Estimation of forest–atmosphere CO2 exchange by eddy covariance and profile techniques

At Hyytiälä, station for measuring forest ecosystem–atmosphere relations (SMEAR II) measurement station in Southern Finland, carbon dioxide concentration measurement system and sampling are designed to meet the demanding requirements of the profile technique over forest. The minimisation of systematic and random errors was achieved by using the same analyser for all levels, by averaging the analyser readings and by switching between lines with frequency enabling sampling from each level for 1/2 min with 6-min intervals to obtain good averaging over turbulent variation of concentration. The fluxes were estimated above a 14 m high forest from measured profiles up to a height of 67 m. At the same time, the eddy covariance (EC) measurements were performed at 23 m. Carbon dioxide fluxes, obtained from the EC system and profile measurements, were inter-compared for periods consisting of several months of observations. Generally, good correspondence between fluxes was observed. Largest discrepancy was observed during early morning transition period of turbulence, which could not be ascribed to limitations of either of the techniques. The profile method provides an alternative method to the EC technique, which is not free of possible error sources. It proved to be a suitable method for flux measurements above forest canopy and can be used for gases lacking fast measurement instrumentation as required by the EC method. For carbon dioxide flux estimation during well-mixed day-time conditions, the average concentrations have to be determined with standard error as small as 0.2–0.3 ppm, thus the average concentration measurement instrumental precision is required to be in the order of 0.1 ppm.