![[logo] US EPA](http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
Biological Endpoints and Thresholds
Inherent in this element is the existence of a viable data management system commensurate with the level of rigor (see Technical Element 9, Data Management). State bioassessment programs should implement index development and threshold selection. Index development can include single or multiple metrics, discriminant models, or other predictive models of the aquatic assemblage. Thresholds are the benchmarks from which the biological condition needed to support designated uses are described. The index is developed and verified by independent data sets. Then the attainment threshold is established and documented. Selecting this threshold is perhaps the most critical aspect in reporting and documenting attainment status. States typically establish this threshold and then add other thresholds to distinguish among higher (e.g., outstanding natural resource waters, excellent warmwater habitat, or excellent/good habitat) and lower assessment categories (e.g., limited resource waters, fair/poor/very poor).
Numerous methods are available for analyzing biological indicator data to assess attainment status, including both univariate and multivariate analysis techniques. The most common method of data analysis is use of a multimetric index, which combines several biological variables into a single, unitless index. These variables, or metrics, are characteristics of the biota that change in some predictable way with increased human influence. Use of multiple metrics to assess biological conditions maximizes the information available regarding the functions and processes of aquatic assemblages. For a metric to be of value, it must be ecologically relevant to the biological assemblage or index under study and to the specified program objectives, and it must be sensitive to stressors. RIVPACS and its derivatives are empirical (statistical) models that predict the aquatic macroinvertebrates that would be expected to occur at a site in the absence of environmental stress. A comparison of the invertebrates predicted to occur at the test sites with those actually collected (i.e., Observed/Expected) provides a measure of biological impairment. The predicted taxa list also provides a “target” community to measure the success of any remediation measures taken to rectify identified impacts.
Regardless of approach, the primary purpose of an analytical threshold is to establish levels of biological quality that can be used in determining attainment or non-attainment of the designated use. USEPA recommends that each State establish its analytical threshold based on either index values from a statistical distribution of candidate reference sites or a discriminant model from a range of aquatic life conditions that includes reference conditions. Estimates of variance, such as standard deviation, and power analysis can assist in determining how many assessment levels an index may represent. States and Tribes must carefully document their rationale for selecting thresholds, including those that define gradations in quality or attainment status such as "good/fair/poor" or "full/partial attainment/nonattainment." The thresholds should promote straightforward decisions when biological data are compared against them to facilitate water quality management decisions. State decisions for applying the threshold also need to be fully documented.
Some Frequently Asked Questions
Question: Are thresholds set for reference conditions or for aquatic
life uses?
Answer: Both. In the first instance, a threshold for reference condition
anchors the top tier of the BCG. This threshold will likely be commensurate
with one of the highest aquatic life use designations. In the second instance,
thresholds are established to differentiate attainable biocriteria for each
aquatic life use. As an example, thresholds were set in Ohio for two designated
uses, Warmwater Habitat (WWH) and Modified Warmwater Habitat (MWH). The 25th
percentile of least disturbed regional reference sites was used to set numeric
biocriteria for the WWH use designation (Figure below); an exception was the
HELP ecoregion where there were no least disturbed reference sites available.
The 25th percentile of modified reference sites that represent best attainable
conditions for channelized streams was used to set numeric biocriteria for
the MWH use designation.
Question: Why is the 25th percentile of a reference site population used to set thresholds?
Answer: Other percentiles can be used. However, the 25th percentile is relatively conservative and reflects the idea that the lower quality reference sites from a population of “least disturbed” may not be reference sites. Choosing the lowest quality reference site to set thresholds is not sufficiently robust to provide consistently accurate assessments.
Learn more about biological endpoints and thesholds:
- Best Practices for Identifying Reference Condition in Mid-Atlantic Streams (August 2006) (8 pp, 1.7MB, About PDF)
- Biological Monitoring and Assessment: Using Multimetric Indexes Effectively (155 pp, 10.7MB About PDF)
- Development and Application of Multimetric Indexes in Developing Biological Indicators: Lessons Learned from Mid-Atlantic Streams
- Linking Human Disturbance To Biological Change in Developing Biological Indicators: Lessons Learned from Mid-Atlantic Streams
- Multivariate Statstical Analysis Often Overlook Biological Knowledge in Lessons Learned for Bioassessment
- Integrating Multiple Factors in Lessons Learned for Bioassessment
- Metrics and the Index of Biotic Integrity
- Multimetric Indices to Prepare and Analyze Data
- Present Results for Decision-Making