Application of Ecological Indicators

Author(s): Gerald J. Niemi and Michael E. McDonald

Source: Annual Review of Ecology, Evolution, and Systematics, Vol. 35 (2004), pp. 89-111
Published by: Annual Reviews
Stable URL: http://www.jstor.org/stable/30034111 .
Accessed: 18/05/2013 05:42
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Annual Reviews is collaborating with JSTOR to digitize, preserve and extend access to Annual Review of
Ecology, Evolution, and Systematics.

http://www.jstor.org

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

Annu. Rev. Ecol. Evol. Syst. 2004. 35:89-111

doi: 10.1146/annurev.ecolsys.35.112202.30000005
Firstpublishedonline as a Review in Advance on June 10, 2004

OF ECOLOGICAL
INDICATORS*
APPLICATION
GeraldJ.Niemi1and MichaelE.McDonald2
'NaturalResourcesResearchInstituteand

of
of Biology,University
Department
Minnesota,Duluth,Minnesota55811;email:gniemi@d.umn.edu
andAssessment
Protection
2U.S.Environmental
Monitoring
Agency,Environmental
email:
20191;
McDonald.Michael@epa.gov
Program,Reston,Virginia
Key Words assessment,condition,monitoring,responses,stressors
0 Abstract Ecologicalindicatorshave widespreadappealto scientists,environmentalmanagers,and the generalpublic. Indicatorshave long been used to detect
changesin nature,but the scientificmaturationin indicatordevelopmentprimarily
has occurredin the past40 years.Currently,indicatorsaremainlyused to assess the
conditionof the environment,as early-warningsignals of ecological problems,and
as barometersfor trendsin ecologicalresources.Use of ecologicalindicatorsrequires
clearlystatedobjectives;the recognitionof spatialandtemporalscales; assessments
of statisticalvariability,precision,andaccuracy;linkageswith specificstressors;and
couplingwitheconomicandsocialindicators.Legislativelymandateduse of ecological
accords.
indicatorsoccursin manycountriesworldwideandis includedin international
As scientificadvancementsandinnovationin the developmentanduse of ecological
indicatorscontinuethroughapplicationsof molecularbiology,computertechnology
and
suchas geographicinformationsystems,datamanagementsuchas bioinformatics,
remotesensing,ourabilityto applyecologicalindicatorsto detectsignalsof environmentalchangewill be substantiallyenhanced.

INTRODUCTION
Humanstryingto understandthe currentconditionor predictthe futurecondition
of ecosystems have often resortedto simple, easily interpretedsurrogates.Often
these surrogateshavebeen indicatorsthatallow humansto isolatekey aspectsof the
environmentfrom an overwhelmingarrayof signals [NationalResearchCouncil
(NRC) 2000]. Early humansused indicatorslike seasonal migratorymovements
of animals or floweringby spring flora to provide insight into changing environmental conditions. The first reference to environmentalindicators is attributed
to Plato, who cited the impacts of human activity on fruit tree harvest (Rapport
1992). Morrison(1986) reviewedthe work of Clements(1920) and noted thatthe
*TheU.S. Government
hastherightto retaina nonexclusive,
royalty-free
licensein andto
anycopyrightcoveringthispaper.
89

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

90

NIEMI0 MCDONALD
conceptof indicatorsfor plantandanimalcommunitiescan be tracedto the 1600s.
Clements's (1920) work set the scientific stage for using plants as indicatorsof
physical processes, changes to soil conditions, and other factors. In the 1920s,
indicatorswere also being successfullyused to determinechangingenvironmental
conditions, such as waterclarity (Rapport1992) or air qualitywith "thecanaryin
the mine" (Burrell& Siebert 1916), which we continueto use (VanBiema 1995).
One of the more elaborateearly environmentalindicatorswas the saprobiansystem (Kolkwitz & Marsson 1908), which used benthic and planktonicplants as
indicatorspecies for classifying streamdecompositionzones.
The past 40 years have seen a rapidaccelerationof scientificinterestin the developmentandapplicationof ecological indicators.This focus on indicatorsstems
from the need to assess ecological condition in making regulatory,stewardship,
sustainability,or biodiversitydecisions. Forexample,the CleanWaterAct of 1972
requiresthat each state produce a reportevery two years on the condition of all
its waters to the U.S. EnvironmentalProtectionAgency (US EPA) for Congress.
Decisions regardingsustainabilityand biodiversityinvolve both researchandpolicy issues (e.g., Mann & Plummer 1999, Ostrom et al. 1999, Tilman 1999). In
the United States, this researchhas been legislatively mandatedto variousfederal
agencies, in particularto the U.S. Departmentof Agriculturethrough the National ForestManagementAct of 1976, to the U.S. Departmentof Interior(1980,
Parsons2004), andto the US EPA (2002b). These mandateshaveresultedfromthe
increasingconcern for the loss of species, deterioratingwater quality and quantity, sustainabilityof resource use, climate change, and overall condition of the
environment.This interesthas generatedmany new books, articles, and reviews
on ecological indicators(e.g., McKenzieet al. 1992; US EPA2002b,c), as well as
a new journal(Ecological Indicatorsin 2001).
The public has increasinglydemandeda betteraccountingof the conditionor
health of the environmentand whether it is improving or getting worse (Heinz
Center 1999; www.heinzctr.org/ecosystems).Developing scientificallydefensible
indicatorsto establish environmentalbaselines and trendsis a universalneed at
a variety of levels. For instance, federal governmentsin the United States and
Canada(EnvironmentCanadaand US EPA 2003), Europe (www.eionet.eu.int),
and Australia(www.csiro.au/csiro/envind/index.htm)
have developed or are defor
routine
on
veloping programs
reporting ecological indicators.Recent international accords(e.g., RIO Accord) have demandedan accountingand reportingof
indicatorson the stateof the environment.The Montr6alProcess (www.mpci.org)
representing12 countrieswas establishedin 1994 to developand implementinternationallyagreedupon criteriaandindicatorsfor the conservationand sustainable
managementof temperateand boreal forests. In 2003, US EPA (2003a) released
its first state of the environmentreport(www.epa.gov/indicators/roe/index.htm).
As the world human population continues to increase exponentially (Cohen
2003), and with consequent environmentaldemands,the applicationsof indicators to determinestatus and trends in environmentalcondition will continue to
grow.

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

ECOLOGICAL
INDICATORS

91

DEFINITIONS
Earlyuses of indicatorsprimarilyreflectedenvironmentalconditions,andthe terms
environmentalandecological indicatorshave oftenbeen used interchangeably.Environmentalindicatorsshouldreflectall the elements of the causal chainthatlinks
humanactivitiesto theirultimateenvironmentalimpactsandthe societalresponses
to these impacts(Smeets & Weterings1999). Ecological indicatorsarethen a subset of environmentalindicatorsthatapplyto ecological processes (NRC 2000). For
policy makers,the amountof ecological datais often overwhelming.Environmental indicatorsarean attemptto reducethe informationoverload,isolate key aspects
of the environmentalcondition,documentlarge-scalepatterns,andhelp determine
appropriateactions(Niemeijer2002). An exampleof a large-scale,policy-relevant
environmentalindicatoris the environmentalsustainabilityindex (ESI). The ESI
was developed to allow quantitativeinternationalcomparisonsof environmental
conditions (WorldEconomic Forum 2002). The ESI has five major categories:
environmentalsystems, reducingenvironmentalstresses,reducinghumanvulnerability, social and institutionalcapacity,and global stewardship.In 2001 the ESI
includedinformationon 68 indicatorswithin these categoriesfrom 142 countries
(WorldEconomic Forum2002).
Ecological indicatorsembodyvariousdefinitionsof ecology, such as the "interactionsthatdeterminethe distributionandabundanceof organisms"(Krebs 1978),
or morebroadlythe "structureandfunctionof nature"(Odum1963). Thus,they are
often primarilybiological andrespondto chemical,physical, and otherbiological
(e.g., introducedspecies) phenomena.We have chosen to combine the definitions
of the US EPA(2002b) andthe hierarchyof Noss (1990), andwe defineecological
indicatorsas: measurablecharacteristicsof the structure(e.g., genetic, population,
habitat, and landscape pattern),composition (e.g., genes, species, populations,
communities, and landscape types), or function (e.g., genetic, demographic/life
history,ecosystem, and landscapedisturbanceprocesses) of ecological systems.
Ecological indicatorsare derivedfrom measurementsof the currentcondition
of ecological systems in the field and are either used directly or combined into
one or more summaryvalues (US EPA 2002b). These ecological indicatorscan
be aggregatedinto ecological attributeswith reportingcategories, such as biotic
condition,chemical andphysical characteristics,ecological processes, and disturbance (Harwellet al. 1999, US EPA2002b). Ecosystemdisturbancecan be natural
(e.g., fire, wind, and drought)and partof the functionalattributesof ecosystems
(Noss 1990), or it can be anthropogenic.Ecological indicatorshave been applied
in manyways in the contextof both naturaldisturbancesandanthropogenicstress.
However,the primaryrole of ecological indicatorsis to measurethe responseof the
ecosystem to anthropogenicdisturbances,but not necessarily to identify specific
anthropogenicstress(es) causing impairment(US EPA 2002b). These indicators
have been referredto as "state indicators"in the State of the Lakes Ecosystem
Conference(SOLEC),which is a joint effort of Canadaand the United States to
develop indicatorsfor the GreatLakes (EnvironmentCanadaand US EPA 2003).

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

92

NIEMI0 MCDONALD
SOLECdefines state indicatorsas response variables(e.g., fish, bird, amphibian
populations)and pressureindicatorsas the stressors(e.g., phosphorusconcentrations, atmosphericdepositionof toxic chemicals, or waterlevel fluctuations).
In this review we focus on ecological indicators,but clearly they can be integratedwith the broaderissues of ecosystem health (Rapportet al. 1998) and
ultimately with economic indicators (Milon & Shogren 1995) to be even more
useful for making policy decisions. There is a continuingdebate on how to accomplishthis integration.A commongoal of linkingeconomic andenvironmental
indicatorsis often based on the concept of sustainability.Forexample,Ekinset al.
(2003) provided a frameworkfor linking economic, social, and environmental
sustainability.Their approachidentifiedhow economic and social options were
constrainedif critical environmentalfunctions were sustained.Lawn (2003) explored the theoreticalfoundationof several indexes of sustainability,including
the Index of SustainableEconomic Welfareand the Genuine ProgressIndicator.
He found that these indexes were theoreticallysound, but more robustvaluation
methodswere necessary.Althoughprogressis being made,thereareno indicators
that link economic, social, or environmentaltrendsin a way thatis meaningfulto
the public.

USE OF ECOLOGICAL
INDICATORS
Ecological indicatorsare primarilyused either to assess the condition of the environment(e.g., as an early-warningsystem) or to diagnose the cause of environmental change (Dale & Beyeler 2001). The widespreaddecline of the peregrine
falcon (Falco peregrinus)in the 1950s is an excellent example of both uses. The
catastrophicdecline of the species served as an early-warningsignal of problems
in the environment,and researchon the cause of the decline led to the diagnosis
of widespreadcontaminationby chlorinatedhydrocarbonssuch as DDT (Ratcliffe
1980). The widespreaddecline of amphibianshas also been viewed as an earlywarningsignal of problemsin the environment,yet furtherresearchhas failed to
identify a specific cause for the decline. Amphibiandeclines are likely due to a
varietyof factors,includinghabitatchange, global climate change, chemical contamination,disease and pathogens,invasive species, and commercialexploitation
(Blaustein & Wake 1995, Semlitsch 2003).
The informationgatheredby ecological indicatorscan also be used to forecast futurechanges in the environment,to identify actions for remediation,or, if
monitoredover time, to identify changes or trendsin indicators(Figure 1). As the
complexity of the system being monitoredincreases (e.g., greaterspatial scales
and levels of biological organization)or as the temporalscale increases,the cost
of gathering,analyzing, and reportingon indicatorsincreases. Complexity also
arises from the need to quantifylinkages between specific stressorsand ecological indicators(Table 1). In the few cases in which such relationshipshave been
determined,these ecological indicatorsare often considereddiagnostic;however,
these linkages have seldom been made (Suter et al. 2002). A major challenge

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

ECOLOGICAL
INDICATORS

93

Monitoring networks

(decreasecosts4 decreaseinformation)
Landscape
Ecosystem
Community
Population
scale Individual
Tissue
Cell
Spatial
Gene

Condition
Diagnosis
Forecasting
Remediation
Change/Trends

Temporal
scale

Scientific understanding

(increaseinformation
- increasecosts)

Figure 1 Illustrationof the suite of ecologicalindicators(left) for which a suite of

assessmentcapabilities(right)aredesired.Constraintson the developmentof ecological indicatorsat all levels for all assessmentendpointsaredue to a lack of scientific
andthe predominanceof policiesrequiringlow cost monitoring.Goals
understanding
in applicationsgenerallyinclude a compromisebetweencost-effectivenessand the
abilityto defendtheecologicalindicatorscientificallyat the spatialandtemporalscale
to answerthe desiredmanagementobjectives.
appropriate
continuesto be the difficultyof discerningspecific stressor-responserelationships
in a multiple stressorenvironmentand the difficultyof separatinganthropogenic
from naturalsources of variation(Niemi et al. 2004).
Ecological indicatorsare usually developed by scientists and focused on aspects of ecosystems they believe are importantfor the assessment of condition.
However,environmentalmanagersand policy makersrequireindicatorsthat are
understoodby the public (Schiller et al. 2001). Ideally,policy-relevantindicators
would allow:(a) assessmentof bothexisting andemergingproblems;(b) diagnosis
of the anthropogenicstressorsleading to impairments;(c) establishmentof trends
in condition for measuringenvironmentalpolicy and programperformance;and
(d) ease of communicationto the public. Besides capturingthe complexities of an
ecosystem andbeing easy to communicate,an indicatorshould also be easily and
routinelymeasurable(Dale & Beyeler 2001). Moreover,the cost of monitoringand
subsequentanalyses is also a considerationfor state and federal agencies. Classificationsof indicatorsthatinclude scientific performance,policy relevance,and
public acceptancehave been proposed(Noss 1990, Cairnset al. 1993). However,
the final choice of indicatorsshould dependon the questionsbeing asked and the
quality of the science supportingthe indicator.
Frost et al. (1992) suggest that ecological indicatorsshould tradeoff two potentially contradictoryendpoints. They should be sensitive enough to react in a
detectableway when a system is affectedby anthropogenicstress,andthey should
also remainreasonablypredictablein unperturbedecosystems. McGeoch (1998)

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

94

NIEMIa MCDONALD
TABLE1 Examplesof ecologicalresponsesto naturalandanthropogenic
stress
Stress
Naturaldisturbance
Forestfire
Drought
Herbivory
Wind
Anthropogenicstress
Acidrain
Eutrophication
Introduced
species
Sedimentation
Logging
Heavymetals
Urbanization
Airpollution
Airquality

Ecologicalresponse

Reference

Landscape
pattern
Birdpopulations
andlitter
Vegetation
Foreststands

Turner
et al. 1994
Blakeet al. 1994
McInneset al. 1992
Foster1988

Feathermoss
Aquaticmacrophytes
Birdpopulations
Shrubs
Landscape
pattern
Mosquitogenefrequencies
Birdguilds
Plantspecies
Lichens

Hutchinson
& Scott1988
Kangaset al. 1982
Savidge1984
Johnston2003
Franklin
& Forman1987
Guttman1994
O'Connellet al. 2000
Stolte& Mangis1992
Kinnunen
et al. 2003

provided an extensive list of suggested criteria to consider in the selection of

bioindicatorsthat included cost, species abundance,baseline data on species biology, and sensitivity to stress. Indicatorselection will always be a compromise
among many factorsand must be optimizedfor the intendedpurpose.
There has been a strong recent interest in reportingon the ecological condition of the environment(Heinz Center 2002, US EPA 2003a) for the purposes
of planning,management,and public reporting(US EPA 2003a). To accomplish
these goals, the ecological indicatorsmust be able to detect anthropogenicchange
against a backgroundof naturalvariability.At issue is that ecological indicators
at the populationor communitylevels are not tightly coupled to the primarybiological effects of stressors,which results in a slower responsetime, high natural
variability,andlow sensitivity(Jenkins& Sanders1992). Confoundingthis further,
communitiesandpopulationsarerespondingto manyotherfactors,some of which
are not necessarilystressorrelated.In some cases it may be our attemptsat aggregating the datathataffect change detection.Frostet al. (1992) foundthatthe level
of taxonomic aggregationof the zooplanktonpopulationsaffected the detection
of change in an acid sensitive lake. Naturalvariabilitydecreased with increasresponse.At
ing taxonomicaggregation,but sensitivityhad a less straightforward
intermediatelevels of taxonomicaggregation,zooplanktonpopulationsexhibited
the highest sensitivity (Frostet al. 1992). Thus, some intermediatelevel of taxonomic aggregationmay be requiredto optimize the trade-offbetween sensitivity
and variabilityin producinga useful ecological indicator.
At the organismlevel, importantphysiologicalprocessescanchangein response
to anthropogenicstress (e.g., growth, fecundity,developmentalrates). However,

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

ECOLOGICAL
INDICATORS

95

the initial biological response often occurs at the cellular or subcellularlevel.

This fact has advancedthe use of biomarkersto detect physiologic condition and
exposure to stressors(Jenkins& Sanders 1992). These biochemical and cellular
indicatorstend to be more sensitive to contaminantsand more responsive than
higherlevel indicators.Metallothioneininductionis an example.Metallothioneins
are low molecularweight, metal binding proteinsthat are involved in homeostasis regulationand compartmentalizationof essential metals (e.g., Brouweret al.
1989). In the presenceof excess metals,inductionof metallothioneinsis enhanced
for detoxification(Hamer1986). The drawbackto this strongindicatorof anthropogenic insultis in its interpretationin the broaderecological context.Forinstance,
it is unclearwhetherthe biomarkeris relatedto the conditionor populationof the
species.
The questions,goals, and objectivesof a monitoringprogramdeterminewhich
ecological indicatorsareused (Dixon et al. 1998). Ecological indicatorshave been
appliedfrom the level of the gene (e.g., Rublee et al. 2001) to the landscape(e.g.,
Lausch& Herzog2002) (Table2). Researchersneed to recognizewhich partof the
ecological indicatorspectrumis relevantto the objectives of their investigation.
For example, is the indicatoran early-warningsystem that may be related to a
specific stress?Is the indicatora measurementof the condition of the ecological
system? Is it importantto know the cause of any change in the indicator?Clearly
statedgoals and objectives are essential (Yoccoz et al. 2001).
Most ecological monitoringprogramsare based on an aggregationof selected
sites (Olsen et al. 1999), andresearchersoften inferregionaltrendsfromthe accumulationof these site-specifictrends(Urquhartet al. 1998). Manyof these studies
are useful for establishingtemporalvariabilityand mechanisticrelationships,but
largerregionaltrendscannotbe compiled from these site-specific dataunless the
sites were selected in a representativeand unbiased manner within the region
of interest (Urquhartet al. 1998). Thus, researchersmust integratethe selection
of ecological indicatorsfor examining large-scale spatial trendsin an ecosystem
within an appropriatestatisticaldesign (McDonaldet al. 2004).

INDICATORSPECIES
Most applicationsof ecological indicatorshave focused at the species level owing to concerns arising from endangeredspecies and species conservationissues
(Fleishmanet al. 2001). For instance,Noss (1990) statedthat "theuse of indicator
species to monitoror assess environmentalconditionsis a firmlyestablishedtradition in ecology, environmentaltoxicology, pollutioncontrol, agriculture,forestry,
and wildlife and range management."The measurementof an indicatorspecies
assumes that a single species representsmany species with similarecological requirements(Landreset al. 1988). Typically,ecological indicatorspecies tend to
be from the macrofloraand macrofauna,especially aquatic macroinvertebrates,
fish, birds, and vascularplants. The primaryreasons for theiruse are: (a) relative

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

96

NIEMI M MCDONALD

thathavebeenappliedwithindifferentecologicallevelsof
TABLE2 Examplesof indicators
(modifiedfromNoss 1990andUS EPA2003)
organization
Example

Type

References

Speciesdifferentiation Rudiet al. 2000

Andersonet al. 1989
Immuneresponse
et al. 2003
Metalconcentration Pdrez-Lop6z
Butterflies
MacNally& Fleischman
2002
Browderet al. 2002
Birds
Populations
Floristicquality
Communities
Lopez& Fennessy2002
Whittieret al. 2002
Lakes
Ecosystems
Lausch& Herzog2002
Landuse/cover
Landscapetypes
Ameset al. 1973
rates
Mutation
Genetic processes
Sierszen& Frost1990
rate
Behavior
Feeding
et al. 2003
Hausner
traits
Life history
Species
Underwood& Roth 2002
processes Productivity
Demographic
Marwoodet al. 2001
Growth
Ecosystem processes
Dixit et al. 1992
Diatoms
Landscapeprocesses
Geneticstructure
genotypic Baird et al. 1990
Zooplankton
differentiation
Birdguilds
Croonquist& Brooks
Populationstructure

Compositional Genes
Cell and subcellular
Tissue
Species

Functional

Structural

Integrative

Habitatphysiognomy
Landscapepatterns
Index of biotic integrity
AMOEBA
Multivariate
Species assemblages

Foreststructure
Fragmentation
Fish
Multipletaxa
Biomarkers
Beetles

Index of environmental Multipleindices

integrity

1991
Lindenmayeret al. 2000
O'Neill et al. 1988
Karr1981
ten Brink et al. 1991
Cormier& Racine 1992
Dufr6ne& Legendre
1997
Paul 2003

ease of identification,(b) interestto the public, (c) relativeease of measurement,

(d) relatively large numberof species with known responses to disturbance,and
(e) relativelylow cost.
We use indicatorspecies as a generalterm to refer to approachesthat use one
or a few species to "indicate"condition or a response to stress that may apply
to other species with similar ecological requirements.Lawton & Gaston (2001)
suggest that indicatorspecies are used in three distinct ways: (a) to reflect the
biotic or abiotic state of the environment;(b) to reveal evidence for the impacts
of environmentalchange;or (c) to indicatethe diversityof other species, taxa, or
communitieswithin an area. The first two reflect the common uses of indicators
as measuresof conditionand the diagnosis of potentialcause(s) of environmental

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

INDICATORS
ECOLOGICAL

97

change. The thirdexpands the concept of indicatorsto incorporatethe idea of a

single species serving as a surrogatefor many other species. This idea has been
largelyuntestedandhas been the focus of muchdebateandcriticismin applications
(see below). Because of the criticism,many new approachesand termshave been
developed to refine the indicator species concept. These include focal species,
umbrellaspecies, flagship species, or guilds as indicators(Verner1984, Landres
et al. 1988, Lambeck 1997, Simberloff 1998, Noss 1999).
The termfocal species has been used in many ways in the literature.For example, Cox et al. (1994) identified44 focal species to serve as umbrellaor indicator
species of biological diversityin Florida.Lambeck(1997) identifiedfocal species
as a subset of the total pool of species in a landscape. Carrollet al. (2001) used
carnivoresas focal species in regional conservationplanning because their distributionalpatternsreflected regional-scalepopulationprocesses. There is not a
consistent definitionof a focal species, except when they are selected by various
meansas the "focus"of study.Focal species tendto differfromindicatorspecies in
thatthey do not necessarilyserve to measureecological conditionnor do they convey a stress-responserelationship.The focal species concept has generally been
used in conservationplanning, landscape ecology, and protection of biological
diversity.
Fleischmanet al. (2001) define umbrellaspecies as those "whose conservation
confers a protectiveumbrellato numerousco-occurringspecies." They also point
out that"blurreddiscriminationsbetween umbrellaand indicatorspecies have led
to misunderstandingsover how umbrella species should be selected." Flagship
species arethose thathave largepublic appeal,such as charismaticmegafaunalike
bearsandtigers.The guild concepthas been exploredas an alternativeto indicator
species both in wildlife managementand in determiningregional condition(e.g.,
Verner1984, O'Connellet al. 2000). Guildswere originallydefinedby Root (1967)
as a "groupof species that exploit the same class of environmentalresourcesin
a similar way"; Verner (1984) concluded that the guild concept held promise
but requiredmore testing. O'Connell et al. (2000) distinguished 16 behavioral
and physiological response guilds of birds and were able to combine their bird
community data into a bird community index (BCI). They related the BCI to
landscapecondition and change from forestedto nonforestedareas.
Lambeck (1997) expanded on the concepts of umbrellaand focal species to
incorporatemore specific responses to landscapeand managementregimes. His
analysis focused on identifying focal species with the most demandingsurvival
requirementsfor severalparametersthreatenedby anthropogenicstressors.Noss
(1999) furtherextendedandcombinedthese conceptsfor indicator,focal, andumbrella species in a forest managementcontext. In his approach,indicatorspecies
were representedby a suite of focal species, each of which was defined by different attributesthat had to be presentin the landscapeto retainthe biota. Landscape attributesincluded: (a) area-limitedspecies, (b) dispersal-limitedspecies,
(c) resource-limitedspecies, (d) process-limited species, (e) keystone species,
(f) narrowendemic species, and (g) special cases such as flagship species that

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

98

NIEMIE MCDONALD
areof public concernin the region.Noss (1999) suggestedthatat least for the first
fourcategoriesumbrellaspecies could be definedthatarethe most sensitiveto the
landscapeattribute.
Birds have been the primaryfocus for most terrestrialapplicationsof indicator
species, but insects hold greatpromisebecause of theirspecies richness,biomass,
and role in ecosystem functioning.McGeoch (1998) recognizedthe potentialapplications of insects as indicatorspecies and defined their use as environmental,
ecological, or biodiversityindicators.Researchershave attemptedto examinevertebrates as possible umbrella species for insects, especially for butterflies.For
example, Rubinoff(2001) analyzedthe use of a bird species, the Californiagnatcatcher(Poliotila californica), as a potentialumbrellaspecies for three species of
butterflies.However, the gnatcatcherwas a poor indicatorprimarilybecause of
its ubiquityin the landscapestudied.Insects and othermicrofaunaoffer excellent
potentialas indicatorspecies. They areof limiteduse in terrestrialsystemsbecause
of the cost of samplingand processingandbecause thereis limited acceptanceby
resourcemanagers,politicians, and the generalpublic.
Researchershave developedotherindexes to providemore holistic approaches
to ecological condition.These indexes range from simple diversityindexes, such
as the Shannonand WienerIndex (Shannon& Weaver 1949), to multimetricindexes (e.g., Karr1981, Kerans& Karr1994, Karr2000, Simon 2003). Multimetric
ecological indicatorsare sets of mathematicallyaggregatedor weighted indicators (US EPA 2000a, Kurtzet al. 2001) that combine attributesof entire biotic
communitiesinto a useful measureof condition (US EPA 2002a). The US EPA
has recently used an index for biotic integrity(IBI) for estuarineinvertebratesas
one of the indicatorsin the assessment of the condition of the nation's estuaries
(US EPA2004). Because of the increasinguse of multimetricandotherindicators,
researchershave developed specific guidelines for evaluatingtheir performance
(US EPA 2000a).
Another aspect of ecological indicatorsis whetherto use them as relative or
absolute measures. As a relative measure, the initial measurementbecomes the
baseline for comparisonof futuremeasures.Most monitoringagencies prefer or
requirea more quantitativebenchmarkfor measuringand regulatingchanges in
ecosystems. These benchmarkor referenceconditionscan be definedas the conditions of ecological resourcesunder minimal contemporaryhumandisturbance
(McDonald et al. 2004). As these conditions are often not available for direct
measurement,models and historicinformationare often invokedas best approximations.However,the selectionof referenceconditionsremainsproblematic(NRC
1990).
In summary,focal species representthose selected as a focus for a specific
investigation.There is no consistent definitionof focal species, but the concept
has been expandedfor use in conservationand management.Focal species have
been used to identify potential indicator species when there is a desire to describe ecological condition or measure the response to a disturbance.Either a
focal species or an indicatorspecies may serve as an umbrellaspecies if the goals

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

ECOLOGICAL
INDICATORS

99

are to monitoror manageone species as a surrogatefor otherspecies or to identify

conservationareasfor preservation.Focal, indicator,or umbrellaspecies could be
flagshipspecies if they have a high profileor interestto the public. Moreover,any
of them could be keystone species (sensu Paine 1969a,b) if they are particularly
importantin establishingor maintainingkey ecological processes or structurefor
other species within an ecosystem (Simberloff 1998). Before any investigation,
researchersmustclearly definethese termsandrigorouslytest whetherthe species
can fulfill its purposeas an indicator,umbrella,or keystone species.

IN APPLICATIONSOF ECOLOGICAL
COMPLEXITIES
INDICATORS
Monitoring for ecosystem or resource managementoften requiresdata about a
specific site or sites, whereaspublic policy decisions typicallyrequireinformation
across broadergeographicalregions (Olsen et al. 1999). Many of the existing
ecological monitoringprogramsare periodically or continually used at certain
sites, which may lead to a betterunderstandingof the temporalvariabilityat the
site but may not be representativeof a largerarea (Urquhartet al. 1999). Thus,
ecological indicatorsare needed to assess statusand trendsin ecological systems
and to diagnose cause(s) of declining condition across a range of spatial and
temporalscales (Kratzet al. 1995, NRC 2000, Dale & Beyeler 2001, Niemi et al.
2004).
Each ecological indicatorrespondsover differentspatial and temporalscales;
thus,the contextof these scales mustbe explicitly statedfor each ecological indicator.Furthermore,understandingthe responsevariabilityin ecological indicatorsis
essentialfor theireffective use (US EPA2002c). Withoutsuch an understanding,it
is impossibleto differentiatemeasurementerrorfromchangingcondition,or ananthropogenicsignal from backgroundvariation.Workhas begun on understanding
how naturaland anthropogenicvariabilityof indicatorscan affect statusand trend
detection,but it is closely tied to differentstatisticaldesign considerations(Larsen
et al. 1995). Specific monitoringdesigns and indicatorscan be implementedto
detect changes across temporaland spatialscales (US EPA 1997, 2002d).
In general, as one moves up levels of organizationfrom cellular phenomena
to landscape processes, the spatial and temporal scales of applicationincrease
immensely. Similarly, as larger spatial and temporal scales are considered, the
linkage to specific stressorscan be either obscured or refined depending on the
stressor.For example,one of the largestandmost successful monitoringprograms
is the U.S. Geological Survey Breeding Bird Survey (BBS), which has gathered
dataovera 38-yearperiod(1966 to present)(Saueret al. 2003). TheBBS is intended
to indicatebreedingbird species trendsover relativelylargeregional and national
spatial scales. Thus, researchersmust exercise caution in interpretingresults for
specific regions or combining results from differentregions (James et al. 1996).
In contrast,nesting tree swallows (Tachycinetabicolor) are an effective wildlife

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

100

NIEMIMMCDONALD
indicatorspecies of sedimentchemical contaminantproblems.Because nestlings
are fed flying adult insects, which typically have aquaticearly life histories, the
uptakeof chemicals by nestlings can be relatedto sedimentchemical levels near
the nesting site (Nichols et al. 1995, Jones 2003). Changesin bird trendsfrom the
BBS over large areas are powerfulbecause of the large numberof sample routes
and the a prioriexperimentaldesign, but the causes of changes in species trends
are speculative. In contrast, contaminantuptake in nestling tree swallows and
potentialrisk to wildlife can clearly be connected to food supplies derivedfrom
sediment. Many of the same problems exist for multimetricindexes commonly
used to assess conditionof surfacewatersacross largeregions. These indexes can
distinguishdegradedsites from sites with little or no humanimpact, but they do
not diagnose the causes of impairmentby themselves (US EPA 2003b).

INDICATORSOF
IN ECOLOGICAL
ADVANCEMENTS
BIOLOGICALCOMMUNITIES
Historically,ecological indicatorswere primarilybased on parametersassociated
with individualspecies (e.g., presence)or simple communitymetrics(e.g., species
richnessor diversity).However,manyof these indicatorsdid not fully representthe
entirebiological communityof organismspresent.Hence, Karr(1981) introduced
the IBI using streamfish communities. This index was a numericalsummation
of subsets of the fish communityfrom one area comparedwith a suitable reference area. These reference areas ideally representedareas that were naturalor
undisturbedfrom the same geographic area and with the same general ecological condition. Karr(1981) calculated the IBI using fish community data for a
specific area and subdivided these data into 12 metrics, including the number
of individualsand species found in the sample, the relative abundanceof guilds
(e.g., carnivores),specific species in the sample,andothercategories(e.g., sunfish
species). The IBI was expressedas deviationsfromthe suitablereferenceareasuch
that largervalues representedcommunitiessimilarto the referencearea,whereas
lower values representedareasthatdeviatefromthe reference,potentiallybecause
of stress. The IBI has received considerableattentionand applicationover the
past 20 years, including applicationsto fish (Fausch et al. 1984, Angermeier&
Karr1986, Karret al. 1986, Simon & Emery 1995), macroinvertebrates
(Kerans&
et
al.
communities
al.
et
Klemm
2001,
Karr1994,
(Simon
DeKeyser
2003), plant
et al. 2003), aquaticcommunities(Simon et al. 2000), and birds (O'Connellet al.

2000).

Many other multimetricindexes have evolved over the past 20 years, such
as the Hilsenhoff biotic index (Hilsenhoff 1982) and biological response signatures (Simon 2003). In contrastto multimetricindexes, multivariateindexes
(Reynoldsonet al. 1997, Karr2000) are statisticalanalyses of the biological communityusing a host of multivariatetechniques,such as principalcomponentsanalysis (O'Connoret al. 2000), canonical correspondenceanalysis (Kingston et al.

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

ECOLOGICAL
INDICATORS

101

1992), and combinationsof multivariateanalyses (Dufr6ne & Legendre 1997).

For example, O'Connor et al. (2000) integratedinformationfrom five different
taxonomic groups (diatoms,benthos, zooplankton,fish, and birds) to provide an
index of the ecological conditionof lakes. Theirapproachwas effective in relating
the gross condition of the lakes across taxa, but it was also effective in identifying a differentialresponse by fish to nearshoreconditions. Dufr6ne & Legendre
(1997) used a combinationof multivariateanalyses of carabidbeetle community
data to determineindicatorspecies and species assemblages for groups of sites.
Theirapproachalso includes a randomizationprocedureto test the significanceof
the indicatorvalues.
Many analyticalapproachesof biological communitydata are currentlybeing
developed, tested, and used for ecological indicators. For instance, Andreasen
et al. (2001) and Paul (2003) have recently introducedindexes of ecological and
environmentalintegrity,respectively. These indexes combine informationfrom
severallevels (e.g., biological communities,habitat,expertopinion)into an overall
measureof integrity.The explorationand debate of these approacheswill likely
continuein the future.

CRITICISMSOF INDICATORS
Virtually all attemptsto use ecological indicatorshave been heavily criticized,
and many criticisms are well deserved. For instance, many existing monitoring
programsof indicatorssuffer from two deficiencies: lack of well-articulatedobjectives and neglect of different sources of error(Yoccoz et al. 2001). Indicator
species havebeen especially criticizedin the contextof forestmanagement-related
issues (Landreset al. 1988, Landres1992, Niemi et al. 1997, Rolstad et al. 2002,
Failing & Gregory2003). Many of these criticisms have focused on the lack of:
(a) identificationof the appropriatecontext(spatialandtemporal)for the indicator,
(b) a conceptualframeworkfor what the indicatoris indicating,(c) integrationof
science and values, and (d) validationof the indicator.
Many of these criticismshave led to more focused efforts on individualspecies
and to the developmentof additionalconcepts such as focal species or umbrella
species (Lambeck 1997, Fleischman et al. 2001). Roberge & Angelstam (2004)
recently reviewed the umbrellaspecies concept and concluded that multispecies
strategieswere more compelling. Lawler et al. (2003) evaluatedseveral indicator groups (e.g., birds, fish, mammals, and mussels) to test whether one group
could providehabitatprotectionfor othertaxa in a large areaof the Mid-Atlantic
region of the United States. No single taxonomic indicatorgroup could provide
adequateprotectionfor anothergroup,especially for at-riskspecies withineach of
the groups.The failurewas likely attributableto the narrowgeographicrangesand
restrictedhabitatdistributionof rare species. Hence, informationon rare species
and those thatare at risk was essential,yet gatheringdataon rarespecies is generally difficult,time-consuming,and expensive. In contrastto the indicatorspecies

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

102

NIEMI0 MCDONALD
approach,Manley et al. (2004) evaluatedan innovative,multispeciesmonitoring
approachthat included all terrestrialvertebratespecies over a large ecoregional
scale (7 million ha). The design of this comprehensiveapproachreducedthe emphasis on indicatorspecies because the spatialcoverage allowed many species to
be adequatelymonitored.A fundamentalproblemwith these approachescontinues to be the inabilityto link species presenceor relativeabundancewith relevant
aspects of habitatquality (VanHome 1983), such as productivity.
Many of these same criticisms apply to indexes (Suter 1993). Indexes have
been viewed as oversimplificationsand generalizationsof biological processes, in
which importantdatacan be lost (May 1985). There are also concernsabouthow
these indexes are calibrated(Seegert 2001) and whether or how they are evaluated across gradients(US EPA 2000a). Despite such criticisms,these indexes can
play an importantmanagementrole by helping characterizeecological condition
(Rakocinskiet al. 1997).
Ecological indicators span broad levels of biological, spatial, and temporal
organizationwithin ecosystems. When establishing a monitoringprogramand
selecting indicators,it is imperativethat researchersarticulatea clear statement
of goals. Once the goals are unambiguouslystated, the scientific soundness and
objectivityof the indicatorbecome a centralissue (Niemeijer2002). Researchers
must recognize these complexities and limitationsin the applicationand use of
ecological indicators(Dale & Beyeler 2001). However,havingeffective indicators
is only one componentof the problem.Sound programdesign and effective data
management,analysis, synthesis, and interpretationare also needed to implement
monitoringand assessment programssuccessfully (NRC 1990). In the past five
years, many publications have provided excellent guidance on how ecological
indicatorscan be improved,includingdocumentsby the NRC (2000) andUS EPA
(2000a, 2002b).

INDICATORS
FUTUREOF ECOLOGICAL
AND CONCLUSIONS
Advances in science and technology at all levels of biological organizationwill
greatly improveour ability to apply ecological indicatorsin the future.Recently
developedtechniquesin molecularbiology such as biomarkershave provenuseful
in rapididentificationof problemsin ecological systems causedby pollutionstress
(e.g., Cormier& Racine 1992, Huggett et al. 1992). For example, Arcand-Hoy
& Metcalfe (1999) found that both fluorescentaromaticcompoundsin bile and
in fish could be used to detect exposureto
hepatic ethoxyresorufin-O-deethylase
the
lowerGreatLakes.Evendon& Depledge
in
polynucleararomatichydrocarbons
(1997) identifiedthe potentialusefulness of genetically susceptiblepopulationsto
environmentalcontaminants.Paerl et al. (2003) have recently used diagnostic
photopigmentsof variousphytoplanktongroupsas ecological indicatorsto detect
changesin nutrients,noxious algal blooms, andoverallwaterquality.Investigators

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

ECOLOGICAL
INDICATORS

103

are optimisticaboutapplyingmoleculartechniquesto addressspecific ecological

problems and to act as early-warningsignals of potential problems. However,
researchis necessaryto illustratehow these techniquescan be scaled up to address
environmentalproblems over large regions. There is tremendouspotential for
applicationof these new techniquesto providereal-time,remotelysensedcondition
assessmentsof environmentalproblems(Kerr& Ostrovsky2003).
Global positioning systems (GPS), geographic information systems (GIS),
remote-sensingtechnology,andcomputerhardwareandsoftwarehold greatpotential for advancingthe science of ecological indicators.GPS allows precise location
of repeatedfield measurements,thus reducingerrorsassociatedwith spatialvariation. GIS providesunprecedentedabilities to organize,analyze, synthesize, and
displayinformationgatheredin the field overboth space andtime. Remote-sensing
technology has also advancedsubstantiallyin resolutionfrom 30 m to <4 m resolution (e.g., Kerr & Ostrovsky2003, Clark et al. 2004). Database storage and
software to manipulatethese data have increased exponentially in the past ten
years and have resulted in the new field of bioinformatics.These techniques in
combinationwith data gatheredin the field or combined with existing databases
have proven effective in a myriad of applications, such as change detection in
forest systems (Wolter& White 2002), mappingbiodiversitypatterns(Stockwell
& Peterson2003), and forecastinganimaldistributionsand abundanceover large
regions (Venieret al. 2004).
Researchershaveaddressedmanyof thecriticismsandfailuresthathaveplagued
the applicationsof ecological indicators,resultingin substantialimprovementsin
assessing condition in many areas (e.g., US EPA 1998, 2000b, 2003c). Guidelines for ecological indicatordevelopmentneed to be heeded (Kurtzet al. 2001).
Dependingon the indicator'suse and the spatialscales of application,experimental design considerationsare crucial for appropriateinferences once the data are
gathered(Urquhartet al. 1998, Olsen et al. 1999, Danz et al. 2004).
Of increasinginterestis the integrationof environmentalindicatorswith other
well-known economic and social indicatorslike the gross nationalproductor the
consumerprice index (Milon & Shogren 1995, NRC 1997). The InternationalSociety for Ecological Economics, which recentlybeganpublishingthejournalEcological Economics,was formedpartiallyto integratethis thinkinginto a "transdiscipline"aimed at developinga sustainableworld (www.ecologicaleconomics.org/
about/index.htm).Moreover, a variety of authorshave emphasized the need to
consider humanhealth and link it to environmentalhealth (Pimentel et al. 2000,
Karr2002), as well as to establishan economic valuationsystem for ecological resources (Costanza1997, Daily 1997) or for ecological sustainability(Armsworth
& Roughgarden2003). The motivation for this integrationstems largely from
managers'need to betterquantifyecological changes resultingfrom such issues
as global climate change; species extinction rates; contaminatedair, water, and
soil; declining fish populations;humanconflicts over resourcessuch as water;and
the emergenceof new diseases (e.g., Pimentelet al. 2000, Brown2003, Karr2002)
in relevanthumansocial andeconomic terms.Clearly,the generalpublic currently

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

104

NIEMIE MCDONALD
has a paucity of informationon which to judge the ecological condition of the
environmentor how the condition might relate to human health or to the economy. Yet, with such information,individualsmake daily and long-termdecisions
on the basis of health indicators(e.g., blood pressure),economic indicators(e.g.,
NASDAQ, Dow Jones IndustrialAverage), and environmentalindicators (e.g.,
weatherforecasts).Despite three decades of discussion of the integrationof economic andecological indicators,therearelimitedapplicationsof integratedanalysis (Milon & Shogren 1995). US EPA's (2003a) state of the environmentreportin
2003 is one of the firststeps in informingthe public of the ecological conditionof
the nation'sresources.Futurereportson integratedand understandablemeasures
will be welcome additionsas indicatorsof environmentalsustainability,but their
acceptanceand impactson policy and public opinion will have to be determined.
Strongpublic interestand legislativemandatesexist at local, state, federal,and
internationallevels to understandthe condition, trends,and cause for change in
our ecosystems. A largearrayof ecological indicatorsareavailablefor application
to environmentalproblems;moreover,the numberof tools and techniques that
are available is rapidly increasing. Therein lies the challenge for the future:to
select appropriatemonitoringdesigns and ecological indicatorsthat will provide
convincingscientificunderpinningsfor managementand policy decisions on realworld problems.

ACKNOWLEDGMENTS
Althoughthe researchdescribedin this articlehas been fundedwholly or in partby
the United StatesEnvironmentalProtectionAgency's Science to Achieve Results
(STAR) programthroughcooperativeagreementR828675-00 to the University
of Minnesota,it has not been subjectedto the agency's requiredpeer and policy
review and thereforedoes not necessarilyreflect the views of the agency, and no
official endorsementshouldbe inferred.We thankJames Cox, RobertHowe, and
LucindaJohnsonfor comments on an earlierversion of this manuscript.This is
publication number 359 of the Center for Waterand the Environment,Natural
Resources ResearchInstitute,Universityof Minnesota,Duluth.
The Annual Reviewof Ecology,Evolution,and Systematicsis online at
http://ecolsys.annualreviews.org
LITERATURE
CITED
tions exposedto copperin vitro.J. Aquat.
Ames BN, Lee F, DurstonW. 1973. An improvedbacterialtestsystemforthedetection Anim.Health1:57-61
of mutagensandcarcino- AndreasenJK, O'Neill RV,Noss R, Slosser
andclassification
for the developNC. 2001. Considerations
gens.Proc.Natl.Acad.Sci. USA70:782-86
of
mentof a terrestrial
index ecologicalinAndersonDP, Dixon OW, BodammerJE,
Lizio EF. 1989. Suppressionof antibody- tegrity.Ecol.Indic.1:21-35
producingcells in rainbowtroutspleensec- AngermeierPL, KarrJR. 1986.Applyingan

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

ECOLOGICALINDICATORS
index of biotic integritybased on streamfish
communities:considerationsin samplingand
interpretation.North Am. J. Fish. Manage.
6:418-29
Arcand-HoyLD, MetcalfeCD. 1999. Biomarkers of exposure of brown bullheads(Ameiurus nebulosus)to contaminantsin the lower
GreatLakes, North America. Environ.Toxicol. Chem. 18:740-49
Armsworth PR, RoughgardenJE. 2003. The
economic value of ecological stability.Proc.
Natl. Acad. Sci. USA 100(12):7147-51
BairdDJ, BarberI, CalowP. 1990. Clonalvariation in generalresponsesof Daphnia magna
Straus to toxic stress. I: Chroniclife-history
effects. Funct. Ecol. 4:399-408
Blake JG,HanowskiJM,Niemi GJ,Collins PT.
1994. Annual variationin bird populations
of mixed conifer-northernhardwoodforests.
Condor96:381-99
Blaustein AR, Wake DB. 1995. The puzzle of
declining amphibianpopulations. Sci. Am.
272:52-57
BrouwerM, WingeDR, GrayWR. 1989. Structural and functional diversity of coppermetallothioneinfrom the American lobster,
Homarusamericanus.J. Inorg.Biochem.35:
289-303
Browder SF, Johnson DH, Ball IJ. 2002. Assemblages of breedingbirds as indicatorsof
grasslandcondition.Ecol. Indic. 2:257-70
Brown LR. 2003. Plan B: Rescuing a Planet
Under Stress and a Civilizationin Trouble.
New York/London:WW Norton
Burrell GA, Siebert FM. 1916. Gases found
in coal mines. Miner's Circular14. Bureau
Mines, US Dep. Inter.,Washington,DC
CairnsJ Jr,McCormickPV, NiederlehnerBR.
1993. A proposedframeworkfor developing
indicatorsof ecosystem health.Hydrobiologia 263:1-44
Carroll C, Noss RF, Paquet PC. 2001. Carnivores as focal species for conservationplanning in the Rocky Mountain region. Ecol.
Appl. 11:961-80
ClarkDB, ReadJM,ClarkML, CruzAM, Dotti
MF, et al. 2004. Applicationof 1-m and 4-m
resolutionsatellitedatato ecological studies

105

of tropical rain forests. Ecol. Appl. 14:6174

Clements FC. 1920. Plant Indicators. Washington, DC: CarnegieInst.
Cohen JE. 2003. Human population:the next
half century.Science 302:1172-75
CormierSM, Racine RN. 1992. Biomarkersof
environmentalexposureandmultivariateapproachesfor assessmentandmonitoring.See
McKenzie et al. 1992, 1:229-42
Costanza R, d'Arge R, deGroot R, FarberS,
Grasso M, et al. 1997. The value of the
world's ecosystem services and naturalcapital. Nature 387:253-60
Cox J, Kautz R, MacLaughlinM, Gilbert T.
1994. Closing the gaps in Florida's wildlife
habitat conservationsystem. Rep., Off. Environ. Serv., Fla. Game Fresh Water Fish
Comm., Tallahassee
CroonquistMJ, Brooks RP. 1991. Use of avian
and mammalianguilds as indicatorsof cumulative impacts in riparian-wetlandareas.
J. Environ.Manage. 15:701-14
Daily GC, ed. 1997. Nature'sServices:Societal
Dependence on Natural Ecosystems.Washington, DC: Island
Dale VH, Beyeler SC. 2001. Challengesin the
development and use of ecological indicators. Ecol. Indic. 1:3-10
Danz NP, Regal RR, Niemi GJ, BradyVJ, HollenhorstT, et al. 2004. Environmentallystratified samplingdesign for the developmentof
GreatLakesenvironmentalindicators.J. Environ.Monitor Assess. In press
DeKeyser ES, Kirby DR, Ell MJ. 2003. An
index of plant community integrity: development of methodology for assessing
prairiewetlandplant communities.Ecol. Indic. 3:119-33
Dixit SS, Smol JP, Kingston JC, Charles DE
1992. Diatoms:powerful indicatorsof environmentalchange.Environ.Sci. Tech.26:2233
Dixon PM, Olsen AR, Kahn BM. 1998. Measuring trends in ecological resources.Ecol.
Appl. 8:225-27
Dufr6ne M, Legendre P. 1997. Species assemblages and indicator species: the need

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

106

NIEMI 0 MCDONALD

for a flexible asymmetricalapproach.Ecol.

Monogr.67:345-66
Ekins P, Simon S, Deutsch L, Folke C, DeGrootR. 2003. A frameworkfor the practical
applicationof the conceptsof criticalnatural
capitaland strongsustainability.Ecol. Econ.
44:165-85
EnvironmentCanadaand US EPA. 2003. State
of the GreatLakes2003. EPA905-R-03-004,
Governmentsof Canadaand the US
Evenden AJ, Depledge MH. 1997. Genetic
susceptibility in ecosystems: the challenge
for ecotoxicology. Environ.Health Perspect.
105(Suppl.4):849-54
FailingL, GregoryRS. 2003. Tencommonmistakes in designingbiodiversityindicatorsfor
forestpolicy.J. Environ.Manage.68:121-32
Fausch KD, KarrJR, YantPR. 1984. Regional
application of an index of biotic integrity
based on stream-fish communities. Trans.
Am. Fish. Soc. 113:39-55
Fleischman E, Blair RB, Murphy DD. 2001.
Empirical validation of a method for umbrellaspecies selection.Ecol.Appl. 11:1489501
Foster DR. 1988. Species and standresponses
to catastrophicwind in centralNew England,
USA. J. Ecol. 76:135-51
FranklinJF,FormanRTT.1987. Creatinglandscape patternsby forest cutting: ecological
consequences and principles.Landsc. Ecol.
1:5-18
Frost TM, Carpenter SR, Kratz TK. 1992.
Choosing ecological indicators: effects of
taxonomic aggregation on sensitivity to
stress and naturalvariability.See McKenzie
et al. 1992, 1:215-27
GuttmanSI. 1994. Populationgenetic structure
andecotoxicology.Environ.HealthPerspect.
102(Suppl. 12):97-100
HamerDH. 1986. Metallothionein.Annu.Rev.
Biochem. 55:913-51
Harwell MA, Myers V, Young T, BartuskaA,
GassmanN, et al. 1999. A frameworkfor an
ecosystem integrityreportcard. BioScience
49:543-56
HausnerVH, Yoccoz NG, Ims RA. 2003. Selecting indicator traits for monitoring land

use impacts:birds in northerncoastal birch

forests. Ecol. Appl. 13:999-1012
Heinz Center.1999. Designing a reporton the
state of the nation's ecosystems: selected
measurementsfor croplands, forests, and
coasts and oceans. Rep., H John Heinz III
Center,Washington,DC
Heinz Center.2002. The State of the Nation's
Ecosystems: Measuring the Lands, Waters,
and Living Resources of the United States.
New York:CambridgeUniv. Press
Hilsenhoff WL. 1982. Using a biotic index to
evaluatewaterqualityofstreams.Tech.Bull.
No. 132, Dep. NaturalResources, Madison,
WI
Huggett RJ, Kimerle RA, Mehrle PM Jr,
Bergman HL, eds. 1992. Biomarkers:Biochemical, Physiological, and Histological
Markersof AnthropogenicStress. Boca Raton, FL: Lewis
HutchinsonTC, Scott MG. 1988. The response
of feathermoss, Pleuroziumschreberi,to 5
years of simulated acid precipitationin the
Canadianboreal forest. Can. J. Bot. 66:8288
James FC, McCulloch CE, Wiedenfeld DA.
1996. New approachesto the analysisof population trendsin land birds. Ecology 77:1327
Jenkins KD, Sanders BM. 1992. Monitoring
with biomarkers:a multi-tieredframework
for evaluating the ecological impacts of
contaminants. See McKenzie et al. 1992,
2:1279-93
JohnstonCA. 2003. Shrubspecies as indicators
of wetlandsedimentation.Wetlands23:91120
Jones J. 2003. Tree swallows (Tachycinetabicolor): A new model organism?Auk 120:
591-99
Kangas P, Autio H, Hallifors G, Luther H,
Niemi A, et al. 1982. A generalmodel of the
decline of Fucus vesiculosus at Tvarminne,
south coast of Finland in 1977-1981. Acta
Bot. Fenn. 118:1-27
Karr JR. 1981. Assessment of biological integrity using fish communities. Fisheries
6:21-27

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

ECOLOGICALINDICATORS

KarrJR. 2000. Health, integrity,and biological assessment: the importanceof measuring whole things. In Ecological Integrity:
IntegratingEnvironment,Conservation,and
Health, ed. D Pimentel,L Westra,RF Noss,
pp. 209-26. Washington,DC: Island
Karr JR. 2002. Understanding the consequences of human actions: indicators from
GNP to IBI. In JustEcological Integrity:The
Ethics of MaintainingPlanetary Life, ed. P
Miller, L Westra,pp. 98-110. Lanham,MD:
Rowman& Littlefield
KarrJR, FauschKD, AngermeierPL, YantPR,
Schlosser IJ. 1986. Assessing biological integrity in runningwaters: a method and its
rationale.Spec. Publ. 5, Ill. Nat. Hist. Surv.,
Champaign
KeransBL, KarrJR. 1994. A benthic index of
biotic integrity(B-IBI) for riversof the Tennessee Valley.Ecol. Appl. 4:768-85
Kerr JT, Ostrovsky M. 2003. From space to
species: ecological applicationsfor remote
sensing. TrendsEcol. Evol. 18:299-305
KingstonJC, Birks HJB, UutalaAJ, Cumming
BF, Smol JP. 1992. Assessing trendsin fishery resourcesandlake wateraluminumfrom
paleolimnological analyses of siliceous algae. Can. J. Fish. Aquat.Sci. 49:116-27
Kinnunen H, Holopainen T, KiirenlampiL.
2003. Sources of error in epiphytic lichen
variablesmappedas bioindicators:needs to
modify the Finnish standard.Ecol. Indic.
3:1-11
Klemm DJ, Blocksom KA, Fulk FA, Herlihy
AT,Hughes R, et al. 2003. Developmentand
evaluationof a macroinvertebratebiotic integrity index (MBII) for regionally assessing Mid-Atlantic Highlands streams. Environ. Manage. 31:656-69
Kolkwitz R, Marsson M. 1908. Okologie der
pflanzlichen Saprobien. Ber Dt. Bot. Ges.
26:505-19
KratzTK, MagnusonJJ, Bayley P, Benson BJ,
Barish CW, et al. 1995. Temporaland spatial variabilityas neglected ecosystem properties: lessons learned from 12 American
ecosystems. In Evaluatingand Monitoring
the Health of Large-ScaleEcosystems,ed. D

107

Rapport,P Calow, pp. 359-83. New York:

Springer-Verlag
Krebs CJ. 1978. Ecology: The Experimental Analysis of Distributionand Abundance.
New York:Harper& Row. 2nd ed.
KurtzJC,JacksonLE, FisherWS. 2001. Strategies for evaluatingindicatorsbased on guidelines from the Environmental Protection
Agency's Office of Research and Development. Ecol. Indic. 1:49-60
Lambeck RJ. 1997. Focal species: a multispecies umbrella for nature conservation.
Conserv.Biol. 11:849-56
LandresPB, VernerJ, ThomasJW. 1988. Ecological uses of vertebrateindicatorspecies: a
critique. Conserv.Biol. 2:1-13
Landres PB. 1992. Ecological indicators:
panacea or liability? See McKenzie et al.
1992, 2:1295-318
LarsenDP,UrquhartNS, KuglerDL. 1995. Regional scale trend monitoring of indicators
of trophic condition of lakes. WaterResour
Bull. 31:117-40
Lausch A, Herzog F. 2002. Applicability of
landscapemetricsforthe monitoringof landscape change:issues of scale, resolutionand
interpretability.Ecol. Indic. 2:3-15
Lawler JJ, White D, Sifneos JC, Master LL.
2003. Rare species and the use of indicator
groups for conservationplanning. Conserv.
Biol. 17:875-82
Lawn PA. 2003. A theoretical foundation to
supportthe Index of SustainableEconomic
Welfare(ISEW),GenuineProgressIndicator
(GPI), andotherrelatedindexes. Ecol. Econ.
44:105-18
LawtonJH,GastonKJ.2001. Indicatorspecies.
In EncyclopediaofBiodiversity,ed. S Levin,
3:437-50. San Diego: Academic
LindenmayerDB, Margules CR, Botkin DB.
2000. Indicatorsof biodiversityfor ecologically sustainableforest management.Conserv.Biol. 14:941-50
Lopez RD, Fennessy MS. 2002. Testing the
floristic quality assessment index as an indicator of wetland condition. Ecol. Appl.
12:487-97
MacNally R, Fleishman E. 2002. Using

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

108

NIEMI M MCDONALD

'indicator'species to model species richness:

model development and predictions. Ecol.
Appl. 12:79-92
Manley PN, Zielinski WJ, Schlesinger MD,
Mori SR. 2004. Evaluation of a multiplespecies approachto monitoringspecies at the
ecoregional scale. Ecol. Appl. 14:296-310
Mann CC, PlummerML. 1999. Call for "sustainability"in forests sparks a fire. Science
283:1996-98
Marwood CA, Solomon KR, GreenbergBM.
2001. Chlorophyllfluorescenceas a bioindicator of effects on growthin aquaticmacrophytes from mixtures of polycyclic aromatichydrocarbons.Environ.Toxicol.Chem.
20:890-98
May RM. 1985. Evolution of pesticide resistance. Nature 315:12-13
McDonald M, Blair R, Bolgrien D, Brown B,
Dlugosz J, et al. 2004. The Environmental
Protection Agency's EnvironmentalMonitoring and Assessment Program.In EnvironmentalMonitoring,ed. GB Wiersma,pp.
649-68. New York:CRC
McGeochMA. 1998. The selection,testing and
applicationof terrestrialinsects as bioindicators. Biol. Rev.73:181-201
McInnes PF, Naiman RJ, Pastor J, Cohen Y.
1992. Effects of moose browsingon vegetation andlitterof theborealforest,Isle Royale,
Michigan,USA. Ecology 73:2059-75
McKenzie DH, Hyatt DE, McDonald VJ, eds.
1992. Ecological Indicators,Vols. 1, 2. London/New York:Elsevier Applied Science
Milon JW, Shogren JF, eds. 1995. Integrating
Economic and Ecological Indicators.Westport, CT:Praeger.214 pp.
MorrisonML. 1986. Birdpopulationsas indicators of environmentalchange.In CurrentOrnithology, ed. RF Johnston,3:429-51. New
York:Plenum
National Research Council. 1990. Managing
Troubled Waters. Washington, DC: Natl.
Acad.
National Research Council. 1997. Building a
Foundationfor Sound EnvironmentalDecisions. Washington,DC: Natl. Acad.
National Research Council. 2000. Ecological

Indicatorsfor the Nation. Washington,DC:

Natl. Acad.
Nichols JW, Larsen CP, McDonald ME,
Niemi GJ, Ankley GT. 1995. Bioenergeticsbased model for accumulationof polychlorinatedbiphenylsby nestling tree swallows,
Tachycinetabicolor. Environ.Sci. Technol.
29:604-12
Niemeijer D. 2002. Developing indicators
for environmentalpolicy: data-drivenand
theory-drivenapproachesexamined by example. Environ.Sci. Pol. 5:91-103
Niemi GJ,HanowskiJM,LimaAR, Nicholls T,
WeilandN. 1997. A critical analysis on the
use of indicator species in management.J.
Wildl.Manage. 61:1240-52
Niemi GJ, WardropDH, Brooks RP, Anderson S, Brady VJ, et al. 2004. Rationale for
a new generationof ecological indicatorsfor
coastal waters.Environ.Health Perspect.In
press
Noss RF. 1990. Indicatorsfor monitoringbiodiversity:a hierarchicalapproach.Conserv.
Biol. 4:355-64
Noss RF. 1999. Assessing and monitoring
forest biodiversity:a suggested framework
and indicators.For Ecol. Manage. 115:135-

46
O'Connell TJ, Jackson LE, Brooks RP. 2000.
Birdguilds as indicatorsof ecological condition in the CentralAppalachians.Ecol. Appl.
10:1706-21
O'Connor RJ, Walls TE, Hughes RM. 2000.
Using multiple taxonomic groups to index
the ecological condition of lakes. Environ.
Monit.Assess. 61:207-28
Odum EP. 1963. Ecology. New York: Holt,
Rinehart,& Winston
Olsen AR, SedranskJ, EdwardsD, GotwayCA,
Liggett W, et al. 1999. Statisticalissues for
monitoringecological and naturalresources
in the United States.Environ.Monit.Assess.
54:1-45
O'Neill RV, KrummelJR, GardnerRH, Sugihara G, Jackson B, et al. 1988. Indices of
landscapepattern.Landsc.Ecol. 1:153-62
Ostrom E, Burger J, Field CB, Norgarrd
RB, Policansky D. 1999. Revisiting the

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

ECOLOGICALINDICATORS
commons: local lessons, global challenges.
Science 284:278-82
PaerlHW,ValdesLM, PinckneyJL,PiehlerMF,
Dyble J, MoisanderPH. 2003. Phytoplankton
photopigmentsas indicatorsof estuarineand
coastal eutrophication.BioScience 53:95364
Paine RT. 1969a. A note on trophiccomplexity
and community stability.Am. Nat. 103:9193
Paine RT. 1969b. The Pisaster-Tegulainteraction: prey patches,predatorfood preference,
and intertidalcommunitystructure.Ecology
50:950-61
ParsonsDJ. 2004. Supportingbasic ecological
researchin US nationalparks:challengesand
opportunities.Ecol. Appl. 14:5-13
Paul JF. 2003. Developing and applying an index of environmentalintegrity for the US
Mid-Atlantic region. J. Environ. Manage.
67:175-85
Pdrez-L6pezM, Alonso J, N6via-ValifiasMC,
Melgar MJ. 2003. Assessment of heavy
metal contaminationof seawaterand marine
limpet, Patella vulgata L., from Northwest
Spain. J. Environ.Sci. Health, Part A: Tox.
Hazard.Subst.Environ.Eng. 38:2845-56
PimentelD, WestraL, Noss RF,eds. 2000. Ecological Integrity: Integrating Environment,
Conservation,and Health. Washington,DC:
Island
RakocinskiCF, Brown SS, Gaston GR, Heard
RW,WalkerWW, SummersJK. 1997. Macrobenthic responses to natural and contaminant-relatedgradientsin northernGulf
of Mexico estuaries. Ecol. Appl. 7:127898
RapportDJ. 1992. Evolution of indicators of
ecosystem health.See McKenzieet al. 1992,
1:121-34
RapportD, CostanzaR, EpsteinPR, GaudetC,
Levins R. 1998. EcosystemHealth. Malden,
MA: Blackwell Sci. 72 pp.
RatcliffeDA. 1980. ThePeregrineFalcon. Calton, UK: Poyser
Reynoldson TB, Norris RH, Resh VH, Day
KE, Rosenberg DM. 1997. The reference
condition: a comparisonof multimetricand

109

multivariate approaches to assess waterquality impairmentusing benthic macroinvertebrates.J. North Am. Benthol. Soc. 16:
833-52
Roberge J, Angelstam P. 2004. Usefulness of
the umbrellaspecies concept as a conservation tool. Conserv.Biol. 18:76-85
Rolstad J, GjerdeI, GundersenVS, Setersdal
M. 2002. Use of indicatorspecies to assess
forest continuity: a critique. Conserv.Biol.
16:253-57
Root RB. 1967. The niche exploitationpattern
of the blue-gray gnatcatcher.Ecol. Monogr
37:317-50
Rubinoff D. 2001. Evaluating the California
gnatcatcheras an umbrella species for conservationof southernCaliforniacoastal sage
scrub.Conserv.Biol. 15:1374-83
Rublee PA, KemptonJW, Schaefer EF, Allen
C, Harris J, et al. 2001. Use of molecular
probes to assess geographic distributionof
Pfiesteria species. Environ.Health Perspect.
109:765-67
Rudi K, SkulbergOM, SkulbergR, Jakobsen
KS. 2000. Application of sequence-specific
labeled 16S rRNA gene oligonucleotide
probesfor geneticprofilingof cyanobacterial
abundanceand diversityby arrayhybridization. Appl. Environ.Microbiol.66:4004-11
Sauer JR, Hines JE, Fallon J. 2003. The North
AmericanBreedingBird Survey,Resultsand
Analysis 1966-2002. Version2003.1, USGS
PatuxentWildl. Res. Cent., Laurel,MD
Savidge JA. 1984. Guam: paradise lost for
wildlife. Biol. Conserv.30:305-17
SchillerA, HunsakerCT,KaneMA, Wolfe AK,
Dale VH, et al. 2001. Communicatingecological indicatorsto decision makersandthe
public. Conserv.Ecol. 5:19
Seegert G. 2001. The development, use, and
misuse of biocriteriawith an emphasis on
index of biotic integrity.Environ.Sci. Pol.
3:51-58
SemlitschRD. 2003. AmphibianConservation.
Washington,DC: SmithsonianInst.
ShannonCE,WeaverW. 1949. TheMathematical Theoryof Communication.Urbana:Univ.
Illinois Press

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

110

NIEMI MMCDONALD

Sierszen ME, Frost TM. 1990. Effects of lake ten Brink BJE, Hosper SH, Colijn F. 1991. A
acidification on zooplankton feeding rates
quantitativemethod for descriptionand assessment of ecosystems: the AMOEBA apand selectivity. Can. J. Fish. Aquat. Sci. 47:
772-79
proach.Mar Pollut.Bull. 23:265-70
SimberloffD. 1998. Flagships, umbrellas,and Tilman D. 1999. Diversity by default.Science
283:495-96
keystones: Is single-species management
Turner
era?
Biol.
Conserv.
in
the
MG, Hargrove WH, Gardner RH,
landscape
pass6
Romme WH. 1994. Effects of fire on land83:247-57
Simon TP, ed. 2003. Biological Response Sigscape heterogeneityin YellowstoneNational
natures: Indicator Patterns Using Aquatic
Park,Wyoming.J. Veg.Sci. 5:731-42
UnderwoodRJ, Roth RR. 2002. Demographic
Communities.Boca Raton,FL: CRC
variablesare poor indicatorsof wood thrush
Simon TP, Emery EB. 1995. Modificationand
assessment of an index of biotic integrity
productivity.Condor 104:92-102
to quantify water resource quality in Great UrquhartNS, Paulsen SG, Larsen DP. 1998.
Rivers.Regul.Rivers:Res. Manage. 11:283Monitoring for policy-relevant regional
trendsover time. Ecol. Appl. 8:246-57
98
Simon TP,JankowskiR, MorrisC. 2000. Mod- US Dep. Inter. 1980. Standardsfor the develification of an index of biotic integrity for
opment of habitat suitabilityindex models.
Ecol. Serv.ManualNo. 103. Div. Ecol. Serv.,
assessing vernal ponds and small palustrine
US Dep. Inter.Fish Wildl.Serv.,Washington,
wetlands using fish, crayfish, and amphibDC
ian assemblagesalong southernLakeMichiEPA. 1997. An ecological assessment
US
3:407Health
Manage.
gan.Aquat.Ecosyst.
of the United States Mid-Atlantic region.
18
SimonTP,StewartPM, RothrockPL. 2001. DeEPA/620/R-97/130,Washington,DC
velopmentof an index of biotic integrityfor US EPA. 1998. Conditionof the Mid-Atlantic
estuaries. EPA 600-R-98-147, Washington,
plant assemblages (P-IBI) in southernLake
DC
Health
Manage.
Michigan. Aquat. Ecosys.
US EPA. 2000a. Evaluationguidelinesfor eco4:293-309
Smeets E, WeteringsR. 1999. Environmental
logical indicators. EPA/620/R-99/005, ResearchTrianglePark,NC.109 pp. httpJi/www.
indicators: typology and overview. Tech.
Environ.
Eur.
25,
ecosystemindicators.org/wg/publication/Jack
Agency, Copenhagen,
Rep.
Den. http://reports.eea.eu.int:80/TEC25/en/ son_KurtzFisher.pdf
US EPA. 2000b. Mid-Atlantic highlands
tech_25_text.pdf
streams assessment. EPA/903/R-00/015,
Stockwell D, Peterson AT. 2003. Comparison
of resolution of methods used in mapping
Washington,DC. 74 pp. http://www.epa.gov/
from
maia/pdf/MAHAStreams.pdf
point-occurrence
biodiversitypatterns
US EPA. 2002a. Environmentalmonitoring
data.Ecol. Indic. 3:213-21
and assessment program research stratStolte KW,MangisDR. 1992. Identificationand
use of plant species as ecological indicators
egy. EPA 620/R-02/002, ResearchTriangle
of air pollution stress in nationalparkunits.
Park,NC. 78 pp. http://www.epa.gov/nheerl/
See McKenzie et al. 1992, 1:373-92
emap/files/emap_research_strategy.pdf
Suter GW II. 1993. A critique of ecosystem US EPA. 2002b. A SAB report: a framework for assessing and reporting on ecohealth concepts and indices. Environ. Toxicol. Chem. 12:1533-39
logical condition. EPA-SAB-EPEC-02-009,
SuterGW II, NortonSB, CormierSM. 2002. A
Washington, DC. 142 pp. http://www.epa.
of
obthe
causes
for
inferring
methodology
gov/sab/pdf/epec02009.pdf
served impairmentsin aquatic ecosystems. US EPA.2002c. Biological indicatorvariability
and streamprogramintegration:a Maryland
Environ.Toxicol.Chem.21:1101-11

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions

ECOLOGICALINDICATORS

111

case study. EPA/903/R-02/008, Washing- Venier LA, Pearce JE, McKee JE, McKenney
DW, Niemi GJ. 2004. Climate and satelliteton, DC. 92 pp. http://www.epa.gov/bio
indicators/pdf/biological-indicator_md_.pdf derived land cover for predicting breeding
US EPA. 2002d. Mid-Atlantic integrated asbird distributionin the GreatLakes basin. J.
sessment (MAIA) estuaries 1997-1998.
Biogeogr 31:315-31
VernerJ. 1984. The guild concept applied to
EPA/620/R-02-003, Washington,DC
US EPA. 2003a. EPA's draft report on the
management of bird populations. Environ.
environment-technical document. EPA/
Manage. 8:1-14
600-R-03-050, Washington, DC. 453 pp. WhittierTR, Paulsen SG, LarsenDP, Peterson
SA, HerlihyAT,KaufmannPR. 2002. Indicahttp://www.epa.gov/indicators
US EPA. 2003b. Developing biological inditorsof ecological stressandtheirextentin the
cators: lessons learned from Mid-Atlantic
populationof northeasternlakes: a regional
streams. EPA/903/R-03/003, Washington,
assessment.BioScience 52:235-47
DC. 52 pp. http://www.epa.gov/bioindica Wolter PT, White MA. 2002. Recent forest cover type transitions and landscape
tors/pdf/MAIAJessonslearnedbiology.pdf
US EPA. 2003c. Response of surface water
structuralchanges in northeastMinnesota.
Landsc.Ecol. 17:133-55
chemistry to Clean Air Act amendmentsof
1990. EPA/620/R-03/001,Washington,DC. WorldEconomic Forum.2002. 2002 Environ74 pp.
mental sustainability index. Annu. Meet.,
US EPA. 2004. Draft national coastal condiGlobal Leaders for TomorrowTask Force,
tion reportII. EPA 620/R-03/002, WashingWorldEcon. Forum,Yale Univ. Cent. Environ. Law Policy, and ColumbiaUniv. Cent.
ton, DC. 362 pp. http://www.epa.gov/owow/
oceans/nccr2/downloads.html
Int. Earth Sci. Inf. Netw., New York http://
Van Biema D. 1995. Prophetof poison. Time
www.ciesin.org/indicators/ESI/downloads.
145:26-33
html
VanHome B. 1983. Density as a misleadingin- Yoccoz NG, Nichols JD, Boulinier T. 2001.
dicatorof habitatquality.J. Wildl.Manage.
Monitoring of biological diversity in space
47:893-901
and time. TrendsEcol. Evol. 16:446-53

This content downloaded from 93.118.247.69 on Sat, 18 May 2013 05:42:23 AM

All use subject to JSTOR Terms and Conditions