Proposal for depol calibration: use and store eta instead of V*

[HolgerPollyNet]

In the frame of Picasso 3.0 I would propose to not calculate and store any longer the V* but simply the eta.

The one need to calculate V* during the depol calculation, but with this approach the calibration is independent of the used transmission ratios. Furthermore, the implementation of the GHK procedure would be straight forward without the need for a reprocessing of calibration measurements.

eta=SQRT([+45_s/+45_g] x [-45_s/-45_g]) while v*=((1+Dg)/(1+Ds)) x eta.

In this way, the long term stability of the calibration could be also easier monitored,

@Moritz-TROPOS formulas correct?

[Moritz-TROPOS]

Yes, the formulas are correct. The GHK formalism needs just the calibration constant eta and no longer V*.

[ZPYin]

Eta from polarization calibration has been implemented. Polarization calibration constant (V*) was still kept in Picasso v3.0. But it has been replaced with eta in all respective plots and netCDF files, i.e., long-term calibration plot.

Due to the discard of V, `polly_default.jsonneeds to be changed as well. New keywords ofpolCaliEta{wavelength}andpolCaliEtaStd{wavelength}` were added to configure default eta-s. See below:

https://github.com/PollyNET/Pollynet_Processing_Chain/blob/e44ee7177ae42cc04c0e17d7329d8b3a5bd8f6d7/lib/config/polly_global_defaults.json#L2-L9

Old keywords of depolCaliConst{wavelength} and depolCaliConstStd{wavelength} was still kept, but with no actual effects on Picasso.

After pull this commit to your local machine or to the server, all polly default files should be modified to add the new keywords. But it will still keep running without any modification, as the default values of polCaliEta and polCaliEtaStd can be inherited from polly_global_default.json. polly_global_default.json is a new feature for Picasso v3.0, which is silimar to polly_global_config.json file. It serves as a template for all polly defaults.

[HolgerPollyNet]

@Moritz-TROPOS you will like it.