| 1 |
C$Header: /u/gcmpack/MITgcm/pkg/opps/opps_calc.F,v 1.1 2004/09/16 11:28:16 mlosch Exp $ |
| 2 |
C$Name: $ |
| 3 |
|
| 4 |
#include "OPPS_OPTIONS.h" |
| 5 |
|
| 6 |
CBOP |
| 7 |
C !ROUTINE: OPPS_CALC |
| 8 |
|
| 9 |
C !INTERFACE: ====================================================== |
| 10 |
subroutine OPPS_CALC( |
| 11 |
U tracerEnv, |
| 12 |
I wVel, |
| 13 |
I kMax, nTracer, nTracerInuse, |
| 14 |
I I, J, bi, bj, myTime, myIter, myThid ) |
| 15 |
|
| 16 |
C !DESCRIPTION: \bv |
| 17 |
C /=====================================================================\ |
| 18 |
C | SUBROUTINE OPPS_CALC | |
| 19 |
C | o Compute all OPPS fields defined in OPPS.h | |
| 20 |
C |=====================================================================| |
| 21 |
C | This subroutine is based on the routine 3dconvection.F | |
| 22 |
C | by E. Skyllingstad (?) | |
| 23 |
C | plenty of modifications to make it work: | |
| 24 |
C | - removed many unused parameters and variables | |
| 25 |
C | - turned everything (back) into 1D code | |
| 26 |
C | - pass variables, that are orginially in common blocks: | |
| 27 |
C | maxDepth | |
| 28 |
C | - pass vertical velocity, set in OPPS_INTERFACE | |
| 29 |
C | - do not use convadj for now (whatever that is) | |
| 30 |
C | - changed two .LT. 0 to .LE. 0 statements (because of possible | |
| 31 |
C | division) | |
| 32 |
C | - replaced statement function state1 by call to a real function | |
| 33 |
C | - removed range check, actually moved it up to OPPS_INTERFACE | |
| 34 |
C | - avoid division by zero: if (Wd.EQ.0) dt = ...1/Wd | |
| 35 |
C | - cleaned-up debugging | |
| 36 |
C | - replaced local dz and GridThickness by global drF | |
| 37 |
C | - replaced 1/dz by 1*recip_drF | |
| 38 |
C | - replaced 9.81 with gravity (=9.81) | |
| 39 |
C | - added a lot of comments that relate code to equation in paper | |
| 40 |
C | (Paluszkiewicz+Romea, 1997, Dynamics of Atmospheres and Oceans, | |
| 41 |
C | 26, pp. 95-130) | |
| 42 |
C | - included passive tracer support. This is the main change and may | |
| 43 |
C | not improve the readability of the code because of the joint | |
| 44 |
C | treatment of active (theta, salt) and passive tracers. The array | |
| 45 |
C | tracerEnv(Nr,2+PTRACERS_num) contains | |
| 46 |
C | theta = tracerEnv(:,1), | |
| 47 |
C | salt = tracerEnv(:,2), and | |
| 48 |
C | ptracers = tracerEnv(:,3:PTRACERS_num+2). | |
| 49 |
C | All related array names have been changed accordingly, so that | |
| 50 |
C | instead of Sd(Nr) and Td(Nr) (plume salinity and temperature), we | |
| 51 |
C | have Pd(Nr,nTracer) (tracer in plume), with Sd(:) = Pd(:,2), | |
| 52 |
C | Td(:) = Pd(:,1), etc. | |
| 53 |
C | o TODO: | |
| 54 |
C | clean up the logic of the vertical loops and get rid off the | |
| 55 |
C | GOTO statements | |
| 56 |
C \=====================================================================/ |
| 57 |
IMPLICIT NONE |
| 58 |
C |
| 59 |
C-------------------------------------------------------------------- |
| 60 |
|
| 61 |
C \ev |
| 62 |
|
| 63 |
C !USES: ============================================================ |
| 64 |
#include "SIZE.h" |
| 65 |
#include "EEPARAMS.h" |
| 66 |
#include "PARAMS.h" |
| 67 |
#include "OPPS.h" |
| 68 |
#include "FFIELDS.h" |
| 69 |
#include "GRID.h" |
| 70 |
|
| 71 |
EXTERNAL DIFFERENT_MULTIPLE |
| 72 |
LOGICAL DIFFERENT_MULTIPLE |
| 73 |
|
| 74 |
C !INPUT PARAMETERS: =================================================== |
| 75 |
c Routine arguments |
| 76 |
c bi, bj - array indices on which to apply calculations |
| 77 |
c myTime - Current time in simulation |
| 78 |
|
| 79 |
INTEGER I, J, bi, bj, KMax, nTracer, nTracerInUse |
| 80 |
INTEGER myThid, myIter |
| 81 |
_RL myTime |
| 82 |
_RL tracerEnv(Nr,nTracer),wVel(Nr) |
| 83 |
|
| 84 |
#ifdef ALLOW_OPPS |
| 85 |
C !LOCAL VARIABLES: ==================================================== |
| 86 |
c Local constants |
| 87 |
C imin, imax, jmin, jmax - array computation indices |
| 88 |
C msgBuf - Informational/error meesage buffer |
| 89 |
CHARACTER*(MAX_LEN_MBUF) msgBuf |
| 90 |
INTEGER K, K2, K2m1, K2p1, ktr |
| 91 |
INTEGER ntime,nn,kmx,ic |
| 92 |
INTEGER maxDepth |
| 93 |
|
| 94 |
_RL wsqr,oldflux,newflux,entrainrate |
| 95 |
_RL pmix |
| 96 |
_RL D1,D2,state1 |
| 97 |
_RL dz1,dz2 |
| 98 |
_RL radius,StartingFlux |
| 99 |
_RL dtts,dt |
| 100 |
C Arrays |
| 101 |
_RL Paa(Nr,nTracer) |
| 102 |
_RL wda(Nr), mda(Nr), pda(Nr,nTracer) |
| 103 |
C |
| 104 |
C Pd, Wd - tracers, vertical velocity in plume |
| 105 |
C Md - plume mass flux (?) |
| 106 |
C Ad - fractional area covered by plume |
| 107 |
C Dd - density in plume |
| 108 |
C De - density of environment |
| 109 |
C PlumeEntrainment - |
| 110 |
_RL Ad(Nr),Wd(Nr),Dd(Nr),Md(Nr) |
| 111 |
_RL De(Nr) |
| 112 |
_RL PlumeEntrainment(Nr) |
| 113 |
_RL Pd(Nr,nTracer) |
| 114 |
CEOP |
| 115 |
|
| 116 |
|
| 117 |
C-- Check to see if should convect now |
| 118 |
C IF ( DIFFERENT_MULTIPLE(cAdjFreq,myTime,myTime-deltaTClock) ) THEN |
| 119 |
IF ( .true. ) THEN |
| 120 |
C local initialization |
| 121 |
|
| 122 |
C Copy some arrays |
| 123 |
dtts = deltaTtracer |
| 124 |
C |
| 125 |
C start k-loop |
| 126 |
C |
| 127 |
|
| 128 |
DO k=1,KMax-1 |
| 129 |
c |
| 130 |
c initialize the plume T,S,density, and w velocity |
| 131 |
c |
| 132 |
DO ktr=1,nTracerInUse |
| 133 |
Pd(k,ktr) = tracerEnv(k,ktr) |
| 134 |
ENDDO |
| 135 |
Dd(k)=state1(Pd(k,2),Pd(k,1),i,j,k,bi,bj,myThid) |
| 136 |
De(k)=Dd(k) |
| 137 |
CML print *, 'ml-opps:', i,j,k,tracerEnv(k,2),tracerEnv(k,1), |
| 138 |
CML & Dd(k),Pd(k,1),Pd(k,2) |
| 139 |
CML compute vertical velocity at cell centers from GCM velocity |
| 140 |
Wd(k)= - .5*(wVel(K)+wVel(K+1)) |
| 141 |
CML( |
| 142 |
CML avoid division by zero |
| 143 |
CML IF (Wd(K) .EQ. 0.D0) Wd(K) = 2.23e-16 |
| 144 |
CML) |
| 145 |
c |
| 146 |
c guess at initial top grid cell vertical velocity |
| 147 |
c |
| 148 |
CML Wd(k) = 0.03 |
| 149 |
c |
| 150 |
c these estimates of initial plume velocity based on plume size and |
| 151 |
c top grid cell water mass |
| 152 |
c |
| 153 |
c Wd(k) = 0.5*drF(k)/(dtts*FRACTIONAL_AREA) |
| 154 |
c Wd(k) = 0.5*drF(k)/dtts |
| 155 |
c |
| 156 |
wsqr=Wd(k)*Wd(k) |
| 157 |
PlumeEntrainment(k) = 0.0 |
| 158 |
c |
| 159 |
c |
| 160 |
c |
| 161 |
#ifdef ALLOW_OPPS_DEBUG |
| 162 |
IF ( OPPSdebugLevel.GE.debLevB ) THEN |
| 163 |
WRITE(msgBuf,'(A,I3)') |
| 164 |
& 'S/R OPPS_CALC: doing old lowerparcel', k |
| 165 |
CALL PRINT_MESSAGE( msgBuf, standardMessageUnit, |
| 166 |
& SQUEEZE_RIGHT , 1) |
| 167 |
ENDIF |
| 168 |
#endif /* ALLOW_OPPS_DEBUG */ |
| 169 |
radius=PlumeRadius |
| 170 |
StartingFlux=radius*radius*Wd(k)*Dd(k) |
| 171 |
oldflux=StartingFlux |
| 172 |
|
| 173 |
dz2=DrF(k) |
| 174 |
DO k2=k,KMax-1 |
| 175 |
D1=state1( Pd(k2,2), Pd(k2,1),i,j,k2+1,bi,bj,myThid) |
| 176 |
D2=state1( tracerEnv(k2+1,2), tracerEnv(k2+1,1), |
| 177 |
& i,j,k2+1,bi,bj,myThid) |
| 178 |
De(k2+1)=D2 |
| 179 |
c |
| 180 |
c To start downward, parcel has to initially be heavier than environment |
| 181 |
c but after it has started moving, we continue plume until plume tke or |
| 182 |
c flux goes negative |
| 183 |
c |
| 184 |
CML & _hFacC(i,j,k-1,bi,bj) |
| 185 |
CML & *_hFacC(i,j,k,bi,bj) .GT. 0. |
| 186 |
CML & .AND. |
| 187 |
IF (D2-D1 .LT. STABILITY_THRESHOLD.or.k2.ne.k) THEN |
| 188 |
dz1=dz2 |
| 189 |
dz2=DrF(k2+1) |
| 190 |
c |
| 191 |
C find mass flux according to eq.(3) from paper by vertical integration |
| 192 |
c |
| 193 |
newflux=oldflux+e2*radius*Wd(k2)*Dd(k2)* |
| 194 |
& .5*(dz1+dz2) |
| 195 |
CML print *, 'ml-opps:', i,j,k,oldflux,newflux,e2,radius, |
| 196 |
CML & Wd(k2),Dd(k2),Pd(k2,1),Pd(k2,2),dz1,dz2 |
| 197 |
c |
| 198 |
PlumeEntrainment(k2+1) = newflux/StartingFlux |
| 199 |
c |
| 200 |
IF(newflux.LE.0.0) then |
| 201 |
#ifdef ALLOW_OPPS_DEBUG |
| 202 |
IF ( OPPSdebugLevel.GE.debLevA ) THEN |
| 203 |
WRITE(msgBuf,'(A,I3)') |
| 204 |
& 'S/R OPPS_CALC: Plume entrained to zero at level ', k2 |
| 205 |
CALL PRINT_MESSAGE( msgBuf, standardMessageUnit, |
| 206 |
& SQUEEZE_RIGHT , 1) |
| 207 |
ENDIF |
| 208 |
#endif /* ALLOW_OPPS_DEBUG */ |
| 209 |
maxdepth = k2 |
| 210 |
if(maxdepth.eq.k) goto 1000 |
| 211 |
goto 1 |
| 212 |
endif |
| 213 |
c |
| 214 |
c entrainment rate is basically a scaled mass flux dM/M |
| 215 |
c |
| 216 |
entrainrate = (newflux - oldflux)/newflux |
| 217 |
oldflux = newflux |
| 218 |
c |
| 219 |
c |
| 220 |
c mix var's are the average environmental values over the two grid levels |
| 221 |
c |
| 222 |
DO ktr=1,nTracerInUse |
| 223 |
pmix=(dz1*tracerEnv(k2,ktr)+dz2*tracerEnv(k2+1,ktr)) |
| 224 |
& /(dz1+dz2) |
| 225 |
Pd(k2+1,ktr)=Pd(k2,ktr) |
| 226 |
& - entrainrate*(pmix - Pd(k2,ktr)) |
| 227 |
ENDDO |
| 228 |
c |
| 229 |
c compute the density at this level for the buoyancy term in the |
| 230 |
c vertical k.e. equation |
| 231 |
c |
| 232 |
Dd(k2+1)=state1(Pd(k2+1,2),Pd(k2+1,1),i,j,k2+1,bi,bj,myThid) |
| 233 |
c |
| 234 |
c next, solve for the vertical velocity k.e. using combined eq. (4) |
| 235 |
c and eq (5) from the paper |
| 236 |
c |
| 237 |
#ifdef ALLOW_OPPS_DEBUG |
| 238 |
IF ( OPPSdebugLevel.GE.debLevA ) THEN |
| 239 |
WRITE(msgBuf,'(A,3E12.4,I3)') |
| 240 |
& 'S/R OPPS_CALC: Dd,De,entr,k ',Dd(k2),De(k2),entrainrate,k2 |
| 241 |
CALL PRINT_MESSAGE( msgBuf, standardMessageUnit, |
| 242 |
& SQUEEZE_RIGHT , 1) |
| 243 |
ENDIF |
| 244 |
#endif /* ALLOW_OPPS_DEBUG */ |
| 245 |
CML insert Eq. (4) into Eq. (5) to get something like this for wp^2 |
| 246 |
wsqr = wsqr - wsqr*abs(entrainrate)+ gravity* |
| 247 |
& (dz1*(Dd(k2)-De(k2))/De(k2) |
| 248 |
& +dz2*(Dd(k2+1)-De(k2+1))/De(k2+1)) |
| 249 |
c |
| 250 |
c if negative k.e. then plume has reached max depth, get out of loop |
| 251 |
c |
| 252 |
IF(wsqr.LE.0.0)then |
| 253 |
maxdepth = k2 |
| 254 |
#ifdef ALLOW_OPPS_DEBUG |
| 255 |
IF ( OPPSdebugLevel.GE.debLevA ) THEN |
| 256 |
WRITE(msgBuf,'(A,I3)') |
| 257 |
& 'S/R OPPS_CALC: Plume velocity went to zero at level ', k2 |
| 258 |
CALL PRINT_MESSAGE( msgBuf, standardMessageUnit, |
| 259 |
& SQUEEZE_RIGHT , 1) |
| 260 |
WRITE(msgBuf,'(A,4A14)') |
| 261 |
& 'S/R OPPS_CALC: ', 'wsqr', 'entrainrate', |
| 262 |
& '(Dd-De)/De up', '(Dd-De)/De do' |
| 263 |
CALL PRINT_MESSAGE( msgBuf, standardMessageUnit, |
| 264 |
& SQUEEZE_RIGHT , 1) |
| 265 |
WRITE(msgBuf,'(A,4E14.6)') |
| 266 |
& 'S/R OPPS_CALC: ', wsqr, entrainrate, |
| 267 |
& (Dd(k2)-De(k2))/De(k2), (Dd(k2+1)-De(k2+1))/De(k2+1) |
| 268 |
CALL PRINT_MESSAGE( msgBuf, standardMessageUnit, |
| 269 |
& SQUEEZE_RIGHT , 1) |
| 270 |
ENDIF |
| 271 |
#endif /* ALLOW_OPPS_DEBUG */ |
| 272 |
if(maxdepth.eq.k) goto 1000 |
| 273 |
goto 1 |
| 274 |
endif |
| 275 |
Wd(k2+1)=sqrt(wsqr) |
| 276 |
C |
| 277 |
C compute a new radius based on the new mass flux at this grid level |
| 278 |
C from Eq. (4) |
| 279 |
C |
| 280 |
radius=sqrt(newflux/(Wd(k2)*Dd(k2))) |
| 281 |
ELSE |
| 282 |
maxdepth=k2 |
| 283 |
if(maxdepth.eq.k) goto 1000 |
| 284 |
GOTO 1 |
| 285 |
ENDIF |
| 286 |
ENDDO |
| 287 |
c |
| 288 |
c plume has reached the bottom |
| 289 |
c |
| 290 |
MaxDepth=kMax |
| 291 |
c |
| 292 |
1 CONTINUE |
| 293 |
c |
| 294 |
Ad(k)=FRACTIONAL_AREA |
| 295 |
IC=0 |
| 296 |
c |
| 297 |
c start iteration on fractional area, not used in OGCM implementation |
| 298 |
c |
| 299 |
c |
| 300 |
DO IC=1,Max_ABE_Iterations |
| 301 |
c |
| 302 |
c |
| 303 |
c next compute the mass flux beteen each grid box using the entrainment |
| 304 |
c |
| 305 |
Md(k)=Wd(k)*Ad(k) |
| 306 |
c |
| 307 |
DO k2=k+1,maxDepth |
| 308 |
Md(k2)=Md(k)*PlumeEntrainment(k2) |
| 309 |
#ifdef ALLOW_OPPS_DEBUG |
| 310 |
IF ( OPPSdebugLevel.GE.debLevA ) THEN |
| 311 |
WRITE(msgBuf,'(A,2E12.4,I3)') |
| 312 |
& 'S/R OPPS_CALC: Md, Wd, and k are ',Md(k2),Wd(k2),k2 |
| 313 |
CALL PRINT_MESSAGE( msgBuf, standardMessageUnit, |
| 314 |
& SQUEEZE_RIGHT , 1) |
| 315 |
ENDIF |
| 316 |
#endif /* ALLOW_OPPS_DEBUG */ |
| 317 |
ENDDO |
| 318 |
c |
| 319 |
c Now move on to calculate new temperature using flux from |
| 320 |
c Td, Sd, Wd, ta, sa, and we. Values for these variables are at |
| 321 |
c center of grid cell, use weighted average to get boundary values |
| 322 |
c |
| 323 |
c use a timestep limited by the GCM model timestep and the maximum plume |
| 324 |
c velocity (CFL criteria) |
| 325 |
c |
| 326 |
c |
| 327 |
c calculate the weighted wd, td, and sd |
| 328 |
c |
| 329 |
dt = dtts |
| 330 |
do k2=k,maxDepth-1 |
| 331 |
IF ( Wd(K2) .NE. 0. _d 0 ) dt = min(dt,drF(k2)/Wd(k2)) |
| 332 |
c |
| 333 |
c time integration will be integer number of steps to get one |
| 334 |
c gcm time step |
| 335 |
c |
| 336 |
ntime = nint(0.5*int(dtts/dt)) |
| 337 |
if(ntime.eq.0) then |
| 338 |
ntime = 1 |
| 339 |
endif |
| 340 |
c |
| 341 |
c make sure area weighted vertical velocities match; in other words |
| 342 |
c make sure mass in equals mass out at the intersection of each grid |
| 343 |
c cell. Eq. (20) |
| 344 |
c |
| 345 |
mda(k2) = (md(k2)*drF(k2)+md(k2+1)*drF(k2+1))/ |
| 346 |
* (drF(k2)+drF(k2+1)) |
| 347 |
c |
| 348 |
wda(k2) = (wd(k2)*drF(k2)+wd(k2+1)*drF(k2+1))/ |
| 349 |
* (drF(k2)+drF(k2+1)) |
| 350 |
c |
| 351 |
DO ktr = 1, nTracerInUse |
| 352 |
Pda(k2,ktr) = Pd(k2,ktr) |
| 353 |
Paa(k2,ktr) = tracerEnv(k2+1,ktr) |
| 354 |
ENDDO |
| 355 |
c |
| 356 |
enddo |
| 357 |
dt = min(dt,dtts) |
| 358 |
#ifdef ALLOW_OPPS_DEBUG |
| 359 |
IF ( OPPSdebugLevel.GE.debLevA ) THEN |
| 360 |
WRITE(msgBuf,'(A,F14.4)') |
| 361 |
& 'S/R OPPS_CALC: time step = ', dt |
| 362 |
CALL PRINT_MESSAGE( msgBuf, standardMessageUnit, |
| 363 |
& SQUEEZE_RIGHT , 1) |
| 364 |
ENDIF |
| 365 |
#endif /* ALLOW_OPPS_DEBUG */ |
| 366 |
DO ktr=1,nTracerInUse |
| 367 |
Pda(maxdepth,ktr) = Pd(maxdepth,ktr) |
| 368 |
ENDDO |
| 369 |
C |
| 370 |
kmx = maxdepth-1 |
| 371 |
do nn=1,ntime |
| 372 |
C |
| 373 |
C top point |
| 374 |
C |
| 375 |
DO ktr = 1,nTracerInUse |
| 376 |
tracerEnv(k,ktr) = tracerEnv(k,ktr)- |
| 377 |
& (mda(k)*(Pda(k,ktr)-Paa(k,ktr)))*dt*recip_drF(k) |
| 378 |
ENDDO |
| 379 |
c |
| 380 |
c now do inner points if there are any |
| 381 |
c |
| 382 |
CML if(Maxdepth-k.gt.1) then |
| 383 |
CML This if statement is superfluous |
| 384 |
CML IF ( k .LT. Maxdepth-1 ) THEN |
| 385 |
CML DO k2=k+1,Maxdepth-1 |
| 386 |
CML mda(maxDepth) = 0. |
| 387 |
DO k2=k+1,kmx |
| 388 |
k2m1 = max(k,k2-1) |
| 389 |
k2p1 = max(k2+1,maxDepth) |
| 390 |
c |
| 391 |
DO ktr = 1,nTracerInUse |
| 392 |
tracerEnv(k2,ktr) = tracerEnv(k2,ktr) + |
| 393 |
& (mda(k2m1)*(Pda(k2m1,ktr)-Paa(k2m1,ktr)) |
| 394 |
& -mda(k2) *(Pda(k2,ktr) -Paa(k2,ktr)) ) |
| 395 |
& *dt*recip_drF(k2) |
| 396 |
ENDDO |
| 397 |
ENDDO |
| 398 |
CML This if statement is superfluous |
| 399 |
CML ENDIF |
| 400 |
C |
| 401 |
C bottom point |
| 402 |
C |
| 403 |
DO ktr=1,nTracerInUse |
| 404 |
tracerEnv(kmx+1,ktr) = tracerEnv(kmx+1,ktr)+ |
| 405 |
& mda(kmx)*(Pda(kmx,ktr)-Paa(kmx,ktr))*dt*recip_drF(kmx+1) |
| 406 |
ENDDO |
| 407 |
c |
| 408 |
c set the environmental temp and salinity to equal new fields |
| 409 |
c |
| 410 |
DO ktr=1,nTracerInUse |
| 411 |
DO k2=1,kmx |
| 412 |
paa(k2,ktr) = tracerEnv(k2+1,ktr) |
| 413 |
ENDDO |
| 414 |
ENDDO |
| 415 |
c |
| 416 |
c end loop on number of time integration steps |
| 417 |
c |
| 418 |
enddo |
| 419 |
ENDDO |
| 420 |
999 continue |
| 421 |
C |
| 422 |
C count convection event in this grid cell |
| 423 |
C |
| 424 |
OPPSconvectCount(I,J,K,bi,bj) = |
| 425 |
& OPPSconvectCount(I,J,K,bi,bj) + 1. _d 0 |
| 426 |
C |
| 427 |
C jump here if k = maxdepth or if level not unstable, go to next |
| 428 |
C profile point |
| 429 |
C |
| 430 |
1000 continue |
| 431 |
c |
| 432 |
C |
| 433 |
C end of k-loop |
| 434 |
C |
| 435 |
ENDDO |
| 436 |
|
| 437 |
C-- End IF (DIFFERENT_MULTIPLE) |
| 438 |
ENDIF |
| 439 |
|
| 440 |
RETURN |
| 441 |
END |
| 442 |
_RL FUNCTION STATE1(sLoc,tLoc,I,J,KREF,bi,bj,mythid) |
| 443 |
C !DESCRIPTION: \bv |
| 444 |
C *===============================================================* |
| 445 |
C | o SUBROUTINE STATE1 |
| 446 |
C | Calculates rho(S,T,p) |
| 447 |
C | It is absolutely necessary to compute |
| 448 |
C | the full rho and not sigma=rho-rhoConst, because |
| 449 |
C | density is used as a scale factor for fluxes and velocities |
| 450 |
C *===============================================================* |
| 451 |
C \ev |
| 452 |
|
| 453 |
C !USES: |
| 454 |
IMPLICIT NONE |
| 455 |
C == Global variables == |
| 456 |
#include "SIZE.h" |
| 457 |
#include "EEPARAMS.h" |
| 458 |
#include "PARAMS.h" |
| 459 |
#include "EOS.h" |
| 460 |
#include "GRID.h" |
| 461 |
#include "DYNVARS.h" |
| 462 |
|
| 463 |
C !INPUT/OUTPUT PARAMETERS: |
| 464 |
C == Routine arguments == |
| 465 |
INTEGER I,J,kRef,bi,bj,myThid |
| 466 |
_RL tLoc,sLoc |
| 467 |
|
| 468 |
C !LOCAL VARIABLES: |
| 469 |
C == Local variables == |
| 470 |
_RL rhoLoc, dRho |
| 471 |
_RL pLoc |
| 472 |
_RL t1, t2, t3, t4, s1, s3o2, p1, p2, sp5, p1t1 |
| 473 |
_RL rfresh, rsalt, rhoP0 |
| 474 |
_RL bMfresh, bMsalt, bMpres, BulkMod |
| 475 |
_RL rhoNum, rhoDen, den, epsln |
| 476 |
PARAMETER ( epsln = 0.D0 ) |
| 477 |
|
| 478 |
character*(max_len_mbuf) msgbuf |
| 479 |
|
| 480 |
CMLC estimate pressure from depth at cell centers |
| 481 |
CML mtoSI = gravity*rhoConst |
| 482 |
CML pLoc = ABS(rC(kRef))*mtoSI |
| 483 |
|
| 484 |
IF ( buoyancyRelation .EQ. 'OCEANIC' ) THEN |
| 485 |
C in Z coordinates the pressure is rho0 * (hydrostatic) Potential |
| 486 |
IF ( useDynP_inEos_Zc ) THEN |
| 487 |
C---------- |
| 488 |
C NOTE: For now, totPhiHyd only contains the Potential anomaly |
| 489 |
C since PhiRef is not available for Atmos and has not (yet) |
| 490 |
C been added in S/R DIAGS_PHI_HYD |
| 491 |
C---------- |
| 492 |
pLoc = rhoConst*( totPhiHyd(i,j,kRef,bi,bj) |
| 493 |
& -rC(kRef)*gravity |
| 494 |
& )*maskC(i,j,kRef,bi,bj) |
| 495 |
ELSE |
| 496 |
pLoc = -rhoConst*rC(kRef)*gravity*maskC(i,j,kRef,bi,bj) |
| 497 |
ENDIF |
| 498 |
ELSEIF ( buoyancyRelation .EQ. 'OCEANICP' ) THEN |
| 499 |
C in P coordinates the pressure is just the coordinate of |
| 500 |
C the tracer point |
| 501 |
pLoc = rC(kRef)* maskC(i,j,kRef,bi,bj) |
| 502 |
ENDIF |
| 503 |
|
| 504 |
rhoLoc = 0. _d 0 |
| 505 |
rhoP0 = 0. _d 0 |
| 506 |
bulkMod = 0. _d 0 |
| 507 |
rfresh = 0. _d 0 |
| 508 |
rsalt = 0. _d 0 |
| 509 |
bMfresh = 0. _d 0 |
| 510 |
bMsalt = 0. _d 0 |
| 511 |
bMpres = 0. _d 0 |
| 512 |
rhoNum = 0. _d 0 |
| 513 |
rhoDen = 0. _d 0 |
| 514 |
den = 0. _d 0 |
| 515 |
|
| 516 |
t1 = tLoc |
| 517 |
t2 = t1*t1 |
| 518 |
t3 = t2*t1 |
| 519 |
t4 = t3*t1 |
| 520 |
|
| 521 |
s1 = sLoc |
| 522 |
|
| 523 |
IF ( equationOfState .EQ. 'LINEAR' ) THEN |
| 524 |
|
| 525 |
dRho = rhoNil-rhoConst |
| 526 |
rhoLoc=rhoNil* ( |
| 527 |
& sBeta *(sLoc-sRef(kRef)) |
| 528 |
& - tAlpha*(tLoc-tRef(KREF)) ) + dRho |
| 529 |
|
| 530 |
ELSEIF (equationOfState.EQ.'POLY3') THEN |
| 531 |
|
| 532 |
C this is not correct, there is a field eosSig0 which should be use here |
| 533 |
C but I DO not intent to include the reference level in this routine |
| 534 |
WRITE(*,'(a)') |
| 535 |
& ' FIND_RHO_SCALAR: for POLY3, the density is not' |
| 536 |
WRITE(*,'(a)') |
| 537 |
& ' computed correctly in this routine' |
| 538 |
rhoLoc = 0. _d 0 |
| 539 |
|
| 540 |
ELSEIF ( equationOfState(1:5).EQ.'JMD95' |
| 541 |
& .OR. equationOfState.EQ.'UNESCO' ) THEN |
| 542 |
C nonlinear equation of state in pressure coordinates |
| 543 |
|
| 544 |
s3o2 = s1*SQRT(s1) |
| 545 |
|
| 546 |
p1 = pLoc*SItoBar |
| 547 |
p2 = p1*p1 |
| 548 |
|
| 549 |
C density of freshwater at the surface |
| 550 |
rfresh = |
| 551 |
& eosJMDCFw(1) |
| 552 |
& + eosJMDCFw(2)*t1 |
| 553 |
& + eosJMDCFw(3)*t2 |
| 554 |
& + eosJMDCFw(4)*t3 |
| 555 |
& + eosJMDCFw(5)*t4 |
| 556 |
& + eosJMDCFw(6)*t4*t1 |
| 557 |
C density of sea water at the surface |
| 558 |
rsalt = |
| 559 |
& s1*( |
| 560 |
& eosJMDCSw(1) |
| 561 |
& + eosJMDCSw(2)*t1 |
| 562 |
& + eosJMDCSw(3)*t2 |
| 563 |
& + eosJMDCSw(4)*t3 |
| 564 |
& + eosJMDCSw(5)*t4 |
| 565 |
& ) |
| 566 |
& + s3o2*( |
| 567 |
& eosJMDCSw(6) |
| 568 |
& + eosJMDCSw(7)*t1 |
| 569 |
& + eosJMDCSw(8)*t2 |
| 570 |
& ) |
| 571 |
& + eosJMDCSw(9)*s1*s1 |
| 572 |
|
| 573 |
rhoP0 = rfresh + rsalt |
| 574 |
|
| 575 |
C secant bulk modulus of fresh water at the surface |
| 576 |
bMfresh = |
| 577 |
& eosJMDCKFw(1) |
| 578 |
& + eosJMDCKFw(2)*t1 |
| 579 |
& + eosJMDCKFw(3)*t2 |
| 580 |
& + eosJMDCKFw(4)*t3 |
| 581 |
& + eosJMDCKFw(5)*t4 |
| 582 |
C secant bulk modulus of sea water at the surface |
| 583 |
bMsalt = |
| 584 |
& s1*( eosJMDCKSw(1) |
| 585 |
& + eosJMDCKSw(2)*t1 |
| 586 |
& + eosJMDCKSw(3)*t2 |
| 587 |
& + eosJMDCKSw(4)*t3 |
| 588 |
& ) |
| 589 |
& + s3o2*( eosJMDCKSw(5) |
| 590 |
& + eosJMDCKSw(6)*t1 |
| 591 |
& + eosJMDCKSw(7)*t2 |
| 592 |
& ) |
| 593 |
C secant bulk modulus of sea water at pressure p |
| 594 |
bMpres = |
| 595 |
& p1*( eosJMDCKP(1) |
| 596 |
& + eosJMDCKP(2)*t1 |
| 597 |
& + eosJMDCKP(3)*t2 |
| 598 |
& + eosJMDCKP(4)*t3 |
| 599 |
& ) |
| 600 |
& + p1*s1*( eosJMDCKP(5) |
| 601 |
& + eosJMDCKP(6)*t1 |
| 602 |
& + eosJMDCKP(7)*t2 |
| 603 |
& ) |
| 604 |
& + p1*s3o2*eosJMDCKP(8) |
| 605 |
& + p2*( eosJMDCKP(9) |
| 606 |
& + eosJMDCKP(10)*t1 |
| 607 |
& + eosJMDCKP(11)*t2 |
| 608 |
& ) |
| 609 |
& + p2*s1*( eosJMDCKP(12) |
| 610 |
& + eosJMDCKP(13)*t1 |
| 611 |
& + eosJMDCKP(14)*t2 |
| 612 |
& ) |
| 613 |
|
| 614 |
bulkMod = bMfresh + bMsalt + bMpres |
| 615 |
|
| 616 |
C density of sea water at pressure p |
| 617 |
rhoLoc = rhoP0/(1. _d 0 - p1/bulkMod) - rhoConst |
| 618 |
|
| 619 |
ELSEIF ( equationOfState.EQ.'MDJWF' ) THEN |
| 620 |
|
| 621 |
sp5 = SQRT(s1) |
| 622 |
|
| 623 |
p1 = pLoc*SItodBar |
| 624 |
p1t1 = p1*t1 |
| 625 |
|
| 626 |
rhoNum = eosMDJWFnum(0) |
| 627 |
& + t1*(eosMDJWFnum(1) |
| 628 |
& + t1*(eosMDJWFnum(2) + eosMDJWFnum(3)*t1) ) |
| 629 |
& + s1*(eosMDJWFnum(4) |
| 630 |
& + eosMDJWFnum(5)*t1 + eosMDJWFnum(6)*s1) |
| 631 |
& + p1*(eosMDJWFnum(7) + eosMDJWFnum(8)*t2 |
| 632 |
& + eosMDJWFnum(9)*s1 |
| 633 |
& + p1*(eosMDJWFnum(10) + eosMDJWFnum(11)*t2) ) |
| 634 |
|
| 635 |
|
| 636 |
den = eosMDJWFden(0) |
| 637 |
& + t1*(eosMDJWFden(1) |
| 638 |
& + t1*(eosMDJWFden(2) |
| 639 |
& + t1*(eosMDJWFden(3) + t1*eosMDJWFden(4) ) ) ) |
| 640 |
& + s1*(eosMDJWFden(5) |
| 641 |
& + t1*(eosMDJWFden(6) |
| 642 |
& + eosMDJWFden(7)*t2) |
| 643 |
& + sp5*(eosMDJWFden(8) + eosMDJWFden(9)*t2) ) |
| 644 |
& + p1*(eosMDJWFden(10) |
| 645 |
& + p1t1*(eosMDJWFden(11)*t2 + eosMDJWFden(12)*p1) ) |
| 646 |
|
| 647 |
rhoDen = 1.0/(epsln+den) |
| 648 |
|
| 649 |
rhoLoc = rhoNum*rhoDen - rhoConst |
| 650 |
|
| 651 |
ELSEIF( equationOfState .EQ. 'IDEALG' ) THEN |
| 652 |
C |
| 653 |
ELSE |
| 654 |
WRITE(msgbuf,'(3A)') |
| 655 |
& ' STATE1 : equationOfState = "', |
| 656 |
& equationOfState,'"' |
| 657 |
CALL PRINT_ERROR( msgbuf, mythid ) |
| 658 |
STOP 'ABNORMAL END: S/R STATE1 in OPPS_CALC' |
| 659 |
ENDIF |
| 660 |
|
| 661 |
state1 = rhoLoc + rhoConst |
| 662 |
|
| 663 |
#endif /* ALLOW_OPPS */ |
| 664 |
RETURN |
| 665 |
END |
| 666 |
|
| 667 |
|
| 668 |
#undef OPPS_ORGCODE |
| 669 |
#ifdef OPPS_ORGCODE |
| 670 |
c Listed below is the subroutine for use in parallel 3-d circulation code. |
| 671 |
c It has been used in the parallel semtner-chervin code and is now being used |
| 672 |
c In the POP code. The subroutine is called nlopps (long story to explain why). |
| 673 |
|
| 674 |
c I've attached the version of lopps that we've been using in the simulations. |
| 675 |
c There is one common block that is different from the standard model commons |
| 676 |
c (countc) and it is not needed if the array convadj is not used. The routine |
| 677 |
c does need "kmp" which is why the boundc common is included. For massively |
| 678 |
c parallel codes (like POP) we think this will work well when converted from a |
| 679 |
c "slab" (i=is,ie) to a column, which just means removing the "do i=is,ie" loop. c There are differences between this |
| 680 |
c code and the 1-d code and the earlier scheme implemented in 3-d models. These c differences are described below. |
| 681 |
|
| 682 |
|
| 683 |
subroutine nlopps(j,is,ie,ta,sa,gcmdz) |
| 684 |
c |
| 685 |
parameter (imt = 361 , jmt = 301 , km = 30 ) |
| 686 |
c |
| 687 |
c Nlopps: E. Skyllingstad and T. Paluszkiewicz |
| 688 |
c |
| 689 |
c Version: December 11, 1996 |
| 690 |
c |
| 691 |
c Nlopps: This version of lopps is significantly different from |
| 692 |
c the original code developed by R. Romea and T. Paluskiewicz. The |
| 693 |
c code uses a flux constraint to control the change in T and S at |
| 694 |
c each grid level. First, a plume profile of T,S, and W are |
| 695 |
c determined using the standard plume model, but with a detraining |
| 696 |
c mass instead of entraining. Thus, the T and S plume |
| 697 |
c characteristics still change, but the plume contracts in size |
| 698 |
c rather than expanding ala classical entraining plumes. This |
| 699 |
c is heuristically more in line with large eddy simulation results. |
| 700 |
c At each grid level, the convergence of plume velocity determines |
| 701 |
c the flux of T and S, which is conserved by using an upstream |
| 702 |
c advection. The vertical velocity is balanced so that the area |
| 703 |
c weighted upward velocity equals the area weighted downdraft |
| 704 |
c velocity, ensuring mass conservation. The present implementation |
| 705 |
c adjusts the plume for a time period equal to the time for 1/2 of |
| 706 |
c the mass of the fastest moving level to move downward. As a |
| 707 |
c consequence, the model does not completely adjust the profile at |
| 708 |
c each model time step, but provides a smooth adjustment over time. |
| 709 |
c |
| 710 |
c |
| 711 |
|
| 712 |
|
| 713 |
c |
| 714 |
|
| 715 |
c include "params.h" |
| 716 |
c include "plume_fast_inc.h" |
| 717 |
c include "plume_fast.h" |
| 718 |
c #include "loppsd.h" |
| 719 |
|
| 720 |
real ta(imt,km),sa(imt,km),gcmdz(km),dz(km) |
| 721 |
real pdensity,wsqr,oldflux,newflux,entrainrate,adtemp |
| 722 |
REAL Del,D,dza1,dza2,kd,kd1,Smix,Thmix,PlumeS,PlumeT,PlumeD |
| 723 |
c |
| 724 |
c |
| 725 |
|
| 726 |
INTEGER i,j,k |
| 727 |
clfh |
| 728 |
integer is,ie,k2 |
| 729 |
clfh |
| 730 |
REAL D1,D2,state1,Density |
| 731 |
REAL dz1,dz2 |
| 732 |
REAL radius,StartingFlux |
| 733 |
real ttemp(km),stemp(km),taa(km),saa(km) |
| 734 |
real wda(km),tda(km),sda(km),mda(km) |
| 735 |
real dtts,dt,sumo,sumn |
| 736 |
integer ntime,nn,kmx,ic |
| 737 |
c |
| 738 |
c |
| 739 |
LOGICAL debug,done |
| 740 |
INTEGER MAX_ABE_ITERATIONS |
| 741 |
PARAMETER(MAX_ABE_ITERATIONS=1) |
| 742 |
REAL PlumeRadius |
| 743 |
REAL STABILITY_THRESHOLD |
| 744 |
REAL FRACTIONAL_AREA |
| 745 |
REAL MAX_FRACTIONAL_AREA |
| 746 |
REAL VERTICAL_VELOCITY |
| 747 |
REAL ENTRAINMENT_RATE |
| 748 |
REAL e2 |
| 749 |
PARAMETER ( PlumeRadius = 100.D0 ) |
| 750 |
PARAMETER ( STABILITY_THRESHOLD = -1.E-4 ) |
| 751 |
PARAMETER ( FRACTIONAL_AREA = .1E0 ) |
| 752 |
PARAMETER ( MAX_FRACTIONAL_AREA = .8E0 ) |
| 753 |
PARAMETER ( VERTICAL_VELOCITY = .02E0 ) |
| 754 |
PARAMETER ( ENTRAINMENT_RATE = -.05E0 ) |
| 755 |
PARAMETER ( e2 = 2.E0*ENTRAINMENT_RATE ) |
| 756 |
! Arrays. |
| 757 |
REAL Ad(km),Sd(km),Td(km),Wd(km),Dd(km),Md(km) |
| 758 |
REAL Se(km),Te(km),We(km),De(km) |
| 759 |
REAL PlumeEntrainment(km) |
| 760 |
REAL GridThickness(km) |
| 761 |
|
| 762 |
c |
| 763 |
c input kmp through a common block |
| 764 |
c |
| 765 |
common / boundc / wsx(imt,jmt),wsy(imt,jmt),hfs(imt,jmt), |
| 766 |
1 ple(imt,jmt),kmp(imt,jmt),kmq(imt,jmt) |
| 767 |
cwmseas |
| 768 |
& ,wsx1(imt,jmt),wsy1(imt,jmt) |
| 769 |
1 ,wsx2(imt,jmt),wsy2(imt,jmt) |
| 770 |
c |
| 771 |
c input the variables through a common |
| 772 |
c |
| 773 |
logical problem |
| 774 |
common /countc/ convadj(imt,jmt,km),ics,depth(km),problem |
| 775 |
|
| 776 |
|
| 777 |
c-----may want to setup an option to get this only on first call |
| 778 |
c otherwise it is repetive |
| 779 |
c griddz is initialize by call to setupgrid |
| 780 |
c |
| 781 |
c |
| 782 |
|
| 783 |
dtts = 2400 |
| 784 |
c |
| 785 |
do k=1,km |
| 786 |
dz(k) = 0.01*gcmdz(k) |
| 787 |
enddo |
| 788 |
c |
| 789 |
do k=1,km |
| 790 |
GridThickness(k) = dz(k) |
| 791 |
enddo |
| 792 |
c |
| 793 |
c modified to loop over slab |
| 794 |
c |
| 795 |
DO i=is,ie |
| 796 |
|
| 797 |
numgridpoints=kmp(i,j) |
| 798 |
c |
| 799 |
c go to next column if only 1 grid point or on land |
| 800 |
c |
| 801 |
if(numgridpoints.le.1) goto 1100 |
| 802 |
c |
| 803 |
c loop over depth |
| 804 |
c |
| 805 |
c |
| 806 |
debug = .false. |
| 807 |
c |
| 808 |
c first save copy of initial profile |
| 809 |
c |
| 810 |
DO k=1,NumGridPoints |
| 811 |
stemp(k)=sa(i,k) |
| 812 |
ttemp(k)=ta(i,k) |
| 813 |
c |
| 814 |
c do a check of t and s range, if out of bounds set flag |
| 815 |
c |
| 816 |
if(problem) then |
| 817 |
write(0,*)"Code in trouble before this nlopps call" |
| 818 |
return |
| 819 |
endif |
| 820 |
c |
| 821 |
if(sa(i,k).gt.40..or.ta(i,k).lt.-4.0) then |
| 822 |
problem = .true. |
| 823 |
write(0,*)"t out of range at j ",j |
| 824 |
debug = .true. |
| 825 |
return |
| 826 |
endif |
| 827 |
ENDDO |
| 828 |
|
| 829 |
if(debug) then |
| 830 |
write(*,*)"T and S Profile at ",i,j |
| 831 |
write(*,*)(ta(i,k),sa(i,k),k=1,NumGridPoints) |
| 832 |
endif |
| 833 |
|
| 834 |
DO k=1,NumGridPoints-1 |
| 835 |
c |
| 836 |
c initialize the plume T,S,density, and w velocity |
| 837 |
c |
| 838 |
Sd(k)=stemp(k) |
| 839 |
Td(k)=ttemp(k) |
| 840 |
Dd(k)=state1(stemp(k),ttemp(k),k) |
| 841 |
De(k)=Dd(k) |
| 842 |
c Wd(k)=VERTICAL_VELOCITY |
| 843 |
c |
| 844 |
c guess at initial top grid cell vertical velocity |
| 845 |
c |
| 846 |
Wd(k) = 0.03 |
| 847 |
c |
| 848 |
c these estimates of initial plume velocity based on plume size and |
| 849 |
c top grid cell water mass |
| 850 |
c |
| 851 |
c Wd(k) = 0.5*dz(k)/(dtts*FRACTIONAL_AREA) |
| 852 |
c Wd(k) = 0.5*dz(k)/dtts |
| 853 |
c |
| 854 |
wsqr=Wd(k)*Wd(k) |
| 855 |
PlumeEntrainment(k) = 0.0 |
| 856 |
c |
| 857 |
c |
| 858 |
c |
| 859 |
if(debug) write(0,*) 'Doing old lowerparcel' |
| 860 |
radius=PlumeRadius |
| 861 |
StartingFlux=radius*radius*Wd(k)*Dd(k) |
| 862 |
oldflux=StartingFlux |
| 863 |
|
| 864 |
dz2=GridThickness(k) |
| 865 |
DO k2=k,NumGridPoints-1 |
| 866 |
D1=state1(Sd(k2),Td(k2),k2+1) |
| 867 |
D2=state1(stemp(k2+1),ttemp(k2+1),k2+1) |
| 868 |
De(k2+1)=D2 |
| 869 |
c |
| 870 |
c To start downward, parcel has to initially be heavier than environment |
| 871 |
c but after it has started moving, we continue plume until plume tke or |
| 872 |
c flux goes negative |
| 873 |
c |
| 874 |
IF (D2-D1 .LT. STABILITY_THRESHOLD.or.k2.ne.k) THEN |
| 875 |
dz1=dz2 |
| 876 |
dz2=GridThickness(k2+1) |
| 877 |
c |
| 878 |
c define mass flux according to eq. 4 from paper |
| 879 |
c |
| 880 |
newflux=oldflux+e2*radius*Wd(k2)*Dd(k2)*0.50* |
| 881 |
. (dz1+dz2) |
| 882 |
c |
| 883 |
PlumeEntrainment(k2+1) = newflux/StartingFlux |
| 884 |
c |
| 885 |
IF(newflux.LT.0.0) then |
| 886 |
if(debug) then |
| 887 |
write(0,*)"Plume entrained to zero at ",k2 |
| 888 |
endif |
| 889 |
maxdepth = k2 |
| 890 |
if(maxdepth.eq.k) goto 1000 |
| 891 |
goto 1 |
| 892 |
endif |
| 893 |
c |
| 894 |
c entrainment rate is basically a scaled mass flux dM/M |
| 895 |
c |
| 896 |
entrainrate = (newflux - oldflux)/newflux |
| 897 |
oldflux = newflux |
| 898 |
c |
| 899 |
c |
| 900 |
c mix var's are the average environmental values over the two grid levels |
| 901 |
c |
| 902 |
smix=(dz1*stemp(k2)+dz2*stemp(k2+1))/(dz1+dz2) |
| 903 |
thmix=(dz1*ttemp(k2)+dz2*ttemp(k2+1))/(dz1+dz2) |
| 904 |
c |
| 905 |
c first compute the new salinity and temperature for this level |
| 906 |
c using equations 3.6 and 3.7 from the paper |
| 907 |
c |
| 908 |
c |
| 909 |
c |
| 910 |
sd(k2+1)=sd(k2) - entrainrate*(smix - sd(k2)) |
| 911 |
td(k2+1)=td(k2) - entrainrate*(thmix - td(k2)) |
| 912 |
c |
| 913 |
c |
| 914 |
c compute the density at this level for the buoyancy term in the |
| 915 |
c vertical k.e. equation |
| 916 |
c |
| 917 |
Dd(k2+1)=state1(Sd(k2+1),Td(k2+1),k2+1) |
| 918 |
c |
| 919 |
c next, solve for the vertical velocity k.e. using combined eq. 4 |
| 920 |
c and eq 5 from the paper |
| 921 |
c |
| 922 |
if(debug) then |
| 923 |
write(0,*)"Dd,De,entr,k ",Dd(k2),De(k2),entrainrate,k2 |
| 924 |
endif |
| 925 |
c |
| 926 |
wsqr = wsqr - wsqr*abs(entrainrate)+ 9.81* |
| 927 |
. (dz1*(Dd(k2)-De(k2))/De(k2) |
| 928 |
. +dz2*(Dd(k2+1)-De(k2+1))/De(k2+1)) |
| 929 |
c |
| 930 |
c if negative k.e. then plume has reached max depth, get out of loop |
| 931 |
c |
| 932 |
IF(wsqr.LT.0.0)then |
| 933 |
maxdepth = k2 |
| 934 |
if(debug) then |
| 935 |
write(0,*)"Plume velocity went to zero at ",k2 |
| 936 |
endif |
| 937 |
if(maxdepth.eq.k) goto 1000 |
| 938 |
goto 1 |
| 939 |
endif |
| 940 |
Wd(k2+1)=sqrt(wsqr) |
| 941 |
c |
| 942 |
c compute a new radius based on the new mass flux at this grid level |
| 943 |
c |
| 944 |
radius=sqrt(newflux/(Wd(k2)*Dd(k2))) |
| 945 |
ELSE |
| 946 |
maxdepth=k2 |
| 947 |
if(maxdepth.eq.k) goto 1000 |
| 948 |
GOTO 1 |
| 949 |
ENDIF |
| 950 |
ENDDO |
| 951 |
c |
| 952 |
c plume has reached the bottom |
| 953 |
c |
| 954 |
MaxDepth=NumGridPoints |
| 955 |
c |
| 956 |
1 continue |
| 957 |
c |
| 958 |
Ad(k)=FRACTIONAL_AREA |
| 959 |
IC=0 |
| 960 |
c |
| 961 |
c start iteration on fractional area, not used in OGCM implementation |
| 962 |
c |
| 963 |
c |
| 964 |
DO IC=1,Max_ABE_Iterations |
| 965 |
c |
| 966 |
c |
| 967 |
c next compute the mass flux beteen each grid box using the entrainment |
| 968 |
c |
| 969 |
92 continue |
| 970 |
Md(k)=Wd(k)*Ad(k) |
| 971 |
c |
| 972 |
DO k2=k+1,MaxDepth |
| 973 |
Md(k2)=Md(k)*PlumeEntrainment(k2) |
| 974 |
if(debug) then |
| 975 |
write(0,*)"Md, Wd, and k are ",Md(k2),Wd(k2),k2 |
| 976 |
endif |
| 977 |
ENDDO |
| 978 |
c |
| 979 |
c Now move on to calculate new temperature using flux from |
| 980 |
c Td, Sd, Wd, ta, sa, and we. Values for these variables are at |
| 981 |
c center of grid cell, use weighted average to get boundary values |
| 982 |
c |
| 983 |
c use a timestep limited by the GCM model timestep and the maximum plume |
| 984 |
c velocity (CFL criteria) |
| 985 |
c |
| 986 |
c |
| 987 |
c calculate the weighted wd, td, and sd |
| 988 |
c |
| 989 |
dt = dtts |
| 990 |
do k2=k,maxdepth-1 |
| 991 |
dt = min(dt,dz(k2)/wd(k2)) |
| 992 |
c |
| 993 |
c time integration will be integer number of steps to get one |
| 994 |
c gcm time step |
| 995 |
c |
| 996 |
ntime = nint(0.5*int(dtts/dt)) |
| 997 |
if(ntime.eq.0) then |
| 998 |
ntime = 1 |
| 999 |
endif |
| 1000 |
c |
| 1001 |
c make sure area weighted vertical velocities match; in other words |
| 1002 |
c make sure mass in equals mass out at the intersection of each grid |
| 1003 |
c cell. |
| 1004 |
c |
| 1005 |
mda(k2) = (md(k2)*dz(k2)+md(k2+1)*dz(k2+1))/ |
| 1006 |
* (dz(k2)+dz(k2+1)) |
| 1007 |
c |
| 1008 |
wda(k2) = (wd(k2)*dz(k2)+wd(k2+1)*dz(k2+1))/ |
| 1009 |
* (dz(k2)+dz(k2+1)) |
| 1010 |
c |
| 1011 |
tda(k2) = td(k2) |
| 1012 |
sda(k2) = sd(k2) |
| 1013 |
c |
| 1014 |
taa(k2) = ttemp(k2+1) |
| 1015 |
saa(k2) = stemp(k2+1) |
| 1016 |
c |
| 1017 |
enddo |
| 1018 |
dt = min(dt,dtts) |
| 1019 |
if(debug) then |
| 1020 |
write(0,*)"Time step is ", dt |
| 1021 |
endif |
| 1022 |
tda(maxdepth) = td(maxdepth) |
| 1023 |
sda(maxdepth) = sd(maxdepth) |
| 1024 |
c |
| 1025 |
c do top and bottom points first |
| 1026 |
c |
| 1027 |
kmx = maxdepth-1 |
| 1028 |
do nn=1,ntime |
| 1029 |
|
| 1030 |
ttemp(k) = ttemp(k)- |
| 1031 |
* (mda(k)*(tda(k)-taa(k)))*dt*recip_drF(k) |
| 1032 |
c |
| 1033 |
stemp(k) = stemp(k)- |
| 1034 |
* (mda(k)*(sda(k)-saa(k)))*dt*recip_drF(k) |
| 1035 |
c |
| 1036 |
c |
| 1037 |
c now do inner points if there are any |
| 1038 |
c |
| 1039 |
if(Maxdepth-k.gt.1) then |
| 1040 |
do k2=k+1,Maxdepth-1 |
| 1041 |
c |
| 1042 |
ttemp(k2) = ttemp(k2) + |
| 1043 |
* (mda(k2-1)*(tda(k2-1)-taa(k2-1))- |
| 1044 |
* mda(k2)*(tda(k2)-taa(k2))) |
| 1045 |
* *dt*recip_drF(k2) |
| 1046 |
|
| 1047 |
c |
| 1048 |
stemp(k2) = stemp(k2) + |
| 1049 |
* (mda(k2-1)*(sda(k2-1)-saa(k2-1))- |
| 1050 |
* mda(k2)*(sda(k2)-saa(k2))) |
| 1051 |
* *dt*recip_drF(k2) |
| 1052 |
|
| 1053 |
c |
| 1054 |
enddo |
| 1055 |
endif |
| 1056 |
ttemp(kmx+1) = ttemp(kmx+1)+ |
| 1057 |
* (mda(kmx)*(tda(kmx)-taa(kmx)))* |
| 1058 |
* dt*recip_drF(kmx+1) |
| 1059 |
c |
| 1060 |
stemp(kmx+1) = stemp(kmx+1)+ |
| 1061 |
* (mda(kmx)*(sda(kmx)-saa(kmx)))* |
| 1062 |
* dt*recip_drF(kmx+1) |
| 1063 |
c |
| 1064 |
c set the environmental temp and salinity to equal new fields |
| 1065 |
c |
| 1066 |
do k2=1,maxdepth-1 |
| 1067 |
taa(k2) = ttemp(k2+1) |
| 1068 |
saa(k2) = stemp(k2+1) |
| 1069 |
enddo |
| 1070 |
c |
| 1071 |
c end loop on number of time integration steps |
| 1072 |
c |
| 1073 |
enddo |
| 1074 |
ENDDO |
| 1075 |
999 continue |
| 1076 |
c |
| 1077 |
c assume that it converged, so update the ta and sa with new fields |
| 1078 |
c |
| 1079 |
c if(i.gt.180.and.j.gt.200.and.i.lt.240) then |
| 1080 |
c write(*,*)"Converged ",i,j,k,maxdepth,ttemp(k+1),ta(i,k+1) |
| 1081 |
c endif |
| 1082 |
do k2=k,maxdepth |
| 1083 |
convadj(i,j,k2) = convadj(i,j,k2) + (ttemp(k2)- |
| 1084 |
* ta(i,k2)) |
| 1085 |
sa(i,k2) = stemp(k2) |
| 1086 |
ta(i,k2) = ttemp(k2) |
| 1087 |
c if(i.gt.180.and.j.gt.200.and.i.lt.240) then |
| 1088 |
c write(*,*)"convadj ",convadj(i,j,k2) |
| 1089 |
c endif |
| 1090 |
c |
| 1091 |
c see if nlopps messed up |
| 1092 |
c |
| 1093 |
if(sa(i,k).gt.40..or.ta(i,k).lt.-4.0) then |
| 1094 |
problem = .true. |
| 1095 |
write(0,*)"t out of range at j after adjust",j |
| 1096 |
debug = .true. |
| 1097 |
endif |
| 1098 |
c |
| 1099 |
enddo |
| 1100 |
c |
| 1101 |
c jump here if k = maxdepth or if level not unstable, go to next |
| 1102 |
c profile point |
| 1103 |
c |
| 1104 |
1000 continue |
| 1105 |
c |
| 1106 |
c |
| 1107 |
c end loop on k, move on to next possible plume |
| 1108 |
c |
| 1109 |
ENDDO |
| 1110 |
1100 continue |
| 1111 |
c |
| 1112 |
c i loop |
| 1113 |
c |
| 1114 |
ENDDO |
| 1115 |
END |
| 1116 |
|
| 1117 |
#endif /* OPPS_ORGCODE */ |
Parent Directory
| 
Revision Log
| 
Revision Graph