Environmental disaster risk analysis using probabilistic models : a case study in Kepulauan Bangka Belitung Province

Abstract The increasing frequency and severity of environmental disasters, such as floods, air pullution, and wildfires pose significant challenges to global sustainability and human resilience. Kepulauan Bangka Belitung, an archipelagic province in Indonesia, faces increasing vulnerability to hydrometeorological disasters, particularly seasonal flooding driven by deforestation, mining activities, and inadequate urban drainage systems. These challenges are compounded by climate variability and rapid land use changes, necessitating robust analytical frameworks for risk assessment and mitigation planning. This study employs probabilistic modeling, integrating probability theory, statistical distributions, and Monte Carlo simulation to quantify flood risk and economic losses across coastal and urban zones. Using historical flood records, rainfall data, and spatial exposure maps, the analysis estimates event probabilities and expected damages, revealing annual losses exceeding IDR 18 billion in high-risk districts such as Pangkalpinang and Belitung Timur. Simulation outputs include probabilistic flood depth distributions and spatial risk maps, supporting evidence based recommendations for infrastructure upgrades, early warning systems, and land use regulation. The findings underscore the utility of probabilistic approaches in enhancing disaster preparedness, informing adaptive policy, and guiding sustainable development in vulnerable island regions.