MYNN/Cloud Radiative Feedback causes segmentation error

[gpriftis]

Hi,

I am running a simulation over 3 domains (9-,3-,1-km) to examine the role of cloud radiative feedback on coastal convection. Following other threads, I have included the following:

1) icloud = 0
2) cu_rad_feedback = .false.,.false.,.false.
3) shallowcu_forced_ra = .false.,.false.,.false.
4) icloud_bl = 0

I am getting a segmentation error which originates form the MYNN coupling with radiation flag (rsl error file). When the flag icloud_bl=0 the simulation fails, whereas when I switch it to icloud_bl=1 the simulation is running fine.

I am using the MYNN as a PBL scheme, and I wasn't expecting this error regarding the icloud_bl. Any help or thoughts would be appreciated.

 

[Ming Chen]

There is no error message in your rsl file. Can you check all your rsl files, and see whether you can find some messages that are helpful for debugging?

 

[gpriftis]

I increased the debug_level to 1000 but I still don't get any errors. I looked through the files and didn't really find any message that gives a clue of what's going on. I am attaching the last rsl.error file from the run. All the rsl files stop at the same output message:
d01 2020-08-15_00:00:00 in MYNNPBL

I also reduced the radt from 9 to 3 for testing, but the run still fails. It actually writes the file wrfout_d01_2020-08-15_00:00:00 and then if fails. Shouln't all the initial files be written, i.e. wrfout_d01_2020-08-15_00:00:00, wrfout_d02_2020-08-15_00:00:00, wrfout_d03_2020-08-15_00:00:00, regardless of the flag icloud_bl?

 

[Ming Chen]

Increasing the debug level won't really give you more information. Please keep it to be debug_level = 0.

For this case, would you please recompile WRF in debug mode (i.e., ./clean -a and ./configure -D), then rerun the case. The log file should tell excatly when and where the model first crashes. This may give you some hints what is wrong.