
Can animals move across landscapes? This network models a landscape as a terrestrial mammal might experience it, with green patches of habitat separated by the least-costly routes for movement.
Assessing landscape connectivity for highly-mobile animals
We are involved in several projects with the goal of understanding how features of landscapes may influence the movements of highly-mobile animals. Both woodland caribou (in Saskatchewan) and bumble bees (in Alberta) figure prominently.
The organizing questions in this work are:
Can animals move to where they need to go (e.g. at various stages of their life history, or in response to environmental change)?
Which landscape features promote or inhibit movement?
How can we model landscape connectivity effectively for highly-mobile animals?
What types of connectivity models and tools best support planning decisions?
These questions have been at the core of our research for some time. We have published a new method and associated software (grains of connectivity) for modelling highly-mobile animal movement, as well as tested these ideas in various publications using caribou genetic, telemetry and simulated data. See also our review on landscape graph modelling.

Can we read the landscape from spatial patterns in genetic relationships? These simulated data analysed with MEMGENE (developed in our lab) shows that we can. The red structure slows the movements of individuals and their offspring (circles) over generations. Circles of similar size and colour show the genetic patterns.
Applied and experimental landscape genetics
As animals move they take their genes with them. A mountain, a river, a road, or the lack of suitable habitat might get in their way, reducing movement and slowing gene flow. If this happens over multiple generations, we may be able to read this landscape signal from spatial autocorrelation in the genetic relationships among individual animals.
Our current landscape genetics research focuses around these questions:
Under what landscape and demographic conditions can we reliably “read the landscape” from genetic data?
Can we develop methods to improve the detection of spatial genetic pattern and the power of landscape genetic inference?
What can genetic data tell us about landscape connectivity in species of conservation concern (e.g. caribou and pollinating insects)?
Towards these goals, we have published a method and associated software (MEMGENE) to extract spatial genetic patterns and improve inference in Methods in Ecology and Evolution. We have also developed a landscape genetic simulator (POPSCAPE) to support simulation experiments. Our landscape genetic work on caribou, testing landscape connectivity predictions for these animals has also appeared in Molecular Ecology.