What causes `wrf.exe` to terminate after running for a period of time?

[xinyuchen]

I have a three-level nested domain with grid spacings of 27 km, 9 km, and 3 km. I reduced the time step from 90 s to 54 s, but the simulation still aborted. Worse yet, it terminated at an even earlier simulation time than before. I have also modified the cu_physics options, testing the combinations 16,16,0, 16,0,0, 1,1,0 and 1,0,0; all of these cases ended prematurely. I adjusted mp_physics from 8 to 6 as well. At present, I cannot identify the root cause of this issue. Could anyone help me troubleshoot it?Attached are my namelist.input, rsl.error.0000 and namelist.wps. Besides, the termination always occurs around 04:30 to 05:50 on day 9.5. Does this indicate that anomalies arise in the simulation during this period?

 

[Ming Chen]

@ xinyuchen

Your namelist.input looks fine except that:

&bdy_control


spec_bdy_width           = 8,


spec_zone                = 1,


relax_zone               = 6,


specified                = .true.,  .false.,  .false.,


nested                   = .false.,   .true.,   .true.,


/

Please change it to :

&bdy_control


spec_bdy_width           = 8,


spec_zone                = 1,


relax_zone               = 7,


specified                = .true.,  .false.,  .false.,


nested                   = .false.,   .true.,   .true.,


/

In addition, for delx =27km, time_step can be 150. Your current setting of 54 is way too small.

Please change the above options, then try again. If the case still crashes, pleasre recompile WRF in debig mode, then run the case again and save the log file for me to take a look.