SUMMARYImplementation of the EU Nitrate Directive in Germany will result in nitrogen (N) balance surpluses being restricted to 60 kg N/ha averaged over 3 years, starting in 2009. With N surpluses of more than 100 kg N/ha, winter oilseed rape (OSR) is a main contributor to N balance surpluses in OSR-based crop rotations in northern Germany. The exact calculation of N fertilization rates therefore becomes increasingly important in order to meet the target of less than 60 kg N/ha N balance average surplus over 3 years at a farm level. Currently, soil mineral nitrogen (SMN) at the beginning of spring growth is commonly used as an indicator for calculation of N fertilization rates in spring. However, amounts of SMN at the beginning of spring growth under OSR are usually low and canopy N is only taken into account to a very limited extent. This might lead to N fertilization rates exceeding the optimum N fertilization rate (N_opt). In the present study, the effects of SMN in spring and of canopy N in autumn and spring on N_opt were investigated. Multi-site field trials producing different crop canopies, as a result of two sowing dates and two autumn N fertilization levels, with five spring N fertilization levels (0–280 kg N/ha) were carried out in 2005/06 and 2006/07.N_opt in spring was estimated by quadratic response functions using the combine-harvested seed yield data from the spring N fertilization treatments. Regression analyses revealed no relationship between N_opt and SMN at the beginning of spring growth or canopy N at the beginning of spring growth. In contrast, a significant negative correlation between N_opt and canopy N at the end of autumn growth was found. Based on the results of the present study, it is sensible to take autumn canopy N into account when calculating N fertilization rates in spring. If canopy N in autumn is high (>50 kg N/ha), as a consequence, N fertilization rates should be reduced.