HAL Id: hal-01631349
https://hal.archives-ouvertes.fr/hal-01631349
Submitted on 14 May 2018
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of
sci-entific research documents, whether they are
pub-lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Quantitative criteria for choosing targets and indicators
for sustainable use of ecosystems
Axel G. Rossberg, Laura Uusitalo, Torsten Berg, Anastasija Zaiko, Anne
Chenuil, Maria C. Uyarra, Angel Borja, Christopher P. Lynam
To cite this version:
Axel G. Rossberg, Laura Uusitalo, Torsten Berg, Anastasija Zaiko, Anne Chenuil, et al.. Quantitative
criteria for choosing targets and indicators for sustainable use of ecosystems. Ecological Indicators,
Elsevier, 2017, 72, pp.215-224. �10.1016/j.ecolind.2016.08.005�. �hal-01631349�
EcologicalIndicators72(2017)215–224
ContentslistsavailableatScienceDirect
Ecological
Indicators
jou rn al h om ep a g e :w w w . e l s e v i e r . c o m / l o c a t e / e c o l i n d
Quantitative
criteria
for
choosing
targets
and
indicators
for
sustainable
use
of
ecosystems
Axel
G.
Rossberg
a,∗,
Laura
Uusitalo
b,
Torsten
Berg
c,
Anastasija
Zaiko
d,
Anne
Chenuil
e,
María
C.
Uyarra
f,
Angel
Borja
f,
Christopher
P.
Lynam
gaCentreforEnvironment,FisheriesandAquacultureScience(Cefas),PakefieldRoad,LowestoftNR330HT,UKandSchoolofBiologicalandChemical
Sciences,QueenMaryUniversityofLondon,327MileEndRd,LondonE1,UK
bMarineResearchCentre,FinnishEnvironmentInstitute(SYKE).Mechelininkatu34a,P.O.Box140,FI-00251Helsinki,Finland cMariLimAquaticResearchGmbH,Heinrich-Wöhlk-Straße14,24232Schönkirchen,Germany
dMarineScienceandTechnologyCenter,KlaipedaUniversity,H.Manto84,LT92294,Klaipeda,Lithuania eAix-MarseilleUniv,UnivAvignon,CNRS,IRD,IMBE,Marseille,France
fAZTI-Tecnalia,HerreraKaia,Portualdeas/n,20100Pasaia,Spain
gCentreforEnvironment,FisheriesandAquacultureScience(Cefas),PakefieldRoad,LowestoftNR330HT,UK
a
r
t
i
c
l
e
i
n
f
o
Articlehistory:
Received12December2015
Receivedinrevisedform4August2016 Accepted6August2016
Availableonline20August2016 Keywords:
Goodenvironmentalstatus MarineStrategyFrameworkDirective Sustainableuse
Assessment Ecologicalindicators
a
b
s
t
r
a
c
t
Wide-ranging,indicator-basedassessmentsoflarge,complexecosystemsareplayinganincreasingrole
inguidingenvironmentalpolicyandmanagement.AnexampleistheEU’sMarineStrategyFramework
Directive,whichrequiresMemberStatestotakemeasurestoreach“goodenvironmentalstatus”(GES)
inEuropeanmarinewaters.However,formulationofindicatortargetsconsistentwiththeDirective’s
high-levelpolicygoalofsustainableusehasprovenchallenging.Wedevelopaspecific,quantitative
interpretationoftheconceptsofGESandsustainableuseintermsofindicatorsandassociatedtargets,
bysharplydistinguishingbetweencurrentusestosatisfycurrentsocietalneedsandpreferences,and
unknownfutureuses.Wearguethatconsistenttargetstosafeguardfutureusesderivefroma
require-mentthatanyenvironmentalstateindicatorshouldrecoverwithinadefinedtime(e.g.30years)to
itspressure-freerangeofvariationwhenallpressuresarehypotheticallyremoved.Withinthese
con-straints,specifictargetsforcurrentusesshouldbeset.Routestoimplementationofthisproposalfor
indicatorsoffish-communitysizestructure,populationsizeofselectedspecies,eutrophication,impacts
ofnon-indigenousspecies,andgeneticdiversityarediscussed.Importantpolicyimplicationsarethat(a)
indicatortargetranges,whichmaybewiderthannaturalranges,systematicallyandrationallyderive
fromourproposal;(b)becauserelevantstateindicatorstendtorespondslowly,correspondingpressures
shouldalsobemonitoredandassessed;(c)supportofcurrentusesandsafeguardingoffutureusesare
distinctmanagementgoals,theyrequiredifferenttypesoftargets,decisionprocesses,andmanagement
philosophies.
©2016PublishedbyElsevierLtd.
1. Introduction
1.1. Fromqualitativetoquantitativecriteriaforindicator selection
Ecologicalindicatorsareincreasinglybeingusedinrule-based managementschemeswhereindicatorvaluesoutsidetheir
respec-∗ Correspondingauthor.
E-mailaddresses:a.rossberg@qmul.ac.uk(A.G.Rossberg),
Laura.Uusitalo@ymparisto.fi,laura.uusitalo@iki.fi(L.Uusitalo),berg@marilim.de
(T.Berg),anastasija.zaiko@jmtc.ku.lt(A.Zaiko),anne.chenuil@imbe.fr(A.Chenuil),
mcuyarra@azti.es(M.C.Uyarra),aborja@azti.es(A.Borja),chris.lynam@cefas.co.uk
(C.P.Lynam).
tivetargetrangestriggermanagementaction.Thequestionwhich propertiesecologicalindicatorsshouldhaveforthispurposehas oftenbeenaddressedintheliterature(Elliott,2011Queirósetal., 2016;RiceandRochet,2005).Anexamplerelevantforassessment andmanagementofmarineecosystemsisthesetofcriteria pro-posedbyICES(2001),whichformsthebasisoftheRiceandRochet (2005)criteria.Theserelatetoconcreteness,theoreticalbasis, pub-licawareness,cost,measurability,representationthroughhistoric data,sensitivity,responsiveness,andspecificityofindicators.Alist byElliott(2011)containing18criteriagoesbeyondtheRiceand Rochet (2005)list,in requiring that indicators(and monitoring parameters)shouldbeanticipatory,broadlyapplicableand inte-grativeoverspaceandtime,interpretable,havelowredundancy,
http://dx.doi.org/10.1016/j.ecolind.2016.08.005
benon-destructive,time-boundedandtimely.Foradetailedreview andanalysisofindicatorselectioncriteria,seeQueirósetal.(2016). However,practicallyallpublishedspecificationsofdesiderata forecologicalindicatorsandtheirmanagementtargetsremainat aqualitativelevel,despitecontainingsomequantitative compo-nents(e.g.reasonablecostincomparisonwithexpectedbenefits). Thishastheadvantageofflexibilitytoaccommodatevariationin preferencesand prioritiesof differentstakeholder groups—after all,policiesmanagehumanactivitiesratherthanthemarine envi-ronment(Elliott,2013).However,expertscanvarywidelyintheir findingswhenevaluatingindicatorsaccordingtothesame crite-ria(Riceand Rochet,2005),which questionstheideathatsuch criteriaprovideanobjectivebasisforindicatorselection.Another disadvantageisthatthescientificproblemofdevelopingindicators andmonitoringprogramsandthescientificandsocietalchallenge offindingappropriatetargetranges fortheseindicatorsremain vaguely specified.This maylead toinconsistencies in specified targetranges,inefficientuseoflimitedmonitoringcapacity,and uncertaintyaboutthemostappropriateuseofresearchcapacityfor refiningindicatorsandtargetsorfillingpotentialgapsinindicator suites(Borjaetal.,2012).
Ideally,aquantitative,generic,andbroadlyacceptedframework wasavailableforchoosingindicatorsandsettingtargets,somaking thisaresearchanddevelopmenttasktodeliveraproduct accord-ingtospecifications,ratherthanasocialprocessoffindingcommon positionsinanuncertainspace.Suchaquantitativeframeworkdoes currentlynotexist.Environmentalpolicydocumentstendto spec-ifytheiroverallhigh-levelobjectivesinaqualitativelanguage.The purposeofthiscontributionistopropose,asawayforward,a quan-titativeinterpretationofthisqualitativelanguage,whichcanthen betestedforpoliticalacceptance.Beingdeliberatelyconstructed buildingonjustafewgenericprinciples,ourproposalisnecessarily somewhatabstractandrigid,andsoshouldnotbemisunderstood asadirectprescriptionofpolicy.Moreplausibly,itwillserveasa scientificallyanchoredorientationpointforpoliticaldecision mak-ing.
Asa specificpolicydocument whichiscurrentlywidely dis-cussedinEurope,wechosetofocushereontheMarineStrategy FrameworkDirective(MSFD;EC,2008)oftheEuropeanUnion(EU). Theprinciplesbeinginvokedforsettingtargetsarenotconsistent withinthecommunityimplementingtheMSFD.ForCochraneetal. (2010),forexample,thetargetisanecosystemnearlyunperturbed byhumans,ICES(2014a)primarilyrequirethatecosystem func-tionsarenotdegraded,Rogersetal.(2010)andICES(2014b)refer toabundancesthatcanrecoverfromperturbationorhavebeen observedtobehistoricallystable,andPietetal.(2010)interpret the“safebiologicallimits”offishstocksasthoseproducing maxi-mumsustainableyield.Weshallhereconcentrateonpolicyneeds undertheMSFD.However,theframeworkweproposedmightbe generallyusefulforlinkingassessmentsofaquaticorterrestrial ecosystemstohigh-levelpolicygoals.
1.2. Theconceptofsustainableuse
TheMSFDrequires fromEUmember statestodetermine, in acollaborative manner,specificenvironmentaltargetsand cor-respondingquantitativeindicatorsthattogetherrepresent“good environmentalstatus”(GES).ItdefinesGESas:
theenvironmentalstatusofmarinewaterswheretheseprovide ecologicallydiverseanddynamicoceansandseaswhichareclean, healthyandproductivewithintheirintrinsicconditions,andthe useofthemarineenvironmentisatalevelthatissustainable,thus safeguardingthepotentialforusesandactivitiesbycurrentand futuregenerations[...].
Table1
Comparisonofconceptsofweaklyandstronglysustainableuse.
Weaklysustainableuse Stronglysustainable use
Typesofrelevant services
Societalchoice Aprioriunknown Valueofservicesused Mostlyknown Unknownoruncertain Valuetobepreserved Anthropogeniccapital
plusnaturalcapital
Naturalcapital Natureoftypicaltarget Thepoint
correspondingto optimallong-termuse
Therangeallowing timelyrecovery Management
philosophy
Optimalcontrol(asin controltheory)
Limitationofpressures
Thelastpassageisavariationofthedefinitionofsustainable
developmentfromtheBrundtlandReport(WorldCommissionon
EnvironmentandDevelopment,1987):
Sustainabledevelopmentisdevelopmentthatmeetstheneeds ofthepresentwithoutcompromisingtheabilityoffuture genera-tionstomeettheirownneeds.
Important is that this definition recognizes that needs of future generations might be different from current needs. By referringto“thepotentialforuses andactivitiesby[...] future generations”,theMSFDfollowsthistradition.Uncertaintyabout futureuses,andsovalues,ofresourcesnaturallyleadstostrong notionsofsustainability1thataimatindependentmaintenanceor
enhancementofvariousformsofnaturalandnon-naturalcapital (Figge, 2005).Contrastingly, weaksustainabilitypermits substi-tutionofnaturalwithmanufacturedcapital,implicitlyassuming goodknowledge oftheirrespectivefuture values (Figge,2005). Correspondingly,wesayhere“stronglysustainable”foruseofthe environmentthatdoesnotconstrainusagechoicesandcapabilities offuturegenerations,and“weaklysustainable”forusethatsimply canbecontinuedindefinitelyinitscurrentform(conceivableare evenweakernotions).Thedistinctionbetweenthetwoconceptsis brieflysummarizedinTable1.
Thebest-knownexampleofusageof“sustainable”inourweak sense in the marine ecology context is “maximum sustainable yield”(MSY). Management for MSY alone does not necessarily implysustainabilitybythestrongerdefinition,becausechanges tothewiderecosystemresultingfromexploitationmaybe irre-versible.TheMSFDreferstoweaklysustainableuse,forexample throughtheadjective“productive”intheGESdefinitionaboveand inaclarifyingCommissionDecision(EC, 2010), whichexplicitly specifiesexploitationatMSYasatarget.
Our considerationshere concentrateon stronglysustainable use,thusmarkingthelimitswithinwhichweaklysustainableuse options can be explored. From above considerations it follows thatconstraintsimposedbystrongsustainabilitywillgenerallybe weakerthanthosefollowingfromspecificweaklysustainableuse objectives;apotentialsourceofconfusiontokeepinmind.
Theoperationalizationofthestrongconceptofsustainableuse inthecontextofmarinemanagementhasbeensubjectofextensive discussionintheworkoftheInternationalCouncilforthe Explo-rationoftheSeas(ICES,2005,2010,2013).ICESarguedthat,since theneedsandpreferencesoffuturegenerationsareunknownto us,sustainableusemeansnottoperturb theecosystemtosuch adegreethatrecoveryfromtheseperturbationsisimpossibleor unacceptablyslow(seealsoFAO,2009).Inotherwords,under sus-tainableusethesystemmustremaincapableofrecoveringtoan unperturbedstateoveranacceptabletimespan.
1Othersmotivatestrongsustainabilitybynon-substitutabilityofcriticalnatural
capital,incomprehensionofnaturalsystems,irreversibilityoflosses,andethically (DietzandNeumayer,2007).
A.G.Rossbergetal./EcologicalIndicators72(2017)215–224 217
Whenmakingthisideaoperational,twopointsneedbearingin mind.Firstly,sincethemanagementobjectiveissustainableuse inthepresent ratherthan inthepast,theunperturbed stateis notnecessarilyahistoricorpre-historicstate,butthestatethat wouldbereachedinthelongtermifallanthropogenicpressures wereremoved.Secondly,theunperturbedstateitselfisnotfixed butundergoesnaturalfluctuations.
Developingaquantitativeinterpretationofsustainableuse,ICES (2010)proposedtofocusindicatorselectiononecosystem compo-nentsthat(1)areunderpressureand(2)forwhichrecoveryfrom pressuresissloworimpossible.Indicatorsarethenchosento quan-tifythestateofthesecomponentsorfeatures,called“vulnerable components”below,andthepressuresonthem.Thismethod, how-ever,leavesopentheproblemofderivingtargetvaluesforthese indicators.
HeretheapproachofICESisthereforereversed.Aruleis pro-posedforsettingtargetrangesforarbitraryquantitativeindicators ofecosystemstatesuchthat,forecosystemcomponentsthatare notvulnerableinthesenseabove,thetargetswill“automatically” bemetunderalmostallcircumstances,whileindicatorsrelatingto vulnerablecomponentsareeasilydrivenoutoftheirtargetranges underinappropriatemanagement, which is then interpretedas unsustainableuse.Thatis,theruleforsettingtargetranges implic-itlyselectsindicatorscriticalformonitoringsustainableuse,and these implicitly identify vulnerable ecosystem components, so focusingassessmentsandmanagementonprotectingthelatter.
Theselectedstateindicatorsarecomplementedwithasetof correspondingpressureindicators,andpotentiallywithadditional indicators quantifyingstate along causalchainslinking anthro-pogenicpressurestovulnerableecosystemcomponents.
2. Theproposal
2.1. Choosingtargetrangesforstateindicators
Theruleforchoosingindicatortargetrangesproposedhere con-tainsasinglefreeparameter,thelongestacceptablemeanrecovery timeR(precisely:thelargestacceptableexpectationvalueoftime torecovery).ThevalueofRisamatterofsocietalchoice.Itcouldbe related,e.g.,tothedurationofpolicycyclesorthehumanlifecycle. AccordingtoadefinitionbytheFAO(2009),forexample, ‘signifi-cantadverseimpacts’onecosystemswilltypicallyhaverecovery times exceeding 5–20 years. Consistent use of the same value ofRwhensettingtargetrangesfordifferentindicatorsimproves consistencyamongmanagementgoals.Societymightrequire com-parisonsoftheimplicationsofdifferentchoicesofR inorderto makeaninformeddecisiononitsnumericalvalue.Weproposethat, toremainconsistentwithintergenerationalfreedomofchoice,R shouldnotexceedtheapproximatehumangenerationtimeof30 years,andassumeR≈30yearsinexampleswediscuss.
Now,letIstandforanyunivariateindicatorofecosystemstate. Theindicatorishereunderstoodasbeingdefineddirectlyinterms ofecosystem statevariables,rather thanby aprotocol to mea-surethese.Withoutanthropogenicpressures,thevalueofIwould relaxtoandthennaturallyfluctuatearoundsometypicalvalue.The resultingdistributionofvaluesIcanbecalleditspressure-free,and, inthissense,naturaldistribution.
Onecandefineanaturalrangeofvariation
Ilow,Ihigh,for exam-plebychosingIlowasthe2.5%quantileofthenaturaldistribution,
andIhighasthe97.5%quantile.Undernaturalconditions,the
indi-catoristheninthenaturalrange95%ofalltimes.Becausedirect observationdatacorrespondingtonaturalorpristineconditions doesnotnecessarilyexist,inferentialmethodstodetermine nat-uralrangeswilloftenberequired.Wenowproposetochoosethe targetrangeforanyindicatorastherangeofvaluesfromwherethe
Fig.1.Illustrationofproposedapproachforchoosingtargetranges.Thetargetrange ofanindicatorisdeterminedastherangeofvaluesfromwhichittakes,onaverage, atmostatimeRtoreachthenaturalrangeinahypotheticalsituationwithout anthropogenicpressures.Dottedlinesindicatethewidthofthetargetrange,dashed lineshypotheticalaveragerelaxationtrajectories,thegreyareathenaturalrange, andtheraggedsolidlineaconceivabletrajectoryoftheindicatorforanecosystem instronglysustainableuse.Inpractice,thetargetrangemayneedtobenarrowed totakemeasurementuncertaintyandmodeluncertaintyintoaccount.
meantimetoreachthenaturalrangewhenallpressuresare, hypo-thetically,removedisnotlargerthantheacceptablemeanrecovery timeR.TheideaisillustratedinFig.1.Managementunderthisrule impliesthat,afteranaveragetransitionperiodR,future genera-tionscanusethecorrespondingecosystemcomponentinanyform thatwouldhavebeenalmostcertainlypossibleundernatural con-ditions,provided“almostcertain”isinterpretedasmeaning95% probability.
Theindicator’snatural rangedepends onexternalfactors,in thecaseoftheMSFDdescribedas“theassociatedphysiographic, geographic,geologicalandclimaticfactors”.Complicating,Earth’s climateisonatrajectoryofdirectedlong-termchange,andthe nat-uralrangecorrespondingtocurrentclimaticconditionsgradually changes.Targetrangesshouldbechosensuchthatrelaxationto thenaturalrangewithinRonaverageispossibleeventhoughit changesovertime.
2.2. Choosingrelevantstateindicators
Byourproposal,allaspectsofecosystemstatearepotentially relevant.Theseincluding,e.g.,thephysicalseascape,water tem-peratureandflows,chemicalwatercomposition,thestructuring elementsoftheecosystemsuchashabitat-formingspecies,top predators,andkeyresourcespecies,butalsoendangeredspecies, groupsorhabitats,andhigh-levelpropertiessuchasspecies rich-ness,communitybiomassandproduction.Iffollowsfromourrule ofchoosing targetrangesthatamong thesethestateindicators thatarerelevantinpractice(below“relevantindicators”)arethose whichareoutsidetheirtargetrangesorlikelytobepushedoutof theirtargetrangesbyprevalentorforeseeableanthropogenic pres-sures.Setsofcandidatestateindicatorscaninitiallybescanned forrelevancebyaskingiftheirrecoverytothenaturalrangecan conceivablylastlongerthanR.
2.3. Choosingrelevantpressureindicatorsandtheirtargets
Weproposetochoosethecombinedtargetrangesofpressure indicatorsinsuchawaythat,whenpressuresaremaintained indef-initelywithintargetranges,allecosystemstateindicatorsreturnto theirtargetrangesandthenremainwithintheserangesduring95%
oftime.2Tocopewithempiricaluncertaintyoverpressure-state
relationships,an adaptivemanagement schemewhere pressure targetrangesareiterativelyrevisedbasedonobservedchangesin statewilloftenbeadequate.Analogouslytothestateindicators, relevantpressureindicatorsarethosewhichareoutsideorlikely tobebroughtoutsidetheirtargetranges,andtheycanbefoundby asimilarscanningprocedure.
2.4. Causalrelationsandsupportingindicators
Somevulnerable ecosystemcomponentsarenot ornot only affected by direct anthropogenic pressures, but also indirectly via causal chains through other ecosystem components (Borja etal.,2010b).Awell-knownexamplearechangesinpopulations at highertrophic levelscaused, through bottom-up control, by populations at lower trophic levels, in turn influenced, e.g. by fluvialnutrientinput.Ifpressure-staterelationshipsalongthese causalchainsarenotwellunderstood,monitoringofintermediate ecosystemcomponents,e.g.abundanceofprimaryorsecondary producers,canplayanimportantroleinsupportingdecision mak-ingby managers.Existence of causal“webs”rather than linear chainsheightensthisneed(Borjaetal.,2010b).Effectivesupporting indicatorswillhavecomparativelywell-understoodcausallinksto bothanthropogenicpressuresandvulnerableecosystem compo-nents,somaximisingtheinformationoncausalrelationsbetween pressuresandstates.Targetrangesforsuchsupportingstate indica-torscanbedeterminedfollowingthesamelogicasthosefordirect anthropogenicpressures.
2.5. Suitesofindicatorsandcorrelationsbetweenindicators Toadequatelycapturethestatus ofcomplexmarine ecosys-tems,largesetsofindicatorsareoftenproposed.Thequestionthen arisesbywhichcriterionpotentiallyredundantindicatorscouldbe identifiedandeliminated.Withinthepresentframework,anatural answerarisesasfollows:considerasituationwhere,undercurrent andforeseeablepressures,someformulapredictsthevaluesofone indicatorIinasuiteofstateindicatorsfromthoseoftheother indi-catorsuptoadifferenceD.ThenIcanbereplacedbyDinthissuite withoutlossofinformation.WhenDisnotarelevantindicatorby ourproposal,D(andI)canberemovedfromthesuite.
Situationscanalsoarisewhererelevantstateindicatorsare eco-logicallycoupledsothatthemeanrelaxationtimeofoneindicator dependsonthevaluesofotherindicators,butthecouplingisnot strongenoughto justifydisregarding anyof them bythe logic above.Wesuggesttwowaysofdealingwiththissituation:(1)to setthetargetrangesofsuchindicatorsdependingonthecurrent valuesofotherindicators,or(2)tofindtargetrangesforall indi-catorssuchthat,aslongasallarewithintargetranges,eachwill relaxtoitsnaturalrangewithinR,nomatterwhatthevaluesofthe others.Bothoptionsreducetoouroriginalproposalifindicators areuncoupled.Option2mightleadtonarrowertargetranges,but ismoreeasilyadministered.
2.6. Precautionarybuffers
Aprecautionaryapproachtomanagementcanbeimplemented followinglogicverysimilartothatappliedintraditionalfisheries management(ICES,1998):afterdeterminingthetargetrangefor
2 A5%probabilityoffailingtomeetthetargetforthestateindicatormustbe
admittedforconsistencywiththe5%probabilitythatstateindicatorsfalloutside thenaturalrangeeveninabsenceofpressures.Toseethis,considerstateindicators withrecoveryratesmuchslowerthan1/R,forwhichthetargetrangebecomes essentiallyidenticaltothenaturalrange(Appendix,Fig.2).
Fig.2. Dependenceoftargetrangeforstronglysustainableuse(hatched&greyarea) onindicatorrelaxationtimeTforthelinearmodelEq.(1).Thenaturalrange(grey area)isshownforcomparison.Calculationassumesacoefficientofvariationforthe naturaldistributionof0.1.
anindicator,itisnarroweddowntotakemeasurementerrorsin determiningitsvalueandmodeluncertaintiesinthe determina-tionofthetargetrangeintoaccount.Modeluncertaintiescanaffect determinationofbothnaturalrangeandmeanrecoverytime.
Whenquantitativeestimatesofmeasurementandmodel uncer-taintyareavailable,theprecautionarytargetrangecouldbechosen sothat(1)meanrecoverytimeremains≤Ralsowhentakingboth kindsofuncertaintyintoaccountand(2)thecorrectindicatorvalue willbewithinthecorrecttargetrangein,say,atleast95%ofcases. Dependingonthecircumstances,oneortheotherconditionwill bestronger.3Managementaimingtorespecttargetrangesof
sev-eralindicators,whiletakinguncertaintyinsystemdynamicsinto account,couldmakeuseoftheviabilitykernelmethod(Curyetal., 2005;Mullonetal.,2004),whichworksindependentofthecriteria bywhichtargetrangesaredefined.
Forpressureindicators,notonlyuncertaintyintargetrangesof subsequentlyaffectedstateindicatorsneedstobeconsidered,but alsointhepressure-staterelationshipsandtheactualmagnitude ofpressures.
Itisaneconomicdecisiontobalancethecostsofmonitoringand researchtoimproveknowledgeofpressure-staterelationswiththe opportunitycostsofwiderprecautionarybufferswhen uncertain-tiesarehigh.
2.7. Isourscienceready?
Theimportanceofrecoverytimesforthemanagementofmarine resourceshaslongbeenrecognisedintheliterature(Borjaetal., 2010a;Duarteetal.,2013;Verdonschotetal.,2012).The quanti-tativeapplicationofthisconceptforindicatorselectionproposed hereisjustthelogicalextensionofthislineofthought,andcan buildonrichpreviousresearchdeterminingrecoverytimesand modellingrecoveryprocesses.
Thedemandsofourproposalontheaccuracyatwhichrecovery timescanbedeterminedmightbecomparativelylow.Asshownin Appendix(Fig.2),rathercoarseestimateswilloftenbesufficient, eitherbecauserecoveryis fastcompared toR andsothetarget rangetoowidetoberelevant,orbecauserecoveryissoslowthat littlevariationbeyondthenaturalrangeistolerated.
3Thefirstconditionislikelytobestrongerforstatechangesinvolvingextinctions
orecosystembi-stability,thesecondconditioninsituationswheremeanrecovery timeisasmoothfunctionoftheindicatorvalue.
A.G.Rossbergetal./EcologicalIndicators72(2017)215–224 219
3. Examples
Nextweapplyourcriteriatoseveraltypesofcandidate indica-torstoexplorefeasibilityandlikelypracticalimplicationsofour proposal.Ineach caseweestimatethemagnitudeofrelaxation timesand/ortheapproximatewidthsoftargetranges,and,based onthis,identifythecandidatesrelevantforstronglysustainable use.Whilethefocusisonindicatorslikelytopassthistest,not allcandidatesweconsiderdo.Overall,wefindthatsufficient eco-logicalunderstandingisavailabletocarryouttheproposal,and thatcomputationofreliabletargetrangeswouldbepossiblewith moderateextraeffort.
3.1. TheLargeFishIndicator
TheLargeFishIndicator(LFI)isdefinedastheproportionby biomassoffishcaughtin agiven surveythatare longerthan a definedlengththreshold.FortheNorthSeademersalfish commu-nity,sampledbytheInternationalBottomTrawlSurveyinquarter 1,theagreedlengththresholdis40cm(Greenstreetetal.,2011).A targetrangeLFI≥0.3haspreviouslybeensetonthebasisof pre-1980dataandtheviewthattheearly1980swere“thelastperiod whenscienceexpertsconsideredfishingtobegenerally[weakly] sustainableintheNorthSea”(Greenstreetetal.,2011).Because recoveryoffishcommunitysizestructurehasbeenshowntobe slow(Fungetal.,2013;Rossberg,2012;Shephardetal.,2013,2012), itisdesirabletoidentifyatargetrangeconsistentwithstrong sus-tainability.
ThenaturalrangeofvariabilityoftheLFIisnotknown,but sim-ulationstudies(Fungetal.,2013;ICES,2011)predictthatindicator valuesof0.5ormorecouldbereachedifpressureswherelower. Withoutanyfishing,simulationsbyFungetal.(2013,Fig.S5a) pre-dictindicatorvaluescloseto0.8.Assumingacoefficientofvariation forLFIof0.05initsnaturaldistribution,sothatthe2.5%quantile correspondstoabout90%ofthemeanundisturbedvalue, simula-tionsbyFungetal.(2013,Fig.7)predictthatrecoveryfromLFI≈0.5 wouldtakearound30yearsandrecoveryfromLFI≈0.25around 35–40years.ThissuggeststhatLFI≥0.3isareasonabletargetrange ifRisontheorderof30years.
Besides being a state indicator for a vulnerable ecosystem component (fish community size structure), the LFI also sig-nals pressures on marine biodiversity. Specifically, prolonged unselective fishing at a rate such that LFI remains near 0.25 leadstoextirpationofnearly athird ofalllargefishspecies in simulationsFungetal.(2013,Fig.6a).Theseextirpationscould rep-resentdeclinesofvulnerablecomponentsoflocalbiodiversity,even whentheydonotimpederecoveryofLFIitself.
3.2. Indicatorspecies 3.2.1. Generalconsiderations
Theuseofpopulationsizes(orthecorrelatedspatialextent) of selected “indicator species” as indicators for community or environmentalstatushasdrawnscepticismfrombothecologists (Lindenmayer andLikens,2010)andjurists(Kellyand Caldwell, 2013).Ourproposalsupportsthisscepticism:populationsizesof speciesin communitiestendtofluctuate,andexhibitlittle ten-dency,ifat all,toreverttoa preferredvalue(Kalyuzhnyetal., 2014;Korhonenetal.,2010).Onlongertimescalesthisleadsto thewell-documentedspeciesturnover(Magnusonetal.,1994).The naturalrangeofvariationofspeciespopulationsizesthusextends fromfairlylargevalues(Sec.14.6)downtoeffectivelyzero. Corre-spondingindicatorswouldnotberelevantinthesenseusedhere. Thisdoes,however,notprecludetherelevanceofcommunity-level indicatorsderivedfrompopulationsizesorpresence/absenceof memberspecies(FaithandPollock,2014).Infact,alphadiversityis
knowntobesensitivetopressuresbutinunperturbedcommunities remarkablystablethroughtime(Vellendetal.,2013),as theoreti-callyexpectedfromacontrolofalphadiversitythroughstructural stabilityconstraints(Rossberg,2013).
Populationsizeorextentofanindividualspeciescan poten-tiallybearelevantindicatorwhenthisspeciesisunderaparticular, manageablepressure,whenthespeciesisvulnerabletoglobalor regionalextinction(fromwhichrecoverywouldbeslowor impos-sible),orwhenthesetofitsactualorpossiblecompetitorsisso smallthatnaturalspeciesturnovercannotunfold.Fortop preda-tors, all three of these criteria arelikely tobe satisfied,which justifiestheuseofspecies-levelindicatorsinthiscase,asillustrated bythenextexample.
3.2.2. Abundanceofsealsasanindicator
Bountyhunting,encouragedinordertodecreasethe mortal-ityoffish,causedthecollapseoftheBalticgreyseal(Halichoerus grypus)populationfromapproximately80,000–100,000 individu-alsintheearly1900stoca.20,000individualsin1940s(Elmgren, 2001;HardingandHärkönen,1999).Ceasedhuntingdidnotresult inrecoveryofthepopulation,however.Mostprobablydueto envi-ronmentalpollutionharmingreproduction,thepopulationfurther decreasedtoapproximately2000inthelate1970s(Boedekeretal., 2002;HardingandHärkönen,1999).Asthesepressureshavebeen relieved or removed since the early 1990s, thepopulation has increasedto ca.28,000 individuals today (Harding et al.,2007; HardingandHärkönen,1999;Härkönenetal.,2013).The popula-tiongrowthratehasbeen>10%yearlybetweentheearly1990sand mid-2000s,butsloweddowntoabout6%inthe2010s(Härkönen etal.,2013).
TheBalticMarineEnvironmentProtectionCommission (HEL-COM)monitorsthesealpopulationsizeandgrowthratethrough acoreindicator(Härkönenetal.,2013).Atargethasbeensetfor thepopulationgrowthrateto≥10%yearly,butnonefor popu-lationsize. Inadditiontohunting andenvironmentalpollution, anthropogenicthreatstosealsincludedrowninginfishinggear, anddecreaseinfoodqualityandspreadofparasitesduetochanges in thefoodweb.A populationsizeof80,000–100,000 (Harding and Härkönen, 1999)can be used as an estimateof the natu-ral rangefor theBalticSeagrey seal.Assuming a constant10% yearlypopulationgrowthrate,apopulationsizeof5050 individ-ualswouldbeenoughtorebuildthepopulationtoNlow=80,000
individualsinR=30years,andr=6%yearlygrowthwouldrequire 15,800 individuals ormore. Assuming, more realistically, logis-tic growth with a carrying capacity of K=100,000 individuals, oneobtainsalowerlimitofthestronglysustainablepopulation target range of Nlim=K
1+erR
K/N low−1 −1=40,000 indi-viduals.More detailedmodelsmighttakedependencies,e.g.on foodavailability,intoaccountasexplainedinSection2.5.
Asthesealpopulationhasincreased,predationonvaluablefish anddamagescausedbysealstofishinggearareincreasinglyseen asproblems(Holmaetal.,2014;Varjopuro,2011).Ontheother hand,ithasbeenproposedthatabundantsealpopulationscould boosttourismincoastalcommunities.Findingabalancebetween competingservicesandusesofthemarineecosystemhasbeen rec-ognizedasachallengetobesolved(e.g.theECOSEALproject,http:// www.ecosealproject.eu/).Byourproposal,theultimatelytargeted sizeoftheBalticgreysealpopulationshouldnotliebelowNlimto
beconsistentwithstrongsustainability. 3.3. Secchidepth
Eutrophication is one of the major pressures at sea, where it affects several other ecosystem components: the food web, sea-floorintegrity, and biodiversity (Cloern, 2001). Increasesof
phytoplankton biomass are primarily caused by anthropogenic nutrientenrichmentinthewater.OneofthekeyaimsoftheBaltic SeaActionPlanisa“BalticSeaunaffectedbyeutrophication”,and twoindicatorsrelatedtothisaimarewaterclarity(Secchidepth) andchlorophyllaconcentration,whichareusedasproxiesfor phy-toplanktonabundance.
Secchidepthmeasurementsfrom1900to1920inthenorthern BalticSearangebetween5and15m,withmeanvaluesaround9m (Fleming-LehtinenandLaamanen,2012).Thiscanbeconsidered thenaturalrange,asanthropogenicnutrientloadingwaslowat thattime.Secchidepthinthesebasinshassincedecreased, reach-ing2–9mduringthelastdecade(Fleming-LehtinenandLaamanen, 2012).Thischangeisconcurrentwithincreasesinnutrientloading andnutrientconcentrationsinthewater.HELCOMtargetsfor Sec-chidepthin thevariousbasinsof theBalticSearangebetween 5.5–8.5m (Fleming-Lehtinen et al., 2014). Thesetargets areset basedontheprincipleofallowing25%deviationfromthe undis-turbedstate.
Whileanthropogenicnutrientenrichmentisthemajordriver for nutrient concentrations in the water, eutrophication abate-mentiscomplicatedbyinternalloading,a processthatrecycles sedimentednutrientsbacktothewatercolumn(Pitkänenetal., 2001).Internalloadingformsaviciouscycle(Vahteraetal.,2007), asitincreasesinnon-oxygenatedsediments,whichagainincrease duetoincreasedsedimentationofphytoplanktonbiomass.Internal loadingcandelaythedeclineofnutrientinthewateraftera reduc-tioninanthropogenicinput.Asimilardelaymustbeexpectedfor Secchidepth.
Modelssuggestthatresponsetimesofnutrientconcentrations areoftheorderof40years(Ahlviketal.,2014;Kiirikkietal.,2006; NeumannandSchernewski,2008).Linkingthesemodelsto empiri-calmodelsforSecchidepth(SavchukandWulff,2007),quantitative targetrangesforSecchidepthconsistentwithstrong sustainabil-itycouldbederivedtoinformtheongoingdebateontargetsetting (Ahtiainenetal.,2014).
3.4. Geneticdiversity
Operationalindicatorstoquantifygeneticdiversitywithin pop-ulationshavebeendefinedsincethe1990s(Chenuil,2006;Petit etal.,1998).Lossofgeneticdiversityisofconcernbecauseofits detrimentalimpacts onpopulationresilience (Frankham, 2005; Frankhametal.,2002;Schwartzetal.,2007).Geneticdiversitywill declinesharplyduringperiodsofsmallpopulationsize,and lab-oratoryandfieldstudieshavedocumentednegativeresponsesto variousenvironmentalandanthropogenicpressures(Ozerovetal., 2013;Piniet al.,2011; Tarisetal.,2006).Naturalvariability in populationsandtheenvironmentscanbeexpectedtodetermine naturalvariabilityingeneticdiversity.Recoverydynamicsoflocal geneticdiversityisunderstoodtoresultfromtwoprocesses, muta-tionandimmigration,whichexhibitcontrastingdynamics.Therate ofaccumulationofmutationsisproportionaltotheproductofthe mutationrateperlocuspergeneration,theeffectivepopulation size(Wright,1938),andinversegenerationtime(Baeretal.,2007; Kimura,1984).Forhigherorganisms,e.g.vertebrates, correspond-ingtimescaleseasilyexceed30years.Withregularimmigration fromneighboringordistantpopulations, recoverycanbemuch faster.Hence,geneticdiversitycanbearelevantindicatorforsmall populationsoflong-livingspecies,inparticularwhentheseare rel-ativelyisolatedorexperiencesimilarpressuresoverbroadspatial scales.
3.5. Non-indigenousspeciesindicators
Finally,weconsideranimportantexampleforwhich applica-tionoftheproposedframeworkisnotobvious:choicesandtarget
rangesforpressureandstateindicatorsrelatedtonon-indigenous species(NIS).Invasion of NISisoftenirreversibleand sodirect recoveryimpossible (Thresherand Kuris,2004).Yet, compared withnaturalspeciesturnover,thefactalonethatNISinvadelocal communitiesandtherecompetewithnativeresidentsmight,at regionallevel,notbeanissue(lossofglobalbiodiversitythrough homogenizationofcommunitiesnotwithstanding).However, inva-sionsbyNISdifferfromspeciesturnoverbynaturaldispersionin beingmorelikelytogothroughaphaseofrapidpopulation expan-sionwithstrongimpactsontheecosystem.Attheclimaxofthis expansionphasetheaffectedecosystemcanbedrivenoutofits naturalrangeofvariation,butthesedisruptionsdifferfromcaseto case.Fortunately,thereismountingevidencethattheexpansion phaseisgenerallyfollowedbyanadjustmentphaseatwhichthe invader’spopulationanditsimpactontheecosystemdeclinetoless disruptive,incasesevenbeneficial,levels(Blackburnetal.,2011; Reiseetal.,2006;Zaikoetal.,2014).
OurproposalcanbeadaptedtothecaseofNISifoneassumes thisboomandbustscenariotobetherule(Williamson,1997),while disregardingcaseswherethelong-termimpactsremainhigh com-paredtothoseofnaturalturnover.Onecantheninterprettherate ofNISarrivalsinanecosystemasthepressure,andtheaggregated disruptiveimpactsNIScausebeforereachingtheirlateadjustment phasesastheresultingchangeinstate.Theimpactscanbe consid-eredstronglysustainableifthesedisruptionswould,withoutnew arrivalsofNIS,onaveragedeclinewithintimeRtolevelstypicalfor naturalturnover.
Thequantificationofthelevelofdisruptionsiscomplicatedby theidiosyncrasyofNISimpacts.Amongframeworkssuggestedfor quantifyingbioinvasionimpacts(Coppetal.,2009;Molnaretal., 2008;Nentwigetal., 2010), theBiologicalPollutionLevel(BPL) assessmentmethod(Oleninetal.,2007)hasbeenrecommended asarobustandstandardizedindicatorinthecontextoftheMSFD (Oleninetal.,2010).Ithasbeentestedinassessmentsoftheimpacts ofsingleandmultipleNISatvariousscales(Oleninaetal.,2010; Zaikoetal.,2011).However,nounambiguoustargetrangehasbeen proposedforit,yet.
Recoverytimesinthisinterpretationdependonthepatternof boomandbustcycles,whichmayvarydependingonintrinsicor extrinsicfactors(StrayerandMalcom,2006).Forzebramusselsin Irishfreshwaterecosystems,forexample,Zaikoetal.(2014) doc-umentrecoverytimesontheorderofonlytenyearssincearrival andfiveyearsaftermaximumimpact.
4. Generalimplications
4.1. Targetrangescandifferfromnaturalranges
Targetrangesforindicatorsarefrequentlychosenasthe indi-catorvaluesthoughttorepresentecosystemsunperturbedoronly slightlyperturbedbyhumaninterference,see,e.g.theEuropean WaterFrameworkDirective(EC,2000).Thepresentproposal sup-portsthis approach for ecosystem components withrelaxation ratesmuch slowerthan1/R (seeAppendix, Fig.2).For compo-nentsthatrelaxfaster,theproposalleadstobroaderindicatortarget rangesand,crucially,supportsthisbyasimplerationale.
4.2. Importanceofpressureindicators
Ifanindicatorhasalongrelaxationtime,itscurrentvaluecanbe interpretedasrepresentingthecumulativeeffectofpressuresover acorrespondingtimespan(seeAppendix,Observation1).The indi-catorvaluecanchangerapidlyonlywhen,temporarily,pressures farexceedthelevelcorrespondingtostronglysustainableuse.If pressuresbecomeweakerorareentirelyremoved,theimpactof
A.G.Rossbergetal./EcologicalIndicators72(2017)215–224 221
previouscumulativepressuresinitiallyremainsand onlyslowly fadesawayastheindicatorrecovers.Theanalogyto“mining”has beeninvoked(Herricketal.,2006).Effectivenessofmanagement onshortertimescalesmustthereforebeassessednotonlydirectly throughstateindicators,butalsoindirectlybasedoncorresponding pressureindicators.Hence,pressureindicatorshaveaparticularly importantroletoplayinthepresentinterpretationofsustainable use.
Apatternoneexpectstoseefrequentlyintimeseriesofstate indicatorsforvulnerableecosystemcomponentsisarapiddecline inphasesofunmanagedoveruse,followedbyslowrecoverytoa baselineaftermanagementbecameeffective(Duarteetal.,2013). Recoveryatthesamerateascollapsecannotgenerallybeexpected. Symmetricpatternsofdeclineandrecoveryaremore characteris-ticofrapidlyrecoveringindicatorsornaturalfluctuationsunder managedsustainableuse.
4.3. Signal-to-noiseratio,monitoringintensityandcosts
Relevantstateindicatorshavenarrowrangesofnatural variabil-ity,andyetareresponsivetolastingpressures.Inthelanguageof engineering,theirsignal-to-noiseratioishigh.
Duetotheinherentlyslowdynamicsofrelevantindicatorsand theirhighsignal-to-noiseratio,monitoringintensitydoesnotneed tobehigh,unlessthereareconcernsthatpresentanthropogenic pressuresleadtorapidchangesinindicatorvalues.Relevantstate indicatorsthereforetendhavecomparativelylowmonitoringcosts. 4.4. Exceptionalityofrelevantindicators
Mathematicalconsiderationssuggestthat,potentialstate indi-catorswithlongrelaxationtimestendhavebroadnaturalranges ofvariability (Appendix,Observation 5),becausetheyintegrate theimpactsofnaturalfluctuationsoverlongtime.Co-occurrence ofslowdynamicsandsmallvariability,asrequiredforindicator “relevance”,impliesthatunderlyingecosystempropertiesremain mostlyunaffectedbytheinherentvariabilityofotherproperties. Oftenthiswillbethecasebecauseindicatordynamicsisgoverned bygeneralecologicalorphysicalprinciples(e.g.conservationlaws) thatinhibitstrongfluctuations.Indicatorsofhighrelevancebythe presentproposalthereforecanbeexpectedtoberatheruncommon amongconceivablestateindicatorsatlarge.
Whenindicatordynamicsaregovernedbygeneralecological orphysicalprinciplesitisoftenpossibletoapproximatedynamics andresponsestopressuresbysimplemanagementmodels.These managementmodelscaninformchoicesofpressureindicatorsand theirtargetranges,aswellasmanagementpracticestoensure sus-tainableuse.Wethereforeexpectthatrelevantindicatorsbythe presentproposalareamongthoseforwhicheffectivemanagement schemescanrathereasilybedeveloped.
4.5. Theimportanceofspecificusetargets
Itisdesirablethat,withintheboundariesofstrong sustainabil-ity,themarineecosystemprovideshighlevelsofservicestosociety. Theseshouldbeusedsustainablyintheweaksense.Theparticular mixofservices,however,dependsonsocietalpreferences.Some societiesmighthavestrongpreferencesforrecreationaluses; oth-ersmightvaluedecompositionofpollutantshigher.Management targetsforweaklysustainableusesandcorrespondingindicators arethereforeunlikelytoderivefromsimplegeneralcriteria.The problemismuchmorecomplex(Elliott,2011),yetaddressingitis paramount,becausethemanagementobjectiveofstrong sustain-abilityonitsownisinsufficientforachievingthesocietalbenefits itismeanttoenable.
Returningtotheanalogybetweentheprecautionaryapproach tofisheriesmanagementandourproposalhere,thehistoric les-sonmustberecalledthattheboundariesofstrong-sustainability targetrangesmighteffectivelybecomemanagementtargetsinthe policyprocess,withdetrimentaleffectsforecosystemfunctioning andservices.ItwasnotlongafterICES(1998)establishedtheir for-mulationoftheprecautionaryapproachthatofficialICESadvice warnedofthisissue(ICES,2002),increasinglysosince2004:
Riskaversion,basedonthe precautionaryapproach,definesthe boundaries of management decisions for sustainable fisheries. Withintheseboundariessocietymaydefineobjectivesrelatingto benefits suchas maximised long-termyield, economicbenefits, or other ecosystem services. The achievement of such objec-tives maybeevaluatedagainstanotherset ofreference points, targetreferencepoints,whichmaybemeasuredinsimilar dimen-sionsaslimitreferencepointsbutwhichmayalsorelatetomoney, food,employment,orotherdimensionsofsocietalobjectives.[...] settingtargetsforfisheriesmanagementinvolvessocio-economic considerations.Therefore,ICESdoesnotproposevaluesforTarget ReferencePoints[...].Thismeansthat[...]exploitationofmost stocksislikelytobesub-optimal,i.e.thelong-termyieldislower thanitcouldbe.
[...]Managersareinvitedtodeveloptargetsandassociated man-agementstrategies
ICES(2004),originalemphasis OnlyrecentlyMSYasauseobjectivewasincorporatedintothe CommonFisheriesPolicy(EU,2013).
4.6. Alignmentwithprevailingapproaches
Comparisonoftheapproachlaidoutherewithcommonly pro-posedqualitativecriteria forchoosingindicators(Queirósetal., 2016)showsthem tobeeitheralignedwiththesecriteriaorto beunrelatedtothem.Anexampleforgoodalignmentisthe crite-rionofcost-efficiency,which,asexplainedabove,isexpectedtobe naturallysatisfiedbymanyindicatorsforthestateofvulnerable ecosystemcomponents.Examples forcriteria thatappear unre-latedtothecurrentproposalaretheconcretenessandtheeasy interpretabilityofthemetricsused(Elliott,2011).Theunrelated criteriacanbetakenintoaccountalongsidethoseproposedhere.
Theonlycriterionforindicatorselectionthatisfrequently men-tionedintheliteraturebutperhapsincompatiblewiththepresent proposalistheresponsivenessofindicatorstomanagement mea-sures.Weproposedtoaddressthisusingpressureindicatorsand othersupportingindicators.
Our proposaldevelopsearlier suggestionsforan operational definitionofGESpresentedbyBorjaetal.(2013)byseparatingthe characterizationsofweaklyandstronglysustainableuse.Another distinctionofourproposalfromthecurrentgeneralunderstanding istherecognitionthatnotallcharacteristicsofecosystemsare nat-urallyresilient(i.e.recoverrapidlyandpredictablyfrompressures). Managementshould payattentionto potentialfurther deterio-rationofresilience,butofprimaryconcernshouldbeecosystem componentsforwhichresilienceisnaturallylow.
5. Conclusionsandpolicyimplications
Weproposedasystematic,quantitativeapproachtoselect indi-catorsandtheirtargetrangesforthepurposeofassessingstrong sustainabilityof ecosystemuse. Theapproach offersarationale forimprovingconsistencyamongtargetsandfocusinginvestments intoindicatorresearchandmonitoring.Toclose,wehighlightthree overarchingimplicationsoftheproposalthatarelikelytostandout infuturedevelopmentsofMSFDandsimilarpolicyinstruments.
Firstly,proposalsfortargetsofMSFDindicatorsoftenstillaim atrestoringnaturalornear-naturalecosystemstates.Thisisnot alwaysnecessarywhenthepolicygoalissustainableuse.Herewe providedanargumentforthechoiceofalternative,broadertarget ranges.
Secondly,relevantstateindicators,byourproposal,willalmost alwaysbepairedwithcorrespondingpressureindicatorsorsetsof pressureindicators.Situationswhereeitherastateorapressure indicatoraresufficienttocharacterisethestatusofanecosystem componentarethosewheretherelevantrecoverytimesare com-parativelysmall (Appendix,Observation3),implying thatthese ecosystemcomponentsarelikelytoberesilienttopressuresand thereforenotofprimaryconservationconcern(Appendix, Obser-vation4).
Thirdly,thesettingofindicatortargetrangesforstrongly sus-tainable use and of target ranges or values corresponding to particularuseobjectivesshouldbeclearlydistinguishedinthe pol-icyprocess.Authorityforsettingthesetypesoftargetsmighteven beassignedtodifferentbodies.Anexamplewheresucha separa-tionisdefactoinplaceisEUfisheriesmanagement.TheCommon FisheriesPolicy (EU,2013)now regulatesthe settingof fishing quotasinaccordancewithMSYobjectives,whilerespecting envi-ronmentalconstraintsaredefined,among others,by theMSFD. TheMSFD,inturn,leavesroomforpragmaticfisheries manage-ment.Thetwo policyinstrumentsareadministeredbydifferent departmentsoftheEuropeanCommission.
6. Appendix:mathematicalanalyses
InthisAppendixaminimalmathematicmodelisintroducedthat describesrelaxationofstateindicatorstosomenaturalrangeand responsivenessofstateindicatorstopressuresandenvironmental fluctuations.Themodelisthenanalyzedmathematicallyinorder todevelopanunderstandingofthegeneralrelationshipsbetween stateindicatordynamics,theirresponsivenesstopressures,andthe implicationsforindicatortargetranges.
Inthemodel, theindicatorvaluechanges becauseof(i) nat-uralrecoverytoa valuecorrespondingtoanundisturbedstate, (ii)externalpressuresand(iii)uncontrollednaturalfluctuations. Specifically,itassumesadependenceofthevalueofanindicator I (t) ontimettofollow
dI (t) dt =−
[I (t) −I0]
T −cP (t) +noise. (1)
Thismodelisa directtranslationofourgeneral understand-ingofindicatordynamics:Theindicatorvaluechanges(“dI (t) /dt”) because of (“=”) natural recovery (“− [I (t) −I0]/T”) toa value
correspondingtoanundisturbedstate(“I0”),becauseofexternal
pressures(“P (t)”)andbecauseofuncontrollednaturalfluctuations (“noise”).Itislegitimatetothinkofthethreetermsontheright handsidetobemechanicallyindependentcontributionswith mag-nitudescontrolledbyindependentmechanisms,sothatthevalues oftheconstantsTandcandthestrengthofthenoiseare indepen-dentparameters.Equationsofthetypeabovearemathematically wellstudied.Anexcellentexpositionoftherelevantmathematics ineasilyaccessibleformcanfoundinthebookbyGardiner(1990). Theconstant cdenotesthesensitivityoftheindicatortothe pressureP (t).Thevalueofthisconstantcaninprinciplebe deter-mined by monitoring the rate at which the indicator changes (dI (t) /dt) when suddenly a largeconstant pressure P (t) =P is applied.Thevalueofcthenfollowsasc≈−
dI (t) /dt/P.Itcan bepositiveornegative.Forsimplicity,cishereassumedpositive, sothattheindicatordeclineswhenapressureisapplied.TheparameterT denotestherelaxationtimeconstantofthe indicator.Whennoiseisnegligible,Tisthetimeittakesthe
indica-torI (t) toreducethedistancetotheequilibriumI0fromitscurrent
valueto40%(=exp (−1))ofthisvalueinabsenceofpressures. ThesolutionofEq.(1)is
I (t) =I0+ t
−∞exp
− (t−) /T[−cP () +noise ()] d. (2)ThedeviationofI (t) fromI0isproportionaltoaweightedsum
overpreviouspressuresandpreviousnoise,withweightsdecaying exponentiallyasexp
− (t−) /T,wheredenotespointsintime inthepast(i.e.beforet).Thisweightfactorisoftheorderof mag-nitudeof1overanapproximatetimespanT,andthendecaysto smallervalues.Whenthe“noise”isnegligibleandaconstantpressureP (t) =P isappliedoveratimethatislongcomparedtoT,theindicatorwill eventuallyrelaxtoaconstantvalue
I (t) =Ieq=I0−TcP (t) . (3)
Whenpressurechangesthoughtimebutthesechangesareslow comparedtoT,thisformulaisstillagoodapproximation.
Observation1Eq.(3)impliesthat,ingeneral,largerelaxation timesT imply ahighsensitivity oftheequilibrium valueIeq to
pressures.
Observation2ForpressuresthatchangeslowlycomparedtoT, thereisadirectfunctionalrelationship(herelinear)betweenthe pressurePandthestateindicatorI (t).
Mostkindsofpressuresarenotexpectedtoremainconstant orapproximatelyconstantoverthetimeR.Withthisinmind,we arriveat
Observation3DirectfunctionalrelationsbetweenpressureP (t) andstateindicatorsI (t) holdonlyforstateindicatorswith relax-ationtimesTconsiderablyshorterthanR.
The“noise”terminEq.(1)describesenvironmentaleffectsthat drivenaturalfluctuationsintheindicatorvalue.4Inthepresenceof
noisetheindicatordoesnotreachtheequilibriumvalueIeqgiven
byEq.(3)whenthepressureisconstantorabsent,butfluctuates aroundthisvalue.Thewidthoftherangeoffluctuation(whichis, forthepresentmodel,independentofpressureP)increasesnot onlywithincreasingstrengthofthe”noise”,but,complicating,also withincreasingautocorrelationinthesefluctuations:theslower thesefluctuations,thestrongertheirimpactonI (t).5Yet,asa
gen-eralrule,itfollows,byEq.(2),fromtheadditivityoftheeffectsof noiseonI (t) overarecenttimeintervalofapproximatedurationT, andtherandomnessof thenoise (bydefinition),thatthemean squareddeviationofI (t) fromIeq resultingfromnoiseincreases
asT.Thissupportsthefollowing
Observation4Allelseequal,indicatorswithlargerrelaxation timesTtendtohavewidernaturalrangesofvariation.
Fortypicalforms ofthenoise, thedistributionofI (t) inthe absenceofpressuresfollowsanormaldistributionwithmeanI0.
Ifisthestandarddeviationofthisdistribution,thenaturalrange accordingtothedefinitionaboveisgivenbyIlow=I0−1.96and
Ihigh=I0+1.96.
Theproblemofcomputingthemeantimetorecoveryis math-ematicallyaproblemofcomputingthemeanfirstpassagetimeof aunivariaterandomprocess.Inthespecialcasethat“noise”inthe
4The“noise”termisassumedtohavealong-termmeanofzero.Ifnot,thiscan
beenforcedbyadjustingthevalueofI0.
A.G.Rossbergetal./EcologicalIndicators72(2017)215–224 223
modelaboveiswhitenoise,themeanfirstpassagetimeforreaching IlowfromastartingvalueI1<IlowforP=0is(Gardiner,1990)
2T −1.96 (I1−I0)/ exp y2 21−erf
y √ 2
dy, (4)
witherf (x) denotingthe so-called errorfunction. Thelower boundoftheindicatortargetrangeisthevalueofI1forwhichthe
expressionaboveequalsR.Fig.2 illustratetheresulting depen-denceofI1onT.
AscanbeseeninFig.2,thetargetrangequicklybecomesvery widewhenTislessthenabouthalfaslargeasR,anddiffersonly littlefromthe naturalrange for T>10R.Theactual valueofT thereforetypicallymattersonlywhenitiswithinabout0.5Rto 10R.
ForrelaxationtimesT muchsmallerthanthemaximalmean recoverytimeR,correspondingto (I1−I0) muchlargerthan,noise
canbedisregardedandthepressure-freerelaxationofI (t) approxi-matedbyasimple,exponentialrelaxation.ForthecasethatI (t) =I1
at t=0, one gets I (t) −I0= (I1−I0) exp
−t/T
. Interpreting I1asthelowerboundofthetargetrange,theconditionI (R) =Ilow
thenleadstoI1=I0−1.96exp
R/T
.Correspondingly,the con-ditionforsustainableusebecomesI0−1.96exp R/T<I (t) < I0+1.96exp R/T. AcknowledgementsTheauthorsthankSimonGreenstreet,CristinaHerbon,Simon Jennings,TizianaLuisetti,LucillePaltriguera,andChristianWilson forcommentsonpreviousversionsofthispaper.Thisworkhas resultedfromtheDEVOTES(DEVelopmentOfinnovativeToolsfor understandingmarinebiodiversity andassessingGood Environ-mentalStatus)projectfundedbytheEUunderthe7thFramework Programme,‘TheOceanofTomorrow’Theme(No.308392),www. devotes-project.eu.Further,A.G.R.waspartiallyfundedbythe Nat-uralEnvironmentResearch Counciland theUKDepartment for Food,Environment and Rural Affairs (Defra)within theMarine Ecosystems ResearchProgram (MERP), C.P.L. byDefra (M1228), A.Z.by BIO C3withinthejointBalticSeaResearch and Devel-opmentProgramme(EU7th andResearch Councilof Lithuania, BONUS-1/2014),andM.C.U.bytheSpanishProgrammefortalent andemployabilityinI+D+i‘TorresQuevedo’.
References
Ahlvik,L.,Ekholm,P.,Hyytiäinen,K.,Pitkänen,H.,2014.Aneconomic-ecological modeltoevaluateimpactsofnutrientabatementintheBalticSea.Environ. Model.Softw.55,164–175,http://dx.doi.org/10.1016/j.envsoft.2014.01.027. Ahtiainen,H.,Artell,J.,Elmgren,R.,Hasselström,L.,Håkansson,C.,2014.BalticSea
nutrientreductions−Whatshouldweaimfor?J.Environ.Manage.145,9–23,
http://dx.doi.org/10.1016/j.jenvman.2014.05.016.
Baer,C.F.,Miyamoto,M.M.,Denver,D.R.,2007.Mutationratevariationin multicellulareukaryotes:causesandconsequences.Nat.Rev.Genet.8, 619–631,http://dx.doi.org/10.1038/nrg2158.
Blackburn,T.M.,Pyˇsek,P.,Bacher,S.,Carlton,J.T.,Duncan,R.P.,Jaroˇsík,V.,Wilson, J.R.U.,Richardson,D.M.,2011.Aproposedunifiedframeworkforbiological invasions.TrendsEcol.Evol.26,333–339,http://dx.doi.org/10.1016/j.tree. 2011.03.023.
Boedeker,D.,Benke,H.,AndersenNorden,O.,Strempel,R.,2002.MarineMammals (EnvironmentoftheBalticSeaArea1994–98).BSEP82b,pp.171–173.
Borja,Á.,Dauer,D.M.,Elliott,M.,Simenstad,C.A.,2010a.Medium-andlong-term recoveryofestuarineandcoastalecosystems:patterns,ratesandrestoration effectiveness.EstuariesCoasts33,1249–1260,http://dx.doi.org/10.1007/ s12237-010-9347-5.
Borja,Á.,Elliott,M.,Carstensen,J.,Heiskanen,A.-S.,vandeBund,W.,2010b. Marinemanagement−TowardsanintegratedimplementationoftheEuropean MarineStrategyFrameworkandtheWaterFrameworkDirectives.Mar.Pollut. Bull.60,2175–2186,http://dx.doi.org/10.1016/j.marpolbul.2010.09.026. Borja,Á.,Dauer,D.M.,Grémare,A.,2012.Theimportanceofsettingtargetsand
referenceconditionsinassessingmarineecosystemquality.Ecol.Indic.12, 1–7,http://dx.doi.org/10.1016/j.ecolind.2011.06.018.
Borja,A.,Elliott,M.,Andersen,J.H.,Cardoso,A.C.,Carstensen,J.,Ferreira,J.G., Heiskanen,A.-S.,Marques,J.C.,Neto,J.M.,Teixeira,H.,Uusitalo,L.,Uyarra,M.C., Zampoukas,N.,2013.Goodenvironmentalstatusofmarineecosystems:what isitandhowdoweknowwhenwehaveattainedit?Mar.Pollut.Bull.76, 16–27,http://dx.doi.org/10.1016/j.marpolbul.2013.08.042.
Chenuil,A.,2006.Choosingtherightmoleculargeneticmarkersforstudying biodiversity:frommolecularevolutiontopracticalaspects.Genetica127, 101–120,http://dx.doi.org/10.1007/s10709-005-2485-1.
Cloern,J.E.,2001.Ourevolvingconceptualmodelofthecoastaleutrophication problem.Mar.Ecol.Prog.Ser.210,223–253.
Cochrane,S.K.J.,Connor,D.W.,Nilsson,P.,Mitchell,I.,Reker,J.,Franco,J.,Valavanis, V.,Moncheva,S.,Ekebom,J.,Nygaard,K.,SerrãoSantos,R.,Naberhaus,I., Packeiser,T.,vandeBund,W.,Cardoso,A.C.,2010.MarineStrategyFramework DirectiveTaskGroup1Report:BiologicalDiversity.JRC,Ispra,Italy.
Copp,G.H.,Vilizzi,L.,Mumford,J.,Fenwick,G.V.,Godard,M.J.,Gozlan,R.E.,2009. CalibrationofFISK,aninvasivenessscreeningtoolfornonnativefreshwater fishes.RiskAnal.29,457–467,http://dx.doi.org/10.1111/j.1539-6924.2008. 01159.x.
Cury,P.M.,Mullon,C.,Garcia,S.M.,Shannon,L.J.,2005.Viabilitytheoryforan ecosystemapproachtofisheries.ICESJ.Mar.Sci.J.Cons.62,577–584,http:// dx.doi.org/10.1016/j.icesjms.2004.10.007.
Dietz,S.,Neumayer,E.,2007.WeakandstrongsustainabilityintheSEEA:Concepts andmeasurement.Ecol.Econ.61,617–626,http://dx.doi.org/10.1016/j. ecolecon.2006.09.007(SpecialIssueonEnvironmentalAccounting:
IntroducingtheSystemofIntegratedEnvironmentalandEconomicAccounting 2003SEEA-2003S.I.).
Duarte,C.M.,Borja,A.,Carstensen,J.,Elliott,M.,Krause-Jensen,D.,Marbà,N.,2013. Paradigmsintherecoveryofestuarineandcoastalecosystems.Estuaries Coasts,1–11,http://dx.doi.org/10.1007/s12237-013-9750-9.
EC,2000.Directive2000/60/ECoftheEuropeanParliamentandoftheCouncilof 23October2000establishingaframeworkforCommunityactioninthefieldof waterpolicy.Off.J.Eur.CommunitiesL327,1–72.
EC,2008.Directive2008/56/ECoftheEuropeanParliamentandoftheCouncilof 17June2008establishingaframeworkforcommunityactioninthefieldof marineenvironmentpolicy(MarineStategyFrameworkDirective).Off.J.Eur. UnionL164,19–40.
EC,2010.CommissionDecisionof1September2010oncriteriaand
methodologicalstandardsongoodenvironmentalstatusofmarinewaters.Off. J.Eur.UnionL232,14–24.
EU,2013.Regulation(EU)No1380/2013oftheEuropeanParliamentandofthe CouncilontheCommonFisheriesPolicy.Off.J.Eur.UnionL354,22–61.
Elliott,M.,2011.Marinescienceandmanagementmeanstacklingexogenic unmanagedpressuresandendogenicmanagedpressures−Anumberedguide. Mar.Pollut.Bull.62,651–655,http://dx.doi.org/10.1016/j.marpolbul.2010.11. 033.
Elliott,M.,2013.The10-tenetsforintegrated,successfulandsustainablemarine management.Mar.Pollut.Bull.74,1–5,http://dx.doi.org/10.1016/j.marpolbul. 2013.08.001.
Elmgren,R.,2001.UnderstandinghumanimpactontheBalticecosystem: changingviewsinrecentdecades.AMBIOJ.Hum.Environ.30,222–231,http:// dx.doi.org/10.1579/0044-7447-30.4.222.
FAO,2009.InternationalGuidelinesfortheManagementofDeep-seaFisheriesin theHighSeas.FoodandAgricultureOrganizationoftheUnitedNations,Rome.
Faith,D.P.,Pollock,L.J.,2014.Phylogeneticdiversityandthesustainableuseof biodiversity.In:Verdade,L.M.,Lyra-Jorge,M.C.,Pi ˜na,C.I.(Eds.),Applied EcologyandHumanDimensionsinBiologicalConservation.Springer,Berlin Heidelberg,pp.35–52.
Figge,F.,2005.Capitalsubstitutabilityandweaksustainabilityrevisited:the conditionsforcapitalsubstitutioninthepresenceofrisk.Environ.Values, 185–201.
Fleming-Lehtinen,V.,Laamanen,M.,2012.Long-termchangesinSecchidepthand theroleofphytoplanktoninexplaininglightattenuationintheBalticSea. Estuar.Coast.ShelfSci.102–103,1–10,http://dx.doi.org/10.1016/j.ecss.2012. 02.015.
V.Fleming-Lehtinen,M.Pyhälä,M.Laamanen,E.Łysiak-Pastuszak,M.Carstens, J.-M.Leppänen,W.Leujak,G.Nausch,2014.Waterclarity−HELCOMCore IndicatorReportOnline.
Frankham,R.,Briscoe,D.A.,Ballou,J.D.,2002.IntroductiontoConservation Genetics.UniversityPress,Cambridge.
Frankham,R.,2005.Geneticsandextinction.Biol.Conserv.126,131–140,http:// dx.doi.org/10.1016/j.biocon.2005.05.002.
Fung,T.,Farnsworth,K.D.,Shephard,S.,Reid,D.G.,Rossberg,A.G.,2013.Whythe sizestructureofmarinecommunitiescanrequiredecadestorecoverfrom fishing.Mar.Ecol.Prog.Ser.484,155–171.
Gardiner,C.W.,1990.HandbookofStochasticMethods,2nded.Springer,Berlin.
Greenstreet,S.P.R.,Rogers,S.I.,Rice,J.C.,Piet,G.J.,Guirey,E.J.,Fraser,H.M.,Fryer, R.J.,2011.DevelopmentoftheEcoQOfortheNorthSeafishcommunity.ICESJ. Mar.Sci.J.Cons.68,1–11.
Härkönen,T.,Galatius,A.,Bräeger,S.,Karlsson,O.,Ahola,M.,2013.CoreIndicator ofBiodiversity.PopulationGrowthRateAbundanceandDistributionofMarine Mammals.HELCOM.
Harding,K.C.,Härkönen,T.J.,1999.DevelopmentintheBalticGreySeal (Halichoerusgrypus)andRingedSeal(Phocahispida)Populationsduringthe 20thCentury.Ambio28,619–627.
Harding,K.C.,Härkönen,T.,Helander,B.,Karlsson,O.,2007.StatusofBalticgrey seals:populationassessmentandextinctionrisk.NAMMCOSci.Publ.6,33,
http://dx.doi.org/10.7557/3.2720.
Herrick,S.F.,Hill,K.,Reiss,C.,2006.Anoptimalharvestpolicyfortherecently renewedUnitedStatesPacificsardinefishery.In:Hannesson,R.,Barange,M., Herrick,S.F.(Eds.),ClimateChangeandtheEconomicsoftheWorld’sFisheries: ExamplesofSmallPelagicStocks.EdwardElgarPublishing,pp.126–150.
Holma,M.,Lindroos,M.,Oinonen,S.,2014.Theeconomicsofconflictinginterests: northernBalticsalmonfisheryadaptiontograysealabundance.Nat.Resour. Model.27,275–299,http://dx.doi.org/10.1111/nrm.12034.
ICES,2013.ReportoftheWorkingGroupontheEcosystemEffectsofFishing Activities(WGECO)(ICESDocumentNoCM2013/ACOM:25).Copenhagen. ICES,2014.ReportoftheWorkshoptoreviewthe2010CommissionDecisionon
criteriaandmethodologicalstandardsongoodenvironmentalstatus(GES)of marinewaters;Descriptor6Copenhagen.
ICES,2014.ReportoftheWorkshoptoreviewthe2010CommissionDecisionon criteriaandmethodologicalstandardsongoodenvironmentalstatus(GES)of marinewaters;Descriptor4FoodwebsCopenhagen.
ICES,2011.ReportoftheWorkingGroupontheEcosystemEffectsofFishing Activities(WGECO)(ICESDocumentNo.CM2011/ACOM:24).Copenhagen. ICES,2010ReportoftheWorkingGroupontheEcosystemEffectsofFishing
Activities(WGECO)(ICESDocumentNoCM2010/ACOM:23).Copenhagen. ICES,2005.ReportoftheWorkingGroupontheEcosystemEffectsofFishing
Activities(WGECO)(No.ACE:04).Copenhagen.
ICES,2004.ReportoftheICESAdvisoryCommitteeonFisheryManagementand AdvisoryCommitteeonEcosystems(ICESAdviceVolume1,Number2) Copenhagen.
ICES,2002.TheFormofICESAdvice(ICESCooperativeResearchReportNo.255) Copenhagen.
ICES,2001.ReportoftheWorkingGroupontheEcosystemEffectsofFishing Activities(NoACME:09).Copenhagen.
ICES,1998.ReportofthePrecautionaryApproachtoFisheriesManagement(ICES DocumentNoCM1998/ACFM:10).Copenhagen.
Kalyuzhny,M.,Seri,E.,Chocron,R.,Flather,C.H.,Kadmon,R.,Shnerb,N.M.,2014. Nicheversusneutrality:adynamicalanalysis.Am.Nat.184,439–446,http:// dx.doi.org/10.1086/677930.
Kelly,R.P.,Caldwell,M.R.,2013.Notsupportedbycurrentscience:theNational ForestManagementActandthelessonsofenvironmentalmonitoringforthe futureofpublicresourcesmanagement.Stanf.Environ.LawJ.32,151–212.
Kiirikki,M.,Lehtoranta,J.,Inkala,A.,Pitkänen,H.,Hietanen,S.,Hall,P.O.J., Tengberg,A.,Koponen,J.,Sarkkula,J.,2006.Asimplesedimentprocess descriptionsuitablefor3D-ecosystemmodelling—Developmentandtesting intheGulfofFinland.J.Mar.Syst.61,55–66,http://dx.doi.org/10.1016/j. jmarsys.2006.02.008.
Kimura,M.,1984.TheNeutralTheoryofMolecularEvolution.UniversityPress, Cambridge.
Korhonen,J.J.,Soininen,J.,Hillebrand,H.,2010.Aquantitativeanalysisoftemporal turnoverinaquaticspeciesassemblagesacrossecosystems.Ecology91, 508–517.
Lindenmayer,D.B.,Likens,G.E.,2010.Directmeasurementversussurrogate indicatorspeciesforevaluatingenvironmentalchangeandbiodiversityloss. Ecosystems14,47–59,http://dx.doi.org/10.1007/s10021-010-9394-6. Magnuson,J.J.,Benson,B.J.,McLain,A.S.,1994.Insightsonspeciesrichnessand
turnoverfromlong-termecologicalresearch:fishesinnorthtemperatelakes. Am.Zool.34,437–451.
Molnar,J.L.,Gamboa,R.L.,Revenga,C.,Spalding,M.D.,2008.Assessingtheglobal threatofinvasivespeciestomarinebiodiversity.Front.Ecol.Environ.6, 485–492,http://dx.doi.org/10.1890/070064.
Mullon,C.,Cury,P.,Shannon,L.,2004.Viabilitymodeloftrophicinteractionsin marineecosystems.Nat.Resour.Mode.17,71–102,http://dx.doi.org/10.1111/ j.1939-7445.2004.tb00129.x.
Nentwig,W.,Kühnel,E.,Bacher,S.,2010.Agenericimpact-scoringsystemapplied toalienmammalsinEurope.Conserv.Biol.J.Soc.Conserv.Biol.24,302–311,
http://dx.doi.org/10.1111/j.1523-1739.2009.01289.x.
Neumann,T.,Schernewski,G.,2008.EutrophicationintheBalticSeaandshiftsin nitrogenfixationanalyzedwitha3Decosystemmodel.J.Mar.Syst.74, 592–602,http://dx.doi.org/10.1016/j.jmarsys.2008.05.003.
Olenina,I.,Wasmund,N.,Hajdu,S.,Jurgensone,I.,Gromisz,S.,Kownacka,J., Toming,K.,Vaici ¯ut ˙e,D.,Olenin,S.,2010.Assessingimpactsofinvasive phytoplankton:theBalticSeacase.Mar.Pollut.Bull.60,1691–1700,http://dx. doi.org/10.1016/j.marpolbul.2010.06.046.
Olenin,S.,Alemany,F.,Cardoso,A.C.,Gollasch,S.,Goulletquer,P.,Lehtiniemi,M., McCollin,T.,Minchin,D.,Miossec,L.,OcchipintiAmbrogi,A.,Ojaveer,H., Jensen,K.R.,Stankiewicz,M.,Wallentinus,I.,Aleksandrov,B.,2010.Marine StrategyFrameworkDirective-TaskGroup2Report.Non-indigenousSpecies. OfficeforOfficialPublicationsoftheEuropeanCommunities,Luxembourg(44 pp.).
Olenin,S.,Minchin,D.,Daunys,D.,2007.Assessmentofbiopollutioninaquatic ecosystems.Mar.Pollut.Bull.55,379–394,http://dx.doi.org/10.1016/j. marpolbul.2007.01.010.
Ozerov,M.Y.,Veselov,A.E.,Lumme,J.,Primmer,C.R.,2013.Temporalvariationof geneticcompositioninAtlanticsalmonpopulationsfromtheWesternWhite SeaBasin:influenceofanthropogenicfactors?BMCGenet.14,88,http://dx. doi.org/10.1186/1471-2156-14-88.
Petit,R.J.,Mousadik,A.,elPons,O.,1998.Identifyingpopulationsforconservation onthebasisofgeneticmarkers.Conserv.Biol.12,844–855.
Piet,G.,Albella,A.,Aro,E.,Farrugio,H.,Lleonart,J.,Lordan,C.,Mesnil,B.,Petrakis, G.,Pusch,C.,Radu,G.,Rätz,H.-J.,2010.MarineStrategyFrameworkDirective TaskGroup3Report:CommerciallyExploitedFishandShellfish(NoEUR 24316EN−2010).JRC,Ispra,Italy.
Pini,J.,Planes,S.,Rochel,E.,Lecchini,D.,Fauvelot,C.,2011.Geneticdiversityloss associatedtohighmortalityandenvironmentalstressduringtherecruitment stageofacoralreeffish.CoralReefs30,399–404,http://dx.doi.org/10.1007/ s00338-011-0718-6.
Pitkänen,H.,Lehtoranta,J.,Räike,A.,2001.Internalnutrientfluxescounteract decreasesinexternalload:thecaseoftheestuarialeasternGulfofFinland, BalticSea.AMBIOJ.Hum.Environ.30,195–201,http://dx.doi.org/10.1579/ 0044-7447-30.4.195.
Queirós,A.M.,Strong,J.A.,Mazik,K.,Carstensen,J.,Bruun,J.,Somerfield,P.J.,Bruhn, A.,Ciavatta,S.,Flo,E.,Bizsel,N.,Özaydinli,M.,Chuˇsev,R.,Muxika,I.,Nygård,H., Papadopoulou,N.,Pantazi,M.,Krause-Jensen,D.,2016.Anobjective frameworktotestthequalityofcandidateindicatorsofgoodenvironemtal status.Front.Mar.Sci.3,73,http://dx.doi.org/10.3389/fmars.2016.00073. Reise,K.,Olenin,S.,Thieltges,D.W.,2006.ArealiensthreateningEuropeanaquatic
coastalecosystems?Helgol.Mar.Res.60,77–83,http://dx.doi.org/10.1007/ s10152-006-0024-9.
Rice,J.C.,Rochet,M.-J.,2005.Aframeworkforselectingasuiteofindicatorsfor fisheriesmanagement.ICESJ.Mar.Sci.J.Cons.62,516–527,http://dx.doi.org/ 10.1016/j.icesjms.2005.01.003.
Rogers,S.,Casini,M.,Cury,P.,Heath,M.,Irigoien,X.,Kuosa,H.,Scheidat,M.,Skov, H.,Stergiou,K.,Trenkel,V.,Wikner,J.,Yunev,O.,2010.MarineStrategy FrameworkDirectiveTaskGroup4Report:FoodWebs(NoEUR24343EN− 2010).JRC,Ispra,Italy.
Rossberg,A.G.,2013.FoodWebsandBiodiversity:Foundations,ModelsData. Wiley.
Rossberg,A.G.,2012.Acompleteanalytictheoryforstructureanddynamicsof populationsandcommunitiesspanningwiderangesinbodysize.Adv.Ecol. Res.46,429–522.
Savchuk,O.P.,Wulff,F.,2007.ModelingtheBalticseaeutrophicationinadecision supportsystem.AMBIOJ.Hum.Environ.36,141–148,
10.1579/0044-7447(2007)36[141:MTBSEI]2.0.CO;2.
Schwartz,M.K.,Luikart,G.,Waples,R.S.,2007.Geneticmonitoringasapromising toolforconservationandmanagement.TrendsEcol.Evol.22,25–33,http://dx. doi.org/10.1016/j.tree.2006.08.009.
Shephard,S.,Fung,T.,Houle,J.E.,Farnsworth,K.D.,Reid,D.G.,Rossberg,A.G.,2012.
Size-selectivefishingdrivesspeciescompositionintheCelticSea.ICESJ.Mar. Sci.69,223–234.
Shephard,S.,Fung,T.,Rossberg,A.G.,Farnsworth,K.D.,Reid,D.G.,Greenstreet, S.P.R.,Warnes,S.,2013.ModellingrecoveryofCelticSeademersalfish communitysize-structure.Fish.Res.140,91–95,http://dx.doi.org/10.1016/j. fishres.2012.12.010.
Strayer,D.L.,Malcom,H.M.,2006.Long-termdemographyofazebramussel (Dreissenapolymorpha)population.Freshw.Biol.51,117–130,http://dx.doi. org/10.1111/j.1365-2427.2005.01482.x.
Taris,N.,Ernande,B.,McCombie,H.,Boudry,P.,2006.Phenotypicandgenetic consequencesofsizeselectionatthelarvalstageinthePacificoyster (Crassostreagigas).J.Exp.Mar.Biol.Ecol.333,147–158,http://dx.doi.org/10. 1016/j.jembe.2005.12.007.
Thresher,R.E.,Kuris,A.M.,2004.Optionsformanaginginvasivemarinespecies. Biol.Invasions6,295–300,http://dx.doi.org/10.1023/B:BINV.0000034598. 28718.2e.
Vahtera,E.,Conley,D.J.,Gustafsson,B.G.,Kuosa,H.,Pitkänen,H.,Savchuk,O.P., Tamminen,T.,Viitasalo,M.,Voss,M.,Wasmund,N.,Wulff,F.,2007.Internal ecosystemfeedbacksenhancenitrogen-fixingcyanobacteriabloomsand complicatemanagementintheBalticSea.AMBIOJ.Hum.Environ.36, 186–194,http://dx.doi.org/10.1579/0044-7447(2007)36[186:IEFENC]2.0.CO;2. Varjopuro,R.,2011.Co-existenceofsealsandfisheries?Adaptationofacoastal
fisheryforrecoveryoftheBalticgreyseal.Mar.Policy35,450–456,http://dx. doi.org/10.1016/j.marpol.2010.10.023(TheHumanDimensionsofNorthern MarineMammalManagementInATimeOfRapidChange).
Vellend,M.,Baeten,L.,Myers-Smith,I.H.,Elmendorf,S.C.,Beauséjour,R.,Brown, C.D.,Frenne,P.D.,Verheyen,K.,Wipf,S.,2013.Globalmeta-analysisrevealsno netchangeinlocal-scaleplantbiodiversityovertime.Proc.Natl.Acad.Sci.110, 19456–19459,http://dx.doi.org/10.1073/pnas.1312779110.
Verdonschot,P.F.M.,Spears,B.M.,Feld,C.K.,Brucet,S.,Keizer-Vlek,H.,Borja,A., Elliott,M.,Kernan,M.,Johnson,R.K.,2012.Acomparativereviewofrecovery processesinrivers,lakes,estuarineandcoastalwaters.Hydrobiologia704, 453–474,http://dx.doi.org/10.1007/s10750-012-1294-7.
Williamson,M.,1997.BiologicalInvasions,1996edition.Springer,London,New York.
WorldCommissiononEnvironmentandDevelopment,1987.OurCommonFuture, OxfordPaperbacks.OxfordUniversityPress,Oxford,NewYork.
Wright,S.,1938.Sizeofpopulationandbreedingstructureinrelationtoevolution. Science87,430–431.
Zaiko,A.,Lehtiniemi,M.,Narˇsˇcius,A.,Olenin,S.,2011.Assessmentofbioinvasion impactsonaregionalscale:acomparativeapproach.Biol.Invasions13, 1739–1765,http://dx.doi.org/10.1007/s10530-010-9928-z.
Zaiko,A.,Minchin,D.,Olenin,S.,2014.Thedayaftertomorrow:anatomyofanr strategistaquaticinvasion.Aquat.Invasions9,145–155.
