Ecosystem Knowledge Should Replace Coexistence and Stability Assumptions in Ecological Network Modelling.

Area of Science:

• Ecology
• Conservation Biology
• Computational Biology

Background:

• Quantitative population modeling is crucial for conservation but often lacks data.
• Traditional models rely on theoretical ecosystem stability, which may not reflect reality.
• Expert knowledge offers a valuable alternative for data-limited ecological modeling.

Purpose of the Study:

• To develop a data-free population modeling framework using expert-elicited knowledge.
• To create a Bayesian algorithm that removes unrealistic model parameters based on expert input.
• To demonstrate an alternative to theoretical assumptions in ecological modeling for conservation.

Main Methods:

• Developed a Bayesian algorithm for data-free population modeling.
• Systematically removed model parameters deemed impossible by expert knowledge.
• Applied the framework to an ordinary differential equation model, incorporating expert constraints on population size and change rate.

Main Results:

• Expert-derived information leads to more realistic population dynamics than theoretical stability assumptions.
• The new algorithm avoids excessive computational costs.
• The framework significantly impacts population predictions and conservation decision-making.

Conclusions:

• Expert knowledge and field observations are more reliable than theoretical properties for data-free ecological modeling.
• This approach enhances the precision and confidence in conservation predictions.
• The method offers a trustworthy alternative for ecosystem management and decision-making in data-scarce situations.