|Chin, David - UNIV OF MIAMI|
|Sakura-Lemessy, Donna - ALBANY STATE UNIV|
|Gay, Paige - UNIV OF GA|
Submitted to: Journal of Water Resources Planning and Management
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: June 26, 2009
Publication Date: April 15, 2010
Citation: Chin, D.A., Sakura-Lemessy, D., Bosch, D.D., Gay, P. 2010. Enhanced Estimation of Terrestrial Loadings for TMDLs: Normalization Approach. Journal of Water Resources Planning and Management. 136(3)357-365. Interpretive Summary: Based upon U.S. Environmental Protection Agency data, over 40% of the assessed waters in the United States do not meet applicable water quality standards. This has resulted in the formation of the total maximum daily load (TMDL) program, a program designed to asses these waters and to develop plans for improving their conditions. The assessments are typically based upon observed data and computer simulations produced using deterministic natural resource models. However, the assessments and the subsequent plans normally do not account for model and data uncertainty. A protocol for incorporating these uncertainties into the assessment is demonstrated by application to fecal coliforms (FCs) concentrations in streamflow in the Little River Experimental Watershed in South Georgia, U.S.A. A demonstration using the U.S. Environmental Protection Agency’s (EPA) HSPF Model (Hydrologic Simulation Program-Fortran) suggests that interflow concentrations of FCs would need to be reduced by at least 90% to meet conventional water-quality standards with a risk of less than 10%.
Technical Abstract: TMDL implementation plans to remediate pathogen-impaired streams are usually based on deterministic terrestrial fate and transport (DTFT) models. A novel protocol is proposed that can effectively, efficiently, and explicitly capture the predictive uncertainty of DTFT models used to establish terrestrial loadings in support of TMDLs. The proposed protocol is implemented in three phases. In the first phase the DTFT model is calibrated using flow and water quality measurements, in the second phase the loading parameters associated with TMDL implementation are varied between their upper and lower practical bounds and the predictive responses of the DTFT model are recorded, and in the third phase the predictive responses are combined with the frequency distribution of the observed contaminant concentrations to extract the probability distribution of concentration reductions as a function of terrestrial loading reductions. The protocol is demonstrated by application to the Little River Experimental Watershed in Georgia, where it is shown that interflow concentrations of fecal coliforms (FCs) would need to be reduced by at least 90% to meet conventional water-quality standards with a risk of less than 10%.