[NDSL] GFDL Single moment microphysics port to NDSL #1041
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
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
FlorianDeconinck
added
the
0 diff trivial
FlorianDeconinck
changed the title
FlorianDeconinck
requested changes
Dec 30, 2024
.DS_Store
Outdated
There was a problem hiding this comment.
I don't see these files in my vs code to remove them
...physics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/GFDL_1M_driver/GFDL_1M_driver.py
Show resolved
Hide resolved
| self.melting_mask_2 = quantity_factory.zeros( | ||
| [X_DIM, Y_DIM, Z_DIM], "n/a", dtype=bool | ||
| ) | ||
| self.current_k_level = quantity_factory.zeros( |
There was a problem hiding this comment.
more complicated than that. this is the double k loop we talked about briefly. its not executed in the test case, but I fear we will run into a case where this is turned on pretty quickly, so I'm leaving the code there. it works when I force it to execute, but I obviously can't validate it.
THIS_K sort of solves the problem, but not entirely. Katrina and I spent a while brainstorming this and this was the best (albeit still not good) solution we could come up with that actually runs. if/when we need to tackle this, we can talk and figure out something better
...physics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/GFDL_1M_driver/GFDL_1M_driver.py
Show resolved
Hide resolved
...physics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/GFDL_1M_driver/GFDL_1M_driver.py
Show resolved
Hide resolved
...omp/GEOSphysics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/GFDL_1M_driver/icloud.py
Outdated
Show resolved
Hide resolved
...Sphysics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/GFDL_1M_driver/terminal_fall.py
Show resolved
Hide resolved
...Sphysics_GridComp/GEOSmoist_GridComp/pyMoist/pyMoist/GFDL_1M/GFDL_1M_driver/terminal_fall.py
Show resolved
Hide resolved
| @@ -47,6 +47,7 @@ def QSat_Float_Liquid( | |||
| PL: Float = -999.0, | |||
| DQ_trigger: bool = False, | |||
| ): | |||
| DQ = 0.0 | |||
There was a problem hiding this comment.
Is this a qsat fix? If so it needs to be PR'ed separately since it's shared code
| export PACE_TEST_N_THRESHOLD_SAMPLES=0 | ||
| export PACE_DACE_DEBUG=True | ||
| export GT4PY_COMPILE_OPT_LEVEL=0 | ||
| python -m pytest -s \ |
There was a problem hiding this comment.
Revert from line 11 downwards
FlorianDeconinck
changed the title
…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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
GFDL single moment microphysics port to NDSL of the driver, corresponding to
gfdl_cloud_microphys_driverin the source Fortran.dace:cpubackendPartial 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) inwarm_rainandicloud.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.