You are here:
Assessing the sensitivity of coral reef condition indicators to local and global stressors with Bayesian networks
Citation:
Carriger, J., W. Fisher, AND S. Yee. Assessing the sensitivity of coral reef condition indicators to local and global stressors with Bayesian networks. 3rd Annual BayesiaLab Conference, Fairfax, VA, October 04 - 10, 2015.
Impact/Purpose:
Abstract for poster presentation at BayesiaLab User Conference
Description:
Coral reefs are highly valued ecosystems that are currently imperiled. Although the value of coral reefs to human societies is only just being investigated and better understood, for many local and global economies coral reefs are important providers of ecosystem services that support cultural, social, market and non-use values. Reef stressors are from both local and global sources. Shallow water reefs are exposed to a variety of conventional and non-conventional pollutants from the coastal zones as well as increasing ocean temperature and acidification. Given the importance of coral reef ecosystems and the many policy challenges, a Bayesian network approach can be beneficial for evaluating the risk to coral reefs from land and sea. To this end, we used available data to evaluate the overlap between local stressors (overfishing, watershed-based pollution, marine-based pollution, and coastal development threats), global stressors (acidification and thermal stress) and management effectiveness with indicators of coral reef health (coral health index, live coral cover, population bleaching, colony bleaching and recently killed corals). Each of the coral indicators had Bayesian networks constructed at a global level and a regional level for Pacific, Atlantic, Australia, Middle East, Indian Ocean, and Southeast Asia coral reef locations with available data. In addition, Bayesian networks were constructed for each of the above scenarios using assumed causal and acausal connections between variables. Sensitivity analysis helped evaluate the relationships between different stressors and reef condition indicators. Validation statistics and dependency evaluations indicated issues with the constructed networks. However, the Bayesian network approach helped to explore the interrelationships among existing databases used for policy development in coral reef management, examine the uncertainties in the relationships between stressors and reef condition, and identify knowledge gaps in the information sources.