EO-based hazard flooding maps - South Sudan, 2012-2025

Flooding and health care service disruption in South Sudan.

EO-based hazard flooding maps. 

The dataset covers the period 2012–2025 and provides hazard flooding maps derived from VIIRS data at 375 m/pixel, where each pixel represents the fraction of water coverage. The maps cover an area of interest (AOI) of 189.856 km2  (almost 30% of total South Sudan surface). A hydrologically conditioned Digital Elevation Model (HydroDEM) combined with the HAND algorithm was used to generate binary flood maps at 90 m/pixel resolution. Each product represents a five-day composite, capturing temporal dynamics of flooded areas. The provided hazard maps have four layers: flood frequency (how often a pixel was flooded in a year), flood duration (how many days it was inundated that year), count valid (the number of valid observations for that point) and max consecutive (the number of maximum consecutive flooded days). 

VIIRS’ moderate spatial resolution (375 m/pixel) restricts its ability to capture small or narrow inundated areas, especially in urban or complex terrains. As an optical–infrared instrument, VIIRS cannot penetrate clouds, making flood detection difficult during cloudy or rainy periods when floods are most likely to occur. Mixed land–water pixels further introduce uncertainty, particularly along flood boundaries or in vegetated regions.