CO2 flux spatial variability in a tropical reservoir in the Central Amazonia

Monday, 15 December 2014
Raoni aquino silva de Santana1, Roseilson Souza do Vale2, Julio Tota3, Scott D Miller4, Rardiles Branches Ferreira Jr3, Eliane Gomes Alves5, Sarah Suely Alves Batalha3 and Rodrigo Augusto Ferreira de Souza6, (1)National Institute for Amazon Research (INPA), manaus, Brazil, (2)INPA National Institute of Amazonian Research, Manaus, Brazil, (3)Federal University of Western Para, Santarem, Brazil, (4)SUNY Albany, Albany, NY, United States, (5)INPA National Institute of Amazonian Research, Climate and Environment Department, Manaus, Brazil, (6)Organization Not Listed, Washington, DC, United States
The carbon budget over water surfaces in the Amazon has an important role in the total budget of this greenhouse gas a regional and global scale. However, more accurate estimates of the spatial and temporal distribution of the CO2 flux over those water surfaces are still required. In this context, this study aims to understand the spatial distribution of CO2 flux in the Balbina hydroelectric reservoir, located at Presidente Figueiredo city, Amazonas, Brazil. The floating chamber method was used to measure and calculate the CO2 flux. This method coup a chamber of known volume with an infrared gas analyzer (LiCor, LI-840A). Measurements were performed at 1 Hz during 20-30 minutes at 5 different points of the reservoir, four upstream (two near the edge and two in the middle) and one downstream of the dam. At all locations the surface water was supersaturated in pCO2 and fluxes were from the water to the atmosphere. The maximum CO2 flux observed was 1.2 μmol m-2 s-1 at the center point of the reservoir upstream the dam. The minimum CO2 flux was 0.05 μmol m-2 s-1, observed near the edge on the upstream side of the dam. On average, CO2 fluxes were larger downstream of the dam, 0.7 μmol m-2 s-1, compared to upstream, 0.45 μmol m-2 s-1. This pattern is consistent with that found in previous studies at this site using other flux estimation methods, and is consistent with turbulent mixing promoted by the water turbine. However, the mean CO2 flux for all measured points using the chambers, 0.47 μmol m-2 s-1, was much lower than those previously found using other methods. The reason for the difference between methods is unclear. In situ deployment of multiple flux estimation methods would be valuable, as would longer periods of measurements.