Use of direct gradient analysis to study the climate-vegetation relationships in Galicia, Spain

This paper reports a bioclimatic analysis of plant species in Galicia, NW Spain. A set of floristic data obtained from 150 plots located at euclimatopes (sites with monitored climate) was analysed using di- rect gradient analysis and clustering with respect to the 8 climatic variables thought to play a major role in regulating the distribution of the species considered in the study area. Principal component analysis (PCA) and hierarchical clustering were based on a matrix of species by climatic variables. Indicator taxa for the variables were identified on the basis of their Indicator values (Brisse & Grandjouan 1978) and grouped by cluster analysis. The groups produced were compatible with the results of principal com- ponent analyis and the frequential analysis of the species, which identified their phytoclimatic nature. The groups were then characterized by determining their climatic positions and indicator values with respect to the chief climatic variables. The first three PCA axes, which were associated with Oceanity, Mean minimum temperatures and the temperature range in the coldest month, together accounted for 97.2 ~o of the variance of the data.