Ensemble Learning for Urban Flood Segmentation Through the Fusion of Multi-Spectral Satellite Data with Water Spectral Indices Using Row-Wise Cross Attention DOI Creative Commons
Han Xu, Alan Woodley

Remote Sensing, Journal Year: 2024, Volume and Issue: 17(1), P. 90 - 90

Published: Dec. 29, 2024

In post-flood disaster analysis, accurate flood mapping in complex riverine urban areas is critical for effective risk management. Recent studies have explored the use of water-related spectral indices derived from satellite imagery combined with machine learning (ML) models to achieve this purpose. However, relying solely on can lead these overlook crucial contextual features, making it difficult distinguish inundated other similar features like shadows or wet roads. To address this, our research explores a novel approach improve segmentation by integrating row-wise cross attention (CA) module ML ensemble learning. We apply method analysis Brisbane Floods 2022, utilizing 4-band PlanetScope and indices. Applied as pre-processing step, CA fuses band index into each peak-flood image using operation. This process amplifies subtle differences between floodwater characteristics while preserving complete landscape information. The CA-fused datasets are then fed proposed model, which constructed four classic models. A soft voting strategy averages their binary predictions determine final classification pixel. Our demonstrates that enhance sensitivity individual areas, generally improving accuracy. experimental results reveal model achieves high accuracy (approaching 100%) dataset. may be affected overfitting, indicates evaluating additional reduced study encourages further optimize validate its generalizability various contexts.

Language: Английский

Flood Susceptibility Mapping: Integrating Machine Learning and GIS for Enhanced Risk Assessment DOI Creative Commons
Zelalem Demissie,

Prashant Rimal,

Wondwosen M. Seyoum

et al.

Applied Computing and Geosciences, Journal Year: 2024, Volume and Issue: 23, P. 100183 - 100183

Published: Aug. 3, 2024

Flooding presents a formidable challenge in the United States, endangering lives and causing substantial economic damage, averaging around $5 billion annually. Addressing this issue improving community resilience is imperative. This project employed machine learning techniques publicly available data to explore factors influencing flooding develop flood susceptibility maps at various spatial resolutions. Six algorithms, including Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), K-nearest neighbor (KNN), Adaptive Boosting (Ada Boost), Extreme Gradient (XGB) were used. Geospatial datasets comprising thirteen predictor variables 1528 inventory collected since 1996 analyzed. The are rainfall, elevation, slope, aspect, flow direction, accumulation, Topographic Wetness Index (TWI), distance from nearest stream, evapotranspiration, land cover, impervious surface, surface temperature, hydrologic soil group. Five hundred twenty-eight non-flood points randomly created using stream buffer for two scenarios. A total of 2964 classified into flooded (1) non-flooded (0) categories used as target. Overall, testing results showed that XGB RF models performed relatively well both cases over multiple resolutions compared other models, with an accuracy ranging 0.82 0.97. Variable importance analysis depicted such streams, type, surfaces significantly affected prediction, suggesting strong association underlying driving process. improved performance variation susceptible areas across scenarios considering appropriate non-flooding training critical developing flood-susceptibility models. Furthermore, tree-based ensemble algorithms like XG boost stack generalization approach can help achieve robustness model where being evaluated.

Language: Английский

Citations

14
Flood susceptibility mapping in the Yom River Basin, Thailand: stacking ensemble learning using multi-year flood inventory data DOI Creative Commons
Gen Long,

Sarintip Tantanee,

Korakod Nusit

et al.

Geocarto International, Journal Year: 2025, Volume and Issue: 40(1)

Published: Feb. 10, 2025

Language: Английский

Citations

2
Novel optimized deep learning algorithms and explainable artificial intelligence for storm surge susceptibility modeling and management in a flood-prone island DOI

Mohammed J. Alshayeb,

Hoang Thi Hang, Ahmed Ali A. Shohan

et al.

Natural Hazards, Journal Year: 2024, Volume and Issue: 120(6), P. 5099 - 5128

Published: Feb. 2, 2024

Language: Английский

Citations

9
Innovative strategies for pollution assessment in Northern Bangladesh: Mapping pollution areas and tracing metal(loid)s sources in various soil types DOI Creative Commons
Abdullah Al Yeamin, Md. Yousuf Mia,

Shahidur R. Khan

et al.

PLoS ONE, Journal Year: 2025, Volume and Issue: 20(2), P. e0311270 - e0311270

Published: Feb. 3, 2025

This study assessed the risks of soil pollution by heavy metals in Chilmari Upazila, northern Bangladesh, using static environmental resilience (Pi) model soil. Geostatistical modeling and self-organizing maps (SOM) identified areas spatial patterns, while a positive matrix factorization (PMF) revealed sources. The results showed that average concentrations Cr, Pb As were well above background levels. Agricultural industrial soils mainly contaminated with according to Nemerow Pollution Index (NPI), Ecological Risk (ER) Pi Index. Over 70% sites co-contamination was particularly high. A one-way ANOVA significant correlations between Pb, Cu Zn levels human activities. PMF analysis effluents, agrochemicals lithogenic sources main contributors contamination 16%, 41% 43%, respectively. SOM three distinct patterns (Pb-Zn, Cr-Cu-Ni Co-Mn-As), which are consistent results. These emphasize need for stringent measures reduce emissions remediate order improve quality food security.

Language: Английский

Citations

1
Enhancing flood susceptibility mapping in Meghna River basin by introducing ensemble Naive Bayes with stacking algorithms DOI Creative Commons
Abu Reza Md. Towfiqul Islam,

Md. Uzzal Mia,

Nílson Augusto Villa Nova

et al.

Geomatics Natural Hazards and Risk, Journal Year: 2025, Volume and Issue: 16(1)

Published: Feb. 13, 2025

Language: Английский

Citations

1
Comparison of Hydrological Modeling, Artificial Neural Networks and Multi-Criteria Decision Making Approaches for Determining Flood Source Areas DOI
Erfan Mahmoodi, M. Azari, Mohammad Taghi Dastorani

et al.

Water Resources Management, Journal Year: 2024, Volume and Issue: 38(13), P. 5343 - 5363

Published: June 26, 2024

Language: Английский

Citations

5
Flood susceptibility mapping to improve models of species distributions DOI Creative Commons
Elham Ebrahimi, Miguel B. Araújo, Babak Naimi

et al.

Ecological Indicators, Journal Year: 2023, Volume and Issue: 157, P. 111250 - 111250

Published: Nov. 16, 2023

As significant ecosystem disturbances flooding events are expected to increase in both frequency and severity due climate change, underscoring the critical need understand their impact on biodiversity. In this study, we employ advanced remote sensing machine learning methodologies investigate effects of biodiversity, from individual species broader ecological communities. Specifically, utilized Sentinel-1 synthetic aperture radar (SAR) images an ensemble machine-learning algorithms derive a flood susceptibility indicator. Our primary objective is potential benefits incorporating susceptibility, as proxy for risk, into distribution models (SDMs). By doing so, aim improve performance SDMs gain deeper insights consequences floods Within biodiverse landscape Zagros Mountains, crucial Irano-Anatolian biodiversity hotspots, examined sensitivity mammals, amphibians, reptiles' distributions flooding. analysis compared that combined with variables against relying solely variables. The results indicate inclusion significantly improves capacity explain map 67% our study region. Notably, amphibians mammals more profoundly affected by reptiles. highlights importance predictor variable baseline characterization distributions. will obviously depend regional context studied but its relevance likely change. summary, research demonstrates integration potent approach advance data science, monitoring, conservation face climate-induced

Language: Английский

Citations

11
Enhancing spatial resolution of satellite soil moisture data through stacking ensemble learning techniques DOI Creative Commons

Mohammad Sadegh Tahmouresi,

Mohammad Hossein Niksokhan,

Amir Houshang Ehsani

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Oct. 26, 2024

Language: Английский

Citations

4
Flood hazard monitoring and modeling systems for improving climate risk management using machine learning and geospatial models in the Hennops River catchment, Centurion, South Africa DOI Creative Commons
Paidamwoyo Mhangara, Eskinder Gidey,

Matilda Mbazo

et al.

Discover Sustainability, Journal Year: 2025, Volume and Issue: 6(1)

Published: Jan. 2, 2025

Climate change has adversely affected precipitation patterns, leading to increased flooding. However, in most African countries, conventional methods of flood hazard monitoring have hindered risk-reduction measures due operational challenges, technological constraints, and data gaps. To address these issues, robust models Earth observation products that can enhance climate-driven impact assessments need be widely implemented across the continent. This study aimed model risks within Hennops River Catchment area Centurion, South Africa, using Support Vector Machine (SVM), Random Forest (RF), Topographic Wetness Index (TWI), Normalized Difference Water (NDWI) from period 2016–2022. achieve this, we obtained Sentinel-2A Landsat images United States Geological Survey Archive processed them SVM RF models, along with TWI NDWI. The findings indicate frequencies every two years climate change, which causes changes intensity, frequency. Consequently, areas low elevations ranging less than 1305–1430 m catchment are at a higher risk flooding because their proximity River. These locations more likely experience severe they flat or elevation, causing runoff ground accumulate pose greater threat residents. also revealed large number built-up contributed exhibit an average accuracy > 70 percent. research improves flood-hazard understanding builds resilient communities around Catchment.

Language: Английский

Citations

0
PSO-random forest approach to enhance flood-prone area identification: using ground and remote sensing data (case study: Ottawa-Gatineau) DOI

Maedeh Mosalla Tabari,

Hamid Ebadi,

Zahra Alizadeh Zakaria

et al.

Earth Science Informatics, Journal Year: 2025, Volume and Issue: 18(2)

Published: Jan. 30, 2025

Language: Английский

Citations

0