Abstract
Executive Summary As recent as 2005 the main ecological and evolutionary processes underlying the origin and maintenance of (botanical) diversity gradients was considered as one of the 25 most important fundamental questions in science still unanswered. The present study aims at further understanding diversity gradients by capitalizing on recent developments that uncovered a number of ecological and evolutionary processes that play a role in explaining diversity gradients; including ecological filtering under both current and past climatic conditions, dispersal limitation, and phylogenetic patterns of niche- and functional trait evolution. The objectives of the present study are to evaluate the role(s) and relative contribution of these processes that lead to the present formation of botanical diversity gradients, and to extrapolate this to modelled future climatic conditions (in the Sundaland hotspot). In this project, we first will reconstruct the gradient of botanical diversity with the use of species distribution models (SDMs). SDMs predict the potential distribution of species by interpolating identified relationships between presence-only herbarium collection data, and high spatial resolution (here 5 arc-minutes; c. 10×10 km) environmental data. We will introduce the use of dispersal kernels, calibrated to functional traits of dispersal capacity, to correct for the maximum dispersal distance from candidate glacial refugia since the last glacial maximum, 21k years ago. The candidate glacial refugia themselves will be identified by hind-casting the SDMs onto the ecological conditions of this strong environmental filter of the past. We will also take dispersal barriers, defined as unsuitable habitat conditions, into account. This will result in SDMs that more closely reflect realized species? distributions than before. Superimposing the independent SDMs will allow reconstruction of a more accurate gradient of botanical diversity. An added advantage of the use of SDMs is that for every locality the species composition can be predicted, which in turn allows the reconstruction of patterns of phylogenetic community structure and distribution of functional traits. Partitioning the variance in the diversity gradient and phylogenetic community structure simultaneously to functional traits of species and to environmental filters will allow to assess which part of the variance in the diversity gradient can be attributed to phylogenetic relatedness, and which part to environmental filters and functional traits. Finally, we will project the realized SDMs to future IPCC climate scenarios to identify those areas where most species can survive in the future, and thereby guide conservation efforts.
Netherlands