Optimal sampling design for spatial capture-recapture

Area of Science:

• Ecology
• Wildlife Biology
• Conservation Science

Background:

• Spatial capture-recapture (SCR) is standard for estimating wildlife population density.
• Trap number and spatial arrangement critically influence SCR precision.
• Current sampling design advice is heuristic and lacks empirical validation.

Purpose of the Study:

• To develop a genetic algorithm for optimizing SCR sampling designs.
• To create near-optimal designs minimizing objective functions for density estimation.
• To improve the accuracy and precision of population size estimates.

Main Methods:

• A genetic algorithm was employed to optimize sampling designs.
• Optimization criteria were based on model-based capture probabilities.
• Simulations were used to compare optimized designs against existing recommendations.

Main Results:

• Optimized designs demonstrated reduced bias and improved precision and accuracy in population size estimation.
• The genetic algorithm approach significantly outperformed designs based on current recommendations.
• Simulations confirmed the superiority of the proposed optimization method.

Conclusions:

• The proposed genetic algorithm offers a robust method for generating customized, near-optimal SCR sampling designs.
• This approach enhances the reliability of wildlife density estimates for conservation and research.
• The method is accessible to practitioners via the R package oSCR.