| 1 |
SUBROUTINE coefkz(nsrf, knon, paprs, pplay, ksta, ksta_ter, ts, rugos, u, v, & |
| 2 |
t, q, qsurf, pcfm, pcfh) |
| 3 |
|
| 4 |
! Auteur(s) F. Hourdin, M. Forichon, Z.X. Li (LMD/CNRS) date: 1993/09/22 |
| 5 |
! (une version strictement identique a l'ancien modele) |
| 6 |
! Objet: calculer le coefficient du frottement du sol (Cdrag) et les |
| 7 |
! coefficients d'echange turbulent dans l'atmosphere. |
| 8 |
|
| 9 |
USE indicesol, ONLY : is_oce |
| 10 |
USE dimphy, ONLY : klev, klon, max |
| 11 |
USE iniprint, ONLY : prt_level |
| 12 |
USE suphec_m, ONLY : rcpd, rd, retv, rg, rkappa, rlstt, rlvtt, rtt |
| 13 |
USE yoethf_m, ONLY : r2es, r5ies, r5les, rvtmp2 |
| 14 |
USE fcttre, ONLY : dqsatl, dqsats, foede, foeew, qsatl, qsats, thermcep |
| 15 |
USE conf_phys_m, ONLY : iflag_pbl |
| 16 |
|
| 17 |
IMPLICIT none |
| 18 |
|
| 19 |
! Arguments: |
| 20 |
! nsrf-----input-I- indicateur de la nature du sol |
| 21 |
! knon-----input-I- nombre de points a traiter |
| 22 |
! paprs----input-R- pression a chaque intercouche (en Pa) |
| 23 |
! pplay----input-R- pression au milieu de chaque couche (en Pa) |
| 24 |
! ts-------input-R- temperature du sol (en Kelvin) |
| 25 |
! rugos----input-R- longeur de rugosite (en m) |
| 26 |
! u--------input-R- vitesse u |
| 27 |
! v--------input-R- vitesse v |
| 28 |
! q--------input-R- vapeur d'eau (kg/kg) |
| 29 |
|
| 30 |
! itop-----output-I- numero de couche du sommet de la couche limite |
| 31 |
! pcfm-----output-R- coefficients a calculer (vitesse) |
| 32 |
! pcfh-----output-R- coefficients a calculer (chaleur et humidite) |
| 33 |
|
| 34 |
INTEGER knon, nsrf |
| 35 |
REAL ts(klon) |
| 36 |
REAL paprs(klon, klev+1), pplay(klon, klev) |
| 37 |
REAL u(klon, klev), v(klon, klev), q(klon, klev) |
| 38 |
REAL, intent(in):: t(klon, klev) ! temperature (K) |
| 39 |
REAL rugos(klon) |
| 40 |
|
| 41 |
REAL pcfm(klon, klev), pcfh(klon, klev) |
| 42 |
INTEGER itop(klon) |
| 43 |
|
| 44 |
! Quelques constantes et options: |
| 45 |
|
| 46 |
REAL cepdu2, ckap, cb, cc, cd, clam |
| 47 |
PARAMETER (cepdu2 =(0.1)**2) |
| 48 |
PARAMETER (CKAP=0.4) |
| 49 |
PARAMETER (cb=5.0) |
| 50 |
PARAMETER (cc=5.0) |
| 51 |
PARAMETER (cd=5.0) |
| 52 |
PARAMETER (clam=160.0) |
| 53 |
REAL ratqs ! largeur de distribution de vapeur d'eau |
| 54 |
PARAMETER (ratqs=0.05) |
| 55 |
LOGICAL richum ! utilise le nombre de Richardson humide |
| 56 |
PARAMETER (richum=.TRUE.) |
| 57 |
REAL ric ! nombre de Richardson critique |
| 58 |
PARAMETER(ric=0.4) |
| 59 |
REAL prandtl |
| 60 |
PARAMETER (prandtl=0.4) |
| 61 |
REAL kstable ! diffusion minimale (situation stable) |
| 62 |
! GKtest |
| 63 |
! PARAMETER (kstable=1.0e-10) |
| 64 |
REAL ksta, ksta_ter |
| 65 |
!IM: 261103 REAL kstable_ter, kstable_sinon |
| 66 |
!IM: 211003 cf GK PARAMETER (kstable_ter = 1.0e-6) |
| 67 |
!IM: 261103 PARAMETER (kstable_ter = 1.0e-8) |
| 68 |
!IM: 261103 PARAMETER (kstable_ter = 1.0e-10) |
| 69 |
!IM: 261103 PARAMETER (kstable_sinon = 1.0e-10) |
| 70 |
! fin GKtest |
| 71 |
REAL mixlen ! constante controlant longueur de melange |
| 72 |
PARAMETER (mixlen=35.0) |
| 73 |
INTEGER isommet ! le sommet de la couche limite |
| 74 |
PARAMETER (isommet=klev) |
| 75 |
LOGICAL tvirtu ! calculer Ri d'une maniere plus performante |
| 76 |
PARAMETER (tvirtu=.TRUE.) |
| 77 |
LOGICAL opt_ec ! formule du Centre Europeen dans l'atmosphere |
| 78 |
PARAMETER (opt_ec=.FALSE.) |
| 79 |
|
| 80 |
! Variables locales: |
| 81 |
|
| 82 |
INTEGER i, k, kk !IM 120704 |
| 83 |
REAL zgeop(klon, klev) |
| 84 |
REAL zmgeom(klon) |
| 85 |
REAL zri(klon) |
| 86 |
REAL zl2(klon) |
| 87 |
|
| 88 |
REAL u1(klon), v1(klon), t1(klon), q1(klon), z1(klon) |
| 89 |
REAL pcfm1(klon), pcfh1(klon) |
| 90 |
|
| 91 |
REAL zdphi, zdu2, ztvd, ztvu, zcdn |
| 92 |
REAL zscf |
| 93 |
REAL zt, zq, zdelta, zcvm5, zcor, zqs, zfr, zdqs |
| 94 |
REAL z2geomf, zalh2, zalm2, zscfh, zscfm |
| 95 |
REAL t_coup |
| 96 |
PARAMETER (t_coup=273.15) |
| 97 |
!IM |
| 98 |
LOGICAL check |
| 99 |
PARAMETER (check=.false.) |
| 100 |
|
| 101 |
! contre-gradient pour la chaleur sensible: Kelvin/metre |
| 102 |
REAL gamt(2:klev) |
| 103 |
real qsurf(klon) |
| 104 |
|
| 105 |
LOGICAL appel1er |
| 106 |
SAVE appel1er |
| 107 |
|
| 108 |
! Fonctions thermodynamiques et fonctions d'instabilite |
| 109 |
REAL fsta, fins, x |
| 110 |
LOGICAL zxli ! utiliser un jeu de fonctions simples |
| 111 |
PARAMETER (zxli=.FALSE.) |
| 112 |
|
| 113 |
fsta(x) = 1.0 / (1.0+10.0*x*(1+8.0*x)) |
| 114 |
fins(x) = SQRT(1.0-18.0*x) |
| 115 |
|
| 116 |
DATA appel1er /.TRUE./ |
| 117 |
|
| 118 |
!-------------------------------------------------------------------- |
| 119 |
|
| 120 |
IF (appel1er) THEN |
| 121 |
if (prt_level > 9) THEN |
| 122 |
print *, 'coefkz, opt_ec:', opt_ec |
| 123 |
print *, 'coefkz, richum:', richum |
| 124 |
IF (richum) print *, 'coefkz, ratqs:', ratqs |
| 125 |
print *, 'coefkz, isommet:', isommet |
| 126 |
print *, 'coefkz, tvirtu:', tvirtu |
| 127 |
appel1er = .FALSE. |
| 128 |
endif |
| 129 |
ENDIF |
| 130 |
|
| 131 |
! Initialiser les sorties |
| 132 |
|
| 133 |
DO k = 1, klev |
| 134 |
DO i = 1, knon |
| 135 |
pcfm(i, k) = 0.0 |
| 136 |
pcfh(i, k) = 0.0 |
| 137 |
ENDDO |
| 138 |
ENDDO |
| 139 |
DO i = 1, knon |
| 140 |
itop(i) = 0 |
| 141 |
ENDDO |
| 142 |
|
| 143 |
! Prescrire la valeur de contre-gradient |
| 144 |
|
| 145 |
if (iflag_pbl.eq.1) then |
| 146 |
DO k = 3, klev |
| 147 |
gamt(k) = -1.0E-03 |
| 148 |
ENDDO |
| 149 |
gamt(2) = -2.5E-03 |
| 150 |
else |
| 151 |
DO k = 2, klev |
| 152 |
gamt(k) = 0.0 |
| 153 |
ENDDO |
| 154 |
ENDIF |
| 155 |
|
| 156 |
IF ( nsrf .NE. is_oce ) THEN |
| 157 |
kstable = ksta_ter |
| 158 |
ELSE |
| 159 |
kstable = ksta |
| 160 |
ENDIF |
| 161 |
|
| 162 |
! Calculer les géopotentiels de chaque couche |
| 163 |
|
| 164 |
DO i = 1, knon |
| 165 |
zgeop(i, 1) = RD * t(i, 1) / (0.5 * (paprs(i, 1) + pplay(i, 1))) & |
| 166 |
* (paprs(i, 1) - pplay(i, 1)) |
| 167 |
ENDDO |
| 168 |
DO k = 2, klev |
| 169 |
DO i = 1, knon |
| 170 |
zgeop(i, k) = zgeop(i, k-1) & |
| 171 |
+ RD * 0.5*(t(i, k-1)+t(i, k)) / paprs(i, k) & |
| 172 |
* (pplay(i, k-1)-pplay(i, k)) |
| 173 |
ENDDO |
| 174 |
ENDDO |
| 175 |
|
| 176 |
! Calculer le frottement au sol (Cdrag) |
| 177 |
|
| 178 |
DO i = 1, knon |
| 179 |
u1(i) = u(i, 1) |
| 180 |
v1(i) = v(i, 1) |
| 181 |
t1(i) = t(i, 1) |
| 182 |
q1(i) = q(i, 1) |
| 183 |
z1(i) = zgeop(i, 1) |
| 184 |
ENDDO |
| 185 |
|
| 186 |
CALL clcdrag(klon, knon, nsrf, zxli, & |
| 187 |
u1, v1, t1, q1, z1, & |
| 188 |
ts, qsurf, rugos, & |
| 189 |
pcfm1, pcfh1) |
| 190 |
!IM $ ts, qsurf, rugos, |
| 191 |
|
| 192 |
DO i = 1, knon |
| 193 |
pcfm(i, 1)=pcfm1(i) |
| 194 |
pcfh(i, 1)=pcfh1(i) |
| 195 |
ENDDO |
| 196 |
|
| 197 |
! Calculer les coefficients turbulents dans l'atmosphere |
| 198 |
|
| 199 |
DO i = 1, knon |
| 200 |
itop(i) = isommet |
| 201 |
ENDDO |
| 202 |
|
| 203 |
DO k = 2, isommet |
| 204 |
DO i = 1, knon |
| 205 |
zdu2=MAX(cepdu2, (u(i, k)-u(i, k-1))**2 & |
| 206 |
+(v(i, k)-v(i, k-1))**2) |
| 207 |
zmgeom(i)=zgeop(i, k)-zgeop(i, k-1) |
| 208 |
zdphi =zmgeom(i) / 2.0 |
| 209 |
zt = (t(i, k)+t(i, k-1)) * 0.5 |
| 210 |
zq = (q(i, k)+q(i, k-1)) * 0.5 |
| 211 |
|
| 212 |
! calculer Qs et dQs/dT: |
| 213 |
|
| 214 |
IF (thermcep) THEN |
| 215 |
zdelta = MAX(0., SIGN(1., RTT-zt)) |
| 216 |
zcvm5 = R5LES*RLVTT/RCPD/(1.0+RVTMP2*zq)*(1.-zdelta) & |
| 217 |
+ R5IES*RLSTT/RCPD/(1.0+RVTMP2*zq)*zdelta |
| 218 |
zqs = R2ES * FOEEW(zt, zdelta) / pplay(i, k) |
| 219 |
zqs = MIN(0.5, zqs) |
| 220 |
zcor = 1./(1.-RETV*zqs) |
| 221 |
zqs = zqs*zcor |
| 222 |
zdqs = FOEDE(zt, zdelta, zcvm5, zqs, zcor) |
| 223 |
ELSE |
| 224 |
IF (zt .LT. t_coup) THEN |
| 225 |
zqs = qsats(zt) / pplay(i, k) |
| 226 |
zdqs = dqsats(zt, zqs) |
| 227 |
ELSE |
| 228 |
zqs = qsatl(zt) / pplay(i, k) |
| 229 |
zdqs = dqsatl(zt, zqs) |
| 230 |
ENDIF |
| 231 |
ENDIF |
| 232 |
|
| 233 |
! calculer la fraction nuageuse (processus humide): |
| 234 |
|
| 235 |
zfr = (zq+ratqs*zq-zqs) / (2.0*ratqs*zq) |
| 236 |
zfr = MAX(0.0, MIN(1.0, zfr)) |
| 237 |
IF (.NOT.richum) zfr = 0.0 |
| 238 |
|
| 239 |
! calculer le nombre de Richardson: |
| 240 |
|
| 241 |
IF (tvirtu) THEN |
| 242 |
ztvd =( t(i, k) & |
| 243 |
+ zdphi/RCPD/(1.+RVTMP2*zq) & |
| 244 |
*( (1.-zfr) + zfr*(1.+RLVTT*zqs/RD/zt)/(1.+zdqs) ) & |
| 245 |
)*(1.+RETV*q(i, k)) |
| 246 |
ztvu =( t(i, k-1) & |
| 247 |
- zdphi/RCPD/(1.+RVTMP2*zq) & |
| 248 |
*( (1.-zfr) + zfr*(1.+RLVTT*zqs/RD/zt)/(1.+zdqs) ) & |
| 249 |
)*(1.+RETV*q(i, k-1)) |
| 250 |
zri(i) =zmgeom(i)*(ztvd-ztvu)/(zdu2*0.5*(ztvd+ztvu)) |
| 251 |
zri(i) = zri(i) & |
| 252 |
+ zmgeom(i)*zmgeom(i)/RG*gamt(k) & |
| 253 |
*(paprs(i, k)/101325.0)**RKAPPA & |
| 254 |
/(zdu2*0.5*(ztvd+ztvu)) |
| 255 |
|
| 256 |
ELSE ! calcul de Ridchardson compatible LMD5 |
| 257 |
|
| 258 |
zri(i) =(RCPD*(t(i, k)-t(i, k-1)) & |
| 259 |
-RD*0.5*(t(i, k)+t(i, k-1))/paprs(i, k) & |
| 260 |
*(pplay(i, k)-pplay(i, k-1)) & |
| 261 |
)*zmgeom(i)/(zdu2*0.5*RCPD*(t(i, k-1)+t(i, k))) |
| 262 |
zri(i) = zri(i) + & |
| 263 |
zmgeom(i)*zmgeom(i)*gamt(k)/RG & |
| 264 |
*(paprs(i, k)/101325.0)**RKAPPA & |
| 265 |
/(zdu2*0.5*(t(i, k-1)+t(i, k))) |
| 266 |
ENDIF |
| 267 |
|
| 268 |
! finalement, les coefficients d'echange sont obtenus: |
| 269 |
|
| 270 |
zcdn=SQRT(zdu2) / zmgeom(i) * RG |
| 271 |
|
| 272 |
IF (opt_ec) THEN |
| 273 |
z2geomf=zgeop(i, k-1)+zgeop(i, k) |
| 274 |
zalm2=(0.5*ckap/RG*z2geomf & |
| 275 |
/(1.+0.5*ckap/rg/clam*z2geomf))**2 |
| 276 |
zalh2=(0.5*ckap/rg*z2geomf & |
| 277 |
/(1.+0.5*ckap/RG/(clam*SQRT(1.5*cd))*z2geomf))**2 |
| 278 |
IF (zri(i).LT.0.0) THEN ! situation instable |
| 279 |
zscf = ((zgeop(i, k)/zgeop(i, k-1))**(1./3.)-1.)**3 & |
| 280 |
/ (zmgeom(i)/RG)**3 / (zgeop(i, k-1)/RG) |
| 281 |
zscf = SQRT(-zri(i)*zscf) |
| 282 |
zscfm = 1.0 / (1.0+3.0*cb*cc*zalm2*zscf) |
| 283 |
zscfh = 1.0 / (1.0+3.0*cb*cc*zalh2*zscf) |
| 284 |
pcfm(i, k)=zcdn*zalm2*(1.-2.0*cb*zri(i)*zscfm) |
| 285 |
pcfh(i, k)=zcdn*zalh2*(1.-3.0*cb*zri(i)*zscfh) |
| 286 |
ELSE ! situation stable |
| 287 |
zscf=SQRT(1.+cd*zri(i)) |
| 288 |
pcfm(i, k)=zcdn*zalm2/(1.+2.0*cb*zri(i)/zscf) |
| 289 |
pcfh(i, k)=zcdn*zalh2/(1.+3.0*cb*zri(i)*zscf) |
| 290 |
ENDIF |
| 291 |
ELSE |
| 292 |
zl2(i)=(mixlen*MAX(0.0, (paprs(i, k)-paprs(i, itop(i)+1)) & |
| 293 |
/(paprs(i, 2)-paprs(i, itop(i)+1)) ))**2 |
| 294 |
pcfm(i, k)=sqrt(max(zcdn*zcdn*(ric-zri(i))/ric, kstable)) |
| 295 |
pcfm(i, k)= zl2(i)* pcfm(i, k) |
| 296 |
pcfh(i, k) = pcfm(i, k) /prandtl ! h et m different |
| 297 |
ENDIF |
| 298 |
ENDDO |
| 299 |
ENDDO |
| 300 |
|
| 301 |
! Au-dela du sommet, pas de diffusion turbulente: |
| 302 |
|
| 303 |
DO i = 1, knon |
| 304 |
IF (itop(i)+1 .LE. klev) THEN |
| 305 |
DO k = itop(i)+1, klev |
| 306 |
pcfh(i, k) = 0.0 |
| 307 |
pcfm(i, k) = 0.0 |
| 308 |
ENDDO |
| 309 |
ENDIF |
| 310 |
ENDDO |
| 311 |
|
| 312 |
END SUBROUTINE coefkz |
Parent Directory
| 
Revision Log