Abstract
AbstractCapture–recapture methods are a common tool in ecological statistics, which have been extended to spatial capture–recapture models for data accompanied by location information. However, standard formulations of these models can be unwieldy and computationally intractable for large spatial scales, many individuals, and/or activity center movement. We provide a cumulative series of methods that yield dramatic improvements in Markov chain Monte Carlo (MCMC) estimation for two examples. These include removing unnecessary computations, integrating out latent states, vectorizing declarations, and restricting calculations to the locality of individuals. Our approaches leverage the flexibility provided by the nimble R package. In our first example, we demonstrate an improvement in MCMC efficiency (the rate of generating effectively independent posterior samples) by a factor of 100. In our second example, we reduce the computing time required to generate 10,000 posterior samples from 4.5 h down to five minutes, and realize an increase in MCMC efficiency by a factor of 25. These approaches can also be applied generally to other spatially indexed hierarchical models. We provide R code for all examples, an executable web‐appendix, and generalized versions of these techniques are made available in the nimbleSCR R package.
Highlights
Capture–recapture methods are the primary tools for estimating abundance and demographic parameters in populations of wild animals (Williams et al 2002)
We demonstrate techniques to decrease the overall algorithm runtime, while increasing Markov chain Monte Carlo (MCMC) mixing to improve the accuracy of posterior inferences
We describe each of these techniques, and nimble code corresponding to each cumulative refinement appears in Appendix S1
Summary
Capture–recapture methods are the primary tools for estimating abundance and demographic parameters in populations of wild animals (Williams et al 2002) These methods rely on statistical modeling of longitudinal encounter histories of individuals in a population, where repeated observations (individuals seen or not seen) within short (closed) periods provide information about population density and structure, and repeated observations over longer (open) periods provide information about demographic rates such as mortality, recruitment, and maturation. Closed SCR models have proved to provide more precise and robust estimates of population densities than nonspatial models (Royle et al 2014), and enable estimation of the distribution of individuals within study areas and parameters relating to individuals’ spacing behavior (Reich and Gardner 2014, Sutherland et al 2015, Royle et al 2016). As opposed to “apparent survival” which confounds mortality and emigration, these SCR models can estimate life-history traits and other population processes in a more mechanistic way than non-spatial models
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
