Citation
Understanding relationships among ecosystem services across spatial scales and over time

Material Information

Title:
Understanding relationships among ecosystem services across spatial scales and over time
Series Title:
Jiangxiao Qiu et al 2018 Environ. Res. Lett. 13 054020
Creator:
Qiu, Jiangxiao
Publisher:
Environmental Research Letters
Publication Date:
Language:
English
Physical Description:
Journal Article

Notes

Abstract:
Sustaining ecosystem services (ES), mitigating their tradeoffs and avoiding unfavorable future trajectories are pressing social-environmental challenges that require enhanced understanding of their relationships across scales. Current knowledge of ES relationships is often constrained to one spatial scale or one snapshot in time. In this research, we integrated biophysical modeling with future scenarios to examine changes in relationships among eight ES indicators from 2001–2070 across three spatial scales—grid cell, subwatershed, and watershed. We focused on the Yahara Watershed (Wisconsin) in theMidwestern United States—an exemplar for many urbanizing agricultural landscapes. Relationships among ES indicators changed over time; some relationships exhibited high interannual variations (e.g. drainage vs. food production, nitrate leaching vs. net ecosystem exchange) and even reversed signs over time (e.g. perennial grass production vs. phosphorus yield). Robust patterns were detected for relationships among some regulating services (e.g. soil retention vs. water quality) across three spatial scales, but other relationships lacked simple scaling rules. This was especially true for relationships of food production vs. water quality, and drainage vs. number of days with runoff >10 mm, which differed substantially across spatial scales. Our results also showed that local tradeoffs between food production and water quality do not necessarily scale up, so reducing local tradeoffs may be insufficient to mitigate such tradeoffs at the watershed scale. We further synthesized these cross-scale patterns into a typology of factors that could drive changes in ES relationships across scales: (1) effects of biophysical connections, (2) effects of dominant drivers, (3) combined effects of biophysical linkages and dominant drivers, and (4) artificial scale effects, and concluded with management implications. Our study highlights the importance of taking a dynamic perspective and accounting for spatial scales in monitoring and management to sustain future ES.
Acquisition:
Collected for University of Florida's Institutional Repository by the UFIR Self-Submittal tool. Submitted by Jiangxiao Qiu.

Record Information

Source Institution:
University of Florida Institutional Repository
Holding Location:
University of Florida
Rights Management:
Copyright Creator/Rights holder. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item is only available as the following downloads:


Full Text

PAGE 2

Environ.Res.Lett. 13 (2018)054020 https://doi.org/10.1088/1748-9326/aabb87 LETTERUnderstandingrelationshi psamongecosystemservices acrossspatialscalesandovertimeJiangxiaoQiu1 2 9 ,StephenRCarpenter3 ,EricGBooth4 5 ,MelissaMotew6 ,SamuelCZipper7 8 ChristopherJKucharik5 6 ,StevenPLoheideII4 andMonicaGTurner2 1SchoolofForestResourcesandConservation,FortLauderdaleRes earchandEducationCenter,UniversityofFlorida,Davie,FL33314, UnitedStatesofAmerica2DepartmentofIntegrativeBiology,UniversityofWisco nsin-Madison,Madison,WI53706,UnitedStatesofAmerica3CenterforLimnology,UniversityofWisconsin-M adison,Madison,WI53706,UnitedStatesofAmerica4DepartmentofCivilandEnvironmentalEngineering,University ofWisconsin-Madison,Madison,WI53706,UnitedStatesofAmerica5DepartmentofAgronomy,UniversityofWisconsin-M adison,Madison,WI53706,UnitedStatesofAmerica6CenterforSustainabilityandtheGlobalEnvironment,Universit yofWisconsin-Madison,Madison,WI53706,UnitedStatesofAmerica7DepartmentofCivilEngineering,UniversityofVictoria,Victoria,BC,Canada8DepartmentofEarthandPlanetarySciences,McGillUniversity,Montreal,QC,Canada9Authortowhomanycorrespondenceshouldbeaddressed. OPENACCESSRECEIVED19December2017REVISED21March2018ACCEPTEDFORPUBLICATION4April2018PUBLISHED3May2018 Originalcontentfrom thisworkmaybeused underthetermsofthe CreativeCommons Attribution3.0licence Anyfurtherdistribution ofthisworkmust maintainattributionto theauthor(s)andthe titleofthework,journal citationandDOI. E-mail: qiuj@u.edu Keywords: tradeoffs,synergies,sustainability,social-ecologicalsystems ,futurescenarios,biophysicalmodeling,agriculturallandscape Supplementarymaterialforthisarticleisavailable online Abstract Sustainingecosystemservices( ES),mitigatingtheirtradeoffsandavoidingunfavorablefuture trajectoriesarepressingsocial-environmentalcha llengesthatrequireenhancedunderstandingoftheir relationshipsacrossscales.CurrentknowledgeofESrelationshipsisoftenconstrainedtoonespatial scaleoronesnapshotintime.Inthisresearch,we integratedbiophysicalmodelingwithfuture scenariostoexaminechangesinrelationshipsam ongeightESindicatorsfrom2001–2070acrossthree spatialscales—gridcell,subwatershed,andwa tershed.WefocusedontheYaharaWatershed (Wisconsin)intheMidwesternUnitedStates—anexemplarformanyurbanizingagricultural landscapes.RelationshipsamongESindicatorschangedovertime;somerelationshipsexhibitedhigh interannualvariations(e.g.drainagevs.foodproduction,nitrateleachingvs.netecosystemexchange) andevenreversedsignsovertime(e.g.perennialgrassproductionvs.phosphorusyield).Robust patternsweredetectedforrelationshipsamongsomeregulatingservices(e.g.soilretentionvs.water quality)acrossthreespatialscales,butotherrel ationshipslackedsimplescalingrules.Thiswas especiallytrueforrelationshipsoffoodproductionvs.waterquality,anddrainagevs.numberofdays withrunoff>10mm,whichdifferedsubstantiallyacrosssp atialscales.Ourresultsalsoshowedthat localtradeoffsbetweenfoodproductionandwaterqualitydonotnecessarilyscaleup,soreducing localtradeoffsmaybeinsufcienttomitigatesu chtradeoffsatthewatershedscale.Wefurther synthesizedthesecross-scalepatternsintoaty pologyoffactorsthatcoulddrivechangesinES relationshipsacrossscales:(1)effectsofbiophysicalconnections,(2)effectsofdominantdrivers,(3) combinedeffectsofbiophysicallinkagesanddominantdrivers,and(4)articialscaleeffects,and concludedwithmanagementimplications.Ourstudyhighlightstheimportanceoftakingadynamic perspectiveandaccountingforspatialscalesin monitoringandmanagementtosustainfutureES. IntroductionHumanactivitieshavesubstantiallytransformedour biospheretopromotedesirableecosystemgoodsand services(ES)(e.g.timberandagriculturalproducts) (Kareiva etal 2007 ,EllisandRamankutty 2008 Foley etal 2011 ).Suchefforts,whilecrucialformeetingdemandsofagrowingpopulation,canleadto unintendedconsequencesforotherESthatareequally ifnotmoreimportant.Forexample,theMillennium EcosystemAssessmentrevealedthat,attheglobalscale, provisioningservicessuchascrops,livestock,and 2018TheAuthor(s).PublishedbyIOPPublishingLtd

PAGE 3

Environ.Res.Lett. 13 (2018)054020 Table1. Biophysicalindicators(andcorrespondingunits)of eightecosystemservicesincludedinthisresearch. EcosystemserviceBiophysicalindicatorofecosystemserviceUnit ProvisioningES CropproductionAnnualtotalcrop(corn,soybean,wheat)yield(buacŠ1)bPerennialgrassproductionAnnualtotalforagecropsandperennialgrass(alfalfa,hay,pasture)yieldkghaŠ1FreshwatersupplyAnnualtotaldrainagemm RegulatingES GroundwaterqualityaAnnualtotalnitrate(NO3-N)leachedatthebottomofsoilprolekghaŠ1Surface-waterqualityaAnnualtotalphosphorusyieldinrunoffkghaŠ1FloodregulationaAnnualnumberofdayswithrunoff > 10mmdays ClimateregulationaAnnualnetecosystemexchange(NEE)ofcarbonMgChaŠ1SoilretentionaAnnualtotalsedimentyieldinrunoffthaŠ1 aDenotesthatecosystemserviceisquantiedusinganinverseindicator,wherethegreaterthenumericvalueoftheindicatormeansthelower theserviceprovided.b1bushel/acre=87LhaŠ1.aquaculture,havebeenincreasingoverthepast50 years,whereasmostregulatingserviceslikediseaseregulation,waterpurication,andpollinationhavebeen declining(MEA 2005a ,Carpenter etal 2009 ).These resultsarenotsurprisingbecauseESinteractincomplexandsometimesnonlinearways(Bennett etal 2009 ,Koch etal 2009 ,QiuandTurner 2015 ),and thusdeliberatechangesinoneEScansimultaneously alterothers.Thedegradationofregulatingservices raisesspecialconcernsfromtheresearchandpolicy communities,becauseitmaycompromiselong-term ecosystemresilienceandleadtochangesthattakeus beyondasafeoperatingspaceforhumanity(Carpenter etal 2009 ,Steffen etal 2015 ).Hence,itisimperativeto understandrelationshipsamongEStosustainmultipleservices,manageundesirabletradeoffs,andforestall ecologicalsurprises. PriorresearchhasdenedtypesofESrelationships, andelucidatedunderpinningmechanisms(Rodr guez etal 2006 ,Bennett etal 2009 ,Cord etal 2017 ).Major relationshipsinclude:(1) tradeoffs ,inwhichoneESis reducedbecauseofincreaseduseorsupplyofanother; and(2) synergies ,wheremultipleESareenhanced simultaneously.RecentstudiesalsosuggestedthatES canhaveconstrainteffects,whereoneESisimposingupperlimitsonanotherES(Hao etal 2017 ). EmpiricalstudieshavedocumentedESrelationships acrossarangeofecosystemsandscales(e.g.RaudseppHearne etal 2010 ,Goldstein etal 2012 ,Haines-Young etal 2012 ,Maes etal 2012 ,QiuandTurner 2013 Howe etal 2014 ,Meacham etal 2016 ,Zheng etal 2016 ).However,twoknowledgegapshamperprogress onkeyresearchfrontiersinESsustainability. First,ESrelationshipsreportedpreviouslywereoften associatedwithoneparticularspatialscale,andonly afewstudieshaveexploredchangesinrelationships acrossdifferentspatialscales(Scholes etal 2013 Raudsepp-HearneandPeterson 2016 ).Nonetheless, ithasbeensuggestedthatESrelationshipsfromone spatialscalemaynottranslatetootherscales,and simpleextrapolationmayleadtomisinformedactions andunwantedoutcomes(Peters etal 2006 ,Costanza etal 2007 ,Anderson etal 2009 ,Holland etal 2011 Scholesetal 2013 ).Second,manyempiricalstudies havefocusedonsnapshotsintimeandnotconsideredtemporalchanges.However,otherstudieshave highlighteddynamicnatureofESandtheirinteractions(MEA 2005b ,Qiu etal 2018 ,Renard etal 2015 TomschaandGergel 2016 ,Spake etal 2017 ). Scaleandthesearchforscalinglawsinbiologicalandecologicalsystemshaslongintriguedscientists (AllenandStarr 1982 ,O ’ Neill 1986 ,Wiens 1989 ,Levin 1992 ,Whittaker 1999 ,Gardner etal 2001 ,Wu 2004 Wu etal 2006 ,Sims etal 2008 ).Onecommonissue isthemismatchbetweenscalesofecologicalprocesses andhumanobservationsandmanagement(Schneider 2001 ,Scholes 2017 ),andwhetherandhowobserved ecologicalphenomenoncanbescaled(Wu etal 2006 ). ScalemismatchesareespeciallyproblematicforES researchbecausetheproduction,distribution,and managementofESaredeterminedbymyriadsocialecologicalprocessesandstructures,eachwithdistinct scales(Cumming etal 2006 ,Andersson etal 2015 Raudsepp-HearneandPeterson 2016 ).Inaddition,ES provisionmaybeaffectedbyprocessesoperatingat differentspatialandtemporalscales,leadingtocomplexcross-scaleinteractions(Heffernan etal 2014 ,Rose etal 2017 ).AlthoughitisoftenassumedthatESand theirrelationshipsvaryacrossscales,quantitativelytestingthisassumptionwithmultipleESisrare.Such empiricalevidenceisneededtotestexpectationsand elucidatemechanismsunderlyingchangesinESrelationshipsacrossscalesofanalysis.Itcouldalsoimprove thecapacitytopredictcriticalESchanges(Clark etal 2001 ),andsustainablymanagemultipleES. Inthisstudy,wequantiedspatial-temporal dynamicsofaportfoliooffood,water,andbiogeochemicalES(table 1 )inanurbanizingagricultural landscape(YaharaWatershed,Wisconsin,USA),and analyzedchangesinESrelationships.WethensynthesizedatypologyofwhyESrelationshipsmayvaryacross scales,andexempliedthistypologywithourresults. Detailedstudyregiondescriptioncanbefoundinsupplementarymaterials(SM)availableat stacks.iop.org/ ERL/13/054020/mmedia .ESwereestimatedfrom 2001 Š 2070usingprocess-basedsimulationmodels underfourplausiblefuturescenariosthatvariedin social,political,economicandbiophysicaldrivers 2

PAGE 4

Environ.Res.Lett. 13 (2018)054020 (Carpenter etal 2015 ,Booth etal 2016 ).Theuseof scenariosandgriddedmodelsimulationsallowedus toanalyzelong-termchangesinESrelationshipsand testwhetherspatialscaleofanalysismattersoverawide rangeoffuturesocial-environmentalconditions.MaterialsandmethodsQuantifyingspatial-temporaldynamicsofES WequantiedindicatorsofeightESat220 220m spatialresolutionusingsimulationresultsfroman integratedspatiallyexplicitmodel—Agro-IBIS(AgroecosystemIntegratedBIosphereSimulator)(Foley etal 1996 ,Kucharik etal 2000 ,Kucharik 2003 ).Selected indicatorscapturekeyecologicalprocessesthatunderlieproduction/conditionofeachES(table 1 ).For example,weuseddrainageasanindicatorforfreshwatersupply,becausedrainageiscriticalforreplenishing aquifersthataretheprimaryfreshwatersourcesin thisregion.Nitrateleachingandphosphorusyield wereusedas(inverse)indicatorsforwaterquality, because(1)nitrateisthemostubiquitouscontaminant ofgroundwaterwithdetrimentalimpactsonhuman health,and(2)phosphorusfromagriculturalorurban runoffsisthemajorthreattosurface-waterquality, especiallyinagriculture-dominatedwatersheds(Qiu andTurner 2013 ). Agro-IBISisaprocess-basedmodelthatsimulates continuousdynamicsofterrestrialecosystemprocesses,biogeochemistry,waterandenergybalances, andhasbeencalibratedandvalidatedextensivelyfor performanceinnaturalandmanagedsystemsinthe MidwesternUnitedStates(DonnerandKucharik 2003 KucharikandTwine 2007 ,MotewandKucharik 2013 ). Inthisresearch,weusedanupdatedversionofAgroIBISthatincludednewlydevelopedphosphorusand sedimentmodules(Motew etal 2017 ).Watershedscalephosphorus,sediment,andstreamowprocesses werecalibratedandevaluatedagainsthistoricaldata withsatisfactoryperformance(Soylu etal 2014 Zipper etal 2015 ,Motew etal 2017 ). Weperformedsimulationsfrom2001 Š 2070,where 2001 Š 2010wasconsideredasthebaselineforcomparison,and2011 Š 2070weresimulatedunderfour scenariosthatcontrastedinsocial,political,economic andbiophysicaldrivers(Carpenter etal 2015 ,Booth etal 2016 ,Wardropper etal 2016 ).Completescenario narrativesareavailableat Yahara2070.org .Abriefsynopsisofdrivingquestions,climateandland-usedrivers foreachscenarioisprovidedbelow:  AcceleratedInnovation(AI) — ‘ Whatifweprioritizedtechnologicalsolutionstoourenvironmental changes? ’ isthedrivingquestioninAI.Booming green-andbio-technologyincreasespopulationand urbanfootprints,andmajorefcienciesaregained inagriculture.Thisscenariohastheleastextremeclimatechange,characterizedbymorefrequentheavy rainfalleventsandwarmingof 2Cby2070.  AbandonmentandRenewal(AR) — ‘ Whatifweare notpreparedforescalatingenvironmentalchanges? ’ isthemajorquestioninAR.ARexploresconsequencesduetosocietalunpreparednessforclimate changes,whereaseriesofcatastrophiceventsin 2030sreducesthewatershedpopulation > 90%, causingfarmlandabandonment,urbandeteriorationandincreasednaturalvegetation.Thisscenario hasthemostextremeclimatechange,withooding andextremeheatduringthe2030sandwarmingof 5.5C.  ConnectedCommunities(CC) – ‘ Whatif,collectively,weshiftedourvaluestowardscommunity andsustainability? ’ istheoverarchingthemeofCC. Urbanfootprintshrinksduetoincreasedurbandensityandconversionofturfgrasstorestoredprairies andurbanfarms.Dietshiftsleadtotransitionsfrom row-cropstoamixofpasture,vegetablesandfruits, andsmallgrains.ClimatechangeinCCisintermediatebetweenARandAI,withheavyrainfalleventsand droughtincreasinginfrequencyand3.5Cwarming.  NestedWatersheds(NW) – ‘ Whatifwereformhow wegovernfreshwaterresourcestobetterprotect them? ’ isthesalientfeatureofNW.InNW,governanceiscenteredonnationalwaterandfood securities.Urbanlandsremainrelativelyconstant, buttaxdisincentivesforintensiveagriculturereduce row-cropsandpromotepracticesthatsupportclean andsufcientwater.Thisscenariohascomparable climatechangetoCC,withmorefrequentprecipitationextremesand4Cwarming. Scenariosofferarangeofsocial-environmental conditionstotestwhetherandhowscalemattersfor ESrelationships.Basedonscenarionarratives,we producedspatial-temporalchangesofmajordrivers (climate,landuse/cover,nutrients)(Booth etal 2016 ). Thesedriverswerespatially-explicitandtemporally dynamic,andwereinputintoAgro-IBIStosimulate long-termESdynamics(Qiu etal 2018 ).Allscenario driversweredetailedinBooth etal ( 2016 ),andhere wehighlightedlanduse/coverandclimateinSM.This workdiffersfromandbuildsuponpriorfoundational workreferencedabovebyanalyzingchangesinESrelationshipsacrossscales. Analyticalframework RelationshipsamongESindicatorswereanalyzed atthreespatialscales—grid-cell,subwatershed,and watershed(gure 1 )usingpairwisecorrelations.We chosepairwisecorrelationsbecausetheyallowfor analyzingtemporalchangesinESrelationshipsand facilitatecomparisonsacrossspatialscales.Sincethere are28possibleESpairs,welimitedouranalysestoa subsetof12thatwerepreviouslyreportedasprominenttradeoffsorsynergiesinagriculturallandscapes (Power 2010 ,QiuandTurner 2013 ).Atthegrid-cell scale(gure 1 ( a )),wegeneratedarandomsampleof 3

PAGE 5

Environ.Res.Lett. 13 (2018)054020 Figure1. Analyticalframeworkforexaminingrelationshipsamongecosystemservicesacrossthreespatialscales:( A )grid-cell;( B ) subwatershed;and( C )watershed,over2001 Š 2070period.Panel A showsrelationshipsamongpairedecosystemsservicescalculated annuallybasedonarandomsampleof220 220mgridcellsacrossthelandscape;correlationcoefcientswerethenplottedagainst timetodemonstratetemporaldynamicsofrelationships.Inpanel B ,relationshipsamongpairedecosystemserviceswerecalculated annuallyatthesubwatershedscales,andthenplottedagainsttime.Panel C showsemergentrelationshipsbetweenpairedecosystem servicesatthewatershedscaleastheyevolveovertime,andthecolorgradientfromlightesttodarkestrepresentsthetimedimension from2001 Š 2070.Themiddlecolumnrepresentsthedeterminationofecosy stemservicerelationshipsforagivenyear,andthethird columnrepresentsthedynamicsofecosystemservicerelationshipsoveratimeperiod. 3000grid-cellsacrossthelandscape,andextractedestimatesofESindicatorsforeachcellfollowingQiuand Turner( 2013 ).WethencomputedpairwiseSpearmancorrelationsannuallybasedonrandomlysampled cells,andplottedcorrelationcoefcientsovertime. Atthesubwatershedscale(gure 1 ( b )),wecomputed ESindicatorsat100second-ordersubwatersheds(Qiu andTurner 2015 )annuallybysummingoraveraging(dependingontheES)biophysicaloutputs,and thencalculatedpairwiseSpearmancorrelationcoefcientsandplottedovertime.Atthewatershedscale (gure 1 ( c )),werstsummedoraveraged(depending ontheES)theESindicatorsannuallyandcalculated watershedmeansat5yearintervals(i.e.2011 Š 2015, 2016–2020...2066 Š 2070),thenplottedeachpairofES indicatorsinatwo-dimensionalspacewithtimecolorcoded.SynergiesaresuggestedifbothESincreaseover time,andtradeoffsareindicatedifoneESincreases 4

PAGE 6

Environ.Res.Lett. 13 (2018)054020 Figure2. Changesinrelationshipsbetweencropandperennialgrassproductionacrossthreespatialscales( A )grid-cell,( B )subwatershed,and( C )watershedfrom2001 Š 2070underfourfuturescenarios.Scenarioswerecolor-codedinpanels A and B ,withthickcolor linesasSpearmancorrelationcoefcientandcolorribbonsas95%condenceinterval,onthebasisofbootstrapapproachwith1000 iterations.Panel C showstemporalchangesintheindicatorsofcropandperennialgrassproduction,calculatedatthewatershedscale at5yearintervals(i.e.2011 Š 2015,2016 Š 2020...2066 Š 2070).Foragivenscenario(coloredcircle),thegradientfromlightesttodarkest representsthetimedimensionfrom2011 Š 2015to2066 Š 2070.Solidblackcirclesarethebaselineestimates(averaged2001 Š 2010)for comparison. astheotherdecreases.Suchanapproachcananalyze emergentlandscapedynamicsofESrelationshipsatthe watershedscale(Qiu etal 2018 ).AllanalyseswereperformedinRstatisticalsoftware3.3.1(RCoreTeam 2016 ).ResultsCropandperennialgrassproductionshowedconsistentnegativerelationshipsovertimeatthegrid-cell scale,butwerepositivelycorrelatedatthesubwatershedscale(gure 2 ).Atthewatershedlevel,crop productionagainshowedtradeoffswithperennialgrass productioninallscenarios,exceptforARinwhich cropandperennialgrassproductionbothdeclined. Persistenttradeoffsbetweencropproductionand waterqualitywerefoundatgrid-cellandsubwatershedscalesacrossallscenarios,indicatedbypositive correlationsofcropyieldwithnitrateleachingand phosphorusyield(inverseindicatorsofwaterquality)(gures 3 ( a )and( c )and 4 ( a )and( c )).Atthe watershedscale,cropproduction—waterqualitytradeoffsappearedinthreescenarios(gures 5 ( a )and( c )), butnotinAIwheresynergiesemergedovertime. Interestingly,tradeoffsbetweencropproductionand waterqualityatthegrid-cellscalediminishedover timeundertwoscenarios(NWandAR);nevertheless, alsointhesetwoscenarios,suchtradeoffspersisted andintensiedatthewatershedscale.Ontheother hand,cropproduction—waterqualitytradeoffsintensiedovertimeinAIscenarioatthegrid-cellscale,but suchtradeoffswerenotevidentatwatershedscale.For relationshipsbetweencropproductionanddrainage (indicatoroffreshwatersupply),noclearpatternswere detectedacrossallspatialscales(gures 3 ( e ), 4 ( e ) and 5 ( e )).Theirrelationshipsshiftedbetweenpositiveandnegativeovertimewithlargeinterannual variationsinallscenarios. Consistenttradeoffsbetweenperennialgrassproductionandgroundwaterqualitywerefoundat grid-cellandsubwatershedscalesunderallscenarios (gures 3 ( b )and 4 ( b )),indicatedbypositivecorrelationsbetweengrassyieldandnitrateleaching.However, atthewatershedscale,thistradeoffappearedinonly twoscenarios(gure 5 ( b )).InNWandCC,synergies emergedovertimebetweenperennialgrassproductionandgroundwaterquality;interestingly,alsoinNW andCC,tradeoffsbetweenthesetwoESintensied atthegrid-cellscale(gure 3 ( b )).Tradeoffsbetween perennialgrassproductionandsurface-waterquality wereevidentatgrid-cellandsubwatershedscalesat thestartofthesimulationperiod,indicatedbypositive associationsbetweengrassandphosphorusyields(gures 3 ( d )and 4 ( d )).However,suchtradeoffsdeclined overtimeinmostscenarios.Atthewatershedscale, perennialgrassproductionandsurface-waterqualitywererelatedassynergiesintwoscenarios(NW andCC),butastradeoffsintheothertwo(gure 5 ( d )).Similartocropproduction,relationships betweenperennialgrassproductionanddrainagewere highlyvariablewithlargeinterannualvariationsat allspatialscales(gures 3 ( f ), 4 ( f )and5 ( f )). Forrelationshipsamongwaterandbiogeochemical ES,ourresultsdemonstratedconsistentnegativerelationshipsbetweendrainageandnumberofdayswith runoff > 10mm(inverseindicatorofoodregulation) atgrid-cellandsubwatershedscales(gures 6 ( a )and 7 ( a )),butpositiverelationshipsatthewatershedscale (gure 8 ( a )).However,relationshipsofdrainagevs. nitrateleaching,nitrateleachingvs.phosphorusyield, andphosphorusvs.sedimentyieldremainedconsistentlypositiveacrossallspatialscales(gures 6 ( b )–( d ), 7 ( b )–( d )and 8 ( b )–( d )).Eventhoughthestrength ofthesepositiverelationshipsdeclinedovertime undercertainscenariosatgrid-cellorsubwatershed scales,theirstrongpositiverelationshipsatthewatershedscaleweremaintained.Inaddition,consistent 5

PAGE 7

Environ.Res.Lett. 13 (2018)054020 Figure3. Changesinrelationshipsbetweenfoodproductionandfreshwaterecosystemservicesatthegrid-cellscalefrom2001 Š 2070 underfourfuturescenarios.Scenarioswerecolor-coded,withthic kcolorlinesasSpearmancorrelationcoefcientandcolorribbons as95%condenceinterval,onthebasisofbootstrapapproachwith1000iterations. positiverelationships(albeitwithlargeinterannual variations)werefoundbetweennitrateleachingand netecosystemexchange(NEE;inverseindicatorofclimateregulation)acrossthreespatialscalesundermost scenarios(gures 6 ( e ), 7 ( e )and 8 ( e )).Similarly,relationshipsbetweenphosphorusyieldandNEEremained positiveacrossallspatialscalesformostsimulation periodsunderallscenarios(gure 6 ( f ), 7 ( f )and 8 ( f )).DiscussionManagingmultipleESsustainablyrequiresimproved understandingofscale-dependentrelationships.Our researchintegratedstate-of-the-artbiophysicalmodelingwithscenariostotestconsistencyofrelationships foreightESovera70yearperiodunderarange ofsocial-ecologicalchanges.MostESrelationships werenotstaticovertime,withlargeinterannualvariationsorsuddenchangesforcertainpairsofES. Whilerelationshipsamongsomeregulatingservices wererobustacrossspatialscales(e.g.waterqualityvs. soilretention),othersvariedsubstantially.Relationshipsbetweenfoodproductionandwaterqualitywere inconsistentacrossscales:localrelationshipsdidnot applyatbroaderscales,andsometimeshadopposite patterns.OurresultssuggestcautionwhenextrapolatingESrelationshipsfromonescaletoanother,and underscoretheimportanceofaccountingforspatial andtemporalscalesinmonitoringandmanagingmultipleES(Sun etal 2016 ,Spake etal 2017 ). AtypologyofESrelationshipsacrossspatialscales Bennett etal ( 2009 )suggestedtwomechanismsfor ESrelationships:(1)interactionsamongES,and(2) effectsofdominantdriversonmultipleES.Such typologieswereinstrumentalandusedtoidentifylinkagesbetweenbiodiversityandES(e.g.Ricketts etal 2016 ).Basedonearlierresearchandourndings, 6

PAGE 8

Environ.Res.Lett. 13 (2018)054020 Figure4. Changesinrelationshipsbetweenfoodproductionandfreshwaterecosystemservicesatthesubwatershedscalefrom 2001 Š 2070underfourfuturescenarios.Scenarioswerecolor-coded,wi ththickcolorlinesasSpearmancorrelationcoefcientand colorribbonsas95%condenceinterval,onthebasisofbootstrapapproachwith1000iterations. weproposedfourpossibleexplanationsofwhyES relationshipsmaydifferacrossscales(gure 9 ):(1) effectsofbiophysicalconnections;(2)effectsofdominantdrivers;(3)combinedeffectsofbiophysical linkagesanddominantdrivers;(4)articialscale effects.Wethenexempliedandsubstantiatedthis typologywithourresults. Effectsofbiophysicalconnections. ESrelationships canresultfromtheirbiophysicalconnections(i.e.level ofES1affectslevelofES2,orviceversa)(gure 9 ( a )).Changingscalesofanalysismayenhance, reverseordiminishapparentrelationshipsamongES, becausebiophysicalconnectionsunderlyinginteractionsamongecologicalprocessesandservicescanbe scale-dependent(e.g.scale-dependenceofpollinatorplantinteractions;Garc aandChacoff 2007 ). OurresultsshowedthatrelationshipsamongcertainwaterandbiogeochemicalESwereconsistent andpredictableacrossscales(gures 5 – 7 ).Positive relationshipsofdrainagevs.nitrateleaching,andphosphorusyieldvs.sedimentyieldareassociatedwith biophysicalprocessesthatlinktheseESatallspatialscales(gure 9 ( a );biophysicalconnectionsremain unchangedacrossscales).Consistentrelationships betweendrainageandnitrateleachingacrossscales suggestinherenttradeoffsoffreshwatersupplyand groundwaterquality,andthuschallengestoenhance thesefreshwaterEStogether.Thisresultcorresponds wellwithndingsfromotherstudies(e.g.Nangia etal 2008 ,Carlson etal 2011 ).Consistentsynergiesbetween surface-waterqualityandsoilretentionsuggestopportunitiestoco-managethesetwoESsimultaneouslyat differentspatialscales.Anotherexampleofthistypologyisconsistentsynergiesbetweenclimateregulation andwaterqualityacrossspatialscales,whicharealso duetothatbiophysicalconnectionsunderlyingthese twoESremainunchangedacrossscales(gure 9 ( a )). Specically,increasedcarbonuptakemeanshighernet 7

PAGE 9

Environ.Res.Lett. 13 (2018)054020 Figure5. Changesinrelationshipsbetweenfoodproductionandfreshwaterecosystemservicesatthewatershedscalefrom2001 Š 2070 underfourfuturescenarios.Scenarioswerecolor-coded,andtemporalchangesintheindicatorsofpairedecosystemserviceswere calculatedatthewatershedscaleat5yearintervals(i.e.2011 Š 2015,2016 Š 2020...2066 Š 2070).Foragivenscenario(coloredcircle), thegradientfromlightesttodarkestrepresentsthetimedimensionfrom2011 Š 2015to2066 Š 2070.Solidblackcirclesarethebaseline estimates(averaged2001 Š 2010)forcomparison.Pleasenotethat y -axeswerereversedforecosystemservicesquantiedusinginverse indicators(i.e.thehighertheindicator,thelowertheprovisionofservice,suchasnitrateleachingandphosphorusyield). primaryproductionandmorenutrientuptakefrom soils,therebyreducingtheriskofnutrientlosses(e.g. nitratefromtherootzone,andphosphorusfromsoils). Otherstudiesalsorevealedsimilarsynergiesbetween carbonstorage/sequestrationandwaterqualityatvariedscales(Raudsepp-Hearne etal 2010 ,Holland etal 2011 ,Turner etal 2014 ). Effectsofdominantdrivers. ESrelationshipscan resultfromeffectsofdominantdriversthatsimultaneouslycontrolmultipleES(gure 9 ( b )).Yetitis possiblethatthemagnitudeofdrivereffectsorkindof dominantdriverschangeacrossscales,thusalteringES relationships.Forexample,priorresearchhasrevealed thatdominantcontrollingabioticfactorsforecosystem processessuchasnutrienttransport,decomposition, andcarbonandnitrogendynamics,differacross spatialscales(Jones etal 2006 ,ManzoniandPorporato 2009 ,Bradford etal 2014 ). Ourresultsshowedthatrelationshipsbetweenfood productionandwaterqualitydifferedacrossspatial scales.Persistenttradeoffsatgrid-cellandsubwatershedscalesarelikelyduetolocaleffectsofdominant drivers—nutrientandmanureapplications—atsmall spatialscales(Motew etal 2017 ).However,atthe watershedscale,suchtradeoffscanbemitigatedor evenshifttosynergies,reectingcumulativeeffects ofmultipledominantdriversactinginconcertat largespatialscales(gure 9 ( b );dominantdrivers changedacrossscales).Specically,proactivemanagementandland-usetransitions,technologicaladvances, andlessextremeclimatechangescaninteractto reducenutrientapplication,increaseplantnutrient 8

PAGE 10

Environ.Res.Lett. 13 (2018)054020 Figure6. Changesinrelationshipsamongwaterandbiogeochemicalservicesatthegrid-cellscalefrom2001 Š 2070underfourfuture scenarios.Scenarioswerecolor-coded,withthickcolorlinesasSpearmancorrelationcoefcientandcolorribbonsas95%condence interval,onthebasisofbootstrapapproachwith1000iterations. uptake,mitigatenutrientl oss,andthusreducetradeoffsoffoodproductionandwaterquality(Randall etal 1997 ,McIsaac etal 2010 ,Asbjornsen etal 2014 ). Whileconsistentwithpriorresearchthatidentied localtradeoffsoffoodproductionandwaterquality(Raudsepp-Hearne etal 2010 ,QiuandTurner 2013 ),oursimulatedresultsprovidedfurtherevidencethatthiswell-recognizedlocaltradeoffcan bealleviatedatbroadspatialscales.However,our resultsalsosuggestedthatreducinglocaltradeoffs betweencropproductionandwaterqualitymaybe insufcienttomitigatetheirtradeoffsatlargerscales whereclimateeffectsaredominant(Carpenter etal 2017 ). Ontheotherhand,dominantdriversofESrelationshipscouldalsoremainunalteredacrossscales(gure 9 ( b )),asevidencedbyrelationshipsbetweendrainage andfoodproductionthatwerehighlyvariablewith largeinterannualvariations.Consistentpatterns betweenthesetwoESlikelyreectthefactthat climateremainsthedominantdriveracrossscales. Additionally,thetightcouplingofnitrateleaching vs.phosphorusyieldislikelyassociatedwithapplied nutrientsandmanurethatcontainhighlevelsof 9

PAGE 11

Environ.Res.Lett. 13 (2018)054020 Figure7. Changesinrelationshipsamongwaterandbiogeochemicalservicesatthesubwatershedscalefrom2001 Š 2070underfour futurescenarios.Scenarioswerecolor-coded,withthickcolorlinesasSpearmancorrelationcoefcientandcolorribbonsas95% condenceinterval,onthebasisofbootstrapapproachwith1000iterations. bothnutrients,andprecipitationastheprimarycontrol(gure 9 ( b ));dominantdriversremainunchanged acrossscales).Consistentsynergiesbetweensurfaceandgroundwaterqualityindicatorshighlightthe importanceofconsideringsurfaceandgroundwater asanintegratedhydrologicalandbiogeochemicalcontinuumforenhancingmanagementeffectivenessacross scales(e.g.leveragingwaterpolicesandlandscapemanagementtoimprovebothsurface-andgroundwater) (QiuandTurner 2015 ,Qiu etal 2017 ). Combinedeffectsofbiophysicallinkagesanddominantdrivers. Changingscalescanalterdominant driversaswellasbiophysicallinkagesofES,thus affectingtheirrelationships.Relationshipsbetween drainageandnumberofdayswithrunoff > 10mm shiftedfromnegativeatlocalscalestopositiveat thewatershedscale(gures 5 – 7 ),reectingcombinedeffectsofbiophysicallinkagesanddominant drivers(gure 9 ( c )).Specically,atgrid-celland subwatershedscales,biophysicalprocessesofrunoffinltrationpartitioningdrivenegativerelationships betweendrainageandextremerunoffdaysonayearly basis(i.e.areaswithmoreinltrationnecessarilyhad lessrunoff,andviceversa)(Craig etal 2010 );whereas 10

PAGE 12

Environ.Res.Lett. 13 (2018)054020 Figure8. Changesinrelationshipsamongwaterandbiogeochemicalservicesatthewatershedscalefrom2001 Š 2070underfourfuture scenarios.Scenarioswerecolor-coded,andtemporalchangesintheindicatorsofpairedserviceswerecalculatedatthewatershedscale at5yearintervals(i.e.2011 Š 2015,2016 Š 2020 ƒ 2066 Š 2070).Foragivenscenario(coloredcircle),thegradientfromlightesttodarkest representsthetimedimensionfrom2011 Š 2015to2066 Š 2070.Solidblackcirclesarethebaselineestimates(averaged2001 Š 2010)for comparison.Pleasenotethat y -axeswerereversedforservicesquantiedusinginverseindicators. atthewatershedscalewhenmanyyearsareconsidered,precipitationemergesasthedominantdriver oftheirpositiverelationships(i.e.moreprecipitation ledtomorewateravailableforbothdrainageand runoff). Articialscaleeffects. ESrelationshipscanalso changeduetoarticialscaleeffects(gure 9 ( d )),e.g. fromsimplealterationstospatialresolutionand/or extentofanalysis(Wu 2004 ).Ourstudyshowedthat cropandperennialgrassproductionwerenegatively correlatedatthegrid-cellscale(i.e.atagiventime,a pixeloflandcanbedevotedtoeithercroporperennialgrass,butnotboth),whilebothcanbeachieved atalargerspatialscale(e.g.subwatershed)viaamix oflanduses.Suchscaleeffectsduetospatialextent ofanalysismayoftenbemanifestedthroughmutual exclusivityofresources(e.g.landuse/cover)dominatingESprovisionwhichvaryacrossscales.Whileour exampleisrelatedtoESdependentonland,thisarticialscaleeffectcouldbroadlyapplytoESdominatedby otherresources(e.g.water)whosemutualexclusivity changesacrossscales. 11

PAGE 13

Environ.Res.Lett. 13 (2018)054020 Figure9. Conceptualdiagramillustratingthetypologythroughwhichchangesinscalesmayalterecosystemservicerelationships:( A ) Effectsofbiophysicalconnections,wherethestrength,magnitudea nddirectionofconnectionsthatunderlietherelationshipsamong servicesmaychangefromonescaletoanother;( B )Effectsofdominantdrivers,whereshiftingscalescouldaltereitherthemagnitude orkindsofdriversineffect,andthusaffectrelationshipsamongservices;( C )Combinedeffectsofbiophysicallinkagesanddominant drivers;( D )Articialscaleeffects,whererelationshipsamongservicesmaychangeacrossscalesasaresultofsimplealterationsto spatialresolutionand/orextentofanalysis.Pleasenotethattheexamplesdemonstratedhereareforillustrativepurposesandnotan exhaustivelistofallpossiblechanges. TemporaldynamicsofESrelationships TemporaldynamicsofESrelationshipsreected responsestosocial-environmentalchanges(gures S1–S2)entailedineachscenario(Qiu etal 2018 ) thatcanshapetheprovisionandinteractionsofES. Changesinrelationshipscanbeabruptwithhigh interannualvariations(e.g.drainagevs.foodproduction,drainagevs.numberofdayswithrunoff > 10mm).AbruptchangesinESrelationshipsseemed toalignwellwithtimingofsubstantialalterations inlandcoverandassociatedmanagement(gures S1–S2).Highinterannualvariability,ontheother hand,ispossiblyassociatedwithweathereffects.Some ESrelationshipsdivergedamongscenarios;e.g.at grid-cellandsubwatershedscales,thestrengthof tradeoffsbetweencropproductionandwaterquality increasedincertainscenariosbutdeclinedinothers.Sometimes,thenatureofrelationshipscouldeven bereversed;e.g.perennialgrassproduction—surfacewaterqualitytradeoffsshiftedtosynergiestowardsthe endofthesimulation. Managementandpolicyimplications Ourresearchprovidesmanagementandpolicyimplications.First,managementatonespatialscaledo notnecessarilyproducesimilarsynergiesortradeoffsatotherscales.Althoughrobustpatternsexist forrelationshipsamongcertainregulatingservices, notallhavesimplescalingrules.Thisisespecially trueforrelationshipsbetweenfoodproductionand waterquality,wherelocalchangesinrelationshipsdo notnecessarilyscaleuptowatershedwhereadifferentsetofdriversareoperating(gures 3 – 5 ).Hence, eld-specicmanagementpractices(e.g.nutrientor stormwatermanagement)toreducelocaltradeoffs mightnotbesufcienttomitigatesuchtradeoffsat 12

PAGE 14

Environ.Res.Lett. 13 (2018)054020 thelandscapescale(Arabi etal 2006 ,Ahiablame etal 2012 ).Rather,managersanddecision-makers needtoconsiderthedrivers,social-ecologicalcomplexities,andmechanismsofESdynamicsthatare appropriatetoscalesofwatersheds.Intheeventof dataandtimeconstraints,ourresultsdidprovideinitialdocumentationtomanagersonwhichESmight berobustandpredictableacrossscales,andwhich onesarelikelytobesensitivetochangesinspatial scales.Forexample,synergiesamongwaterquality,soilretentionandclimateregulationESacross scalessuggestthatmanagementandpolicyresponses (e.g.afforestation,covercrops,conservationtillage) atdifferentscalesmayleadtosimilarsynergistic outcomes. Timealsoplaysanimportantrole.Analyzing ESrelationshipsatasingletime,asinearlierstudies,wouldemphasizeeffectsofspatialvariabilityof drivers(e.g.landuse/cover,managementpractices), butoverlookeffectsofdriverswhosetemporalvariationsplayamorecriticalrole(e.g.precipitation). Associal-environmentalconditionschange,EStradeoffsandsynergiesmayalsovaryintheirmagnitude anddirections.Hence,timelyassessmentandmonitoringofESandtheirrelationshipsareneeded,and canhelpavoidsurprisingtradeoffsandtakeadvantage ofemergingsynergies.Italsopointstothenecessitytoleveragelong-termmonitoringprograms(e.g. Long-TermEcologicalResearch,NationalEcological ObservationNetwork,andCriticalZoneObservatoriesintheUnitedStates)forESresearch.Such long-termandextensiveeffortscouldrevealhowES relationshipschangeovertime,whatfactorsdrive theirdynamics,andanytimelagsorlegacyeffects tobetterguidemanagementstrategiesforsustaining multipleES.AcknowledgmentsThisresearchwassupportedbyNationalScienceFoundationWaterSustainabilityandClimateProgram (DEB-1038759),NorthTemperateLakesLong-Term EcologicalResearch(DEB-1440297),andWisconsin AlumniResearchFoundati onBridgetotheFuture funding.MGTalsoacknowledgessupportfromthe VilasTrustoftheUniversityofWisconsin-Madison. JQalsoacknowledgesUSDANationalInstituteof FoodandAgriculture,Hatch(FLA-FTL-005640)and McIntire-Stennis(1014703)projectsforpartialnancialsupportofthiswork.Publicationofthisarticle wasfundedinpartbytheUniversityofFlorida OpenAccessPublishingFund.Specialthanksto JennySeifertandElizabethKatt-Reindersforthe developmentofscenarios,andPavelPinkasfor assistanceinparallelcomputingofmodelsimulations.Wethankanonymousreviewersandtheeditor forhelpfulcommentsonanearlierversionofthe manuscript.ORCIDiDsJiangxiaoQiu https://orcid.org/0000-0002-37415213 StephenRCarpenter https://orcid.org/0000-00018097-8700 EricGBooth https://orcid.org/0000-0003-21916627 MelissaMotew https://orcid.org/0000-0003-16864754 SamuelCZipper https://orcid.org/0000-0002-87355757 ChristopherJKucharik https://orcid.org/0000-00020400-758X LoheideStevenPII https://orcid.org/0000-00031897-0163 MonicaGTurner https://orcid.org/0000-0003-19032822ReferencesAhiablameLM,EngelBAandChaubeyI2012Effectivenessoflow impactdevelopmentpractices:literaturereviewand suggestionsforfutureresearch WaterAirSoilPollut. 223 4253–73 AllenTFHandStarrTB1982 HierarchyPerspectivesforEcological Complexity (Chicage,IL:UniversityofChicagoPress) AndersonBJ etal 2009Spatialcovariancebetweenbiodiversityand otherecosystemservicepriorities J.Appl.Ecol. 46 888–96 AnderssonE etal 2015Scaleandcontextdependenceofecosystem serviceprovidingunits Ecosyst.Serv. 12 157–64 ArabiM,GovindarajuRS,HantushMMandEngelBA2006Role ofwatershedsubdivisiononmodelingtheeffectivenessofbest managementpracticeswithSWAT J.Am.WaterResour. Assoc. 42 513–28 AsbjornsenH etal 2014Targetingperennialvegetationin agriculturallandscapesforenhancingecosystemservices Renew.Agric.FoodSyst. 29 101–25 BennettEM,PetersonGDandGordonLJ2009Understanding relationshipsamongmultipleecosystemservices Ecol.Lett. 12 1394–404 BoothEG etal 2016Fromqualitativetoquantitativeenvironmental scenarios:translatingstorylinesintobiophysicalmodeling inputsatthewatershedscale Environ.Model.Softw. 85 80–97 BradfordMA,WarrenIIRJ,BaldrianP,CrowtherTW,Maynard DS,OldeldEE,WiederWR,WoodSAandKingJR2014 Climatefailstopredictwooddecompositionatregionalscales Nat.Clim.Change 4 625–30 CarlsonMA,LohseKA,McIntoshJCandMcLainJET2011 Impactsofurbanizationongro undwaterqualityandrecharge inasemi-aridalluvialbasin J.Hydrol. 409 196–211 CarpenterSR etal 2015Plausiblefuturesofasocial-ecological system:Yaharawatershed,Wisconsin,USA Ecol.Soc. 20 10 CarpenterSR,BoothEGandKucharikCJ2017Extreme precipitationandphosphorusloadsfromtwoagricultural watersheds Limnol.Oceanogr. ( https://doi.org/10.1002/ lno.10767 ) CarpenterSR etal 2009Scienceformanagingecosystemservices: beyondthemillenniumecosystemassessment Proc.NatlAcad. Sci. 106 1305–12 ClarkJS etal 2001Ecologicalforecasts:anemergingimperative Science 293 657–60 CordAF etal 2017Towardssystematicanalysesofecosystem servicetrade-offsandsynergies:mainconcepts,methodsand theroadahead Ecosyst.Serv. 28 264–72 13

PAGE 15

Environ.Res.Lett. 13 (2018)054020 CostanzaR,FisherB,MulderK,LiuSandChristopherT2007 Biodiversityandecosystemservices:amulti-scaleempirical studyoftherelationshipbetweenspeciesrichnessandnet primaryproduction Ecol.Econ. 61 478–91 CraigJR,LiuGandSoulisED2010Runoff–inltration partitioningusinganupscaledGreen–Amptsolution Hydrol. Process. 24 2328–34 CummingGS,CummingDHandRedmanCL2006Scale mismatchesinsocial-ecologicalsystems:causes, consequences,andsolutions Ecol.Soc. 11 14 DonnerSDandKucharikCJ2003Evaluatingtheimpactsofland managementandclimatevariabilityoncropproductionand nitrateexportacrosstheUpperMississippiBasin Glob. Biogeochem.Cycles 17 1085 EllisECandRamankuttyN2008Puttingpeopleinthemap: anthropogenicbiomesoftheworld Front.Ecol.Environ. 6 439–47 FoleyJA etal 1996Anintegratedbiospheremodeloflandsurface processes,terrestrialcarbonbalance,andvegetationdynamics Glob.Biogeochem.Cycles 10 603–28 FoleyJA etal 2011Solutionsforacultivatedplanet Nature 478 337–42 GardnerRH,KempWM,KennedyVSandPetersenJE2001 ScalingRelationsinExperimentalEcology (NewYork: ColumbiaUniversityPress) Garc aDandChacoffNP2007Scale-dependenteffectsofhabitat fragmentationonhawthornpollination,frugivory,andseed predation Conserv.Biol. 21 400–11 GoldsteinJH etal 2012Integratingecosystem-servicetradeoffs intoland-usedecisions Proc.NatlAcad.Sci. 109 7565–70 Haines-YoungR,PotschinMandKienastF2012Indicatorsof ecosystemservicepotentialatEuropeanscales:mapping marginalchangesandtrade-offs Ecol.Indic. 21 39–53 HaoRF,YuDYandWuJ2017Relationshipbetweenpaired ecosystemservicesinthegrasslandandagro-pastoral transitionalzoneofChinausingtheconstraintlinemethod Agric.Ecosyst.Environ. 240 171–81 HeffernanJB etal 2014Macrosystemsecology:understanding ecologicalpatternsandprocessesatcontinentalscales Front. Ecol.Environ. 12 5–14 HollandRA etal 2011Spatialcovariationbetweenfreshwaterand terrestrialecosystemservices Ecol.Appl. 21 2034–48 HoweC,SuichH,ViraBandMaceGM2014Creatingwin-wins fromtrade-offs?Ecosystemservicesforhumanwell-being:a meta-analysisofecosystemservicetrade-offsandsynergiesin therealworld Glob.Environ.Change 28 263–75 JonesKB etal 2006Multiscalerelationshipsbetweenlandscape characteristicsandnitrogenconcentrationsinstreams Scaling andUncertaintyAnalysisinEcology (Dordrecht:Springer) pp205–24 KareivaP,WattsS,McDonaldRandBoucherT2007 Domesticatednature:shapinglandscapesandecosystemsfor humanwelfare Science 316 1866–9 KochEWetal 2009Non-linearityinecosystemservices:temporal andspatialvariabilityincoastalprotection Front.Ecol. Environ. 7 29–37 KucharikCJ2003Evaluationofaprocess-basedAgro-Ecosystem Model(Agro-IBIS)acrosstheUSCornBelt:simulationsofthe interannualvariabilityinmaizeyield EarthInteract. 7 1–33 KucharikCJ etal 2000Testingtheperformanceofadynamic globalecosystemmodel:waterbalance,carbonbalance,and vegetationstructure Glob.Biogeochem.Cycles 14 795–825 KucharikCJandTwineTE2007Residue,respiration,and residuals:evaluationofadynamicagroecosystemmodelusing eddyuxmeasurementsandbiometricdata Agric.Forest Meteorol. 146 134–58 LevinSA1992Theproblemofpatternandscaleinecology:the RobertHMacArthurawardlecture Ecology 73 1943–67 MaesJ,ParacchiniML,ZulianG,DunbarMBandAlkemadeR 2012Synergiesandtrade-offsbetweenecosystemservice supply,biodiversity,andhabitatconservationstatusinEurope Biol.Conserv. 155 1–12 ManzoniSandPorporatoA2009Soilcarbonandnitrogen mineralization:theoryandmodelsacrossscales Soil.Biol. Biochem. 41 1355–79 McIsaacGF,DavidMBandMitchellCA2010 Miscanthus and switchgrassproductionincentralIllinois:impactson hydrologyandinorganicnitrogenleaching J.Environ.Qual. 39 1790–9 MEA2005a EcosystemsandHumanWell-being:Synthesis (Washington,DC:IslandPress) MEA2005b EcosystemsandHumanWell-being:Scenarios (Washington,DC:IslandPress) MeachamM,QueirozC,Norstr ¨ omAandPetersonG2016 Social-ecologicaldriversofmultipleecosystemservices:what variablesexplainpatternsofecosystemservicesacrossthe Norrstr ¨ omdrainagebasin? Ecol.Soc. 21 14 MotewM etal 2017TheinuenceoflegacyPonlakewaterquality inaMidwesternAgriculturalWatershed Ecosystems 20 1468–82 MotewMMandKucharikCJ2013Climate-inducedchangesin biomedistribution,NPP,andhydrologyintheUpper MidwestUS:acasestudyforpotentialvegetation J.Geophys. Res.Biogeosci. 118 248–64 NangiaV,deFraitureCandTurralH2008Waterquality implicationsofraisingcropwaterproductivity Agric.Water Manage. 95 825–35 O ’ NeillRV1986 AHierarchicalConceptofEcosystems (Princeton UniversityPress) PetersDP,YaoJ,HuennekeLF,GibbensRP,HavstadKM, HerrickJE,RangoAandSchlesingerWH2006A frameworkandmethodsforsimplifyingcomplexlandscapes toreduceuncertaintyinpredictions ScalingandUncertainty AnalysisinEcology (Dordrecht:Springer)pp131–46 PowerAG2010Ecosystemservicesandagriculture:tradeoffsand synergies Phil.Trans.R.Soc.Lond.BBiol.Sci. 365 2959–71 QiuJ etal 2018Scenariosrevealpathwaystosustainfuture ecosystemservicesinanagriculturallandscapeEcol.Appl. 28 119–34 QiuJandTurnerMG2013Spatialinteractionsamongecosystem servicesinanurbanizingagriculturalwatershed Proc.Natl Acad.Sci. 110 12149–54 QiuJandTurnerMG2015Importanceoflandscapeheterogeneity insustaininghydrologicecosystemservicesinanagricultural watershed Ecosphere 6 1–19 QiuJ,WardropperCB,RissmanARandTurnerMG2017Spatial tbetweenwaterqualitypoliciesandhydrologicecosystem servicesinanurbanizingagriculturallandscape Landsc.Ecol. 32 59–75 RCoreTeam2016R:Alanguageandenvironmentforstatistical computing RFoundationforStatisticalComputing,Vienna, Austria ( www.R-project.org/ ) RandallGW,HugginsDR,RusselleMP,FuchsDJ,NelsonWW andAndersonJL1997Nitratelossesthroughsubsurfacetile drainageinconservationreserveprogram,alfalfa,androw cropsystems J.Environ.Qual. 26 1240–7 Raudsepp-HearneCandPetersonG2016Scaleandecosystem services:howdoobservation,management,andanalysisshift withscale—lessonsfromQu ebec Ecol.Soc. 21 16 Raudsepp-HearneC,PetersonGDandBennettEM2010 Ecosystemservicebundlesforanalyzingtradeoffsindiverse landscapes Proc.NatlAcad.Sci. 107 5242–7 RenardD,RhemtullaJMandBennettEM2015Historical dynamicsinecosystemservicebundles Proc.NatlAcad.Sci. 112 13411–6 RickettsTH etal 2016Disaggregatingtheevidencelinking biodiversityandecosystemservices Nat.Commun. 7 13106 Rodr guezJP etal 2006Trade-offsacrossspace,time,and ecosystemservices Ecol.Soc. 11 28 RoseKC etal 2017Historicalfoundationsandfuturedirectionsin macrosystemsecology Ecol.Lett. 20 147–57 SchneiderDC2001Theriseoftheconceptofscaleinecology BioScience 51 545–53 14

PAGE 16

Environ.Res.Lett. 13 (2018)054020 ScholesR,ReyersB,BiggsR,SpierenburgMandDuriappahA 2013Multi-scaleandcross-scaleassessmentsof social–ecologicalsystemsandtheirecosystemservices Curr. Opin.Environ.Sustain. 5 16–25 ScholesRJ2017Takingthemumbooutofthejumbo:progress towardsarobustbasisforecologicalscaling Ecosystems 20 4–13 SimsDW etal 2008Scalinglawsofmarinepredatorsearch behaviour Nature 451 1098–102 SoyluME,KucharikCJandLoheideIISP2014Inuenceof groundwateronplantwateruseandproductivity: developmentofanintegratedecosystem–variablysaturated soilwaterowmodel Agric.ForestMeteorol. 189190 198–210 SpakeR etal 2017Unpackingecosystemservicebundles:towards predictivemappingofsynergiesandtrade-offsbetween ecosystemservices Glob.Environ.Change. 47 37–50 SteffenW etal 2015Planetaryboundaries:guidinghuman developmentonachangingplanet Science 347 1259855 SunZ,LiuZ,HeCandWuJ2016Multi-scaleanalysisof ecosystemservicetrade-offsinurbanizingdrylandsofChina:a casestudyintheHohhot-Baotou-Ordos-Yulinregion Acta Ecol.Sin. 36 4881–91 TomschaSandGergelS2016Ecosystemservicetrade-offsand synergiesmisunderstoodwithoutlandscapehistory Ecol.Soc. 21 43 TurnerKG,OdgaardMV,BcherPK,DalgaardTandSvenning J-C2014BundlingecosystemservicesinDenmark:trade-offs andsynergiesinaculturallandscape Landsc.UrbanPlan. 125 89–104 WardropperCB,GillonS,MaseAS,McKinneyEA,CarpenterSR andRissmanAR2016Localperspectivesandglobal archetypesinscenariodevelopment Ecol.Soc. 21 12 WhittakerRJ1999Ecology:scaling,energeticsanddiversity Nature 401 865–6 WiensJA1989Spatialscalinginecology Funct.Ecol. 3 385–97 WuJ2004Effectsofchangingscaleonlandscapepatternanalysis: scalingrelations Landsc.Ecol. 19 125–38 WuJ,JonesKB,LiHandLoucksOL2006 Scalingand UncertaintyAnalysisinEcology:MethodsandApplications (Dordrecht:Springer) ZhengH etal 2016Usingecosystemservicetrade-offstoinform waterconservationpoliciesandmanagementpractices Front. Ecol.Environ. 14 527–32 ZipperSC,SoyluME,BoothEGandLoheideSP2015Untangling theeffectsofshallowgroundwaterandsoiltextureasdriversof subeld-scaleyieldvariability WaterResour.Res. 51 6338–58 15