Community structure determines the predictability of population collapse

Area of Science:

• Ecological dynamics
• Systems biology
• Biodiversity research

Background:

• Early warning signals (EWS) are proposed predictors of biological system collapse.
• Existing EWS research primarily focuses on single-species populations, limiting predictive power for complex ecological systems.
• Understanding EWS in a community context is crucial for reliable ecological forecasting.

Purpose of the Study:

• To investigate the reliability of abundance-based and trait-based EWS in predicting population collapse within simulated ecological communities.
• To explore the relationship between EWS and stability-landscape metrics (e.g., basin of attraction).
• To determine if community-level EWS can forecast species-level collapses.

Main Methods:

• Reconciled quantitative trait evolution with Lotka-Volterra equations to simulate population dynamics in mutualistic and predator-prey communities.
• Generated EWS from both population- and community-level abundance and trait data.
• Analyzed stability-landscape metrics, including basin width and depth.

Main Results:

• Abundance-based EWS and stability-landscape metrics (basin width/depth) are fundamentally linked.
• Species' community position influences their ability to display trait and abundance-based EWS, identifying potential indicator species.
• Trait-based EWS were generally less reliable than abundance-based EWS for predicting species collapse.
• Community-level abundance-based EWS demonstrated reliability comparable to species-level EWS in forecasting collapses.

Conclusions:

• A holistic framework combining abundance-based EWS and stability-landscape theory can enhance forecasting of species loss in communities.
• Community-level abundance-based EWS offer a robust tool for predicting species collapses.
• Identifying indicator species within communities is possible through analyzing EWS and stability metrics.