Hurricanes have deep and complex impacts on the structural dynamics and ecology of forests. Mangroves encounter hurricanes more than any other forest ecosystems, and lack of appropriate data and complexity of the issue hinders a proper understanding of forest regeneration and succession after hurricane impact. The Sundarbans, the largest single tract of mangrove forest in the world, frequently encounter hurricanes, but their impact on this forest is yet unclear. In this study, we used Landsat 8 satellite images to assess the impact of hurricane Bulbul (5-10 Nov 2019), which made landfall near the Sundarbans mangrove coast. First, using Google Earth Engine, Landsat 8 Surface Reflectance scenes between the months of January-March 2018 were used to make forest cover classification using classification and regression tree (CART) algorithm and classification accuracy were determined. Two vegetation indices were calculated: (i) NDVI and (ii) NDWI before and after the hurricane. Furthermore, Disturbance Index was calculated using the difference between pre and post-hurricane incidents.
Results/Conclusions
From the classification, the mangrove forest cover was 5608.91 sq km. The classification accuracy was satisfactory, with an overall accuracy of 0.91 and a Kappa coefficient of 0.87. The disturbance occurrences were prevalent on the western part of the Sundarbans, as it made landfall in the Sagar Island near the Indian part of the Sundarbans. Despite the maximum wind speed of 110-120 km hr⁻¹, the hurricane Bulbul impacted 1.33 to 4.33 % forest area. Forest departments of Bangladesh and India stated that there was low tide during the cyclone passage, which might have reduced the impacts. Among the indices, NDWI showed higher sensitivity to hurricane impacts than NDVI. Past studies showed that the use of the NDWI facilitates a greater understanding of the hurricane's impacts and quantitative measurements. This study also demonstrates that NDWI can be more useful in detecting hurricane disturbances for mangrove forests. Further, the results of this study, along with field measurements, will help in understanding the complex interaction between hurricanes and mangrove forests.