This article illustrates a statistical approach for deriving farm economic impacts of policy options aiming to reduce nitrate diffuse pollution. Building upon Fezzi et al. (2008), who assess the costs of Water Framework Directive-related measures on farm accounts data, we estimate regression models allowing such costs to be predicted for any region for which land use patterns and livestock numbers are known. We derive economic impacts in terms of changes in farm gross margin for (a) reducing inorganic fertilizer application, (b) reducing livestock stocking rates, and (c) converting arable land to ungrazed grassland. A case study of the agriculturally diverse Yorkshire Derwent catchment, in the North of England, demonstrates the overall approach. In addition, for this illustration, we combine these cost estimates with prediction of the water quality changes arising from each measure, derived via an integrated hydrological model of the Derwent. This allows a comparison of cost-effectiveness. Finally, we implement our spatially explicit approach to target the measures to specific subcatchments identified as being of particular environmental policy interest.