Dept. of Computer Science & Engineering and Convergence Engineering for Intelligent Drone, XR Research Center, Sejong University, Seoul, Republic of Korea.
Dept. of Computer Science & Engineering and Convergence Engineering for Intelligent Drone, XR Research Center, Sejong University, Seoul, Republic of Korea; Future & Smart Construction Division, Korea Institute of Civil Engineering and Building Technology, Republic of Korea.
Sci Total Environ. 2023 May 15;873:162285. doi: 10.1016/j.scitotenv.2023.162285. Epub 2023 Feb 17.
Floods are the natural disaster that occurs most frequently due to the weather and causes the most widespread destruction. The purpose of the proposed research is to analyze flood susceptibility mapping (FSM) in the Sulaymaniyah province of Iraq. This study employed a genetic algorithm (GA) to fine-tune parallel ensemble-based machine learning algorithms (random forest (RF) and bootstrap aggregation (Bagging)). Four machine learning algorithms (RF, Bagging, RF-GA, and Bagging-GA) were used to build FSM in the study area. To provide inputs into parallel ensemble-based machine learning algorithms, we gathered and processed data from meteorological (Rainfall), satellite image (flood inventory, normalized difference vegetation index (NDVI), aspect, land cover, altitude, stream power index (SPI), plan curvature, topographic wetness index (TWI), slope) and geographic sources (geology). For this research, Sentinel-1 synthetic aperture radar (SAR) satellite images were utilized to locate flooded areas and create an inventory map of floods. To train and validate the model, we employed 70 % and 30 % of 160 selected flood locations, respectively. Multicollinearity, frequency ratio (FR), and Geodetector methods were used for data preprocessing. Four metrics were utilized to assess the FSM performance: the root mean square error (RMSE), the area under the receiver-operator characteristic curve (AUC-ROC), the Taylor diagram, and the seed cell area index (SCAI). The results exhibited that all the suggested models have high accuracy of prediction, but the performance of Bagging-GA (RMSE (Train = 0.1793, Test = 0.4543)) was slightly better than RF-GA (RMSE (Train = 0.1803, Test = 0.4563)), Bagging (RMSE (Train = 0.2191, Test = 0.4566)), and RF (RMSE (Train = 0.2529, Test = 0.4724)). According to the ROC index, the Bagging-GA model (AUC = 0.935) was the most accurate in flood susceptibility modeling, followed by the RF-GA (AUC = 0.904), the Bagging (AUC = 0.872), and the RF (AUC = 0.847) models. The study's identification of high-risk flood zones and the most significant factors contributing to flooding make it a helpful resource for flood management.
洪水是最常由天气引起并造成最广泛破坏的自然灾害。本研究旨在分析伊拉克苏莱曼尼亚省的洪水易感性图（FSM）。本研究采用遗传算法（GA）微调基于并行集成的机器学习算法（随机森林（RF）和自举聚合（Bagging））。在研究区域中使用了四种机器学习算法（RF、Bagging、RF-GA 和 Bagging-GA）来构建 FSM。为了为基于并行集成的机器学习算法提供输入，我们从气象（降雨量）、卫星图像（洪水清单、归一化差异植被指数（NDVI）、方位、土地覆盖、海拔、水流功率指数（SPI）、平面曲率、地形湿度指数（TWI）、坡度）和地理来源（地质）中收集和处理数据。对于这项研究，我们利用了 Sentinel-1 合成孔径雷达（SAR）卫星图像来定位洪水淹没区域并创建洪水清单图。我们分别使用 70%和 30%的 160 个选定的洪水位置来训练和验证模型。多共线性、频率比（FR）和地理探测器方法用于数据预处理。使用了四个指标来评估 FSM 性能：均方根误差（RMSE）、接收器操作特征曲线（ROC）下的面积（AUC-ROC）、泰勒图和种子单元格面积指数（SCAI）。结果表明，所有建议的模型都具有很高的预测精度，但 Bagging-GA（RMSE（Train=0.1793，Test=0.4543））的性能略优于 RF-GA（RMSE（Train=0.1803，Test=0.4563））、Bagging（RMSE（Train=0.2191，Test=0.4566））和 RF（RMSE（Train=0.2529，Test=0.4724））。根据 ROC 指数，Bagging-GA 模型（AUC=0.935）在洪水易感性建模中最准确，其次是 RF-GA（AUC=0.904）、Bagging（AUC=0.872）和 RF（AUC=0.847）模型。本研究确定了高风险洪水区域和对洪水最有影响的因素，这使其成为洪水管理的有用资源。
