A local evaluation of the individual state‐space to scale up Bayesian spatial capture–recapture

Abstract

Spatial capture–recapture models (SCR) are used to estimate animal density and to investigate a range of problems in spatial ecology that cannot be addressed with traditional nonspatial methods. Bayesian approaches in particular offer tremendous flexibility for SCR modeling. Increasingly, SCR data are being collected over very large spatial extents making analysis computational intensive, sometimes prohibitively so. To mitigate the computational burden of large‐scale SCR models, we developed an improved formulation of the Bayesian SCR model that uses local evaluation of the individual state‐space (LESS). Based on prior knowledge about a species’ home range size, we created square evaluation windows that restrict the spatial domain in which an individual's detection probability (detector window) and activity center location (AC window) are estimated. We used simulations and empirical data analyses to assess the performance and bias of SCR with LESS. LESS produced unbiased estimates of SCR parameters when the AC window width was ≥5σ (σ: the scale parameter of the half‐normal detection function), and when the detector window extended beyond the edge of the AC window by 2σ. Importantly, LESS considerably decreased the computation time needed for fitting SCR models. In our simulations, LESS increased the computation speed of SCR models up to 57‐fold. We demonstrate the power of this new approach by mapping the density of an elusive large carnivore—the wolverine (Gulo gulo)—with an unprecedented resolution and across the species’ entire range in Norway (> 200,000 km2). Our approach helps overcome a major computational obstacle to population and landscape‐level SCR analyses. The LESS implementation in a Bayesian framework makes the customization and fitting of SCR accessible for practitioners working at scales that are relevant for conservation and management.

Estimating and forecasting spatial population dynamics of apex predators using transnational genetic monitoring

“…This challenge is amplified in our analysis due to the size of the problem (number of individuals, spatial and temporal extent). For this reason, we implemented additional approaches to substantially reduce computation time, including 1) a binomial observation model that allowed substantial reduction of the number of detectors (and therefore runtime) without compromising the precision and accuracy of model estimates (48) and 2) removing unnecessary evaluation of the likelihood whenever the distance between a detector and a predicted AC location was larger than a distance threshold (49). This distance was adjusted for each species and sex to maximize the efficiency of the local evaluation.…”

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

“…Bayesian SCR models are often preferred over maximum likelihood approaches because they allow for customized analyses and are highly flexible (Royle et al 2014, Proffitt et al 2015, Whittington et al 2018. These advantages can, however, result in excessive computational demand, making them inefficient for many analysis problems (Milleret et al 2018(Milleret et al , 2019. Despite large improvements in computer processing speed and ease of parallel processing, it is common for models to run for days or even a month.…”

Estimating abundance with interruptions in data collection using open population spatial capture–recapture models

“…We also estimated year‐specific p 0 to account for annual variation in sampling intensity. To increase computing efficiency, we used a local evaluation of the state space to reduce the number of detectors considered for each individual during the model fit (Milleret et al 2019, Sutherland et al 2019). Searches were conducted continuously from 2013 to 2017, which allowed us to introduce different artificial gaps in the data time series, while having a reference point (scenario without gaps: 11111).…”