Background and aims – Monitoring plots are important tools for forest management and can help to predict the directions of changes in species composition in a changing climate. Understanding ecological processes in forest communities is one of the most important goals for a proper management of these ecosystems and for biodiversity conservation. Methods – The compositional data recorded within an inventory were used here to describe plant community composition and investigate how it is affected by climatic or other ecological factors. Six large forests along an altitudinal gradient owned and managed by the regional administration of Tuscany were chosen as study sites. 109 plots of 400 m ₂ were randomly located according to a stratified sampling design, with the number of plots in each forest related to forest size. The number of stems and DBH of all plants with DBH > 3 cm were recorded for each woody species within each plot. In addition, the frequency of understory species was recorded by twelve subplots of 0.25 m ₂, randomly located within each quadrant of each plot. Key results – Elevation, soil water capacity, minimum precipitation, annual precipitation/temperature ratio and maximum temperature were found to be the explanatory variables for the gradient of woody species composition. After removing plantation plots from analyses, only elevation, soil water capacity and sum of temperature of the months over 5ºC were retained as explanatory variables of this gradient. Elevation, soil organic matter and the sum of winter precipitation were the only environmental variables that were related to the compositional gradient of understory species. Conclusions – Present species composition of these forests is unlikely to remain in a next future according to the IPCC climate predictions, whose moderate scenarios predict in the Mediterranean area a severe decrease in precipitations and a raise of 3–4 degrees on average temperature.

CLIMATE; DCA; FOREST ECOLOGY; FOREST ECOSYSTEMS; GRADIENT ANALYSIS; VEGETATION ANALYSIS