| 1 | SUBROUTINE surflayer_interpol(ngrid,nlay,pz0, |
|---|
| 2 | & pg,pz,pu,pv,wmax,pts,ph,z_out,Teta_out,u_out,ustar,tstar,L_mo) |
|---|
| 3 | IMPLICIT NONE |
|---|
| 4 | !======================================================================= |
|---|
| 5 | ! |
|---|
| 6 | ! Subject: interpolation of temperature and velocity norm in the surface layer |
|---|
| 7 | ! by recomputation of surface layer quantities (Richardson, Prandtl, Reynolds, u*, teta*) |
|---|
| 8 | ! ------- |
|---|
| 9 | ! |
|---|
| 10 | ! Author: Arnaud Colaitis 05/08/11 |
|---|
| 11 | ! ------- |
|---|
| 12 | ! |
|---|
| 13 | ! Arguments: |
|---|
| 14 | ! ---------- |
|---|
| 15 | ! |
|---|
| 16 | ! inputs: |
|---|
| 17 | ! ------ |
|---|
| 18 | ! ngrid size of the horizontal grid |
|---|
| 19 | ! pg gravity (m s -2) |
|---|
| 20 | ! pz(ngrid,nlay) height of layers |
|---|
| 21 | ! pu(ngrid,nlay) u component of the wind |
|---|
| 22 | ! pv(ngrid,nlay) v component of the wind |
|---|
| 23 | ! pts(ngrid) surface temperature |
|---|
| 24 | ! ph(ngrid) potential temperature T*(p/ps)^kappa |
|---|
| 25 | ! |
|---|
| 26 | ! |
|---|
| 27 | !======================================================================= |
|---|
| 28 | ! |
|---|
| 29 | !----------------------------------------------------------------------- |
|---|
| 30 | ! Declarations: |
|---|
| 31 | ! ------------- |
|---|
| 32 | |
|---|
| 33 | #include "comcstfi.h" |
|---|
| 34 | #include "callkeys.h" |
|---|
| 35 | |
|---|
| 36 | ! Arguments: |
|---|
| 37 | ! ---------- |
|---|
| 38 | |
|---|
| 39 | INTEGER, INTENT(IN) :: ngrid,nlay |
|---|
| 40 | REAL, INTENT(IN) :: pz0(ngrid) |
|---|
| 41 | REAL, INTENT(IN) :: pg,pz(ngrid,nlay) |
|---|
| 42 | REAL, INTENT(IN) :: pu(ngrid,nlay),pv(ngrid,nlay) |
|---|
| 43 | REAL, INTENT(IN) :: wmax(ngrid) |
|---|
| 44 | REAL, INTENT(IN) :: pts(ngrid),ph(ngrid,nlay) |
|---|
| 45 | REAL, INTENT(IN) :: z_out ! output height (in m above surface) |
|---|
| 46 | REAL, INTENT(OUT) :: Teta_out(ngrid),u_out(ngrid)! interpolated fields at z_out : potential temperature and norm(uv) |
|---|
| 47 | REAL, INTENT(OUT) :: ustar(ngrid), tstar(ngrid) ! u* and teta* |
|---|
| 48 | REAL, INTENT(OUT) :: L_mo(ngrid) ! Monin-Obukhov length |
|---|
| 49 | |
|---|
| 50 | ! Local: |
|---|
| 51 | ! ------ |
|---|
| 52 | |
|---|
| 53 | INTEGER ig |
|---|
| 54 | |
|---|
| 55 | REAL karman,nu |
|---|
| 56 | DATA karman,nu/.41,0.001/ |
|---|
| 57 | SAVE karman,nu |
|---|
| 58 | |
|---|
| 59 | ! Local(2): |
|---|
| 60 | ! --------- |
|---|
| 61 | REAL zout |
|---|
| 62 | |
|---|
| 63 | REAL rib(ngrid) ! Bulk Richardson number |
|---|
| 64 | REAL fm(ngrid) ! stability function for momentum |
|---|
| 65 | REAL fh(ngrid) ! stability function for heat |
|---|
| 66 | REAL z1z0,z1z0t ! ratios z1/z0 and z1/z0T |
|---|
| 67 | |
|---|
| 68 | ! phim = 1+betam*zeta or (1-bm*zeta)**am |
|---|
| 69 | ! phih = alphah + betah*zeta or alphah(1.-bh*zeta)**ah |
|---|
| 70 | REAL betam, betah, alphah, bm, bh, lambda |
|---|
| 71 | ! ah and am are assumed to be -0.25 and -0.5 respectively |
|---|
| 72 | |
|---|
| 73 | REAL cdn(ngrid),chn(ngrid) ! neutral momentum and heat drag coefficient |
|---|
| 74 | REAL pz0t ! initial thermal roughness length. (local) |
|---|
| 75 | REAL ric ! critical richardson number |
|---|
| 76 | REAL reynolds(ngrid) ! reynolds number for UBL |
|---|
| 77 | REAL prandtl(ngrid) ! prandtl number for UBL |
|---|
| 78 | REAL pz0tcomp(ngrid) ! computed z0t |
|---|
| 79 | REAL ite |
|---|
| 80 | REAL residual,zcd0,z1 |
|---|
| 81 | REAL pcdv(ngrid),pcdh(ngrid) |
|---|
| 82 | ! For output : |
|---|
| 83 | |
|---|
| 84 | REAL zu2(ngrid) ! Large-scale wind at first layer |
|---|
| 85 | !----------------------------------------------------------------------- |
|---|
| 86 | ! couche de surface: |
|---|
| 87 | ! ------------------ |
|---|
| 88 | |
|---|
| 89 | c Init : |
|---|
| 90 | |
|---|
| 91 | L_mo(:)=0. |
|---|
| 92 | ustar(:)=0. |
|---|
| 93 | tstar(:)=0. |
|---|
| 94 | zout=z_out |
|---|
| 95 | |
|---|
| 96 | reynolds(:)=0. |
|---|
| 97 | |
|---|
| 98 | ! New formulation (AC) : |
|---|
| 99 | |
|---|
| 100 | ! phim = 1+betam*zeta or (1-bm*zeta)**am |
|---|
| 101 | ! phih = alphah + betah*zeta or alphah(1.-bh*zeta)**ah |
|---|
| 102 | ! am=-0.25, ah=-0.5 |
|---|
| 103 | |
|---|
| 104 | pz0t = 0. ! for the sake of simplicity |
|---|
| 105 | pz0tcomp(:) = 0.1*pz0(:) |
|---|
| 106 | rib(:)=0. |
|---|
| 107 | pcdv(:)=0. |
|---|
| 108 | pcdh(:)=0. |
|---|
| 109 | |
|---|
| 110 | ! this formulation assumes alphah=1., implying betah=betam |
|---|
| 111 | ! We use Dyer et al. parameters, as they cover a broad range of Richardson numbers : |
|---|
| 112 | bm=16. !UBL |
|---|
| 113 | bh=16. !UBL |
|---|
| 114 | alphah=1. |
|---|
| 115 | betam=5. !SBL |
|---|
| 116 | betah=5. !SBL |
|---|
| 117 | lambda=(sqrt(bh/bm))/alphah |
|---|
| 118 | ric=betah/(betam**2) |
|---|
| 119 | |
|---|
| 120 | DO ig=1,ngrid |
|---|
| 121 | |
|---|
| 122 | ite=0. |
|---|
| 123 | residual=abs(pz0tcomp(ig)-pz0t) |
|---|
| 124 | |
|---|
| 125 | do while((residual .gt. 0.01*pz0(ig)) .and. (ite .lt. 10.)) |
|---|
| 126 | |
|---|
| 127 | pz0t=pz0tcomp(ig) |
|---|
| 128 | |
|---|
| 129 | if ((pu(ig,1) .ne. 0.) .or. (pv(ig,1) .ne. 0.)) then |
|---|
| 130 | |
|---|
| 131 | ! Classical Richardson number formulation |
|---|
| 132 | |
|---|
| 133 | ! rib(ig) = (pg/ph(ig,1))*pz(ig,1)*(ph(ig,1)-pts(ig)) |
|---|
| 134 | ! & /(pu(ig,1)*pu(ig,1) + pv(ig,1)*pv(ig,1)) |
|---|
| 135 | |
|---|
| 136 | ! Richardson number formulation proposed by D.E. England et al. (1995) |
|---|
| 137 | |
|---|
| 138 | ! IF((pu(ig,1)*pu(ig,1) + pv(ig,1)*pv(ig,1) .lt. 1.) |
|---|
| 139 | ! & .and. (wmax(ig) .gt. 0.)) THEN |
|---|
| 140 | zu2(ig)= |
|---|
| 141 | ! & (MAX(pu(ig,1)*pu(ig,1) + pv(ig,1)*pv(ig,1),wmax(ig)**2)) |
|---|
| 142 | & ( pu(ig,1)*pu(ig,1) + pv(ig,1)*pv(ig,1)) |
|---|
| 143 | ! & pu(ig,1)*pu(ig,1) + pv(ig,1)*pv(ig,1) |
|---|
| 144 | ! ELSE |
|---|
| 145 | ! zu2(ig)=pu(ig,1)*pu(ig,1) + pv(ig,1)*pv(ig,1) |
|---|
| 146 | ! ENDIF |
|---|
| 147 | |
|---|
| 148 | if(.not.callrichsl) then |
|---|
| 149 | rib(ig) = (pg/ph(ig,1)) |
|---|
| 150 | & *sqrt(pz(ig,1)*pz0(ig)) |
|---|
| 151 | & *(((log(pz(ig,1)/pz0(ig)))**2)/(log(pz(ig,1)/pz0t))) |
|---|
| 152 | & *(ph(ig,1)-pts(ig))/(zu2(ig)+6.) |
|---|
| 153 | else |
|---|
| 154 | rib(ig) = (pg/ph(ig,1)) |
|---|
| 155 | ! & *pz(ig,1)*pz0(ig)/sqrt(pz(ig,1)*pz0t) |
|---|
| 156 | & *sqrt(pz(ig,1)*pz0(ig)) |
|---|
| 157 | & *(((log(pz(ig,1)/pz0(ig)))**2)/(log(pz(ig,1)/pz0t))) |
|---|
| 158 | & *(ph(ig,1)-pts(ig))/(zu2(ig) + (0.5*wmax(ig))**2) |
|---|
| 159 | |
|---|
| 160 | ! & /(MAX(pu(ig,1)*pu(ig,1) + pv(ig,1)*pv(ig,1),wmax(ig)**2)) |
|---|
| 161 | ! & /( pu(ig,1)*pu(ig,1) + pv(ig,1)*pv(ig,1) + wmax(ig)**2) |
|---|
| 162 | |
|---|
| 163 | endif |
|---|
| 164 | |
|---|
| 165 | else |
|---|
| 166 | print*,'warning, infinite Richardson at surface' |
|---|
| 167 | print*,pu(ig,1),pv(ig,1) |
|---|
| 168 | rib(ig) = ric ! traiter ce cas ! |
|---|
| 169 | endif |
|---|
| 170 | |
|---|
| 171 | z1z0=pz(ig,1)/pz0(ig) |
|---|
| 172 | z1z0t=pz(ig,1)/pz0t |
|---|
| 173 | |
|---|
| 174 | cdn(ig)=karman/log(z1z0) |
|---|
| 175 | cdn(ig)=cdn(ig)*cdn(ig) |
|---|
| 176 | chn(ig)=cdn(ig)*log(z1z0)/log(z1z0t) |
|---|
| 177 | |
|---|
| 178 | ! Stable case : |
|---|
| 179 | if (rib(ig) .gt. 0.) then |
|---|
| 180 | ! From D.E. England et al. (95) |
|---|
| 181 | prandtl(ig)=1. |
|---|
| 182 | if(rib(ig) .lt. ric) then |
|---|
| 183 | ! Assuming alphah=1. and bh=bm for stable conditions : |
|---|
| 184 | fm(ig)=((ric-rib(ig))/ric)**2 |
|---|
| 185 | fh(ig)=((ric-rib(ig))/ric)**2 |
|---|
| 186 | else |
|---|
| 187 | ! For Ri>Ric, we consider Ri->Infinity => no turbulent mixing at surface |
|---|
| 188 | fm(ig)=0. |
|---|
| 189 | fh(ig)=0. |
|---|
| 190 | endif |
|---|
| 191 | ! Unstable case : |
|---|
| 192 | else |
|---|
| 193 | ! From D.E. England et al. (95) |
|---|
| 194 | fm(ig)=sqrt(1.-lambda*bm*rib(ig)) |
|---|
| 195 | fh(ig)=(1./alphah)*((1.-lambda*bh*rib(ig))**0.5)* |
|---|
| 196 | & (1.-lambda*bm*rib(ig))**0.25 |
|---|
| 197 | prandtl(ig)=alphah*((1.-lambda*bm*rib(ig))**0.25)/ |
|---|
| 198 | & ((1.-lambda*bh*rib(ig))**0.5) |
|---|
| 199 | endif |
|---|
| 200 | |
|---|
| 201 | reynolds(ig)=karman*sqrt(fm(ig)) |
|---|
| 202 | & *sqrt(zu2(ig)+(0.5*wmax(ig))**2) |
|---|
| 203 | ! & *sqrt(pu(ig,1)**2 + pv(ig,1)**2) |
|---|
| 204 | & *pz0(ig)/(log(z1z0)*nu) |
|---|
| 205 | pz0tcomp(ig)=pz0(ig)*exp(-karman*7.3* |
|---|
| 206 | & (reynolds(ig)**0.25)*(prandtl(ig)**0.5)) |
|---|
| 207 | |
|---|
| 208 | |
|---|
| 209 | residual = abs(pz0t-pz0tcomp(ig)) |
|---|
| 210 | ite = ite+1 |
|---|
| 211 | ! if(ite .eq. 10) then |
|---|
| 212 | ! print*, 'iteration max reached' |
|---|
| 213 | ! endif |
|---|
| 214 | ! print*, "iteration nnumber, residual",ite,residual |
|---|
| 215 | |
|---|
| 216 | enddo ! of while |
|---|
| 217 | |
|---|
| 218 | pz0t=0. |
|---|
| 219 | |
|---|
| 220 | ! Drag computation : |
|---|
| 221 | |
|---|
| 222 | pcdv(ig)=cdn(ig)*fm(ig) |
|---|
| 223 | pcdh(ig)=chn(ig)*fh(ig) |
|---|
| 224 | |
|---|
| 225 | ENDDO |
|---|
| 226 | |
|---|
| 227 | ! endif !of if callrichsl |
|---|
| 228 | |
|---|
| 229 | ! Large-scale wind at first layer (without gustiness) and |
|---|
| 230 | ! u* theta* computation |
|---|
| 231 | DO ig=1,ngrid |
|---|
| 232 | |
|---|
| 233 | if (rib(ig) .ge. ric) then |
|---|
| 234 | ustar(ig)=0. |
|---|
| 235 | tstar(ig)=0. |
|---|
| 236 | else |
|---|
| 237 | |
|---|
| 238 | ! ustar(ig)=karman*sqrt(fm(ig)*zu2(ig))/(log(pz(ig,1)/pz0(ig))) |
|---|
| 239 | ! tstar(ig)=karman*fh(ig)*(ph(ig,1)-pts(ig)) |
|---|
| 240 | ! & /(log(pz(ig,1)/pz0tcomp(ig))*sqrt(fm(ig))) |
|---|
| 241 | |
|---|
| 242 | !simpler definition of u* and teta*, equivalent to the formula above : |
|---|
| 243 | |
|---|
| 244 | ustar(ig)=sqrt(pcdv(ig))*sqrt(zu2(ig)) |
|---|
| 245 | tstar(ig)=-pcdh(ig)*(pts(ig)-ph(ig,1)) |
|---|
| 246 | & /sqrt(pcdv(ig)) |
|---|
| 247 | |
|---|
| 248 | if ((tstar(ig) .lt. -50) .and. callrichsl) then |
|---|
| 249 | print*, fh(ig),rib(ig),(ph(ig,1)-pts(ig)) |
|---|
| 250 | & ,log(pz(ig,1)/pz0tcomp(ig)),sqrt(fm(ig)) |
|---|
| 251 | endif |
|---|
| 252 | endif |
|---|
| 253 | ENDDO |
|---|
| 254 | |
|---|
| 255 | ! Monin Obukhov length : |
|---|
| 256 | |
|---|
| 257 | DO ig=1,ngrid |
|---|
| 258 | if (rib(ig) .gt. ric) then |
|---|
| 259 | L_mo(ig)=0. |
|---|
| 260 | else |
|---|
| 261 | L_mo(ig)=pts(ig)*(ustar(ig)**2)/(pg*karman*tstar(ig)) !as defined here, L is positive for SBL, negative for UBL |
|---|
| 262 | endif |
|---|
| 263 | ENDDO |
|---|
| 264 | |
|---|
| 265 | DO ig=1,ngrid |
|---|
| 266 | IF(zout .ge. pz(ig,1)) THEN |
|---|
| 267 | zout=1. |
|---|
| 268 | print*, 'Warning, z_out should be less than the first |
|---|
| 269 | & vertical grid-point' |
|---|
| 270 | print*, 'z1 =',pz(ig,1) |
|---|
| 271 | print*, 'z_out=',z_out |
|---|
| 272 | print*, 'z_out has been changed to 1m |
|---|
| 273 | & and computation is resumed' |
|---|
| 274 | ENDIF |
|---|
| 275 | |
|---|
| 276 | IF(zout .lt. pz0tcomp(ig)) THEN |
|---|
| 277 | zout=pz0tcomp(ig) |
|---|
| 278 | print*, 'Warning, z_out should be more than the thermal |
|---|
| 279 | & roughness length' |
|---|
| 280 | print*, 'z0 =',pz0tcomp(ig) |
|---|
| 281 | print*, 'z_out=',z_out |
|---|
| 282 | print*, 'z_out has been changed to z0t |
|---|
| 283 | & and computation is resumed' |
|---|
| 284 | ENDIF |
|---|
| 285 | ENDDO |
|---|
| 286 | |
|---|
| 287 | print*, 'interpolation of u and teta at z_out=',zout |
|---|
| 288 | |
|---|
| 289 | DO ig=1,ngrid |
|---|
| 290 | IF (L_mo(ig) .gt. 0.) THEN |
|---|
| 291 | u_out(ig)=(ustar(ig)/karman)*log(zout/pz0(ig)) + |
|---|
| 292 | & 5.*(ustar(ig)/(karman*L_mo(ig)))*(zout-pz0(ig)) |
|---|
| 293 | Teta_out(ig)=pts(ig)+(tstar(ig)/(prandtl(ig)*karman)) |
|---|
| 294 | & *log(zout/pz0tcomp(ig)) + |
|---|
| 295 | & 5.*(tstar(ig)/(prandtl(ig)*karman*L_mo(ig))) |
|---|
| 296 | & *(zout-pz0tcomp(ig)) |
|---|
| 297 | ELSEIF (L_mo(ig) .lt. 0.) THEN |
|---|
| 298 | |
|---|
| 299 | IF(L_mo(ig) .gt. -1000.) THEN |
|---|
| 300 | |
|---|
| 301 | u_out(ig)=(ustar(ig)/karman)*( |
|---|
| 302 | & 2.*atan((1.-16.*zout/L_mo(ig))**0.25) |
|---|
| 303 | & -2.*atan((1.-16.*pz0(ig)/L_mo(ig))**0.25) |
|---|
| 304 | & -2.*log(1.+(1.-16.*zout/L_mo(ig))**0.25) |
|---|
| 305 | & +2.*log(1.+(1.-16.*pz0(ig)/L_mo(ig))**0.25) |
|---|
| 306 | & - log(1.+sqrt(1.-16.*zout/L_mo(ig))) |
|---|
| 307 | & + log(1.+sqrt(1.-16.*pz0(ig)/L_mo(ig))) |
|---|
| 308 | & + log(zout/pz0(ig)) |
|---|
| 309 | & ) |
|---|
| 310 | |
|---|
| 311 | Teta_out(ig)=pts(ig)+(tstar(ig)/(prandtl(ig)*karman))*( |
|---|
| 312 | & 2.*log(1.+sqrt(1.-16.*pz0tcomp(ig)/L_mo(ig))) |
|---|
| 313 | & -2.*log(1.+sqrt(1.-16.*zout/L_mo(ig))) |
|---|
| 314 | & + log(zout/pz0tcomp(ig)) |
|---|
| 315 | & ) |
|---|
| 316 | |
|---|
| 317 | ELSE |
|---|
| 318 | |
|---|
| 319 | ! We have to treat the case where L is very large and negative, |
|---|
| 320 | ! corresponding to a very nearly stable atmosphere (but not quite !) |
|---|
| 321 | ! i.e. teta* <0 and almost zero. We choose the cutoff |
|---|
| 322 | ! corresponding to L_mo < -1000 and use a 3rd order taylor expansion. The difference |
|---|
| 323 | ! between the two expression for z/L = -1/1000 is -1.54324*10^-9 |
|---|
| 324 | ! (we do that to avoid using r*4 precision, otherwise, we get -inf values) |
|---|
| 325 | |
|---|
| 326 | u_out(ig)=(ustar(ig)/karman)*( |
|---|
| 327 | & (4./L_mo(ig))*(zout-pz0(ig)) |
|---|
| 328 | & + (20./(L_mo(ig))**2)*(zout**2-pz0(ig)**2) |
|---|
| 329 | & + (160./(L_mo(ig))**3)*(zout**3-pz0(ig)**3) |
|---|
| 330 | & + log(zout/pz0(ig)) |
|---|
| 331 | & ) |
|---|
| 332 | |
|---|
| 333 | Teta_out(ig)=pts(ig)+(tstar(ig)/(prandtl(ig)*karman))*( |
|---|
| 334 | & (8./L_mo(ig))*(zout-pz0tcomp(ig)) |
|---|
| 335 | & + (48./(L_mo(ig))**2)*(zout**2-pz0tcomp(ig)**2) |
|---|
| 336 | & + (1280./(3.*(L_mo(ig))**3))*(zout**3-pz0tcomp(ig)**3) |
|---|
| 337 | & + log(zout/pz0tcomp(ig)) |
|---|
| 338 | & ) |
|---|
| 339 | |
|---|
| 340 | ENDIF |
|---|
| 341 | ELSE |
|---|
| 342 | u_out(ig)=0. |
|---|
| 343 | Teta_out(ig)=pts(ig) |
|---|
| 344 | ENDIF |
|---|
| 345 | ENDDO |
|---|
| 346 | |
|---|
| 347 | ! when using convective adjustment without thermals, a vertical potential temperature |
|---|
| 348 | ! profile is assumed up to the thermal roughness length. Hence, theoretically, theta |
|---|
| 349 | ! interpolated at any height in the surface layer is theta at the first level. |
|---|
| 350 | |
|---|
| 351 | IF ((.not.calltherm).and.(calladj)) THEN |
|---|
| 352 | Teta_out(:)=ph(:,1) |
|---|
| 353 | ENDIF |
|---|
| 354 | |
|---|
| 355 | ! Usefull diagnostics for the interpolation routine : |
|---|
| 356 | |
|---|
| 357 | ! |
|---|
| 358 | ! call WRITEDIAGFI(ngrid,'Ri', |
|---|
| 359 | ! & 'Richardson','m', |
|---|
| 360 | ! & 2,rib) |
|---|
| 361 | ! |
|---|
| 362 | ! call WRITEDIAGFI(ngrid,'z0T', |
|---|
| 363 | ! & 'thermal roughness length','m', |
|---|
| 364 | ! & 2,pz0tcomp) |
|---|
| 365 | ! call WRITEDIAGFI(ngrid,'z0', |
|---|
| 366 | ! & 'roughness length','m', |
|---|
| 367 | ! & 2,pz0) |
|---|
| 368 | |
|---|
| 369 | |
|---|
| 370 | RETURN |
|---|
| 371 | END |
|---|
