Automated eddy covariance data quality control for long-term measurements

Estimation of matter and energy exchange using the eddy covariance method is often organized into regional or global networks. To achieve comparability among sites, it is important to standardize and specify the methodology used. Currently, quality control (QC) is one of the most time-demanding steps in data processing within the Czech Carbon Observation System. Although manual QC (MQC) enables consideration of more complex test applications, it is often difficult to document. The aim of this study was to establish an automated QC (AQC) scheme based on available literature and post-processing software and test its effectivity and reliability on sites comprising an agroecosystem and a mature European beech forest. AQC successfully flagged low-quality CO2 fluxes and provided estimates of net ecosystem productivity similar to estimates based on MQC. The tests' efficiency was particularly high for the agroecosystem, where AQC removed 13% less data than did MQC. We conclude that the adopted AQC displays satisfactory performance, especially for sites with low canopy heights.