Hydrologic and non-point source pollution risk analysis on agricultural watersheds

Gabriel G Vazquez-Amabile, Purdue University


Non-point source (NPS) pollution caused by nutrients and pesticides is a problem of significant interest in agricultural watersheds. The main goal of this study was the development of a decision support system (DSS) to compute hydrologic and NPS pollution risk in agricultural watersheds using a comprehensive hydrologic model. The SWAT model (Soil Water Assessment Tool) was selected for this purpose. SWAT was validated to predict daily and monthly in-stream atrazine concentration in the St. Joseph River watershed in northeast Indiana using nine years of data. Even though the model was somewhat inaccurate for predicting atrazine levels at specific points, SWAT performed well in predicting the general atrazine concentration trend in streams over time for daily and monthly time intervals. The estimation of the timing of atrazine application was very important in the calibration and validation processes and proved to be a key input to predict the amount and timing of pesticide released to streams. The model was deemed suitable to conduct NPS pollution risk analyses for atrazine by computation of exceedance probability curves and thematic and probability maps at a basin scale. Time series models combined with SWAT were explored to perform one-month-ahead forecasts of groundwater table depth and streamflow. After being calibrated and validated, SWAT was run to generate 50 years of records to make it possible to fit time series models to streamflow and groundwater monthly series without using long historical records. This strategy was validated in four watersheds in northeast Indiana and two watersheds in southeast Indiana, and it proved to be useful to forecast groundwater table depth for areas that lack long series of records. Finally, an analytical tool, named AGRIRISK, was proposed as an add-on application integrated to AVSWAT2000 to create a DSS. SWAT-AGRIRISK is intended to be used in agricultural watersheds to perform hydrologic and NPS pollution risk analyses, the later as a complement to the GLEAMS (Groundwater Loading Effects of Agricultural Management Systems)-NAPRA (National Agricultural Pesticide Risk Analysis) system which cannot be readily applied to large watersheds. The system has been conceived to quantify and communicate the risk in the decision-making processes at regional levels, comparing soils and management scenarios.




Engel, Purdue University.

Subject Area

Agricultural engineering|Hydrology|Soil sciences

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server