Modelling cyclone-induced flood impact assessment and quantifying the effect on biophysical dynamics employing geospatial techniques in South-West of Bangladesh

Flood Damage Assessment Costal resilience Vegetation dynamics Remote sensing Spatial model
["Miah, Md Tanvir","Hasan, Md. Rakibul","Fariha, Jannatun Nahar","Tammi, Jarin Jannati","Raiyan, Raiyan","Jodder, Pankaj Kanti","Mishu, Remon Ahmed","Usha, Salima Ahamed","Hossain, Md Zakir","Rahaman, Khan Rubayet"] 2025-06-01 期刊论文
(10)
This article investigates the influence of climatic and geographical characteristics in south-western region of Bangladesh on the temporal dynamics of post-cyclone impacts, with a critical focus on biophysical contexts. By quantitatively assessing the environmental consequences of cyclones Amphan (2020), Yaas (2021), Mocha (2023) and Remal (2024), the study offers a nuanced understanding of flood damage extent and vegetation health, measured through advanced remote sensing and geospatial techniques. Using Sentinel-1 (GRD) and Sentinel-2 (MSI) satellite imageries from 2020 to 2024, the study has examined post-cyclone changes of vegetation health and flood damage extent using available indices such as Normalized Difference Vegetation Index (NDVI) and Soil-Adjusted Vegetation Index (SAVI). The results exhibit substantial spatial disparities occurred due to the cyclone events, with NDVI variations ranging from - 0.124 to 0.546 (Amphan), - 0.033 to 0.498 (Mocha), - 0.086 to 0.458 (Yaas), and - 0.061 to 0.362 (Remal), indicating significant ecological stress. Corresponding SAVI changes ranged from - 0.001 to 0.396 (Amphan), - 0.029 to 0.338 (Mocha), - 0.002 to 0.345 (Yaas), and - 0.0524 to 0.269 (Remal). Negative indices underscore potential vegetation degradation, while positive values indicate resilience or post-cyclone recovery. Furthermore, flood damage analysis indicates to a more severe and unevenly distributed impact than previously recognized, particularly in areas with pre-existing vulnerabilities with the damage extent variations between - 35.918 to - 2.0093 (Amphan), - 35.334 to - 4.4059 (Mocha), - 34.806 to - 0.94921 (Yaas), and - 48.469 to 0.00255 (Remal). The Geographically Weighted Regression (GWR), model demonstrates a robust relationship, with r2 values of 0.894, 0.889, 0.899, and 0.95, indicating that approximately 85% of the ecological changes are driven by fluctuations of vegetation due to flood. The insight from this research provides a foundation of flood damage assessment technique occurred by cyclones in a short span of time to aid immediate policy recommendations to enhance resilience in remote areas of the coastal regions of Bangladesh.
来源平台:NATURAL HAZARDS