Issues outputting microphysics variables to wrfout

[jalter]

WRF3.7.1 Thompson Scheme
I am trying to output prr_wau, pri_wfz, prr_rcw, prs_scw, prg_scw, prg_gcw from the Thompson Microphysics scheme and write it to the wrfout files.
compile and real are both successful, but it always stop when it run wrf.exe
and rsl files do not show where is error, ONLY rsl.error.0042 shows "SIGSEGV: Segmentation fault"
it bother me a long time, please tell me the problems

 

[jalter]

module+Registry+solve+namelist+rsl

 

[kwerner]

Hi,
There are many CFL errors in your rsl* files. You can find these by issuing:

grep cfl rsl.error.*

This means that the model has become unstable - typically due to complex terrain. Take a look at this FAQ that addresses CFL errors:
https://forum.mmm.ucar.edu/phpBB3/viewtopic.php?f=73&t=133

 

[jalter]

thank you for your advice,
as you suggest, I set smooth_cg_topo = .true. and epssm = 0.2
when I use "grep cfl rsl.error.*", the cfl error didn't show again. However ,it still stopped .
I try to set time_step as 15,12,9,6,3 ,it both didn't work.

 

[kwerner]

Hi,
Can you please set debug_level = 0 and rerun this? In recent versions of the code we have actually removed that variable from the namelist as it is rarely helpful and just fills the RSL files with a lot of junk, making them difficult to read through. It can actually sometimes make them so large that they use up the disk space, causing the model to fail. It's unlikely that is what is happening here, but I'd like to see the RSL files without the junk. Once it fails, please package the files back up and attach them again, along with your most recent namelist. Thanks!

 

[jalter]

When I set debug_level = 0 , wrf.exe has stopped so quickly

 

[kwerner]

Hi,
Okay, it was stopping immediately before also, but wasn't printing out all the junk. Unfortunately there is still not much explanation in the rsl files. Can you do a test with a version of WRF that has not been modified (just the fresh out of the box code) with the same input and namelist to see if it's able to run?
- If it does run without stopping, then we know the problem is with your modifications. If that's the case, I would recommend starting with the clean code and making changes slowly, compiling, doing quick tests to verify if everything runs okay, and the move on to another small modification until you track down the problem.
- If it still stops immediately, then because it stops at the beginning, it is likely an issue with the input data. Check to make sure you aren't missing any data at any levels, and that the input files are not corrupt or have NaNs in them.