That woodland Caribou were released during the anthropause from a ‘landscape of fear’ caused by heli-skiing is not supported by available evidence

Abstract

The disruption in socio-economic activity during the COVID-19 pandemic, commonly labeled the “anthropause,” provided an opportunity to observe the effects of a pause in human-related activities on ecological systems (e.g., Naidoo & Burton, 2020; Gaiser et al., 2022; Perkins, Shilling, & Collinson, 2022). However, this treatment of reduced activity was not under the spatial or temporal control of researchers, and therefore, estimated effects could be confounded in time by unobserved factors that varied over the course of the study.

In British Columbia, Canada, heli-ski operators suspended or severely curtailed their activities for the 2020–2021 season in response to international border closures and public health requirements (HeliCat Canada, 2022). Gill et al. (2023) analyzed the late winter movements of woodland caribou (Rangifer tarandus caribou) before, during, and after the anthropause. They reported that caribou home ranges were largest during the anthropause, and inferred that the reduction in heli-skiing activity released caribou from a “landscape of fear” (Bleicher, 2017). However, Gill et al.'s (2023) analysis identified a year effect, but not evidence that heli-skiing was the cause.

Mean home range size during the late winter anthropause year of 2020–21 was indeed larger than in years before or after. But Gill et al. (2023) found that home range overlap with heli-ski tenures was not a significant covariate, meaning that ranges were larger during the anthropause regardless of their pre- or post-anthropause exposure to heli-ski activity. With nearly a third of home ranges occurring entirely outside tenures, their data are well-suited for re-analysis as a Before-After-Control-Impact design, with ranges located outside tenures serving as a quasi-control group (untreated but not randomly assigned). I conducted this analysis (Appendix S1) and found a larger increase in home range size during the anthropause among “control” caribou than among “impact” caribou that had home ranges that overlapped tenures (P = 0.01, β = −1.03). Post-anthropause, the resumption of heli-skiing was associated with a reduction in home range size that did not differ among “control” and “impact” caribou (P = 0.28, β = −0.31; Fig. 1). Thus, changes in home range size from before, during the anthropause treatment, and after, did not provide evidence of a heli-ski effect.

As Gill et al. (2023) noted, tenure overlap is only a coarse measure of exposure to heli-skiing because the distribution of flights and skiing is not uniform in space or time within tenures. But outside tenures, caribou would not have been subject to landings, take-offs, or skiing during years of normal operation, but some might have experienced some helicopter overflights. Detailed flight and skiing data might better resolve heli-ski effects but, as presented, the evidence does not support the definitive conclusion that “heli-skiing reduced caribou home range sizes” (Gill et al., 2023: 6).

But for heli-skiing, why might home ranges be larger during the anthropause year? Gill et al. (2023) recognized the potential confounding of weather effects and presented snowfall, temperature, and precipitation trends. They found no evidence of exceptional circumstances during the anthropause year; however, they did find that the length of time caribou spent on late winter ranges varied by herd and year. Mountain caribou in the interior wet-belt of British Columbia migrate seasonally, moving to lower elevations in early winter as snow accumulates but returning to the subalpine when snow consolidates. This return marks the beginning of the late winter season, which persists until spring conditions again trigger movements to lower elevations to capitalize on early green-up (Environment Canada, 2014). Thus the differences in late winter season lengths likely reflect differences among years and herds with respect to some combination of weather and snow conditions.

Gill et al. (2023) used these seasonal migrations to define the lengths of late winter seasons, which varied between 34 and 99 days (Gill et al., 2023: Table S1.2). Home range size and the lengths of the winter seasons were correlated (τ = 0.12, P = 0.01; Appendix S1; Fig. 2). Gill et al.'s (2023) global model, revised to include late winter season length instead of year, generated a nearly identical fit to the original model (ΔAIC = 1.0; Appendix S1). Therefore, the length of the late winter season confounded the changes in mean home range size by year interpreted by Gill et al. (2023) as a heli-ski effect.

Without evidence of a heli-ski effect, the conclusion that caribou were released from a “landscape of fear” during the anthropause is not currently supported. This example highlights the importance of adjusting for potential confounding in time that can occur in “natural” experiments, such as the COVID-19 pandemic anthropause (Rosen, 2021).

The author received a consulting fee for the drafting and editing of this manuscript from HeliCat Canada, the trade association for heli and cat-ski operators in Canada.