Flood susceptibility mapping using support vector regression and hyper‐parameter optimization

Abstract

Floods are both complex and destructive, and in most parts of the world cause injury, death, loss of agricultural land, and social disruption. Flood susceptibility (FS) maps are used by land‐use managers and land owners to identify areas that are at risk from flooding and to plan accordingly. This study uses machine learning ensembles to produce objective and reliable FS maps for the Haraz watershed in northern Iran. Specifically, we test the ability of the support vector regression (SVR), together with linear kernel (LK), base classifier (BC), and hyper‐parameter optimization (HPO), to identify flood‐prone areas in this watershed. We prepared a map of 201 past floods to predict future floods. Of the 201 flood events, 151 (75%) were used for modeling and 50 (25%) were used for validation. Based on the relevant literature and our field survey of the study area, 10 effective factors were selected and prepared for flood zoning. The results show that three of the 10 factors are most important for predicting flood‐sensitive areas, specifically and in order of importance, slope, distance to the river and river. Additionally, the SVR‐HPO model, with area under the curve values of 0.986 and 0.951 for the training and testing phases, outperformed the other two tested models.

Transferability of machine learning-based modeling frameworks across flood events for hindcasting maximum river flood depths in coastal watersheds

“…Consequently, these features exhibit a limited impact on the model predictive performance when compared to other factors. The spatial average of distance to rivers and HAND have limited variability within our watershed and might The finding about the less importance of rainfall in flood estimation concurs with the results reported in the study by Salvati et al (2023) in pinpointing vulnerable regions within a non-coastal medium-sized watershed. The study suggests that rainfall may have a lower impact on flood occurrences or flood depth estimations compared to other influential factors.…”

A comparative spatial analysis of flood susceptibility mapping using boosting machine learning algorithms in Rathnapura, Sri Lanka

“…The flood probability values were then categorized into five susceptibility classes, very low, low, moderate, high, and very high, by applying the natural breaks classification method (Aydin & Iban, 2022; Duwal et al, 2023) (Figure 5). The natural breaks classification method creates meaningful and distinct grouping in the data based on inherent patterns while minimizing the variance within classes and maximizing the variance between classes (Salvati et al, 2023). Despite the variations in the model performances and flood susceptibility extent, all the models indicated that the low‐altitude regions situated along the rivers and residential areas and paddy fields in the central, western, some parts of southeastern and northern regions of the study area are at high risk of flooding.…”

“…In order to address the shortcomings of conventional models, data‐driven approaches have been introduced, which extract information based on data provided, without relying on pre‐established assumptions or a deep understanding of the underlying physical processes and facilitate rapid spatial data analysis (Duwal et al, 2023; Zhu & Zhang, 2022). Over the past few years, numerous machine learning (ML) algorithms such as support vector machine (SVM) (Salvati et al, 2023; Shafapour Tehrany et al, 2017), random forest (RF) (Schmidt et al, 2020; Zhao et al, 2018), decision tree (DT) (Tehrany, Jones, & Shabani, 2019), boosted regression tree (Abedi et al, 2022), artificial neural network (ANN), (Pirnia et al, 2019), adaptive network‐based fuzzy inference system (ANFIS) (Hong et al, 2018), gradient boosting (GB) (Band et al, 2020), eXtreme Gradient Boosting (XGBoost) and AdaBoost (AB) (Aydin & Iban, 2022), have been effectively utilized to evaluate flood susceptibility.…”

Risk Identification of Mountain Torrent Hazard Using Machine Learning and Bayesian Model Averaging Techniques

“…The results indicated that RF has the best predictive performance for landslides and floods, with an AUC of 94.9% and 98.7%. Salvati et al [24] constructed a watershed flood-prone area identification model using support vector regression and three optimization methods (linear kernel, basic classifier, and hyperparameter optimization) based on 201 historical flood maps. The results demonstrated that this method effectively identified floods in the watershed, and the AUC of the test results was above 0.9.…”