Remove dust_a1 radiative effects

Hi,

I am currently trying to remove the radiative effects of dust_a1 in CESM2.1.5 but am running into issues and do not know the rrtmg code so well. I have seen other similar issues on the forum such as here and here but they were either not answered or I could not apply the proposed solutions.

I am aware of the possibility to define a new radiation call using "rad_diag_1" where dust_a1 is removed from the radiation by creating a new mode without it. However, I cannot use this because I need to remove it completely. I tried a similar approach but changing "rad_climate" instead of "rad_diag_1". However, I get the following error: ERROR: init_modes ERROR: cannot find physics buffer field dst_c1.

Another option would be to change the physical properties of dust, but I don't know which ones to change exactly.

I guess that I am not the first one to try something similar and was therefore wondering if anyone could help me with this problem.

Thanks in advance for the time and the help !

Paul

Hi Paul,
This entry may help:
Best,
Rebecca

Hi Paul--
I don't entirely understand what you are trying to achieve, but I think you are aware that if you work on rad_diag_1 you are isolating the radiative effect diagnostic of dst_a1 as a history output. It is not really eliminating the radiative effects of dst_a1 on the met fields in the model (i.e., dst_a1 still warms the neigbhouring air and cools the Earth at the TOA in CAM). I don't know how to modify the RRTMG code to take the radiative effect of dst_a1 away either, so I think what you suggested about removing the optical properties of dust might be a good way.
1. If you look at mode_defs in your case under CaseDocs/atm_in, you will see something like
'A:dst_a1:N:dst_c1:dust:/glade/campaign/cesm/cesmdata/inputdata/atm/cam/physprops/dust_aeronet_rrtmg_c141106.nc:+'
this might be the file you want to modify. You can use python or NCO to set a bunch of variables to zero, e.g., hygroscopicity, refindex_real_aer_sw, refindex_im_aer_sw, and so on. Doing this, you keep dst_a1 transporting in CAM but they have little interaction with radiation (if you set the refractive indices to zero) and warm clouds (if you set hygroscopicity to zero). Let me know if you care about mixed phase clouds and ice clouds, and we can look at whether you use hetfrz and other details.
2. Another easy way to do it may be just to completely eliminate dst_a1 from the dust emission, so no dst_a1 exists in CAM. Of course this will not only take away the radiative effects but also change other things like total AOD and PM2.5 history fields.
Best,
Danny

Hey Danny,

Thanks for the help! I actually managed to do it by skipping the part of the code where the dust is read in modal_aer_opt.F90. Before that, I also tried to change the properties in 'A:dst_a1:N:dst_c1:dust:/glade/campaign/cesm/cesmdata/inputdata/atm/cam/physprops/dust_aeronet_rrtmg_c141106.nc:+' but it didn't work for me.

Best,
Paul

Hi Paul-
Right, that actually sounds like a better plan! I haven't thought about that. Since modal_aer_opt.F90 is the module that reads in the hygroscopicity and refractive indices of aerosols, editing that .F90 file can let you skip dust optical effects while keeping dust aerosols in other parts of the model. Nice!
Take care,
Danny