Spatial capture–recapture models allowing Markovian transience or dispersal

Spatial capture–recapture (SCR) models are arelatively recent development in quantitative ecology, andthey are becoming widely used to model density in studiesof animal populations using camera traps, DNA samplingand other methods which produce spatially explicit individualencounter information. One of the core assumptionsof SCR models is that individuals possess home ranges thatare spatially stationary during the sampling period. Formany species, this assumption is unlikely to be met and,even for species that are typically territorial, individualsmay disperse or exhibit transience at some life stages. Inthis paper we first conduct a simulation study to evaluatethe robustness of estimators of density under ordinary SCRmodels when dispersal or transience is present in thepopulation. Then, using both simulated and real data, wedemonstrate that such models can easily be described in theBUGS language providing a practical framework for theiranalysis, which allows us to evaluate movement dynamicsof species using capture–recapture data. We find that whileestimators of density are extremely robust, even to pathologicallevels of movement (e.g., complete transience), theestimator of the spatial scale parameter of the encounterprobability model is confounded with the dispersal/transiencescale parameter. Thus, use of ordinary SCR modelsto make inferences about density is feasible, but interpretationof SCR model parameters in relation to movementshould be avoided. Instead, when movement dynamics areof interest, such dynamics should be parameterizedexplicitly in the model