Welcome to the new DiscussCESM forum!
We are still working on the website migration, so you may experience downtime during this process.

Existing users, please reset your password before logging in here: https://xenforo.cgd.ucar.edu/cesm/index.php?lost-password/

control the aerosol constituent calculating optical properties

Henry

Henry
New Member
Hi there,

I want to run CAM5.0 without any aerosol effect except only black carbon, which means removing all kinds of aerosols but only calculating optical properties(absorption) of black carbon. I am trying to compare the absorption enhancement between all aerosols and only black carbon. Maybe change any namelist variable?
Any suggestion is much appreciated. Thanks a lot.

Kind regards,
Henry
 

brianpm

Member
Henry, you can do this with namelist changes. Take your control run's atmospheric namelist (atm_in) and remove all the non-black carbon from the emissions files (ext_frc_specifier, srf_emis_specifier). It is valid to completely remove entries from those lists if the files don't contain any emissions you want. That will take care of the prescribed emissions.

Then you have to decide what you want to do about sea-salt and dust emissions, which are calculated diagnostically. To remove sea-salt emissions, you can set the namelist parameter `seasalt_emis_scale` to zero. For dust, you can set the namelist parameter `dust_emis_fact`, but I'm not sure whether you can get the dust emission to go to identically zero because that factor seems to end up in a denominator in `./components/cam/src/chemistry/modal_aero/dust_model.F90`. I didn't have to deal with this in aquaplanets because there's no source of dust (see: Error - Cookies Turned Off). One way you can definitely turn off the dust emission is to do a source modification of dust_model.F90 and multiply cflx by zero (see the last few lines of that file).
 

Henry

Henry
New Member
Thanks for your reply.
I do find the namelist parameter "dust_emis_fact", but I do not find "seasalt_emis_scale" and dust_model.F90 in the .../cam/src/chemistry/modal_aero/. Is that about my version of CAM? I'm running CESM1.2.2-CAM5.0.

Henry
 

brianpm

Member
Yes, that is because you are using an old version.

It looks like dust_emis_fact is treated the same in CESM1, in the sense that it is in a denominator. It looks like the code is in cesm1_2_2/models/atm/cam/src/chemistry/bulk_aero/dust_intr.F90 and can be modified (maybe around line 880). Looking at it again, it looks like the *easiest* way to remove dust would be to zero-out the entire soil erodibility file. I DO NOT KNOW whether that has an impact on other aspects of the simulation, in particular if the land model makes use of it.

To get around seasalt, you will need to find where the emission is calculated. The only place that looks right to me is in ./atm/cam/src/chemistry/bulk_aero/progseasalts_intr.F90. I'm not sure if that is the code that is used in CAM5.
 

Henry

Henry
New Member
Thanks for your reply.
I set "dust_emis_fact" to zero, and just like you said, the model would not complete. So I modified source code multiplying "cflx" by zero even turning "sflx" to zero in dust_intr.F90. Then I output the optical properties and found that AODDUST and burdenDUST is a little bit smaller than before but I thought these optical properties are supposed to be zero. What do you think about it?
Thanks again!

Henry
 

brianpm

Member
The AOD of dust won't be zero unless the burden is zero. I wonder whether your initial condition has some dust still. If so, you can write the dust burden to a high-frequency output file and see if it is decreasing over time. If not, then my next guess would be that the modification is not being used. Possible causes for that would be that (a) the source code modification isn't being picked up in the re-build step because it is in the wrong location ($CASE/SourceMods), or (b) the zeroing out of cflx is not correct.

It is probably worth a test to see if you can get dust down to zero by modifying the soil erodibility file.
 

CESM researcher

HW doctor
New Member
Thanks for your reply.
I set "dust_emis_fact" to zero, and just like you said, the model would not complete. So I modified source code multiplying "cflx" by zero even turning "sflx" to zero in dust_intr.F90. Then I output the optical properties and found that AODDUST and burdenDUST is a little bit smaller than before but I thought these optical properties are supposed to be zero. What do you think about it?
Thanks again!

Henry
Hello, Henry,

Recently, I encountered a similar problem. I want to eliminate the effects of dust and seasalt emissions, but for files dust_intr.f90 and Progseasalts_intr.f90. I wonder if the modification is correct. Could you give me some guidance? Modify the code in the picture.Finally, I looked at the DSTFLXT indicator on lnd, the total surface dust emissions, to check that the source code was modified correctly. Am I doing this correctly?(figure4)
 

Attachments

  • dust.png
    dust.png
    28.5 KB · Views: 0
  • sea.png
    sea.png
    57.9 KB · Views: 0
  • sea1.png
    sea1.png
    33 KB · Views: 0
  • mmexport1638953630460.jpg
    mmexport1638953630460.jpg
    17.7 KB · Views: 0
Top