Skip to content

OM5 crashes in debug mode because land%mom_rough is set to zero #9

@nikizadehgfdl

Description

@nikizadehgfdl

When I try OM5 in debug mode it gets a division by zero:

forrtl: error (73): floating divide by zero                                                                             
Image              PC                Routine            Line        Source                                              
libc.so.6          000014DE0F33B900  Unknown               Unknown  Unknown                                             
fms_OM5_cobv3_com  000000000811974C  monin_obukhov_int         130  monin_obukhov_inter.inc                             
fms_OM5_cobv3_com  0000000006285095  monin_obukhov_mod          49  monin_obukhov.inc                                   
fms_OM5_cobv3_com  00000000005EFEDB  surface_flux_mod_         386  surface_flux.F90                                    
fms_OM5_cobv3_com  00000000004AB544  atm_land_ice_flux        1242  atm_land_ice_flux_exchange.F90                      
fms_OM5_cobv3_com  000000000043A214  full_coupler_mod_        2006  full_coupler_mod.F90                                
fms_OM5_cobv3_com  0000000000413252  MAIN__                    481  coupler_main.F90          

I tracked this to land%mom_rough being set to zero and passed to Monin-Obukhov scheme which devides by it.
The following (line of code in land_null [https://github.com/NOAA-GFDL/land_null/blob/main/land_model.F90#L354])
land2cplr%rough_mom = 0.0
Indeed, setting this to 1.0 instead of 0 makes the debug run go fine.
Does anyone know a realistic order of magnitude for land%rough_mom that this can be set to?
Also, should the value matter for ocean-ice runs/answers at all?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions