Flood susceptibility modeling based on new hybrid intelligence model: Optimization of XGboost model using GA metaheuristic algorithm

Advances in Space Research, Journal Year: 2022, Volume and Issue: 69(9), P. 3301 - 3318

Published: Feb. 22, 2022

Language: Английский

---

Snow avalanche susceptibility mapping using novel tree-based machine learning algorithms (XGBoost, NGBoost, and LightGBM) with eXplainable Artificial Intelligence (XAI) approach

Stochastic Environmental Research and Risk Assessment, Journal Year: 2023, Volume and Issue: 37(6), P. 2243 - 2270

Published: March 13, 2023

Language: Английский

---

Hybrid river stage forecasting based on machine learning with empirical mode decomposition

Applied Water Science, Journal Year: 2024, Volume and Issue: 14(3)

Published: Feb. 15, 2024

Abstract The river stage is certainly an important indicator of how the water level fluctuates overtime. Continuous control can help build early warning floods along rivers and streams. Hence, forecasting stages up to several days in advance very constitutes a challenging task. Over past few decades, use machine learning paradigm investigate complex hydrological systems has gained significant importance, one promising areas investigations. Traditional situ measurements, which are sometime restricted by existing handicaps especially terms regular access any points alongside streams rivers, be overpassed modeling approaches. For more accurate stages, we suggest new framework based on learning. A hybrid approach was developed combining techniques, namely random forest regression (RFR), bootstrap aggregating (Bagging), adaptive boosting (AdaBoost), artificial neural network (ANN), with empirical mode decomposition (EMD) provide robust model. singles models were first applied using only data without preprocessing, following step, decomposed into intrinsic functions (IMF), then used as input variables. According obtained results, proposed showed improved results compared standard RFR EMD for which, error performances metrics drastically reduced, correlation index increased remarkably great changes models’ have taken place. RFR_EMD, Bagging_EMD, AdaBoost_EMD less than ANN_EMD model, had higher R≈0.974, NSE≈0.949, RMSE≈0.330 MAE≈0.175 values. While RFR_EMD Bagging_EMD relatively equal exhibited same accuracies AdaBoost_EMD, superiority obvious. model shows potential signal learning, serve basis insights forecasting.

Language: Английский

---

Comparative study of machine learning methods and GR2M model for monthly runoff prediction

Ain Shams Engineering Journal, Journal Year: 2022, Volume and Issue: 14(4), P. 101941 - 101941

Published: Sept. 6, 2022

Monthly runoff time-series estimation is imperative information for water resources planning and development projects. This article aims to comparatively investigate the applicability of machine learning (ML) methods (i.e., Random Forest (RF), M5 model tree (M5), Support Vector Regression with polynomial kernel function (SVR-poly), radial (SVR-rbf)) GR2M simulating monthly hydrograph. The models experimented at six stations in Thailand's Southern basin. Four performance criteria, including Nash-Sutcliffe Efficiency (NSE), Correlation Coefficient (r), Overall Index (OI), Combined (CI), were utilized comparison. finding results revealed that a low correlation coefficient (r) between input output data sets, ML algorithms showed superior GR2M. In particular, SVR-rbf outstanding over other methods. It expressed could manage problem low-quality simulate under limited available data.

Language: Английский

---

Toward the development of deep learning analyses for snow avalanche releases in mountain regions

Geocarto International, Journal Year: 2021, Volume and Issue: 37(25), P. 7855 - 7880

Published: Sept. 27, 2021

Snow avalanches impose a considerable threat to infrastructure and human safety in snow bound mountain areas. Nevertheless, the spatial prediction of has received little research attention many vulnerable parts world, particularly developing countries. The present study investigates applicability stand alone convolutional neural network (CNN) model, as deep learning approach, along with two metaheuristic algorithms including grey wolf optimization (CNN-GWO) imperialist competitive algorithm (CNN-ICA) avalanche modelling Darvan watershed, Iran. analysis was based on thirteen potential drivers occurrence an inventory map previously documented occurrences. efficiency models' performance evaluated by Area Under Receiver Operating Characteristic curve (AUC) Root Mean Square Error (RMSE). CNN-ICA model yielded highest accuracy both training (AUC= 0.982, RMSE = 0.067) validation 0.972, 0.125) steps, followed CNN-GWO (AUC 0.975 for training, 0.18 AUC 0.968 validation, 0.157 validation). However, standalone CNN showed lower goodness-of-fit 0.864, 0.22) predictive 0.811, 0.330). approach utilized this is broadly applicable identifying areas where hazard likely be high mitigation measures or corresponding land use planning should prioritized.

Language: Английский

---

GIS-Based Spatial Modeling of Snow Avalanches Using Analytic Hierarchy Process. A Case Study of the Šar Mountains, Serbia

Atmosphere, Journal Year: 2022, Volume and Issue: 13(8), P. 1229 - 1229

Published: Aug. 3, 2022

Snow avalanches are one of the most devastating natural hazards in highlands that often cause human casualties and economic losses. The complex process modeling terrain susceptibility requires application modern methods software. prediction this study is based on use geographic information systems (GIS), remote sensing, multicriteria analysis—analytic hierarchy (AHP) territory Šar Mountains (Serbia). Five indicators (lithological, geomorphological, hydrological, vegetation, climatic) were processed, where 14 criteria analyzed. results showed approximately 20% investigated area highly susceptible to 24% has a medium susceptibility. Based results, settlements avalanche protection measures should be applied have been singled out. obtained data can will help local self-governments, emergency management services, mountaineering services mitigate material losses from snow avalanches. This first research Republic Serbia deals with GIS-AHP spatial avalanches, methodology used tested other high mountainous regions.

Language: Английский

---

Snow avalanche susceptibility mapping from tree-based machine learning approaches in ungauged or poorly-gauged regions

CATENA, Journal Year: 2023, Volume and Issue: 224, P. 106997 - 106997

Published: Feb. 14, 2023

Language: Английский

---

Flood Susceptibility Mapping Using Information Fusion Paradigm Integrated with Decision Trees

Water Resources Management, Journal Year: 2024, Volume and Issue: 38(13), P. 5365 - 5383

Published: July 11, 2024

Abstract Accurate estimation of flood-damaged zones in a watershed is prominent guiding framework for developing sustainable strategies. For these purposes, several flood conditioning factor values at flooded and non-flooded points are extracted, those analyzed using decision tree algorithms eight novel information fusion techniques to get more reliable susceptibility mapping. The belief function leaf nodes the fused by named Dempster-Shafer (DS), Fuzzy Gamma Overlay (FGO), Hesitant Weighted Averaging (HFWA), Geometric (HFWG), Ordered (HFWOA), HFWOG, Closeness coefficient (C c ) Euclidean Manhattan distances. extracted from generated maps validated receiver operating characteristics (ROC) curve parameters, seed cell area index (SCAI) classified levels. under ROC (AUROC) training process 0.997 DS, HFWA, HFWOA, C -Euclidean, 0.996 -Manhattan, 0.995 FGO 0.994 HFWG HFWOG. AUROC testing 0.951 0.945 FGO, 0.943 HFWG, 0.941 True Skill Statistics 0.962 0.870 processes. Although present excellent performance, SCAI versus classes fitted assess prediction capabilities further. HFWA HFWOG have first- second-best performances on estimations. Hence, paradigm can be employed combine factors based robust classification method predictions potential levels utilize them land use construction planning management.

Language: Английский

---

Detection of areas prone to flood-induced landslides risk using certainty factor and its hybridization with FAHP, XGBoost and deep learning neural network

Geocarto International, Journal Year: 2021, Volume and Issue: 37(25), P. 7303 - 7338

Published: Aug. 26, 2021

This article is intended to assess the flood-induced landslide susceptibility in Indian state of Assam. study area has high frequency and severity landslides that are triggered by heavy rainfall floods. In order obtain results, two machine learning models (XGBoost DLNN) one fuzzy-multi-criteria decision-making methods (FAHP) were combined with certainty factor (CF) bivariate statistic model. Firstly, 16 predictors 198 locations prepared, this data set being split into training (70%) validating sets (30%). The analysis results shows region's most prone occurrence can be found southern part, while those less these phenomena generally located northern part area. Receiver Operating Characteristic (ROC) curve indicator XGBoost-CF performance model (area under [AUC] = 0.977), followed FAHP-CF (AUC 0.976), DLNN-CF (AUC= 0.974) CF 0.963).

Language: Английский

---

Identification and Assessment of Avalanche Hazards in Aerxiangou Section of Duku Expressway in TianShan Mountainous Region Based on Unmanned Aerial Vehicle Photography

Research in Cold and Arid Regions, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 1, 2025

Language: Английский

---