Theoretical perspectives on evolution of long-distance dispersal and the example of specialized pests

Long-distance dispersal (LDD) - dispersal beyond the bounds of the local patch or cluster of conspecifics - will be most advantageous in landscapes in which large areas of suitable habitat are consistently available at long distances from established populations. We review conditions under which LDD will be selected and conclude that biotic interactions, and in particular specialized natural enemies, are likely to be one of the most important factors selecting for LDD in many species. We use simple spatially implicit and spatially explicit models to illustrate how such pests affect the evolutionarily stable strategy (ESS) for investment in LDD. Patches currently occupied by parents are more likely to be infected than distant, potentially unoccupied, patches, thus advantaging dispersal. Patchy infestations also result in higher variance in reproductive success among patches, which alone selects for increased among-patch dispersal. Both of these effects increase with the strength of the impact of infestation, and with the number of species competing for space in the community. We discuss the potential of different types of models and analytical tools to capture the impacts of pests on the evolution of LDD, and conclude that, even simple models can illustrate the general relationship between pest pressure and LDD advantage, but only spatially explicit simulation models can fully elucidate the resulting ecological and evolutionary dynamics. In conclusion, we consider the potential role of selection for LDD in the spread of invasive species, and in long-term responses to habitat fragmentation and range shifts.