Eutrophication of coastal waters is a recognised problem in estuaries around the globe. To analyse cross-boundary water quality management, such as protection of the Baltic Sea, economic nutrient abatement models commonly operate on a large scale, grouping river systems to large catchment areas. Theoretical deliberation suggests that modelling abatement in such a way removes the opportunity of targeting measures to the most vulnerable regions within the catchment, while overestimating the capacity of abatement measures in the upstream areas. However, the implications on designing environmental policy depend on catchment characteristics and the stringency of the abatement targets. In this study a model of the catchment area is built with zones that are an increasing distance away from the coast to show what kind of bias in the optimal abatement set is caused by the assumption of spatial homogeneity. By solving the model with and without the zones for good ecological status at the South-West coast of Sweden, it is shown that while assuming homogeneity prevents from perceiving the abatement measures where they would be the most effective, it also leads to ignoring spatial limitations that are more relevant to a subset of abatement measures, such as the wetlands and buffer zones. Following from the relatively large nitrogen abatement targets, the overall effect of assuming catchment homogeneity is to underestimate the abatement costs and to overestimate the potential of wetlands to reduce nitrogen.