Report on optimum PFTs definition

Abstract

This document reports on the numerical techniques applied to define plant functional types (PFTs) for Mediterranean forests. The analysis presented here uses data gathered at the leaf, individual and stand level. The key assumption is that the way suites of functional characters coordinate can reveal life history strategies, that are associated with the local environmental conditions. In order to avoid an "a-priory" PFTs definition based on taxonomy (this analysis has been made in deliverable 7) we use a multivariate technique to define the PFTs based on their trait values. The outputs of this technique are then compared with the a classic taxonomy-based definition. By defining these new PFTs we provide to the vegetation modelling community a set of Mediterranean PFTs with parameter (traits) estimates that can be readily used when simulating forest dynamics following the classic approach of representing diversity with PFT, i.e. the first-generation of vegetation dynamics models. In an ecological perspective we validate to what extent taxonomy can be used to categorize species in terms of their structure and function. All analyses were made with the R programming language (R Development Core Team, 2015).