Range expansion is a fundamental population-level process that plays an essential role in establishing past, present, and future species distributions. Understanding the dynamics of range expansion is increasingly important in the current era of anthropogenic change, where species distributions are being modified due to climate change, land-use change, conservation efforts, and invasions by noxious pests. However, range expansion dynamics are difficult to predict; expansions are complex and highly variable processes shaped by ecological and evolutionary forces. Understanding how these forces interact to drive range expansion dynamics has only recently begun to be investigated. Longstanding theory indicates that the speed of range expansion is determined by the ecological processes of dispersal (the rate at which individuals move from one spatial location to another) and the low-density reproductive rate (the rate at which individuals produce offspring in environments where conspecific densities are low). Recent theory has suggested that populations at the leading-edge of expanding ranges are subject to evolutionary forces that can rapidly modify their dispersal and reproductive rates in ways that make range expansions faster and more variable. Dispersal provides a means for individuals in an expanding population to become spatially sorted by dispersal ability. This ‘spatial sorting’ is expected to cause the over-representation of highly dispersive individuals at the leading edge of an expansion, increasing the probability of non-random mating structured by dispersal ability. If dispersal is heritable, highly dispersive individuals at the leading edge are expected to pass dispersal-related traits to their offspring, an effect which increases the speed of range expansion over multiple generations. Furthermore, relative fitness advantages caused by reduced conspecific competition at the low-density range edge may result in two additional selective mechanisms: selection for increased reproductive rates (natural selection) and selection for increased dispersal ability (‘spatial selection’). Developing useful expectations for the dynamics of range expansion thus requires detailed investigations into how these phenomena interact, and how they may be modified by additional population-level processes common to biological systems.

Chapter One of this thesis provides one of the first experimental tests of spatial sorting, using laboratory populations of the bean beetle Callosobruchus maculatus. It finds clear evidence that spatial sorting increases the speed and variability of range expansions. It identifies the rapid evolution of dispersal ability as having caused the increase in speed, and suggests that the random accumulation of genotypes at the leading edge (i.e., ‘gene surfing’) is responsible for the increase in variability. Chapter Two measures the heritability of dispersal and low-density per-capita reproductive rate in C. maculatus, as well as the genetic and environmental correlations between these traits, and builds a simulation model to test how these correlations affect the speed and variability of range expansion. It demonstrates that range expansion speed and variability have a strong dependence on the magnitude and sign of genetic and environmental correlations in these traits, with more positive correlations generating faster and more variable expansions. Chapter Three uses a simulation model to explore how the outcome of spatial sorting is dependent on ecological processes related to population growth and dispersal. It shows that increasing the probability of long-distance dispersal increases the speed and variability associated with spatial sorting; that Allee effects (reduced per-capita reproductive rates in low-density populations) decrease variability and generate smaller increases in range expansion speeds; and that these processes interact with each other and the degree to which dispersal is heritable to ultimately determine range expansion dynamics. Overall, this thesis finds that spatial sorting is a highly robust evolutionary mechanism. Spatial sorting increases the speed of range expansion over a wide range of conditions, including some conditions (such as Allee effects and low dispersal heritability) where I hypothesized that it might be rendered ineffective. This thesis also find that the dynamics of range expansion are highly variable, and the magnitude of this variability is dependent on a myriad of ecological and evolutionary factors. Given these results, I expect spatial sorting to be a common and highly variable phenomenon in natural systems. My research suggests that making useful predictions about range expansion dynamics requires a detailed accounting of ecological and evolutionary forces, particularly those that have been shown to modify variability in the range expansion process.