USEPA biomonitoring and bioindicator concepts needed to evaluate the biological
integrity of aquatic systems
James M. Lazorchak, Brian H. Hill, Barbara S. Brown, Frank H. McCormick, Virginia Engle, David J. Lattier,
Mark J. Bagley, Michael B. Griffith, Anthony F. Maciorowski, and Greg P. Toth
Abstract
This chapter presents the current uses, concepts and anticipated future directions of bio-monitoring and bioindicators in the regulatory and research programs of the United StatesEnvironmental Protection Agency (USEPA). The chapter provides a historical look on howbiomonitoring and bioindicators evolved in the USEPA or its predecessor agencies from the 1960s – 1980s, then describes two current key biomonitoring and bioindicator programs,the USEPA Office of Research and Development’s Environmental Monitoring and Assess-ment Program (EMAP) and USEPA’s Office of Water’s Biocriteria Program. The remainderof the chapter is organized hierarchically beginning with concepts and monitoring approachesusing fish, macroinvertebrates, and periphyton assemblages, and functional ecosystemmeasures. The assemblage approaches are followed by current research and regulatory use of whole organism toxicity testing assessments for measuring contamination in aquaticenvironments and remediation assessment. The chapter includes existing and proposed activ-ities in the use of real-time biomonitoring to assess biological exposures to contaminants andother environmental changes. A new approach that uses small and large adult whole fish tissueas a bioindicator for assessing potential contaminant exposures to wildlife is presented,followed by a description of new research in molecular approaches to biomonitoring andbioindicators through measures of gene expression, use of microarrays and measures ofgenetic diversity.
1. Overview of USEPA’s current use of biomonitoring in regulatoryand research programs
This chapter presents the United States Environmental Protection Agency’s (USEPA)current uses, concepts and anticipated future directions of biomonitoring in regulatoryand research programs. The terms biomonitoring and bioindicator used in this chapterwill generally follow Markert et al. (1999): “Biomonitoring is a method of observing
the impact of external factors on ecosystems and their development over a period, orof ascertaining differences between one location and another. A biomonitor or bio-indicator is an organism (or a part of an organism or a community of organisms) thatcontains information on the quality of the environment (or part of the environment).A biomonitor, on the other hand, is an organism (or a part of an organism or a commu-nity of organisms) that contains information on the quantitative aspects of the qualityof the environment.” This chapter covers the USEPA use of (1) fish, macroinverte-brates, and periphyton surveys and assessments; (2) toxicity testing in laboratories andin the field; (3) fish tissue assessment; (4) microbial measures as functional ecosystemmeasures; (5) molecular measures; and (6) real-time biological monitoring as quanti-tative or qualitative measures of ecosystem health or ecosystem integrity. Figure 1depicts an exposure paradigm that we will generally follow. The diagram starts on the left with how exposure moves to effects, starting at the molecular level and moving up to the ecosystem level. On the far right of the diagram are the differentbiological measures that are currently used to measure the effect at the correspondinghierarchal level.
The USEPA or its predecessor agencies began in the 1960s using fish and macroin-vertebrate surveys to gather basic information on the existing fauna and to observechanges from year to year on large interstate navigable waters (Mason, et al., 1971).In these early years, information on water quality impacts in large rivers led to surveysin major waterways across the United States to document and quantify these impacts.Collection locations were usually selected upstream of major cities so that their faunal
832 J.M. Lazorchak et al.
Figure 1. Exposure Onset Hierarchy. Modified from presentation made at the 20th Annual Meeting ofthe Society of Environmental Toxicology by Mary Haasch.
3571 MARKERT3 23/10/02 9:46 am Page 832
characteristics would reflect environmental conditions rather than specific sources ofmunicipal or industrial pollution. In the late 1960s and early 1970s, the USEPA beganexamining interstate large rivers and smaller streams and testing effluents for toxicity.These studies were employed to (1) measure the toxicity of specific pollutants oreffluents to individual species or communities of aquatic organisms under naturalconditions; (2) detect violations of water quality standards; (3) evaluate the trophicstatus of waters; and (4) determine long term trends in water quality (Weber, 1973).USEPA field studies were mostly conducted by national enforcement programs andlaboratory studies were conducted principally by national research laboratories(Weber, 1973). USEPA laboratory studies were undertaken to measure the effects ofknown or potentially deleterious substances on aquatic organisms, to estimate “safe”concentrations, and to determine some basic environmental requirements, such astemperature, pH, and dissolved oxygen, using important and sensitive species ofaquatic organisms.
After the passage of the Federal Water Pollution Control Act Amendments of 1972[referred to hereafter as the CWA (Clean Water Act)], the focus of biomonitoring wasto collect information to assess the goal of restoring and maintaining the chemical,physical and biological integrity of the Nation’s waters. In the years that followed the passage of these amendments, there were a number of deferring opinions on what was meant by integrity. Some felt that integrity of water meant, “ the capabilityof supporting and maintaining a balanced, integrated, adaptive community of organ-isms having a composition and diversity comparable to that of the natural habitats ofthe region”(USEPA, 1977). Others defined integrity of water as “ the maintenance of the community structure and function characteristic of a particular locale or deemedsatisfactory to society” and “ Integrity as used is intended to convey a concept thatrefers to a condition in which the natural structure and function of ecosystems is main-tained” (USEPA, 1977). All these interpretations of integrity require some assessmentof the biological health of an aquatic system. There was a basic change in the USEPA’semphasis of achieving better water quality from one solely based on numerical waterquality standards for chemicals or physical conditions, to one that utilized a combi-nation of numerical standards and a technology-based approach (permit-drivenwastewater treatment technology) to achieving the integrity goal (USEPA, 1977). The CWA amendments of 1977 also directed the USEPA and States to collect bio-monitoring information for a number of purposes: (1) basic water monitoring forassessing the status of water quality conditions meeting chemical and biologicalcriteria and trend monitoring; (2) development of National Water Quality Criteria,which were to be used to set both chemical as well as biological water quality stan-dards, and (3) compliance monitoring of permit conditions for effluents or non-pointsources.
In the late 1980s, the USEPA began to restructure monitoring programs from onethat emphasized compliance monitoring of permit conditions, to one that emphasizedenvironmental results. The USEPA published qualitative and semi-quantitative bio-assessment protocols designed to provide basic aquatic life data for planning andmanagement purposes, such as screening, site ranking, and trend monitoring (USEPA,1989). These protocols were fundamental assessment techniques to generate basicinformation on ambient physical, chemical and biological conditions.
2. Biological monitoring and use of bioindicators in USEPA biocriteriaprogram
In the 1990s, the trend toward measuring environmental results was enhanced withpassage of the Governmental Performance and Results Act (GPRA) of 1993. UnderGPRA, the USEPA established environmental performance objectives based onecological outcomes, such as “By 2005, conserve and enhance the ecological healthof the nations waters and aquatic resources so that 75% of waters will support healthyaquatic communities.” Such objectives required the development of broader indicatorsfocused on assemblages of organisms and their supporting habitats, as well as “health”of individual organisms, i.e., shifting from organismal lethality to more subtle impacts.Therefore, bioindicators for monitoring and assessing environmental condition wouldbe evaluated against “expected” (or reference) sites in the natural environment, asopposed to control (non-dosed) samples in the laboratory. During this same time framethe USEPA encouraged States to first adopt narrative biological standards into Statewater quality standards (Gibson, 1991). Biological criteria can be used by States toconfirm impairment from known and unknown sources of impact, determine supportof designated aquatic life use classifications and provide a tool to expand monitoringand assessment programs expansion from source control to overall resource manage-ment (Yoder and Rankin, 1995). Some states use biological criteria to better delineateand protect aquatic life use classifications and in the enforcement of water quality stan-dards (Yoder and Rankin, 1995). Further refinement of narrative criteria into numericalcriteria or expectations of community structure and function in a least disturbed condi-tion (or reference condition) were considered a next logical progression (Gibson,1991). Patterns in community response to stress are then used to determine biologicalintegrity and ecological function (Karr and Dudley, 1981). Biological criteria, there-fore, supplement, rather than replace chemical and toxicological endpoints. They arebased on the premise that the structure and function of an aquatic community withina specific habitat provide important information about the quality of surface waters(USEPA, 1989). The USEPA provided additional guidelines and standardized proce-dures for using benthic macroinvertebrates and fish in developing biocriteria (USEPA,1990, 1993a). These procedures were more quantitative techniques for collecting,processing, and identifying specimens and also included taxonomic references.
3. Use of biomonitoring in risk assessments in the pesticide regulatoryprograms
Ecological risk assessment methods and procedures Federal Insecticide, Fungicide, andRodenticide Act (FIFRA) are detailed elsewhere (40 Code of Federal Regulations(CFR) 158.130; 40 CFR 158.145; Urban and Cook, 1986; SETAC, 1994; Touart and Maciorowski, 1997), and are briefly described here. Existing methods pre-dateEPA’s ecological risk assessment framework (USEPA, 1992) and guidelines (USEPA,1996). However, two pesticide case studies (carbofuran, synthetic pyrethroids) wereused in the Agency’s state-of-the-practice for ecological risk assessment preparedduring the guidelines development process (USEPA, 1993b). Generally, ecological risk
834 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 834
assessments for pesticide registration are prospective estimates based on single activeingredients and use-sites (e.g., corn, wheat, ornamental plants, etc.). The scope andcomplexity of pesticide risk assessments will vary with the specific chemical and itsuse, but a tiered, iterative approach is generally used. The tiers progress through simplerisk quotients derived from laboratory fate, transport, and toxicity data in lower tiers,to a weight-of-the-evidence approach in higher tiers (Tables 2 and 3).
Exposure analysis consists of a preliminary or comprehensive fate and transportassessment (Table 1) based on registrant submitted data. The exposure analysisprovides exposure profiles and estimated environmental concentrations (EEC) for thepesticide use (e.g. corn, cotton, wheat, etc.). Note that ECCs may be derived from fourestimation procedures ranging from simple to complex . The ecological effects analysis(Table 2) is also tiered. Tier I provides an acute toxicity profile for birds, fish,mammals, and invertebrates. Tier II provides a sub-chronic and chronic toxicity (No-Observed-Effect Concentration or NOEC) profile and bioaccumulation potential forthe same test species. Depending upon the hazard and exposure characteristics of aparticular pesticide and use pattern, Tier II analyses may be conducted for all repre-sentative taxa, or may focus on either aquatic or terrestrial species. When warranted,Tier III effects analysis is used to refine NOEC and bioaccumulation estimates.
Following exposure and effects analysis, ecological risk is estimated as a functionof ecotoxicological effects and environmental exposure using the quotient method(Table 2). A number of risk quotients are calculated (e.g., acute avian, acute fish, acuteinvertebrate, chronic avian, chronic fish, chronic invertebrate, etc.) and compared to regulatory risk criteria (e.g., presumption of acceptable risk, presumption of un-acceptable risk, etc.). Traditionally, if regulatory criteria were exceeded, a high riskpotential was assumed to exist for the pesticide-use combination. If registrants wishedto refute a presumption of risk finding, Tier IV effects analysis consisting of fieldstudies, simulated field studies, or other special studies could be conducted (Touart,1988; Fite et al., 1988). The types of ecological studies that have been required includequalitative avian mortality screening studies, pond studies, mesocosm studies, andmicrocosm studies.
Table 1. Sample assessment questions for the assessment of stream condition pertaining tostream fish assemblages.
What % of stream miles (and spatial distribution) have fish assemblages that differ from“reference” condition as measured by:
� Species richness?� Number of species sensitive to human disturbance?� Percentage of individuals tolerant of chemical or habitat disturbance?� Percentage of non-indigenous individuals?� Cumulative index of biotic integrity based on fish assemblage?
What % of stream miles support coldwater vs. warmwater fisheries as determined by the fishspecies?
(Modified from US EPA (1998a) and McCormick and Peck (2000).
3571 MARKERT3 23/10/02 9:46 am Page 835
Although field studies and simulated field studies represent one type of bio-monitoring, they are presently rarely used due to regulatory policy changes in 1992(SETAC, 1994). Such studies provided useful information for evaluating short-termimpacts of pesticides, but provided little information regarding longer-term impacts,latent effects, recovery of individuals and populations, or alterations of higher dimen-sional ecological phenomena. Further, many of the acute and chronic effects observedin field studies could be predicted from the acute and chronic laboratory data sed inrisk assessment.
836 J.M. Lazorchak et al.
Table 2. Generalized exposure analysis and assessment methods and procedures used inprospective ecological risk screens of pesticides.a
Preliminary Exposure Analysis includes simple laboratory tests and models to provide aninitial fate profile for pesticide (Hydrolysis and photolysis in soil and water, aerobic andanaerobic soil metabolism, and mobility).
Fate and Transport Assessment provides a comprehensive profile of the chemical(persistence, mobility, leachability, binding capacity, degradates) and may include fielddissipation studies, published literature, other field monitoring data; groundwater studies; and modeled surface water estimates.
Estimated Environmental Concentrations (EEC) are derived during the exposure analysis or comprehensive fate and transport assessment. There are four EEC estimationprocedures.
Level 1: A direct application, high exposure model designed to estimate direct exposure to a non-flowing, shallow water (<15 cm) system.
Level 2: Adds simple drift or runoff exposure variables such as drainage basin size, surface area of receiving water, average depth, pesticide solubility, surface runoff, or spray drift loss which attenuate the Level 1 direct application model estimate.
Level 3: Computer runoff and aquatic exposure simulation models. A loading model(SWRBB-WQb, PRZMc,etc.) is used to estimate field losses of pesticide associated with surface runoff and erosion, which then serves as a input to a partitioning model(EXAMS IId) to estimate sorbed and dissolved residue concentrations. Simulations arebased on either reference environment scenarios or environmental scenarios derived from typical pesticide use circumstances.
Level 4: Stochastic modeling where ECCS are expressed as exceedance probabilities for the environment, field, and cropping conditions.
a For additional details regarding environmental fate data requirements see 40 CFR §158.130, SETAC (1994); Touart (1995).
b Simulator for Water Resources in Rural Basins–Water Quality.c Pesticide Root Zone Model.d Exposure Analysis Modeling System.
3571 MARKERT3 23/10/02 9:46 am Page 836
4. Biological monitoring and use of bioindicators in the USEPAEnvironmental Monitoring and Assessment Program (EMAP) surfacewaters and estuarine program
In 1989, the USEPA initiated the Environmental Monitoring and Assessment Program(EMAP), an integrated multi-resource program designed to develop methods toestimate the condition of the Nation’s ecological resources at various geographic scales
Table 3. Generalized ecological effects analysis and risk quotient methods and procedures usedin prospective risk screens of pesticides.
Tier I Effects Analysis provides acute toxicity values and dose-response information(mammalian and avian acute oral LD50; avian dietary LC50; seedling emergence andvegetative vigor EC25; honey bee acute contact LD50; and additional wild mammal, estuarine,and plant tests depending on pesticide use category.
Tier II Effects Analysis provides sub-chronic and chronic toxicity values (NOEC) includingavian reproduction studies; special avian or mammal studies; fish early life stage studies;invertebrate life cycle studies; and a fish bioacumulation factor.
Tier III Effects Analysis provides refined NOEC estimates for chronic toxicity that mayinclude a fish full life cycle test, aquatic organism accumulation, or food chain transfer tests
The Quotient Method is used to provide a set of acute and chronic risk quotients (RQ) forfish, birds, invertebrates, plants and endangered species. The RQs are calculated by dividingexposure (EEC) by hazard (LD50 or LC50 or NOEC). Risk quotients are then compared toregulatory risk criteria as follows.
Presumption of Presumption of risk Presumption of unacceptable riskacceptable risk that may be mitigated
by restricted use Nonendangered Endangered species
Tier IV Effects Analysis allows registrants to rebut a presumption of risk derived fromlaboratory studies by performing field or simulated field studies, including qualitativeterrestrial field studies, farm pond studies, mesocosm studies, or other special studies.
a For additional details regarding ecological effects data requirements see 40 CFR §158.145 Subdivision E;Urban and Cook (1986); SETAC (1994); Touart (1995).
3571 MARKERT3 23/10/02 9:46 am Page 837
over long periods of time. (Messer et al., 1991). Within EMAP, an indicator wasdefined as any environmental measurement that can be used to quantitatively estimatethe condition of ecological resources, the magnitude of stress, the exposure of biolog-ical components to stress, or the amount of change in condition. “Condition indicators”provided quantitative information on the state of ecological resources of interest, asubset of which were biotic indicators which estimated the condition of a biologicalcomponent of a resource (Barber, 1994). Such indicators were specifically associatedwith previously identified environmental values of interest and the assessmentendpoints that represented those environmental values.
For freshwater systems, EMAP initiated its first effort in the Mid-Atlantic Highlandsecoregion, (Whittier and Paulsen, 1992; USEPA, 1997). The study was identified asthe Mid-Atlantic Highlands Assessment (MAHA) and was conducted to develop anddemonstrate EMAP approaches such as probability-based survey designs and appro-priate indicators of ecological condition as applied to address specific regionalassessment questions of interest to the USEPA. The monitoring framework for MAHAused a regional-scale probability-based survey design to select sampling sites. Thisdesign permits unbiased inferences the subset of sites where samples and data arecollected with known certainty to explicitly defined populations of ecological resourceunits (Larsen, 1995, 1997; Diaz-Ramos et al., 1996). For MAHA, populations weredefined based on the total length of streams. For example, the design allows one toestimate the total length of streams in the target population (e.g., all permanent streamsappearing on a particular scale of map) which meet some criteria (e.g., all first-ordertarget streams, all target streams within a specific ecoregion, etc.). The distribution ofindicator scores can then be examined for these defined populations to determine theestimated length of stream characterized by a particular set of indicator values, withassociated uncertainty in these estimates represented by confidence bounds.
A similar study was undertaken in 1997 and 1998 in the Mid-Atlantic Estuaries,spanning the Delaware and Chesapeake Estuaries and the Atlantic coastal bays. Thepurpose of the study was to fill information gaps identified during preparation of theCondition of the Mid Atlantic Estuaries report (USEPA, 1998a) and to demonstratehow to integrate different institutional monitoring programs. The study integratedmonitoring programs from the Chesapeake Bay Program, the Delaware River BasinCommission, the National Oceanic and Atmospheric Administration, and EMAP intoa compatible design using a stratified probability-based design and a core suite of indi-cators which were measured by all programs (USEPA, 1998a). The core suite ofindicators included measures to characterize habitat (e.g., salinity, temperature, grainsize), stressors (e.g., toxics, nutrients, dissolved oxygen), and biological response (e.g.,benthic community, fish community, chlorophyll a). Consistent methods and samplingdesign over the programs allowed the results across programs to be aggregated to gaina better understanding of the regional scale condition. For the estuarine component ofthe program, the biotic indicators selected for use or development included benthicspecies composition and biomass, fish community composition, contaminants in fishflesh and shellfish, the gross pathology and histopathology of fish, and (for linkagebetween biology and stressor) sediment chemistry and toxicity (Holland, 1990). EMAPbuilt on the pioneering work of Karr (1981) and others to develop multimetric indicesof biotic integrity. Such indices would represent the response of biological communi-
838 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 838
ties to environmental stressors by not only quantifying the current condition of theecosystem but also by integrating the effects of multiple anthropogenic and naturalstressors over time.
5. Community and ecosystem measures
5.1. Fish monitoring and bioindicators (freshwater)
In support of the biocriteria program, USEPA has conducted research on the devel-opment fish biotic indices and evaluated various components of such indices. Fishspecies exhibit diverse evolutionary, morphological, ecological, and behavioral adap-tations to their natural habitat and thus are particularly effective indicators of thecondition of aquatic systems (Karr et al., 1986; Fausch et al., 1990; Simon and Lyons,1995). The biological characteristics of stream fish assemblages, including the capa-bility to integrate the effects of a variety of stressors across different time scales andlevels of ecological organization, and the importance and familiarity of fishes to thegeneral public, make them conducive to the development of an indicator of ecologicalcondition (Karr et al., 1986; Simon, 1991; Simon and Lyons, 1995, and US EPA,1999).
5.1. Multimetric approaches
Multimetric indicators such as the Index of Biotic Integrity (IBI) represent a means tointegrate various structural and functional attributes of an ecosystem and provide anoverall assessment of ecosystem condition (Fausch et al., 1990; Karr, 1991; Karr andChu, 1997). Structural and functional attributes of the fish assemblage (derived fromspecies presence/absence and relative abundance data) are aggregated into metriccategories (taxonomic composition, abundance and individual condition, trophic, andreproductive function) that are hypothesized to respond predictably to increasing inten-sities of human disturbance (Karr et al., 1986; Karr, 1991, Barbour et al., 1995; Hughesand Oberdorff, 1999). Candidate metrics are tested for responsiveness to biotic orabiotic conditions resulting from increasing human disturbance, and their biologicalimportance (Hughes et al., 1998; McCormick and Peck, 2000; McCormick et al.,2001). The IBI was originally developed with 12 metrics (Karr, 1981), but IBI’s havesubsequently been developed with fewer and more metrics (Hughes et al., 1998;Halliwell et al.,1999; Moyle and Marchetti, 1999; McCormick et al., 2001). Responsevalues for each metric selected are transformed to a metric score based on the degreeof deviation of the response value from that expected at a similar site under condi-tions of minimal human disturbance. The individual metric scores are then aggregatedto produce a multimetric index score in which a higher score indicates better ecolog-ical condition (i.e., closer to the expected condition when human disturbance isminimal). Possible causes of poor ecological condition may be identified (althoughspecific cause-effect relationships cannot always be ascertained) by examining corre-lations between the index or its component metrics and various measures of ecosystemstress. More detailed descriptions of the general approach used to develop multimetric
indices can be found in Hughes et al. (1998), US EPA (1999), McCormick and Peck(2000), and McCormick et al. (2001).
5.1. Indicator development
Several factors should be considered in the indicator development for use in biomon-itoring and biocriteria programs (Yoder and Rankin, 1995; McCormick and Peck,2000).
5.1.2.1. Conceptual relevance of the indicatorThe design of monitoring studies should be driven in part by a series of specific assess-ment questions related to the condition of stream resources. The indicator should belinked to identified assessment questions, should contribute information to addressmultiple assessment questions, and should complement other potential indicators(Table 1). The nature of the question suggests that an appropriate indicator would focusat the assemblage level and consist of multiple components to address the variousaspects of the questions. The indicator is also useful in that the basic fish species andabundance data used to develop it can also be used with little or no additional effortto address other assessment questions of interest. These subsidiary questions are rele-vant to a separate societal value of interest to the MAHA study, fishery health.McCormick and Peck (2000) graphically represented conceptual relationships betweenmajor structural components and processes to illustrate possible routes of exposurefrom anthropogenic stressors (Fig. 2).
5.1.2.2. Feasibility of implementationCollection of field data at an individual sampling site is based on standard approachesfor stream fish assemblages (Ohio EPA, 1987; Lyons, 1992; McCormick, 1993). Fishassemblage sampling is conducted using a combination of gear types (electrofishingand seining), standardized sampling times and distances (Ohio EPA, 1997; McCormickand Hughes, 1998; US EPA, 1999). McCormick and Peck (2000) presented the resultsof a pilot study on wadeable streams in the Interior Highlands and Central Lowlandsthat showed that 90% of the species in a reach were sampled by single-pass, backpackelectrofishing over a distance equal to 40 times the mean channel width (Fig. 4). EMAPdocumented protocols for sampling wadeable and non-wadeable streams that weredeveloped in the Mid-Atlantic region of the United States but have been used, withsome modifications, in the Southern Rocky Mountains, California’s Central Valley,Interior Highlands and the Great Plains (Lazorchak et al., 1998). An appropriate qualityassurance program can be developed and implemented for the indicator and monitoringframework using available resources and techniques (e.g., Chaloud and Peck, 1994).
5.1.2.3. Sources of errorMcCormick and Peck (2000) addressed the different types of errors that can affecteither the measurement data or the development of indicator values from measure-ment data. Measurement-related errors of field collection data, in terms of number of species collected, species composition, and number of individuals, cannot be estimateddirectly for the indicator by collecting replicate samples during a single visit to a site
(Fore et al., 1996) but must be addressed by professional ichthyologists. The other crit-ical source of error in measurement data is incorrect identifications of fish species.Various means of controlling this source of error include the collection and confirma-tion of voucher specimens, using personnel experienced in fish identification andadditional training in field identification of regional fishes. McCormick and Peck(2000) presented a more quantitative evaluation of five types of errors related to fieldidentification of fish species. Transcription errors occur when the wrong species (orspecies code) is recorded on the field data form. The remaining four relate to actualerrors in species identification and include a cumulative estimate of errors for all
842 J.M. Lazorchak et al.
1991
Gower General Similarity Coefficient
E01AE01E03E12AE13E13BE20E22E05E10E12
0.82 0.85 0.88 0.91 0.94 0.97 1
1992
Gower General Similarity Coefficient
E01AE01E12AE03E12E05E10E11E13BE13E20E22
0.85 0.875 0.9 0.925 0.95 0.975 1
Figure 3. Community similarity dendrograms (based on for Eagle River periphyton assemblages in 1991and 1992.
3571 MARKERT3 23/10/02 9:46 am Page 842
species, errors specific to groups of fishes that are difficult to identify to the specieslevel in the field (e.g., sculpins, genus Cottus, and a cyprinid genus Nocomis), anderrors at the genus level.
5.1.2.4. Sources of varianceIt is important to identify the components and magnitude of variance that affect theability of the indicator to detect differences in condition among sites. Among-site vari-ance is variation due to differences in the indicator value among a sample of streamsites. This component represents the environmental “signal” to be detected and inter-preted with respect to an ecological condition. Extraneous variance consists of theremaining temporal and measurement-related variation. Collectively, these componentsrepresent “noise” that inhibit the ability to detect and interpret the environmental“signal” and include the extent to which regional-scale effects (e.g., climate, hydrol-ogy) and temporal variance affect the ability to detect differences among sites.
Figure 4. Regression of PO42� concentrations against downstream distance for old growth (top) and
harvested (bottom) watersheds in northwestern California.
3571 MARKERT3 23/10/02 9:46 am Page 843
5.1.2.5. Geographic factorsEnvironmental assessment is potentially affected by selection of appropriategeographic and temporal scales (McCormick et al., 2000, 2001). Multi-metric indica-tors developed for a particular geographic area and scale of monitoring effort shouldnot be applied to other scales of monitoring or other geographic areas without evalu-ation and modification. Within a biogeographic province, spatial and temporalvariability is relatively low. However, some seasonal variability associated withspawning activity may confound assessments if no consistent index period (a specifictime frame for sampling, i.e April 15–May 30 for spring low flow or July 1–September1 for summer low flow) is selected for sampling. Numbers of species vary with ecore-gion, drainage basin, and watershed size. Understanding the patterns of geographicvariation in the structure of fish assemblages is crucial to developing a comprehensiveassessment of stream conditions. Understanding the influence of geographic factors instructuring fish assemblages is crucial to developing a comprehensive assessment ofstream conditions. The variability in responses at different spatial and temporal scalesmay affect the interpretation of bioassessment endpoints and has important implica-tions for large-scale monitoring programs. Establishing reference conditions forMAHA streams required identifying the local factors (e.g., stream size, gradient,temperature, substrate composition, and habitat complexity) that control fish assem-blage structure in minimally disturbed streams. McCormick et al. (2000; 2001) foundno substantial differences in the range or general distribution of fish assemblageresponse values across ecoregions.
5.2. Macroinvertebrate monitoring and bioindicators (freshwater)
Although Karr’s (1981) IBI was originally developed for fish assemblages, the utilityof macroinvertebrate assemblage structure for describing the integrity of aquaticecosystems has been widely recognized (Kerans and Karr, 1994; DeShon, 1995;Barbour et al., 1996; Fore et al., 1996). Their role in aquatic food webs as primaryconsumers of producers (i.e., periphyton) and decomposers (i.e., heterotrophic bacteriaand fungi) and as prey for secondary and tertiary consumers (i.e., fish) make macroin-vertebrates important to the community’s total integrity. As a result, measurements ofmacroinvertebrate assemblages have been an integral part of monitoring biologicalconditions of streams and lakes both in the Environmental Monitoring and AssessmentProgram (EMAP) Mid-Atlantic pilot study, in regional EMAP (R-EMAP) studies, andin state monitoring programs.
In the multimetric approach, assemblage structure is summarized with simplenumerical measures of an assemblage’s attributes called metrics. To create an index,selected metrics are calculated and scored using a standardized scale (i.e., a continuousscale from 1–10 or a discrete scale of 1, 3, or 5). Then, the scores are summed. Manymetrics have been proposed for use in macroinvertebrate IBIs that measure differentcategories of assemblage attributes (USEPA, 1999). Richness measures include thetotal number of species or genera or the number of species or genera in selectedtaxonomic groups, such as Ephemeroptera, Plecoptera, and Trichoptera (EPT) orChironomidae (i.e., midges). Evenness measures assess the other component of diver-sity, the relative dominance of the most abundant taxa in the assemblage. Composition
844 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 844
measures assess the abundance of a taxa group, such as EPT or Chironomidae, relativeto total macroinvertebrate abundance or the abundance of a tribe, subfamily, or familyrelative to its larger taxonomic grouping (ex., % of tanytarisinid midges to Chiron-omidae or % of Hydropsychidae to Trichoptera). Tolerance measures assess taxa rich-ness or relative abundance of macroinvertebrates considered tolerant or intolerant ofdifferent environmental stressors or the abundance weighted average tolerance of themacroinvertebrate assemblage to a stressor gradient (ex., Hilsenhoff’s Biotic Index for organic pollution). Feeding, habitat, and life cycle measures assess taxa richness or relative abundances of macroinvertebrates with specific functional feeding (i.e.,scrapers, shredders, predators), habitat use (i.e., clingers, burrowers), or life cycle (i.e.,univoltine, semivoltine) adaptations to the habitat template of the ecosystem(Southwood, 1977; Townsend and Hildrew, 1994). Similarity measures assess compo-sitional resemblance of the macroinvertebrate assemblage to that expected underreference conditions for a region. However, similarity measures have been underuti-lized because they require identification of an expected assemblage by comparison withdata from a reference site or predicted by modeling.
One major objective of a bioassessment is diagnosis of the anthropogenic stressorsat a site. Selection of metrics for incorporation into macroinvertebrate indices of bioticintegrity has been largely based on general observations on the response of assem-blages to increasing perturbation (Kerans and Karr, 1994; Fore et al., 1996; USEPA,1999). However, few metrics have been tested for their relation to specific stressorgradients (Wallace et al., 1996; Carlisle and Clements, 1999). Testing is needed toapply the multimetric approach to diagnosis of the causes of decreased biotic integrityat individual sites (Griffith et al., 2001a).
The development of field sampling designs that employ multiple reference andpolluted sites has been proposed as an alternative to the traditional upstream versusdownstream approach used in most biomonitoring studies (Clements et al., 2000).Spatially-extensive monitoring programs can characterize ecological conditions withinan ecoregion and provide the necessary background information to evaluate futurechanges in water quality. Clements, et al. (2000) used physicochemical characteristics,heavy metal concentrations, and benthic macroinvertebrate community structure datafrom 95 sites in the Southern Rocky Mountain ecoregion in Colorado collected in1995–1996 as part of the USEPA R-EMAP program. Most sites (82%) were selectedusing a systematic, randomized sampling design. Each site was placed into one of fourmetal categories (background, low, medium, and high metals), based on the cumula-tive criterion unit (CCU), which was defined as the ratio of the instream metalconcentration to the USEPA criterion concentration, summed for all metals measured.A CCU of 1.0 represents a conservative estimate of the total metal concentration that,when exceeded, is likely to cause harm to aquatic organisms. Although the CCU wasless than 2.0 at most (66.3%) sites, values exceeded 10.0 at 13 highly polluted stations.Differences among metal categories were highly significant for most measures ofmacroinvertebrate abundance and all measures of species richness. Clements et al.(2000) observed the greatest effects on several species of heptageniid mayflies(Ephemeroptera: Heptageniidae), which were highly sensitive to heavy metals andwere reduced by >75% at moderately polluted stations. The influence of taxonomicaggregation on responses to metals was also greatest for mayflies. In general, total
abundance of mayflies and abundance of heptageniids were better indicators of metalpollution than abundance of dominant mayfly taxa. Heavy metal concentration was themost important predictor of benthic community structure at these sites. Because of theubiquitous distribution of heavy metal pollution in the Southern Rocky Mountainecoregion, we concluded that potential effects of heavy metals should be consideredwhen investigating large scale spatial patterns of benthic macroinvertebrate commu-nities in Colorado’s mountain streams.
Griffith et al. (2001a) conducted multivariate analyses of R-EMAP data from thesame mineralized belt of the Southern Rockies Ecoregion in Colorado and the CentralValley Ecoregion in California have suggested that various metrics respond differentlyto environmental stressor gradients. Richness and evenness measures were correlatedwith dissolved and sediment metal concentrations in Rocky Mountain streams vari-ously affected by metal mining. Richness, evenness, composition, and feedingmeasures were generally uncorrelated with alterations in dominant taxa related toincreased dissolved cation and anion concentrations associated with irrigation runoffin Central Valley streams (Griffith et al., 2001b), but specifically designed toleranceor similarity measures are likely to be more sensitive to this chemical gradient.Richness and composition measures, particularly for EPT taxa, were correlated withalterations in substrates, in-stream habitats, and riparian structure and shading associ-ated with agriculture (i.e., livestock grazing in Colorado or irrigated row crops inCalifornia) in both ecoregions (Griffith et al., 2001a, 2001b). These differences illus-trate the potential to create indices of biotic integrity composed of diagnostic metricsfor specific stressor gradients.
5.3. Macroinvertebrate monitoring and bioindicators (marine)
Benthic indices of environmental condition were developed, tested, and validated for each of the biogeographic regions that were defined for the EMAP estuarinemonitoring program (Table 4). The EMAP approach to development of multimetric
846 J.M. Lazorchak et al.
Table 4. References for benthic indices of estuarine condition that were used by EMAP.
Biogeographic province Geographic range References
Virginian Province Cape Cod, MA to Weisberg et al. (1993)Chesapeake Bay, VA Schimmel et al. (1994)
Paul et al. (1999)
Carolinian Province Cape Henry, VA to Hyland et al. (1996)St. Lucie Inlet, FL Hyland et al. (1998)
West Indian Province Indian River, FL to Macauley et al. (2002)Tampa Bay, FL
Louisianian Province Anclote Key, FL to Summers et al. (1993)Rio Grande, TX Engle et al. (1994)
Engle and Summers (1999)Macauley et al. (1999)
3571 MARKERT3 23/10/02 9:46 am Page 846
benthic indices in estuaries was loosely adapted from Karr’s (1981) IBI approach. Aset of sites were identified as having known condition with regard to stressors thatwould potentially elicit a response in the benthic macroinvertebrate community.Although the critical levels of the stressors varied among provinces, typically, dis-solved oxygen, sediment toxicity, and sediment contaminants were used to identifyreference and degraded sites. In addition, sites were chosen to represent the range ofnatural conditions (e.g., salinity, sediment type) found within a province. A suite ofbenthic community components were then chosen to represent the range of potentialresponses to stressors. The list typically included measures of abundance, biomass, anddiversity as well as proportional abundances of taxonomic, trophic, or functionalgroups. Various multivariate methods such as discriminant and multiple regressionanalysis were used to identify a subset of components from this list that best discrim-inated between the reference and degraded sites. An index was then created as a linearcombination of the subset of components weighted by their proportional contributionto the multivariate model. The benthic index was used to classify sites of unknowncondition. Using the EMAP design and analysis procedures, the proportion of estu-arine area with reference or degraded benthic condition was calculated for eachbiogeographical province.
The benthic index for the Virginian Province comprised Gleason’s D based uponinfauna and epifauna (normalized for salinity), abundance of tubificid oligochaetes(normalized for salinity), and abundance of spionid polychaetes. The overall efficiencyfor correct classification using this index was 86% for both reference and degradedsites. A four-year assessment of benthic condition using this index indicated that 25 ± 3% of the estuarine area in the Virginian Province was impacted (Paul et al.,1999). A benthic IBI for the Carolinian Province included the following metrics: (1)mean abundance, (2) mean number of taxa, (3) 100 – % abundance of the top twonumerical dominants, and 4) % abundance of pollution-sensitive taxa. This indexcorrectly classified 93% of the development sites and 75% of the validation sites. In1995, 21% of the estuarine area of the Carolinian Province was classified as degradedusing this index (Hyland et al., 1998). The West Indian Province was only sampled in1995; therefore, the benthic index that was developed is preliminary and no validationhas occurred. The benthic index was composed of the abundance of gastropods andall molluscs, total abundance of all organisms, and the proportion of polychaetes thatwere spionids. Using this index, 33±11% of the estuarine area was classified asdegraded (Macauley et al., 2002). In the Louisianian Province the benthic index repre-sented a linear combination of five metrics: proportion of expected diversity(Shannon-Wiener H′ normalized for salinity), mean abundance of tubificidoligochaetes, and the proportional abundances of capitellid polychaetes, bivalves, andamphipods. The average classification efficiency for this index was 74% for degradedsites and 77% for reference sites. Degraded benthic condition occurred in 23 ± 7% ofthe estuarine resource in the Louisianian Province using this index (Macauley et al.,1999).
The need is increasing for biological indicators that are diagnostic for multiple,combined, and often unidentified stressors. EMAP has evolved into Coastal 2000, aprogram designed to transfer technology to the States to assist with monitoring designand indicator development. This technology transfer will enable the States to improve
their reporting capability for 305(b) and 303(d) Clean Water Act requirements. In addi-tion, EPA, through the STAR program, is funding research on the development oftools to evaluate the health of estuaries and the Great Lakes with a particular emphasison diagnostic indicators that are applicable over large geographic scales.
5.2. Anticipated marine biomonitoring research activities
One innovative measure to assess the health and integrity of coastal ecosystems is toidentify and count the bottom-dwelling organisms living in the sediment. Theseanimals, which are a major food source for many fish, create intricate tubes and tunnelsin the sediment to depths as much as three feet. A healthy sediment is characterizedby a high degree of tube and tunnel formation and, by contrast, an impacted sedimenthas fewer large, deep burrowing animal and their associated tubes and tunnels.Traditionally, sediment health is determined by collecting, identifying and countingthese organisms, but this procedure requires specialized training and is labor-intensiveand time-consuming. Computer Axial Tomography (CAT) imaging offers a rapid cost-effective alternative to this traditional method by quantifying the burrows and tunnelsin sediment cores. Scientists first collect intact mud cores from an estuary, using cylin-drical plastic tubes pushed into the sediment. The cores are tightly sealed at the topand bottom and transported to a hospital for CAT imaging. The resulting image dataare stored on magnetic tape and may be analyzed on a personal computer back at thelaboratory. A three-dimensional image of tubes and tunnels within the core can bequantified, and these measures can be used to identify, monitor and assess the effectsof human activities on sediment habitats. Because medical CAT imaging scanners arelocated throughout the world, this technique could be widely available for environ-mental managers to evaluate the health of sediments.
5.3. Measures of periphyton assemblage structure and ecosystem function
Efforts to use periphyton assemblage structure and ecosystems functions for thebiological monitoring of aquatic ecosystems fall into two broad categories: measuresof assemblage structure (taxa richness and diversity, assemblage similarity, the rela-tive abundances of indicator taxa, chlorophyll and biomass) and measures ofcommunity function, which can be further divided into organismal-level measures(cellular integrity, growth, photosynthesis, cellular respiration, enzyme activity) andcommunity-level measures (primary productivity, community respiration, nutrientuptake). Researchers from the USEPA have used measures of periphyton assemblagestructure and ecosystem function to monitor biological condition of streams and lakesunder two programs: Superfund and the Environmental Monitoring and AssessmentProgram (EMAP). The USEPA researchers measured structural and functional res-ponses of stream communities to elevated heavy metals related to mining activities inour Superfund assessments of the Eagle River, Colorado. USEPA also measured peri-phyton assemblage structure and ecosystem function in EMAP and regional EMAP(R-EMAP) studies of the ecological conditions of streams in the Appalachians, andused diatom assemblage structure in the assessment of ecological conditions in NewJersey lakes.
848 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 848
5.3. Structural measures of periphyton assemblages
Measurement and analysis of assemblage structure is the mainstay of biologicalmonitoring programs. Assemblage structure may be measured as lists of the totalnumber of species present within an assemblage, abundance of indicator species, or asan aggregate index derived from other attributes of the assemblage structure.
5.3.1.1. Species richness and diversity and assemblage similaritySpecies diversity has two components: species richness and evenness. Several studieshave used species richness to monitor stream assemblage responses to disturbance. Itis generally assumed that richness is inversely related to environmental stressors, andseveral researchers have documented decreases in diatom species richness as a resultof stream contamination by organic enrichment, metals, and pesticides (Lange-Bertalot, 1979; Crossey and La Point, 1988; Whitton et al., 1991). USEPA’s work insupport of Superfund and EMAP found poor correlations of species richness withhuman disturbance gradients (Hill et al., 2000a, b; 2001). Species better adapted to theprevailing environmental conditions will have an advantage resulting in an unevendistribution of individuals among taxa. Evenness is often reported as % dominance ofthe assemblage by single species, and results from studies employing diatom assem-blages have indicated that dominance increases with nutrient enrichment (Stevensonand Pan, 1999) and metal contamination (Crossey and LaPoint,1988). Superfund andEMAP studies have found similar nutrient and metal relationships (Hill et al., 2000a,2001), as well as correlations with watershed land uses (Hill et al., 2000b; Hill andKurtenbach, 2001). Assemblage similarity, the degree of compositional agreementamong the species in two or more assemblages along an environmental gradient, areparticularly suited for identifying changes in assemblage structure relative to thedistance from the source of perturbation, and may be more sensitive to low level stressthan are diversity indices (Hellawell, 1977). Medley and Clements (1998) reported thatassemblage similarity was better related to metal concentrations in the stream thanwere diversity indices. Hill et al. (2000b) found similar results in Superfund studieson the Eagle River, Colorado.
5.3.1.2. Relative abundance of indicator speciesIndicator species are used to assess current or recent environmental conditions. Therelative abundances of indicator taxa provides a measure of not only how well thosetaxa tolerate existing environmental conditions, but also provides an indirect measureof those environmental conditions. Shifts in the relative abundances of diatom taxahave been used for monitoring aquatic ecosystem contamination by heavy metals(Medley and Clements, 1998) and watershed land use changes (Kutka and Richards,1996). USEPA used diatoms as indicators of these stressors in Superfund assessmentsof the Eagle River, Colorado (Hill et al., 2000b), and in regional assessments of streamand lake water quality (Pan et al., 1996; Hill et al., 2000a, 2001; Hill and Kurtenbach,2001). Figure 3 is an example on how community similarity dendrograms were usedto analyze Eagle River periphyton assemblages in 1991 and 1992.
5.3.1.3. ChlorophyllChlorophyll a concentration has been widely used to assess nutrient enrichment ofstreams, even in regional-scale studies (Leland, 1995; Pan et al., 1999, 2000). USEPAdid not, however, find a significant relationship between chlorophyll a content of peri-phyton and stream chemistry, habitat, or watershed land use in our EMAP studies ofAppalachian streams (Hill et al., 2000a, 2001).
5.3.1.4. BiomassOne of the simplest measures of aquatic plant assemblage structure is standing cropof biomass. The relationship between standing crop and water quality, however, is noteasily interpreted. Clark et al. (1979) compared methods of estimating periphytonbiomass in response to chemical perturbations in stream mesocosms, and found thatno one method was consistently better than any of the others in detecting the impactof copper, chromium, and chloride contaminations, but found that biomass colonizingclean substrates was depressed in response to these contaminants. Hill et al. (2000b)reported no significant effects of heavy metals on periphyton biomass in theirSuperfund assessment of the Eagle River, Colorado. They found only weak correla-tions between periphyton biomass and stream chemistry, channel substrate size, andwatershed land use in our EMAP studies of Appalachian streams (Hill et al., 2000a,2001).
5.3.1.5. Cellular integrityMeasures of cellular integrity fall into two broad categories, those related to morpho-logical changes in cell structure and those related to changes in cell membranepermeability. Few researchers have used changes in cellular structure to monitor phys-iological condition of the cells or to predict water quality. Analysis of five species offossil diatoms collected from Mono Lake, California, revealed a large percentage ofdeformed individuals, possibly related to the transition of this lake from freshwater toalkaline, brackish waters (Solladay, 1994). USEPA reported significantly increasednumbers of deformed Fragilaria frustules with increasing dissolved metal concentra-tions in our Superfund assessment of the Eagle River, Colorado (McFarland et al.,1997).
5.3.2. Functional measures of plant assemblages
Recent research has been critical of the reliance on structural measures of biotic condi-tions to assess aquatic ecosystem integrity. From an ecosystem managementperspective structural measures are proving to be less reliable than previously thought.It has been argued that the high spatial and temporal variability exhibited by bioticassemblages preclude the use of population data alone as indicators of anthropogenicdisturbances, and resource managers are urged to exercise caution in the use these data.Ecosystem research over the past several years has increasingly focused on functionalparameters rather than the more traditional structural metrics. Emphasis on system-level functional roles may not answer population-level questions, but it permitsclustering of genetically and taxonomically diverse groups into functional guilds.Functional indicators are less likely to be constrained by regionally restricted biota.
850 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 850
Thus, functional approaches lead to a more global view of stream ecosystems, a viewthat is much less variable than one based only on taxa inhabiting stream communities.Hunsaker et al. (1990) argued that for regional ecological risk assessments to be effec-tive, the system must be functionally defined, with the spatio-temporal boundaries ofthe system set by functional attributes of the communities inhabiting the system.Assessments that are functionally based are likely to have greater applicability acrossregions (Hunsaker et al., 1990).
5.3.2.1. PhotosynthesisOne of the most direct measures of plant physiology is photosynthesis. This non-taxo-nomic integrator of physiological condition is responsive to changes in environmentalcondition, and can be accurately measured either by carbon incorporation or oxygenevolution. The actual mechanism of photosynthetic inhibition varies by chemical, butmost inhibitors fall into three categories, those that interrupt electron transport activity,those that alter the structure of chloroplasts, and those that reduce chlorophyll concen-trations within the chloroplast. Most herbicides and organochlorine andorganophosphate pesticides inhibit photosynthesis by blocking electron transport.Adjusting photosynthesis for chlorophyll a per unit mass allows for comparisons ofcommunities with differing levels of biomass, resulting in lower variance componentsto these measures. Hill et al. (1997), found depressed periphyton photosynthesis in themetal-impacted Eagle River, Colorado, during an investigation for the USEPASuperfund program. In their Superfund assessment of a metal-impacted river, Hill etal. (1997) also found significantly reduced quantum yield (photosynthesis adjusted forsolar radiation) and assimilation ratio (chlorophyll-adjusted photosynthesis) in streamswith elevated dissolved metal concentrations.
5.3.2.2. RespirationRespiration in aquatic plants is often overlooked in the assessment of physiologicalcondition and responses to perturbations. Respiration integrates most cellular functionsand indirectly measures impacts to all cell systems. The mechanisms of respiratoryinhibition or stimulation by chemical substances are poorly understood, but electrontransfer within glycolysis and the Krebs cycle seems likely points of action. Inregional-scale EMAP surveys of Appalachian, Rocky Mountain, and CaliforniaCentral Valley streams (Hill et al.,1998, 2000c), we found similar rates of respirationamong these diverse regions, and reported significant correlations between respirationand stream chemistry and habitat variables.
5.3.2.3. Microbial enzyme activityThe use of microbial enzyme activity to assess the integrity of aquatic ecosystems arelatively new idea. The lack of a substantial microbial history in ecosystem assess-ments stems largely from the lack of understanding of the microbial assemblage withinthe ecosystem. Through its role in detritus processing, the microbial assemblage inte-grates carbon and nutrient cycling within the process of energy flow throughecosystems. Because of its role in ecosystem function, the microbial assemblage maybe the best indicator of overall ecosystem process integrity and any change in micro-bial metabolic rates may be construed as an impact. Since microbial metabolic
pathways are dependent on respiration, respiration should be a sensitive indicator ofthe condition of the stream microbial assemblage. Dehydrogenase activity has beenused to measure the effects of metabolic activity of stream microbial communities andtheir responses to physical and chemical disturbances (Burton and Lanza, 1987; Burtonet al., 1987; Blenkinsopp and Lock, 1990, 1992). Hill et al., 2001b compared the O2
depletion method with DHA in EMAP studies of Appalachian streams. Hill et al.(2000a) reported that APA was positively correlated with riparian zone agriculture,and negatively correlated with indices of human disturbances in the riparian zones.
5.3.2.4. Community metabolismCommunity metabolism (primary productivity and respiration) is a commonlymeasured functional attribute of stream ecosystems. That metabolism is not used moreoften in monitoring may be linked to the perception that its response to environmentalconditions is too variable and thus is of limited use for assessing a streams responseto environmental conditions. Hill et al. (1997) found significant differences in commu-nity metabolism between metal-impacted and reference sites in their Superfundresearch on the Eagle River, Colorado.
5.3.2.5. Nutrient uptake and spiralingNutrient spiraling, defined as spatially-dependent nutrient cycling in stream ecosys-tems (Elwood et al., 1983), links the concept of nutrient cycling with unidirectionalflow. Nutrient cycles of ecosystems are viewed as either closed (i.e., an atom ofnutrient is continuously recycled within the ecosystem) or open (i.e., an atom of anutrient is cycled within the system, but is eventually exported from the system). Mostecosystems are considered to be open, though the degree of openness may depend onthe relative time scale used to analyze nutrient cycles. Because of the unidirectionalflow, nutrient cycling was never considered as an attribute of streams. A basic differ-ence between spiraling and cycling in an open system is that spiraling moves thenutrient downstream within the same system rather than losing it from the system.That is, transport occurs as a part of the nutrient cycle rather than as an alternative toit. Newbold et al. (1981) developed an index of nutrient spiraling known as spirallength, defined as the average downstream distance associated with one complete cycleof a nutrient atom. Under steady-state conditions spiral length is expressed as the ratioof total downstream transport of a nutrient to nutrient utilization. It appears that uptakelength accounts for as much as 98% of spiral length (Newbold et al., 1983; Mulhollandet al., 1990), and that uptake length may be measures downstream depletion of pulseadditions of non-radioactive nutrients (Stream Solute Workshop, 1990; Webster et al.,1991). Webster et al. (1991) reported decreased PO4
�3 retention in logged streams,resulting in longer uptake lengths, and attributed this to biotic and abiotic changes inthe stream. As part of the Agency’s stream monitoring methods development research,Hill and McCormick (2001) found differences in NH4
+1 and PO4�3 uptake in streams
draining harvested and old growth watershed, and attributed these results to differ-ences in biotic activity and transient channel storage. Figure 4 shows the regressionof PO4
2� concentrations against downstream distance for old growth and harvestedwatersheds in our northwestern California sites.
852 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 852
6. Toxicity assessments
6.1. Point source toxicity assessment
Whole effluent toxicity testing (WET) is defined as “the aggregate toxic effect of aneffluent measured directly by an aquatic toxicity test” [USEPA Regulations, 54 FR23868 at 23895; June 2, 1989]. Aquatic toxicity test methods designed specifically formeasuring WET and receiving water toxicity have been codified in USEPA regula-tions (40 CFR part 136 [60 FR 53529; October 16, 1995]). These WET test methodsemploy a suite of standardized freshwater, marine, and estuarine tests using plants,invertebrates, and vertebrates to estimate acute and short-term chronic toxicity of efflu-ents and receiving waters. Specific test procedures for conducting WET and receivingwater tests are included in USEPA, 1993c and USEPA, 1994a. These three methodmanuals (WET method manuals) were incorporated by reference into USEPA 40 CFRpart 136 in 1995. As regulations, use of these methods and adherence to the specifictest procedures outlined in the WET method manuals is required when monitoringWET under the National Pollutant Discharge Elimination System (NPDES).
6.2. Receiving water toxicity assessment
The USEPA conducted a field toxicity study in order to determine if laboratory esti-mates of safe concentrations of pollutants were valid for protection of real streams(Geckler et al., 1976). The study was conducted on Shayler Run, in Clermont County,Ohio and examined at the effects of copper on stream biota. Copper was added to thestream for 33 months to maintain a concentration of 120 �g/L, a concentration thatwas expected to adversely affect some fish species but not others. The stream alsoreceived sewage effluent containing a variety of compounds known to affect acutecopper bioavailability. All but one abundant species of fish and four of the five mostabundant macroinvertebrate species were adversely affected by exposure to copper atthis concentration. Direct effects on fish were death, avoidance, and restrictedspawning. Acute and chronic tests with copper were also conducted in standard labo-ratory conditions and streamside with fathead minnows. This study concluded thatlaboratory derived data could be used to predict toxic effects in a natural stream situ-ation. In general, the toxicity of copper was underestimated by the laboratory databecause of avoidance of fish to copper was not measured by laboratory exposures(Geckler et al., 1976). Indirect effects on fish, as a result of the effects of copper onthe aquatic food chain, could not be demonstrated. More recently the USEPA has usedmethods similar to the WET methods to assess toxicity in receiving waters. In the fallof 1995 and spring of 1997 the USEPA Region VIII collected physicochemical andtoxicity information from the Clear Creek watershed in central Colorado. The purposeof this investigation was to evaluate the relative advantages of an ecotoxicologicalapproach for identifying residual contaminant sources and evaluating established cleanup goals within a watershed. Ceriodaphnia and fathead minnow 48-hr acute toxicitytests and metal analyses were performed on 32 stream samples collected in 1995 and37 stream samples collected in 1997 from the Clear Creek watershed. Stream waterwas shipped overnight to the USEPA Aquatic Research Facility in Cincinnati for
Ceriodaphnia and fathead minnow toxicity testing and to the USEPA Region VIIILaboratory in Denver, Colorado for metal analyses. Both profile tests (100% streamwater) and definitive toxicity tests (stream samples serially diluted with moderatelyhard reconstituted water) were performed. Ceriodaphnia toxicity results (LC50s and100% stream water) in 1995 and 1997 showed a similar trend throughout the water-shed; the upper 20 miles of the mainstem of Clear Creek had LC50s >20% streamwater, while the lower 20 miles had LC50s <10% stream water. Converting the toxi-city results to a No Observed Acute Effect Level (NOAEL), in metal concentrations,was demonstrated presented as an alternative approach for evaluating clean up goalswithin an entire watershed (Fig. 5). Figure 5 shows the results of zinc metal analyses,calculated USEPA National Water Quality Criteria and NOAELs calculated fromCeriodaphnia and fathead minnow tests. Although the NOAELs are higher thannational criteria they can be used for interim clean-up goals and can be used to tractprogress in eventually meeting the Criteria. As part of a Superfund InnovativeTechnology Evaluation (SITE) Program, the USEPA evaluated a remediation tech-nology that was put in place at the Summitville Mine Superfund Site in southernColorado. The technology evaluated was a successive alkalinity producing system(SAPS) for removing high concentrations of metals (aluminum, copper, iron,manganese, and zinc). Two treated and one untreated water sample were evaluatedusing a series of acute aquatic toxicity tests with Pimephales promelas, the fatheadminnow and Ceriodaphnia dubia, and chronic aquatic toxicity tests with Onco-rhynchus mykiss, rainbow trout. All tests used moderately hard reconstituted water asthe control and dilution water. The P. promelas used in this study were three days old,the C. dubia were <24 H old, and the rainbow trout O. mykiss used were 18 days old, 5 days post swimup, provided. The trout tests were conducted at 15 C, the twoother species were tested at 20 C. Ceriodaphnia were more sensitive than rainbowtrout, than the fathead minnow. Both treated samples reduced toxicity by 7–8-fold forCeriodaphnia, 10-fold for rainbow trout, and about 5-fold for the fathead minnow.However, a substantial amount of toxicity remained. A 100-fold more reduction in theconcentration of metals would need to be achieved to remove acute toxicity to rainbowtrout, a 1000-fold reduction in metals in both treatments would be needed to removeacute toxicity to Ceriodaphnia and a 50-fold reduction for no acute effects to fatheadminnows.
6.3. Sediment toxicity assessment
As part of the EMAP Surface water program, sediment samples were collected toassess toxicity on a Regional scale in streams and rivers of the Mid-Atlantic U.S. in1994, 1997 and 1998 and in the Southern Mineralized Zone Ecoregion in the ColoradoRocky Mountains in 1994 and 1995. Sample sites were selected randomly using aprobability design so that the results could be inferred for the entire region. Sedimentswere collected from each site by scooping 11 small samples of surficial sedimentswithin a 150–800 m long sample reach (reach length was proportional to stream width)and composited in a bucket. Samples were then placed in a plastic bag and shippedon ice back to the lab and stored at 4–6°C. A 7-day Hyalella azteca, amphipod, lethalityand growth method was used to assess the toxicity of all sediment samples. During
854 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 854
the spring of 1994 in Mid-Atlantic streams, an estimated 3890 km of stream length(2.1% of the 188,700 km in the target population) were found to have toxic sediment(survival or growth significantly less than (p = 0.05)) the control). During the summersof 1997 and 1998 in Mid-Atlantic streams and river sediments, an estimated 5610 km(2.2%) of the 250,500 km of target length were found to be toxic. In the 1994/1995Southern Mineralized Zone Ecoregion of the Colorado Rockies, an estimated 10.2%or 673 km of streams had toxic sediment.
Sediment toxicity assessments have also been used to evaluate the success of reme-diation of contaminated sediments (Tabak et al., 2000). Freshwater and marinesediment toxicity tests were used to measure baseline toxicity of sediment samplescollected from New Jersey/New York Harbor (NJ/NY) (minimally contaminated) andEast River (PAH-contaminated) sediment (ERC). Four freshwater toxicity tests wereused: (1) amphipod (Hyalella azteca) mortality and growth tests (a standard 10-dayUSEPA method and two 7-day exposure methods (one using the standard amount ofsediment, 100 ml; one using a reduced sediment volume, 17 ml) – the reduced volumefreshwater amphipod test was developed and used in this study since existing volumerequirements of the USEPA standard method exceeded the amounts available fromenhanced or natural attenuation treatment); (2) a 7-day aquatic worm (Lumbriculusvariegatus) mortality and budding test; (3) a 7/8-day fathead minnow (Pimephalespromelas) embryo-larval survival and teratogenic test (FHM-EL); and (4) a 4-dayvascular aquatic plant (Lemna minor) frond number/growth/chlorophyll a test (Duck-weed). Two marine tests were also used: (1) an amphipod (Ampelisca abdita)10-daymortality test and a sheepshead minnow (Cyprinodon variegatus) embryo-larval sedi-ment test (SHM-EL). ERC sediments were found to be highly toxic to all freshwater
Average Zinc Level By SiteCeriodaphnia & Fathead Ecorestoration Goal for Clear Creek Basin
Figure 5. Figure demonstrating an approach to setting ecorestoration goals as interim goals for achievingsite specific water quality standards.
3571 MARKERT3 23/10/02 9:46 am Page 855
and marine organisms tested whereas the NJ/NY, sample showed no significant toxi-city to the marine amphipod, but was slightly toxic to the freshwater worm and tofreshwater and marine fish. For all tests with freshwater organisms and the one marineamphipod no survival was found in any of the tests except for one of the freshwateramphipod tests (55%). The ERC sediment significantly reduced frond production (�58.3%) and chlorophyll a levels (�35.4%) in the freshwater duckweed test. Todetermine the cause of toxicity in the sediments, five sediment manipulations wereperformed: (1) a sediment purge procedure, where 2 to 4 volumes of lab water werereplaced over the sediment in a 24-h period; (2) a sediment dilution procedure where,grade 40 silica sand was mixed with PAH-contaminated sediments on a weight:weightbasis; (3) a sediment aeration procedure, where sediment samples were aerated byadding 80 ml of sediment (140 g) to a 250 ml glass graduated cylinder and 120 ml ofoverlying water followed by aeration for 24–48 h; 4) an Ambersorb TreatmentProcedure, where PAH-contaminated sediment samples were treated with two typesof organic removal resins – Ambersorb 563 (AS 563) and Ambersorb 572 (AS 572);and (5) an Amberlite Treatment Procedure where IRC-718, an inorganic removal resin,was mixed with PAH-contaminated sediments. Results showed that freshwateramphipod survival was improved with the sediment aeration procedure and with 8%AS 563 and AS 572 treatments. Toxicity can also be reduced with the sediment dilu-tion technique (100-fold). These manipulations revealed that hydrogen sulfide, organiccompounds and inorganic compounds (metals) were factors in ERC sediment toxicity.
For the estuarine component of the EMAP program, sediment toxicity tests includedthe 10-day acute test method on Ampelisca abdita, as well as Ampelisca verilli, andMicrotox. For the four-year period from 1990–93 sediment toxicity (survival < 80%)using Ampelisca abdita was observed in 9±2% of the bottom sediments in estuarinearea in the Virginian Province (Paul et al. 1999). In 1994, 3.2±6.2% of the estuarinearea of the Carolinian Province showed sediment toxicity; in 1995, 0% was observed(Hyland et al., 1998). Sediment toxicity was also observed in the estuarine resourcein the Louisianian Province (Macauley et al. 1999).
6.4. Tissue contaminants
The USEPA conducted a national screening-level investigation in 1987 to determinethe prevalence of selected bioaccumulative pollutants in fish, and to correlate elevatedfish tissue contaminant levels with pollutant sources. Game fish and bottom-dwellingfish were collected from 314 locations thought to be influenced by various point andnonpoint sources, and the fish tissue samples were analyzed to determine levels ofselected contaminants. A list of 60 target analytes was developed for the study,including dioxins and furans, PCBs, pesticides and herbicides, mercury, and severalother organic compounds. Results of the 1987 study indicated that target analytes werepresent in fish tissue at many of the sampling sites, and some of the contaminantsoccurred at levels posing potential human health risks.
In 1992–1994 the EMAP-Surface Waters program conducted a survey of 167 lakesin the Northeastern United States and analyzed whole fish composite samples for cont-aminants, including Al, As, Cd, Cr, Cu, Fe, Hg, Ni, Pb, Se, and Zn. Using values forfish tissue contaminant levels that pose consumption risk derived from the literature
856 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 856
and hazards assessment models, methylmercury (MeHg) was determined to be themetal contaminant of regional concern to fish consumers: 26% of lakes contained fishwith MeHg exceeding a critical value of 0.2 �g/g, which implies risk to human con-sumers. Compared to USEPA wildlife values, 54% and 98% of lakes contain fish withMeHg exceeding critical values (0.1 �g/g; 0.02 �g/g) which implies risk to piscivo-rous mammals and birds, respectively. The other metals analyzed appeared to be atsafe levels on a regional scale, and of only localized concern with regard to humanhealth (Yeardley et al., 1998). In the EMAP 1993/1994 Mid-Atlantic Region assess-ment, fish assemblages from first- through third-order streams were dominated bysmall, short-lived fish (minnows (Cyprinidae), darters (Percidae), and sculpins(Cottidae)) that were more widely distributed and abundant than the large speciestypically chosen for tissue contaminant studies (suckers (Catostomidae), trout(Salmonidae), bass and sunfish (Centrarchidae), and carp (Cyprinidae)). Whole fishhomogenate concentrations from small fish species exceeded the detection limits forcontaminants such as mercury, DDT and PCBs in 50 to 65% of the stream lengthassessed and yielded greater regional estimates of wildlife exposure to mercury, DDT,dieldrin and chlordane than large fish species. Whole fish homogenate residues in largespecies exceeded the detection limits in only 32 to 38% of the stream length butregional estimates of wildlife exposure to PCBs were higher in larger fish. To makeregional estimates of wildlife exposure to whole fish contaminants, USEPA developedwildlife values to reflect a potential threshold for toxic effects from chlordane, DDTand metabolites, dieldrin, endrin, mercury and PCBs exposure, using the approachdescribed in USEPA’s Great Lakes Water Quality Initiative. Whole fish homogenateconcentrations of mercury, PCBs, DDT and metabolites, chlordane and dieldrinexceeded one or more of the wildlife values in both small and large fish species. Thisstudy concluded that, their greater distribution and higher estimates of contaminationindicate that small, short-lived fish may be an excellent choice as target species forconducting regional fish contaminant studies (Lazorchak et al., 2002). USEPA alsoused the same EMAP data but looked at all 56 analytes and used toxicological bench-mark values for the belted Kingfisher (USEPA, 2001). Twenty two of the 56 analyteshad median values that were above detection limits for either small or large fish. Allsites from which samples were taken showed exposure to at least one contaminant.Seventy sites (100%) exceeded at least one of the 16 benchmark values and 22 sites(31.4%) exceeded four or more benchmark values (USEPA, 2001).
In 2000, the USEPA initiated work on three year national study of chemical residuesin fish tissue, designed to expand the scope of the 1987 study. The new study is statis-tically designed and provided screening level data on fish tissue contaminants from agreater number of water bodies than were sampled in 1987. The 2000 study expandedthe scope of the 1987 study which focused on chemical residues in fish tissue nearpoint source discharges. The year 2000–2002 study intended: (1) to provide informa-tion on the national distribution of selected Persistent Bioaccumulative Toxics (PBT)residues in game fish and bottom-dwelling fish in lakes and reservoirs of the conti-nental United States (excluding the Great Lakes); (2) to include lakes and reservoirsselected according to a probability design; (3) to involve the collection of fish fromthose randomly-selected lakes and reservoirs over a three year survey period; (4) willnot be used to set fish consumption advisories; and (5) to include the analysis of fish
tissue for PBT chemicals selected from the USEPA’s multi-media candidate PBT listof 451 chemicals and a list of 130 chemicals from several contemporary fish and bioac-cumulation studies. Lakes and reservoirs were chosen as the target population becausethey are accumulative environments where contamination is detectable, provideimportant sport fisheries nationwide, offer other recreational (non-fishing) access andopportunities, and occur in agricultural, urban, and less-developed areas, so that asso-ciations with each primary use may be determined. Lakes and reservoirs were the focusrather than other water body types due to the fact that fish consumption advisoriesrepresent 16.5% of the Nation’s total lake acres (plus 100% of the Great Lakes),compared to 8.2% of the Nation’s total river miles.
7. Real-time biological monitoring approaches
Assessing the environmental exposures of nonpoint sources, municipal and industrialpoint sources and storm water emanating from large metropolitan areas on largeaquatic ecosystems presents many new and complex challenges. Unlike traditionalpoint source discharges, storm water discharges and nonpoint sources of major metro-politan areas are sporadic and vary in intensity creating temporally and spatiallyvariable shock loadings to receiving waters. Consequently, traditional assessment tech-niques which rely solely on sampling and characterization of the water column areineffective in determining exposures to organisms in ecosystems that are temporallyand spatially variable. In addition, the traditional acute and chronic water column andsediment toxicity tests (USEPA, 1993c; USEPA, 1994, a, b) used in toxicity testingare not without weaknesses when applied to variable exposures. Foremost among theweaknesses are concerns of how well, if at all, these methods mimic exposure ofaquatic organisms during episodic events. Understanding interactions among toxicants(additivity, synergism, antagonism, etc.) is simply inadequate to predict biologicalresponses (Waller et al., 1996).
Aquatic organisms have been shown to generate bioelectric signals which propa-gate into surrounding water. These signals can be recorded as rhythmic analog signalsrepresentative of specific movement activities (e.g. gill beats, heart rates, etc.). In addi-tion, gape measurements (the degree to which a bivalve is open or closed) have beenused with clams and mussels (bivalves) as a means to determine the status of theseorganisms. Utilizing appropriate statistical techniques and accompanying electronics,changes in bioelectric action potential (BAP) responses of fish and gape in bivalvescan be detected, processed, and continuously recorded. They have also been used indetecting water quality induced stress in aquatic organisms (Waller et al., 1996). Inresponse to toxic stress or unfavorable environmental conditions, bivalves (clams andmussels) may temporarily avoid exposure by closing their valves. Indeed this behavioris the basis of current biomonitoring approaches where value-movement or “gape” isused as a measure of environmental stress or early warning to developing toxicity inaquatic systems (Waller et al., 1996 and Allen et al., 1996). Gape activity in livebivalves is the result of integrated responses of muscle/nerve functions, specificallyadductor muscle contraction. Monitoring gape position as open or closed (“catch”state) provides an all-or-none quantal response that in the individual bivalve is a
858 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 858
qualitative phenomenon. Graded responses, such as a measured contraction of adduc-tor, cardiac or foot muscles, are readily quantifiable, and provide higher levels ofsensitivity to environmental stress analysis. Graded muscle responses are readilydetected as bioelectric action potentials using electrocardiographic (EKG) techniques,but have limited utility in practice because they require that electrodes (probes) beinserted and are thus invasive and destructive to the bivalve.
A fish gill ventilatory activity monitor developed for remote field operations hasbeen in use since the late 1970s and early 1980s (Morgan et al., 1978; Morgan et al.,1979a and b). Until the late 1980s, researchers were primarily interested in monitoringbioelectric events from aquatic animals in terms of the rate in which they occurred,i.e. the number of events per unit time. Although rate monitoring provided usefulinformation on the state of the organism, considerable losses in sensitivity and infor-mation were experienced when looking only at heart and breathing rates. To accountfor these losses the system has been redesigned to digitally record the bioelectricsignals generated by muscle and nerve responses and to process the data by digitalprocessing (DSP) utilities.
The USEPA in cooperation with the University of North Texas, Denton, Texas and Tennessee Technical University, Cookesville, Tennessee installed a biologicalmonitoring station on the Little Miami River at Miamiville, OH. The river is desig-nated as a National and State Scenic River. The purpose of this station was to monitorthe water quality of the Little Miami River near Cincinnati using physical/chemical
Figure 6. Schematic diagram of how real-time biological and chemical data are collected, transmittedand received.
3571 MARKERT3 23/10/02 9:46 am Page 859
and biological (fish and clams) sensors. The University of North Texas has placedclams in the river and has monitored their behavior while Tennessee TechnologicalUniversity monitors caged fish “breathing” characteristics. The clam biomonitoringsystem was made up of three major components; Data Collection, Processing andCommunication, and Power (Fig. 6). The data collection component consisted of botha biological and physical/chemical aspect. The biological aspect consisted of 15 clams.Gape was measured once per minute using proximity sensors and a stainless steelproxy attached to a shell of a clam. Physical/chemical data, temperature, pH, conduc-tivity and dissolved oxygen, were also collected in real time using a multiprobe. Plotsof these data over 24 hour periods were posted on a website (www.ias.unt.edu/~jallen/littlemiami/Clam_Page.html). Data were collected and processed using a microcom-puter. The computer also telemetered the data to the University of North Texas AquaticToxicology Laboratory. A digital cellular modem was used to connect to the internetand pass the data for further analysis, display, and archiving. The field installation waspowered using four, 90 amphr batteries. A solar array maintains the charge on thebattery during the daylight hours. One finding of the study was that deployment inflowing systems with quickly fluctuating depths presented problems in operation andmaintenance. In addition, uploading data from the two biosensors required moreenergy and cellular time than anticipated. However, there was success for up to 60continuous days of operation and some baseline levels were established as well as anunderstanding of clam behavior under these conditions.
A USEPA Environmental Monitoring for Public Access and Community Tracking(EMAPCT) funded real-time biological monitoring project has been in operation in2000 and 2001 on the Chicamacomico River, Maryland. The U.S. Army Center forEnvironmental Health Research (USACEHR) developed an automated fish monitoringsystem, known as the Real Time Environmental Protection System (REPS). REPS wasdesigned to detect harmful water quality conditions in the Chesapeake Bay and otherwaterways. In cooperation with the Maryland Department of Natural Resources, aportable REPS facility has been monitoring the water at a potential site of toxicPfiesteria activity on the Chicamacomico River. REPS complements other on-goingmonitoring efforts to give early warning of potential risks to human and ecologicalhealth http://www.aquaticpath.umd.edu/empact/index.html . The REPS has been incontinuous operation on the Chicamacomico River at Drawbridge, MD from Julythrough November 2000, generating data on water quality and fish behavior. The REPSdid not detect severe stress resulting in fish death at any time during the monitoringperiod. Short-term fish stress events occurred occasionally, generally during stormevents, but fish recovered quickly and no Pfiesteria were detected in water samplesautomatically collected during these events.
8. Molecular approaches
8.1. Measures of gene expression
Expression of certain genes in aquatic organisms is the cellular reaction to environ-mental action. Molecular biology offers sensitive and expedient tools for the detection
860 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 860
of exposure to environmental stressors. Molecular approaches provide the means fordetection of the “first cellular event(s)” in response to environmental changes – specif-ically, immediate changes in gene expression. Environmental exposure monitoringusing gene activity as an indicator is based on the hypothesis that sub-cellular eventsresulting from an organism’s contact with chemical milieus are manifested far inadvance of those effects observed at higher levels of biological organization.Specifically, this approach involves detection of changes in gene transcription and rela-tive levels of tissue-specific messenger RNA (mRNA), which occur as a result of directcontact with xenobiotic chemicals present in the environment. Protein products thatare synthesized in response to environmental change represent the terminal aspect ina multi-step biochemical pathway that is replete with diverse cellular control mecha-nisms. Most of these studies in teleosts have centered on the single chemical/singlegene response (i.e., �-naphthaflavone/P450IA1 and estradiol/vitellogenin). RecentUSEPA studies describe the quantifiable induction of vitellogenin gene transcriptionin common carp (Cyprinus carpio) (Lattier, et al., 2001), and fathead minnows(Pimephales promelas) (Lattier et al., 2002) as indicators of environmental estrogens.This scheme largely ignores the reality of environmental complexity, such as chem-ical fate and transport, synergism of chemical mixtures, multiple genes competing forlimited intracellular pools of transcription co-factors, and gene induction profilesresulting from chronically exposed organisms. Emerging technologies, such as differ-ential display and microarray DNA chips, provide a means to detect differences ininestimable gene products induced by the above scenarios. When applied judiciously,and in concert with higher order ecological analyses, molecular biology can providethe important link between immediate environmental exposure and long term biolog-ical, community and population effects.
Using the commonly found USEPA toxicological standard model, fathead minnow(Pimephales promelas), and the technique to detect changes in total cellular RNAs(differential display), we will establish the temporal global gene expression profilechanges in embryos, resulting from exposure to several individual compounds. Thesewill include a potent inducer of P450IA1, an identified metal, an estrogenic compound,and a target pesticide. At the dose and time wherein each inducer causes the greatestalteration in the mRNA expression profile, a subtractive cDNA libraries will beconstructed. Subtractive cloning is a powerful technique for isolating genes expressedor present in one cell population but not in another. Constitutively transcribed mRNAs,or those that are common between two cell or tissue types, are selectively removed.The remaining mRNA arises from genes that are uniquely expressed under specificenvironmental, developmental, or disease conditions. For the purpose of our indicatorinitiative, individual, inducer-specific subtractive libraries (ostensibly representingonly those up- or down-regulated gene products resulting from a specific exposure)will then be combined, and arrayed in total on DNA chip(s), thus providing probes fordetecting exposure induced changes in gene expression.
The platform for DNA arrayed on chips is accomplished through nanofabrication.Probe DNA is typically attached to standard 1″ × 3″ microscope slides by photolith-ography or other covalent chemistry, and spotted by “pins” at a density of 75–125 _mon-center with 0.1–2.0 nanograms of probe DNA per spot. Current technology offersthe ability to place between 15,000 and 45,000 probe sequences on a single array.
Following fabrication of the chip, mRNA (cDNA) of exposed organisms and that ofthe control animals is isolated, and differentially labeled with the fluorescent dyes Cy3[green] and Cy5 [red] in enzymatic reactions. Equal amounts of the fluorescent taggednucleic acid, from both control and exposed animals, will be combined and simulta-neously hybridized to complementary probes covalently bound to the chip(s).Following this competitive hybridization, the fluorescent dyes present on the arrayedchip are then induced to an excitation state by two independent lasers. The chip is then‘read’ with a dual wavelength scanner at 632 and 532 nanometers for the red and greenlabels, respectively. In the resulting pseudo color image, the green Cy3 and red Cy5signals are overlaid. The range of spectral color intensities, detected on the DNA spots,indicates whether a gene in question was expressed exclusively in one cell type(absolute red), or the other (absolute green). A yellow spot indicates equal intensityfor the dyes, and suggests that a gene was expressed equally in both cell types.Comparison of patterns of gene expression between unexposed organisms and thoseexposed to environmental mixtures will permit investigators to determine genes thatare activated, inhibited, or uncharacteristically modulated by multiple chemical stres-sors present in binary, ternary, and more complex mixtures.
This approach will facilitate understanding of the synergistic nature of expressedgenes, and multi-gene pathways, when challenged with chemical mixtures, establishpatterns of expressed genes that correlate with single or complex toxicant exposures,and provide information about the relative bioavailable concentrations of environ-mental stressors. This approach will also provide the ability to differentiate geneexpression patterns influenced by chronic chemical exposure, when compared to genesthat are activated by acute onset exposure. Embryo-specific expression patterns usingindividual compounds and environmental mixtures will then be indexed, providing‘expressed sequence tags’, gene expression patterns, and a means to structurally char-acterized differentially expressed cDNAs. Early developmental stages of Pimephalespromelas were chosen for analysis based on the following rationale: (i) This speciesis ubiquitous, and represents the USEPA toxicological standard model. (ii) The numberof genes expressed during plastic early development far exceeds expression profiles inlater life stages. (iii) Disruption of genes critical in early development could influencereproductive outcomes, and pose deleterious consequences to communities and popu-lations. Gene expression analyses of early developmental stages provide unparalleledmeans by which to link patterns of gene expression with whole animal and populationeffects. (iv) Use of this developmental stage may comply with future Agency guide-lines restricting vivisection and/or whole animal studies.
The fathead minnow is an excellent freshwater fish to be used as a model for testinggene expression for two reasons, (1) its wide distribution in North America, and (2)its long history of use in acute and chronic testing of contaminants, effluents andreceiving waters by North American and European ecotoxicologists. The fatheadminnow is a popular bait fish, and the ease with which it is propagated has led to itswidespread introduction both within and outside the native range of the species. It hasbeen so widely distributed in the eastern and southwestern United States by bait trans-portation that it is difficult to determine its original range. The presumed nativedistribution (Lee et al., 1980) extended from the Great Slave Lake in the northwest toNew Brunswick, in eastern Canada, southward throughout the Mississippi valley in
862 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 862
the United States, to southern Chihuahua in Mexico. Distribution records for thisspecies also now include Oregon, and the Central Valley and other locations inCalifornia (Mount, 1968), but there are no records for British Columbia. This speciesis found in a wide range of habitats. It is most abundant in muddy brooks, streams,creeks, ponds, and small lakes, is uncommon or absent in streams of moderate andhigh gradients and in most of the larger and deeper impoundments, and is tolerant ofhigh temperature and turbidity, and low oxygen concentrations.
The fathead minnow is unrivaled in toxicological research concerning the effectsof pollution on freshwater resources. Tolerance to adverse conditions and ease ofspawning makes the fathead minnow ideal for laboratory culture. Brood stock can bemaintained in spawning condition year-round, ensuring a constant supply of larval fishfor toxicity testing purposes. The fathead has been used since the 1960s (Mount, 1968;Norberg and Mount, 1985; Pickering and Lazorchak, 1995; Pickering et al., 1996) togenerate acute life cycle, short-term chronic and embryo-larval toxicity informationon various chemicals, municipal and industrial discharges and receiving waters. Sincethe 1970s more standardized tests have been developed and used for the fatheadminnow throughout North America and Europe than any other freshwater fish. TheUSEPA has standardized acute, short-term chronic and embryo-larval survival andteratogenicity toxicity methods (USEPA, 1993, 1994 a). The American Society ofTesting and Materials has published a guide for acute and chronic fathead short-termchronic fathead minnow testing (ASTM, 1998). Standard methods have also beenpublished describing acute and chronic methods for fathead minnows (APHA, 1998).Environment Canada has published a method for testing the larval growth and survivalof fathead minnows (Environment Canada, 1992) and in Europe the Organization forEconomic Cooperation and Development (OECD) has adopted acute and short-termchronic methods using fathead minnows (OECD, 1984, 1992).
8.2. Measures of genetic diversity
Molecular analysis of intraspecific genetic diversity provides a logical extension ofpreviously described approaches to measure biological integrity with fish assemblagedata. Although studies of molecular genetic diversity at the USEPA are in theirinfancy, there are a number of compelling reasons to believe that molecular geneticmeasures will ultimately provide highly useful bioindicators.
8.2.1. Rationale for development of a genetic diversity indicator
Biodiversity is usually defined in terms of three hierarchically related components:genetic diversity, species diversity, and ecosystem diversity. Genetic diversity is basalin this hierarchy, so erosion of biodiversity at this level will eventually impact diver-sity at species and ecosystem levels. For this reason, it has been argued that a fullydeveloped indicator of genetic diversity may provide a sensitive and efficient measureof ecosystem health, especially when ecosystems are exposed to long-term, low-levelchronic stressors (Bickham et al., 2000). Genetic diversity is fundamentally a trait ofbiological populations, and significant changes in genetic diversity reflect import-ant population-level changes. Genetic erosion may involve loss of diversity within
individual populations, often as a consequence of small (effective) population size, orit may involve loss of diversity among populations, which may reflect removal ofprevious barriers to effective migration. Both types of genetic erosion can increaseextinction risk. Loss of diversity among populations is of particular concern wherelocally adapted populations are introgressed with foreign, nonadaptive genes, loweringpopulation fitness (Waser, 1993). Loss of diversity within small populations may putthem at immediate risk from inbreeding depression, but they are also at risk over thelong term because of loss of adaptive potential (Lande, 1999). Current rates of anthro-pogenic change are unprecedented and populations with little genetic diversity maynot be able to adapt in response to these changes quickly enough to avoid extinction(Lande and Shannon, 1996; Lynch, 1996; Lande, 1999).
In addition to its utility as a direct measure of ecosystem condition, a genetic diver-sity indicator will provide fundamental data that enhances the value and interpretationof other ecological assessment data, such as those obtained from landscape analysesand species assemblage data. An under-appreciated problem for ecological assessmentsis how to define the appropriate ecological unit of analysis. For any one species, thebiological population is the most logical assessment unit for questions of biologicalintegrity. Characterization of population genetic structure is one of the simplest appli-cations of genetic diversity analyses, and delineation of distinct populations and theirgeographic boundaries is a tractable problem using genetic methods. For assessmentsof biological communities, a multi-species analysis of genetic diversity for differentguilds and life-history types will provide the most meaningful information for definingassessment units.
The notion that genetic diversity should be responsive to changes in environmentalcondition is not new, and attempts to relate the variability of molecular genetic markersto specific aquatic stressors date back more than 30 years. These studies include bothfield surveys and controlled laboratory experiments of fish populations, and have eval-uated the effects of metals, acidity, pesticides, radionuclides, and complex effluents(reviewed by Gillespie and Guttman, 1999). Taken as a whole, this body of studiespresents overwhelming evidence that molecular genetic diversity can be substantiallyaltered by deteriorated environmental conditions. Attempts to evaluate genetic diver-sity in relation to environmental quality at large, regional scales are noticeably absentfrom the literature. In fact, most field surveys have compared a relatively small numberof populations in contaminant-exposed and “reference” sites. However, since referencesites are often chosen based on ecological, rather than genetic criteria, it is oftenunclear whether genetic differences between reference and exposed sites reflect recentchanges or long-standing genetic differences.
8.2.2. USEPA genetic diversity research
Current EPA research is designed to assess the utility of incorporating a genetic diver-sity indicator into large-scale assessment and monitoring efforts. To avoid inherentproblems in defining reference sites, a “landscape genetics” approach is used to makeappropriate comparisons. Under this strategy, regional patterns of genetic similaritiesamong populations are compared with current and historical geographic informationto infer patterns of dispersal and evolutionary divergence among populations. Levels
864 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 864
of genetic diversity within populations are compared to patterns of evolutionary diver-gence among populations to differentiate recent from historical genetic changes.Ecological data for each site is compared to inferred patterns of recent genetic changein order to evaluate possible environmental or anthropogenic factors that initiatedgenetic changes. An ongoing pilot analysis focuses on genetics of cyprinid stream fish,particularly the central stoneroller (Campostoma anomalum), creek chub (Semotilusatromaculatus), and blacknose dace (Rhynichtys atratulus ) in the eastern half of theUSA. Two different types of genetic assays are being evaluated. The first utilizes DNAfingerprinting technologies such as randomly amplified polymorphic DNA (RAPD;Welsh and McClelland, 1990; Williams et al., 1990) and amplified fragment lengthpolymorphisms (AFLP; Vos et al., 1995) to identify differences between individualsat large numbers of anonymous DNA loci. The advantage of these techniques is thatgenetic profiles can be generated relatively quickly, with little developmental effortand modest molecular biological expertise. The second approach under evaluationentails sequence analysis of mitochondrial DNA combined with length-polymorphismanalysis of nuclear DNA microsatellites (also called simple-sequence repeats).Microsatellite markers are extremely polymorphic, often segregating for dozens ofalleles within populations, and therefore are extremely sensitive indicators of changesin genetic diversity within populations (Luikart et al., 1998). Mitochondrial DNAprovides a relatively straightforward interpretation of evolutionary relationships amongpopulations (Avise, 1994), so it should be highly complementary to microsatellitemarkers using the landscape genetics approach.
Both of these strategies target genetic loci that are expected to be selectively neutral,on average. A valid concern with both approaches is whether the patterns observedfor these genetic markers are comparable to patterns at more ecologically and evolu-tionarily relevant (fitness-influencing) loci (Lynch, 1996; Rodriguez-Clark, 1999).Future research plans call for targeted analysis of large numbers of protein-coding andregulatory loci, building on an expanding database of knowledge of gene functionresulting from the various fish genome projects (e.g., Detrich et al., 1999), andincreased analytical capabilities resulting from USEPA gene expression research.
Acknowledgments
We would like to thank Teresa Ruby of CSC Graphics Support, c/o USEPA,Cincinnati, Ohio, for her preparation of Figure 1 and her patience in the many revi-sions the authors made to it.
APHA, 1998. Standard Methods for the Examination of Water and Wastewater, 20th edn. American PublicHealth Association, Washington, DC.
ASTM, 1998. Biological effects and environmental fate; Section 11: water and environmental technology.In American Society for Testing and Materials Annual Book of ASTM Standards, 11.05, AmericanSociety for Testing and Materials, West Conshohocken, PA.
Avise, J.C., 1994. Molecular Markers, Natural History and Evolution. Chapman and Hall, New York.Barber, M.C., 1994. Indicator Development Strategy. EPA/620/R-94/022. US Environmental Protection
Agency, Office of Research and Development, North Carolina.Barbour, M.T., Stribling, S.B., Karr, J.R., 1995. Multimetric approach for establishing biocriteria and
measuring biological condition. In: Davis, W.S., Simon, T.P. (Eds), Biological Assessment and Criteria:Tools for Water Resource Planning and Decision Making. Lewis Publishers, Boca Raton, pp. 63–77.
Barbour, M.T., Gerritsen, J., Griffith, G.E., Frydenborg, R., Mccarron, E., White, J.S., Bastian, M.L., 1996.A framework for biological criteria for Florida streams using benthic macroinvertebrates. Journal of theNorth American Benthological Society 15, 185–211.
Bickham, J.W., Sandhu, S., Hebert, P.D.N., Chikhi, L., Athwal, R., 2000. Effects of chemical contami-nants on genetic diversity in natural populations: implications for biomonitoring and ecotoxicology.Mutation Research 463, 33–51.
Blenkinsopp, S.A., Lock, M.A., 1990. The measurement of electron transport system activity in riverbiofilms. Water Research 24, 441–445.
Blenkinsopp, S.A., Lock, M.A., 1992. Impact of storm-flow on electron transport system activity in riverbiofilms. Freshwater Biology 27, 397–404.
Boston, H.L., Hill, W.R., Stewart, A.J., 1991. Evaluating direct toxicity and food-chain effects in aquaticsystems using natural periphyton communities. In: Gorsuch, J.W., Lower, W.R., Wang, W., Lewis,M.A. (Eds), Plants for Toxicity Assessments, Vol. 2. ASTM STP 1115, American Society for Testingand Materials, Philadelphia, pp. 126–145.
Broberg, A., 1985. A modified method for studies of electron transport system activity in freshwater sedi-ments. Hydrobiologia 120, 181–187.
Burton, G.A., Lanza, G.R., 1987. Aquatic microbial activity and macrofaunal profile of an Oklahoma stream.Water Research 21, 1173–1182.
Burton, G.A., Drotar, A., Lazorchak, J.M., Bahls, L.L., 1987. Relationship of microbial activity andCeriodaphnia responses to mining impacts on the Clark Fork River, Montana. Archives forEnvironmental Contamination and Toxicology 16, 523–530.
Cairns, J., Niederlehner, B.R., Smith, E.R., 1992. The emergence of functional attributes as endpoints inecotoxicology. In: Burton, G.A. (Ed.), Sediment Toxicity Assessment. Lewis Publishers, Chelsea, MI,pp. 111–128.
Carlisle, D.M., Clements, W.H., 1999. Sensitivity and variability of metics used in biological assessmentsof running waters. Environmental Toxicology and Chemistry 18, 285–291.
Carter, J., 1990. A new Eunotia and its great morphological variations under stress caused by a habitatloaded with copper salts. Ouvrage dedie a H. Germain, Koletz 1990, 13–17.
Chaloud, D.J., Peck, D.V. (Eds), 1994. Environmental Monitoring and Assessment Program: IntegratedQuality Assurance Project Plan for the Surface Waters Resource Group. EPA/600/X-91/080, Revision2.00. US Environmental Protection Agency, Las Vegas.
Chapman, D. (Ed.), 1992. Water Quality Assessment. Chapman and Hall, London.Charles, D.F., 1985. Relationships between surface sediment diatoms assemblages and lakewater charac-
and Measurement of Periphyton Communities: A Review. Special Technical Publication 690, AmericanSociety for Testing and Materials, Philadelphia, pp. 116–141.
Clements, W.H., Carlisle, D.M., Lazorchak, J.M., Johnson, P.C., 2000. The role of heavy metals in struc-turing benthic macroinvertebrate communities in Colorado’s mountain streams. Eco. Appl. 10, 2,626–638.
Crossey, M.J., LaPoint, T.W., 1988. A comparison of periphyton assemblage structural and functional res-ponses to heavy metals. Hydrobiologia 162, 109–121.
Descy, J.P., 1979. A new approach to water quality estimation using diatoms. Nova Hedwigia 64, 305–323.DeShon, J.E., 1995. Development and application of the invertebrate community index (ICI). In: Davis,
W.S., Simon, T.P. (Eds), Biological Assessment and Criteria: Tools for Water Resource Planning andDecision Making. Lewis Publishers, Boca Raton, pp. 217–243.
Detrich, H.W., Westerfield, M.,, Zon, L.I., 1999. The Zebrafish: Genetics and Genomics. Methods in CellBiology, 60. Academic Press, San Diego.
866 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 866
Diaz-Ramos, S., Stevens Jr., D.L., Olsen, A.R., 1996. EMAP Statistical Methods Manual. EPA 620/R-96/002. US Environmental Protection Agency, Office of Research and Development, Washington, DC.
Dixit, S.S., Smol, J.P., Kingston, J.C., Charles, D.F., 1992. Diatoms: powerful indicators of environmentalchange. Environmental Science and Technology 26, 22–33.
Elwood, J.W., Newbold, J.D., O’Neill, R.V., Van Winkle, W., 1983. Resource spiraling: an operationalparadigm for analyzing lotic ecosystems. In: Fontaine III, T.D., Bartell, S.M. (Eds), Dynamics of LoticEcosystems. Ann Arbor Science Publishers, Ann Arbor, pp. 3–27
Engle, V.D., Summers, J.K., Gaston, G.R., 1994. A benthic index of environmental condition of Gulf ofMexico estuaries. Estuaries 17, 372–384.
Engle, V.D., Summers, J.K., 1999. Refinement, validation, and application of a benthic condition index forGulf of Mexico estuaries. Estuaries 22, 624–635.
Environment Canada, 1992. Biological Test Method: Tests of Larval Growth and Survival Using FatheadMinnows. EPS 1/RM/22. Environmental Protection Series, Ottawa, Ontario.
Fausch, K.D., Lyons, J., Karr, J.R., Angermeier, P.L., 1990. Fish communities as indicators of environ-mental degradation. In: Adams, S.M. (Ed.), Biological Indicators of Stress in Fish. American FisheriesSociety Symposium 8. Bethesda, pp. 123–144.
Fore, L.S., Karr, J.R., Wisseman, R.W., 1996. Assessing invertebrate responses to human activities: eval-uating alternative approaches. Journal of the North American Benthological Society 15, 212–231.
Geckler, J.R., Horning, W.B., Neiheisel, T.M., Pickering, Q.H., Robinson, E.L., Stephan, C.E., 1976.Validity of Laboratory Tests for Predicting Copper Toxicity in Streams. US Environmental ProtectionAgency, Dulth, MN. EPA-600/3–76–116.
Geller, W., 1984. A toxicity monitor using the weakly electric fish Gnathonemus petersi. Wat. Res. 18,1285–1290.
Gibson Jr, G.R., 1991 Biological criteria: research and regulation. Proceedings of a Symposium. USEnvironmental Protection Agency, Office of Water, Washington, DC EPA-440/5–91–005.
Gillespie, R.B., Guttman, S.I., 1999. Chemical-induced changes in the genetic structure of populations:effects on allozymes. In: Forbes, V.E. (Ed.), Genetics and Ecotoxicology. Taylor & Francis,Philadelphia, pp. 55–77.
Griffith, M.B., Kaufmann, P.R., Herlihy, A.T., Hill, B.H., 2001a. Analysis of macroinvertebrate assem-blages in relation to environmental gradients in Rocky Mountain streams. Ecological Applications 11,489–505.
Griffith, M.B., Husby, P., Hall, R.K., Kaufmann, P.R., Hill, B.H., 2001b. Analysis of macroinvertebrateassemblages in lotic habitats of California’s Central Valley. Freshwater Biology, in review.
Hall, R.I., Smol, J.P., 1992. A weighted-averaging regression and calibration model for inferrring totalphosphorus concentration from diatoms in British Columbia (Canada) lakes. Freshwater Biology 27,417–434.
Halliwell, D.B., Langdon, R.W., Daniels, R.A., Kurtenbach, J.P., Jacobsen, R.A., 1999. Classification offreshwater fish species of the Northeastern United States for use in the development of indices of bioticintegrity, with regional applications. In: Simon, T.P. (Ed.), Assessing the Sustainability and BiologicalIntegrity of Water Resources Using Fish Communities. CRC Press, Boca Raton, pp. 301–338.
Ham, K.D., Peterson, M.J., 1994. Effect of fluctuating low-level chlorine concentrations on valve-move-ment behavior of the Asiatic clam (Corbicula fluminea). Env. Tox. Chem. 13 (3), 493–498.
Healey, F.P., Henzel, L.L., 1979. Fluorometric measurements of alkaline phosphatase activity in algae.Freshwater Biology 9, 429–439.
Hellawell, J.M., 1977. Change in natural and managed ecosystems: detection, measurement and assess-ment. Proceedings of the Royal Academy of London, B. Biological Sciences 197, 31–56.
Hill, B.H., Lazorchak, J.M., McCormick, F.H., Willingham, W.T., 1997. The effects of elevated metals onbenthic community metabolism in a Rocky Mountain stream. Environmental Pollution 95, 183–190.
Hill, B.H., Herlihy, A.T., Kaufmann, P.R., Sinsabaugh, R.L., 1998. Sediment microbial respiration in asynoptic survey of mid-Atlantic region streams. Freshwater Biology 39, 493–501.
Hill, B.H., McCormick, F.H., 2000. Nutrient uptake and community metabolism in streams drainingharvested and old growth watersheds: a preliminary assessment. In: Guldin, J. (Ed.), Proceedings of the
Symposium on Ecosystem Management Research in the Ouachita and Ozark Mountains, GeneralTechnical Report No. XX, Southern Research Station, USDA Forest Service, Oxford, MS, pp. xx–xx.
Hill, B.H., Herlihy, A.T., Kaufmann, P.R., Stevenson, R.J., McCormick, F.H., Burch Johnson, C., 2000a.Use of periphyton assemblage data as an index of biotic integrity. Journal of the North AmericanBenthological Society 19, 50–67.
Hill, B.H., Willingham, W.T., Parrish, L.P., McFarland, B.H., 2000b. Periphyton assemblage responses toelevated metal concentrations in a Rocky Mountain stream. Hydrobiologia 428, 161–169.
Hill, B.H., Hall, R.K., Husby, P., Herlihy, A.T., Dunne, M., 2000c. Interregional comparisons of sedimentmicrobial respiration in streams. Freshwater Biology 44, 213–222.
Hill, B. H., Kurtenbach, J.P., 2001. Correlations of sedimentary diatoms with watershed land-use and limno-logical conditions in New Jersey lakes. Lake and Reservoir Management, in review
Hill, B.H., Stevenson, R.J., Pan, Y., Herlihy, A.T., Kaufmann, P.R., Burch Johnson, C., 2001. Comparisonof correlations between environmental characteristics and stream diatom assemblages characterized atgenus and species levels. Journal of the North American Benthological Society 20, 299–310.
Holland, A.F. (Ed.), 1990. Near Coastal Program Plan for 1990: Estuaries. EPA/600/4–90/033. USEnvironmental Protection Agency, Environmental Research Laboratory, Office of Research andDevelopment, Narragansett, Rhode Island.
Hughes, R.M., 1995. Defining acceptable biological status by comparing with reference conditions. In:Davis, W.S., Simon, T.P. (Eds), Biological Assessment and Criteria: Tools for Water Resource Planningand Decision Making. Lewis Publishers, Boca Raton, pp. 31–48.
Hughes, R.M., Kaufmann, P.R., Herlihy, A.T., Kincaid, T.M., Reynolds, L., Larsen, D.P., 1998. A processfor developing and evaluating indices of fish assemblage integrity. Canadian Journal of Fisheries andAquatic Sciences 55, 1618–1631.
Hughes, R.M., Oberdorff, T., 1999. Applications of IBI concepts and metrics towaters outside the UnitedStates and Canada. In: Simon, T.P. (Ed.), Assessing the Sustainability and Biological Integrity of WaterResources Using Fish Communities. CRC Press, Boca Raton, pp. 79–93.
Hyland, J.L., Herrlinger, T.J., Snoots, T.R., Ringwood, A.H., Van Dolah, R.F., Hackney, C.T., Nelson,G.A., Rosen, J.S., Kokkinakis, S.A., 1996. Environmental quality of estuaries of the CarolinianProvince, 1994. Annual statistical summary for the 1994 EMAP-Estuaries Demonstration Project in theCarolinian Province. NOAA Technical Memorandum NOS ORCA 97. NOAA/NOS, Office of OceanResources Conservation and Assessment, Silver Spring, MD.
Hyland, J.L., Balthis, L., Hackney, C.T., McRae, G., Ringwood, A.H., Snoots, T.R., Van Dolah, R.F.,Wade, T.L., 1998. Environmental quality of estuaries of the Carolinian Province, 1995. Annual statis-tical summary for the 1995 EMAP-Estuaries Demonstration Project in the Carolinian Province. NOAATechnical Memorandum NOS ORCA 123. NOAA/NOS, Office of Ocean Resources Conservation andAssessment, Silver Spring, MD.
Karr, J.B., 1981. Assessment of biotic integrity using fish communities. Fisheries 66, 21–27.Karr, J.R., Fausch, K.D., Angermeier, P.L., Yant, P.R., Schlosser, I.J., 1986. Assessing biological integrity
in running waters: a method and its rationale. Illinois Natural History Survey Special Publication 5,Champaign, IL.
Karr, J.R., 1991. Biological integrity, a long neglected aspect of water resource management. EcologicalApplications 1, 66–84.
Karr, J.R., Dudley, D.R., 1981. Ecological perspective on water quality goals. Environmental Management5, 55–68.
Karr, J.R., Heidinger, R.C., Helmer, E.H., 1985. Sensitivity of the index of biotic integrity to changes inchlorine and ammonia levels from wastewater treatment facilities. Journal of the Water PollutionControl Federation 57, 912–915.
Karr, J.R., Chu, E.W., 1997. Biological Monitoring and Assessment: Using Multimetric IndexesEffectively. EPA 235/R97/001. University of Washington, Seattle.
Kerans, B.L., Karr, J.R., 1994. A benthic index of biotic integrity (B-IBI) for rivers of the Tennessee Valley.Ecological Applications 4, 768–785.
Kutka, F.J., Richards., C., 1996. Relating diatom assemblage structure to stream habitat quality. Journal ofthe North American Benthological Society 15, 469–480.
868 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 868
Lande, R., 1999. Extinction risks from anthropogenic, ecological, and genetic factors. In: Landweber, L.F.,Dobson, A.P. (Eds), Genetics and the Extinction of Species. Princeton University Press. Princeton, NJ,pp. 1–22.
Lande, R., Shannon, S., 1996. The role of genetic variability in adaptation and population persistence in achanging environment. Evolution 50, 434–437.
Lange-Bertalot, H., 1979. Pollution tolerance of diatoms as a criterion for water quality estimation. NovaHedwigia 64, 285–304.
Lanza G.R., Cairns J., 1972. Physio-morphological effects of abrupt thermal stress on diatoms. Transactionsof the American Microscopical Society 91, 276–298.
Larsen, D.P., 1995. The role of ecological sample surveys in the implementation of biocriteria. In: W.S.Davis and T.P. Simon (Eds), Biological Assessment and Criteria: Tools for Water Resource Planningand Decision Making. Lewis Publishers, Boca Raton, pp. 287–300.
Lattier, D.L., Gordon, D.A., Burks, D.J., Toth, G.P., 2001. Vitellogenin gene transcription: a relative quan-titative exposure indicator of environmental estrogens. Environmental Toxicology and Chemistry 20,1979–1985.
Lazorchak, J.M., Klemm, D.J., Peck, D.V. (Eds), 1998. Environmental Monitoring and AssessmentProgram-Surface Waters: Field Operations and Methods for Measuring the Ecological Condition ofWadeable Streams. US Environmental Protection Agency, Cincinnati.
Lazorchak, J.M., McCormick, F.H., Herlihy, A.T., Henry, T.R., 2002. Fish tissue contamination in streamsof the mid-atlantic region: an approach to regional indicator selection and assessment. Submitted J. Environ. Toxicol. Chem.
Lee, D.S., Shute, J.R., 1980. Pimephales promelas. In: Lee, D.S., Gilbert, C.R., Hocutt, C.H., Jenkins, R.E.,McAllister, D.E., Stauffer Jr, R. (Eds), Atlas of North American Freshwater Fishes, Publication Number1980–12, North Carolina State Museum of Natural History, Raleigh, North Carolina.
Leland, H.V., 1995. Distribution of phytobenthos in the Yakima River basin, Washington, in relation togeology, land use, and other environmental factors. Canadian Journal of Fisheries and Aquatic Sciences52, 1108–1129.
Leopold, A., 1941. Lakes in relation to terrestrial life patterns. In: ???? Eds ????, A Symposium onHydrobiology. University of Wisconsin Press, Madison, pp. 17–22.
Lowe, R.L., 1974. Environmental requirements and pollution tolerance of freshwater diatoms. EPA-670/4–74–005, United States Environmental Protection Agency, Cincinnati.
Luikart, G., Sherwin, W.B., Steele, B.M., Allendorf, F.W., 1998. Usefulness of molecular markers fordetecting population bottlenecks via monitoring genetic change. Molecular Ecology, 7, 963–974.
Lynch, M., 1996. A quantitative-genetic perspective on conservation issues. In: Avise, J.C. (Ed.),Conservation Genetics: Case Histories from Nature. Chapman and Hall, New York, pp. 471–501.
Lyons, J.S., 1992. The length of stream to sample with a towed electrofishing unit when fish species rich-ness is estimated. North American Journal of Fisheries Management 12, 198–203.
Macauley, J.M., Summers, J.K., Engle, V.D., Harwell, L.C., 2002. Ecological Condition of South FloridaEstuaries. J. Aquat. Ecosyst. Stress Recovery, in press.
Macauley, J.M., Summers, J.K., Engle, V.D., 1999. Estimating the ecological condition of the estuaries ofthe Gulf of Mexico. Environmental Monitoring and Assessment 57, 59–83.
Markert, B., Wappelhorst, O., Weckert, V., Herpin, U., Siewers, U., Friese, K., 1999. The use of bio-indicators for monitoring the heavy-metal status of the environment. Journal of Radioanalytical NuclearChemistry 240 (2), 425–429.
Marshall, J.S., Mellinger, D.L., 1980. Dynamics of cadmium-stressed plankton communities. CanadianJournal of Fisheries and Aquatic Sciences 37, 403–414.
Mason Jr, W.T., Lewis, P.A., Anderson, J.B., 1971. Macroinvertebrate collections and water quality moni-toring in the Ohio River Basin 1963–1967. US Environmental Protection Agency, Office of TechnicalPrograms, Ohio Basin Region and Analytical Quality Control Laboratory, Water Quality Office.Cincinnati.
McCormick, F.H., Fish Communities as indicators of stream condition, 1993. In: Hughes, R.M. (Ed.),EMAP Streams Bioassessment Workshop. Report of the Proceedings. EPA/600/R-93/138. United StatesEnvironmental Protection Agency, Corvallis.
McCormick, F.H., Hughes, R.M., 1998. Aquatic vertebrate indicator. In: Klemm, D.J., Lazorchak, J.M.,Peck, D.V. (Eds), Environmental Monitoring and Assessment Program-Surface Waters: Field Opera-tions and Methods for Measuring the Ecological Condition of Wadeable Streams. US EnvironmentalProtection Agency, Cincinnati.
McCormick, F.H., Peck, D.V., 2000. Fish indicator development: metrics and indices of biotic integrity.In: Jackson, L., Fischer, W., Kurtz, J. (Eds), EPA /620/R-99/005.
McCormick, F.H., Peck, D.V. Larsen, D.P., 2000. A comparison of ecological classification hierarchies formid-Atlantic stream fish assemblages. Journal of the North American Benthological Society 19, 385–404.
McCormick, F.H., Hughes, R.M., Kaufmann, P.R., Herlihy, A.T., Peck, D.V., Stoddard, J.L., 2001.Development of an index of biotic integrity for the mid-Atlantic Highlands Assessment Project.Transactions of the American Fisheries Society 130, 857–877.
McFarland, B.H, Hill, B.H., Willingham, W.T., 1997. Abnormal Fragilaria spp. found in streams impactedby mine drainage. Journal of Freshwater Ecology 12, 141–149.
Medley, C.N., Clements, W.H., 1998. Responses of diatom communities to heavy metals in streams: theinfluence of longitudinal variation. Ecological Applications 8, 633–644.
Messer, J.J., Linthurst, R.A., Overton, W.S., 1991. An EPA program for monitoring ecological status andtrends. Environmental Monitoring and Assessment 17, 67–78.
Morgan, E.L., et al., 1978. Biological water quality monitoring from remote stations and NOAA GOESSatellite. In: Proc. 4th Joint Conf. on Sensing Environmental Pollutants. American Chemical Society,New Orleans, November 1977, pp. 885–887.
Morgan, E.L., et al. 1979a. New developments in automated biosensing from remote water quality stationsand satellite data retrieval for recourse management. In: Proc. 3rd World Congress on Water Resources,International Water Resources Association, Mexico City, 23–27 April, pp. 3534–3542.
Morgan, E.L. et al., 1979b. Automated biomonitoring applications to remote water quality stations andsatellite data retrieval: new developments in achieving real-time biosensing for watershed management.Proceedings 9th Intersociety Conf. on Environ. Systems. Aerospace Div. Amer. Soc. Mech. Engineers,San Francisco, 16–19 July.
Morgan, E.L., Young, R.C., Wright Jr, J.R., 1988a. Developing portable computer-automated biomoni-toring for a regional water quality surveillance network. In: Gruber, ??., Diamond, J. (Eds), AutomatedBiomonitoring. Ellis Horwood, Chichester.
Morgan, E.L., Wright Jr, J.R., Young, R.C., 1988b. Developing portable automated biomonitoring systemfor aquatic hazard evaluation. Special Technical Publication 988, American Society for Testing andMaterials. Philadelphia.
Morgan, E.L., Yokley Jr, P., Rausina, G., Wright Jr, J.R., McFadden, J.F., Red, J.T., 1989. A toxicity testprotocol for mature bivalve mussels using automated biological monitoring. In: Pesticides in Terrestrialand Aquatic Environments, Proceedings of a National Research Conference, Virginia Water ResourcesResearch Center, Virginia Polytechnic Institute and State University, Blacksburg.
Morgan, E.L., et al., 1991. Remote biosensing applications in environmental assessment. EnvironmentalAuditor 2 (4), 213–227.
Morgan, E.L., Waller, W.T., 1993. Detecting aquatic ecosystem change on regional scales: a proposed inte-grated remote-automated biosensing, GIS based network. Bull. Ecol. Soc. Am. 73 (4) (abstract).
Morgan, E.L. et al., 1995. Stress detection in bivalve mollusc using non-invasive bioelectric monitoring ofmyoneural behavior. In: 2nd World Congress of Soc. Environ. Toxicol. and Chem., GlobalEnvironmental Protection: Science, Politics, and Common Sense. 5–9 November 1995, Vancouver(abstract).
Mount, D.I., 1968. Chronic toxicity of copper to fathead minnows (Pimephales promelas Rafinesque).Water Research 2, 214–223.
Moyle, P.B., Marchetti, M.P., 1999. Applications of indices of biotic integrity to California streams andwatersheds. In: Simon, T.P. (Ed.), Assessing the Sustainability and Biological Integrity of WaterResources Using Fish Communities. CRC Press, Boca Raton, pp. 367–380.
870 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 870
Mulholland, P.J., Rosemond, A.D., 1992. Periphyton response to longitudinal nutrient depletion in a wood-land stream: evidence of upstream-downstream linkage. Journal of the North American BenthologicalSociety 11, 405–419.
Mulholland, P.J., Steinman, A.D., Elwood, J.W., 1990. Measurement of phosphorus uptake length instreams: comparison of radiotracer and stable PO4 releases. Canadian Journal of Fisheries and AquaticSciences 47, 2351–2357.
Newbold, J.D., Elwood, J.W., O’Neill, R.V., Van Winkle, W., 1981. Measuring nutrient spiraling instreams. Canadian Journal of Fisheries and Aquatic Sciences 38, 860–863.
Newbold, J.D., Elwood, J.W., O’Neill, R.V., Sheldon, A.L., 1983. Phosphorus dynamics in a woodlandstream ecosystem: a study of nutrient spiraling. Ecology 64, 1249–1265.
Newman, R.M., Perry, J.A., Tam, E., Crawford, R.L., 1987. Effects of chronic chlorine exposure on litterprocessing in outdoor experimental streams. Freshwater Biology 18, 415–428.
Niemi, G.J., Detenbeck, N.E., Perry, J.A., 1993. Comparative analysis of variables to measure recoveryrates in streams. Environmental Toxicology and Chemistry 12, 1541–1547.
Norberg, T.J., Mount, D.I., 1985. A new fathead minnow (Pimephales promelas) subchronic toxicity test.Environmental Toxicology and Chemistry 4, 711–718.
OECD, 1984. Guidelines for testing of chemicals. Section 2, Effects on Biotic Systems. Technical GuideNo. 204, Fish, Prolonged Toxicity Test: 14-Day Study. Organization for Economic Cooperation andDevelopment, Paris.
OECD, 1992. Guidelines for testing of chemicals. Section 2, Effects on Biotic Systems. Technical GuideNo. 210, Fish, Early-Life Stage Toxicity Test. Organization for Economic Cooperation and Develop-ment, Paris.
Ohio Environmental Protection Agency, 1987. Biological Criteria for the Protection of Aquatic Life, Vol.2. Users manual for biological field assessment of Ohio surface waters. Division of Water QualityMonitoring and Assessment, Columbus.
Omernik, J.M., 1987. Ecoregions of the conterminous United States. Annals of the Association of AmericanGeographers 77, 118–125.
Omernik, J.M., Griffith, G.E., 1991. Ecological regions versus hydrologic units: frameworks for managingwater quality. Journal of Soil and Water Conservation 46, 334–340.
Omernik, J.M., 1995. Ecoregions, a spatial framework for environmental management. In: Davis, W.S.,Simon, T.P. (Eds), Biological Assessment and Criteria: Tools for Water Resource Planning andDecision Making. Lewis Publishers, Boca Raton, pp. 49–62.
Pan, Y., Stevenson, R.J., Hill, B.H., Herlihy, A.T., Collins, G.B., 1996. Using diatoms as indicators ofecological conditions in lotic systems: a regional assessment. Journal of the North AmericanBenthological Society 15, 481–495.
Pan, Y., Stevenson, R.J., Hill, B.H., Kaufmann, P.R., Herlihy, A.T., 1999. Spatial patterns and ecologicaldeterminants of benthic algal assemblages in mid-Atlantic streams, USA. Journal of Phycology 35,460–468.
Pan, Y., Stevenson, R.J., Hill, B.H., Herlihy, A.T., 2000. Ecoregions and benthic diatom assemblages inmid-Atlantic Highland streams, USA. Journal of the North American Benthological Society 19,518–540.
Pickering, Q.H., Lazorchak, J.M., 1995. Evaluation of the robustness of the fathead minnow, Pimephalespromelas, larval survival and growth test, USEPA method 1000.0. Environmental Toxicology andChemistry 14, 653–659.
Pickering, Q.H., Lazorchak, J.M., Winks, K.L., 1996. Subchronic sensitivity of one-, four-, and seven-day-old fathead minnow (Pimephales promelas) larvae to five toxicants. Environmental Toxicology andChemistry 15, 353–359.
Rodriguez-Clark, K.M., 1999. Genetic theory and evidence supporting current practices in captive breedingfor conservation. In: Landweber, L.F., Dobson, A.P. (Eds), Genetics and the Extinction of Species.Princeton University Press. Princeton, pp. 47–74.
Sayler, G.S., Sherrill, T.W., Perkins, R.E., Mallory, L.M., Shiaris, M.P., Pedersen, D., 1982. Impact ofcoal-coking effluent on sediment microbial communities: a multivariate approach. Applied and Environ-mental Microbiology 44, 1118–1129.
SETAC, 1994 Final Report: Aquatic Risk Assessment and Mitigation Dialogue Group. Pensacola, FL.Society of Environmental Toxicology and Chemistry, SETAC Foundation for Environmental Education.
Simon, T.P., 1991. Development of ecoregion expectations for the index of biotic integrity. I. Central CornBelt Plain. EPA 905/9–90–005. US Environmental Protections Agency, Region V EnvironmentalSciences Division, Chicago, Illinois.
Simon, T.P., Lyons, J., 1995. Application of the index of biotic integrity to evaluate water resourcesintegrity in freshwater ecosystems. In: Davis, W.S., Simon, T.P. (Eds), Biological Assessment andCriteria: Tools for Water Resource Planning and Decision Making. Lewis Publishers, Boca Raton, pp.245–262.
Sloof, W., 1979. Detection limits of a biological monitoring system based on fish respiration. Bull. Environ.Contam. Toxicol. 23, 517–523.
Slooff, W., De Zwart, D., Marquenie, J.M., 1983. Detection limits of a biological monitoring system forchemical water pollution based on mussel activity. Bull. Environ. Contam. Toxicol. 30, 400–405.
Solladay, J.D., 1994. Morphological variation in fossil diatoms from Mono Lake. In: Kociolek, J.P. (Ed.),Proceedings of the 11th International Diatom Symposium, Memoirs of the California Academy ofSciences 17, pp. 337–348.
Southwood, T.R.E., 1977. Habitat, the templet for ecological strategies. Journal of Animal Ecology 46,337–365.
Stevenson, R.J., Pan, Y., 1999. Assessing ecological conditions in rivers and streams with diatoms. In:Stoermer, E.F., Smol, J.P. (Eds), The Diatoms: Applications to the Environmental and Earth Sciences.Cambridge University Press, Cambridge, pp. 11–40.
Stream Solute Workshop, 1990. Concepts and methods for assessing solute dynamics in stream ecosys-tems. Journal of the North American Benthological Society 9, 95–119.
Tabak, H.H., Lazorchak, J.M., Khodadoust, A.P., Anita, J.E., Suidan, M.T., 2000. Studies on in-situ bio-remediation of PAH contaminated sediments: bioavailability, biodegradation, and toxicity issues.Proceedings of the Second Battelle International Conference on Remediation of Chlorinated andRecalcitrant Compounds, 22–25 May, Monterey, California, in press. ??Eds??
Touart, LW., 1988. Aquatic mesocosm tests to support pesticide registration. Hazard Evaluation DivisionTechnical Guidance Document. Office of Pesticide Programs, US EPA, Washington, DC. EPA540/09–88–035.
Touart, L.W., Maciorowski, A.F., 1997. Information needs for pesticide registration in the United States.Ecol. Appl. 74, 1086–1093.
Townsend, C.R., Hildrew, A.G., 1994. Species traits in relation to a habitat templet for river systems.Freshwater Biology 31, 265–275.
Tu, C.M., 1981. Effects of pesticides on activities of enzymes and microorganisms in clay soil. Journal ofEnvironmental Science and Health B 16, 179–191.
Tyler, G., 1976. Heavy metal pollution, phosphatase activity, and mineralization of organic phosphorus inforest soils. Soil Biology and Biochemistry 8, 327–332.
USEPA, 1977. The Integrity of Water. Proceeedings of a Symposium. US Environmental ProtectionAgency. Office of Water and Hazardous Materials. Washington, DC Stock Number 055–001–01068–1,pp 1–230.
USEPA, 1978. The Selenastrum capricornutum Printz Algal Assay Bottle Test. US EnvironmentalProtection Agency, Corvallis, Oregon. EPA-600/9–78–018,
USEPA, 1989. Rapid Bioassessment Protocols for Use in Streams and Rivers. Benthic Macroinvertebratesand Fish. US Environmental Protection Agency Office of Water. Washington, DC. EPA 440/4–89/001.
USEPA, 1990. Macroinvertebrate Field and Laboratory Methods for Evaluating the Biological Integrity ofSurface Waters. US Environmental Protection Agency. Washington DC. EPA/600/4–90/030.
USEPA, 1991. Technical Support Document for Water Quality-based Toxics Control, 2nd edn. USEnvironmental Protection Agency. Office of Water, Washington, DC. Report 505/2–90/001.
USEPA, 1992. Framework for Ecological Risk Assessment. Risk Assessment Forum, Office of Researchand Development, Washington, DC. EPA/600/R-92/001.
USEPA, 1993a. Fish Field and Laboratory Methods for Evaluating the Biological Integrity of SurfaceWaters. US Environmental Protection Agency. Washington DC. EPA/600/R-92/111.
872 J.M. Lazorchak et al.
3571 MARKERT3 23/10/02 9:46 am Page 872
USEPA, 1993b. A Review of Ecological Assessment Case Studies from a Risk Assessment Perspective.Risk Assessment Forum, Office of Research and Development, US Environmental Protection Agency,Washington, DC. EPA/630/R-92/005.
USEPA, 1993c. Methods for Measuring the Acute Toxicity of Effluents and Receiving Waters to Fresh-water and Marine and Marine Organisms. US Environmental Protection Agency Report. Office ofResearch and Development, Washington, DC. EPA/600/4–90/027F.
USEPA, 1994a. Short-term Methods for Estimating the Chronic Toxicity of Effluents and Receiving Watersto Freshwater Organisms (Lewis, P.A., Klemm, D.J., Lazorchak, J.M., Eds). US Environmental Pro-tection Agency Environmental Monitoring Systems Laboratory, Cincinnati. Report EPA/600/4–91/002
USEPA, 1994b. Methods for Measuring the Toxicity and Bioaccumulation of Sediment-associatedContaminants with Freshwater Invertebrate. US Environmental Protection Agency. Office of Researchand Development, Washington DC. Report 600/R-94–024.
USEPA, 1996. Proposed guidelines for ecological risk assessment. Federal Register 61(175), 47552–47631.USEPA, 1998a. Condition of the Mid Atlantic Estuaries. Washington, DC. EPA/600/R-98/147.USEPA, 1998b. National Water Quality Inventory: 1996 Report to Congress. Washington, DC. EPA/841/R-
97/008.USEPA, 1999. Rapid Bioassessment Protocols for Use in Streams and Wadeable Rivers: Periphyton,
Benthic Macroinvertebrates and Fish, 2nd edn. Office of Water, US Environmental Protection Agency,Washington, DC. EPA 841-B-99–002.
USEPA, 2001. A Survey of Fish Contamination in Small Wadeable Streams in the Mid-Atlantic Region.Office of Research and Development, Washington DC. February. EPA/600/R-00/107.
Vos, P., Hogers, R., Bleeker, M.R., van de Lee, T., Hornes, M., Fritjers, A., Pot, J., Peleman, J., Kuiper,M., Zabeau, M., 1995. AFLP: a new technique for DNA fingerprinting. Nucleic Acids Research 23,4407–4414.
Wallace, J.B., Grubaugh, J.W., Whiles, M.R., 1996. Biotic indices and stream ecosystem processes: resultsfrom an experimental study. Ecological Applications 6, 140–151.
Waller, W.T., Acevedo, M.F., Allen, H.J., Schwalm, F.U., 1996. University of North Texas, Institute ofApplied Sciences, Denton. Technical Report No. 172. Texas Water Resources Institute, The TexasA&M University System College Station.
Waser, N.M., 1993. Population structure, optimal outcrossing, and assortative mating in angiosperms. In:Thornhill, N.W. (Ed.), The Natural History of Inbreeding and Outbreeding: Theoretical and EmpiricalPerspectives. University of Chicago Press, Chicago, pp. 1–13.
Weber, C.I. (Eds), 1973. Biological Field and Laboratory Methods for Measuring the Quality of SurfaceWaters and Effluents. US Environmental Protection Agency, National Enforcement Research Center,Cincinnati. EPA-670/4-73-001.
Webster, J.R., 1975. Analysis of potassium and calcium dynamics in ecosystems on three southernAppalachian watersheds of contrasting vegetation. Ph.D. Dissertation, University of Georgia, Athens.
Webster, J.R., D’Angelo, D.J., Peters, G.T., 1991. Nitrate and phosphate uptake in streams at CoweetaHydrologic Laboratory. Verhandlungen der Internationalen Vereinengen Limnologie 24, 1681–1686.
Welsh, J., McClelland, M., 1990. Fingerprinting genomes using PCR with arbitrary primers. Nucleic AcidsResearch 18, 7213–7218.
Whittier, T.R., Paulsen, S.G., 1992. The surface waters component of the Environmental Monitoring andAssessment Program (EMAP): an overview. Journal of Aquatic Ecosystem Health 1, 119–126.
Whitton, B.A., Rott, E., Friedrich, G. (Eds), 1991. Use of Algae for Monitoring Rivers. Institut für Botanik,Universität Innsbruck, Innsbruck.
Williams, J.G., Kubelik, A.R., Livak, K.J., Rafalski, J.A., Tingey, S.V., 1990. DNA polymorphisms ampli-fied by arbitrary primers are useful as genetic markers. Nucleic Acids Research 18, 6531–6535.
Yoder, C.O., Rankin, E.T., 1995. Biological criteria program development and implementation in Ohio.In: Davis, W.S., Simon, T.P. (Eds), Biological Assessment and Criteria: Tools for Water ResourcePlanning and Decision Making. Lewis Publishers, Boca Raton, pp. 109–144.
Yeardley Jr, Roger B. Lazorchak, J.M., Paulsen, S., 1998. Elemental fish tissue contamination in north-eastern US lakes: evaluation of an approach to regional assessment. J. Envir. Tox. Chem. 17 (9),1875–1884.
