Suitability of arable weeds as indicator organisms to evaluate species conservation effects of management in agricultural ecosystems

The overall objective of this study is to examine the application of arable weeds as indicator organisms of biodiversity in agro-ecosystems to evaluate species conservation effects of management practices. Both investigations of interactions between weeds with heterotrophic consumers and strong overall correlations between the number of weed species and the total species diversity indicate that arable weeds are "key species", the loss of which leads to serious changes in the remaining biocoenosis via habitat and food chain relations. The assessment of the value of management measures for species conservation presupposes a strong relation of target organisms to land use practice. In arable fields, the high percentage of dormant seeds reduces the relevance of single cultivation measures against weed populations and emphasizes the significance of long lasting management including cultivation systems, crop rotations, or field edge effects. A comparison of the number of weed species found at different times and frequencies of sampling indicates-especially in herbicide treated fields-the importance of two recording times-one before the application and one before harvest. This ensures a better estimation of the total species spectrum. To evaluate plant species diversity in fields the number of characteristic arable weeds is proposed. In contrast to the total number of species, the typical arable weeds do not include species frequently occurring outside fields. Thus, several highly noxious species like Cirsium arvense, Elymus repens and Galium aparine are not positively valued. Differences in rarity and usefulness could lead to a more sophisticated evaluation of single weed species and the species spectrum. The use of different number of species as threshold values for different soil types cannot be recommended, however.