Bayesian Network-Based Stochastic Management of Marine Protected Areas: Assessing the Risk Impact of Human Activities on Marine Pollution

Abstract

Globally, the sustainable utilization of marine resources and the preservation of marine ecological environments have led to the establishment of marine protected areas (MPAs), which aim to protect biodiversity. However, various factors such as inappropriate anthropogenic activities, illegal fishing, and pollution result in failed conservation efforts. This study referenced literature on MPA and integrated the risks perceived by the industry, the government, and academia to use a stochastic approach for building a risk assessment model based on the Bayesian network for MPAs. The model can help managers select governance measures under random conditions. Using natural and anthropogenic conditions for simulations, this study discovered that increasing the MPA management fund and law enforcement capacity and limiting the number of tourists are short-term strategies for mitigating the impact of fishing. Mid-term and long-term strategies include raising environmental awareness and community awareness, which are time consuming. Limiting the number of tourists is a short-term solution for coral bleaching. Mid-term and long-term solutions include increasing the MPA management fund and promoting environmental education. The results demonstrated that MPA management in different environments produces different outcomes. This study combined various risk perceptions to provide managerial and decision-making models for MPA managers; thus, the information and tools in this study can be used for protecting MPAs in the future.

Keywords: Marine Protected Areas, Stochastic Management, Stakeholder engagement, Multiple Stressors, Pollution Probabilistic Indicators