| 1 |
molod |
1.1 |
subroutine step_fizhi_corr (myTime, myIter, myThid) |
| 2 |
|
|
c---------------------------------------------------------------------- |
| 3 |
|
|
c Subroutine step_fizhi_corr - 'Wrapper' routine to advance |
| 4 |
|
|
c the physics state and make the new value. |
| 5 |
|
|
c At this point, increment with the "correction term" |
| 6 |
|
|
c which includes the dynamics tendency and the integral |
| 7 |
|
|
c constraint to enforce agreement with the dynamics state |
| 8 |
|
|
c Also: Set up "bi, bj loop" and some timers and clocks here. |
| 9 |
|
|
c |
| 10 |
|
|
c Call:phys2dyn (4) (interpolate physics state to dynamics grid |
| 11 |
|
|
c for use in the correction terms) |
| 12 |
|
|
c AtoC (convert physics state on dynamics grid to C-Grid) |
| 13 |
|
|
c CtoA (convert correction term on dynamics grid to A-Grid) |
| 14 |
|
|
c dyn2phys (4) (interpolate A-Grid correction term to physics grid) |
| 15 |
|
|
c step_physics (advance physics state by correction term) |
| 16 |
|
|
c----------------------------------------------------------------------- |
| 17 |
|
|
implicit none |
| 18 |
|
|
#include "CPP_OPTIONS.h" |
| 19 |
|
|
#include "SIZE.h" |
| 20 |
|
|
#include "GRID.h" |
| 21 |
|
|
#include "fizhi_SIZE.h" |
| 22 |
|
|
#include "land_SIZE.h" |
| 23 |
|
|
#include "DYNVARS.h" |
| 24 |
|
|
#include "fizhi_coms.h" |
| 25 |
|
|
#include "gridalt_mapping.h" |
| 26 |
|
|
#include "EEPARAMS.h" |
| 27 |
|
|
#include "SURFACE.h" |
| 28 |
|
|
|
| 29 |
|
|
integer myTime, myIter, myThid |
| 30 |
|
|
|
| 31 |
|
|
c pe on dynamics and physics grid refers to bottom edge |
| 32 |
|
|
_RL pephy(1-OLx:sNx+Olx,1-Oly:sNy+Oly,Nrphys+1,nSx,nSy) |
| 33 |
|
|
_RL pedyn(1-Olx:sNx+Olx,1-Oly:sNy+Oly,Nr+1,nSx,nSy) |
| 34 |
|
|
_RL windphy(1-Olx:sNx+Olx,1-Oly:sNy+Oly,Nrphys,nSx,nSy) |
| 35 |
|
|
_RL udyntemp(1-Olx:sNx+Olx,1-Oly:sNy+Oly,Nr,nSx,nSy) |
| 36 |
|
|
_RL vdyntemp(1-Olx:sNx+Olx,1-Oly:sNy+Oly,Nr,nSx,nSy) |
| 37 |
|
|
_RL thdyntemp(1-Olx:sNx+Olx,1-Oly:sNy+Oly,Nr,nSx,nSy) |
| 38 |
|
|
_RL sdyntemp(1-Olx:sNx+Olx,1-Oly:sNy+Oly,Nrphys,nSx,nSy) |
| 39 |
|
|
_RL uphytemp(1-Olx:sNx+Olx,1-Oly:sNy+Oly,Nrphys,nSx,nSy) |
| 40 |
|
|
_RL vphytemp(1-Olx:sNx+Olx,1-Oly:sNy+Oly,Nrphys,nSx,nSy) |
| 41 |
|
|
_RL thphytemp(1-Olx:sNx+Olx,1-Oly:sNy+Oly,Nrphys,nSx,nSy) |
| 42 |
|
|
_RL sphytemp(1-Olx:sNx+Olx,1-Oly:sNy+Oly,Nrphys,nSx,nSy) |
| 43 |
|
|
|
| 44 |
|
|
integer i, j, L, Lbotij, bi, bj |
| 45 |
|
|
integer im1, im2, jm1, jm2, idim1, idim2, jdim1, jdim2 |
| 46 |
molod |
1.2 |
_RL dt |
| 47 |
molod |
1.1 |
|
| 48 |
|
|
im1 = 1-OLx |
| 49 |
|
|
im2 = sNx+OLx |
| 50 |
|
|
jm1 = 1-OLy |
| 51 |
|
|
jm2 = sNy+OLy |
| 52 |
|
|
idim1 = 1 |
| 53 |
|
|
idim2 = sNx |
| 54 |
|
|
jdim1 = 1 |
| 55 |
|
|
jdim2 = sNy |
| 56 |
molod |
1.2 |
dt = 1. |
| 57 |
molod |
1.1 |
|
| 58 |
|
|
do bj = myByLo(myThid), myByHi(myThid) |
| 59 |
|
|
do bi = myBxLo(myThid), myBxHi(myThid) |
| 60 |
|
|
|
| 61 |
|
|
C Construct Pressures on physics and dynamics grids |
| 62 |
|
|
do j = 1,sNy |
| 63 |
|
|
do i = 1,sNx |
| 64 |
|
|
do L = 1,Nr |
| 65 |
|
|
pedyn(i,j,L,bi,bj) = 0. |
| 66 |
|
|
enddo |
| 67 |
|
|
enddo |
| 68 |
|
|
enddo |
| 69 |
|
|
do j = 1,sNy |
| 70 |
|
|
do i = 1,sNx |
| 71 |
|
|
Lbotij = ksurfC(i,j,bi,bj) |
| 72 |
|
|
if(Lbotij.ne.0.) |
| 73 |
|
|
. pedyn(i,j,Lbotij,bi,bj) = Ro_surf(i,j,bi,bj) + etaH(i,j,bi,bj) |
| 74 |
|
|
enddo |
| 75 |
|
|
enddo |
| 76 |
|
|
do j = 1,sNy |
| 77 |
|
|
do i = 1,sNx |
| 78 |
|
|
Lbotij = ksurfC(i,j,bi,bj) |
| 79 |
|
|
do L = Lbotij+1,Nr+1 |
| 80 |
|
|
pedyn(i,j,L,bi,bj) = pedyn(i,j,L-1,bi,bj) - |
| 81 |
|
|
. drF(L-1)*hfacC(i,j,L-1,bi,bj) |
| 82 |
|
|
enddo |
| 83 |
|
|
c Do not use a zero field as the top edge pressure for interpolation |
| 84 |
|
|
if(pedyn(i,j,Nr+1,bi,bj).lt.1.e-5) |
| 85 |
|
|
. pedyn(i,j,Nr+1,bi,bj) = 1.e-5 |
| 86 |
|
|
enddo |
| 87 |
|
|
enddo |
| 88 |
|
|
|
| 89 |
|
|
do j = 1,sNy |
| 90 |
|
|
do i = 1,sNx |
| 91 |
|
|
pephy(i,j,1,bi,bj)=Ro_surf(i,j,bi,bj) + etaH(i,j,bi,bj) |
| 92 |
|
|
do L = 2,Nrphys+1 |
| 93 |
|
|
pephy(i,j,L,bi,bj)=pephy(i,j,L-1,bi,bj)-dpphys(i,j,L-1,bi,bj) |
| 94 |
|
|
enddo |
| 95 |
|
|
c Do not use a zero field as the top edge pressure for interpolation |
| 96 |
|
|
if(pephy(i,j,Nrphys+1,bi,bj).lt.1.e-5) |
| 97 |
|
|
. pephy(i,j,Nrphys+1,bi,bj) = 1.e-5 |
| 98 |
|
|
enddo |
| 99 |
|
|
enddo |
| 100 |
|
|
c |
| 101 |
|
|
c Create a wind magnitude field on the physics grid - |
| 102 |
|
|
do L = 1,Nrphys |
| 103 |
|
|
do j = 1,sNy |
| 104 |
|
|
do i = 1,sNx |
| 105 |
|
|
windphy(i,j,L,bi,bj) = sqrt(uphy(i,j,L,bi,bj)*uphy(i,j,L,bi,bj) |
| 106 |
|
|
. + vphy(i,j,L,bi,bj)*vphy(i,j,L,bi,bj)) |
| 107 |
|
|
enddo |
| 108 |
|
|
enddo |
| 109 |
|
|
enddo |
| 110 |
|
|
|
| 111 |
|
|
c Compute correction term (new dyn state-phys state to dyn) on physics grid: |
| 112 |
|
|
c First: interp physics state to dynamics grid |
| 113 |
|
|
call phys2dyn(uphy,pephy,im1,im2,jm1,jm2,Nrphys,Nsx,Nsy, |
| 114 |
|
|
. 1,sNx,1,sNy,bi,bj,pedyn,ksurfC,Nr,nlperdyn,udyntemp) |
| 115 |
|
|
call phys2dyn(vphy,pephy,im1,im2,jm1,jm2,Nrphys,Nsx,Nsy, |
| 116 |
|
|
. 1,sNx,1,sNy,bi,bj,pedyn,ksurfC,Nr,nlperdyn,vdyntemp) |
| 117 |
|
|
call phys2dyn(thphy,pephy,im1,im2,jm1,jm2,Nrphys,Nsx,Nsy, |
| 118 |
|
|
. 1,sNx,1,sNy,bi,bj,pedyn,ksurfC,Nr,nlperdyn,thdyntemp) |
| 119 |
|
|
call phys2dyn(sphy,pephy,im1,im2,jm1,jm2,Nrphys,Nsx,Nsy, |
| 120 |
|
|
. 1,sNx,1,sNy,bi,bj,pedyn,ksurfC,Nr,nlperdyn,sdyntemp) |
| 121 |
|
|
|
| 122 |
|
|
enddo |
| 123 |
|
|
enddo |
| 124 |
|
|
|
| 125 |
|
|
c Second: Convert physics state on dynamics grid to C-Grid |
| 126 |
|
|
|
| 127 |
|
|
call AtoC(myThid,udyntemp,vdyntemp,maskC,im1,im2,jm1,jm2,Nr, |
| 128 |
|
|
. Nsx,Nsy,1,sNx,1,sNy,udyntemp,vdyntemp) |
| 129 |
|
|
|
| 130 |
|
|
c Third: Subtract Phys state on dyn. grid from new dynamics state |
| 131 |
|
|
do bj = myByLo(myThid), myByHi(myThid) |
| 132 |
|
|
do bi = myBxLo(myThid), myBxHi(myThid) |
| 133 |
|
|
|
| 134 |
|
|
do L = 1,Nr |
| 135 |
|
|
do j = jdim1,jdim2 |
| 136 |
|
|
do i = idim1,idim2 |
| 137 |
|
|
udyntemp(i,j,L,bi,bj)=uvel(i,j,L,bi,bj)-udyntemp(i,j,L,bi,bj) |
| 138 |
|
|
vdyntemp(i,j,L,bi,bj)=vvel(i,j,L,bi,bj)-vdyntemp(i,j,L,bi,bj) |
| 139 |
|
|
thdyntemp(i,j,L,bi,bj)=theta(i,j,L,bi,bj)-thdyntemp(i,j,L,bi,bj) |
| 140 |
|
|
sdyntemp(i,j,L,bi,bj)=salt(i,j,L,bi,bj)-sdyntemp(i,j,L,bi,bj) |
| 141 |
|
|
enddo |
| 142 |
|
|
enddo |
| 143 |
|
|
enddo |
| 144 |
|
|
|
| 145 |
|
|
enddo |
| 146 |
|
|
enddo |
| 147 |
|
|
|
| 148 |
|
|
c Fourth: Convert correction terms to A-Grid |
| 149 |
|
|
call CtoA(myThid,udyntemp,vdyntemp,maskW,maskS,im1,im2,jm1,jm2, |
| 150 |
|
|
. Nr,Nsx,Nsy,1,sNx,1,sNy,udyntemp,vdyntemp) |
| 151 |
|
|
|
| 152 |
|
|
c Fifth: Interpolate correction terms to physics grid |
| 153 |
|
|
do bj = myByLo(myThid), myByHi(myThid) |
| 154 |
|
|
do bi = myBxLo(myThid), myBxHi(myThid) |
| 155 |
|
|
|
| 156 |
|
|
call dyn2phys(udyntemp,pedyn,im1,im2,jm1,jm2,Nr,Nsx,Nsy,1,sNx, |
| 157 |
molod |
1.2 |
. 1,sNy,bi,bj,windphy,pephy,ksurfC,Nrphys,nlperdyn,0,uphytemp) |
| 158 |
molod |
1.1 |
call dyn2phys(vdyntemp,pedyn,im1,im2,jm1,jm2,Nr,Nsx,Nsy,1,sNx, |
| 159 |
molod |
1.2 |
. 1,sNy,bi,bj,windphy,pephy,ksurfC,Nrphys,nlperdyn,0,vphytemp) |
| 160 |
molod |
1.1 |
call dyn2phys(thdyntemp,pedyn,im1,im2,jm1,jm2,Nr,Nsx,Nsy,1,sNx, |
| 161 |
|
|
. 1,sNy,bi,bj,windphy,pephy,ksurfC,Nrphys,nlperdyn,0,thphytemp) |
| 162 |
|
|
call dyn2phys(sdyntemp,pedyn,im1,im2,jm1,jm2,Nr,Nsx,Nsy,1,sNx, |
| 163 |
|
|
. 1,sNy,bi,bj,windphy,pephy,ksurfC,Nrphys,nlperdyn,0,sphytemp) |
| 164 |
molod |
1.2 |
|
| 165 |
molod |
1.1 |
c Last: Increment physics state by the correction term |
| 166 |
molod |
1.2 |
call step_physics(uphy,vphy,thphy,sphy,dt,im1,im2,jm1,jm2, |
| 167 |
|
|
. Nrphys,Nsx,Nsy,1,sNx,1,sNy,bi,bj, |
| 168 |
|
|
. uphytemp,vphytemp,thphytemp,sphytemp) |
| 169 |
molod |
1.1 |
|
| 170 |
|
|
enddo |
| 171 |
|
|
enddo |
| 172 |
|
|
|
| 173 |
|
|
return |
| 174 |
|
|
end |
Parent Directory
| 
Revision Log
| 
Revision Graph