Jump to Main Content
An ecosystem-based approach and Bayesian modelling to inform coastal planning: A case study of Manly, Australia
- Domínguez-Tejo, Elianny, Metternicht, Graciela
- Environmental science & policy 2019 v.101 pp. 72-86
- Bayesian theory, accountability, anthropogenic activities, case studies, coastal zone management, coasts, databases, decision making, decision support systems, ecosystems, issues and policy, models, monitoring, planning, probability, prototypes, recreation areas, socioeconomics, surveys, watersheds, New South Wales
- Managing coastal areas under an Ecosystem Based Approach–Marine Spatial Planning framework acknowledges the complexity associated with the need to address multiple environmental and socioeconomic issues. The development of efficient management plans is critical to the implementation success of the framework; in this regard, unresolved challenges remain for measuring the effectiveness of planning plans and monitoring implementation progress. This paper describes the development of a Bayesian Belief Network as a prototype Decision Support Tool to assist coastal planning in the catchment areas of the Sydney Harbour, New South Wales, Australia. The model was co-designed with local managers, underpinned by the Drivers-Pressures-States-Impacts-Responses analytical framework to identify key coastal cause-effect relationships, and by the Recreational Opportunity Spectrum framework to account for significant recreational areas. The Bayesian Belief Network was structured on a conceptualisation of the relationships between key pressures affecting coastal management targets (biological areas and human activities) and their impacts on the state of the variables, with emphasis on the beach ecosystem. The socio-economic component of the model consists of predictive socio-economic modelling on preferred beach activities, the assessment of beach recreational settings, and a beach quality survey. Conditional probability tables were derived from local and regional databases. The model structure allows decision makers enhanced understanding of key interactions between management variables, assessment of management scenarios, and increased accountability of planning decisions. Future work on the prototype could expand the model to become a Bayesian Decision Network, through the integration of proposed management actions and their utilities, thereby helping managers identify optimal decisions.