Innovative gap-filling strategy for annual sums of CO2 net ecosystem exchange

The determination of carbon dioxide net ecosystem exchange (NEE) using the eddy-covariance (EC) method has become a fundamental tool for the investigation of the carbon balance of terrestrial ecosystems. This study presents a strategy for the processing, subsequent quality control and gap-filling of carbon dioxide eddy-covariance flux measurements for the derivation of annual sums of NEE. A set of criteria is used for quality assessment and to identify periods with instrumental or methodological failures. The complete evaluation scheme was applied to data recorded above a spruce forest at the FLUXNET-Station Waldstein-Weidenbrunnen (DE-Wei) in 2003. Comparison of this new evaluation scheme to the use of a friction velocity (u∗) threshold criterion of 0.3 m s−1indicates less systematic distribution of data gaps. The number of available high quality night-time measurements increased. This effect was most pronounced during summer, when data is essential for a robust parameterisation of respiratory fluxes. Non-linear regression analysis showed that air temperature and global radiation explain most of the variability of NEE and further seasonal segregation of the data based on an objective method did not significantly improve predictions at this evergreen forest site.