tilt in ssh in lim2 rheology

[gm]

The tilt of the sea surface in the rheology is better represented by the gradient of ssh rather that its approximation using geostrophic current as it is done currently in lim2. This change has already being done in LIM3.

Here we propose the modification same modification in LIM2.

This changes the ice thickness by improving the ice velocity.

The main modification is to replace the following lines:

            ! Computation of the velocity field taking into account the ice-ice interaction.                                 
            ! Terms that are independent of the ice velocity field.
            za1ct(ji,jj) = ztagnx - zcorl(ji,jj) * v_oce(ji,jj) - zgphsx
            za2ct(ji,jj) = ztagny + zcorl(ji,jj) * u_oce(ji,jj) - zgphsy

by

            ! Gradient of the sea surface height
            zgsshx =  (   (ssh_m(ji  ,jj  ) - ssh_m(ji-1,jj  ))/e1u(ji-1,jj  )   &
               &       +  (ssh_m(ji  ,jj-1) - ssh_m(ji-1,jj-1))/e1u(ji-1,jj-1)   ) * 0.5
            zgsshy =  (   (ssh_m(ji  ,jj  ) - ssh_m(ji  ,jj-1))/e2v(ji  ,jj-1)   &
               &       +  (ssh_m(ji-1,jj  ) - ssh_m(ji-1,jj-1))/e2v(ji-1,jj-1)   ) * 0.5

            ! Computation of the velocity field taking into account the ice-ice interaction.                                 
            ! Terms that are independent of the ice velocity field.
            za1ct(ji,jj) = ztagnx - zmass(ji,jj) * grav * zgsshx - zgphsx
            za2ct(ji,jj) = ztagny - zmass(ji,jj) * grav * zgsshy - zgphsy

and add in the header:

   USE sbc_oce        ! surface boundary condition: ocean variables

The resulting routine can be found in attachment. It has been tested in a 100y long run with ORCA2_LIM

Gurvan & Yevgeny

[gm]

limrhg_2.F90 for the v3.2 with the use of ssh_m

[nicolasmartin]

Milestone 2010 Stream 3: Developments & Implementation of new features deleted