Demography and Dispersal: Life Table Response Experiments for Invasion Speed

• Published in: Ecology (Durham) 2003-08, Vol.84 (8), p.1968-1978
• Main Authors: Caswell, Hal, Lensink, Rob, Neubert, Michael G.
• Format: Article
• Language: English
• Subjects: Accipiter nisus; Animal migration; Animal populations; Biological invasions; Breeding; Conservation biology; Demography; Ecological invasion; Ecology; elasticity; Experiments; Ficedula hypoleuca; Flycatchers; Habitat conservation; integrodifference equations; life table response experiment; long-distance dispersal; LTRE; matrix population models; Metapopulation ecology; Modeling; sensitivity; Special Feature: Long-Distance Dispersal; Starlings; Sturnus vulgaris; traveling waves
• ISSN: 0012-9658; 1939-9170
• DOI: 10.1890/02-0100

Long-distance dispersal has many consequences. One of these is the speed with which a population can invade a new habitat. This invasion wave speed can be calculated from information on demography (in the form of a stage-structured matrix population model) and dispersal (in the form of a distribution of stage-specific dispersal distances). Wave speed complements population growth rate (λ) as an index of population performance in conservation biology. Here we present a new way to calculate the sensitivity of wave speed to changes in short- and long-distance dispersal-indeed, to dispersal at any distance-using the order statistics of the dispersal distribution. This makes it possible to examine the effects of changing the distance dispersed by the longest-dispersing individual, the shortest, the median individual, etc. We apply the method to data on the Pied Flycatcher in the Netherlands, the Starling in the United States, and the Sparrowhawk in England and the Netherlands. We find that wave speed is vastly more sensitive to changes in the uppermost percentiles of the dispersal distribution than to changes elsewhere. The sensitivity results make it possible to carry out life table response experiment (LTRE) calculations, which quantify the contribution of demography, short-distance dispersal, and long-distance dispersal to interpopulation differences in wave speed. We find that 30% of the difference in wave speed between the Pied Flycatcher and the Starling is due to demography and about 70% to dispersal. Most of the dispersal effects come from the top 10% of the dispersal distribution. In contrast, 86% of the difference in wave speed between the two populations of Sparrowhawks is due to differences in demography, and about 14% to differences in dispersal. Again, most of the dispersal effects are due to the upper percentiles of the dispersal distribution. The patterns revealed in the analysis of the four populations are very similar, suggesting that the dependence of wave speed on long-distance dispersal may exhibit common patterns of sensitivity and elasticity.