What question did this study set out to answer?

To integrate synthetic aperture radar and ensemble machine learning algorithms to map flood extent, estimate water depth, assess land-use exposure, and model flood susceptibility.

April 30, 2026Journal of African Earth SciencesOpen Access

Integrating SAR remote sensing and machine learning for comprehensive flood risk assessment in Kogi state, Nigeria

View Full Paper

Ask AI

Bookmark

Authors

MRMuhammad Haris Satyawan Rahman JIJohnson Ayomide IbukunFederal University of Technology AOAyomide Emmanuel OlubajuFirst Technical University

Discussion

Member takes

Implication

Geospatial modeling study reveals flood dynamics and land exposure across riverine basins, indicating machine learning reliably evaluates flood risk in data-limited regions.

Key Points

To integrate synthetic aperture radar and ensemble machine learning algorithms to map flood extent, estimate water depth, assess land-use exposure, and model flood susceptibility.
Processed Sentinel-1 SAR, Sentinel-2 multispectral, NASA DEM, and CHIRPS precipitation records in Google Earth Engine and ArcGIS Pro, estimating flood depths via cost-distance interpolation.
Trained Random Forest and eXtreme Gradient Boosting (XGBoost) models on 290 spatial points using nine topographic, hydrological, and environmental conditioning factors.
Flood extent spanned from 161 km² in 2020 to 405 km² in 2022, with peak flood depths reaching 3.82 m in riparian corridors.
Cultivated land accounted for 75–85% of total flooded areas, while built-up land exposure doubled from 5 km² to 10 km² due to urban expansion into floodways.
XGBoost predicted flood susceptibility with 93.1% accuracy (AUC = 0.94), identifying elevation (>50%), river proximity (~22%), and slope (~9%) as the dominant conditioning factors.

Cite This Study

Rahman et al. (2026) studied this question.

synapsesocial.com/papers/6a8694fd8acdfd7e65512c39 https://doi.org/10.1016/j.jafrearsci.2026.106168

View Full Paper

Ask AI

Bookmark

Implication

Geospatial modeling study reveals flood dynamics and land exposure across riverine basins, indicating machine learning reliably evaluates flood risk in data-limited regions.

Integrating SAR remote sensing and machine learning for comprehensive flood risk assessment in Kogi state, Nigeria

Discussion

Member takes

Implication

Key Points

Cite This Study

Integrating SAR remote sensing and machine learning for comprehensive flood risk assessment in Kogi state, Nigeria

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider