Postdoctoral Research Associate, Wolkovich Group
Forestry and Conservation Science
University of British Columbia
3041 - 2424 Main Mall, Vancouver, BC, V6T 1Z4
geoffrey.legault@ubc.ca

Hello! I am interested in a variety of topics in the fields of population biology and theoretical ecology, including: spatial spread, coexistence, synchrony, and life history. To answer questions in these areas, I use or derive mathematical models to make predictions about what is possible. Alternatively, I search for useful models by confronting candidate models with reality (i.e., data).

I am particularly interested in applications of stochastic models as they are an efficient way to account for unknown or unmeasured sources of variation, both of which are common in ecology.

You can read more about my research by scrolling down or clicking here.

News

Curriculum vitae

Google scholar profile | ORCID profile

You can download my CV here (updated 2019-07-02)

Research

Spatial spread and competition

As early as Darwin, ecologists recognized that species ranges were driven by both abiotic and biotic factors. However, models of species' range dynamics routinely incorporate only abiotic factors (e.g., species distribution models).

With Brett Melbourne and Alan Hastings, I am using experimental microcosms of the flour beetles Tribolium castaneum and Tribolium confusum (see below) to test how competition affects their spatio-temporal range dynamics.

gen1234.png

Figure 1: Four generations of beetles (black dots) spreading from patch 1 to patch 7 in a one-dimensional landscape.

Age and size at maturity

Age and size at maturity are co-varying life history traits strongly associated with fitness. As a result, there has been considerable interest in understanding how they co-vary and evolve in response to factors such as temperature or predation. The vast majority of life history studies are deterministic, meaning they do not explicitly account for individual variation that is independent of genetic and/or environmental factors. However, non-genetic and non-environmental variation in age and size has been observed in many organisms, meaning deterministic approaches are incomplete.

In collaboration with Joel Kingsolver, I am developing a continuous-time stochastic model describing the joint distribution of an insect's age and size at maturity. The model is based on the developmental biology of Manduca sexta.

stochastic_dev_mass.png

Figure 2: Contour plot of the joint distribution of maturation time (age) and mass (size) for different parameter values (colors). Darker shades represent more likely combinations of maturation time and mass.

Nonhomogeneous SSA

Stochastic population models are used to account for unexplained variation in population dynamics. One of the most flexible methods for simulating these models in continuous-time is Gillespie's stochastic simulation algorithm (SSA).

Importantly, the SSA assumes that demography is homogeneous, meaning that demographic rates (e.g. birth rate) must be constant over time. Given that many demographic rates are known to change over an individual's lifespan and may also respond to externally varying environmental conditions, this assumption is incompatible with nature.

I developed a nonhomogeneous version of Gillespie's SSA (SSA+) that allows demography to respond to environmental change. You can read the paper about it here.

ssan_ssaplus_exponential.png

Figure 3: Population sizes predicted by an exponential growth model where birth responds to a fluctuating environment, simulated with the SSA (red) and SSA+ (blue).

Publications

Google scholar profile | ORCID profile

If you would like a reprint of a paper listed below, send me an email.

  1. Legault, G., Fox, J., Melbourne, B. A. (2019) Demographic stochasticity alters expected outcomes in experimental and simulated non‚Äźneutral communities. Oikos. In press data
  1. Legault, G., Melbourne, B. A. (2019) Accounting for environmental change in continuous-time stochastic population models. Theoretical Ecology 12(1): 31-48 link
    • Recommended ("Very Good") by Faculty of 1000 Prime Access the recommendation on F1000Prime
  1. Contributing author (2018) "Direct and indirect drivers of land degradation and restoration" in Assessment Report on Land Degradation and Restoration. Intergovernmental Platform on Biodiversity and Ecosystem Services (IPBES) link
  1. Legault, G., Cusa, M. (2015) Temperature and delayed snowmelt jointly affect the vegetative and reproductive phenologies of four sub-Arctic plants. Polar Biology 38: 1701-1711 link
  1. Fox, J., Legault, G., Vasseur, D., Einarson, J. (2013) Nonlinear effect of dispersal rate on spatial synchrony of predator-prey cycles. PLoS ONE 8(11): e379527 link
  1. Legault, G., Weis, A. (2013) The impact of snow accumulation on a heath spider community in a sub-Arctic landscape. Polar Biology 36: 885-894 link

In revision

  • Dallas, T., Legault, G., Melbourne, B. A., Hastings, A. M. Context-dependent dispersal in two species communities.
  • Bullock, M., Legault, G., Melbourne, B. A. Interspecific chemical competition between Tribolium castaneum and Tribolium confusum reduces fecundity and hastens development time.

Submitted

  • Legault, G., Kingsolver J. G. A stochastic model for predicting age and mass at maturity of insects.

In preparation

  • Legault, G., Bitters, M. E., Hastings, A. M., Melbourne, B. A. Competition slows spatial spread and shapes range boundaries in experimental landscapes
  • Legault, G., Riley, M. J., Melbourne, B. A. Intrinsic dispersal ability and environment affect trait evolution during range expansion

Resources

Data

Emacs

I do many of my tasks (e.g., R coding, writing, tracking RSS feeds, this website) within the venerable text editor Emacs.

As I am sometimes asked about getting started with Emacs, I've created a basic introduction to it, available here.