LARGE-SCALE CONVECTIVE CIRCULATION IN THE LAKE CHAD BASIN

In Lake Chad basin, the Quaternary phreatic Aquifer (named hereafter QPA) is impacted by large closed piezometric anomalies as depressions (with a water table depth of ~60 m) and domes, which have a depth of ~15 m [1]. The classic hypothesis to explain the formation of these piezometric depressions is that the water deficit generated by evapotranspiration may not be compensated by lateral groundwater flow [2]. Comparison of the piezometric levels with a free-air gravity map shows a correlation of the location of the piezometric anomalies with the sedimentary infill of the basin. Moreover, bottom hole temperature (BHT) profiles [3] suggest that the thermal structure of the basin is impacted by a ~4 km deep large-scale convective circulation. Indeed, three others deep aquifers have been described, which are separated from the QPA by a clay-rich formation. These observations suggest that the sedimentary structure of the Lake Chad basin may impact the formation of the QPA piezometric anomalies by the development of a large convective circulation of the aquifers.

The classic model implicitly assumes that the QPA is hydrodynamically separated from the others aquifers by the clay-rich formation. However, during this study, some giant polygons have been discovered that presume the vertical connectivity of the different aquifers. Accordingly with these observations, a numerical 2D convective model is developed, which shows that beneath the depressions, a cold descending convective current sucks the QPA while beneath the domes, a warm ascending one creates an overpressure. Variations of the water geochemistry composition between the aquifers is compatible with our convective model if we take into account the cation exchanges and the sulphate-reducing bacteria processes in the clay-rich formation.

[1] Schneider (1966), Carte hydrogéologique au 1/500 000, B.R.G.M. [2] Aranyossy and Ndiaye (1993), J. Water Sci, 6, 81-96. [3] Nwankwo and Ekine (2010), Scientia Africana, 9 (1), 37-45.