Good morning,

I would like to calculate the wind stresses over sea ice in order to send them to the coupler and the ocean model I'm using. The ocean model I use expects the windstresses over sea ice to be scaled down by the momentum transfer coefficient for sea ice. As far as I can see, NEMO doesn't require this and does this scaling itself. Therefore in the OIFS-FESOM2 coupled model I have too large windstresses over sea ice. Before I implement this from scratch, I wanted to quickly ask if a piece of code for this is available somewhere. I'm asking because I found the following section in surface_fields_mix.F90:


surface_fields_mix.F90
! * Group VG=VCLIG: ice-coupler diagnostic fields:
!   ky: currently not used, missing setup in su_surf_flds.F90
TYPE TYPE_SFL_VCLIG
  TYPE(TYPE_SURF_MTL_2D),POINTER :: YICFR   ! sea-ice fraction
  TYPE(TYPE_SURF_MTL_2D),POINTER :: YSOUP   ! upward solar flux over sea-ice
  TYPE(TYPE_SURF_MTL_2D),POINTER :: YIRUP   ! upward IR flux over sea-ice
  TYPE(TYPE_SURF_MTL_2D),POINTER :: YCHSS   ! sensible heat over sea-ice
  TYPE(TYPE_SURF_MTL_2D),POINTER :: YEVAP   ! evaporation over sea-ice
  TYPE(TYPE_SURF_MTL_2D),POINTER :: YTAUX   ! U-component of stress over sea-ice
  TYPE(TYPE_SURF_MTL_2D),POINTER :: YTAUY   ! V-component of stress over sea-ice
  TYPE(TYPE_SURF_MTL_2D),POINTER :: YVG(:) => NULL()
END TYPE TYPE_SFL_VCLIG

Best regards,

Jan


3 Comments

  1. Glenn Carver

    Hi Jan,

    Those pointers are not used in IFS.

    We think there might be some code in the wave model that will compute a drag coefficient over sea-ice but I need to check first. Will be in touch.

    Glenn

  2. Jan Streffing

    Hello Glenn,

    I found that code in surf/module/vupdz0s_mod.F90

    There are 8 different drag coefficients that can go into the calculation of PUSTRTI. I guess it mixes them base on the surface type fraction? I think I might add two fields that contain only the windstresses for the sea ice and ocean separately. That's what the ocean model I work with expects to get at least.


    Cheers,

    Jan

  3. Glenn Carver

    Hi Jan,

    Apologies for the delay in replying, we've been busy.

    There is some code in IFS that might help if you have not already found it. Look in ifs/phys_ec/ec_phys.F90, between lines 1193 "IF (LWCOU) THEN" and the "ENDIF" on line 1223, this is where the computation is done for the coupling to the wave model. The PSURF%PSD_WW array is written to with the neutral 10m wind values before used by the wave model. The wave model does not use the drag coefficient from the model directly, instead using the 10m neutral wind. This is because for mixed land/sea points, the drag can change by an order of 10 due to the orography over land. The wind by contrast changes much more smoothly.  The same applies over mixed sea-ice/ocean points.

    If that's not quite the detail you need, let me know and I will ask someone.

    Regards,  Glenn