Simulation-based validation of spatial capture-recapture models: A case study using mountain lions

Abstract

Spatial capture-recapture (SCR) models have improved the ability to estimate densities of rare and elusive animals. However, SCR models have seldom been validated even as model formulations diversify and expand to incorporate new sampling methods and/or additional sources of information on model parameters. Information on the relationship between encounter probabilities, sources of additional information, and the reliability of density estimates, is rare but crucial to assessing reliability of SCR-based estimates. We used a simulation-based approach that incorporated prior empirical work to assess the accuracy and precision of density estimates from SCR models using spatially unstructured sampling. To assess the consequences of sparse data and potential sources of bias, we simulated data under six scenarios corresponding to three different levels of search effort and two levels of correlation between search effort and animal density. We then estimated density for each scenario using four models that included increasing amounts of information from harvested individuals and telemetry to evaluate the impact of additional sources of information. Model results were sensitive to the quantity of available information: density estimates based on low search effort were biased high and imprecise, whereas estimates based on high search effort were unbiased and precise. A correlation between search effort and animal density resulted in a positive bias in density estimates, though the bias decreased with increasingly informative datasets. Adding information from harvested individuals and telemetered individuals improved density estimates based on low and moderate effort but had negligible impact for datasets resulting from high effort. We demonstrated that density estimates from SCR models using spatially unstructured sampling are reliable when sufficient information is provided. Accurate density estimates can result if empirical-based simulations such as those presented here are used to develop study designs with appropriate amounts of effort and information sources.

Estimating Coyote Densities with Local, Discrete Bayesian Capture‐Recapture Models

“…This increase in density and abundance result from the non-local model potentially biasing the effective sampling area low because encounter probabilities decline as more traps are included that are outside of an individual's trap detection range. Paterson et al (2019) also noted the potential for biased abundance estimates in SCR models when detection probabilities are lower but in reference to inadequate search effort and correlation between search effort and density. Thus, our result may suggest that future SCR analyses should implement a local evaluation approach, especially if they have large study areas (or trapping grids) relative to animal activity areas.…”

Integrated Carnivore‐Ungulate Management: A Case Study in West‐Central Montana

“…The number of sampling-days, level of search effort, number of samples included in analysis, number of individuals identified, and number of spatial recaptures varied across study areas and time periods (Table 3). The number of individuals identified and the number of spatial captures in each study area and time period, together with previous simulation-based work on the same study design, suggested that each dataset was adequate to result in unbiased spatial-capture recapture abundance estimates (Table 3; Paterson et al 2019a). We radio-collared a total of 15 animals (9 female, 6 male) in the Bitterroot study area and 9 (6 female, 3 male) in the Clark Fork study area.…”

“…The SCR model component for the distribution of animal activity centers in space included a covariate representing a separately estimated mountain lion resource selection function (RSF; Robinson et al 2015). Previous work suggested this RSF was a strong predictor of activity centers (Proffitt et al 2015, Paterson et al 2019a. The RSF predicted the relative probability of mountain lion use during winter (1 Dec-15 Apr) and was developed based on GPS location data from 85 independent-aged collared male and female mountain lions from 9 different study areas in Montana (Robinson et al 2015).…”

“…where α was the baseline detection probability, Sex i is an indicator variable (female = 0, male = 1), and effort j t , is the effort expended in cell j during occasion t. We set p ijt 0 to zero for any trapping cell in which no effort was expended. We included the locations of harvested individuals as if they arose from the detection process; previous work has suggested this introduces negligible bias into estimates (Paterson et al 2019a) and can improve SCR-based estimates of abundance (Proffitt et al 2015). We censored these animals from the model for all occasions following harvest to avoid a bias in detection probabilities and violating the population closure assumption.…”

“…Estimating carnivore population size is challenging because they often occur at low densities, are wide ranging and difficult to detect, and violate the population closure assumption, which is important to traditional capture‐recapture population estimation. Spatial‐capture recapture modeling provides an efficient method of estimating carnivore population size (Proffitt et al 2015, Boulanger et al 2018, Paterson et al 2019 a ) that incorporates the spatial organization of individuals through estimation of trap‐specific capture probabilities while also accounting for detection probability (Efford and Fewster 2013, Royle et al 2013, Efford 2014). These methodologies allow wildlife managers to monitor carnivore population abundances before and after implementing harvest prescriptions and to evaluate if harvest prescriptions meet the management objectives.…”