Despite the intense strain that various socioeconomic factors place on agriculture in the United States, few models can accurately depict the cross-system interactions occurring on and around farms. This research attempts to address the disconnect between policies, models, and the processes that they are representing and regulating by using the Soil and Water Assessment Tool+ (SWAT+) to simulate the coastal agricultural region of Eastern North Carolina. We used SWAT+ to simulate farming practices and soil properties in this agriculturally productive, coastal watershed. Our research question was: How can we use SWAT+ to simulate the effects of differing soil properties on crop yield and nitrate loss? To answer this question, we ran numerous trials within SWAT+ model with varying soil properties and analyzed their effect on crop yield and nitrate loss. Our validation and calibration processes were primarily in formed by USDA North Carolina yield results. We found that the model’s output was most sensitive to changes in parameters relating to soil structure and partitioning of precipitated water (curve number and curve number soil water factor). The curve number parameters were the most effective in optimizing yield across the watershed’s top three crops: corn, cotton, and soybean, as well as decreasing nitrate lost from each crop. Our model was more sensitive to the curve number parameters than the parameter relating to water availability (surface runoff lag). These results can be used to support land use management practices that address the utility of improved soil structure on crop yield and reduced nitrate loss. Therefore, our results suggest that land use management practices that increased soil health are at a confluence of interests between farmers, environmentalists, and consumers alike. These results can more accurately inform agricultural policies that reduce the strain that socio-economic and ecological processes currently place on agriculture.