search

geoschem / Cloud-J

Cloud-J is multi-scattering eight-stream radiative transfer model for solar radiation based on Fast-J. It was originally developed by Michael J. Prather.
GNU General Public License v3.0
3 stars 7 forks source link

Updates for GEOS-Chem #2

Closed lizziel closed 6 months ago

lizziel commented 1 year ago

Name and Institution (Required)

Name: Lizzie Lundgren Institution: Harvard University

Describe the update

This PR brings into the main branch all remaining changes made during development of Cloud-J for use in GEOS-Chem. This includes changes for using GEOS-Chem with Cloud-J photolysis in multiple configurations: GEOS-Chem Classic, GCHP, GEOS, and CESM. It also includes several bug fixes for compatibility with GNU and Intel compilers, and one Cloud-J standalone bug fix for computing ice water path input to Cloud_JX. This PR also includes a change in expected spacing of entries in input file FJX_scat-aer.dat which will affect Cloud-J standalone users who have custom LUTs outside of the ones stored in this repository.

Expected changes

See the Changelog (new file added in this PR) for details of all changes in Cloud-J.

Reference(s)

None

Tagging @pratherUCI for review

lizziel commented 7 months ago

Merging this in is on the back-burner because it will break compatibility with the Cloud-J standalone tables files.

lizziel commented 6 months ago

This PR is almost ready for review. I pushed the latest updates for GEOS-Chem in offline, GEOS, and CESM and now need to address the resulting conflicts in the standalone model.

lizziel commented 6 months ago

This PR is now ready for review. I confirmed that the changes cause no differences in Cloud-J standalone output when using default settings.

lizziel commented 6 months ago

Thanks @yantosca for your detailed review! I tried to keep the Cloud-J as close to the original as possible to avoid divergence. Your comments will be very helpful when we go through and do general improvements to the code.

@pratherUCI, once we start making general code improvements it will be more difficult to bring in updates you have based on older versions. Would you be able to start working off of this version (once merged) moving forward to avoid problems down the road?

pratherUCI commented 6 months ago

I went thru all the changes. They make sense and indeed help. I approve. Michael

There is no reason why the standalone version that I continue to develop could not now begin with the new version.

There are two remaining issues for us: 1) The next version of Cloud-J standalone is v8.0c, it has been changed (minorly) to include H2O absorption. At the time I also went through all the old commons and removed dead variables (as you did). I will check my list. So how is best to proceed. BTW, this code enabled me to include NO2 absorption (small effects vs H2O). The major change is in one input table, a few lines of code, and then the clean up. This is not urgent, the only change is the H2O abs. Prather, Michael (2023). An updated cloud-overlap photolysis module for atmospheric chemistry models, UCI Cloud-J v8.0, with near-UV H2O absorption [Dataset]. Dryad. https://doi.org/10.7280/D1Q398

2) I have intentionally ignored development for the aerosol code until there was a reason to update. GEOS Chem should have updated aerosol tables that address rel hum, etc. We need someone who is interested in the aerosol simulations and I can work with them.

Michael

lizziel commented 6 months ago

Hi @pratherUCI, thanks for reviewing the code. I will create a new GitHub issue to discuss putting in the next version of Cloud-J.