Help with fldread.F90 for surface boundary conditions

Versions v4.2.x
1

Hi all,
I used to force my configuration with yearly climatological emp fields at daily frequency, which worked fine.
Now I’m trying to go to fields at monthly frequency, making use of the time-interpolation implemented in fldread.F90. The simulation runs, but is constantly forced with the January values from my forcing.nc file, so I have no seasonal cycle.
I assume it is a problem with the structure of my forcing file, but can’t find examples for how it should look. Here the output from ncdump -h:

netcdf emp_monthly_1deg {
dimensions:
        t = 12 ;
        y = 199 ;
        x = 52 ;
variables:
        double emp(t, y, x) ;
                emp:_FillValue = 0. ;
}

and my namelist_cfg:

&namsbc_flx    !   surface boundary condition : flux formulation        (ln_flx =T)
!-----------------------------------------------------------------------
   cn_dir      = './'      !  root directory for the fluxes data location
   !
   !                 !  file name                ! frequency  ! variable  ! time interp.!  clim  !  monthly/yearly  
   sn_emp  = 'emp_monthly_1deg' ,         -1       ,  'emp'     , .true.         , .true. ,         'yearly'  
/

I would appreciate any help with this!
Thanks,
David

2

Hi David,

Can you try to set the time dimension of your netcdf file to “UNLIMITED” ?
I think this could the be cause of your problem.

Hope this helps,

1 Like
3

thank you very much! That did the trick