Direct measurements of CO2 flux below a spruce forest canopy

We applied the eddy covariance technique below the canopy of a mature Norway spruce (Picea abies) forest in SE Germany, with a dense ground cover of the grasses Calamagrostis villosa and Deschampsia flexuosa. Measurements were carried out during three field campaigns between May and November 1999. Data quality analysis led to the exclusion of around 61% of flux measurements. Chamber measurements of soil CO2 efflux at the same site showed that the micro-meteorological measurements systematically under-estimate the night-time CO2 efflux on average by about 41%. By correcting the eddy covariance flux on the basis of chamber measurements, flux measurements over a larger area could be obtained, and so it was possible to calculate the daytime gross assimilation flux for the ground vegetation averaged for each of four 4-week periods. Separation of fluxes according to grass species (by wind sectors) showed no detectable difference in CO2 flux, but scarcity of data following quality tests meant that comparisons were limited to selected periods only. Light response curves of the corrected flux values indicate different saturation values for the gross assimilation rates in each period, which reflect the phenological development of the grasses, as well as differences in environmental conditions. On the basis of mean day-time assimilation by ground vegetation, the contribution to the total gross assimilation of the stand was estimated as between 3 and 12% at different times of the year. About 30% of the carbon emitted annually as CO2 from the soil is estimated to be assimilated by the ground vegetation (ca. 170 g C m(-2) year(-1)). (C) 2004 Elsevier B.V. All rights reserved.