Force model with wind stress

[aryapaul]

Normally I force the model with wind speed at 10m. What changes do I need to do if I want to force the model with wind stress ? Actually I am developing LETKF data assimilation for ROMS. It is using ensemble atmospheric fluxes and I am provided with wind stress rather than wind speed. I am not finding any suitable variable name in the var_info.dat as well.

[kate]

Here is the varinfo.dat section for wind stress:

Code: Select all

!------------------------------------------------------------------------------
!  Momentum stress variables.
!------------------------------------------------------------------------------
!

'sustr'                                            ! Input/Output
  'surface u-momentum stress'
  'newton meter-2'                                 ! Input:  [N/m2]
  'surface u-momentum stress, scalar, series'      ! [m2/s2]
  'sms_time'                                       ! Output: [N/m2]
  'idUsms'
  'u2dvar'
  1.0d0

'svstr'                                            ! Input/Output
  'surface v-momentum stress'
  'newton meter-2'                                 ! Input:  [N/m2]
  'surface v-momentum stress, scalar, series'      ! [m2/s2]
  'sms_time'                                       ! Output: [N/m2]
  'idVsms'
  'v2dvar'
  1.0d0

[aryapaul]

Thanks for the reply. I have recompiled the model using the following cppdefs and forced the model with wind stress ( not wind speed ).

#undef TS_MPDATA
#undef INLINE_2DIO
#define SPONGE
#define UV_QDRAG
#define VISC_GRID
#define UV_ADV
#define TS_A6VADVECTION
#define TS_A4HADVECTION
#define UV_C4ADVECTION
#define UV_VIS2
#define UV_VIS4
#define UV_SMAGORINSKY
#define TS_DIF2
#define TS_DIF4
#define TS_SMAGORINSKY
#define UV_COR
#define DJ_GRADPS
#define MIX_GEO_TS
#define MIX_S_UV
#define MIX_S_TS
#define NONLIN_EOS
#define SALINITY
#define SOLVE3D
#define MASKING
#undef SPLINES
#undef SSH_TIDES
#undef UV_TIDES
#undef RAMP_TIDES
#define SOLAR_SOURCE
#define SRELAXATION
#define CURVGRID
#define AVERAGES
#define AVERAGES_FLUXES

#undef BULK_FLUXES
#ifdef BULK_FLUXES
# undef LONGWAVE_OUT
# undef LONGWAVE
# define EMINUSP
#endif

# undef MY25_MIXING
# ifdef MY25_MIXING
# undef N2S2_HORAVG
# define KANTHA_CLAYSON
# undef UV_LOGDRAG
# define K_C2ADVECTION
# undef K_C4ADVECTION
# endif
#undef GLS_MIXING
#if defined GLS_MIXING
# define KANTHA_CLAYSON
# define N2S2_HORAVG
# define CRAIG_BANNER
# define CHARNOK
# undef ZOS_HSIG
# undef TKE_WAVEDISS
#endif

Earlier, when I was forcing the model with wind speeds, I have simply had BULK_FLUXES ON and the SST were fine.

But when forced with wind stress & the above cppdef.h, I am seeing that the surface temperatures are warming by as much as 3-4 degrees. Is it that I need to switch on/off something else as well ?

[kate]

You don't want both of these on:

Code: Select all

#define MIX_GEO_TS
#define MIX_S_TS

The second is evil except over a flat bottom.

For salinity, you are nudging to SSS. You must be providing surface temperature flux with the wind stress, right? If so, it is a good idea to include a dQ/dSST term as well, which sort of acts as a nudging to the SST used when computing the temperature flux. Otherwise, you applying fluxes appropriate to one SST while having a different SST. With bulk fluxes, the flux is appropriate to your SST.

You seem to have turned off all the non-background vertical mixing so any oddness at the surface will not be diffused down.