Flood susceptibility mapping in an arid region of Pakistan through ensemble machine learning model

Abstract

Floods are among the most destructive natural hazards. Therefore, their prediction is pivotal for flood management and public safety. Factors contributing to flooding are different for every region as they depend upon the characteristics of each region. Therefore, this study evaluated the factors contributing to flood and the precise location of high and very high flood susceptibility regions in Karachi. A new ensemble model (LR-SVM-MLP) is introduced to develop the susceptibility map and evaluate influencing factors. This ensemble model was formed by employing a stacking ensemble on Logistic Regression (LR), Support Vector Machine (SVM), and Multi-Layer Perceptron (MLP). A spatial database was generated for the Karachi watershed, which included; twelve conditioning factors as independent variables, 652 flood points and the same number of non-flood points as dependent variables. This data was then randomly divided into 70% and 30% to train and validate models, respectively. To analyse the collinearity among factors and to scrutinize each variable's predictive power, multicollinearity test and Information Gain Ratio were applied, respectively. After training, the models were evaluated on various statistical measures and compared with benchmark models. Results revealed that the proposed ensemble model outperformed Logistic Regression (LR), Support Vector Machine (SVM), and Multi-Layer Perceptron (MLP) and produced a precise and accurate map. Results of the ensemble model showed 99% accuracy in training and 98% accuracy in validation datasets. This ensemble model can be used by flood management authorities and the government to contribute to future research studies.

Google Earth Engine and Machine Learning for Flash Flood Exposure Mapping—Case Study: Tetouan, Morocco

“…To evaluate a multiclass classification, the instances correctly and incorrectly classified for each category must be displayed in a very well-organized tabular representation called a confusion matrix of the predicted class labels against the actual class labels (Table 6). The classification product and associated validation sample can be cross-tabulated to determine a variety of metrics, such as overall accuracy (15), class producers' accuracy (16), class users' accuracy (17), and kappa statistic (18)…”

“…Moreover, Al-Areeq et al's [16] research used a logistic model tree (LT), a Bagging ensemble (BE), k-nearest neighbors, and a kernel support vector machine to map Jeddah, Saudi Arabia's flood vulnerability. In Karachi, Pakistan, by combining a novel set model of Multi-Layer Perceptron, Support Vector Machine, and Logistic Regression (LR), which additionally assesses influencing variables, Yaseen et al [17] created a flood susceptibility map.…”

Assessment of Inundation Susceptibility in the Context of Climate Change, Based on Machine Learning and Remote Sensing: Case Study in Vinh Phuc Province of Vietnam

“…Inundation inventory preparation is essential in building a inundation susceptibility model and describes the relationships between past inundation events and conditioning factors (Nguyen et al 2022b, Yaseen et al 2022. In this study, 106 inundation points were collected from different sources such as the province's Department of Natural Resources and Environment, field missions in 2021 and 2022, and Sentinel 1A imagery.…”

Comparison of Machine Learning Algorithms for Flood Susceptibility Mapping

“…Moreover, they generated 14 conditioning factors as independent variables. Yaseen et al [38] also developed an ensemble FSM method using a combination of three machine learning models, including Logistic Regression, SVM, and MLP. They employed 12 causative factors for FSM.…”