| 1 |
adcroft |
1.28 |
C $Header: /u/gcmpack/models/MITgcmUV/model/src/calc_gs.F,v 1.27.2.4 2001/04/09 18:52:45 adcroft Exp $ |
| 2 |
cnh |
1.25 |
C $Name: $ |
| 3 |
cnh |
1.1 |
|
| 4 |
cnh |
1.17 |
#include "CPP_OPTIONS.h" |
| 5 |
cnh |
1.1 |
|
| 6 |
adcroft |
1.28 |
#define COSINEMETH_III |
| 7 |
|
|
#undef ISOTROPIC_COS_SCALING |
| 8 |
|
|
|
| 9 |
cnh |
1.1 |
CStartOfInterFace |
| 10 |
|
|
SUBROUTINE CALC_GS( |
| 11 |
|
|
I bi,bj,iMin,iMax,jMin,jMax,k,kM1,kUp,kDown, |
| 12 |
adcroft |
1.28 |
I xA,yA,uTrans,vTrans,rTrans,maskUp, |
| 13 |
adcroft |
1.21 |
I KappaRS, |
| 14 |
adcroft |
1.23 |
U fVerS, |
| 15 |
cnh |
1.16 |
I myCurrentTime, myThid ) |
| 16 |
cnh |
1.1 |
C /==========================================================\ |
| 17 |
|
|
C | SUBROUTINE CALC_GS | |
| 18 |
adcroft |
1.7 |
C | o Calculate the salt tendency terms. | |
| 19 |
cnh |
1.1 |
C |==========================================================| |
| 20 |
|
|
C | A procedure called EXTERNAL_FORCING_S is called from | |
| 21 |
|
|
C | here. These procedures can be used to add per problem | |
| 22 |
adcroft |
1.7 |
C | E-P flux source terms. | |
| 23 |
cnh |
1.1 |
C | Note: Although it is slightly counter-intuitive the | |
| 24 |
|
|
C | EXTERNAL_FORCING routine is not the place to put | |
| 25 |
|
|
C | file I/O. Instead files that are required to | |
| 26 |
|
|
C | calculate the external source terms are generally | |
| 27 |
|
|
C | read during the model main loop. This makes the | |
| 28 |
|
|
C | logisitics of multi-processing simpler and also | |
| 29 |
|
|
C | makes the adjoint generation simpler. It also | |
| 30 |
|
|
C | allows for I/O to overlap computation where that | |
| 31 |
|
|
C | is supported by hardware. | |
| 32 |
|
|
C | Aside from the problem specific term the code here | |
| 33 |
|
|
C | forms the tendency terms due to advection and mixing | |
| 34 |
|
|
C | The baseline implementation here uses a centered | |
| 35 |
|
|
C | difference form for the advection term and a tensorial | |
| 36 |
|
|
C | divergence of a flux form for the diffusive term. The | |
| 37 |
|
|
C | diffusive term is formulated so that isopycnal mixing and| |
| 38 |
|
|
C | GM-style subgrid-scale terms can be incorporated b simply| |
| 39 |
|
|
C | setting the diffusion tensor terms appropriately. | |
| 40 |
|
|
C \==========================================================/ |
| 41 |
|
|
IMPLICIT NONE |
| 42 |
|
|
|
| 43 |
|
|
C == GLobal variables == |
| 44 |
|
|
#include "SIZE.h" |
| 45 |
|
|
#include "DYNVARS.h" |
| 46 |
|
|
#include "EEPARAMS.h" |
| 47 |
|
|
#include "PARAMS.h" |
| 48 |
|
|
#include "GRID.h" |
| 49 |
cnh |
1.8 |
#include "FFIELDS.h" |
| 50 |
cnh |
1.1 |
|
| 51 |
|
|
C == Routine arguments == |
| 52 |
adcroft |
1.7 |
C fVerS - Flux of salt (S) in the vertical |
| 53 |
cnh |
1.1 |
C direction at the upper(U) and lower(D) faces of a cell. |
| 54 |
|
|
C maskUp - Land mask used to denote base of the domain. |
| 55 |
|
|
C xA - Tracer cell face area normal to X |
| 56 |
|
|
C yA - Tracer cell face area normal to X |
| 57 |
|
|
C uTrans - Zonal volume transport through cell face |
| 58 |
|
|
C vTrans - Meridional volume transport through cell face |
| 59 |
adcroft |
1.18 |
C rTrans - Vertical volume transport through cell face |
| 60 |
cnh |
1.1 |
C bi, bj, iMin, iMax, jMin, jMax - Range of points for which calculation |
| 61 |
|
|
C results will be set. |
| 62 |
adcroft |
1.7 |
C myThid - Instance number for this innvocation of CALC_GT |
| 63 |
cnh |
1.1 |
_RL fVerS (1-OLx:sNx+OLx,1-OLy:sNy+OLy,2) |
| 64 |
|
|
_RS xA (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 65 |
|
|
_RS yA (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 66 |
|
|
_RL uTrans(1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 67 |
|
|
_RL vTrans(1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 68 |
cnh |
1.12 |
_RL rTrans(1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 69 |
cnh |
1.1 |
_RS maskUp(1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 70 |
cnh |
1.14 |
_RL KappaRS(1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 71 |
adcroft |
1.7 |
INTEGER k,kUp,kDown,kM1 |
| 72 |
cnh |
1.1 |
INTEGER bi,bj,iMin,iMax,jMin,jMax |
| 73 |
adcroft |
1.18 |
_RL myCurrentTime |
| 74 |
cnh |
1.1 |
INTEGER myThid |
| 75 |
|
|
CEndOfInterface |
| 76 |
|
|
|
| 77 |
|
|
C == Local variables == |
| 78 |
|
|
C I, J, K - Loop counters |
| 79 |
adcroft |
1.7 |
INTEGER i,j |
| 80 |
|
|
LOGICAL TOP_LAYER |
| 81 |
|
|
_RL afFacS, dfFacS |
| 82 |
adcroft |
1.19 |
_RL df4 (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 83 |
adcroft |
1.23 |
_RL fZon (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 84 |
|
|
_RL fMer (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 85 |
|
|
_RL af (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 86 |
|
|
_RL df (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 87 |
heimbach |
1.20 |
|
| 88 |
|
|
#ifdef ALLOW_AUTODIFF_TAMC |
| 89 |
|
|
C-- only the kUp part of fverS is set in this subroutine |
| 90 |
|
|
C-- the kDown is still required |
| 91 |
|
|
fVerS(1,1,kDown) = fVerS(1,1,kDown) |
| 92 |
adcroft |
1.27 |
#endif |
| 93 |
heimbach |
1.20 |
DO j=1-OLy,sNy+OLy |
| 94 |
|
|
DO i=1-OLx,sNx+OLx |
| 95 |
|
|
fZon(i,j) = 0.0 |
| 96 |
|
|
fMer(i,j) = 0.0 |
| 97 |
|
|
fVerS(i,j,kUp) = 0.0 |
| 98 |
|
|
ENDDO |
| 99 |
|
|
ENDDO |
| 100 |
cnh |
1.1 |
|
| 101 |
|
|
afFacS = 1. _d 0 |
| 102 |
|
|
dfFacS = 1. _d 0 |
| 103 |
adcroft |
1.7 |
TOP_LAYER = K .EQ. 1 |
| 104 |
cnh |
1.1 |
|
| 105 |
|
|
C--- Calculate advective and diffusive fluxes between cells. |
| 106 |
|
|
|
| 107 |
adcroft |
1.19 |
C o Zonal tracer gradient |
| 108 |
|
|
DO j=1-Oly,sNy+Oly |
| 109 |
|
|
DO i=1-Olx+1,sNx+Olx |
| 110 |
adcroft |
1.28 |
fZon(i,j) = _recip_dxC(i,j,bi,bj)*xA(i,j) |
| 111 |
|
|
& *(salt(i,j,k,bi,bj)-salt(i-1,j,k,bi,bj)) |
| 112 |
|
|
#ifdef COSINEMETH_III |
| 113 |
|
|
& *sqCosFacU(j,bi,bj) |
| 114 |
|
|
#endif |
| 115 |
adcroft |
1.19 |
ENDDO |
| 116 |
|
|
ENDDO |
| 117 |
|
|
C o Meridional tracer gradient |
| 118 |
|
|
DO j=1-Oly+1,sNy+Oly |
| 119 |
|
|
DO i=1-Olx,sNx+Olx |
| 120 |
adcroft |
1.28 |
fMer(i,j) = _recip_dyC(i,j,bi,bj)*yA(i,j) |
| 121 |
|
|
& *(salt(i,j,k,bi,bj)-salt(i,j-1,k,bi,bj)) |
| 122 |
|
|
#ifdef ISOTROPIC_COS_SCALING |
| 123 |
|
|
#ifdef COSINEMETH_III |
| 124 |
|
|
& *sqCosFacV(j,bi,bj) |
| 125 |
|
|
#endif |
| 126 |
|
|
#endif |
| 127 |
adcroft |
1.19 |
ENDDO |
| 128 |
|
|
ENDDO |
| 129 |
|
|
|
| 130 |
|
|
C-- del^2 of S, needed for bi-harmonic (del^4) term |
| 131 |
|
|
IF (diffK4S .NE. 0.) THEN |
| 132 |
|
|
DO j=1-Oly+1,sNy+Oly-1 |
| 133 |
|
|
DO i=1-Olx+1,sNx+Olx-1 |
| 134 |
|
|
df4(i,j)= _recip_hFacC(i,j,k,bi,bj) |
| 135 |
|
|
& *recip_drF(k)/_rA(i,j,bi,bj) |
| 136 |
|
|
& *( |
| 137 |
adcroft |
1.28 |
& +( fZon(i+1,j)-fZon(i,j) ) |
| 138 |
|
|
& +( fMer(i,j+1)-fMer(i,j) ) |
| 139 |
adcroft |
1.19 |
& ) |
| 140 |
|
|
ENDDO |
| 141 |
|
|
ENDDO |
| 142 |
|
|
ENDIF |
| 143 |
|
|
|
| 144 |
cnh |
1.1 |
C-- Zonal flux (fZon is at west face of "salt" cell) |
| 145 |
|
|
C Advective component of zonal flux |
| 146 |
|
|
DO j=jMin,jMax |
| 147 |
|
|
DO i=iMin,iMax |
| 148 |
|
|
af(i,j) = |
| 149 |
|
|
& uTrans(i,j)*(salt(i,j,k,bi,bj)+salt(i-1,j,k,bi,bj))*0.5 _d 0 |
| 150 |
|
|
ENDDO |
| 151 |
|
|
ENDDO |
| 152 |
adcroft |
1.19 |
C o Diffusive component of zonal flux |
| 153 |
cnh |
1.1 |
DO j=jMin,jMax |
| 154 |
|
|
DO i=iMin,iMax |
| 155 |
adcroft |
1.28 |
df(i,j) = -diffKhS*xA(i,j)*_recip_dxC(i,j,bi,bj)* |
| 156 |
|
|
& (salt(i,j,k,bi,bj)-salt(i-1,j,k,bi,bj)) |
| 157 |
|
|
& *CosFacU(j,bi,bj) |
| 158 |
cnh |
1.1 |
ENDDO |
| 159 |
|
|
ENDDO |
| 160 |
adcroft |
1.21 |
#ifdef ALLOW_GMREDI |
| 161 |
heimbach |
1.22 |
IF (useGMRedi) CALL GMREDI_XTRANSPORT( |
| 162 |
adcroft |
1.21 |
I iMin,iMax,jMin,jMax,bi,bj,K, |
| 163 |
|
|
I xA,salt, |
| 164 |
|
|
U df, |
| 165 |
|
|
I myThid) |
| 166 |
|
|
#endif |
| 167 |
adcroft |
1.19 |
C o Add the bi-harmonic contribution |
| 168 |
|
|
IF (diffK4S .NE. 0.) THEN |
| 169 |
|
|
DO j=jMin,jMax |
| 170 |
|
|
DO i=iMin,iMax |
| 171 |
|
|
df(i,j) = df(i,j) + xA(i,j)* |
| 172 |
|
|
& diffK4S*(df4(i,j)-df4(i-1,j))*_recip_dxC(i,j,bi,bj) |
| 173 |
adcroft |
1.28 |
#ifdef COSINEMETH_III |
| 174 |
|
|
& *sqCosFacU(j,bi,bj) |
| 175 |
|
|
#else |
| 176 |
|
|
& *CosFacU(j,bi,bj) |
| 177 |
|
|
#endif |
| 178 |
adcroft |
1.19 |
ENDDO |
| 179 |
|
|
ENDDO |
| 180 |
|
|
ENDIF |
| 181 |
cnh |
1.1 |
C Net zonal flux |
| 182 |
|
|
DO j=jMin,jMax |
| 183 |
|
|
DO i=iMin,iMax |
| 184 |
|
|
fZon(i,j) = afFacS*af(i,j) + dfFacS*df(i,j) |
| 185 |
|
|
ENDDO |
| 186 |
|
|
ENDDO |
| 187 |
|
|
|
| 188 |
|
|
C-- Meridional flux (fMer is at south face of "salt" cell) |
| 189 |
|
|
C Advective component of meridional flux |
| 190 |
|
|
DO j=jMin,jMax |
| 191 |
|
|
DO i=iMin,iMax |
| 192 |
|
|
C Advective component of meridional flux |
| 193 |
|
|
af(i,j) = |
| 194 |
|
|
& vTrans(i,j)*(salt(i,j,k,bi,bj)+salt(i,j-1,k,bi,bj))*0.5 _d 0 |
| 195 |
|
|
ENDDO |
| 196 |
|
|
ENDDO |
| 197 |
|
|
C Diffusive component of meridional flux |
| 198 |
|
|
DO j=jMin,jMax |
| 199 |
|
|
DO i=iMin,iMax |
| 200 |
adcroft |
1.28 |
df(i,j) = -diffKhS*yA(i,j)*_recip_dyC(i,j,bi,bj)* |
| 201 |
|
|
& (salt(i,j,k,bi,bj)-salt(i,j-1,k,bi,bj)) |
| 202 |
|
|
& *CosFacV(j,bi,bj) |
| 203 |
cnh |
1.1 |
ENDDO |
| 204 |
|
|
ENDDO |
| 205 |
adcroft |
1.21 |
#ifdef ALLOW_GMREDI |
| 206 |
heimbach |
1.22 |
IF (useGMRedi) CALL GMREDI_YTRANSPORT( |
| 207 |
adcroft |
1.21 |
I iMin,iMax,jMin,jMax,bi,bj,K, |
| 208 |
|
|
I yA,salt, |
| 209 |
|
|
U df, |
| 210 |
|
|
I myThid) |
| 211 |
|
|
#endif |
| 212 |
adcroft |
1.19 |
C o Add the bi-harmonic contribution |
| 213 |
|
|
IF (diffK4S .NE. 0.) THEN |
| 214 |
|
|
DO j=jMin,jMax |
| 215 |
|
|
DO i=iMin,iMax |
| 216 |
|
|
df(i,j) = df(i,j) + yA(i,j)* |
| 217 |
|
|
& diffK4S*(df4(i,j)-df4(i,j-1))*_recip_dyC(i,j,bi,bj) |
| 218 |
adcroft |
1.28 |
#ifdef ISOTROPIC_COS_SCALING |
| 219 |
|
|
#ifdef COSINEMETH_III |
| 220 |
|
|
& *sqCosFacV(j,bi,bj) |
| 221 |
|
|
#else |
| 222 |
|
|
& *CosFacV(j,bi,bj) |
| 223 |
|
|
#endif |
| 224 |
|
|
#endif |
| 225 |
adcroft |
1.19 |
ENDDO |
| 226 |
|
|
ENDDO |
| 227 |
|
|
ENDIF |
| 228 |
|
|
|
| 229 |
cnh |
1.1 |
C Net meridional flux |
| 230 |
|
|
DO j=jMin,jMax |
| 231 |
|
|
DO i=iMin,iMax |
| 232 |
|
|
fMer(i,j) = afFacS*af(i,j) + dfFacS*df(i,j) |
| 233 |
|
|
ENDDO |
| 234 |
|
|
ENDDO |
| 235 |
|
|
|
| 236 |
adcroft |
1.28 |
C-- Vertical flux ( fVerS(,,kUp) is at upper face of "Tracer" cell ) |
| 237 |
|
|
C o Advective component of vertical flux : assume W_bottom=0 (mask) |
| 238 |
|
|
C Note: For K=1 then KM1=1 this gives a barZ(S) = S |
| 239 |
adcroft |
1.7 |
C (this plays the role of the free-surface correction) |
| 240 |
adcroft |
1.28 |
IF ( rigidLid .AND. TOP_LAYER) THEN |
| 241 |
|
|
DO j=jMin,jMax |
| 242 |
|
|
DO i=iMin,iMax |
| 243 |
|
|
af(i,j) = 0. |
| 244 |
|
|
ENDDO |
| 245 |
|
|
ENDDO |
| 246 |
|
|
ELSEIF ( rigidLid ) THEN |
| 247 |
|
|
DO j=jMin,jMax |
| 248 |
|
|
DO i=iMin,iMax |
| 249 |
|
|
af(i,j) = rTrans(i,j)* |
| 250 |
|
|
& (salt(i,j,k,bi,bj)+salt(i,j,kM1,bi,bj))*0.5 _d 0 |
| 251 |
|
|
ENDDO |
| 252 |
|
|
ENDDO |
| 253 |
|
|
ELSE |
| 254 |
|
|
DO j=jMin,jMax |
| 255 |
|
|
DO i=iMin,iMax |
| 256 |
|
|
af(i,j) = rTrans(i,j)*( |
| 257 |
|
|
& maskC(i,j,kM1,bi,bj)* |
| 258 |
|
|
& (salt(i,j,k,bi,bj)+salt(i,j,kM1,bi,bj))*0.5 _d 0 |
| 259 |
|
|
& +(maskC(i,j,k,bi,bj)-maskC(i,j,kM1,bi,bj))* |
| 260 |
|
|
& salt(i,j,k,bi,bj) ) |
| 261 |
|
|
ENDDO |
| 262 |
cnh |
1.1 |
ENDDO |
| 263 |
adcroft |
1.28 |
ENDIF |
| 264 |
adcroft |
1.21 |
C o Diffusive component of vertical flux |
| 265 |
|
|
C Note: For K=1 then KM1=1 and this gives a dS/dr = 0 upper |
| 266 |
adcroft |
1.7 |
C boundary condition. |
| 267 |
adcroft |
1.21 |
IF (implicitDiffusion) THEN |
| 268 |
|
|
DO j=jMin,jMax |
| 269 |
|
|
DO i=iMin,iMax |
| 270 |
|
|
df(i,j) = 0. |
| 271 |
|
|
ENDDO |
| 272 |
cnh |
1.1 |
ENDDO |
| 273 |
adcroft |
1.21 |
ELSE |
| 274 |
adcroft |
1.7 |
DO j=jMin,jMax |
| 275 |
|
|
DO i=iMin,iMax |
| 276 |
adcroft |
1.21 |
df(i,j) = - _rA(i,j,bi,bj)*( |
| 277 |
|
|
& KappaRS(i,j,k)*recip_drC(k) |
| 278 |
cnh |
1.13 |
& *(salt(i,j,kM1,bi,bj)-salt(i,j,k,bi,bj))*rkFac |
| 279 |
adcroft |
1.7 |
& ) |
| 280 |
|
|
ENDDO |
| 281 |
|
|
ENDDO |
| 282 |
|
|
ENDIF |
| 283 |
adcroft |
1.21 |
|
| 284 |
|
|
#ifdef ALLOW_GMREDI |
| 285 |
heimbach |
1.22 |
IF (useGMRedi) CALL GMREDI_RTRANSPORT( |
| 286 |
adcroft |
1.21 |
I iMin,iMax,jMin,jMax,bi,bj,K, |
| 287 |
|
|
I maskUp,salt, |
| 288 |
|
|
U df, |
| 289 |
|
|
I myThid) |
| 290 |
|
|
#endif |
| 291 |
|
|
|
| 292 |
adcroft |
1.18 |
#ifdef ALLOW_KPP |
| 293 |
adcroft |
1.21 |
C-- Add non-local KPP transport term (ghat) to diffusive salt flux. |
| 294 |
heimbach |
1.22 |
IF (useKPP) CALL KPP_TRANSPORT_S( |
| 295 |
adcroft |
1.21 |
I iMin,iMax,jMin,jMax,bi,bj,k,km1, |
| 296 |
adcroft |
1.28 |
I KappaRS, |
| 297 |
adcroft |
1.21 |
U df ) |
| 298 |
|
|
#endif |
| 299 |
adcroft |
1.18 |
|
| 300 |
cnh |
1.1 |
C Net vertical flux |
| 301 |
|
|
DO j=jMin,jMax |
| 302 |
|
|
DO i=iMin,iMax |
| 303 |
adcroft |
1.28 |
fVerS(i,j,kUp) = afFacS*af(i,j) + dfFacS*df(i,j)*maskUp(i,j) |
| 304 |
cnh |
1.1 |
ENDDO |
| 305 |
|
|
ENDDO |
| 306 |
|
|
|
| 307 |
|
|
C-- Tendency is minus divergence of the fluxes. |
| 308 |
|
|
C Note. Tendency terms will only be correct for range |
| 309 |
|
|
C i=iMin+1:iMax-1, j=jMin+1:jMax-1. Edge points |
| 310 |
|
|
C will contain valid floating point numbers but |
| 311 |
|
|
C they are not algorithmically correct. These points |
| 312 |
|
|
C are not used. |
| 313 |
adcroft |
1.28 |
DO j=jMin,jMax |
| 314 |
|
|
DO i=iMin,iMax |
| 315 |
cnh |
1.1 |
gS(i,j,k,bi,bj)= |
| 316 |
adcroft |
1.28 |
& -_recip_hFacC(i,j,k,bi,bj)*recip_drF(k) |
| 317 |
|
|
& *recip_rA(i,j,bi,bj) |
| 318 |
cnh |
1.1 |
& *( |
| 319 |
|
|
& +( fZon(i+1,j)-fZon(i,j) ) |
| 320 |
|
|
& +( fMer(i,j+1)-fMer(i,j) ) |
| 321 |
cnh |
1.12 |
& +( fVerS(i,j,kUp)-fVerS(i,j,kDown) )*rkFac |
| 322 |
cnh |
1.1 |
& ) |
| 323 |
|
|
ENDDO |
| 324 |
|
|
ENDDO |
| 325 |
|
|
|
| 326 |
cnh |
1.16 |
C-- External forcing term(s) |
| 327 |
|
|
CALL EXTERNAL_FORCING_S( |
| 328 |
|
|
I iMin,iMax,jMin,jMax,bi,bj,k, |
| 329 |
|
|
I myCurrentTime,myThid) |
| 330 |
cnh |
1.1 |
|
| 331 |
|
|
RETURN |
| 332 |
|
|
END |
Parent Directory
| 
Revision Log
| 
Revision Graph