Quality Control and Error Analysis of Eddy Covariance Measurements Over a Small Lake Surrounded by Forest

Abstract:

We present eddy covariance fluxes of carbon dioxide, sensible and latent heat measured during an open-water period. The measurements are carried out at Lake Kuivajärvi (Hyytiälä, southern Finland), a small boreal lake (0.63 km²), surrounded by forest. The measurement platform, including the EC system and other auxiliary measurements, is located approximately 1.8 km and 0.8 km from the Northern and Southern shorelines, respectively. Standard quality control criteria were applied and the 30 min fluxes were flagged accordingly (Foken and Wichura, 1996). The steady state test was more effective in removing the CO₂ flux than H and LE fluxes, the fraction of non-stationary records being 35%, 7% and 5% respectively. Similarly, the total relative random error (ΔF) for CO₂ flux was about twice that estimated for energy fluxes. Median value of random error (δF) for CO₂ was 0.14 μmol m^-2 s^-1, corresponding to 26% of the observed flux. If only 30 min periods with the best quality (flag=0) are included, median values of δF and ΔF were 0.11 μmol m^-2 s^-1 and 20% respectively, showing that, on average, more conservative flux quality criteria lead to lower flux random uncertainty. The estimated values of ΔF are close to the ones reported in other ecosystems (e.g. Finkeisten and Sims, 2001). However, different diurnal course of random errors was found for energy and CO₂ fluxes over the lake and the surrounding forest. The implications on extending standard quality criteria (including also those based on friction velocity, atmospheric stability, etc, which are routinely used for land-based flux towers) to EC flux measurements over freshwater ecosystems are further analysed and discussed.

References
Aubinet et al, 2012, Springer
Foken and Wichura, 1996, Agric. For. Meteorol., 78, 83-105
Finkeisten and Sims, 2001, J. Geophys. Res. -Atmos., 106, 3503-3509