Eddy covariance methodologies revisited for urban measurements

The eddy covariance method (EC) is the state of the art measurement technique for determining surface-atmosphere interactions in the atmospheric boundary layer through the determination of turbulent fluxes of mass, momentum and heat. The fluxes of sensible and latent heat form a part of the urban energy balance together with net radiation, the anthropogenic heat flux and the storage heat flux. The latter two are hard to define and thus are often considered as the residual of the energy balance. Consequently, small errors in turbulent heat flux calculations propagate to the residual. This is a challenge as most interest is directed to the determination of the anthropogenic and heat storage fluxes. In spite of its popularity among researches, the EC calculation procedures have not yet reached consistency within the EC community. Several generally accepted but arbitrary choices are to be made when fluxes are calculated. Our objective is to quantify the magnitude of the effect of such choices using long term EC data from the urban SMEAR III station (Station for Measuring Ecosystem-Atmosphere Relationships) in Helsinki, Finland. The measurements include over four years of data measured both with open- and closed-path infrared analyzers making instrumentation intercomparison possible.

Several different calculation procedure steps were observed to have a significant effect on the final turbulent flux values. The choice of spectral correction of latent heat flux measurements using a closed-path analyzer can lead up to a 20% underestimation of the flux. For sensible heat and latent heat with an open-path analyzer, the underestimation stays below 5%. Furthermore, shortening the averaging period from 60 minutes to 5 minutes causes also an up to 20% loss of flux. Moreover, a strong dependency of the closed-path water vapor lag time on relative humidity was observed. Allowing longer lag times than traditional results in higher latent heat fluxes and better data quality as 10%-units more data pass the quality checks. Finally, iteration in flux corrections affected the fluxes by 5% when the atmosphere is stably stratified, but the effect was negligible for unstable stratification. To summarize, numerous calculation procedure choices have an impact on the sensible and latent heat fluxes and the effect on the energy balance residual can be several tens of percents.