Problem with LBC at simple analytical case

[pelijas]

Hi everyone!

I'm trying to make a simple analytical test with the last ROMS version.
I want to reproduce a chanel with Clamped conditions at East and West sides, in particular for the u velocity.

This is my configuration for
ana_m3obc.h
ueast(j,k) = uwest(j,k) = 0.1 m/s
ana_m2obc=ana_fsobc = 0

Tracers
T = cte 17ºC
S = cte 35
For all domain in time

Conditions LBC


! W S E N
! e o a o
! s u s r
! t t t t
! h h
!
! 1 2 3 4

LBC(isFsur) == Cla Clo Cla Clo ! free-surface
LBC(isUbar) == Cla Clo Cla Clo ! 2D U-momentum
LBC(isVbar) == Cla Clo Cla Clo ! 2D V-momentum
LBC(isUvel) == Cla Clo Cla Clo ! 3D U-momentum
LBC(isVvel) == Cla Clo Cla Clo ! 3D V-momentum
LBC(isMtke) == Cla Clo Cla Clo ! mixing TKE

LBC(isTvar) == Cla Clo Cla Clo \ ! temClaature
Cla Clo Cla Clo ! salinity

I have 2 problems:

1 At time step = 0 we can see that the LBC at the 4 corners of the domain doesn't have the value that is expected to be.(green color)
The corners have the mean value between m3obc at the LBC and the next value in the interior grid point. (0.1+0)/2 = 0.5 m/s
See u_initial.jpg


2 The Clamped "profile" disapear at the second time step, beeing the same value that for the initial conditions = 0 m/s
See the uwestobc_t0.jpg and uwestobc_t1.jpg

So my questions are:
a) why this mismatch for the first problem? Is not posible to combine Clamped and Close conditions?
b) How I have to do it to make a Clamped profile at west and east side??

Any help would be appreciated

[kate]

I'm not sure exactly how the 2D and 3D momentum BCs play together, but if you want barotropic flow, you should be setting ana_m2obc to match that of ana_m3obc.

To see what is going on with the corners, the place to look is u3dbc_im.F:

Code: Select all

        IF (DOMAIN(ng)%SouthWest_Corner(tile)) THEN
          IF (LBC_apply(ng)%south(Istr  ).and.                          &   
     &        LBC_apply(ng)%west (Jstr-1)) THEN
            DO k=1,N(ng)
              u(Istr,Jstr-1,k,nout)=0.5_r8*(u(Istr+1,Jstr-1,k,nout)+    &   
     &                                      u(Istr  ,Jstr  ,k,nout))
            END DO
          END IF
        END IF

It looks like LBC_apply is True unless you are nesting, so you just get the average of the two nearest edge values.

[pelijas]

Thanks Kate,

I will take a look at that.