Building matrix population models when individuals are non-identifiable

Area of Science:

• Ecology
• Evolutionary Biology
• Population Dynamics

Background:

• Matrix Population Models (MPMs) are crucial tools in ecology and evolution for analyzing life cycles.
• Traditional MPM parameterization requires identifiable individuals and complete data on survival, fertility, and stage development times.
• Data limitations, especially with non-identifiable individuals or incomplete cohort data, pose challenges for accurate MPM construction.

Purpose of the Study:

• To present a simplified procedure for parameterizing MPMs with non-identifiable individual data from cohorts.
• To develop a method that does not require external estimation of stage development times, a common source of error.
• To demonstrate the applicability and advantages of the new procedure using a real-world dataset.

Main Methods:

• A novel procedure for MPM parameterization using cohort data from non-identifiable individuals.
• The method bypasses the need for estimating stage residence times.
• Validation using a laboratory cohort dataset of Eratyrus mucronatus.

Main Results:

• The proposed procedure successfully parameterizes MPMs with non-identifiable individuals.
• It eliminates the requirement for stage development time data, simplifying the process.
• The method yields identical MPM estimates compared to traditional methods when stage durations are known and accurate.

Conclusions:

• This simplified MPM parameterization method enhances accessibility and accuracy, particularly for non-identifiable individual data.
• It offers a robust alternative that reduces potential errors associated with stage development time estimation.
• The procedure is broadly applicable and beneficial even when individual data is identifiable.