Antoine Gardarin and Muriel Valantin-MorisonPlease use the format "First name initials family name" as in "Marie S. Curie, Niels H. D. Bohr, Albert Einstein, John R. R. Tolkien, Donna T. Strickland"
<p>1. Ground dwelling arthropods are affected by agricultural practices, and analyses of their responses to different crop management are required. The sampling efficiency of pitfall traps has been widely studied in natural ecosystems. In arable agroecosystems, arthropod communities are more simplified than in natural ones and sampling techniques need to be adjusted to these specific conditions. In particular, the ability to distinguish between simplified communities and the sampling effort required have been little investigated. We evaluated the suitability of pitfall traps for characterizing the effects of arable cropping systems on the taxonomic and functional composition of spider and carabid communities. 2. In a field experiment comparing three cropping systems, we compared the effects of two pitfall trap diameters, the type of preserving fluid used in pitfall traps and the sampling effort on six metrics describing communities: activity-density, richness and community weighted mean (CWM) of body size, each one for carabid and spiders. 3. Trap size affected the observed composition of carabid and spider communities, with large traps yielding a higher relative proportion of spiders, and a higher species richness and CWM body size for both taxa. The type of preserving fluid had no marked effect on any of the metrics considered. In the case of our experiment conducted in arable crops, simulations with various sampling efforts showed that only very different communities could be significantly distinguished with less than ten traps per field or less than 30 field-year replicates. The relationship between the arthropod community differences and the minimum sampling effort required to detect it was similar for activity-density and species or genus richness metrics. Fewer traps were required to find differences between cropping systems for CWM body size than for other metrics. For the three arable cropping systems studied here, carabid activity-density, carabid CWM body size and spider genus richness were the variables better distinguishing between cropping systems with the smallest sampling effort. 4. The minimum sampling effort required for community comparisons under different arable cropping systems was smaller for functional composition than for activity-density in case of spiders and richness in case of carabids. The trap design, arthropod community metrics and crops selected were the principal levers for optimizing the trade-off between sampling effort and the ability to detect arthropod community responses to cropping systems.</p>
activity-density, arthropod community, computer simulation, cropping system, CWM body size, ground beetle, pitfall trap, sampling effort, species richness, spider
Agroecology, Biodiversity, Biological control, Community ecology
