Open access Bayesian Belief Networks for estimating the hydrodynamics and shoreline response behind fringing reefs subject to climate changes and reef degradation

• Published in: Environmental modelling & software : with environment data news 2019-09, Vol.119, p.327-340
• Main Authors: Baldock, T.E., Shabani, B., Callaghan, D.P.
• Format: Article
• Language: English
• Subjects: Bayesian analysis; Bayesian belief networks; Beaches; Belief networks; Climate change; Coastal erosion; Coastal morphology; Communication tools; Computational fluid dynamics; Configurations; Coral reefs; Degradation; Elevation; End users; Fluid flow; Fluid mechanics; Hydrodynamics; Interfaces; Open access; Recession; Reefs; Sea level; Sea level rise; Shorelines; User interfaces
• ISSN: 1364-8152; 1873-6726
• DOI: 10.1016/j.envsoft.2019.07.001

Reef-protected beaches are vulnerable to the effects of sea level rise and degradation of their associated fringing reefs. The SWAN hydrodynamic wave model is combined with classical theory describing the planform of beaches in equilibrium with the wave forcing to estimate the reef top hydrodynamics and the shoreline configuration in the lee of the reefs. Open access Bayesian Belief Networks with high accuracy and simple user interfaces have been built to communicate the results. The BBN enable end users to access all the model results and to compare different scenario to determine how changes in the wave climate or reef elevation change the shoreline configuration. The results show that recession of the shoreline in the lee of fringing reefs due to sea level rise may be much greater than that expected on open coast beaches. Loss of reef flat elevation can also lead to severe shoreline erosion. •Bayesian Belief Networks are constructed for end-users to model complex coastal processes.•Models include the hydrodynamics and shoreline response behind fringing reefs.•The impact of climate change and reef degradation is considered.•Results are communicated using open-access BBNKeywords.•End-user experience was used to develop accurate and transparent networks.