Generalized spatial mark–resight models with an application to grizzly bears

Abstract

Abstract1. The high cost associated with capture-recapture studies presents a major challenge when monitoring and managing wildlife populations. Recently developed spatial mark-resight (SMR) models were proposed as a cost-effective alternative because they only require a single marking event. However, existing SMR models ignore the marking process and make the tenuous assumption that marked and unmarked populations have the same encounter probabilities. This assumption will be violated in most situations because the marking process results in different spatial distributions of marked and unmarked animals.2. We developed a generalized SMR model that includes sub-models for the marking and resighting processes, thereby relaxing the assumption that marked and unmarked populations have the same spatial distributions and encounter probabilities.3. Our simulation study demonstrated that conventional SMR models produce biased density estimates with low credible interval coverage (CIC) when marked and unmarked animals had differing spatial distributions. In contrast, generalized SMR models produced unbiased density estimates with correct CIC in all scenarios.4. We applied our SMR model to grizzly bear (Ursus arctos) data where the marking process occurred along a transportation route through Banff and Yoho National Parks, Canada. Twenty-two grizzly bears were trapped, fitted with radiocollars and then detected along with unmarked bears on 214 remote cameras. Closed population density estimates (posterior median ± 1 SD) averaged from 2012 to 2014 were much lower for conventional SMR models (7.4 ± 1.0 bears per 1,000 km 2 ) than for generalized SMR models (12.4 ± 1.5). When compared to previous DNA-based estimates, conventional SMR estimates erroneously suggested a 51% decline in density. Conversely, generalized SMR estimates were similar to previous estimates, indicating that the grizzly bear population was relatively stable. Synthesis and applications.Mark-resight studies often cost less than capture-recapture studies, but require that marked and unmarked animals have equal encounter rates. Generalized spatial mark-resight models relax this assumption by including sub-models for both the marking and resighting processes. They produce unbiased density estimates even when marked and unmarked animals have differing spatialThis is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Spatial Capture–Mark–Resight Estimation of Animal Population Density

“…For the SMR puma dataset of Kelly et al (), we show in Web Appendix E that the pseudolikelihood approach gives similar estimates to Bayesian methods (Sollmann et al, ; Royle et al, ). Recent work by Whittington et al () has extended the Bayesian approach to include spatial capture–mark–resight models in our terminology, with promising results. We consider the methodologies to be complementary and mutually illuminating.…”

“…We do not believe these factors caused significant bias in our estimates of rat density, but we had no independent check. Whittington et al () analyzed comparable data from a study of grizzly bears ( Ursus arctos ). They found it difficult to fit models to count data by MCMC sampling, a problem they attributed to (non‐spatial) overdispersion.…”

“…The need to specify the spatial distribution of marked animals is a weakness of mark–resight models, and mis‐specification can be damaging (Web Table 4; Royle et al, ; Whittington et al, ). Under some conditions the distribution of marked animals may be known or reasonably assumed.…”

Evaluating spatially explicit density estimates of unmarked wildlife detected by remote cameras

“…Alternatively, using reliable visible marks would ensure identifiability of marked and unmarked individuals, eliminating this source of bias (Rich et al., ; Whittington, Hebblewhite, & Chandler, ). However, independently marking a subset of individuals obviates one or both of the non‐invasive and cost‐saving benefits of using remote cameras for density estimation.…”

Unifying population and landscape ecology with spatial capture–recapture

“…In practice, many population studies employ both capture-recapture sampling and also telemetry of a sample of individuals. Combining these two data sources to develop population level models of movement and resource selection is an important topic of ongoing research (Royle et al 2016, Linden et al 2017b, Whittington et al 2017). …”