandandgravelarestillthemostemployedsubstrateinCW,alotofLCAstudiesfull-scaleCWswithonlysandandgravelassubstratematerials.Asmall-scaleCWwithadifferentsubstrateandanextendedaerationactivatedsludgewastewatertreatment(ASTS)werecomparedusingLCAinruralareasinEstonia,includingbothconstructionandoperationphases(Lopsik,2013).Impact2002+andReCiPemethodswerebothusedtoassessthewastewatertreatmentsystems.BothmethodsfoundthattheCWwithlightweightexpandedclayaggregate(LWA)assubstratematerialdominatesinmostenvironmentalimpactindicatorsandconstitutesonaverage83.64%ofthetotalimpactinthecomparativecharacterizationwithImpact2002+methodand75.09%ofthetotalimpactwithReCiPemethod.AnincreaseinASTSimpactswithReCiPemethodcomparedwithImpact2002+methodwasduetothreecategoriesthatarerelatedtodischargedeffluentinsteadofoneinthelattermethod.ItwasfoundthatthemainnegativeimpactofCWistheuseofLWAduringCWconstructionphase,andtoalesserextentduetotheimpactofEPDMrubber(ethylenepropylenedienemonomer),PVCpipes,rubbleelectricalcableandlanduse.Fortheextendedaerationactivatedsludgetreatmentsystem,themainnegativeimpactis

69causedbytheuseofelectricityandeffluentquality,wherebythelatterdominatesintheeutrophicationcategory.PineffluentwashigherinASTSthanCW.IftheLWAsubstratematerialoftheCWisreplacedwithsandandgravelduringsensitivityanalysis,theaverageenvironmentalimpactsoftheCWwerefoundtoreduceby22.91%.However,theeffluentqualitymightdecreaseduetolimitedP-sorptioncapacityofsandandgravelasaforementioned.Inanalternativescenario,CWcouldobtainanegativevalueintheglobalwarmingcategoryasplantuptakeisgreaterthantheemissionsfromconstructionandoperationphasesofthetreatmentsystems.Fromthisstudy,theimpactof1PEofCWislargerthantheimpactofactivatedsludgetreatmentmainlyduetothelargeenergyinputforLWAsubstrateduringconstructionphase.AsCWonlyrequiresrelativelylesselectricitytooperatecomparedtoWWTP,theimpactoftheoperationphaseismarginal.TheoperationimpactofASTShasalargerimpactduetoitselectricityusage.Hence,thisLCAstudysuggestedthatselectingafiltermaterialwithlittleenergyconsumptionduringproductiondoesaltertheenvironmentalimpactgreatlyduringCWconstructionphase.VegetationisalsoimportantincarbonsequestrationwhichcouldoffsetCO2releaseduringconstructionofCW.AnotherLCAforcomparingconventionalCWandmicrobialfuelcellcoupledCWhasalsoshownthattheuseoflessenvironmentallyfriendlymaterialshasledtotwicetheenvironmentalimpactintheabioticdepletioncategory,directlyaffectingtheoverallenvironmentalimpact(Corbellaetal.,2017)(tobediscussedinsection3.3.2.1).3.2.3LCAofaFieldTidalFlowCWwithDewateredAlumsludgeassubstrateAccordingtopreviousstudies,itwasreportedthatdewateredalumsludge(DAS)hasgoodP-sorptioncapacity(Babatundeetal.,2010).Itisaby-productfromconventionalWWTPthatiscommonlydisposedintothelandfill.Wangetal.(2018)

70haveperformedLCAonaset-upDASbasedCWwithtidalflowoperationstrategytoincreaseP-removalefficiencyandatthesametime,intensifyTNremovalefficiencybycreatingalternatingaerobicandanoxicconditions.Tidalflow(alsoknownasfill-anddrain)isanoperationalstrategytoletCWbedtobeintermittentlysaturatedandairisdrawnintooxygenatethebiofilms.Swinewastewaterwaspumpedintoafour-stagedfieldtidalflowCWsystem.Duringsensitivityanalysis,noenergyinputwasincludedforDASproductionandthereforeavaryingamountofDAShasnosignificantsensitivityinallenvironmentalimpactcategories.Havingthepre-existingknowledgethatconstructionmaterialwouldaffectthetotalimpactincategorysuchasGlobalWarmingPotential(GWP),thissensitivityanalysishasshownthattheuseofDASassubstratehasreducedtheenvironmentalimpactthatwouldhaveotherwiseexertedbylessenvironmentallyfriendlymaterials.However,DAStransportationwasthemainsourceofNOxemission.Therefore,ifaCWislocatednexttoaWWTPforcomplementarypurpose,thiswouldgreatlyreducetheNOxemission.Inthisstudy,electricitywasfoundtodominatealltheenvironmentalimpactswhilewaterpumpsusedinthetidalflowsetupwerethedominantelectricityconsumer.AlthoughP.australissequesteredcarbonandwasoftenknowntooffsetcarbonemissionfromtheCW,theCO2emittedfromtheelectricitygenerationhashugelyovertakenthesequesteredamount,accountingfor60-80%oftheadverseimpactsinfossildepletionpotential,GWP,CO2andSO2emissions.TidalflowCWshouldthereforebecomparedwithothermethodsthatincreasetreatmentperformanceofCWtoidentifythebestdesignforsustainablewastewatertreatment.AlthoughDAShasnoenergyinputduringitsproduction,DAS’sdownstreamusageasfertilizershavetobetreatedwithacidchemicalsandbiotoxicitytestshavetobeperformedtoensureitssafeapplicationoncrops(Kluczkaetal.,2017;Zhaoetal.,

712009).The“afterlife”ofDASshouldbeconsideredifweaimtoclosethematerialconsumptionloop.3.2.4LCAofbiofuelproductionthroughCWThepossibilityofgrowingbioenergyfeedstockasCW’svegetationhasbeenexploredbyseveralstudies.Asthereisongoingdebateofbiofueltakingupspacethatwouldotherwiseuseforfoodcrops,growingbioenergycropsinspacethatisusedforwastewatertreatmentcouldbeanattractivealternative.Life-cycleassessmentwascarriedoutbyLiuetal.(2012)toestimatetheenergybalanceandgreenhousegas(GHG)emissionsforbiofuelproductionbyCW.Itsassessmentinvolvesstagesfromseedlinggrowthtoharvest,consideringallconstructionandoperationstagesandenergyconsumedintransportationandinconvertingcropstobiofuels.OtherdataregardingbiofuelproductionandwastewatertreatmentmethodswerealsocollectedandcomparedagainsttheCWsystem(Table1).Asshownintable1,CW’smaximumannualproductionofabove-groundbiofuel(bioenergyproduction)islowerthanthemicroalgaesystem,buthigherthantheothers.Table1.Comparisonofbiofuelproductionandwastewatertreatmentsystems.(Source:Liuetal.,2012).Note:NA,notavailable;–,notthisitem;GHGsincludeCO2,CH4andN2O.*Megagram(1Mg=106g)CO2equivalentperhectareperyearforbiofuelproduction.EcosystemsandkilogramsofCO2equivalentperkilogramofnitrogenremovalforwastewatertreatmentsystems.†103USperhectareperyearforbiofuelproductionecosystemsandUSperkilogramofnitrogen.

72ForCWtofurtherincreaseitsbioenergyyield,threefactorshavetobecarefullyconsidered.Nitrogensupplyfromwastewaterhastobesubstantialtosupportthegrowthofplants,optimizinghydrologicflowpatternandselectingproductiveplantspecies.Averagebiofuelproductionwasdeterminedtobeone-thirdhigherinSFwetlandsthanFWSwetlandsduetoitsmesichabitatssuitableforvariousplantspecies,whileFWSonlyfavouredaquaticplantswithlowerbiomass.VFCWisgenerallyselectedforhigherproductivityduetoamoreuniformgrowthenvironmentanditssuitabilityforproducingcellulosicbiofuel.Asmentionedinthevegetationselectionsection,plantssuchassugarcaneandcornthatarehighinsugarandstarchcanbeusedtoproducefirst-generationbiofuels;howeverplantswithhighholocellulosecontentsuchasPhragmitessp.andA.donax(Giantreed)arealsoshowntobesuitablecandidatesforsecond-generationbiofuelproductioninCWs(Zhaoetal.,2014);bothofthesespecieshavehighbiomasscontent,calorificvaluesandhighNphyto-uptake.InthereviewconductedbyLiuetal.(2012),A.donaxrankedfirstintermsofproductivitywith1836GJha-1yr-1whichiscomparabletoreportedexistingbiofuelplantsintropicalregions(1,628GJha-1yr-1forPennisetumpurpureumand1,850GJha-1yr-1forEchinochloapolystachy)(Somervilleetal.,2010).P.australisrankedsecondasshowninFigure6andisknownforitstoleranceforwiderrangeofclimaticconditionsandflowpatterns(Pagteretal.,2005).

73Fig.6.Plantspeciesandtheirproductivity.(Matchinglowercaselettersindicatenon-significantdifferences(P>0:05)betweenplantspecies.Forbandcthelineandsquarewithintheboxrepresentthemedianandmeanvalues;bottombars,bottomandtopedgesofboxes,andtopbarsrepresent5%,25%,75%and95%ofalldata,respectively.Opencirclesoutsidetheboxesrepresentextremevalues(Source:Liuetal.,2012).Asshownintable1,CWsactuallyhavehigheraveragenetenergybalance(NEB=energyoutput-input)whencomparedtoothersixbiofuelproductionsystems(corn,soybean,switchgrass,microalgaeandLIHDgrassland(low-inputhigh-diversitymixturesofnativegrasslandperennials)),butasenergyinputduringconstructionisalothigherinCWs,theNEBratioisaboutmid-levelcomparedtotheothers.ForGHGemissions,CO2emissionsarerelativelyhighinCWsduringinfrastructureconstruction,biomassproductionandharvesting.N2OemissionfromCWisalsohighduetohighNconcentration,emissionisapproximately4600%higherthanmicroalgaeand1800%ofLIHDgrassland.However,CO2sequestrationisalsohigherinCWsthantheyemitandishigherthantheotherfivebiofuelsystems(Table1).AstheprimaryaimofCWsiswastewatertreatmentandthattheirhighGHGemissionsaremainlyduetothisfunction,iftakingWWTPsasbaseline,CWscould

74offsetGHGemissionsbyreductionof591kgequivalentCO2perkgofNremoval(Table1).MicroalgaeaswastewatertreatmentsystemshaveahighernetGHGemissionsthanCWastheylacklong-termcarbonsequestrationeventhoughtheyemitlessGHG.Economicallyaswastewatertreatmentsystems,CWsarecompetitiveasitcostsone-thirdtoone-halflessthanthecostofconventionalWWTPperkgofNremovedinChina(Table1).However,formoreengineeredCWswithmorepumpingsandcomplicateddesigns,thecostwillincrease.Nevertheless,ifweonlyconsiderbiomassharvestingasthecostofbiofuelproduction,assumingtheconversioncostofcellulosicfeedstocktoethanolisthesameinallbiofuelsystems,thebiofuelproductioncostofCWs(takingtheWWTPasbaseline)onlyaccountsfor0.5%oftotalcost,makingCWsveryeconomicallycompetitiveamongotherbiofuelproductionsystems.Cost-benefitratioofCWsisalsohigherthanothersystemsasshownintable1asitoffersanumberofecosystemservices.Overall,accordingtothelifecycleassessmentandliteraturereviewconductedbyLiuetal.(2012),couplingwastewatertreatmentandbiofuelproductionthroughCWisenvironmentallyandeconomicallyfeasiblewhencomparedtoWWTPandotherbiofuelproductionsystems.However,dependingonthesizeandscatteringsofCW,acentralisedcollectionandethanolconversionstationhastobeestablished;perhapslocallyconvertedenergystationcouldbesetupnearCWstominimiseenergyinputintransportationandoperatinglargescalefacilities.AnotherLCAofenergybalanceandGHGemissionwasconductedonaVFCWbiofuelsysteminZhoushan,China(Liuetal.,2019).SubstrateusedinthisVFCWismainlysandandgravel;sevenplantspecieswereselectedfortheirhighbiomassproductivity(Liuetal.,2009),includingP.australis,T.latifolia,C.papyrus,C.indica,

75Phalarisarundinacea,A.donax,andGlyceriamaxima.Firstly,maximumplantabovegroundbiomassinthisstudyreached90,000kgha-1yr-1whichisalmostthehighestbiomassproductivity,secondtoEchinochloapolystachyaontheAmazonfloodplain.IntermsofNEBratio,theVFCWisaroundmid-levelamongswitchgrass,corn,soybeanandLIHDgrasslandbiofuelproductionsystems.However,ifonlybiofuelproductionisconsideredinCW,noNfertilizerhastobeappliedunlikeotherfirstandsecond-generationbiofuelsystems,makingitoverallN-negative.ForLCAofenvironmentaleffectsofCW,thereisanetGHGreductionof8.8MgCO2equivalentha-1yr-1fromthisVFCWbiofuelsystem;consideringarelativelyhighCH4emissionandsmallN2Oemissionfrommicrobialproduction,thiswasoffsetbyanoverallhighCO2sequestration,resultinginanetGHGreductioninglobalwarmingpotentialindicator(GWP).UsingWWTPasbaseline,WWTPhasnoenergyoutputandenergyinputis40timesmorethanCW,CWcouldintheorysave7649.1Gjha-1yr-1.CWalsoemits596.6MgCO2ha-1yr-1lessthanWWTPandemitsthreeordersofmagnitudeofCH4lessthanWWTP,intotal,CWcanreduceGHGemissionsto2714timeslessthanthatoftheWWTP.IfCWsweretoreplaceWWTPastheconventionalapproachtowastewatertreatment,thetotalGWPreductionintensityfortheCWwillbe86,050.5MgCO2eq.ha-1yr-1.SubstratesubstitutionusingmaterialswithbetterP-sorptionandwithlowerenergyinputsuchasindustrialbyproductscouldpotentiallyfurtherdecreasetheoverallenergyinput.Nonetheless,abovegroundplantbiomassinthisstudyconductedinasubtropicalregioncouldbeoverestimatingthevaluesastheyarenotexpectedtobeashighinplaceswithcoolerclimates.Nevertheless,theabovetwoLCAshaveprovidedpositiveresultstowardstheuseofcombinedCWbiofuelsystem,givinganinsighttowardstheenvironmental,ecologicalandeconomicbenefitsofbiofuelproductionthroughCW.

763.3Reductioninfootprint(Landrequirement)ofConstructedwetlandSeveralresearchstudieshavebeenconductedtoexplorethepossibilityofreducingtheCWarearequirement(footprint)whileachievinggoodpollutantremovalefficiency(Corbellaetal.,2017;Kantawanichkuletal.,2005;Prost-BoucleandMolle,2012).TheEnvironmentalProtectionAgency(EPA)designrecommendsreducingthefootprintto1m2PE−1(PEreferstopopulationequivalent)fortertiarytreatmentand0.5m2PE−1forstormwatertreatment(Ilyas&Masih,2017).However,Vymazaletal.(2011)havereviewedthatVFCWsaregenerally1-3m2PE−1andHFCWsarearound5m2PE−1forremovalofTotalSolidRemoval(TSS)andorganicmatter.Butsomestudieshaveshownthatthe1-3m2PE−1mightnotbeenoughfornutrientremovaltomeetnationalstandards(Babatundeetal.,2008).Ontheotherhand,conventionalwastewatertreatmentplantonlyhasafootprintof0.2-0.5m2PE−1(vonSperling,1996).ToreducethesurfaceareaofaCW,therateofpollutantremovalhastobeimprovedinordertoensureCW’streatmentperformance.However,asCWsarenotinacontrolledenvironment,studiesoftenpresentdiscrepanciesamongeachotherduetothenumberofunverifiedparametersthatcontributetotherateofpollutantremoval.Studieswereconductedonwastewaterwithdifferentstartingpollutantconcentration,hydraulicloadingandretentiontime,indifferentgeologicallocationswithvariousclimatesandinvariousrangesofdurationtime.Forexample,CWsinsubtropicalclimateregionsinSouthernChinamayhaveasmallerareathanNorthernChinaastheyhavehighertemperatureandhumiditywhichisfavorableforvegetationgrowth,reactionratewouldalsobefaster(LiandWang,2006).Itisthereforeextremelyhard

77tomakeaccuratecomparisonsandderiveastandard“bestsetofparameters”.Otherstudieswereconductedinlaboratoryscalewhichdoesnotnecessarilyreflecttheefficiencyofafull-scalesystem.Hence,itisveryimportanttotestthedesignfirstbeforeimplementingitinfull-scale.Nevertheless,toreduceCWfootprint,commonrate-limitingfactorsofthereactionsinCWwerereviewed.Table2presentssomeofthefactorsthatreducepollutantremovalefficiencyandexampletechniquesfromstudieswerereviewedtosolvetheserate-limingfactors.Table2:Factorsthatreducepollutantremovalefficiencyandtheirpotentialsolutions.FactorsthatreduceCWefficiencySolutionTechniquesReferencesLandavailabilityHybridsystemsandothersmartdesignsIlyasandMasih,2017;Kantawanichkuletal.,2003Oxygenavailability(a)Tidalflow(b)EffluentRecirculation(c)ArtificialAeration(d)MicrobialFuelCells(MFCs)(a)Changetal.,2014(b)IlyasandMasih,2017(c)Nivalaetal.,2013b;Resendeetal.,2019(d)Corbellaetal.,2017TemperatureThermalInsulationWuetal.,2015aCarbondeficiencyindenitrification(a)step-feeding(b)Additionoforganiccarbon(c)Additionoforganicsubstrates(a)Wuetal.,2015a(b)Chenetal.,2011;Linetal.,2002(c)Wuetal.,2015aRe-releaseofnutrientsbydecomposingvegetationMultipleharvestofbiomassWuetal.,2015a

78MicrobialdegradationBioaugmentationZhaoetal.,2017LimitedelectronacceptorsMicrobialfuelcellCorbellaetal.,2017Clogging(a)Additionoflocalearthworms(e.g.Pheretimapeguana)(b)Uselargersizedmedia(morethan5mmindiameter)(a)Nuengjamnongetal.,2011(b)Wuetal.,2015a3.3.1FactorsthatreduceCWefficiency:LandavailabilityVFCWsgenerallyhasasmallerfootprintthanHFCWs.HFCWssupportlimitednitrificationreactionduetoitslowoxygentransfercapacity;VFCWsareoperatedwithintermittentlyloadingwhichallowsoxygentocomeintocontactwithsubstratemediaforalongerperiodoftimeandgravitationalenergydrivesthedrainage.VFCWsalsogivetheoptionofnotrequiringpre-treatmentsinceorganicsolidcansettleonthesubstrateandwillbedewateredandaerobicallyreactwithmicroorganisms(Stefanakisetal.,2014).VFCWs,thus,requirelessareathanHFCWs.However,nitratelevelisrelativelyhighinVFCWsasthedenitrificationprocessoccursinanenvironmentwithDOlessthan0.5mgL-1.HFCWsprovidealowoxygenenvironmentnearthesubstratelayerandhenceareeffectiveinremovingtotalnitrogencontent.Hence,HCW,acombinationofHFCWandVFCWwithonestackingoveranother,wouldhavebetterremovalefficiencyasitcombinesbothnitrificationanddenitrificationprocessesinoneandthus,smallerfootprint.Ina2017review,IlyasandMasihhavecomparedseveralstudiesofdifferentHCWfootprintandpollutantremovalrate;Kantawanichkuletal.(2003)designedtwoHCWwithonetypeofCWstackingoveranotherandviceversa.Theremovalefficiencywassimilarinbothsystemsand

79footprintwasdecreasedto1.8m2PE−1inaVFCWstackingoverHFCWdesign.VFCWwasshowntocontributemoretonitrificationandHFCWwaspromisingindenitrification.Thisstudyalsohighlightsthesignificanceofhydraulicloadingrate(HLR);lowHLRof(3cm/d)hashigherremovalefficienciesthan6and12cm/d.TotalPhosphorus(TF)removalratealsodecreasesinhighHLRof12cm/dduetolimitedadsorptioncapacityofsandmedia.StackingofVFCWoverHFCWisattractiveinareaswithlandscarcityasthedesignhasgoodremovalefficiencyandrelativelysmallfootprint.However,thisdesignismoredifficulttobuildandrequiremorefrequentmaintenance.3.3.2FactorsthatreduceCWefficiency:OxygenavailabilityOxygenisoneofthemostcrucialrate-limitingfactors.Itallowsmicroorganismstocompletebiodegradation;bacteriasuchasNitrosomonasandNitrobacteriahelpconvertammoniumtonitrateinanaerobicprocessknownasnitrification.Asatmosphericdiffusionisslow,theoxygendemandexertedbytheincomingwastewaterexceedstheamountofoxygenavailablewithinthesystem.OptimizationofDissolvedoxygen(DO)inCWwithtechniquessuchastidalflow(TF),effluentrecirculation(ER)andartificialaeration(AA)werestudiedalongwiththeirabilitytoreduceCWfootprint(Ilyas&Masih,2017).Tidalflow(alsoknownasfill-anddrain)isanoperationalstrategytoletCWbedtobeintermittentlysaturatedandairisdrawnintooxygenatethebiofilms.Thisoperationalsodependsonflooddrainratios,C/Nratioandsubstratecharacteristics.Coarsesubstratesintheupperlayercouldhelppreventclogging.Longerdrainandshorterfloodedtimemightbepositivetonitrification,butlittleinfluenceonPremovalwasshowninastudyconductedbyChangetal.in2014.

80Effluentrecirculationinvolvespumpingpartoftheeffluentbacktotheinflow,thisaddsoxygenintothesystemandgreatlyincreaseswastewatercontactwithaerobicmicroorganismsandimprovespollutant’sremoval.Artificialaerationrequirestheuseofairpumpandblowertoaddoxygenintothesystem.This,however,consumestwicetheamountofpowercomparedtotidalflowsystemwetlands(Ilyas&Masih,2017).IntermittentAAispreferredtoprovideproperratioaerobicandanoxicconditionsforallreactionstooccur.ThisalsoconsumeslessenergythanContinuousAA.Oxygenconsumptionrate(OCR)wasmeasuredinIntensifiedaeratedsystemswithbothVFCWandHFCWandwascomparedagainstnon-aeratedsystems(Nivalaetal.,2013b);intensifiedsystemswerefoundtohavethehighestOCRbyfarandhasthebestammonium-NRemovalrate.Resendeetal.(2019)havecomparedaHybridCW(VFCW-HFCW)withanaeratedVFCWusinglifecycleimpactassessment;thehybridsystemhasthehighestimpactinGHGdirectionemissionsfromseptictankandaeratedVFCWconsumeshighamountofelectricityduringoperationwhichcausesdominationinallimpactcategories.Theaeratedsystem,however,performsbetterinpollutant-removalefficiency.Theuseofwaterandairpumpsintheabovemethodshowevercomesatacostofincreaseinenergyconsumptionduringoperation.FootprintandremovalefficiencywerecomparedbetweendifferentCWwithdifferentaerationmethodsinFigure7and8.

81Fig.7.FootprintofthethreeaerationmethodsofdifferentCWs.RedcirclesindicateCWswiththethreelowestfootprint.(Source:AdaptedfromIlyas&Masih,2017).Fig.8.ComparisonofremovalefficienciesandfootprintbyaerationmethodofdifferentCWs.RedcirclesindicateCWswiththebestcombinationofremovalefficiencyandlowestfootprint.(Source:AdaptedfromIlyas&Masih,2017).Figure7hasshownthatiffootprintisthemajorindicatorforCWselection,VFCWwouldbethemostobviouschoicewithER-VFCWandTF-VFCWhavingthesmallestfootprint.However,figure8indicatesthatER-VFCWhasthelowestefficiencyinpollutantremoval.HighesttreatmentefficiencywasdemonstratedbyER-

82HCW.Overall,ifbothfootprintandremovalefficiencyareequallyimportant,TF-VFCW,AA-HCW,andER-HCWcouldberecommendedasthebestoptions(Ilyas&Masih,2017).3.3.2.1LCAofMicrobialFuelCellcoupledCWAnothernewinnovativeapproachtoincreasetreatmentperformanceofCWisbymergingmicrobialfuelcells(MFCs)intoCW.MFCisabioelectrochemicaldevicethatgenerateselectricityviaaredoxgradient.Themicrobesattheanodewilloxidiseorganiccompounds,theelectronsarethentransferredtothecathodeviaanexternalcircuittoreactwithanoxidant(likeoxygen)andprotonstogeneratewatermolecules.ThenaturallygeneratedredoxgradientbetweenupperaerobiclayerandloweranaerobiclayerinaHFCW,providesfavourableconditionsfortheimplementationofMFCsinCWsystems(Srivastavaetal.,2019).TheextraelectronacceptorsintheformofaconductivematerialintheanaerobiclayerofHFCWcanhelptopromoteanaerobicreactions;MFCsstimulateorganicwasteremovalbyfosteringmoreefficientdegradationpathways,thusimprovingoveralltreatmentefficiency(Fig.9).Yadavetal.,(2012)haveshownthatapowerdensityof15.7mW/m2andCODremovalof74.9%fromasyntheticdye-containingwastewaterwereachievedwithaCW-MFCsystem.LifecycleassessmentofthreescenarioswascarriedoutbyCorbellaetal.(2017):(1)conventionalCWsystem,(2)CW-MFCsystemwithagravel-basedanode,and(3)CW-MFCsystemwithagraphite-basedanode.Theconstructionphaseofallthreescenarioswerefoundtoaccountfor88-95%ofthetotalimpactintheabioticdepletioncategory,whichisinaccordancewithpreviousstudies(Dixonetal.,2003;Fuchsetal.,2011).IfelectricityproducedbyMFCsisusedinsteadofusingelectricityfromthepowergrid,potentialenvironmentalimpactwouldbereducedbyabout3%inallimpactcategories.Nonetheless,gravel-basedanodeinCW-MFCwasshowntohave

83twicetheenvironmentalimpactintheabioticdepletioncategorythanotherscenariosduetotheuseoflargeamountsoflessenvironmentallyfriendlymaterial,suchasstainlesssteelmeshattheanode.ThisindicatestheimportantrolethatconstructionmaterialplaysintheoverallenvironmentalimpactoftheCW.ConventionalCWdominatedtheglobalwarmingcategoryduetotheemissionofCH4bymethanogenicmicrobes,whereasMFCsactuallyhelpreduceCH4emissionunderanaerobicconditions.Overall,CW-MFCwithgraphite-basedanodeappearedtobethemostenvironmentallyfriendlyoptionwith20%footprintreductioncomparedtoconventionalCWs.Fig.9.MicrobialfuelcelltoCW-MFC(Source:Srivastavaetal.,2019).3.3.3FactorsthatreduceCWefficiency:TemperatureWarmtemperaturewillincreasetherateofchemicalreaction.ThetopofsubsurfaceCWcanbecoveredbyinsulatingmulchmaterialssuchasbarkagriculturalstraw,wastecompost,topreventfreezingintemperatezoneswithlowertemperature(Wuetal.,2015a).Thiswillnotbenecessaryforsubtropicalclimateregionsandalso,initialinvestmentcostwillbehigherifinsulationwasadded.

843.3.4FactorsthatreduceCWefficiency:CarbondeficiencyindenitrificationDuringdenitrificationforTNremoval,carbonsourceisrequiredformicrobes.Forwastewaterwithhighnitratesandloworganicmatter,carbondeficiencycouldbetheratelimitingfactor.Step-feedingisaprocessthatintroducesgradationalinflowofwastewater,thishelpsincreasetheC/Nratio.WastewaterwillalsobeevenlydistributedintheCWwithmorethanoneinputpoint(Wuetal,2015a).Additionoforganiccarbonisawaytoimprovedenitrification.StudieshaveshownthataddingfructoseandanaerobiclitterleachatehasresultedinTNremovalofover90%(Chenetal.,2011;Linetal.,2002).Highermolecularcarbohydratessuchaslitterleachateispreferredaslowmolecularcarbohydratescouldbeconsumedbymicrobesforotherprocesses.3.3.5FactorsthatreduceCWefficiency:Re-releaseofnutrientsbydecomposingvegetationAlthoughthemainremovalofcontaminantshighlydependsonmicroorganism,averysmallpercentagewasduetovegetationuptakeofpollutantsortransformedcontaminantsintotheirsystem.Whenvegetationdecays,nutrientswillbere-releasedintotheCW.HarvestofabovegroundplantswasshowntobebeneficialonTNandNH4+-Nremovalcapacity(Zhuetal.,2012).3.3.6FactorsthatreduceCWefficiency:MicrobialdegradationAsmicroorganismsplayakeypartinremovalofpollutants,treatmentperformancecanbeimprovedviabioaugmentation-theadditionofhighlyconcentratedandspecializedmicrobialcultures.Thistechniqueismainlyemployedforcomplicatedpollutantssuchasthosethatarefoundinpollutedriversorleachatesites.Experimental

85studiesmustbecarriedouttofindthemostcompetentstrainwithhighremovalefficiency.UsingdifferentLCAmethods,bioaugmentedCWwasshowntohavebothbetter(e-Balance)andworse(CMLanalysis)environmentalimpactsthannon-bioaugmentedCW(Zhaoetal.,2017).E-balanceputsmoreemphasisonpollutantremoval,thereforedecreasingimpactsineutrophicationandCODcategoryinbioaugmentedCWandsoaffectstheoverallimpact.CMLanalysisismorecomprehensive,inoculateproductionhasgreatlyincreasedimpactinglobalwarmingpotential(GWP)asthemainsourceofenergycomesfromcoal.Inasensitivityanalysiswhere,naturalgasisusedasenergysourceinsteadofcoal,GWPandabioticdepletionoffossilfuels(ADF)wereevidentlydecreasedbyhalftotwo-thirdsoftheimpact.Ifdomesticwastewateristreated,bioaugmentationmightnotbenecessary.3.3.7FactorsthatreduceCWefficiency:CloggingCloggingpresentsamajorproblemasoneofthemajorgoalsindesigningaCWisthatitwouldrequireaslittlemaintenanceaspossible.Cloggingbysolidcontentswouldlowerthedesignatedloadingrateandthesubsequenttreatmentefficiency.Usingcoarsesubstrate(morethan5mmindiameter)attheupperlayerisonewayofreducingthechanceofclogging.Anotherwayistheadditionofearthworms(e.g.Pheretimapeguana).Earthwormsdirectlyingestanddigestorganicmaterialsandindirectlyinfluencenitrogencycling(Wuetal.,2015a).Itwasshownthatearthwormshelptoreducesludgeproductiononthesurfaceofswinewastewaterwetlands40%byvolume,whichmeansthecostoftreatingCWsludgewillbeloweredtoo(Nuengjamnongetal.,2011).

863.4.Constructedwetlandsinhighdensityurbansettlements:casestudyofMacaoInsection3.1to3.3,reviewsregardingthesubstrate,vegetationandlinermaterialforconstructedwetlandsetupwerediscussed.ExistingLCAstudiesforenvironmentalimpactsofdifferentmaterialsandCWsystemwerealsoexamined,alongwithCWfootprintreductionstrategiesinordertodesignaCWmodelthathaslowenvironmentalimpact,smalllandrequirementandgoodpollutantremovalefficiency.ThisinformationwillbeusedinthefollowingsectiontodesignaCWmodelforMacaoSARinChina.3.4.1BackgroundinformationofMacaoMacaoisaspecialadministrativeregiononthewesternsideofthePearlRiverestuaryinSouthernChina.Ithasthreeislands:MacaoPeninsula,TaipaandColoane,wherebythelattertwoislandsarereclaimedintoone,withthereclamationareaknownasCotaiReclamationzone.AccordingtothelatestestimatesfromtheUN’sWorldPopulationProspects(UnitedNations,2019),itspopulationhasreached640,000asof2019.Withjust32.9km2,populationdensityisestimatedtobe21,420peopleperkm2,makingitthesecondmostdenselypopulatedarea/regionintheworld.However,with85%ofthepopulationlivingintheMacauPeninsula,themainislandwithanareaofjust9.3km2(excludingNewDistrictZoneA),populationdensityisestimatedtobearound58,000perkm2(Minetal.,2011).Someareas,accordingtoastudydonebyMinetal.(2011),haveapopulationdensityover120,000perkm2(Fig.10).Insuchahighly-denseurbanenvironmentwithhighvolumesofdailywastewatereffluent,effectiveandsustainablewastewatertreatmentisfundamentaltoreduceenvironmentalimpactsandtocirculateresourcesforreuseinthenetworkofecosystems.

87Fig.10.PopulationdensitymapofMacaoPeninsula.P1-P17indicatesdifferentregionsofMacao,colour-codedwithspecificpopulationdensityrange(perkm2).(Source:Minetal.,2011).Asof2020,thereareatotaloffiveconventionalwastewatertreatmentplants(WWTP)inMacao,oneislocatedoneachofthethreeislands,anotheroneisforindustrialwastewatertreatmentpurposeandthelastoneisasmall-scalewastewatertreatmentstationspecificallyfortreatingwastewaterfromtheMacaoInternationalAirport.Thecombinedplantsandstationsarecapableoftreating356,000m3ofwastewaterdaily(DSPA,2020).Inthepastdecade,waterconsumptionhasincreasedduetothepopulationincreaseandtheincreasedconsumptionfromtheentertainmentfacilitiesintheCotaiReclamationZone.Thecommercialanddomesticsectoraccountsfor88.7%ofbilledwaterconsumptioninMacaoin2018,whereastheindustrialandpublicsectoraccountsfor5%and6.3%respectively;MacaoPeninsula(themainisland)aloneaccountsfor500-500030000-5000050000-120000>1200005000-30000

8862.6%ofwaterconsumption,andtheCotaireclamationzonemakesup17.6%(DSPA,2018).Duetotheincreaseinwaterconsumption,thevolumeoftreatedwastewaterhasalsoincreased;themeanvaluesofthedailytreatedwastewatervolumebythecombinedWWTPinMacaofrom2009-2018isshowninFigure11.ThemeanvalueofthecombinedtreatedwastewatervolumeofMacaohasincreasedby5.9%in2018comparedtothemeanvaluein2017(Table3).Fig.11Dailymeantreatedwastewatervolumefrom2009-2018.(Source:DSPA,2018)Notes:1.Percentageshownisthevariationbetween2017and2018.2.Thegreylineisthetrendlineofthetotaltreatedvolumeofwastewater3.ThetreatedvolumeofTaipaWWTPincludeswastewaterfromMacauInternationalairport4.TIPWWTS-TransborderIndustrialParkWastewaterTreatmentStation

89Table3.NumericaldataofdailymeantreatedwastewatervolumebyWWTPandWWTS(wastewatertreatmentstation)inMacaobetween2017and2018.(Source:DSPA,2018)Originally,theWWTPintheMacaoPeninsulawasbuilttotreataround140,000m3ofdailywastewaterinfluent(approximately4,320,000m3monthly)byconventionalprimaryandsecondarytreatmentprocesses(DSPA,2020).In2019,theaveragemonthlywastewaterinfluentoftheMacaoPeninsulaWWTPwas4,391,192m3(DSPA,2019);intheory,theplantshouldbeabletotreatallofthewastewaterinfluent.However,onaverageonly50%ofthewastewaterinfluentin2019underwentsecondarybiologicaltreatment;theremaininghalfofthewastewateronlywentthroughprimarytreatment(coarsegridfiltrationandsedimentation)andwasdirectlypipedoutintothecoastalwaters(DSPA,2019)(Fig.12).ThisisduetotheincapabilityoftheWWTPtotreattheoriginallycalculatedwastewatervolume.Therefore,inpractice,halfofthewastewaterinfluentoftheMacaoPeninsulaWWTPhasnotbeentreatedbiologicallybeforedischargingintothesurroundingwaterbodies.Inaddition,theeutrophicationindexrecordedatseveralmonitoringpointshasincreasedin2018comparedwiththe2017data(DSPA,2018).

90Fig.12.Percentageof2019wastewaterinfluentofMacaoPeninsulaWWTPthatunderwentsecondarytreatment.(Source:DataretrievedfromDSPA,2019)Hence,withtheincreaseinwaterconsumptionandtheincapabilityoftheMacaoPeninsulaWWTPtocopewiththecurrentwastewaterinfluentvolume;anurgentexpansionplanoralternativesourceofwastewatertreatmenttechnologyhastobeexplored.Urbanconventionalwastewatertreatmentplants(WWTPs)consumeahighamountofenergyinput.MostoftheWWTPsintheworldaredesignedtomeettherequirementsoftheeffluentqualitywithoutconsideringenergyrequirements.Electricityinputinwastewatertreatmentactuallyaccountsfor3%oftheworld’selectricityconsumption(Lietal.,2015).Therefore,WWTPexpansioninMacaoisunsustainableas:(1)Itincreasescarbonemissionsandfurtherexacerbatesglobalwarming;(2)expansionwouldbeanever-endingprocessaspopulationcontinuestoincreasewhileavailablelandremainsscarceinMacao.

91Accordingtothe“DevelopmentPlanoftheGuangdong-HongKong-MacaoGreaterBayArea”,weshouldadoptaninnovative,greenandlow-carbondevelopmentmodelbyreducingenergyandmaterialconsumption,recyclingresourcesandimplementingwaterconservationactions.ComparedtoWWTP,constructedwetlands(CW)wereoftenshownbylifecycleassessmentsasaverylow-impactandcost-effectivealternativetechnology(Machadoetal.,2007;Roux.etal.,2010).Economically,CWsarecompetitiveasitcostsone-thirdtoone-halflessthanWWTPperkgofNremovedinChina(Liuetal.,2012).Therefore,wewouldexplorethepossibilityofCWimplementationinMacaointhefollowingsections.Firstly,calculationsofCWdimensionsfortheprimary-treatedwastewaterwouldbeperformed;aCWdesignforMacaoimplementationwouldthenbeproposedbasedonthefindingsintheprevioussections,finallylocationswouldbesuggestedforpilotCWimplementation.3.4.2CWdimensioningforMacaoForVFCWsinwarmerclimates,studieshaveshownthatlowersurfaceareaisrequiredforthesametreatmentefficiency(Langergraberetal.,2007).Intemperateclimates,amaximumorganicloadingrateof20gCODm-²·dcanbeapplied.Weaimtoexaminethesurfacearearequiredtotreatthe70,000m3ofdailywastewaterthatonlyundergoesPrimarytreatmentintheMacaoPeninsulaWWTPwhichthenwillbedirectlypipedoutintothecoastalwaters.Accordingtosection2.2intheaforementionedmethodology,thecalculationforaVFCWthatsecondarytreats70,000m3ofwastewaterisasfollows:•Qi(Dailyinfluentflow)=70,000m3d-1=70,000,000ld-1

92AverageDailyrawCODconcentrationin2019intheWWTP=378mg(DSEC,2019b)In2019,theaverageeffluentfromtheMacaoPeninsulais205mgl-1,whichexceedsthemaximumdailyCODemissionstandardof150mgl-1(DSEC,2019b).Assumethatone-thirdofCODconcentrationisremovedinprimarytreatment(DWA,2017),theCODconcentrationofthepost-primarytreatedwastewaterwouldbe:•378mgCODl-1x(⅔)=252mgCODl-1=0.252gCODl-1•OrganicLoad:70,000,000ld-1x0.252gCODl-1=17,640,000gCODd-1AccordingtotheGermanguideline(DWA,2017),20gCODm-²d-1oforganicloadingratecanbeassumedfortemperateclimateifsandisusedassubstrate.Thiscanbeincreasedifappliedtowarmertropicalclimates.Stefanakisetal.(2014)haveindicatedthatorganicloadingratecanincreaseupto80gCODm-²d-1inwarmerclimatessuchasthatshowninCzechRepublic(VymazalandKröpfelová,2008).Onaverage,MacaohasahighermonthlytemperaturecomparedtoCzechRepublic,anOLRof80gCODm-²d-1willbeassumed.EstimatedVFCWsurfacearea=•17,640,000gCODd-1÷80gCODm-²d-1=220,500m2=0.2205km2MacaoPeninsula’ssizeisapproximately9.3km2,0.2205ofVFCWaccountsforaround2.37%ofitsland.TheareaperPE(Populationequivalent)iscalculatedasfollows:AveragedailyrawBOD5concentrationin2019intheMacaoPeninsulaWWTP=178mgl-1(DSEC,2019b)

93Assumethatone-thirdofBOD5concentrationisremovedinprimarytreatment(DWA,2017),theBOD5concentrationofthepost-primarytreatedwastewaterwouldbe:•178mgl-1x(⅔)=119mgBOD5l-1=0.119gBOD5l-1Assumethat1PEis60gBODd-1(Langergraberetal.,2007)andeffluentis70,000,000ld-1:•70,000,000ld-1x0.119gBOD5l-1=8,330,000gBOD5d-1•8,330,000gBOD5d-1÷60gBOD5d-1=138,833PETheareaperPEofa0.2205km2ofVFCWs:•220,500m2÷138,833PE=1.59m2PE-1Asaforementioned,intensifiedCWdesignscouldhelptoreducetheoverallCWlandrequirement.Tidalflow-VFCW(2.1±1.8m2PE−1),artificialaeration-HCW(3.0±2.1m2PE−1),andeffluentrecirculation-HCW(4.0±4.3m2PE−1)aresuggestedasthebestoptionsforgoodwastewatertreatmentperformancewithsmallfootprint(Ilyas&Masih,2017).VFCWwouldbeconsideredfirstasitwouldrequirelessmaterialsandpartstobuildcomparedtoHCW.Astudywasconductedtoinvestigatetheperformanceresponseoflab-scaleTidalflow-VFCWandconventionalVFCWtodifferentorganicloadingrates(OLR)(Duetal.,2016).IthasbeenfoundthatincreasedOLRpromotedCODandNH4+-NremovalefficiencyinTF-VFCWwithOLRof751.2gm-²d-1.ForconventionalVFCW,oxygenwithinthesystemisnotsufficienttosupportboththeexcessorganicreactionandnitrification,thereforeonly40%ofNH4+-NremovalwasobtainedunderOLR751.2m-²d-1,thisalsointurnleadtoadropofTNremovalduetodecreaseinNO3-inconventionalVFCW.Therefore,wecouldproposeahigherOLRthantheprevious80gm-²d-1forTF-VFCW.However,asDuetal.(2016)conductedthestudyinlab-scale,a

94pilotscalesystemshouldbetestedasproblemssuchascloggingwouldariseduetohigherOLR.Apilot-scalestudyconductedbySunetal.(2006)hasachievedanareaof0.3m2PE−1forTF-VFCWwithorganicloadingrateof330gCODm-²d-1.IfweuseOLR330gCODm-²d-1,asmallersurfacearearequirementcanbeachieved:EstimatedTF-VFCWsurfacearea=•17,640,000gCODd-1÷330gCODm-²d-1=53,455m2=0.0535km2MacaoPeninsula’ssizeisapproximately9.3km2,0.0535ofVFCWaccountsforaround0.58%ofitsland.Estimatedarea(m2PE−1)=•53,455m2÷138,833PE=0.39m2PE−1AsconventionalWWTPhasanestimatedlandrequirementof0.2-0.5m2PE−1(vonSperling,1996),thisTF-VFCWwith0.39m2PE−1makesithighlycompetitiveintermsoflandrequirement.Nevertheless,cloggingpresentsamajorproblemespeciallyifhighOLRisused.OneofthemajorgoalsindesigningaCWisthatitwouldrequireaslittlemaintenanceaspossible.Topreventandreduceclogging,wecouldadoptthestrategiesinSection3.3.3.4.3FinalconstructedwetlanddesignforMacaoAfinalCWdesignforMacaowouldbeproposedinthissection.Ideally,itwouldhavearelativelysmallfootprint,recoverablematerialswithlessenvironmentalimpactandgenerateslesssecondarywaste.

953.4.3.1CWdesignguidelinesfromLCAstudiesBeforeselectingCWmaterialsforMacao,someimportantfactorswerehighlightedfromtheLCAstudiesinSection3.2.AlthoughnotmanyLCAstudiesarereviewed,wecanstillobtaininstructiveCWdesignguidelinesfromtheLCAstudiesdiscussed:1)VFCWhaslessenvironmentalimpactandsmallerfootprintthanHFCW.2)ConstructionphaserequireshigherenergyconsumptionthanOperationphase,thereforeexertsgreaterenvironmentalimpact.3)Filtermaterialusedcouldimmenselyaffectoverallenvironmentalimpactasshowninsensitivityanalysis.4)Transportation,secondtoconstructionmaterials,alsocontributesgreatlytooverallenergyconsumption.5)Duringoperationphase,aerationtechniquesthatrequireelectricityforpumping(e.g.tidalflow,artificialaeration)couldbethemaincontributortoenvironmentalimpact.6)CouplingbiofuelproductiontoCWcouldleadtonegativenetenergybalancewithenergyoutput.ItcouldalsomakeCWevenmoreeconomicallyfeasible.Therefore,whendesigningaCWforMacao,weshouldchooseVFCWoverHFCWespeciallywhenlandisscarceintheMacaoPeninsula.AsshowninthecalculationinthepreviousCWdimensioningsection,tofurtherreducetheareaperPErequiredforCW,TF-VFCWisconsideredtobecompetitiveintermsoffootprint.However,fromtheLCAstudies,aerationtechniquesthatrequireelectricityisoneofthemaincontributorsofenvironmentalimpactinoperationphase.Abalanceshouldthereforebeconsideredbetweenreducingfootprinttoaviablesizeandminimizingenvironmentalimpact.However,withoutintensifiedtechniques,itwouldbeinfeasibletoimplementCWsindenseMacao.AssuggestedintheLCAstudies,CWoperationphaseexertsless

96environmentalimpactthanconstructionphase,insteadtheselectedCWmaterialsandtheirtransportation(localavailabilityofmaterials)arethemaincontributorsthataffectstheoverallenvironmentalimpactoftheCW;therefore,inthiscasestudy,intensifiedCWdesignsthatrequireelectricitytooperatewouldbeconsidered.MoresustainablefiltermaterialsthatareavailableinMacaowillbeselectedinthenextsection.3.4.3.2SubstrateselectionforpilotCWinMacaoInordertocouplewastewatermanagementandcirculareconomywhenselectingasubstrateforaCWinMacao,wehavetofocusonselectingsubstratesfortheir:(1)pollutantremovalcapacityand(2)potentialtoachievecircularwatermanagement.AsidefromfocusingonP-removalduringsubstrateselection,weshouldalsoconsidertheenergyinput,localavailabilityandsecondarywastegenerationwhenchoosingasuitablesubstrateforCW.Table4hassummedupsomeofthemainfindingsinsection3.1regardingdifferentmaterialsinP-retention,Precoveryandfilterrecyclability.

97Table4.Propertiesofdifferentsubstrates.MaterialMaximumP-retention(gPkg-1/Removal%)Energyinput****AvailabilityinMacao*****P-recoveryforplantuptake(Method)FilterrecyclabilitystudytypeReferenceIndustrialby-productBlastfurnaceslag420.0(a)(gPkg-1)0NY(b)(usewithoutprocessing)Y(c)batch(a)(a)MannandBavor,1993(b)Hylanderetal.,2006(c)Drizoetal.,2002Hydratedsedimentofoilshaleash65.0(a)(gPkg-1)10NY(b)(usewithoutprocessing)---batch(a)(a)Kaasiketal.,2008(b)Koivetal,2012Unheatedredmud113.9(a)(gPkg-1)0NN*(b)N(b)batch(a)(a)Lietal.,2006b(b)Yangetal,2013CAAC57.0(a)(gPkg-1)0YY**(b)(usewithoutprocessing)---pilot(a)(a)RenmanandRenman,2012(b)Bergetal.,2006;Mkumboetal.,2019De-wateredalumsludge0.0423(a)(gPkg-1)0N(b)Y**(c)(addingH2SO4)---column(a)(a)Babatundeetal,2010(b)DSPA,2020(c)Zhaoetal.,2009

98Magnesiumslagparticle90%(a)(PO43−-PremovalEfficiency)0NY(b)(struviteprecipitation;usewithoutprocessing)---column(a)(a)Etteretal.,2011(b)Tangetal,2017FragmentedMoleanoslimestone(FML)61%(a)(PO43−-PremovalEfficiency)0N------Pilot(a)(a)Mateusatal.,2012NaturalmaterialsSand0.117(a)(gPkg-1)0Y***(b)------full-scale(a)(a)Vohlaetal.,2007(b)DSEC,2019cGravel4.33%(PO43−-PremovalEfficiency)0Y***(b)------full-scale(a)(a)Korkusuzetal,2005(b)DSEC,2019cDolomite9.7-52.9(a)(gPkg-1)0NN*(b)N(b)batch(a)(a)Karacaetal.,2004(b)Yangetal,2013Zeolite2.15(a)(gPkg-1)0N------batch(a)(a)SakadevanandBavor,1998

99Man-madeproductsLab-madeLWA12(a)(gPkg-1)Manufacturedat1200oC(b)N------batch(a)(a)JenssenandKrogstad,2003(b)EPA,1985Tabletporousmaterial(TPM)4.39(a)(gPkg-1)Manufacturedat600oC(a)NY**(a)(HCladdition)---column(a)(a)Yangetal,2013HeatedRedmud345.5(a)(gPkg-1)Manufacturedat700oC(a)NN*(b)N(b)batch(a)(a)Lietal.,2006b(b)Yangetal,2013Biochar0.208(a)(gPkg-1)Manufacturedat400oC(a)NY**(b)(usewithoutprocessing)---pilot(a)(a)Boltonetal.,2019(b)Taghizadeh-Toosietal.,2012Note:OnlymaximumP-retentiondataofeachmaterialisshown(ifgPkg-1isnotusedintheresearchpaper,P-removal%ispresentedinstead).“Studytype”referstostudiesconductedonP-retentioncapacity.*Extremelydifficulttoseparate,furtherpotexperimentsrequiredtotestitsdirectapplicationasfertilizers.**Furtherpotexperimentsrequiredtotestitsdirectapplicationasfertilizers..***AvailableinMacao,butmostlyimportedfromChina(DSEC,2019c).****Onlyproductionenergyinputisconsidered(transportationandextractionenergyinputarenotconsidered).*****Otherwisespecified,dataavailableinthe2018IndustrialSurveybytheMacaoStatisticsandCensusServiceDepartment(DSEC,2019a)Y:YesN:No---norelatedstudieswerefound.

100Threetypesofsubstrate:industrialby-product,naturalmaterialsandman-madeproductswereshowninTable4.ForP-retentioncapacity,blastfurnaceslagandheatedredmudperformedthebest.However,heatedredmudrequirestwohoursofheatingat700oC(Lietal.,2006b)whichincreasestheenergyinputandthushasagreaterenvironmentalimpact.Energyinputshownintable4onlyconsidersenergyrequiredinproductionanddoesnotincludeenergyinputduringdiggingandtransportation.Theidealsubstrateshouldrequireminimalornoinputofenergytoobtain;therefore,man-madeproductsarenotrecommendedastheywouldonlypropagateenvironmentalproblems.Onthecontrary,selectingindustrialby-productswouldhelptoreduceupstreamwastegenerationwithoutfurtherenergyinput;howevernotalloptionsareavailableinMacao.AstherearenoheavyindustrialactivitiesinMacao,onlyCAACisavailablelocallyduetobuildingconstruction(DSEC,2019a).CAACwastestedtoobtainover90%ofP-removalcapacity(RenmanandRenman,2012).Thoughoftenavailableinwastewatertreatmentplants,DASisnotavailableasMacaowastewatertreatmentplantsdonotusealuminumforcoagulation(DSPA,2020).BlastfurnaceslagismostlyproducedinNorthernandmid-China,asearchonlineshowsthattheclosestwholesalersareatleastaseven-hourdrivefromQuanzhouintheFujianProvinceofChina;thefuelrequiredfortransportationwouldincreasetheenvironmentalimpactduringCWconstructionphase.TosupportacirculareconomyandtoreducesecondarywastegenerationbyCWs,thesaturatedfiltershouldberecycledandfurtherselectedforP-recovery.AlthoughmostmaterialsallowP-recovery,someprocessesrequirehigherenergyinputcomparedtoothers.Forexample,DASrequiresadditionofsulphuricacidtorecoverPandfurtherneutralizationbeforeitcanbeusedforsoilamendmentorfertilizers(Nodirectapplicationstudywasfound),thisconsumeschemicalreagentthatrequiresenergyinputduringproduction(Zhaoetal.,2009);

101whereasblastfurnaceslag,oilshaleash,MgslagparticleandpotentiallybiocharandCAACcouldbeappliedwithoutfurtherprocessingandenergyinput(Bergetal.,2006;Hylanderetal.,2006;Koivetal.,2012;Taghizadeh-Toosietal.,2012;Tangetal.,2017).Saturatedblastfurnaceslagwasshowntobeabletorecycleasfiltermedia(Drizoetal.,2002),butfurtherexperimentswouldhavetobeperformedforothermaterials.Thisproducesminimalsolidwasteandisabletorecyclenutrientsbacktotheland,completingthenutrientcycle(Tangetal.,2017).Overall,weshouldconsiderindustrialby-productsforsubstrateselectiontoreduceenergyinput.Secondly,minimalenergyinputforP-recoveryfromsaturatedfiltermediashouldalsobeconsideredtofurtherdecreasenetenergyinput.Thirdly,locally-sourcedby-productsshouldbeusedtoreduceenergyinputduringtransportation.CAACpotentiallysatisfiedallthecriterias:ithasgoodP-retentioncapacity,couldbesourcedinMacaoandrequiresminimalenergyinputforbothproductionandP-recovery,essentiallyrecyclingPandclosingthenutrientcycle.Therefore,CAACwouldberecommendedassubstrateforpilotCWsinMacao.3.4.3.3VegetationSelectionforpilotCWinMacaoAsdiscussedabove,vegetationcoulduptakenutrientsfromwastewater,sequestercarbonandevenoffsetgreenhousegasemissions.SelectingspeciesthataretoleranttofloodandareabletobeconvertedtobioenergywilladdtothelistoffreeecosystemservicesthatCWprovides.Harvestedvegetationwouldnotgoto“waste”asabsorbednutrientswouldnotre-enterCWasplantdecayandbioenergycouldbeproduced,incorporatingtheconceptofcircularsustainability.AsMacaohasbeenpronetofloodingduringtyphoonseasons,pilotCWswouldrequirediverseflood-tolerantspecies.Table5hassummarizedsomeofthefindingsinsection3.1.2,differentcriteriaforvegetationselectionwereshown.

102Table5.PropertiesofdifferentvegetationspeciesVegetationspecies(a)TestedtotolerateCODconcentration(mgl-1)(b)NPhyto-uptake(gm-2)(c)HemicelluloseandCellulosecontent(%)(d)Biomass(kgm-2)(e)Calorificvalues(MJkg-1)(f)Bioenergyfeedstock(Max.Productivity,GJha-1yr-1)(g)Floodtolerance*(h)Invasiveness**AestheticvaluesReference(a-hreferstodifferentcolumncatergory)Typhalatifolia(Bulrush)201030.4601.7618326tolerantaggressiveperennialwithflowers(a)Calheirosetal.,2007(b)Rebaqueetal.,2017(c)Ciceketal.,2006(d)Maddisonetal.,2009(e,f)Liuetal.,2012(g)Gucker,2008(h)Davis,1995Phragmitesaustralis(Commonreed)201073.79562.7181500toleranthighlyaggressiveperennialgrass(a)Calheirosetal.,2007(b,c,d)Zhaoetal.,2014(e,f)Liuetal.,2012(g)Pagteretal.,2005(h)Davis,1995

103Arundodonax(Giantreed)42588.87605.59181836tolerant,canre-establisheasilyfollowingstormhighlyaggressivereed-likegrass(a)Calheirosetal.,2012(b,c,d)Zhaoetal.,2014(e,f)Liuetal.,2012(g)Mannetal.,2013(h)Lambertetal.,2010Cannaindica(Cannalily)110045451.329.521064toleranttowidetemperature,butnotwaterloggingeasilybecomeinvasivetosensitiveareasfloweringornamentalplants(a)Haritashetal.,2015(b,c,d)Zhaoetal.,2014(e,f)Liuetal.,2012(g)Zamoraetal,2019(h)BatianoffandButler,2002Irispseudacorus(YellowIris)1156------3.3---170tolerantaggressivefloweringornamentalplants(a)Tangahuetal.,2019(d)HernándezCrespoetal.,2015(f)Liuetal.,2012(g)SchlüterandCrawford,2001

104(h)Hayasakaetal.,2018Saccharumofficinarum(Sugarcane)713---NANA23.2684tolerantgenerallynon-invasivePerennialgrass(a)Diresetal.,2019(e)Lomeda-DeMesaetal.,2019(f)Longetal.,2015(g)Viatoretal,2012(h)Cheavegatti-Gianottoetal.,2011Zeamays(Corn)1450---NANA18331tolerantgenerallynon-invasivegrass(a)JosephSahayarayanetal.,2019(e)Ambrosioetal.,2017(f)Longetal.,2015(g)Zaidietal.,2004(h)OECD,2003*Assumetoleranceatmaturestage,notearlyseedlingstage**AllspeciesarenotfoundinthedatabaseofDSPA(EnviornmentalProtectionBureauMacao)andnotfoundinthedatabaseofthebook"".Theinvasivenessisestimatedaccordingtothespecies'nature.NA-Notapplicable---Relatedstudieswerenotfound.

105StudieshaveshownthatallthespeciesinTable5weretoleranttoCODconcentrationabove378mg/LwhichwasthedailyaverageCODconcentrationofthewastewaterenteringMacaoPeninsulawastewaterplantin2019(DSEC,2019b).Sugarcaneandcornwerefirst-generationbiofuelsandarenon-invasivespecieswhiletheothersaresecond-generationbiofuelsandareallinvasivewitheasyseedpropagationespeciallyafterfloodevents.Theyarealltoleranttofloodingforashortperiodoftimeandareallassumedtobenon-nativetoMacao.SincenoneofthemarenativetoMacao,selectingvegetationwouldhavetobedonecarefullyashighlyinvasivespeciescouldleadtodecreaseinlocalbiodiversity.BothP.australisandA.donaxwereshowntohavethehighestNphyto-uptakeinplantsandhavethehighestmaximumproductivityasbioenergyfeedstock,andC.indicarankedthirdintermsofNphyto-uptakeandbioenergyfeedstockandisanornamentalplant.However,P.australis,A.donaxandC.indicaareallinvasiveinnatureandarenotnativetoMacao.Therefore,theycouldnotbeconsideredforCW.Ifnon-invasivespecieswerepreferred,sugarcaneandcorncanbeselectedbuttheirNphyto-uptakewouldneedtobefurtherinvestigatedasnoexistingdataisavailable.Theirproductivityisalsolowerthanthethreementionedinvasivespecies.Asamixtureofspecieswouldbemorefloodresilient,amixcornandsugarcanecanbothbeconsidered.Overall,selectingplantssuchassugarcaneandcornthatarenotinvasive,canbeusedasbioenergyfeedstockandarefloodresilientwouldincorporatetheideaofcircularwatermanagement,energyproductionintourbanwastewatertreatmentinMacao,butfurtherstudiesintheirpollutantremovalcapacityhavetobeperformed.

1063.4.3.4LinerSelectionforpilotCWinMacaoForlinerselection,insteadofusingHDPEgeomembraneorPVClinerthatwereshowntoincreaseenvironmentalimpact(Gkikaetal.,2015),wecouldusecompactedclayorbyproductsuchasFlue-gasdesulfurization(FGD)materialasanalternative,butfurthertoxicitystudieshavetobeperformedonFGD’stoxicity.3.4.3.5FinalCWdesignforMacaoSofar,wehaveevaluatedthemaincomponentsforconstructingaCWinMacao:substratematerials,vegetationspeciesandlinermaterials.FromdifferentstudiesandLCAanalysis,wehaveconcludedwithselectionsthatwouldhavetheleastenvironmentalimpactandweremostsuitedforMacaoimplementation:Substrate:Crushedautoclavedaeratedconcrete(CAAC)Vegetation:Sugarcaneand/orcornLiner:CompactedclayCAACcouldbeselectedassubstrateduetoitslocalavailability,P-removalcapacityanditsby-productnature.Itcouldpotentiallybeusedasfertilizer.Sugarcaneandcorncanbeselectedforbioenergyfeedstocktocouplebiofuelproductionwithwastewatermanagement.Forlinerselection,compactedclaycouldbeusedinsteadofPVCliners.Inthiscase,overallenvironmentalimpactofCWshouldbeminimized.However,Macaohasthesecondhighestpopulationdensityintheworld(UnitedNations,2019).TheconceptofimplementingconstructedwetlandstothisconcretecityofMacaohasneverbeenproposed.ItwaspreviouslythoughtinfeasibletouseCWasawayoftreatingwastewaterinurbanareas;however,byreviewingsomeofthemostrecentCWengineeringdesignsinsection3.3,CWswereshowntoreducetosizethatismuchclosertotheconventionalapproach(IlyasandMasih,2017).

107Over-engineering,however,shouldbeavoidedasdesignsshouldbekeptassimpleaspossiblewithminimalmaintenance(Davis,1995).AccordingtoIlyas&Masih(2017),tidalflow-VFCW(2.1±1.8m2PE−1),artificialaeration-HCW(3.0±2.1m2PE−1),andeffluentrecirculation-HCW(4.0±4.3m2PE−1)arecalculatedasthebestoptionsforgoodwastewatertreatmentperformancewithsmallfootprint.Tokeepthedesignassimpleaspossible,wewouldrecommendTF-VFCW.ThisdesignwascalculatedtooccupylessareaperPEinsection3.4.2thanVFCW.Also,itdoesn’trequirehorizontalfilterorstackingofbeds,thereforewouldrequirelessconstructionmaterialsthanHCW;subsequently,environmentalimpactwouldbetheleastasconstructionmaterialsisthemaincontributortoenvironmentalimpactaccordingtoseveralLCAstudies(Corbellaetal.,2017;Gkikaetal.,2015).Moreover,TFconsumes50%lesselectricitycomparedtoAAoperations(Ilyas&Masih,2017);electricityconsumptionisanothergreatcontributortoenvironmentalimpactsinCWoperationphase(Babatundeetal.,2010).Therefore,withthebasisofLCAstudies,TF-VFCWshouldhavetheleastenvironmentalimpactoutofthethreeintensifiedCWdesigns.Nevertheless,anLCAstudyhasshownthataTF-VFCWstillconsumesahighamountofelectricity(2.36kWhperFU(1m3))andtheCO2emittedfromtheelectricitygenerationhasovertakenthesequesteredamountfromplants(Wangetal.,2018).Nonetheless,theTF-VFCWconsumptionofelectricityislowincomparisontoWWTP.AZhouShanWWTPwithsimilardailytreatmentdesignscale(150,000m2)astheMacaoPeninsulaWWTP(140,000m2)wascomparedwithaconventional1000m2VFCW(Liuetal.,2019);itwasreportedthattheWWTPhasnoenergyoutputandenergyinputis40timesmorethanthefull-scaleVFCW.Also,theVFCWemits596.6MgCO2ha-1yr-1lessthanWWTPandemitsthreeordersofmagnitudeofCH4lessthanWWTP.Intotal,theVFCWcanreduceGHGemissionsby2714timeslessthanthatof

108theWWTP(Liuetal.,2019).Therefore,thoughrequireelectricitytooperate,TF-VFCWstillhaslessenvironmentalimpactthanWWTP,butwithincreasedefficiencyandlowerlandrequirementthanconventionalVFCW.TofurtherincreaseTF-VFCW’sefficiency,wecouldcombinethisdesignwiththosetechniquesdiscussedinsection3.3:a)Multipleharvestofbiomassb)AlayerofgravelsisaddedontopoftheCW.c)Manualadditionofearthwormsonthesurfaceofthewetlandbed(e.g.Pheretimapeguana)Sinceweplantocouplebiofuelproductionwithwastewatermanagement,harvestingofvegetationisexpected,thiswouldalsopreventthere-releaseoftheuptakednutrientsintotheCW.BioaugmentedCWwouldincreaseCWmicroorganismsandincreasepollutantremovalefficiency,howeveritcouldincreaseimpactinglobalwarmingpotential(GWP)duringinoculation(Zhaoetal.,2017).Therefore,forpilotCWimplementation,bioaugmentationisnotyetrecommended.Finally,cloggingsometimesposesaproblemforsubsurfaceCWswhichincreasesmaintenancecostandcausestreatmentefficiencytodecrease.Coarsesubstratesontheupperlayercouldhelppreventclogging.Athinlayeroflargegravelsthataremorethan>10mmindiametercanbeapplied(Dotroetal.,2017).Analternativewayistheadditionofearthworms.Earthwormsdirectlyingestanddigestorganicmaterials;Itwasshownthatearthwormshelptoreducesludgeproductiononthesurfaceofwetlands40%byvolume,inturnreducingthecostoftreatingCWsludge(Nuengjamnongetal.2011).EarthwormsarefoundalmostworldwideandareabundantinChina.LocalearthwormscanbemanuallyspreadevenlyoverthesurfaceoftheTF-

109VFCWwhichtheywouldburrowintothebedastheyaresensitivetosunlight,foodintheformofwastewaterisabundantinthebed.Overall,ouroptimizedTF-VFCWdesignisshowninfigure13:Fig.13.OptimisedTF-VFCWtoreducelandrequirementandincreasetreatmentefficiency.(Source:AdaptedfromWuetal.,2015a)OuroptimizedTF-VFCWwouldhaveatoplayerofgravelsandearthwormswereappliedtoreduceclogging.CAACby-productandcompactedclaylayerwouldbechosenassubstrateandlinerrespectively.Sugarcaneand/orcorncouldbeusedforvegetation,theirperiodicharvestallowsbiofuelproductiontobeachieved.ThisCWdesignaimedtohavetheleastenvironmentalimpactpossiblewhilebeingintensifiedtoreducelandrequirementsothatitcanbeimplementedinurbanareas.Itisaimedto

110coupleurbanwastewatermanagement,materialcircularsustainabilitywithbioenergyproduction.3.4.4LocationsforCWimplementationinMacaoInsection3.4.2,itwascalculatedthatthecombinedarea0.0535km2ofTF-VFCWcouldtheoreticallytreat70,000m3ofprimary-treateddailydomesticwastewatertoarequiredeffluentstandard.0.0535km2isequivalentto0.58%ofMacaoPeninsula’sarea.LargeVFCW(>100P.E.)commonlyhasdividedsurfaceareaintosmalleronessothattheycanbeindependentlyloaded.Ideally,thesenumberofTF-VFCWsshouldbelocatednexttotheWWTPtoreceivetheprimary-treatedwastewater.However,evenwiththereducedsize,0.0535km2ofunoccupiedlandisnotavailablenexttotheWWTP.Therefore,insuchahighlydensecity,adecentralizedTF-VFCWsystemshouldbeconsidered.AccordingtotheMacaostatisticsdepartment(DSEJ,2019),thecombinedareaofgreenspacefortrafficinfrastructureinMacaoPeninsulais211,475m2;andthecombinedareaofgreenspaceforleisureandrecreationis665,760m2.Thiscouldtheoreticallycovertherequired53,455m2(0.0535km2)ofTF-VFCWs.However,decentralizedsystemwouldrequireextraspaceforsedimentationtanksasthewastewaterwouldnothaveundergoneprimarytreatmentiftheywerenotsetupnexttotheWWTP.Also,realistically,notallofthesegreenspacesfortrafficinfrastructureandleisurearesuitableforCWimplementation;someoftheseareascontainoldtreesthatsupportanetworkofurbanecosystems,whileothersaresimplyforpublicrecreation.ToidentifylocationsforimplementingourTF-VFCWsinMacao,weshouldconsiderthefollowing:(a)implementattheopenareasorparksavailableinMacaoPeninsula;

111(b)CWsshouldideallybelocatedatnearbywaterbodiessothattreatedwatercanbedirectlydischargedwithaminimaleffluentpipingsystem;(c)CWsshouldideallybelocatednearplacesthatcanreusetreatedwastewater;(d)Macaoispronetofloodingduringtyphoonsummerseason,domesticCWsshouldthereforeavoidimplementingonlow-lyingareas.AsshowninFigure14a,thereareatotalof608greenpatchesinMacaoPeninsula,ofwhich365(over50%)arebetween100and1000m2(Minetal.,2011).AccordingtotheAustrianGuideline(ÖNORM,2009),themaximumVFbedsizeisdeterminedtobe400m2forevendistributionofwastewater.Theopengreenspaceofsizerangingfrom100m2tomorethan50,000m2inMacaoPeninsulaisshowninFig14b(modifiedfromMinetal.,2011).

112Fig.14.(A)Numberofgreenpatcheswithindifferentsizecategories(adaptedfromMinetal.,2011)(B)GreenpatchesinMacaoPeninsulaoverlaidwithstormwaterrunoffchannelsandSSW-1affectedarea.(Source:AdaptedfromLietal.,2019;Minetal.,2011)AsshowninFigure15,themostflood-proneareaofMacauiswherePortoInteriorandHongKungTempleislocated,thisareawillbeaffectedbytheleastseverestormsurgelevel(SSW-1)(Lietal.,2019).TheSSW-1affectedareaisoverlaidontopofFigure14balongwiththestormwaterrunoffchannelsidentifiedbyLietal.(2019).PilotTF-VFCWsshouldnotbeconsideredtobesetupattheselocationstoavoidfloodingatSSW-1.HongKungTemple

113Fig.15.Spatialextentaffectedbyurbanfloods:SSW-1floodingsimulationresults(0-0.5mfloods)(Source:adaptedfromLietal.,2019).AsshowninFig.14b,openspacesuchasthesurroundinggreenpatchesoftheMountFortresscanpotentiallybetestedforCWimplementation.Generally,therearenotmanyopenareasalongtheWesternStripofMacaoPeninsula,anotherlargeropenspaceisCamoespark(between10,000to50,000m2)whichcanbesuggestedforapplyingsmalldecentralisedTF-VFCWs.FerreiraAmaralSquareisaroundaboutwhichcouldbesuggestedforTF-VFCW;treatedwatercanthenbeeasilypipedoutintothenearbySaiVanlake.OtheropenspacesincludeGardenoftheArtsandDr.Carlosd’AssumpçãoPark,theseparkscontaintoiletswherebywastewatercanbetreatedbytheTF-VFCWs,treatedeffluentcanthenberecycledforwateringplantsorflushingtoilets.Otherscouldbesetupalongroadside(withlittlepublicaccess)andsmallerroundaboutsfortreatingdomesticwastewatergeneratedbynearbyfive-storeyresidentialbuildings.Extensiveplanninganddesigningthatincludesre-arrangingpipingsystemshastobeperformedbeforeanypilotsystemisimplemented.PatanePortoInteriorHongKungTemplePonteeHorta0-0.5mFloodsBuildingLegend

114Theoretically,rooftopCWscouldhelptosolvetheproblemofopenspacelimitation;however,asMacaoPeninsulahaveahighnumberoffive-storeybuildingswherebyrooftopsareprivatelyusedbytheresidentstraditionally,publicCWimplementationwouldnotbefeasible.Nevertheless,thecombinedgreenspacefrominfrastructureandrecreation(0.8772km2)couldtheoreticallyaccommodatethecombinedareaofTF-VFCW0.0535km2andtheirsedimentationtankstotreat70,000m3ofwastewaterdaily.Inaworstcasescenario,assumethesedimentationtankswouldcausetherequiredTF-VFCWs’surfaceareatodoubleto0.107km2,thisisstillonlyaround12%ofthecombinedgreenspaceinMacao.Moreover,TF-VFCWscouldbeincorporatedintoexistinggreenspaceandcouldstillbeaestheticallypleasingtothepublicandactasrecreationalareas.Therefore,althoughitwillbeextremelychallengingtoimplementurbanCWs,itisnotcompletelyimpossible.FirstpilotTF-VFCWshouldbeimplementedinparkstotreatwastewaterfrompublictoiletstocheckitsfunctionalityandpublicacceptance.SomesuggestedlocationsfromFigure14bincludeMountfortress,Camoespark,FerreiraAmaralSquare,GardenoftheArtsandDr.Carlosd’AssumpçãoPark.

1154.CONCLUSIONSANDFUTUREWORK4.1.GeneralconclusionsItiswidelyacceptedthatCWsexertlessenvironmentalimpactthanconventionalWWTPs;however,inorderfordecision-makerstoestimateandcomparetheoverallenvironmentalimpactofdifferentCWsystems,thecircularmanagementandsuitabilityofeachcomponentwithintheCWmustbeextensivelyreviewed.Thisprojectaimstoevaluateandselectalowenvironmental-impactsubstrate,vegetationandlinermaterialforaconstructedwetlandbasedonexistingLifecycleassessmentsandotheravailableresearch.Ourstudyhasexploredtheconceptofcircularsustainabilitywithinthescopeofurbanwastewatermanagement,attemptedtoprovidenature-basedsolutionstocloseurbannutrientcyclesinsteadofgeneratingsecondarywastematerials.Uponobtaininganoverviewofdifferentenvironmentalimpactsofvariousmaterialsfromproductionphasetorecyclingmethods,themostsustainablematerialsweresuggested:Crushedautoclavedaeratedconcrete(substrate),amixtureofsugarcane,corn(vegetation)andcompactedclay(liner).ThiscombinationofmaterialswouldberecommendedforimplementinginapilotCWinMacao.CWhasalwaysthoughttotakeupmorelandthanconventionalWWTPs,makingitdifficulttoimplementinurbanareaswhichmostlyrelyonhardengineeringprojects.Wethereforeexploredintensifiedstrategiestotacklerate-limitingfactorsinCW’streatmentperformance;thiswouldinturnreducethearearequiredtotreatthesamevolumeofwastewater.AfinalCWdesignwithintensifiedstrategies:TF-VFCWwithatoplayerofgravels,earthwormadditionalongwiththechosensustainablematerialsarethenproposedforimplementation.Periodicharvestofsugarcaneandcornissuggestedduringoperationphase.ACWsystemhasneverbeenproposedinMacao,

116oursuggestedCWmodelwouldbetheveryfirsttheoreticalattempttotreatdomesticwastewaterinthiscitywiththesecondhighestpopulationdensityintheworld.WithourproposedfinalCWdesign,itwasfoundthatadecentralisedTF-VFCWnetworkcouldtheoreticallybeimplementedinMacao’scombinedopenandgreenareastotreat70,000m3ofdomesticwastewaterdaily.Currently,this70,000m3ofwastewateronlyunderwentPrimarytreatmentwhichisinsufficienttoremovepollutantsthatleadtoeutrophication.TheTF-VFCWscouldbeincorporatedintoexistinggreenareasforrecreationalandeveneducationalpurpose.ThearearequiredperPEisverycompetitivewithconventionalWWTP.However,furtherresearchintothecity’ssewagedrainagesystemandthecatchmentareaofeachdecentralisedCWhastobeperformedtoexaminethefeasibilityoftreatinglocaldomesticwastewater.Nevertheless,ourresearchoutcomeswouldenabledecision-makerstolookatwastewatermanagementinadifferentperspectiveotherthantraditionalhard-engineeringprojects.Nature-basedsolutions(NBS)oftenhelptoaddressmultipleurbanissuesandsimultaneouslyprovideenvironmental,socialandeconomicbenefitstobuildresilienceinthecity.Intheurbancontext,constructedwetlandsareaninterestingNBStosupportastrategyofresource-orientedwatermanagement.Ourstudywouldprovideaninsightintocouplingnature-basedsolutionswithurbanwastewatermanagement,bioenergyproductionandatthesametime,achievecircularmaterialmanagement.OurstudyhopestoprovideasteppingstoneforMacaotodevelopintoalow-carbonsmartcitythatincorporatesinnovativesolutionstocirculateresourcesandreduceenergyconsumption.

1174.2.FutureworkOurTF-VFCWmodelandthecalculatedrequiredareaisanestimatednumber.Anumberofexperimentswouldhavetobeperformedtoverifyitsfunctionality.Firstly,alaboratoryandapilotscalesystemwithorganicloadingrate330gCODm-²d-1ofourTF-VFCWdesignwouldhavetobeestablishedtotestthepollutantremovalefficiency.SaturatedCAACsubstrateshouldbetestedforitsdirectapplicationasfertilizertoplantswithP-deficientsoil.SugarcaneandcornfromthepilotscalesystemshouldbetestedfortheirNphyto-uptakeasnoexistingdataisavailable;theirbiomassshouldbeweighedtoestimatethegrowthrateintheTF-VFCWmodel.Sugarcontentshouldalsobemeasuredtoestimatesugarcaneandcorns’biofuelproductivity.Inaddition,higherOLRof330gCODm-²d-1couldleadtoproblemssuchascloggingandleadstodysfunctionalCWandincreaseinmaintenancecost,theadditionofearthwormsandtoplayerofgravelshavetobetestedifthesestrategiescouldalleviateclogging.Ifnoissuesweredetected,anevenhigherOLRshouldbetestedforthepossibilityoffurtherreducingtheTF-VFCWmodel’sarea.Alife-cycleassessmentoftheproposedpilotTF-VFCWdesignhastobeperformedtoexaminetheenvironmentalimpactofthewholepilotsystemindetail;furtheradjustmentscouldthenbemadetooptimizethemodel.AlthoughMacaohassufficientgreenandopenareastoaccommodatethecalculatedTF-VFCWs,amorein-depthGISanalysishastobeperformedtoidentifymoresuitablelocationsforCWimplementation.ExtensiveresearchonrearrangingexistingundergrounddomesticdrainagesystemhastobeperformedtotestthefeasibilityoftheCWimplementationintheseGIS-identifiedlocations.Nevertheless,asthepopulationdensityvarieswithinMacaoPeninsula,“catchmentareas”wouldhave

118tobeidentifiedsothatvariousCWsofsufficientsizecouldbecalculatedforvaryingwastewatervolumesindifferentareas/catchmentareas.Finally,asapplyingCWsinurbanareasisstillarelativelynewconcept,citizenswouldneedtofamiliarizewiththisnature-basedtechnology.ApilotCWshouldfirstbesetupinapublicgardenasdemonstrationandtoseekpublicacceptance.Toursandtalksshouldalsobeheldatthesegardensaspubliceducationprogramstoraiseawarenessintheimportanceofwaterconservation,itscircularmanagementandthemultiplebenefitsofnature-basedsolutions.SuccessfulintegrationofCWsintoaconcretejunglecouldbeawayforpeopletofeelconnectedtonature,andtoinspirepublicsupportforfurtherinclusionofgreensystemsinfutureurbanplanning.

1195.REFERENCESAfzal,M.,Arslan,M.,Müller,J.A.,Shabir,G.,Islam,E.,Tahseen,R.,...&Khan,Q.M.(2019).Floatingtreatmentwetlandsasasuitableoptionforlarge-scalewastewatertreatment.NatureSustainability,2(9),863-871.Ahn,C.,Mitsch,W.J.,&Wolfe,W.E.(2001).EffectsofrecycledFGDlinermaterialonwaterqualityandmacrophytesofconstructedwetlands:Amesocosmexperiment.WaterResearch,35(3),633-642.Alufasi,R.,Gere,J.,Chakauya,E.,Lebea,P.,Parawira,W.,&Chingwaru,W.(2017).Mechanismsofpathogenremovalbymacrophytesinconstructedwetlands.EnvironmentalTechnologyReviews,6(1),135-144.Ambrosio,R.,Pauletti,V.,Barth,G.,Povh,F.P.,Silva,D.A.D.,&Blum,H.(2017).Energypotentialofresidualmaizebiomassatdifferentspacingsandnitrogendoses.CiênciaeAgrotecnologia,41(6),626-633.Arias,C.A.,DelBubba,M.,&Brix,H.(2001).Phosphorusremovalbysandsforuseasmediainsubsurfaceflowconstructedreedbeds.Waterresearch,35(5),1159-1168.Babatunde,A.O.,Zhao,Y.Q.,O'neill,M.,&O'sullivan,B.(2008).Constructedwetlandsforenvironmentalpollutioncontrol:areviewofdevelopments,researchandpracticeinIreland.EnvironmentInternational,34(1),116-126.Babatunde,A.O.,Zhao,Y.Q.,&Zhao,X.H.(2010).Alumsludge-basedconstructedwetlandsystemforenhancedremovalofPandOMfromwastewater:concept,designandperformanceanalysis.Bioresourcetechnology,101(16),6576-6579.Balke,K.D.,&Zhu,Y.(2008).Naturalwaterpurificationandwatermanagementbyartificialgroundwaterrecharge.JournalofZhejiangUniversitySCIENCEB,9(3),221-226.Batianoff,G.N.,&Butler,D.W.(2002).Assessmentofinvasivenaturalizedplantsinsouth-eastQueensland.PlantProtectionQuarterly,17(1),27-34.Berg,U.,Donnert,D.,Weidler,P.G.,Kaschka,E.,Knoll,G.,&Nüesch,R.(2006).Phosphorusremovalandrecoveryfromwastewaterbytobermorite-seededcrystallisationofcalciumphosphate.WaterScienceandTechnology,53(3),131-138.BerlinerWasserbetriebe(n.d.).-PumpingStationsandSewageTreatmentPlants.Accessibleat:ttps://www.bwb.de/content/en/html/2177.php.Accessedon23.03.2020Bolton,L.,Joseph,S.,Greenway,M.,Donne,S.,Munroe,P.,&Marjo,C.E.(2019).Phosphorusadsorptionontoanenrichedbiocharsubstrateinconstructedwetlandstreatingwastewater.EcologicalEngineering:X,1,100005.Bosak,V.,VanderZaag,A.,Crolla,A.,Kinsley,C.,&Gordon,R.(2016).Performanceofaconstructedwetlandandpretreatmentsystemreceivingpotatofarmwashwater.Water,8(5),183.Brix,H.(1997).Domacrophytesplayaroleinconstructedtreatmentwetlands?WaterSci.Technol.35,11–17.

120Brix,H.,Arias,C.A.,&DelBubba,M.(2001).Mediaselectionforsustainablephosphorusremovalinsubsurfaceflowconstructedwetlands.Waterscienceandtechnology,44(11-12),47-54.Brogowski,Z.,&Renman,G.(2004).Characterizationofopokaasabasisforitsuseinwastewatertreatment.PolishJournalofEnvironmentalStudies,13(1),15-20.Burton,F.,Tchobanoglous,G.,&Stensel,H.D.(2003).Metcalf&EddyInc.-WastewaterEngineering:TreatmentandReuse.Calheiros,C.S.,Rangel,A.O.,&Castro,P.M.(2007).Constructedwetlandsystemsvegetatedwithdifferentplantsappliedtothetreatmentoftannerywastewater.Waterresearch,41(8),1790-1798.Calheiros,C.S.,Quitério,P.V.,Silva,G.,Crispim,L.F.,Brix,H.,Moura,S.C.,&Castro,P.M.(2012).UseofconstructedwetlandsystemswithArundoandSarcocorniaforpolishinghighsalinitytannerywastewater.Journalofenvironmentalmanagement,95(1),66-71.Calheiros,C.S.C.,Almeida,C.M.R.,&Mucha,A.M.(2018).Chapter8:Multiservicesandfunctionsofconstructedwetlands.InWetlandFunction,Services,ImportanceandThreats;Halicki,W.,Ed.;NovaSciencePublishers,Inc.NewYork,NY,USA,pp.269–298.ISBN978-1-53613-562-6.Castellar,J.A.D.C.,Formosa,J.,Chimenos,J.M.,Canals,J.,Bosch,M.,Rosell,J.R.,...&Arias,C.A.(2019).CrushedAutoclavedAeratedConcrete(CAAC),aPotentialReactiveFilterMediumforEnhancingPhosphorusRemovalinNature-BasedSolutions—PreliminaryBatchStudies.Water,11(7),1442.Chang,Y.,Wu,S.,Zhang,T.,Mazur,R.,Pang,C.,&Dong,R.(2014).Dynamicsofnitrogentransformationdependingondifferentoperationalstrategiesinlaboratory-scaletidalflowconstructedwetlands.Scienceofthetotalenvironment,487,49-56.Cheavegatti-Gianotto,A.,deAbreu,H.M.C.,Arruda,P.,BespalhokFilho,J.C.,Burnquist,W.L.,Creste,S.,...&deFátimaGrossi-de-Sá,M.(2011).Sugarcane(SaccharumXofficinarum):areferencestudyfortheregulationofgeneticallymodifiedcultivarsinBrazil.Tropicalplantbiology,4(1),62-89.Chen,G.Q.,Shao,L.,Chen,Z.M.,Li,Z.,Zhang,B.,Chen,H.,&Wu,Z.(2011).Low-carbonassessmentforecologicalwastewatertreatmentbyaconstructedwetlandinBeijing.EcologicalEngineering,37(4),622-628.Cicek,N.,Lambert,S.,Venema,H.D.,Snelgrove,K.R.,Bibeau,E.L.,&Grosshans,R.(2006).Nutrientremovalandbio-energyproductionfromNetley-LibauMarshatLakeWinnipegthroughannualbiomassharvesting.BiomassandBioenergy,30(6),529-536.Corbella,C.,Puigagut,J.,&Garfí,M.(2017).Lifecycleassessmentofconstructedwetlandsystemsforwastewatertreatmentcoupledwithmicrobialfuelcells.Scienceofthetotalenvironment,584,355-362.Crites,R.W.(1988).Designmanual:Constructedwetlandsandaquaticplantsystemsformunicipalwastewatertreatment.USEnvironmentalProtectionAgency,OfficeofResearchandDevelopment,CenterforEnvironmentalResearchInformation.

121Davis,L.(1995).Ahandbookofconstructedwetlands:Aguidetocreatingwetlandsfor:agriculturalwastewater,domesticwastewater,coalminedrainage,stormwater.IntheMid-AtlanticRegion.Volume1:Generalconsiderations.USDA-NaturalResourcesConservationService.De-Bashan,L.E.,&Bashan,Y.(2004).Recentadvancesinremovingphosphorusfromwastewateranditsfutureuseasfertilizer(1997–2003).Waterresearch,38(19),4222-4246.Dires,S.,Birhanu,T.,&Ambelu,A.(2019).Useofbrokenbricktoenhancetheremovalofnutrientsinsubsurfaceflowconstructedwetlandsreceivinghospitalwastewater.WaterScienceandTechnology,79(1),156-164.Dixon,A.,Simon,M.,&Burkitt,T.(2003).Assessingtheenvironmentalimpactoftwooptionsforsmall-scalewastewatertreatment:comparingareedbedandanaeratedbiologicalfilterusingalifecycleapproach.EcologicalEngineering,20(4),297-308.Dotro,G.,Langergraber,G.,Molle,P.,Nivala,J.,Puigagut,J.,Stein,O.,&VonSperling,M.(2017).Treatmentwetlands(Vol.7).London,UK:IWApublishing.DrainageServiceDepartment-WastewaterTreatment.(n.d.).RetrievedJune10,2020,fromhttps://www.dsd.gov.hk/Documents/AnnualReports/0405/EN/ch1/ch1.htmDrizo,A.,Comeau,Y.,Forget,C.,&Chapuis,R.P.(2002).Phosphorussaturationpotential:aparameterforestimatingthelongevityofconstructedwetlandsystems.Environmentalscience&technology,36(21),4642-4648.Dragone,G.,Fernandes,B.D.,Vicente,A.A.,&Teixeira,J.A.(2010).Thirdgenerationbiofuelsfrommicroalgae.DSEC.(StatisticsandCensusServiceDepartment)(2019a).IndustrialSurvey2018.Accessibleat:https://www.dsec.gov.mo/getAttachment/051e6aa9-e561-49e2-a0cb-79cfdd457397/C_IND_FR_2018_Y.aspxAccessedon22.08.2020DSEC.(StatisticsandCensusServiceDepartment)(2019b).EnvironmentalStatistics2019.Accessibleat:https://www.dsec.gov.mo/Statistic/Social/EnvironmentStatistics/2019%E5%B9%B4%E7%92%B0%E5%A2%83%E7%B5%B1%E8%A8%88.aspx?lang=zh-MO.Accessedon20.05.2020DSEC.(StatisticsandCensusServiceDepartment)(2019c).YearbookofStatistics2019.Accessibleat:https://www.dsec.gov.mo/getAttachment/a2b8aa77-3b58-4f44-b5cf-ea003a02cc4f/C_AE_PUB_2019_Y.aspx.Accessedon22.08.2020DSPA.(EnvironmentalProtectionBureau)(2018).ReportontheStateoftheEnvironmentofMacao2018.Accessibleat:https://www.dspa.gov.mo/Publications/StateReport/2018/2018_en.pdfAccessedon22.08.2020DSPA.(EnvironmentalProtectionBureau)(2019).EnvironmentalData2019.Accessibleat:https://www.dspa.gov.mo/envdata.aspx.Accessedon20.05.2020

122DSPA.(EnvironmentalProtectionBureau)(2020).Accessibleat:https://www.dspa.gov.mo/place2_3.aspx.Accessedon20.05.2020Du,X.,Shi,C.,&Ma,F.(2016).Influenceofintermittentaerationandorganicloadingrateonlab-scaleconstructedwetlandsystemstreatingsyntheticwastewater.DesalinationandWaterTreatment,57(21),9651-9659.DWA(2017)GrundsätzefürBemessung,BauundBetriebvonKläranlagenmitbepflanztenundunbepflanztenFilternzurReinigunghäuslichenundkommunalenAbwassers,inGerman.(Principlesofdesign,constructionandoperationofplantedandunplantedfiltersfortreatmentofdomesticwastewater).DeutscheVereinigungfürWasserwirtschaft,AbwasserundAbfalle.V.(DWA):Hennef,Germany.EC(EuropeanComission)(2020).Communicationfromthecommissiontotheeuropeanparliament,thecouncil,theeuropeaneconomicandsocialcommitteeandthecommitteeoftheregions.AnewCircularEconomyActionPlan.ForacleanerandmorecompetitiveEuropeBrussels,11.3.2020.COM(2020)98finalEPA(EnvironmentalProtectionAgency)(1985).CalcinersAndDryersInMineralIndustries-BackgroundInformationForProposedStandards,EPA-450/3-85-025a,U.S.EnvironmentalProtectionAgency,ResearchTrianglePark,NC,October1985.EPA,U.(2000).WastewaterTechnologyFactSheetWetlands:SubsurfaceFlow.EtterB,TilleyE,KhadkaR,UdertKM(2011)Low-coststruviteproductionusingsource-separatedurineinNepal.WaterRes45:852–862Fan,Y.,Li,Y.,Li,H.,&Cheng,F.(2018).Evaluatingheavymetalaccumulationandpotentialrisksinsoil-plantsystemsappliedwithmagnesiumslag-basedfertilizer.Chemosphere,197,382-388.Faeth,P.(2012).USEnergysecurityandwater:Thechallengesweface.Environment:ScienceandPolicyforSustainableDevelopment,54(1),4-19.Fuchs,V.J.,Mihelcic,J.R.,&Gierke,J.S.(2011).Lifecycleassessmentofverticalandhorizontalflowconstructedwetlandsforwastewatertreatmentconsideringnitrogenandcarbongreenhousegasemissions.Waterresearch,45(5),2073-2081.Gao,R.Y.,Shao,L.,Li,J.S.,Guo,S.,Han,M.Y.,Meng,J.,...&Lin,C.(2012).Comparisonofgreenhousegasemissionaccountingforaconstructedwetlandwastewatertreatmentsystem.Ecologicalinformatics,12,85-92.García-Pérez,A.,Harrison,M.,&Grant,B.(2009).Recirculatingverticalflowconstructedwetland:greenalternativetotreatingbothhumanandanimalsewage.Journalofenvironmentalhealth,72(4),17-21.García-Pérez,A,A.,Harrison,M.,&Grant,B.(2011).Recirculatingverticalflowconstructedwetlandforon-sitesewagetreatment:anapproachforasustainableecosystem.JournalofWaterandEnvironmentTechnology,9(1),39-46.García-Pérez,A.,Harrison,M.,Grant,B.,&Chivers,C.(2013).Microbialanalysisandchemicalcompositionofmaize(Zeamays,L.)growingonarecirculatingverticalflowconstructedwetlandtreatingsewageon-site.Biosystemsengineering,114(3),351-356.

123Garfí,M.,Flores,L.,&Ferrer,I.(2017).Lifecycleassessmentofwastewatertreatmentsystemsforsmallcommunities:activatedsludge,constructedwetlandsandhighratealgalponds.JournalofCleanerProduction,161,211-219.Ghermandi,A.,Bixio,D.,&Thoeye,C.(2007).Theroleoffreewatersurfaceconstructedwetlandsaspolishingstepinmunicipalwastewaterreclamationandreuse.ScienceoftheTotalEnvironment,380(1-3),247-258.Gkika,D.,Gikas,G.D.,&Tsihrintzis,V.A.(2015).Environmentalfootprintofconstructedwetlandstreatingwastewater.JournalofEnvironmentalScienceandHealth,PartA,50(6),631-638.Gorgoglione,A.,&Torretta,V.(2018).Sustainablemanagementandsuccessfulapplicationofconstructedwetlands:Acriticalreview.Sustainability,10(11),3910.Gottschall,N.,Boutin,C.,Crolla,A.,Kinsley,C.,&Champagne,P.(2007).Theroleofplantsintheremovalofnutrientsataconstructedwetlandtreatingagricultural(dairy)wastewater,Ontario,Canada.Ecologicalengineering,29(2),154-163.Greenaway,M.,&Woolley,A.(2001).Changesinplantbiomassandnutrientremovalover3yearsinaconstructedwetlandinCairns,Australia.WaterScienceandTechnology,44(11-12),303-310.Gucker,CoreyL.(2008).Typhalatifolia.In:FireEffectsInformationSystem,[Online].U.S.DepartmentofAgriculture,ForestService,RockyMountainResearchStation,FireSciencesLaboratory(Producer).Available:https://www.fs.fed.us/database/feis/plants/graminoid/typlat/all.html.AccessedonAugust22,2020.Haritash,A.K.,Sharma,A.,&Bahel,K.(2015).ThepotentialofCannalilyforwastewatertreatmentunderIndianconditions.Internationaljournalofphytoremediation,17(10),999-1004.Hathaway,J.M.,&Hunt,W.F.(2011).Evaluationoffirstflushforindicatorbacteriaandtotalsuspendedsolidsinurbanstormwaterrunoff.Water,Air,&SoilPollution,217(1-4),135-147.Hayasaka,D.,Fujiwara,S.,&Uchida,T.(2018).ImpactsofinvasiveIrispseudacorusL.(yellowflag)establishinginanabandonedurbanpondonnativesemi-wetlandvegetation.JournalofIntegrativeAgriculture,17(8),1881-1887.Haynes,R.J.(2015).Useofindustrialwastesasmediainconstructedwetlandsandfilterbeds—prospectsforremovalofphosphateandmetalsfromwastewaterstreams.Criticalreviewsinenvironmentalscienceandtechnology,45(10),1041-1103.Hernández-Crespo,C.,Oliver,N.,Bixquert,J.,Gargallo,S.,&Martín,M.(2016).ComparisonofthreeplantsinasurfaceflowconstructedwetlandtreatingeutrophicwaterinaMediterraneanclimate.Hydrobiologia,774(1),183-192.Hinck,J.E.,Blazer,V.S.,Schmitt,C.J.,Papoulias,D.M.,&Tillitt,D.E.(2009).Widespreadoccurrenceofintersexinblackbasses(Micropterusspp.)fromUSrivers,1995–2004.AquaticToxicology,95(1),60-70.Hylander,L.D.,&Simán,G.(2001).Plantavailabilityofphosphorussorbedtopotentialwastewatertreatmentmaterials.Biologyandfertilityofsoils,34(1),42-48.

124Hylander,L.D.,Kietlińska,A.,Renman,G.,&Simán,G.(2006).Phosphorusretentioninfiltermaterialsforwastewatertreatmentanditssubsequentsuitabilityforplantproduction.Bioresourcetechnology,97(7),914-921.Huang,X.,Liu,C.,Wang,Z.,Gao,C.,Zhu,G.,&Liu,L.(2013).TheEffectsofDifferentSubstratesonAmmoniumRemovalinConstructedWetlands:AComparisonofTheirPhysicochemicalCharacteristicsandAmmonium‐OxidizingProkaryoticCommunities.CLEAN–Soil,Air,Water,41(3),283-290.Ilyas,H.,&Masih,I.(2017).Intensificationofconstructedwetlandsforlandareareduction:areview.EnvironmentalScienceandPollutionResearch,24(13),12081-12091.IWA(InternationalWaterAssociation).(2016).TheIWAprinciplesforwaterwisecities.Jenssen,P.D.,&Krogstad,T.(2003).Designofconstructedwetlandsusingphosphorussorbinglightweightaggregate(LWA).AdvancesinEcologicalSciences,11,259-272.Jia,H.,Ma,H.,&Wei,M.(2011).Urbanwetlandplanning:acasestudyintheBeijingcentralregion.EcologicalComplexity,8(2),213-221.Jia,H.,Wang,Z.,Zhen,X.,Clar,M.,&Shaw,L.Y.(2017).China’sspongecityconstruction:Adiscussionontechnicalapproaches.FrontiersofEnvironmentalScience&Engineering,11(4),18.Johansson,L.(1997).TheuseofLECA(lightexpandedclayaggregrates)fortheremovalofphosphorusfromwastewater.WaterScienceandTechnology,35(5),87-93.Johansson,L.,&Gustafsson,J.P.(2000).Phosphateremovalusingblastfurnaceslagsandopoka-mechanisms.Waterresearch,34(1),259-265.JosephSahayarayan,J.,Ramasamy,V.,&Kandasamy,K.(2019).TREATMENTOFTEXTILEWASTEWATERUSINGVERTICALFLOWCONSTRUCTEDWETLANDWITHPLANTEDALTERNANTHERASESSILISANDZEAMAYS.InternationalJournalofAdvancedResearch,7(12),731–741.doi:10.21474/ijar01/10202Kaasik,A.,Vohla,C.,Motlep,R.,Mander,Ü.,&Kirsimäe,K.(2008).Hydratedcalcareousoil-shaleashaspotentialfiltermediaforphosphorusremovalinconstructedwetlands.Waterresearch,42(4-5),1315-1323.KadlecR.H.&WallaceS.D.(2009)TreatmentWetlands,SecondEdition.CRCPress,BocaRaton,FL,USA.Kantawanichkul,S.,S.Somprasert,U.Aekasin,&R.BrianE.Shutes."Treatmentofagriculturalwastewaterintwoexperimentalcombinedconstructedwetlandsystemsinatropicalclimate."WaterScienceandTechnology48,no.5(2003):199-205.Kantawanichkul,S.,&Somprasert,S.(2005).Usingacompactcombinedconstructedwetlandsystemtotreatagriculturalwastewaterwithhighnitrogen.WaterScienceandTechnology,51(9),47-53.Karaca,S.,Gürses,A.,Ejder,M.,Ac.&ıkyıldız,M.(2004).Kineticmodelingofliquid-phaseadsorptionofphosphateondolomite.J.ColloidInterfaceSci.277,257–263.

125Karim,M.R.,Manshadi,F.D.,Karpiscak,M.M.,Gerba,C.P.(2004)Thepersistenceandremovalofentericpathogensinconstructedwetlands.WaterResearch.38:7,1831-1837.Kindberg,L.,&Energy,N.F.(2010).Anintroductiontobioenergy:feedstocks,processesandproducts.ATTRA.Kluczka,J.,Zołotajkin,M.,Ciba,J.,&Staroń,M.(2017).Assessmentofaluminumbioavailabilityinalumsludgeforagriculturalutilization.Environmentalmonitoringandassessment,189(8),422.Koiv,M.,Vohla,C.,Mõtlep,R.,Liira,M.,Kirsimäe,K.,&Mander,Ü.(2009).Theperformanceofpeat-filledsubsurfaceflowfilterstreatinglandfillleachateandmunicipalwastewater.EcologicalEngineering,35(2),204-212.Koiv,M.,Ostonen,I.,Vohla,C.,Mõtlep,R.,Liira,M.,Lohmus,K.,...&Mander,Ü.(2012).Reusepotentialofphosphorus-richfiltermaterialsfromsubsurfaceflowwastewatertreatmentfiltersforforestsoilamendment.Hydrobiologia,692(1),145-156.Konnerup,D.,Koottatep,T.,&Brix,H.(2009).Treatmentofdomesticwastewaterintropical,subsurfaceflowconstructedwetlandsplantedwithCannaandHeliconia.Ecologicalengineering,35(2),248-257.Korkusuz,E.A.,Beklioğlu,M.,&Demirer,G.N.(2005).Comparisonofthetreatmentperformancesofblastfurnaceslag-basedandgravel-basedverticalflowwetlandsoperatedidenticallyfordomesticwastewatertreatmentinTurkey.EcologicalEngineering,24(3),185-198.Lambert,A.M.,Dudley,T.L.,&Saltonstall,K.(2010).Ecologyandimpactsofthelarge-staturedinvasivegrassesArundodonaxandPhragmitesaustralisinNorthAmerica.InvasivePlantScienceandManagement,3(4),489-494.Langergraber,G.,Dotro,G.,Nivala,J.,Rizzo,A.,&Stein,O.R.(Eds.).(2019).WetlandTechnology:PracticalInformationontheDesignandApplicationofTreatmentWetlands.IWAPublishing.Ledon,Y.C.,Rivas,A.,López,D.,&Vidal,G.(2017).Life-cyclegreenhousegasemissionsassessmentandextendedexergyaccountingofahorizontal-flowconstructedwetlandformunicipalwastewatertreatment:AcasestudyinChile.EcologicalIndicators,74,130-139.LiL,&WangQ.Q.(2006a)ThedevelopmentofconstructedwetlandsinChina.Availablewww.Chinacitywater.org.Accessed12Oct2008(inChinese)Li,Y.,Liu,C.,Luan,Z.,Peng,X.,Zhu,C.,Chen,Z.,...&Jia,Z.(2006b).Phosphateremovalfromaqueoussolutionsusingrawandactivatedredmudandflyash.Journalofhazardousmaterials,137(1),374-383.Li,W.W.,Yu,H.Q.,&Rittmann,B.E.(2015).Chemistry:reusewaterpollutants.Nature,528(7580),29-31.

126Li,L.,Yang,J.,Lin,C.Y.,Chua,C.T.,Wang,Y.,Zhao,K.,...&Wang,P.(2018).FieldsurveyofTyphoonHato(2017)andacomparisonwithstormsurgemodelinginMacau.Li,K.,&Zhou,L.(2019).TheInfluenceofUrbanFloodingonResidents’DailyTravel:ACaseStudyofMacauwithProposedAmeliorativeStrategies.Water,11(9),1825.Lin,Y.,Zhao,Y.,Ruan,X.,Barzee,T.J.,Zhang,Z.,Kong,H.,&Zhang,X.(2019).ThePotentialofConstructedWetlandPlantsforBioethanolProduction.BioEnergyResearch,1-7.LindB,BanZ,&BydénS(2000)Nutrientrecoveryfromhumanurinebystruvitecrystallizationwithammoniaadsorptiononzeoliteandwollastonite.BioresourTechnol73:169–174Liu,D.,Ge,Y.,Chang,J.,Peng,C.,Gu,B.,Chan,G.Y.,&Wu,X.(2009).ConstructedwetlandsinChina:recentdevelopmentsandfuturechallenges.FrontiersinEcologyandtheEnvironment,7(5),261-268.Liu,D.,Wu,X.,Chang,J.,Gu,B.,Min,Y.,Ge,Y.,...&Wu,J.(2012).Constructedwetlandsasbiofuelproductionsystems.NatureClimateChange,2(3),190-194.Liu,L.,Liu,Y.H.,Liu,C.X.,Wang,Z.,Dong,J.,Zhu,G.F.,&Huang,X.(2013).PotentialeffectandaccumulationofveterinaryantibioticsinPhragmitesaustralisunderhydroponicconditions.Ecologicalengineering,53,138-143.Liu,D.,Zou,C.,&Xu,M.(2019).Environmental,Ecological,andEconomicBenefitsofBiofuelProductionUsingaConstructedWetland:ACaseStudyinChina.Internationaljournalofenvironmentalresearchandpublichealth,16(5),827.Lomeda-DeMesa,R.A.P.,Soriano,A.N.,Marquez,A.R.D.,&Adornado,A.P.(2019).CharacterizationofTorrefiedBiomassfromSugarcane(Saccharumofficinarum)BagasseBlendedwithSemiraraCoal.InE3SWebofConferences(Vol.120,p.02002).EDPSciences.Long,S.P.,Karp,A.,Buckeridge,M.S.,Davis,S.C.,Jaiswal,D.,Moore,P.H.,...&Vonshakh,A.(2015).Feedstocksforbiofuelsandbioenergy.SouzaGM,VictoriaR,JolyCetal..,etal..,eds.Bioenergy&Sustainability:BridgingtheGaps.Paris:ScientificCommitteeonProblemsoftheEnvironment(SCOPE),302-47.Lopsik,K.(2013).Lifecycleassessmentofsmall-scaleconstructedwetlandandextendedaerationactivatedsludgewastewatertreatmentsystem.InternationalJournalofEnvironmentalScienceandTechnology,10(6),1295-1308.Lua,A.C.,Yang,T.,&Guo,J.(2004).Effectsofpyrolysisconditionsonthepropertiesofactivatedcarbonspreparedfrompistachio-nutshells.Journalofanalyticalandappliedpyrolysis,72(2),279-287.Machado,A.P.,Urbano,L.,Brito,A.G.,Janknecht,P.,Salas,J.J.,&Nogueira,R.(2007).Lifecycleassessmentofwastewatertreatmentoptionsforsmallanddecentralizedcommunities.WaterScienceandTechnology,56(3),15-22.

127Maddison,M.,Soosaar,K.,Mauring,T.,&Mander,Ü.(2009).ThebiomassandnutrientandheavymetalcontentofcattailsandreedsinwastewatertreatmentwetlandsfortheproductionofconstructionmaterialinEstonia.Desalination,246(1-3),120-128.Mahmood,Q.,Pervez,A.,Zeb,B.S.,Zaffar,H.,Yaqoob,H.,Waseem,M.,&Afsheen,S.(2013).Naturaltreatmentsystemsassustainableecotechnologiesforthedevelopingcountries.BioMedresearchinternational,2013.MalcolmPirnieEngineers,Inc.(2008).Statewideassessmentofenergyusebythemunicipalwaterandwastewatersector.FINALREPORT08-17.NewYorkStateEnergyResearchandDevelopmentAuthority.Accessibleat:https://wwww.nyserda.ny.gov.Accessedon23.03.2020Mander,Ü.,Dotro,G.,Ebie,Y.,Towprayoon,S.,Chiemchaisri,C.,Nogueira,S.F.,...&Mitsch,W.J.(2014).Greenhousegasemissioninconstructedwetlandsforwastewatertreatment:areview.EcologicalEngineering,66,19-35.Manderso,T.M.(2018).DeterminationoftheVolumeofFlowEqualizationBasininWastewaterTreatmentSystem.CivilandEnvironmentalResearch,10(4),34-41.Mann,R.A.,&Bavor,H.J.(1993).Phosphorusremovalinconstructedwetlandsusinggravelandindustrialwastesubstrata.WaterScienceandTechnology,27(1),107-113.Mann,J.J.,Barney,J.N.,Kyser,G.B.,&DiTomaso,J.M.(2013).Miscanthus×giganteusandArundodonaxshootandrhizometoleranceofextrememoisturestress.GcbBioenergy,5(6),693-700.Mannina,G.,&Viviani,G.(2009).Separateandcombinedsewersystems:along-termmodellingapproach.Waterscienceandtechnology,60(3),555-565.Mannina,G.,Ekama,G.A.,Ødegaard,H.,&Olsson,G.(Eds.).(2018).AdvancesinWastewaterTreatment.iwapublishing.Masi,F.,Rizzo,A.,&Regelsberger,M.(2018).Theroleofconstructedwetlandsinanewcirculareconomy,resourceoriented,andecosystemservicesparadigm.Journalofenvironmentalmanagement,216,275-284.Mateus,D.M.,&Pinho,H.J.(2010).PhosphorusRemovalbyExpandedClay—SixYearsofPilot‐ScaleConstructedWetlandsExperience.Waterenvironmentresearch,82(2),128-137.Mateus,D.M.,Vaz,M.M.,&Pinho,H.J.(2012).Fragmentedlimestonewastesasaconstructedwetlandsubstrateforphosphorusremoval.EcologicalEngineering,41,65-69.Mateus,D.M.,Vaz,M.M.,Capela,I.,&Pinho,H.J.(2014).Sugarcaneasconstructedwetlandvegetation:Preliminarystudies.Ecologicalengineering,62,175-178.McGrane,S.J.(2016).Impactsofurbanisationonhydrologicalandwaterqualitydynamics,andurbanwatermanagement:areview.HydrologicalSciencesJournal,61(13),2295-2311.Metcalf&Eddy,Burton,F.L.,Stensel,H.D.,&Tchobanoglous,G.(2003).Wastewaterengineering:treatmentandreuse.McGrawHill.

128Mena,J.,Rodriguez,L.,Nuñez,J.,Fernández,F.J.,&Villaseñor,J.(2008).Designofhorizontalandverticalsubsurfaceflowconstructedwetlandstreatingindustrialwastewater.WITTransactionsonEcologyandtheEnvironment,111,555-564.MillenniumEcosystemAssessment(2005).Ecosystemsandhumanwell-being(Vol.5,p.563).UnitedStatesofAmerica:Islandpress.Min,L.,Fangying,G.,Jiawei,F.,Meixuan,S.,&He,Z.(2011).Thesustainableapproachtothegreenspacelayoutinhighdensityurbanenvironment:acasestudyofMacaupeninsula.ProcediaEngineering,21,922-928.Mkumbo,S.,Mwegoha,W.J.,&Kihampa,C.(2019).TheeffectofAutoclavedAeratedConcreteontheSurvivalandGrowthofLaunaeacornutaandSporobolusjacquemontiigrowninHeavyMetalPollutedSoils:AGreenhouseStudy.Neset,T.S.S.,&Cordell,D.(2012).Globalphosphorusscarcity:identifyingsynergiesforasustainablefuture.JournaloftheScienceofFoodandAgriculture,92(1),2-6.Nivala,J.,Headley,T.,Wallace,S.,Bernhard,K.,Brix,H.,vanAfferden,M.,&Müller,R.A.(2013a).Comparativeanalysisofconstructedwetlands:ThedesignandconstructionoftheecotechnologyresearchfacilityinLangenreichenbach,Germany.EcologicalEngineering,61,527-543.Nivala,J.,Wallace,S.,Headley,T.,Kassa,K.,Brix,H.,vanAfferden,M.,&Müller,R.(2013b).Oxygentransferandconsumptioninsubsurfaceflowtreatmentwetlands.EcologicalEngineering,61,544-554.Norton,S.(2014).Removalmechanismsinconstructedwastewaterwetlands.ONLINE:http://home.eng.iastate.edu/~tge/ce421-521/stephen.pdf.Nuengjamnong,C.,Chiarawatchai,N.,Polprasert,C.,&Otterpohl,R.(2011).Treatingswinewastewaterbyintegratingearthwormsintoconstructedwetlands.JournalofEnvironmentalScienceandHealthPartA,46(7),800-804.NYCDEP(NewYorkCityDepartmentofEnvironmentalProtection)(n.d.).WastewaterTreatmentSystem.Accessibleat:https://semspub.epa.gov/work/02/206372.pdf.Accessedon23.03.2020OECD.(2003).ConsensusDocumentonthebiologyofZeamayssubsp.mays(Maize).SeriesonHarmonisationofRegulatoryOversightinBiotechnology(ENV/JM/MONO(2003)11),27,1-49.ÖNORM(2009)ÖNORMB2505:BepflanzteBodenfilter(Pflanzenkläranlagen)-Anwendung,Bemessung,BauundBetrieb(Subsurfaceflowconstructedwetlands-Application,dimensioning,installation,andoperation)[inGerman].ÖsterreichischesNormungsinstitut:Vienna,Austria.Pagter,M.,Bragato,C.,&Brix,H.(2005).ToleranceandphysiologicalresponsesofPhragmitesaustralistowaterdeficit.AquaticBotany,81(4),285-299.Pakenas,L.J.(1995).Energyefficiencyinmunicipalwastewatertreatmentplants:technologyassessment.NewYorkStateEnergyResearchandDevelopmentAuthority.Pant,H.K.,Reddy,K.R.,&Lemon,E.(2001).Phosphorusretentioncapacityofrootbedmediaofsub-surfaceflowconstructedwetlands.EcologicalEngineering,17(4),345-355.

129Pescod,M.B.(1992).Wastewatertreatmentanduseinagriculture.Retrievedfromhttp://www.fao.org/3/t0551e/t0551e05.htmPrasad,R.(n.d.).PhosphorusBasics:UnderstandingPhosphorusFormsandTheirCyclingintheSoil.Retrievedfromhttps://www.aces.edu/blog/topics/crop-production/understanding-phosphorus-forms-and-their-cycling-in-the-soil/Prochaska,C.A.,&Zouboulis,A.I.(2006).Removalofphosphatesbypilotvertical-flowconstructedwetlandsusingamixtureofsandanddolomiteassubstrate.EcologicalEngineering,26(3),293-303.Prost-Boucle,S.,&Molle,P.(2012).Recirculationonasinglestageofverticalflowconstructedwetland:treatmentlimitsandoperationmodes.EcologicalEngineering,43,81-84.Pucher,B.,&Langergraber,G.(2019).TheStateoftheArtofClogginginVerticalFlowWetlands.Water,11(11),2400.Rani,S.H.C.,Din,M.,Md,F.,Yusof,M.,Mohd,B.,&Chelliapan,S.(2011).OverviewofSubsurfaceConstructedWetlandsApplicationinTropicalClimates.UniversalJournalofEnvironmentalResearch&Technology,1(2).Rebaque,D.,Martínez-Rubio,R.,Fornalé,S.,García-Angulo,P.,Alonso-Simón,A.,Álvarez,J.M.,...&Encina,A.(2017).Characterizationofstructuralcellwallpolysaccharidesincattail(Typhalatifolia):Evaluationaspotentialbiofuelfeedstock.Carbohydratepolymers,175,679-688.Renman,G.,&Renman,A.(2012).Sustainableuseofcrushedautoclavedaeratedconcrete(CAAC)asafiltermediuminwastewaterpurification.WASCON2012Conferenceproceedings.7s.Resende,J.D.,Nolasco,M.A.,&Pacca,S.A.(2019).Lifecycleassessmentandcostingofwastewatertreatmentsystemscoupledtoconstructedwetlands.Resources,ConservationandRecycling,148,170-177.Rigby,H.,Pritchard,D.,Collins,D.,Walton,K.,&Penney,N.(2013).Theuseofalumsludgetoimprovecerealproductiononanutrient-deficientsoil.Environmentaltechnology,34(11),1359-1368.Roux,P.,Boutin,C.,Risch,E.,&Heduit,A.(2010,October).LifeCycleenvironmentalAssessment(LCA)ofsanitationsystemsincludingsewerage:Caseofverticalflowconstructedwetlandsversusactivatedsludge.Ryu,H.D.,Lim,C.S.,Kang,M.K.,&Lee,S.I.(2012).EvaluationofstruviteobtainedfromsemiconductorwastewaterasafertilizerincultivatingChinesecabbage.Journalofhazardousmaterials,221,248-255.Sakadevan,K.,&Bavor,H.J.(1998).Phosphateadsorptioncharacteristicsofsoils,slagsandzeolitetobeusedassubstratesinconstructedwetlandsystems.WaterRes.32,393–399.Schlüter,U.,&Crawford,R.M.(2001).Long‐termanoxiatoleranceinleavesofAcoruscalamusL.andIrispseudacorusL.JournalofexperimentalBotany,52(364),2213-2225.

130Smith,B.R.(2009).Re-thinkingwastewaterlandscapes:combininginnovativestrategiestoaddresstomorrow'surbanwastewatertreatmentchallenges.WaterScienceandTechnology,60(6),1465-1473.Somerville,C.,Youngs,H.,Taylor,C.,Davis,S.C.,&Long,S.P.(2010).Feedstocksforlignocellulosicbiofuels.science,329(5993),790-792.Srivastava,P.,Yadav,A.K.,Garaniya,V.,&Abbassi,R.(2019).Constructedwetlandcoupledmicrobialfuelcelltechnology:developmentandpotentialapplications.InMicrobialElectrochemicalTechnology(pp.1021-1036).Elsevier.Stefanakis,A.,Akratos,C.S.,&Tsihrintzis,V.A.(2014).Verticalflowconstructedwetlands:eco-engineeringsystemsforwastewaterandsludgetreatment.Newnes.Stefanakis,A.I.,Prigent,S.,&Breuer,R.(2018).Integratedproducedwatermanagementinadesertoilfieldusingwetlandtechnologyandinnovativereusepractices.ConstructedWetlandsforIndustrialWastewaterTreatment;Stefanakis,AI,Ed.;OhnWiley&SonsLtd.:Chichester,UK,25-42.Stefanakis,A.I.(2019).TheRoleofConstructedWetlandsasGreenInfrastructureforSustainableUrbanWaterManagement.Sustainability,11(24),6981.Strang,T.J.,&Wareham,D.G.(2006).Phosphorusremovalinawaste-stabilizationpondcontaininglimestonerockfilters.JournalofEnvironmentalEngineeringandScience,5(6),447-457.Streubel,J.D.(2011).Biochar:Itscharacterizationandutilityforrecoveringphosphorusfromanaerobicdigesteddairyeffluent.Sun,G.,Zhao,Y.,Allen,S.,&Cooper,D.(2006).Generating“tide”inpilot‐scaleconstructedwetlandstoenhanceagriculturalwastewatertreatment.EngineeringinLifeSciences,6(6),560-565.Taghizadeh-Toosi,A.,Clough,T.J.,Sherlock,R.R.,&Condron,L.M.(2012).Biocharadsorbedammoniaisbioavailable.Plantandsoil,350(1-2),57-69.Talboys,P.J.,Heppell,J.,Roose,T.,Healey,J.R.,Jones,D.L.,&Withers,P.J.(2016).Struvite:aslow-releasefertiliserforsustainablephosphorusmanagement?.Plantandsoil,401(1-2),109-123.Tang,X.,Wu,M.,Li,R.,&Wang,Z.(2017).Prospectofrecoveringphosphorusinmagnesiumslag-packedwetlandfilter.EnvironmentalScienceandPollutionResearch,24(29),22808-22815.Tangahu,B.V.,Ningsih,D.A.,Kurniawan,S.B.,&Imron,M.F.(2019).StudyofBODandCODremovalinbatikwastewaterusingscirpusgrossusandIrispseudacoruswithintermittentexposuresystem.JournalofEcologicalEngineering,20(5).Tarrago,M.,Garcia-Valles,M.,Aly,M.H.,&Martínez,S.(2017).Valorizationofsludgefromawastewatertreatmentplantbyglass-ceramicproduction.CeramicsInternational,43(1),930-937.Thameswater,2020-Thesewagetreatmentprocess.(n.d.).RetrievedJune10,2020,fromhttps://cycles.thameswater.co.uk/Accessible/The-sewage-treatment-process

131Tran,N.H.,Reinhard,M.,&Gin,K.Y.H.(2018).Occurrenceandfateofemergingcontaminantsinmunicipalwastewatertreatmentplantsfromdifferentgeographicalregions-areview.Waterresearch,133,182-207.Trazzi,P.A.,Leahy,J.J.,Hayes,M.H.,&Kwapinski,W.(2016).Adsorptionanddesorptionofphosphateonbiochars.JournalofEnvironmentalChemicalEngineering,4(1),37-46.UnitedNations(2016).InternationalDecadeforAction,"WaterforSustainableDevelopment",2018-2028.reviseddraftresolution.A/C.2/71/L.12/Rev.1.Availableat:https://digitallibrary.un.org/record/849767/files/A_C-2_71_L-12_Rev-1-EN.pdf(Accessed:8October2020).UnitedNations,DepartmentofEconomicandSocialAffairs,PopulationDivision(2019).WorldPopulationProspects2019:DataBooket.ST/ESA/SER.A/424.Vanraes,P.,Nikiforov,A.Y.,&Leys,C.(2016).Electricaldischargeinwatertreatmenttechnologyformicropollutantdecomposition.Plasmascienceandtechnology—progressinphysicalstatesandchemicalreactions,429.Viator,R.P.,WhiteJr,P.M.,Hale,A.J.,&Waguespack,H.L.(2012).Screeningfortolerancetoperiodicfloodingforcanegrownforsucroseandbioenergy.BiomassandBioenergy,44,56-63.Vohla,C.,Alas,R.,Nurk,K.,Baatz,S.,&Mander,Ü.(2007).Phosphorusretentioncapacityinahorizontalsubsurfaceflowconstructedwetland.Sci.TotalEnviron.380,66–74.Vohla,C.,Kõiv,M.,Bavor,H.J.,Chazarenc,F.,&Mander,Ü.(2011).Filtermaterialsforphosphorusremovalfromwastewaterintreatmentwetlands—Areview.EcologicalEngineering,37(1),70-89.vonSperling,M.(1996).Comparisonamongthemostfrequentlyusedsystemsforwastewatertreatmentindevelopingcountries.Waterscienceandtechnology,33(3),59-72.Vymazal,Jan(2006)RemovalofNutrientsinVariousTypesofConstructedWetlands.ScienceoftheTotalEnvironment,380,48-65.Vymazal,J.(2011a).Constructedwetlandsforwastewatertreatment:fivedecadesofexperience.Environmentalscience&technology,45(1),61-69.Vymazal,J.(2011b).Plantsusedinconstructedwetlandswithhorizontalsubsurfaceflow:areview.Hydrobiologia,674(1),133-156.Vymazal,J.(2013).Emergentplantsusedinfreewatersurfaceconstructedwetlands:areview.Ecologicalengineering,61,582-592.Vymazal,J.,&Kröpfelová,L.(2008).Wastewatertreatmentinconstructedwetlandswithhorizontalsub-surfaceflow(Vol.14).Springerscience&businessmedia.Wang,T.,Liu,R.,O’Meara,K.,Mullan,E.,&Zhao,Y.(2018).AssessmentofaFieldTidalFlowConstructedWetlandinTreatmentofSwineWastewater:LifeCycleApproach.Water,10(5),573.

132WWAP(UnitedNationsWorldWaterAssessmentProgramme)(2017).TheUnitedNationsWorldWaterDevelopmentReport2017:Wastewater,TheUntappedResource.Paris,UNESCO.Accessibleat:https://unesdoc.unesco.org/ark:/48223/pf0000247153.Accessedon23.03.2020WWAP(UnitedNationsWorldWaterAssessmentProgramme)/UN-Water(2018)TheUnitedNationsWorldWaterDevelopmentReport2018:Nature-BasedSolutionsforWater.Paris,UNESCO.Weber,K.P.,&Legge,R.L.(2008).Pathogenremovalinconstructedwetlands.Wetlands:Ecology,ConservationandRestoration,176-211.Wilderer,P.A.,&Schreff,D.(2000).Decentralizedandcentralizedwastewatermanagement:achallengefortechnologydevelopers.WaterScienceandTechnology,41(1),1-8.Wu,H.,Zhang,J.,Li,C.,Fan,J.,&Zou,Y.(2013).Massbalancestudyonphosphorusremovalinconstructedwetlandmicrocosmstreatingpollutedriverwater.CLEAN–Soil,Air,Water,41(9),844-850.Wu,H.,Fan,J.,Zhang,J.,Ngo,H.H.,Guo,W.,Liang,S.,...&Liu,H.(2015a).Strategiesandtechniquestoenhanceconstructedwetlandperformanceforsustainablewastewatertreatment.EnvironmentalScienceandPollutionResearch,22(19),14637-14650.Wu,J.,Xu,D.,He,F.,He,J.,&Wu,Z.(2015b).Comprehensiveevaluationofsubstratesinvertical-flowconstructedwetlandsfordomesticwastewatertreatment.WaterPracticeandTechnology,10(3),625-632.Wu,H.,Zhang,J.,Ngo,H.H.,Guo,W.,Hu,Z.,Liang,S.,...&Liu,H.(2015c).Areviewonthesustainabilityofconstructedwetlandsforwastewatertreatment:designandoperation.Bioresourcetechnology,175,594-601.Xu,J.,Zhang,J.,Xie,H.,Li,C.,Bao,N.,Zhang,C.,&Shi,Q.(2010).PhysiologicalresponsesofPhragmitesaustralistowastewaterwithdifferentchemicaloxygendemands.EcologicalEngineering,36(10),1341-1347.Yadav,A.K.,Dash,P.,Mohanty,A.,Abbassi,R.,&Mishra,B.K.(2012).Performanceassessmentofinnovativeconstructedwetland-microbialfuelcellforelectricityproductionanddyeremoval.EcologicalEngineering,47,126-131.Yaghi,N.,&Hartikainen,H.(2013).Enhancementofphosphorussorptionontolightexpandedclayaggregatesbymeansofaluminumandironoxidecoatings.Chemosphere,93(9),1879-1886.Yan,Y.,&Xu,J.(2014).ImprovingwinterperformanceofconstructedwetlandsforwastewatertreatmentinnorthernChina:areview.Wetlands,34(2),243-253.Yang,S.,Zhao,Y.,Chen,R.,Feng,C.,Zhang,Z.,Lei,Z.,&Yang,Y.(2013).Anoveltabletporousmaterialdevelopedasadsorbentforphosphateremovalandrecycling.Journalofcolloidandinterfacescience,396,197-204.Zaidi,P.H.,Rafique,S.,Rai,P.K.,Singh,N.N.,&Srinivasan,G.(2004).Tolerancetoexcessmoistureinmaize(ZeamaysL.):susceptiblecropstagesandidentificationoftolerantgenotypes.FieldCropsResearch,90(2-3),189-202.

133Zamora,S.,Marín-Muñíz,J.L.,Nakase-Rodríguez,C.,Fernández-Lambert,G.,&Sandoval,L.(2019).WastewaterTreatmentbyConstructedWetlandEco-Technology:InfluenceofMineralandPlasticMaterialsasFilterMediaandTropicalOrnamentalPlants.Water,11(11),2344.Zhao,X.H.,&Zhao,Y.Q.(2009).InvestigationofphosphorusdesorptionfromP-saturatedalumsludgeusedasasubstrateinconstructedwetland.SeparationandPurificationTechnology,66(1),71-75.Zhao,F.,Liu,C.,Rafiq,M.T.,Ding,Z.,Zeng,Z.,Aziz,R.,&Yang,X.(2014).Screeningwetlandplantsfornutrientuptakeandbioenergyfeedstockproduction.InternationalJournalofAgricultureandBiology,16(1).Zhao,X.,Yang,J.,Zhang,X.,Wang,L.,&Ma,F.(2017).Evaluationofbioaugmentationusingmultiplelifecycleassessmentapproaches:acasestudyofconstructedwetland.Bioresourcetechnology,244,407-415.Zhao,Y.,Damgaard,A.,&Christensen,T.H.(2018).Bioethanolfromcornstover–areviewandtechnicalassessmentofalternativebiotechnologies.ProgressinEnergyandCombustionScience,67,275-291.Zhu,T.,Jenssen,P.D.,Maehlum,T.,&Krogstad,T.(1997).Phosphorussorptionandchemicalcharacteristicsoflightweightaggregates(LWA)-potentialfiltermediaintreatmentwetlands.WaterScienceandTechnology,35(5),103.Zhu,J.,Hu,W.,Hu,L.,Deng,J.,Li,Q.,&Gao,F.(2012).Variationintheefficiencyofnutrientremovalinapilot-scalenaturalwetland.Wetlands,32(2),311-319.