Riverine flood potential assessment using metaheuristic hybrid machine learning algorithms DOI Creative Commons
Matej Vojtek, Saeid Janizadeh, Jana Vojteková

et al.

Journal of Flood Risk Management, Journal Year: 2023, Volume and Issue: 16(3)

Published: March 31, 2023

Abstract This study presents the performance of stand‐alone and novel hybrid models combining feed‐forward neural network (FFNN) extreme gradient boosting (XGB) with genetic algorithm (GA) optimization to determine riverine flood potential at a local spatial scale, which is represented by Gidra river basin, Slovakia. Eleven factors robust inventory database, consisting 10,000 non‐flood locations, were used. Using FFNN, XGB, GA‐FFNN GA‐XGB models, 16.5%, 11.0%, 17.1%, 12.3% studied respectively, characterized high very potential. The applied resulted in accuracy, that is, AUC = 0.93 case FFNN model 0.96 XGB model. GA was able raise value for 0.94 0.97, respectively. results this can be useful, especially, identification areas highest floods within next updating Preliminary Flood Risk Assessment, being carried out based on EU Floods Directive.

Language: Английский

Spatio-temporal graph convolutional networks for road network inundation status prediction during urban flooding DOI Creative Commons
Faxi Yuan,

Yuanchang Xu,

Qingchun Li

et al.

Computers Environment and Urban Systems, Journal Year: 2022, Volume and Issue: 97, P. 101870 - 101870

Published: Aug. 22, 2022

Language: Английский

Citations

44
Predicting flood damage probability across the conterminous United States DOI Creative Commons
Elyssa L. Collins, Georgina M. Sanchez, Adam Terando

et al.

Environmental Research Letters, Journal Year: 2022, Volume and Issue: 17(3), P. 034006 - 034006

Published: Feb. 21, 2022

Abstract Floods are the leading cause of natural disaster damages in United States, with billions dollars incurred every year form government payouts, property damages, and agricultural losses. The Federal Emergency Management Agency oversees delineation floodplains to mitigate but disparities exist between locations designated as high risk where flood occur due land use climate changes incomplete floodplain mapping. We harnessed publicly available geospatial datasets random forest algorithms analyze spatial distribution underlying drivers damage probability (FDP) caused by excessive rainfall overflowing water bodies across conterminous States. From this, we produced first spatially complete map FDP for nation, along explicit standard errors four selected cities. trained models using historical reported events ( n = 71 434) a suite predictors (e.g. severity, climate, socio-economic exposure, topographic variables, soil properties, hydrologic characteristics). developed independent each unit code level 2 watershed generated 100 m pixel. Our model classified or no an average area under curve accuracy 0.75; however, performance varied environmental conditions, certain cover classes forest) resulting higher error rates than others wetlands). results identified hotspots multiple regional scales, probabilities common both inland coastal regions. highest tended be areas low elevation, close proximity streams, extreme precipitation, urban road density. Given rapid changes, our study demonstrates efficient approach updating estimates nation.

Language: Английский

Citations

41
Interpretable machine learning for predicting urban flash flood hotspots using intertwined land and built-environment features DOI Creative Commons
Zhewei Liu, Tyler Felton, Ali Mostafavi

et al.

Computers Environment and Urban Systems, Journal Year: 2024, Volume and Issue: 110, P. 102096 - 102096

Published: March 13, 2024

Language: Английский

Citations

14
Satellite Video Remote Sensing for Flood Model Validation DOI Creative Commons
Christopher Masafu, Richard Williams

Water Resources Research, Journal Year: 2024, Volume and Issue: 60(1)

Published: Jan. 1, 2024

Abstract Satellite‐based optical video sensors are poised as the next frontier in remote sensing. Satellite offers unique advantage of capturing transient dynamics floods with potential to supply hitherto unavailable data for assessment hydraulic models. A prerequisite successful application models is their proper calibration and validation. In this investigation, we validate 2D flood model predictions using satellite video‐derived extents velocities. Hydraulic simulations a event 5‐year return period (discharge 722 m 3 s −1 ) were conducted Hydrologic Engineering Center—River Analysis System Darling River at Tilpa, Australia. To extract from studied event, use hybrid transformer‐encoder, convolutional neural network (CNN)‐decoder deep network. We evaluate influence test‐time augmentation (TTA)—the transformations on test image ensembles, during inference. employ Large Scale Particle Image Velocimetry (LSPIV) non‐contact‐based river surface velocity estimation sequential frames. When validating segmented extents, critical success index peaked 94% an average relative improvement 9.5% when TTA was implemented. show that significant value network‐based segmentation, compensating aleatoric uncertainties. The correlations between LSPIV velocities reasonable averaged 0.78. Overall, our investigation demonstrates space‐based studying dynamics.

Language: Английский

Citations

12
Integrating machine learning and geospatial data analysis for comprehensive flood hazard assessment DOI Creative Commons
Chiranjit Singha, Vikas Kumar Rana,

Quoc Bao Pham

et al.

Environmental Science and Pollution Research, Journal Year: 2024, Volume and Issue: 31(35), P. 48497 - 48522

Published: July 20, 2024

Flooding is a major natural hazard worldwide, causing catastrophic damage to communities and infrastructure. Due climate change exacerbating extreme weather events robust flood modeling crucial support disaster resilience adaptation. This study uses multi-sourced geospatial datasets develop an advanced machine learning framework for assessment in the Arambag region of West Bengal, India. The inventory was constructed through Sentinel-1 SAR analysis global databases. Fifteen conditioning factors related topography, land cover, soil, rainfall, proximity, demographics were incorporated. Rigorous training testing diverse models, including RF, AdaBoost, rFerns, XGB, DeepBoost, GBM, SDA, BAM, monmlp, MARS algorithms, undertaken categorical mapping. Model optimization achieved statistical feature selection techniques. Accuracy metrics model interpretability methods like SHAP Boruta implemented evaluate predictive performance. According area under receiver operating characteristic curve (AUC), prediction accuracy models performed around > 80%. RF achieves AUC 0.847 at resampling factor 5, indicating strong discriminative AdaBoost also consistently exhibits good ability, with values 0.839 10. indicated precipitation elevation as most significantly contributing area. Most pointed out southern portions highly susceptible areas. On average, from 17.2 18.6% hazards. In analysis, various nature-inspired algorithms identified selected input parameters assessment, i.e., elevation, precipitation, distance rivers, TWI, geomorphology, lithology, TRI, slope, soil type, curvature, NDVI, roads, gMIS. As per analyses, it found that rivers play roles decision-making process assessment. results majority building footprints (15.27%) are high very risk, followed by those low risk (43.80%), (24.30%), moderate (16.63%). Similarly, cropland affected flooding this categorized into five classes: (16.85%), (17.28%), (16.07%), (16.51%), (33.29%). However, interdisciplinary contributes towards hydraulic hydrological management.

Language: Английский

Citations

12
Analyzing Common Social and Physical Features of Flash-Flood Vulnerability in Urban Areas DOI
Natalie Coleman,

Allison Clarke,

Miguel Esparza

et al.

International Journal of Disaster Risk Reduction, Journal Year: 2025, Volume and Issue: unknown, P. 105437 - 105437

Published: March 1, 2025

Language: Английский

Citations

1
Influencing Factors and Risk Assessment of Precipitation-Induced Flooding in Zhengzhou, China, Based on Random Forest and XGBoost Algorithms DOI Open Access
Xun Liu, Peng Zhou,

Yi-Chen Lin

et al.

International Journal of Environmental Research and Public Health, Journal Year: 2022, Volume and Issue: 19(24), P. 16544 - 16544

Published: Dec. 9, 2022

Due to extreme weather phenomena, precipitation-induced flooding has become a frequent, widespread, and destructive natural disaster. Risk assessments of have thus popular area research. In this study, we studied the severe that occurred in Zhengzhou, Henan Province, China, July 2021. We identified 16 basic indicators, random forest algorithm was used determine contribution each indicator Zhengzhou flood. then optimised selected indicators introduced XGBoost construct risk index assessment model flooding. Our results four primary for study area: total rainfall three consecutive days, daily rainfall, vegetation cover, river system. The storm flood evaluation constructed from 12 indicators: elevation, slope, water system index, night-time light brightness, land-use type, proportion arable land area, gross regional product, elderly population, medical rescue capacity. After streamlining bottom terms rate, it had best performance, with an accuracy rate reaching 91.3%. Very high-risk areas accounted 11.46% 27.50% respectively, their distribution more significantly influenced by extent heavy direction systems, types; medium-risk largest, accounting 33.96% area; second-lowest-risk low-risk together 27.09%. highest were Erqi, Guanchenghui, Jinshui, Zhongyuan, Huizi Districts western part Xinmi City; these should be given priority attention during disaster monitoring early warning prevention control.

Language: Английский

Citations

22
Critical facility accessibility and road criticality assessment considering flood-induced partial failure DOI Creative Commons

Utkarsh Gangwal,

A.R. Siders, Jennifer A. Horney

et al.

Sustainable and Resilient Infrastructure, Journal Year: 2022, Volume and Issue: 8(sup1), P. 337 - 355

Published: Nov. 25, 2022

This paper examines communities' accessibility to critical facilities such as hospitals, emergency medical services, and shelters when facing flooding. We use travel speed reduction account for flood-induced partial road failure. A modified betweenness centrality metric is also introduced calculate the criticality of roads connecting communities facilities. The proposed model are applied Delaware network under 100-year floods. highlights severe facility access loss risk due flood isolation mapped post-flooding suggests a significant time increase reveals disparities among communities, especially vulnerable groups long-term care residents. identified that vital results this research can help inform targeted infrastructure investment decisions hazard mitigation strategies contribute equitable community resilience enhancement.

Language: Английский

Citations

21
Assessing the compound flood risk in coastal areas: Framework formulation and demonstration DOI
Mahjabeen Fatema Mitu, Giulia Sofia, Xinyi Shen

et al.

Journal of Hydrology, Journal Year: 2023, Volume and Issue: 626, P. 130278 - 130278

Published: Oct. 2, 2023

Language: Английский

Citations

12
A Systematic Literature Review on Classification Machine Learning for Urban Flood Hazard Mapping DOI
Maelaynayn El Baida,

Mohamed Hosni,

Farid Boushaba

et al.

Water Resources Management, Journal Year: 2024, Volume and Issue: 38(15), P. 5823 - 5864

Published: Aug. 3, 2024

Language: Английский

Citations

5