An integrative computational methodology is developed for the management of non-point source pollution from watersheds. The method is based on an interface between evolutionary algorithms (EAs) and a comprehensive watershed simulation model known as Soil and Water Assessment Tool (SWAT). The associated decision support model is capable of identifying optimal land use patterns to satisfy environmental and economic related objectives. The suitability of SWAT to the study is carefully investigated, and a global sensitivity analysis model is developed in order to identify the most influential simulation parameters that need to be calibrated. An automatic calibration model, which is based on a genetic algorithm (GA), is used to improve the accuracy of daily streamflow and sediment yield predictions by SWAT. The Generalized Likelihood Uncertainty Estimation methodology is implemented to investigate uncertainty of SWAT estimates, accounting for errors due to model structure, input data, and model parameters. Finally, the calibrated SWAT is linked with a GA for single objective evaluations, and with a Strength Pareto Evolutionary Algorithm for multiobjective optimization. The model can be operated at small spatial scales, such as a farm field, or on a larger watershed scale. Application of the decision support model to a demonstration watershed located in southern Illinois reveals the capability of the model in achieving its intended goals. However, the model is found to be computationally demanding as a direct consequence of repeated SWAT simulations during the search for optimal land use patterns. An artificial neural network (ANN) is developed to mimic SWAT outputs and ultimately replace it during the search for an optimal solution. The replacement resulted in 86 percent reduction in computational time. The ANN model is trained using a hybrid of evolutionary programming (EP) and the back propagation (BP) algorithms. The hybrid algorithm was found to be more effective and efficient than either EP or BP alone. Overall, this study demonstrates the powerful and multifaceted role that EAs, artificial intelligence techniques, and comprehensive simulation models can play in solving complex and realistic problems within science and engineering.