!-------------------------------------- LICENCE BEGIN ------------------------------------
!Environment Canada - Atmospheric Science and Technology License/Disclaimer,
! version 3; Last Modified: May 7, 2008.
!This is free but copyrighted software; you can use/redistribute/modify it under the terms
!of the Environment Canada - Atmospheric Science and Technology License/Disclaimer
!version 3 or (at your option) any later version that should be found at:
!http://collaboration.cmc.ec.gc.ca/science/rpn.comm/license.html
!
!This software is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY;
!without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
!See the above mentioned License/Disclaimer for more details.
!You should have received a copy of the License/Disclaimer along with this software;
!if not, you can write to: EC-RPN COMM Group, 2121 TransCanada, suite 500, Dorval (Quebec),
!CANADA, H9P 1J3; or send e-mail to service.rpn@ec.gc.ca
!-------------------------------------- LICENCE END --------------------------------------
***s/r adw_main_1_wnd_ad - ADJ of adw_main_1_wnd
*
#include "model_macros_f.h"
*
subroutine adw_main_1_wnd_ad ( F_u, F_v, F_w, F_nit, F_njt, F_nk ) 1,4
*
implicit none
*
integer F_nit, F_njt, F_nk
*
real F_u(F_nit,F_njt,F_nk),
% F_v(F_nit,F_njt,F_nk),
% F_w(F_nit,F_njt,F_nk)
*
*author
* monique tanguay
*
*revision
* v2_31 - Tanguay M. - initial MPI version
* v3_00 - Tanguay M. - adapt to restructured adw_main
* v3_11 - Tanguay M. - AIXport+Opti+OpenMP for TLM-ADJ
* v3_21 - Tanguay M. - Revision Openmp
* v3_30 - Tanguay M. - Validation for LAM version
*
*language
* fortran 77
*
*object
* see id section
*
*ADJ of
*arguments
*______________________________________________________________________
* | | |
* NAME | DESCRIPTION | I/O |
*--------------|-------------------------------------------------|-----|
* F_u | u components of winds on advection grid | o |
* F_v | v components of winds on advection grid | o |
* F_w | w components of winds on advection grid | o |
* | | |
* F_nit | \ total number of points in x,y direction in | i |
* F_njt | / advection grid (including halos) | i |
* | | |
* F_nk | number of levels | i |
*______________|_________________________________________________|_____|
*
*implicits
#include "glb_ld.cdk"
#include "glb_pil.cdk"
#include "adw.cdk"
#include "lun.cdk"
#include "inuvl.cdk"
#include "vth.cdk"
*
*modules
integer vmmlod, vmmget, vmmuld
external vmmlod, vmmget, vmmuld
************************************************************************
integer pnerr, pnlkey1(30), pnlod
*
integer i, j, k, i0, in, j0, jn, n, dest_ni
real*8 ONE_8,sc_8,ZERO_8
parameter( ONE_8 = 1.0 )
parameter( ZERO_8 = 0.0 )
*
real, dimension(:,:,:,:) ,allocatable :: in0
*
************************************************************************
if (Lun_debug_L) write (Lun_out,1000)
*
if (G_lam) allocate ( in0(LDIST_SHAPE,l_nk,3) )
*
pnlkey1(1) = VMM_KEY(uth)
pnlkey1(2) = VMM_KEY(vth)
pnlkey1(3) = VMM_KEY(psdth)
*
pnerr = vmmlod(pnlkey1,3)
*
pnerr = VMM_GET_VAR(uth)
pnerr = VMM_GET_VAR(vth)
pnerr = VMM_GET_VAR(psdth)
*
!$omp parallel private(i0,in,j0,jn,sc_8)
*
***********************************************************************
* ADJ of
* Adjust wind fields to advection grid
***********************************************************************
if (.not.G_lam) then
*
if ( l_north ) then
*
call adw_pols_ad (F_w,Adw_wx_8,Adw_njc+1,Adw_nic,Adw_halox,
% Adw_njc,Adw_haloy,l_nk)
*
call adw_polw_ad (F_u,F_v,Adw_cx_8,Adw_sx_8,Adw_wx_8,
% Adw_sy_8,Adw_njc+1,Adw_nic,Adw_halox,Adw_njc,Adw_haloy,l_nk)
*
endif
*
if ( l_south ) then
*
call adw_pols_ad (F_w,Adw_wx_8,0,Adw_nic,Adw_halox,Adw_njc,
% Adw_haloy,l_nk)
*
call adw_polw_ad (F_u,F_v,Adw_cx_8,Adw_sx_8,Adw_wx_8,
% Adw_sy_8,0,Adw_nic,Adw_halox,Adw_njc,Adw_haloy,l_nk)
*
endif
*
endif
*
if (G_lam) then
n=0
dest_ni=l_ni
else
n=999
dest_ni=G_ni
endif
*
if (G_lam) then
!$omp do
do k=1,l_nk
do j=l_miny,l_maxy
do i=l_minx,l_maxx
in0(i,j,k,1) = psdth(i,j,k)
in0(i,j,k,2) = vth(i,j,k)
in0(i,j,k,3) = uth(i,j,k)
end do
end do
end do
!$omp end do
endif
*
!$omp single
call rpn_comm_adj_halox ( psdth, LDIST_DIM, l_ni, l_nj, l_nk,
% Adw_halox, Adw_haloy, G_periodx, G_periody, F_w, 1-Adw_halox,
% Adw_nic+Adw_halox, 1-Adw_haloy, Adw_njc+Adw_haloy, dest_ni, n)
*
call rpn_comm_adj_halox ( vth, LDIST_DIM, l_ni, l_nj, l_nk,
% Adw_halox, Adw_haloy, G_periodx, G_periody, F_v, 1-Adw_halox,
% Adw_nic+Adw_halox, 1-Adw_haloy, Adw_njc+Adw_haloy, dest_ni, n)
*
call rpn_comm_adj_halox ( uth, LDIST_DIM, l_ni, l_nj, l_nk,
% Adw_halox, Adw_haloy, G_periodx, G_periody, F_u, 1-Adw_halox,
% Adw_nic+Adw_halox, 1-Adw_haloy, Adw_njc+Adw_haloy, dest_ni, n)
!$omp end single
*
if (G_lam) then
!$omp do
do k=1,l_nk
do j=l_miny,l_maxy
do i=l_minx,l_maxx
psdth(i,j,k) = in0(i,j,k,1) + psdth(i,j,k)
vth(i,j,k) = in0(i,j,k,2) + vth(i,j,k)
uth(i,j,k) = in0(i,j,k,3) + uth(i,j,k)
C in0(i,j,k,1) = ZERO_8
C in0(i,j,k,2) = ZERO_8
C in0(i,j,k,3) = ZERO_8
end do
end do
end do
!$omp end do
endif
*
* Zero adjoint variables
* ---------------------
!$omp do
do k = 1, F_nk
do j = 1, F_njt
do i = 1, F_nit
F_w(i,j,k) = ZERO_8
F_v(i,j,k) = ZERO_8
F_u(i,j,k) = ZERO_8
enddo
enddo
enddo
!$omp enddo
*
***********************************************************************
* ADJ of
* Image to component
***********************************************************************
i0=1
in=l_ni
j0=1
jn=l_nj
if (G_lam) then
if (l_west) i0 = 3
if (l_east) in = l_niu - 1
endif
*
!$omp do
do k = 1,l_nk
do j = j0,jn
*
sc_8 = ONE_8/Adw_cy_8(j)
*
do i = i0,in
*
F_u(i,j,k) = sc_8 * uth(i,j,k) + F_u(i,j,k)
uth(i,j,k) = ZERO_8
*
enddo
enddo
enddo
!$omp enddo
*
i0=1
in=l_ni
j0=1
jn=l_nj
if (G_lam) then
if (l_south) j0 = 3
if (l_north) jn = l_njv - 1
endif
*
!$omp do
do k = 1,l_nk
do j = j0,jn
*
sc_8 = ONE_8/Adw_cy_8(j)
*
do i = i0,in
*
F_v(i,j,k) = sc_8 * vth(i,j,k) + F_v(i,j,k)
vth(i,j,k) = ZERO_8
*
enddo
enddo
enddo
!$omp enddo
*
***********************************************************************
* ADJ of
* Interpolate advection winds to geopotential grid
* F_u and F_v are used as work space
***********************************************************************
!$omp do
do k=1,l_nk
*
i0 = 1
in = l_ni
j0 = 1
jn = l_njv
if (l_south) j0 = 3
if (l_north) jn = l_njv - 1
*
if (.not.G_lam) then
*
if (l_north) then
do i = i0, in
vth(i,jn+1,k) = inuvl_wyvy3_8(jn+1,3) * F_v(i,jn+1,k) + vth(i,jn+1,k)
vth(i,jn ,k) = inuvl_wyvy3_8(jn+1,2) * F_v(i,jn+1,k) + vth(i,jn ,k)
vth(i,jn-1,k) = inuvl_wyvy3_8(jn+1,1) * F_v(i,jn+1,k) + vth(i,jn-1,k)
F_v(i,jn+1,k) = ZERO_8
*
vth (i,jn+1,k)= inuvl_wyvy3_8(jn+2,2) * F_v(i,jn+2,k) + vth(i,jn+1,k)
vth (i,jn ,k)= inuvl_wyvy3_8(jn+2,1) * F_v(i,jn+2,k) + vth(i,jn ,k)
F_v(i,jn+2,k) = ZERO_8
end do
endif
if (l_south) then
do i = i0, in
vth(i,j0, k) = inuvl_wyvy3_8(j0-1,4) * F_v(i,j0-1,k) + vth(i,j0, k)
vth(i,j0-1,k) = inuvl_wyvy3_8(j0-1,3) * F_v(i,j0-1,k) + vth(i,j0-1,k)
vth(i,j0-2,k) = inuvl_wyvy3_8(j0-1,2) * F_v(i,j0-1,k) + vth(i,j0-2,k)
F_v(i,j0-1,k) = ZERO_8
*
vth(i,j0-1,k) = inuvl_wyvy3_8(j0-2,4) * F_v(i,j0-2,k) + vth(i,j0-1,k)
vth(i,j0-2,k) = inuvl_wyvy3_8(j0-2,3) * F_v(i,j0-2,k) + vth(i,j0-2,k)
F_v(i,j0-2,k) = ZERO_8
end do
endif
*
endif
*
do j = jn, j0, -1
do i = in, i0, -1
vth(i,j+1,k) = inuvl_wyvy3_8(j,4) * F_v(i,j,k) + vth(i,j+1,k)
vth(i,j ,k) = inuvl_wyvy3_8(j,3) * F_v(i,j,k) + vth(i,j ,k)
vth(i,j-1,k) = inuvl_wyvy3_8(j,2) * F_v(i,j,k) + vth(i,j-1,k)
vth(i,j-2,k) = inuvl_wyvy3_8(j,1) * F_v(i,j,k) + vth(i,j-2,k)
F_v(i,j, k) = ZERO_8
end do
end do
*
i0 = 1
in = l_niu
j0 = 1
jn = l_nj
if (G_lam) then
if (l_west) i0 = 3
if (l_east) in = l_niu - 1
endif
*
do j = jn,j0,-1
do i = in,i0,-1
uth(i+1,j,k) = inuvl_wxux3_8(i,4) * F_u(i,j,k) + uth(i+1,j,k)
uth(i ,j,k) = inuvl_wxux3_8(i,3) * F_u(i,j,k) + uth(i ,j,k)
uth(i-1,j,k) = inuvl_wxux3_8(i,2) * F_u(i,j,k) + uth(i-1,j,k)
uth(i-2,j,k) = inuvl_wxux3_8(i,1) * F_u(i,j,k) + uth(i-2,j,k)
F_u(i, j,k) = ZERO_8
end do
end do
*
end do
!$omp enddo
*
!$omp single
call rpn_comm_adj_halo (vth,LDIST_DIM,l_ni,l_njv,G_nk,
% G_halox,G_haloy,G_periodx,G_periody,l_ni,0)
!$omp end single
*
* Zero vth halo
* -------------
!$omp do
do k=1,G_nk
do j=l_miny,0
do i=l_minx,l_maxx
vth(i,j,k) = ZERO_8
end do
end do
do j=l_njv+1,l_maxy
do i=l_minx ,l_maxx
vth(i,j,k) = ZERO_8
end do
end do
do i=l_minx,0
do j=l_miny,l_maxy
vth(i,j,k) = ZERO_8
end do
end do
do i=l_ni+1,l_maxx
do j=l_miny,l_maxy
vth(i,j,k) = ZERO_8
end do
end do
end do
!$omp enddo
*
!$omp single
call rpn_comm_adj_halo (uth,LDIST_DIM,l_niu,l_nj, G_nk,
% G_halox,G_haloy,G_periodx,G_periody,l_ni,0)
!$omp end single
*
* Zero uth halo
* -------------
!$omp do
do k=1,G_nk
do j=l_miny,0
do i=l_minx,l_maxx
uth(i,j,k) = ZERO_8
end do
end do
do j=l_nj+1,l_maxy
do i=l_minx,l_maxx
uth(i,j,k) = ZERO_8
end do
end do
do i=l_minx,0
do j=l_miny,l_maxy
uth(i,j,k) = ZERO_8
end do
end do
do i=l_niu+1,l_maxx
do j=l_miny ,l_maxy
uth(i,j,k) = ZERO_8
end do
end do
end do
!$omp enddo
*
!$omp end parallel
*
pnerr = vmmuld(-1,0)
*
if (G_lam) deallocate ( in0 )
*
1000 format(3X,'ADJ of PREPARE WINDS: (S/R ADW_MAIN_1_WND_AD)')
return
end