| 1 |
cnh |
1.1 |
#include "CPP_EEOPTIONS.h" |
| 2 |
|
|
|
| 3 |
|
|
CStartOfInterFace |
| 4 |
|
|
SUBROUTINE CALC_QGPV( |
| 5 |
|
|
I bi, bj, iMin,iMax,jMin,jMax, pH, nsquare, |
| 6 |
|
|
I K13, K23, |
| 7 |
|
|
I myTime, myIter, |
| 8 |
|
|
O q, qr, qs, KpvU, KpvV, gUpv, gVpv, dqdyU, |
| 9 |
|
|
I myThid ) |
| 10 |
|
|
|
| 11 |
|
|
C /==========================================================\ |
| 12 |
|
|
C | SUBROUTINE CALC_QGPV | |
| 13 |
|
|
C | o Calculate the quasigeostrophic potential vorticity | |
| 14 |
|
|
C | o and eddy flux terms of qgpv etc | |
| 15 |
|
|
C |==========================================================| |
| 16 |
|
|
C | Richard Wardle 7/98 | |
| 17 |
|
|
C | Daniel Jamous 1/00 N^2 entered as an argument | |
| 18 |
|
|
C | instead of being calculated from | |
| 19 |
|
|
C | temperature | |
| 20 |
|
|
C | Updated to take into account | |
| 21 |
|
|
C | arbitrary topography | |
| 22 |
|
|
C | 3/00 compute PV fluxes using a local | |
| 23 |
|
|
C | N^2 (as opposed to an horizontal | |
| 24 |
|
|
C | average)--see key N2local | |
| 25 |
|
|
C | IMPORTANT NOTE: For now, the routine has been tested | |
| 26 |
|
|
C | only when there are no interpolation of gsf and dTdz | |
| 27 |
|
|
C | at horizontal boundaries, when qgpv is only made of | |
| 28 |
|
|
C | qs the stretching term, and when the Kpv are constants. | |
| 29 |
|
|
C | For a more general case, some more coding might be | |
| 30 |
|
|
C | necessary. | |
| 31 |
|
|
C \==========================================================/ |
| 32 |
|
|
C |
| 33 |
|
|
C == Global variables == |
| 34 |
|
|
#include "SIZE.h" |
| 35 |
|
|
#include "DYNVARS.h" |
| 36 |
|
|
#include "GRID.h" |
| 37 |
|
|
#include "EEPARAMS.h" |
| 38 |
|
|
#include "PARAMS.h" |
| 39 |
|
|
#include "CG2D.h" |
| 40 |
|
|
#include "FFIELDS.h" |
| 41 |
|
|
C |
| 42 |
|
|
C ========= Local variables that we may want to export ============= |
| 43 |
|
|
C q - quasigeostrophic potential vorticity |
| 44 |
|
|
C ================================================================== |
| 45 |
|
|
C |
| 46 |
|
|
_RL pH (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 47 |
|
|
_RL q (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 48 |
|
|
_RL qf (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 49 |
|
|
_RL qr (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 50 |
|
|
_RL qs (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 51 |
|
|
_RL KpvV (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 52 |
|
|
_RL KpvU (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 53 |
|
|
_RL gUpv (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr,nSx,nSy) |
| 54 |
|
|
_RL gVpv (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr,nSx,nSy) |
| 55 |
|
|
_RL nsquare (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr,nSx,nSy) |
| 56 |
|
|
_RL K13 (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 57 |
|
|
_RL K23 (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 58 |
|
|
C |
| 59 |
|
|
INTEGER bi,bj,iMin,iMax,jMin,jMax |
| 60 |
|
|
INTEGER myThid |
| 61 |
|
|
INTEGER myIter |
| 62 |
|
|
_RL myTime |
| 63 |
|
|
C |
| 64 |
|
|
CEndOfInterface |
| 65 |
|
|
C |
| 66 |
|
|
C define local variables here: |
| 67 |
|
|
_RL ptotal (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 68 |
|
|
_RL sfn (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 69 |
|
|
_RL gsfn (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 70 |
|
|
_RL tmpphi (lShare8,MAX_NO_THREADS) |
| 71 |
|
|
_RL tmpphi_area (lShare8,MAX_NO_THREADS) |
| 72 |
|
|
c _RL tbarxy (Nr) |
| 73 |
|
|
_RL sfnbarxy (Nr) |
| 74 |
|
|
_RL dTdz (Nr) |
| 75 |
|
|
_RL dTdz3d (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 76 |
|
|
_RL dVdx (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 77 |
|
|
_RL dUdy (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 78 |
|
|
_RL dVdxbarxy (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 79 |
|
|
_RL dUdybarxy (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 80 |
|
|
_RL dgsfndz (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 81 |
|
|
_RL arr (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr+1) |
| 82 |
|
|
_RL arrprime (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr+1) |
| 83 |
|
|
_RL arr2 (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 84 |
|
|
C |
| 85 |
|
|
_RL dqdyV (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 86 |
|
|
_RL dqdyT (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 87 |
|
|
_RL dqdyU (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 88 |
|
|
C |
| 89 |
|
|
_RL dqdxU (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 90 |
|
|
_RL dqdxT (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 91 |
|
|
_RL dqdxV (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 92 |
|
|
C |
| 93 |
|
|
_RL pvFacT (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 94 |
|
|
_RL pvFacU (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 95 |
|
|
_RL pvFacV (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 96 |
|
|
C |
| 97 |
|
|
_RL d2gsfndz2 (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 98 |
|
|
_RL interp1 (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 99 |
|
|
_RL d2Tdz2 (Nr) |
| 100 |
|
|
_RL interp2 (Nr) |
| 101 |
|
|
C |
| 102 |
|
|
_RL pmask (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 103 |
|
|
_RL umask (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 104 |
|
|
_RL vmask (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 105 |
|
|
C |
| 106 |
|
|
_RL num_i (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 107 |
|
|
_RL num_b (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 108 |
|
|
_RL denom (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 109 |
|
|
_RL kbotindex (1-OLx:sNx+OLx,1-OLy:sNy+OLy) |
| 110 |
|
|
|
| 111 |
|
|
INTEGER i,j,k |
| 112 |
|
|
INTEGER iG, jG |
| 113 |
|
|
INTEGER ku, kv, kup1, kbot |
| 114 |
|
|
ctest |
| 115 |
|
|
CHARACTER*(MAX_LEN_MBUF) suff |
| 116 |
|
|
ctest |
| 117 |
|
|
_RL tmpphiS |
| 118 |
|
|
_RL Kpvref |
| 119 |
|
|
_RL tmp, fac |
| 120 |
|
|
_RL mldu, mldv |
| 121 |
|
|
_RL int1, int2, int3, int4, int5, int6 |
| 122 |
|
|
_RL int2km1, signtest |
| 123 |
|
|
|
| 124 |
|
|
C ================================================================== |
| 125 |
|
|
C-- iG and jG are the global indices |
| 126 |
|
|
C ================================================================== |
| 127 |
|
|
Kpvref = 1000. |
| 128 |
|
|
C ================================================================== |
| 129 |
|
|
C-- Compute mask at pressure and velocity points |
| 130 |
|
|
C ================================================================== |
| 131 |
|
|
|
| 132 |
|
|
DO k=1,Nr |
| 133 |
|
|
DO j=iMin,jMax |
| 134 |
|
|
DO i=jMin,iMax |
| 135 |
|
|
pmask(i,j,k) = 1. |
| 136 |
|
|
IF (_hFacC(i,j,k,bi,bj).eq.0.) pmask(i,j,k)=0. |
| 137 |
|
|
ENDDO |
| 138 |
|
|
ENDDO |
| 139 |
|
|
DO j=iMin,jMax |
| 140 |
|
|
DO i=jMin,iMax |
| 141 |
|
|
umask(i,j,k) = pmask(i-1,j,k)*pmask(i,j,k) |
| 142 |
|
|
vmask(i,j,k) = pmask(i,j-1,k)*pmask(i,j,k) |
| 143 |
|
|
ENDDO |
| 144 |
|
|
ENDDO |
| 145 |
|
|
ENDDO |
| 146 |
|
|
|
| 147 |
|
|
DO j=iMin,jMax |
| 148 |
|
|
DO i=jMin,iMax |
| 149 |
|
|
kbotindex(i,j) = 0. |
| 150 |
|
|
IF ( pmask(i,j,Nr) .eq. 1. ) THEN |
| 151 |
|
|
kbotindex(i,j) = float(Nr) |
| 152 |
|
|
ELSE |
| 153 |
|
|
DO K = Nr-1,1,-1 |
| 154 |
|
|
IF ( pmask(i,j,k+1) .EQ. 0. .AND. |
| 155 |
|
|
& pmask(i,j,k) .EQ. 1. ) THEN |
| 156 |
|
|
kbotindex(i,j) = float(k) |
| 157 |
|
|
ENDIF |
| 158 |
|
|
ENDDO |
| 159 |
|
|
ENDIF |
| 160 |
|
|
ENDDO |
| 161 |
|
|
ENDDO |
| 162 |
|
|
|
| 163 |
|
|
C |
| 164 |
|
|
C ================================================================== |
| 165 |
|
|
C Compute the geostrophic streamfunction: gsf ====================== |
| 166 |
|
|
C ================================================================== |
| 167 |
|
|
C |
| 168 |
|
|
C -------v-------- |
| 169 |
|
|
C | | |
| 170 |
|
|
C | | |
| 171 |
|
|
C u x u Streamfunction is located at p points |
| 172 |
|
|
C | gsfn | |
| 173 |
|
|
C | | |
| 174 |
|
|
C -------v-------- |
| 175 |
|
|
C |
| 176 |
|
|
C ================================================================== |
| 177 |
|
|
C Compute the total pressure field: |
| 178 |
|
|
DO k=1,Nr |
| 179 |
|
|
DO j=iMin,jMax |
| 180 |
|
|
DO i=jMin,iMax |
| 181 |
|
|
ptotal(i,j,k) = pH(i,j,k) + |
| 182 |
|
|
& cg2d_x(i,j,bi,bj) * (gBaro * rhonil) |
| 183 |
|
|
ENDDO |
| 184 |
|
|
ENDDO |
| 185 |
|
|
ENDDO |
| 186 |
|
|
C Compute the streamfunction: |
| 187 |
|
|
DO k=1,Nr |
| 188 |
|
|
DO j=iMin,jMax |
| 189 |
|
|
DO i=iMin,iMax |
| 190 |
|
|
sfn(i,j,k) = ptotal(i,j,k) / ( rhonil * f0 ) |
| 191 |
|
|
ENDDO |
| 192 |
|
|
ENDDO |
| 193 |
|
|
ENDDO |
| 194 |
|
|
C |
| 195 |
|
|
C Compute the streamfunction: |
| 196 |
|
|
DO k=1,Nr |
| 197 |
|
|
DO j=iMin,jMax |
| 198 |
|
|
DO i=iMin,iMax |
| 199 |
|
|
sfn(i,j,k) = ptotal(i,j,k) / ( rhonil * f0 ) |
| 200 |
|
|
ENDDO |
| 201 |
|
|
ENDDO |
| 202 |
|
|
ENDDO |
| 203 |
|
|
C ================================================================== |
| 204 |
|
|
C 1. Evaluate the global horizontal mean of sf: |
| 205 |
|
|
c write(0,*)'Evaluate the horizontal mean' |
| 206 |
|
|
do k=1,Nr |
| 207 |
|
|
tmpphi(1,myThid)=0. |
| 208 |
|
|
tmpphi_area(1,myThid)=0. |
| 209 |
|
|
do j=1,sNy |
| 210 |
|
|
do i=1,sNx |
| 211 |
|
|
tmpphi(1,myThid) = tmpphi(1,myThid) + sfn(i,j,k)* |
| 212 |
|
|
& _rA(i,j,bi,bj)*pmask(i,j,k) |
| 213 |
|
|
tmpphi_area(1,myThid) = tmpphi_area(1,myThid) + |
| 214 |
|
|
& _rA(i,j,bi,bj)*pmask(i,j,k) |
| 215 |
|
|
enddo |
| 216 |
|
|
enddo |
| 217 |
|
|
_GLOBAL_SUM_R8( tmpphi, myThid ) |
| 218 |
|
|
_GLOBAL_SUM_R8( tmpphi_area, myThid ) |
| 219 |
|
|
sfnbarxy(k)=tmpphi(1,myThid)/tmpphi_area(1,myThid) |
| 220 |
|
|
c write(0,*)'level :',k,' horizontal mean :',sfnbarxy(k) |
| 221 |
|
|
enddo |
| 222 |
|
|
C |
| 223 |
|
|
C 2. Subtract sgbarxy from sf to give the geostrophic streamfunction: |
| 224 |
|
|
DO k=1,Nr |
| 225 |
|
|
DO j=jMin,jMax |
| 226 |
|
|
DO i=iMin,iMax |
| 227 |
|
|
gsfn(i,j,k) = ( sfn(i,j,k) - sfnbarxy(k) ) * pmask(i,j,k) |
| 228 |
|
|
ENDDO |
| 229 |
|
|
ENDDO |
| 230 |
|
|
ENDDO |
| 231 |
|
|
C ================================================================== |
| 232 |
|
|
C-- Interpolate gsf at horizontal boundaries: |
| 233 |
|
|
C ================================================================== |
| 234 |
|
|
C-- Vertical gradient of gsfn: |
| 235 |
|
|
DO k=2,Nr |
| 236 |
|
|
DO j=jMin,jMax |
| 237 |
|
|
DO i=iMin,iMax |
| 238 |
|
|
dgsfndz(i,j,k) = recip_drC(k)*( gsfn(i,j,k-1)-gsfn(i,j,k) ) |
| 239 |
|
|
ENDDO |
| 240 |
|
|
ENDDO |
| 241 |
|
|
ENDDO |
| 242 |
|
|
C-- Vertical gradient of dgsfndz: |
| 243 |
|
|
DO k=2,Nr |
| 244 |
|
|
DO j=jMin,jMax |
| 245 |
|
|
DO i=iMin,iMax |
| 246 |
|
|
d2gsfndz2(i,j,k) = recip_drC(k)*(dgsfndz(i,j,k)-dgsfndz(i,j,k+1)) |
| 247 |
|
|
ENDDO |
| 248 |
|
|
ENDDO |
| 249 |
|
|
ENDDO |
| 250 |
|
|
C-- |
| 251 |
|
|
DO k=2,3 |
| 252 |
|
|
DO j=jMin,jMax |
| 253 |
|
|
DO i=iMin,iMax |
| 254 |
|
|
d2gsfndz2(i,j,k) = d2gsfndz2(i,j,4) |
| 255 |
|
|
ENDDO |
| 256 |
|
|
ENDDO |
| 257 |
|
|
ENDDO |
| 258 |
|
|
C |
| 259 |
|
|
C-- Evaluate the gradients of gsfn |
| 260 |
|
|
DO j=jMin,jMax |
| 261 |
|
|
DO i=iMin,iMax |
| 262 |
|
|
interp1(i,j,3) = dgsfndz(i,j,4) + drC(3)*d2gsfndz2(i,j,3) |
| 263 |
|
|
ENDDO |
| 264 |
|
|
ENDDO |
| 265 |
|
|
C |
| 266 |
|
|
DO j=jMin,jMax |
| 267 |
|
|
DO i=iMin,iMax |
| 268 |
|
|
interp1(i,j,2) = interp1(i,j,3) + drC(2)*d2gsfndz2(i,j,2) |
| 269 |
|
|
ENDDO |
| 270 |
|
|
ENDDO |
| 271 |
|
|
DO k=4,Nr |
| 272 |
|
|
DO j=jMin,jMax |
| 273 |
|
|
DO i=iMin,iMax |
| 274 |
|
|
interp1(i,j,k) = dgsfndz(i,j,k) |
| 275 |
|
|
ENDDO |
| 276 |
|
|
ENDDO |
| 277 |
|
|
ENDDO |
| 278 |
|
|
C ================================================================== |
| 279 |
|
|
C ================================================================== |
| 280 |
|
|
C === Global mean temperature profile: dTdz3d ===================== |
| 281 |
|
|
C ================================================================== |
| 282 |
|
|
C Evaluate the global horizontal mean profile of nsquare |
| 283 |
|
|
c write(0,*)'mean N2' |
| 284 |
|
|
do k=1,Nr |
| 285 |
|
|
tmpphi(1,myThid)=0. |
| 286 |
|
|
tmpphi_area(1,myThid)=0. |
| 287 |
|
|
do j=1,sNy |
| 288 |
|
|
do i=1,sNx |
| 289 |
|
|
tmpphi(1,myThid)=tmpphi(1,myThid)+nsquare(i,j,k,bi,bj)* |
| 290 |
|
|
& _rA(i,j,bi,bj)*pmask(i,j,k) |
| 291 |
|
|
tmpphi_area(1,myThid)=tmpphi_area(1,myThid) + |
| 292 |
|
|
& _rA(i,j,bi,bj)*pmask(i,j,k) |
| 293 |
|
|
enddo |
| 294 |
|
|
enddo |
| 295 |
|
|
_GLOBAL_SUM_R8( tmpphi, myThid ) |
| 296 |
|
|
_GLOBAL_SUM_R8( tmpphi_area, myThid ) |
| 297 |
|
|
dTdz(K)=tmpphi(1,myThid)/tmpphi_area(1,myThid) |
| 298 |
|
|
c write(0,*)'level :',k,' horizontal mean :',dTdz(k) |
| 299 |
|
|
enddo |
| 300 |
|
|
C |
| 301 |
|
|
C Vertical gradient of tbarxy |
| 302 |
|
|
c DO k=2,Nr |
| 303 |
|
|
c dTdz(k) = recip_drC(k)*( tbarxy(k-1)-tbarxy(k) ) |
| 304 |
|
|
c ENDDO |
| 305 |
|
|
C Avoid dividing by zero: |
| 306 |
|
|
dTdz(1) = dTdz(2) |
| 307 |
|
|
C ================================================================== |
| 308 |
|
|
C interpolate dTdz at horizontal boundaries: |
| 309 |
|
|
C ================================================================== |
| 310 |
|
|
C-- Vertical gradient of dTdz: |
| 311 |
|
|
DO k=2,Nr |
| 312 |
|
|
DO j=jMin,jMax |
| 313 |
|
|
DO i=iMin,iMax |
| 314 |
|
|
d2Tdz2(k) = recip_drC(k)*(dTdz(k)-dTdz(k+1)) |
| 315 |
|
|
ENDDO |
| 316 |
|
|
ENDDO |
| 317 |
|
|
ENDDO |
| 318 |
|
|
C-- |
| 319 |
|
|
DO k=2,3 |
| 320 |
|
|
DO j=jMin,jMax |
| 321 |
|
|
DO i=iMin,iMax |
| 322 |
|
|
d2Tdz2(k) = d2Tdz2(4) |
| 323 |
|
|
ENDDO |
| 324 |
|
|
ENDDO |
| 325 |
|
|
ENDDO |
| 326 |
|
|
C |
| 327 |
|
|
C-- Evaluate the gradients of gsfn |
| 328 |
|
|
DO j=jMin,jMax |
| 329 |
|
|
DO i=iMin,iMax |
| 330 |
|
|
interp2(3) = dTdz(4) + drC(3)*d2Tdz2(3) |
| 331 |
|
|
ENDDO |
| 332 |
|
|
ENDDO |
| 333 |
|
|
C |
| 334 |
|
|
DO j=jMin,jMax |
| 335 |
|
|
DO i=iMin,iMax |
| 336 |
|
|
interp2(2) = interp2(3) + drC(2)*d2Tdz2(2) |
| 337 |
|
|
ENDDO |
| 338 |
|
|
ENDDO |
| 339 |
|
|
C |
| 340 |
|
|
DO k=4,Nr |
| 341 |
|
|
DO j=jMin,jMax |
| 342 |
|
|
DO i=iMin,iMax |
| 343 |
|
|
interp2(k) = dTdz(k) |
| 344 |
|
|
ENDDO |
| 345 |
|
|
ENDDO |
| 346 |
|
|
ENDDO |
| 347 |
|
|
C |
| 348 |
|
|
C ================================================================== |
| 349 |
|
|
C Create a threedimensional array of dTdz: |
| 350 |
|
|
DO k=1,Nr |
| 351 |
|
|
DO j=jMin,jMax |
| 352 |
|
|
DO i=iMin,iMax |
| 353 |
|
|
dTdz3d(i,j,k) = dTdz(k) |
| 354 |
|
|
c dTdz3d(i,j,k) = interp2(k) |
| 355 |
|
|
ENDDO |
| 356 |
|
|
ENDDO |
| 357 |
|
|
ENDDO |
| 358 |
|
|
C ================================================================== |
| 359 |
|
|
C ================================================================== |
| 360 |
|
|
C ======== Calculate the quasigeostrophic potential vorticity ====== |
| 361 |
|
|
C ================================================================== |
| 362 |
|
|
C |
| 363 |
|
|
C q = quasigeostrophic potential vorticity |
| 364 |
|
|
C |
| 365 |
|
|
C -------v-------- |
| 366 |
|
|
C | | |
| 367 |
|
|
C | | |
| 368 |
|
|
C u x u qgpv is located at p points |
| 369 |
|
|
C | qgpv | |
| 370 |
|
|
C | | |
| 371 |
|
|
C -------v-------- |
| 372 |
|
|
C |
| 373 |
|
|
C q = f + (gsf)_xx + (gsf)_yy + fo^2 ( (gsf)_z / N^2 )_z |
| 374 |
|
|
C |
| 375 |
|
|
C = qf + qr + qs |
| 376 |
|
|
C |
| 377 |
|
|
C ============================== |
| 378 |
|
|
C 1: Coriolis |
| 379 |
|
|
C ============================== |
| 380 |
|
|
C although fCori is defined at u-points they are at the same |
| 381 |
|
|
C latitude as p-points and so do not need to be inerpolated: |
| 382 |
|
|
DO j=jMin,jMax |
| 383 |
|
|
DO i=iMin,iMax |
| 384 |
|
|
qf(i,j) = fCori(i,j,bi,bj) |
| 385 |
|
|
ENDDO |
| 386 |
|
|
ENDDO |
| 387 |
|
|
C |
| 388 |
|
|
C ============================== |
| 389 |
|
|
C 2: relative |
| 390 |
|
|
C ============================== |
| 391 |
|
|
C !! for now use gradients of velocity field !! |
| 392 |
|
|
C ============================== |
| 393 |
|
|
C-- Zonal gradient of vVel |
| 394 |
|
|
C ============================== |
| 395 |
|
|
DO k=1,Nr |
| 396 |
|
|
DO j=jMin,jMax |
| 397 |
|
|
DO i=iMin,iMax |
| 398 |
|
|
dVdx(i,j,k) = _recip_dxV(i,j,bi,bj) * |
| 399 |
|
|
& ( vVel(i,j,k,bi,bj) - vVel(i-1,j,k,bi,bj) ) |
| 400 |
|
|
ENDDO |
| 401 |
|
|
ENDDO |
| 402 |
|
|
ENDDO |
| 403 |
|
|
C |
| 404 |
|
|
C-- Interpolate term to center point |
| 405 |
|
|
DO k=1,Nr |
| 406 |
|
|
DO j=jMin,jMax |
| 407 |
|
|
DO i=iMin,iMax |
| 408 |
|
|
iG = myXGlobalLo-1+(bi-1)*sNx+I |
| 409 |
|
|
C |
| 410 |
|
|
dVdxbarxy(i,j,k) = 0.25 * |
| 411 |
|
|
& ( dVdx(i,j ,k) + dVdx(i+1,j ,k) |
| 412 |
|
|
& + dVdx(i,j+1,k) + dVdx(i+1,j+1,k) ) |
| 413 |
|
|
C |
| 414 |
|
|
C-- free-slip b.c.'s: |
| 415 |
|
|
IF ( iG .EQ. 1 ) THEN |
| 416 |
|
|
dVdxbarxy(i,j,k) = 0.5 * |
| 417 |
|
|
& ( dVdx(i+1,j ,k) + dVdx(i+1,j+1,k) ) |
| 418 |
|
|
ENDIF |
| 419 |
|
|
IF ( iG .EQ. Nx-1 ) THEN |
| 420 |
|
|
dVdxbarxy(i,j,k) = 0.5 * |
| 421 |
|
|
& ( dVdx(i,j,k) + dVdx(i,j+1,k) ) |
| 422 |
|
|
ENDIF |
| 423 |
|
|
C |
| 424 |
|
|
ENDDO |
| 425 |
|
|
ENDDO |
| 426 |
|
|
ENDDO |
| 427 |
|
|
C-------------------------------------------------------- |
| 428 |
|
|
C ============================== |
| 429 |
|
|
C-- Meridional gradient of uVel |
| 430 |
|
|
C ============================== |
| 431 |
|
|
DO k=1,Nr |
| 432 |
|
|
DO j=jMin,jMax |
| 433 |
|
|
DO i=iMin,iMax |
| 434 |
|
|
dUdy(i,j,k) = _recip_dyU(i,j,bi,bj) * |
| 435 |
|
|
& ( uVel(i,j,k,bi,bj) - uVel(i,j-1,k,bi,bj) ) |
| 436 |
|
|
ENDDO |
| 437 |
|
|
ENDDO |
| 438 |
|
|
ENDDO |
| 439 |
|
|
C |
| 440 |
|
|
C-- Interpolate term to center point |
| 441 |
|
|
DO k=1,Nr |
| 442 |
|
|
DO j=jMin,jMax |
| 443 |
|
|
DO i=iMin,iMax |
| 444 |
|
|
jG = myYGlobalLo-1+(bj-1)*sNy+J |
| 445 |
|
|
C |
| 446 |
|
|
dUdybarxy(i,j,k) = 0.25 * |
| 447 |
|
|
& ( dUdy(i,j ,k) + dUdy(i+1,j ,k) |
| 448 |
|
|
& + dUdy(i,j+1,k) + dUdy(i+1,j+1,k) ) |
| 449 |
|
|
C |
| 450 |
|
|
C-- free-slip b.c.'s |
| 451 |
|
|
IF ( jG .EQ. 1 ) THEN |
| 452 |
|
|
dUdybarxy(i,j,k) = 0.5 * |
| 453 |
|
|
& ( dUdy(i, j+1,k) + dUdy(i+1,j+1,k) ) |
| 454 |
|
|
ENDIF |
| 455 |
|
|
IF ( jG .EQ. Ny-1 ) THEN |
| 456 |
|
|
dUdybarxy(i,j,k) = 0.5 * |
| 457 |
|
|
& ( dUdy(i, j,k) + dUdy(i+1,j,k) ) |
| 458 |
|
|
ENDIF |
| 459 |
|
|
C |
| 460 |
|
|
ENDDO |
| 461 |
|
|
ENDDO |
| 462 |
|
|
ENDDO |
| 463 |
|
|
C-- ================================= |
| 464 |
|
|
DO k=1,Nr |
| 465 |
|
|
DO j=jMin,jMax |
| 466 |
|
|
DO i=iMin,iMax |
| 467 |
|
|
qr(i,j,k) = ( dVdxbarxy(i,j,k) - dUdybarxy(i,j,k) ) |
| 468 |
|
|
ENDDO |
| 469 |
|
|
ENDDO |
| 470 |
|
|
ENDDO |
| 471 |
|
|
C-- ================================= |
| 472 |
|
|
C 3. stretching |
| 473 |
|
|
C-- ================================= |
| 474 |
|
|
C-- ====a. Vertical gradient of gsfn: |
| 475 |
|
|
DO k=2,Nr |
| 476 |
|
|
DO j=jMin,jMax |
| 477 |
|
|
DO i=iMin,iMax |
| 478 |
|
|
dgsfndz(i,j,k) = recip_drC(k)*( gsfn(i,j,k-1)-gsfn(i,j,k) )*pmask(i,j,k) |
| 479 |
|
|
ENDDO |
| 480 |
|
|
ENDDO |
| 481 |
|
|
ENDDO |
| 482 |
|
|
C |
| 483 |
|
|
C-- ====b. Evaluate (gsf)_z / N^2 |
| 484 |
|
|
DO k=2,Nr |
| 485 |
|
|
DO j=jMin,jMax |
| 486 |
|
|
DO i=iMin,iMax |
| 487 |
|
|
c arr(i,j,k) = interp1(i,j,k) |
| 488 |
|
|
arr(i,j,k) = dgsfndz(i,j,k) |
| 489 |
|
|
& / dTdz3d(i,j,k) |
| 490 |
|
|
ENDDO |
| 491 |
|
|
ENDDO |
| 492 |
|
|
ENDDO |
| 493 |
|
|
C set to zero in the upper and lower layers: |
| 494 |
|
|
C to include the pv sheets |
| 495 |
|
|
DO j=jMin,jMax |
| 496 |
|
|
DO i=iMin,iMax |
| 497 |
|
|
arr(i,j,1) = 0. |
| 498 |
|
|
arr(i,j,Nr+1) = 0. |
| 499 |
|
|
ENDDO |
| 500 |
|
|
ENDDO |
| 501 |
|
|
C |
| 502 |
|
|
C-- ====c. Evaluate z-derivative of (gsf)_z / N^2 |
| 503 |
|
|
DO k=1,Nr |
| 504 |
|
|
DO j=jMin,jMax |
| 505 |
|
|
DO i=iMin,iMax |
| 506 |
|
|
arr2(i,j,k) = recip_drF(k)*( arr(i,j,k)-arr(i,j,k+1) ) |
| 507 |
|
|
ENDDO |
| 508 |
|
|
ENDDO |
| 509 |
|
|
ENDDO |
| 510 |
|
|
C |
| 511 |
|
|
C-- ====d. mulitply by f0^2 |
| 512 |
|
|
DO k=1,Nr |
| 513 |
|
|
DO j=jMin,jMax |
| 514 |
|
|
DO i=iMin,iMax |
| 515 |
|
|
qs(i,j,k) = ( f0**2. ) * arr2(i,j,k) * pmask(i,j,k) |
| 516 |
|
|
ENDDO |
| 517 |
|
|
ENDDO |
| 518 |
|
|
ENDDO |
| 519 |
|
|
C ================================================================== |
| 520 |
|
|
C ================================================================== |
| 521 |
|
|
C Sum for q: the quasigeostrophic potential vorticity |
| 522 |
|
|
C ================================================================== |
| 523 |
|
|
DO k=1,Nr |
| 524 |
|
|
DO j=jMin,jMax |
| 525 |
|
|
DO i=iMin,iMax |
| 526 |
|
|
c q(i,j,k)= qf(i,j) + qr(i,j,k) + qs(i,j,k) |
| 527 |
|
|
q(i,j,k) = qs(i,j,k) |
| 528 |
|
|
ENDDO |
| 529 |
|
|
ENDDO |
| 530 |
|
|
ENDDO |
| 531 |
|
|
c write(0,*) (q(10,j,2),j=1,sNy) |
| 532 |
|
|
|
| 533 |
|
|
C ================================================================== |
| 534 |
|
|
C ================================================================== |
| 535 |
|
|
C-- |
| 536 |
|
|
IF ( pvForcing ) THEN |
| 537 |
|
|
C-- |
| 538 |
|
|
C-- ================================================================= |
| 539 |
|
|
C-- ================Compute the PV gradients========================= |
| 540 |
|
|
C-- ================================================================= |
| 541 |
|
|
C-- |
| 542 |
|
|
IF ( N2local ) THEN |
| 543 |
|
|
C |
| 544 |
|
|
C Interpolate K13 and K23 at horizontal u and v positions but vertical |
| 545 |
|
|
C w level |
| 546 |
|
|
DO k=1,Nr |
| 547 |
|
|
DO j=jMin,jMax |
| 548 |
|
|
DO i=iMin,iMax |
| 549 |
|
|
arr(i,j,k) = - 0.5 * |
| 550 |
|
|
& ( K23(i-1,j,k)+K23(i,j,k) ) * umask(i,j,k) |
| 551 |
|
|
arrprime(i,j,k) = - 0.5 * |
| 552 |
|
|
& ( K13(i,j-1,k)+K13(i,j,k) ) * vmask(i,j,k) |
| 553 |
|
|
ENDDO |
| 554 |
|
|
ENDDO |
| 555 |
|
|
ENDDO |
| 556 |
|
|
DO j=jMin,jMax |
| 557 |
|
|
DO i=iMin,iMax |
| 558 |
|
|
arr(i,j,Nr+1) = 0. |
| 559 |
|
|
arrprime(i,j,Nr+1) = 0. |
| 560 |
|
|
ENDDO |
| 561 |
|
|
ENDDO |
| 562 |
|
|
C Take z-derivative of arr and arrprime |
| 563 |
|
|
DO k=1,Nr |
| 564 |
|
|
DO j=jMin,jMax |
| 565 |
|
|
DO i=iMin,iMax |
| 566 |
|
|
dqdyU(i,j,k) = fCori(i,j,bi,bj) * recip_drF(k) * |
| 567 |
|
|
& (arr(i,j,k)-arr(i,j,k+1)) |
| 568 |
|
|
dqdxV(i,j,k) = 0.5 * ( fCori(i,j-1,bi,bj)+fCori(i,j,bi,bj) ) * recip_drF(k) * |
| 569 |
|
|
& (arrprime(i,j,k)-arrprime(i,j,k+1)) |
| 570 |
|
|
IF ( PVsheetmld ) THEN |
| 571 |
|
|
c ku = int( min(mldindex(i-1,j),mldindex(i,j)) ) |
| 572 |
|
|
c mldu = min( mld(i-1,j),mld(i,j) ) |
| 573 |
|
|
ku = int( max(mldindex(i-1,j),mldindex(i,j)) ) |
| 574 |
|
|
mldu = max( mld(i-1,j),mld(i,j) ) |
| 575 |
|
|
IF ( k .le. ku .AND. mldu .ne. 0. ) |
| 576 |
|
|
& dqdyU(i,j,k) = fCori(i,j,bi,bj)*(-arr(i,j,ku+1))/mldu |
| 577 |
|
|
c kv = int( min(mldindex(i,j-1),mldindex(i,j)) ) |
| 578 |
|
|
c mldv = min( mld(i,j-1),mld(i,j) ) |
| 579 |
|
|
kv = int( max(mldindex(i,j-1),mldindex(i,j)) ) |
| 580 |
|
|
mldv = max( mld(i,j-1),mld(i,j) ) |
| 581 |
|
|
IF ( k .le. kv .AND. mldv .ne. 0. ) |
| 582 |
|
|
& dqdxV(i,j,k) = 0.5*(fCori(i,j-1,bi,bj)+fCori(i,j,bi,bj)) |
| 583 |
|
|
& *(-arrprime(i,j,kv+1))/mldv |
| 584 |
|
|
ENDIF |
| 585 |
|
|
IF ( betaterm ) THEN |
| 586 |
|
|
dqdyU(i,j,k) = dqdyU(i,j,k) + dfdy(i,j,bi,bj)*umask(i,j,k) |
| 587 |
|
|
ENDIF |
| 588 |
|
|
ENDDO |
| 589 |
|
|
ENDDO |
| 590 |
|
|
ENDDO |
| 591 |
|
|
C |
| 592 |
|
|
ELSE |
| 593 |
|
|
C |
| 594 |
|
|
C ======================== |
| 595 |
|
|
C-- Meridional gradient of q |
| 596 |
|
|
C ======================== |
| 597 |
|
|
c write(0,*)'Meridional gradient of q' |
| 598 |
|
|
c-- At V-points: |
| 599 |
|
|
DO k=1,Nr |
| 600 |
|
|
DO j=jMin,jMax |
| 601 |
|
|
DO i=iMin,iMax |
| 602 |
|
|
dqdyV(i,j,k) = _recip_dyU(i,j,bi,bj) |
| 603 |
|
|
& * (q(i,j,k)-q(i,j-1,k)) |
| 604 |
|
|
ENDDO |
| 605 |
|
|
ENDDO |
| 606 |
|
|
ENDDO |
| 607 |
|
|
C |
| 608 |
|
|
C-- ---------------------------- |
| 609 |
|
|
C-- Interpolate term to U points |
| 610 |
|
|
C-- ---------------------------- |
| 611 |
|
|
C-- ::T-points first |
| 612 |
|
|
DO k=1,Nr |
| 613 |
|
|
DO j=jMin,jMax |
| 614 |
|
|
DO i=iMin,iMax |
| 615 |
|
|
jG = myYGlobalLo-1+(bj-1)*sNy+J |
| 616 |
|
|
dqdyT(i,j,k) = 0. |
| 617 |
|
|
tmp = vmask(i,j,k) + vmask(i,j+1,k) |
| 618 |
|
|
IF ( tmp .ne. 0. ) THEN |
| 619 |
|
|
dqdyT(i,j,k) = ( dqdyV(i,j,k)*vmask(i,j,k) + dqdyV(i,j+1,k)*vmask(i,j+1,k) )/( vmask(i,j,k) + vmask(i,j+1,k) ) |
| 620 |
|
|
ENDIF |
| 621 |
|
|
C |
| 622 |
|
|
C-- bondaries: |
| 623 |
|
|
c IF ( jG .EQ. 1 ) THEN |
| 624 |
|
|
c dqdyT(i,j,k) = dqdyV(i, j+1,k) |
| 625 |
|
|
c ENDIF |
| 626 |
|
|
c IF ( jG .EQ. sNy-1 ) THEN |
| 627 |
|
|
c dqdyT(i,j,k) = dqdyV(i,j,k) |
| 628 |
|
|
c ENDIF |
| 629 |
|
|
C-- |
| 630 |
|
|
ENDDO |
| 631 |
|
|
ENDDO |
| 632 |
|
|
ENDDO |
| 633 |
|
|
C-- ::U-points |
| 634 |
|
|
DO k=1,Nr |
| 635 |
|
|
DO j=jMin,jMax |
| 636 |
|
|
DO i=iMin,iMax |
| 637 |
|
|
iG = myXGlobalLo-1+(bi-1)*sNx+I |
| 638 |
|
|
dqdyU(i,j,k) = 0. |
| 639 |
|
|
IF ( umask(i,j,k) .ne. 0. ) THEN |
| 640 |
|
|
dqdyU(i,j,k) = ( dqdyT(i,j,k)*pmask(i,j,k) + dqdyT(i-1,j,k)*pmask(i-1,j,k) )/( pmask(i,j,k) + pmask(i-1,j,k) ) |
| 641 |
|
|
ENDIF |
| 642 |
|
|
C |
| 643 |
|
|
C-- bondaries: |
| 644 |
|
|
c IF ( iG .EQ. 1 ) THEN |
| 645 |
|
|
c dqdyU(i,j,k) = dqdyT(i+1,j,k) |
| 646 |
|
|
c ENDIF |
| 647 |
|
|
c IF ( iG .EQ. sNx-1 ) THEN |
| 648 |
|
|
c dqdyU(i,j,k) = dqdyT(i,j,k) |
| 649 |
|
|
c ENDIF |
| 650 |
|
|
C-- |
| 651 |
|
|
ENDDO |
| 652 |
|
|
ENDDO |
| 653 |
|
|
ENDDO |
| 654 |
|
|
|
| 655 |
|
|
C =================== |
| 656 |
|
|
C-- Zonal gradient of q |
| 657 |
|
|
C =================== |
| 658 |
|
|
c-- At U-points: |
| 659 |
|
|
c write(0,*)'Zonal gradient of q' |
| 660 |
|
|
DO k=1,Nr |
| 661 |
|
|
DO j=jMin,jMax |
| 662 |
|
|
DO i=iMin,iMax |
| 663 |
|
|
dqdxU(i,j,k) = _recip_dxC(i,j,bi,bj) |
| 664 |
|
|
& * ( q(i,j,k)-q(i-1,j,k) ) |
| 665 |
|
|
ENDDO |
| 666 |
|
|
ENDDO |
| 667 |
|
|
ENDDO |
| 668 |
|
|
C |
| 669 |
|
|
C-- ---------------------------- |
| 670 |
|
|
C-- Interpolate term to V points |
| 671 |
|
|
C-- ---------------------------- |
| 672 |
|
|
C-- ::T-points first |
| 673 |
|
|
DO k=1,Nr |
| 674 |
|
|
DO j=jMin,jMax |
| 675 |
|
|
DO i=iMin,iMax |
| 676 |
|
|
iG = myXGlobalLo-1+(bi-1)*sNx+I |
| 677 |
|
|
dqdxT(i,j,k) = 0. |
| 678 |
|
|
tmp = umask(i,j,k) + umask(i+1,j,k) |
| 679 |
|
|
IF ( tmp .ne. 0. ) THEN |
| 680 |
|
|
dqdxT(i,j,k) = ( dqdxU(i,j,k)*umask(i,j,k) + dqdxU(i+1,j,k)*umask(i+1,j,k) )/( umask(i,j,k) + umask(i+1,j,k) ) |
| 681 |
|
|
ENDIF |
| 682 |
|
|
C |
| 683 |
|
|
C-- bondaries: |
| 684 |
|
|
c IF ( iG .EQ. 1 ) THEN |
| 685 |
|
|
c dqdxT(i,j,k) = dqdxU(i+1,j,k) |
| 686 |
|
|
c ENDIF |
| 687 |
|
|
c IF ( iG .EQ. sNx-1 ) THEN |
| 688 |
|
|
c dqdxT(i,j,k) = dqdxU(i ,j,k) |
| 689 |
|
|
c ENDIF |
| 690 |
|
|
C-- |
| 691 |
|
|
ENDDO |
| 692 |
|
|
ENDDO |
| 693 |
|
|
ENDDO |
| 694 |
|
|
C-- ::V-points |
| 695 |
|
|
DO k=1,Nr |
| 696 |
|
|
DO j=jMin,jMax |
| 697 |
|
|
DO i=iMin,iMax |
| 698 |
|
|
jG = myYGlobalLo-1+(bj-1)*sNy+J |
| 699 |
|
|
dqdxV(i,j,k) = 0. |
| 700 |
|
|
IF ( vmask(i,j,k) .ne. 0. ) THEN |
| 701 |
|
|
dqdxV(i,j,k) = ( dqdxT(i,j,k)*pmask(i,j,k) + dqdxT(i,j-1,k)*pmask(i,j-1,k) )/( pmask(i,j,k) + pmask(i,j-1,k) ) |
| 702 |
|
|
ENDIF |
| 703 |
|
|
C |
| 704 |
|
|
C-- bondaries: |
| 705 |
|
|
c IF ( jG .EQ. 1 ) THEN |
| 706 |
|
|
c dqdxV(i,j,k) = dqdxT(i, j+1,k) |
| 707 |
|
|
c ENDIF |
| 708 |
|
|
c IF ( jG .EQ. sNy-1 ) THEN |
| 709 |
|
|
c dqdxV(i,j,k) = dqdxT(i,j,k) |
| 710 |
|
|
c ENDIF |
| 711 |
|
|
C-- |
| 712 |
|
|
ENDDO |
| 713 |
|
|
ENDDO |
| 714 |
|
|
ENDDO |
| 715 |
|
|
C |
| 716 |
|
|
ENDIF |
| 717 |
|
|
C-- |
| 718 |
|
|
C =============== |
| 719 |
|
|
C-- Evaluate Kpv : |
| 720 |
|
|
C =============== |
| 721 |
|
|
C-- |
| 722 |
|
|
IF ( N2local .AND. betaterm ) THEN |
| 723 |
|
|
c IF ( N2local ) THEN |
| 724 |
|
|
c |
| 725 |
|
|
DO i=iMin,iMax |
| 726 |
|
|
DO j=jMin,jMax |
| 727 |
|
|
kbot = int(kbotindex(i,j)) |
| 728 |
|
|
c zonal component |
| 729 |
|
|
int1 = 0. |
| 730 |
|
|
int2 = 0. |
| 731 |
|
|
int3 = 0. |
| 732 |
|
|
ku = int( max(mldindex(i-1,j),mldindex(i,j)) ) |
| 733 |
|
|
DO k=1,ku |
| 734 |
|
|
int1 = int1 + delz(k)*dqdyU(i,j,k) |
| 735 |
|
|
ENDDO |
| 736 |
|
|
IF ( ku .LT. kbot ) int3 = delz(kbot)*dqdyU(i,j,kbot) |
| 737 |
|
|
IF ( ku .LT. kbot-1 ) THEN |
| 738 |
|
|
DO k=ku+1,kbot-1 |
| 739 |
|
|
int2 = int2 + delz(k)*dqdyU(i,j,k) |
| 740 |
|
|
ENDDO |
| 741 |
|
|
ENDIF |
| 742 |
|
|
qymld(i,j) = int1 |
| 743 |
|
|
qyint(i,j) = int2 |
| 744 |
|
|
qybot(i,j) = int3 |
| 745 |
|
|
c meridional component |
| 746 |
|
|
int4 = 0. |
| 747 |
|
|
int5 = 0. |
| 748 |
|
|
int6 = 0. |
| 749 |
|
|
kv = int( max(mldindex(i,j-1),mldindex(i,j)) ) |
| 750 |
|
|
DO k=1,kv |
| 751 |
|
|
int4 = int4 + delz(k)*dqdxV(i,j,k) |
| 752 |
|
|
ENDDO |
| 753 |
|
|
IF ( kv .LT. kbot ) int6 = delz(kbot)*dqdxV(i,j,kbot) |
| 754 |
|
|
IF ( kv .LT. kbot-1 ) THEN |
| 755 |
|
|
DO k=kv+1,kbot-1 |
| 756 |
|
|
int5 = int5 + delz(k)*dqdxV(i,j,k) |
| 757 |
|
|
ENDDO |
| 758 |
|
|
ENDIF |
| 759 |
|
|
qxmld(i,j) = int4 |
| 760 |
|
|
qxint(i,j) = int5 |
| 761 |
|
|
qxbot(i,j) = int6 |
| 762 |
|
|
c compute the K |
| 763 |
|
|
num_i(i,j) = int3*int4 - int1*int6 |
| 764 |
|
|
num_b(i,j) = int1*int5 - int2*int4 |
| 765 |
|
|
denom(i,j) = int2*int6 - int3*int5 |
| 766 |
|
|
IF ( denom(i,j) .ne. 0. ) THEN |
| 767 |
|
|
ratioqy(i,j) = num_i(i,j)/denom(i,j) |
| 768 |
|
|
ratio_b(i,j) = num_b(i,j)/denom(i,j) |
| 769 |
|
|
ENDIF |
| 770 |
|
|
IF ( abs(ratioqy(i,j)-25.) .ge. 25. .OR. |
| 771 |
|
|
& abs(ratio_b(i,j)-25.) .ge. 25. ) THEN |
| 772 |
|
|
fac = 0. |
| 773 |
|
|
ELSE |
| 774 |
|
|
fac = 1. |
| 775 |
|
|
ENDIF |
| 776 |
|
|
DO k=1,kbot-1 |
| 777 |
|
|
IF ( k .le. ku ) THEN |
| 778 |
|
|
KpvU(i,j,k) = fac * Kpvref * umask(i,j,k) |
| 779 |
|
|
ELSE |
| 780 |
|
|
KpvU(i,j,k) = fac * ratioqy(i,j) * Kpvref * umask(i,j,k) |
| 781 |
|
|
ENDIF |
| 782 |
|
|
IF ( k .le. kv ) THEN |
| 783 |
|
|
KpvV(i,j,k) = fac * Kpvref * vmask(i,j,k) |
| 784 |
|
|
ELSE |
| 785 |
|
|
KpvV(i,j,k) = fac * ratioqy(i,j) * Kpvref * vmask(i,j,k) |
| 786 |
|
|
ENDIF |
| 787 |
|
|
ENDDO |
| 788 |
|
|
KpvU(i,j,kbot) = fac * ratio_b(i,j) * Kpvref * umask(i,j,kbot) |
| 789 |
|
|
KpvV(i,j,kbot) = fac * ratio_b(i,j) * Kpvref * vmask(i,j,kbot) |
| 790 |
|
|
ENDDO |
| 791 |
|
|
ENDDO |
| 792 |
|
|
c |
| 793 |
|
|
ELSE |
| 794 |
|
|
c |
| 795 |
|
|
DO i=iMin,iMax |
| 796 |
|
|
DO j=jMin,jMax |
| 797 |
|
|
pvFacT(i,j) = 0. _d 0 |
| 798 |
|
|
ENDDO |
| 799 |
|
|
ENDDO |
| 800 |
|
|
C ========================= |
| 801 |
|
|
C-- Lateral variation of Kpv: |
| 802 |
|
|
DO i=iMin,iMax |
| 803 |
|
|
DO j=jMin,jMax |
| 804 |
|
|
crmw pvFacT(i,j) = qp2(i,j,bi,bj) * 1.0 _d 0 |
| 805 |
|
|
pvFacT(i,j) = 1.0 _d 0 |
| 806 |
|
|
ENDDO |
| 807 |
|
|
ENDDO |
| 808 |
|
|
C-- |
| 809 |
|
|
C ::At U-point |
| 810 |
|
|
DO i=iMin,iMax |
| 811 |
|
|
DO j=jMin,jMax |
| 812 |
|
|
iG = myXGlobalLo-1+(bi-1)*sNx+I |
| 813 |
|
|
C-- |
| 814 |
|
|
pvFacU(i,j) = 0.5 * ( pvFacT(i,j) + pvFacT(i+1,j) ) |
| 815 |
|
|
C-- |
| 816 |
|
|
C-- bondaries: |
| 817 |
|
|
IF ( iG .EQ. 1 ) THEN |
| 818 |
|
|
pvFacU(i,j) = pvFacT(i+1,j) |
| 819 |
|
|
ENDIF |
| 820 |
|
|
C-- |
| 821 |
|
|
ENDDO |
| 822 |
|
|
ENDDO |
| 823 |
|
|
C-- |
| 824 |
|
|
C ::At V-point |
| 825 |
|
|
DO i=iMin,iMax |
| 826 |
|
|
DO j=jMin,jMax |
| 827 |
|
|
jG = myYGlobalLo-1+(bj-1)*sNy+J |
| 828 |
|
|
C-- |
| 829 |
|
|
pvFacV(i,j) = 0.5 * ( pvFacT(i,j) + pvFacT(i,j+1) ) |
| 830 |
|
|
C-- |
| 831 |
|
|
C-- bondaries: |
| 832 |
|
|
IF ( jG .EQ. 1 ) THEN |
| 833 |
|
|
pvFacV(i,j) = pvFacT(i,j+1) |
| 834 |
|
|
ENDIF |
| 835 |
|
|
C-- |
| 836 |
|
|
ENDDO |
| 837 |
|
|
ENDDO |
| 838 |
|
|
C-- |
| 839 |
|
|
C ========================= |
| 840 |
|
|
DO K=1,Nr |
| 841 |
|
|
DO i=iMin,iMax |
| 842 |
|
|
DO j=jMin,jMax |
| 843 |
|
|
KpvU(i,j,K) = Kpvref * pvFacU(i,j) |
| 844 |
|
|
KpvV(i,j,K) = Kpvref * pvFacV(i,j) |
| 845 |
|
|
c IF ( I .EQ. 10 .AND. J .EQ. 10) THEN |
| 846 |
|
|
c write(0,*) 't0', Kpvref, pvFacU(i,j), KpvU(i,j,K) |
| 847 |
|
|
c ENDIF |
| 848 |
|
|
ENDDO |
| 849 |
|
|
ENDDO |
| 850 |
|
|
ENDDO |
| 851 |
|
|
c |
| 852 |
|
|
ENDIF |
| 853 |
|
|
C-- |
| 854 |
|
|
C ========================================= |
| 855 |
|
|
C-- |
| 856 |
|
|
C ======================= |
| 857 |
|
|
C-- Evaluate gUpv and gVpv: |
| 858 |
|
|
C ======================= |
| 859 |
|
|
C-- |
| 860 |
|
|
c write(0,*)' Evaluate gUpv and gVpv' |
| 861 |
|
|
C |
| 862 |
|
|
DO k=1,Nr |
| 863 |
|
|
DO j=jMin,jMax |
| 864 |
|
|
DO i=iMin,iMax |
| 865 |
|
|
gUpv(i,j,k,bi,bj) = - KpvU(i,j,K) * dqdyU(i,j,k) |
| 866 |
|
|
c gUpv(i,j,k,bi,bj) = - Kpvref * ( dqdyU(i,j,k) - dfdy(i,j,bi,bj) ) |
| 867 |
|
|
gVpv(i,j,k,bi,bj) = KpvV(i,j,k) * dqdxV(i,j,k) |
| 868 |
|
|
c gVpv(i,j,k,bi,bj) = Kpvref * dqdxV(i,j,k) |
| 869 |
|
|
ENDDO |
| 870 |
|
|
ENDDO |
| 871 |
|
|
ENDDO |
| 872 |
|
|
C |
| 873 |
|
|
C-- |
| 874 |
|
|
ELSE !!!!!!!(pvForcing = .FALSE.) |
| 875 |
|
|
C-- |
| 876 |
|
|
IF ( GMadv ) THEN |
| 877 |
|
|
C |
| 878 |
|
|
C Compute horizontal eddy-induced velocities and save them in |
| 879 |
|
|
C arrays gUpv, gVpv |
| 880 |
|
|
DO k=1,Nr |
| 881 |
|
|
DO j=jMin,jMax |
| 882 |
|
|
DO i=iMin,iMax |
| 883 |
|
|
arr(i,j,k) = - Kpvref * 0.5 * |
| 884 |
|
|
& ( K13(i-1,j,k)+K13(i,j,k) ) * umask(i,j,k) |
| 885 |
|
|
arrprime(i,j,k) = - Kpvref * 0.5 * |
| 886 |
|
|
& ( K23(i,j-1,k)+K23(i,j,k) ) * vmask(i,j,k) |
| 887 |
|
|
ENDDO |
| 888 |
|
|
ENDDO |
| 889 |
|
|
ENDDO |
| 890 |
|
|
DO j=jMin,jMax |
| 891 |
|
|
DO i=iMin,iMax |
| 892 |
|
|
arr(i,j,Nr+1) = 0. |
| 893 |
|
|
arrprime(i,j,Nr+1) = 0. |
| 894 |
|
|
ENDDO |
| 895 |
|
|
ENDDO |
| 896 |
|
|
C Take z-derivative of arr and arrprime |
| 897 |
|
|
DO k=1,Nr |
| 898 |
|
|
DO j=jMin,jMax |
| 899 |
|
|
DO i=iMin,iMax |
| 900 |
|
|
KpvU(i,j,k) = 0. _d 0 |
| 901 |
|
|
KpvV(i,j,k) = 0. _d 0 |
| 902 |
|
|
gUpv(i,j,k,bi,bj) = recip_drF(k) * (arr(i,j,k)-arr(i,j,k+1)) |
| 903 |
|
|
gVpv(i,j,k,bi,bj) = recip_drF(k) * (arrprime(i,j,k)-arrprime(i,j,k+1)) |
| 904 |
|
|
ENDDO |
| 905 |
|
|
ENDDO |
| 906 |
|
|
ENDDO |
| 907 |
|
|
C |
| 908 |
|
|
ELSE |
| 909 |
|
|
C |
| 910 |
|
|
DO k=1,Nr |
| 911 |
|
|
DO j=jMin,jMax |
| 912 |
|
|
DO i=iMin,iMax |
| 913 |
|
|
KpvU(i,j,k) = 0. _d 0 |
| 914 |
|
|
KpvV(i,j,k) = 0. _d 0 |
| 915 |
|
|
gUpv(i,j,k,bi,bj) = 0. _d 0 |
| 916 |
|
|
gVpv(i,j,k,bi,bj) = 0. _d 0 |
| 917 |
|
|
ENDDO |
| 918 |
|
|
ENDDO |
| 919 |
|
|
ENDDO |
| 920 |
|
|
C |
| 921 |
|
|
ENDIF |
| 922 |
|
|
C-- |
| 923 |
|
|
ENDIF |
| 924 |
|
|
|
| 925 |
|
|
ctest |
| 926 |
|
|
_BARRIER |
| 927 |
|
|
_BEGIN_MASTER( myThid ) |
| 928 |
|
|
|
| 929 |
|
|
WRITE(suff,'(I10.10)') myIter |
| 930 |
|
|
|
| 931 |
|
|
C-- Write model fields |
| 932 |
|
|
CALL WRITE_FLD_XY_RL('kbot.',suff,kbotindex,myIter,myThid) |
| 933 |
|
|
c CALL WRITE_FLD_XY_RL('mldindex.',suff,mldindex,myIter,myThid) |
| 934 |
|
|
c CALL WRITE_FLD_XY_RL('mld.',suff,mld,myIter,myThid) |
| 935 |
|
|
c CALL WRITE_FLD_XY_RL('qymld.',suff,qymld,myIter,myThid) |
| 936 |
|
|
c CALL WRITE_FLD_XY_RL('qyint.',suff,qyint,myIter,myThid) |
| 937 |
|
|
c CALL WRITE_FLD_XY_RL('qybot.',suff,qybot,myIter,myThid) |
| 938 |
|
|
c CALL WRITE_FLD_XY_RL('qxmld.',suff,qxmld,myIter,myThid) |
| 939 |
|
|
c CALL WRITE_FLD_XY_RL('qxint.',suff,qxint,myIter,myThid) |
| 940 |
|
|
c CALL WRITE_FLD_XY_RL('qxbot.',suff,qxbot,myIter,myThid) |
| 941 |
|
|
c CALL WRITE_FLD_XY_RL('ratioqy.',suff,ratioqy,myIter,myThid) |
| 942 |
|
|
c CALL WRITE_FLD_XY_RL('ratio_b.',suff,ratio_b,myIter,myThid) |
| 943 |
|
|
c CALL WRITE_FLD_XYZ_RL( 'dqdyU.',suff,dqdyU,myIter,myThid) |
| 944 |
|
|
c CALL WRITE_FLD_XYZ_RL( 'dqdxV.',suff,dqdxV,myIter,myThid) |
| 945 |
|
|
c CALL WRITE_FLD_XYZ_RL( 'KpvU.',suff,KpvU,myIter,myThid) |
| 946 |
|
|
c CALL WRITE_FLD_XYZ_RL( 'KpvV.',suff,KpvV,myIter,myThid) |
| 947 |
|
|
c CALL WRITE_FLD_XYZ_RL( 'gUpv.',suff,gUpv,myIter,myThid) |
| 948 |
|
|
c CALL WRITE_FLD_XYZ_RL( 'gVpv.',suff,gVpv,myIter,myThid) |
| 949 |
|
|
c CALL WRITE_FLD_XY_RL('num_i.',suff,num_i,myIter,myThid) |
| 950 |
|
|
c CALL WRITE_FLD_XY_RL('num_b.',suff,num_b,myIter,myThid) |
| 951 |
|
|
c CALL WRITE_FLD_XY_RL('denom.',suff,denom,myIter,myThid) |
| 952 |
|
|
|
| 953 |
|
|
_END_MASTER( myThid ) |
| 954 |
|
|
_BARRIER |
| 955 |
|
|
ctest |
| 956 |
|
|
C-- |
| 957 |
|
|
C ========================================================= |
| 958 |
|
|
c DO k=1,Nr |
| 959 |
|
|
c DO j=jMin,jMax |
| 960 |
|
|
c DO i=iMin,iMax |
| 961 |
|
|
c write(0,*) 'TTTT ', Kpvref, pvFacU(i,j) |
| 962 |
|
|
c ENDDO |
| 963 |
|
|
c ENDDO |
| 964 |
|
|
c ENDDO |
| 965 |
|
|
C-- |
| 966 |
|
|
C ========================================================= |
| 967 |
|
|
|
| 968 |
|
|
RETURN |
| 969 |
|
|
END |
Parent Directory
| 
Revision Log
| 
Revision Graph