Height for Raman Channel LC

[martin-rdz]

@cristoferjimenez just mentioned, that it might be more stable to take the lidar constant for the Raman channels from the same height as there elastic pendants and not search for the best height intervall only in that channel.

Not sure how straigtforward it would be to include that option or if we should make it a default. Opinions @HolgerPollyNet ?

[cristoferjimenez]

I just realized that it might be better the other way around, first to get the LC from the Raman channels, as they are noisier, and take the height range as output to enter it later on when the function mean_stable is applied for the elastic channel.

If it helps, in the MATLAB version it would be (as example 532 and 607):

[LC_raman_607, range_cal, lcStd] = mean_stable(LC_raman_532, PollyConfig.LCMeanWindow, PollyConfig.LCMeanMinIndx, PollyConfig.LCMeanMaxIndx);
and then:
[LC_raman_532, ~, lcStd] = mean_stable(LC_raman_607, 10000, range_cal(1), range_cal(end));

The 10000 is just a value larger than the height range given, so that it just averages the whole range.

[HolgerPollyNet]

Whatever approach we do, we need to make sure that we have enough SNR in the Raman channels. Thus, I am a bit sceptical to use the height of the elastic channel. This might - especially at daytime, cause large errors.
In general, I think we don't need the lidar constant of the Raman channels but the ratio of the elastic to Raman. Is it wise to calculate this calibration (of the ratio) or better calibrate the channels itself?

[cristoferjimenez]

In general the ratio of lidar constants profile looks more like a constant compared to them individually, so I would also be en favor of calculating it directly and apply the function mean_stable to this ratio.