Spatial dependence of gas transfer velocity in reservoirs: implications for diffusive flux estimates
Reservoirs may be significant sources of greenhouse gases (GHG) such as CO2 and CH4, but the magnitude of these emissions remains uncertain. One key requirement of diffusive flux estimates is combining robust CO2 and CH4 gas concentration gradients at the water-air interface with realistic gas transfer velocities (k600). There exist many models for estimating k600, primarily based on wind. However, these models do not perform well in large reservoirs with complex shapes. This study aimed to improve current k600 estimation approaches for such reservoirs. We focus on the La Romaine complex, composed of 3 consecutive reservoirs along the La Romaine River in Northeastern Québec. In these reservoirs, we performed 594 flux measurements using floating chambers in 11 field campaigns over four years and calculated the k600 from these fluxes. We also measured wind, morphometry, and other complementary variables. We found that besides wind, the location within the reservoir based on depth and distance to shore strongly modulates k600 and that this effect was reservoir-dependent. Hence, we created a k600 model based on these three variables and modeled k600 to detailed grids on each of the reservoirs. We had previously modeled the CO2, and CH4 surface water gas concentrations for each of these grid cells, which, combined with the modeled k600 layer, yielded a detailed spatially and temporally resolved cartography of diffusive fluxes for both gases in the reservoirs. The improved k600 estimation approach is transferable to other reservoirs worldwide and will help narrow the uncertainties of reservoir GHG fluxes.
Pascal Bodmer, Pedro M. Barbosa, Felipe Rust et Paul A. del Giorgio
| < Précédent | Suivant > |
|---|