Contrasting diversity patterns of epigeic arthropods between grasslands of high and low agronomic potential

Abstract in Undetermined Increasing demand for food, fuel and fibre promotes the intensification of land-use, particularly in areas favourable for agricultural production. In less-favourable areas, more wildlife-friendly farming systems are often either abandoned or under pressure of conversion, e.g. for bioenergy production. This raises the question, to which extent areas of different agronomicpotential contribute to regional biodiversity. To approach this question on a regional scale, we established our study within a region where sites of high and lowagronomicpotential (AP) alternate on a small spatial scale. We selected 13 high-AP and 13 low-AP grasslands to quantify the contribution of these classes to the regional diversity of four epigeicarthropod taxa (ants, springtails, functional groups of ground beetles, and spiders). The regional diversity (γ) was partitioned into species richness per site (α-diversity), diversity among sites within one class (βwithin-diversity), and diversity between the two classes (βbetween-diversity). The β-diversity generally accounted for the largest share of the γ-diversity, with patterns of diversity components being highly taxon- and class-specific. Carnivorous carabids had a higherα-diversity at high-AP sites. Ants, springtails, and cursorial spiders had a higherβwithin-diversity in low-AP grasslands. Low-AP sites also harboured many more species that occurred exclusively in one grassland class. We conclude that grasslands that may be unfavourable for agricultural production contributed more to regional diversity of epigeicarthropods than favourable grasslands. We therefore suggest that future agricultural schemes should promote arthropod biodiversity by specifically targeting agri-environment schemes or other wildlife-friendly farming approaches to areas of lowagronomicpotential, since this bears the greatest potential to preserve a comparatively high species turnover (β-diversity) and in consequence high regional diversity