Problem when adding NO emission to ext_frc_specifier

[koichi omi]

Hello,

I'm trying to calculate the case with NO emission from aircraft to investigate its effects.
The calculation conditions are below.

COMPSET: FWHIST
RES: f09_f09_mg17
RUN_STARTDATE: 1979_0101
STOP_N = 13
STOP_OPTION = nmonths

I added NO emission file to "ext_frc_specifier" in the $CASEROOT/CaseDocs/atm_in.
However, when I execute this case, the calculation failed to converge.

Is there any suggestion to fix this problem?

Sincerely,

 

[mmills]

Try reducing the model's dynamical time step. The number of dynamics time steps per physics time step is set by the namelist variable fv_nsplit. By default it is set to 8. You can try doubling it to 16. If that doesn't work, keep doubling. It may be possible to get to a stable condition, after which you can reduce fv_nsplit again.

The only restriction on the value of fv_nsplit is that it must be a multiple of fv_nspltrac (default value 2), which in turn must be a multiple of fv_nspltvrm (default value 2). Sometimes it can help to increase these other two namelist variables at the same time. For example, you could try putting these changes in your user_nl_cam:

fv_nsplit = 128
fv_nspltrac = 128
fv_nspltvrm = 128

 

[koichi omi]

Thank you so much for your advice!

I tried the following setting.

fv_nsplit = 4096
fv_nspltrac = 4096
fv_nspltvrm = 4096

But the calculation failed to converge.
Is there any other suggestion to fix this problem?

Or, is doubling "fv_nsplit" the only way to fix this problem?

 

[mmills]

So the model runs fine if you don't add the NO emissions? Perhaps you are adding a very large amount. Have you checked units to make sure your emissions make sense? Can you try starting with a reduced level of NO emissions?

 

[koichi omi]

Thank you for your reply.

The model runs successfully when I don't add the NO emission.
In addition, the model also runs successfully when I add one-third of the amount I want to add.
So, perhaps NO emission I added was a little large.

Is there any suggestion to run the model with increased level of NO emission?

 

[mmills]

You could try increasing the emissions steadily to allow chemistry to adjust.