A unified parametric regression model for recapture studies with random removals in continuous time

Area of Science:

• Ecology
• Statistical Ecology
• Population Dynamics

Background:

• Accurate population size estimation is crucial for wildlife management and ecological research.
• Existing methods for population estimation often have limitations, particularly with complex sampling designs or removal processes.
• Developing more efficient and robust estimators is an ongoing challenge in ecological studies.

Purpose of the Study:

• To develop a novel and more efficient population size estimator.
• To incorporate random removals into the recapturing process for greater realism.
• To evaluate the performance of the proposed estimator through simulations and empirical data.

Main Methods:

• Combining conditional likelihood based on counting processes with a Horvitz-Thompson estimator.
• Developing a statistical framework that accommodates random removals in mark-recapture studies.
• Conducting simulation studies to compare the proposed estimator with existing methods.
• Applying the estimator to real-world data from bird banding and small mammal studies.

Main Results:

• The proposed population size estimator demonstrates improved efficiency compared to existing methods.
• The estimator performs well under various simulation scenarios, including those with random removals.
• Empirical examples show the practical applicability and effectiveness of the new method.
• The approach provides a valuable tool for ecological data analysis.

Conclusions:

• The integrated approach offers a statistically sound and more efficient method for population size estimation.
• This estimator enhances the analysis of mark-recapture data, especially when removals occur.
• The findings contribute to advancing quantitative methods in ecological monitoring and conservation.