Workshop on Plant Dispersal and Migration Modeling

Global environmental change is causing shifts in the geographical locations of habitats suitable for particular plant species. While it is established that the future distributions of plant species will be strongly influenced by the ability of plants to migrate to sites of suitable habitat, our ability to predict potential and actual migration rates is rudimentary. This workshop organized by the Global Change and Terrestrial Ecosystems (GCTE) core project of the International Geosphere-Biosphere Program provided scientists with interests and expertise in global change and plant migration with a forum for developing a new collaborative synthesis of understanding on long distance dispersal and migration modeling. This grant from the U.S. Department of Energy, Office of Biological and Environmental Research, provided partial support for the workshop by supporting the participation of U.S. scientists. Rationale Global environmental change is causing shifts in the geographical locations of habitats suitable for particular plant species (Davis and Zabinkski 1992). While it is established that the future distributions of plant species will be strongly influenced by the ability of plants to migrate to sites of suitable habitat (Pitelka et al. 1997), our ability to predict potential and realized (as influenced by habitat fragmentation and habitat loss) migration rates is rudimentary. A work group organized by the Global Change and Terrestrial Ecosystems (GCTE) core project of the International Geosphere-Biosphere Program has been tasked with developing capacity to predict plant migration rates in the context of global change. The group aims to (1) improve our empirical knowledge of long distance dispersal, (2) use empirically based dispersal functions to predict plant migration rates as influenced by landscape heterogeneity and (3) develop techniques for scaling up dispersal algorithms. It is the group’s premise that these goals can be achieved by developing an intimate relationship between natural history knowledge, data, parameter estimation and modeling. Capacity to predict potential migration rates has improved markedly recently (Clark et al. 1988, 1999). Although we now understand the processes that determine potential migration rates it should be noted that our empirical knowledge of these processes remains poor. This is because the great majority of quantitative dispersal studies have concentrated on local dispersal and have explicitly ignored the rare (e 5%) long distance (an order of magnitude greater than the modal dispersal distance) dispersal events that determine potential migration rates (Cain et al. in press). Most reports of rare long distance dispersal have been anecdotal. An immediate task is therefore to improve and synthesize our quantitative knowledge of long distance dispersal. While progress has been made in developing techniques for predicting potential migration rates our ability to predict how habitat loss, fragmentation and other forms of environmental heterogeneity determine realized migration rates is poor. Most studies of migration in heterogeneous landscapes have been theoretical (e.g. Gardner et al. 1987), while empirical work