| 1 |
C $Header: /u/gcmpack/MITgcm/pkg/kpp/kpp_calc.F,v 1.35 2007/04/19 15:49:11 mlosch Exp $ |
| 2 |
C $Name: $ |
| 3 |
|
| 4 |
#include "KPP_OPTIONS.h" |
| 5 |
|
| 6 |
CBOP |
| 7 |
C !ROUTINE: KPP_CALC |
| 8 |
|
| 9 |
C !INTERFACE: ========================================================== |
| 10 |
subroutine KPP_CALC( |
| 11 |
I bi, bj, myTime, myThid ) |
| 12 |
|
| 13 |
C !DESCRIPTION: \bv |
| 14 |
C /==========================================================\ |
| 15 |
C | SUBROUTINE KPP_CALC | |
| 16 |
C | o Compute all KPP fields defined in KPP.h | |
| 17 |
C |==========================================================| |
| 18 |
C | This subroutine serves as an interface between MITGCMUV | |
| 19 |
C | code and NCOM 1-D routines in kpp_routines.F | |
| 20 |
C \==========================================================/ |
| 21 |
IMPLICIT NONE |
| 22 |
|
| 23 |
c======================================================================= |
| 24 |
c |
| 25 |
c written by : jan morzel, august 11, 1994 |
| 26 |
c modified by : jan morzel, january 25, 1995 : "dVsq" and 1d code |
| 27 |
c detlef stammer, august, 1997 : for MIT GCM Classic |
| 28 |
c d. menemenlis, july, 1998 : for MIT GCM UV |
| 29 |
c |
| 30 |
c compute vertical mixing coefficients based on the k-profile |
| 31 |
c and oceanic planetary boundary layer scheme by large & mcwilliams. |
| 32 |
c |
| 33 |
c summary: |
| 34 |
c - compute interior mixing everywhere: |
| 35 |
c interior mixing gets computed at all interfaces due to constant |
| 36 |
c internal wave background activity ("fkpm" and "fkph"), which |
| 37 |
c is enhanced in places of static instability (local richardson |
| 38 |
c number < 0). |
| 39 |
c Additionally, mixing can be enhanced by adding contribution due |
| 40 |
c to shear instability which is a function of the local richardson |
| 41 |
c number |
| 42 |
c - double diffusivity: |
| 43 |
c interior mixing can be enhanced by double diffusion due to salt |
| 44 |
c fingering and diffusive convection (ifdef "kmixdd"). |
| 45 |
c - kpp scheme in the boundary layer: |
| 46 |
c |
| 47 |
c a.boundary layer depth: |
| 48 |
c at every gridpoint the depth of the oceanic boundary layer |
| 49 |
c ("hbl") gets computed by evaluating bulk richardson numbers. |
| 50 |
c b.boundary layer mixing: |
| 51 |
c within the boundary layer, above hbl, vertical mixing is |
| 52 |
c determined by turbulent surface fluxes, and interior mixing at |
| 53 |
c the lower boundary, i.e. at hbl. |
| 54 |
c |
| 55 |
c this subroutine provides the interface between the MIT GCM UV and the |
| 56 |
c subroutine "kppmix", where boundary layer depth, vertical |
| 57 |
c viscosity, vertical diffusivity, and counter gradient term (ghat) |
| 58 |
c are computed slabwise. |
| 59 |
c note: subroutine "kppmix" uses m-k-s units. |
| 60 |
c |
| 61 |
c time level: |
| 62 |
c input tracer and velocity profiles are evaluated at time level |
| 63 |
c tau, surface fluxes come from tau or tau-1. |
| 64 |
c |
| 65 |
c grid option: |
| 66 |
c in this "1-grid" implementation, diffusivity and viscosity |
| 67 |
c profiles are computed on the "t-grid" (by using velocity shear |
| 68 |
c profiles averaged from the "u,v-grid" onto the "t-grid"; note, that |
| 69 |
c the averaging includes zero values on coastal and seafloor grid |
| 70 |
c points). viscosity on the "u,v-grid" is computed by averaging the |
| 71 |
c "t-grid" viscosity values onto the "u,v-grid". |
| 72 |
c |
| 73 |
c vertical grid: |
| 74 |
c mixing coefficients get evaluated at the bottom of the lowest |
| 75 |
c layer, i.e., at depth zw(Nr). these values are only useful when |
| 76 |
c the model ocean domain does not include the entire ocean down to |
| 77 |
c the seafloor ("upperocean" setup) and allows flux through the |
| 78 |
c bottom of the domain. for full-depth runs, these mixing |
| 79 |
c coefficients are being zeroed out before leaving this subroutine. |
| 80 |
c |
| 81 |
c------------------------------------------------------------------------- |
| 82 |
|
| 83 |
c global parameters updated by kpp_calc |
| 84 |
c KPPviscAz - KPP eddy viscosity coefficient (m^2/s) |
| 85 |
c KPPdiffKzT - KPP diffusion coefficient for temperature (m^2/s) |
| 86 |
c KPPdiffKzS - KPP diffusion coefficient for salt and tracers (m^2/s) |
| 87 |
c KPPghat - Nonlocal transport coefficient (s/m^2) |
| 88 |
c KPPhbl - Boundary layer depth on "t-grid" (m) |
| 89 |
c KPPfrac - Fraction of short-wave flux penetrating mixing layer |
| 90 |
|
| 91 |
c-- KPP_CALC computes vertical viscosity and diffusivity for region |
| 92 |
c (-2:sNx+3,-2:sNy+3) as required by CALC_DIFFUSIVITY and requires |
| 93 |
c values of uVel, vVel, surfaceForcingU, surfaceForcingV in the |
| 94 |
c region (-2:sNx+4,-2:sNy+4). |
| 95 |
c Hence overlap region needs to be set OLx=4, OLy=4. |
| 96 |
c \ev |
| 97 |
|
| 98 |
C !USES: =============================================================== |
| 99 |
#include "SIZE.h" |
| 100 |
#include "EEPARAMS.h" |
| 101 |
#include "PARAMS.h" |
| 102 |
#include "DYNVARS.h" |
| 103 |
#include "KPP.h" |
| 104 |
#include "KPP_PARAMS.h" |
| 105 |
#include "FFIELDS.h" |
| 106 |
#include "GRID.h" |
| 107 |
#include "GAD.h" |
| 108 |
#ifdef ALLOW_SHELFICE |
| 109 |
# include "SHELFICE.h" |
| 110 |
#endif /* ALLOW_SHELFICE */ |
| 111 |
#ifdef ALLOW_AUTODIFF_TAMC |
| 112 |
#include "tamc.h" |
| 113 |
#include "tamc_keys.h" |
| 114 |
#else /* ALLOW_AUTODIFF_TAMC */ |
| 115 |
integer ikppkey |
| 116 |
#endif /* ALLOW_AUTODIFF_TAMC */ |
| 117 |
|
| 118 |
EXTERNAL DIFFERENT_MULTIPLE |
| 119 |
LOGICAL DIFFERENT_MULTIPLE |
| 120 |
|
| 121 |
C !INPUT PARAMETERS: =================================================== |
| 122 |
c Routine arguments |
| 123 |
c bi, bj - array indices on which to apply calculations |
| 124 |
c myTime - Current time in simulation |
| 125 |
|
| 126 |
INTEGER bi, bj |
| 127 |
INTEGER myThid |
| 128 |
_RL myTime |
| 129 |
|
| 130 |
#ifdef ALLOW_KPP |
| 131 |
|
| 132 |
C !LOCAL VARIABLES: ==================================================== |
| 133 |
c Local constants |
| 134 |
c minusone, p0, p5, p25, p125, p0625 |
| 135 |
c imin, imax, jmin, jmax - array computation indices |
| 136 |
|
| 137 |
_RL minusone |
| 138 |
parameter( minusone=-1.0) |
| 139 |
_KPP_RL p0 , p5 , p25 , p125 , p0625 |
| 140 |
parameter( p0=0.0, p5=0.5, p25=0.25, p125=0.125, p0625=0.0625 ) |
| 141 |
integer imin ,imax ,jmin ,jmax |
| 142 |
parameter(imin=2-OLx,imax=sNx+OLx-1,jmin=2-OLy,jmax=sNy+OLy-1) |
| 143 |
|
| 144 |
c Local arrays and variables |
| 145 |
c work? (nx,ny) - horizontal working arrays |
| 146 |
c ustar (nx,ny) - surface friction velocity (m/s) |
| 147 |
c bo (nx,ny) - surface turbulent buoyancy forcing (m^2/s^3) |
| 148 |
c bosol (nx,ny) - surface radiative buoyancy forcing (m^2/s^3) |
| 149 |
c shsq (nx,ny,Nr) - local velocity shear squared |
| 150 |
c at interfaces for ri_iwmix (m^2/s^2) |
| 151 |
c dVsq (nx,ny,Nr) - velocity shear re surface squared |
| 152 |
c at grid levels for bldepth (m^2/s^2) |
| 153 |
c dbloc (nx,ny,Nr) - local delta buoyancy at interfaces |
| 154 |
c for ri_iwmix and bldepth (m/s^2) |
| 155 |
c Ritop (nx,ny,Nr) - numerator of bulk richardson number |
| 156 |
c at grid levels for bldepth |
| 157 |
c vddiff (nx,ny,Nrp2,1)- vertical viscosity on "t-grid" (m^2/s) |
| 158 |
c vddiff (nx,ny,Nrp2,2)- vert. diff. on next row for salt&tracers (m^2/s) |
| 159 |
c vddiff (nx,ny,Nrp2,3)- vert. diff. on next row for temperature (m^2/s) |
| 160 |
c ghat (nx,ny,Nr) - nonlocal transport coefficient (s/m^2) |
| 161 |
c hbl (nx,ny) - mixing layer depth (m) |
| 162 |
c kmtj (nx,ny) - maximum number of wet levels in each column |
| 163 |
c z0 (nx,ny) - Roughness length (m) |
| 164 |
c zRef (nx,ny) - Reference depth: Hmix * epsilon (m) |
| 165 |
c uRef (nx,ny) - Reference zonal velocity (m/s) |
| 166 |
c vRef (nx,ny) - Reference meridional velocity (m/s) |
| 167 |
|
| 168 |
_RL worka ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy ) |
| 169 |
integer work1 ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy ) |
| 170 |
_KPP_RL work2 ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy ) |
| 171 |
_KPP_RL work3 ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy ) |
| 172 |
_KPP_RL ustar ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy ) |
| 173 |
_KPP_RL bo ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy ) |
| 174 |
_KPP_RL bosol ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy ) |
| 175 |
_KPP_RL shsq ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy, Nr ) |
| 176 |
_KPP_RL dVsq ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy, Nr ) |
| 177 |
_KPP_RL dbloc ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy, Nr ) |
| 178 |
_KPP_RL Ritop ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy, Nr ) |
| 179 |
_KPP_RL vddiff( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy, 0:Nrp1, mdiff ) |
| 180 |
_KPP_RL ghat ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy, Nr ) |
| 181 |
_KPP_RL hbl ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy ) |
| 182 |
cph( |
| 183 |
_KPP_RL TTALPHA( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy, Nrp1 ) |
| 184 |
_KPP_RL SSBETA ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy, Nrp1 ) |
| 185 |
cph) |
| 186 |
#ifdef KPP_ESTIMATE_UREF |
| 187 |
_KPP_RL z0 ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy ) |
| 188 |
_KPP_RL zRef ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy ) |
| 189 |
_KPP_RL uRef ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy ) |
| 190 |
_KPP_RL vRef ( 1-OLx:sNx+OLx, 1-OLy:sNy+OLy ) |
| 191 |
#endif /* KPP_ESTIMATE_UREF */ |
| 192 |
|
| 193 |
_KPP_RL tempvar2 |
| 194 |
integer i, j, k, kp1, km1, im1, ip1, jm1, jp1 |
| 195 |
|
| 196 |
#ifdef KPP_ESTIMATE_UREF |
| 197 |
_KPP_RL tempvar1, dBdz1, dBdz2, ustarX, ustarY |
| 198 |
#endif |
| 199 |
|
| 200 |
#ifdef ALLOW_AUTODIFF_TAMC |
| 201 |
act1 = bi - myBxLo(myThid) |
| 202 |
max1 = myBxHi(myThid) - myBxLo(myThid) + 1 |
| 203 |
act2 = bj - myByLo(myThid) |
| 204 |
max2 = myByHi(myThid) - myByLo(myThid) + 1 |
| 205 |
act3 = myThid - 1 |
| 206 |
max3 = nTx*nTy |
| 207 |
act4 = ikey_dynamics - 1 |
| 208 |
ikppkey = (act1 + 1) + act2*max1 |
| 209 |
& + act3*max1*max2 |
| 210 |
& + act4*max1*max2*max3 |
| 211 |
#endif /* ALLOW_AUTODIFF_TAMC */ |
| 212 |
CEOP |
| 213 |
|
| 214 |
c Check to see if new vertical mixing coefficient should be computed now? |
| 215 |
IF ( DIFFERENT_MULTIPLE(kpp_freq,myTime,deltaTClock) |
| 216 |
1 .OR. myTime .EQ. startTime ) THEN |
| 217 |
|
| 218 |
c----------------------------------------------------------------------- |
| 219 |
c prepare input arrays for subroutine "kppmix" to compute |
| 220 |
c viscosity and diffusivity and ghat. |
| 221 |
c All input arrays need to be in m-k-s units. |
| 222 |
c |
| 223 |
c note: for the computation of the bulk richardson number in the |
| 224 |
c "bldepth" subroutine, gradients of velocity and buoyancy are |
| 225 |
c required at every depth. in the case of very fine vertical grids |
| 226 |
c (thickness of top layer < 2m), the surface reference depth must |
| 227 |
c be set to zref=epsilon/2*zgrid(k), and the reference value |
| 228 |
c of velocity and buoyancy must be computed as vertical average |
| 229 |
c between the surface and 2*zref. in the case of coarse vertical |
| 230 |
c grids zref is zgrid(1)/2., and the surface reference value is |
| 231 |
c simply the surface value at zgrid(1). |
| 232 |
c----------------------------------------------------------------------- |
| 233 |
|
| 234 |
c------------------------------------------------------------------------ |
| 235 |
c density related quantities |
| 236 |
c -------------------------- |
| 237 |
c |
| 238 |
c work2 - density of surface layer (kg/m^3) |
| 239 |
c dbloc - local buoyancy gradient at Nr interfaces |
| 240 |
c g/rho{k+1,k+1} * [ drho{k,k+1}-drho{k+1,k+1} ] (m/s^2) |
| 241 |
c dbsfc (stored in Ritop to conserve stack memory) |
| 242 |
c - buoyancy difference with respect to the surface |
| 243 |
c g * [ drho{1,k}/rho{1,k} - drho{k,k}/rho{k,k} ] (m/s^2) |
| 244 |
c ttalpha (stored in vddiff(:,:,:,1) to conserve stack memory) |
| 245 |
c - thermal expansion coefficient without 1/rho factor |
| 246 |
c d(rho{k,k})/d(T(k)) (kg/m^3/C) |
| 247 |
c ssbeta (stored in vddiff(:,:,:,2) to conserve stack memory) |
| 248 |
c - salt expansion coefficient without 1/rho factor |
| 249 |
c d(rho{k,k})/d(S(k)) (kg/m^3/PSU) |
| 250 |
c------------------------------------------------------------------------ |
| 251 |
|
| 252 |
CALL TIMER_START('STATEKPP [KPP_CALC]', myThid) |
| 253 |
CALL STATEKPP( |
| 254 |
I ikppkey, bi, bj, myThid |
| 255 |
O , work2, dbloc, Ritop |
| 256 |
O , TTALPHA, SSBETA |
| 257 |
& ) |
| 258 |
CALL TIMER_STOP ('STATEKPP [KPP_CALC]', myThid) |
| 259 |
|
| 260 |
DO k = 1, Nr |
| 261 |
DO j = 1-OLy, sNy+OLy |
| 262 |
DO i = 1-OLx, sNx+OLx |
| 263 |
ghat(i,j,k) = dbloc(i,j,k) |
| 264 |
ENDDO |
| 265 |
ENDDO |
| 266 |
ENDDO |
| 267 |
|
| 268 |
#ifdef KPP_SMOOTH_DBLOC |
| 269 |
c horizontally smooth dbloc with a 121 filter |
| 270 |
c smooth dbloc stored in ghat to save space |
| 271 |
c dbloc(k) is buoyancy gradientnote between k and k+1 |
| 272 |
c levels therefore k+1 mask must be used |
| 273 |
|
| 274 |
DO k = 1, Nr-1 |
| 275 |
CALL KPP_SMOOTH_HORIZ ( |
| 276 |
I k+1, bi, bj, |
| 277 |
U ghat (1-OLx,1-OLy,k) ) |
| 278 |
ENDDO |
| 279 |
|
| 280 |
#endif /* KPP_SMOOTH_DBLOC */ |
| 281 |
|
| 282 |
#ifdef KPP_SMOOTH_DENS |
| 283 |
c horizontally smooth density related quantities with 121 filters |
| 284 |
CALL KPP_SMOOTH_HORIZ ( |
| 285 |
I 1, bi, bj, |
| 286 |
U work2 ) |
| 287 |
DO k = 1, Nr |
| 288 |
CALL KPP_SMOOTH_HORIZ ( |
| 289 |
I k+1, bi, bj, |
| 290 |
U dbloc (1-OLx,1-OLy,k) ) |
| 291 |
CALL KPP_SMOOTH_HORIZ ( |
| 292 |
I k, bi, bj, |
| 293 |
U Ritop (1-OLx,1-OLy,k) ) |
| 294 |
CALL KPP_SMOOTH_HORIZ ( |
| 295 |
I k, bi, bj, |
| 296 |
U TTALPHA(1-OLx,1-OLy,k) ) |
| 297 |
CALL KPP_SMOOTH_HORIZ ( |
| 298 |
I k, bi, bj, |
| 299 |
U SSBETA(1-OLx,1-OLy,k) ) |
| 300 |
ENDDO |
| 301 |
#endif /* KPP_SMOOTH_DENS */ |
| 302 |
|
| 303 |
DO k = 1, Nr |
| 304 |
km1 = max(1,k-1) |
| 305 |
DO j = 1-OLy, sNy+OLy |
| 306 |
DO i = 1-OLx, sNx+OLx |
| 307 |
|
| 308 |
c zero out dbloc over land points (so that the convective |
| 309 |
c part of the interior mixing can be diagnosed) |
| 310 |
dbloc(i,j,k) = dbloc(i,j,k) * maskC(i,j,k,bi,bj) |
| 311 |
& * maskC(i,j,km1,bi,bj) |
| 312 |
ghat(i,j,k) = ghat(i,j,k) * maskC(i,j,k,bi,bj) |
| 313 |
& * maskC(i,j,km1,bi,bj) |
| 314 |
Ritop(i,j,k) = Ritop(i,j,k) * maskC(i,j,k,bi,bj) |
| 315 |
& * maskC(i,j,km1,bi,bj) |
| 316 |
if(k.eq.nzmax(i,j,bi,bj)) then |
| 317 |
dbloc(i,j,k) = p0 |
| 318 |
ghat(i,j,k) = p0 |
| 319 |
Ritop(i,j,k) = p0 |
| 320 |
endif |
| 321 |
|
| 322 |
c numerator of bulk richardson number on grid levels |
| 323 |
c note: land and ocean bottom values need to be set to zero |
| 324 |
c so that the subroutine "bldepth" works correctly |
| 325 |
Ritop(i,j,k) = (zgrid(1)-zgrid(k)) * Ritop(i,j,k) |
| 326 |
|
| 327 |
END DO |
| 328 |
END DO |
| 329 |
END DO |
| 330 |
|
| 331 |
cph( |
| 332 |
cph this avoids a single or double recomp./call of statekpp |
| 333 |
CADJ store work2 = comlev1_kpp, key = ikppkey |
| 334 |
#ifdef KPP_AUTODIFF_EXCESSIVE_STORE |
| 335 |
CADJ store dbloc, Ritop, ghat = comlev1_kpp, key = ikppkey |
| 336 |
CADJ store vddiff = comlev1_kpp, key = ikppkey |
| 337 |
CADJ store TTALPHA, SSBETA = comlev1_kpp, key = ikppkey |
| 338 |
#endif |
| 339 |
cph) |
| 340 |
|
| 341 |
c------------------------------------------------------------------------ |
| 342 |
c friction velocity, turbulent and radiative surface buoyancy forcing |
| 343 |
c ------------------------------------------------------------------- |
| 344 |
c taux / rho = surfaceForcingU (N/m^2) |
| 345 |
c tauy / rho = surfaceForcingV (N/m^2) |
| 346 |
c ustar = sqrt( sqrt( taux^2 + tauy^2 ) / rho ) (m/s) |
| 347 |
c bo = - g * ( alpha*surfaceForcingT + |
| 348 |
c beta *surfaceForcingS ) / rho (m^2/s^3) |
| 349 |
c bosol = - g * alpha * Qsw * drF(1) / rho (m^2/s^3) |
| 350 |
c------------------------------------------------------------------------ |
| 351 |
|
| 352 |
c initialize arrays to zero |
| 353 |
DO j = 1-OLy, sNy+OLy |
| 354 |
DO i = 1-OLx, sNx+OLx |
| 355 |
ustar(i,j) = p0 |
| 356 |
bo (I,J) = p0 |
| 357 |
bosol(I,J) = p0 |
| 358 |
END DO |
| 359 |
END DO |
| 360 |
|
| 361 |
DO j = jmin, jmax |
| 362 |
jp1 = j + 1 |
| 363 |
DO i = imin, imax |
| 364 |
ip1 = i+1 |
| 365 |
work3(i,j) = |
| 366 |
& (surfaceForcingU(i,j,bi,bj) + surfaceForcingU(ip1,j,bi,bj)) * |
| 367 |
& (surfaceForcingU(i,j,bi,bj) + surfaceForcingU(ip1,j,bi,bj)) + |
| 368 |
& (surfaceForcingV(i,j,bi,bj) + surfaceForcingV(i,jp1,bi,bj)) * |
| 369 |
& (surfaceForcingV(i,j,bi,bj) + surfaceForcingV(i,jp1,bi,bj)) |
| 370 |
END DO |
| 371 |
END DO |
| 372 |
cph( |
| 373 |
CADJ store work3 = comlev1_kpp, key = ikppkey |
| 374 |
cph) |
| 375 |
DO j = jmin, jmax |
| 376 |
jp1 = j + 1 |
| 377 |
DO i = imin, imax |
| 378 |
ip1 = i+1 |
| 379 |
|
| 380 |
if ( work3(i,j) .lt. (phepsi*phepsi*drF(1)*drF(1)) ) then |
| 381 |
ustar(i,j) = SQRT( phepsi * p5 * drF(1) ) |
| 382 |
else |
| 383 |
tempVar2 = SQRT( work3(i,j) ) * p5 |
| 384 |
ustar(i,j) = SQRT( tempVar2 ) |
| 385 |
endif |
| 386 |
|
| 387 |
END DO |
| 388 |
END DO |
| 389 |
|
| 390 |
CML#ifdef ALLOW_SHELFICE |
| 391 |
CMLC For the pbl parameterisation to work underneath the ice shelves |
| 392 |
CMLC it needs to know the surface (ice-ocean) fluxes. However, masking |
| 393 |
CMLC and indexing problems make this part of the code not work |
| 394 |
CMLC underneath the ice shelves and the following lines are only here |
| 395 |
CMLC to remind me that this still needs to be sorted out. |
| 396 |
CML IF ( useShelfIce ) THEN |
| 397 |
CML DO j = jmin, jmax |
| 398 |
CML jp1 = j + 1 |
| 399 |
CML DO i = imin, imax |
| 400 |
CML ip1 = i+1 |
| 401 |
CML k = kTopC(I,J,bi,bj) |
| 402 |
CML bo(I,J) = - gravity * |
| 403 |
CML & ( TTALPHA(I,J,K) * ( maskC(I,J,1,bi,bj) |
| 404 |
CML & * ( surfaceForcingT(i,j,bi,bj) |
| 405 |
CML & + surfaceForcingTice(i,j,bi,bj) ) |
| 406 |
CML & + shelficeForcingT(i,j,bi,bj) ) |
| 407 |
CML & + SSBETA(I,J,K) * ( maskC(I,J,1,bi,bj) |
| 408 |
CML & * surfaceForcingS(i,j,bi,bj) |
| 409 |
CML & + shelficeForcingS(i,j,bi,bj) ) ) |
| 410 |
CML & / work2(I,J) |
| 411 |
CML |
| 412 |
CML bosol(I,J) = gravity * TTALPHA(I,J,1) * Qsw(i,j,bi,bj) |
| 413 |
CML & * maskC(I,J,1,bi,bj) * recip_Cp*recip_rhoConst |
| 414 |
CML & / work2(I,J) |
| 415 |
CML |
| 416 |
CML END DO |
| 417 |
CML END DO |
| 418 |
CML ELSE |
| 419 |
CML#endif /* ALLOW_SHELFICE */ |
| 420 |
DO j = jmin, jmax |
| 421 |
jp1 = j + 1 |
| 422 |
DO i = imin, imax |
| 423 |
ip1 = i+1 |
| 424 |
|
| 425 |
bo(I,J) = - gravity * |
| 426 |
& ( TTALPHA(I,J,1) * (surfaceForcingT(i,j,bi,bj)+ |
| 427 |
& surfaceForcingTice(i,j,bi,bj)) + |
| 428 |
& SSBETA(I,J,1) * surfaceForcingS(i,j,bi,bj) ) |
| 429 |
& / work2(I,J) |
| 430 |
|
| 431 |
bosol(I,J) = gravity * TTALPHA(I,J,1) * Qsw(i,j,bi,bj) * |
| 432 |
& recip_Cp*recip_rhoConst |
| 433 |
& / work2(I,J) |
| 434 |
|
| 435 |
END DO |
| 436 |
END DO |
| 437 |
CML#ifdef ALLOW_SHELFICE |
| 438 |
CML ENDIF |
| 439 |
CML#endif /* ALLOW_SHELFICE */ |
| 440 |
|
| 441 |
cph( |
| 442 |
CADJ store ustar = comlev1_kpp, key = ikppkey |
| 443 |
cph) |
| 444 |
|
| 445 |
c------------------------------------------------------------------------ |
| 446 |
c velocity shear |
| 447 |
c -------------- |
| 448 |
c Get velocity shear squared, averaged from "u,v-grid" |
| 449 |
c onto "t-grid" (in (m/s)**2): |
| 450 |
c dVsq(k)=(Uref-U(k))**2+(Vref-V(k))**2 at grid levels |
| 451 |
c shsq(k)=(U(k)-U(k+1))**2+(V(k)-V(k+1))**2 at interfaces |
| 452 |
c------------------------------------------------------------------------ |
| 453 |
|
| 454 |
c initialize arrays to zero |
| 455 |
DO k = 1, Nr |
| 456 |
DO j = 1-OLy, sNy+OLy |
| 457 |
DO i = 1-OLx, sNx+OLx |
| 458 |
shsq(i,j,k) = p0 |
| 459 |
dVsq(i,j,k) = p0 |
| 460 |
END DO |
| 461 |
END DO |
| 462 |
END DO |
| 463 |
|
| 464 |
c dVsq computation |
| 465 |
|
| 466 |
#ifdef KPP_ESTIMATE_UREF |
| 467 |
|
| 468 |
c Get rid of vertical resolution dependence of dVsq term by |
| 469 |
c estimating a surface velocity that is independent of first level |
| 470 |
c thickness in the model. First determine mixed layer depth hMix. |
| 471 |
c Second zRef = espilon * hMix. Third determine roughness length |
| 472 |
c scale z0. Third estimate reference velocity. |
| 473 |
|
| 474 |
DO j = jmin, jmax |
| 475 |
jp1 = j + 1 |
| 476 |
DO i = imin, imax |
| 477 |
ip1 = i + 1 |
| 478 |
|
| 479 |
c Determine mixed layer depth hMix as the shallowest depth at which |
| 480 |
c dB/dz exceeds 5.2e-5 s^-2. |
| 481 |
work1(i,j) = nzmax(i,j,bi,bj) |
| 482 |
DO k = 1, Nr |
| 483 |
IF ( k .LT. nzmax(i,j,bi,bj) .AND. |
| 484 |
& maskC(I,J,k,bi,bj) .GT. 0. .AND. |
| 485 |
& dbloc(i,j,k) / drC(k+1) .GT. dB_dz ) |
| 486 |
& work1(i,j) = k |
| 487 |
END DO |
| 488 |
|
| 489 |
c Linearly interpolate to find hMix. |
| 490 |
k = work1(i,j) |
| 491 |
IF ( k .EQ. 0 .OR. nzmax(i,j,bi,bj) .EQ. 1 ) THEN |
| 492 |
zRef(i,j) = p0 |
| 493 |
ELSEIF ( k .EQ. 1) THEN |
| 494 |
dBdz2 = dbloc(i,j,1) / drC(2) |
| 495 |
zRef(i,j) = drF(1) * dB_dz / dBdz2 |
| 496 |
ELSEIF ( k .LT. nzmax(i,j,bi,bj) ) THEN |
| 497 |
dBdz1 = dbloc(i,j,k-1) / drC(k ) |
| 498 |
dBdz2 = dbloc(i,j,k ) / drC(k+1) |
| 499 |
zRef(i,j) = rF(k) + drF(k) * (dB_dz - dBdz1) / |
| 500 |
& MAX ( phepsi, dBdz2 - dBdz1 ) |
| 501 |
ELSE |
| 502 |
zRef(i,j) = rF(k+1) |
| 503 |
ENDIF |
| 504 |
|
| 505 |
c Compute roughness length scale z0 subject to 0 < z0 |
| 506 |
tempVar1 = p5 * ( |
| 507 |
& (uVel(i, j, 1,bi,bj)-uVel(i, j, 2,bi,bj)) * |
| 508 |
& (uVel(i, j, 1,bi,bj)-uVel(i, j, 2,bi,bj)) + |
| 509 |
& (uVel(ip1,j, 1,bi,bj)-uVel(ip1,j, 2,bi,bj)) * |
| 510 |
& (uVel(ip1,j, 1,bi,bj)-uVel(ip1,j, 2,bi,bj)) + |
| 511 |
& (vVel(i, j, 1,bi,bj)-vVel(i, j, 2,bi,bj)) * |
| 512 |
& (vVel(i, j, 1,bi,bj)-vVel(i, j, 2,bi,bj)) + |
| 513 |
& (vVel(i, jp1,1,bi,bj)-vVel(i, jp1,2,bi,bj)) * |
| 514 |
& (vVel(i, jp1,1,bi,bj)-vVel(i, jp1,2,bi,bj)) ) |
| 515 |
if ( tempVar1 .lt. (epsln*epsln) ) then |
| 516 |
tempVar2 = epsln |
| 517 |
else |
| 518 |
tempVar2 = SQRT ( tempVar1 ) |
| 519 |
endif |
| 520 |
z0(i,j) = rF(2) * |
| 521 |
& ( rF(3) * LOG ( rF(3) / rF(2) ) / |
| 522 |
& ( rF(3) - rF(2) ) - |
| 523 |
& tempVar2 * vonK / |
| 524 |
& MAX ( ustar(i,j), phepsi ) ) |
| 525 |
z0(i,j) = MAX ( z0(i,j), phepsi ) |
| 526 |
|
| 527 |
c zRef is set to 0.1 * hMix subject to z0 <= zRef <= drF(1) |
| 528 |
zRef(i,j) = MAX ( epsilon * zRef(i,j), z0(i,j) ) |
| 529 |
zRef(i,j) = MIN ( zRef(i,j), drF(1) ) |
| 530 |
|
| 531 |
c Estimate reference velocity uRef and vRef. |
| 532 |
uRef(i,j) = p5 * |
| 533 |
& ( uVel(i,j,1,bi,bj) + uVel(ip1,j,1,bi,bj) ) |
| 534 |
vRef(i,j) = p5 * |
| 535 |
& ( vVel(i,j,1,bi,bj) + vVel(i,jp1,1,bi,bj) ) |
| 536 |
IF ( zRef(i,j) .LT. drF(1) ) THEN |
| 537 |
ustarX = ( surfaceForcingU(i, j,bi,bj) + |
| 538 |
& surfaceForcingU(ip1,j,bi,bj) ) * p5 |
| 539 |
& *recip_drF(1) |
| 540 |
ustarY = ( surfaceForcingV(i,j, bi,bj) + |
| 541 |
& surfaceForcingV(i,jp1,bi,bj) ) * p5 |
| 542 |
& *recip_drF(1) |
| 543 |
tempVar1 = ustarX * ustarX + ustarY * ustarY |
| 544 |
if ( tempVar1 .lt. (epsln*epsln) ) then |
| 545 |
tempVar2 = epsln |
| 546 |
else |
| 547 |
tempVar2 = SQRT ( tempVar1 ) |
| 548 |
endif |
| 549 |
tempVar2 = ustar(i,j) * |
| 550 |
& ( LOG ( zRef(i,j) / rF(2) ) + |
| 551 |
& z0(i,j) / zRef(i,j) - z0(i,j) / rF(2) ) / |
| 552 |
& vonK / tempVar2 |
| 553 |
uRef(i,j) = uRef(i,j) + ustarX * tempVar2 |
| 554 |
vRef(i,j) = vRef(i,j) + ustarY * tempVar2 |
| 555 |
ENDIF |
| 556 |
|
| 557 |
END DO |
| 558 |
END DO |
| 559 |
|
| 560 |
DO k = 1, Nr |
| 561 |
DO j = jmin, jmax |
| 562 |
jm1 = j - 1 |
| 563 |
jp1 = j + 1 |
| 564 |
DO i = imin, imax |
| 565 |
im1 = i - 1 |
| 566 |
ip1 = i + 1 |
| 567 |
dVsq(i,j,k) = p5 * ( |
| 568 |
$ (uRef(i,j) - uVel(i, j, k,bi,bj)) * |
| 569 |
$ (uRef(i,j) - uVel(i, j, k,bi,bj)) + |
| 570 |
$ (uRef(i,j) - uVel(ip1,j, k,bi,bj)) * |
| 571 |
$ (uRef(i,j) - uVel(ip1,j, k,bi,bj)) + |
| 572 |
$ (vRef(i,j) - vVel(i, j, k,bi,bj)) * |
| 573 |
$ (vRef(i,j) - vVel(i, j, k,bi,bj)) + |
| 574 |
$ (vRef(i,j) - vVel(i, jp1,k,bi,bj)) * |
| 575 |
$ (vRef(i,j) - vVel(i, jp1,k,bi,bj)) ) |
| 576 |
#ifdef KPP_SMOOTH_DVSQ |
| 577 |
dVsq(i,j,k) = p5 * dVsq(i,j,k) + p125 * ( |
| 578 |
$ (uRef(i,j) - uVel(i, jm1,k,bi,bj)) * |
| 579 |
$ (uRef(i,j) - uVel(i, jm1,k,bi,bj)) + |
| 580 |
$ (uRef(i,j) - uVel(ip1,jm1,k,bi,bj)) * |
| 581 |
$ (uRef(i,j) - uVel(ip1,jm1,k,bi,bj)) + |
| 582 |
$ (uRef(i,j) - uVel(i, jp1,k,bi,bj)) * |
| 583 |
$ (uRef(i,j) - uVel(i, jp1,k,bi,bj)) + |
| 584 |
$ (uRef(i,j) - uVel(ip1,jp1,k,bi,bj)) * |
| 585 |
$ (uRef(i,j) - uVel(ip1,jp1,k,bi,bj)) + |
| 586 |
$ (vRef(i,j) - vVel(im1,j, k,bi,bj)) * |
| 587 |
$ (vRef(i,j) - vVel(im1,j, k,bi,bj)) + |
| 588 |
$ (vRef(i,j) - vVel(im1,jp1,k,bi,bj)) * |
| 589 |
$ (vRef(i,j) - vVel(im1,jp1,k,bi,bj)) + |
| 590 |
$ (vRef(i,j) - vVel(ip1,j, k,bi,bj)) * |
| 591 |
$ (vRef(i,j) - vVel(ip1,j, k,bi,bj)) + |
| 592 |
$ (vRef(i,j) - vVel(ip1,jp1,k,bi,bj)) * |
| 593 |
$ (vRef(i,j) - vVel(ip1,jp1,k,bi,bj)) ) |
| 594 |
#endif /* KPP_SMOOTH_DVSQ */ |
| 595 |
END DO |
| 596 |
END DO |
| 597 |
END DO |
| 598 |
|
| 599 |
#else /* KPP_ESTIMATE_UREF */ |
| 600 |
|
| 601 |
DO k = 1, Nr |
| 602 |
DO j = jmin, jmax |
| 603 |
jm1 = j - 1 |
| 604 |
jp1 = j + 1 |
| 605 |
DO i = imin, imax |
| 606 |
im1 = i - 1 |
| 607 |
ip1 = i + 1 |
| 608 |
dVsq(i,j,k) = p5 * ( |
| 609 |
$ (uVel(i, j, 1,bi,bj)-uVel(i, j, k,bi,bj)) * |
| 610 |
$ (uVel(i, j, 1,bi,bj)-uVel(i, j, k,bi,bj)) + |
| 611 |
$ (uVel(ip1,j, 1,bi,bj)-uVel(ip1,j, k,bi,bj)) * |
| 612 |
$ (uVel(ip1,j, 1,bi,bj)-uVel(ip1,j, k,bi,bj)) + |
| 613 |
$ (vVel(i, j, 1,bi,bj)-vVel(i, j, k,bi,bj)) * |
| 614 |
$ (vVel(i, j, 1,bi,bj)-vVel(i, j, k,bi,bj)) + |
| 615 |
$ (vVel(i, jp1,1,bi,bj)-vVel(i, jp1,k,bi,bj)) * |
| 616 |
$ (vVel(i, jp1,1,bi,bj)-vVel(i, jp1,k,bi,bj)) ) |
| 617 |
#ifdef KPP_SMOOTH_DVSQ |
| 618 |
dVsq(i,j,k) = p5 * dVsq(i,j,k) + p125 * ( |
| 619 |
$ (uVel(i, jm1,1,bi,bj)-uVel(i, jm1,k,bi,bj)) * |
| 620 |
$ (uVel(i, jm1,1,bi,bj)-uVel(i, jm1,k,bi,bj)) + |
| 621 |
$ (uVel(ip1,jm1,1,bi,bj)-uVel(ip1,jm1,k,bi,bj)) * |
| 622 |
$ (uVel(ip1,jm1,1,bi,bj)-uVel(ip1,jm1,k,bi,bj)) + |
| 623 |
$ (uVel(i, jp1,1,bi,bj)-uVel(i, jp1,k,bi,bj)) * |
| 624 |
$ (uVel(i, jp1,1,bi,bj)-uVel(i, jp1,k,bi,bj)) + |
| 625 |
$ (uVel(ip1,jp1,1,bi,bj)-uVel(ip1,jp1,k,bi,bj)) * |
| 626 |
$ (uVel(ip1,jp1,1,bi,bj)-uVel(ip1,jp1,k,bi,bj)) + |
| 627 |
$ (vVel(im1,j, 1,bi,bj)-vVel(im1,j, k,bi,bj)) * |
| 628 |
$ (vVel(im1,j, 1,bi,bj)-vVel(im1,j, k,bi,bj)) + |
| 629 |
$ (vVel(im1,jp1,1,bi,bj)-vVel(im1,jp1,k,bi,bj)) * |
| 630 |
$ (vVel(im1,jp1,1,bi,bj)-vVel(im1,jp1,k,bi,bj)) + |
| 631 |
$ (vVel(ip1,j, 1,bi,bj)-vVel(ip1,j, k,bi,bj)) * |
| 632 |
$ (vVel(ip1,j, 1,bi,bj)-vVel(ip1,j, k,bi,bj)) + |
| 633 |
$ (vVel(ip1,jp1,1,bi,bj)-vVel(ip1,jp1,k,bi,bj)) * |
| 634 |
$ (vVel(ip1,jp1,1,bi,bj)-vVel(ip1,jp1,k,bi,bj)) ) |
| 635 |
#endif /* KPP_SMOOTH_DVSQ */ |
| 636 |
END DO |
| 637 |
END DO |
| 638 |
END DO |
| 639 |
|
| 640 |
#endif /* KPP_ESTIMATE_UREF */ |
| 641 |
|
| 642 |
c shsq computation |
| 643 |
DO k = 1, Nrm1 |
| 644 |
kp1 = k + 1 |
| 645 |
DO j = jmin, jmax |
| 646 |
jm1 = j - 1 |
| 647 |
jp1 = j + 1 |
| 648 |
DO i = imin, imax |
| 649 |
im1 = i - 1 |
| 650 |
ip1 = i + 1 |
| 651 |
shsq(i,j,k) = p5 * ( |
| 652 |
$ (uVel(i, j, k,bi,bj)-uVel(i, j, kp1,bi,bj)) * |
| 653 |
$ (uVel(i, j, k,bi,bj)-uVel(i, j, kp1,bi,bj)) + |
| 654 |
$ (uVel(ip1,j, k,bi,bj)-uVel(ip1,j, kp1,bi,bj)) * |
| 655 |
$ (uVel(ip1,j, k,bi,bj)-uVel(ip1,j, kp1,bi,bj)) + |
| 656 |
$ (vVel(i, j, k,bi,bj)-vVel(i, j, kp1,bi,bj)) * |
| 657 |
$ (vVel(i, j, k,bi,bj)-vVel(i, j, kp1,bi,bj)) + |
| 658 |
$ (vVel(i, jp1,k,bi,bj)-vVel(i, jp1,kp1,bi,bj)) * |
| 659 |
$ (vVel(i, jp1,k,bi,bj)-vVel(i, jp1,kp1,bi,bj)) ) |
| 660 |
#ifdef KPP_SMOOTH_SHSQ |
| 661 |
shsq(i,j,k) = p5 * shsq(i,j,k) + p125 * ( |
| 662 |
$ (uVel(i, jm1,k,bi,bj)-uVel(i, jm1,kp1,bi,bj)) * |
| 663 |
$ (uVel(i, jm1,k,bi,bj)-uVel(i, jm1,kp1,bi,bj)) + |
| 664 |
$ (uVel(ip1,jm1,k,bi,bj)-uVel(ip1,jm1,kp1,bi,bj)) * |
| 665 |
$ (uVel(ip1,jm1,k,bi,bj)-uVel(ip1,jm1,kp1,bi,bj)) + |
| 666 |
$ (uVel(i, jp1,k,bi,bj)-uVel(i, jp1,kp1,bi,bj)) * |
| 667 |
$ (uVel(i, jp1,k,bi,bj)-uVel(i, jp1,kp1,bi,bj)) + |
| 668 |
$ (uVel(ip1,jp1,k,bi,bj)-uVel(ip1,jp1,kp1,bi,bj)) * |
| 669 |
$ (uVel(ip1,jp1,k,bi,bj)-uVel(ip1,jp1,kp1,bi,bj)) + |
| 670 |
$ (vVel(im1,j, k,bi,bj)-vVel(im1,j, kp1,bi,bj)) * |
| 671 |
$ (vVel(im1,j, k,bi,bj)-vVel(im1,j, kp1,bi,bj)) + |
| 672 |
$ (vVel(im1,jp1,k,bi,bj)-vVel(im1,jp1,kp1,bi,bj)) * |
| 673 |
$ (vVel(im1,jp1,k,bi,bj)-vVel(im1,jp1,kp1,bi,bj)) + |
| 674 |
$ (vVel(ip1,j, k,bi,bj)-vVel(ip1,j, kp1,bi,bj)) * |
| 675 |
$ (vVel(ip1,j, k,bi,bj)-vVel(ip1,j, kp1,bi,bj)) + |
| 676 |
$ (vVel(ip1,jp1,k,bi,bj)-vVel(ip1,jp1,kp1,bi,bj)) * |
| 677 |
$ (vVel(ip1,jp1,k,bi,bj)-vVel(ip1,jp1,kp1,bi,bj)) ) |
| 678 |
#endif |
| 679 |
END DO |
| 680 |
END DO |
| 681 |
END DO |
| 682 |
|
| 683 |
cph( |
| 684 |
#ifdef KPP_AUTODIFF_EXCESSIVE_STORE |
| 685 |
CADJ store dvsq, shsq = comlev1_kpp, key = ikppkey |
| 686 |
#endif |
| 687 |
cph) |
| 688 |
|
| 689 |
c----------------------------------------------------------------------- |
| 690 |
c solve for viscosity, diffusivity, ghat, and hbl on "t-grid" |
| 691 |
c----------------------------------------------------------------------- |
| 692 |
|
| 693 |
c precompute background vertical diffusivities, which are needed for |
| 694 |
c matching diffusivities at bottom of KPP PBL |
| 695 |
CALL CALC_3D_DIFFUSIVITY( |
| 696 |
I bi,bj,1-Olx,sNx+OLx,1-Oly,sNy+OLy, |
| 697 |
I GAD_SALINITY, .FALSE., .FALSE., |
| 698 |
O KPPdiffKzS(1-Olx,1-Oly,1,bi,bj), |
| 699 |
I myThid) |
| 700 |
CALL CALC_3D_DIFFUSIVITY( |
| 701 |
I bi,bj,1-Olx,sNx+OLx,1-Oly,sNy+OLy, |
| 702 |
I GAD_TEMPERATURE, .FALSE., .FALSE., |
| 703 |
O KPPdiffKzT(1-Olx,1-Oly,1,bi,bj), |
| 704 |
I myThid) |
| 705 |
|
| 706 |
DO j = 1-OLy, sNy+OLy |
| 707 |
DO i = 1-OLx, sNx+OLx |
| 708 |
work1(i,j) = nzmax(i,j,bi,bj) |
| 709 |
work2(i,j) = Fcori(i,j,bi,bj) |
| 710 |
END DO |
| 711 |
END DO |
| 712 |
CALL TIMER_START('KPPMIX [KPP_CALC]', myThid) |
| 713 |
CALL KPPMIX ( |
| 714 |
I mytime, mythid |
| 715 |
I , work1, shsq, dVsq, ustar |
| 716 |
I , bo, bosol, dbloc, Ritop, work2 |
| 717 |
I , KPPdiffKzS(1-Olx,1-Oly,1,bi,bj) |
| 718 |
I , KPPdiffKzT(1-Olx,1-Oly,1,bi,bj) |
| 719 |
I , ikppkey |
| 720 |
O , vddiff |
| 721 |
U , ghat |
| 722 |
O , hbl ) |
| 723 |
CALL TIMER_STOP ('KPPMIX [KPP_CALC]', myThid) |
| 724 |
|
| 725 |
c----------------------------------------------------------------------- |
| 726 |
c zero out land values and transfer to global variables |
| 727 |
c----------------------------------------------------------------------- |
| 728 |
|
| 729 |
DO j = jmin, jmax |
| 730 |
DO i = imin, imax |
| 731 |
DO k = 1, Nr |
| 732 |
km1 = max(1,k-1) |
| 733 |
KPPviscAz(i,j,k,bi,bj) = vddiff(i,j,k-1,1) * maskC(i,j,k,bi,bj) |
| 734 |
& * maskC(i,j,km1,bi,bj) |
| 735 |
KPPdiffKzS(i,j,k,bi,bj)= vddiff(i,j,k-1,2) * maskC(i,j,k,bi,bj) |
| 736 |
& * maskC(i,j,km1,bi,bj) |
| 737 |
KPPdiffKzT(i,j,k,bi,bj)= vddiff(i,j,k-1,3) * maskC(i,j,k,bi,bj) |
| 738 |
& * maskC(i,j,km1,bi,bj) |
| 739 |
KPPghat(i,j,k,bi,bj) = ghat(i,j,k) * maskC(i,j,k,bi,bj) |
| 740 |
& * maskC(i,j,km1,bi,bj) |
| 741 |
END DO |
| 742 |
k = 1 |
| 743 |
#ifdef ALLOW_SHELFICE |
| 744 |
if ( useShelfIce ) k = kTopC(i,j,bi,bj) |
| 745 |
#endif /* ALLOW_SHELFICE */ |
| 746 |
KPPhbl(i,j,bi,bj) = hbl(i,j) * maskC(i,j,k,bi,bj) |
| 747 |
|
| 748 |
END DO |
| 749 |
END DO |
| 750 |
|
| 751 |
#ifdef KPP_SMOOTH_VISC |
| 752 |
c horizontal smoothing of vertical viscosity |
| 753 |
DO k = 1, Nr |
| 754 |
CALL SMOOTH_HORIZ ( |
| 755 |
I k, bi, bj, |
| 756 |
U KPPviscAz(1-OLx,1-OLy,k,bi,bj) ) |
| 757 |
END DO |
| 758 |
_EXCH_XYZ_R8(KPPviscAz , myThid ) |
| 759 |
#endif /* KPP_SMOOTH_VISC */ |
| 760 |
|
| 761 |
#ifdef KPP_SMOOTH_DIFF |
| 762 |
c horizontal smoothing of vertical diffusivity |
| 763 |
DO k = 1, Nr |
| 764 |
CALL SMOOTH_HORIZ ( |
| 765 |
I k, bi, bj, |
| 766 |
U KPPdiffKzS(1-OLx,1-OLy,k,bi,bj) ) |
| 767 |
CALL SMOOTH_HORIZ ( |
| 768 |
I k, bi, bj, |
| 769 |
U KPPdiffKzT(1-OLx,1-OLy,k,bi,bj) ) |
| 770 |
END DO |
| 771 |
_EXCH_XYZ_R8(KPPdiffKzS , myThid ) |
| 772 |
_EXCH_XYZ_R8(KPPdiffKzT , myThid ) |
| 773 |
#endif /* KPP_SMOOTH_DIFF */ |
| 774 |
|
| 775 |
cph( |
| 776 |
cph crucial: this avoids full recomp./call of kppmix |
| 777 |
CADJ store KPPhbl = comlev1_kpp, key = ikppkey |
| 778 |
cph) |
| 779 |
|
| 780 |
C Compute fraction of solar short-wave flux penetrating to |
| 781 |
C the bottom of the mixing layer. |
| 782 |
DO j=1-OLy,sNy+OLy |
| 783 |
DO i=1-OLx,sNx+OLx |
| 784 |
worka(i,j) = KPPhbl(i,j,bi,bj) |
| 785 |
ENDDO |
| 786 |
ENDDO |
| 787 |
CALL SWFRAC( |
| 788 |
I (sNx+2*OLx)*(sNy+2*OLy), minusone, |
| 789 |
I mytime, mythid, |
| 790 |
U worka ) |
| 791 |
DO j=1-OLy,sNy+OLy |
| 792 |
DO i=1-OLx,sNx+OLx |
| 793 |
KPPfrac(i,j,bi,bj) = worka(i,j) |
| 794 |
ENDDO |
| 795 |
ENDDO |
| 796 |
|
| 797 |
ENDIF |
| 798 |
|
| 799 |
#endif /* ALLOW_KPP */ |
| 800 |
|
| 801 |
RETURN |
| 802 |
END |
| 803 |
|
| 804 |
subroutine KPP_CALC_DUMMY( |
| 805 |
I bi, bj, myTime, myThid ) |
| 806 |
C /==========================================================\ |
| 807 |
C | SUBROUTINE KPP_CALC_DUMMY | |
| 808 |
C | o Compute all KPP fields defined in KPP.h | |
| 809 |
C | o Dummy routine for TAMC |
| 810 |
C |==========================================================| |
| 811 |
C | This subroutine serves as an interface between MITGCMUV | |
| 812 |
C | code and NCOM 1-D routines in kpp_routines.F | |
| 813 |
C \==========================================================/ |
| 814 |
IMPLICIT NONE |
| 815 |
|
| 816 |
#include "SIZE.h" |
| 817 |
#include "EEPARAMS.h" |
| 818 |
#include "PARAMS.h" |
| 819 |
#include "KPP.h" |
| 820 |
#include "KPP_PARAMS.h" |
| 821 |
#include "GRID.h" |
| 822 |
#include "GAD.h" |
| 823 |
|
| 824 |
c Routine arguments |
| 825 |
c bi, bj - array indices on which to apply calculations |
| 826 |
c myTime - Current time in simulation |
| 827 |
|
| 828 |
INTEGER bi, bj |
| 829 |
INTEGER myThid |
| 830 |
_RL myTime |
| 831 |
|
| 832 |
#ifdef ALLOW_KPP |
| 833 |
|
| 834 |
c Local constants |
| 835 |
integer i, j, k |
| 836 |
|
| 837 |
DO j=1-OLy,sNy+OLy |
| 838 |
DO i=1-OLx,sNx+OLx |
| 839 |
KPPhbl (i,j,bi,bj) = 1.0 |
| 840 |
KPPfrac(i,j,bi,bj) = 0.0 |
| 841 |
DO k = 1,Nr |
| 842 |
KPPghat (i,j,k,bi,bj) = 0.0 |
| 843 |
KPPviscAz (i,j,k,bi,bj) = viscAr |
| 844 |
ENDDO |
| 845 |
ENDDO |
| 846 |
ENDDO |
| 847 |
|
| 848 |
CALL CALC_3D_DIFFUSIVITY( |
| 849 |
I bi,bj,1-Olx,sNx+OLx,1-Oly,sNy+OLy, |
| 850 |
I GAD_SALINITY, .FALSE., .FALSE., |
| 851 |
O KPPdiffKzS(1-Olx,1-Oly,1,bi,bj), |
| 852 |
I myThid) |
| 853 |
CALL CALC_3D_DIFFUSIVITY( |
| 854 |
I bi,bj,1-Olx,sNx+OLx,1-Oly,sNy+OLy, |
| 855 |
I GAD_TEMPERATURE, .FALSE., .FALSE., |
| 856 |
O KPPdiffKzT(1-Olx,1-Oly,1,bi,bj), |
| 857 |
I myThid) |
| 858 |
|
| 859 |
#endif |
| 860 |
RETURN |
| 861 |
END |
Parent Directory
| 
Revision Log
| 
Revision Graph