Abstract aerosol interfaces extended for bulk representation #1404
Conversation
modified: src/chemistry/aerosol/aerosol_properties_mod.F90 modified: src/chemistry/aerosol/carma_aerosol_properties_mod.F90 modified: src/chemistry/aerosol/modal_aerosol_properties_mod.F90 modified: src/physics/cam/aer_rad_props.F90 modified: src/physics/cam/aerosol_optics_cam.F90
new file: src/physics/cam/bam_optics_diags_mod.F90 modified: src/physics/cam/aer_rad_props.F90 modified: src/physics/cam/aerosol_optics_cam.F90
modified: src/chemistry/aerosol/bulk_aerosol_properties_mod.F90 modified: src/chemistry/aerosol/hygrocoreshell_aerosol_optics_mod.F90 modified: src/chemistry/aerosol/hygroscopic_aerosol_optics_mod.F90 modified: src/chemistry/aerosol/insoluble_aerosol_optics_mod.F90 modified: src/physics/cam/aerosol_optics_cam.F90
new file: src/chemistry/aerosol/hygro_aerosol_optics_mod.F90 modified: src/chemistry/aerosol/aerosol_properties_mod.F90 modified: src/chemistry/aerosol/bulk_aerosol_properties_mod.F90 modified: src/chemistry/aerosol/carma_aerosol_properties_mod.F90 modified: src/chemistry/aerosol/hygroscopic_aerosol_optics_mod.F90 modified: src/chemistry/aerosol/modal_aerosol_properties_mod.F90 modified: src/physics/cam/aer_rad_props.F90 modified: src/physics/cam/aerosol_optics_cam.F90
modified: src/chemistry/aerosol/bulk_aerosol_properties_mod.F90 modified: src/chemistry/aerosol/hygro_aerosol_optics_mod.F90 modified: src/physics/cam/aer_rad_props.F90 modified: src/physics/cam/aerosol_optics_cam.F90
new file: src/chemistry/aerosol/volcrad_aerosol_optics_mod.F90 modified: src/chemistry/aerosol/aerosol_properties_mod.F90 modified: src/chemistry/aerosol/bulk_aerosol_properties_mod.F90 modified: src/chemistry/aerosol/carma_aerosol_properties_mod.F90 modified: src/chemistry/aerosol/modal_aerosol_properties_mod.F90 modified: src/physics/cam/aer_rad_props.F90 modified: src/physics/cam/aerosol_optics_cam.F90
modified: src/physics/cam/aer_rad_props.F90 modified: src/physics/cam/bam_optics_diags_mod.F90
modified: src/chemistry/aerosol/bulk_aerosol_properties_mod.F90 modified: src/physics/cam/aer_rad_props.F90 modified: src/physics/cam/aerosol_optics_cam.F90
renamed: src/physics/cam/bam_optics_diags_mod.F90 -> src/physics/cam/aer_vis_diag_mod.F90 modified: src/physics/cam/aer_rad_props.F90 modified: src/physics/cam/aer_vis_diag_mod.F90 modified: src/physics/cam/aerosol_optics_cam.F90
modified: src/physics/cam/aer_rad_props.F90
modified: src/physics/cam/aerosol_optics_cam.F90
modified: src/chemistry/aerosol/bulk_aerosol_properties_mod.F90 modified: src/chemistry/aerosol/hygro_aerosol_optics_mod.F90 modified: src/chemistry/aerosol/hygrocoreshell_aerosol_optics_mod.F90 modified: src/chemistry/aerosol/hygroscopic_aerosol_optics_mod.F90 modified: src/chemistry/aerosol/insoluble_aerosol_optics_mod.F90 modified: src/chemistry/aerosol/volcrad_aerosol_optics_mod.F90 modified: src/physics/cam/aer_vis_diag_mod.F90
|
|
||
| nspecies(:) = 1 | ||
|
|
||
| alogsig(:) = log(2._r8) |
There was a problem hiding this comment.
Should this be made as a parameter and then assigned to alogsig?
There was a problem hiding this comment.
What will we gain be inserting a parameter here?
There was a problem hiding this comment.
What will we gain be inserting a parameter here?
I'm just not sure where this magic number is coming from, so a comment for at least the value would help. It seems to be the same value as CARMA, and MAM has species-specific deviations.
There was a problem hiding this comment.
This may never be used for bulk aerosols. The settings are adopted from CARMA properties. Comments are added.
| ! call endrun('ERROR: bulk_aerosol_properties_mod%amcube not yet implemented') | ||
| amcube = -huge(1._r8) |
There was a problem hiding this comment.
Maybe add a comment here that amcube cannot be defined for bulk aerosols since there are no number concentrations in the bulk representation?
I understand endrun cannot be called here because this implements a pure function so maybe a comment of why the intentionally invalid value is returned here will be useful, to distinguish between 'does not exist for bulk representation' and 'not yet implemented'
There was a problem hiding this comment.
It seems to me that the need to look for workarounds like this is a sign that something is wrong with the abstract aerosol class structure. In this case, the abstract properties class is requiring information that is not present in all aerosol models.
In a perfect world, all code that used %amcube (and the other problematic methods below) would reside in the appropriate subclass and all uses would be restricted to that subclass (which could still be a superclass of final models if a set of aerosol models shared a property such as amcube).
I don't see CAM (or CAM-SIMA) getting ready for such a clean class hierarchy soon so in the interim, I think it would be best if the abstract base class (aerosol_properties) provided a default, overridable method to produce an invalid value. I am happy to present this change as a PR to this PR branch so you can evaluate it.
Speaking of an invalid value, what happens if <bulk_aero_prop_obj>%amcube gets called? The code where it is called has no value checks.
There was a problem hiding this comment.
@jimmielin and @gold2718
Many of these undefined methods (such as amcube) could probably be implemented for bulk aerosols. They are not needed for optical properties of prescribed bulk aerosols, which is the purpose of this PR. In the interest of time, I did not make the effort. They can be implemented later if needed.
There was a problem hiding this comment.
@fvitt, thanks for the explanation. Would it be possible to set the value to NaN instead of -huge? That way we would know if any not-implemented method was used in active code.
There was a problem hiding this comment.
One aspect I struggled when reviewing the PR is that perhaps there could be better distinction of what is not implemented yet (as @fvitt notes as the scope of this PR) versus what cannot be physically defined for bulk aerosols (as @gold2718 notes as "information that is not present in all aerosol models"). Either -huge or NaN serve as good magic numbers for me but I believe the comments should make the intent of the code clear.
There was a problem hiding this comment.
Now using NaN, as requested by @gold2718 .
|
|
||
| nullify(num) | ||
|
|
||
| call endrun('ERROR: bulk_aerosol_state_mod%get_ambient_num not yet implemented') |
There was a problem hiding this comment.
I would suggest returning the error as this cannot be defined for bulk
| call endrun('ERROR: bulk_aerosol_state_mod%get_ambient_num not yet implemented') | |
| call endrun('ERROR: bulk_aerosol_state_mod%get_ambient_num cannot be defined for bulk representation') |
|
|
||
| nullify(num) | ||
|
|
||
| call endrun('ERROR: bulk_aerosol_state_mod%get_cldbrne_num not yet implemented') |
There was a problem hiding this comment.
Similarly I would suggest returning the error as this cannot be defined for bulk instead of not yet implemented (as it never will)
| call endrun('ERROR: bulk_aerosol_state_mod%get_cldbrne_num not yet implemented') | |
| call endrun('ERROR: bulk_aerosol_state_mod%get_cldbrne_num cannot be defined for bulk representation') |
| @@ -0,0 +1,118 @@ | |||
| !------------------------------------------------------------------------------- | |||
| ! Insoluble (non-hygrostatic) aerosol optical properties | |||
There was a problem hiding this comment.
Nitpick for typo
| ! Insoluble (non-hygrostatic) aerosol optical properties | |
| ! Insoluble (non-hygroscopic) aerosol optical properties |
| ! call endrun('ERROR: bulk_aerosol_properties_mod%amcube not yet implemented') | ||
| amcube = -huge(1._r8) |
There was a problem hiding this comment.
It seems to me that the need to look for workarounds like this is a sign that something is wrong with the abstract aerosol class structure. In this case, the abstract properties class is requiring information that is not present in all aerosol models.
In a perfect world, all code that used %amcube (and the other problematic methods below) would reside in the appropriate subclass and all uses would be restricted to that subclass (which could still be a superclass of final models if a set of aerosol models shared a property such as amcube).
I don't see CAM (or CAM-SIMA) getting ready for such a clean class hierarchy soon so in the interim, I think it would be best if the abstract base class (aerosol_properties) provided a default, overridable method to produce an invalid value. I am happy to present this change as a PR to this PR branch so you can evaluate it.
Speaking of an invalid value, what happens if <bulk_aero_prop_obj>%amcube gets called? The code where it is called has no value checks.
modified: src/chemistry/aerosol/bulk_aerosol_properties_mod.F90
modified: src/chemistry/aerosol/aerosol_properties_mod.F90 modified: src/chemistry/aerosol/bulk_aerosol_properties_mod.F90 modified: src/chemistry/aerosol/carma_aerosol_properties_mod.F90 modified: src/chemistry/aerosol/hygro_aerosol_optics_mod.F90 modified: src/chemistry/aerosol/insoluble_aerosol_optics_mod.F90 modified: src/chemistry/aerosol/modal_aerosol_properties_mod.F90 modified: src/physics/cam/aerosol_optics_cam.F90
3 participants
For the purpose of aerosol optical properties input into radiative transfer, aerosol interfaces have been extended for bulk aerosols.
closes #1400
Roundoff level answer changing.