Appraisal of flood susceptibility of Hooghly basin, India using Shannon entropy (SE) and fuzzy analytical hierarchy process (FAHP)

Environmental Earth Sciences, Journal Year: 2024, Volume and Issue: 83(16)

Published: Aug. 1, 2024

Language: Английский

---

Flood modeling using HEC-RAS 2D and IBER 2D: a comparative study

Water Science & Technology Water Supply, Journal Year: 2024, Volume and Issue: 24(9), P. 3061 - 3076

Published: Aug. 8, 2024

ABSTRACT Floods have become a major risk in urban areas. Identifying areas at of flooding has crucial to reducing this and protecting lives property. Various softwares are used identify flooding. In work, comparative study was made between two well-known software flood modeling, HEC-RAS 2D IBER 2D. The comparison included water depth, flow velocity, the extent El Bayadh City Algeria for return periods 50, 100, 1,000 years. Despite existence some differences results depths velocities compared, showed good agreement softwares. gave higher values than three estimating provided velocity estimation. flooded almost identical, relative deviation compared varies 1 3%. A map produced most vulnerable This can be help mitigate

Language: Английский

---

SAR-driven flood inventory and multi-factor ensemble susceptibility modelling using machine learning frameworks

Geomatics Natural Hazards and Risk, Journal Year: 2024, Volume and Issue: 15(1)

Published: Oct. 16, 2024

Climate change has substantially increased both the occurrence and intensity of flood events, particularly in Indian subcontinent, exacerbating threats to human populations economic infrastructure. The present research employed novel ML models—LR, SVM, RF, XGBoost, DNN, Stacking Ensemble—developed Python environment leveraged 18 flood-influencing factors delineate flood-prone areas with precision. A comprehensive inventory, obtained from Sentinel-1 Synthetic Aperture Radar (SAR) data using Google Earth Engine (GEE) platform, provided empirical for entire model training validation. Model performance was assessed precision, recall, F1-score, accuracy, ROC-AUC metrics. results highlighted Ensemble's superior predictive ability (0.965), followed closely by, XGBoost (0.934), DNN (0.929), RF (0.925), LR (0.921), SVM (0.920) respectively, establishing feasibility applications disaster management. maps depicting susceptibility flooding generated by current provide actionable insights decision-makers, city planners, authorities responsible management, guiding infrastructural community resilience enhancements against risks.

Language: Английский

---

On the application of machine learning into flood modeling: data consideration and modeling algorithm

SSRN Electronic Journal, Journal Year: 2023, Volume and Issue: unknown

Published: Jan. 1, 2023

This article reviews the literature on application of Machine Learning (ML) to identify flood-prone areas, covering studies published since 2013. The review focuses data considerations, such as specifics study area and conditioning factors, well ML algorithms used flooding areas. 100 scientific articles were analyzed through a wide scope geographical ranging from arid tropical climates small catchments large river basins, evaluate influence features, historical flood occurrences, climatic patterns, urbanization, availability susceptibility modeling (FSM). Iran, India, China, Vietnam are most frequently studied locations. slope land, topographic wetness index, land use cover, rainfall levels distance rivers key factors in at least 61% reviewed articles. Furthermore, employed can be categorized into various types: statistical, kernel-based, tree-based, Neural Network (NN)-based, ensemble, hybrid approaches. NN-based models, Long Short-Term Memory (LSTM) Recurrent Networks (RNNs), excel solving high-dimensional problems but face challenges related reliability overfitting. Kernel-based approaches require optimal configuration trial-and-error process, while tree-based models offer simplicity less prone overfitting, although they may precise. Among these, ensemble generally outperform traditional methods, despite their own limitations. These methods primarily focus event-based floods, limiting ability make real-time predictions due lack time-series data. Additionally, restrictions given consistency validity. They often inconsistent data, where conditions input parameter values not aligned time space. discrepancy undermines models' reliability. Consistent valid datasets imperative for accurate model development.

Language: Английский

---

Flood subsidence susceptibility mapping using persistent scatterer SAR interferometry technique coupled with novel metaheuristic approaches from Jeddah, Saudi Arabia

Neural Computing and Applications, Journal Year: 2024, Volume and Issue: 36(26), P. 15961 - 15980

Published: May 20, 2024

Language: Английский

---