Post-Cyclone Vegetation Recovery and Change Detection in the Sundarbans Mangrove Forest Using Landsat-Derived NDVI and SAVI Indices

This study investigates the long-term impact of Cyclone Sidr (2007) on the vegetation dynamics of the Sundarbans mangrove forest in Bangladesh. Using multi-temporal Landsat 7 ETM + imagery from 2007, 2008, and 2023, vegetation cover changes were analyzed through Normalized Difference Vegetation Index (NDVI) and Soil-Adjusted Vegetation Index (SAVI). Five vegetation classes namely water bodies, bare soil, sparse, intermediate and dense vegetation were derived using NDVI thresholds and change detection analysis. Results indicate a substantial decrease in dense vegetation from 77.07% in 2007 to 57.37% in 2008, followed by gradual recovery to 72.53% by 2023. Cyclone Sidr caused a dramatic increase in sparse vegetation and water bodies due to flooding and forest damage. Accuracy assessment using ground-truth data and Google Earth observations yielded kappa coefficients of 0.81, 0.87, and 0.76 for 2007, 2008, and 2023 respectively. The findings demonstrate the resilience and regeneration capacity of mangrove ecosystems but also emphasize anthropogenic stressors, including land use change. These insights inform conservation strategies, disaster risk reduction, and forest monitoring efforts in cyclone-prone coastal zones.