The consequences of spatial environmental variability on dispersal and on the spatial distribution of species

A central goal in ecology is to identify and understand causal factors that lead to the expansion or contraction of species ranges. Spatio - temporal population dynamics depend on biotic and abiotic environmental conditions, local demography, dispersal behaviour, and phenotypic variation. In particular understanding dispersal behaviour turns out to be a tough problem, because complex feedback loops between dispersal, local demography, and individual variations can arise. Furthermore, previous attempts to understand dispersal by reducing the complexity either in space or time have often resulted in a disregard of these feedbacks. The aim of this thesis was to investigate the influence of dispersal on spatio-temporal population dynamics with models and experiments that explicitly consider the multi causality of dispersal. The thesis is composed of three different studies: Firstly, for an active dispersing species, a plausible factor affecting dispersal behaviour could be personal information. Birds, for example, might gather information on future nest sites and, as a result, individuals differ in the amount and quality of information they possess for use in reaching a dispersal decision. We manipulated the information available to flycatchers (Ficedula hypoleuca) in a field experiment and we found that individuals which were longer exposed to the information altered dispersal behaviour to a greater extent, but only at a local spatial extent. Secondly, models of sex-biased dispersal rarely take space into account. With a computer simulation model, we showed that acknowledging the spatial distribution of the sexes has consequences for the evolutionary outcome of the model leading to selection of more similar dispersal behaviour among the two sexes. Thirdly, models of invasion spread rate often ignore the dependency of dispersal on environmental heterogeneity. We expanded on a reaction-diffusion model to improve this deficit and show that the invasion dynamics of an ecto-parasite (deer ked, Lipoptena cervi) is dependent on the local density of its main host, moose (Alces alces), across its Finnish range. In conclusion, these studies point at the necessity to consider interactions between dispersal and environmental variability, feedbacks between causal factor of dispersal, and realistic assumptions about space and time in order to solve the conundrum of factors determining the spatio-temporal distribution of species.Ei saatavill