feat: aether cubed sphere model_mod #992

hkershaw-brown · 2025-10-30T13:44:11Z

Description:

Please provide a summary of the change and include any relevant motivation and context.

This pull request is the aether cubed sphere development from @jlaucar. The history is squashed to a single commit to remove the various binary files, unused content, & DART_LAB merged content that was used for teaching an in person session in Michigan.
If you need to retrieve anything from the history, the branch aether_michigan is available here: https://github.com/hkershaw-brown/DART/tree/aether_michigan Note is the branch is called aether_michigan because it is was used to teach DART & aether at Michigan.

I have not made code changes from Jeff's work, this pull request contains:

aether cubed sphere model_mod
New qty: QTY_SLANT_GPS_VTEC
New fwd-op: get_expected_slant_gps_vtec. Note sections in code
"THIS SUBROUTINE NEEDS ADDITIONAL INPUT FROM AETHER SCIENTISTS"
developer test: test_aether_grid
matlab tools for aether cubed sphere: perturb, plot lat lon, plot filter. text file of demo documentation (taught at Michigan?)
public debugging routines for quad_utils

Note not looking at cube_sphere_grid_tools.f90 on advice of Jeff so this is unreviewed.

Fixes issue

fixes #665 #666

Types of changes

Bug fix (non-breaking change which fixes an issue)
New feature (non-breaking change which adds functionality)
Breaking change (fix or feature that would cause existing functionality to not work as expected)
Documentation update

Documentation changes needed?

My change requires a change to the documentation.
- I have updated the documentation accordingly.

Tests

Please describe any tests you ran to verify your changes.
None, I am just making this pull request for JLA.

Checklist for merging

Updated changelog entry
Documentation updated
Update conf.py

Checklist for release

Merge into main
Create release from the main branch with appropriate tag
Delete feature-branch

Testing Datasets

I believe Jeff has test data sets.

Dataset needed for testing available upon request
Dataset download instructions included
No dataset needed

assimilation_code/modules/observations/space_quantities_mod.f90

developer_tests/aether_grid/test_aether_grid.f90

developer_tests/aether_grid/work/input.nml

models/aether_cube_sphere/aether_to_dart.f90

models/aether_cube_sphere/cube_sphere_grid_tools.f90

models/aether_cube_sphere/model_mod.f90

models/aether_cube_sphere/perturb_aether_ensemble.m

models/aether_cube_sphere/readme.rst

hkershaw-brown · 2025-10-30T16:19:11Z

I can't "request changes" as the author so I'm putting the pull request on draft to mean "request changes"

models/aether_cube_sphere/work/input.nml

models/utilities/quad_utils_mod.f90

observations/forward_operators/obs_def_upper_atm_mod.f90

models/aether_cube_sphere/work/demo_documentation.txt

models/aether_cube_sphere/transform_state_mod.f90

models/aether_cube_sphere/work/input.nml

developer_tests/aether_grid/test_aether_grid.f90

developer_tests/aether_grid/work/quickbuild.sh

models/aether_cube_sphere/model_mod.f90

hkershaw-brown · 2025-10-31T13:09:14Z

Note not looking at cube_sphere_grid_tools.f90 on advice of Jeff so this is unreviewed.

models/aether_cube_sphere/model_mod.f90

hkershaw-brown · 2025-11-11T16:00:10Z

Looks good Jeff, just one question, and a couple of tidy-up code todos

question: f10.7 MSG vs ALLMSG to make sure the message is printed (see feat: aether cubed sphere model_mod #992 (comment))
test_aether_grid (executable) needs adding to the .gitignore file
there are some unused routines in model_mod:
'use ... only :' routines from '../model_mod.f90' NOT USED:
E_ERR
get_index_for_quantity
grid_to_lat_lon
is_point_in_quad
is_point_in_triangle
lat_lon_to_col_index
lat_lon_to_grid
lat_lon_to_xyz
register_module
to_upper

New qty: QTY_SLANT_GPS_VTEC New fwd-op: get_expected_slant_gps_vtec. Note sections in code "THIS SUBROUTINE NEEDS ADDITIONAL INPUT FROM AETHER SCIENTISTS" developer test: test_aether_grid matlab tools for aether cubed sphere: perutb, plot lat lon, plot filter. text file of demo documentation (taught at Michigan?) debugging routines for quad_utils

Updated calls to error_handler that used these.

files from the developer_tests directory. Changed the quickbuild.sh in aether_cube_sphere/work to include test_aether_grid as a sequential program. Moved the test_grid_box subroutine from model_mod to test_aether_grid. Made three additional public entities in model_mod that are needed for test_grid_box. These are the grid geometry basic quantities np and ncenter_altitudes and the routine get_state_index. Changed to using utilities instead of mpi_utilities in test_aether_grid.

compatabilty with ifort.

get the list of state variables and their clamping ranges from the namelist.

…art.

…ties_mod as possible. Tests show that this bitwise duplicates the results from the old aether_michigan branch in the final Aether ions/neutrals file space. The filter_input and filter_output files are not bitwise because they now use double variables and the order of various fields has been changed.

Since we don't know much about roundoff sensitivity for Aether at this point, this made sense as a scientific precaution. If file size became an issue, this could be changed back. There are now a handful of last bit differences in the final aether ions and neutrals files.

…nged to more direct netcdf_utilities calls for that read.

the original vtec subroutine.

to make F10.7 DA work.

…handler