The new program 2002-2005

The current project is coordinated by the LA and conducted in collaboration with the UMR EPOC, the university of Bordeaux 1, the "Laboratoire Environnement Hydreco" and the "Laboratoire de Mécanismes des transferts en Géologie" of the Toulouse 3 University. It largely relies on previous studies... It includes an experimental part to have a better documentation on the carbon cycle at the level of a reservoir, in particular for the kinetics of degradation of the submerged organic material, and the evaluation of the carbon production by photosynthesis and its potential contribution to the gas production. Laboratory experiments of are underway for studying the degradation kinetics. Field experiments are also conducted in Guiana and Brazil to study the natural emissions of submerged areas before the impoundment.

The second part of this work is focused on the development of a predictive model of GHG emissions which may be used, for instance, for impact studies before the building of dams in tropical zone. The initial developments are based on SYMPHONY hydrodynamic code developed at the LA. This finite differences model is used in a 2D version. It allows to reproduce in a very satisfying way, the dynamics of the reservoir on the period of interest (95-03). Different biogeochemical modules enable to establish GHG balance are under development. We are expecting to move toward a more advanced stage thanks to the development of a 3D finite volume model with uncovering bench able to reproduce in a realistic way, the complex geometry of the reservoir and to simulate correctly the emission of dissolved CO2, CH4 and O2, and GHG flux to the atmosphere. This model will be validated thanks to the temporal series of data obtained at Petit Saut since its impounding. Once validated, at the end of the project, this model will be used to produce a new estimation of methane emissions ( more generally of GHG emissions) by artificial reservoirs in tropical areas, inventoried in the " World Atlas of Big Dams", after classification according to determining criterions. The model proposed should work at the end of 2004 and the whole objectives should be reached in the late 2005.

Hydrological forcings

• Qi : inlet flow
• Q0: outlet flow
• Qr-Qe = rainfall minus evaporation

• Rn: radiation (sensible heat fux)
• U* friction velocity (momentum flux)

• inlet/outlet fluxes of dissolved carbonaceous species
• F1: emission flux (CH4, CO2) at the reservoir bottom, P1 emission from labile carbon sedimentation
• F2 : CH4 flux above the oxycline (after methane oxidation)
• F3 :diffusive flux (CH4, CO2) at the reservoir surface
• F4: bubbling flux (CH4, CO2)
• F5 : flux from degassing at the reservoir outlet*

Schematic representation of the modelling of Petit Saut reservoir and of dynamic and biogeochemical process controlling the GG's emissions and the methane oxidation in the dam reservoir. For the initial results see Delon et al (Workshop 2004)