no/partial/strong slip condition is actually free slip in one-point straits

[davestorkey]

BE CAREFUL !!! Due to dynamic behaviour of this ticket creation page, it is highly recommend to set first all other fields before writing the ticket description below.
If you have lost your draft after an unwanted reload, you can click on the link 'Restore Form' in the contextual menu upper right to recover it.
Remove these lines after reading.

Context

Analysis

The boundary condition on velocities tangential to topography is set via the value of fmask initialised in dom_msk as follows (NEMO 4 code):

Calculate from tmask:

               fmask(ji,jj,jk) = tmask(ji,jj  ,jk) * tmask(ji+1,jj  ,jk)   &
                  &            * tmask(ji,jj+1,jk) * tmask(ji+1,jj+1,jk)

Apply slip condition:

      IF( rn_shlat /= 0 ) THEN      ! Not free-slip lateral boundary condition
         !
         ALLOCATE( zwf(jpi,jpj) )
         !
         DO jk = 1, jpk
            zwf( : , : ) = fmask(:,:,jk)         
            DO jj = 2, jpjm1
               DO ji = fs_2, fs_jpim1   ! vector opt.
                  IF( fmask(ji,jj,jk) == 0._wp ) THEN
                     fmask(ji,jj,jk) = rn_shlat * MIN( 1._wp , MAX( zwf(ji+1,jj), zwf(ji,jj+1),   &
                        &                                           zwf(ji-1,jj), zwf(ji,jj-1)  )  )
                  ENDIF
               END DO
            END DO

This selects fmask points at the topography-sea boundary by looking for points where fmask is zero but has a neighbouring fmask which is one. But this criterion doesn't work for single-point channels. A better criterion would be to check the neighbouring umask and vmask values:

Instead of:

MAX( zwf(ji+1,jj), zwf(ji,jj+1),zwf(ji-1,jj), zwf(ji,jj-1) )

this would be better:

MAX( umask(ji,jj,jk), umask(ji,jj+1,jk), vmask(ji,jj,jk), vmask(ji+1,jj,jk) )

Maybe this was deliberate? But if you want no-slip with free-slip in channels it is probably better to do that explicitly via a spatially varying shlat parameter.

Fix

...

[davestorkey]

In 10648:

Error: Failed to load processor CommitTicketReference

No macro or processor named 'CommitTicketReference' found

[gm]

Yes Dave, I agree. I never though of having a at least 2 ocean grid point channel since I don't think its a good idea to have such a channel.
Nevertheless, it may occur in a user configuration. Your solution is the one to be implemented.

Gurvan

Note that using usrdef_fmask you can always set the boundary condition you want.

[clem]

In 13270:

Error: Failed to load processor CommitTicketReference

No macro or processor named 'CommitTicketReference' found

[clem]

In 13271:

Error: Failed to load processor CommitTicketReference

No macro or processor named 'CommitTicketReference' found

[clem]

fixed also in the trunk from r13472