Wet deposition issue caused by area RESUSPENSION in middle levels

[shrtri2494]

Your name

Shruti Tripathi

Your affiliation

IITB

Please provide a clear and concise description of your question or discussion topic.

I am running the nested simulation for January 2022, and the GeosChem Version is 14.1.1. First, I encountered an error on 20/01/2022, and I tried to resolve this by changing the corrupted files of HEMCO offline (seasalt, bvoc and soilNOx) #2093 and Prevent divide by zero in sulfur chemistry mentioned in #1698. I have attached the GC log file for this error. After solving this, I got "WETDEP: ERROR at 80 6 44 for species 2 in area RESUSPENSION in middle levels" at 21/01/2022 00:05. In #1217 mentioned changing the time step will solve this error, but my time step was already set to 300/600.

I tried to debug this error using gdb, but it ran successfully from 21/01/2022 to 24/01/2022, whereas in my main run directory, I got an error at 21/01/2022 00:05 when i tried to run from 21/01/2022 to 24/01/2022.
I have checked the Hemco config, History.rc, and Geoschem config files from the debug directory and my main directory, and they were all identical.

Here, I am attaching three GC log files

1. For Nan error
2. For resuspension error
3. successful run for December 2021
   GC_Log Files.zip

Please help me solve this error so I can run the nested simulation for January successfully.

[yantosca]

Thanks for writing @shrtri2494 and thanks for your patience. We are just now back from the holidays.

It seems like the wetdep error is a side-effect of a non-convergence error in chemistry. The KPP Rosenbrock solver cannot integrate the mechanism forward at certain grid boxes and returns with error. This can sometimes happen in nested-grid simulations when the boundary conditions contain unphysical values (and these get copied into the region where chemistry is being done).

Please see a potential workaround here (I just added this to ReadTheDocs):

• https://geos-chem.readthedocs.io/en/latest/geos-chem-shared-docs/supplemental-guides/error-guide.html#integrator-error-code-5-stop-at-integrate-kpp