Analyzing dynamic species abundance distributions using generalized linear mixed models

Area of Science:

• Ecology and Evolutionary Biology
• Population Dynamics
• Ecological Modeling

Background:

• Understanding ecological community dynamics and their response to environmental change is crucial.
• Traditional population dynamic models struggle to incorporate vital ecological parameters like environmental noise and density regulation.
• Integrating key species characteristics into community dynamics models presents a significant challenge.

Purpose of the Study:

• To demonstrate the application of generalized linear mixed models (GLMMs) for fitting dynamic species abundance distributions.
• To provide ecological interpretations for random effects within GLMMs, linking them to environmental stochasticity and species-specific variations.
• To assess the accuracy of parameter estimation in relation to density regulation strength.

Main Methods:

• Utilized intercept-only generalized linear mixed models with various random effects to model dynamic species abundance.
• Interpreted random effects to represent general and species-specific responses to temporal/spatial environmental stochasticity and variations in growth rate/carrying capacity.
• Employed simulations to evaluate estimation accuracy based on density regulation strength.

Main Results:

• Successfully fitted dynamic species abundance distributions using GLMMs with ecologically meaningful random effects.
• Demonstrated the estimation of population dynamic parameters and covariances, including statistical uncertainties, for fish and bat communities.
• Identified species heterogeneity as the primary driver of spatial and temporal community similarity in both case studies.

Conclusions:

• GLMMs provide a flexible framework for analyzing complex ecological community dynamics and incorporating key population parameters.
• Species heterogeneity plays a critical role in shaping community structure and similarity over space and time.
• The developed modeling approach enhances our ability to detect and understand changes in ecological communities under environmental pressures.