This paper reports a methodology to define and select basic activities for spatially variable N-management, referred to as management tracks. Their main purpose is to support decision making whether or not to apply variable nitrogen fertilisation. The methodology is based on biophysical simulation of crop growth and nitrogen leaching (WAVE) in combination with economic optimisation (linear programming) and enables a normative environmental-economic evaluation of site specific N-management to be made. The partial results of a case study with an input-intensive and an input-extensive crop (ware potato and winter wheat, respectively) showed that site specific nitrogen management led to positive returns over variable costs compared to uniform N-application, conditional on the validity of the WAVE model used in simulating yield effects. The investments that could be allowed for at maximum were 6,300 Dfl and 13,500 Dfl for winter wheat and ware potato, respectively, assuming application to an area of 100 ha. A pollution tax or a tax on nitrogen designed to internalise pollution costs in agricultural production raises these maximum amounts to 7,600 and 33,700 Dfl, respectively. Practical feasibility requires site-specific nitrogen management to be integrated with time specific management since optimal N-tracks were found to be highly weather dependent. Besides, spatially variable management cannot be achieved unless good farming practices (soil testing and crop scouting) are already in place.