| 1 |
! |
module divergf_m |
| 2 |
! $Header: /home/cvsroot/LMDZ4/libf/dyn3d/divergf.F,v 1.1.1.1 2004/05/19 12:53:05 lmdzadmin Exp $ |
|
| 3 |
! |
IMPLICIT NONE |
| 4 |
SUBROUTINE divergf(klevel,x,y,div) |
|
| 5 |
c |
contains |
| 6 |
c P. Le Van |
|
| 7 |
c |
SUBROUTINE divergf(klevel, x, y, div) |
| 8 |
c ********************************************************************* |
|
| 9 |
c ... calcule la divergence a tous les niveaux d'1 vecteur de compos. |
! From libf/dyn3d/divergf.F, version 1.1.1.1 2004/05/19 12:53:05 |
| 10 |
c x et y... |
|
| 11 |
c x et y etant des composantes covariantes ... |
! P. Le Van |
| 12 |
c ********************************************************************* |
|
| 13 |
use dimens_m |
! Calcule la divergence à tous les niveaux d'un vecteur de |
| 14 |
use paramet_m |
! composantes x et y. x et y sont des composantes covariantes. |
| 15 |
use comgeom |
|
| 16 |
IMPLICIT NONE |
USE comgeom, ONLY: apoln, apols, cuvsurcv_2d, cvusurcu_2d, unsaire_2d |
| 17 |
c |
USE dimens_m, ONLY: iim, jjm |
| 18 |
c x et y sont des arguments d'entree pour le s-prog |
USE filtreg_m, ONLY: filtreg |
| 19 |
c div est un argument de sortie pour le s-prog |
|
| 20 |
c |
INTEGER, intent(in):: klevel |
| 21 |
c |
REAL, intent(in):: x(iim + 1, jjm + 1, klevel), y(iim + 1, jjm, klevel) |
| 22 |
c --------------------------------------------------------------------- |
real, intent(out):: div(iim + 1, jjm + 1, klevel) ! in (unit of x, y) m-2 |
| 23 |
c |
|
| 24 |
c ATTENTION : pendant ce s-pg , ne pas toucher au COMMON/scratch/ . |
! Variables locales : |
| 25 |
c |
|
| 26 |
c --------------------------------------------------------------------- |
INTEGER l, i, j |
| 27 |
c |
|
| 28 |
c .......... variables en arguments ................... |
!------------------------------------------------------------ |
| 29 |
c |
|
| 30 |
INTEGER klevel |
DO l = 1, klevel |
| 31 |
REAL x( ip1jmp1,klevel ),y( ip1jm,klevel ),div( ip1jmp1,klevel ) |
forall (i = 2:iim + 1, j = 2:jjm) div(i, j, l) = cvusurcu_2d(i, j) & |
| 32 |
INTEGER l,ij |
* x(i, j, l) - cvusurcu_2d(i - 1, j) * x(i - 1, j , l) & |
| 33 |
c |
+ cuvsurcv_2d(i, j - 1) * y(i, j - 1, l) - cuvsurcv_2d(i, j) & |
| 34 |
c ............... variables locales ......................... |
* y(i, j, l) |
| 35 |
|
|
| 36 |
REAL aiy1( iip1 ) , aiy2( iip1 ) |
div(1, 2:jjm, l) = div(iim + 1, 2:jjm, l) |
| 37 |
REAL sumypn,sumyps |
|
| 38 |
c ................................................................... |
! Calcul aux pôles |
| 39 |
c |
div(:, 1, l) = - SUM(cuvsurcv_2d(:iim, 1) * y(:iim, 1, l)) / apoln |
| 40 |
REAL SSUM |
div(:, jjm + 1, l) = SUM(cuvsurcv_2d(:iim, jjm) * y(:iim, jjm, l)) & |
| 41 |
c |
/ apols |
| 42 |
c |
end DO |
| 43 |
DO 10 l = 1,klevel |
|
| 44 |
c |
CALL filtreg(div, direct = .true., intensive = .false.) |
| 45 |
DO ij = iip2, ip1jm - 1 |
|
| 46 |
div( ij + 1, l ) = |
DO l = 1, klevel |
| 47 |
* cvusurcu( ij+1 ) * x( ij+1,l ) - cvusurcu( ij ) * x( ij , l) + |
div(:, 2:jjm, l) = div(:, 2:jjm, l) * unsaire_2d(:, 2:jjm) |
| 48 |
* cuvsurcv(ij-iim) * y(ij-iim,l) - cuvsurcv(ij+1) * y(ij+1,l) |
ENDDO |
| 49 |
ENDDO |
|
| 50 |
c |
END SUBROUTINE divergf |
| 51 |
c .... correction pour div( 1,j,l) ...... |
|
| 52 |
c .... div(1,j,l)= div(iip1,j,l) .... |
end module divergf_m |
|
c |
|
|
CDIR$ IVDEP |
|
|
DO ij = iip2,ip1jm,iip1 |
|
|
div( ij,l ) = div( ij + iim,l ) |
|
|
ENDDO |
|
|
c |
|
|
c .... calcul aux poles ..... |
|
|
c |
|
|
DO ij = 1,iim |
|
|
aiy1(ij) = cuvsurcv( ij ) * y( ij , l ) |
|
|
aiy2(ij) = cuvsurcv( ij+ ip1jmi1 ) * y( ij+ ip1jmi1, l ) |
|
|
ENDDO |
|
|
sumypn = SSUM ( iim,aiy1,1 ) / apoln |
|
|
sumyps = SSUM ( iim,aiy2,1 ) / apols |
|
|
c |
|
|
DO ij = 1,iip1 |
|
|
div( ij , l ) = - sumypn |
|
|
div( ij + ip1jm, l ) = sumyps |
|
|
ENDDO |
|
|
10 CONTINUE |
|
|
c |
|
|
|
|
|
CALL filtreg( div, jjp1, klevel, 2, 2, .TRUE., 1 ) |
|
|
|
|
|
c |
|
|
DO l = 1, klevel |
|
|
DO ij = iip2,ip1jm |
|
|
div(ij,l) = div(ij,l) * unsaire(ij) |
|
|
ENDDO |
|
|
ENDDO |
|
|
c |
|
|
RETURN |
|
|
END |
|
Parent Directory
| 
Revision Log
| 
Patch