Set gas-species profiles in TUV-x prior to calculating rate constants #174
Conversation
| air_pressure_thickness, rate_parameters, & | ||
| errmsg, errcode) | ||
|
|
||
| ! TODO(jiwon) this might not be correct because it doesn't know the index |
There was a problem hiding this comment.
Hi @peverwhee, is this code still valid (line 116 - 122)? We discussed that the sequence of the constituents array is not guaranteed. Do we now need to search for the index of the MICM constituents before attempting to retrieve the molar mass?
do i_elem = 1, size(molar_mass_arr)
call constituent_props(i_elem)%molar_mass(molar_mass_arr(i_elem), errcode, errmsg)
if (errcode /= 0) then
errmsg = "[MUSICA Error] Unable to get molar mass."
return
end if
end do
| !> Molar mass value of dry air is obtained from 'CAM-SIMA/src/utils/std_atm_profile.F90' | ||
| ! TODO(jiwon) | ||
| real(kind_phys), parameter, public :: MOLAR_MASS_DRY_AIR = 0.0289644_kind_phys ! kg mol-1 |
There was a problem hiding this comment.
Hi @mattldawson and @nusbaume, I get the molar mass value of dry air from CAM-SIMA/src/utils/std_atm_profile.F90. Is it okay to define it here? We discussed about passing parameters from CAM-SIMA, but I'm still unsure about the approach we're going to take.
There was a problem hiding this comment.
Hi @boulderdaze, you should be able to read in the molar mass of dry air as an input argument with the standard name molecular_weight_of_dry_air, which will be in units of g mol-1. There is also a composition-dependent version as well, but that is only used in WACCM and WACCM-X configurations (although if you would like to use that quantity instead just let me know).
The reason we generally suggest passing this value in as an argument is because some host models and host model configurations might have slightly different values for the dry air molar mass, especially if they are configured for high-top, paleoclimate, or exoplanet simulations, and so by bringing it in as an argument we won't have to worry about molar mass value mismatches.
Anyways, hopefully that helps? If for some reason that method does not work for your particular setup here just let me know. Thanks!
There was a problem hiding this comment.
Ah that makes sense, thank you for the additional details! I will use molecular_weight_of_dry_air as an input.
| !> Configures gas species. | ||
| ! Note: The user can customize this function to configure gas species in the system | ||
| ! This function allocates memory for the gas species that is needed in tuvx_init and tuvx_run. | ||
| ! The user is responsible for deallocating the memory. | ||
| subroutine configure_gas_species(constituent_props, gas_species_group, errmsg, errcode) | ||
| use ccpp_const_utils, only: ccpp_const_get_idx | ||
| use ccpp_constituent_prop_mod, only: ccpp_constituent_prop_ptr_t | ||
|
|
||
| type(ccpp_constituent_prop_ptr_t), intent(in) :: constituent_props(:) | ||
| type(gas_species_t), allocatable, intent(out) :: gas_species_group(:) | ||
| character(len=512), intent(out) :: errmsg | ||
| integer, intent(out) :: errcode | ||
|
|
||
| ! local variable | ||
| integer :: num_gas_species = 3 | ||
| real(kind_phys), parameter :: SCALE_HEIGHT_DRY_AIR = 8.01_kind_phys ! km | ||
| real(kind_phys), parameter :: SCALE_HEIGHT_O2 = 7.0_kind_phys ! km | ||
| real(kind_phys), parameter :: SCALE_HEIGHT_O3 = 7.0_kind_phys ! km | ||
| real(kind_phys) :: molar_mass_O2 ! kg mol-1 | ||
| real(kind_phys) :: molar_mass_O3 ! kg mol-1 | ||
| integer, parameter :: INDEX_NOT_KNOWN = -9999 | ||
| integer :: index_air, index_O2, index_O3 | ||
|
|
||
| allocate(gas_species_group(num_gas_species)) ! need to deallocate | ||
|
|
||
| ! TODO(jiwon) - I commented out this block of code that searches for the molar mass | ||
| ! of air and instead hard-coded it using the value imported from CAM-SIMA. | ||
| ! Should we register air in the register phase like cloud liquid water content, | ||
| ! or should we include it in the configuration file, or is it good as it is? | ||
|
|
||
| ! Air | ||
| ! Find the index of the species and create a 'gas_species_t' for it | ||
| ! call ccpp_const_get_idx(constituent_props, "air", index_air, errmsg, errcode) | ||
| ! write(*,*) "air index: ", index_air | ||
| ! if (errcode /= 0) return | ||
|
|
||
| ! call constituent_props(index_air)%molar_mass(molar_mass_air, errcode, errmsg) | ||
| ! if (errcode /= 0) return | ||
|
|
There was a problem hiding this comment.
Hi @mattldawson, I was trying to make the gas species configurable, allowing the user to provide their own species and define a function. Do you have any feedback on this implementation? Also, air is not included in the species.json, so it ends up being hard-coded. Could you share your thoughts on that in my comment?
Currently in draft for questions and discussions