American Society of Naturalists

A membership society whose goal is to advance and to diffuse knowledge of organic evolution and other broad biological principles so as to enhance the conceptual unification of the biological sciences.

“Within reach? Habitat availability as a function of individual mobility and spatial structuring”

Posted on

Jason Matthiopoulos, John Fieberg, Geert Aarts, Frédéric Barraquand, and Bruce E. Kendall (June 2020)

Habitat availability in terms of organism mobility and habitat structure. Geographical constraints in niche spaces

Read the Article (Just Accepted)

Abstract

Organisms need access to particular habitats for their survival and reproduction. However, even if all necessary habitats are available within the broader environment, they may not all be easily reachable from the position of a single individual. Many Species Distribution Models (SDMs) consider populations in environmental (or niche) space, hence overlooking this fundamental aspect of geographical accessibility. Here, we develop a formal way of thinking about habitat availability in environmental spaces by describing how limitations in accessibility can cause animals to experience a more limited or, simply, different mixture of habitats than those more broadly available. We develop an analytical framework for characterizing constrained habitat availability based on the statistical properties of movement and environmental autocorrelation. Using simulation experiments, we show that our general statistical representation of constrained availability is a good approximation of habitat availability for particular realizations of landscape-organism interactions. We present two applications of our approach, one to the statistical analysis of habitat preference (using step-selection functions to analyze harbor seal telemetry data) and a second that derives theoretical insights about population viability from knowledge of the underlying environment. Analytical expressions for habitat availability, such as those we develop here, can yield gains in analytical speed, biological realism and conceptual generality by allowing us to formulate models that are habitat-sensitive, without needing to be spatially explicit.