Here, the local continuity equation is calculated in order to obtain the grid-related vertical velocity . An layer-integrated equation for this quantity is given as equation (25) which has been derived from the differential formulation (3).
Since the kinematic boundary condition must hold (and is used for the derivation of (25)), the grid-related vertical velocity at the surface muzst be zero, i.e. . This is a good consistence check for the mode splitting, since this is only fulfilled if the vertically integrated continuity equation (which is the sea surface elevation equation calculated on the micro time step) and this local continuity equation are compatible.
The physical vertical velocity is then recalculated from the grid-related vertical velocity by means of (32), ... which should soon be coded in the routine tow in the directory futils. USES:
use domain, only: imin,imax,jmin,jmax,kmax #if defined(SPHERICAL) || defined(CURVILINEAR) use domain, only: arcd1,dxv,dyu #else use domain, only: dx,dy,ard1 #endif use variables_3d, only: dt,kmin,uu,vv,ww,ho,hn #define CALC_HALO_WW #ifndef CALC_HALO_WW use domain, only: az use halo_zones, only: update_3d_halo,wait_halo,z_TAG #endif use getm_timers, only: tic, toc, TIM_WWMOMENTUM, TIM_WWMOMENTUMH $ use omp_lib IMPLICIT NONEREVISION HISTORY:
Original author(s): Hans Burchard & Karsten BoldingLOCAL VARIABLES:
REALTYPE :: dtm1 integer :: i,j,k