The effect of dispersal heterogeneity in bioinvasions

Abstract

The spread against a dominant flow of a marine invasive species with pelagic larval stage is a complex ecological phenomenon. This upstream spread involves the interplay between a number of processes including demography and dispersal of organisms, and variability in these processes. The main objectives of this thesis are to study the connection between dispersal, demography and spread in this setting and to understand the effect of variability in dispersal on spreading speeds. As a case study, I used the invasive green crab, Carcinus maenas, which has maintained a relatively consistent rate of spread for over 100 years covering a wide range of temperate latitudes and local hydrological environments along the Atlantic coast of North America. I developed a stage-structured integrodifference equation modelling framework to link spreading speed to underlying demographic and dispersal processes. First, simple kernels (namely Normal and Laplace) were used to model the larval dispersal. Then, a mechanistic kernel with behaviour was incorporated into the stage-structured integrodifference equation model. The temporal variability of the kernel was parameterized by using a particle-tracking submodel embedded into a 3-dimensional hydrodynamic model of the Gulf of St. Lawrence. Results indicated that the relationship between demography and dispersal was compensatory for all the three kernels. Sensitivity analysis indicated that larval dispersal had more effect on spreading speed than any demographic parameter. Our simulation implied that there was a spatial structure in the larval dispersal. Further, when dispersal parameters vary with time, using the time-averaged dispersal would underestimate the upstream invasion spread rates. Thus, accounting for spatial and annual variations in dispersal in population models is important to enhance understanding of spatial dynamics and population spread rates. My research thus contributed to (i) applied science by ranking different possible strategies in the management of a marine invasive species, (ii) theoretical ecology by developing tools to incorporate biophysical and behavioural features into the dispersal component of an integrodifference equation model, and by showing that the year-to-year dispersal variability increases the upstream spreading speed.