Comparison of gas analyzers for quantifying eddy covariance fluxes- results from an irrigated alfalfa field in Davis, CA

Abstract:

The eddy covariance technique requires simultaneous, rapid measurements of wind components and scalars (e.g., water vapor, carbon dioxide) to calculate the vertical exchange due to turbulent processes. The technique has been used extensively as a non-intrusive means to quantify land-atmosphere exchanges of mass and energy. A variety of sensor technologies and gas sampling designs have been tried. Gas concentrations are commonly measured using infrared or laser absorption spectroscopy. Open-path sensors directly sample the ambient environment but suffer when the sample volume is obstructed (e.g., rain, dust). Closed-path sensors utilize pumps to draw air into the analyzer through inlet tubes which can attenuate the signal. Enclosed-path sensors are a newer, hybrid of the open- and closed-path designs where the sensor is mounted in the environment and the sample is drawn through a short inlet tube with short residence time. Five gas analyzers were evaluated as part of this experiment: open-path LI-COR 7500A, enclosed-path LI-COR 7200, closed-path Picarro G2311-f, open-path Campbell Scientific IRGASON, and enclosed-path Campbell Scientific EC155. We compared the relative performance of the gas analyzers over an irrigated alfalfa field in Davis, CA. The field was host to a range of ancillary measurements including below-ground sensors, and a weighing lysimeter. The crop was flood irrigated and harvested monthly. To compare sensors, we evaluated the half-hour mean and variance of gas concentrations (or mole densities). Power spectra for the gas analyzers and turbulent fluxes (from a common sonic anemometer) were also calculated and analyzed. Eddy covariance corrections will be discussed as they relate to sensor design (e.g., density corrections, signal attenuation).