Regression test for the S2S configuration without CICE coupling

Submitted by Maria.Aristizabal on Mon, 02/14/2022 - 21:13
Forums
Build Dependencies

Hello,

I recently run the regression test cpld_2threads P8 test, Fully coupled FV3-CCPP-MOM6-CICE-CMEPS-WW3 system - C96MX100 - 2 threads

Is it possible to run a configuration that is not coupled to ice model CICE? In other words, is it possible to run the configuration FV3-CCPP-MOM6-CMEPS?

Thanks!

Maria

Permalink

Submitted by mark.potts on Mon, 02/14/2022 - 22:54

You *might* be able to get the model to build without CICE by commenting out the CICE line in cmake/configure_apps.cmake under the S2S section, I have no idea if it would actually work. 

Permalink

Submitted by Maria.Aristizabal on Tue, 02/15/2022 - 19:20

Hi Mark,

I did what you suggested: " get the model to build without CICE by commenting out the CICE line in cmake/configure_apps.cmake under the S2S section". The ufs-weather-model does build without including any CICE component, but at run time it fails because it can not find CICE. So it seems that the current set up of the ufs-weather-model does need CICE to be able to run. Do you know if at some point in the development of the S2S application, the capability of running without the ice model was available? I am asking because for the hafs system we need to couple fv3 to mom6 but we don't need to couple to any ice model.

Thanks for your help!

I am copying here the error message:

20220215 101458.182 ERROR            PET180 EARTH_GRID_COMP.F90:519 Not valid  -  No component cice6 found
20220215 101458.182 ERROR            PET180 EARTH Grid Comp:src/addon/NUOPC/src/NUOPC_Driver.F90:764 Not valid  - Passing error in return code
20220215 101458.182 ERROR            PET180 EARTH Grid Comp:src/addon/NUOPC/src/NUOPC_Driver.F90:457 Not valid  - Passing error in return code
20220215 101458.182 ERROR            PET180 UFS.F90:386 Not valid  - Aborting UFS
20220215 101458.182 INFO             PET180 Finalizing ESMF

Permalink

Submitted by jong.kim on Wed, 02/16/2022 - 22:05

Hi Maria,

This is Jong. BTW what I am thinking is a bit hardwired appraoch if you have time restriction. It depends on how you will prescribe sea ice conditions in your coupled model configuration. One option is that you can directly tweak CICE time stepping main driver to by-pass and kick in your prescribed sea ice condition file. 

Thanks

-Jong

Permalink

Submitted by Maria.Aristizabal on Thu, 02/17/2022 - 20:14

Hi Jong,

Thanks for the reply. We do not have any prescribed sea-ice condition file. What we need is not to couple to any ice model because the areas of interest to the hafs system are not covered by ice at any point in time.

Best,

Maria

Permalink

Submitted by jong.kim on Thu, 02/17/2022 - 22:04

Maria.

If your case is for HAFS system, some people are working on coupling MOM6 to HAFS system. I am wondering if you discuss with them. Let me know if you want to open a thread of email to catch the update.

Thanks

-Jong 

  • 50 views

Copyright © 2023. All rights reserved.