Spatial Capture–Recapture Models

Abstract

There has been a rapid growth in spatial capture–recapture (SCR) methods in the last decade. This paper provides an overview of existing SCR models and suggestions on how they might develop in future. The core of the paper is a likelihood framework that synthesises existing SCR models. This is used to illustrate similarities and differences between models. The key difference between conventional capture–recapture models and SCR models is that the latter include a spatial point process model for individuals’ locations and allow capture probability to depend on location. This extends the kinds of inferences that can be drawn from capture–recapture surveys, allowing them to address questions of a fundamentally spatial nature, relating to animal distribution, habitat preference, movement patterns, spatial connectivity of habitats and dependence of demographic parameters on spatial variables.