Surface eddy flux, cospectral similarity, and energy budget closure

This study discusses the importance of low frequencies in the determination of turbulent surface layer fluxes from towers. It is based on the analysis of the eddy covariance technique to estimate the momentum, heat, moisture and carbon dioxide fluxes using data collected over a deciduous and a boreal forest during midday periods. There is an average contribution of about 17% due to long period signals, ranging from 4 to 30 minutes in convective conditions. One possible cause of this contribution is the presence of large eddies in the convective boundary layer (CBL) passing the micrometeorological tower. Spectral analysis indicates that not only the horizontal but also the vertical wind spectra are being influenced by the CBL eddies. For the data presented a single cospectral similarity function describing the turbulent fluxes has been found. Several running mean periods have been used to assess the effect of the Reynolds' mean removal procedure on flux estimates. Traditional methods of data conditioning are tested and compared with the fluxes obtained from the running mean filter results. A procedure is introduced to correct fluxes obtained with short averaging intervals. The correction is successful in achieving observational energy balance closure at two dissimilar forested sites