Query on Regional Masking of Sea Salt Emissions in CESM Simulations

[Sarin]

Hi, I am conducting regional aerosol perturbation experiments in CESM2.1.5 using the F2000climo compset resolution to study the influence of aerosol teleconnections on regional climate. I have already successfully implemented regional masking of prescribed aerosol emissions through modifications to the emission NetCDF files, namely to BC, POM and SO4. However, I am now trying to understand how to handle interactively generated natural aerosols, particularly sea salt, which do not appear to be controlled through standard prescribed emission inventories. From the atm_in file, I identified seasalt_emis_scale = 1.00D0, which appears to be a global scaling parameter rather than a regionally configurable mask.

My main question is whether there is any recommended workflow in CAM6/MAM4 for suppressing sea salt only within selected geographic regions while retaining fully interactive aerosol transport and chemistry elsewhere. Any guidance, references, or example implementations would be greatly appreciated.

 

[Danny Leung]

Hi Sarin--
This probably requires some code changes in CESM. A good reference is dust emission, which uses a namelist variable called soil_erod_file that contains a path to a mask in a netcdf dataset. Here defines soil_erod_file and here documents the path to the mask. The namelist variable soil_erod_file is obtained from soil_erod_mod.F90 and
dust emissions are then masked by soil_erod_file in dust_model.F90.
So, you can do the same thing for seasalt emissions in seasalt_model.F90, if you have your own netcdf file for your own mask.
Let me know if you need further clarifications!
Danny

 

[Sarin]

Hi Sarin--
This probably requires some code changes in CESM. A good reference is dust emission, which uses a namelist variable called soil_erod_file that contains a path to a mask in a netcdf dataset. Here defines soil_erod_file and here documents the path to the mask. The namelist variable soil_erod_file is obtained from soil_erod_mod.F90 and
dust emissions are then masked by soil_erod_file in dust_model.F90.
So, you can do the same thing for seasalt emissions in seasalt_model.F90, if you have your own netcdf file for your own mask.
Let me know if you need further clarifications!
Danny

Hi Danny,

Thank you very much for the helpful suggestion and references. I ended up implementing it by modifying the modal aerosol sea salt emission specifically, chemistry/modal_aero/seasalt_model.F90 and chemistry/modal_aero/aero_model.F90.

I modified the ocean fraction term (ocnfrc) used in the sea salt emission calculation and introduced a regional mask inside the seasalt_emis()routine. I replaced the original ocean fraction with a masked local copy (ocnfrc_local). Within the selected region, the local ocean fraction was set to zero, thereby suppressing online sea salt generation only within the masked domain while leaving the rest of the global sea salt source unchanged.

I verified that the surface emissions (ncl_a1SF, ncl_a2SF..) became zero over the masked Arctic region. So the edit seems to be working. I realise this may not yet be the most elegant or generalised approach, particularly since the masking is currently hard-coded through lat-lon bounds (which I will change for each experiment) rather than using an external mask file or namelist framework. I am attaching the modified source files (uploading as .txt as I am not able to upload the .F90 file) used for the implementation. I would be very grateful if you could let me know whether this approach appears scientifically and technically reasonable from your perspective, or if there are any concerns/improvements you would recommend before proceeding to longer climate integrations.

Warm regards,

Sarin T.S.
Doctoral Student
Aerosol, Air Quality and Climate Lab
School of Earth, Ocean and Climate Sciences (SEOCS)
Indian Institute of Technology Bhubaneswar