Spatial Capture-Recapture with Partial Identity: An Application to Camera Traps

Abstract

Camera trapping surveys frequently capture individuals whose identity is only known from a single flank. The most widely used methods for incorporating these partial identity individuals into density analyses discard some of the partial identity capture histories, reducing precision, and while not previously recognized, introducing bias. Here, we present the spatial partial identity model (SPIM), which uses the spatial location where partial identity samples are captured to probabilistically resolve their complete identities, allowing all partial identity samples to be used in the analysis. We show that the SPIM out-performs other analytical alternatives. We then apply the SPIM to an ocelot data set collected on a trapping array with double-camera stations and a bobcat data set collected on a trapping array with single-camera stations. The SPIM improves inference in both cases and in the ocelot example, individual sex determined from photographs is used to further resolve partial identities, one of which is resolved to near certainty. The SPIM opens the door for the investigation of trapping designs that deviate from the standard 2 camera design, the combination of other data types between which identities cannot be deterministically linked, and can be extended to the problem of partial genotypes. * Corresponding Author 1 2 AUGUSTINE ET AL.1. Introduction. The inferential goal of capture-recapture studies is to estimate population density, D, or abundance, N , in the presence of imperfect detection. Individuals are either naturally or manually marked and subjected to repeated capture attempts in order to estimate their capture probability and thus D or N . Generally, capture-recapture models for wildlife species regard the individual identity of each capture event as known; however in practice, the identities of individuals for some capture events can be ambiguous or erroneous. In live-capture studies, tags can be lost. In camera trapping studies, researchers often obtain partial identity samples -left-only and right-only photographs that cannot be deterministically linked. In genetic capture-recapture studies, partial genotypes and allelic dropout can lead to partial identification or misidentification, respectively. Statistical models have been developed to address the problem of imperfect identification in live capture using double tagging (e.g. Wimmer et al., 2013) and in camera trap and genetic capture-recapture studies by regarding the complete identification of partial or potentially erroneous samples as latent and specifying models for both the capture-recapture process and the imperfect observation process conditional on the capture (e.g. McClintock et al., 2013;Bonner and Holmberg, 2013;Wright et al., 2009). However, relatively little attention has been paid to one of the most important determinants of sample identity-the spatial location where it was collected. The identity of ambiguous samples should more likely match the identity of other samples collected closer together in space than those collecte...

Evaluating spatially explicit density estimates of unmarked wildlife detected by remote cameras

“…When marked and unmarked individuals cannot be reconciled with certainty, density will be overestimated—as observed here—because the number of unmarked individuals is inflated by misidentification of marked animals. Extending partial identity models (Augustine et al., ) that address irreconcilability of two individual identities produced by the same individual (e.g. DNA and individual markings) to address correlation among spatially proximate marked and unmarked detections might reduce this bias, while preserving gains in precision from incorporating individual detections.…”

Generalized spatial mark–resight models with an application to grizzly bears

“…u | n u , y obs ), but the latent encounter histories can be sampled from their posterior distribution using Markov chain Monte Carlo (MCMC) sampling (see Appendix ). This flexible approach for updating encounter histories can be easily expanded to include other covariates such as sex or age of marked, unknown or unmarked animals that can be used to further limit the pool of potential individuals and thus increase precision of parameter estimates (Augustine et al., ). Conditional on the partially observed encounter histories, y R can be modelled with σ and resighting encounter rates using standard SCR observation models.…”

Integrating multiple genetic detection methods to estimate population density of social and territorial carnivores

“…Both Mexican gray wolves and red wolves are annually radiomonitored and attempts to estimate their population densities and abundances have either failed because of sparse detection data or resulted in imprecise and possibly biased estimates (Adams et al 2003, Piaggio et al 2016, Seamster et al 2016, Hinton et al 2017. Spatial partial identity models that probabilistically link spatial detections of partial individuality to improve precision of spatial capture-recapture density estimates (Augustine et al 2018a) have been recently extended to incorporate partial genotypes from noninvasive genetic detection data (Augustine et al 2018b). Spatial partial identity models that probabilistically link spatial detections of partial individuality to improve precision of spatial capture-recapture density estimates (Augustine et al 2018a) have been recently extended to incorporate partial genotypes from noninvasive genetic detection data (Augustine et al 2018b).…”