Day 4

Detailed paper information

Back to list

Paper title Spatial variability of water reflectance in Sahelian lakes and reservoirs: perspectives for monitoring water quality at the regional scale by Sentinel2
  1. Manuela Grippa Géosciences Environnement Toulouse, Toulouse Speaker
  2. Elodie Robert CNRS/LETG
  3. Amadou Abdourhamane Touré Université Abdou-Moumouni, Niamey, Niger
  4. Moussa Boubacar Moussa Géoscience Environnement Toulouse (Université Toulouse 3, CNRS, IRD, CNES), Toulouse, France
  5. Hedwige Nikiema University Ki-Zerbo
  6. Bruno Lartiges Géoscience Environnement Toulouse (Université Toulouse 3, CNRS, IRD, CNES), Toulouse, France
  7. Guillaume Morin INRAE - PACA
  8. Laurent Kergoat Geosciences Environnement Toulouse, GET, CNRS, University of Toulouse
Form of presentation Poster
  • A7. Hydrology and Water Cycle
    • A7.06 EO for monitoring water quality and ecological status in inland waters
Abstract text Monitoring water turbidity in water bodies provides useful information on hydrological processes occurring at the watershed scale as well as on the state of aquatic ecosystems including bacteriological contamination. Quantification of suspended sediment is also important for reservoirs management since it allows to monitor silting that can affect dams functioning while providing important information for water treatment. Remote sensing provides a useful tool for monitoring inland water at the regional scale but only recent satellites provide the spatial and temporal resolution necessary to follow the dynamics of small water bodies.

This study is focused on the Sahelian region where ponds, lakes and reservoirs play a major role for populations. Given their small size, their important temporal variability, and the scarcity of in-situ monitoring network, their dynamics and water quality information is not available at the regional scale. In addition, Sahelian water bodies are very reactive to climate and human forcing and display complex and sometimes unexpected behaviours, like increasing trends in water area across the Sahel, which question their future evolution in a context of environmental changes and demographic increase.
We explore the capability of Sentinel2 optical sensor MSI to retrieve information on waterbodies variability at the large scale, using Google Earth Engine to processes several Sentinel2 tiles. Overall, 1672 Sahelian lakes are analysed and compared to other 5666 lakes in semi-arid regions worldwide.

Water reflectance in the visible and NIR bands significantly vary across different lakes and can reach extremely high values (higher than 0.4 in the NIR band) in some lakes, as for example for several lakes located in Niger which are among the brightest in the world.
In-situ measurements over some of these lakes highlights the high concentration of suspended particulate matter (SPM) which increases water reflectance. In addition, SPM are mainly composed of fine kaolinites , that display low absorption coefficient and so high reflectance. Finally the important fraction of very fine mineral particles (a major volumetric mode is found at 200-300 nanometers) may induce an increased diffusion and a higher back-scattering, which both contribute to increase reflectance. High aerosols conditions and sunglint effects are efficiently masked by the image processing and the post-processing applied (based on thresholds on the MNDWI index and on the reflectance in the blue band) and do not significantly affect the results reported.

At the regional scale the brightest lakes are identified as relatively small lakes situated in area with low vegetation cover, where erosion and sediment transport is more likely. However, the contrary isn’t not always true and low reflectance values can be encountered on small lakes in low vegetated areas, as it happens, for example, for lakes fed by water table or by the flooding of the Niger river.
Observations by high resolution remote sensors such as Sentinel2 are thus an efficient tool to derive information on water color spatial variability in relation to eco-hydrological characteristics at the regional scale.