Size and composition of the weed seedbank after 7 years of different cover-crop-maize management systems

The objective of this study was to obtain detailed information on the long-term weed suppression potential of four winter soil cover types included in an arable crop system managed at various input levels. We used weed seedbank size and composition to assess weed suppression potential. A field experiment was established in 1993 as a split-split-plot design with four replications, including two tillage systems [a conventional system (CS) including ploughing in the cover crops and a low-input system (LIS) including no tillage with surface mulching of the cover crops] in the main plots, three mineral nitrogen fertilization rates for the main crop in the sub-plots and four soil cover types (main crop residue, rye, crimson clover and subterranean clover) in the sub-sub-plots. Seedbank sampling took place in winter 2000/01. The weed seedbank was analysed with the seedling emergence method. Data were analysed using ANOVA and multivariate techniques. Results indicated that the seedbank density in the LIS was about five times higher than in the conventional input system. In the CS, use of a rye cover crop resulted in a lower seedbank density with respect to the crop residue treatment (-25%), whereas in the LIS the subterranean clover cover crop decreased weed seedbank density as compared with the other cover crops and the crop residue treatment (-22% on average). Differences in species composition were mainly related to tillage system. Implications for cover crop management and the development of sustainable cropping systems are discussed.