Spatial variability more influential than soil pH and land relief on thermophilous vegetation in overgrown coppice oak forests

The overgrown coppice oak forests that cover the southern slopes beneath the foothills of the Sudetes (Silesia, Central Europe) are considered to be Euro-Siberian steppic woods with a Quercus sp. habitat (91I0): a priority habitat in the European Union, according to the Natura 2000 system. In subcontinental parts of Central Europe, thermophilous oak forest vegetation is found extrazonally; its presence, in the study site, is related to previous coppice management. In this paper we explore the differentiation of the vegetation caused both by land-relief derived variables (potential heat load, slope inclination and exposition, soil depth) and soil pH, as well as spatial processes. The data on the vegetation were collected from 117 regularly arranged sampling plots, located in three mountain ranges. The vegetation consisted of a mixture of species considered as typical for different habitats (mesophilous forests, acidophilous forests, thermophilous oak forests, grassland, thermophilous fringes and mesophilous mantle) and was relatively rich in species. Many of the species found were rare and are protected in Poland. The results of the bioindication, on the basis of Ellenberg indicator values, suggest the pH gradient to be the most important, followed by the insolation/moisture gradient, to the differentiation of the studied vegetation. The thermophilous oak forests seem to occupy the niche between acidophilous and mesophilous forest. However the decomposition of spatial variation, assessed on the basis of semivariance values of the vegetation similarity coefficient (frequency index), emphasizes a strong differentiation of vegetation between sites and mountain ranges. The results of canonical correspondence analysis, performed on a spatially stratified sub-set of the data, revealed a stronger effect caused by spatial variation (32.7% of explained species variation) than environmental variables, such as soil pH and potential heat load (13.1%). Since the shared variation was low (1.8%), it showed a strong influence of spatial processes, revealing the effect of the local species pool.