Agricultural fertilizer contributes substantially to nitrogen pollution throughout the world, leading to many negative impacts including ecological dead zones. Alternative crop management practices, such as cover crops and perennial crops, can limit nitrogen pollution. To optimize land use changes to meet potential nitrate reduction scenarios, we develop a flexible geo-spatial economic framework, balancing nitrate reduction with reductions in farm profit. Cover crop and perennial crop patterns at the hydrologic response unit (HRU) level are simulated with a novel and more realistic management unit approach by combining Soil and Water Assessment Tool model outputs and an economic programming model. We apply our framework to a major Minnesota River Basin watershed, the Cottonwood River watershed, in the state of Minnesota, USA, finding that strategically located cover crops can be used to achieve significant nitrate effluent reduction, if perennial crops are optimally placed as a part of a collective effort. A large driver of nitrate pollution and reduction in profit is yearly variation, a proxy for precipitation volume-indicating that climate change may be particularly impactful in areas where climate change models predict significant changes in precipitation patterns.