Learning from a long-term crop rotation experiment

A crop rotation experiment was established in 1996/97 at three locations representing differ-ent soil types and climates. Three factors were tested: i) crop rotation with different propor-tions of N2-fixing crops, ii) with and without a catch crop, and iii) with and without animal manure. A green manure crop increased yields in the following cereal crops, but at the rota-tional level, total yields were larger in crop rotations without a green manure crop. There were positive effects of animal manure and catch crops on yield. However, except for the coarse sandy soil, the yield effects of catch crops and animal manure decreased over time when a grass-clover green manure was included in the rotation. The problems with perennial weeds increased over time depending on crop rotation and use of catch crops. This stresses the im-portance of considering long-term effects in the evaluation of crop management measures.