ROMS currents to SWAN

[jacopo]

in mct_roms_swan.h I see:

Code: Select all

          CASE ('Ubar')                   ! 2D U-momentum

            CALL ROMS_export2d (ng, tile,                               &
     &                          id, gtype, scale, add_offset,           &
     &                          LBi, UBi, LBj, UBj,                     &
#ifdef SOLVE3D
     &                          OCEAN(ng)%u(:,:,N(ng),NOUT),            &
#else
     &                          OCEAN(ng)%ubar(:,:,KOUT),               &
#endif
     &                          Fields(id)%ExpMin, Fields(id)%ExpMax,   &
     &                          Asize, A,                               &
     &                          status)

Does this mean that ifdef SOLVE3D ROMS is passing the surface velocity, not 2D currents?

-J

[jcwarner]

yes. For 2D it sends ubar. For 3D it sends the surface currents. However, for 3D this perhaps should be some weighted average over a depth ~ Hsig, with an exponential decay.

[jacopo]

good news, I was going to implement the same thing, but it's already there

However, for 3D this perhaps should be some weighted average over a depth ~ Hsig, with an exponential decay.

Indeed, but including TKE from wave breaking makes this OK anyway.

[wulunyu]

I used the wave related depth weighted velocity as the doppler velocity from George L. Mellor
《A Surface Wave Model for Coupling with Numerical Ocean Circulation Models》.
I can send you if you want it.

yes. For 2D it sends ubar. For 3D it sends the surface currents. However, for 3D this perhaps should be some weighted average over a depth ~ Hsig, with an exponential decay.

[jcwarner]

hey. i happended to come across that paper for a different reason, but i have it in front of me.
Can you help me dig thru and point me to what you are referrring to?
-j

[wulunyu]

I referred to equation 6 ---The Doppler (or advective) velocity.
I used this velocity to advect the action density(N)in SWAN.
I still use the same one when calculating the relative wind speed(uwind-ucurrent) for use in source term, which I think surface current velocity should be a better choice.

Allen