Flash-flood susceptibility mapping based on XGBoost, random forest and boosted regression trees

Abedi, Rahebeh; Costache, Romulus; Shafizadeh‐Moghadam, Hossein; Pham, Quoc Bao

doi:10.1080/10106049.2021.1920636

Cited by 161 publications

(51 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As for the equally strong performance by XGB, Rampali et al [58] obtained comparable results, finding it to have the highest predictive power in their work on flood risk assessment in India. Correspondingly, Abedi et al [56] in their study in Romania also noted its high performance, although, much like in our study, better than RF. These results are in line with our observations, showing that the selected algorithms can provide sufficiently accurate models for flood prediction in the region.…”

Section: Floods Prediction In the Ourika Watershedsupporting

confidence: 88%

“…XGB and RF were found to be the most reliable for predicting flooding, with AUC scores above 99%. The superior performance of RF in particular over other algorithms has been highlighted in other studies [52][53][54][55][56]. Indeed, it has been shown to be robust to noise and outliers, which are some of the common problems in flood susceptibility modeling.…”

Section: Floods Prediction In the Ourika Watershedmentioning

confidence: 94%

See 1 more Smart Citation

Spatial prediction of flood-susceptible zones in the Ourika watershed of Morocco using machine learning algorithms

Meliho

Khattabi

Zejli

et al. 2022

ACI

View full text Add to dashboard Cite

PurposeThe purpose of the paper is to predict mapping of areas vulnerable to flooding in the Ourika watershed in the High Atlas of Morocco with the aim of providing a useful tool capable of helping in the mitigation and management of floods in the associated region, as well as Morocco as a whole.Design/methodology/approachFour machine learning (ML) algorithms including k-nearest neighbors (KNN), artificial neural network, random forest (RF) and x-gradient boost (XGB) are adopted for modeling. Additionally, 16 predictors divided into categorical and numerical variables are used as inputs for modeling.FindingsThe results showed that RF and XGB were the best performing algorithms, with AUC scores of 99.1 and 99.2%, respectively. Conversely, KNN had the lowest predictive power, scoring 94.4%. Overall, the algorithms predicted that over 60% of the watershed was in the very low flood risk class, while the high flood risk class accounted for less than 15% of the area.Originality/valueThere are limited, if not non-existent studies on modeling using AI tools including ML in the region in predictive modeling of flooding, making this study intriguing.

show abstract

Section: Floods Prediction In the Ourika Watershedsupporting

confidence: 88%

Section: Floods Prediction In the Ourika Watershedmentioning

confidence: 94%

Spatial prediction of flood-susceptible zones in the Ourika watershed of Morocco using machine learning algorithms

Meliho

Khattabi

Zejli

et al. 2022

ACI

View full text Add to dashboard Cite

show abstract

“…Moreover, EGB is a boosting technique known to be a robust feature in data mining algorithms, resulting in better performance than the model LR with an AUC value of 86%. Our ndings are also very close to the results ofAbedi et al (2021). They were used four models named "Classi cation and Regression Tree (CART), RF, Boosted Regression Trees (BRT), and XGBoost".…”

supporting

confidence: 84%

Optimal flood susceptibility model based on performance comparisons of LR, EGB, and RF algorithms

2022

View full text Add to dashboard Cite

The Wadi El-Matulla, located in the eastern desert of Egypt, is the most important water basin. The Qift -Qusayr highway (west-east direction) and the Cairo -Aswan eastern desert highway (north-south) pass through the basin. Many urban areas (villages and industrial areas) and agricultural lands are located at the outlet of these basins. In addition, the basin has a promising potential for future economic and urban development as it is located within the Golden Triangle (governmental megaproject). The current study investigates ood hazard modeling and its impact on the area. To determine the optimal algorithm for ood susceptibility mapping, performance comparisons of three techniques were conducted: logistic regression (LR), extreme gradient boosting (EGB), and random forest (RF). Remote sensing, topographic, geologic, and meteorological data were used with the help of eld visits to provide spatial database and inventory datasets required by the models. The models' performances were assessed using ve statistical indices, ROC, overall accuracy (OAC), kappa index (Ka), RMSE, and mean absolute error (MAE). Our ndings indicate that RF is the optimal algorithm for ood susceptibility mapping, followed by EGB and LR. RF provides high performance with high values of ROC (93%), OAC (88%), and Kappa index (0.85) and the lowest values of RMSE (0.34) and MAE (0.12). Finally, the RF model was veri ed using sentinel-1 images for real oods in 2016 and 2021, and it provides good agreement. The optimal model could help the decision makers and planners to protect the existing facilities and plan the future projects in nonood prone areas.

show abstract

“…At present, many studies have concluded that the RF algorithm has achieved better simulation results in gully erosion susceptibility mapping [20,37,61,62]. The XGBoost algorithm has been widely used in landslide susceptibility mapping [49], flash-flood susceptibility mapping [63], and groundwater vulnerability predictive mapping [64], and excellent simulation results have been achieved. However, there has been less application in gully erosion susceptibility mapping, and further research is still needed to fully evaluate the application of the XGBoost algorithm in gully mapping.…”

Section: Variable Importance and Gesm Model Comparisonmentioning

confidence: 99%

Gully Erosion Susceptibility Mapping in Highly Complex Terrain Using Machine Learning Models

Yang

Wang

Pang

et al. 2021

IJGI

View full text Add to dashboard Cite

Gully erosion is the most severe type of water erosion and is a major land degradation process. Gully erosion susceptibility mapping (GESM)’s efficiency and interpretability remains a challenge, especially in complex terrain areas. In this study, a WoE-MLC model was used to solve the above problem, which combines machine learning classification algorithms and the statistical weight of evidence (WoE) model in the Loess Plateau. The three machine learning (ML) algorithms utilized in this research were random forest (RF), gradient boosted decision trees (GBDT), and extreme gradient boosting (XGBoost). The results showed that: (1) GESM were well predicted by combining both machine learning regression models and WoE-MLC models, with the area under the curve (AUC) values both greater than 0.92, and the latter was more computationally efficient and interpretable; (2) The XGBoost algorithm was more efficient in GESM than the other two algorithms, with the strongest generalization ability and best performance in avoiding overfitting (averaged AUC = 0.947), followed by the RF algorithm (averaged AUC = 0.944), and GBDT algorithm (averaged AUC = 0.938); and (3) slope gradient, land use, and altitude were the main factors for GESM. This study may provide a possible method for gully erosion susceptibility mapping at large scale.

show abstract

Flash-flood susceptibility mapping based on XGBoost, random forest and boosted regression trees

Cited by 161 publications

References 54 publications

Spatial prediction of flood-susceptible zones in the Ourika watershed of Morocco using machine learning algorithms

Spatial prediction of flood-susceptible zones in the Ourika watershed of Morocco using machine learning algorithms

Optimal flood susceptibility model based on performance comparisons of LR, EGB, and RF algorithms

Gully Erosion Susceptibility Mapping in Highly Complex Terrain Using Machine Learning Models

Contact Info

Product

Resources

About