Description:

This is the second year of funding for the New England SPARROW (Spatially Referenced Regressions on Watershed Attributes) model. Funds in the first year (along with funds allocated for projects supporting Nutrient-Criteria development) were used to analyze regional results from the national SPARROW model. Databases and modeling results from the national model were used to plan development of a New England SPARROW model. The New England SPARROW model will be developed using funds from programs supporting TMDLs, Nutrient-Criteria Development, and the Long Island Sound (LIS) program, as well as matching funds from the USGS. The SPARROW Model is a useful tool for pollutant load and source assessments. The strength of the model is that it defines an empirical relation between actual in-stream water-quality measurements with watershed conditions and pollution sources to determine predicted pollutant loads in streams and the source of the loads. The model is also a tool for making predictions of pollutant concentrations and loads and pollutant sources where you have no existing water quality information. Because SPARROW is a statistical model, it will also provide confidence intervals for each prediction to help assess the effectiveness of the model. The proposed New England SPARROW model will provide assessments for nitrogen and phosphorus throughout the entire New England area. Approximately 50,000 stream segments will be assessed; the watersheds for these stream segments average about 3-4 sq. mile in size. SPARROW will provide a level of detail for a regional model that would be cost and time prohibitive with other existing deterministic water-quality models. Specifically, the New England SPARROW model will provide the following information that would be useful to the USEPA Region 1 TMDL program: -Estimated mean annual loads of nitrogen and phosphorus in all New England stream segments for the mid-1990s time period. This includes streams with both existing and no water-quality data, and confidence intervals for all determinations and predictions. Estimated mean annual concentrations will also be available. -Estimates for the amounts of N and P contributed by pollutant sources in each stream segment and confidence intervals for these source estimates. Sources to be assessed in the model include major land use types (urban, forest, wetland, and agricultural), point sources, atmospheric deposition (for N only), agricultural fertilizer and manure applications, and human population. Watershed characteristics and their impacts on pollutant loads that will also be assessed include precipitation, soils, reservoirs and lakes, and watershed slope. -Estimates of the contributions from individual stream segments to downstream stream N and P loads within the basin and to coastal waters. This will include estimates of how much one state is providing to the stream pollutant loads in another state. In summary, we believe that SPARROW can provide first-cut determinations of pollutant sources in streams and determine the potential complexities of detailed TMDL modeling needs. For larger river basins and regional analysis, SPARROW can provide the necessary information for TMDL determinations that more conventional deterministic models may or may not be able to provide

Record Details: