Using Bayesian Networks to link Environmental Flows to Ecosystem Services in the Murray-Darling Basin, Australia

Session H2: Water Resources Management and Planning - Modeling and Software for Improving Decisions and Engaging Stakeholders

In water resources management and planning, it is important to understand both (1.) the linkages between environmental flows and ecosystem condition as well as (2.) the linkages between ecosystem condition and ecosystem services. Bayesian Networks (BNs) are an integration-based modeling tool that could be suitable for representing and quantifying these linkages. To evaluate this, we developed a BN, using the Murray-Darling Basin (Australia) as a case study. The Murray-Darling Basin has both agricultural and ecological values, and has more than a century of water resource development and planning. Recently, ecosystem services have been proposed as a means to progressing water resource planning methods in the basin, and a means to engaging stakeholders in the planning process. On the basis of existing Ecological Character Descriptions of the 16 Ramsar sites of the Murray-Darling Basin, complimented with expert knowledge, we developed four sub- networks: ecosystem condition; regulating services; provisioning services; and cultural services. Our preliminary results showed that Bayesian Networks are very useful as integration tool to visualize and quantify the linkages between environmental flows and ecosystem services. However, the complexity and size of the combined Bayesian Network might be problematic for stakeholder engagement and communication. Extensions of Bayesian Networks, such as Object-Oriented Bayesian Networks (OOBNs) and Dynamic Bayesian Network (DBNs) might be even more suitable in this problem field, especially for communicating the complexity and multi-dimensional nature of the problem to stakeholders.