| 1 |
|
| 2 |
#include "CTRL_CPPOPTIONS.h" |
| 3 |
#ifdef ALLOW_OBCS |
| 4 |
# include "OBCS_OPTIONS.h" |
| 5 |
#endif |
| 6 |
|
| 7 |
subroutine ctrl_obcsvol( |
| 8 |
I mytime, |
| 9 |
I myiter, |
| 10 |
I mythid |
| 11 |
& ) |
| 12 |
|
| 13 |
c ================================================================== |
| 14 |
c SUBROUTINE ctrl_obcsvol |
| 15 |
c ================================================================== |
| 16 |
c |
| 17 |
c o volumetrically balance the control vector contribution. |
| 18 |
c o Assume the calendar is identical |
| 19 |
c for all open boundaries. Need to save the barotropic adjustment |
| 20 |
c velocity so it can be used in all ctrl_getobcs files. |
| 21 |
c ================================================================== |
| 22 |
c SUBROUTINE ctrl_obcsvol |
| 23 |
c ================================================================== |
| 24 |
|
| 25 |
implicit none |
| 26 |
|
| 27 |
c == global variables == |
| 28 |
|
| 29 |
#include "EEPARAMS.h" |
| 30 |
#include "SIZE.h" |
| 31 |
#include "PARAMS.h" |
| 32 |
#include "GRID.h" |
| 33 |
#include "DYNVARS.h" |
| 34 |
#ifdef ALLOW_OBCS |
| 35 |
# include "OBCS.h" |
| 36 |
#endif |
| 37 |
|
| 38 |
#ifdef ALLOW_CALENDAR |
| 39 |
# include "cal.h" |
| 40 |
#endif |
| 41 |
#include "cost.h" |
| 42 |
#include "ctrl.h" |
| 43 |
#include "ctrl_dummy.h" |
| 44 |
#include "optim.h" |
| 45 |
|
| 46 |
c == routine arguments == |
| 47 |
|
| 48 |
integer myiter |
| 49 |
_RL mytime |
| 50 |
integer mythid |
| 51 |
|
| 52 |
c == local variables == |
| 53 |
|
| 54 |
integer bi,bj |
| 55 |
integer i,j,k |
| 56 |
integer itlo,ithi |
| 57 |
integer jtlo,jthi |
| 58 |
integer jmin,jmax |
| 59 |
integer imin,imax |
| 60 |
integer irec |
| 61 |
integer il |
| 62 |
integer iobcs |
| 63 |
integer ip1 |
| 64 |
integer jp1 |
| 65 |
integer nrec |
| 66 |
integer ilfld |
| 67 |
integer igg |
| 68 |
|
| 69 |
_RL sumvol |
| 70 |
_RL sumarea |
| 71 |
_RL tmpflux |
| 72 |
_RL tmparea |
| 73 |
_RL dummy |
| 74 |
_RL gg |
| 75 |
_RL tmpx |
| 76 |
_RL tmpy |
| 77 |
character*(80) fnamefldn |
| 78 |
character*(80) fnameflds |
| 79 |
character*(80) fnamefldw |
| 80 |
character*(80) fnameflde |
| 81 |
|
| 82 |
logical doglobalread |
| 83 |
logical ladinit |
| 84 |
logical obcsnfirst, obcsnchanged |
| 85 |
integer obcsncount0, obcsncount1 |
| 86 |
_RL obcsnfac |
| 87 |
|
| 88 |
#ifdef ECCO_VERBOSE |
| 89 |
character*(MAX_LEN_MBUF) msgbuf |
| 90 |
#endif |
| 91 |
|
| 92 |
c == external functions == |
| 93 |
|
| 94 |
integer ilnblnk |
| 95 |
external ilnblnk |
| 96 |
|
| 97 |
c == end of interface == |
| 98 |
|
| 99 |
jtlo = mybylo(mythid) |
| 100 |
jthi = mybyhi(mythid) |
| 101 |
itlo = mybxlo(mythid) |
| 102 |
ithi = mybxhi(mythid) |
| 103 |
jmin = 1 |
| 104 |
jmax = sny |
| 105 |
imin = 1 |
| 106 |
imax = snx |
| 107 |
|
| 108 |
c-- Read tiled data. |
| 109 |
doglobalread = .false. |
| 110 |
ladinit = .false. |
| 111 |
|
| 112 |
cgg Assume the number of records is the same for |
| 113 |
cgg all boundaries. |
| 114 |
|
| 115 |
|
| 116 |
#ifdef BALANCE_CONTROL_VOLFLUX_GLOBAL |
| 117 |
tmpflux= 0. d 0 |
| 118 |
tmparea= 0. d 0 |
| 119 |
sumarea= 0. d 0 |
| 120 |
sumvol = 0. d 0 |
| 121 |
|
| 122 |
#ifdef ECCO_VERBOSE |
| 123 |
_BEGIN_MASTER( mythid ) |
| 124 |
write(msgbuf,'(a)') ' ' |
| 125 |
call print_message( msgbuf, standardmessageunit, |
| 126 |
& SQUEEZE_RIGHT , mythid) |
| 127 |
write(msgbuf,'(a)') ' ' |
| 128 |
call print_message( msgbuf, standardmessageunit, |
| 129 |
& SQUEEZE_RIGHT , mythid) |
| 130 |
write(msgbuf,'(a,i9.8)') |
| 131 |
& ' ctrl_obcsvol: number of records to process: ',nrec |
| 132 |
call print_message( msgbuf, standardmessageunit, |
| 133 |
& SQUEEZE_RIGHT , mythid) |
| 134 |
write(msgbuf,'(a)') ' ' |
| 135 |
call print_message( msgbuf, standardmessageunit, |
| 136 |
& SQUEEZE_RIGHT , mythid) |
| 137 |
_END_MASTER( mythid ) |
| 138 |
#endif |
| 139 |
|
| 140 |
c-- Get the counters, flags, and the interpolation factor. |
| 141 |
call ctrl_GetRec( 'xx_obcsn', |
| 142 |
O obcsnfac, obcsnfirst, obcsnchanged, |
| 143 |
O obcsncount0,obcsncount1, |
| 144 |
I mytime, myiter, mythid ) |
| 145 |
|
| 146 |
if (optimcycle .ge. 0) then |
| 147 |
c |
| 148 |
ilfld=ilnblnk( xx_obcsn_file ) |
| 149 |
write(fnamefldn(1:80),'(2a,i10.10)') |
| 150 |
& xx_obcsn_file(1:ilfld),'.', optimcycle |
| 151 |
ilfld=ilnblnk( xx_obcss_file ) |
| 152 |
write(fnameflds(1:80),'(2a,i10.10)') |
| 153 |
& xx_obcss_file(1:ilfld),'.',optimcycle |
| 154 |
ilfld=ilnblnk( xx_obcsw_file ) |
| 155 |
write(fnamefldw(1:80),'(2a,i10.10)') |
| 156 |
& xx_obcsw_file(1:ilfld),'.',optimcycle |
| 157 |
ilfld=ilnblnk( xx_obcse_file ) |
| 158 |
write(fnameflde(1:80),'(2a,i10.10)') |
| 159 |
& xx_obcse_file(1:ilfld),'.',optimcycle |
| 160 |
c |
| 161 |
else |
| 162 |
print* |
| 163 |
print*,' ctrl_obcsvol: optimcycle has a wrong value.' |
| 164 |
print*,' optimcycle = ',optimcycle |
| 165 |
print* |
| 166 |
stop ' ... stopped in ctrl_obcsvol.' |
| 167 |
endif |
| 168 |
|
| 169 |
c-- Loop over records. For north boundary, we only need V velocity. |
| 170 |
|
| 171 |
if ( obcsnfirst ) then |
| 172 |
|
| 173 |
shiftvel(1) = 0. d0 |
| 174 |
shiftvel(2) = 0. d0 |
| 175 |
|
| 176 |
#ifdef ALLOW_OBCSN_CONTROL |
| 177 |
jp1 = 0 |
| 178 |
|
| 179 |
call active_read_xz(fnamefldn,tmpfldxz,(obcsncount0-1)*nobcs+4, |
| 180 |
& doglobalread, |
| 181 |
& ladinit, optimcycle, mythid |
| 182 |
& , xx_obcsn_dummy ) |
| 183 |
|
| 184 |
c-- Loop over this thread's tiles. |
| 185 |
do bj = jtlo,jthi |
| 186 |
do bi = itlo,ithi |
| 187 |
|
| 188 |
tmpflux = 0. d0 |
| 189 |
tmparea = 0. d0 |
| 190 |
|
| 191 |
do k = 1, Nr |
| 192 |
do i = imin,imax |
| 193 |
j = Ob_Jn(I,bi,bj) |
| 194 |
if (maskS(i,j+jp1,k,bi,bj) .ne. 0.) then |
| 195 |
cgg -- Don't let the corners contribute to the volume flux. |
| 196 |
if(ob_iw(j,bi,bj).ne.i .and. ob_ie(j,bi,bj).ne.i) then |
| 197 |
tmpx = tmpfldxz(i,k,bi,bj) |
| 198 |
cgg -- Positive is flux in. |
| 199 |
tmpflux = tmpflux - tmpx* delZ(k)*dxg(i,j+jp1,bi,bj) |
| 200 |
tmparea = tmparea + delZ(k) * dxg(i,j+jp1,bi,bj) |
| 201 |
endif |
| 202 |
endif |
| 203 |
enddo |
| 204 |
enddo |
| 205 |
|
| 206 |
sumarea = sumarea+ tmparea |
| 207 |
sumvol = sumvol + tmpflux |
| 208 |
enddo |
| 209 |
enddo |
| 210 |
#endif |
| 211 |
|
| 212 |
#ifdef ALLOW_OBCSS_CONTROL |
| 213 |
jp1 = 1 |
| 214 |
|
| 215 |
call active_read_xz(fnameflds,tmpfldxz,(obcsncount0-1)*nobcs+4, |
| 216 |
& doglobalread, |
| 217 |
& ladinit, optimcycle, mythid |
| 218 |
& , xx_obcss_dummy ) |
| 219 |
|
| 220 |
c-- Loop over this thread's tiles. |
| 221 |
do bj = jtlo,jthi |
| 222 |
do bi = itlo,ithi |
| 223 |
|
| 224 |
tmpflux = 0. d 0 |
| 225 |
tmparea = 0. d 0 |
| 226 |
|
| 227 |
do k = 1, Nr |
| 228 |
do i = imin,imax |
| 229 |
j = Ob_Js(I,bi,bj) |
| 230 |
if (maskS(i,j+jp1,k,bi,bj) .ne. 0.) then |
| 231 |
cgg -- Don't let the corners contribute to the volume flux. |
| 232 |
if (ob_iw(j,bi,bj).ne.i .and.ob_ie(j,bi,bj).ne.i) then |
| 233 |
tmpx = tmpfldxz(i,k,bi,bj) |
| 234 |
cgg -- Positive is flux in. |
| 235 |
tmpflux = tmpflux + tmpx* delZ(k)*dxg(i,j+jp1,bi,bj) |
| 236 |
tmparea = tmparea + delZ(k) * dxg(i,j+jp1,bi,bj) |
| 237 |
endif |
| 238 |
endif |
| 239 |
enddo |
| 240 |
enddo |
| 241 |
sumarea = sumarea+ tmparea |
| 242 |
sumvol = sumvol + tmpflux |
| 243 |
enddo |
| 244 |
enddo |
| 245 |
#endif |
| 246 |
|
| 247 |
#ifdef ALLOW_OBCSW_CONTROL |
| 248 |
ip1 = 1 |
| 249 |
|
| 250 |
call active_read_yz( fnamefldw, tmpfldyz, |
| 251 |
& (obcsncount0-1)*nobcs+3, doglobalread, |
| 252 |
& ladinit, optimcycle, mythid |
| 253 |
& , xx_obcsw_dummy ) |
| 254 |
|
| 255 |
c-- Loop over this thread's tiles. |
| 256 |
do bj = jtlo,jthi |
| 257 |
do bi = itlo,ithi |
| 258 |
|
| 259 |
c-- Determine the weights to be used. |
| 260 |
tmpflux = 0. d 0 |
| 261 |
tmparea = 0. d 0 |
| 262 |
|
| 263 |
do k = 1, Nr |
| 264 |
do j = jmin,jmax |
| 265 |
i = ob_iw(j,bi,bj) |
| 266 |
if (maskW(i+ip1,j,k,bi,bj) .ne. 0.) then |
| 267 |
cgg -- Don't let the corners contribute to the volume flux. |
| 268 |
if (ob_jn(i,bi,bj).ne.j .and. ob_js(i,bi,bj).ne.j)then |
| 269 |
tmpy = tmpfldyz(j,k,bi,bj) |
| 270 |
cgg -- Positive is flux in. |
| 271 |
tmpflux = tmpflux + tmpy* delZ(k)*dyg(i+ip1,j,bi,bj) |
| 272 |
tmparea = tmparea + delZ(k)*dyg(i+ip1,j,bi,bj) |
| 273 |
endif |
| 274 |
endif |
| 275 |
enddo |
| 276 |
enddo |
| 277 |
sumarea =sumarea + tmparea |
| 278 |
sumvol = sumvol + tmpflux |
| 279 |
enddo |
| 280 |
enddo |
| 281 |
#endif |
| 282 |
|
| 283 |
#ifdef ALLOW_OBCSE_CONTROL |
| 284 |
ip1 = 0 |
| 285 |
|
| 286 |
call active_read_yz( fnameflde, tmpfldyz, |
| 287 |
(obcsncount0-1)*nobcs+3, doglobalread, |
| 288 |
ladinit, optimcycle, mythid |
| 289 |
& , xx_obcse_dummy ) |
| 290 |
|
| 291 |
c-- Loop over this thread's tiles. |
| 292 |
do bj = jtlo,jthi |
| 293 |
do bi = itlo,ithi |
| 294 |
|
| 295 |
c-- Determine the weights to be used. |
| 296 |
tmpflux = 0. d 0 |
| 297 |
tmparea = 0. d 0 |
| 298 |
|
| 299 |
do k = 1, Nr |
| 300 |
do j = jmin,jmax |
| 301 |
i = ob_ie(j,bi,bj) |
| 302 |
if (maskW(i+ip1,j,k,bi,bj) .ne. 0.) then |
| 303 |
cgg -- Don't let the corners contribute to the volume flux. |
| 304 |
if (ob_jn(i,bi,bj).ne.j .and. ob_js(i,bi,bj).ne.j)then |
| 305 |
tmpy = tmpfldyz(j,k,bi,bj) |
| 306 |
cgg -- Positive is flux in. |
| 307 |
tmpflux = tmpflux - tmpy* delZ(k)*dyg(i+ip1,j,bi,bj) |
| 308 |
tmparea = tmparea + delZ(k) *dyg(i+ip1,j,bi,bj) |
| 309 |
endif |
| 310 |
endif |
| 311 |
enddo |
| 312 |
enddo |
| 313 |
sumarea = sumarea+ tmparea |
| 314 |
sumvol = sumvol + tmpflux |
| 315 |
enddo |
| 316 |
enddo |
| 317 |
#endif |
| 318 |
|
| 319 |
c-- Do the global summation. |
| 320 |
_GLOBAL_SUM_R8( sumvol, mythid ) |
| 321 |
_GLOBAL_SUM_R8( sumarea,mythid ) |
| 322 |
|
| 323 |
shiftvel(2) = sumvol /sumarea |
| 324 |
endif |
| 325 |
cgg End of the obcsnfirst loop. |
| 326 |
|
| 327 |
if ( ( obcsnfirst) .or. (obcsnchanged)) then |
| 328 |
|
| 329 |
cgg Swap the value. |
| 330 |
shiftvel(1) = shiftvel(2) |
| 331 |
|
| 332 |
sumvol = 0. d0 |
| 333 |
sumarea= 0. d0 |
| 334 |
|
| 335 |
#ifdef ALLOW_OBCSN_CONTROL |
| 336 |
jp1 = 0 |
| 337 |
|
| 338 |
call active_read_xz(fnamefldn,tmpfldxz,(obcsncount1-1)*nobcs+4, |
| 339 |
& doglobalread, |
| 340 |
& ladinit, optimcycle, mythid |
| 341 |
& , xx_obcsn_dummy ) |
| 342 |
c-- Loop over this thread's tiles. |
| 343 |
do bj = jtlo,jthi |
| 344 |
do bi = itlo,ithi |
| 345 |
|
| 346 |
tmpflux = 0. d0 |
| 347 |
tmparea = 0. d0 |
| 348 |
|
| 349 |
do k = 1, Nr |
| 350 |
do i = imin,imax |
| 351 |
j = Ob_Jn(I,bi,bj) |
| 352 |
if (maskS(i,j+jp1,k,bi,bj) .ne. 0.) then |
| 353 |
cgg -- Don't let the corners contribute to the volume flux. |
| 354 |
if (ob_iw(j,bi,bj).ne.i .and. ob_ie(j,bi,bj).ne.i)then |
| 355 |
tmpx = tmpfldxz(i,k,bi,bj) |
| 356 |
cgg -- Positive is flux in. |
| 357 |
tmpflux = tmpflux - tmpx* delZ(k)*dxg(i,j+jp1,bi,bj) |
| 358 |
tmparea = tmparea + delZ(k) * dxg(i,j+jp1,bi,bj) |
| 359 |
endif |
| 360 |
endif |
| 361 |
enddo |
| 362 |
enddo |
| 363 |
|
| 364 |
sumarea = sumarea+ tmparea |
| 365 |
sumvol = sumvol + tmpflux |
| 366 |
enddo |
| 367 |
enddo |
| 368 |
#endif |
| 369 |
|
| 370 |
#ifdef ALLOW_OBCSS_CONTROL |
| 371 |
jp1 = 1 |
| 372 |
|
| 373 |
call active_read_xz(fnameflds,tmpfldxz,(obcsncount1-1)*nobcs+4, |
| 374 |
& doglobalread, |
| 375 |
& ladinit, optimcycle, mythid |
| 376 |
& , xx_obcss_dummy ) |
| 377 |
|
| 378 |
c-- Loop over this thread's tiles. |
| 379 |
do bj = jtlo,jthi |
| 380 |
do bi = itlo,ithi |
| 381 |
|
| 382 |
tmpflux = 0. d 0 |
| 383 |
tmparea = 0. d 0 |
| 384 |
|
| 385 |
do k = 1, Nr |
| 386 |
do i = imin,imax |
| 387 |
j = Ob_Js(I,bi,bj) |
| 388 |
if (maskS(i,j+jp1,k,bi,bj) .ne. 0.) then |
| 389 |
cgg -- Don't let the corners contribute to the volume flux. |
| 390 |
if (ob_iw(j,bi,bj).ne.i .and. ob_ie(j,bi,bj).ne.i)then |
| 391 |
tmpx = tmpfldxz(i,k,bi,bj) |
| 392 |
cgg -- Positive is flux in. |
| 393 |
tmpflux = tmpflux + tmpx* delZ(k)*dxg(i,j+jp1,bi,bj) |
| 394 |
tmparea = tmparea + delZ(k) * dxg(i,j+jp1,bi,bj) |
| 395 |
endif |
| 396 |
endif |
| 397 |
enddo |
| 398 |
enddo |
| 399 |
sumarea = sumarea+ tmparea |
| 400 |
sumvol = sumvol + tmpflux |
| 401 |
enddo |
| 402 |
enddo |
| 403 |
#endif |
| 404 |
|
| 405 |
#ifdef ALLOW_OBCSW_CONTROL |
| 406 |
ip1 = 1 |
| 407 |
|
| 408 |
call active_read_yz( fnamefldw, tmpfldyz, |
| 409 |
& (obcsncount1-1)*nobcs+3, doglobalread, |
| 410 |
& ladinit, optimcycle, mythid |
| 411 |
& , xx_obcsw_dummy ) |
| 412 |
|
| 413 |
c-- Loop over this thread's tiles. |
| 414 |
do bj = jtlo,jthi |
| 415 |
do bi = itlo,ithi |
| 416 |
|
| 417 |
c-- Determine the weights to be used. |
| 418 |
tmpflux = 0. d 0 |
| 419 |
tmparea = 0. d 0 |
| 420 |
|
| 421 |
do k = 1, Nr |
| 422 |
do j = jmin,jmax |
| 423 |
i = ob_iw(j,bi,bj) |
| 424 |
if (maskW(i+ip1,j,k,bi,bj) .ne. 0.) then |
| 425 |
cgg -- Don't let corners contribute. |
| 426 |
if (ob_jn(i,bi,bj).ne.j .and. ob_js(i,bi,bj).ne.j)then |
| 427 |
tmpy = tmpfldyz(j,k,bi,bj) |
| 428 |
cgg -- Positive is flux in. |
| 429 |
tmpflux = tmpflux + tmpy* delZ(k)*dyg(i+ip1,j,bi,bj) |
| 430 |
tmparea = tmparea + delZ(k)*dyg(i+ip1,j,bi,bj) |
| 431 |
endif |
| 432 |
endif |
| 433 |
enddo |
| 434 |
enddo |
| 435 |
sumarea =sumarea + tmparea |
| 436 |
sumvol = sumvol + tmpflux |
| 437 |
enddo |
| 438 |
enddo |
| 439 |
#endif |
| 440 |
|
| 441 |
#ifdef ALLOW_OBCSE_CONTROL |
| 442 |
ip1 = 0 |
| 443 |
|
| 444 |
call active_read_yz( fnameflde, tmpfldyz, |
| 445 |
(obcsncount1-1)*nobcs+3, doglobalread, |
| 446 |
ladinit, optimcycle, mythid |
| 447 |
& , xx_obcse_dummy ) |
| 448 |
|
| 449 |
c-- Loop over this thread's tiles. |
| 450 |
do bj = jtlo,jthi |
| 451 |
do bi = itlo,ithi |
| 452 |
|
| 453 |
c-- Determine the weights to be used. |
| 454 |
tmpflux = 0. d 0 |
| 455 |
tmparea = 0. d 0 |
| 456 |
|
| 457 |
do k = 1, Nr |
| 458 |
do j = jmin,jmax |
| 459 |
i = ob_ie(j,bi,bj) |
| 460 |
if (maskW(i+ip1,j,k,bi,bj) .ne. 0.) then |
| 461 |
cgg -- Don't let the corners contribute to the volume flux. |
| 462 |
if (ob_jn(i,bi,bj).ne.j .and. ob_js(i,bi,bj).ne.j)then |
| 463 |
tmpy = tmpfldyz(j,k,bi,bj) |
| 464 |
cgg -- Positive is flux in. |
| 465 |
tmpflux = tmpflux - tmpy* delZ(k)*dyg(i+ip1,j,bi,bj) |
| 466 |
tmparea = tmparea + delZ(k) *dyg(i+ip1,j,bi,bj) |
| 467 |
endif |
| 468 |
endif |
| 469 |
enddo |
| 470 |
enddo |
| 471 |
sumarea = sumarea+ tmparea |
| 472 |
sumvol = sumvol + tmpflux |
| 473 |
enddo |
| 474 |
enddo |
| 475 |
#endif |
| 476 |
|
| 477 |
c-- Do the global summation. |
| 478 |
_GLOBAL_SUM_R8( sumvol, mythid ) |
| 479 |
_GLOBAL_SUM_R8( sumarea,mythid ) |
| 480 |
|
| 481 |
shiftvel(2) = sumvol /sumarea |
| 482 |
endif |
| 483 |
cgg End of the obcsnfirst, obcsnchanged loop. |
| 484 |
|
| 485 |
#endif |
| 486 |
|
| 487 |
return |
| 488 |
end |
| 489 |
|
| 490 |
|
| 491 |
|
| 492 |
|
| 493 |
|
| 494 |
|
| 495 |
|
Parent Directory
| 
Revision Log
| 
Revision Graph