Remove CAM dependencies and add CCPP layer for PUMAS v1 routines #71
Conversation
… using framework script.
and change subroutine dummy argument declarations to not specify dimension sizes.
| @@ -1,4 +1,7 @@ | |||
| ==================================== | |||
| .. pumas_cam-release_v1.38 | |||
| January 15, 2025 - Removed all CAM dependencies from core PUMAS routines except wv_sat_methods. Also added a Common Community Physics Package (CCPP) wrapper for micro_pumas_v1.F90, including two new CCPP metadata files. A new CAM tag will be made which calls the wrapper instead of PUMAS directly, although this isn't technically required. | |||
There was a problem hiding this comment.
This is kind of silly, but could you add the name of the wrapper file (micro_pumas_ccpp.F90) and possibly the new calling order for CAM-SIMA in here?
| @@ -147,8 +149,6 @@ subroutine proc_rates_allocate(this, psetcols, nlev, ncd, warm_rain, errstring) | |||
| ! Routine to allocate the elements of the proc_rates DDT | |||
| !-------------------------------------------------------------- | |||
|
|
|||
| use cam_abortutils, only: endrun | |||
|
|
|||
There was a problem hiding this comment.
The fact that this is just here and never used makes me giggle. It's a clean up that was needed for all end uses.
| #process rates DDT to allow it | ||
| #to be passed between PUMAS and | ||
| #any CCPP intersitial schemes. | ||
| ################################# |
There was a problem hiding this comment.
Maybe add a note that this doesn't work yet. Or a Todo comment so we can track the todos.
| @@ -1,6 +1,6 @@ | |||
| module tau_neural_net_quantile | |||
|
|
|||
| use shr_kind_mod, only: r8=>shr_kind_r8 | |||
There was a problem hiding this comment.
So, this type is changed, but it's not used below. There are a few literals in this file where the typing is not specified. It would be really nice if you could add "_r8" to the real literals in this file and the typing for the integers, too, possibly?
| @@ -86,8 +88,6 @@ module micro_pumas_utils | |||
| avg_diameter_vec | |||
|
|
|||
| ! 8 byte real and integer | |||
There was a problem hiding this comment.
Let's remove this comment too
| [micro_accre_enhan_in] | ||
| standard_name = microphysics_accretion_enhancement_factor | ||
| long_name = microphysics accretion enhancement factor | ||
| units = 1 |
| intent = in | ||
| [micro_pdel_in] | ||
| standard_name = microphysics_air_pressure_thickness | ||
| long_name = microphysics air pressure thickness |
There was a problem hiding this comment.
"microphysics air pressure layer thickness"?
| [micro_qsatfac_in] | ||
| standard_name = microphysics_subgrid_cloud_water_saturation_scaling_factor | ||
| long_name = microphysics subgrid cloud water saturation scaling factor | ||
| units = 1 |
There was a problem hiding this comment.
Another factor - units none?
| intent = out | ||
| [micro_prect_out] | ||
| standard_name = microphysics_lwe_large_scale_precipitation_rate_at_surface | ||
| long_name = microphysics LWE large scale precipitation rate at surface |
There was a problem hiding this comment.
Ok, is LWE liquid water equivalent? I can see that being used here because this total precip rate includes rain and snow, but I feel like total should be in the long name and LWE is confusing. We actually don't use the LWE acronym in PUMAS at all.
| [micro_lamcrad_out] | ||
| standard_name = microphysics_cloud_particle_size_distribution_slope_parameter | ||
| long_name = microphysics cloud particle size distribution slope (lambda) parameter | ||
| units = 1 |
There was a problem hiding this comment.
I'm ok with these being "1" but I've written so many comments now that it feels disingenuous.
This PR removes all explicit uses of CAM and/or CESM share modules, and instead adds internal PUMAS routines to accomplish the same goals. The one exception is for calls to
wv_sat_methodswhich will need to remain a dependency for now (but can likely be CCPP-ized as well later). Part of this effort included removing the no-longer-usedmicro_pumas_data.F90file.This PR also adds a CCPP Fortran module and metadata file to enable CCPP-compliant calls to
micro_pumas_initandmicro_pumas_tend, as well as a stub metadata file for theproc_rates_typeDDT, which will need to be visible to the CCPP-framework in order for it to be passed to external CCPP schemes (e.g. diagnostics).Testing:
Successfully passed all aux_cam tests on Derecho with the Intel compiler using
cam6_4_055baselines. Also passed all aux_pumas tests except for the one test failure described in #70