A Bayesian Modelling Framework for Integration of Ecosystem Services into Freshwater Resources Management

Area of Science:

• Environmental Science
• Water Resource Management
• Ecological Modeling

Background:

• Models integrating multiple stressors and ecosystem services (ES) are limited.
• Water resource management requires tools to assess the impact of interventions on ES delivery.

Purpose of the Study:

• To develop a methodology for constructing a Bayesian Belief Network (BBN) integrating catchment and water quality models with expert knowledge.
• To support the integration of ES into water resource management and assess the impact of different management options.

Main Methods:

• Developed a BBN combining model outputs, data, and expert knowledge elicited through small group workshops.
• Assessed four selected ES under management scenarios: no-change, riparian management, and altered livestock numbers.
• Analyzed expert agreement and disagreement, and performed sensitivity analysis on expert information.

Main Results:

• Riparian management and decreased livestock numbers improved the investigated ES to varying degrees compared to a no-change scenario.
• Sensitivity analysis indicated expert disagreements primarily occurred in low-probability situations, minimally impacting overall results.
• High expert agreement was observed for more probable, likely situations, enhancing model applicability.

Conclusions:

• The developed BBN methodology effectively supports ES integration into water resource management.
• Despite complexity and multiple stressors, the model aids decision-making by focusing on the nature of solutions (e.g., riparian or livestock management).
• Expert knowledge, particularly when in agreement for likely scenarios, is crucial for practical catchment management support.