cloudinfo and target classification

[HenniHG]

I'm currently working with the cloudinfo.nc file and was wondering if the cloud information are derived via V1 or V2 of the target classification. For the example day I was looking at (29 Oct 2021), there are cloud base heights down to 500m in the plot of the target classification V2

However, the minimum cloud base height, which I get from the cloudinfo file 2021_10_29_Fri_CPV_18_00_01_cloudinfo.zip is around 800m, which looks similar to the target classification V1

So I guess, V1 was used. If yes, is it possible to derive the cloudinfo file from the V2 of the target classification? Because I'm especially interested in the boundary layer clouds in the overlap region.

[HolgerPollyNet]

@HenniHG Thanks, now I understand!

@ZPYin I guess this can easily be done, right?

[ZPYin]

Yes, you are right (@HenniHG ). Right now, the cloudinfo was derived from V1 results.
But if you are interested in V2, it can also be easily implemented. As suggested by Holger (see below), you can replace the variable tcMaskV1 by tcMaskV2 (the easiest way). Re-run the program, then you can get the results. The cloudinfo file will be generated by V2.

[HolgerPollyNet]

I guess, you have three options:
Option 1:
quick and dirty, you just exchange V1 by v2 here:
[ [clBaseH, clTopH, clPh, clPhProb] = cloudGeoExtract(data.mTime, data.height, tcMaskV1, ...]

(

Pollynet_Processing_Chain/lib/interface/picassoProcV3.m

Line 4376 in 1db40c4

[clBaseH, clTopH, clPh, clPhProb] = cloudGeoExtract(data.mTime, data.height, tcMaskV1, ...

)

Option 2:
More sustainable:
You copy these lines and create a V2 product, i.e. [clBaseHv2, clTopHv2, clPhv2, clPhProbv2]= ...

and then store it separately:

1. add to data here:

https://github.com/PollyNET/Pollynet_Processing_Chain/blob/1db40c492f706dcb795a06a36e8af32a56b7f615/lib/interface/picassoProcV3.m#L4782C1-L4785C26

2. make a new save function using these lines:

Pollynet_Processing_Chain/lib/interface/picassoProcV3.m

Line 4952 in 1db40c4

case 'cloudinfo'

Option 3: Add a new cloud screening mode here (middle way of the ones above):

https://github.com/PollyNET/Pollynet_Processing_Chain/blob/1db40c492f706dcb795a06a36e8af32a56b7f615/lib/interface/picassoProcV3.m#L4381C1-L4381C40

PollyConfig.cloudScreenMode == 3 and then do step 1 but without changing the output variable names.

I guess the last one is the best/fastest to be done by you which is sustainable in a way too.