TECHNICAL APPENDIX G

MULTIMETRIC APPROACH FOR DESCRIBING ECOLOGICAL CONDITIONS

[Much of the text and several figures in this issue paper were taken from Barbour and others (1995) and U.S. Environmental Agency (1994).]

The Intergovernmental Task Force on Monitoring Water Quality (ITFM) supports the national goal of using multimetric approaches for biological data in combination with information on physical and chemical indicators to assess water quality. Multimetric approaches to water-quality assessment, where locally modified, applied, and nationally compared, are a recommended component of a national assessment of the ecological condition of natural resources. The following steps are necessary to accomplish this goal nationally: establishment of reference conditions in the context of ecoregions/subecoregions; further development of information about the interrelations among biological, chemical, and physical characteristics of ecosystems; recognition of when local or regional modifications to the approach are needed; recognition that reference conditions are necessary to assess community-level responses at sites of interest; and establishment of a mechanism that allows data to be aggregated at appropriate regional or national levels.

The Multimetric Approach

The accurate assessment of biological integrity requires a method that integrates biotic responses by examining patterns and processes from individual to ecosystem levels. Classical approaches select some biological attribute that refers to a narrow range of perturbations or conditions. Many ecological studies focus on a limited number of parameters that may include one or more of the following: species distributions, abundance trends, standing crop, and production estimates. Usually, parameters are interpreted separately with a summary statement about the overall health of the system. This approach is limited in its usefulness because the attributes emphasized may not reflect overall ecological health (Karr and others, 1986). This is analogous to the removal of single-species toxicity testing from "environmental realism" and the low applicability for assessments of system-level responses (Buikema and Voshell, 1993).

An alternative approach is to define an array of metrics, each of which provides information on a biological attribute and, when integrated, functions as an overall indicator of biological conditions. The strength of a multimetric approach is its ability to integrate information from individual, population, community, and ecosystem levels and to evaluate, with reference to biogeography, a single ecologically based index of water quality (Karr and others, 1986; Plafkin and others, 1989; Karr, 1991; Barbour and others, 1995). Multimetric assessments provide detection capability over a broader range and nature of stressors and give a more complete picture of biological condition than single biological indicators. The Ohio Environmental Protection Agency (1987) suggested that combined strengths of metrics minimize any individual weaknesses.

Metrics
The validity of an integrated assessment that uses multiple metrics is supported by the use of metrics firmly rooted in sound ecological principles (Karr and others, 1986; Fausch and others, 1990; Lyons, 1992). A metric or biological attribute is some feature or characteristic of the biotic assemblage that reflects ambient conditions, especially the influence of human actions. A composite of appropriate metrics provides an accurate reflection of the biological condition at a study site. A large number of metrics have been used (for example, see Karr and others, 1986; Fausch and others, 1990; Kay, 1990; Noss, 1990; Karr, 1991), and each is essentially a hypothesis about the relations between an instream condition and human influences (Fausch and others, 1990). Gray (1989) stated that the three best-documented responses to environmental stressors are reduction in species richness, change in species composition to dominance by opportunistic species, and reduction in mean size of organisms. However, because each feature responds to different stressors, the best approach to assessment is to incorporate many attributes into the assessment process. These metrics can be surrogate measures of more complicated elements and processes as long as they have a strong ecological foundation and enable the biologist to ascertain the attainment or nonattainment of biological criteria, designated uses, or some other statement on ecological condition.

A number of metrics have been developed and subsequently tested in field surveys of benthic macroinvertebrate and fish assemblages (Karr, 1991). Because metrics have been recommended for fish assemblages (Karr, 1981; Karr and others, 1986) and benthic macroinvertebrates (Ohio Environmental Protection Agency, 1987; Plafkin and others, 1989; Barbour and others, 1992; Karr and Kerans, 1992), they will not be reviewed extensively here. A list of fish assemblage metrics used in the Index of Biotic Integrity (IBI) is presented in table 1, which includes local variations used in regional IBI applications.

Table 1. Index of Biotic Integrity metrics used in various regions of North America

Benthic metrics have undergone similar evolutionary developments and are documented in the Invertebrate Community Index (ICI) (Ohio Environmental Protection Agency, 1987), the Rapid Bioassessment Protocols (RBP's) (Shackleford, 1988; Plafkin and others, 1989; Barbour and others, 1992; Hayslip, 1993) and the benthic IBI (Kerans and others, 1992). Metrics used in these indices are surrogate measures of elements and processes of the macroinvertebrate assemblage. Although several of these indices are regionally developed, some are more broadly based and may be appropriate for use in various regions of the country. Selected metrics are listed by specific approach in table 2. Winget and Mangum (1979) and Mangum (1986) developed and tested the Biotic Condition Index (BCI), which is a metric similar to the Hilsenhoff Biotic Index (Hilsenhoff, 1987). The BCI incorporates characteristics of aquatic insect taxonomic diversity with tolerance characteristics on the basis of stream gradient, substrate composition, total alkalinity, and sulfate (U.S. Forest Service, 1989).

Table 2. Examples of metric suites used for analysis of macroinvertebrate assemblages

Figure 1 illustrates a conceptual structure for attributes of a biotic assemblage in an integrated assessment that reflects overall biological condition. A number of these attributes can be characterized by metrics within four general classes--community structure, taxonomic composition, individual condition, and biological processes.

Figure 1. Organizational structure of the attributes that should be incorporated into biological assessments.

Community structure can be measured by the variety and distribution of individuals among taxa. Taxa richness, or the number of distinct taxa, reflects the diversity within a sample of an assemblage. Multimetric uses of taxa richness as a key metric include the ICI (Ohio Environmental Protection Agency, 1987), the Fish IBI (Karr and others, 1986), the Invertebrate IBI (Kerans and others, 1992), and the RBP's (Plafkin and others, 1989). Taxonomic richness also is recommended as critical information in assays of natural phytoplankton assemblages (Schelske, 1984). Taxa richness usually is species level but also can be evaluated at designated groupings of taxa, often at higher taxonomic levels (that is, genus, family, order) in assessments of invertebrate assemblages.

Relative abundance of taxa refers to the number of individuals of one taxon compared with that within the entire sample. Dominance, which is measured as percent composition of the dominant taxon (Barbour and others, 1992), is an indicator of community balance or lack thereof. Dominance is an important indicator when the most significant taxa are eliminated from the assemblage or if the food source is altered. Dominants-in-common (Shackleford, 1988) is a comparison with reference conditions to evaluate the extent to which dominance may reflect human influence.

Taxonomic composition can be characterized by several classes of information, such as identity and sensitivity. Identity is the knowledge of individual taxa and associated ecological principles and environmental requirements. Key taxa, which are those of special interest or are ecologically important, provide information that is important to the identity of the targeted assemblages. The presence of exotics or nuisance species may be an important aspect of biotic interactions that relates to identity and sensitivity. Sensitivity refers to the numbers of pollutant-tolerant and pollutant-intolerant species in the sample. The ICI (Ohio Environmental Protection Agency, 1987) and the RBP's (Plafkin and others, 1989) use a metric based on species tolerance values. A similar metric for fish assemblages is included in the IBI (table 1).

Recognition of rare, endangered, or important taxa provides additional legal support for remediation activities or recommendations. Species status for response guilds of bird assemblages (for example, whether they are threatened or endangered, native or introduced, or of some commercial or recreational value) also relates to the composition class of metrics (Brooks and others, 1991).

Individual condition metrics are those that refer to the degradation of physical or physiological health of individual organisms. This type of metric is not commonly used for benthic macroinvertebrates; examples of fish metrics for individual condition are "percent individuals diseased" and "percent individuals with fin rot."

The functional aspects of biological processes can be divided into several categories as potential metrics. Trophic dynamics encompass functional feeding groups and relate to the energy source for the system, the identity of the herbivores and carnivores, the presence of detritivores in the system, and the relative representation of the functional groups. Abundance estimates are surrogate measures of standing crop and density that can relate to contaminant and enrichment problems.

Inferences on the biological condition can often be drawn from a knowledge of the capacity of the system to support the survival and propagation of the top carnivore. This attribute can be a surrogate measure for predation rate. Without stable food dynamics, populations of the top carnivore reflect stressed conditions. Likewise, if production at a site is considered to be high on the basis of organism abundance or biomass and if high production is natural for the habitat type under study (as per reference conditions), then biological conditions would be considered to be good. Fitness is the capacity of an individual or population to maximize reproductive success by the production of viable offspring (Price, 1975) and figures significantly in recruitment rate. Life cycle success, therefore, should include age-specific birth and death rates.

Process metrics have been developed for a number of different assemblages. For example, table 1 indicates at least seven IBI metrics that deal with trophic status or feeding behavior in fish, which focuses on insectivores, omnivores, or herbivores. Also, the number or density of individuals of fish in a sample (or an estimate of standing crop) is a measure of production and, thus, in the function class of metrics. Additional information is gained from density measures when they are considered to be relative to size or age distribution. Three RBP metrics for benthic macroinvertebrates focus on functional feeding groups (table 2; Plafkin and others, 1989; Barbour and others, 1992). Brooks and others (1991) used trophic level as one category for rating avian assemblages. It may not be necessary to establish metrics for every attribute of the targeted assemblage. However, the integration of information from several attributes, especially a grouping of metrics representative of the four major classes of attributes (fig. 1), would improve and strengthen the overall bioassessment.

Development of Metrics

The development of appropriate metrics follows definition of the taxa to be sampled, the biological characteristics at reference conditions, and, to a certain extent, the anthropogenic influences being assessed. In many situations, because multiple stressors impact ecological resources, a specific cause-and-effect assessment may be difficult. However, change over sets of metrics in response to perturbation by certain stressors (or sets thereof) may be used as response signatures (Yoder, 1991). A broad approach for program-directed development of metrics may be modeled after Fausch and others (1990), Holland (1990), Barbour and others (1992), or Karr and Kerans (1992). Candidate metrics (fig. 2) are selected on the basis of knowledge of aquatic systems, flora and fauna, literature reviews, and historical data. Candidate metrics are then evaluated for efficacy and validity for implementation into the bioassessment program. Less-robust metrics or those not well founded in ecological principles are eliminated in this research process. Metrics with little or no relation to stressors are rejected. Core metrics are those remaining that provide information useful in differentiating among sites that have good- and poor-quality biotic characteristics. Core metrics should be selected to represent diverse aspects of structure, composition, individual health, or processes of the aquatic biota. Together, they should form the foundation for a sound integrated analysis of the biotic condition to judge the attainment of biological integrity. Thus, metrics that reflect community characteristics are appropriate in biocriteria programs if their relevance can be demonstrated, if the response range can be verified and documented, and if the potential for program application exists. Regional variation in metric details are expected, but the general principles used to define metrics need to be consistent over wide geographic areas (Miller and others, 1988).

Figure 2. Tiered metric development process (Adopted from Holland, 1990).

Calibration of Metrics

Pilot studies or small-scale research may be needed to define, evaluate, and calibrate metrics. Metrics can be calibrated by using controlled prospective studies (Jongman and others, 1987); that is, by evaluating the response of metric values to varying levels of stressors. Sites must be carefully selected for controlled prospective studies so that a wide range of suspected stressors on the stream ecosystems can be included. In general, impaired sites are selected because single and combined stressors have impacted them. The selected impaired sites and the reference sites are the basis for the development of an empirical model of metric response to stressors.

Metrics can be evaluated following model development. Candidate metrics that do not respond to any of the stressors expected in a region may be eliminated. Metrics also are evaluated for variability with respect to responsiveness; those with high variability compared with the range of response should be used with caution. A more-detailed discussion of metric calibration is provided in Technical Appendix F.

Rating the Metrics
Once the reference condition is established from a compiled set of reference sites, the expectations for each metric can be delineated. Certain metrics may exhibit a continuum of expectations that are dependent on specific physical attributes of the reference streams. For example, the total number of fish species changes as a function of stream size estimated by stream order or watershed area for a number of undisturbed reference sites (Fausch, and others 1984). When reference site data are plotted, the points produce a distinct right triangle, the hypotenuse of which approximates the upper limit of species richness. Fausch and others (1984) suggested that a line with a slope fit to include about 95 percent of the sites is an appropriate approximation of a maximum line of expectations for the metric in question. When different stream classes have different expectations in metric values and a covariat, such as drainage area, exists that produces a monotonic response in a metric, a plot of survey data for each stream class versus the covariante may be useful (fig. 3).

As shown in figure 3, the area on the graph beneath the maximum line can then be trisected or quadrisected to assign scores to a range of metric values. It should be noted that as drainage area increases, there is a leveling or diminishing rate of increase in the number of species, which accounts for the bending of the lines. Even so, the upper line represents the maximum-species richness across the range of drainage area (Yoder and Rankin, 1995). The scores provide the transformation of values to a consistent measurement scale to group information from several metrics for analysis. An alternative is to calculate the median, 25th, and 75th percentiles and display the results in a box and whisker graph (fig. 4). For each metric, the sites are sorted by stream class (for example, ecoregion, stream type, and so forth) and plotted to ascertain the spread in data and the ability to discriminate among classes. If such a representation of the data does not allow discrimination of the classes, then it will not be necessary to develop a separate biocriterion for each class; that is, a single criterion will be applicable to a set of sites that represent different physical classes. Conversely, if differences in the biological attribute are apparent and appear to correspond to the classification, then separate criteria are necessary.

Figure 3. Examples of the technique used to calibrate the Index of Biotic Integrity (IBI) and the Invertebrate Community Index (ICI) for the drainage area dependent metrics of each index. The number of fish species (A) and number of mayfly taxa (B) vs. drainage area demonstrate the use of the 95 percent maximum-species richness line and the trisection and quadrisection methods used to establish the IBI and the ICI scoring criteria (Yoder and Rankin 1995).

For each metric, which is based on the distribution of metric values in the reference data base, scoring categories are developed on the basis of different percentiles of the observed range of individual metrics. For example, a reference data base has a maximum taxa richness value of 28 and a median value of 21. The scoring categories, which use the 50th percentile as the most appropriate threshold, would appear as follows


  Metric value ranges      Bioassessment          Condition category,  
      points                                        by percentiles
        21                       6                       50th
     14 -- 20                    4                    25th -- 49th
      7 -- 13                    2                    13th -- 24th 
      0 -- 6                     0                      <12th

Figure 4. Metric value and stream class to ascertain the spread in data and the ability to discriminate among classes.

With these types of categories established for all metrics, calculated values from test-site samples can be compared with the reference-based criteria for assignment of bioassessment scores. An alternative to assigning scores is to calculate the percent deviation from the maximum species richness line for each value obtained in calculating the metric from biological data collected at a site. In this approach, assessment of acceptability would be based on the percentage of reference value.
Aggregation
After defining the lower limit of the highest nonimpaired category and dividing the remainder of the value range into one or more impaired categories (fig. 3), actual metric values are substituted for the percentile limits of those category ranges. The ranges of metric values are put into scoring tables that provide the ability to associate bioassessment scores to individual metrics (for example, tables 3, 4), thus "normalizing" the values. The Ohio Environmental Protection Agency (1987) established tables that are based on some decided-upon percentiles as discussed above. As shown in table 3, they recognize three categories of metric scoring ranges for fish-assemblage data collected at nonwadeable (boat) sites.

After scoring all metrics for each of the sites, aggregation of these normalized metric scores is possible. By assuming equal weighting among metrics, a simple summation can accomplish aggregation. If the contribution of one or more metrics needs to be emphasized or increased over the remainder owing to, perhaps, specific recognition of known problems (habitat degradation or point-source discharges) and expected responses, then individual metrics can have a weighting factor incorporated. The weighting factor can be applied to either the calculated metric value or the normalized metric score.

Table 3. Index of Biotic Integrity metrics and scoring criteria based on fish-community data from more than 300 reference sites throughout Ohio applicable only to boat (nonwadeable) sites

[IBI, Index of Biotic Integrity, <, less than; >, greater than; �, less than or equal to.  Table modified from Ohio Environmental Protection Agency (1987).  For further informtion on metrics,  see Ohio Environmental Protection Agency (1987)]

            IBI metrics                             Scoring criteria 
                                                 5         3          1

Total number of species                         >20     10 - 20     <10
Percentage of round-bodied suckers              >38     19 - 38     <19
Number of sunfish species                        >3      2 - 3       <2
Number of sucker species                         >5      3 - 5       <3
Number of intolerant species                     >3      2 - 3       <2
Percentage of tolerant species                  <15     15 - 27     >27
Percentage of omnivores                         <16     16 - 28     >28
Percentage of insectivores                      >54     27 - 54     <27
Percentage of top carnivores                    >10      5 - 10      <5
Percentage of simple lithophils(1)              >50     25 - 50     <25
Percentage of deformities, eroded fins, 
    lesions, and tumors anomalies                <0.5  0.5 - 3.0    >3.0
Fish numbers                                   <200    200 - 450  >450

(1)For sites in a drainage area of less than or equal to 600 square miles; for sites in a drainage area greater than 600 square miles, scoring categories vary with drainage area.

Table 4. Bioassessment scoring criteria developed for Rapid Bioassessment Protocols benthic macroinvertebrate metrics based on 300-organism subsamples of double-composite square-meter kicknet samples from the Sandusky River in Ohio.

Unresolved Issues

The multimetric approach to biological assessment has been criticized because the reduction of taxonomic composition and abundance data to a handful of indices loses the rich information in the original data. Often, these criticisms do not consider how the indices are to be employed. Management acts on a small handful of societal actions; for example, regulation of point sources, controlling urban runoff, and fisheries management. Biological assessment must reduce the complexity of the ecosystem in such a way that management can act. For example, it is unrealistic to expect that the species composition of harpacticoid copepodes will be "managed" in streams. Final decisions on impact/no impact or management actions are not made on the single aggregated value alone; rather if comparisons to established reference values indicate an impairment in biological condition, then component parameters (or metrics) are examined for their individual effects on the aggregated value.

A larger issue is the statistical distribution, behavior, and uncertainty of indices and metrics generated in the multimetric approach. This issue will be resolved as the approach is increasingly adopted and data are generated and analyzed. Currently, the most pressing need is for side-by-side comparison of different analytical approaches; for example, multimetric assessment, multivariate community ordination, and multiple regression.

It is important to understand the effects of various stressors on the behavior of specific metrics. An often-stated concern is that IBI values will be misleading unless the sensitivity of the monitored populations to specific pollutants are well characterized. These concerns are often directed at the use of tolerance values inferred from incomplete field observations. Nonetheless, field fisheries biologists who have extensive local experience do, in fact, know the distribution and ecological requirements of the resident fish species. The general concept of integrating tolerance information with distributional data has been used successfully in a variety of situations (Karr and others, 1986; Mangum, 1986; Hilsenhoff, 1987; Ohio Environmental Protection Agency, 1987; Plafkin and others, 1989; F.A. Mangum and D.A. Duff, U.S. Forest Service, written commun., 1992).

References

Barbour, M.T., Plafkin, J.L., Bradley, B.P., Graves, C.G., and Wisseman, R.W., 1992, Evaluation of EPA's rapid bioassessment benthic metrics--Metric redundancy and variability among reference stream sites: Journal of Environmental Toxicology and Chemistry, v. 11, no. 4, p. 437-449.

Barbour, M.T., Stribling, J.B., Karr, J.R., 1995, Multimetric approach for establishing biocriteria and measuring biological condition, chap. 6 of Davis, W.S. and Simon, T.P., eds. Biological assessment and criteria--Tools for water resources planning and decision making: Boca Raton, Fla., Lewis Publishers, p. 63-77.

Brooks, R.P., Croonquist, M.J., D'Silva, E.T., Gallagher, J.E., and Arnold, D.E., 1991, Selection of biological indicators for integrating assessments of wetland, stream, and riparian habitats, in Biological criteria--Research and regulation: Washington, D.C., U.S. Environmental Protection Agency, Office of Water, EPA-440/5-91-005, p. 81--89.

Buikema, Jr., A.J., and Voshell, Jr., J.R., 1993, Toxicity studies using freshwater benthic macroinvertebrates, in D. M. Rosenberg and V. H. Resh, eds., Freshwater biomonitoring and benthic macroinvertebrates: New York, Chapman and Hall, chap. 10, p. 344-398.

Fausch, D.D., Karr, J.R., and Yant, P.R., 1984, Regional application of an index of biotic integrity based on stream fish communities: Transactions of the American Fisheries Society, v. 113, p. 39-55.

Fausch, K.D., Lyons, J., Karr, J.R., and Angermeier, P.L., 1990, Fish communities as indicators of environmental degradation: American Society Symposium, v. 8, p. 123-144.

Gray, J.S., 1989, Effects of environmental stress on species rich assemblages: Biological Journal of the Linnean Society, v. 37, p. 19-32.

Hayslip, G.A., 1993, EPA Region 10 in-stream biological monitoring hand book (for wadable streams in the Pacific Northwest): Seattle, Wash., U.S. Environmental Protection Agency, Environmental Services Division, EPA-910/9-92-013, 75p.

Hilsenhoff, W.L., 1987, An improved biotic index of organic stream pollution: Great Lakes Entomology, v. 20, p.31-39.

Holland, A.F., ed., 1990, Near Coastal Program Plan for 1990--Estuaries: Washington, D.C., U.S. Environmental Protection Agency, Environmental Research Laboratory, Office of Research and Development, EPA-600/9-90-033, 259 p.

Hughes, R.M., and Gammon, J.R., 1987, Longitudinal changes in fish assemblages and water quality in the Willamette River, Oregon: Transactions of the American Fisheries Society, v. 116, p. 196-209.

Jongman, R.H.G., terBraak, C.J.F., and van Tongeren, O.F.R, eds., 1987, Data analysis in community and landscape ecology: Wageningen, the Netherlands, Pudoc, 299 p.

Karr, J.R., 1981, Assessment of biotic integrity using fish communities: Fisheries, v. 6, no. 6, p. 21-27.

___1991, Biological integrity--A long-neglected aspect of water resource management: Ecological Applications, v. 1, p. 66-84.

Karr, J.R., Fausch, K.D., Angermeier, P.L., Yant, P.R., and Schlosser, I.J., 1986, Assessing biological integrity in running waters--A method and its rationale: Illinois Natural History Survey Special Publication 5, 28 p.

Karr, J.R. and Kerans, B.L., 1992, Components of biological integrity--Their definition and use in development of an invertebrate IBI, in Midwest Pollution Control Biologists Meeting, Chicago, Ill., 1991, Proceedings: U.S. Environmental Protection Agency, Region V, EPA-905/R-92-003, p. 1-16.

Kay, J.J., 1990. A non-equilibrium thermodynamic framework for discussing ecosystem integrity, in Edwards, C.J., and Regier, H.A., eds., An ecosystem approach to the integrity of the Great Lakes in turbulent times: Ann Arbor, Mich., Great Lakes Fisheries Commission Special Publication 90, p. 209-237.

Kerans, B.L., Karr, J.R., and Ahlstedt, S.A., 1992. Aquatic invertebrate assemblages--Spatial and temporal differences among sampling protocols: Journal of the North American Benthological Society, v. 11, no. 4, p.377-390.

Lyons, J., 1992, Using the index of biotic integrity (IBI) to measure environmental quality in warmwater streams of Wisconsin: U.S. Department of Agriculture, U.S. Forest Service, General Technical Report NC-149, 51p.

Mangum, F.A., 1986, Macroinvertebrates, chap. 32 of Cooperriderr, A.Y., and others, eds., Inventory and monitoring of wildlife habitat: Bureau of Land Management, 858 p.

Miller, D.L., Leonard, P.A., Hughes, R.M., Karr, J.R., Moyle, P.B., Schrader, L.H., Thompson, B.A., Daniels, R.A., Fausch, K.D., Fitzhugh, G.A., Gammon, J.R., Hailliwell, D.B., Angermeier, P.L., and Orth, D.J., 1988, Regional applications of an Index of Biotic Integrity for use in water resource management: Fisheries, v. 13, no. 5, p. 12-20.

Moyle, P.B., Brown, L.R., and Herbold, B., 1986, Final report on development and preliminary test of indices of biotic integrity for California: Corvallis, Oreg., U.S. Environmental Protection Agency, Environmental Research Laboratory, 17 p.

Noss, R.F., 1990, Indicators for monitoring biodiversity--A hierarchical approach: Conservation Biology, v. 4, p.355-364.

Ohio Environmental Protection Agency, 1987, Biological criteria for the protection of aquatic life: Ohio Environmental Protection Agency, Division of Water Quality Monitoring and Assessment, Surface Water Section, v. 1, 44 p.; v. 2, 251 p.; v. 3, 57 p.

Plafkin, J.L., Barbour, M.T., Porter, K.D., Gross, S.K., and Hughes, R.M., 1989, Rapid bioassessment protocols for use in streams and rivers--Benthic macroinvertebrates and fish: U.S. Environmental Protection Agency, Office of Water, Washington, D.C., EPA-440/4-89-001, 190 p.

Price, P.W., 1975, Insect ecology: New York, John Wiley and Sons, 514 p.1

Schelske, C.L., 1984, In situ and natural phytoplankton assemblage bioassays, chap. 2 of Shubert, L.E., ed., Algae as ecological indicators: London, Academic Press, p. 15-47.

Shackleford, B., 1988, Rapid bioassessments of lotic macroinvertebrate communities--Biocriteria development: Arkansas Department of Pollution Control and Ecology, 45 p.

Steedman, R.J., 1988, Modification and assessment of an index of biotic integrity to quantify stream quality in southern Ontario: Canadian Journal of Fisheries and Aquatic Science, v. 45, p. 492-501.

U.S. Environmental Protection Agency, 1994, Biological criteria.--Technical guidance for streams and small rivers: Washington, D.C., U.S. Environmental Protection Agency, Office of Water, EPA 822-B-94-001, 162 p.

U.S. Forest Service, 1989, Fisheries habitat surveys handbook, chap. 5 of Aquatic macroinvertebrate surveys: U.S. Forest Service, Intermountain Region Fisheries and Wildlife Management R--4 FSH 2609.23, 48 p.

Winget, R.N. and Mangum, F.A., 1979, Biotic condition index--Integrated biological, physical, and chemical stream parameters for management: Ogden Utah, U.S. Forest Service, Intermountain Region. 50 p.

Yoder, C.O., 1991, The integrated biosurvey as a tool for evaluation of aquatic life use attainment and impairment in Ohio surface waters, in Biological criteria--Research and regulation: Proceedings of a Symposium, U.S. Environmental Protection Agency, Office of Water, Washington, D.C., EPA 440-5-91-005, p. 110-122.

Yoder, C.O., and Rankin, E., 1995, Biological criteria program development and implementation in Ohio; chap. 9 of Davis, W.S., and Simon, T.P., eds., Biological assessment and criteria are tools for water resource planning and decision making: Boca Raton, Fla., Lewis Publishers, p. 109-144.

Return to ITFM Report Appendixes Table of Contents

Please e-mail comments to lkendrix@usgs.gov

Last modified: Fri Nov 8 16:05:26 1996