Feature request: Add comments for units in KPP/commonIncludeVars.H

[jhaskinsPhD]

Name and Institution (Required)

Name: Jessica Haskins
Institution: University of Utah

Description of your issue or question

Hi, I'm using the KPP files in creating my GEOS-Chem Emulator to generate F0AM compliant mechanisms based on past GEOS-Chem versions. In doing so, I need to make sure the units of all the global variables used in KPP rate functions are translated properly to F0AM, but the documentation of those global variables is a bit ambiguous in the commonIncludeVars.H, gckpp_Global.f90 and gckpp_Parameters.F90 files.

Could you let me know the units of the following variables & add that to the documentation of that file?

• PRESS (assuming its hPa because that's what PMID & all StateMet pressure variables are in)
• RELHUM (is this % or fraction?)
• H2O (assuming is mole/mole?)

[yantosca]

Thanks for writing @jhaskinsPhD. These are all set in routine Set_Kpp_Gridbox_Values in the GeosCore/fullchem_mod.F90 module:

    !========================================================================
    ! Populate global variables in gckpp_Global.F90
    !========================================================================

    ! Solar quantities
    SUNCOS          = State_Met%SUNCOSmid(I,J)

    ! Pressure and density quantities
    NUMDEN          = State_Met%AIRNUMDEN(I,J,L)
    H2O             = State_Met%AVGW(I,J,L) * NUMDEN
    PRESS           = Get_Pcenter( I, J, L )

    ! Temperature quantities
    TEMP            = State_Met%T(I,J,L)
    INV_TEMP        = 1.0_dp   / TEMP
    TEMP_OVER_K300  = TEMP     / 300.0_dp
    K300_OVER_TEMP  = 300.0_dp / TEMP
    SR_TEMP         = SQRT( TEMP )
    FOUR_R_T        = 4.0_dp * CON_R    * TEMP
    FOUR_RGASLATM_T = 4.0_dp * RGASLATM * TEMP
    EIGHT_RSTARG_T  = 8.0_dp * RSTARG   * TEMP

    ! Relative humidity quantities
    CONSEXP         = 17.2693882_dp * (TEMP - 273.16_dp) / (TEMP - 35.86_dp)
    VPRESH2O        = CONSVAP * EXP( CONSEXP ) / TEMP
    RELHUM          = ( H2O / VPRESH2O ) * 100_dp

PRESS is in hPa because that is what GEOS-Chem uses. I think it is only used for pressure-dependent rate-law reactions as specified in KPP/fullchem/fullchem_RateLawFuncs.F90. My guess is that the units shouldn't matter as long as you are consistent.

RELHUM looks like a fraction percent since it is multiplied by 100.

Also from Headers/state_met_mod.F90:

     REAL(fp), POINTER :: AIRNUMDEN     (:,:,:) ! Dry air density [molec/cm3]
     REAL(fp), POINTER :: AVGW          (:,:,:) ! Water vapor volume mixing
                                                !  ratio [vol H2O/vol dry air]

Lastly, H2O looks like it is in molec/cm3 (mixing ratio * number density).

We can add an update to the common IncludeVars.H to specifically list the units of these data.

[jhaskinsPhD]

Thanks! A comment would be great so I can parse all future files to make the units are consistent in future versions!

[yantosca]

Thanks @jhaskinsPhD. I've updated the title of this issue to be a feature request so that it'll remind us to add the comments in. That can go into our "no-diff-to-benchmark" branch.

[yantosca]

@jhaskinsPhD, I am closing out this issue since PR @2061 has been merged into the GEOS-Chem "no-diff-to-benchmark" development stream. It iwll be merged into a future version shortly.