| 1 |
SUBROUTINE coefkz(nsrf, knon, paprs, pplay, |
| 2 |
cIM 261103 |
| 3 |
. ksta, ksta_ter, |
| 4 |
cIM 261103 |
| 5 |
. ts, rugos, |
| 6 |
. u,v,t,q, |
| 7 |
. qsurf, |
| 8 |
. pcfm, pcfh) |
| 9 |
use dimens_m |
| 10 |
use indicesol |
| 11 |
use dimphy |
| 12 |
use iniprint |
| 13 |
use YOMCST |
| 14 |
use yoethf |
| 15 |
use fcttre |
| 16 |
use conf_phys_m |
| 17 |
IMPLICIT none |
| 18 |
c====================================================================== |
| 19 |
c Auteur(s) F. Hourdin, M. Forichon, Z.X. Li (LMD/CNRS) date: 19930922 |
| 20 |
c (une version strictement identique a l'ancien modele) |
| 21 |
c Objet: calculer le coefficient du frottement du sol (Cdrag) et les |
| 22 |
c coefficients d'echange turbulent dans l'atmosphere. |
| 23 |
c Arguments: |
| 24 |
c nsrf-----input-I- indicateur de la nature du sol |
| 25 |
c knon-----input-I- nombre de points a traiter |
| 26 |
c paprs----input-R- pression a chaque intercouche (en Pa) |
| 27 |
c pplay----input-R- pression au milieu de chaque couche (en Pa) |
| 28 |
c ts-------input-R- temperature du sol (en Kelvin) |
| 29 |
c rugos----input-R- longeur de rugosite (en m) |
| 30 |
c u--------input-R- vitesse u |
| 31 |
c v--------input-R- vitesse v |
| 32 |
c t--------input-R- temperature (K) |
| 33 |
c q--------input-R- vapeur d'eau (kg/kg) |
| 34 |
c |
| 35 |
c itop-----output-I- numero de couche du sommet de la couche limite |
| 36 |
c pcfm-----output-R- coefficients a calculer (vitesse) |
| 37 |
c pcfh-----output-R- coefficients a calculer (chaleur et humidite) |
| 38 |
c====================================================================== |
| 39 |
c |
| 40 |
c Arguments: |
| 41 |
c |
| 42 |
INTEGER knon, nsrf |
| 43 |
REAL ts(klon) |
| 44 |
REAL paprs(klon,klev+1), pplay(klon,klev) |
| 45 |
REAL u(klon,klev), v(klon,klev), t(klon,klev), q(klon,klev) |
| 46 |
REAL rugos(klon) |
| 47 |
c |
| 48 |
REAL pcfm(klon,klev), pcfh(klon,klev) |
| 49 |
INTEGER itop(klon) |
| 50 |
c |
| 51 |
c Quelques constantes et options: |
| 52 |
c |
| 53 |
REAL cepdu2, ckap, cb, cc, cd, clam |
| 54 |
PARAMETER (cepdu2 =(0.1)**2) |
| 55 |
PARAMETER (CKAP=0.4) |
| 56 |
PARAMETER (cb=5.0) |
| 57 |
PARAMETER (cc=5.0) |
| 58 |
PARAMETER (cd=5.0) |
| 59 |
PARAMETER (clam=160.0) |
| 60 |
REAL ratqs ! largeur de distribution de vapeur d'eau |
| 61 |
PARAMETER (ratqs=0.05) |
| 62 |
LOGICAL richum ! utilise le nombre de Richardson humide |
| 63 |
PARAMETER (richum=.TRUE.) |
| 64 |
REAL ric ! nombre de Richardson critique |
| 65 |
PARAMETER(ric=0.4) |
| 66 |
REAL prandtl |
| 67 |
PARAMETER (prandtl=0.4) |
| 68 |
REAL kstable ! diffusion minimale (situation stable) |
| 69 |
! GKtest |
| 70 |
! PARAMETER (kstable=1.0e-10) |
| 71 |
REAL ksta, ksta_ter |
| 72 |
cIM: 261103 REAL kstable_ter, kstable_sinon |
| 73 |
cIM: 211003 cf GK PARAMETER (kstable_ter = 1.0e-6) |
| 74 |
cIM: 261103 PARAMETER (kstable_ter = 1.0e-8) |
| 75 |
cIM: 261103 PARAMETER (kstable_ter = 1.0e-10) |
| 76 |
cIM: 261103 PARAMETER (kstable_sinon = 1.0e-10) |
| 77 |
! fin GKtest |
| 78 |
REAL mixlen ! constante controlant longueur de melange |
| 79 |
PARAMETER (mixlen=35.0) |
| 80 |
INTEGER isommet ! le sommet de la couche limite |
| 81 |
PARAMETER (isommet=klev) |
| 82 |
LOGICAL tvirtu ! calculer Ri d'une maniere plus performante |
| 83 |
PARAMETER (tvirtu=.TRUE.) |
| 84 |
LOGICAL opt_ec ! formule du Centre Europeen dans l'atmosphere |
| 85 |
PARAMETER (opt_ec=.FALSE.) |
| 86 |
|
| 87 |
c |
| 88 |
c Variables locales: |
| 89 |
c |
| 90 |
INTEGER i, k, kk !IM 120704 |
| 91 |
REAL zgeop(klon,klev) |
| 92 |
REAL zmgeom(klon) |
| 93 |
REAL zri(klon) |
| 94 |
REAL zl2(klon) |
| 95 |
|
| 96 |
REAL u1(klon), v1(klon), t1(klon), q1(klon), z1(klon) |
| 97 |
REAL pcfm1(klon), pcfh1(klon) |
| 98 |
c |
| 99 |
REAL zdphi, zdu2, ztvd, ztvu, zcdn |
| 100 |
REAL zscf |
| 101 |
REAL zt, zq, zdelta, zcvm5, zcor, zqs, zfr, zdqs |
| 102 |
REAL z2geomf, zalh2, zalm2, zscfh, zscfm |
| 103 |
REAL t_coup |
| 104 |
PARAMETER (t_coup=273.15) |
| 105 |
cIM |
| 106 |
LOGICAL check |
| 107 |
PARAMETER (check=.false.) |
| 108 |
c |
| 109 |
c contre-gradient pour la chaleur sensible: Kelvin/metre |
| 110 |
REAL gamt(2:klev) |
| 111 |
real qsurf(klon) |
| 112 |
c |
| 113 |
LOGICAL appel1er |
| 114 |
SAVE appel1er |
| 115 |
c |
| 116 |
c Fonctions thermodynamiques et fonctions d'instabilite |
| 117 |
REAL fsta, fins, x |
| 118 |
LOGICAL zxli ! utiliser un jeu de fonctions simples |
| 119 |
PARAMETER (zxli=.FALSE.) |
| 120 |
c |
| 121 |
fsta(x) = 1.0 / (1.0+10.0*x*(1+8.0*x)) |
| 122 |
fins(x) = SQRT(1.0-18.0*x) |
| 123 |
c |
| 124 |
DATA appel1er /.TRUE./ |
| 125 |
c |
| 126 |
IF (appel1er) THEN |
| 127 |
if (prt_level > 9) THEN |
| 128 |
WRITE(lunout,*)'coefkz, opt_ec:', opt_ec |
| 129 |
WRITE(lunout,*)'coefkz, richum:', richum |
| 130 |
IF (richum) WRITE(lunout,*)'coefkz, ratqs:', ratqs |
| 131 |
WRITE(lunout,*)'coefkz, isommet:', isommet |
| 132 |
WRITE(lunout,*)'coefkz, tvirtu:', tvirtu |
| 133 |
appel1er = .FALSE. |
| 134 |
endif |
| 135 |
ENDIF |
| 136 |
c |
| 137 |
c Initialiser les sorties |
| 138 |
c |
| 139 |
DO k = 1, klev |
| 140 |
DO i = 1, knon |
| 141 |
pcfm(i,k) = 0.0 |
| 142 |
pcfh(i,k) = 0.0 |
| 143 |
ENDDO |
| 144 |
ENDDO |
| 145 |
DO i = 1, knon |
| 146 |
itop(i) = 0 |
| 147 |
ENDDO |
| 148 |
|
| 149 |
c |
| 150 |
c Prescrire la valeur de contre-gradient |
| 151 |
c |
| 152 |
if (iflag_pbl.eq.1) then |
| 153 |
DO k = 3, klev |
| 154 |
gamt(k) = -1.0E-03 |
| 155 |
ENDDO |
| 156 |
gamt(2) = -2.5E-03 |
| 157 |
else |
| 158 |
DO k = 2, klev |
| 159 |
gamt(k) = 0.0 |
| 160 |
ENDDO |
| 161 |
ENDIF |
| 162 |
cIM cf JLD/ GKtest |
| 163 |
IF ( nsrf .NE. is_oce ) THEN |
| 164 |
cIM 261103 kstable = kstable_ter |
| 165 |
kstable = ksta_ter |
| 166 |
ELSE |
| 167 |
cIM 261103 kstable = kstable_sinon |
| 168 |
kstable = ksta |
| 169 |
ENDIF |
| 170 |
cIM cf JLD/ GKtest fin |
| 171 |
c |
| 172 |
c Calculer les geopotentiels de chaque couche |
| 173 |
c |
| 174 |
DO i = 1, knon |
| 175 |
zgeop(i,1) = RD * t(i,1) / (0.5*(paprs(i,1)+pplay(i,1))) |
| 176 |
. * (paprs(i,1)-pplay(i,1)) |
| 177 |
ENDDO |
| 178 |
DO k = 2, klev |
| 179 |
DO i = 1, knon |
| 180 |
zgeop(i,k) = zgeop(i,k-1) |
| 181 |
. + RD * 0.5*(t(i,k-1)+t(i,k)) / paprs(i,k) |
| 182 |
. * (pplay(i,k-1)-pplay(i,k)) |
| 183 |
ENDDO |
| 184 |
ENDDO |
| 185 |
c |
| 186 |
c Calculer le frottement au sol (Cdrag) |
| 187 |
c |
| 188 |
DO i = 1, knon |
| 189 |
u1(i) = u(i,1) |
| 190 |
v1(i) = v(i,1) |
| 191 |
t1(i) = t(i,1) |
| 192 |
q1(i) = q(i,1) |
| 193 |
z1(i) = zgeop(i,1) |
| 194 |
ENDDO |
| 195 |
c |
| 196 |
CALL clcdrag(klon, knon, nsrf, zxli, |
| 197 |
$ u1, v1, t1, q1, z1, |
| 198 |
$ ts, qsurf, rugos, |
| 199 |
$ pcfm1, pcfh1) |
| 200 |
cIM $ ts, qsurf, rugos, |
| 201 |
C |
| 202 |
DO i = 1, knon |
| 203 |
pcfm(i,1)=pcfm1(i) |
| 204 |
pcfh(i,1)=pcfh1(i) |
| 205 |
ENDDO |
| 206 |
c |
| 207 |
c Calculer les coefficients turbulents dans l'atmosphere |
| 208 |
c |
| 209 |
DO i = 1, knon |
| 210 |
itop(i) = isommet |
| 211 |
ENDDO |
| 212 |
|
| 213 |
|
| 214 |
DO k = 2, isommet |
| 215 |
DO i = 1, knon |
| 216 |
zdu2=MAX(cepdu2,(u(i,k)-u(i,k-1))**2 |
| 217 |
. +(v(i,k)-v(i,k-1))**2) |
| 218 |
zmgeom(i)=zgeop(i,k)-zgeop(i,k-1) |
| 219 |
zdphi =zmgeom(i) / 2.0 |
| 220 |
zt = (t(i,k)+t(i,k-1)) * 0.5 |
| 221 |
zq = (q(i,k)+q(i,k-1)) * 0.5 |
| 222 |
c |
| 223 |
c calculer Qs et dQs/dT: |
| 224 |
c |
| 225 |
IF (thermcep) THEN |
| 226 |
zdelta = MAX(0.,SIGN(1.,RTT-zt)) |
| 227 |
zcvm5 = R5LES*RLVTT/RCPD/(1.0+RVTMP2*zq)*(1.-zdelta) |
| 228 |
. + R5IES*RLSTT/RCPD/(1.0+RVTMP2*zq)*zdelta |
| 229 |
zqs = R2ES * FOEEW(zt,zdelta) / pplay(i,k) |
| 230 |
zqs = MIN(0.5,zqs) |
| 231 |
zcor = 1./(1.-RETV*zqs) |
| 232 |
zqs = zqs*zcor |
| 233 |
zdqs = FOEDE(zt,zdelta,zcvm5,zqs,zcor) |
| 234 |
ELSE |
| 235 |
IF (zt .LT. t_coup) THEN |
| 236 |
zqs = qsats(zt) / pplay(i,k) |
| 237 |
zdqs = dqsats(zt,zqs) |
| 238 |
ELSE |
| 239 |
zqs = qsatl(zt) / pplay(i,k) |
| 240 |
zdqs = dqsatl(zt,zqs) |
| 241 |
ENDIF |
| 242 |
ENDIF |
| 243 |
c |
| 244 |
c calculer la fraction nuageuse (processus humide): |
| 245 |
c |
| 246 |
zfr = (zq+ratqs*zq-zqs) / (2.0*ratqs*zq) |
| 247 |
zfr = MAX(0.0,MIN(1.0,zfr)) |
| 248 |
IF (.NOT.richum) zfr = 0.0 |
| 249 |
c |
| 250 |
c calculer le nombre de Richardson: |
| 251 |
c |
| 252 |
IF (tvirtu) THEN |
| 253 |
ztvd =( t(i,k) |
| 254 |
. + zdphi/RCPD/(1.+RVTMP2*zq) |
| 255 |
. *( (1.-zfr) + zfr*(1.+RLVTT*zqs/RD/zt)/(1.+zdqs) ) |
| 256 |
. )*(1.+RETV*q(i,k)) |
| 257 |
ztvu =( t(i,k-1) |
| 258 |
. - zdphi/RCPD/(1.+RVTMP2*zq) |
| 259 |
. *( (1.-zfr) + zfr*(1.+RLVTT*zqs/RD/zt)/(1.+zdqs) ) |
| 260 |
. )*(1.+RETV*q(i,k-1)) |
| 261 |
zri(i) =zmgeom(i)*(ztvd-ztvu)/(zdu2*0.5*(ztvd+ztvu)) |
| 262 |
zri(i) = zri(i) |
| 263 |
. + zmgeom(i)*zmgeom(i)/RG*gamt(k) |
| 264 |
. *(paprs(i,k)/101325.0)**RKAPPA |
| 265 |
. /(zdu2*0.5*(ztvd+ztvu)) |
| 266 |
c |
| 267 |
ELSE ! calcul de Ridchardson compatible LMD5 |
| 268 |
c |
| 269 |
zri(i) =(RCPD*(t(i,k)-t(i,k-1)) |
| 270 |
. -RD*0.5*(t(i,k)+t(i,k-1))/paprs(i,k) |
| 271 |
. *(pplay(i,k)-pplay(i,k-1)) |
| 272 |
. )*zmgeom(i)/(zdu2*0.5*RCPD*(t(i,k-1)+t(i,k))) |
| 273 |
zri(i) = zri(i) + |
| 274 |
. zmgeom(i)*zmgeom(i)*gamt(k)/RG |
| 275 |
cSB . /(paprs(i,k)/101325.0)**RKAPPA |
| 276 |
. *(paprs(i,k)/101325.0)**RKAPPA |
| 277 |
. /(zdu2*0.5*(t(i,k-1)+t(i,k))) |
| 278 |
ENDIF |
| 279 |
c |
| 280 |
c finalement, les coefficients d'echange sont obtenus: |
| 281 |
c |
| 282 |
zcdn=SQRT(zdu2) / zmgeom(i) * RG |
| 283 |
c |
| 284 |
IF (opt_ec) THEN |
| 285 |
z2geomf=zgeop(i,k-1)+zgeop(i,k) |
| 286 |
zalm2=(0.5*ckap/RG*z2geomf |
| 287 |
. /(1.+0.5*ckap/rg/clam*z2geomf))**2 |
| 288 |
zalh2=(0.5*ckap/rg*z2geomf |
| 289 |
. /(1.+0.5*ckap/RG/(clam*SQRT(1.5*cd))*z2geomf))**2 |
| 290 |
IF (zri(i).LT.0.0) THEN ! situation instable |
| 291 |
zscf = ((zgeop(i,k)/zgeop(i,k-1))**(1./3.)-1.)**3 |
| 292 |
. / (zmgeom(i)/RG)**3 / (zgeop(i,k-1)/RG) |
| 293 |
zscf = SQRT(-zri(i)*zscf) |
| 294 |
zscfm = 1.0 / (1.0+3.0*cb*cc*zalm2*zscf) |
| 295 |
zscfh = 1.0 / (1.0+3.0*cb*cc*zalh2*zscf) |
| 296 |
pcfm(i,k)=zcdn*zalm2*(1.-2.0*cb*zri(i)*zscfm) |
| 297 |
pcfh(i,k)=zcdn*zalh2*(1.-3.0*cb*zri(i)*zscfh) |
| 298 |
ELSE ! situation stable |
| 299 |
zscf=SQRT(1.+cd*zri(i)) |
| 300 |
pcfm(i,k)=zcdn*zalm2/(1.+2.0*cb*zri(i)/zscf) |
| 301 |
pcfh(i,k)=zcdn*zalh2/(1.+3.0*cb*zri(i)*zscf) |
| 302 |
ENDIF |
| 303 |
ELSE |
| 304 |
zl2(i)=(mixlen*MAX(0.0,(paprs(i,k)-paprs(i,itop(i)+1)) |
| 305 |
. /(paprs(i,2)-paprs(i,itop(i)+1)) ))**2 |
| 306 |
pcfm(i,k)=sqrt(max(zcdn*zcdn*(ric-zri(i))/ric, kstable)) |
| 307 |
pcfm(i,k)= zl2(i)* pcfm(i,k) |
| 308 |
pcfh(i,k) = pcfm(i,k) /prandtl ! h et m different |
| 309 |
ENDIF |
| 310 |
ENDDO |
| 311 |
ENDDO |
| 312 |
c |
| 313 |
c Au-dela du sommet, pas de diffusion turbulente: |
| 314 |
c |
| 315 |
DO i = 1, knon |
| 316 |
IF (itop(i)+1 .LE. klev) THEN |
| 317 |
DO k = itop(i)+1, klev |
| 318 |
pcfh(i,k) = 0.0 |
| 319 |
pcfm(i,k) = 0.0 |
| 320 |
ENDDO |
| 321 |
ENDIF |
| 322 |
ENDDO |
| 323 |
c |
| 324 |
RETURN |
| 325 |
END |
Parent Directory
| 
Revision Log