Quantifying species richness at multiple spatial scales in a Natura 2000 network

Even if the establishment of nature reserves is to date a reality and the increase of protected areas is going to grow year after year, monitoring programs aiming to assess the effectiveness of the established protected areas for biodiversity conservation are still needed. That is the case for the Natura 2000 network in Europe, for which monitoring methods and programs are not yet well-established. A probabilistic sampling procedure is proposed and tested for quantifying and monitoring plant species diversity within a local network of protected areas, namely the Natura 2000 network in the Siena Province, Italy. On the basis of a sampling strategy of one 100 m 2 plot randomly located in each 1 km +ù 1 km cell, four Sites of Community Importance (SCIs) were investigated in 2005. The gradients in species composition at the plot scale were largely related to elevation and forest cover. The species richness values of the four SCIs were compared by means of sample-based rarefaction curves. Then, additive partitioning of species richness was applied to determine the most important spatial components in determining the total species richness of the network. Compositional differences among the plots within each SCI were the most responsible of the total species richness. These methodologies can be adopted for assessing plant species richness within a large region or within a reserve network and, if combined with additive partitioning, they can be used as a set of large scale indicators of species diversit