[NDSL] GFDL Single moment microphysics port to NDSL by CharlesKrop · Pull Request #1041 · GEOS-ESM/GEOSgcm_GridComp

CharlesKrop · 2024-12-30T19:07:02Z

Summary

GFDL single moment microphysics port to NDSL of the driver, corresponding to gfdl_cloud_microphys_driver in the source Fortran.

Port all code for FP configuration
Partial validation (see below) on one-step numerical regression with dace:cpu backend
Integration into GEOS

Partial Validation

GFDL_1M_driver created. Primary components: fall_speed, terminal_fall, warm_rain, icloud. All components are considered correct on their own, with minor errors (order 10^-9/10^-10 or <100 ULP in most cases) in warm_rain and icloud.

Putting these components together results in divergence in nearly all variables. Most are minor, but a few (e.g PRCP_RAIN ) are significant. These errors are most likely the accumulation of the small errors from the aforementioned driver components as the driver runs its iterations (in the test case, these functions were called 12 times).

Configuration ported

During construction numerous unexecuted code paths were discovered, controlled by boolean keywords. This keyword control pattern has been replicated, but unexecuted code was not carried over. Instead, if one of these paths would be triggered, the keywords will prevent execution.

…actory

…iver

Written but unverified: warm_rain, i_cloud, iqs1, iqs2, wqs1 Written, verified, but incorrect: all tables. Errors are small (< 100 ULP) for tables. des tables (which are just the gradients of the tables themselves) have larger errors, but none of these errors cause significant problems in later calculations, so I am putting them aside for now.

…if, before major else). warm rain does not verify. only fails because of implicit rain "function" call. this "function" runs perfectly within the terminal_fall port (called three times with different data of similar order of magnitude), but fails within warm_rain

…ic, DQVDTmic, DQLDTmic, DQIDTmic, PRCP_ICE, PFI_LS DQIDTmic and DUDTmic errors are small in magnitude not tremendous concern. Others are a problem

…heck_flags error handling capabilities. Driver passes when testing only the first loop (with the exception of DQADTmic), errors grow with each subsequent loop, compounding to some pretty serious errors after the final iteration. Cause unknown

…iver

FlorianDeconinck

First look. More docstrings on all functions and rename of the generic "loop" functions + inline coments

.DS_Store

...physics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/GFDL_1M_driver/GFDL_1M_driver.py

...omp/GEOSphysics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/GFDL_1M_driver/icloud.py

...Sphysics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/GFDL_1M_driver/terminal_fall.py

GEOSagcm_GridComp/GEOSphysics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/saturation/qsat.py

GEOSagcm_GridComp/GEOSphysics_GridComp/GEOSmoist_GridComp/pyMoist/tests/scripts/run_tests.sh

…er port Fix all string `cffi.FFI.CDATA` with __annotations__ since py 3.11

…ridComp into dsl/GFDL_1M_driver

…o a class (namelist_constants) moved length constant to driver_constants

…dl_1m_driver

…he rest of the driver

…lose but something is wrong with the extra data loader

…Need to figure out to test outputs that are not also inputs

...cm_GridComp/GEOSphysics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/driver/config.py

...GridComp/GEOSphysics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/driver/constants.py

...GEOSphysics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/driver/terminal_fall/main.py

...physics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/driver/terminal_fall/stencils.py

...sics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/driver/terminal_fall/temporaries.py

...GEOSphysics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/driver/warm_rain/stencils.py

…ist/pyMoist/GFDL_1M/driver/config.py renamed config class Co-authored-by: Tobias Wicky-Pfund <tobias.wicky@meteoswiss.ch>

…ist/pyMoist/GFDL_1M/driver/constants.py Co-authored-by: Tobias Wicky-Pfund <tobias.wicky@meteoswiss.ch>

…ridComp into dsl/GFDL_1M_driver

FlorianDeconinck

Almost there:

Configuration is textbook dataclass, let's get it there
Interface need to be returned to the non-OSX hack (for now)

...cm_GridComp/GEOSphysics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/driver/config.py

...Comp/GEOSphysics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/interface/cffi_lib/interface.py

Update `.gitignore` to next standard

Tobias greenlit in stand up

CharlesKrop added 11 commits October 11, 2024 14:31

Merge remote-tracking branch 'origin/dsl/develop' into dsl/extradim_f…

0a1ec4a

…actory

Merge remote-tracking branch 'origin/dsl/develop' into dsl/extradim_f…

9490578

…actory

Merge remote-tracking branch 'origin/dsl/develop' into dsl/GFDL_1M_dr…

c78e946

…iver

Assembled, functioning GFDL_1M_driver. Errors remain: DQADTmic, DUDTm…

99ba9a5

…ic, DQVDTmic, DQLDTmic, DQIDTmic, PRCP_ICE, PFI_LS DQIDTmic and DUDTmic errors are small in magnitude not tremendous concern. Others are a problem

Updates to the driver

63b9d46

Clean up

434b244

Merge remote-tracking branch 'origin/dsl/develop' into dsl/GFDL_1M_dr…

425dc6c

…iver

pre-commit run and corrected errors

15dc561

CharlesKrop requested review from FlorianDeconinck and twicki December 30, 2024 19:07

CharlesKrop self-assigned this Dec 30, 2024

CharlesKrop requested review from a team as code owners December 30, 2024 19:07

FlorianDeconinck changed the title ~~GFDL_1M_driver~~ [NDSL] GFDL_1M_driver Dec 30, 2024

FlorianDeconinck added the 0 diff trivial The changes in this pull request are trivially zero-diff (documentation, build failure, &c.) label Dec 30, 2024

FlorianDeconinck changed the title ~~[NDSL] GFDL_1M_driver~~ [NDSL] GFDL Single moment driver port to NDSL Dec 30, 2024

FlorianDeconinck requested changes Dec 30, 2024

View reviewed changes

FlorianDeconinck changed the title ~~[NDSL] GFDL Single moment driver port to NDSL~~ [NDSL] GFDL Single moment microphysics port to NDSL Dec 30, 2024

FlorianDeconinck and others added 5 commits December 30, 2024 17:28

Add pymoist_GFDL_1M_driver hook and interface for microphysics driv…

5875e82

…er port Fix all string `cffi.FFI.CDATA` with __annotations__ since py 3.11

docstrings

2151fbb

Merge branch 'dsl/GFDL_1M_driver' of github.com:CharlesKrop/GEOSgcm_G…

94a9f46

…ridComp into dsl/GFDL_1M_driver

refactored GFDL_1M_driver __init__ to move calculation of constants t…

7685b6b

…o a class (namelist_constants) moved length constant to driver_constants

capitalized constants

a83b6d3

CharlesKrop requested a review from FlorianDeconinck December 31, 2024 22:41

update to python bridge to work with GFDL driver

05bdd2b

CharlesKrop requested a review from a team as a code owner January 8, 2025 02:03

Refactored bridge, created independent call & function cascade for gf…

721e0c0

…dl_1m_driver

GEOS-ESM deleted a comment from FlorianDeconinck Feb 21, 2025

CharlesKrop added 3 commits February 24, 2025 18:57

Major reorganization of warm_rain. Minor reorganization/renaming in t…

2dc822e

…he rest of the driver

new translate test design, implemented on warm_rain, first attempt. c…

1595619

…lose but something is wrong with the extra data loader

New translate tests. All are funcitonal EXCEPT the main driver test. …

69f7768

…Need to figure out to test outputs that are not also inputs

twicki previously requested changes Feb 26, 2025

View reviewed changes

CharlesKrop and others added 9 commits February 26, 2025 21:29

Update GEOSagcm_GridComp/GEOSphysics_GridComp/GEOSmoist_GridComp/pyMo…

18a7098

…ist/pyMoist/GFDL_1M/driver/config.py renamed config class Co-authored-by: Tobias Wicky-Pfund <tobias.wicky@meteoswiss.ch>

changed i32 to int

136d102

Update GEOSagcm_GridComp/GEOSphysics_GridComp/GEOSmoist_GridComp/pyMo…

2fe3bcd

…ist/pyMoist/GFDL_1M/driver/constants.py Co-authored-by: Tobias Wicky-Pfund <tobias.wicky@meteoswiss.ch>

added docstrings for constants

9a0deb4

bunch of updates per Tobias' review

38318db

more changes from tobias' review

cdb315f

Merge branch 'dsl/GFDL_1M_driver' of github.com:CharlesKrop/GEOSgcm_G…

922b41c

…ridComp into dsl/GFDL_1M_driver

changed organization of constant files

6eafc57

stencilified the saturation table calculations

9daa8d6

CharlesKrop requested review from FlorianDeconinck and twicki March 4, 2025 23:38

CharlesKrop added 4 commits March 4, 2025 18:38

last round of changes per Tobias' first round of comments

207ccdf

pre-commit checks

0cec3e1

classes to dataclasses where appropriate

3b69483

linting

e88411e

FlorianDeconinck requested changes Mar 7, 2025

View reviewed changes

CharlesKrop added 2 commits March 7, 2025 14:33

autodetection of library_dirs for the pymoist interface

b4a4e94

pre-commit

3d2b8a6

CharlesKrop requested a review from FlorianDeconinck March 7, 2025 19:35

CharlesKrop and others added 3 commits March 7, 2025 14:47

MicrophysicsConfiguration class --> dataclass

86412cd

Remove unneeded make

971a307

Update `.gitignore` to next standard

Lint

1ab3622

FlorianDeconinck approved these changes Mar 7, 2025

View reviewed changes

FlorianDeconinck merged commit a049807 into GEOS-ESM:dsl/develop Mar 7, 2025
3 of 6 checks passed

Conversation

CharlesKrop commented Dec 30, 2024 • edited by FlorianDeconinck Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Partial Validation

Configuration ported

Uh oh!

FlorianDeconinck left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

FlorianDeconinck left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

CharlesKrop commented Dec 30, 2024 •

edited by FlorianDeconinck

Loading