Land use change is a major driver of biodiversity patterns, therefore conservation management in cultivated landscapes should seek to optimize land use diversity. Especially under changing environmental conditions there is an increasing need of identifying management options for preserving biodiversity. However, the design of historical data sets is often inappropriate for detecting biodiversity responses to ongoing rapid changes. Here, we present an approach to quantify plant species diversity and relate it to land use diversity.

 Data mining took place at the landscape scale in two mountainous regions of Central Europe, differing in natural and cultural history. Within these landscapes a representative systematic rectangular grid (7 x 7 plots of 1 ha) was sampled. At each plot polygons of uniform land use were mapped and presence-absence data of plant species were recorded.
Plant species diversity differed significantly between landscapes: species richness and within-landscape beta diversity in the calcareous mosaic landscape was higher than in the siliceous mosaic landscape. Land use diversity explained the significant variation in species richness. The relationship between plant species diversity and land use diversity is consistent in different cultivated landscapes and on multiple scales.
The chosen sampling approach integrates the advantage of random but grid-based sampling with land use polygon specific information. This enables not only to investigate also similarity pattern (in land use and species composition) but also an integration of further information on the patch scale, if needed.