Shared physics YSU PBL - Githubissues

wrf-model / WRF

The official repository for the Weather Research and Forecasting (WRF) model

Other

1.18k stars 658 forks source link

Shared physics YSU PBL #1991

Closed weiwangncar closed 5 months ago

weiwangncar commented 5 months ago

TYPE: enhancement, new feature

KEYWORDS: shared physics, ysu pbl

SOURCE: internal

DESCRIPTION OF CHANGES: Add YSU PBL scheme from the shared physics directory, which is ccpp-compliant.

Solution: The CCPP-compliant YSU PBL code will be residing in phys/physics_mmm/ directory. Currently it is a copy of the code from the future MMM-physics repo.

LIST OF MODIFIED FILES: M Registry/Registry.EM_COMMON M arch/postamble M arch/preamble M clean M main/depend.common M phys/Makefile M phys/module_bl_ysu.F M phys/module_pbl_driver.F M phys/module_physics_init.F A phys/physics_mmm/bl_ysu.F90

TESTS CONDUCTED:

Option runs before and after the change do not produce identical results with 'configure -d'. It does produce bit-for-bit results when running with 36, 12 and 1 processors.
The Jenkins tests have passed.

RELEASE NOTE: Add YSU PBL scheme from the future shared MMM-physics repository.

weiwangncar commented 5 months ago

The regression test results:

Test Type              | Expected  | Received |  Failed
= = = = = = = = = = = = = = = = = = = = = = = =  = = = =
Number of Tests        : 23           24
Number of Builds       : 60           57
Number of Simulations  : 158           150        0
Number of Comparisons  : 95           86        0

Failed Simulations are: 
None
Which comparisons are not bit-for-bit: 
None

smileMchen commented 5 months ago

@weiwangncar Wei, Option runs before and after the change do not produce identical results with 'configure -d', --- is this a concern? If the physics scheme itself is not changed, I expect that the results should be the same. What could be the possible reason that the results are not identical? Thanks.

weiwangncar commented 5 months ago

@weiwangncar Wei, Option runs before and after the change do not produce identical results with 'configure -d', --- is this a concern? If the physics scheme itself is not changed, I expect that the results should be the same. What could be the possible reason that the results are not identical? Thanks.

Yes, it is somewhat a concern. One could argue that the newly polished YSU routine may be more robust. I do plan to follow up on this a bit more during the testing phase.

Plantain commented 3 months ago

Does a clean build actually succeed after these changes? After this patchset we receive: module_bl_ysu.f90(4): error #7002: Error in opening the compiled module file. Check INCLUDE paths. [BL_YSU] use bl_ysu

It looks like it also has problems on Jenkins, but I can't see the build output.

weiwangncar commented 3 months ago

@Plantain Our regression tests have worked. Is your system case-sensitive? Can your computer tell .F90 from .f90 files?