Scholarlink

<p>Floods are among the most devastating natural hazards in northeastern Bangladesh, where complex hydrological settings and intense monsoon rainfall frequently cause widespread inundation. This study develops a probabilistic flood hazard and exposure assessment framework for Golapganj Upazila, Sylhet district, by integrating flood frequency analysis, 2D hydraulic modeling (HEC-RAS), and GIS-based exposure evaluation. Six probability distribution functions, Normal, Lognormal, Pearson Type III, Log-Pearson Type III, Gumbel, and Generalized Extreme Value, were applied to discharge and water-level data, and the best-fit distributions were identified using Chi-squared, Kolmogorov–Smirnov, and Anderson–Darling tests. The resulting design discharges and water levels for multiple return periods were used as boundary inputs for the hydraulic model. Model calibration (2020) and validation (2021) at Fenchuganj (SW174) and Sylhet (SW267) stations showed excellent performance (R<sup>2</sup> &gt; 0.98, NSE &gt; 0.98, RMSE ≈ 0.25, PBIAS &lt;  ± 3, RSR &lt; 0.12). Validation against the 2022 Sylhet flood produced a Jaccard Index of 65%, F1-score of 78.8%, and AUC = 0.806, confirming high predictive accuracy. Hazard maps generated for 2-, 5-, 10-, 20-, 50-, and 100-year return periods indicated flood extent increasing from 64.77% to 79.24%, with maximum water depth rising from 6.81&#xa0;m to 9.37&#xa0;m. Exposure analysis revealed that up to 264,656 people, 100.39 km<sup>2</sup> of agricultural land, 160 educational institutions, and 394.16&#xa0;km of roads are affected under the 100-year scenario. This integrated probabilistic-hydraulic framework offers a robust and transferable approach for quantifying flood hazards and exposures in data-scarce regions, such as northeastern Bangladesh.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Probabilistic flood hazard and exposure assessment using hydraulic modelling and geospatial analysis: insights from Golapganj Upazila, Sylhet district, Bangladesh

  • 期刊论文

  • 发表日期：

    2026年5月26日

查看全文
  • Md. Mahfuzar Rahman,
  • Ismat Tahsin Mahmood,
  • Md. Jonaid Al-Habib

摘要

Floods are among the most devastating natural hazards in northeastern Bangladesh, where complex hydrological settings and intense monsoon rainfall frequently cause widespread inundation. This study develops a probabilistic flood hazard and exposure assessment framework for Golapganj Upazila, Sylhet district, by integrating flood frequency analysis, 2D hydraulic modeling (HEC-RAS), and GIS-based exposure evaluation. Six probability distribution functions, Normal, Lognormal, Pearson Type III, Log-Pearson Type III, Gumbel, and Generalized Extreme Value, were applied to discharge and water-level data, and the best-fit distributions were identified using Chi-squared, Kolmogorov–Smirnov, and Anderson–Darling tests. The resulting design discharges and water levels for multiple return periods were used as boundary inputs for the hydraulic model. Model calibration (2020) and validation (2021) at Fenchuganj (SW174) and Sylhet (SW267) stations showed excellent performance (R2 > 0.98, NSE > 0.98, RMSE ≈ 0.25, PBIAS <  ± 3, RSR < 0.12). Validation against the 2022 Sylhet flood produced a Jaccard Index of 65%, F1-score of 78.8%, and AUC = 0.806, confirming high predictive accuracy. Hazard maps generated for 2-, 5-, 10-, 20-, 50-, and 100-year return periods indicated flood extent increasing from 64.77% to 79.24%, with maximum water depth rising from 6.81 m to 9.37 m. Exposure analysis revealed that up to 264,656 people, 100.39 km2 of agricultural land, 160 educational institutions, and 394.16 km of roads are affected under the 100-year scenario. This integrated probabilistic-hydraulic framework offers a robust and transferable approach for quantifying flood hazards and exposures in data-scarce regions, such as northeastern Bangladesh.