Likelihood analysis of spatial capture‐recapture models for stratified or class structured populations

Abstract

2015. Likelihood analysis of spatial capture-recapture models for stratified or class structured populations. Ecosphere 6(2):22. http://dx.Abstract. We develop a likelihood analysis framework for fitting spatial capture-recapture (SCR) models to data collected on class structured or stratified populations. Our interest is motivated by the necessity of accommodating the problem of missing observations of individual class membership. This is particularly problematic in SCR data arising from DNA analysis of scat, hair or other material, which frequently yields individual identity but fails to identify the sex. Moreover, this can represent a large fraction of the data and, given the typically small sample sizes of many capture-recapture studies based on DNA information, utilization of the data with missing sex information is necessary. We develop the class structured likelihood for the case of missing covariate values, and then we address the scaling of the likelihood so that models with and without class structured parameters can be formally compared regardless of missing values. We apply our class structured model to black bear data collected in New York in which sex could be determined for only 62 of 169 uniquely identified individuals. The models containing sex-specificity of both the intercept of the SCR encounter probability model and the distance coefficient, and including a behavioral response are strongly favored by log-likelihood. Estimated population sex ratio is strongly influenced by sex structure in model parameters illustrating the importance of rigorous modeling of sex differences in capture-recapture models.

Potential for camera-traps and spatial mark-resight models to improve monitoring of the critically endangered West African lion (Panthera leo)

“…Accounting for sex bias in movement in a larger scale study can improve precision and inference about specific habitat selection, and this information could be particularly useful as barriers to female dispersal were recently determined to be a limiting factor in eastern African lion subpopulation persistence (Dolrenry et al 2014). It would follow that consideration of age or stage as an individual covariate in a large scale application of an SMR approach may also elucidate conservation priorities to maintain population persistence and genetic connectivity in an increasingly fragmented landscape (Royle et al 2015).…”

From distance sampling to spatial capture–recapture

“…The sex of individuals is an example. Royle et al (2015) and Efford (2010) developed methods for this case.…”

Comparing spatial capture–recapture modeling and nest count methods to estimate orangutan densities in the Wehea Forest, East Kalimantan, Indonesia

“…Since male and female orangutans are known to have different ranging patterns , we employed models that allowed for population substructuring, specifically sex structuring, to take into account the possibility of differences in space use and encounter rate for males and females (Royle et al, 2015). We considered sex specificity of the parameters σ and p 0 , fitting a set of four models which included: null (no sex effects), p(sex) (sex-specific effects on p 0 , which represents the baseline encounter probability), σ(sex) (sex-specific effects on σ, which represents the extent of space use by individuals), and both (sex-specific effects on p 0 and σ).…”

“…This buffer around the sample locations is at least 4 times the estimated value of σ. We carried out all modeling using the program R as described in Royle et al (2015). All code and scripts used are available as supplemental material to that paper at Ecological Archives, http://dx.doi.org/10.…”

“…Camera trapping is becoming a preferred method for studying rare and elusive species (e.g., O'Connell et al, 2010). Recent advances in statistical techniques, namely spatial capture-recapture modeling or SCR (Borchers and Efford, 2008;Royle and Young, 2008;Efford, 2011;Royle et al, 2013Royle et al, , 2015, allow the calculation of population density from 'captures' of individual animals obtained using camera traps. SCR models have an advantage over conventional capture-recapture (CR) models in that they allow for flexible trap arrangement (e.g., grid vs. linear arrangements that do not require even spacing across the study area; Efford and Fewster, 2013;Tobler and Powell, 2013) and can incorporate both individual-level covariates (e.g., sex or age class) as well as station level covariates (e.g., road vs. trail or habitat; Sollmann et al, 2011).…”