The thesis consists of four papers all dealing with phosphorus (P) regulation in the agricultural sector from either a theoretic or an empirical approach. Reducing P losses to the aquatic environment is important because P and nitrogen can damage the aquatic system by fuelling excessive algal growth and accelerating eutrophication of lakes and streams. In recent decades loss of P from the agricultural sector has caused eutrophication of streams and lakes across Europe and North America. The sector there seems to be the core of the problem is the intensive and increasing livestock sector which generates manure P well in excess of crop requirements. Because application of manure (and the regulation thereof) is primarily determined by nitrogen crop nutrition, livestock farmers are not motivated to reduce their surplus P applications. Continued application of P in excess of crop requirements causes P stocks to build up in fields and, over time, this increases the risk of losing P to the aquatic environment.

The first paper identifies the most important parameters to include in a dynamic model where farm profit is maximized over time whilst taking soil-P dynamics into account. The second paper completes the modelling framework from paper 1 and analyses how a tax on P surpluses motivates the different farmer types to utilize the soil-P stock and implement measures to reduce P loss. The third and fourth papers are empirical, where a farm profit maximization model is developed for an area in the catchment area of Odense Fjord, Denmark. The two papers investigate two different general incentives for reducing soil P losses: a tax on mineral-fertilizer P and a tax on P surplus and a targeted tax on the phosphorus index (PI) for erosion. The PI estimates the risk of P being lost to the environment from critical source areas taking spatial parameters into account. With reference to the joint findings of all four papers it is concluded that a tax on P surplus can be a core element in a close-to-efficient policy regulating P losses from the agricultural sector. However, the general tax fails to regulate high risk areas efficiently and could therefore be supplemented with a targeted policy for implementation of filter strips at high risk fields.