Investigating Physical Controls on Methane and Carbon Dioxide Fluxes over Reservoirs Using the Eddy Covariance Method

Reservoirs are a globally important source of carbon to the atmosphere. Several recent studies have found that both carbon dioxide (CO2) and methane (CH4) emissions from reservoirs are currently being underestimated by up to 50%. This underestimation is due to inadequate characterization of both spatial variability (e.g. ebullition and CO2 surface water concentration hot spots) and temporal variability (e.g. diurnal patterns, seasonal differences, and pulses driven by weather events or other disturbances). Use of the eddy covariance technique to measure CO2 and CH4 fluxes over reservoirs can help address the issues of spatial and temporal coverage. Here we present results from two eddy covariance measurement campaigns monitoring CO2 and CH4 fluxes over reservoirs in southwestern Ohio, US. The first campaign examined the effects of water level drawdown on reservoir CH4 ebullition. The eddy covariance results showed a clear response of CH4 emissions to the change in water level, increasing from a baseline of 3440 mg CH4 m-2 d-1 to a maximum of 6740 mg CH4 m-2 d-1 during the drawdown. These results agreed well with the emission rates measured via bubble samplers deployed in the vicinity of the tower. Conversely, the CO2 fluxes did not show a strong response to the drawdown. The eddy covariance system was deployed for a longer period of time during a second campaign at a mid-sized (2.4 km2) lake. Analyses of diurnal patterns in CO2 and CH4 emissions as well as emission response to synoptic events will be presented. Our results contribute to the ongoing effort to better interpret and scale-up CH4 and CO2 emissions from reservoirs.