Abstract
Adopting a multifaceted approach of biodiversity is believed to capture different aspects of the ecosystem functioning and it is thus advised for conservation prioritisation, especially for anthropogenic ecosystems but this key topic has never been conducted for the Mediterranean tree assemblages, despite their ecological importance. We explored how the multi-faceted diversity of woody plant assemblages, as measured by taxonomic (TD), functional (FD) and phylogenetic (PD) diversities, are distributed over space in the French part of the Mediterranean biome, and to which extend they are spatially correlated to each other, in order to understand whether one facet can be used as a proxy for another, especially for conservation management purposes. We analysed data from 5885 plots over the study area. We estimated several FD types by considering separately (i) regeneration, (ii) morphology traits (iii) modalities on species ecological properties, and finally (iv) considering all traits together. We used the Rao quadratic entropy to estimate the TD, FD and PD diversity facets. We tested for the links and spatial correlation (congruence) levels among these facets, using simultaneous au-toregressive (SAR) models and partial Mantel tests. Spatial structure varied among diversity facets and spatial autocorrelation patterns were identified for all diversity indices from 30 to 50 km distances. We observed a functional convergence and a phylogenetic divergence within tree assemblages comparing to the ones expected given the regional species pool, indicating that even in tree communities with functional similarities, phylogenetic diversity may be high. PD was zero to slightly congruent to FD, regardless the type of functional traits considered. The highest SAR slope (=0.3) and partial Mantel test (=0.2) were revealed between the PD and the FD based on species morphological traits, but still remained considerably low. Each one of the diversity indicators reflected a different tree community spatial pattern. Functional diversity patterns varied according to the type of traits considered. Using only taxonomic indices may be misleading for responding to Mediterranean tree assemblages conservation needs and additional information about the species phylogeny and functional responses to disturbance pressures should be considered in large scale analyses.