Geospatial assessment of climate driven migration in the southwest coastal region of Bangladesh

Abstract

Climate-induced migration poses a growing challenge in Bangladesh’s southwest coastal region. This study aims to identify climate migration hotspots in Barguna District by analyzing both slow-onset (sea-level rise, salinity intrusion, temperature increase, erratic rainfall, waterlogging, and land degradation) and sudden-onset events (cyclones, floods, and riverbank erosion). A mixed-method framework integrating geospatial analysis, remote sensing, and machine learning was applied. Principal component analysis (PCA) explained 92.8% of the climate variance, whereas support vector machine (SVM) classification via Landsat 8 bands (5, 6, 7) effectively mapped migration risk zones with high accuracy (RMSS = 0.89). The key findings revealed significant climate trends: the temperature rose at 0.022 °C/year, the annual rainfall increased by 8.39 mm/year, the sea level rose at 4.19 mm/year, and the peak salinity reached 3.5 ds/m. Sudden hazards, including 30 cyclones (1875–2020), floods (3.91 m depth in 2020), and shoreline erosion (25.95 km lost), further triggered displacement. Migration is highest in areas exposed to riverbank erosion (228.61 sq. km), intense cyclones (229.09 sq. km), and overlapping cyclone and flood hazards (475.93 sq. km). Very high-risk zones (270.49 sq. km; 17.97%) are concentrated in Patharghata, Barguna Sadar, and Betagi. The results offer a spatial decision-support tool for policymakers to strengthen climate adaptation, disaster planning, and migration governance in coastal Bangladesh.