Geo-spatial Analysis for Flash Flood Susceptibility Mapping in the North-East Haor (Wetland) Region in Bangladesh

Autor: Kazi Redwan Shabab, Mizbah Ahmed Sresto, Md. Nazmul Haque, Md. Mustafa Saroar, Sharmin Siddika
Rok vydání: 2021
Předmět:
Zdroj: Earth Systems and Environment. 5:365-384
ISSN: 2509-9434
2509-9426
DOI: 10.1007/s41748-021-00221-w
Popis: The north-eastern haor (wetland) area of Bangladesh is extremely vulnerable to flash flooding which damages crops, property, and infrastructures. The main objective of this research is to develop and test a methodology for delineation of flash flood hazard zones in the north-eastern haor (wetland) region of Bangladesh which would help to identify and prepare against the flash flood. After reviewing relevant literature, multiple indicators and indexes including slope, soil type, rainfall, land use land cover, drainage density, distance from the river, normalized difference vegetation index, topographic wetness index, and digital elevation model were used to assess the flood susceptibility in the study region. The Landsat 8 images and data from other secondary sources are used in AHP and GIS platforms to obtain the results. The findings show that among the seven haor districts, the susceptibility of flood hazard differs significantly. Significant areas from Brahmanbaria, Kishoreganj, Netrokona, Sunamganj district have been emerged as highly susceptible to flash flood hazard. About 756,970 ha areas were identified as high flood-prone areas. The other three haor districts such as Habiganj, Moulavibazar, and Sylhet are less susceptible to flash floods. About 255,587-ha areas were identified as low flood-prone. Model validity was tested through receiver operating characteristics (ROC) curve. In the ROC curve, the area under the curve (AUC) is 0.839 which means the model’s predictive power is 84%. Similarly, the Wilcoxon signed rank test and the Friedman test also show that the model is valid. Furthermore, the correlation analysis between flood hazard map and inundation map (based on 2017s flood) was also performed where a moderately good R2 value (0.72) was obtained which means the susceptibility map has delineated the flood-prone areas correctly in 73% of cases. In the changing scenario of climate, the risk of flash floods might increase even more. Therefore, the findings could help the policymakers and planners in developing a preparedness system that would limit the property loss and wealth in the north-eastern haor region of Bangladesh.
Databáze: OpenAIRE
Externí odkaz: