Devi Kamalini, Reddy Chundi Chenna, Rahul Kandakatla, Khuntia Jnana Ranjan, Das Bhabani Shankar
Department of Civil Engieering, National Institute of Technology Warangal, Warangal, 506004, India.
Department of Civil Engieering, Chaitanya Bharathi Institute of Technology, Hyderabad, 500075, Telangana, India.
Sci Rep. 2025 Aug 2;15(1):28253. doi: 10.1038/s41598-025-13025-z.
Flood risk assessment (FRA) is a process of evaluating potential flood damage by considering vulnerability of exposed elements and consequences of flood events through risk analysis which recommends the mitigation measures to reduce the impact of floods. This flood risk analysis is a technique used to identify and rank the level of flood risk through modeling and spatial analysis. In the present study, Musi River in the Osmansagar basin is taken in to consideration to evaluate the flood risk, which is located at Hyderabad. The input data collected for the study encompasses Hydrological and Meteorological datasets from Gandipet Guage station in Hyderabad, raster grid data for Osmansagar basin along with several indicators data influencing flood vulnerability. The primary research objective is to conduct a quantitative assessment of the Flood vulnerability index (FVI), to develop a comprehensive flood risk map and to evaluate the magnitude of damaging flood parameters, inundated volume and to analyze the regions inundated in the study area. In risk analysis, FVI determines the degree of which an area is susceptible to the negative impact of flood through various influencing indicators, Flood hazard map segregate the regions based on flood risk level through spatial analysis in Arc-GIS. A part of this study includes an integrated methodology for assessing flood inundation using Quantum Geographic Information Systems (QGIS) data modelling for spatial analysis, Hydraulic Engineering Center's River Analysis System (HEC-RAS) hydraulic modelling for unsteady flow analysis and a machine learning technique i.e. XGBoost, to enhance the accuracy and efficiency of flood risk assessment. Subsequently, inundation map produced using HEC-RAS is superimposed with building footprints to identify vulnerable structures. The results obtained by risk analysis using hydraulic modeling, GIS analysis, and machine learning technique illustrates the flood vulnerability, areas having high flood risk and inundated volume along with predicted flood levels for next 10 years. These findings demonstrate the efficiency of the holistic approach in identifying vulnerability, flood-prone areas and evaluating potential impacts on infrastructure and communities. The outcomes of the study assist the decision-makers to gain valuable insights into flood risk management strategies.
洪水风险评估（FRA）是一个通过风险分析来评估潜在洪水损害的过程，该分析考虑暴露元素的脆弱性和洪水事件的后果，并推荐减轻洪水影响的缓解措施。这种洪水风险分析是一种通过建模和空间分析来识别洪水风险水平并进行排序的技术。在本研究中，考虑了位于海得拉巴德的奥斯曼萨加尔盆地的穆西河，以评估洪水风险。为该研究收集的输入数据包括来自海得拉巴德甘地佩特水位站的水文和气象数据集、奥斯曼萨加尔盆地的栅格网格数据以及影响洪水脆弱性的若干指标数据。主要研究目标是对洪水脆弱性指数（FVI）进行定量评估，绘制全面的洪水风险地图，评估破坏性洪水参数的大小、淹没体积，并分析研究区域内被淹没的区域。在风险分析中，FVI通过各种影响指标确定一个地区易受洪水负面影响的程度，洪水危险地图通过Arc-GIS中的空间分析根据洪水风险水平对区域进行划分。本研究的一部分包括一种综合方法，用于使用量子地理信息系统（QGIS）进行空间分析的数据建模、水利工程中心的河流分析系统（HEC-RAS）进行非恒定流分析的水力建模以及一种机器学习技术即XGBoost，以提高洪水风险评估的准确性和效率。随后，将使用HEC-RAS生成的淹没图与建筑物足迹叠加，以识别易受影响的结构。通过水力建模、GIS分析和机器学习技术进行风险分析得到的结果说明了洪水脆弱性、洪水高风险区域和淹没体积以及未来10年的预测洪水水位。这些发现证明了整体方法在识别脆弱性、洪水易发地区以及评估对基础设施和社区的潜在影响方面的有效性。该研究结果有助于决策者深入了解洪水风险管理策略。
