| 1 |
C $Header: /u/gcmpack/MITgcm/pkg/generic_advdiff/gad_dst2u1_adv_r.F,v 1.3 2006/06/18 23:31:35 jmc Exp $ |
| 2 |
C $Name: $ |
| 3 |
|
| 4 |
#include "GAD_OPTIONS.h" |
| 5 |
|
| 6 |
CBOP |
| 7 |
C !ROUTINE: GAD_DST2U1_ADV_R |
| 8 |
|
| 9 |
C !INTERFACE: ========================================================== |
| 10 |
SUBROUTINE GAD_DST2U1_ADV_R( |
| 11 |
I bi,bj,k, advectionScheme, |
| 12 |
I deltaTloc, rTrans, wFld, |
| 13 |
I tracer, |
| 14 |
O wT, |
| 15 |
I myThid ) |
| 16 |
|
| 17 |
C !DESCRIPTION: |
| 18 |
C Calculates the area integrated vertical flux due to advection |
| 19 |
C of a tracer using second-order Direct Space and Time (DST-2) |
| 20 |
C interpolation (=Lax-Wendroff) or simple 1rst order upwind scheme. |
| 21 |
|
| 22 |
C !USES: =============================================================== |
| 23 |
IMPLICIT NONE |
| 24 |
#include "SIZE.h" |
| 25 |
#include "GRID.h" |
| 26 |
#include "GAD.h" |
| 27 |
|
| 28 |
C !INPUT PARAMETERS: =================================================== |
| 29 |
C bi,bj :: tile indices |
| 30 |
C k :: vertical level |
| 31 |
C advectionScheme :: advection scheme to use: either 2nd Order DST |
| 32 |
C or 1rst Order Upwind |
| 33 |
C deltaTloc :: local time-step (s) |
| 34 |
C rTrans :: vertical volume transport |
| 35 |
C wFld :: vertical flow |
| 36 |
C tracer :: tracer field |
| 37 |
C myThid :: thread number |
| 38 |
INTEGER bi,bj,k |
| 39 |
INTEGER advectionScheme |
| 40 |
_RL deltaTloc |
| 41 |
_RL rTrans(1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 42 |
_RL wFld (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 43 |
_RL tracer(1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 44 |
INTEGER myThid |
| 45 |
|
| 46 |
C !OUTPUT PARAMETERS: ================================================== |
| 47 |
C wT :: vertical advective flux |
| 48 |
_RL wT (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 49 |
|
| 50 |
C !LOCAL VARIABLES: ==================================================== |
| 51 |
C i,j :: loop indices |
| 52 |
C km1 :: =max( k-1 , 1 ) |
| 53 |
C rLimit :: centered (vs upwind) fraction |
| 54 |
C wLoc :: velocity, vertical component |
| 55 |
C wCFL :: Courant-Friedrich-Levy number |
| 56 |
INTEGER i,j,km1 |
| 57 |
_RL wLoc, wCFL, rLimit, wAbs |
| 58 |
CEOP |
| 59 |
|
| 60 |
rLimit = 0. _d 0 |
| 61 |
IF ( advectionScheme.EQ.ENUM_DST2 ) rLimit = 1. _d 0 |
| 62 |
|
| 63 |
km1=MAX(1,k-1) |
| 64 |
|
| 65 |
IF ( k.LE.1 .OR. k.GT.Nr) THEN |
| 66 |
DO j=1-Oly,sNy+Oly |
| 67 |
DO i=1-Olx,sNx+Olx |
| 68 |
wT(i,j) = 0. |
| 69 |
ENDDO |
| 70 |
ENDDO |
| 71 |
ELSE |
| 72 |
DO j=1-Oly,sNy+Oly |
| 73 |
DO i=1-Olx,sNx+Olx |
| 74 |
|
| 75 |
wLoc = wFld(i,j) |
| 76 |
c wLoc = rTrans(i,j)*recip_rA(i,j,bi,bj) |
| 77 |
wCFL = ABS(wLoc*deltaTloc*recip_drC(k)) |
| 78 |
|
| 79 |
c wT(i,j) = maskC(i,j,km1,bi,bj)*( |
| 80 |
c & rTrans(i,j)*(tracer(i,j,km1)+tracer(i,j,k))*0.5 _d 0 |
| 81 |
c & + ( 1. _d 0 - rLimit*(1. _d 0 - wCFL) )*ABS(rTrans(i,j)) |
| 82 |
c & *(tracer(i,j,km1)-tracer(i,j,k))*0.5 _d 0*rkSign |
| 83 |
c & ) |
| 84 |
C-- above formulation produces large truncation error when: |
| 85 |
C 1rst.O upWind and w*rkSign > 0 & |tracer(k-1)| << |tracer(k)| |
| 86 |
C or w*rkSign < 0 & |tracer(k-1)| >> |tracer(k)| |
| 87 |
C-- change to a more robust expression: |
| 88 |
wAbs = ABS(rTrans(i,j))*rkSign |
| 89 |
& *( 1. _d 0 - rLimit*(1. _d 0 - wCFL) ) |
| 90 |
wT(i,j) = maskC(i,j,km1,bi,bj)*( |
| 91 |
& ( rTrans(i,j)+wAbs )* 0.5 _d 0 * tracer(i,j,km1) |
| 92 |
& + ( rTrans(i,j)-wAbs )* 0.5 _d 0 * tracer(i,j,k) |
| 93 |
& ) |
| 94 |
ENDDO |
| 95 |
ENDDO |
| 96 |
ENDIF |
| 97 |
|
| 98 |
RETURN |
| 99 |
END |
Parent Directory
| 
Revision Log
| 
Revision Graph