em_les

[廊桥遗梦]

I want to run the d03 file in the official em_ les, but it has been unsuccessful all the time. As a beginner, are there any steps? Could it be due to the namelist? Do I just need to change max dom to 3?

 

[kwerner]

Can you provide a bit more information? Are you getting an error, or just not getting d03 output files? In order to process 3 domains, you will need to set max_dom = 3, but you will also need to set up all namelist variables for 3 domains, as well.

If you're receiving an error, please attach your namelist.input file, as well as your error log. Thanks!

 

[廊桥遗梦]

Thanks for your answer!This is my namelist and log. There has been an error since the ideal.exe step.I'm very confused. Here it also reminds me that there was a problem with the MPI operation. When I installed it, I compiled the parallel version according to the official tutorial.

 

[kwerner]

Thanks for sharing those files. The error message the rsl.error.0000 file shows is :

--- ERROR: ghg_input available only for these radiation schemes: CAM, RRTM, RRTMG, RRTMG_fast
           And the LW and SW schemes must be reasonably paired together:
           OK = CAM LW with CAM SW
           OK = RRTM, RRTMG LW or SW, RRTMG_fast LW or SW may be mixed
  --- ERROR: e_vert differs on grid ids  2 and  1. Set vert_refine_method or make e_vert consistent.
  --- ERROR: e_vert differs on grid ids  3 and  2. Set vert_refine_method or make e_vert consistent.
-------------- FATAL CALLED ---------------
FATAL CALLED FROM FILE:  <stdin>  LINE:    2807
NOTE:       2 namelist settings are wrong. Please check and reset these options
-------------------------------------------

Although the namelist parameter "ghg_input" is not listed in your namelist, this parameter is turned on by default, so add the following to the &physics namelist record:

&physics
ghg_input = 0

It also looks like the default namelist.input file for the em_les case has the variable e_vert listed as you do (e_vert = 40, 100, 41). I'm not sure why that is and it's something I'll look into. Please set e_vert to the same value for each domain:

e_vert = 40, 40, 40

As for the MPI messages you see in your rsl* file, I think those are just standard MPI print-outs for your system. Because the simulation attempts to run with MPI and because you did get the rsl* files, I think your MPI is installed correctly, so you can ignore those prints.

 

[廊桥遗梦]

thank you!I would try it again.