Satellite-derived flood inundation over South Sudan: a multi-sensor assessment

South Sudan has experienced recurrent and widespread flooding, particularly following the extreme 2019–2020 flood events. Satellite-derived flood products are increasingly used for flood monitoring and risk assessment, yet their accuracies vary across different land-cover and hydrological conditions, especially in regions with extensive wetlands, dense vegetation, and persistent cloud cover. This study presents a comparative assessment of widely used satellite-derived flood inundation products across South Sudan and the Sudd wetlands from 2019 to 2024, examining their spatiotemporal consistency and sensitivity to environmental conditions. Monthly flood extent products derived from MODIS, VIIRS, CYGNSS, Sentinel-1, and SWOT were evaluated across wet and dry seasons and under major land-cover classes, including open water, wetlands, vegetation, croplands, rangelands, and built-up areas. All products captured the seasonal flooding dynamics, with inundation increasing during the May-November rainy season and peaking between September and November but differed substantially in flood extent and spatial distribution. MODIS and CYGNSS detected the largest inundation extents, with rainy-season averages of 72,914 km2 and 67,186 km2, respectively, during 2019 to 2021 and showed greater sensitivity to inundation beneath dense vegetation. Sentinel-1 and VIIRS performed more consistently over open water, with rainy-season averages of 42,479 km2 and 43,381 km2, respectively, but underestimated flooding in vegetated areas and exhibited increased false detections over burned surfaces. SWOT detected the largest flood extents overall, averaging 119,001 km2 during 2023–2024 rainy season, while maintaining the seasonal patterns identified by other products. These findings highlight sensor-specific limitations affecting flood monitoring accuracy and demonstrate that product selection should be aligned with land cover and seasonal conditions. The observed patterns of sensor accuracies are relevant to other vegetated and flood-prone wetland environments with similar hydrological and land-cover characteristics.

Authors

Publication Details

Journal
Figshare
Published
2026-11-12
DOI
https://doi.org/10.6084/m9.figshare.34994139
Primary Topic
Flood Risk Assessment and Management
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Satellite-derived flood inundation over South Sudan: a multi-sensor assessment

Elisabeth Stephens, Douglas Mulangwa, Cynthia Gerlein‐Safdi, Ross Maidment et al.
Figshare
Flood Risk Assessment and Management
article

Satellite-derived flood inundation over South Sudan: a multi-sensor assessment

Elisabeth Stephens, Douglas Mulangwa, Cynthia Gerlein‐Safdi, Ross Maidment, Rogério Bonifácio, Md Shahriar Pervez, Evet Naturinda, Emily Black
article en

Abstract

South Sudan has experienced recurrent and widespread flooding, particularly following the extreme 2019–2020 flood events. Satellite-derived flood products are increasingly used for flood monitoring and risk assessment, yet their accuracies vary across different land-cover and hydrological conditions, especially in regions with extensive wetlands, dense vegetation, and persistent cloud cover. This study presents a comparative assessment of widely used satellite-derived flood inundation products across South Sudan and the Sudd wetlands from 2019 to 2024, examining their spatiotemporal consistency and sensitivity to environmental conditions. Monthly flood extent products derived from MODIS, VIIRS, CYGNSS, Sentinel-1, and SWOT were evaluated across wet and dry seasons and under major land-cover classes, including open water, wetlands, vegetation, croplands, rangelands, and built-up areas. All products captured the seasonal flooding dynamics, with inundation increasing during the May-November rainy season and peaking between September and November but differed substantially in flood extent and spatial distribution. MODIS and CYGNSS detected the largest inundation extents, with rainy-season averages of 72,914 km2 and 67,186 km2, respectively, during 2019 to 2021 and showed greater sensitivity to inundation beneath dense vegetation. Sentinel-1 and VIIRS performed more consistently over open water, with rainy-season averages of 42,479 km2 and 43,381 km2, respectively, but underestimated flooding in vegetated areas and exhibited increased false detections over burned surfaces. SWOT detected the largest flood extents overall, averaging 119,001 km2 during 2023–2024 rainy season, while maintaining the seasonal patterns identified by other products. These findings highlight sensor-specific limitations affecting flood monitoring accuracy and demonstrate that product selection should be aligned with land cover and seasonal conditions. The observed patterns of sensor accuracies are relevant to other vegetated and flood-prone wetland environments with similar hydrological and land-cover characteristics.

Figshare
Climate action, Clean water and sanitation
Flood Risk Assessment and Management
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon