| 1 | SUBROUTINE boundary_conds( range ) |
|---|
| 2 | |
|---|
| 3 | !------------------------------------------------------------------------------! |
|---|
| 4 | ! Actual revisions: |
|---|
| 5 | ! ----------------- |
|---|
| 6 | ! Boundary conditions for temperature adjusted for coupled runs, |
|---|
| 7 | ! bugfixes for the radiation boundary conditions at the outflow: radiation |
|---|
| 8 | ! conditions are used for every substep, phase speeds are calculated for the |
|---|
| 9 | ! first Runge-Kutta substep only and then reused, several index values changed |
|---|
| 10 | ! |
|---|
| 11 | ! |
|---|
| 12 | ! Former revisions: |
|---|
| 13 | ! ----------------- |
|---|
| 14 | ! $Id: boundary_conds.f90 107 2007-08-17 13:54:45Z raasch $ |
|---|
| 15 | ! |
|---|
| 16 | ! 95 2007-06-02 16:48:38Z raasch |
|---|
| 17 | ! Boundary conditions for salinity added |
|---|
| 18 | ! |
|---|
| 19 | ! 75 2007-03-22 09:54:05Z raasch |
|---|
| 20 | ! The "main" part sets conditions for time level t+dt instead of level t, |
|---|
| 21 | ! outflow boundary conditions changed from Neumann to radiation condition, |
|---|
| 22 | ! uxrp, vynp eliminated, moisture renamed humidity |
|---|
| 23 | ! |
|---|
| 24 | ! 19 2007-02-23 04:53:48Z raasch |
|---|
| 25 | ! Boundary conditions for e(nzt), pt(nzt), and q(nzt) removed because these |
|---|
| 26 | ! gridpoints are now calculated by the prognostic equation, |
|---|
| 27 | ! Dirichlet and zero gradient condition for pt established at top boundary |
|---|
| 28 | ! |
|---|
| 29 | ! RCS Log replace by Id keyword, revision history cleaned up |
|---|
| 30 | ! |
|---|
| 31 | ! Revision 1.15 2006/02/23 09:54:55 raasch |
|---|
| 32 | ! Surface boundary conditions in case of topography: nzb replaced by |
|---|
| 33 | ! 2d-k-index-arrays (nzb_w_inner, etc.). Conditions for u and v remain |
|---|
| 34 | ! unchanged (still using nzb) because a non-flat topography must use a |
|---|
| 35 | ! Prandtl-layer, which don't requires explicit setting of the surface values. |
|---|
| 36 | ! |
|---|
| 37 | ! Revision 1.1 1997/09/12 06:21:34 raasch |
|---|
| 38 | ! Initial revision |
|---|
| 39 | ! |
|---|
| 40 | ! |
|---|
| 41 | ! Description: |
|---|
| 42 | ! ------------ |
|---|
| 43 | ! Boundary conditions for the prognostic quantities (range='main'). |
|---|
| 44 | ! In case of non-cyclic lateral boundaries the conditions for velocities at |
|---|
| 45 | ! the outflow are set after the pressure solver has been called (range= |
|---|
| 46 | ! 'outflow_uvw'). |
|---|
| 47 | ! One additional bottom boundary condition is applied for the TKE (=(u*)**2) |
|---|
| 48 | ! in prandtl_fluxes. The cyclic lateral boundary conditions are implicitly |
|---|
| 49 | ! handled in routine exchange_horiz. Pressure boundary conditions are |
|---|
| 50 | ! explicitly set in routines pres, poisfft, poismg and sor. |
|---|
| 51 | !------------------------------------------------------------------------------! |
|---|
| 52 | |
|---|
| 53 | USE arrays_3d |
|---|
| 54 | USE control_parameters |
|---|
| 55 | USE grid_variables |
|---|
| 56 | USE indices |
|---|
| 57 | USE pegrid |
|---|
| 58 | |
|---|
| 59 | IMPLICIT NONE |
|---|
| 60 | |
|---|
| 61 | CHARACTER (LEN=*) :: range |
|---|
| 62 | |
|---|
| 63 | INTEGER :: i, j, k |
|---|
| 64 | |
|---|
| 65 | REAL :: c_max, denom |
|---|
| 66 | |
|---|
| 67 | |
|---|
| 68 | IF ( range == 'main') THEN |
|---|
| 69 | ! |
|---|
| 70 | !-- Bottom boundary |
|---|
| 71 | IF ( ibc_uv_b == 0 ) THEN |
|---|
| 72 | ! |
|---|
| 73 | !-- Satisfying the Dirichlet condition with an extra layer below the |
|---|
| 74 | !-- surface where the u and v component change their sign |
|---|
| 75 | u_p(nzb,:,:) = -u_p(nzb+1,:,:) |
|---|
| 76 | v_p(nzb,:,:) = -v_p(nzb+1,:,:) |
|---|
| 77 | ELSE |
|---|
| 78 | u_p(nzb,:,:) = u_p(nzb+1,:,:) |
|---|
| 79 | v_p(nzb,:,:) = v_p(nzb+1,:,:) |
|---|
| 80 | ENDIF |
|---|
| 81 | DO i = nxl-1, nxr+1 |
|---|
| 82 | DO j = nys-1, nyn+1 |
|---|
| 83 | w_p(nzb_w_inner(j,i),j,i) = 0.0 |
|---|
| 84 | ENDDO |
|---|
| 85 | ENDDO |
|---|
| 86 | |
|---|
| 87 | ! |
|---|
| 88 | !-- Top boundary |
|---|
| 89 | IF ( ibc_uv_t == 0 ) THEN |
|---|
| 90 | u_p(nzt+1,:,:) = ug(nzt+1) |
|---|
| 91 | v_p(nzt+1,:,:) = vg(nzt+1) |
|---|
| 92 | ELSE |
|---|
| 93 | u_p(nzt+1,:,:) = u_p(nzt,:,:) |
|---|
| 94 | v_p(nzt+1,:,:) = v_p(nzt,:,:) |
|---|
| 95 | ENDIF |
|---|
| 96 | w_p(nzt:nzt+1,:,:) = 0.0 ! nzt is not a prognostic level (but cf. pres) |
|---|
| 97 | |
|---|
| 98 | ! |
|---|
| 99 | !-- Temperature at bottom boundary. |
|---|
| 100 | !-- In case of coupled runs (ibc_pt_b = 2) the temperature is given by |
|---|
| 101 | !-- the sea surface temperature of the coupled ocean model. |
|---|
| 102 | IF ( ibc_pt_b == 0 ) THEN |
|---|
| 103 | DO i = nxl-1, nxr+1 |
|---|
| 104 | DO j = nys-1, nyn+1 |
|---|
| 105 | pt_p(nzb_s_inner(j,i),j,i) = pt(nzb_s_inner(j,i),j,i) |
|---|
| 106 | ENDDO |
|---|
| 107 | ENDDO |
|---|
| 108 | ELSEIF ( ibc_pt_b == 1 ) THEN |
|---|
| 109 | DO i = nxl-1, nxr+1 |
|---|
| 110 | DO j = nys-1, nyn+1 |
|---|
| 111 | pt_p(nzb_s_inner(j,i),j,i) = pt_p(nzb_s_inner(j,i)+1,j,i) |
|---|
| 112 | ENDDO |
|---|
| 113 | ENDDO |
|---|
| 114 | ENDIF |
|---|
| 115 | |
|---|
| 116 | ! |
|---|
| 117 | !-- Temperature at top boundary |
|---|
| 118 | IF ( ibc_pt_t == 0 ) THEN |
|---|
| 119 | pt_p(nzt+1,:,:) = pt(nzt+1,:,:) |
|---|
| 120 | ELSEIF ( ibc_pt_t == 1 ) THEN |
|---|
| 121 | pt_p(nzt+1,:,:) = pt_p(nzt,:,:) |
|---|
| 122 | ELSEIF ( ibc_pt_t == 2 ) THEN |
|---|
| 123 | pt_p(nzt+1,:,:) = pt_p(nzt,:,:) + bc_pt_t_val * dzu(nzt+1) |
|---|
| 124 | ENDIF |
|---|
| 125 | |
|---|
| 126 | ! |
|---|
| 127 | !-- Boundary conditions for TKE |
|---|
| 128 | !-- Generally Neumann conditions with de/dz=0 are assumed |
|---|
| 129 | IF ( .NOT. constant_diffusion ) THEN |
|---|
| 130 | DO i = nxl-1, nxr+1 |
|---|
| 131 | DO j = nys-1, nyn+1 |
|---|
| 132 | e_p(nzb_s_inner(j,i),j,i) = e_p(nzb_s_inner(j,i)+1,j,i) |
|---|
| 133 | ENDDO |
|---|
| 134 | ENDDO |
|---|
| 135 | e_p(nzt+1,:,:) = e_p(nzt,:,:) |
|---|
| 136 | ENDIF |
|---|
| 137 | |
|---|
| 138 | ! |
|---|
| 139 | !-- Boundary conditions for salinity |
|---|
| 140 | IF ( ocean ) THEN |
|---|
| 141 | ! |
|---|
| 142 | !-- Bottom boundary: Neumann condition because salinity flux is always |
|---|
| 143 | !-- given |
|---|
| 144 | DO i = nxl-1, nxr+1 |
|---|
| 145 | DO j = nys-1, nyn+1 |
|---|
| 146 | sa_p(nzb_s_inner(j,i),j,i) = sa_p(nzb_s_inner(j,i)+1,j,i) |
|---|
| 147 | ENDDO |
|---|
| 148 | ENDDO |
|---|
| 149 | |
|---|
| 150 | ! |
|---|
| 151 | !-- Top boundary: Dirichlet or Neumann |
|---|
| 152 | IF ( ibc_sa_t == 0 ) THEN |
|---|
| 153 | sa_p(nzt+1,:,:) = sa(nzt+1,:,:) |
|---|
| 154 | ELSEIF ( ibc_sa_t == 1 ) THEN |
|---|
| 155 | sa_p(nzt+1,:,:) = sa_p(nzt,:,:) |
|---|
| 156 | ENDIF |
|---|
| 157 | |
|---|
| 158 | ENDIF |
|---|
| 159 | |
|---|
| 160 | ! |
|---|
| 161 | !-- Boundary conditions for total water content or scalar, |
|---|
| 162 | !-- bottom and top boundary (see also temperature) |
|---|
| 163 | IF ( humidity .OR. passive_scalar ) THEN |
|---|
| 164 | ! |
|---|
| 165 | !-- Surface conditions for constant_humidity_flux |
|---|
| 166 | IF ( ibc_q_b == 0 ) THEN |
|---|
| 167 | DO i = nxl-1, nxr+1 |
|---|
| 168 | DO j = nys-1, nyn+1 |
|---|
| 169 | q_p(nzb_s_inner(j,i),j,i) = q(nzb_s_inner(j,i),j,i) |
|---|
| 170 | ENDDO |
|---|
| 171 | ENDDO |
|---|
| 172 | ELSE |
|---|
| 173 | DO i = nxl-1, nxr+1 |
|---|
| 174 | DO j = nys-1, nyn+1 |
|---|
| 175 | q_p(nzb_s_inner(j,i),j,i) = q_p(nzb_s_inner(j,i)+1,j,i) |
|---|
| 176 | ENDDO |
|---|
| 177 | ENDDO |
|---|
| 178 | ENDIF |
|---|
| 179 | ! |
|---|
| 180 | !-- Top boundary |
|---|
| 181 | q_p(nzt+1,:,:) = q_p(nzt,:,:) + bc_q_t_val * dzu(nzt+1) |
|---|
| 182 | ENDIF |
|---|
| 183 | |
|---|
| 184 | ! |
|---|
| 185 | !-- Lateral boundary conditions at the inflow. Quasi Neumann conditions |
|---|
| 186 | !-- are needed for the wall normal velocity in order to ensure zero |
|---|
| 187 | !-- divergence. Dirichlet conditions are used for all other quantities. |
|---|
| 188 | IF ( inflow_s ) THEN |
|---|
| 189 | v_p(:,nys,:) = v_p(:,nys-1,:) |
|---|
| 190 | ELSEIF ( inflow_n ) THEN |
|---|
| 191 | v_p(:,nyn,:) = v_p(:,nyn+1,:) |
|---|
| 192 | ELSEIF ( inflow_l ) THEN |
|---|
| 193 | u_p(:,:,nxl) = u_p(:,:,nxl-1) |
|---|
| 194 | ELSEIF ( inflow_r ) THEN |
|---|
| 195 | u_p(:,:,nxr) = u_p(:,:,nxr+1) |
|---|
| 196 | ENDIF |
|---|
| 197 | |
|---|
| 198 | ! |
|---|
| 199 | !-- Lateral boundary conditions for scalar quantities at the outflow |
|---|
| 200 | IF ( outflow_s ) THEN |
|---|
| 201 | pt_p(:,nys-1,:) = pt_p(:,nys,:) |
|---|
| 202 | IF ( .NOT. constant_diffusion ) e_p(:,nys-1,:) = e_p(:,nys,:) |
|---|
| 203 | IF ( humidity .OR. passive_scalar ) q_p(:,nys-1,:) = q_p(:,nys,:) |
|---|
| 204 | ELSEIF ( outflow_n ) THEN |
|---|
| 205 | pt_p(:,nyn+1,:) = pt_p(:,nyn,:) |
|---|
| 206 | IF ( .NOT. constant_diffusion ) e_p(:,nyn+1,:) = e_p(:,nyn,:) |
|---|
| 207 | IF ( humidity .OR. passive_scalar ) q_p(:,nyn+1,:) = q_p(:,nyn,:) |
|---|
| 208 | ELSEIF ( outflow_l ) THEN |
|---|
| 209 | pt_p(:,:,nxl-1) = pt_p(:,:,nxl) |
|---|
| 210 | IF ( .NOT. constant_diffusion ) e_p(:,:,nxl-1) = e_p(:,:,nxl) |
|---|
| 211 | IF ( humidity .OR. passive_scalar ) q_p(:,:,nxl-1) = q_p(:,:,nxl) |
|---|
| 212 | ELSEIF ( outflow_r ) THEN |
|---|
| 213 | pt_p(:,:,nxr+1) = pt_p(:,:,nxr) |
|---|
| 214 | IF ( .NOT. constant_diffusion ) e_p(:,:,nxr+1) = e_p(:,:,nxr) |
|---|
| 215 | IF ( humidity .OR. passive_scalar ) q_p(:,:,nxr+1) = q_p(:,:,nxr) |
|---|
| 216 | ENDIF |
|---|
| 217 | |
|---|
| 218 | ENDIF |
|---|
| 219 | |
|---|
| 220 | ! |
|---|
| 221 | !-- Radiation boundary condition for the velocities at the respective outflow |
|---|
| 222 | IF ( outflow_s ) THEN |
|---|
| 223 | |
|---|
| 224 | c_max = dy / dt_3d |
|---|
| 225 | |
|---|
| 226 | DO i = nxl-1, nxr+1 |
|---|
| 227 | DO k = nzb+1, nzt+1 |
|---|
| 228 | |
|---|
| 229 | ! |
|---|
| 230 | !-- Calculate the phase speeds for u,v, and w. In case of using a |
|---|
| 231 | !-- Runge-Kutta scheme, do this for the first substep only and then |
|---|
| 232 | !-- reuse this values for the further substeps. |
|---|
| 233 | IF ( intermediate_timestep_count == 1 ) THEN |
|---|
| 234 | |
|---|
| 235 | denom = u_m_s(k,0,i) - u_m_s(k,1,i) |
|---|
| 236 | |
|---|
| 237 | IF ( denom /= 0.0 ) THEN |
|---|
| 238 | c_u(k,i) = -c_max * ( u(k,0,i) - u_m_s(k,0,i) ) / denom |
|---|
| 239 | IF ( c_u(k,i) < 0.0 ) THEN |
|---|
| 240 | c_u(k,i) = 0.0 |
|---|
| 241 | ELSEIF ( c_u(k,i) > c_max ) THEN |
|---|
| 242 | c_u(k,i) = c_max |
|---|
| 243 | ENDIF |
|---|
| 244 | ELSE |
|---|
| 245 | c_u(k,i) = c_max |
|---|
| 246 | ENDIF |
|---|
| 247 | |
|---|
| 248 | denom = v_m_s(k,1,i) - v_m_s(k,2,i) |
|---|
| 249 | |
|---|
| 250 | IF ( denom /= 0.0 ) THEN |
|---|
| 251 | c_v(k,i) = -c_max * ( v(k,1,i) - v_m_s(k,1,i) ) / denom |
|---|
| 252 | IF ( c_v(k,i) < 0.0 ) THEN |
|---|
| 253 | c_v(k,i) = 0.0 |
|---|
| 254 | ELSEIF ( c_v(k,i) > c_max ) THEN |
|---|
| 255 | c_v(k,i) = c_max |
|---|
| 256 | ENDIF |
|---|
| 257 | ELSE |
|---|
| 258 | c_v(k,i) = c_max |
|---|
| 259 | ENDIF |
|---|
| 260 | |
|---|
| 261 | denom = w_m_s(k,0,i) - w_m_s(k,1,i) |
|---|
| 262 | |
|---|
| 263 | IF ( denom /= 0.0 ) THEN |
|---|
| 264 | c_w(k,i) = -c_max * ( w(k,0,i) - w_m_s(k,0,i) ) / denom |
|---|
| 265 | IF ( c_w(k,i) < 0.0 ) THEN |
|---|
| 266 | c_w(k,i) = 0.0 |
|---|
| 267 | ELSEIF ( c_w(k,i) > c_max ) THEN |
|---|
| 268 | c_w(k,i) = c_max |
|---|
| 269 | ENDIF |
|---|
| 270 | ELSE |
|---|
| 271 | c_w(k,i) = c_max |
|---|
| 272 | ENDIF |
|---|
| 273 | |
|---|
| 274 | ! |
|---|
| 275 | !-- Save old timelevels for the next timestep |
|---|
| 276 | u_m_s(k,:,i) = u(k,0:1,i) |
|---|
| 277 | v_m_s(k,:,i) = v(k,1:2,i) |
|---|
| 278 | w_m_s(k,:,i) = w(k,0:1,i) |
|---|
| 279 | |
|---|
| 280 | ENDIF |
|---|
| 281 | |
|---|
| 282 | ! |
|---|
| 283 | !-- Calculate the new velocities |
|---|
| 284 | u_p(k,-1,i) = u(k,-1,i) - dt_3d * tsc(2) * c_u(k,i) * & |
|---|
| 285 | ( u(k,-1,i) - u(k,0,i) ) * ddy |
|---|
| 286 | |
|---|
| 287 | v_p(k,0,i) = v(k,0,i) - dt_3d * tsc(2) * c_v(k,i) * & |
|---|
| 288 | ( v(k,0,i) - v(k,1,i) ) * ddy |
|---|
| 289 | |
|---|
| 290 | w_p(k,-1,i) = w(k,-1,i) - dt_3d * tsc(2) * c_w(k,i) * & |
|---|
| 291 | ( w(k,-1,i) - w(k,0,i) ) * ddy |
|---|
| 292 | |
|---|
| 293 | ENDDO |
|---|
| 294 | ENDDO |
|---|
| 295 | |
|---|
| 296 | ! |
|---|
| 297 | !-- Bottom boundary at the outflow |
|---|
| 298 | IF ( ibc_uv_b == 0 ) THEN |
|---|
| 299 | u_p(nzb,-1,:) = -u_p(nzb+1,-1,:) |
|---|
| 300 | v_p(nzb,0,:) = -v_p(nzb+1,0,:) |
|---|
| 301 | ELSE |
|---|
| 302 | u_p(nzb,-1,:) = u_p(nzb+1,-1,:) |
|---|
| 303 | v_p(nzb,0,:) = v_p(nzb+1,0,:) |
|---|
| 304 | ENDIF |
|---|
| 305 | w_p(nzb,-1,:) = 0.0 |
|---|
| 306 | |
|---|
| 307 | ! |
|---|
| 308 | !-- Top boundary at the outflow |
|---|
| 309 | IF ( ibc_uv_t == 0 ) THEN |
|---|
| 310 | u_p(nzt+1,-1,:) = ug(nzt+1) |
|---|
| 311 | v_p(nzt+1,0,:) = vg(nzt+1) |
|---|
| 312 | ELSE |
|---|
| 313 | u_p(nzt+1,-1,:) = u(nzt,-1,:) |
|---|
| 314 | v_p(nzt+1,0,:) = v(nzt,0,:) |
|---|
| 315 | ENDIF |
|---|
| 316 | w_p(nzt:nzt+1,-1,:) = 0.0 |
|---|
| 317 | |
|---|
| 318 | ENDIF |
|---|
| 319 | |
|---|
| 320 | IF ( outflow_n ) THEN |
|---|
| 321 | |
|---|
| 322 | c_max = dy / dt_3d |
|---|
| 323 | |
|---|
| 324 | DO i = nxl-1, nxr+1 |
|---|
| 325 | DO k = nzb+1, nzt+1 |
|---|
| 326 | |
|---|
| 327 | ! |
|---|
| 328 | !-- Calculate the phase speeds for u,v, and w. In case of using a |
|---|
| 329 | !-- Runge-Kutta scheme, do this for the first substep only and then |
|---|
| 330 | !-- reuse this values for the further substeps. |
|---|
| 331 | IF ( intermediate_timestep_count == 1 ) THEN |
|---|
| 332 | |
|---|
| 333 | denom = u_m_n(k,ny,i) - u_m_n(k,ny-1,i) |
|---|
| 334 | |
|---|
| 335 | IF ( denom /= 0.0 ) THEN |
|---|
| 336 | c_u(k,i) = -c_max * ( u(k,ny,i) - u_m_n(k,ny,i) ) / denom |
|---|
| 337 | IF ( c_u(k,i) < 0.0 ) THEN |
|---|
| 338 | c_u(k,i) = 0.0 |
|---|
| 339 | ELSEIF ( c_u(k,i) > c_max ) THEN |
|---|
| 340 | c_u(k,i) = c_max |
|---|
| 341 | ENDIF |
|---|
| 342 | ELSE |
|---|
| 343 | c_u(k,i) = c_max |
|---|
| 344 | ENDIF |
|---|
| 345 | |
|---|
| 346 | denom = v_m_n(k,ny,i) - v_m_n(k,ny-1,i) |
|---|
| 347 | |
|---|
| 348 | IF ( denom /= 0.0 ) THEN |
|---|
| 349 | c_v(k,i) = -c_max * ( v(k,ny,i) - v_m_n(k,ny,i) ) / denom |
|---|
| 350 | IF ( c_v(k,i) < 0.0 ) THEN |
|---|
| 351 | c_v(k,i) = 0.0 |
|---|
| 352 | ELSEIF ( c_v(k,i) > c_max ) THEN |
|---|
| 353 | c_v(k,i) = c_max |
|---|
| 354 | ENDIF |
|---|
| 355 | ELSE |
|---|
| 356 | c_v(k,i) = c_max |
|---|
| 357 | ENDIF |
|---|
| 358 | |
|---|
| 359 | denom = w_m_n(k,ny,i) - w_m_n(k,ny-1,i) |
|---|
| 360 | |
|---|
| 361 | IF ( denom /= 0.0 ) THEN |
|---|
| 362 | c_w(k,i) = -c_max * ( w(k,ny,i) - w_m_n(k,ny,i) ) / denom |
|---|
| 363 | IF ( c_w(k,i) < 0.0 ) THEN |
|---|
| 364 | c_w(k,i) = 0.0 |
|---|
| 365 | ELSEIF ( c_w(k,i) > c_max ) THEN |
|---|
| 366 | c_w(k,i) = c_max |
|---|
| 367 | ENDIF |
|---|
| 368 | ELSE |
|---|
| 369 | c_w(k,i) = c_max |
|---|
| 370 | ENDIF |
|---|
| 371 | |
|---|
| 372 | ! |
|---|
| 373 | !-- Swap timelevels for the next timestep |
|---|
| 374 | u_m_n(k,:,i) = u(k,ny-1:ny,i) |
|---|
| 375 | v_m_n(k,:,i) = v(k,ny-1:ny,i) |
|---|
| 376 | w_m_n(k,:,i) = w(k,ny-1:ny,i) |
|---|
| 377 | |
|---|
| 378 | ENDIF |
|---|
| 379 | |
|---|
| 380 | ! |
|---|
| 381 | !-- Calculate the new velocities |
|---|
| 382 | u_p(k,ny+1,i) = u(k,ny+1,i) - dt_3d * tsc(2) * c_u(k,i) * & |
|---|
| 383 | ( u(k,ny+1,i) - u(k,ny,i) ) * ddy |
|---|
| 384 | |
|---|
| 385 | v_p(k,ny+1,i) = v(k,ny+1,i) - dt_3d * tsc(2) * c_v(k,i) * & |
|---|
| 386 | ( v(k,ny+1,i) - v(k,ny,i) ) * ddy |
|---|
| 387 | |
|---|
| 388 | w_p(k,ny+1,i) = w(k,ny+1,i) - dt_3d * tsc(2) * c_w(k,i) * & |
|---|
| 389 | ( w(k,ny+1,i) - w(k,ny,i) ) * ddy |
|---|
| 390 | |
|---|
| 391 | ENDDO |
|---|
| 392 | ENDDO |
|---|
| 393 | |
|---|
| 394 | ! |
|---|
| 395 | !-- Bottom boundary at the outflow |
|---|
| 396 | IF ( ibc_uv_b == 0 ) THEN |
|---|
| 397 | u_p(nzb,ny+1,:) = -u_p(nzb+1,ny+1,:) |
|---|
| 398 | v_p(nzb,ny+1,:) = -v_p(nzb+1,ny+1,:) |
|---|
| 399 | ELSE |
|---|
| 400 | u_p(nzb,ny+1,:) = u_p(nzb+1,ny+1,:) |
|---|
| 401 | v_p(nzb,ny+1,:) = v_p(nzb+1,ny+1,:) |
|---|
| 402 | ENDIF |
|---|
| 403 | w_p(nzb,ny+1,:) = 0.0 |
|---|
| 404 | |
|---|
| 405 | ! |
|---|
| 406 | !-- Top boundary at the outflow |
|---|
| 407 | IF ( ibc_uv_t == 0 ) THEN |
|---|
| 408 | u_p(nzt+1,ny+1,:) = ug(nzt+1) |
|---|
| 409 | v_p(nzt+1,ny+1,:) = vg(nzt+1) |
|---|
| 410 | ELSE |
|---|
| 411 | u_p(nzt+1,ny+1,:) = u_p(nzt,nyn+1,:) |
|---|
| 412 | v_p(nzt+1,ny+1,:) = v_p(nzt,nyn+1,:) |
|---|
| 413 | ENDIF |
|---|
| 414 | w_p(nzt:nzt+1,ny+1,:) = 0.0 |
|---|
| 415 | |
|---|
| 416 | ENDIF |
|---|
| 417 | |
|---|
| 418 | IF ( outflow_l ) THEN |
|---|
| 419 | |
|---|
| 420 | c_max = dx / dt_3d |
|---|
| 421 | |
|---|
| 422 | DO j = nys-1, nyn+1 |
|---|
| 423 | DO k = nzb+1, nzt+1 |
|---|
| 424 | |
|---|
| 425 | ! |
|---|
| 426 | !-- Calculate the phase speeds for u,v, and w. In case of using a |
|---|
| 427 | !-- Runge-Kutta scheme, do this for the first substep only and then |
|---|
| 428 | !-- reuse this values for the further substeps. |
|---|
| 429 | IF ( intermediate_timestep_count == 1 ) THEN |
|---|
| 430 | |
|---|
| 431 | denom = u_m_l(k,j,1) - u_m_l(k,j,2) |
|---|
| 432 | |
|---|
| 433 | IF ( denom /= 0.0 ) THEN |
|---|
| 434 | c_u(k,j) = -c_max * ( u(k,j,1) - u_m_l(k,j,1) ) / denom |
|---|
| 435 | IF ( c_u(k,j) < 0.0 ) THEN |
|---|
| 436 | c_u(k,j) = 0.0 |
|---|
| 437 | ELSEIF ( c_u(k,j) > c_max ) THEN |
|---|
| 438 | c_u(k,j) = c_max |
|---|
| 439 | ENDIF |
|---|
| 440 | ELSE |
|---|
| 441 | c_u(k,j) = c_max |
|---|
| 442 | ENDIF |
|---|
| 443 | |
|---|
| 444 | denom = v_m_l(k,j,0) - v_m_l(k,j,1) |
|---|
| 445 | |
|---|
| 446 | IF ( denom /= 0.0 ) THEN |
|---|
| 447 | c_v(k,j) = -c_max * ( v(k,j,0) - v_m_l(k,j,0) ) / denom |
|---|
| 448 | IF ( c_v(k,j) < 0.0 ) THEN |
|---|
| 449 | c_v(k,j) = 0.0 |
|---|
| 450 | ELSEIF ( c_v(k,j) > c_max ) THEN |
|---|
| 451 | c_v(k,j) = c_max |
|---|
| 452 | ENDIF |
|---|
| 453 | ELSE |
|---|
| 454 | c_v(k,j) = c_max |
|---|
| 455 | ENDIF |
|---|
| 456 | |
|---|
| 457 | denom = w_m_l(k,j,0) - w_m_l(k,j,1) |
|---|
| 458 | |
|---|
| 459 | IF ( denom /= 0.0 ) THEN |
|---|
| 460 | c_w(k,j) = -c_max * ( w(k,j,0) - w_m_l(k,j,0) ) / denom |
|---|
| 461 | IF ( c_w(k,j) < 0.0 ) THEN |
|---|
| 462 | c_w(k,j) = 0.0 |
|---|
| 463 | ELSEIF ( c_w(k,j) > c_max ) THEN |
|---|
| 464 | c_w(k,j) = c_max |
|---|
| 465 | ENDIF |
|---|
| 466 | ELSE |
|---|
| 467 | c_w(k,j) = c_max |
|---|
| 468 | ENDIF |
|---|
| 469 | |
|---|
| 470 | ! |
|---|
| 471 | !-- Swap timelevels for the next timestep |
|---|
| 472 | u_m_l(k,j,:) = u(k,j,1:2) |
|---|
| 473 | v_m_l(k,j,:) = v(k,j,0:1) |
|---|
| 474 | w_m_l(k,j,:) = w(k,j,0:1) |
|---|
| 475 | |
|---|
| 476 | ENDIF |
|---|
| 477 | |
|---|
| 478 | ! |
|---|
| 479 | !-- Calculate the new velocities |
|---|
| 480 | u_p(k,j,0) = u(k,j,0) - dt_3d * tsc(2) * c_u(k,j) * & |
|---|
| 481 | ( u(k,j,0) - u(k,j,1) ) * ddx |
|---|
| 482 | |
|---|
| 483 | v_p(k,j,-1) = v(k,j,-1) - dt_3d * tsc(2) * c_v(k,j) * & |
|---|
| 484 | ( v(k,j,-1) - v(k,j,0) ) * ddx |
|---|
| 485 | |
|---|
| 486 | w_p(k,j,-1) = w(k,j,-1) - dt_3d * tsc(2) * c_w(k,j) * & |
|---|
| 487 | ( w(k,j,-1) - w(k,j,0) ) * ddx |
|---|
| 488 | |
|---|
| 489 | ENDDO |
|---|
| 490 | ENDDO |
|---|
| 491 | |
|---|
| 492 | ! |
|---|
| 493 | !-- Bottom boundary at the outflow |
|---|
| 494 | IF ( ibc_uv_b == 0 ) THEN |
|---|
| 495 | u_p(nzb,:,-1) = -u_p(nzb+1,:,-1) |
|---|
| 496 | v_p(nzb,:,-1) = -v_p(nzb+1,:,-1) |
|---|
| 497 | ELSE |
|---|
| 498 | u_p(nzb,:,-1) = u_p(nzb+1,:,-1) |
|---|
| 499 | v_p(nzb,:,-1) = v_p(nzb+1,:,-1) |
|---|
| 500 | ENDIF |
|---|
| 501 | w_p(nzb,:,-1) = 0.0 |
|---|
| 502 | |
|---|
| 503 | ! |
|---|
| 504 | !-- Top boundary at the outflow |
|---|
| 505 | IF ( ibc_uv_t == 0 ) THEN |
|---|
| 506 | u_p(nzt+1,:,-1) = ug(nzt+1) |
|---|
| 507 | v_p(nzt+1,:,-1) = vg(nzt+1) |
|---|
| 508 | ELSE |
|---|
| 509 | u_p(nzt+1,:,-1) = u_p(nzt,:,-1) |
|---|
| 510 | v_p(nzt+1,:,-1) = v_p(nzt,:,-1) |
|---|
| 511 | ENDIF |
|---|
| 512 | w_p(nzt:nzt+1,:,-1) = 0.0 |
|---|
| 513 | |
|---|
| 514 | ENDIF |
|---|
| 515 | |
|---|
| 516 | IF ( outflow_r ) THEN |
|---|
| 517 | |
|---|
| 518 | c_max = dx / dt_3d |
|---|
| 519 | |
|---|
| 520 | DO j = nys-1, nyn+1 |
|---|
| 521 | DO k = nzb+1, nzt+1 |
|---|
| 522 | |
|---|
| 523 | ! |
|---|
| 524 | !-- Calculate the phase speeds for u,v, and w. In case of using a |
|---|
| 525 | !-- Runge-Kutta scheme, do this for the first substep only and then |
|---|
| 526 | !-- reuse this values for the further substeps. |
|---|
| 527 | IF ( intermediate_timestep_count == 1 ) THEN |
|---|
| 528 | |
|---|
| 529 | denom = u_m_r(k,j,nx) - u_m_r(k,j,nx-1) |
|---|
| 530 | |
|---|
| 531 | IF ( denom /= 0.0 ) THEN |
|---|
| 532 | c_u(k,j) = -c_max * ( u(k,j,nx) - u_m_r(k,j,nx) ) / denom |
|---|
| 533 | IF ( c_u(k,j) < 0.0 ) THEN |
|---|
| 534 | c_u(k,j) = 0.0 |
|---|
| 535 | ELSEIF ( c_u(k,j) > c_max ) THEN |
|---|
| 536 | c_u(k,j) = c_max |
|---|
| 537 | ENDIF |
|---|
| 538 | ELSE |
|---|
| 539 | c_u(k,j) = c_max |
|---|
| 540 | ENDIF |
|---|
| 541 | |
|---|
| 542 | denom = v_m_r(k,j,nx) - v_m_r(k,j,nx-1) |
|---|
| 543 | |
|---|
| 544 | IF ( denom /= 0.0 ) THEN |
|---|
| 545 | c_v(k,j) = -c_max * ( v(k,j,nx) - v_m_r(k,j,nx) ) / denom |
|---|
| 546 | IF ( c_v(k,j) < 0.0 ) THEN |
|---|
| 547 | c_v(k,j) = 0.0 |
|---|
| 548 | ELSEIF ( c_v(k,j) > c_max ) THEN |
|---|
| 549 | c_v(k,j) = c_max |
|---|
| 550 | ENDIF |
|---|
| 551 | ELSE |
|---|
| 552 | c_v(k,j) = c_max |
|---|
| 553 | ENDIF |
|---|
| 554 | |
|---|
| 555 | denom = w_m_r(k,j,nx) - w_m_r(k,j,nx-1) |
|---|
| 556 | |
|---|
| 557 | IF ( denom /= 0.0 ) THEN |
|---|
| 558 | c_w(k,j) = -c_max * ( w(k,j,nx) - w_m_r(k,j,nx) ) / denom |
|---|
| 559 | IF ( c_w(k,j) < 0.0 ) THEN |
|---|
| 560 | c_w(k,j) = 0.0 |
|---|
| 561 | ELSEIF ( c_w(k,j) > c_max ) THEN |
|---|
| 562 | c_w(k,j) = c_max |
|---|
| 563 | ENDIF |
|---|
| 564 | ELSE |
|---|
| 565 | c_w(k,j) = c_max |
|---|
| 566 | ENDIF |
|---|
| 567 | |
|---|
| 568 | ! |
|---|
| 569 | !-- Swap timelevels for the next timestep |
|---|
| 570 | u_m_r(k,j,:) = u(k,j,nx-1:nx) |
|---|
| 571 | v_m_r(k,j,:) = v(k,j,nx-1:nx) |
|---|
| 572 | w_m_r(k,j,:) = w(k,j,nx-1:nx) |
|---|
| 573 | |
|---|
| 574 | ENDIF |
|---|
| 575 | |
|---|
| 576 | ! |
|---|
| 577 | !-- Calculate the new velocities |
|---|
| 578 | u_p(k,j,nx+1) = u(k,j,nx+1) - dt_3d * tsc(2) * c_u(k,j) * & |
|---|
| 579 | ( u(k,j,nx+1) - u(k,j,nx) ) * ddx |
|---|
| 580 | |
|---|
| 581 | v_p(k,j,nx+1) = v(k,j,nx+1) - dt_3d * tsc(2) * c_v(k,j) * & |
|---|
| 582 | ( v(k,j,nx+1) - v(k,j,nx) ) * ddx |
|---|
| 583 | |
|---|
| 584 | w_p(k,j,nx+1) = w(k,j,nx+1) - dt_3d * tsc(2) * c_w(k,j) * & |
|---|
| 585 | ( w(k,j,nx+1) - w(k,j,nx) ) * ddx |
|---|
| 586 | |
|---|
| 587 | ENDDO |
|---|
| 588 | ENDDO |
|---|
| 589 | |
|---|
| 590 | ! |
|---|
| 591 | !-- Bottom boundary at the outflow |
|---|
| 592 | IF ( ibc_uv_b == 0 ) THEN |
|---|
| 593 | u_p(nzb,:,nx+1) = -u_p(nzb+1,:,nx+1) |
|---|
| 594 | v_p(nzb,:,nx+1) = -v_p(nzb+1,:,nx+1) |
|---|
| 595 | ELSE |
|---|
| 596 | u_p(nzb,:,nx+1) = u_p(nzb+1,:,nx+1) |
|---|
| 597 | v_p(nzb,:,nx+1) = v_p(nzb+1,:,nx+1) |
|---|
| 598 | ENDIF |
|---|
| 599 | w_p(nzb,:,nx+1) = 0.0 |
|---|
| 600 | |
|---|
| 601 | ! |
|---|
| 602 | !-- Top boundary at the outflow |
|---|
| 603 | IF ( ibc_uv_t == 0 ) THEN |
|---|
| 604 | u_p(nzt+1,:,nx+1) = ug(nzt+1) |
|---|
| 605 | v_p(nzt+1,:,nx+1) = vg(nzt+1) |
|---|
| 606 | ELSE |
|---|
| 607 | u_p(nzt+1,:,nx+1) = u_p(nzt,:,nx+1) |
|---|
| 608 | v_p(nzt+1,:,nx+1) = v_p(nzt,:,nx+1) |
|---|
| 609 | ENDIF |
|---|
| 610 | w(nzt:nzt+1,:,nx+1) = 0.0 |
|---|
| 611 | |
|---|
| 612 | ENDIF |
|---|
| 613 | |
|---|
| 614 | |
|---|
| 615 | END SUBROUTINE boundary_conds |
|---|
