Comparison of eddy covariance and automatic chamber‐based methods for measuring carbon flux

Abstract

AbstractAn automatic chamber (AC) is an alternative method to the eddy covariance (EC) technique for estimating of CO2 fluxes. However, few direct comparisons between two techniques have been performed on nature alpine meadows. In order to determine diurnal pattern of net ecosystem exchange (NEE) of CO2 in summer on alpine meadows, and test the NEE measurement performance of AC under site conditions, we conducted simultaneous measurements of NEE using both EC and AC methods. The NEEAC can be estimated with the information from the initial slope of the time series of CO2 mixing ratio. We found that the application of linear regression, when photosynthesis processes were considered, could lead to serious deviations, even if closure times were short. The linear regression, however, was adequate for estimating CO2 fluxes when photosynthesis processes were not involved in, at least for short chamber closing times. Using appropriate fitting method, the fluxes of AC were relatively close to that of EC, but presented a daytime underestimation and nighttime overestimation. With the subsequent friction velocity (u*) filtering (<m s–1), the nighttime overestimation was eliminated, and the fluxes of AC showed a similar diurnal pattern of NEE with that of EC, but an overall underestimation all day long. Here, we may provide a side evidence of the overestimation of chamber fluxes at low atmospheric turbulence. Finally, we assume that the overall underestimation of AC is caused by the absence of detection of effective volume of chamber.