Application of the eddy-covariance method under the canopy at a boreal forest site in central Sweden

Sammanfattning: The eddy-covariance (EC) method was applied at 1.5 m height, for the period 2007-2010 at Norunda, central Sweden, for calculating energy (sensible and latent heat) and carbon dioxide (CO2) fluxes. The aim of this study was to assess the applicability of EC method above the forest floor. In 2007 an open-path EC system was used and in 2008, 2009 and 2010 a closed-path system was used. The energy and CO2 fluxes showed a clear annual pattern, with maximum values occurring in the summer season or at the beginning of September. The values of the sensible heat (H) flux were low, below 10 Wm-2, while the latent heat (LE) flux had annual daily average maximum values of 48 Wm-2 (2007), 18 Wm-2 (2008) and 27 Wm-2 (2009 and 2010). The mean annual CO2 flux values were 2.5 µmol m-2s-1 (2007 and 2008), 3 µmol m-2s-1 in 2009 and 2.8 µmol m-2s-1 in 2010. The fluxes measured in 2009 and 2010 were on average higher than the ones measured in 2008, a possible consequence of the forest thinning operation, which allowed more net radiation to reach the ground and caused an increase in turbulence above the forest floor. LE and CO2 fluxes measured with the open-path system had the biggest values and the largest variability under the entire study periods. The summer daily patterns of the energy fluxes showed higher LE than H throughout the entire 24 hours cycle. Daily summertime H flux was positive during 2009 and 2010, however a decrease was observed during afternoon, when the flux became negative. CO2 flux was very variable in time in 2007 and during nights, a consequence of both the use of an open-path system and the data selection turbulence criterion, linked to the standard deviation of the vertical wind velocity. A mid-day minimum was observed for the summer CO2 flux in 2009 and 2010, caused by photosynthesis inside the footprint area of the EC system. A comparison of EC and soil chamber data showed that CO2 fluxes measured by both methods followed in general the same patterns; however there was a discrepancy between the recorded values. On average, the EC fluxes accounted for 39 to 52% of the soil chamber fluxes. These results are consistent to other studies. In order to have a complete understanding of the applicability of the EC method above the forest floor at Norunda, further studies are necessary in order to analyze the energy balance closure, the spectral corrections for the raw EC data and the treatment of CO2 flux under very low turbulent conditions.