A spatial open-population capture-recapture model

Area of Science:

• Ecology
• Wildlife Biology
• Statistical Modeling

Background:

• Capture-recapture models are crucial for wildlife population estimation.
• Existing models often struggle with open populations and spatial dynamics.
• Integrating spatial information can improve demographic parameter estimates.

Purpose of the Study:

• To develop a novel spatial open-population capture-recapture model.
• To extend existing non-spatial and spatially explicit models.
• To assess the impact of spatial modeling on survival estimate accuracy.

Main Methods:

• A two-dimensional Poisson point process models the superpopulation.
• Birth, death, and dispersal are incorporated using Markovian algorithms and dispersal kernels.
• Distance-dependent sampling and integration over unknown animal locations are used for likelihood estimation.

Main Results:

• The model was successfully applied to brushtail possum (Trichosurus vulpecula) data.
• Spatial modeling significantly reduced bias in capture-recapture survival estimates.
• The model demonstrated robustness to dispersal kernel misspecification.

Conclusions:

• Spatial open-population capture-recapture models offer substantial improvements in ecological studies.
• The developed model provides a flexible framework for analyzing wildlife population dynamics.
• An associated R package facilitates the application of these advanced methods.