In my research I want to study the ensemble spread of ERA5 EDA on a synoptic timescale. For interannual comparability I want to control for the effect of the improved observation system on the ensemble spread.
In Hersbach et al., 2020 it says
”The perturbations of observations are sampled from a zero-mean Gaussian distribution with variance equal to the expected variances of the observation errors.”
Is there a way to get these expected variances?
Best,
Henry Schoeller