Semi-Arid Brasil Water Analysis with Swot and other satellites .

Principle Investigator: Marielle Gosset (Geosciences Environnement Toulouse (GET))

Co-Investigator(s): Rafel Reis, Alfredo Ribeiro

Collaborator(s): DAniel Moreira, Jean-François Cretaux, Eduardo Martins, Renaud Hostache

---

In the semi-arid Ceara (Brazil), the water resource is disseminated through more than twenty thousand reservoirs of all sizes from state monitored reservoirs larger than 5.000 hectares, down to small farm reservoirs of less than 1 ha. Over the years, and as the region underwent several severe droughts, the number of small reservoirs has grown anarchically. The very large number and reduced size of these “açudes” make it challenging to inventory them, analyze the hydrological processes and to monitor the amount of water they hold. On the other hand, there is more and more evidence that these small water bodies have a considerable impact on the water balance of the basin they belong to, and on the water resource available to fill in the state’s reservoirs.

While the largest reservoirs in Ceara are monitored daily by the state, with in situ measurement, the vast majorities of other reservoirs do not benefit from regular measurements, and their number is too large to consider their systematic in situ observation. New techniques need to be proposed to survey a large number of reservoirs despite their small sizes. Remote sensing is a very promising approach and can play a central role for this in the future, considering the new satellite missions devoted to hydrology which are already operational (like the COPERNICUS Sentinels) or expected soon. In particular the Surface Water and Ocean Topography (SWOT) launched in december 2022 is bringing new promises: it is providing the first global monitoring of surface water elevation and extent based on high resolution Ka-Band interferometric altimetry.

To take full advantage of these new data, the processing of the satellite data and images will need to be optimized for the characteristics of the semi-arid Nordeste, coping with the challenge of detecting very small hydrological objects in a dry environment and with substantial cloud cover during the rainy season. Methods based on artificial intelligence and machine learning could help pushing the limits of what is feasible with satellite imagery for water resources monitoring.

During the first semester of 2023, the calibration orbit of SWOT, offering daily revisit, covered the western part of Ceara. In order to exploit this data a subset of eight reservoirs was equipped for intense in situ monitoring with pressure gauges to measure water surface elevation (WSE) and drone/bathymetry survey to obtain a high resolution DEM of the basins. Additional reservoirs have been equipped in the instrumented experimental basin of Forquilha under the science orbit. In addition to these SWOT specific equipment, Ceara’s state data base provides historical monitoring of about 200 reservoirs, information on water usage and a hydrological network including a weather radar for high resolution rainfall monitoring.

This rich data base will be used to i) assess satellite information and SWOT products and ii) analyse the hydrological behaviour of the reservoirs and their (positive or negative ?) role in the water budget and resource.