Reply to Efford on‘Integrating resource selection information with spatial capture-recapture’

Protection status, human disturbance, snow cover and trapping drive density of a declining wolverine population in the Canadian Rocky Mountains

Protected areas are important in species conservation, but high rates of human-caused mortality outside their borders and increasing popularity for recreation can negatively affect wildlife populations. We quantified wolverine (Gulo gulo) population trends from 2011 to 2020 in > 14,000 km² protected and non-protected habitat in southwestern Canada. We conducted wolverine and multi-species surveys using non-invasive DNA and remote camera-based methods. We developed Bayesian integrated models combining spatial capture-recapture data of marked and unmarked individuals with occupancy data. Wolverine density and occupancy declined by 39%, with an annual population growth rate of 0.925. Density within protected areas was 3 times higher than outside and declined between 2011 (3.6 wolverines/1000 km²) and 2020 (2.1 wolverines/1000 km²). Wolverine density and detection probability increased with snow cover and decreased near development. Detection probability also decreased with human recreational activity. The annual harvest rate of ≥ 13% was above the maximum sustainable rate. We conclude that humans negatively affected the population through direct mortality, sub-lethal effects and habitat impacts. Our study exemplifies the need to monitor population trends for species at risk-within and between protected areas-as steep declines can occur unnoticed if key conservation concerns are not identified and addressed.

Confronting spatial capture-recapture models with realistic animal movement simulations

Spatial capture-recapture (SCR) models have emerged as a robust method to estimate the population density of mobile animals. However, model evaluation has generally been based on data simulated from simplified representations of animal space use. Here, we generated data from animal movement simulated from a mechanistic individual-based model, in which movement emerges from the individual's response to a changing environment (i.e., from the bottom-up), driven by key ecological processes (e.g., resource memory and territoriality). We drew individual detection data from simulated movement trajectories and fitted detection data sets to a basic, resource selection and transience SCR model, as well as their variants accounting for resource-driven heterogeneity in density and detectability. Across all SCR models, abundance estimates were robust to multiple, but low-degree violations of the specified movement processes (e.g., resource selection). SCR models also successfully captured the positive effect of resource quality on density. However, covariate models failed to capture the finer scale effect of resource quality on detectability and space use, which may be a consequence of the low temporal resolution of SCR data sets and/or model misspecification. We show that home-range size is challenging to infer from the scale parameter alone, compounded by reliance on conventional measures of "true" home-range size that are highly sensitive to sampling regime. Additionally, we found the transience model challenging to fit, probably due to data sparsity and violation of the assumption of normally distributed inter-occasion movement of activity centers, suggesting that further development of the model is required for general applicability. Our results showed that further integration of complex movement into SCR models may not be necessary for population estimates of abundance when the level of individual heterogeneity induced by the underlying movement process is low, but appears warranted in terms of accurately revealing finer scale patterns of ecological and movement processes. Further investigation into whether this holds true in populations with other types of realistic movement characteristics is merited. Our study provides a framework to generate realistic SCR data sets to develop and evaluate more complex movement processes in SCR models.

Precision and bias of spatial capture-recapture estimates: A multi-site, multi-year Utah black bear case study

Spatial capture-recapture (SCR) models are powerful analytical tools that have become the standard for estimating abundance and density of wild animal populations. When sampling populations to implement SCR, the number of unique individuals detected, total recaptures, and unique spatial relocations can be highly variable. These sample sizes influence the precision and accuracy of model parameter estimates. Testing the performance of SCR models with sparse empirical data sets typical of low-density, wide-ranging species can inform the threshold at which a more integrated modeling approach with additional data sources or additional years of monitoring may be required to achieve reliable, precise parameter estimates. Using a multi-site, multi-year Utah black bear (Ursus americanus) capture-recapture data set, we evaluated factors influencing the uncertainty of SCR structural parameter estimates, specifically density, detection, and the spatial scale parameter, sigma. We also provided some of the first SCR density estimates for Utah black bear populations, which ranged from 3.85 to 74.33 bears/100 km² . Increasing total detections decreased the uncertainty of density estimates, whereas an increasing number of total recaptures and individuals with recaptures decreased the uncertainty of detection and sigma estimates, respectively. In most cases, multiple years of data were required for precise density estimates (<0.2 coefficient of variation [CV]). Across study areas there was an average decline in CV of 0.07 with the addition of another year of data. One sampled population with very high estimated bear density had an atypically low number of spatial recaptures relative to total recaptures, apparently inflating density estimates. A complementary simulation study used to assess estimate bias suggested that when <30% of recaptured individuals were spatially recaptured, density estimates were unreliable and ranged widely, in some cases to >3 times the simulated density. Additional research could evaluate these requirements for other density scenarios. Large numbers of individuals detected, numbers of spatial recaptures, and precision alone may not be sufficient indicators of parameter estimate reliability. We provide an evaluation of simple summary statistics of capture-recapture data sets that can provide an early signal of the need to alter sampling design or collect auxiliary data before model implementation to improve estimate precision and accuracy.