[1] | 1 | MODULE diffusion_u_mod |
---|
| 2 | |
---|
| 3 | !------------------------------------------------------------------------------! |
---|
[484] | 4 | ! Current revisions: |
---|
[1] | 5 | ! ----------------- |
---|
[106] | 6 | ! |
---|
[1002] | 7 | ! |
---|
[1] | 8 | ! Former revisions: |
---|
| 9 | ! ----------------- |
---|
[3] | 10 | ! $Id: diffusion_u.f90 1002 2012-09-13 15:12:24Z raasch $ |
---|
[39] | 11 | ! |
---|
[1002] | 12 | ! 1001 2012-09-13 14:08:46Z raasch |
---|
| 13 | ! arrays comunicated by module instead of parameter list |
---|
| 14 | ! |
---|
[979] | 15 | ! 978 2012-08-09 08:28:32Z fricke |
---|
| 16 | ! outflow damping layer removed |
---|
| 17 | ! kmym_x/_y and kmyp_x/_y change to kmym and kmyp |
---|
| 18 | ! |
---|
[668] | 19 | ! 667 2010-12-23 12:06:00Z suehring/gryschka |
---|
| 20 | ! nxl-1, nxr+1, nys-1, nyn+1 replaced by nxlg, nxrg, nysg, nyng |
---|
| 21 | ! |
---|
[392] | 22 | ! 366 2009-08-25 08:06:27Z raasch |
---|
| 23 | ! bc_ns replaced by bc_ns_cyc |
---|
| 24 | ! |
---|
[110] | 25 | ! 106 2007-08-16 14:30:26Z raasch |
---|
| 26 | ! Momentumflux at top (uswst) included as boundary condition, |
---|
| 27 | ! i loop is starting from nxlu (needed for non-cyclic boundary conditions) |
---|
| 28 | ! |
---|
[77] | 29 | ! 75 2007-03-22 09:54:05Z raasch |
---|
| 30 | ! Wall functions now include diabatic conditions, call of routine wall_fluxes, |
---|
| 31 | ! z0 removed from argument list, uxrp eliminated |
---|
| 32 | ! |
---|
[39] | 33 | ! 20 2007-02-26 00:12:32Z raasch |
---|
| 34 | ! Bugfix: ddzw dimensioned 1:nzt"+1" |
---|
| 35 | ! |
---|
[3] | 36 | ! RCS Log replace by Id keyword, revision history cleaned up |
---|
| 37 | ! |
---|
[1] | 38 | ! Revision 1.15 2006/02/23 10:35:35 raasch |
---|
| 39 | ! nzb_2d replaced by nzb_u_outer in horizontal diffusion and by nzb_u_inner |
---|
| 40 | ! or nzb_diff_u, respectively, in vertical diffusion, |
---|
| 41 | ! wall functions added for north and south walls, +z0 in argument list, |
---|
| 42 | ! terms containing w(k-1,..) are removed from the Prandtl-layer equation |
---|
| 43 | ! because they cause errors at the edges of topography |
---|
| 44 | ! WARNING: loops containing the MAX function are still not properly vectorized! |
---|
| 45 | ! |
---|
| 46 | ! Revision 1.1 1997/09/12 06:23:51 raasch |
---|
| 47 | ! Initial revision |
---|
| 48 | ! |
---|
| 49 | ! |
---|
| 50 | ! Description: |
---|
| 51 | ! ------------ |
---|
| 52 | ! Diffusion term of the u-component |
---|
[51] | 53 | ! To do: additional damping (needed for non-cyclic bc) causes bad vectorization |
---|
| 54 | ! and slows down the speed on NEC about 5-10% |
---|
[1] | 55 | !------------------------------------------------------------------------------! |
---|
| 56 | |
---|
[56] | 57 | USE wall_fluxes_mod |
---|
| 58 | |
---|
[1] | 59 | PRIVATE |
---|
| 60 | PUBLIC diffusion_u |
---|
| 61 | |
---|
| 62 | INTERFACE diffusion_u |
---|
| 63 | MODULE PROCEDURE diffusion_u |
---|
| 64 | MODULE PROCEDURE diffusion_u_ij |
---|
| 65 | END INTERFACE diffusion_u |
---|
| 66 | |
---|
| 67 | CONTAINS |
---|
| 68 | |
---|
| 69 | |
---|
| 70 | !------------------------------------------------------------------------------! |
---|
| 71 | ! Call for all grid points |
---|
| 72 | !------------------------------------------------------------------------------! |
---|
[1001] | 73 | SUBROUTINE diffusion_u |
---|
[1] | 74 | |
---|
[1001] | 75 | USE arrays_3d |
---|
[1] | 76 | USE control_parameters |
---|
| 77 | USE grid_variables |
---|
| 78 | USE indices |
---|
| 79 | |
---|
| 80 | IMPLICIT NONE |
---|
| 81 | |
---|
| 82 | INTEGER :: i, j, k |
---|
[978] | 83 | REAL :: kmym, kmyp, kmzm, kmzp |
---|
[1001] | 84 | |
---|
[75] | 85 | REAL, DIMENSION(nzb:nzt+1,nys:nyn,nxl:nxr) :: usvs |
---|
[1] | 86 | |
---|
[56] | 87 | ! |
---|
| 88 | !-- First calculate horizontal momentum flux u'v' at vertical walls, |
---|
| 89 | !-- if neccessary |
---|
| 90 | IF ( topography /= 'flat' ) THEN |
---|
[75] | 91 | CALL wall_fluxes( usvs, 1.0, 0.0, 0.0, 0.0, nzb_u_inner, & |
---|
[56] | 92 | nzb_u_outer, wall_u ) |
---|
| 93 | ENDIF |
---|
| 94 | |
---|
[106] | 95 | DO i = nxlu, nxr |
---|
[1001] | 96 | DO j = nys, nyn |
---|
[1] | 97 | ! |
---|
| 98 | !-- Compute horizontal diffusion |
---|
| 99 | DO k = nzb_u_outer(j,i)+1, nzt |
---|
| 100 | ! |
---|
| 101 | !-- Interpolate eddy diffusivities on staggered gridpoints |
---|
[978] | 102 | kmyp = 0.25 * & |
---|
| 103 | ( km(k,j,i)+km(k,j+1,i)+km(k,j,i-1)+km(k,j+1,i-1) ) |
---|
| 104 | kmym = 0.25 * & |
---|
| 105 | ( km(k,j,i)+km(k,j-1,i)+km(k,j,i-1)+km(k,j-1,i-1) ) |
---|
[1] | 106 | |
---|
| 107 | tend(k,j,i) = tend(k,j,i) & |
---|
| 108 | & + 2.0 * ( & |
---|
| 109 | & km(k,j,i) * ( u(k,j,i+1) - u(k,j,i) ) & |
---|
| 110 | & - km(k,j,i-1) * ( u(k,j,i) - u(k,j,i-1) ) & |
---|
| 111 | & ) * ddx2 & |
---|
[978] | 112 | & + ( kmyp * ( u(k,j+1,i) - u(k,j,i) ) * ddy & |
---|
| 113 | & + kmyp * ( v(k,j+1,i) - v(k,j+1,i-1) ) * ddx & |
---|
| 114 | & - kmym * ( u(k,j,i) - u(k,j-1,i) ) * ddy & |
---|
| 115 | & - kmym * ( v(k,j,i) - v(k,j,i-1) ) * ddx & |
---|
[1] | 116 | & ) * ddy |
---|
| 117 | ENDDO |
---|
| 118 | |
---|
| 119 | ! |
---|
| 120 | !-- Wall functions at the north and south walls, respectively |
---|
| 121 | IF ( wall_u(j,i) /= 0.0 ) THEN |
---|
[51] | 122 | |
---|
[1] | 123 | DO k = nzb_u_inner(j,i)+1, nzb_u_outer(j,i) |
---|
[978] | 124 | kmyp = 0.25 * & |
---|
| 125 | ( km(k,j,i)+km(k,j+1,i)+km(k,j,i-1)+km(k,j+1,i-1) ) |
---|
| 126 | kmym = 0.25 * & |
---|
| 127 | ( km(k,j,i)+km(k,j-1,i)+km(k,j,i-1)+km(k,j-1,i-1) ) |
---|
[1] | 128 | |
---|
| 129 | tend(k,j,i) = tend(k,j,i) & |
---|
| 130 | + 2.0 * ( & |
---|
| 131 | km(k,j,i) * ( u(k,j,i+1) - u(k,j,i) ) & |
---|
| 132 | - km(k,j,i-1) * ( u(k,j,i) - u(k,j,i-1) ) & |
---|
| 133 | ) * ddx2 & |
---|
| 134 | + ( fyp(j,i) * ( & |
---|
[978] | 135 | kmyp * ( u(k,j+1,i) - u(k,j,i) ) * ddy & |
---|
| 136 | + kmyp * ( v(k,j+1,i) - v(k,j+1,i-1) ) * ddx & |
---|
[1] | 137 | ) & |
---|
| 138 | - fym(j,i) * ( & |
---|
[978] | 139 | kmym * ( u(k,j,i) - u(k,j-1,i) ) * ddy & |
---|
| 140 | + kmym * ( v(k,j,i) - v(k,j,i-1) ) * ddx & |
---|
[1] | 141 | ) & |
---|
[56] | 142 | + wall_u(j,i) * usvs(k,j,i) & |
---|
[1] | 143 | ) * ddy |
---|
| 144 | ENDDO |
---|
| 145 | ENDIF |
---|
| 146 | |
---|
| 147 | ! |
---|
| 148 | !-- Compute vertical diffusion. In case of simulating a Prandtl layer, |
---|
| 149 | !-- index k starts at nzb_u_inner+2. |
---|
[102] | 150 | DO k = nzb_diff_u(j,i), nzt_diff |
---|
[1] | 151 | ! |
---|
| 152 | !-- Interpolate eddy diffusivities on staggered gridpoints |
---|
| 153 | kmzp = 0.25 * & |
---|
| 154 | ( km(k,j,i)+km(k+1,j,i)+km(k,j,i-1)+km(k+1,j,i-1) ) |
---|
| 155 | kmzm = 0.25 * & |
---|
| 156 | ( km(k,j,i)+km(k-1,j,i)+km(k,j,i-1)+km(k-1,j,i-1) ) |
---|
| 157 | |
---|
| 158 | tend(k,j,i) = tend(k,j,i) & |
---|
| 159 | & + ( kmzp * ( ( u(k+1,j,i) - u(k,j,i) ) * ddzu(k+1) & |
---|
| 160 | & + ( w(k,j,i) - w(k,j,i-1) ) * ddx & |
---|
| 161 | & ) & |
---|
| 162 | & - kmzm * ( ( u(k,j,i) - u(k-1,j,i) ) * ddzu(k) & |
---|
| 163 | & + ( w(k-1,j,i) - w(k-1,j,i-1) ) * ddx & |
---|
[667] | 164 | & ) & |
---|
[1] | 165 | & ) * ddzw(k) |
---|
| 166 | ENDDO |
---|
| 167 | |
---|
| 168 | ! |
---|
| 169 | !-- Vertical diffusion at the first grid point above the surface, |
---|
| 170 | !-- if the momentum flux at the bottom is given by the Prandtl law or |
---|
| 171 | !-- if it is prescribed by the user. |
---|
| 172 | !-- Difference quotient of the momentum flux is not formed over half |
---|
| 173 | !-- of the grid spacing (2.0*ddzw(k)) any more, since the comparison |
---|
| 174 | !-- with other (LES) modell showed that the values of the momentum |
---|
| 175 | !-- flux becomes too large in this case. |
---|
| 176 | !-- The term containing w(k-1,..) (see above equation) is removed here |
---|
| 177 | !-- because the vertical velocity is assumed to be zero at the surface. |
---|
| 178 | IF ( use_surface_fluxes ) THEN |
---|
| 179 | k = nzb_u_inner(j,i)+1 |
---|
| 180 | ! |
---|
| 181 | !-- Interpolate eddy diffusivities on staggered gridpoints |
---|
| 182 | kmzp = 0.25 * & |
---|
| 183 | ( km(k,j,i)+km(k+1,j,i)+km(k,j,i-1)+km(k+1,j,i-1) ) |
---|
| 184 | kmzm = 0.25 * & |
---|
| 185 | ( km(k,j,i)+km(k-1,j,i)+km(k,j,i-1)+km(k-1,j,i-1) ) |
---|
| 186 | |
---|
| 187 | tend(k,j,i) = tend(k,j,i) & |
---|
| 188 | & + ( kmzp * ( w(k,j,i) - w(k,j,i-1) ) * ddx & |
---|
| 189 | & ) * ddzw(k) & |
---|
[102] | 190 | & + ( kmzp * ( u(k+1,j,i) - u(k,j,i) ) * ddzu(k+1) & |
---|
[1] | 191 | & + usws(j,i) & |
---|
| 192 | & ) * ddzw(k) |
---|
| 193 | ENDIF |
---|
| 194 | |
---|
[102] | 195 | ! |
---|
| 196 | !-- Vertical diffusion at the first gridpoint below the top boundary, |
---|
| 197 | !-- if the momentum flux at the top is prescribed by the user |
---|
[103] | 198 | IF ( use_top_fluxes .AND. constant_top_momentumflux ) THEN |
---|
[102] | 199 | k = nzt |
---|
| 200 | ! |
---|
| 201 | !-- Interpolate eddy diffusivities on staggered gridpoints |
---|
| 202 | kmzp = 0.25 * & |
---|
| 203 | ( km(k,j,i)+km(k+1,j,i)+km(k,j,i-1)+km(k+1,j,i-1) ) |
---|
| 204 | kmzm = 0.25 * & |
---|
| 205 | ( km(k,j,i)+km(k-1,j,i)+km(k,j,i-1)+km(k-1,j,i-1) ) |
---|
| 206 | |
---|
| 207 | tend(k,j,i) = tend(k,j,i) & |
---|
| 208 | & - ( kmzm * ( w(k-1,j,i) - w(k-1,j,i-1) ) * ddx & |
---|
| 209 | & ) * ddzw(k) & |
---|
| 210 | & + ( -uswst(j,i) & |
---|
| 211 | & - kmzm * ( u(k,j,i) - u(k-1,j,i) ) * ddzu(k) & |
---|
| 212 | & ) * ddzw(k) |
---|
| 213 | ENDIF |
---|
| 214 | |
---|
[1] | 215 | ENDDO |
---|
| 216 | ENDDO |
---|
| 217 | |
---|
| 218 | END SUBROUTINE diffusion_u |
---|
| 219 | |
---|
| 220 | |
---|
| 221 | !------------------------------------------------------------------------------! |
---|
| 222 | ! Call for grid point i,j |
---|
| 223 | !------------------------------------------------------------------------------! |
---|
[1001] | 224 | SUBROUTINE diffusion_u_ij( i, j ) |
---|
[1] | 225 | |
---|
[1001] | 226 | USE arrays_3d |
---|
[1] | 227 | USE control_parameters |
---|
| 228 | USE grid_variables |
---|
| 229 | USE indices |
---|
| 230 | |
---|
| 231 | IMPLICIT NONE |
---|
| 232 | |
---|
| 233 | INTEGER :: i, j, k |
---|
[978] | 234 | REAL :: kmym, kmyp, kmzm, kmzp |
---|
[1] | 235 | |
---|
[1001] | 236 | REAL, DIMENSION(nzb:nzt+1) :: usvs |
---|
| 237 | |
---|
[1] | 238 | ! |
---|
| 239 | !-- Compute horizontal diffusion |
---|
| 240 | DO k = nzb_u_outer(j,i)+1, nzt |
---|
| 241 | ! |
---|
| 242 | !-- Interpolate eddy diffusivities on staggered gridpoints |
---|
[978] | 243 | kmyp = 0.25 * ( km(k,j,i)+km(k,j+1,i)+km(k,j,i-1)+km(k,j+1,i-1) ) |
---|
| 244 | kmym = 0.25 * ( km(k,j,i)+km(k,j-1,i)+km(k,j,i-1)+km(k,j-1,i-1) ) |
---|
[1] | 245 | |
---|
| 246 | tend(k,j,i) = tend(k,j,i) & |
---|
| 247 | & + 2.0 * ( & |
---|
| 248 | & km(k,j,i) * ( u(k,j,i+1) - u(k,j,i) ) & |
---|
| 249 | & - km(k,j,i-1) * ( u(k,j,i) - u(k,j,i-1) ) & |
---|
| 250 | & ) * ddx2 & |
---|
[978] | 251 | & + ( kmyp * ( u(k,j+1,i) - u(k,j,i) ) * ddy & |
---|
| 252 | & + kmyp * ( v(k,j+1,i) - v(k,j+1,i-1) ) * ddx & |
---|
| 253 | & - kmym * ( u(k,j,i) - u(k,j-1,i) ) * ddy & |
---|
| 254 | & - kmym * ( v(k,j,i) - v(k,j,i-1) ) * ddx & |
---|
[1] | 255 | & ) * ddy |
---|
| 256 | ENDDO |
---|
| 257 | |
---|
| 258 | ! |
---|
| 259 | !-- Wall functions at the north and south walls, respectively |
---|
| 260 | IF ( wall_u(j,i) .NE. 0.0 ) THEN |
---|
[51] | 261 | |
---|
| 262 | ! |
---|
| 263 | !-- Calculate the horizontal momentum flux u'v' |
---|
| 264 | CALL wall_fluxes( i, j, nzb_u_inner(j,i)+1, nzb_u_outer(j,i), & |
---|
| 265 | usvs, 1.0, 0.0, 0.0, 0.0 ) |
---|
| 266 | |
---|
[1] | 267 | DO k = nzb_u_inner(j,i)+1, nzb_u_outer(j,i) |
---|
[978] | 268 | kmyp = 0.25 * ( km(k,j,i)+km(k,j+1,i)+km(k,j,i-1)+km(k,j+1,i-1) ) |
---|
| 269 | kmym = 0.25 * ( km(k,j,i)+km(k,j-1,i)+km(k,j,i-1)+km(k,j-1,i-1) ) |
---|
[1] | 270 | |
---|
| 271 | tend(k,j,i) = tend(k,j,i) & |
---|
| 272 | + 2.0 * ( & |
---|
| 273 | km(k,j,i) * ( u(k,j,i+1) - u(k,j,i) ) & |
---|
| 274 | - km(k,j,i-1) * ( u(k,j,i) - u(k,j,i-1) ) & |
---|
| 275 | ) * ddx2 & |
---|
| 276 | + ( fyp(j,i) * ( & |
---|
[978] | 277 | kmyp * ( u(k,j+1,i) - u(k,j,i) ) * ddy & |
---|
| 278 | + kmyp * ( v(k,j+1,i) - v(k,j+1,i-1) ) * ddx & |
---|
[1] | 279 | ) & |
---|
| 280 | - fym(j,i) * ( & |
---|
[978] | 281 | kmym * ( u(k,j,i) - u(k,j-1,i) ) * ddy & |
---|
| 282 | + kmym * ( v(k,j,i) - v(k,j,i-1) ) * ddx & |
---|
[1] | 283 | ) & |
---|
[51] | 284 | + wall_u(j,i) * usvs(k) & |
---|
[1] | 285 | ) * ddy |
---|
| 286 | ENDDO |
---|
| 287 | ENDIF |
---|
| 288 | |
---|
| 289 | ! |
---|
| 290 | !-- Compute vertical diffusion. In case of simulating a Prandtl layer, |
---|
| 291 | !-- index k starts at nzb_u_inner+2. |
---|
[102] | 292 | DO k = nzb_diff_u(j,i), nzt_diff |
---|
[1] | 293 | ! |
---|
| 294 | !-- Interpolate eddy diffusivities on staggered gridpoints |
---|
| 295 | kmzp = 0.25 * ( km(k,j,i)+km(k+1,j,i)+km(k,j,i-1)+km(k+1,j,i-1) ) |
---|
| 296 | kmzm = 0.25 * ( km(k,j,i)+km(k-1,j,i)+km(k,j,i-1)+km(k-1,j,i-1) ) |
---|
| 297 | |
---|
| 298 | tend(k,j,i) = tend(k,j,i) & |
---|
| 299 | & + ( kmzp * ( ( u(k+1,j,i) - u(k,j,i) ) * ddzu(k+1) & |
---|
| 300 | & + ( w(k,j,i) - w(k,j,i-1) ) * ddx & |
---|
| 301 | & ) & |
---|
| 302 | & - kmzm * ( ( u(k,j,i) - u(k-1,j,i) ) * ddzu(k) & |
---|
| 303 | & + ( w(k-1,j,i) - w(k-1,j,i-1) ) * ddx & |
---|
| 304 | & ) & |
---|
| 305 | & ) * ddzw(k) |
---|
| 306 | ENDDO |
---|
| 307 | |
---|
| 308 | ! |
---|
| 309 | !-- Vertical diffusion at the first grid point above the surface, if the |
---|
| 310 | !-- momentum flux at the bottom is given by the Prandtl law or if it is |
---|
| 311 | !-- prescribed by the user. |
---|
| 312 | !-- Difference quotient of the momentum flux is not formed over half of |
---|
| 313 | !-- the grid spacing (2.0*ddzw(k)) any more, since the comparison with |
---|
| 314 | !-- other (LES) modell showed that the values of the momentum flux becomes |
---|
| 315 | !-- too large in this case. |
---|
| 316 | !-- The term containing w(k-1,..) (see above equation) is removed here |
---|
| 317 | !-- because the vertical velocity is assumed to be zero at the surface. |
---|
| 318 | IF ( use_surface_fluxes ) THEN |
---|
| 319 | k = nzb_u_inner(j,i)+1 |
---|
| 320 | ! |
---|
| 321 | !-- Interpolate eddy diffusivities on staggered gridpoints |
---|
| 322 | kmzp = 0.25 * ( km(k,j,i)+km(k+1,j,i)+km(k,j,i-1)+km(k+1,j,i-1) ) |
---|
| 323 | kmzm = 0.25 * ( km(k,j,i)+km(k-1,j,i)+km(k,j,i-1)+km(k-1,j,i-1) ) |
---|
| 324 | |
---|
| 325 | tend(k,j,i) = tend(k,j,i) & |
---|
| 326 | & + ( kmzp * ( w(k,j,i) - w(k,j,i-1) ) * ddx & |
---|
| 327 | & ) * ddzw(k) & |
---|
[102] | 328 | & + ( kmzp * ( u(k+1,j,i) - u(k,j,i) ) * ddzu(k+1) & |
---|
[1] | 329 | & + usws(j,i) & |
---|
| 330 | & ) * ddzw(k) |
---|
| 331 | ENDIF |
---|
| 332 | |
---|
[102] | 333 | ! |
---|
| 334 | !-- Vertical diffusion at the first gridpoint below the top boundary, |
---|
| 335 | !-- if the momentum flux at the top is prescribed by the user |
---|
[103] | 336 | IF ( use_top_fluxes .AND. constant_top_momentumflux ) THEN |
---|
[102] | 337 | k = nzt |
---|
| 338 | ! |
---|
| 339 | !-- Interpolate eddy diffusivities on staggered gridpoints |
---|
| 340 | kmzp = 0.25 * & |
---|
| 341 | ( km(k,j,i)+km(k+1,j,i)+km(k,j,i-1)+km(k+1,j,i-1) ) |
---|
| 342 | kmzm = 0.25 * & |
---|
| 343 | ( km(k,j,i)+km(k-1,j,i)+km(k,j,i-1)+km(k-1,j,i-1) ) |
---|
| 344 | |
---|
| 345 | tend(k,j,i) = tend(k,j,i) & |
---|
| 346 | & - ( kmzm * ( w(k-1,j,i) - w(k-1,j,i-1) ) * ddx & |
---|
| 347 | & ) * ddzw(k) & |
---|
| 348 | & + ( -uswst(j,i) & |
---|
| 349 | & - kmzm * ( u(k,j,i) - u(k-1,j,i) ) * ddzu(k) & |
---|
| 350 | & ) * ddzw(k) |
---|
| 351 | ENDIF |
---|
| 352 | |
---|
[1] | 353 | END SUBROUTINE diffusion_u_ij |
---|
| 354 | |
---|
| 355 | END MODULE diffusion_u_mod |
---|