Development of Methodologies for Incorporating the Copper Biotic Ligand Model (BLM) into Aquatic Life Criteria
(U.S. Environmental Protection Agency
Office of Science and Technology
Health and Ecological Criteria Division)
Great Lakes Environmental Center is supporting EPAs efforts to incorporate the Biotic Ligand Model (BLM; Santore et al., 2000) into the nationally recommended aquatic life criteria for copper. The BLM describes and quantifies the bioavailability of certain metals to aquatic life. EPA is now considering incorporating the BLM directly into the nationally recommended aquatic life criteria for copper, and is investigating methodologies for doing so. GLEC is conducting analyses that will aid the Agency in developing implementation guidance for a BLM-based criteria, and to assemble the additional water chemistry required to calculate BLM-normalized criteria. Specifically, this includes:
- Preparing BLM model input from Table 1 (acute toxicity of copper to freshwater animals) data;
- Testing standard toxicity water to characterize dissolved organic carbon (DOC) and other essential BLM input parameter ([Ca 2+], [Mg 2+], [Na+], alkalinity, and pH) concentrations;
- Identifying and retrieving data for copper BLM input parameters measured in U.S. waters, using NASQAN and other high-quality datasets, and creating a relational database of these data;
- Development and application of probabilistic techniques and statistical analyses to address the following questions:
Since BLM-calculated water quality criteria will be time-variable (according to the variability of the water quality parameters which serve as inputs), what methods can be used to calculate a single fixed value for a site criterion?
How do varying sample sizes affect the uncertainty of a BLM-calculated criterion, and how can this uncertainty be accounted for in a nationally-applicable methodology?
Monte Carlo analysis, using synthetic water quality data based upon observed statistical properties, has been used to address these issues. Using the statistical properties of the surface water datasets, a Latin Hypercube Monte Carlo generator program was used to simulate a large number of events (realizations) for each BLM test scenario. The generator program also applied Iman and Conover's (1982) algorithm for inducing the specified degree of rank correlation between water quality parameters, without altering the statistical distributions of individual parameters. In addition to providing EPA with guidance on these specific issues, analysis of the uncertainty of BLM predictions was also conducted.
Back to Project Experience