The following section provides a justification for the reasons why we are proposing the use of these alternative methods of decision support for TMDL decisionmaking. In early 2001, the General Accounting Office (GAO) highlighted the pervasive lack of data at the State level available to set water-quality standards, to determine what waters are impaired, and to develop TMDLs. Subsequent to the GAO report and after issuance by the EPA of updated TMDL regulations, Congress requested that the National Research Council (NRC) assess the scientific basis of the TMDL program, including:
1) the information required to identify sources of pollutant loadings and their respective contributions to water-quality impairment,
2) the information required to allocate reductions in pollutant loadings among sources,
3) whether such information is available for use by the states and whether such information, if available, is reliable, and
4) if such information is unavailable or is unreliable, what methodologies should be used to obtain such information
|
The following comments represent the consensus opinion of the eight-member NRC committee assembled to complete this task:
• TMDL plans should employ adaptive implementation to ensure that the TMDL program is not halted because of a lack of data and information but rather progresses while better data are collected and analyzed with the intent of improving upon initial TMDL plans.
• The EPA should endorse statistical approaches for the characterization of natural processes in the watershed, proper monitoring design, data analysis, and impairment assessment.
• Biologic criteria should be used in conjunction with physical and chemical criteria to determine whether a water body is meeting its designated use (as defined by the TMDL). Criteria should be defined in terms of magnitude, frequency, and duration. Models should effectively link environmental stressors (and control actions) to biologic responses.
• Uncertainty must be explicitly acknowledged both in the models selected to develop TMDLs and in the results generated by those models.
• The margin of safety concept should account for uncertainties in water-quality data, as well as for the variation in background (natural) water-quality contributions, and should reflect the reliability of the models used for estimating load capacity. A margin of safety should be determined through a formal uncertainty and error-propagation analysis.
• To carry out adaptive implementation, the EPA needs to foster the use of strategies that combine monitoring and modeling and expedite TMDL development.
• The EPA needs to provide guidance on model application so that thorough uncertainty analyses will become a standard component of TMDL studies.
A predictive model useful for TMDL decision support ideally should have the following characteristics:
• The model focuses on the water-quality standard. The model is designed to quantitatively link management options to meaningful response variables. Thus, it is desirable to define the TMDL endpoints (e.g., pollutant sources and standard violation parameter) and incorporate the entire “chain” from stressors to response in the modeling analysis.
• The model is consistent with scientific theory. The model does not err in process characterization.
• Model prediction uncertainty is reported. Given the reality of incomplete information, it makes sense to explicitly acknowledge the levels of prediction uncertainty for various management options. Providing decisionmakers with an understanding of the risks of various options and allowing them to factor this understanding into their decisions.
• The model is appropriate to the complexity of the situation. Simple water quality problems can be addressed with simple models, but complex water-quality problems may or may not require the use of complex models.
• The model is consistent with the amount of data available. Models requiring large amounts of monitoring data should not be used in situations where such data are unavailable.
• The model results are credible to stakeholders. Given the increasing role of stakeholders in the TMDL process, modelers may have to provide more than a cursory explanation of the predictive model.
• Cost for annual model support is an acceptable long-term expense. Given growth and change, water-quality management will not end with the initial TMDL determination. The cost of maintaining and updating the model must be tolerable over the long term.
• The model is flexible enough to allow updates and improvements. Research can be expected to improve scientific understanding, leading to refinements in models. (National Research Council, 2001)
The decision-analytical approach being proposed by the USGS WGSC takes into consideration these recommendations for supporting TMDL decisionmaking. This approach could help solve some of these problems by providing a simple way for watershed participants to analyze various mitigation scenarios to meet their discharge-permit requirements. The approach could be designed to serve the following functions:
1) provide potential market participants and other stakeholders with background information on mercury offsets;
2) provide municipal treatment works and industrial plants with tools for estimating releases of mercury and methyl-mercury loading to surface waters from their operations, exploring reduction options, and estimating the costs of achieving reductions;
3) help market participants identify potential offset projects;
4) track the volume and type of offset within a watershed;
5) share lessons learned about offsets across watersheds where it is being tried or considered;
6) provide information on how offsets can be used to address water-quality problems; and
7) record past successes and failures of offset and other market-based programs.
|
 |
|
| This diagram illustrates the complexities of mercury fate and transformation. The physical uncertainties, mercury loading from mountain sources, the chemical uncertainties, methylation processes that occur in fine-grained sediment and in low-oxygen conditions, and biological uncertainties, food-web dynamics and biomagnification of methylmercury, warrant an alternative model that explicitly includes these uncertainties in the decisionmaking process (Alpers, 2000). |
|
|
|
