Assessmentofwetlandfunctionwhattheyareandwhattheya

AssessmentsofWetlandFunctions:WhatTheyAreandWhatTheyAreNotTHOMASHRUBYWashingtonStateDepartmentofEcologySEAProgramP.O.Box47600Olympia,Washington98504,USAABSTRACT/Manymethodshavebeendevelopedoverthelasttwodecadestoprovideinformationaboutwetlandfunc-tions,buttherehasbeenlittlediscussionofthemodelsandalgorithmsused.Methodsforgeneratinginformationaboutwetlandswereanalyzedtounderstandtheirsimilarities,dif-ferences,andthetypeofinformationprovided.Methodscanfirstbegroupedbythetypeofinformationtheyprovide—classifications,characterizations,ratings,assessments,andevaluations.Methodsthatcharacterize,rate,orassesswet-landsmaygenerateinformationusingoneoftwoconceptualapproaches—logicandmechanistic.Mostmethodsthatgenerateanumericassessmentofperformanceorvalueofwetlandfunctionsrelyonthemechanisticapproachtocon-structingmodels.Rapidassessmentmethodsbasedonmechanisticmodels,however,donotassesstheratesordynamicsofecologicalprocessesoccurringinwetlands.Rather,theyprovideaclearandconcisewayoforganizingourcurrent,andoftensubjective,knowledgeaboutwetlandfunctions.Thisisonelimitationofcurrentmethodsthatisoftenmisunderstoodbothbywetlandmanagersandthesci-entificcommunity.Theadvantagesandlimitationsoftheas-sumptionsandthecomputationalelementsinherentintheseapproachesarediscussedtoprovidewetlandmanagersandregulatorsabetterunderstandingoftheinformationtheyareusing.OveradecadeagoJonKusler(1986)commentedthatmethodsforanalyzingwetlandfunctionsdevel-opeduptothenwereusuallyspeciesspecific,shortterm,andnarrowintheirscope.Therewasaneedforrapid,comprehensiveapproachesthatassessedarangeofwetlandfunctionsandthatcouldbeusedinaregulatoryorplanningcontext,notascientificorresearchone.Theneedisforcomprehensiveinformationthatwillsupportsounddecisionsaboutcompensatorymitigation,acqui-sition,restoration,andassessmentofanthropogenicimpacts(NationalResearchCouncil1995).Manymethodshavebeendevelopedinthelastdecadetofillthisneed.Someareregionalinscope,somenational,andsometrytomeetspecificlocalplanningorregulatoryneeds.Manyaremodificationsofaprecedingmethod.CurrentlytheUSArmyCorpsofEngineersisleadinganationwideefforttodevelopaseriesofregionalfunctionassessmentmethodsbasedonahydrogeomorphicclassificationofwetlands(theHGMapproachtoassessment;SmithandBartoldus1995,Brinson1995,1996).Thisnationwideeffortissignificant.By1996,atleast$5millionhadbeenspentontheproject(C.Rhodes,USEPA,personalcommuni-cation).Althoughthelastdecadehasseenmanynewmeth-odsdevelopedtoassesswetlandfunctions,therehasbeenlittlediscussionoftheassumptionsonwhichthemethodsarebased,therationaleforchoosingthecomputationalmodels,orthetypeofinformationgeneratedbythesemethods.Furthermore,fewoftherapidmethodshaveundergoneavalidation,wheretheresultsofamethodaretestedagainstactualmeasure-mentsoflevelsoffunctions(WorldWildlifeFund1992).Asaresult,wetlandscientistsoftengreetnewassessmentmethodswithskepticismandquestiontheirscientificvalidity(Brinsonandothers1997a).Rapidassessmentmethodsthatgeneratenumericresultsareespeciallysuspectbecausetheyoftenlackadequatedata(Adamus1986).Environmentalmanagers,ontheotherhand,areconfusedaboutthetypeofinformationthemethodsprovide.ArecentmeetingoftheAssociationofStateWetlandManagersinAnnapolis,Maryland,USA,inMarch1997,andarecentlettersbyBrinsonandothers(1997a,b)andHruby(1997),havehighlightedthemisconceptionsthatsomewetlandmanagersmayhaveaboutthetypeofinformationprovidedbyassessmentmethods.Theyhavetroubledifferentiatingamong:(1)thelevelsatwhichwetlandsperformfunctionsandthevaluesgiventothosefunctions;(2)themeaningofindicesgeneratedbydifferentmethods;and(3)thepotentialthatawetlandhastoperformafunctionversustheactuallevelofperformance.UnderstandingandcorrectingmisconceptionsisimportantbecausemanyregulatoryandmanagementKEYWORDS:Wetlands;Functions;Assessment;Models;MethodsEnvironmentalManagementVol.23,No.1,pp.75–85r1999Springer-VerlagNewYorkInc.decisionsarebasedontheresultsofanassessmentmethod.Wrongdecisionsmaybemade,orwetlandsmaybenotbeadequatelyprotected,ifinappropriatemethodsareused(KuslerandNiering1998).Thispaperexaminesexistingassessmentmethodsandproposedmethods(e.g.,theUSArmyCorpsofEngineersHGMapproachtodevelopingmethods)toidentifysomeoftheirbasicassumptionsandcomputa-tionalelements.Myhopeistoprovideabetterunder-standingoftheinformationthatwetlandassessmentmethodsdoanddonotprovide.Thefirstsectiondescribesthetypesofinformationcurrentrapidassessmentmethodsprovide.Thesecondsectiondescribesthetwotypesofcomputationalap-proachescommonlyused.Thethirdsectiondescribesthestructureandelementsofmechanisticassessmentmodelsthatarethebasisofmostcurrentrapidap-proachestowetlandassessment.InformationProvidedbyWetlandAssessmentMethodsMethodsfororganizingourknowledgeaboutwet-landshavebeencalledclassifications,categorizations,characterizations,ratings,assessments,andevaluations.Thesegroupingsaremeanttoindicatethetypeofinformationamethodprovides.Unfortunately,thescientificcommunityhasbeensloppyintheuseofthesetermstotheextentofmistitlingmanyoftheanalyticaltoolsdeveloped[e.g.,thewetlandevaluationtechnique(WE