Lightning Threat Indexes: pass w to model

[SamuelTrahanNOAA]

Description

This is part of a collection of PRs to add lightning threat indexes to the RRFS. The algorithm was originally ported from HRRR to the GFDL_atmos_cubed_sphere NGGPS diagnostics, using the w variable in both implementations. I've moved it to CCPP, but it still needs w in the formulation, so this three-line change sends w to the model.

ufs-community/ufs-weather-model#1632

How Has This Been Tested?

A 3km test case from a time that had some lightning was run and compared to the HRRR output. Developers of the scheme found the results to be sufficiently similar. There'll be more extensive testing after this is merged and running in parallels.

Checklist:

Please check all whether they apply or not

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation -- I don't think there are any documentation changes required.
• My changes generate no new warnings
• Any dependent changes have been merged and published in downstream modules -- no downstream changes

[junwang-noaa]

The code change is to bring w field from dycore to physics. It looks good to me.

[bensonr]

I believe this should be protected with an if-test for hydrostatic as w is only defined for non-hydrostatic simulations.

[SamuelTrahanNOAA]

Thank you for noticing that. Also, @junwang-noaa requested that wgrs only be allocated if lightning_threat is enabled, so I'll put both in the same "if" guard.

[lharris4]

When implemented in the physics this check will also need to be made.

…

On Tue, Feb 28, 2023 at 11:28 AM Rusty Benson ***@***.***> wrote: ***@***.**** commented on this pull request. ------------------------------ In driver/fvGFS/atmosphere.F90 <#243 (comment)> : > @@ -2113,6 +2113,7 @@ subroutine atmos_phys_driver_statein (IPD_Data, Atm_block,flip_vc) IPD_Data(nb)%Statein%tgrs(ix,k) = _DBL_(_RL_(Atm(mygrid)%pt(i,j,k1))) IPD_Data(nb)%Statein%ugrs(ix,k) = _DBL_(_RL_(Atm(mygrid)%ua(i,j,k1))) IPD_Data(nb)%Statein%vgrs(ix,k) = _DBL_(_RL_(Atm(mygrid)%va(i,j,k1))) + IPD_Data(nb)%Statein%wgrs(ix,k) = _DBL_(_RL_(Atm(mygrid)%w(i,j,k1))) I believe this should be protected with an if-test for hydrostatic as w is only defined for non-hydrostatic simulations. — Reply to this email directly, view it on GitHub <#243 (review)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AMUQRVHAESF7WKP2MBXBRQTWZYRT5ANCNFSM6AAAAAAVJ62DUQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.*** com>

[SamuelTrahanNOAA]

When implemented in the physics this check will also need to be made.

The code that uses wgrs is never reached unless lightning_threat .eqv. .true.

[bensonr]

Thanks @SamuelTrahanNOAA. I haven't looked at the IPD_Data%Statein ddt in a while, but if it's a pointer and not a pure allocatable, then this all looks good.

[SamuelTrahanNOAA]

I haven't looked at the IPD_Data%Statein ddt in a while, but if it's a pointer and not a pure allocatable, then this all looks good

From GFS_typedefs.F90:

  type GFS_statein_type
    ...
    real (kind=kind_phys), pointer :: wgrs (:,:)   => null()  !< w component of layer wind
    ...
  end type GFS_statein_type

...
  subroutine statein_create (Statein, IM, Model)
    ...
    if(Model%lightning_threat) then
      allocate (Statein%wgrs   (IM,Model%levs))
    endif

[jkbk2004]

@bensonr @laurenchilutti Can you merge in this pr? All tests are done on ufs-community/ufs-weather-model#1642

[jkbk2004]

thanks! @bensonr