Density-dependent dispersal and the speed of range expansions
Correspondence: Res Altwegg, South African National Biodiversity Institute, P/Bag X7, Claremont 7735, South Africa.
The speed of range expansions, be it invasive species colonizing a new area or species tracking a moving climatic niche, critically depends on dispersal. Models for species' range expansions generally assume dispersal to be independent of local population densities. However, animals often disperse in response to high population size or alternatively may avoid or leave areas of very low population sizes. We explore whether such density dependence in dispersal can safely be ignored when predicting the speed of range expansions.
We use simulations to examine the effect of different forms of density dependence in emigration and immigration on the speed of range expansions. For emigration, we consider linear and nonlinear forms of positive density dependence, negative density dependence at low population densities and constant emigration rates. For immigration, we consider options where individuals avoid crowded patches, are attracted to the presence of conspecifics or settle independent of local density.
The speed of range expansion was slowest when emigration was strongly positively related to density (higher emigration at higher densities) and when individuals avoided settling in low-density patches. It tended to be fastest under negatively density-dependent emigration (higher emigration at lower densities). These results were consistent across two different life histories and different levels of carrying capacity.
Our results suggest that considering density-dependent dispersal and the mechanisms leading to it are important for correctly predicting species' rates of spread. Organisms with a tendency to aggregate, for example, by relying on conspecific attraction in settlement and emigrating mainly in response to high local densities, are predicted to be least likely to expand their ranges and most at risk from spatial shifts in their climatic niches.