RIBX never exceeds RIC, RIB error

[smeech84]

-------------- FATAL CALLED ---------------
FATAL CALLED FROM FILE: <stdin> LINE: 425
RIBX never exceeds RIC, RIB(i,kte) = NaN THETAV(i,1) = 275.0700 MOL= -27.71101 TCONV = NaN WST = NaN KMIX = 2 UST = 8.1822090E-02 TST = -1.3693518E-02 U,V = -0.4795572 -2.350616 I,J= 20 6


I haven't seen this error before and am having a hard time trying to resolve it. I've tried reducing time step and using adaptive time step but no dice. It seems to stop after about 14-15 hours each time. I 've used this physics configuration on other time periods and didn't run into any issues.

I have read from other error posts about this problem that it was related to nudging parameters. However, I am not using nudging.

Has anyone else come across this? Namelist attached.

Thanks!

 

[Ming Chen]

This error message indicates that something went wrong in the physics. I looked at your namelist.input, and I have a few concerns about the physics options:

(1) ra_lw_physics = 4,4,4, 1,
ra_sw_physics = 1, 1, 1, 1,
radt = 20, 20, 20, 20,

Can you change to

ra_lw_physics = 4,4,4, 1,
ra_sw_physics = 4, 4, 4, 1,
radt = 30, 30, 30, 20,

(2) We recommend to keep PBL, SFCLAY and SURFACE schemes consistent for Pleim scheme, that is

sf_sfclay_physics = 7,7,7, 1,
sf_surface_physics = 7, 7, 7, 2,
bl_pbl_physics = 7,7,7, 1,

You can us different combinations of the above options, however, it may not always work, just as what you have sen in your failed case.


(3) Plan turn off cumulus scheme for D03.

 

[smeech84]

Hi Ming thank you for the reply.

This physics scheme is part of a 20 member ensemble that the US Air Force runs on their systems. So, this is just one of the members that failed at a certain time step in a month long simulation. They seem to run this all the time with seemingly no problems. I have also run this physics set up for other months with no problems.

I can change the radt as it is not specified in their ensemble but I don't think I can change the other options you have suggested. Is there anything else I may be able to try to get the model to get through this time period?

Thanks!

 

[Ming Chen]

This case failed after 14-15 hours of simulation, which indicates that the input data (wrfinpout and wrfbdy) are correct. It is definitely something wrong in the physics and only for this specific case. I would suggest to save the latest 'correct' wrfrst file, use the 'restart' capability of WRF to 'resdtart' the case, and save wrfout files every time step. Then we can check these wrfout files, find when and where the problem first occurs, and track back to find the reason.
It is a tedious and painful process to debug a specific case, I have to say.

 

[sreya]

Can you explain me as to why some combination of physics options fail?