Water, energy and CO2 exchange over a seasonally flooded forest in the Sahel.

Wednesday, 17 December 2014
Laurent Kergoat1, Valerie Le Dantec2, Franck Timouk1, Pierre Hiernaux1, Eric Mougin1, Grippa Manuela3 and Mamadou Diawara1, (1)GET Géosciences Environnement Toulouse, Toulouse, France, (2)Centre d'Etudes Spatiales de la Biosphere, Toulouse Cedex 9, France, (3)GET Géosciences Environnement Toulouse, toulouse, France
In semi-arid areas like the Sahel, perennial water bodies and temporary-flooded lowlands are critical for a number of activities. In some cases, their existence is simply a necessary condition for human societies to establish. They also play an important role in the water and carbon cycle and have strong ecological values.

As a result of the strong multi-decadal drought that impacted the Sahel in the 70' to 90', a paradoxical increase of ponds and surface runoff has been observed ("Less rain, more water in the ponds", Gardelle 2010). In spite of this, there are excessively few data documenting the consequence of such a paradox on the water and carbon cycle. Here we present 2 years of eddy covariance data collected over the Kelma flooded Acacia forest in the Sahel (15.50 °N), in the frame of the AMMA project. The flooded forest is compared to the other major component of this Sahelian landscape: a grassland and a rocky outcrop sites. All sites are involved in the ALMIP2 data/LSM model comparison.

The seasonal cycle of the flooded forest strongly departs from the surroundings grassland and bare soil sites. Before the rain season, the forest displays the strongest net radiation and sensible heat flux. Air temperature within the canopy reaches extremely high values. During the flood, it turns to the lowest sensible heat flux. In fact, due to an oasis effect, this flux is negative during the late flood. Water fluxes turn from almost zero in the dry season to strong evaporation during the flood, since it uses additional energy provided by negative sensible heat flux. The eddy covariance fluxes are consistent with sap flow data, showing that the flood greatly increases the length of the growing season. CO2 fluxes over the forest were twice as large as over the grassland, and the growing season was also longer, giving a much larger annual photosynthesis.

In view of these data and data over surroundings grasslands and bare soil, as well as data from a long-term ecological survey, we discuss the importance of the flooded forest in terms of fluxes and productivity.