Use of global climate and weather reanalysis to fill meteorological time series gaps

Meteorological is an essential input for many terrestrial ecosystem models to simulate energy, water and carbon exchanges between the surface and the atmosphere. A significant improvement in the comprehension of the processes represented in the models is linked to the installation of the eddy covariance (EC) towers.In particular the EC carbon and energy data are utilized by the modelers to develop and parameterize the models and evaluate their  performances. 

Continuous, gap-free main in-situ meteorological time series are crucial for the EC fluxes processing (mainly gap-filling and partitioning) but are also used as input for simulations of many different models. Different approaches exist for filling the gaps present in the meteorological data collected at the EC sites. In the standard FLUXNET processing (ONEFlux) a downscaled approach is used (Vuichard and Papale 2015), which was originally based on the ERA-Interim dataset, and now uses ERA5.

Here we present the results of a modified method where the downscaling has been also compared across three different reanalysis datasets (ERA5 hourly data on single levels and ERA5-Land from Copernicus and a CMCC product at 2.2 km produced for the whole Italian territory). Our results show that the reanalysis product used has an impact on the performance of the overall gap-filling of data and we suggest the implementation of a new strategy in the standardized processing chains.