Abstract
Predicting water quantity and quality from agricultural watersheds is critical for effective management of agroecosystems. The ArcGIS-based Agricultural Policy Environmental eXtender (ArcAPEX) model (version 1501) was employed to simulate streamflow, sediment, and nutrient yield in an agricultural watershed in northeastern Arkansas, and to evaluate scenario-based conservation practices. The ArcAPEX model was set up using topographical, soil, agronomical, and meteorological data. Measured daily data used for model calibration and validation included streamflow and sediment and nutrient yields derived from analysis of in-stream water samples collected from the Little River Ditches watershed in Northeastern Arkansas from 2015 to 2022. Sensitivity analysis, model calibration, and validation were conducted using the APEX auto-calibration and uncertainty estimator. The simulated streamflow and sediment, total phosphorus (TP), and total nitrogen (TN) yields were within the model evaluation criteria with higher reliability. The average annual streamflow and sediment, TP, and TN yields measured from the watershed were 667 mm, 3.39 tons/ha, 4.69 kg/ha, and 5.4 kg/ha and those from the ArcAPEX model simulations were 628 mm, 3.96 ton/ha, 5.41 kg/ha, and 5.69 kg/ha, respectively, during the study period. The scenario-based ArcAPEX model demonstrated that the combination of cover crops and filter strips reduced the runoff, sediment, and nutrient yields most effectively. Implementing cover crops and filter strips across 30% of the watershed reduced sediment yield by 26%, more than any other pollutant. This study validated ArcAPEX as an effective tool for simulating surface runoff and sediment and nutrient yields in agricultural watersheds and predicting the impact of conservation practices, demonstrating its value for complex agroecosystem management.