Abstract
Weeds remain a challenge for organic arable production systems. However, when densities are kept under control, the presence of weed flora can actually provide a range of beneficial functions, both ecologically and agronomically. Additionally, the yield loss potential of the weed community is linked to its species composition. Plant species communities are dynamic, and traits are selected by different filters, of which crop management is one. This sparked the theory that: a greater crop diversity will lead to greater diversity of weed traits, thereby reducing the dominance of single traits and, ultimately, of deleterious weed populations. The application of crop diversity, in the form of intercropping, undersowing, and winter catch crops, could foster this low density but high diversity state. In order to understand this interaction, weed and management data was collected from organic spring cereals fields in 5 different regions in Northern Europe. In addition to the field data, a literature database of problematic weed species was compiled in order to be able to focus on specific weed species or species groups. In several analyses the effect of crop diversity strategies on weed density, diversity, community and traits was studied. First, the crop diversity and site effects, in the form of soil variables, were studied in Germany. Secondly, the regional use of crop diversity, and effects on weeds were studied on organic farms in Denmark, Germany, Finland, Sweden and Latvia. Thirdly, the same dataset was used to test the crop diversity measures for traits selection in the weed community. Results show that direct weed control lowers weed densities, but long term strategies such as winter catch crop use and organic management increase weed diversity, without increasing densities. When looking into trait selection, only a few traits were influenced by crop management and diversity. The traits that were selected for, were not considered problematic. Still, clear and recognizable functional weed groups were observed. The low trait selection is possibly due to the high diversity in well-established organic fields, which increases functional redundancy. These results are promising for both the use of diverse crop management, as well as the implementation of organic management.