Open capture-recapture models with heterogeneity: I. Cormack-Jolly-Seber model

Area of Science:

• Ecology
• Population Biology
• Statistical Modeling

Background:

• Traditional capture-recapture models assume uniform survival and capture probabilities for similar animals.
• This simplification can lead to inaccurate population estimates and biased parameter results.
• Advances in computing and closed population models enable more complex analyses.

Purpose of the Study:

• To present a flexible framework for likelihood-based models incorporating individual heterogeneity.
• To address limitations of traditional assumptions in open population capture-recapture studies.
• To improve accuracy in population parameter estimation and model selection.

Main Methods:

• Developed likelihood-based models allowing for individual variability in survival and capture rates.
• Utilized finite mixture models to represent diverse patterns of individual heterogeneity.
• Conditioned models on the first capture of each animal, including the Cormack-Jolly-Seber model as a special case.
• Employed Akaike's information criterion and likelihood ratio tests for model selection.

Main Results:

• The proposed models accommodate a wide range of individual variation patterns.
• Model selection procedures allow for the assessment of factors influencing survival rates.
• Incorporating individual heterogeneity significantly reduces bias in parameter estimates.
• The framework provides a more robust approach compared to traditional methods.

Conclusions:

• Individual heterogeneity in survival and capture rates is crucial for accurate population studies.
• The presented finite mixture models offer a flexible and powerful tool for ecological research.
• Improved model selection and bias reduction support informed wildlife management and conservation strategies.