Technical note: Flagging Inconsistencies in flux tower data

Global collections of synthesized flux tower data such as FLUXNET have accelerated scientific progress beyond the eddy covariance community. However, there are remaining data quality issues in FLUXNET data that pose challenges for users. Here we present a complementary flux tower data quality control approach based on multiple indications of inconsistency among variables, along with a methodology to detect discontinuities in time series. The final results are for NEE, GPP, TER, LE, H, SW_IN, NETRAD, PPFD, TA, VPD and consist of: (1) flags for daily data values, (2) flags for entire site variables, (3) flags at time stamps that mark large discontinuities in the time series. The quality control is primarily based on combining deviations from a set of predefined relationships among variables. The methodology to detect break points in the time series is based on a non-parametric test for the difference of distributions. Applying the quality control approach to the FLUXNET 2015 dataset reveals (1) frequent and systematic rejections for GPP and TER under dry and hot conditions due to flux partitioning issues that limit our capability to better understanding drought and rain pulse responses, (2) comparatively frequent rejections for radiation variables that are crucial for correctly interpreting and modelling ecosystem fluxes, (3) frequently detected discontinuities in long time series of LE, NEE, and radiation variables that raise caution flags for studying interannual variations and trends.

JUNG Martin;  NELSON Jacob;  MIGLIAVACCA Mirco;  EL-MADANY Tarek S.;  PAPALE Dario;  REICHSTEIN Markus;  WALTER Sophia;  WUTZLER Thomas; 

2024-04-30

COPERNICUS GESELLSCHAFT MBH

JRC136494

1726-4170 (online),   

https://bg.copernicus.org/articles/21/1827/2024/,    https://publications.jrc.ec.europa.eu/repository/handle/JRC136494,   

10.5194/bg-21-1827-2024 (online),