Make AeroTab it's own repository

[Ovewh]

AeroTab, which is the code that generates the optical tables used in Oslo_Aero, currently lives in the cam/tools directory (not in the noresm_develop branch).
Because lacking of documentation and being hard to use, the tables we are using in Oslo_Aero have not really been look at for the past six years.

The code needs to be refactored to become more configurable and better documented, a end goal could be to make a python wrapper and create something like a pyAeroTab. In my opinion this could make AeroTab into a great tool exploring impact of e.g. aerosol processing (aging and hygroscopic growth) on size distribution and a useful tool for e.g. teaching purposes.

TL;DR Initiate a process of refactoring AeroTab and make into its own repository.

[gold2718]

I second this and can also do the set up if desired.
In the end, it folks want to be available with CAM, it can be "brought back" as an external.

Note, making AeroTab its own repository lowers the cost of development since it is much easier to up date the code (e.g., process a pull request). It is easier to review a smaller code base. It would also make it easier to add some tests and checks to make sure everything is always working properly.

[Ovewh]

Yes it would be great if you can set up the workflow.

[gold2718]

Just to check, do you agree with these statements?

• The new repo will be called AeroTab
• The original source for AeroTab will come from cam_cesm2_1_rel_05-Nor_v1.0.5 (NorESM2 release version)
• The AeroTab source will be removed from the NorESM2.1 release branch
• AeroTab repo will start with a single branch, main, containing the original source

Let me know if I have any of these wrong.

[MichaelSchulzMETNO]

@oyvindseland - would you agree also?

[MichaelSchulzMETNO]

That sounds good to have it as a stand alone, but within-NorESMhub github directory.

[oyvindseland]

yes fre. 20. okt. 2023 kl. 16:39 skrev Michael Schulz ***@***.***

…

: @oyvindseland <https://github.com/oyvindseland> - would you agree also? — Reply to this email directly, view it on GitHub <#104 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADZCKBTD7LRZPCFUVHDW6YDYAKELJAVCNFSM6AAAAAA6I3LTC2VHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM3TGMZZHEZTQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[oyvindseland]

Sorry I missed this question. To elaborate. Yes it is good to have it in separate repository in particular since it is entirely independent of CAM / NorESM. There are some sharing of code with Oslo-Aero but embedded in different subroutines.

[Ovewh]

Do you mean code duplication with Oslo_Aero? Or actually sharing subroutines such that it would not compile if it does not know about Oslo_Aero?

[gold2718]

It looks like the dependent code has all been copied into the tools/AeroTab folder.
I think this is mainly commondefinitions.F90 (which has been updated and documented in Oslo_Aero.
I also see a copy of shr_kind_mod.F90.

The code builds fine just using the files in tools/AeroTab.

I think I can go ahead and do the separation as part of the NorESM2.1 release.

[oyvindseland]

No AeroTab does not depend on subroutines in Oslo-Aero so it is code duplication and not code sharing. As long as AeroTab is used for making tables in NorESM one should remember to update the code both places. One example is assumed accommodation coefficient for sulfate and organic vapor deposition on to mineral dust particles.

[gold2718]

This repo now exists, it will probably be included as an external beginning with NorESM2.3 (which will use CAM externals, unlike NorESM2.1).