Movement ecology of endangered caribou during a <scp>COVID</scp>‐19 mediated pause in winter recreation – response to Wilson (2024)

In Gill et al. (2023), we present evidence that the COVID-19 induced reduction in heli-skiing affected space use by southern mountain caribou (SMC). Wilson (2024) re-analyzed the data presented in Gill et al. (2023) and concluded that there was no evidence of a spatial effect of heli-ski tenure on home-range size, and that home-range size was correlated with the duration of time on late-winter range. From this re-analysis, Wilson questions the evidence for the claim that the COVID-19 induced reduction in heli-skiing released SMC from a landscape of fear. We value this constructive criticism, but we argue that there are three main shortcomings to Wilson's (2024) work that challenge the validity of this rebuttal. First, the data used in Wilson (2024) were a subset of the data used by Gill et al. (2023). In our original paper, we fit a single model to 223 SMC home-range sizes over 4 years (i.e., 2018/2019, 2019/2020, 2020/2021 (anthropause) and 2021/2022). In contrast, Wilson (2024) excluded the data from 2018/2019 and split the remaining 3 years of data into two halves (comparing 2019/2020 to the anthropause, and the anthropause to 2021/2022), fitting models to these two subsets independently. These analytical choices reduced the sample size for each model by half, in turn reducing the statistical power and obfuscating conclusions of the effects of heli-skiing on SMC. Not surprisingly, Wilson (2024) did not find significant effects of heli-skiing on home-range size. Indeed, examining the code provided in the supporting information of Wilson (2024) shows that his models ran into convergence issues, which may explain why this re-analysis resulted in parameter estimates and significance levels that differed from Gill et al. (2023). Second, there is an important difference in the coding of overlap between SMC home ranges and heli-ski tenures between Gill et al. (2023) and Wilson (2024). In Gill et al. (2023), we used the proportion of home-range overlap with tenures as a continuous measure of each caribou's potential exposure to heli-skiing. In contrast, Wilson (2024) coded heli-ski exposure as a binary variable. To do this Wilson treated any caribou whose home range overlapped tenures to even the smallest degree (including e.g. overlap of 6.4 × 10−5 km2) as being functionally identical to an animal whose home range was entirely contained within a tenure (Fig. 1). This binary classification likely caused the spatial effect to change. Exactly where heli-skiing occurs within tenures is unknown to anyone aside from the heli-ski operators, a point also recognized by Wilson (2024). Whereas we concluded from the lack of a clear spatial effect that there is still a large amount of uncertainty as to heli-skiing's zone of influence, Wilson (2024) concluded that this uncertainty demonstrated that there was no evidence of a heli-ski effect. Third, we interpret differently Wilson's (2024) resulting correlation between home-range size and skier days. Wilson (2024) states that because there was a correlation between home-range size and the number of days SMC spent on their winter range, that the heli-ski effect was an erroneous confound. We do not dispute Wilson's finding of a correlation between home-range size and days on winter range, however, we disagree with the interpretation. Although based on a small sample size, both home-range size and days on late-winter range are also strongly negatively correlated with the number of skier days (home range: t = −4.6, df = 2, P-value = 0.044, correlation = 96%; days on range: t = −3.5, df = 2, P-value = 0.073, correlation = 93%, [Fig. 2]). Given that Gill et al. (2023) found no meaningful differences in the winter weather among years (a point also recognized by Wilson), this relationship suggests that intense skiing pressure could not only be reducing the size of their home ranges, but also pre-maturely displacing SMC from their late-winter ranges. In other words, we interpret these results as further supporting the hypothesis that heli-skiing is impacting SMC movement. We believe that where the uncertainty lies is in how SMC are responding – as late-winter-range displacement may be an additional impact of heli-skiing – but additional data and analyses are required. Wilson's conclusion that there is insufficient evidence to claim that heli-skiing generates a landscape of fear for SMC relied on (i) an approach with reduced statistical power and (ii) the modification of variables resulting in spurious representation of home-range overlap with tenures. It also relied on Wilson contradicting the results of previous work on the responses of SMC to helicopters and heli-skiers. In Wilson and Wilmshurst (2019), the authors describe how woodland caribou respond to human disturbances through elevated stress hormones (Freeman, 2008) and abandonment of habitat (Lesmerises et al., 2018), further acknowledging that there are energetic costs to disturbances that they avoid. Examining the results of Wilson and Wilmshurst (2019), the authors present evidence of responses by SMC to helicopters and skiers that include ‘Concerned’, ‘Alarmed’ and ‘Very Alarmed’, descriptors that are potentially responses of fear to a stressor. Management actions to support the recovery of species at risk require top-down guidance from regulators, but also bottom-up engagement from those industries or organizations that may be impeding recovery. Though perhaps imperfect (see Palm et al., 2020), there are examples of industry working collaboratively towards the recovery of SMC (Lamb et al., 2022; LetsRideBC, 2024). The heli-ski industry, in comparison, has remained largely beyond regulatory reach. As with most species, uncertainties remain with regards to SMC's fine scale movement and space use, but one of the greatest unknowns is the spatiotemporal use of critical SMC winter habitats by heli-skiing operators. Better data are needed from heli-ski operators to inform the science surrounding the effects of this disturbance on SMC recovery. Avoiding further extinctions and championing wildlife stewardship on public lands should be the goal of all stakeholders operating within SMC ranges. Ensuring that data are available and analyzed correctly is a critical first step in providing the science needed to support recovery in multi-user landscapes.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.