Flood susceptibility modelling using advanced ensemble machine learning models

Geoscience Frontiers, Journal Year: 2020, Volume and Issue: 12(3), P. 101075 - 101075

Published: Oct. 5, 2020

Floods are one of nature's most destructive disasters because the immense damage to land, buildings, and human fatalities. It is difficult forecast areas that vulnerable flash flooding due dynamic complex nature floods. Therefore, earlier identification flood susceptible sites can be performed using advanced machine learning models for managing disasters. In this study, we applied assessed two new hybrid ensemble models, namely Dagging Random Subspace (RS) coupled with Artificial Neural Network (ANN), Forest (RF), Support Vector Machine (SVM) which other three state-of-the-art modelling susceptibility maps at Teesta River basin, northern region Bangladesh. The application these includes twelve influencing factors 413 current former points, were transferred in a GIS environment. information gain ratio, multicollinearity diagnostics tests employed determine association between occurrences influential factors. For validation comparison ability predict statistical appraisal measures such as Freidman, Wilcoxon signed-rank, t-paired Receiver Operating Characteristic Curve (ROC) employed. value Area Under (AUC) ROC was above 0.80 all models. modelling, model performs superior, followed by RF, ANN, SVM, RS, then several benchmark approach solution-oriented outcomes outlined paper will assist state local authorities well policy makers reducing flood-related threats also implementation effective mitigation strategies mitigate future damage.

Language: Английский

---

Spatial prediction of groundwater potential mapping based on convolutional neural network (CNN) and support vector regression (SVR)

Journal of Hydrology, Journal Year: 2020, Volume and Issue: 588, P. 125033 - 125033

Published: May 7, 2020

Language: Английский

---

Flood hazard mapping methods: A review

Journal of Hydrology, Journal Year: 2021, Volume and Issue: 603, P. 126846 - 126846

Published: Aug. 28, 2021

Language: Английский

---

Evaluating urban flood risk using hybrid method of TOPSIS and machine learning

International Journal of Disaster Risk Reduction, Journal Year: 2021, Volume and Issue: 66, P. 102614 - 102614

Published: Oct. 1, 2021

With the growth of cities, urban flooding has increasingly become an issue for regional and national governments. The destructive effects floods are magnified in cities. Accurate models flood susceptibility required to mitigate this hazard mitigation build resilience In paper, we evaluate riskin Jiroft city, Iran, using a combination machine learning decision-making methods. Flood maps were created three state-of-the-art methods (support vector machine, random forest, boosted regression tree). metadata supporting our analysis comprises 218 inundation points variety derived factors: slope aspect, elevation, angle, rainfall, distance streets, rivers, land use/land cover, drainages, drainage density, curve number. We then employed TOPSIS tool vulnerability analysis, which is based on socio-economic factors such as building population history, conditions. Finally, risk map maps. Of tested, forest model yielded most accurate map. results indicate that density drainages important modeling. As might be expected, areas with high or very vulnerable flooding. These show mapping provide insights priority planning management, especially limited hydrological data.

Language: Английский

---

Predicting flood susceptibility using LSTM neural networks

Journal of Hydrology, Journal Year: 2020, Volume and Issue: 594, P. 125734 - 125734

Published: Nov. 8, 2020

Identifying floods and producing flood susceptibility maps are crucial steps for decision-makers to prevent manage disasters. Plenty of studies have used machine learning models produce reliable maps. Nevertheless, most research ignores the importance developing appropriate feature engineering methods. In this study, we propose a local spatial sequential long short-term memory neural network (LSS-LSTM) prediction in Shangyou County, China. The three main contributions study summarized below. First all, it is new perspective use deep technique LSTM prediction. Second, integrate an method with predict susceptibility. Third, implement two optimization techniques data augmentation batch normalization further improve performance proposed method. LSS-LSTM can not only capture attribution information conditioning factors data, but also has powerful modelling capabilities deal relationship floods. experimental results demonstrate that achieves satisfactory (93.75% 0.965) terms accuracy area under receiver operating characteristic (ROC) curve.

Language: Английский

---

Flash Flood Susceptibility Modeling Using New Approaches of Hybrid and Ensemble Tree-Based Machine Learning Algorithms

Remote Sensing, Journal Year: 2020, Volume and Issue: 12(21), P. 3568 - 3568

Published: Oct. 31, 2020

Flash flooding is considered one of the most dynamic natural disasters for which measures need to be taken minimize economic damages, adverse effects, and consequences by mapping flood susceptibility. Identifying areas prone flash a crucial step in hazard management. In present study, Kalvan watershed Markazi Province, Iran, was chosen evaluate susceptibility modeling. Thus, detect flood-prone zones this study area, five machine learning (ML) algorithms were tested. These included boosted regression tree (BRT), random forest (RF), parallel (PRF), regularized (RRF), extremely randomized trees (ERT). Fifteen climatic geo-environmental variables used as inputs models. The results showed that ERT optimal model with an area under curve (AUC) value 0.82. rest models’ AUC values, i.e., RRF, PRF, RF, BRT, 0.80, 0.79, 0.78, 0.75, respectively. model, areal coverage very high moderate susceptible 582.56 km2 (28.33%), portion associated low zones. It concluded topographical hydrological parameters, e.g., altitude, slope, rainfall, river’s distance, effective parameters. will play vital role planning implementation mitigation strategies region.

Language: Английский

---

Flood susceptibility modeling in Teesta River basin, Bangladesh using novel ensembles of bagging algorithms

Stochastic Environmental Research and Risk Assessment, Journal Year: 2020, Volume and Issue: 34(12), P. 2277 - 2300

Published: Sept. 4, 2020

Language: Английский

---

Modeling Pan Evaporation Using Gaussian Process Regression K-Nearest Neighbors Random Forest and Support Vector Machines; Comparative Analysis

Atmosphere, Journal Year: 2020, Volume and Issue: 11(1), P. 66 - 66

Published: Jan. 4, 2020

Evaporation is a very important process; it one of the most critical factors in agricultural, hydrological, and meteorological studies. Due to interactions multiple climatic factors, evaporation considered as complex nonlinear phenomenon model. Thus, machine learning methods have gained popularity this realm. In present study, four Gaussian Process Regression (GPR), K-Nearest Neighbors (KNN), Random Forest (RF) Support Vector (SVR) were used predict pan (PE). Meteorological data including PE, temperature (T), relative humidity (RH), wind speed (W), sunny hours (S) collected from 2011 through 2017. The accuracy studied was determined using statistical indices Root Mean Squared Error (RMSE), correlation coefficient (R) Absolute (MAE). Furthermore, Taylor charts utilized for evaluating mentioned models. results study showed that at Gonbad-e Kavus, Gorgan Bandar Torkman stations, GPR with RMSE 1.521 mm/day, 1.244 1.254 KNN 1.991 1.775 1.577 RF 1.614 1.337 1.316 SVR 1.55 1.262 1.275 mm/day had more appropriate performances estimating PE values. It found Kavus Station input parameters T, W S Torkmen stations RH, accurate predictions proposed precise estimation PE. findings current indicated values may be accurately estimated few easily measured parameters.

Language: Английский

---

Convolutional neural network approach for spatial prediction of flood hazard at national scale of Iran

Journal of Hydrology, Journal Year: 2020, Volume and Issue: 591, P. 125552 - 125552

Published: Sept. 20, 2020

Language: Английский

---

Forest Fire Susceptibility Prediction Based on Machine Learning Models with Resampling Algorithms on Remote Sensing Data

Remote Sensing, Journal Year: 2020, Volume and Issue: 12(22), P. 3682 - 3682

Published: Nov. 10, 2020

This study predicts forest fire susceptibility in Chaloos Rood watershed Iran using three machine learning (ML) models—multivariate adaptive regression splines (MARS), support vector (SVM), and boosted tree (BRT). The utilizes 14 set of predictors derived from vegetation indices, climatic variables, environmental factors, topographical features. To assess the suitability models estimating variance bias estimation, training dataset obtained Natural Resources Directorate Mazandaran province was subjected to resampling cross validation (CV), bootstrap, optimism bootstrap techniques. Using inflation factor (VIF), weight indicating strength spatial relationship occurrence assigned each contributing variable. Subsequently, were trained validated receiver operating characteristics (ROC) area under curve (AUC) curve. Results model based on techniques (non, 5- 10-fold CV, bootstrap) produced AUC values 0.78, 0.88, 0.90, 0.86 0.83 for MARS model; 0.82, 0.89, 0.87, 0.84 SVM 0.91 BRT model. Across individual model, CV performed best with 0.90 0.89. Overall, outperformed other all ramification highest value algorithm. Generally, process enhanced prediction performance models.

Language: Английский

---