Seed dispersal and range dynamics of plants: understanding and predicting the spatial dynamics of serotinous Proteaceae

Seed dispersal affects the rate at which plant populations colonize unoccupied habitat, and the speed at which plant species migrate. In this thesis, I aim (1) to better understand seed dispersal processes, (2) to investigate the consequences of dispersal for the geographical range dynamics of plants, and (3) to forecast the future range size of plant species under climate change. To this end, I developed process-based models for seed dispersal, colonization and range dynamics. These models were parameterised with empirical data for serotinous Proteaceae endemic to the Cape Floristic Region (South Africa). The study species are woody plants with a canopy seed bank and predominantly wind-dispersed seeds. To better understand the seed dispersal of these species, I formulated a process-based model for secondary seed dispersal by wind (the wind-driven movement of seeds along the ground surface). This model was successfully validated in a field experiment in which I released seeds of 7 species that cover the range of seed sizes and seed morphologies typical of the studied Proteaceae. The model predicts a unimodal relationship between dispersal distance and seed size. This indicates that the ecological consequences of secondary seed dispersal by wind are qualitatively different from those of airborne seed movement. To investigate the consequences of seed dispersal for large-scale range dynamics, I derived a model of habitat colonization that links seed dispersal models to data on local abundance and habitat arrangement. This model was used to estimate the colonization ability of 37 Proteaceae species. I found that the extent to which these species fill their potential ranges increases with colonization ability, decreases with the extinction probability of local populations, and is not related to a species' evolutionary age. This suggests that colonization and local extinction shape the range dynamics of the study species on ecological timescales. To forecast how climate change and migration ability will influence the range sizes of 26 Proteaceae species in the year 2050, I combined process-based models for seed dispersal and migration with an existing bioclimatic scenario. Model simulations indicate that informative forecasts of future range size can be made despite the substantial uncertainty inherent in forecasts of population-level migration rates. Although the simulations predict some long-distance dispersal for all study species and use assumptions that upwardly bias migration estimates, they forecast that several species will experience severe range losses or go extinct. This suggests that climate change-integrated conservation strategies should not exclusively rely on the intrinsic migration ability of plants. I nevertheless identify conservation measures by which colonization and migration rates of Proteaceae may be enhanced. The results of this thesis add to a functional understanding of seed dispersal and biogeography, help to predict the future range dynamics of plants, and may prove useful for other areas of ecology, evolutionary and conservation biology