Identifying road avoidance behavior using time‐geography for red deer in Banff National Park, Alberta, Canada

Abstract

Analysis of animal movement as a complex spatiotemporal signal attenuated by behavioural and contextual factors comprises a recent perspective in the time‐geographic study of movement. For their significant ecological and human impacts, animal–roadway interactions have become a particularly important subject matter in this arena. Analyses relying on spatiotemporal aggregation or reductive modelling of the information held in movement trajectories may overlook the influences of behaviour and environmental context. Towards expanding perspectives on animal movement and roadway interactions, this research acknowledges and characterizes the varying influence of temporally dynamic elements in the environmental context at fine spatiotemporal scales. In particular, these elements are hourly traffic volumes and their effect on the probability of animal–roadway interactions. A set of methods from time geography and signal analysis, including the probabilistic space‐time prism, comprehensive probability surface, and cross‐correlation were combined to provide for serial comparison of hourly roadway interaction probabilities and traffic volumes for red deer (Cervus elaphus) tracked in Banff National Park, Alberta, Canada. Results suggest a cyclical, diurnal repulsion in roadway interaction probabilities from periods of higher traffic volume at the hourly scale in the study area, consistent with prior theoretical and empirical findings on ungulates living in similar environmental settings.

Using a Cost-Distance Time-Geographic Approach to Identify Red Deer Habitat Use in Banff National Park, Alberta, Canada

“…While early space-time prisms have served in a range of applied studies, the prevailing understanding holds that interior volumes of space-time prisms are not homogeneous, as movement opportunity is not equally distributed over space and time for the mover [1,11,21,23]. As an early improvement on the binary bounding action of the classical space-time prism, [23] introduced the voxel-based probabilistic spacetime prism, where a mover's chance of having occupied any given voxel location over the time and space elapsed between space-time anchors is assigned as a function of that voxel's deviation from the space-time path (Equation ( 2)).…”

“…The velocity multiplier is a value used to scale the observed velocity of the mover between two consecutive space-time anchors; this parameter is meant to account for the assumed straight-line movements captured in trajectories. Since straight-line, top-speed movements are rare for terrestrial animals' routine traversal in a varied environmental context, the velocity multiplier offers a means to adjust the reachable distances expressed as terms in Equations ( 4) and ( 5), such that prism results do not simply converge to the least-cost space-time path, resulting in a prism having zero volume [1,2,22,23].…”

“…The degree of realism by which we may characterize animal space use is foundationally important to the work of conservationists, biologists, and spatial ecologists interested in understanding patterns and causality in animal movement [1][2][3]. Early work delineated the space of an animal's daily activity in a deterministic manner based on known animal locations [4][5][6].…”

“…Advances in space-time prism construction also introduce the notion of a four-dimensional space-time prism [30] and examine the influence of static spatial context on prism volumes [31,32]. Further still, prism research extends into the examination of the influence that dynamic or temporally modulated factors in the spatial context have on movement [1] and the bounding of prism volumes based on the simulated action of behaviorally informed agent-based models [3]. Advancements in these methods have been demonstrated in application to issues of conservation and wildlife management, enabling better spatiotemporal understandings of animal home range [12], long-distance animal movements [33], and population-level examination of animal habitat use [2].…”

“…Advancements in these methods have been demonstrated in application to issues of conservation and wildlife management, enabling better spatiotemporal understandings of animal home range [12], long-distance animal movements [33], and population-level examination of animal habitat use [2]. In particular, research on animal interactions includes animal movement considering the interaction between environmental context and local choice [34], animal-to-animal interactions [25], and animal-to-roadway interactions [1,20,35]. Clearly, incorporating spatiotemporal dynamics into the study of animal movement has proven productive.…”