Aquatic Life Use Support (ALUS)
Biocriteria Links
ITFM Recommendations for Monitoring
The Intergovernmental Task Force on Monitoring Water Quality (ITFM) was formed in 1992 to develop recommendations on monitoring to achieve more comparable and scientifically defensible information, interpretations, and evaluations of water-quality conditions across the nation. The ITFM comprised both Federal and State agencies responsible for monitoring and assessment programs as well as an associated advisory committee including municipalities, academia, industry, etc.. The ITFM subsequently developed a model for stream monitoring for different types of designated uses based on a combination of biological, physical, and chemical monitoring (Figure 1). The model defines the relationship between parameters that directly measure the condition of the biotic community and its response over time to stressors, such as fish and benthic macroinvertebrate indices, and parameters that measure either stressors or exposure of organisms to stressors, such as levels of pH, nutrients, and toxicants. For streams, EPA recommends that States incorporate ITFM's suite of parameters in their monitoring programs for evaluating attainment of designated uses. These are general recommendations to consider when developing and revising monitoring programs. For example, monitoring for aquatic life use would include the base monitoring program parameters in the box--community level biological data from at least two assemblages, habitat, and physical/chemical field parameters--plus ionic strength, nutrients, and toxicants in water and sediment.
The ITFM in May 1997 became a permanent National Water Quality Monitoring Council to facilitate, among other tasks, the development and implementation of the recommendations on specific methods for measuring the parameters shown in Figure 1. Standard methods for measuring the chemical parameters and conducting toxicity tests are well established among the States, but methods for biological and habitat assessments are not standardized for all types of waterbodies. Recent work by the Ohio EPA suggests that bioassessment methods differ widely in their accuracy and discriminatory power for aquatic life use determinations. Ohio evaluated a hierarchy of bioassessment approaches relevant to differing levels of rigor and confidence. In their State, Ohio EPA found that less intensive bioassessment approaches tend to be accurate in detecting impairment, but may give a false indication of full support in reaches where the methods are not rigorous enough to detect subtle problems.
ITFM recommends that to combine data for assessment, monitoring data produced by different organizations should be comparable, of known quality, available for integration with information from a variety of sources, and easily aggregated spatially and temporally. This is important at a variety of scales, up to and including national assessments. If different methods are similar with respect to the quality of data each produces, then data from those methods may be used interchangeably or together. As data quality (i.e., precision, sensitivity) increases, the confidence in the assessment increases. Data quality objectives should be defined for each method so that assessments can be validated by imposing a known level of confidence in the results.
Monitoring Design
Any monitoring and assessment program begins with setting goals and a monitoring design that can meet those goals. The history of water quality monitoring is replete with programs that could not answer key questions. Examples include:
- A watershed study where the monitoring organization assumes that flow data can be obtained after the fact based on "reference point" measurements from bridges, only to learn later that many streams lack the channel morphometry to develop a stage-discharge relationship;
- An intensive survey where the laboratory's detection levels for metals prove inadequate to detect even concentrations above water quality standards;
- A basin survey where management or the legislature poses the question "What is the statistical trend in biological condition of our streams?" too late to be incorporated into the monitoring design.
EPA has a goal of comprehensively characterizing the Nation's streams,
rivers, lakes, wetlands, estuaries, and shorelines. These assessments
will include monitored and evaluated assessments and may involve
probability-based as well as targeted monitoring. To achieve this
goal, EPA encourages States to incorporate a formal process of goal
setting and monitoring design while meeting their own State-specific
goals. ITFM provides general
guidelines
for the topics to consider in monitoring design in a technical appendix
of its final report.
Data Quality Objectives
The Data Quality Objectives (DQO) process developed by EPA's Quality Assurance Management Staff is a specific approach to monitoring design that has been applied to monitoring programs in all media. The DQO process involves the stakeholders in the program in the design. Stakeholders itemize and clarify the questions being asked of a monitoring program, including the required level of accuracy in the answers. Generally, these questions are stated in quantitative terms ("What are the index of biotic integrity [IBI] and invertebrate community index [ICI] values for wadeable streams in Big River Basin, and what is the trend in IBI across the basin, with 80 percent certainty?"), and statistical methods may be recommended for selecting sites or sampling frequency. For information about DQOs for water quality monitoring contact the Assessment and Watershed Protection Division at (202) 260-7023.
Approaches
To date, States have taken three main approaches to monitoring a large portion of their waterbodies:
- Fixed-station networks with hundreds or thousands of sites (most large networks have been reduced in the past 10 years)
- Rotating basin surveys with a large number of monitoring sites covering thousands of miles of waters (Ohio EPA's bioassessment program)
- Rotating basin surveys with a probabilistic monitoring design; a statistically valid set of sites are selected for sampling in each basin (Delaware's benthic macroinvertebrate program).