Comparisons of different sample air-drying systems for carbon dioxide flux measurements based on eddy covariance in cold environments

In cold region such as sea ice areas, it is difficult to evaluate the carbon dioxide exchange between the surface and atmosphere (CO2 flux) by eddy covariance due to their small flux magnitudes. Drying air samples using a closed-path infrared gas analyzer is effective for measuring small CO2 fluxes. However, using drying equipment leads to the attenuation of turbulent fluctuations, which tends to result in underestimation of CO2 flux. Therefore, it is necessary to survey which dryer is best to minimize this underestimation while effectively drying air samples. In this study, we evaluated the drying ability and persistence using desiccants and a membrane dryer, and the impact of air-drying on CO2 fluxes by simultaneous observations of the drying and non-drying systems. The drying systems with desiccants showed high drying abilities, but their persistence was only a few hours. The drying system with a membrane dryer had a lower drying ability. However, it successfully eliminated the water vapor fluctuation, which was important for accurate CO2 flux measurements. The use of the membrane dryer in the drying system resulted in only 5% underestimation of CO2 fluxes due to the attenuation of CO2 mixing ratio fluctuations, further suggesting its usefulness.