Skip to main content
ARS Home » Southeast Area » Tifton, Georgia » Southeast Watershed Research » Research » Publications at this Location » Publication #166371
Title: USING USDA-AGRICULTURAL RESEARCH SERVICE LONG-TERM WATERSHED HYDROLOGY AND WATER QUALITY OBSERVATIONS TO ESTIMATE THE QUALITY OF MODEL PREDICTIONS OF THE CONSERVATION EFFECTS ASSESSMENT PROGRAM (CEAP)

Author
item Strickland, Timothy
item Sullivan, Dana
item Wauchope, Robert - Don
item Bosch, David
item Potter, Thomas

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 5/30/2004
Publication Date: 8/1/2004
Citation: Strickland, T.C., Sullivan, D.G., Wauchope, R.D., Bosch, D.D., Potter, T.L. 2004. Using USDA-ARS long term watershed hydrology and water quality observations to estimate the quality of model predictions of the conservation effects assessment program (CEAP). (Abstract) American Chemical Society AGRO Division, Washington, D.C., Picogram 67:75 #138.

Interpretive Summary:

Technical Abstract: The CEAP program of the Natural Resources Conservation Service (NRCS) is an assessment, at a large-watershed scale, of the effectiveness of USDA's resource conservation programs in protecting water quality. Hydrologic/water quality models will be used to make these assessments. Several USDA-ARS research groups will 'ground truth' these assessments by developing estimates of the uncertainty of the model estimates. In this presentation we will describe an approach we are developing that takes advantage of the rich database of hydrologic and water quality data available from the Little River Watershed at Tifton, GA. The scientists and collaborators of the ARS Southeast Watershed Research Laboratory (SEWRL) are combining input data sets (e.g., soils, weather, soil conservation practices, agrochemicals use and land use) at higher resolution in both time and space than the national program, in order to estimate the sensitivity of prediction errors of the models being applied to the national-level assessment. Response differences in input data resolution and/or model selection will be used to estimate the confidence on predictions at those larger scales.