Fixed a bug related to sensible/latent heat fluxes for AQM coupling (Issue#945)

NCAR / ccpp-physics

GFS physics for CCPP

Other

58 stars 145 forks source link

Fixed a bug related to sensible/latent heat fluxes for AQM coupling (Issue#945) #951

Closed ChunxiZhang-NOAA closed 2 years ago

ChunxiZhang-NOAA commented 2 years ago

A bug is fixed which is described in Issue#945. Currently, this PR only fixed the bug that caused model crashes due to the unallocated variables dtsfc_cice and dqsfc_cice which are only allocated when cplflx=.true. in GFS_typedefs.F90. There are potential other issues. For instance, when (cplaqm .and. .not.cplflx) is ture, dtsfci_cpl and dqsfci_cpl are only assigned values over ocean (e.g., oceanfrac > 0).

grantfirl commented 2 years ago

Does anyone know if @rmontuoro (the original author of this section) agrees with the solution?

ChunxiZhang-NOAA commented 2 years ago

@grantfirl Maybe we should invite Raffaele to review this code change.

rmontuoro commented 2 years ago

@ChunxiZhang-NOAA, @grantfirl - Thanks. I will take a look and get back to you.

rmontuoro commented 2 years ago

The proposed changes do not address the lack of surface net heat fluxes over land, which need to be provided to AQM. The current formulation also provides improper values on mixed grid points for fractional grids.

A new formulation is proposed in PR #953.

grantfirl commented 2 years ago

@ChunxiZhang-NOAA @rmontuoro Should we close this (and upstream PRs) in favor of merging #953 (and upstream PRs)?

ChunxiZhang-NOAA commented 2 years ago

@grantfirl We can close this PR.

grantfirl commented 2 years ago

Closing this PR in favor of continuing discussion of the solution in #953