Create overview of Aerosol Interface

[mattldawson]

Issue Type

Other (please describe below)

Issue Description

Create a high-level overview of the refactoring work to introduce a common interface for aerosol packages (initially MAM and CARMA)

Ideas

• Include documentation from Francis's All Staff Report on Aerosol Interface
• Include swimlane information if helpful

Will this change answers?

Yes

Will you be implementing this yourself?

Yes

[K20shores]

Bug fix PR that went into CAM is on this branch: https://github.com/NCAR/CAM-ACOM-dev/tree/develop-cloud-bug-fix

[K20shores]

#5 seems to be related to this as well

[K20shores]

From my perspective, we need to answer these questions

1. How do we refactor the aerosols in CAM so that

• Running any aerosol package MAM, CARMA, BAM, GOCART
  • Do all of these need to be targeted?

2. As we do this refactor, some/all of this aerosol functionality becomes available in MUSICA. How do we do that?
3. How do we still support new science functionality as we do this?

• Caveat, new science should happen separately from a refactor, and probably should be done by different people since they are really different efforts
• Do we actually have new science that needs to be implemented right now?

[K20shores]

What are you trying to get out of this, Simone?

• To get phase 1 aerosol refactor implemented to handle cloud chemistry
• Many things are duplicated (modal model data)
  • No reason to have model_aero, carma, because these can be configured with the interface
• CESM is workhorse science model. We want to make implementation of different aerosols easy by refactoring the interface

Did we manage to refactor the cloud chemistry code using the new aerosol interface? And what does that look like?

[K20shores]

At the end of the aerosol refactor, CAM-SIMA can run these packages.

• MAM
• CARMA
• BAM
• GOCART

CAM-SIMA will be able to use MUSICA to run these configurations, but could also have its own implementation of a different model that doesn't use MUSICA.

[K20shores]

Separately, we need to support new scientific development of new aerosol schemes that can be used in CAM (really, CAM-SIMA, and therefore CESM), like MOSAIC.

It would be nice to support both the new scientific development and the refactor at the same time.

[K20shores]

Part of this is handled by ESCOMP#1428, as well as offline work