A SWAT-based optimization tool for obtaining cost-effective strategies for agricultural conservation practice implementation at watershed scales

To address the harmful algal blooms problem in Lake Erie, one solution is to determine the most cost-effective strategies for implementing agricultural best management practices (BMPs) in the Maumee River watershed. An optimization tool, which combines multi-objective optimization algorithms, SWAT (Soil and Water Assessment Tool), and a computational efficient framework, was created to optimally identify agricultural BMPs at watershed scales. The optimization tool was demonstrated in the Matson Ditch watershed, an agricultural watershed in the Maumee River basin considering critical areas (25% of the watershed with the greatest pollutant loadings per area) and the entire watershed. The initial implementation of BMPs with low expenditures greatly reduced pollutant loadings; beyond certain levels of pollutant reductions, additional expenditures resulted in less significant reductions in pollutant loadings. Compared to optimization for the entire watershed, optimization in critical areas can greatly reduce computational time and obtain similar optimization results for initial reductions in pollutant loadings, which were 10% for Dissolved Reactive Phosphorus (DRP) and 38% for Total Phosphorus (TP); however, for greater reductions in pollutant loadings, critical area optimization was less cost-effective. With the target of simultaneously reducing March–July DRP/TP losses by 40%, the optimized scenario that reduced DRP losses by 40% was found to reduce 51.1% of TP; however, the optimized scenario that reduced TP losses by 40% can only decrease 11.3% of DRP. The optimization tool can help stakeholders identify optimal types, quantities, and spatial locations of BMPs that can maximize reductions in pollutant loadings with the lowest BMP costs.