Felix Girard (Géosciences Environnement Toulouse, France); Hedwige Nikiema (Ki-Zerbo University, Ouagadougo, Burkina Faso); Cheikh Faye (Assane Seck University, Ziguinchor, Senegal); Amadou Abdourahame Toure (Abdou Moumouni University, Niamey, Niger); Ibrahim Mainassara (Abdou Moumouni University, Niamey, Niger); Maxime Wubda (Ki-Zerbo University, Ouagadougo, Burkina Faso); Laurent Kergoat (Géosciences Environnement Toulouse, France)

Lakes, reservoirs, and small water bodies play a pivotal role in West African drylands. They are widely spread all over the landscape, which makes them a primary source of water for people and livestock. However, given their usually small size and their significant temporal variability, their hydrological dynamics remain poorly known at the regional scale. Also, these water bodies are very reactive to climate and human forcing, with a complex and sometimes unexpected dynamics, which questions their future evolution in a context of environmental changes and demographic increase.

This work heavily relies on SWOT data to explore the dynamics of water volumes in the thousands of lakes included in the study area. SWOT performance is evaluated using 1) in-situ data collected in Niger, Burkina Faso, and Senegal, and 2) other satellite estimates by altimetery data (Sentinel-3 and ICESat-2) for the lakes below the orbit, and high-resolution Digital Surface Models(Pleiades) acquired when water levels are at their minimum (Girard et al., 2025a).

We show that SWOT water surface elevations are in excellent agreement with in-situ data and with water level estimations by other satellite sensors (Girard et al., 2025b). SWOT-derived water levels are used to analyze the spatial distribution of water stocks and their seasonal dynamics, including water withdrawals from anthropogenic activities, which is poorly known at the regional scale.

Regarding water surface area, SWOT estimates show a general overestimation compared to Sentinel-2 estimates, which makes volume retrievals by SWOT alone challenging in the study region. Water volume changes are therefore estimated by combining SWOT-dervied water levels with water areas estimated by a deep learning approach applied to optical imagery (de Fleury et al., 2025).

Finally, the elevation-area-volume relationships obtained are used to reconstruct past changes in water volume from water area derived from the Landsat archives (1984 to present) over more than 2000 lakes and reservoirs. This reveals changes and trends in the hydrological long-term evolution in relation to environmental changes (i.e., precipitation, temperature, and vegetation cover) and anthropogenic activities (i.e., reservoir management and land use).


References

F. Girard, L. Kergoat , H. Nikiema, M. Wubda, R. Yonaba, T. Fowé, A. Abdourhamane Touré, I. Mainassara, M. de Fleury and M. Grippa (2025a) “Comparison of methods to derive the height-area relationship of shallow lakes in West Africa using remote sensing.” Water Resources Research, 61,
e2024WR037411. https://doi.org/10.1029/2024WR037411

F. Girard, L. Kergoat, I. Mainassara, M. Wubda, H. Nikiema, A. Abdourhamane Touré, J. Renou, M. Vayre, N. Taburet, N. Picot and M. Grippa (2025b) “Performance of the Surface Water and Ocean Topography (SWOT) Mission for Monitoring Small Lakes in West Africa.” IEEE J-STARS, in press.

M. de Fleury, M. Grippa, M. Brandt, R. Fensholt, F. Reiner, G. Matle Kovacs and L. Kergoat (2025). ”Highly turbid and eutrophic small water bodies in West Africa well identified by a CNN U-Net algorithm.” Remote Sensing Applications: Society and Environment https://doi.org/10.1016/j.rsase.2024.101412