An optimization algorithm linked with a nonpoint source (NPS) pollution model can be used to optimize NPS pollution control strategies on a field-by-field basis in a watershed by maximizing NPS pollution reduction and net monetary return. In this paper a methodology is described which integrated a genetic algorithm (GA) (an optimization algorithm) with a continuous simulation, watershed-scale, NPS pollution model, Annualized Agricultural Non-Point Source Pollution model (AnnAGNPS) to optimize the selection of best management practices (BMP) on a field-by-field basis for an entire watershed. To test the methodology, optimization analysis was performed for a U.S. Department of Agriculture experimental watershed in Pennsylvania to identify BMPs that minimized long-term (over a 4-year period) water quality degradation and maximized net farm return on an annual basis. Results indicate that the GA was able to identify BMP schemes that reduced pollutant load by as much as 56% and increased net annual return by 109%.