| 1 | MODULE advec_ws |
|---|
| 2 | |
|---|
| 3 | !--------------------------------------------------------------------------------! |
|---|
| 4 | ! This file is part of PALM. |
|---|
| 5 | ! |
|---|
| 6 | ! PALM is free software: you can redistribute it and/or modify it under the terms |
|---|
| 7 | ! of the GNU General Public License as published by the Free Software Foundation, |
|---|
| 8 | ! either version 3 of the License, or (at your option) any later version. |
|---|
| 9 | ! |
|---|
| 10 | ! PALM is distributed in the hope that it will be useful, but WITHOUT ANY |
|---|
| 11 | ! WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR |
|---|
| 12 | ! A PARTICULAR PURPOSE. See the GNU General Public License for more details. |
|---|
| 13 | ! |
|---|
| 14 | ! You should have received a copy of the GNU General Public License along with |
|---|
| 15 | ! PALM. If not, see <http://www.gnu.org/licenses/>. |
|---|
| 16 | ! |
|---|
| 17 | ! Copyright 1997-2012 Leibniz University Hannover |
|---|
| 18 | !--------------------------------------------------------------------------------! |
|---|
| 19 | ! |
|---|
| 20 | ! Current revisions: |
|---|
| 21 | ! ------------------ |
|---|
| 22 | ! |
|---|
| 23 | ! |
|---|
| 24 | ! Former revisions: |
|---|
| 25 | ! ----------------- |
|---|
| 26 | ! $Id: advec_ws.f90 1222 2013-09-10 14:48:09Z raasch $ |
|---|
| 27 | ! |
|---|
| 28 | ! 1221 2013-09-10 08:59:13Z raasch |
|---|
| 29 | ! wall_flags_00 introduced, which holds bits 32-... |
|---|
| 30 | ! |
|---|
| 31 | ! 1128 2013-04-12 06:19:32Z raasch |
|---|
| 32 | ! loop index bounds in accelerator version replaced by i_left, i_right, j_south, |
|---|
| 33 | ! j_north |
|---|
| 34 | ! |
|---|
| 35 | ! 1115 2013-03-26 18:16:16Z hoffmann |
|---|
| 36 | ! calculation of qr and nr is restricted to precipitation |
|---|
| 37 | ! |
|---|
| 38 | ! 1053 2012-11-13 17:11:03Z hoffmann |
|---|
| 39 | ! necessary expansions according to the two new prognostic equations (nr, qr) |
|---|
| 40 | ! of the two-moment cloud physics scheme: |
|---|
| 41 | ! +flux_l_*, flux_s_*, diss_l_*, diss_s_*, sums_ws*s_ws_l |
|---|
| 42 | ! |
|---|
| 43 | ! 1036 2012-10-22 13:43:42Z raasch |
|---|
| 44 | ! code put under GPL (PALM 3.9) |
|---|
| 45 | ! |
|---|
| 46 | ! 1027 2012-10-15 17:18:39Z suehring |
|---|
| 47 | ! Bugfix in calculation indices k_mm, k_pp in accelerator version |
|---|
| 48 | ! |
|---|
| 49 | ! 1019 2012-09-28 06:46:45Z raasch |
|---|
| 50 | ! small change in comment lines |
|---|
| 51 | ! |
|---|
| 52 | ! 1015 2012-09-27 09:23:24Z raasch |
|---|
| 53 | ! accelerator versions (*_acc) added |
|---|
| 54 | ! |
|---|
| 55 | ! 1010 2012-09-20 07:59:54Z raasch |
|---|
| 56 | ! cpp switch __nopointer added for pointer free version |
|---|
| 57 | ! |
|---|
| 58 | ! 888 2012-04-20 15:03:46Z suehring |
|---|
| 59 | ! Number of IBITS() calls with identical arguments is reduced. |
|---|
| 60 | ! |
|---|
| 61 | ! 862 2012-03-26 14:21:38Z suehring |
|---|
| 62 | ! ws-scheme also work with topography in combination with vector version. |
|---|
| 63 | ! ws-scheme also work with outflow boundaries in combination with |
|---|
| 64 | ! vector version. |
|---|
| 65 | ! Degradation of the applied order of scheme is now steered by multiplying with |
|---|
| 66 | ! Integer wall_flags_0. 2nd order scheme, WS3 and WS5 are calculated on each |
|---|
| 67 | ! grid point and mulitplied with the appropriate flag. |
|---|
| 68 | ! 2nd order numerical dissipation term changed. Now the appropriate 2nd order |
|---|
| 69 | ! term derived according to the 4th and 6th order terms is applied. It turns |
|---|
| 70 | ! out that diss_2nd does not provide sufficient dissipation near walls. |
|---|
| 71 | ! Therefore, the function diss_2nd is removed. |
|---|
| 72 | ! Near walls a divergence correction is necessary to overcome numerical |
|---|
| 73 | ! instabilities due to too less divergence reduction of the velocity field. |
|---|
| 74 | ! boundary_flags and logicals steering the degradation are removed. |
|---|
| 75 | ! Empty SUBROUTINE local_diss removed. |
|---|
| 76 | ! Further formatting adjustments. |
|---|
| 77 | ! |
|---|
| 78 | ! 801 2012-01-10 17:30:36Z suehring |
|---|
| 79 | ! Bugfix concerning OpenMP parallelization. Summation of sums_wsus_ws_l, |
|---|
| 80 | ! sums_wsvs_ws_l, sums_us2_ws_l, sums_vs2_ws_l, sums_ws2_ws_l, sums_wspts_ws_l, |
|---|
| 81 | ! sums_wsqs_ws_l, sums_wssas_ws_l is now thread-safe by adding an additional |
|---|
| 82 | ! dimension. |
|---|
| 83 | ! |
|---|
| 84 | ! 743 2011-08-18 16:10:16Z suehring |
|---|
| 85 | ! Evaluation of turbulent fluxes with WS-scheme only for the whole model |
|---|
| 86 | ! domain. Therefor dimension of arrays needed for statistical evaluation |
|---|
| 87 | ! decreased. |
|---|
| 88 | ! |
|---|
| 89 | ! 736 2011-08-17 14:13:26Z suehring |
|---|
| 90 | ! Bugfix concerning OpenMP parallelization. i_omp introduced, because first |
|---|
| 91 | ! index where fluxes on left side have to be calculated explicitly is |
|---|
| 92 | ! different on each thread. Furthermore the swapping of fluxes is now |
|---|
| 93 | ! thread-safe by adding an additional dimension. |
|---|
| 94 | ! |
|---|
| 95 | ! 713 2011-03-30 14:21:21Z suehring |
|---|
| 96 | ! File reformatted. |
|---|
| 97 | ! Bugfix in vertical advection of w concerning the optimized version for |
|---|
| 98 | ! vector architecture. |
|---|
| 99 | ! Constants adv_mom_3, adv_mom_5, adv_sca_5, adv_sca_3 reformulated as |
|---|
| 100 | ! broken numbers. |
|---|
| 101 | ! |
|---|
| 102 | ! 709 2011-03-30 09:31:40Z raasch |
|---|
| 103 | ! formatting adjustments |
|---|
| 104 | ! |
|---|
| 105 | ! 705 2011-03-25 11:21:43 Z suehring $ |
|---|
| 106 | ! Bugfix in declaration of logicals concerning outflow boundaries. |
|---|
| 107 | ! |
|---|
| 108 | ! 411 2009-12-11 12:31:43 Z suehring |
|---|
| 109 | ! Allocation of weight_substep moved to init_3d_model. |
|---|
| 110 | ! Declaration of ws_scheme_sca and ws_scheme_mom moved to check_parameters. |
|---|
| 111 | ! Setting bc for the horizontal velocity variances added (moved from |
|---|
| 112 | ! flow_statistics). |
|---|
| 113 | ! |
|---|
| 114 | ! Initial revision |
|---|
| 115 | ! |
|---|
| 116 | ! 411 2009-12-11 12:31:43 Z suehring |
|---|
| 117 | ! |
|---|
| 118 | ! Description: |
|---|
| 119 | ! ------------ |
|---|
| 120 | ! Advection scheme for scalars and momentum using the flux formulation of |
|---|
| 121 | ! Wicker and Skamarock 5th order. Additionally the module contains of a |
|---|
| 122 | ! routine using for initialisation and steering of the statical evaluation. |
|---|
| 123 | ! The computation of turbulent fluxes takes place inside the advection |
|---|
| 124 | ! routines. |
|---|
| 125 | ! Near non-cyclic boundaries the order of the applied advection scheme is |
|---|
| 126 | ! degraded. |
|---|
| 127 | ! A divergence correction is applied. It is necessary for topography, since |
|---|
| 128 | ! the divergence is not sufficiently reduced, resulting in erroneous fluxes and |
|---|
| 129 | ! partly numerical instabilities. |
|---|
| 130 | !-----------------------------------------------------------------------------! |
|---|
| 131 | |
|---|
| 132 | PRIVATE |
|---|
| 133 | PUBLIC advec_s_ws, advec_s_ws_acc, advec_u_ws, advec_u_ws_acc, & |
|---|
| 134 | advec_v_ws, advec_v_ws_acc, advec_w_ws, advec_w_ws_acc, & |
|---|
| 135 | ws_init, ws_statistics |
|---|
| 136 | |
|---|
| 137 | INTERFACE ws_init |
|---|
| 138 | MODULE PROCEDURE ws_init |
|---|
| 139 | END INTERFACE ws_init |
|---|
| 140 | |
|---|
| 141 | INTERFACE ws_statistics |
|---|
| 142 | MODULE PROCEDURE ws_statistics |
|---|
| 143 | END INTERFACE ws_statistics |
|---|
| 144 | |
|---|
| 145 | INTERFACE advec_s_ws |
|---|
| 146 | MODULE PROCEDURE advec_s_ws |
|---|
| 147 | MODULE PROCEDURE advec_s_ws_ij |
|---|
| 148 | END INTERFACE advec_s_ws |
|---|
| 149 | |
|---|
| 150 | INTERFACE advec_u_ws |
|---|
| 151 | MODULE PROCEDURE advec_u_ws |
|---|
| 152 | MODULE PROCEDURE advec_u_ws_ij |
|---|
| 153 | END INTERFACE advec_u_ws |
|---|
| 154 | |
|---|
| 155 | INTERFACE advec_u_ws_acc |
|---|
| 156 | MODULE PROCEDURE advec_u_ws_acc |
|---|
| 157 | END INTERFACE advec_u_ws_acc |
|---|
| 158 | |
|---|
| 159 | INTERFACE advec_v_ws |
|---|
| 160 | MODULE PROCEDURE advec_v_ws |
|---|
| 161 | MODULE PROCEDURE advec_v_ws_ij |
|---|
| 162 | END INTERFACE advec_v_ws |
|---|
| 163 | |
|---|
| 164 | INTERFACE advec_v_ws_acc |
|---|
| 165 | MODULE PROCEDURE advec_v_ws_acc |
|---|
| 166 | END INTERFACE advec_v_ws_acc |
|---|
| 167 | |
|---|
| 168 | INTERFACE advec_w_ws |
|---|
| 169 | MODULE PROCEDURE advec_w_ws |
|---|
| 170 | MODULE PROCEDURE advec_w_ws_ij |
|---|
| 171 | END INTERFACE advec_w_ws |
|---|
| 172 | |
|---|
| 173 | INTERFACE advec_w_ws_acc |
|---|
| 174 | MODULE PROCEDURE advec_w_ws_acc |
|---|
| 175 | END INTERFACE advec_w_ws_acc |
|---|
| 176 | |
|---|
| 177 | CONTAINS |
|---|
| 178 | |
|---|
| 179 | |
|---|
| 180 | !------------------------------------------------------------------------------! |
|---|
| 181 | ! Initialization of WS-scheme |
|---|
| 182 | !------------------------------------------------------------------------------! |
|---|
| 183 | SUBROUTINE ws_init |
|---|
| 184 | |
|---|
| 185 | USE arrays_3d |
|---|
| 186 | USE constants |
|---|
| 187 | USE control_parameters |
|---|
| 188 | USE indices |
|---|
| 189 | USE pegrid |
|---|
| 190 | USE statistics |
|---|
| 191 | |
|---|
| 192 | ! |
|---|
| 193 | !-- Set the appropriate factors for scalar and momentum advection. |
|---|
| 194 | adv_sca_5 = 1./60. |
|---|
| 195 | adv_sca_3 = 1./12. |
|---|
| 196 | adv_sca_1 = 1./2. |
|---|
| 197 | adv_mom_5 = 1./120. |
|---|
| 198 | adv_mom_3 = 1./24. |
|---|
| 199 | adv_mom_1 = 1./4. |
|---|
| 200 | ! |
|---|
| 201 | !-- Arrays needed for statical evaluation of fluxes. |
|---|
| 202 | IF ( ws_scheme_mom ) THEN |
|---|
| 203 | |
|---|
| 204 | ALLOCATE( sums_wsus_ws_l(nzb:nzt+1,0:threads_per_task-1), & |
|---|
| 205 | sums_wsvs_ws_l(nzb:nzt+1,0:threads_per_task-1), & |
|---|
| 206 | sums_us2_ws_l(nzb:nzt+1,0:threads_per_task-1), & |
|---|
| 207 | sums_vs2_ws_l(nzb:nzt+1,0:threads_per_task-1), & |
|---|
| 208 | sums_ws2_ws_l(nzb:nzt+1,0:threads_per_task-1) ) |
|---|
| 209 | |
|---|
| 210 | sums_wsus_ws_l = 0.0 |
|---|
| 211 | sums_wsvs_ws_l = 0.0 |
|---|
| 212 | sums_us2_ws_l = 0.0 |
|---|
| 213 | sums_vs2_ws_l = 0.0 |
|---|
| 214 | sums_ws2_ws_l = 0.0 |
|---|
| 215 | |
|---|
| 216 | ENDIF |
|---|
| 217 | |
|---|
| 218 | IF ( ws_scheme_sca ) THEN |
|---|
| 219 | |
|---|
| 220 | ALLOCATE( sums_wspts_ws_l(nzb:nzt+1,0:threads_per_task-1) ) |
|---|
| 221 | sums_wspts_ws_l = 0.0 |
|---|
| 222 | |
|---|
| 223 | IF ( humidity .OR. passive_scalar ) THEN |
|---|
| 224 | ALLOCATE( sums_wsqs_ws_l(nzb:nzt+1,0:threads_per_task-1) ) |
|---|
| 225 | sums_wsqs_ws_l = 0.0 |
|---|
| 226 | ENDIF |
|---|
| 227 | |
|---|
| 228 | IF ( cloud_physics .AND. icloud_scheme == 0 .AND. & |
|---|
| 229 | precipitation ) THEN |
|---|
| 230 | ALLOCATE( sums_wsqrs_ws_l(nzb:nzt+1,0:threads_per_task-1) ) |
|---|
| 231 | ALLOCATE( sums_wsnrs_ws_l(nzb:nzt+1,0:threads_per_task-1) ) |
|---|
| 232 | sums_wsqrs_ws_l = 0.0 |
|---|
| 233 | sums_wsnrs_ws_l = 0.0 |
|---|
| 234 | ENDIF |
|---|
| 235 | |
|---|
| 236 | IF ( ocean ) THEN |
|---|
| 237 | ALLOCATE( sums_wssas_ws_l(nzb:nzt+1,0:threads_per_task-1) ) |
|---|
| 238 | sums_wssas_ws_l = 0.0 |
|---|
| 239 | ENDIF |
|---|
| 240 | |
|---|
| 241 | ENDIF |
|---|
| 242 | |
|---|
| 243 | ! |
|---|
| 244 | !-- Arrays needed for reasons of speed optimization for cache version. |
|---|
| 245 | !-- For the vector version the buffer arrays are not necessary, |
|---|
| 246 | !-- because the the fluxes can swapped directly inside the loops of the |
|---|
| 247 | !-- advection routines. |
|---|
| 248 | IF ( loop_optimization /= 'vector' ) THEN |
|---|
| 249 | |
|---|
| 250 | IF ( ws_scheme_mom ) THEN |
|---|
| 251 | |
|---|
| 252 | ALLOCATE( flux_s_u(nzb+1:nzt,0:threads_per_task-1), & |
|---|
| 253 | flux_s_v(nzb+1:nzt,0:threads_per_task-1), & |
|---|
| 254 | flux_s_w(nzb+1:nzt,0:threads_per_task-1), & |
|---|
| 255 | diss_s_u(nzb+1:nzt,0:threads_per_task-1), & |
|---|
| 256 | diss_s_v(nzb+1:nzt,0:threads_per_task-1), & |
|---|
| 257 | diss_s_w(nzb+1:nzt,0:threads_per_task-1) ) |
|---|
| 258 | ALLOCATE( flux_l_u(nzb+1:nzt,nys:nyn,0:threads_per_task-1), & |
|---|
| 259 | flux_l_v(nzb+1:nzt,nys:nyn,0:threads_per_task-1), & |
|---|
| 260 | flux_l_w(nzb+1:nzt,nys:nyn,0:threads_per_task-1), & |
|---|
| 261 | diss_l_u(nzb+1:nzt,nys:nyn,0:threads_per_task-1), & |
|---|
| 262 | diss_l_v(nzb+1:nzt,nys:nyn,0:threads_per_task-1), & |
|---|
| 263 | diss_l_w(nzb+1:nzt,nys:nyn,0:threads_per_task-1) ) |
|---|
| 264 | |
|---|
| 265 | ENDIF |
|---|
| 266 | |
|---|
| 267 | IF ( ws_scheme_sca ) THEN |
|---|
| 268 | |
|---|
| 269 | ALLOCATE( flux_s_pt(nzb+1:nzt,0:threads_per_task-1), & |
|---|
| 270 | flux_s_e(nzb+1:nzt,0:threads_per_task-1), & |
|---|
| 271 | diss_s_pt(nzb+1:nzt,0:threads_per_task-1), & |
|---|
| 272 | diss_s_e(nzb+1:nzt,0:threads_per_task-1) ) |
|---|
| 273 | ALLOCATE( flux_l_pt(nzb+1:nzt,nys:nyn,0:threads_per_task-1), & |
|---|
| 274 | flux_l_e(nzb+1:nzt,nys:nyn,0:threads_per_task-1), & |
|---|
| 275 | diss_l_pt(nzb+1:nzt,nys:nyn,0:threads_per_task-1), & |
|---|
| 276 | diss_l_e(nzb+1:nzt,nys:nyn,0:threads_per_task-1) ) |
|---|
| 277 | |
|---|
| 278 | IF ( humidity .OR. passive_scalar ) THEN |
|---|
| 279 | ALLOCATE( flux_s_q(nzb+1:nzt,0:threads_per_task-1), & |
|---|
| 280 | diss_s_q(nzb+1:nzt,0:threads_per_task-1) ) |
|---|
| 281 | ALLOCATE( flux_l_q(nzb+1:nzt,nys:nyn,0:threads_per_task-1), & |
|---|
| 282 | diss_l_q(nzb+1:nzt,nys:nyn,0:threads_per_task-1) ) |
|---|
| 283 | ENDIF |
|---|
| 284 | |
|---|
| 285 | IF ( cloud_physics .AND. icloud_scheme == 0 .AND. & |
|---|
| 286 | precipitation ) THEN |
|---|
| 287 | ALLOCATE( flux_s_qr(nzb+1:nzt,0:threads_per_task-1), & |
|---|
| 288 | diss_s_qr(nzb+1:nzt,0:threads_per_task-1), & |
|---|
| 289 | flux_s_nr(nzb+1:nzt,0:threads_per_task-1), & |
|---|
| 290 | diss_s_nr(nzb+1:nzt,0:threads_per_task-1) ) |
|---|
| 291 | ALLOCATE( flux_l_qr(nzb+1:nzt,nys:nyn,0:threads_per_task-1), & |
|---|
| 292 | diss_l_qr(nzb+1:nzt,nys:nyn,0:threads_per_task-1), & |
|---|
| 293 | flux_l_nr(nzb+1:nzt,nys:nyn,0:threads_per_task-1), & |
|---|
| 294 | diss_l_nr(nzb+1:nzt,nys:nyn,0:threads_per_task-1) ) |
|---|
| 295 | ENDIF |
|---|
| 296 | |
|---|
| 297 | IF ( ocean ) THEN |
|---|
| 298 | ALLOCATE( flux_s_sa(nzb+1:nzt,0:threads_per_task-1), & |
|---|
| 299 | diss_s_sa(nzb+1:nzt,0:threads_per_task-1) ) |
|---|
| 300 | ALLOCATE( flux_l_sa(nzb+1:nzt,nys:nyn,0:threads_per_task-1), & |
|---|
| 301 | diss_l_sa(nzb+1:nzt,nys:nyn,0:threads_per_task-1) ) |
|---|
| 302 | ENDIF |
|---|
| 303 | |
|---|
| 304 | ENDIF |
|---|
| 305 | |
|---|
| 306 | ENDIF |
|---|
| 307 | |
|---|
| 308 | END SUBROUTINE ws_init |
|---|
| 309 | |
|---|
| 310 | |
|---|
| 311 | !------------------------------------------------------------------------------! |
|---|
| 312 | ! Initialize variables used for storing statistic quantities (fluxes, variances) |
|---|
| 313 | !------------------------------------------------------------------------------! |
|---|
| 314 | SUBROUTINE ws_statistics |
|---|
| 315 | |
|---|
| 316 | USE control_parameters |
|---|
| 317 | USE statistics |
|---|
| 318 | |
|---|
| 319 | IMPLICIT NONE |
|---|
| 320 | |
|---|
| 321 | ! |
|---|
| 322 | !-- The arrays needed for statistical evaluation are set to to 0 at the |
|---|
| 323 | !-- beginning of prognostic_equations. |
|---|
| 324 | IF ( ws_scheme_mom ) THEN |
|---|
| 325 | sums_wsus_ws_l = 0.0 |
|---|
| 326 | sums_wsvs_ws_l = 0.0 |
|---|
| 327 | sums_us2_ws_l = 0.0 |
|---|
| 328 | sums_vs2_ws_l = 0.0 |
|---|
| 329 | sums_ws2_ws_l = 0.0 |
|---|
| 330 | ENDIF |
|---|
| 331 | |
|---|
| 332 | IF ( ws_scheme_sca ) THEN |
|---|
| 333 | sums_wspts_ws_l = 0.0 |
|---|
| 334 | IF ( humidity .OR. passive_scalar ) sums_wsqs_ws_l = 0.0 |
|---|
| 335 | IF ( cloud_physics .AND. icloud_scheme == 0 .AND. & |
|---|
| 336 | precipitation ) THEN |
|---|
| 337 | sums_wsqrs_ws_l = 0.0 |
|---|
| 338 | sums_wsnrs_ws_l = 0.0 |
|---|
| 339 | ENDIF |
|---|
| 340 | IF ( ocean ) sums_wssas_ws_l = 0.0 |
|---|
| 341 | |
|---|
| 342 | ENDIF |
|---|
| 343 | |
|---|
| 344 | END SUBROUTINE ws_statistics |
|---|
| 345 | |
|---|
| 346 | |
|---|
| 347 | !------------------------------------------------------------------------------! |
|---|
| 348 | ! Scalar advection - Call for grid point i,j |
|---|
| 349 | !------------------------------------------------------------------------------! |
|---|
| 350 | SUBROUTINE advec_s_ws_ij( i, j, sk, sk_char, swap_flux_y_local, & |
|---|
| 351 | swap_diss_y_local, swap_flux_x_local, & |
|---|
| 352 | swap_diss_x_local, i_omp, tn ) |
|---|
| 353 | |
|---|
| 354 | USE arrays_3d |
|---|
| 355 | USE constants |
|---|
| 356 | USE control_parameters |
|---|
| 357 | USE grid_variables |
|---|
| 358 | USE indices |
|---|
| 359 | USE pegrid |
|---|
| 360 | USE statistics |
|---|
| 361 | |
|---|
| 362 | IMPLICIT NONE |
|---|
| 363 | |
|---|
| 364 | INTEGER :: i, ibit0, ibit1, ibit2, ibit3, ibit4, ibit5, ibit6, & |
|---|
| 365 | ibit7, ibit8, i_omp, j, k, k_mm, k_pp, k_ppp, tn |
|---|
| 366 | REAL :: diss_d, div, flux_d, u_comp, v_comp |
|---|
| 367 | #if defined( __nopointer ) |
|---|
| 368 | REAL, DIMENSION(nzb:nzt+1,nysg:nyng,nxlg:nxrg) :: sk |
|---|
| 369 | #else |
|---|
| 370 | REAL, DIMENSION(:,:,:), POINTER :: sk |
|---|
| 371 | #endif |
|---|
| 372 | REAL, DIMENSION(nzb:nzt+1) :: diss_n, diss_r, diss_t, flux_n, & |
|---|
| 373 | flux_r, flux_t |
|---|
| 374 | REAL, DIMENSION(nzb+1:nzt,0:threads_per_task-1) :: swap_diss_y_local, & |
|---|
| 375 | swap_flux_y_local |
|---|
| 376 | REAL, DIMENSION(nzb+1:nzt,nys:nyn,0:threads_per_task-1) :: & |
|---|
| 377 | swap_diss_x_local, & |
|---|
| 378 | swap_flux_x_local |
|---|
| 379 | CHARACTER (LEN = *), INTENT(IN) :: sk_char |
|---|
| 380 | |
|---|
| 381 | ! |
|---|
| 382 | !-- Compute southside fluxes of the respective PE bounds. |
|---|
| 383 | IF ( j == nys ) THEN |
|---|
| 384 | ! |
|---|
| 385 | !-- Up to the top of the highest topography. |
|---|
| 386 | DO k = nzb+1, nzb_max |
|---|
| 387 | |
|---|
| 388 | ibit5 = IBITS(wall_flags_0(k,j,i),5,1) |
|---|
| 389 | ibit4 = IBITS(wall_flags_0(k,j,i),4,1) |
|---|
| 390 | ibit3 = IBITS(wall_flags_0(k,j,i),3,1) |
|---|
| 391 | |
|---|
| 392 | v_comp = v(k,j,i) - v_gtrans |
|---|
| 393 | swap_flux_y_local(k,tn) = v_comp * ( & |
|---|
| 394 | ( 37.0 * ibit5 * adv_sca_5 & |
|---|
| 395 | + 7.0 * ibit4 * adv_sca_3 & |
|---|
| 396 | + ibit3 * adv_sca_1 & |
|---|
| 397 | ) * & |
|---|
| 398 | ( sk(k,j,i) + sk(k,j-1,i) ) & |
|---|
| 399 | - ( 8.0 * ibit5 * adv_sca_5 & |
|---|
| 400 | + ibit4 * adv_sca_3 & |
|---|
| 401 | ) * & |
|---|
| 402 | ( sk(k,j+1,i) + sk(k,j-2,i) ) & |
|---|
| 403 | + ( ibit5 * adv_sca_5 & |
|---|
| 404 | ) * & |
|---|
| 405 | ( sk(k,j+2,i) + sk(k,j-3,i) ) & |
|---|
| 406 | ) |
|---|
| 407 | |
|---|
| 408 | swap_diss_y_local(k,tn) = -ABS( v_comp ) * ( & |
|---|
| 409 | ( 10.0 * ibit5 * adv_sca_5 & |
|---|
| 410 | + 3.0 * ibit4 * adv_sca_3 & |
|---|
| 411 | + ibit3 * adv_sca_1 & |
|---|
| 412 | ) * & |
|---|
| 413 | ( sk(k,j,i) - sk(k,j-1,i) ) & |
|---|
| 414 | - ( 5.0 * ibit5 * adv_sca_5 & |
|---|
| 415 | + ibit4 * adv_sca_3 & |
|---|
| 416 | ) * & |
|---|
| 417 | ( sk(k,j+1,i) - sk(k,j-2,i) ) & |
|---|
| 418 | + ( ibit5 * adv_sca_5 & |
|---|
| 419 | ) * & |
|---|
| 420 | ( sk(k,j+2,i) - sk(k,j-3,i) ) & |
|---|
| 421 | ) |
|---|
| 422 | |
|---|
| 423 | ENDDO |
|---|
| 424 | ! |
|---|
| 425 | !-- Above to the top of the highest topography. No degradation necessary. |
|---|
| 426 | DO k = nzb_max+1, nzt |
|---|
| 427 | |
|---|
| 428 | v_comp = v(k,j,i) - v_gtrans |
|---|
| 429 | swap_flux_y_local(k,tn) = v_comp * ( & |
|---|
| 430 | 37.0 * ( sk(k,j,i) + sk(k,j-1,i) ) & |
|---|
| 431 | - 8.0 * ( sk(k,j+1,i) + sk(k,j-2,i) ) & |
|---|
| 432 | + ( sk(k,j+2,i) + sk(k,j-3,i) ) & |
|---|
| 433 | ) * adv_sca_5 |
|---|
| 434 | swap_diss_y_local(k,tn) = -ABS( v_comp ) * ( & |
|---|
| 435 | 10.0 * ( sk(k,j,i) - sk(k,j-1,i) ) & |
|---|
| 436 | - 5.0 * ( sk(k,j+1,i) - sk(k,j-2,i) ) & |
|---|
| 437 | + sk(k,j+2,i) - sk(k,j-3,i) & |
|---|
| 438 | ) * adv_sca_5 |
|---|
| 439 | |
|---|
| 440 | ENDDO |
|---|
| 441 | |
|---|
| 442 | ENDIF |
|---|
| 443 | ! |
|---|
| 444 | !-- Compute leftside fluxes of the respective PE bounds. |
|---|
| 445 | IF ( i == i_omp ) THEN |
|---|
| 446 | |
|---|
| 447 | DO k = nzb+1, nzb_max |
|---|
| 448 | |
|---|
| 449 | ibit2 = IBITS(wall_flags_0(k,j,i),2,1) |
|---|
| 450 | ibit1 = IBITS(wall_flags_0(k,j,i),1,1) |
|---|
| 451 | ibit0 = IBITS(wall_flags_0(k,j,i),0,1) |
|---|
| 452 | |
|---|
| 453 | u_comp = u(k,j,i) - u_gtrans |
|---|
| 454 | swap_flux_x_local(k,j,tn) = u_comp * ( & |
|---|
| 455 | ( 37.0 * ibit2 * adv_sca_5 & |
|---|
| 456 | + 7.0 * ibit1 * adv_sca_3 & |
|---|
| 457 | + ibit0 * adv_sca_1 & |
|---|
| 458 | ) * & |
|---|
| 459 | ( sk(k,j,i) + sk(k,j,i-1) ) & |
|---|
| 460 | - ( 8.0 * ibit2 * adv_sca_5 & |
|---|
| 461 | + ibit1 * adv_sca_3 & |
|---|
| 462 | ) * & |
|---|
| 463 | ( sk(k,j,i+1) + sk(k,j,i-2) ) & |
|---|
| 464 | + ( ibit2 * adv_sca_5 & |
|---|
| 465 | ) * & |
|---|
| 466 | ( sk(k,j,i+2) + sk(k,j,i-3) ) & |
|---|
| 467 | ) |
|---|
| 468 | |
|---|
| 469 | swap_diss_x_local(k,j,tn) = -ABS( u_comp ) * ( & |
|---|
| 470 | ( 10.0 * ibit2 * adv_sca_5 & |
|---|
| 471 | + 3.0 * ibit1 * adv_sca_3 & |
|---|
| 472 | + ibit0 * adv_sca_1 & |
|---|
| 473 | ) * & |
|---|
| 474 | ( sk(k,j,i) - sk(k,j,i-1) ) & |
|---|
| 475 | - ( 5.0 * ibit2 * adv_sca_5 & |
|---|
| 476 | + ibit1 * adv_sca_3 & |
|---|
| 477 | ) * & |
|---|
| 478 | ( sk(k,j,i+1) - sk(k,j,i-2) ) & |
|---|
| 479 | + ( ibit2 * adv_sca_5 & |
|---|
| 480 | ) * & |
|---|
| 481 | ( sk(k,j,i+2) - sk(k,j,i-3) ) & |
|---|
| 482 | ) |
|---|
| 483 | |
|---|
| 484 | ENDDO |
|---|
| 485 | |
|---|
| 486 | DO k = nzb_max+1, nzt |
|---|
| 487 | |
|---|
| 488 | u_comp = u(k,j,i) - u_gtrans |
|---|
| 489 | swap_flux_x_local(k,j,tn) = u_comp * ( & |
|---|
| 490 | 37.0 * ( sk(k,j,i) + sk(k,j,i-1) ) & |
|---|
| 491 | - 8.0 * ( sk(k,j,i+1) + sk(k,j,i-2) ) & |
|---|
| 492 | + ( sk(k,j,i+2) + sk(k,j,i-3) ) & |
|---|
| 493 | ) * adv_sca_5 |
|---|
| 494 | |
|---|
| 495 | swap_diss_x_local(k,j,tn) = -ABS( u_comp ) * ( & |
|---|
| 496 | 10.0 * ( sk(k,j,i) - sk(k,j,i-1) ) & |
|---|
| 497 | - 5.0 * ( sk(k,j,i+1) - sk(k,j,i-2) ) & |
|---|
| 498 | + ( sk(k,j,i+2) - sk(k,j,i-3) ) & |
|---|
| 499 | ) * adv_sca_5 |
|---|
| 500 | |
|---|
| 501 | ENDDO |
|---|
| 502 | |
|---|
| 503 | ENDIF |
|---|
| 504 | |
|---|
| 505 | flux_t(0) = 0.0 |
|---|
| 506 | diss_t(0) = 0.0 |
|---|
| 507 | flux_d = 0.0 |
|---|
| 508 | diss_d = 0.0 |
|---|
| 509 | ! |
|---|
| 510 | !-- Now compute the fluxes and tendency terms for the horizontal and |
|---|
| 511 | !-- vertical parts up to the top of the highest topography. |
|---|
| 512 | DO k = nzb+1, nzb_max |
|---|
| 513 | ! |
|---|
| 514 | !-- Note: It is faster to conduct all multiplications explicitly, e.g. |
|---|
| 515 | !-- * adv_sca_5 ... than to determine a factor and multiplicate the |
|---|
| 516 | !-- flux at the end. |
|---|
| 517 | |
|---|
| 518 | ibit2 = IBITS(wall_flags_0(k,j,i),2,1) |
|---|
| 519 | ibit1 = IBITS(wall_flags_0(k,j,i),1,1) |
|---|
| 520 | ibit0 = IBITS(wall_flags_0(k,j,i),0,1) |
|---|
| 521 | |
|---|
| 522 | u_comp = u(k,j,i+1) - u_gtrans |
|---|
| 523 | flux_r(k) = u_comp * ( & |
|---|
| 524 | ( 37.0 * ibit2 * adv_sca_5 & |
|---|
| 525 | + 7.0 * ibit1 * adv_sca_3 & |
|---|
| 526 | + ibit0 * adv_sca_1 & |
|---|
| 527 | ) * & |
|---|
| 528 | ( sk(k,j,i+1) + sk(k,j,i) ) & |
|---|
| 529 | - ( 8.0 * ibit2 * adv_sca_5 & |
|---|
| 530 | + ibit1 * adv_sca_3 & |
|---|
| 531 | ) * & |
|---|
| 532 | ( sk(k,j,i+2) + sk(k,j,i-1) ) & |
|---|
| 533 | + ( ibit2 * adv_sca_5 & |
|---|
| 534 | ) * & |
|---|
| 535 | ( sk(k,j,i+3) + sk(k,j,i-2) ) & |
|---|
| 536 | ) |
|---|
| 537 | |
|---|
| 538 | diss_r(k) = -ABS( u_comp ) * ( & |
|---|
| 539 | ( 10.0 * ibit2 * adv_sca_5 & |
|---|
| 540 | + 3.0 * ibit1 * adv_sca_3 & |
|---|
| 541 | + ibit0 * adv_sca_1 & |
|---|
| 542 | ) * & |
|---|
| 543 | ( sk(k,j,i+1) - sk(k,j,i) ) & |
|---|
| 544 | - ( 5.0 * ibit2 * adv_sca_5 & |
|---|
| 545 | + ibit1 * adv_sca_3 & |
|---|
| 546 | ) * & |
|---|
| 547 | ( sk(k,j,i+2) - sk(k,j,i-1) ) & |
|---|
| 548 | + ( ibit2 * adv_sca_5 & |
|---|
| 549 | ) * & |
|---|
| 550 | ( sk(k,j,i+3) - sk(k,j,i-2) ) & |
|---|
| 551 | ) |
|---|
| 552 | |
|---|
| 553 | ibit5 = IBITS(wall_flags_0(k,j,i),5,1) |
|---|
| 554 | ibit4 = IBITS(wall_flags_0(k,j,i),4,1) |
|---|
| 555 | ibit3 = IBITS(wall_flags_0(k,j,i),3,1) |
|---|
| 556 | |
|---|
| 557 | v_comp = v(k,j+1,i) - v_gtrans |
|---|
| 558 | flux_n(k) = v_comp * ( & |
|---|
| 559 | ( 37.0 * ibit5 * adv_sca_5 & |
|---|
| 560 | + 7.0 * ibit4 * adv_sca_3 & |
|---|
| 561 | + ibit3 * adv_sca_1 & |
|---|
| 562 | ) * & |
|---|
| 563 | ( sk(k,j+1,i) + sk(k,j,i) ) & |
|---|
| 564 | - ( 8.0 * ibit5 * adv_sca_5 & |
|---|
| 565 | + ibit4 * adv_sca_3 & |
|---|
| 566 | ) * & |
|---|
| 567 | ( sk(k,j+2,i) + sk(k,j-1,i) ) & |
|---|
| 568 | + ( ibit5 * adv_sca_5 & |
|---|
| 569 | ) * & |
|---|
| 570 | ( sk(k,j+3,i) + sk(k,j-2,i) ) & |
|---|
| 571 | ) |
|---|
| 572 | |
|---|
| 573 | diss_n(k) = -ABS( v_comp ) * ( & |
|---|
| 574 | ( 10.0 * ibit5 * adv_sca_5 & |
|---|
| 575 | + 3.0 * ibit4 * adv_sca_3 & |
|---|
| 576 | + ibit3 * adv_sca_1 & |
|---|
| 577 | ) * & |
|---|
| 578 | ( sk(k,j+1,i) - sk(k,j,i) ) & |
|---|
| 579 | - ( 5.0 * ibit5 * adv_sca_5 & |
|---|
| 580 | + ibit4 * adv_sca_3 & |
|---|
| 581 | ) * & |
|---|
| 582 | ( sk(k,j+2,i) - sk(k,j-1,i) ) & |
|---|
| 583 | + ( ibit5 * adv_sca_5 & |
|---|
| 584 | ) * & |
|---|
| 585 | ( sk(k,j+3,i) - sk(k,j-2,i) ) & |
|---|
| 586 | ) |
|---|
| 587 | ! |
|---|
| 588 | !-- k index has to be modified near bottom and top, else array |
|---|
| 589 | !-- subscripts will be exceeded. |
|---|
| 590 | ibit8 = IBITS(wall_flags_0(k,j,i),8,1) |
|---|
| 591 | ibit7 = IBITS(wall_flags_0(k,j,i),7,1) |
|---|
| 592 | ibit6 = IBITS(wall_flags_0(k,j,i),6,1) |
|---|
| 593 | |
|---|
| 594 | k_ppp = k + 3 * ibit8 |
|---|
| 595 | k_pp = k + 2 * ( 1 - ibit6 ) |
|---|
| 596 | k_mm = k - 2 * ibit8 |
|---|
| 597 | |
|---|
| 598 | |
|---|
| 599 | flux_t(k) = w(k,j,i) * ( & |
|---|
| 600 | ( 37.0 * ibit8 * adv_sca_5 & |
|---|
| 601 | + 7.0 * ibit7 * adv_sca_3 & |
|---|
| 602 | + ibit6 * adv_sca_1 & |
|---|
| 603 | ) * & |
|---|
| 604 | ( sk(k+1,j,i) + sk(k,j,i) ) & |
|---|
| 605 | - ( 8.0 * ibit8 * adv_sca_5 & |
|---|
| 606 | + ibit7 * adv_sca_3 & |
|---|
| 607 | ) * & |
|---|
| 608 | ( sk(k_pp,j,i) + sk(k-1,j,i) ) & |
|---|
| 609 | + ( ibit8 * adv_sca_5 & |
|---|
| 610 | ) * ( sk(k_ppp,j,i)+ sk(k_mm,j,i) ) & |
|---|
| 611 | ) |
|---|
| 612 | |
|---|
| 613 | diss_t(k) = -ABS( w(k,j,i) ) * ( & |
|---|
| 614 | ( 10.0 * ibit8 * adv_sca_5 & |
|---|
| 615 | + 3.0 * ibit7 * adv_sca_3 & |
|---|
| 616 | + ibit6 * adv_sca_1 & |
|---|
| 617 | ) * & |
|---|
| 618 | ( sk(k+1,j,i) - sk(k,j,i) ) & |
|---|
| 619 | - ( 5.0 * ibit8 * adv_sca_5 & |
|---|
| 620 | + ibit7 * adv_sca_3 & |
|---|
| 621 | ) * & |
|---|
| 622 | ( sk(k_pp,j,i) - sk(k-1,j,i) ) & |
|---|
| 623 | + ( ibit8 * adv_sca_5 & |
|---|
| 624 | ) * & |
|---|
| 625 | ( sk(k_ppp,j,i) - sk(k_mm,j,i) ) & |
|---|
| 626 | ) |
|---|
| 627 | ! |
|---|
| 628 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 629 | !-- correction is needed to overcome numerical instabilities caused |
|---|
| 630 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 631 | div = ( u(k,j,i+1) - u(k,j,i) ) * ddx & |
|---|
| 632 | + ( v(k,j+1,i) - v(k,j,i) ) * ddy & |
|---|
| 633 | + ( w(k,j,i) - w(k-1,j,i) ) * ddzw(k) |
|---|
| 634 | |
|---|
| 635 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 636 | ( flux_r(k) + diss_r(k) - swap_flux_x_local(k,j,tn) - & |
|---|
| 637 | swap_diss_x_local(k,j,tn) ) * ddx & |
|---|
| 638 | + ( flux_n(k) + diss_n(k) - swap_flux_y_local(k,tn) - & |
|---|
| 639 | swap_diss_y_local(k,tn) ) * ddy & |
|---|
| 640 | + ( flux_t(k) + diss_t(k) - flux_d - diss_d & |
|---|
| 641 | ) * ddzw(k) & |
|---|
| 642 | ) + sk(k,j,i) * div |
|---|
| 643 | |
|---|
| 644 | swap_flux_y_local(k,tn) = flux_n(k) |
|---|
| 645 | swap_diss_y_local(k,tn) = diss_n(k) |
|---|
| 646 | swap_flux_x_local(k,j,tn) = flux_r(k) |
|---|
| 647 | swap_diss_x_local(k,j,tn) = diss_r(k) |
|---|
| 648 | flux_d = flux_t(k) |
|---|
| 649 | diss_d = diss_t(k) |
|---|
| 650 | |
|---|
| 651 | ENDDO |
|---|
| 652 | ! |
|---|
| 653 | !-- Now compute the fluxes and tendency terms for the horizontal and |
|---|
| 654 | !-- vertical parts above the top of the highest topography. No degradation |
|---|
| 655 | !-- for the horizontal parts, but for the vertical it is stell needed. |
|---|
| 656 | DO k = nzb_max+1, nzt |
|---|
| 657 | |
|---|
| 658 | u_comp = u(k,j,i+1) - u_gtrans |
|---|
| 659 | flux_r(k) = u_comp * ( & |
|---|
| 660 | 37.0 * ( sk(k,j,i+1) + sk(k,j,i) ) & |
|---|
| 661 | - 8.0 * ( sk(k,j,i+2) + sk(k,j,i-1) ) & |
|---|
| 662 | + ( sk(k,j,i+3) + sk(k,j,i-2) ) ) * adv_sca_5 |
|---|
| 663 | diss_r(k) = -ABS( u_comp ) * ( & |
|---|
| 664 | 10.0 * ( sk(k,j,i+1) - sk(k,j,i) ) & |
|---|
| 665 | - 5.0 * ( sk(k,j,i+2) - sk(k,j,i-1) ) & |
|---|
| 666 | + ( sk(k,j,i+3) - sk(k,j,i-2) ) ) * adv_sca_5 |
|---|
| 667 | |
|---|
| 668 | v_comp = v(k,j+1,i) - v_gtrans |
|---|
| 669 | flux_n(k) = v_comp * ( & |
|---|
| 670 | 37.0 * ( sk(k,j+1,i) + sk(k,j,i) ) & |
|---|
| 671 | - 8.0 * ( sk(k,j+2,i) + sk(k,j-1,i) ) & |
|---|
| 672 | + ( sk(k,j+3,i) + sk(k,j-2,i) ) ) * adv_sca_5 |
|---|
| 673 | diss_n(k) = -ABS( v_comp ) * ( & |
|---|
| 674 | 10.0 * ( sk(k,j+1,i) - sk(k,j,i) ) & |
|---|
| 675 | - 5.0 * ( sk(k,j+2,i) - sk(k,j-1,i) ) & |
|---|
| 676 | + ( sk(k,j+3,i) - sk(k,j-2,i) ) ) * adv_sca_5 |
|---|
| 677 | ! |
|---|
| 678 | !-- k index has to be modified near bottom and top, else array |
|---|
| 679 | !-- subscripts will be exceeded. |
|---|
| 680 | ibit8 = IBITS(wall_flags_0(k,j,i),8,1) |
|---|
| 681 | ibit7 = IBITS(wall_flags_0(k,j,i),7,1) |
|---|
| 682 | ibit6 = IBITS(wall_flags_0(k,j,i),6,1) |
|---|
| 683 | |
|---|
| 684 | k_ppp = k + 3 * ibit8 |
|---|
| 685 | k_pp = k + 2 * ( 1 - ibit6 ) |
|---|
| 686 | k_mm = k - 2 * ibit8 |
|---|
| 687 | |
|---|
| 688 | |
|---|
| 689 | flux_t(k) = w(k,j,i) * ( & |
|---|
| 690 | ( 37.0 * ibit8 * adv_sca_5 & |
|---|
| 691 | + 7.0 * ibit7 * adv_sca_3 & |
|---|
| 692 | + ibit6 * adv_sca_1 & |
|---|
| 693 | ) * & |
|---|
| 694 | ( sk(k+1,j,i) + sk(k,j,i) ) & |
|---|
| 695 | - ( 8.0 * ibit8 * adv_sca_5 & |
|---|
| 696 | + ibit7 * adv_sca_3 & |
|---|
| 697 | ) * & |
|---|
| 698 | ( sk(k_pp,j,i) + sk(k-1,j,i) ) & |
|---|
| 699 | + ( ibit8 * adv_sca_5 & |
|---|
| 700 | ) * ( sk(k_ppp,j,i)+ sk(k_mm,j,i) ) & |
|---|
| 701 | ) |
|---|
| 702 | |
|---|
| 703 | diss_t(k) = -ABS( w(k,j,i) ) * ( & |
|---|
| 704 | ( 10.0 * ibit8 * adv_sca_5 & |
|---|
| 705 | + 3.0 * ibit7 * adv_sca_3 & |
|---|
| 706 | + ibit6 * adv_sca_1 & |
|---|
| 707 | ) * & |
|---|
| 708 | ( sk(k+1,j,i) - sk(k,j,i) ) & |
|---|
| 709 | - ( 5.0 * ibit8 * adv_sca_5 & |
|---|
| 710 | + ibit7 * adv_sca_3 & |
|---|
| 711 | ) * & |
|---|
| 712 | ( sk(k_pp,j,i) - sk(k-1,j,i) ) & |
|---|
| 713 | + ( ibit8 * adv_sca_5 & |
|---|
| 714 | ) * & |
|---|
| 715 | ( sk(k_ppp,j,i) - sk(k_mm,j,i) ) & |
|---|
| 716 | ) |
|---|
| 717 | ! |
|---|
| 718 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 719 | !-- correction is needed to overcome numerical instabilities introduced |
|---|
| 720 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 721 | div = ( u(k,j,i+1) - u(k,j,i) ) * ddx & |
|---|
| 722 | + ( v(k,j+1,i) - v(k,j,i) ) * ddy & |
|---|
| 723 | + ( w(k,j,i) - w(k-1,j,i) ) * ddzw(k) |
|---|
| 724 | |
|---|
| 725 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 726 | ( flux_r(k) + diss_r(k) - swap_flux_x_local(k,j,tn) - & |
|---|
| 727 | swap_diss_x_local(k,j,tn) ) * ddx & |
|---|
| 728 | + ( flux_n(k) + diss_n(k) - swap_flux_y_local(k,tn) - & |
|---|
| 729 | swap_diss_y_local(k,tn) ) * ddy & |
|---|
| 730 | + ( flux_t(k) + diss_t(k) - flux_d - diss_d & |
|---|
| 731 | ) * ddzw(k) & |
|---|
| 732 | ) + sk(k,j,i) * div |
|---|
| 733 | |
|---|
| 734 | swap_flux_y_local(k,tn) = flux_n(k) |
|---|
| 735 | swap_diss_y_local(k,tn) = diss_n(k) |
|---|
| 736 | swap_flux_x_local(k,j,tn) = flux_r(k) |
|---|
| 737 | swap_diss_x_local(k,j,tn) = diss_r(k) |
|---|
| 738 | flux_d = flux_t(k) |
|---|
| 739 | diss_d = diss_t(k) |
|---|
| 740 | |
|---|
| 741 | ENDDO |
|---|
| 742 | |
|---|
| 743 | ! |
|---|
| 744 | !-- Evaluation of statistics |
|---|
| 745 | SELECT CASE ( sk_char ) |
|---|
| 746 | |
|---|
| 747 | CASE ( 'pt' ) |
|---|
| 748 | |
|---|
| 749 | DO k = nzb, nzt |
|---|
| 750 | sums_wspts_ws_l(k,tn) = sums_wspts_ws_l(k,tn) + & |
|---|
| 751 | ( flux_t(k) + diss_t(k) ) & |
|---|
| 752 | * weight_substep(intermediate_timestep_count) |
|---|
| 753 | ENDDO |
|---|
| 754 | |
|---|
| 755 | CASE ( 'sa' ) |
|---|
| 756 | |
|---|
| 757 | DO k = nzb, nzt |
|---|
| 758 | sums_wssas_ws_l(k,tn) = sums_wssas_ws_l(k,tn) + & |
|---|
| 759 | ( flux_t(k) + diss_t(k) ) & |
|---|
| 760 | * weight_substep(intermediate_timestep_count) |
|---|
| 761 | ENDDO |
|---|
| 762 | |
|---|
| 763 | CASE ( 'q' ) |
|---|
| 764 | |
|---|
| 765 | DO k = nzb, nzt |
|---|
| 766 | sums_wsqs_ws_l(k,tn) = sums_wsqs_ws_l(k,tn) + & |
|---|
| 767 | ( flux_t(k) + diss_t(k) ) & |
|---|
| 768 | * weight_substep(intermediate_timestep_count) |
|---|
| 769 | ENDDO |
|---|
| 770 | |
|---|
| 771 | CASE ( 'qr' ) |
|---|
| 772 | |
|---|
| 773 | DO k = nzb, nzt |
|---|
| 774 | sums_wsqrs_ws_l(k,tn) = sums_wsqrs_ws_l(k,tn) + & |
|---|
| 775 | ( flux_t(k) + diss_t(k) ) & |
|---|
| 776 | * weight_substep(intermediate_timestep_count) |
|---|
| 777 | ENDDO |
|---|
| 778 | |
|---|
| 779 | CASE ( 'nr' ) |
|---|
| 780 | |
|---|
| 781 | DO k = nzb, nzt |
|---|
| 782 | sums_wsnrs_ws_l(k,tn) = sums_wsnrs_ws_l(k,tn) + & |
|---|
| 783 | ( flux_t(k) + diss_t(k) ) & |
|---|
| 784 | * weight_substep(intermediate_timestep_count) |
|---|
| 785 | ENDDO |
|---|
| 786 | |
|---|
| 787 | END SELECT |
|---|
| 788 | |
|---|
| 789 | END SUBROUTINE advec_s_ws_ij |
|---|
| 790 | |
|---|
| 791 | |
|---|
| 792 | |
|---|
| 793 | |
|---|
| 794 | !------------------------------------------------------------------------------! |
|---|
| 795 | ! Advection of u-component - Call for grid point i,j |
|---|
| 796 | !------------------------------------------------------------------------------! |
|---|
| 797 | SUBROUTINE advec_u_ws_ij( i, j, i_omp, tn ) |
|---|
| 798 | |
|---|
| 799 | USE arrays_3d |
|---|
| 800 | USE constants |
|---|
| 801 | USE control_parameters |
|---|
| 802 | USE grid_variables |
|---|
| 803 | USE indices |
|---|
| 804 | USE statistics |
|---|
| 805 | |
|---|
| 806 | IMPLICIT NONE |
|---|
| 807 | |
|---|
| 808 | INTEGER :: i, ibit9, ibit10, ibit11, ibit12, ibit13, ibit14, ibit15, & |
|---|
| 809 | ibit16, ibit17, i_omp, j, k, k_mm, k_pp, k_ppp, tn |
|---|
| 810 | REAL :: diss_d, div, flux_d, gu, gv, u_comp_l, v_comp, w_comp |
|---|
| 811 | REAL, DIMENSION(nzb:nzt+1) :: diss_n, diss_r, diss_t, flux_n, flux_r, & |
|---|
| 812 | flux_t, u_comp |
|---|
| 813 | |
|---|
| 814 | gu = 2.0 * u_gtrans |
|---|
| 815 | gv = 2.0 * v_gtrans |
|---|
| 816 | ! |
|---|
| 817 | !-- Compute southside fluxes for the respective boundary of PE |
|---|
| 818 | IF ( j == nys ) THEN |
|---|
| 819 | |
|---|
| 820 | DO k = nzb+1, nzb_max |
|---|
| 821 | |
|---|
| 822 | ibit14 = IBITS(wall_flags_0(k,j,i),14,1) |
|---|
| 823 | ibit13 = IBITS(wall_flags_0(k,j,i),13,1) |
|---|
| 824 | ibit12 = IBITS(wall_flags_0(k,j,i),12,1) |
|---|
| 825 | |
|---|
| 826 | v_comp = v(k,j,i) + v(k,j,i-1) - gv |
|---|
| 827 | flux_s_u(k,tn) = v_comp * ( & |
|---|
| 828 | ( 37.0 * ibit14 * adv_mom_5 & |
|---|
| 829 | + 7.0 * ibit13 * adv_mom_3 & |
|---|
| 830 | + ibit12 * adv_mom_1 & |
|---|
| 831 | ) * & |
|---|
| 832 | ( u(k,j,i) + u(k,j-1,i) ) & |
|---|
| 833 | - ( 8.0 * ibit14 * adv_mom_5 & |
|---|
| 834 | + ibit13 * adv_mom_3 & |
|---|
| 835 | ) * & |
|---|
| 836 | ( u(k,j+1,i) + u(k,j-2,i) ) & |
|---|
| 837 | + ( ibit14 * adv_mom_5 & |
|---|
| 838 | ) * & |
|---|
| 839 | ( u(k,j+2,i) + u(k,j-3,i) ) & |
|---|
| 840 | ) |
|---|
| 841 | |
|---|
| 842 | diss_s_u(k,tn) = - ABS ( v_comp ) * ( & |
|---|
| 843 | ( 10.0 * ibit14 * adv_mom_5 & |
|---|
| 844 | + 3.0 * ibit13 * adv_mom_3 & |
|---|
| 845 | + ibit12 * adv_mom_1 & |
|---|
| 846 | ) * & |
|---|
| 847 | ( u(k,j,i) - u(k,j-1,i) ) & |
|---|
| 848 | - ( 5.0 * ibit14 * adv_mom_5 & |
|---|
| 849 | + ibit13 * adv_mom_3 & |
|---|
| 850 | ) * & |
|---|
| 851 | ( u(k,j+1,i) - u(k,j-2,i) ) & |
|---|
| 852 | + ( ibit14 * adv_mom_5 & |
|---|
| 853 | ) * & |
|---|
| 854 | ( u(k,j+2,i) - u(k,j-3,i) ) & |
|---|
| 855 | ) |
|---|
| 856 | |
|---|
| 857 | ENDDO |
|---|
| 858 | |
|---|
| 859 | DO k = nzb_max+1, nzt |
|---|
| 860 | |
|---|
| 861 | v_comp = v(k,j,i) + v(k,j,i-1) - gv |
|---|
| 862 | flux_s_u(k,tn) = v_comp * ( & |
|---|
| 863 | 37.0 * ( u(k,j,i) + u(k,j-1,i) ) & |
|---|
| 864 | - 8.0 * ( u(k,j+1,i) + u(k,j-2,i) ) & |
|---|
| 865 | + ( u(k,j+2,i) + u(k,j-3,i) ) ) * adv_mom_5 |
|---|
| 866 | diss_s_u(k,tn) = - ABS(v_comp) * ( & |
|---|
| 867 | 10.0 * ( u(k,j,i) - u(k,j-1,i) ) & |
|---|
| 868 | - 5.0 * ( u(k,j+1,i) - u(k,j-2,i) ) & |
|---|
| 869 | + ( u(k,j+2,i) - u(k,j-3,i) ) ) * adv_mom_5 |
|---|
| 870 | |
|---|
| 871 | ENDDO |
|---|
| 872 | |
|---|
| 873 | ENDIF |
|---|
| 874 | ! |
|---|
| 875 | !-- Compute leftside fluxes for the respective boundary of PE |
|---|
| 876 | IF ( i == i_omp ) THEN |
|---|
| 877 | |
|---|
| 878 | DO k = nzb+1, nzb_max |
|---|
| 879 | |
|---|
| 880 | ibit11 = IBITS(wall_flags_0(k,j,i),11,1) |
|---|
| 881 | ibit10 = IBITS(wall_flags_0(k,j,i),10,1) |
|---|
| 882 | ibit9 = IBITS(wall_flags_0(k,j,i),9,1) |
|---|
| 883 | |
|---|
| 884 | u_comp_l = u(k,j,i) + u(k,j,i-1) - gu |
|---|
| 885 | flux_l_u(k,j,tn) = u_comp_l * ( & |
|---|
| 886 | ( 37.0 * ibit11 * adv_mom_5 & |
|---|
| 887 | + 7.0 * ibit10 * adv_mom_3 & |
|---|
| 888 | + ibit9 * adv_mom_1 & |
|---|
| 889 | ) * & |
|---|
| 890 | ( u(k,j,i) + u(k,j,i-1) ) & |
|---|
| 891 | - ( 8.0 * ibit11 * adv_mom_5 & |
|---|
| 892 | + ibit10 * adv_mom_3 & |
|---|
| 893 | ) * & |
|---|
| 894 | ( u(k,j,i+1) + u(k,j,i-2) ) & |
|---|
| 895 | + ( ibit11 * adv_mom_5 & |
|---|
| 896 | ) * & |
|---|
| 897 | ( u(k,j,i+2) + u(k,j,i-3) ) & |
|---|
| 898 | ) |
|---|
| 899 | |
|---|
| 900 | diss_l_u(k,j,tn) = - ABS( u_comp_l ) * ( & |
|---|
| 901 | ( 10.0 * ibit11 * adv_mom_5 & |
|---|
| 902 | + 3.0 * ibit10 * adv_mom_3 & |
|---|
| 903 | + ibit9 * adv_mom_1 & |
|---|
| 904 | ) * & |
|---|
| 905 | ( u(k,j,i) - u(k,j,i-1) ) & |
|---|
| 906 | - ( 5.0 * ibit11 * adv_mom_5 & |
|---|
| 907 | + ibit10 * adv_mom_3 & |
|---|
| 908 | ) * & |
|---|
| 909 | ( u(k,j,i+1) - u(k,j,i-2) ) & |
|---|
| 910 | + ( ibit11 * adv_mom_5 & |
|---|
| 911 | ) * & |
|---|
| 912 | ( u(k,j,i+2) - u(k,j,i-3) ) & |
|---|
| 913 | ) |
|---|
| 914 | |
|---|
| 915 | ENDDO |
|---|
| 916 | |
|---|
| 917 | DO k = nzb_max+1, nzt |
|---|
| 918 | |
|---|
| 919 | u_comp_l = u(k,j,i) + u(k,j,i-1) - gu |
|---|
| 920 | flux_l_u(k,j,tn) = u_comp_l * ( & |
|---|
| 921 | 37.0 * ( u(k,j,i) + u(k,j,i-1) ) & |
|---|
| 922 | - 8.0 * ( u(k,j,i+1) + u(k,j,i-2) ) & |
|---|
| 923 | + ( u(k,j,i+2) + u(k,j,i-3) ) ) * adv_mom_5 |
|---|
| 924 | diss_l_u(k,j,tn) = - ABS(u_comp_l) * ( & |
|---|
| 925 | 10.0 * ( u(k,j,i) - u(k,j,i-1) ) & |
|---|
| 926 | - 5.0 * ( u(k,j,i+1) - u(k,j,i-2) ) & |
|---|
| 927 | + ( u(k,j,i+2) - u(k,j,i-3) ) ) * adv_mom_5 |
|---|
| 928 | |
|---|
| 929 | ENDDO |
|---|
| 930 | |
|---|
| 931 | ENDIF |
|---|
| 932 | |
|---|
| 933 | flux_t(0) = 0.0 |
|---|
| 934 | diss_t(0) = 0.0 |
|---|
| 935 | flux_d = 0.0 |
|---|
| 936 | diss_d = 0.0 |
|---|
| 937 | ! |
|---|
| 938 | !-- Now compute the fluxes tendency terms for the horizontal and |
|---|
| 939 | !-- vertical parts. |
|---|
| 940 | DO k = nzb+1, nzb_max |
|---|
| 941 | |
|---|
| 942 | ibit11 = IBITS(wall_flags_0(k,j,i),11,1) |
|---|
| 943 | ibit10 = IBITS(wall_flags_0(k,j,i),10,1) |
|---|
| 944 | ibit9 = IBITS(wall_flags_0(k,j,i),9,1) |
|---|
| 945 | |
|---|
| 946 | u_comp(k) = u(k,j,i+1) + u(k,j,i) |
|---|
| 947 | flux_r(k) = ( u_comp(k) - gu ) * ( & |
|---|
| 948 | ( 37.0 * ibit11 * adv_mom_5 & |
|---|
| 949 | + 7.0 * ibit10 * adv_mom_3 & |
|---|
| 950 | + ibit9 * adv_mom_1 & |
|---|
| 951 | ) * & |
|---|
| 952 | ( u(k,j,i+1) + u(k,j,i) ) & |
|---|
| 953 | - ( 8.0 * ibit11 * adv_mom_5 & |
|---|
| 954 | + ibit10 * adv_mom_3 & |
|---|
| 955 | ) * & |
|---|
| 956 | ( u(k,j,i+2) + u(k,j,i-1) ) & |
|---|
| 957 | + ( ibit11 * adv_mom_5 & |
|---|
| 958 | ) * & |
|---|
| 959 | ( u(k,j,i+3) + u(k,j,i-2) ) & |
|---|
| 960 | ) |
|---|
| 961 | |
|---|
| 962 | diss_r(k) = - ABS( u_comp(k) - gu ) * ( & |
|---|
| 963 | ( 10.0 * ibit11 * adv_mom_5 & |
|---|
| 964 | + 3.0 * ibit10 * adv_mom_3 & |
|---|
| 965 | + ibit9 * adv_mom_1 & |
|---|
| 966 | ) * & |
|---|
| 967 | ( u(k,j,i+1) - u(k,j,i) ) & |
|---|
| 968 | - ( 5.0 * ibit11 * adv_mom_5 & |
|---|
| 969 | + ibit10 * adv_mom_3 & |
|---|
| 970 | ) * & |
|---|
| 971 | ( u(k,j,i+2) - u(k,j,i-1) ) & |
|---|
| 972 | + ( ibit11 * adv_mom_5 & |
|---|
| 973 | ) * & |
|---|
| 974 | ( u(k,j,i+3) - u(k,j,i-2) ) & |
|---|
| 975 | ) |
|---|
| 976 | |
|---|
| 977 | ibit14 = IBITS(wall_flags_0(k,j,i),14,1) |
|---|
| 978 | ibit13 = IBITS(wall_flags_0(k,j,i),13,1) |
|---|
| 979 | ibit12 = IBITS(wall_flags_0(k,j,i),12,1) |
|---|
| 980 | |
|---|
| 981 | v_comp = v(k,j+1,i) + v(k,j+1,i-1) - gv |
|---|
| 982 | flux_n(k) = v_comp * ( & |
|---|
| 983 | ( 37.0 * ibit14 * adv_mom_5 & |
|---|
| 984 | + 7.0 * ibit13 * adv_mom_3 & |
|---|
| 985 | + ibit12 * adv_mom_1 & |
|---|
| 986 | ) * & |
|---|
| 987 | ( u(k,j+1,i) + u(k,j,i) ) & |
|---|
| 988 | - ( 8.0 * ibit14 * adv_mom_5 & |
|---|
| 989 | + ibit13 * adv_mom_3 & |
|---|
| 990 | ) * & |
|---|
| 991 | ( u(k,j+2,i) + u(k,j-1,i) ) & |
|---|
| 992 | + ( ibit14 * adv_mom_5 & |
|---|
| 993 | ) * & |
|---|
| 994 | ( u(k,j+3,i) + u(k,j-2,i) ) & |
|---|
| 995 | ) |
|---|
| 996 | |
|---|
| 997 | diss_n(k) = - ABS ( v_comp ) * ( & |
|---|
| 998 | ( 10.0 * ibit14 * adv_mom_5 & |
|---|
| 999 | + 3.0 * ibit13 * adv_mom_3 & |
|---|
| 1000 | + ibit12 * adv_mom_1 & |
|---|
| 1001 | ) * & |
|---|
| 1002 | ( u(k,j+1,i) - u(k,j,i) ) & |
|---|
| 1003 | - ( 5.0 * ibit14 * adv_mom_5 & |
|---|
| 1004 | + ibit13 * adv_mom_3 & |
|---|
| 1005 | ) * & |
|---|
| 1006 | ( u(k,j+2,i) - u(k,j-1,i) ) & |
|---|
| 1007 | + ( ibit14 * adv_mom_5 & |
|---|
| 1008 | ) * & |
|---|
| 1009 | ( u(k,j+3,i) - u(k,j-2,i) ) & |
|---|
| 1010 | ) |
|---|
| 1011 | ! |
|---|
| 1012 | !-- k index has to be modified near bottom and top, else array |
|---|
| 1013 | !-- subscripts will be exceeded. |
|---|
| 1014 | ibit17 = IBITS(wall_flags_0(k,j,i),17,1) |
|---|
| 1015 | ibit16 = IBITS(wall_flags_0(k,j,i),16,1) |
|---|
| 1016 | ibit15 = IBITS(wall_flags_0(k,j,i),15,1) |
|---|
| 1017 | |
|---|
| 1018 | k_ppp = k + 3 * ibit17 |
|---|
| 1019 | k_pp = k + 2 * ( 1 - ibit15 ) |
|---|
| 1020 | k_mm = k - 2 * ibit17 |
|---|
| 1021 | |
|---|
| 1022 | w_comp = w(k,j,i) + w(k,j,i-1) |
|---|
| 1023 | flux_t(k) = w_comp * ( & |
|---|
| 1024 | ( 37.0 * ibit17 * adv_mom_5 & |
|---|
| 1025 | + 7.0 * ibit16 * adv_mom_3 & |
|---|
| 1026 | + ibit15 * adv_mom_1 & |
|---|
| 1027 | ) * & |
|---|
| 1028 | ( u(k+1,j,i) + u(k,j,i) ) & |
|---|
| 1029 | - ( 8.0 * ibit17 * adv_mom_5 & |
|---|
| 1030 | + ibit16 * adv_mom_3 & |
|---|
| 1031 | ) * & |
|---|
| 1032 | ( u(k_pp,j,i) + u(k-1,j,i) ) & |
|---|
| 1033 | + ( ibit17 * adv_mom_5 & |
|---|
| 1034 | ) * & |
|---|
| 1035 | ( u(k_ppp,j,i) + u(k_mm,j,i) ) & |
|---|
| 1036 | ) |
|---|
| 1037 | |
|---|
| 1038 | diss_t(k) = - ABS( w_comp ) * ( & |
|---|
| 1039 | ( 10.0 * ibit17 * adv_mom_5 & |
|---|
| 1040 | + 3.0 * ibit16 * adv_mom_3 & |
|---|
| 1041 | + ibit15 * adv_mom_1 & |
|---|
| 1042 | ) * & |
|---|
| 1043 | ( u(k+1,j,i) - u(k,j,i) ) & |
|---|
| 1044 | - ( 5.0 * ibit17 * adv_mom_5 & |
|---|
| 1045 | + ibit16 * adv_mom_3 & |
|---|
| 1046 | ) * & |
|---|
| 1047 | ( u(k_pp,j,i) - u(k-1,j,i) ) & |
|---|
| 1048 | + ( ibit17 * adv_mom_5 & |
|---|
| 1049 | ) * & |
|---|
| 1050 | ( u(k_ppp,j,i) - u(k_mm,j,i) ) & |
|---|
| 1051 | ) |
|---|
| 1052 | ! |
|---|
| 1053 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 1054 | !-- correction is needed to overcome numerical instabilities introduced |
|---|
| 1055 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 1056 | div = ( ( u_comp(k) - ( u(k,j,i) + u(k,j,i-1) ) ) * ddx & |
|---|
| 1057 | + ( v_comp + gv - ( v(k,j,i) + v(k,j,i-1 ) ) ) * ddy & |
|---|
| 1058 | + ( w_comp - ( w(k-1,j,i) + w(k-1,j,i-1) ) ) * ddzw(k) & |
|---|
| 1059 | ) * 0.5 |
|---|
| 1060 | |
|---|
| 1061 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 1062 | ( flux_r(k) + diss_r(k) & |
|---|
| 1063 | - flux_l_u(k,j,tn) - diss_l_u(k,j,tn) ) * ddx & |
|---|
| 1064 | + ( flux_n(k) + diss_n(k) & |
|---|
| 1065 | - flux_s_u(k,tn) - diss_s_u(k,tn) ) * ddy & |
|---|
| 1066 | + ( flux_t(k) + diss_t(k) & |
|---|
| 1067 | - flux_d - diss_d ) * ddzw(k) & |
|---|
| 1068 | ) + div * u(k,j,i) |
|---|
| 1069 | |
|---|
| 1070 | flux_l_u(k,j,tn) = flux_r(k) |
|---|
| 1071 | diss_l_u(k,j,tn) = diss_r(k) |
|---|
| 1072 | flux_s_u(k,tn) = flux_n(k) |
|---|
| 1073 | diss_s_u(k,tn) = diss_n(k) |
|---|
| 1074 | flux_d = flux_t(k) |
|---|
| 1075 | diss_d = diss_t(k) |
|---|
| 1076 | ! |
|---|
| 1077 | !-- Statistical Evaluation of u'u'. The factor has to be applied for |
|---|
| 1078 | !-- right evaluation when gallilei_trans = .T. . |
|---|
| 1079 | sums_us2_ws_l(k,tn) = sums_us2_ws_l(k,tn) & |
|---|
| 1080 | + ( flux_r(k) * & |
|---|
| 1081 | ( u_comp(k) - 2.0 * hom(k,1,1,0) ) & |
|---|
| 1082 | / ( u_comp(k) - gu + 1.0E-20 ) & |
|---|
| 1083 | + diss_r(k) * & |
|---|
| 1084 | ABS( u_comp(k) - 2.0 * hom(k,1,1,0) ) & |
|---|
| 1085 | / ( ABS( u_comp(k) - gu ) + 1.0E-20 ) ) & |
|---|
| 1086 | * weight_substep(intermediate_timestep_count) |
|---|
| 1087 | ! |
|---|
| 1088 | !-- Statistical Evaluation of w'u'. |
|---|
| 1089 | sums_wsus_ws_l(k,tn) = sums_wsus_ws_l(k,tn) & |
|---|
| 1090 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 1091 | * weight_substep(intermediate_timestep_count) |
|---|
| 1092 | ENDDO |
|---|
| 1093 | |
|---|
| 1094 | DO k = nzb_max+1, nzt |
|---|
| 1095 | |
|---|
| 1096 | u_comp(k) = u(k,j,i+1) + u(k,j,i) |
|---|
| 1097 | flux_r(k) = ( u_comp(k) - gu ) * ( & |
|---|
| 1098 | 37.0 * ( u(k,j,i+1) + u(k,j,i) ) & |
|---|
| 1099 | - 8.0 * ( u(k,j,i+2) + u(k,j,i-1) ) & |
|---|
| 1100 | + ( u(k,j,i+3) + u(k,j,i-2) ) ) * adv_mom_5 |
|---|
| 1101 | diss_r(k) = - ABS( u_comp(k) - gu ) * ( & |
|---|
| 1102 | 10.0 * ( u(k,j,i+1) - u(k,j,i) ) & |
|---|
| 1103 | - 5.0 * ( u(k,j,i+2) - u(k,j,i-1) ) & |
|---|
| 1104 | + ( u(k,j,i+3) - u(k,j,i-2) ) ) * adv_mom_5 |
|---|
| 1105 | |
|---|
| 1106 | v_comp = v(k,j+1,i) + v(k,j+1,i-1) - gv |
|---|
| 1107 | flux_n(k) = v_comp * ( & |
|---|
| 1108 | 37.0 * ( u(k,j+1,i) + u(k,j,i) ) & |
|---|
| 1109 | - 8.0 * ( u(k,j+2,i) + u(k,j-1,i) ) & |
|---|
| 1110 | + ( u(k,j+3,i) + u(k,j-2,i) ) ) * adv_mom_5 |
|---|
| 1111 | diss_n(k) = - ABS( v_comp ) * ( & |
|---|
| 1112 | 10.0 * ( u(k,j+1,i) - u(k,j,i) ) & |
|---|
| 1113 | - 5.0 * ( u(k,j+2,i) - u(k,j-1,i) ) & |
|---|
| 1114 | + ( u(k,j+3,i) - u(k,j-2,i) ) ) * adv_mom_5 |
|---|
| 1115 | ! |
|---|
| 1116 | !-- k index has to be modified near bottom and top, else array |
|---|
| 1117 | !-- subscripts will be exceeded. |
|---|
| 1118 | ibit17 = IBITS(wall_flags_0(k,j,i),17,1) |
|---|
| 1119 | ibit16 = IBITS(wall_flags_0(k,j,i),16,1) |
|---|
| 1120 | ibit15 = IBITS(wall_flags_0(k,j,i),15,1) |
|---|
| 1121 | |
|---|
| 1122 | k_ppp = k + 3 * ibit17 |
|---|
| 1123 | k_pp = k + 2 * ( 1 - ibit15 ) |
|---|
| 1124 | k_mm = k - 2 * ibit17 |
|---|
| 1125 | |
|---|
| 1126 | w_comp = w(k,j,i) + w(k,j,i-1) |
|---|
| 1127 | flux_t(k) = w_comp * ( & |
|---|
| 1128 | ( 37.0 * ibit17 * adv_mom_5 & |
|---|
| 1129 | + 7.0 * ibit16 * adv_mom_3 & |
|---|
| 1130 | + ibit15 * adv_mom_1 & |
|---|
| 1131 | ) * & |
|---|
| 1132 | ( u(k+1,j,i) + u(k,j,i) ) & |
|---|
| 1133 | - ( 8.0 * ibit17 * adv_mom_5 & |
|---|
| 1134 | + ibit16 * adv_mom_3 & |
|---|
| 1135 | ) * & |
|---|
| 1136 | ( u(k_pp,j,i) + u(k-1,j,i) ) & |
|---|
| 1137 | + ( ibit17 * adv_mom_5 & |
|---|
| 1138 | ) * & |
|---|
| 1139 | ( u(k_ppp,j,i) + u(k_mm,j,i) ) & |
|---|
| 1140 | ) |
|---|
| 1141 | |
|---|
| 1142 | diss_t(k) = - ABS( w_comp ) * ( & |
|---|
| 1143 | ( 10.0 * ibit17 * adv_mom_5 & |
|---|
| 1144 | + 3.0 * ibit16 * adv_mom_3 & |
|---|
| 1145 | + ibit15 * adv_mom_1 & |
|---|
| 1146 | ) * & |
|---|
| 1147 | ( u(k+1,j,i) - u(k,j,i) ) & |
|---|
| 1148 | - ( 5.0 * ibit17 * adv_mom_5 & |
|---|
| 1149 | + ibit16 * adv_mom_3 & |
|---|
| 1150 | ) * & |
|---|
| 1151 | ( u(k_pp,j,i) - u(k-1,j,i) ) & |
|---|
| 1152 | + ( ibit17 * adv_mom_5 & |
|---|
| 1153 | ) * & |
|---|
| 1154 | ( u(k_ppp,j,i) - u(k_mm,j,i) ) & |
|---|
| 1155 | ) |
|---|
| 1156 | ! |
|---|
| 1157 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 1158 | !-- correction is needed to overcome numerical instabilities introduced |
|---|
| 1159 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 1160 | div = ( ( u_comp(k) - ( u(k,j,i) + u(k,j,i-1) ) ) * ddx & |
|---|
| 1161 | + ( v_comp + gv - ( v(k,j,i) + v(k,j,i-1 ) ) ) * ddy & |
|---|
| 1162 | + ( w_comp - ( w(k-1,j,i) + w(k-1,j,i-1) ) ) * ddzw(k) & |
|---|
| 1163 | ) * 0.5 |
|---|
| 1164 | |
|---|
| 1165 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 1166 | ( flux_r(k) + diss_r(k) & |
|---|
| 1167 | - flux_l_u(k,j,tn) - diss_l_u(k,j,tn) ) * ddx & |
|---|
| 1168 | + ( flux_n(k) + diss_n(k) & |
|---|
| 1169 | - flux_s_u(k,tn) - diss_s_u(k,tn) ) * ddy & |
|---|
| 1170 | + ( flux_t(k) + diss_t(k) & |
|---|
| 1171 | - flux_d - diss_d ) * ddzw(k) & |
|---|
| 1172 | ) + div * u(k,j,i) |
|---|
| 1173 | |
|---|
| 1174 | flux_l_u(k,j,tn) = flux_r(k) |
|---|
| 1175 | diss_l_u(k,j,tn) = diss_r(k) |
|---|
| 1176 | flux_s_u(k,tn) = flux_n(k) |
|---|
| 1177 | diss_s_u(k,tn) = diss_n(k) |
|---|
| 1178 | flux_d = flux_t(k) |
|---|
| 1179 | diss_d = diss_t(k) |
|---|
| 1180 | ! |
|---|
| 1181 | !-- Statistical Evaluation of u'u'. The factor has to be applied for |
|---|
| 1182 | !-- right evaluation when gallilei_trans = .T. . |
|---|
| 1183 | sums_us2_ws_l(k,tn) = sums_us2_ws_l(k,tn) & |
|---|
| 1184 | + ( flux_r(k) * & |
|---|
| 1185 | ( u_comp(k) - 2.0 * hom(k,1,1,0) ) & |
|---|
| 1186 | / ( u_comp(k) - gu + 1.0E-20 ) & |
|---|
| 1187 | + diss_r(k) * & |
|---|
| 1188 | ABS( u_comp(k) - 2.0 * hom(k,1,1,0) ) & |
|---|
| 1189 | / ( ABS( u_comp(k) - gu ) + 1.0E-20 ) ) & |
|---|
| 1190 | * weight_substep(intermediate_timestep_count) |
|---|
| 1191 | ! |
|---|
| 1192 | !-- Statistical Evaluation of w'u'. |
|---|
| 1193 | sums_wsus_ws_l(k,tn) = sums_wsus_ws_l(k,tn) & |
|---|
| 1194 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 1195 | * weight_substep(intermediate_timestep_count) |
|---|
| 1196 | ENDDO |
|---|
| 1197 | |
|---|
| 1198 | sums_us2_ws_l(nzb,tn) = sums_us2_ws_l(nzb+1,tn) |
|---|
| 1199 | |
|---|
| 1200 | |
|---|
| 1201 | |
|---|
| 1202 | END SUBROUTINE advec_u_ws_ij |
|---|
| 1203 | |
|---|
| 1204 | |
|---|
| 1205 | |
|---|
| 1206 | !-----------------------------------------------------------------------------! |
|---|
| 1207 | ! Advection of v-component - Call for grid point i,j |
|---|
| 1208 | !-----------------------------------------------------------------------------! |
|---|
| 1209 | SUBROUTINE advec_v_ws_ij( i, j, i_omp, tn ) |
|---|
| 1210 | |
|---|
| 1211 | USE arrays_3d |
|---|
| 1212 | USE constants |
|---|
| 1213 | USE control_parameters |
|---|
| 1214 | USE grid_variables |
|---|
| 1215 | USE indices |
|---|
| 1216 | USE statistics |
|---|
| 1217 | |
|---|
| 1218 | IMPLICIT NONE |
|---|
| 1219 | |
|---|
| 1220 | INTEGER :: i, ibit18, ibit19, ibit20, ibit21, ibit22, ibit23, ibit24, & |
|---|
| 1221 | ibit25, ibit26, i_omp, j, k, k_mm, k_pp, k_ppp, tn |
|---|
| 1222 | REAL :: diss_d, div, flux_d, gu, gv, u_comp, v_comp_l, w_comp |
|---|
| 1223 | REAL, DIMENSION(nzb:nzt+1) :: diss_n, diss_r, diss_t, flux_n, flux_r, & |
|---|
| 1224 | flux_t, v_comp |
|---|
| 1225 | |
|---|
| 1226 | gu = 2.0 * u_gtrans |
|---|
| 1227 | gv = 2.0 * v_gtrans |
|---|
| 1228 | |
|---|
| 1229 | ! |
|---|
| 1230 | !-- Compute leftside fluxes for the respective boundary. |
|---|
| 1231 | IF ( i == i_omp ) THEN |
|---|
| 1232 | |
|---|
| 1233 | DO k = nzb+1, nzb_max |
|---|
| 1234 | |
|---|
| 1235 | ibit20 = IBITS(wall_flags_0(k,j,i),20,1) |
|---|
| 1236 | ibit19 = IBITS(wall_flags_0(k,j,i),19,1) |
|---|
| 1237 | ibit18 = IBITS(wall_flags_0(k,j,i),18,1) |
|---|
| 1238 | |
|---|
| 1239 | u_comp = u(k,j-1,i) + u(k,j,i) - gu |
|---|
| 1240 | flux_l_v(k,j,tn) = u_comp * ( & |
|---|
| 1241 | ( 37.0 * ibit20 * adv_mom_5 & |
|---|
| 1242 | + 7.0 * ibit19 * adv_mom_3 & |
|---|
| 1243 | + ibit18 * adv_mom_1 & |
|---|
| 1244 | ) * & |
|---|
| 1245 | ( v(k,j,i) + v(k,j,i-1) ) & |
|---|
| 1246 | - ( 8.0 * ibit20 * adv_mom_5 & |
|---|
| 1247 | + ibit19 * adv_mom_3 & |
|---|
| 1248 | ) * & |
|---|
| 1249 | ( v(k,j,i+1) + v(k,j,i-2) ) & |
|---|
| 1250 | + ( ibit20 * adv_mom_5 & |
|---|
| 1251 | ) * & |
|---|
| 1252 | ( v(k,j,i+2) + v(k,j,i-3) ) & |
|---|
| 1253 | ) |
|---|
| 1254 | |
|---|
| 1255 | diss_l_v(k,j,tn) = - ABS( u_comp ) * ( & |
|---|
| 1256 | ( 10.0 * ibit20 * adv_mom_5 & |
|---|
| 1257 | + 3.0 * ibit19 * adv_mom_3 & |
|---|
| 1258 | + ibit18 * adv_mom_1 & |
|---|
| 1259 | ) * & |
|---|
| 1260 | ( v(k,j,i) - v(k,j,i-1) ) & |
|---|
| 1261 | - ( 5.0 * ibit20 * adv_mom_5 & |
|---|
| 1262 | + ibit19 * adv_mom_3 & |
|---|
| 1263 | ) * & |
|---|
| 1264 | ( v(k,j,i+1) - v(k,j,i-2) ) & |
|---|
| 1265 | + ( ibit20 * adv_mom_5 & |
|---|
| 1266 | ) * & |
|---|
| 1267 | ( v(k,j,i+2) - v(k,j,i-3) ) & |
|---|
| 1268 | ) |
|---|
| 1269 | |
|---|
| 1270 | ENDDO |
|---|
| 1271 | |
|---|
| 1272 | DO k = nzb_max+1, nzt |
|---|
| 1273 | |
|---|
| 1274 | u_comp = u(k,j-1,i) + u(k,j,i) - gu |
|---|
| 1275 | flux_l_v(k,j,tn) = u_comp * ( & |
|---|
| 1276 | 37.0 * ( v(k,j,i) + v(k,j,i-1) ) & |
|---|
| 1277 | - 8.0 * ( v(k,j,i+1) + v(k,j,i-2) ) & |
|---|
| 1278 | + ( v(k,j,i+2) + v(k,j,i-3) ) ) * adv_mom_5 |
|---|
| 1279 | diss_l_v(k,j,tn) = - ABS( u_comp ) * ( & |
|---|
| 1280 | 10.0 * ( v(k,j,i) - v(k,j,i-1) ) & |
|---|
| 1281 | - 5.0 * ( v(k,j,i+1) - v(k,j,i-2) ) & |
|---|
| 1282 | + ( v(k,j,i+2) - v(k,j,i-3) ) ) * adv_mom_5 |
|---|
| 1283 | |
|---|
| 1284 | ENDDO |
|---|
| 1285 | |
|---|
| 1286 | ENDIF |
|---|
| 1287 | ! |
|---|
| 1288 | !-- Compute southside fluxes for the respective boundary. |
|---|
| 1289 | IF ( j == nysv ) THEN |
|---|
| 1290 | |
|---|
| 1291 | DO k = nzb+1, nzb_max |
|---|
| 1292 | |
|---|
| 1293 | ibit23 = IBITS(wall_flags_0(k,j,i),23,1) |
|---|
| 1294 | ibit22 = IBITS(wall_flags_0(k,j,i),22,1) |
|---|
| 1295 | ibit21 = IBITS(wall_flags_0(k,j,i),21,1) |
|---|
| 1296 | |
|---|
| 1297 | v_comp_l = v(k,j,i) + v(k,j-1,i) - gv |
|---|
| 1298 | flux_s_v(k,tn) = v_comp_l * ( & |
|---|
| 1299 | ( 37.0 * ibit23 * adv_mom_5 & |
|---|
| 1300 | + 7.0 * ibit22 * adv_mom_3 & |
|---|
| 1301 | + ibit21 * adv_mom_1 & |
|---|
| 1302 | ) * & |
|---|
| 1303 | ( v(k,j,i) + v(k,j-1,i) ) & |
|---|
| 1304 | - ( 8.0 * ibit23 * adv_mom_5 & |
|---|
| 1305 | + ibit22 * adv_mom_3 & |
|---|
| 1306 | ) * & |
|---|
| 1307 | ( v(k,j+1,i) + v(k,j-2,i) ) & |
|---|
| 1308 | + ( ibit23 * adv_mom_5 & |
|---|
| 1309 | ) * & |
|---|
| 1310 | ( v(k,j+2,i) + v(k,j-3,i) ) & |
|---|
| 1311 | ) |
|---|
| 1312 | |
|---|
| 1313 | diss_s_v(k,tn) = - ABS( v_comp_l ) * ( & |
|---|
| 1314 | ( 10.0 * ibit23 * adv_mom_5 & |
|---|
| 1315 | + 3.0 * ibit22 * adv_mom_3 & |
|---|
| 1316 | + ibit21 * adv_mom_1 & |
|---|
| 1317 | ) * & |
|---|
| 1318 | ( v(k,j,i) - v(k,j-1,i) ) & |
|---|
| 1319 | - ( 5.0 * ibit23 * adv_mom_5 & |
|---|
| 1320 | + ibit22 * adv_mom_3 & |
|---|
| 1321 | ) * & |
|---|
| 1322 | ( v(k,j+1,i) - v(k,j-2,i) ) & |
|---|
| 1323 | + ( ibit23 * adv_mom_5 & |
|---|
| 1324 | ) * & |
|---|
| 1325 | ( v(k,j+2,i) - v(k,j-3,i) ) & |
|---|
| 1326 | ) |
|---|
| 1327 | |
|---|
| 1328 | ENDDO |
|---|
| 1329 | |
|---|
| 1330 | DO k = nzb_max+1, nzt |
|---|
| 1331 | |
|---|
| 1332 | v_comp_l = v(k,j,i) + v(k,j-1,i) - gv |
|---|
| 1333 | flux_s_v(k,tn) = v_comp_l * ( & |
|---|
| 1334 | 37.0 * ( v(k,j,i) + v(k,j-1,i) ) & |
|---|
| 1335 | - 8.0 * ( v(k,j+1,i) + v(k,j-2,i) ) & |
|---|
| 1336 | + ( v(k,j+2,i) + v(k,j-3,i) ) ) * adv_mom_5 |
|---|
| 1337 | diss_s_v(k,tn) = - ABS( v_comp_l ) * ( & |
|---|
| 1338 | 10.0 * ( v(k,j,i) - v(k,j-1,i) ) & |
|---|
| 1339 | - 5.0 * ( v(k,j+1,i) - v(k,j-2,i) ) & |
|---|
| 1340 | + ( v(k,j+2,i) - v(k,j-3,i) ) ) * adv_mom_5 |
|---|
| 1341 | |
|---|
| 1342 | ENDDO |
|---|
| 1343 | |
|---|
| 1344 | ENDIF |
|---|
| 1345 | |
|---|
| 1346 | flux_t(0) = 0.0 |
|---|
| 1347 | diss_t(0) = 0.0 |
|---|
| 1348 | flux_d = 0.0 |
|---|
| 1349 | diss_d = 0.0 |
|---|
| 1350 | ! |
|---|
| 1351 | !-- Now compute the fluxes and tendency terms for the horizontal and |
|---|
| 1352 | !-- verical parts. |
|---|
| 1353 | DO k = nzb+1, nzb_max |
|---|
| 1354 | |
|---|
| 1355 | ibit20 = IBITS(wall_flags_0(k,j,i),20,1) |
|---|
| 1356 | ibit19 = IBITS(wall_flags_0(k,j,i),19,1) |
|---|
| 1357 | ibit18 = IBITS(wall_flags_0(k,j,i),18,1) |
|---|
| 1358 | |
|---|
| 1359 | u_comp = u(k,j-1,i+1) + u(k,j,i+1) - gu |
|---|
| 1360 | flux_r(k) = u_comp * ( & |
|---|
| 1361 | ( 37.0 * ibit20 * adv_mom_5 & |
|---|
| 1362 | + 7.0 * ibit19 * adv_mom_3 & |
|---|
| 1363 | + ibit18 * adv_mom_1 & |
|---|
| 1364 | ) * & |
|---|
| 1365 | ( v(k,j,i+1) + v(k,j,i) ) & |
|---|
| 1366 | - ( 8.0 * ibit20 * adv_mom_5 & |
|---|
| 1367 | + ibit19 * adv_mom_3 & |
|---|
| 1368 | ) * & |
|---|
| 1369 | ( v(k,j,i+2) + v(k,j,i-1) ) & |
|---|
| 1370 | + ( ibit20 * adv_mom_5 & |
|---|
| 1371 | ) * & |
|---|
| 1372 | ( v(k,j,i+3) + v(k,j,i-2) ) & |
|---|
| 1373 | ) |
|---|
| 1374 | |
|---|
| 1375 | diss_r(k) = - ABS( u_comp ) * ( & |
|---|
| 1376 | ( 10.0 * ibit20 * adv_mom_5 & |
|---|
| 1377 | + 3.0 * ibit19 * adv_mom_3 & |
|---|
| 1378 | + ibit18 * adv_mom_1 & |
|---|
| 1379 | ) * & |
|---|
| 1380 | ( v(k,j,i+1) - v(k,j,i) ) & |
|---|
| 1381 | - ( 5.0 * ibit20 * adv_mom_5 & |
|---|
| 1382 | + ibit19 * adv_mom_3 & |
|---|
| 1383 | ) * & |
|---|
| 1384 | ( v(k,j,i+2) - v(k,j,i-1) ) & |
|---|
| 1385 | + ( ibit20 * adv_mom_5 & |
|---|
| 1386 | ) * & |
|---|
| 1387 | ( v(k,j,i+3) - v(k,j,i-2) ) & |
|---|
| 1388 | ) |
|---|
| 1389 | |
|---|
| 1390 | ibit23 = IBITS(wall_flags_0(k,j,i),23,1) |
|---|
| 1391 | ibit22 = IBITS(wall_flags_0(k,j,i),22,1) |
|---|
| 1392 | ibit21 = IBITS(wall_flags_0(k,j,i),21,1) |
|---|
| 1393 | |
|---|
| 1394 | |
|---|
| 1395 | v_comp(k) = v(k,j+1,i) + v(k,j,i) |
|---|
| 1396 | flux_n(k) = ( v_comp(k) - gv ) * ( & |
|---|
| 1397 | ( 37.0 * ibit23 * adv_mom_5 & |
|---|
| 1398 | + 7.0 * ibit22 * adv_mom_3 & |
|---|
| 1399 | + ibit21 * adv_mom_1 & |
|---|
| 1400 | ) * & |
|---|
| 1401 | ( v(k,j+1,i) + v(k,j,i) ) & |
|---|
| 1402 | - ( 8.0 * ibit23 * adv_mom_5 & |
|---|
| 1403 | + ibit22 * adv_mom_3 & |
|---|
| 1404 | ) * & |
|---|
| 1405 | ( v(k,j+2,i) + v(k,j-1,i) ) & |
|---|
| 1406 | + ( ibit23 * adv_mom_5 & |
|---|
| 1407 | ) * & |
|---|
| 1408 | ( v(k,j+3,i) + v(k,j-2,i) ) & |
|---|
| 1409 | ) |
|---|
| 1410 | |
|---|
| 1411 | diss_n(k) = - ABS( v_comp(k) - gv ) * ( & |
|---|
| 1412 | ( 10.0 * ibit23 * adv_mom_5 & |
|---|
| 1413 | + 3.0 * ibit22 * adv_mom_3 & |
|---|
| 1414 | + ibit21 * adv_mom_1 & |
|---|
| 1415 | ) * & |
|---|
| 1416 | ( v(k,j+1,i) - v(k,j,i) ) & |
|---|
| 1417 | - ( 5.0 * ibit23 * adv_mom_5 & |
|---|
| 1418 | + ibit22 * adv_mom_3 & |
|---|
| 1419 | ) * & |
|---|
| 1420 | ( v(k,j+2,i) - v(k,j-1,i) ) & |
|---|
| 1421 | + ( ibit23 * adv_mom_5 & |
|---|
| 1422 | ) * & |
|---|
| 1423 | ( v(k,j+3,i) - v(k,j-2,i) ) & |
|---|
| 1424 | ) |
|---|
| 1425 | ! |
|---|
| 1426 | !-- k index has to be modified near bottom and top, else array |
|---|
| 1427 | !-- subscripts will be exceeded. |
|---|
| 1428 | ibit26 = IBITS(wall_flags_0(k,j,i),26,1) |
|---|
| 1429 | ibit25 = IBITS(wall_flags_0(k,j,i),25,1) |
|---|
| 1430 | ibit24 = IBITS(wall_flags_0(k,j,i),24,1) |
|---|
| 1431 | |
|---|
| 1432 | k_ppp = k + 3 * ibit26 |
|---|
| 1433 | k_pp = k + 2 * ( 1 - ibit24 ) |
|---|
| 1434 | k_mm = k - 2 * ibit26 |
|---|
| 1435 | |
|---|
| 1436 | w_comp = w(k,j-1,i) + w(k,j,i) |
|---|
| 1437 | flux_t(k) = w_comp * ( & |
|---|
| 1438 | ( 37.0 * ibit26 * adv_mom_5 & |
|---|
| 1439 | + 7.0 * ibit25 * adv_mom_3 & |
|---|
| 1440 | + ibit24 * adv_mom_1 & |
|---|
| 1441 | ) * & |
|---|
| 1442 | ( v(k+1,j,i) + v(k,j,i) ) & |
|---|
| 1443 | - ( 8.0 * ibit26 * adv_mom_5 & |
|---|
| 1444 | + ibit25 * adv_mom_3 & |
|---|
| 1445 | ) * & |
|---|
| 1446 | ( v(k_pp,j,i) + v(k-1,j,i) ) & |
|---|
| 1447 | + ( ibit26 * adv_mom_5 & |
|---|
| 1448 | ) * & |
|---|
| 1449 | ( v(k_ppp,j,i) + v(k_mm,j,i) ) & |
|---|
| 1450 | ) |
|---|
| 1451 | |
|---|
| 1452 | diss_t(k) = - ABS( w_comp ) * ( & |
|---|
| 1453 | ( 10.0 * ibit26 * adv_mom_5 & |
|---|
| 1454 | + 3.0 * ibit25 * adv_mom_3 & |
|---|
| 1455 | + ibit24 * adv_mom_1 & |
|---|
| 1456 | ) * & |
|---|
| 1457 | ( v(k+1,j,i) - v(k,j,i) ) & |
|---|
| 1458 | - ( 5.0 * ibit26 * adv_mom_5 & |
|---|
| 1459 | + ibit25 * adv_mom_3 & |
|---|
| 1460 | ) * & |
|---|
| 1461 | ( v(k_pp,j,i) - v(k-1,j,i) ) & |
|---|
| 1462 | + ( ibit26 * adv_mom_5 & |
|---|
| 1463 | ) * & |
|---|
| 1464 | ( v(k_ppp,j,i) - v(k_mm,j,i) ) & |
|---|
| 1465 | ) |
|---|
| 1466 | ! |
|---|
| 1467 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 1468 | !-- correction is needed to overcome numerical instabilities introduced |
|---|
| 1469 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 1470 | div = ( ( u_comp + gu - ( u(k,j-1,i) + u(k,j,i) ) ) * ddx & |
|---|
| 1471 | + ( v_comp(k) - ( v(k,j,i) + v(k,j-1,i) ) ) * ddy & |
|---|
| 1472 | + ( w_comp - ( w(k-1,j-1,i) + w(k-1,j,i) ) ) * ddzw(k) & |
|---|
| 1473 | ) * 0.5 |
|---|
| 1474 | |
|---|
| 1475 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 1476 | ( flux_r(k) + diss_r(k) & |
|---|
| 1477 | - flux_l_v(k,j,tn) - diss_l_v(k,j,tn) ) * ddx & |
|---|
| 1478 | + ( flux_n(k) + diss_n(k) & |
|---|
| 1479 | - flux_s_v(k,tn) - diss_s_v(k,tn) ) * ddy & |
|---|
| 1480 | + ( flux_t(k) + diss_t(k) & |
|---|
| 1481 | - flux_d - diss_d ) * ddzw(k) & |
|---|
| 1482 | ) + v(k,j,i) * div |
|---|
| 1483 | |
|---|
| 1484 | flux_l_v(k,j,tn) = flux_r(k) |
|---|
| 1485 | diss_l_v(k,j,tn) = diss_r(k) |
|---|
| 1486 | flux_s_v(k,tn) = flux_n(k) |
|---|
| 1487 | diss_s_v(k,tn) = diss_n(k) |
|---|
| 1488 | flux_d = flux_t(k) |
|---|
| 1489 | diss_d = diss_t(k) |
|---|
| 1490 | |
|---|
| 1491 | ! |
|---|
| 1492 | !-- Statistical Evaluation of v'v'. The factor has to be applied for |
|---|
| 1493 | !-- right evaluation when gallilei_trans = .T. . |
|---|
| 1494 | sums_vs2_ws_l(k,tn) = sums_vs2_ws_l(k,tn) & |
|---|
| 1495 | + ( flux_n(k) & |
|---|
| 1496 | * ( v_comp(k) - 2.0 * hom(k,1,2,0) ) & |
|---|
| 1497 | / ( v_comp(k) - gv + 1.0E-20 ) & |
|---|
| 1498 | + diss_n(k) & |
|---|
| 1499 | * ABS( v_comp(k) - 2.0 * hom(k,1,2,0) ) & |
|---|
| 1500 | / ( ABS( v_comp(k) - gv ) +1.0E-20 ) ) & |
|---|
| 1501 | * weight_substep(intermediate_timestep_count) |
|---|
| 1502 | ! |
|---|
| 1503 | !-- Statistical Evaluation of w'v'. |
|---|
| 1504 | sums_wsvs_ws_l(k,tn) = sums_wsvs_ws_l(k,tn) & |
|---|
| 1505 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 1506 | * weight_substep(intermediate_timestep_count) |
|---|
| 1507 | |
|---|
| 1508 | ENDDO |
|---|
| 1509 | |
|---|
| 1510 | DO k = nzb_max+1, nzt |
|---|
| 1511 | |
|---|
| 1512 | u_comp = u(k,j-1,i+1) + u(k,j,i+1) - gu |
|---|
| 1513 | flux_r(k) = u_comp * ( & |
|---|
| 1514 | 37.0 * ( v(k,j,i+1) + v(k,j,i) ) & |
|---|
| 1515 | - 8.0 * ( v(k,j,i+2) + v(k,j,i-1) ) & |
|---|
| 1516 | + ( v(k,j,i+3) + v(k,j,i-2) ) ) * adv_mom_5 |
|---|
| 1517 | |
|---|
| 1518 | diss_r(k) = - ABS( u_comp ) * ( & |
|---|
| 1519 | 10.0 * ( v(k,j,i+1) - v(k,j,i) ) & |
|---|
| 1520 | - 5.0 * ( v(k,j,i+2) - v(k,j,i-1) ) & |
|---|
| 1521 | + ( v(k,j,i+3) - v(k,j,i-2) ) ) * adv_mom_5 |
|---|
| 1522 | |
|---|
| 1523 | |
|---|
| 1524 | v_comp(k) = v(k,j+1,i) + v(k,j,i) |
|---|
| 1525 | flux_n(k) = ( v_comp(k) - gv ) * ( & |
|---|
| 1526 | 37.0 * ( v(k,j+1,i) + v(k,j,i) ) & |
|---|
| 1527 | - 8.0 * ( v(k,j+2,i) + v(k,j-1,i) ) & |
|---|
| 1528 | + ( v(k,j+3,i) + v(k,j-2,i) ) ) * adv_mom_5 |
|---|
| 1529 | |
|---|
| 1530 | diss_n(k) = - ABS( v_comp(k) - gv ) * ( & |
|---|
| 1531 | 10.0 * ( v(k,j+1,i) - v(k,j,i) ) & |
|---|
| 1532 | - 5.0 * ( v(k,j+2,i) - v(k,j-1,i) ) & |
|---|
| 1533 | + ( v(k,j+3,i) - v(k,j-2,i) ) ) * adv_mom_5 |
|---|
| 1534 | ! |
|---|
| 1535 | !-- k index has to be modified near bottom and top, else array |
|---|
| 1536 | !-- subscripts will be exceeded. |
|---|
| 1537 | ibit26 = IBITS(wall_flags_0(k,j,i),26,1) |
|---|
| 1538 | ibit25 = IBITS(wall_flags_0(k,j,i),25,1) |
|---|
| 1539 | ibit24 = IBITS(wall_flags_0(k,j,i),24,1) |
|---|
| 1540 | |
|---|
| 1541 | k_ppp = k + 3 * ibit26 |
|---|
| 1542 | k_pp = k + 2 * ( 1 - ibit24 ) |
|---|
| 1543 | k_mm = k - 2 * ibit26 |
|---|
| 1544 | |
|---|
| 1545 | w_comp = w(k,j-1,i) + w(k,j,i) |
|---|
| 1546 | flux_t(k) = w_comp * ( & |
|---|
| 1547 | ( 37.0 * ibit26 * adv_mom_5 & |
|---|
| 1548 | + 7.0 * ibit25 * adv_mom_3 & |
|---|
| 1549 | + ibit24 * adv_mom_1 & |
|---|
| 1550 | ) * & |
|---|
| 1551 | ( v(k+1,j,i) + v(k,j,i) ) & |
|---|
| 1552 | - ( 8.0 * ibit26 * adv_mom_5 & |
|---|
| 1553 | + ibit25 * adv_mom_3 & |
|---|
| 1554 | ) * & |
|---|
| 1555 | ( v(k_pp,j,i) + v(k-1,j,i) ) & |
|---|
| 1556 | + ( ibit26 * adv_mom_5 & |
|---|
| 1557 | ) * & |
|---|
| 1558 | ( v(k_ppp,j,i) + v(k_mm,j,i) ) & |
|---|
| 1559 | ) |
|---|
| 1560 | |
|---|
| 1561 | diss_t(k) = - ABS( w_comp ) * ( & |
|---|
| 1562 | ( 10.0 * ibit26 * adv_mom_5 & |
|---|
| 1563 | + 3.0 * ibit25 * adv_mom_3 & |
|---|
| 1564 | + ibit24 * adv_mom_1 & |
|---|
| 1565 | ) * & |
|---|
| 1566 | ( v(k+1,j,i) - v(k,j,i) ) & |
|---|
| 1567 | - ( 5.0 * ibit26 * adv_mom_5 & |
|---|
| 1568 | + ibit25 * adv_mom_3 & |
|---|
| 1569 | ) * & |
|---|
| 1570 | ( v(k_pp,j,i) - v(k-1,j,i) ) & |
|---|
| 1571 | + ( ibit26 * adv_mom_5 & |
|---|
| 1572 | ) * & |
|---|
| 1573 | ( v(k_ppp,j,i) - v(k_mm,j,i) ) & |
|---|
| 1574 | ) |
|---|
| 1575 | ! |
|---|
| 1576 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 1577 | !-- correction is needed to overcome numerical instabilities introduced |
|---|
| 1578 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 1579 | div = ( ( u_comp + gu - ( u(k,j-1,i) + u(k,j,i) ) ) * ddx & |
|---|
| 1580 | + ( v_comp(k) - ( v(k,j,i) + v(k,j-1,i) ) ) * ddy & |
|---|
| 1581 | + ( w_comp - ( w(k-1,j-1,i) + w(k-1,j,i) ) ) * ddzw(k) & |
|---|
| 1582 | ) * 0.5 |
|---|
| 1583 | |
|---|
| 1584 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 1585 | ( flux_r(k) + diss_r(k) & |
|---|
| 1586 | - flux_l_v(k,j,tn) - diss_l_v(k,j,tn) ) * ddx & |
|---|
| 1587 | + ( flux_n(k) + diss_n(k) & |
|---|
| 1588 | - flux_s_v(k,tn) - diss_s_v(k,tn) ) * ddy & |
|---|
| 1589 | + ( flux_t(k) + diss_t(k) & |
|---|
| 1590 | - flux_d - diss_d ) * ddzw(k) & |
|---|
| 1591 | ) + v(k,j,i) * div |
|---|
| 1592 | |
|---|
| 1593 | flux_l_v(k,j,tn) = flux_r(k) |
|---|
| 1594 | diss_l_v(k,j,tn) = diss_r(k) |
|---|
| 1595 | flux_s_v(k,tn) = flux_n(k) |
|---|
| 1596 | diss_s_v(k,tn) = diss_n(k) |
|---|
| 1597 | flux_d = flux_t(k) |
|---|
| 1598 | diss_d = diss_t(k) |
|---|
| 1599 | |
|---|
| 1600 | ! |
|---|
| 1601 | !-- Statistical Evaluation of v'v'. The factor has to be applied for |
|---|
| 1602 | !-- right evaluation when gallilei_trans = .T. . |
|---|
| 1603 | sums_vs2_ws_l(k,tn) = sums_vs2_ws_l(k,tn) & |
|---|
| 1604 | + ( flux_n(k) & |
|---|
| 1605 | * ( v_comp(k) - 2.0 * hom(k,1,2,0) ) & |
|---|
| 1606 | / ( v_comp(k) - gv + 1.0E-20 ) & |
|---|
| 1607 | + diss_n(k) & |
|---|
| 1608 | * ABS( v_comp(k) - 2.0 * hom(k,1,2,0) ) & |
|---|
| 1609 | / ( ABS( v_comp(k) - gv ) +1.0E-20 ) ) & |
|---|
| 1610 | * weight_substep(intermediate_timestep_count) |
|---|
| 1611 | ! |
|---|
| 1612 | !-- Statistical Evaluation of w'v'. |
|---|
| 1613 | sums_wsvs_ws_l(k,tn) = sums_wsvs_ws_l(k,tn) & |
|---|
| 1614 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 1615 | * weight_substep(intermediate_timestep_count) |
|---|
| 1616 | |
|---|
| 1617 | ENDDO |
|---|
| 1618 | sums_vs2_ws_l(nzb,tn) = sums_vs2_ws_l(nzb+1,tn) |
|---|
| 1619 | |
|---|
| 1620 | |
|---|
| 1621 | END SUBROUTINE advec_v_ws_ij |
|---|
| 1622 | |
|---|
| 1623 | |
|---|
| 1624 | |
|---|
| 1625 | !------------------------------------------------------------------------------! |
|---|
| 1626 | ! Advection of w-component - Call for grid point i,j |
|---|
| 1627 | !------------------------------------------------------------------------------! |
|---|
| 1628 | SUBROUTINE advec_w_ws_ij( i, j, i_omp, tn ) |
|---|
| 1629 | |
|---|
| 1630 | USE arrays_3d |
|---|
| 1631 | USE constants |
|---|
| 1632 | USE control_parameters |
|---|
| 1633 | USE grid_variables |
|---|
| 1634 | USE indices |
|---|
| 1635 | USE statistics |
|---|
| 1636 | |
|---|
| 1637 | IMPLICIT NONE |
|---|
| 1638 | |
|---|
| 1639 | INTEGER :: i, ibit27, ibit28, ibit29, ibit30, ibit31, ibit32, ibit33, & |
|---|
| 1640 | ibit34, ibit35, i_omp, j, k, k_mm, k_pp, k_ppp, tn |
|---|
| 1641 | REAL :: diss_d, div, flux_d, gu, gv, u_comp, v_comp, w_comp |
|---|
| 1642 | REAL, DIMENSION(nzb:nzt+1) :: diss_n, diss_r, diss_t, flux_n, flux_r, & |
|---|
| 1643 | flux_t |
|---|
| 1644 | |
|---|
| 1645 | gu = 2.0 * u_gtrans |
|---|
| 1646 | gv = 2.0 * v_gtrans |
|---|
| 1647 | |
|---|
| 1648 | ! |
|---|
| 1649 | !-- Compute southside fluxes for the respective boundary. |
|---|
| 1650 | IF ( j == nys ) THEN |
|---|
| 1651 | |
|---|
| 1652 | DO k = nzb+1, nzb_max |
|---|
| 1653 | ibit32 = IBITS(wall_flags_00(k,j,i),0,1) |
|---|
| 1654 | ibit31 = IBITS(wall_flags_0(k,j,i),31,1) |
|---|
| 1655 | ibit30 = IBITS(wall_flags_0(k,j,i),30,1) |
|---|
| 1656 | |
|---|
| 1657 | v_comp = v(k+1,j,i) + v(k,j,i) - gv |
|---|
| 1658 | flux_s_w(k,tn) = v_comp * ( & |
|---|
| 1659 | ( 37.0 * ibit32 * adv_mom_5 & |
|---|
| 1660 | + 7.0 * ibit31 * adv_mom_3 & |
|---|
| 1661 | + ibit30 * adv_mom_1 & |
|---|
| 1662 | ) * & |
|---|
| 1663 | ( w(k,j,i) + w(k,j-1,i) ) & |
|---|
| 1664 | - ( 8.0 * ibit32 * adv_mom_5 & |
|---|
| 1665 | + ibit31 * adv_mom_3 & |
|---|
| 1666 | ) * & |
|---|
| 1667 | ( w(k,j+1,i) + w(k,j-2,i) ) & |
|---|
| 1668 | + ( ibit32 * adv_mom_5 & |
|---|
| 1669 | ) * & |
|---|
| 1670 | ( w(k,j+2,i) + w(k,j-3,i) ) & |
|---|
| 1671 | ) |
|---|
| 1672 | |
|---|
| 1673 | diss_s_w(k,tn) = - ABS( v_comp ) * ( & |
|---|
| 1674 | ( 10.0 * ibit32 * adv_mom_5 & |
|---|
| 1675 | + 3.0 * ibit31 * adv_mom_3 & |
|---|
| 1676 | + ibit30 * adv_mom_1 & |
|---|
| 1677 | ) * & |
|---|
| 1678 | ( w(k,j,i) - w(k,j-1,i) ) & |
|---|
| 1679 | - ( 5.0 * ibit32 * adv_mom_5 & |
|---|
| 1680 | + ibit31 * adv_mom_3 & |
|---|
| 1681 | ) * & |
|---|
| 1682 | ( w(k,j+1,i) - w(k,j-2,i) ) & |
|---|
| 1683 | + ( ibit32 * adv_mom_5 & |
|---|
| 1684 | ) * & |
|---|
| 1685 | ( w(k,j+2,i) - w(k,j-3,i) ) & |
|---|
| 1686 | ) |
|---|
| 1687 | |
|---|
| 1688 | ENDDO |
|---|
| 1689 | |
|---|
| 1690 | DO k = nzb_max+1, nzt |
|---|
| 1691 | |
|---|
| 1692 | v_comp = v(k+1,j,i) + v(k,j,i) - gv |
|---|
| 1693 | flux_s_w(k,tn) = v_comp * ( & |
|---|
| 1694 | 37.0 * ( w(k,j,i) + w(k,j-1,i) ) & |
|---|
| 1695 | - 8.0 * ( w(k,j+1,i) +w(k,j-2,i) ) & |
|---|
| 1696 | + ( w(k,j+2,i) + w(k,j-3,i) ) ) * adv_mom_5 |
|---|
| 1697 | diss_s_w(k,tn) = - ABS( v_comp ) * ( & |
|---|
| 1698 | 10.0 * ( w(k,j,i) - w(k,j-1,i) ) & |
|---|
| 1699 | - 5.0 * ( w(k,j+1,i) - w(k,j-2,i) ) & |
|---|
| 1700 | + ( w(k,j+2,i) - w(k,j-3,i) ) ) * adv_mom_5 |
|---|
| 1701 | |
|---|
| 1702 | ENDDO |
|---|
| 1703 | |
|---|
| 1704 | ENDIF |
|---|
| 1705 | ! |
|---|
| 1706 | !-- Compute leftside fluxes for the respective boundary. |
|---|
| 1707 | IF ( i == i_omp ) THEN |
|---|
| 1708 | |
|---|
| 1709 | DO k = nzb+1, nzb_max |
|---|
| 1710 | |
|---|
| 1711 | ibit29 = IBITS(wall_flags_0(k,j,i),29,1) |
|---|
| 1712 | ibit28 = IBITS(wall_flags_0(k,j,i),28,1) |
|---|
| 1713 | ibit27 = IBITS(wall_flags_0(k,j,i),27,1) |
|---|
| 1714 | |
|---|
| 1715 | u_comp = u(k+1,j,i) + u(k,j,i) - gu |
|---|
| 1716 | flux_l_w(k,j,tn) = u_comp * ( & |
|---|
| 1717 | ( 37.0 * ibit29 * adv_mom_5 & |
|---|
| 1718 | + 7.0 * ibit28 * adv_mom_3 & |
|---|
| 1719 | + ibit27 * adv_mom_1 & |
|---|
| 1720 | ) * & |
|---|
| 1721 | ( w(k,j,i) + w(k,j,i-1) ) & |
|---|
| 1722 | - ( 8.0 * ibit29 * adv_mom_5 & |
|---|
| 1723 | + ibit28 * adv_mom_3 & |
|---|
| 1724 | ) * & |
|---|
| 1725 | ( w(k,j,i+1) + w(k,j,i-2) ) & |
|---|
| 1726 | + ( ibit29 * adv_mom_5 & |
|---|
| 1727 | ) * & |
|---|
| 1728 | ( w(k,j,i+2) + w(k,j,i-3) ) & |
|---|
| 1729 | ) |
|---|
| 1730 | |
|---|
| 1731 | diss_l_w(k,j,tn) = - ABS( u_comp ) * ( & |
|---|
| 1732 | ( 10.0 * ibit29 * adv_mom_5 & |
|---|
| 1733 | + 3.0 * ibit28 * adv_mom_3 & |
|---|
| 1734 | + ibit27 * adv_mom_1 & |
|---|
| 1735 | ) * & |
|---|
| 1736 | ( w(k,j,i) - w(k,j,i-1) ) & |
|---|
| 1737 | - ( 5.0 * ibit29 * adv_mom_5 & |
|---|
| 1738 | + ibit28 * adv_mom_3 & |
|---|
| 1739 | ) * & |
|---|
| 1740 | ( w(k,j,i+1) - w(k,j,i-2) ) & |
|---|
| 1741 | + ( ibit29 * adv_mom_5 & |
|---|
| 1742 | ) * & |
|---|
| 1743 | ( w(k,j,i+2) - w(k,j,i-3) ) & |
|---|
| 1744 | ) |
|---|
| 1745 | |
|---|
| 1746 | ENDDO |
|---|
| 1747 | |
|---|
| 1748 | DO k = nzb_max+1, nzt |
|---|
| 1749 | |
|---|
| 1750 | u_comp = u(k+1,j,i) + u(k,j,i) - gu |
|---|
| 1751 | flux_l_w(k,j,tn) = u_comp * ( & |
|---|
| 1752 | 37.0 * ( w(k,j,i) + w(k,j,i-1) ) & |
|---|
| 1753 | - 8.0 * ( w(k,j,i+1) + w(k,j,i-2) ) & |
|---|
| 1754 | + ( w(k,j,i+2) + w(k,j,i-3) ) ) * adv_mom_5 |
|---|
| 1755 | diss_l_w(k,j,tn) = - ABS( u_comp ) * ( & |
|---|
| 1756 | 10.0 * ( w(k,j,i) - w(k,j,i-1) ) & |
|---|
| 1757 | - 5.0 * ( w(k,j,i+1) - w(k,j,i-2) ) & |
|---|
| 1758 | + ( w(k,j,i+2) - w(k,j,i-3) ) ) * adv_mom_5 |
|---|
| 1759 | |
|---|
| 1760 | ENDDO |
|---|
| 1761 | |
|---|
| 1762 | ENDIF |
|---|
| 1763 | ! |
|---|
| 1764 | !-- The lower flux has to be calculated explicetely for the tendency at |
|---|
| 1765 | !-- the first w-level. For topography wall this is done implicitely by |
|---|
| 1766 | !-- wall_flags_0. |
|---|
| 1767 | k = nzb + 1 |
|---|
| 1768 | w_comp = w(k,j,i) + w(k-1,j,i) |
|---|
| 1769 | flux_t(0) = w_comp * ( w(k,j,i) + w(k-1,j,i) ) * adv_mom_1 |
|---|
| 1770 | diss_t(0) = -ABS(w_comp) * ( w(k,j,i) - w(k-1,j,i) ) * adv_mom_1 |
|---|
| 1771 | flux_d = flux_t(0) |
|---|
| 1772 | diss_d = diss_t(0) |
|---|
| 1773 | ! |
|---|
| 1774 | !-- Now compute the fluxes and tendency terms for the horizontal |
|---|
| 1775 | !-- and vertical parts. |
|---|
| 1776 | DO k = nzb+1, nzb_max |
|---|
| 1777 | |
|---|
| 1778 | ibit29 = IBITS(wall_flags_0(k,j,i),29,1) |
|---|
| 1779 | ibit28 = IBITS(wall_flags_0(k,j,i),28,1) |
|---|
| 1780 | ibit27 = IBITS(wall_flags_0(k,j,i),27,1) |
|---|
| 1781 | |
|---|
| 1782 | u_comp = u(k+1,j,i+1) + u(k,j,i+1) - gu |
|---|
| 1783 | flux_r(k) = u_comp * ( & |
|---|
| 1784 | ( 37.0 * ibit29 * adv_mom_5 & |
|---|
| 1785 | + 7.0 * ibit28 * adv_mom_3 & |
|---|
| 1786 | + ibit27 * adv_mom_1 & |
|---|
| 1787 | ) * & |
|---|
| 1788 | ( w(k,j,i+1) + w(k,j,i) ) & |
|---|
| 1789 | - ( 8.0 * ibit29 * adv_mom_5 & |
|---|
| 1790 | + ibit28 * adv_mom_3 & |
|---|
| 1791 | ) * & |
|---|
| 1792 | ( w(k,j,i+2) + w(k,j,i-1) ) & |
|---|
| 1793 | + ( ibit29 * adv_mom_5 & |
|---|
| 1794 | ) * & |
|---|
| 1795 | ( w(k,j,i+3) + w(k,j,i-2) ) & |
|---|
| 1796 | ) |
|---|
| 1797 | |
|---|
| 1798 | diss_r(k) = - ABS( u_comp ) * ( & |
|---|
| 1799 | ( 10.0 * ibit29 * adv_mom_5 & |
|---|
| 1800 | + 3.0 * ibit28 * adv_mom_3 & |
|---|
| 1801 | + ibit27 * adv_mom_1 & |
|---|
| 1802 | ) * & |
|---|
| 1803 | ( w(k,j,i+1) - w(k,j,i) ) & |
|---|
| 1804 | - ( 5.0 * ibit29 * adv_mom_5 & |
|---|
| 1805 | + ibit28 * adv_mom_3 & |
|---|
| 1806 | ) * & |
|---|
| 1807 | ( w(k,j,i+2) - w(k,j,i-1) ) & |
|---|
| 1808 | + ( ibit29 * adv_mom_5 & |
|---|
| 1809 | ) * & |
|---|
| 1810 | ( w(k,j,i+3) - w(k,j,i-2) ) & |
|---|
| 1811 | ) |
|---|
| 1812 | |
|---|
| 1813 | ibit32 = IBITS(wall_flags_00(k,j,i),0,1) |
|---|
| 1814 | ibit31 = IBITS(wall_flags_0(k,j,i),31,1) |
|---|
| 1815 | ibit30 = IBITS(wall_flags_0(k,j,i),30,1) |
|---|
| 1816 | |
|---|
| 1817 | v_comp = v(k+1,j+1,i) + v(k,j+1,i) - gv |
|---|
| 1818 | flux_n(k) = v_comp * ( & |
|---|
| 1819 | ( 37.0 * ibit32 * adv_mom_5 & |
|---|
| 1820 | + 7.0 * ibit31 * adv_mom_3 & |
|---|
| 1821 | + ibit30 * adv_mom_1 & |
|---|
| 1822 | ) * & |
|---|
| 1823 | ( w(k,j+1,i) + w(k,j,i) ) & |
|---|
| 1824 | - ( 8.0 * ibit32 * adv_mom_5 & |
|---|
| 1825 | + ibit31 * adv_mom_3 & |
|---|
| 1826 | ) * & |
|---|
| 1827 | ( w(k,j+2,i) + w(k,j-1,i) ) & |
|---|
| 1828 | + ( ibit32 * adv_mom_5 & |
|---|
| 1829 | ) * & |
|---|
| 1830 | ( w(k,j+3,i) + w(k,j-2,i) ) & |
|---|
| 1831 | ) |
|---|
| 1832 | |
|---|
| 1833 | diss_n(k) = - ABS( v_comp ) * ( & |
|---|
| 1834 | ( 10.0 * ibit32 * adv_mom_5 & |
|---|
| 1835 | + 3.0 * ibit31 * adv_mom_3 & |
|---|
| 1836 | + ibit30 * adv_mom_1 & |
|---|
| 1837 | ) * & |
|---|
| 1838 | ( w(k,j+1,i) - w(k,j,i) ) & |
|---|
| 1839 | - ( 5.0 * ibit32 * adv_mom_5 & |
|---|
| 1840 | + ibit31 * adv_mom_3 & |
|---|
| 1841 | ) * & |
|---|
| 1842 | ( w(k,j+2,i) - w(k,j-1,i) ) & |
|---|
| 1843 | + ( ibit32 * adv_mom_5 & |
|---|
| 1844 | ) * & |
|---|
| 1845 | ( w(k,j+3,i) - w(k,j-2,i) ) & |
|---|
| 1846 | ) |
|---|
| 1847 | ! |
|---|
| 1848 | !-- k index has to be modified near bottom and top, else array |
|---|
| 1849 | !-- subscripts will be exceeded. |
|---|
| 1850 | ibit35 = IBITS(wall_flags_00(k,j,i),3,1) |
|---|
| 1851 | ibit34 = IBITS(wall_flags_00(k,j,i),2,1) |
|---|
| 1852 | ibit33 = IBITS(wall_flags_00(k,j,i),1,1) |
|---|
| 1853 | |
|---|
| 1854 | k_ppp = k + 3 * ibit35 |
|---|
| 1855 | k_pp = k + 2 * ( 1 - ibit33 ) |
|---|
| 1856 | k_mm = k - 2 * ibit35 |
|---|
| 1857 | |
|---|
| 1858 | w_comp = w(k+1,j,i) + w(k,j,i) |
|---|
| 1859 | flux_t(k) = w_comp * ( & |
|---|
| 1860 | ( 37.0 * ibit35 * adv_mom_5 & |
|---|
| 1861 | + 7.0 * ibit34 * adv_mom_3 & |
|---|
| 1862 | + ibit33 * adv_mom_1 & |
|---|
| 1863 | ) * & |
|---|
| 1864 | ( w(k+1,j,i) + w(k,j,i) ) & |
|---|
| 1865 | - ( 8.0 * ibit35 * adv_mom_5 & |
|---|
| 1866 | + ibit34 * adv_mom_3 & |
|---|
| 1867 | ) * & |
|---|
| 1868 | ( w(k_pp,j,i) + w(k-1,j,i) ) & |
|---|
| 1869 | + ( ibit35 * adv_mom_5 & |
|---|
| 1870 | ) * & |
|---|
| 1871 | ( w(k_ppp,j,i) + w(k_mm,j,i) ) & |
|---|
| 1872 | ) |
|---|
| 1873 | |
|---|
| 1874 | diss_t(k) = - ABS( w_comp ) * ( & |
|---|
| 1875 | ( 10.0 * ibit35 * adv_mom_5 & |
|---|
| 1876 | + 3.0 * ibit34 * adv_mom_3 & |
|---|
| 1877 | + ibit33 * adv_mom_1 & |
|---|
| 1878 | ) * & |
|---|
| 1879 | ( w(k+1,j,i) - w(k,j,i) ) & |
|---|
| 1880 | - ( 5.0 * ibit35 * adv_mom_5 & |
|---|
| 1881 | + ibit34 * adv_mom_3 & |
|---|
| 1882 | ) * & |
|---|
| 1883 | ( w(k_pp,j,i) - w(k-1,j,i) ) & |
|---|
| 1884 | + ( ibit35 * adv_mom_5 & |
|---|
| 1885 | ) * & |
|---|
| 1886 | ( w(k_ppp,j,i) - w(k_mm,j,i) ) & |
|---|
| 1887 | ) |
|---|
| 1888 | |
|---|
| 1889 | ! |
|---|
| 1890 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 1891 | !-- correction is needed to overcome numerical instabilities introduced |
|---|
| 1892 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 1893 | div = ( ( u_comp + gu - ( u(k+1,j,i) + u(k,j,i) ) ) * ddx & |
|---|
| 1894 | + ( v_comp + gv - ( v(k+1,j,i) + v(k,j,i) ) ) * ddy & |
|---|
| 1895 | + ( w_comp - ( w(k,j,i) + w(k-1,j,i) ) ) * ddzu(k+1) & |
|---|
| 1896 | ) * 0.5 |
|---|
| 1897 | |
|---|
| 1898 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 1899 | ( flux_r(k) + diss_r(k) & |
|---|
| 1900 | - flux_l_w(k,j,tn) - diss_l_w(k,j,tn) ) * ddx & |
|---|
| 1901 | + ( flux_n(k) + diss_n(k) & |
|---|
| 1902 | - flux_s_w(k,tn) - diss_s_w(k,tn) ) * ddy & |
|---|
| 1903 | + ( flux_t(k) + diss_t(k) & |
|---|
| 1904 | - flux_d - diss_d ) * ddzu(k+1) & |
|---|
| 1905 | ) + div * w(k,j,i) |
|---|
| 1906 | |
|---|
| 1907 | flux_l_w(k,j,tn) = flux_r(k) |
|---|
| 1908 | diss_l_w(k,j,tn) = diss_r(k) |
|---|
| 1909 | flux_s_w(k,tn) = flux_n(k) |
|---|
| 1910 | diss_s_w(k,tn) = diss_n(k) |
|---|
| 1911 | flux_d = flux_t(k) |
|---|
| 1912 | diss_d = diss_t(k) |
|---|
| 1913 | ! |
|---|
| 1914 | !-- Statistical Evaluation of w'w'. |
|---|
| 1915 | sums_ws2_ws_l(k,tn) = sums_ws2_ws_l(k,tn) & |
|---|
| 1916 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 1917 | * weight_substep(intermediate_timestep_count) |
|---|
| 1918 | |
|---|
| 1919 | ENDDO |
|---|
| 1920 | |
|---|
| 1921 | DO k = nzb_max+1, nzt |
|---|
| 1922 | |
|---|
| 1923 | u_comp = u(k+1,j,i+1) + u(k,j,i+1) - gu |
|---|
| 1924 | flux_r(k) = u_comp * ( & |
|---|
| 1925 | 37.0 * ( w(k,j,i+1) + w(k,j,i) ) & |
|---|
| 1926 | - 8.0 * ( w(k,j,i+2) + w(k,j,i-1) ) & |
|---|
| 1927 | + ( w(k,j,i+3) + w(k,j,i-2) ) ) * adv_mom_5 |
|---|
| 1928 | |
|---|
| 1929 | diss_r(k) = - ABS( u_comp ) * ( & |
|---|
| 1930 | 10.0 * ( w(k,j,i+1) - w(k,j,i) ) & |
|---|
| 1931 | - 5.0 * ( w(k,j,i+2) - w(k,j,i-1) ) & |
|---|
| 1932 | + ( w(k,j,i+3) - w(k,j,i-2) ) ) * adv_mom_5 |
|---|
| 1933 | |
|---|
| 1934 | v_comp = v(k+1,j+1,i) + v(k,j+1,i) - gv |
|---|
| 1935 | flux_n(k) = v_comp * ( & |
|---|
| 1936 | 37.0 * ( w(k,j+1,i) + w(k,j,i) ) & |
|---|
| 1937 | - 8.0 * ( w(k,j+2,i) + w(k,j-1,i) ) & |
|---|
| 1938 | + ( w(k,j+3,i) + w(k,j-2,i) ) ) * adv_mom_5 |
|---|
| 1939 | |
|---|
| 1940 | diss_n(k) = - ABS( v_comp ) * ( & |
|---|
| 1941 | 10.0 * ( w(k,j+1,i) - w(k,j,i) ) & |
|---|
| 1942 | - 5.0 * ( w(k,j+2,i) - w(k,j-1,i) ) & |
|---|
| 1943 | + ( w(k,j+3,i) - w(k,j-2,i) ) ) * adv_mom_5 |
|---|
| 1944 | ! |
|---|
| 1945 | !-- k index has to be modified near bottom and top, else array |
|---|
| 1946 | !-- subscripts will be exceeded. |
|---|
| 1947 | ibit35 = IBITS(wall_flags_00(k,j,i),3,1) |
|---|
| 1948 | ibit34 = IBITS(wall_flags_00(k,j,i),2,1) |
|---|
| 1949 | ibit33 = IBITS(wall_flags_00(k,j,i),1,1) |
|---|
| 1950 | |
|---|
| 1951 | k_ppp = k + 3 * ibit35 |
|---|
| 1952 | k_pp = k + 2 * ( 1 - ibit33 ) |
|---|
| 1953 | k_mm = k - 2 * ibit35 |
|---|
| 1954 | |
|---|
| 1955 | w_comp = w(k+1,j,i) + w(k,j,i) |
|---|
| 1956 | flux_t(k) = w_comp * ( & |
|---|
| 1957 | ( 37.0 * ibit35 * adv_mom_5 & |
|---|
| 1958 | + 7.0 * ibit34 * adv_mom_3 & |
|---|
| 1959 | + ibit33 * adv_mom_1 & |
|---|
| 1960 | ) * & |
|---|
| 1961 | ( w(k+1,j,i) + w(k,j,i) ) & |
|---|
| 1962 | - ( 8.0 * ibit35 * adv_mom_5 & |
|---|
| 1963 | + ibit34 * adv_mom_3 & |
|---|
| 1964 | ) * & |
|---|
| 1965 | ( w(k_pp,j,i) + w(k-1,j,i) ) & |
|---|
| 1966 | + ( ibit35 * adv_mom_5 & |
|---|
| 1967 | ) * & |
|---|
| 1968 | ( w(k_ppp,j,i) + w(k_mm,j,i) ) & |
|---|
| 1969 | ) |
|---|
| 1970 | |
|---|
| 1971 | diss_t(k) = - ABS( w_comp ) * ( & |
|---|
| 1972 | ( 10.0 * ibit35 * adv_mom_5 & |
|---|
| 1973 | + 3.0 * ibit34 * adv_mom_3 & |
|---|
| 1974 | + ibit33 * adv_mom_1 & |
|---|
| 1975 | ) * & |
|---|
| 1976 | ( w(k+1,j,i) - w(k,j,i) ) & |
|---|
| 1977 | - ( 5.0 * ibit35 * adv_mom_5 & |
|---|
| 1978 | + ibit34 * adv_mom_3 & |
|---|
| 1979 | ) * & |
|---|
| 1980 | ( w(k_pp,j,i) - w(k-1,j,i) ) & |
|---|
| 1981 | + ( ibit35 * adv_mom_5 & |
|---|
| 1982 | ) * & |
|---|
| 1983 | ( w(k_ppp,j,i) - w(k_mm,j,i) ) & |
|---|
| 1984 | ) |
|---|
| 1985 | ! |
|---|
| 1986 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 1987 | !-- correction is needed to overcome numerical instabilities introduced |
|---|
| 1988 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 1989 | div = ( ( u_comp + gu - ( u(k+1,j,i) + u(k,j,i) ) ) * ddx & |
|---|
| 1990 | + ( v_comp + gv - ( v(k+1,j,i) + v(k,j,i) ) ) * ddy & |
|---|
| 1991 | + ( w_comp - ( w(k,j,i) + w(k-1,j,i) ) ) * ddzu(k+1) & |
|---|
| 1992 | ) * 0.5 |
|---|
| 1993 | |
|---|
| 1994 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 1995 | ( flux_r(k) + diss_r(k) & |
|---|
| 1996 | - flux_l_w(k,j,tn) - diss_l_w(k,j,tn) ) * ddx & |
|---|
| 1997 | + ( flux_n(k) + diss_n(k) & |
|---|
| 1998 | - flux_s_w(k,tn) - diss_s_w(k,tn) ) * ddy & |
|---|
| 1999 | + ( flux_t(k) + diss_t(k) & |
|---|
| 2000 | - flux_d - diss_d ) * ddzu(k+1) & |
|---|
| 2001 | ) + div * w(k,j,i) |
|---|
| 2002 | |
|---|
| 2003 | flux_l_w(k,j,tn) = flux_r(k) |
|---|
| 2004 | diss_l_w(k,j,tn) = diss_r(k) |
|---|
| 2005 | flux_s_w(k,tn) = flux_n(k) |
|---|
| 2006 | diss_s_w(k,tn) = diss_n(k) |
|---|
| 2007 | flux_d = flux_t(k) |
|---|
| 2008 | diss_d = diss_t(k) |
|---|
| 2009 | ! |
|---|
| 2010 | !-- Statistical Evaluation of w'w'. |
|---|
| 2011 | sums_ws2_ws_l(k,tn) = sums_ws2_ws_l(k,tn) & |
|---|
| 2012 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 2013 | * weight_substep(intermediate_timestep_count) |
|---|
| 2014 | |
|---|
| 2015 | ENDDO |
|---|
| 2016 | |
|---|
| 2017 | |
|---|
| 2018 | END SUBROUTINE advec_w_ws_ij |
|---|
| 2019 | |
|---|
| 2020 | |
|---|
| 2021 | !------------------------------------------------------------------------------! |
|---|
| 2022 | ! Scalar advection - Call for all grid points |
|---|
| 2023 | !------------------------------------------------------------------------------! |
|---|
| 2024 | SUBROUTINE advec_s_ws( sk, sk_char ) |
|---|
| 2025 | |
|---|
| 2026 | USE arrays_3d |
|---|
| 2027 | USE constants |
|---|
| 2028 | USE control_parameters |
|---|
| 2029 | USE grid_variables |
|---|
| 2030 | USE indices |
|---|
| 2031 | USE statistics |
|---|
| 2032 | |
|---|
| 2033 | IMPLICIT NONE |
|---|
| 2034 | |
|---|
| 2035 | INTEGER :: i, ibit0, ibit1, ibit2, ibit3, ibit4, ibit5, ibit6, & |
|---|
| 2036 | ibit7, ibit8, j, k, k_mm, k_pp, k_ppp, tn = 0 |
|---|
| 2037 | #if defined( __nopointer ) |
|---|
| 2038 | REAL, DIMENSION(nzb:nzt+1,nysg:nyng,nxlg:nxrg) :: sk |
|---|
| 2039 | #else |
|---|
| 2040 | REAL, DIMENSION(:,:,:), POINTER :: sk |
|---|
| 2041 | #endif |
|---|
| 2042 | REAL :: diss_d, div, flux_d, u_comp, v_comp |
|---|
| 2043 | REAL, DIMENSION(nzb:nzt) :: diss_n, diss_r, diss_t, flux_n, flux_r, & |
|---|
| 2044 | flux_t |
|---|
| 2045 | REAL, DIMENSION(nzb+1:nzt) :: swap_diss_y_local, swap_flux_y_local |
|---|
| 2046 | REAL, DIMENSION(nzb+1:nzt,nys:nyn) :: swap_diss_x_local, & |
|---|
| 2047 | swap_flux_x_local |
|---|
| 2048 | CHARACTER (LEN = *), INTENT(IN) :: sk_char |
|---|
| 2049 | |
|---|
| 2050 | ! |
|---|
| 2051 | !-- Compute the fluxes for the whole left boundary of the processor domain. |
|---|
| 2052 | i = nxl |
|---|
| 2053 | DO j = nys, nyn |
|---|
| 2054 | |
|---|
| 2055 | DO k = nzb+1, nzb_max |
|---|
| 2056 | |
|---|
| 2057 | ibit2 = IBITS(wall_flags_0(k,j,i),2,1) |
|---|
| 2058 | ibit1 = IBITS(wall_flags_0(k,j,i),1,1) |
|---|
| 2059 | ibit0 = IBITS(wall_flags_0(k,j,i),0,1) |
|---|
| 2060 | |
|---|
| 2061 | u_comp = u(k,j,i) - u_gtrans |
|---|
| 2062 | swap_flux_x_local(k,j) = u_comp * ( & |
|---|
| 2063 | ( 37.0 * ibit2 * adv_sca_5 & |
|---|
| 2064 | + 7.0 * ibit1 * adv_sca_3 & |
|---|
| 2065 | + ibit0 * adv_sca_1 & |
|---|
| 2066 | ) * & |
|---|
| 2067 | ( sk(k,j,i) + sk(k,j,i-1) ) & |
|---|
| 2068 | - ( 8.0 * ibit2 * adv_sca_5 & |
|---|
| 2069 | + ibit1 * adv_sca_3 & |
|---|
| 2070 | ) * & |
|---|
| 2071 | ( sk(k,j,i+1) + sk(k,j,i-2) ) & |
|---|
| 2072 | + ( ibit2 * adv_sca_5 & |
|---|
| 2073 | ) * & |
|---|
| 2074 | ( sk(k,j,i+2) + sk(k,j,i-3) ) & |
|---|
| 2075 | ) |
|---|
| 2076 | |
|---|
| 2077 | swap_diss_x_local(k,j) = -ABS( u_comp ) * ( & |
|---|
| 2078 | ( 10.0 * ibit2 * adv_sca_5 & |
|---|
| 2079 | + 3.0 * ibit1 * adv_sca_3 & |
|---|
| 2080 | + ibit0 * adv_sca_1 & |
|---|
| 2081 | ) * & |
|---|
| 2082 | ( sk(k,j,i) - sk(k,j,i-1) ) & |
|---|
| 2083 | - ( 5.0 * ibit2 * adv_sca_5 & |
|---|
| 2084 | + ibit1 * adv_sca_3 & |
|---|
| 2085 | ) * & |
|---|
| 2086 | ( sk(k,j,i+1) - sk(k,j,i-2) ) & |
|---|
| 2087 | + ( ibit2 * adv_sca_5 & |
|---|
| 2088 | ) * & |
|---|
| 2089 | ( sk(k,j,i+2) - sk(k,j,i-3) ) & |
|---|
| 2090 | ) |
|---|
| 2091 | |
|---|
| 2092 | ENDDO |
|---|
| 2093 | |
|---|
| 2094 | DO k = nzb_max+1, nzt |
|---|
| 2095 | |
|---|
| 2096 | u_comp = u(k,j,i) - u_gtrans |
|---|
| 2097 | swap_flux_x_local(k,j) = u_comp * ( & |
|---|
| 2098 | 37.0 * ( sk(k,j,i) + sk(k,j,i-1) ) & |
|---|
| 2099 | - 8.0 * ( sk(k,j,i+1) + sk(k,j,i-2) ) & |
|---|
| 2100 | + ( sk(k,j,i+2) + sk(k,j,i-3) ) & |
|---|
| 2101 | ) * adv_sca_5 |
|---|
| 2102 | |
|---|
| 2103 | swap_diss_x_local(k,j) = -ABS( u_comp ) * ( & |
|---|
| 2104 | 10.0 * ( sk(k,j,i) - sk(k,j,i-1) ) & |
|---|
| 2105 | - 5.0 * ( sk(k,j,i+1) - sk(k,j,i-2) ) & |
|---|
| 2106 | + ( sk(k,j,i+2) - sk(k,j,i-3) ) & |
|---|
| 2107 | ) * adv_sca_5 |
|---|
| 2108 | |
|---|
| 2109 | ENDDO |
|---|
| 2110 | |
|---|
| 2111 | ENDDO |
|---|
| 2112 | |
|---|
| 2113 | DO i = nxl, nxr |
|---|
| 2114 | |
|---|
| 2115 | j = nys |
|---|
| 2116 | DO k = nzb+1, nzb_max |
|---|
| 2117 | |
|---|
| 2118 | ibit5 = IBITS(wall_flags_0(k,j,i),5,1) |
|---|
| 2119 | ibit4 = IBITS(wall_flags_0(k,j,i),4,1) |
|---|
| 2120 | ibit3 = IBITS(wall_flags_0(k,j,i),3,1) |
|---|
| 2121 | |
|---|
| 2122 | v_comp = v(k,j,i) - v_gtrans |
|---|
| 2123 | swap_flux_y_local(k) = v_comp * ( & |
|---|
| 2124 | ( 37.0 * ibit5 * adv_sca_5 & |
|---|
| 2125 | + 7.0 * ibit4 * adv_sca_3 & |
|---|
| 2126 | + ibit3 * adv_sca_1 & |
|---|
| 2127 | ) * & |
|---|
| 2128 | ( sk(k,j,i) + sk(k,j-1,i) ) & |
|---|
| 2129 | - ( 8.0 * ibit5 * adv_sca_5 & |
|---|
| 2130 | + ibit4 * adv_sca_3 & |
|---|
| 2131 | ) * & |
|---|
| 2132 | ( sk(k,j+1,i) + sk(k,j-2,i) ) & |
|---|
| 2133 | + ( ibit5 * adv_sca_5 & |
|---|
| 2134 | ) * & |
|---|
| 2135 | ( sk(k,j+2,i) + sk(k,j-3,i) ) & |
|---|
| 2136 | ) |
|---|
| 2137 | |
|---|
| 2138 | swap_diss_y_local(k) = -ABS( v_comp ) * ( & |
|---|
| 2139 | ( 10.0 * ibit5 * adv_sca_5 & |
|---|
| 2140 | + 3.0 * ibit4 * adv_sca_3 & |
|---|
| 2141 | + ibit3 * adv_sca_1 & |
|---|
| 2142 | ) * & |
|---|
| 2143 | ( sk(k,j,i) - sk(k,j-1,i) ) & |
|---|
| 2144 | - ( 5.0 * ibit5 * adv_sca_5 & |
|---|
| 2145 | + ibit4 * adv_sca_3 & |
|---|
| 2146 | ) * & |
|---|
| 2147 | ( sk(k,j+1,i) - sk(k,j-2,i) ) & |
|---|
| 2148 | + ( ibit5 * adv_sca_5 & |
|---|
| 2149 | ) * & |
|---|
| 2150 | ( sk(k,j+2,i) - sk(k,j-3,i) ) & |
|---|
| 2151 | ) |
|---|
| 2152 | |
|---|
| 2153 | ENDDO |
|---|
| 2154 | ! |
|---|
| 2155 | !-- Above to the top of the highest topography. No degradation necessary. |
|---|
| 2156 | DO k = nzb_max+1, nzt |
|---|
| 2157 | |
|---|
| 2158 | v_comp = v(k,j,i) - v_gtrans |
|---|
| 2159 | swap_flux_y_local(k) = v_comp * ( & |
|---|
| 2160 | 37.0 * ( sk(k,j,i) + sk(k,j-1,i) ) & |
|---|
| 2161 | - 8.0 * ( sk(k,j+1,i) + sk(k,j-2,i) ) & |
|---|
| 2162 | + ( sk(k,j+2,i) + sk(k,j-3,i) ) & |
|---|
| 2163 | ) * adv_sca_5 |
|---|
| 2164 | swap_diss_y_local(k) = -ABS( v_comp ) * ( & |
|---|
| 2165 | 10.0 * ( sk(k,j,i) - sk(k,j-1,i) ) & |
|---|
| 2166 | - 5.0 * ( sk(k,j+1,i) - sk(k,j-2,i) ) & |
|---|
| 2167 | + sk(k,j+2,i) - sk(k,j-3,i) & |
|---|
| 2168 | ) * adv_sca_5 |
|---|
| 2169 | |
|---|
| 2170 | ENDDO |
|---|
| 2171 | |
|---|
| 2172 | DO j = nys, nyn |
|---|
| 2173 | |
|---|
| 2174 | flux_t(0) = 0.0 |
|---|
| 2175 | diss_t(0) = 0.0 |
|---|
| 2176 | flux_d = 0.0 |
|---|
| 2177 | diss_d = 0.0 |
|---|
| 2178 | |
|---|
| 2179 | DO k = nzb+1, nzb_max |
|---|
| 2180 | |
|---|
| 2181 | ibit2 = IBITS(wall_flags_0(k,j,i),2,1) |
|---|
| 2182 | ibit1 = IBITS(wall_flags_0(k,j,i),1,1) |
|---|
| 2183 | ibit0 = IBITS(wall_flags_0(k,j,i),0,1) |
|---|
| 2184 | |
|---|
| 2185 | u_comp = u(k,j,i+1) - u_gtrans |
|---|
| 2186 | flux_r(k) = u_comp * ( & |
|---|
| 2187 | ( 37.0 * ibit2 * adv_sca_5 & |
|---|
| 2188 | + 7.0 * ibit1 * adv_sca_3 & |
|---|
| 2189 | + ibit0 * adv_sca_1 & |
|---|
| 2190 | ) * & |
|---|
| 2191 | ( sk(k,j,i+1) + sk(k,j,i) ) & |
|---|
| 2192 | - ( 8.0 * ibit2 * adv_sca_5 & |
|---|
| 2193 | + ibit1 * adv_sca_3 & |
|---|
| 2194 | ) * & |
|---|
| 2195 | ( sk(k,j,i+2) + sk(k,j,i-1) ) & |
|---|
| 2196 | + ( ibit2 * adv_sca_5 & |
|---|
| 2197 | ) * & |
|---|
| 2198 | ( sk(k,j,i+3) + sk(k,j,i-2) ) & |
|---|
| 2199 | ) |
|---|
| 2200 | |
|---|
| 2201 | diss_r(k) = -ABS( u_comp ) * ( & |
|---|
| 2202 | ( 10.0 * ibit2 * adv_sca_5 & |
|---|
| 2203 | + 3.0 * ibit1 * adv_sca_3 & |
|---|
| 2204 | + ibit0 * adv_sca_1 & |
|---|
| 2205 | ) * & |
|---|
| 2206 | ( sk(k,j,i+1) - sk(k,j,i) ) & |
|---|
| 2207 | - ( 5.0 * ibit2 * adv_sca_5 & |
|---|
| 2208 | + ibit1 * adv_sca_3 & |
|---|
| 2209 | ) * & |
|---|
| 2210 | ( sk(k,j,i+2) - sk(k,j,i-1) ) & |
|---|
| 2211 | + ( ibit2 * adv_sca_5 & |
|---|
| 2212 | ) * & |
|---|
| 2213 | ( sk(k,j,i+3) - sk(k,j,i-2) ) & |
|---|
| 2214 | ) |
|---|
| 2215 | |
|---|
| 2216 | ibit5 = IBITS(wall_flags_0(k,j,i),5,1) |
|---|
| 2217 | ibit4 = IBITS(wall_flags_0(k,j,i),4,1) |
|---|
| 2218 | ibit3 = IBITS(wall_flags_0(k,j,i),3,1) |
|---|
| 2219 | |
|---|
| 2220 | v_comp = v(k,j+1,i) - v_gtrans |
|---|
| 2221 | flux_n(k) = v_comp * ( & |
|---|
| 2222 | ( 37.0 * ibit5 * adv_sca_5 & |
|---|
| 2223 | + 7.0 * ibit4 * adv_sca_3 & |
|---|
| 2224 | + ibit3 * adv_sca_1 & |
|---|
| 2225 | ) * & |
|---|
| 2226 | ( sk(k,j+1,i) + sk(k,j,i) ) & |
|---|
| 2227 | - ( 8.0 * ibit5 * adv_sca_5 & |
|---|
| 2228 | + ibit4 * adv_sca_3 & |
|---|
| 2229 | ) * & |
|---|
| 2230 | ( sk(k,j+2,i) + sk(k,j-1,i) ) & |
|---|
| 2231 | + ( ibit5 * adv_sca_5 & |
|---|
| 2232 | ) * & |
|---|
| 2233 | ( sk(k,j+3,i) + sk(k,j-2,i) ) & |
|---|
| 2234 | ) |
|---|
| 2235 | |
|---|
| 2236 | diss_n(k) = -ABS( v_comp ) * ( & |
|---|
| 2237 | ( 10.0 * ibit5 * adv_sca_5 & |
|---|
| 2238 | + 3.0 * ibit4 * adv_sca_3 & |
|---|
| 2239 | + ibit3 * adv_sca_1 & |
|---|
| 2240 | ) * & |
|---|
| 2241 | ( sk(k,j+1,i) - sk(k,j,i) ) & |
|---|
| 2242 | - ( 5.0 * ibit5 * adv_sca_5 & |
|---|
| 2243 | + ibit4 * adv_sca_3 & |
|---|
| 2244 | ) * & |
|---|
| 2245 | ( sk(k,j+2,i) - sk(k,j-1,i) ) & |
|---|
| 2246 | + ( ibit5 * adv_sca_5 & |
|---|
| 2247 | ) * & |
|---|
| 2248 | ( sk(k,j+3,i) - sk(k,j-2,i) ) & |
|---|
| 2249 | ) |
|---|
| 2250 | ! |
|---|
| 2251 | !-- k index has to be modified near bottom and top, else array |
|---|
| 2252 | !-- subscripts will be exceeded. |
|---|
| 2253 | ibit8 = IBITS(wall_flags_0(k,j,i),8,1) |
|---|
| 2254 | ibit7 = IBITS(wall_flags_0(k,j,i),7,1) |
|---|
| 2255 | ibit6 = IBITS(wall_flags_0(k,j,i),6,1) |
|---|
| 2256 | |
|---|
| 2257 | k_ppp = k + 3 * ibit8 |
|---|
| 2258 | k_pp = k + 2 * ( 1 - ibit6 ) |
|---|
| 2259 | k_mm = k - 2 * ibit8 |
|---|
| 2260 | |
|---|
| 2261 | |
|---|
| 2262 | flux_t(k) = w(k,j,i) * ( & |
|---|
| 2263 | ( 37.0 * ibit8 * adv_sca_5 & |
|---|
| 2264 | + 7.0 * ibit7 * adv_sca_3 & |
|---|
| 2265 | + ibit6 * adv_sca_1 & |
|---|
| 2266 | ) * & |
|---|
| 2267 | ( sk(k+1,j,i) + sk(k,j,i) ) & |
|---|
| 2268 | - ( 8.0 * ibit8 * adv_sca_5 & |
|---|
| 2269 | + ibit7 * adv_sca_3 & |
|---|
| 2270 | ) * & |
|---|
| 2271 | ( sk(k_pp,j,i) + sk(k-1,j,i) ) & |
|---|
| 2272 | + ( ibit8 * adv_sca_5 & |
|---|
| 2273 | ) * ( sk(k_ppp,j,i)+ sk(k_mm,j,i) ) & |
|---|
| 2274 | ) |
|---|
| 2275 | |
|---|
| 2276 | diss_t(k) = -ABS( w(k,j,i) ) * ( & |
|---|
| 2277 | ( 10.0 * ibit8 * adv_sca_5 & |
|---|
| 2278 | + 3.0 * ibit7 * adv_sca_3 & |
|---|
| 2279 | + ibit6 * adv_sca_1 & |
|---|
| 2280 | ) * & |
|---|
| 2281 | ( sk(k+1,j,i) - sk(k,j,i) ) & |
|---|
| 2282 | - ( 5.0 * ibit8 * adv_sca_5 & |
|---|
| 2283 | + ibit7 * adv_sca_3 & |
|---|
| 2284 | ) * & |
|---|
| 2285 | ( sk(k_pp,j,i) - sk(k-1,j,i) ) & |
|---|
| 2286 | + ( ibit8 * adv_sca_5 & |
|---|
| 2287 | ) * & |
|---|
| 2288 | ( sk(k_ppp,j,i) - sk(k_mm,j,i) ) & |
|---|
| 2289 | ) |
|---|
| 2290 | ! |
|---|
| 2291 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 2292 | !-- correction is needed to overcome numerical instabilities caused |
|---|
| 2293 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 2294 | div = ( u(k,j,i+1) - u(k,j,i) ) * ddx & |
|---|
| 2295 | + ( v(k,j+1,i) - v(k,j,i) ) * ddy & |
|---|
| 2296 | + ( w(k,j,i) - w(k-1,j,i) ) * ddzw(k) |
|---|
| 2297 | |
|---|
| 2298 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 2299 | ( flux_r(k) + diss_r(k) - swap_flux_x_local(k,j) - & |
|---|
| 2300 | swap_diss_x_local(k,j) ) * ddx & |
|---|
| 2301 | + ( flux_n(k) + diss_n(k) - swap_flux_y_local(k) - & |
|---|
| 2302 | swap_diss_y_local(k) ) * ddy & |
|---|
| 2303 | + ( flux_t(k) + diss_t(k) - flux_d - diss_d & |
|---|
| 2304 | ) * ddzw(k) & |
|---|
| 2305 | ) + sk(k,j,i) * div |
|---|
| 2306 | |
|---|
| 2307 | swap_flux_y_local(k) = flux_n(k) |
|---|
| 2308 | swap_diss_y_local(k) = diss_n(k) |
|---|
| 2309 | swap_flux_x_local(k,j) = flux_r(k) |
|---|
| 2310 | swap_diss_x_local(k,j) = diss_r(k) |
|---|
| 2311 | flux_d = flux_t(k) |
|---|
| 2312 | diss_d = diss_t(k) |
|---|
| 2313 | |
|---|
| 2314 | ENDDO |
|---|
| 2315 | |
|---|
| 2316 | DO k = nzb_max+1, nzt |
|---|
| 2317 | |
|---|
| 2318 | u_comp = u(k,j,i+1) - u_gtrans |
|---|
| 2319 | flux_r(k) = u_comp * ( & |
|---|
| 2320 | 37.0 * ( sk(k,j,i+1) + sk(k,j,i) ) & |
|---|
| 2321 | - 8.0 * ( sk(k,j,i+2) + sk(k,j,i-1) ) & |
|---|
| 2322 | + ( sk(k,j,i+3) + sk(k,j,i-2) ) ) * adv_sca_5 |
|---|
| 2323 | diss_r(k) = -ABS( u_comp ) * ( & |
|---|
| 2324 | 10.0 * ( sk(k,j,i+1) - sk(k,j,i) ) & |
|---|
| 2325 | - 5.0 * ( sk(k,j,i+2) - sk(k,j,i-1) ) & |
|---|
| 2326 | + ( sk(k,j,i+3) - sk(k,j,i-2) ) ) * adv_sca_5 |
|---|
| 2327 | |
|---|
| 2328 | v_comp = v(k,j+1,i) - v_gtrans |
|---|
| 2329 | flux_n(k) = v_comp * ( & |
|---|
| 2330 | 37.0 * ( sk(k,j+1,i) + sk(k,j,i) ) & |
|---|
| 2331 | - 8.0 * ( sk(k,j+2,i) + sk(k,j-1,i) ) & |
|---|
| 2332 | + ( sk(k,j+3,i) + sk(k,j-2,i) ) ) * adv_sca_5 |
|---|
| 2333 | diss_n(k) = -ABS( v_comp ) * ( & |
|---|
| 2334 | 10.0 * ( sk(k,j+1,i) - sk(k,j,i) ) & |
|---|
| 2335 | - 5.0 * ( sk(k,j+2,i) - sk(k,j-1,i) ) & |
|---|
| 2336 | + ( sk(k,j+3,i) - sk(k,j-2,i) ) ) * adv_sca_5 |
|---|
| 2337 | ! |
|---|
| 2338 | !-- k index has to be modified near bottom and top, else array |
|---|
| 2339 | !-- subscripts will be exceeded. |
|---|
| 2340 | ibit8 = IBITS(wall_flags_0(k,j,i),8,1) |
|---|
| 2341 | ibit7 = IBITS(wall_flags_0(k,j,i),7,1) |
|---|
| 2342 | ibit6 = IBITS(wall_flags_0(k,j,i),6,1) |
|---|
| 2343 | |
|---|
| 2344 | k_ppp = k + 3 * ibit8 |
|---|
| 2345 | k_pp = k + 2 * ( 1 - ibit6 ) |
|---|
| 2346 | k_mm = k - 2 * ibit8 |
|---|
| 2347 | |
|---|
| 2348 | |
|---|
| 2349 | flux_t(k) = w(k,j,i) * ( & |
|---|
| 2350 | ( 37.0 * ibit8 * adv_sca_5 & |
|---|
| 2351 | + 7.0 * ibit7 * adv_sca_3 & |
|---|
| 2352 | + ibit6 * adv_sca_1 & |
|---|
| 2353 | ) * & |
|---|
| 2354 | ( sk(k+1,j,i) + sk(k,j,i) ) & |
|---|
| 2355 | - ( 8.0 * ibit8 * adv_sca_5 & |
|---|
| 2356 | + ibit7 * adv_sca_3 & |
|---|
| 2357 | ) * & |
|---|
| 2358 | ( sk(k_pp,j,i) + sk(k-1,j,i) ) & |
|---|
| 2359 | + ( ibit8 * adv_sca_5 & |
|---|
| 2360 | ) * ( sk(k_ppp,j,i)+ sk(k_mm,j,i) ) & |
|---|
| 2361 | ) |
|---|
| 2362 | |
|---|
| 2363 | diss_t(k) = -ABS( w(k,j,i) ) * ( & |
|---|
| 2364 | ( 10.0 * ibit8 * adv_sca_5 & |
|---|
| 2365 | + 3.0 * ibit7 * adv_sca_3 & |
|---|
| 2366 | + ibit6 * adv_sca_1 & |
|---|
| 2367 | ) * & |
|---|
| 2368 | ( sk(k+1,j,i) - sk(k,j,i) ) & |
|---|
| 2369 | - ( 5.0 * ibit8 * adv_sca_5 & |
|---|
| 2370 | + ibit7 * adv_sca_3 & |
|---|
| 2371 | ) * & |
|---|
| 2372 | ( sk(k_pp,j,i) - sk(k-1,j,i) ) & |
|---|
| 2373 | + ( ibit8 * adv_sca_5 & |
|---|
| 2374 | ) * & |
|---|
| 2375 | ( sk(k_ppp,j,i) - sk(k_mm,j,i) ) & |
|---|
| 2376 | ) |
|---|
| 2377 | ! |
|---|
| 2378 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 2379 | !-- correction is needed to overcome numerical instabilities introduced |
|---|
| 2380 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 2381 | div = ( u(k,j,i+1) - u(k,j,i) ) * ddx & |
|---|
| 2382 | + ( v(k,j+1,i) - v(k,j,i) ) * ddy & |
|---|
| 2383 | + ( w(k,j,i) - w(k-1,j,i) ) * ddzw(k) |
|---|
| 2384 | |
|---|
| 2385 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 2386 | ( flux_r(k) + diss_r(k) - swap_flux_x_local(k,j) - & |
|---|
| 2387 | swap_diss_x_local(k,j) ) * ddx & |
|---|
| 2388 | + ( flux_n(k) + diss_n(k) - swap_flux_y_local(k) - & |
|---|
| 2389 | swap_diss_y_local(k) ) * ddy & |
|---|
| 2390 | + ( flux_t(k) + diss_t(k) - flux_d - diss_d & |
|---|
| 2391 | ) * ddzw(k) & |
|---|
| 2392 | ) + sk(k,j,i) * div |
|---|
| 2393 | |
|---|
| 2394 | swap_flux_y_local(k) = flux_n(k) |
|---|
| 2395 | swap_diss_y_local(k) = diss_n(k) |
|---|
| 2396 | swap_flux_x_local(k,j) = flux_r(k) |
|---|
| 2397 | swap_diss_x_local(k,j) = diss_r(k) |
|---|
| 2398 | flux_d = flux_t(k) |
|---|
| 2399 | diss_d = diss_t(k) |
|---|
| 2400 | |
|---|
| 2401 | ENDDO |
|---|
| 2402 | ! |
|---|
| 2403 | !-- evaluation of statistics |
|---|
| 2404 | SELECT CASE ( sk_char ) |
|---|
| 2405 | |
|---|
| 2406 | CASE ( 'pt' ) |
|---|
| 2407 | DO k = nzb, nzt |
|---|
| 2408 | sums_wspts_ws_l(k,tn) = sums_wspts_ws_l(k,tn) & |
|---|
| 2409 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 2410 | * weight_substep(intermediate_timestep_count) |
|---|
| 2411 | ENDDO |
|---|
| 2412 | CASE ( 'sa' ) |
|---|
| 2413 | DO k = nzb, nzt |
|---|
| 2414 | sums_wssas_ws_l(k,tn) = sums_wssas_ws_l(k,tn) & |
|---|
| 2415 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 2416 | * weight_substep(intermediate_timestep_count) |
|---|
| 2417 | ENDDO |
|---|
| 2418 | CASE ( 'q' ) |
|---|
| 2419 | DO k = nzb, nzt |
|---|
| 2420 | sums_wsqs_ws_l(k,tn) = sums_wsqs_ws_l(k,tn) & |
|---|
| 2421 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 2422 | * weight_substep(intermediate_timestep_count) |
|---|
| 2423 | ENDDO |
|---|
| 2424 | |
|---|
| 2425 | END SELECT |
|---|
| 2426 | |
|---|
| 2427 | ENDDO |
|---|
| 2428 | ENDDO |
|---|
| 2429 | |
|---|
| 2430 | END SUBROUTINE advec_s_ws |
|---|
| 2431 | |
|---|
| 2432 | |
|---|
| 2433 | !------------------------------------------------------------------------------! |
|---|
| 2434 | ! Scalar advection - Call for all grid points - accelerator version |
|---|
| 2435 | !------------------------------------------------------------------------------! |
|---|
| 2436 | SUBROUTINE advec_s_ws_acc ( sk, sk_char ) |
|---|
| 2437 | |
|---|
| 2438 | USE arrays_3d |
|---|
| 2439 | USE constants |
|---|
| 2440 | USE control_parameters |
|---|
| 2441 | USE grid_variables |
|---|
| 2442 | USE indices |
|---|
| 2443 | USE statistics |
|---|
| 2444 | |
|---|
| 2445 | IMPLICIT NONE |
|---|
| 2446 | |
|---|
| 2447 | CHARACTER (LEN = *), INTENT(IN) :: sk_char |
|---|
| 2448 | |
|---|
| 2449 | INTEGER :: i, ibit0, ibit1, ibit2, ibit3, ibit4, ibit5, ibit6, & |
|---|
| 2450 | ibit7, ibit8, j, k, k_mm, k_mmm, k_pp, k_ppp, tn = 0 |
|---|
| 2451 | |
|---|
| 2452 | REAL :: diss_d, diss_l, diss_n, diss_r, diss_s, diss_t, div, flux_d, & |
|---|
| 2453 | flux_l, flux_n, flux_r, flux_s, flux_t, u_comp, v_comp |
|---|
| 2454 | |
|---|
| 2455 | REAL, INTENT(IN), DIMENSION(nzb:nzt+1,nysg:nyng,nxlg:nxrg) :: sk |
|---|
| 2456 | |
|---|
| 2457 | ! |
|---|
| 2458 | !-- Computation of fluxes and tendency terms |
|---|
| 2459 | !$acc kernels present( ddzw, sk, tend, u, v, w, wall_flags_0, wall_flags_00 ) |
|---|
| 2460 | !$acc loop |
|---|
| 2461 | DO i = i_left, i_right |
|---|
| 2462 | DO j = j_south, j_north |
|---|
| 2463 | !$acc loop vector( 32 ) |
|---|
| 2464 | DO k = nzb+1, nzt |
|---|
| 2465 | |
|---|
| 2466 | ibit2 = IBITS(wall_flags_0(k,j,i),2,1) |
|---|
| 2467 | ibit1 = IBITS(wall_flags_0(k,j,i),1,1) |
|---|
| 2468 | ibit0 = IBITS(wall_flags_0(k,j,i),0,1) |
|---|
| 2469 | |
|---|
| 2470 | u_comp = u(k,j,i) - u_gtrans |
|---|
| 2471 | flux_l = u_comp * ( & |
|---|
| 2472 | ( 37.0 * ibit2 * adv_sca_5 & |
|---|
| 2473 | + 7.0 * ibit1 * adv_sca_3 & |
|---|
| 2474 | + ibit0 * adv_sca_1 & |
|---|
| 2475 | ) * & |
|---|
| 2476 | ( sk(k,j,i) + sk(k,j,i-1) ) & |
|---|
| 2477 | - ( 8.0 * ibit2 * adv_sca_5 & |
|---|
| 2478 | + ibit1 * adv_sca_3 & |
|---|
| 2479 | ) * & |
|---|
| 2480 | ( sk(k,j,i+1) + sk(k,j,i-2) ) & |
|---|
| 2481 | + ( ibit2 * adv_sca_5 & |
|---|
| 2482 | ) * & |
|---|
| 2483 | ( sk(k,j,i+2) + sk(k,j,i-3) ) & |
|---|
| 2484 | ) |
|---|
| 2485 | |
|---|
| 2486 | diss_l = -ABS( u_comp ) * ( & |
|---|
| 2487 | ( 10.0 * ibit2 * adv_sca_5 & |
|---|
| 2488 | + 3.0 * ibit1 * adv_sca_3 & |
|---|
| 2489 | + ibit0 * adv_sca_1 & |
|---|
| 2490 | ) * & |
|---|
| 2491 | ( sk(k,j,i) - sk(k,j,i-1) ) & |
|---|
| 2492 | - ( 5.0 * ibit2 * adv_sca_5 & |
|---|
| 2493 | + ibit1 * adv_sca_3 & |
|---|
| 2494 | ) * & |
|---|
| 2495 | ( sk(k,j,i+1) - sk(k,j,i-2) ) & |
|---|
| 2496 | + ( ibit2 * adv_sca_5 & |
|---|
| 2497 | ) * & |
|---|
| 2498 | ( sk(k,j,i+2) - sk(k,j,i-3) ) & |
|---|
| 2499 | ) |
|---|
| 2500 | |
|---|
| 2501 | u_comp = u(k,j,i+1) - u_gtrans |
|---|
| 2502 | flux_r = u_comp * ( & |
|---|
| 2503 | ( 37.0 * ibit2 * adv_sca_5 & |
|---|
| 2504 | + 7.0 * ibit1 * adv_sca_3 & |
|---|
| 2505 | + ibit0 * adv_sca_1 & |
|---|
| 2506 | ) * & |
|---|
| 2507 | ( sk(k,j,i+1) + sk(k,j,i) ) & |
|---|
| 2508 | - ( 8.0 * ibit2 * adv_sca_5 & |
|---|
| 2509 | + ibit1 * adv_sca_3 & |
|---|
| 2510 | ) * & |
|---|
| 2511 | ( sk(k,j,i+2) + sk(k,j,i-1) ) & |
|---|
| 2512 | + ( ibit2 * adv_sca_5 & |
|---|
| 2513 | ) * & |
|---|
| 2514 | ( sk(k,j,i+3) + sk(k,j,i-2) ) & |
|---|
| 2515 | ) |
|---|
| 2516 | |
|---|
| 2517 | diss_r = -ABS( u_comp ) * ( & |
|---|
| 2518 | ( 10.0 * ibit2 * adv_sca_5 & |
|---|
| 2519 | + 3.0 * ibit1 * adv_sca_3 & |
|---|
| 2520 | + ibit0 * adv_sca_1 & |
|---|
| 2521 | ) * & |
|---|
| 2522 | ( sk(k,j,i+1) - sk(k,j,i) ) & |
|---|
| 2523 | - ( 5.0 * ibit2 * adv_sca_5 & |
|---|
| 2524 | + ibit1 * adv_sca_3 & |
|---|
| 2525 | ) * & |
|---|
| 2526 | ( sk(k,j,i+2) - sk(k,j,i-1) ) & |
|---|
| 2527 | + ( ibit2 * adv_sca_5 & |
|---|
| 2528 | ) * & |
|---|
| 2529 | ( sk(k,j,i+3) - sk(k,j,i-2) ) & |
|---|
| 2530 | ) |
|---|
| 2531 | |
|---|
| 2532 | ibit5 = IBITS(wall_flags_0(k,j,i),5,1) |
|---|
| 2533 | ibit4 = IBITS(wall_flags_0(k,j,i),4,1) |
|---|
| 2534 | ibit3 = IBITS(wall_flags_0(k,j,i),3,1) |
|---|
| 2535 | |
|---|
| 2536 | v_comp = v(k,j,i) - v_gtrans |
|---|
| 2537 | flux_s = v_comp * ( & |
|---|
| 2538 | ( 37.0 * ibit5 * adv_sca_5 & |
|---|
| 2539 | + 7.0 * ibit4 * adv_sca_3 & |
|---|
| 2540 | + ibit3 * adv_sca_1 & |
|---|
| 2541 | ) * & |
|---|
| 2542 | ( sk(k,j,i) + sk(k,j-1,i) ) & |
|---|
| 2543 | - ( 8.0 * ibit5 * adv_sca_5 & |
|---|
| 2544 | + ibit4 * adv_sca_3 & |
|---|
| 2545 | ) * & |
|---|
| 2546 | ( sk(k,j+1,i) + sk(k,j-2,i) ) & |
|---|
| 2547 | + ( ibit5 * adv_sca_5 & |
|---|
| 2548 | ) * & |
|---|
| 2549 | ( sk(k,j+2,i) + sk(k,j-3,i) ) & |
|---|
| 2550 | ) |
|---|
| 2551 | |
|---|
| 2552 | diss_s = -ABS( v_comp ) * ( & |
|---|
| 2553 | ( 10.0 * ibit5 * adv_sca_5 & |
|---|
| 2554 | + 3.0 * ibit4 * adv_sca_3 & |
|---|
| 2555 | + ibit3 * adv_sca_1 & |
|---|
| 2556 | ) * & |
|---|
| 2557 | ( sk(k,j,i) - sk(k,j-1,i) ) & |
|---|
| 2558 | - ( 5.0 * ibit5 * adv_sca_5 & |
|---|
| 2559 | + ibit4 * adv_sca_3 & |
|---|
| 2560 | ) * & |
|---|
| 2561 | ( sk(k,j+1,i) - sk(k,j-2,i) ) & |
|---|
| 2562 | + ( ibit5 * adv_sca_5 & |
|---|
| 2563 | ) * & |
|---|
| 2564 | ( sk(k,j+2,i) - sk(k,j-3,i) ) & |
|---|
| 2565 | ) |
|---|
| 2566 | |
|---|
| 2567 | |
|---|
| 2568 | v_comp = v(k,j+1,i) - v_gtrans |
|---|
| 2569 | flux_n = v_comp * ( & |
|---|
| 2570 | ( 37.0 * ibit5 * adv_sca_5 & |
|---|
| 2571 | + 7.0 * ibit4 * adv_sca_3 & |
|---|
| 2572 | + ibit3 * adv_sca_1 & |
|---|
| 2573 | ) * & |
|---|
| 2574 | ( sk(k,j+1,i) + sk(k,j,i) ) & |
|---|
| 2575 | - ( 8.0 * ibit5 * adv_sca_5 & |
|---|
| 2576 | + ibit4 * adv_sca_3 & |
|---|
| 2577 | ) * & |
|---|
| 2578 | ( sk(k,j+2,i) + sk(k,j-1,i) ) & |
|---|
| 2579 | + ( ibit5 * adv_sca_5 & |
|---|
| 2580 | ) * & |
|---|
| 2581 | ( sk(k,j+3,i) + sk(k,j-2,i) ) & |
|---|
| 2582 | ) |
|---|
| 2583 | |
|---|
| 2584 | diss_n = -ABS( v_comp ) * ( & |
|---|
| 2585 | ( 10.0 * ibit5 * adv_sca_5 & |
|---|
| 2586 | + 3.0 * ibit4 * adv_sca_3 & |
|---|
| 2587 | + ibit3 * adv_sca_1 & |
|---|
| 2588 | ) * & |
|---|
| 2589 | ( sk(k,j+1,i) - sk(k,j,i) ) & |
|---|
| 2590 | - ( 5.0 * ibit5 * adv_sca_5 & |
|---|
| 2591 | + ibit4 * adv_sca_3 & |
|---|
| 2592 | ) * & |
|---|
| 2593 | ( sk(k,j+2,i) - sk(k,j-1,i) ) & |
|---|
| 2594 | + ( ibit5 * adv_sca_5 & |
|---|
| 2595 | ) * & |
|---|
| 2596 | ( sk(k,j+3,i) - sk(k,j-2,i) ) & |
|---|
| 2597 | ) |
|---|
| 2598 | |
|---|
| 2599 | ! |
|---|
| 2600 | !-- indizes k_m, k_mm, ... should be known at these point |
|---|
| 2601 | ibit8 = IBITS(wall_flags_0(k-1,j,i),8,1) |
|---|
| 2602 | ibit7 = IBITS(wall_flags_0(k-1,j,i),7,1) |
|---|
| 2603 | ibit6 = IBITS(wall_flags_0(k-1,j,i),6,1) |
|---|
| 2604 | |
|---|
| 2605 | k_pp = k + 2 * ibit8 |
|---|
| 2606 | k_mm = k - 2 * ( ibit7 + ibit8 ) |
|---|
| 2607 | k_mmm = k - 3 * ibit8 |
|---|
| 2608 | |
|---|
| 2609 | flux_d = w(k-1,j,i) * ( & |
|---|
| 2610 | ( 37.0 * ibit8 * adv_sca_5 & |
|---|
| 2611 | + 7.0 * ibit7 * adv_sca_3 & |
|---|
| 2612 | + ibit6 * adv_sca_1 & |
|---|
| 2613 | ) * & |
|---|
| 2614 | ( sk(k,j,i) + sk(k-1,j,i) ) & |
|---|
| 2615 | - ( 8.0 * ibit8 * adv_sca_5 & |
|---|
| 2616 | + ibit7 * adv_sca_3 & |
|---|
| 2617 | ) * & |
|---|
| 2618 | ( sk(k+1,j,i) + sk(k_mm,j,i) ) & |
|---|
| 2619 | + ( ibit8 * adv_sca_5 & |
|---|
| 2620 | ) * ( sk(k_pp,j,i)+ sk(k_mmm,j,i) ) & |
|---|
| 2621 | ) |
|---|
| 2622 | |
|---|
| 2623 | diss_d = -ABS( w(k-1,j,i) ) * ( & |
|---|
| 2624 | ( 10.0 * ibit8 * adv_sca_5 & |
|---|
| 2625 | + 3.0 * ibit7 * adv_sca_3 & |
|---|
| 2626 | + ibit6 * adv_sca_1 & |
|---|
| 2627 | ) * & |
|---|
| 2628 | ( sk(k,j,i) - sk(k-1,j,i) ) & |
|---|
| 2629 | - ( 5.0 * ibit8 * adv_sca_5 & |
|---|
| 2630 | + ibit7 * adv_sca_3 & |
|---|
| 2631 | ) * & |
|---|
| 2632 | ( sk(k+1,j,i) - sk(k_mm,j,i) ) & |
|---|
| 2633 | + ( ibit8 * adv_sca_5 & |
|---|
| 2634 | ) * & |
|---|
| 2635 | ( sk(k_pp,j,i) - sk(k_mmm,j,i) ) & |
|---|
| 2636 | ) |
|---|
| 2637 | |
|---|
| 2638 | ibit8 = IBITS(wall_flags_0(k,j,i),8,1) |
|---|
| 2639 | ibit7 = IBITS(wall_flags_0(k,j,i),7,1) |
|---|
| 2640 | ibit6 = IBITS(wall_flags_0(k,j,i),6,1) |
|---|
| 2641 | |
|---|
| 2642 | k_ppp = k + 3 * ibit8 |
|---|
| 2643 | k_pp = k + 2 * ( 1 - ibit6 ) |
|---|
| 2644 | k_mm = k - 2 * ibit8 |
|---|
| 2645 | |
|---|
| 2646 | flux_t = w(k,j,i) * ( & |
|---|
| 2647 | ( 37.0 * ibit8 * adv_sca_5 & |
|---|
| 2648 | + 7.0 * ibit7 * adv_sca_3 & |
|---|
| 2649 | + ibit6 * adv_sca_1 & |
|---|
| 2650 | ) * & |
|---|
| 2651 | ( sk(k+1,j,i) + sk(k,j,i) ) & |
|---|
| 2652 | - ( 8.0 * ibit8 * adv_sca_5 & |
|---|
| 2653 | + ibit7 * adv_sca_3 & |
|---|
| 2654 | ) * & |
|---|
| 2655 | ( sk(k_pp,j,i) + sk(k-1,j,i) ) & |
|---|
| 2656 | + ( ibit8 * adv_sca_5 & |
|---|
| 2657 | ) * ( sk(k_ppp,j,i)+ sk(k_mm,j,i) ) & |
|---|
| 2658 | ) |
|---|
| 2659 | |
|---|
| 2660 | diss_t = -ABS( w(k,j,i) ) * ( & |
|---|
| 2661 | ( 10.0 * ibit8 * adv_sca_5 & |
|---|
| 2662 | + 3.0 * ibit7 * adv_sca_3 & |
|---|
| 2663 | + ibit6 * adv_sca_1 & |
|---|
| 2664 | ) * & |
|---|
| 2665 | ( sk(k+1,j,i) - sk(k,j,i) ) & |
|---|
| 2666 | - ( 5.0 * ibit8 * adv_sca_5 & |
|---|
| 2667 | + ibit7 * adv_sca_3 & |
|---|
| 2668 | ) * & |
|---|
| 2669 | ( sk(k_pp,j,i) - sk(k-1,j,i) ) & |
|---|
| 2670 | + ( ibit8 * adv_sca_5 & |
|---|
| 2671 | ) * & |
|---|
| 2672 | ( sk(k_ppp,j,i) - sk(k_mm,j,i) ) & |
|---|
| 2673 | ) |
|---|
| 2674 | ! |
|---|
| 2675 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 2676 | !-- correction is needed to overcome numerical instabilities caused |
|---|
| 2677 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 2678 | div = ( u(k,j,i+1) - u(k,j,i) ) * ddx & |
|---|
| 2679 | + ( v(k,j+1,i) - v(k,j,i) ) * ddy & |
|---|
| 2680 | + ( w(k,j,i) - w(k-1,j,i) ) * ddzw(k) |
|---|
| 2681 | |
|---|
| 2682 | tend(k,j,i) = - ( & |
|---|
| 2683 | ( flux_r + diss_r - flux_l - diss_l ) * ddx & |
|---|
| 2684 | + ( flux_n + diss_n - flux_s - diss_s ) * ddy & |
|---|
| 2685 | + ( flux_t + diss_t - flux_d - diss_d ) * ddzw(k)& |
|---|
| 2686 | ) + div * sk(k,j,i) |
|---|
| 2687 | |
|---|
| 2688 | !++ |
|---|
| 2689 | !-- Evaluation of statistics |
|---|
| 2690 | ! SELECT CASE ( sk_char ) |
|---|
| 2691 | ! |
|---|
| 2692 | ! CASE ( 'pt' ) |
|---|
| 2693 | ! sums_wspts_ws_l(k,tn) = sums_wspts_ws_l(k,tn) & |
|---|
| 2694 | ! + ( flux_t + diss_t ) & |
|---|
| 2695 | ! * weight_substep(intermediate_timestep_count) |
|---|
| 2696 | ! CASE ( 'sa' ) |
|---|
| 2697 | ! sums_wssas_ws_l(k,tn) = sums_wssas_ws_l(k,tn) & |
|---|
| 2698 | ! + ( flux_t + diss_t ) & |
|---|
| 2699 | ! * weight_substep(intermediate_timestep_count) |
|---|
| 2700 | ! CASE ( 'q' ) |
|---|
| 2701 | ! sums_wsqs_ws_l(k,tn) = sums_wsqs_ws_l(k,tn) & |
|---|
| 2702 | ! + ( flux_t + diss_t ) & |
|---|
| 2703 | ! * weight_substep(intermediate_timestep_count) |
|---|
| 2704 | ! |
|---|
| 2705 | ! END SELECT |
|---|
| 2706 | |
|---|
| 2707 | ENDDO |
|---|
| 2708 | ENDDO |
|---|
| 2709 | ENDDO |
|---|
| 2710 | !$acc end kernels |
|---|
| 2711 | |
|---|
| 2712 | END SUBROUTINE advec_s_ws_acc |
|---|
| 2713 | |
|---|
| 2714 | |
|---|
| 2715 | !------------------------------------------------------------------------------! |
|---|
| 2716 | ! Advection of u - Call for all grid points |
|---|
| 2717 | !------------------------------------------------------------------------------! |
|---|
| 2718 | SUBROUTINE advec_u_ws |
|---|
| 2719 | |
|---|
| 2720 | USE arrays_3d |
|---|
| 2721 | USE constants |
|---|
| 2722 | USE control_parameters |
|---|
| 2723 | USE grid_variables |
|---|
| 2724 | USE indices |
|---|
| 2725 | USE statistics |
|---|
| 2726 | |
|---|
| 2727 | IMPLICIT NONE |
|---|
| 2728 | |
|---|
| 2729 | INTEGER :: i, ibit9, ibit10, ibit11, ibit12, ibit13, ibit14, ibit15, & |
|---|
| 2730 | ibit16, ibit17, j, k, k_mm, k_pp, k_ppp, tn = 0 |
|---|
| 2731 | REAL :: diss_d, div, flux_d, gu, gv, v_comp, w_comp |
|---|
| 2732 | REAL, DIMENSION(nzb+1:nzt) :: swap_diss_y_local_u, swap_flux_y_local_u |
|---|
| 2733 | REAL, DIMENSION(nzb+1:nzt,nys:nyn) :: swap_diss_x_local_u, & |
|---|
| 2734 | swap_flux_x_local_u |
|---|
| 2735 | REAL, DIMENSION(nzb:nzt) :: diss_n, diss_r, diss_t, flux_n, flux_r, & |
|---|
| 2736 | flux_t, u_comp |
|---|
| 2737 | |
|---|
| 2738 | gu = 2.0 * u_gtrans |
|---|
| 2739 | gv = 2.0 * v_gtrans |
|---|
| 2740 | |
|---|
| 2741 | ! |
|---|
| 2742 | !-- Compute the fluxes for the whole left boundary of the processor domain. |
|---|
| 2743 | i = nxlu |
|---|
| 2744 | DO j = nys, nyn |
|---|
| 2745 | DO k = nzb+1, nzb_max |
|---|
| 2746 | |
|---|
| 2747 | ibit11 = IBITS(wall_flags_0(k,j,i),11,1) |
|---|
| 2748 | ibit10 = IBITS(wall_flags_0(k,j,i),10,1) |
|---|
| 2749 | ibit9 = IBITS(wall_flags_0(k,j,i),9,1) |
|---|
| 2750 | |
|---|
| 2751 | u_comp(k) = u(k,j,i) + u(k,j,i-1) - gu |
|---|
| 2752 | swap_flux_x_local_u(k,j) = u_comp(k) * ( & |
|---|
| 2753 | ( 37.0 * ibit11 * adv_mom_5 & |
|---|
| 2754 | + 7.0 * ibit10 * adv_mom_3 & |
|---|
| 2755 | + ibit9 * adv_mom_1 & |
|---|
| 2756 | ) * & |
|---|
| 2757 | ( u(k,j,i) + u(k,j,i-1) ) & |
|---|
| 2758 | - ( 8.0 * ibit11 * adv_mom_5 & |
|---|
| 2759 | + ibit10 * adv_mom_3 & |
|---|
| 2760 | ) * & |
|---|
| 2761 | ( u(k,j,i+1) + u(k,j,i-2) ) & |
|---|
| 2762 | + ( ibit11 * adv_mom_5 & |
|---|
| 2763 | ) * & |
|---|
| 2764 | ( u(k,j,i+2) + u(k,j,i-3) ) & |
|---|
| 2765 | ) |
|---|
| 2766 | |
|---|
| 2767 | swap_diss_x_local_u(k,j) = - ABS( u_comp(k) ) * ( & |
|---|
| 2768 | ( 10.0 * ibit11 * adv_mom_5 & |
|---|
| 2769 | + 3.0 * ibit10 * adv_mom_3 & |
|---|
| 2770 | + ibit9 * adv_mom_1 & |
|---|
| 2771 | ) * & |
|---|
| 2772 | ( u(k,j,i) - u(k,j,i-1) ) & |
|---|
| 2773 | - ( 5.0 * ibit11 * adv_mom_5 & |
|---|
| 2774 | + ibit10 * adv_mom_3 & |
|---|
| 2775 | ) * & |
|---|
| 2776 | ( u(k,j,i+1) - u(k,j,i-2) ) & |
|---|
| 2777 | + ( ibit11 * adv_mom_5 & |
|---|
| 2778 | ) * & |
|---|
| 2779 | ( u(k,j,i+2) - u(k,j,i-3) ) & |
|---|
| 2780 | ) |
|---|
| 2781 | |
|---|
| 2782 | ENDDO |
|---|
| 2783 | |
|---|
| 2784 | DO k = nzb_max+1, nzt |
|---|
| 2785 | |
|---|
| 2786 | u_comp(k) = u(k,j,i) + u(k,j,i-1) - gu |
|---|
| 2787 | swap_flux_x_local_u(k,j) = u_comp(k) * ( & |
|---|
| 2788 | 37.0 * ( u(k,j,i) + u(k,j,i-1) ) & |
|---|
| 2789 | - 8.0 * ( u(k,j,i+1) + u(k,j,i-2) ) & |
|---|
| 2790 | + ( u(k,j,i+2) + u(k,j,i-3) ) ) * adv_mom_5 |
|---|
| 2791 | swap_diss_x_local_u(k,j) = - ABS(u_comp(k)) * ( & |
|---|
| 2792 | 10.0 * ( u(k,j,i) - u(k,j,i-1) ) & |
|---|
| 2793 | - 5.0 * ( u(k,j,i+1) - u(k,j,i-2) ) & |
|---|
| 2794 | + ( u(k,j,i+2) - u(k,j,i-3) ) ) * adv_mom_5 |
|---|
| 2795 | |
|---|
| 2796 | ENDDO |
|---|
| 2797 | ENDDO |
|---|
| 2798 | |
|---|
| 2799 | DO i = nxlu, nxr |
|---|
| 2800 | ! |
|---|
| 2801 | !-- The following loop computes the fluxes for the south boundary points |
|---|
| 2802 | j = nys |
|---|
| 2803 | DO k = nzb+1, nzb_max |
|---|
| 2804 | |
|---|
| 2805 | ibit14 = IBITS(wall_flags_0(k,j,i),14,1) |
|---|
| 2806 | ibit13 = IBITS(wall_flags_0(k,j,i),13,1) |
|---|
| 2807 | ibit12 = IBITS(wall_flags_0(k,j,i),12,1) |
|---|
| 2808 | |
|---|
| 2809 | v_comp = v(k,j,i) + v(k,j,i-1) - gv |
|---|
| 2810 | swap_flux_y_local_u(k) = v_comp * ( & |
|---|
| 2811 | ( 37.0 * ibit14 * adv_mom_5 & |
|---|
| 2812 | + 7.0 * ibit13 * adv_mom_3 & |
|---|
| 2813 | + ibit12 * adv_mom_1 & |
|---|
| 2814 | ) * & |
|---|
| 2815 | ( u(k,j,i) + u(k,j-1,i) ) & |
|---|
| 2816 | - ( 8.0 * ibit14 * adv_mom_5 & |
|---|
| 2817 | + ibit13 * adv_mom_3 & |
|---|
| 2818 | ) * & |
|---|
| 2819 | ( u(k,j+1,i) + u(k,j-2,i) ) & |
|---|
| 2820 | + ( ibit14 * adv_mom_5 & |
|---|
| 2821 | ) * & |
|---|
| 2822 | ( u(k,j+2,i) + u(k,j-3,i) ) & |
|---|
| 2823 | ) |
|---|
| 2824 | |
|---|
| 2825 | swap_diss_y_local_u(k) = - ABS ( v_comp ) * ( & |
|---|
| 2826 | ( 10.0 * ibit14 * adv_mom_5 & |
|---|
| 2827 | + 3.0 * ibit13 * adv_mom_3 & |
|---|
| 2828 | + ibit12 * adv_mom_1 & |
|---|
| 2829 | ) * & |
|---|
| 2830 | ( u(k,j,i) - u(k,j-1,i) ) & |
|---|
| 2831 | - ( 5.0 * ibit14 * adv_mom_5 & |
|---|
| 2832 | + ibit13 * adv_mom_3 & |
|---|
| 2833 | ) * & |
|---|
| 2834 | ( u(k,j+1,i) - u(k,j-2,i) ) & |
|---|
| 2835 | + ( ibit14 * adv_mom_5 & |
|---|
| 2836 | ) * & |
|---|
| 2837 | ( u(k,j+2,i) - u(k,j-3,i) ) & |
|---|
| 2838 | ) |
|---|
| 2839 | |
|---|
| 2840 | ENDDO |
|---|
| 2841 | |
|---|
| 2842 | DO k = nzb_max+1, nzt |
|---|
| 2843 | |
|---|
| 2844 | v_comp = v(k,j,i) + v(k,j,i-1) - gv |
|---|
| 2845 | swap_flux_y_local_u(k) = v_comp * ( & |
|---|
| 2846 | 37.0 * ( u(k,j,i) + u(k,j-1,i) ) & |
|---|
| 2847 | - 8.0 * ( u(k,j+1,i) + u(k,j-2,i) ) & |
|---|
| 2848 | + ( u(k,j+2,i) + u(k,j-3,i) ) ) * adv_mom_5 |
|---|
| 2849 | swap_diss_y_local_u(k) = - ABS(v_comp) * ( & |
|---|
| 2850 | 10.0 * ( u(k,j,i) - u(k,j-1,i) ) & |
|---|
| 2851 | - 5.0 * ( u(k,j+1,i) - u(k,j-2,i) ) & |
|---|
| 2852 | + ( u(k,j+2,i) - u(k,j-3,i) ) ) * adv_mom_5 |
|---|
| 2853 | |
|---|
| 2854 | ENDDO |
|---|
| 2855 | ! |
|---|
| 2856 | !-- Computation of interior fluxes and tendency terms |
|---|
| 2857 | DO j = nys, nyn |
|---|
| 2858 | |
|---|
| 2859 | flux_t(0) = 0.0 |
|---|
| 2860 | diss_t(0) = 0.0 |
|---|
| 2861 | flux_d = 0.0 |
|---|
| 2862 | diss_d = 0.0 |
|---|
| 2863 | |
|---|
| 2864 | DO k = nzb+1, nzb_max |
|---|
| 2865 | |
|---|
| 2866 | ibit11 = IBITS(wall_flags_0(k,j,i),11,1) |
|---|
| 2867 | ibit10 = IBITS(wall_flags_0(k,j,i),10,1) |
|---|
| 2868 | ibit9 = IBITS(wall_flags_0(k,j,i),9,1) |
|---|
| 2869 | |
|---|
| 2870 | u_comp(k) = u(k,j,i+1) + u(k,j,i) |
|---|
| 2871 | flux_r(k) = ( u_comp(k) - gu ) * ( & |
|---|
| 2872 | ( 37.0 * ibit11 * adv_mom_5 & |
|---|
| 2873 | + 7.0 * ibit10 * adv_mom_3 & |
|---|
| 2874 | + ibit9 * adv_mom_1 & |
|---|
| 2875 | ) * & |
|---|
| 2876 | ( u(k,j,i+1) + u(k,j,i) ) & |
|---|
| 2877 | - ( 8.0 * ibit11 * adv_mom_5 & |
|---|
| 2878 | + ibit10 * adv_mom_3 & |
|---|
| 2879 | ) * & |
|---|
| 2880 | ( u(k,j,i+2) + u(k,j,i-1) ) & |
|---|
| 2881 | + ( ibit11 * adv_mom_5 & |
|---|
| 2882 | ) * & |
|---|
| 2883 | ( u(k,j,i+3) + u(k,j,i-2) ) & |
|---|
| 2884 | ) |
|---|
| 2885 | |
|---|
| 2886 | diss_r(k) = - ABS( u_comp(k) - gu ) * ( & |
|---|
| 2887 | ( 10.0 * ibit11 * adv_mom_5 & |
|---|
| 2888 | + 3.0 * ibit10 * adv_mom_3 & |
|---|
| 2889 | + ibit9 * adv_mom_1 & |
|---|
| 2890 | ) * & |
|---|
| 2891 | ( u(k,j,i+1) - u(k,j,i) ) & |
|---|
| 2892 | - ( 5.0 * ibit11 * adv_mom_5 & |
|---|
| 2893 | + ibit10 * adv_mom_3 & |
|---|
| 2894 | ) * & |
|---|
| 2895 | ( u(k,j,i+2) - u(k,j,i-1) ) & |
|---|
| 2896 | + ( ibit11 * adv_mom_5 & |
|---|
| 2897 | ) * & |
|---|
| 2898 | ( u(k,j,i+3) - u(k,j,i-2) ) & |
|---|
| 2899 | ) |
|---|
| 2900 | |
|---|
| 2901 | ibit14 = IBITS(wall_flags_0(k,j,i),14,1) |
|---|
| 2902 | ibit13 = IBITS(wall_flags_0(k,j,i),13,1) |
|---|
| 2903 | ibit12 = IBITS(wall_flags_0(k,j,i),12,1) |
|---|
| 2904 | |
|---|
| 2905 | v_comp = v(k,j+1,i) + v(k,j+1,i-1) - gv |
|---|
| 2906 | flux_n(k) = v_comp * ( & |
|---|
| 2907 | ( 37.0 * ibit14 * adv_mom_5 & |
|---|
| 2908 | + 7.0 * ibit13 * adv_mom_3 & |
|---|
| 2909 | + ibit12 * adv_mom_1 & |
|---|
| 2910 | ) * & |
|---|
| 2911 | ( u(k,j+1,i) + u(k,j,i) ) & |
|---|
| 2912 | - ( 8.0 * ibit14 * adv_mom_5 & |
|---|
| 2913 | + ibit13 * adv_mom_3 & |
|---|
| 2914 | ) * & |
|---|
| 2915 | ( u(k,j+2,i) + u(k,j-1,i) ) & |
|---|
| 2916 | + ( ibit14 * adv_mom_5 & |
|---|
| 2917 | ) * & |
|---|
| 2918 | ( u(k,j+3,i) + u(k,j-2,i) ) & |
|---|
| 2919 | ) |
|---|
| 2920 | |
|---|
| 2921 | diss_n(k) = - ABS ( v_comp ) * ( & |
|---|
| 2922 | ( 10.0 * ibit14 * adv_mom_5 & |
|---|
| 2923 | + 3.0 * ibit13 * adv_mom_3 & |
|---|
| 2924 | + ibit12 * adv_mom_1 & |
|---|
| 2925 | ) * & |
|---|
| 2926 | ( u(k,j+1,i) - u(k,j,i) ) & |
|---|
| 2927 | - ( 5.0 * ibit14 * adv_mom_5 & |
|---|
| 2928 | + ibit13 * adv_mom_3 & |
|---|
| 2929 | ) * & |
|---|
| 2930 | ( u(k,j+2,i) - u(k,j-1,i) ) & |
|---|
| 2931 | + ( ibit14 * adv_mom_5 & |
|---|
| 2932 | ) * & |
|---|
| 2933 | ( u(k,j+3,i) - u(k,j-2,i) ) & |
|---|
| 2934 | ) |
|---|
| 2935 | ! |
|---|
| 2936 | !-- k index has to be modified near bottom and top, else array |
|---|
| 2937 | !-- subscripts will be exceeded. |
|---|
| 2938 | ibit17 = IBITS(wall_flags_0(k,j,i),17,1) |
|---|
| 2939 | ibit16 = IBITS(wall_flags_0(k,j,i),16,1) |
|---|
| 2940 | ibit15 = IBITS(wall_flags_0(k,j,i),15,1) |
|---|
| 2941 | |
|---|
| 2942 | k_ppp = k + 3 * ibit17 |
|---|
| 2943 | k_pp = k + 2 * ( 1 - ibit15 ) |
|---|
| 2944 | k_mm = k - 2 * ibit17 |
|---|
| 2945 | |
|---|
| 2946 | w_comp = w(k,j,i) + w(k,j,i-1) |
|---|
| 2947 | flux_t(k) = w_comp * ( & |
|---|
| 2948 | ( 37.0 * ibit17 * adv_mom_5 & |
|---|
| 2949 | + 7.0 * ibit16 * adv_mom_3 & |
|---|
| 2950 | + ibit15 * adv_mom_1 & |
|---|
| 2951 | ) * & |
|---|
| 2952 | ( u(k+1,j,i) + u(k,j,i) ) & |
|---|
| 2953 | - ( 8.0 * ibit17 * adv_mom_5 & |
|---|
| 2954 | + ibit16 * adv_mom_3 & |
|---|
| 2955 | ) * & |
|---|
| 2956 | ( u(k_pp,j,i) + u(k-1,j,i) ) & |
|---|
| 2957 | + ( ibit17 * adv_mom_5 & |
|---|
| 2958 | ) * & |
|---|
| 2959 | ( u(k_ppp,j,i) + u(k_mm,j,i) ) & |
|---|
| 2960 | ) |
|---|
| 2961 | |
|---|
| 2962 | diss_t(k) = - ABS( w_comp ) * ( & |
|---|
| 2963 | ( 10.0 * ibit17 * adv_mom_5 & |
|---|
| 2964 | + 3.0 * ibit16 * adv_mom_3 & |
|---|
| 2965 | + ibit15 * adv_mom_1 & |
|---|
| 2966 | ) * & |
|---|
| 2967 | ( u(k+1,j,i) - u(k,j,i) ) & |
|---|
| 2968 | - ( 5.0 * ibit17 * adv_mom_5 & |
|---|
| 2969 | + ibit16 * adv_mom_3 & |
|---|
| 2970 | ) * & |
|---|
| 2971 | ( u(k_pp,j,i) - u(k-1,j,i) ) & |
|---|
| 2972 | + ( ibit17 * adv_mom_5 & |
|---|
| 2973 | ) * & |
|---|
| 2974 | ( u(k_ppp,j,i) - u(k_mm,j,i) ) & |
|---|
| 2975 | ) |
|---|
| 2976 | ! |
|---|
| 2977 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 2978 | !-- correction is needed to overcome numerical instabilities caused |
|---|
| 2979 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 2980 | div = ( ( u_comp(k) - ( u(k,j,i) + u(k,j,i-1) ) ) * ddx & |
|---|
| 2981 | + ( v_comp + gv - ( v(k,j,i) + v(k,j,i-1 ) ) ) * ddy & |
|---|
| 2982 | + ( w_comp - ( w(k-1,j,i) + w(k-1,j,i-1) ) ) & |
|---|
| 2983 | * ddzw(k) & |
|---|
| 2984 | ) * 0.5 |
|---|
| 2985 | |
|---|
| 2986 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 2987 | ( flux_r(k) + diss_r(k) & |
|---|
| 2988 | - swap_flux_x_local_u(k,j) - swap_diss_x_local_u(k,j) ) * ddx & |
|---|
| 2989 | + ( flux_n(k) + diss_n(k) & |
|---|
| 2990 | - swap_flux_y_local_u(k) - swap_diss_y_local_u(k) ) * ddy & |
|---|
| 2991 | + ( flux_t(k) + diss_t(k) & |
|---|
| 2992 | - flux_d - diss_d & |
|---|
| 2993 | ) * ddzw(k) & |
|---|
| 2994 | ) + div * u(k,j,i) |
|---|
| 2995 | |
|---|
| 2996 | swap_flux_x_local_u(k,j) = flux_r(k) |
|---|
| 2997 | swap_diss_x_local_u(k,j) = diss_r(k) |
|---|
| 2998 | swap_flux_y_local_u(k) = flux_n(k) |
|---|
| 2999 | swap_diss_y_local_u(k) = diss_n(k) |
|---|
| 3000 | flux_d = flux_t(k) |
|---|
| 3001 | diss_d = diss_t(k) |
|---|
| 3002 | ! |
|---|
| 3003 | !-- Statistical Evaluation of u'u'. The factor has to be applied |
|---|
| 3004 | !-- for right evaluation when gallilei_trans = .T. . |
|---|
| 3005 | sums_us2_ws_l(k,tn) = sums_us2_ws_l(k,tn) & |
|---|
| 3006 | + ( flux_r(k) * & |
|---|
| 3007 | ( u_comp(k) - 2.0 * hom(k,1,1,0) ) & |
|---|
| 3008 | / ( u_comp(k) - gu + 1.0E-20 ) & |
|---|
| 3009 | + diss_r(k) * & |
|---|
| 3010 | ABS( u_comp(k) - 2.0 * hom(k,1,1,0) ) & |
|---|
| 3011 | / ( ABS( u_comp(k) - gu ) + 1.0E-20 ) ) & |
|---|
| 3012 | * weight_substep(intermediate_timestep_count) |
|---|
| 3013 | ! |
|---|
| 3014 | !-- Statistical Evaluation of w'u'. |
|---|
| 3015 | sums_wsus_ws_l(k,tn) = sums_wsus_ws_l(k,tn) & |
|---|
| 3016 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 3017 | * weight_substep(intermediate_timestep_count) |
|---|
| 3018 | ENDDO |
|---|
| 3019 | |
|---|
| 3020 | DO k = nzb_max+1, nzt |
|---|
| 3021 | |
|---|
| 3022 | u_comp(k) = u(k,j,i+1) + u(k,j,i) |
|---|
| 3023 | flux_r(k) = ( u_comp(k) - gu ) * ( & |
|---|
| 3024 | 37.0 * ( u(k,j,i+1) + u(k,j,i) ) & |
|---|
| 3025 | - 8.0 * ( u(k,j,i+2) + u(k,j,i-1) ) & |
|---|
| 3026 | + ( u(k,j,i+3) + u(k,j,i-2) ) ) * adv_mom_5 |
|---|
| 3027 | diss_r(k) = - ABS( u_comp(k) - gu ) * ( & |
|---|
| 3028 | 10.0 * ( u(k,j,i+1) - u(k,j,i) ) & |
|---|
| 3029 | - 5.0 * ( u(k,j,i+2) - u(k,j,i-1) ) & |
|---|
| 3030 | + ( u(k,j,i+3) - u(k,j,i-2) ) ) * adv_mom_5 |
|---|
| 3031 | |
|---|
| 3032 | v_comp = v(k,j+1,i) + v(k,j+1,i-1) - gv |
|---|
| 3033 | flux_n(k) = v_comp * ( & |
|---|
| 3034 | 37.0 * ( u(k,j+1,i) + u(k,j,i) ) & |
|---|
| 3035 | - 8.0 * ( u(k,j+2,i) + u(k,j-1,i) ) & |
|---|
| 3036 | + ( u(k,j+3,i) + u(k,j-2,i) ) ) * adv_mom_5 |
|---|
| 3037 | diss_n(k) = - ABS( v_comp ) * ( & |
|---|
| 3038 | 10.0 * ( u(k,j+1,i) - u(k,j,i) ) & |
|---|
| 3039 | - 5.0 * ( u(k,j+2,i) - u(k,j-1,i) ) & |
|---|
| 3040 | + ( u(k,j+3,i) - u(k,j-2,i) ) ) * adv_mom_5 |
|---|
| 3041 | ! |
|---|
| 3042 | !-- k index has to be modified near bottom and top, else array |
|---|
| 3043 | !-- subscripts will be exceeded. |
|---|
| 3044 | ibit17 = IBITS(wall_flags_0(k,j,i),17,1) |
|---|
| 3045 | ibit16 = IBITS(wall_flags_0(k,j,i),16,1) |
|---|
| 3046 | ibit15 = IBITS(wall_flags_0(k,j,i),15,1) |
|---|
| 3047 | |
|---|
| 3048 | k_ppp = k + 3 * ibit17 |
|---|
| 3049 | k_pp = k + 2 * ( 1 - ibit15 ) |
|---|
| 3050 | k_mm = k - 2 * ibit17 |
|---|
| 3051 | |
|---|
| 3052 | w_comp = w(k,j,i) + w(k,j,i-1) |
|---|
| 3053 | flux_t(k) = w_comp * ( & |
|---|
| 3054 | ( 37.0 * ibit17 * adv_mom_5 & |
|---|
| 3055 | + 7.0 * ibit16 * adv_mom_3 & |
|---|
| 3056 | + ibit15 * adv_mom_1 & |
|---|
| 3057 | ) * & |
|---|
| 3058 | ( u(k+1,j,i) + u(k,j,i) ) & |
|---|
| 3059 | - ( 8.0 * ibit17 * adv_mom_5 & |
|---|
| 3060 | + ibit16 * adv_mom_3 & |
|---|
| 3061 | ) * & |
|---|
| 3062 | ( u(k_pp,j,i) + u(k-1,j,i) ) & |
|---|
| 3063 | + ( ibit17 * adv_mom_5 & |
|---|
| 3064 | ) * & |
|---|
| 3065 | ( u(k_ppp,j,i) + u(k_mm,j,i) ) & |
|---|
| 3066 | ) |
|---|
| 3067 | |
|---|
| 3068 | diss_t(k) = - ABS( w_comp ) * ( & |
|---|
| 3069 | ( 10.0 * ibit17 * adv_mom_5 & |
|---|
| 3070 | + 3.0 * ibit16 * adv_mom_3 & |
|---|
| 3071 | + ibit15 * adv_mom_1 & |
|---|
| 3072 | ) * & |
|---|
| 3073 | ( u(k+1,j,i) - u(k,j,i) ) & |
|---|
| 3074 | - ( 5.0 * ibit17 * adv_mom_5 & |
|---|
| 3075 | + ibit16 * adv_mom_3 & |
|---|
| 3076 | ) * & |
|---|
| 3077 | ( u(k_pp,j,i) - u(k-1,j,i) ) & |
|---|
| 3078 | + ( ibit17 * adv_mom_5 & |
|---|
| 3079 | ) * & |
|---|
| 3080 | ( u(k_ppp,j,i) - u(k_mm,j,i) ) & |
|---|
| 3081 | ) |
|---|
| 3082 | ! |
|---|
| 3083 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 3084 | !-- correction is needed to overcome numerical instabilities caused |
|---|
| 3085 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 3086 | div = ( ( u_comp(k) - ( u(k,j,i) + u(k,j,i-1) ) ) * ddx & |
|---|
| 3087 | + ( v_comp + gv - ( v(k,j,i) + v(k,j,i-1 ) ) ) * ddy & |
|---|
| 3088 | + ( w_comp - ( w(k-1,j,i) + w(k-1,j,i-1) ) ) & |
|---|
| 3089 | * ddzw(k) & |
|---|
| 3090 | ) * 0.5 |
|---|
| 3091 | |
|---|
| 3092 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 3093 | ( flux_r(k) + diss_r(k) & |
|---|
| 3094 | - swap_flux_x_local_u(k,j) - swap_diss_x_local_u(k,j) ) * ddx & |
|---|
| 3095 | + ( flux_n(k) + diss_n(k) & |
|---|
| 3096 | - swap_flux_y_local_u(k) - swap_diss_y_local_u(k) ) * ddy & |
|---|
| 3097 | + ( flux_t(k) + diss_t(k) & |
|---|
| 3098 | - flux_d - diss_d & |
|---|
| 3099 | ) * ddzw(k) & |
|---|
| 3100 | ) + div * u(k,j,i) |
|---|
| 3101 | |
|---|
| 3102 | swap_flux_x_local_u(k,j) = flux_r(k) |
|---|
| 3103 | swap_diss_x_local_u(k,j) = diss_r(k) |
|---|
| 3104 | swap_flux_y_local_u(k) = flux_n(k) |
|---|
| 3105 | swap_diss_y_local_u(k) = diss_n(k) |
|---|
| 3106 | flux_d = flux_t(k) |
|---|
| 3107 | diss_d = diss_t(k) |
|---|
| 3108 | ! |
|---|
| 3109 | !-- Statistical Evaluation of u'u'. The factor has to be applied |
|---|
| 3110 | !-- for right evaluation when gallilei_trans = .T. . |
|---|
| 3111 | sums_us2_ws_l(k,tn) = sums_us2_ws_l(k,tn) & |
|---|
| 3112 | + ( flux_r(k) * & |
|---|
| 3113 | ( u_comp(k) - 2.0 * hom(k,1,1,0) ) & |
|---|
| 3114 | / ( u_comp(k) - gu + 1.0E-20 ) & |
|---|
| 3115 | + diss_r(k) * & |
|---|
| 3116 | ABS( u_comp(k) - 2.0 * hom(k,1,1,0) ) & |
|---|
| 3117 | / ( ABS( u_comp(k) - gu ) + 1.0E-20 ) ) & |
|---|
| 3118 | * weight_substep(intermediate_timestep_count) |
|---|
| 3119 | ! |
|---|
| 3120 | !-- Statistical Evaluation of w'u'. |
|---|
| 3121 | sums_wsus_ws_l(k,tn) = sums_wsus_ws_l(k,tn) & |
|---|
| 3122 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 3123 | * weight_substep(intermediate_timestep_count) |
|---|
| 3124 | ENDDO |
|---|
| 3125 | ENDDO |
|---|
| 3126 | ENDDO |
|---|
| 3127 | sums_us2_ws_l(nzb,tn) = sums_us2_ws_l(nzb+1,tn) |
|---|
| 3128 | |
|---|
| 3129 | |
|---|
| 3130 | END SUBROUTINE advec_u_ws |
|---|
| 3131 | |
|---|
| 3132 | |
|---|
| 3133 | !------------------------------------------------------------------------------! |
|---|
| 3134 | ! Advection of u - Call for all grid points - accelerator version |
|---|
| 3135 | !------------------------------------------------------------------------------! |
|---|
| 3136 | SUBROUTINE advec_u_ws_acc |
|---|
| 3137 | |
|---|
| 3138 | USE arrays_3d |
|---|
| 3139 | USE constants |
|---|
| 3140 | USE control_parameters |
|---|
| 3141 | USE grid_variables |
|---|
| 3142 | USE indices |
|---|
| 3143 | USE statistics |
|---|
| 3144 | |
|---|
| 3145 | IMPLICIT NONE |
|---|
| 3146 | |
|---|
| 3147 | INTEGER :: i, ibit9, ibit10, ibit11, ibit12, ibit13, ibit14, ibit15, & |
|---|
| 3148 | ibit16, ibit17, j, k, k_mmm, k_mm, k_pp, k_ppp, tn = 0 |
|---|
| 3149 | |
|---|
| 3150 | REAL :: diss_d, diss_l, diss_n, diss_r, diss_s, diss_t, div, & |
|---|
| 3151 | flux_d, flux_l, flux_n, flux_r, flux_s, flux_t, gu, gv, & |
|---|
| 3152 | u_comp, u_comp_l, v_comp, v_comp_s, w_comp |
|---|
| 3153 | |
|---|
| 3154 | |
|---|
| 3155 | gu = 2.0 * u_gtrans |
|---|
| 3156 | gv = 2.0 * v_gtrans |
|---|
| 3157 | |
|---|
| 3158 | ! |
|---|
| 3159 | !-- Computation of fluxes and tendency terms |
|---|
| 3160 | !$acc kernels present( ddzw, tend, u, v, w, wall_flags_0, wall_flags_00 ) |
|---|
| 3161 | !$acc loop |
|---|
| 3162 | DO i = i_left, i_right |
|---|
| 3163 | DO j = j_south, j_north |
|---|
| 3164 | !$acc loop vector( 32 ) |
|---|
| 3165 | DO k = nzb+1, nzt |
|---|
| 3166 | |
|---|
| 3167 | ibit11 = IBITS(wall_flags_0(k,j,i),11,1) |
|---|
| 3168 | ibit10 = IBITS(wall_flags_0(k,j,i),10,1) |
|---|
| 3169 | ibit9 = IBITS(wall_flags_0(k,j,i),9,1) |
|---|
| 3170 | |
|---|
| 3171 | u_comp_l = u(k,j,i) + u(k,j,i-1) - gu |
|---|
| 3172 | flux_l = u_comp_l * ( & |
|---|
| 3173 | ( 37.0 * ibit11 * adv_mom_5 & |
|---|
| 3174 | + 7.0 * ibit10 * adv_mom_3 & |
|---|
| 3175 | + ibit9 * adv_mom_1 & |
|---|
| 3176 | ) * & |
|---|
| 3177 | ( u(k,j,i) + u(k,j,i-1) ) & |
|---|
| 3178 | - ( 8.0 * ibit11 * adv_mom_5 & |
|---|
| 3179 | + ibit10 * adv_mom_3 & |
|---|
| 3180 | ) * & |
|---|
| 3181 | ( u(k,j,i+1) + u(k,j,i-2) ) & |
|---|
| 3182 | + ( ibit11 * adv_mom_5 & |
|---|
| 3183 | ) * & |
|---|
| 3184 | ( u(k,j,i+2) + u(k,j,i-3) ) & |
|---|
| 3185 | ) |
|---|
| 3186 | |
|---|
| 3187 | diss_l = - ABS( u_comp_l ) * ( & |
|---|
| 3188 | ( 10.0 * ibit11 * adv_mom_5 & |
|---|
| 3189 | + 3.0 * ibit10 * adv_mom_3 & |
|---|
| 3190 | + ibit9 * adv_mom_1 & |
|---|
| 3191 | ) * & |
|---|
| 3192 | ( u(k,j,i) - u(k,j,i-1) ) & |
|---|
| 3193 | - ( 5.0 * ibit11 * adv_mom_5 & |
|---|
| 3194 | + ibit10 * adv_mom_3 & |
|---|
| 3195 | ) * & |
|---|
| 3196 | ( u(k,j,i+1) - u(k,j,i-2) ) & |
|---|
| 3197 | + ( ibit11 * adv_mom_5 & |
|---|
| 3198 | ) * & |
|---|
| 3199 | ( u(k,j,i+2) - u(k,j,i-3) ) & |
|---|
| 3200 | ) |
|---|
| 3201 | |
|---|
| 3202 | u_comp = u(k,j,i+1) + u(k,j,i) |
|---|
| 3203 | flux_r = ( u_comp - gu ) * ( & |
|---|
| 3204 | ( 37.0 * ibit11 * adv_mom_5 & |
|---|
| 3205 | + 7.0 * ibit10 * adv_mom_3 & |
|---|
| 3206 | + ibit9 * adv_mom_1 & |
|---|
| 3207 | ) * & |
|---|
| 3208 | ( u(k,j,i+1) + u(k,j,i) ) & |
|---|
| 3209 | - ( 8.0 * ibit11 * adv_mom_5 & |
|---|
| 3210 | + ibit10 * adv_mom_3 & |
|---|
| 3211 | ) * & |
|---|
| 3212 | ( u(k,j,i+2) + u(k,j,i-1) ) & |
|---|
| 3213 | + ( ibit11 * adv_mom_5 & |
|---|
| 3214 | ) * & |
|---|
| 3215 | ( u(k,j,i+3) + u(k,j,i-2) ) & |
|---|
| 3216 | ) |
|---|
| 3217 | |
|---|
| 3218 | diss_r = - ABS( u_comp - gu ) * ( & |
|---|
| 3219 | ( 10.0 * ibit11 * adv_mom_5 & |
|---|
| 3220 | + 3.0 * ibit10 * adv_mom_3 & |
|---|
| 3221 | + ibit9 * adv_mom_1 & |
|---|
| 3222 | ) * & |
|---|
| 3223 | ( u(k,j,i+1) - u(k,j,i) ) & |
|---|
| 3224 | - ( 5.0 * ibit11 * adv_mom_5 & |
|---|
| 3225 | + ibit10 * adv_mom_3 & |
|---|
| 3226 | ) * & |
|---|
| 3227 | ( u(k,j,i+2) - u(k,j,i-1) ) & |
|---|
| 3228 | + ( ibit11 * adv_mom_5 & |
|---|
| 3229 | ) * & |
|---|
| 3230 | ( u(k,j,i+3) - u(k,j,i-2) ) & |
|---|
| 3231 | ) |
|---|
| 3232 | |
|---|
| 3233 | ibit14 = IBITS(wall_flags_0(k,j,i),14,1) |
|---|
| 3234 | ibit13 = IBITS(wall_flags_0(k,j,i),13,1) |
|---|
| 3235 | ibit12 = IBITS(wall_flags_0(k,j,i),12,1) |
|---|
| 3236 | |
|---|
| 3237 | v_comp_s = v(k,j,i) + v(k,j,i-1) - gv |
|---|
| 3238 | flux_s = v_comp_s * ( & |
|---|
| 3239 | ( 37.0 * ibit14 * adv_mom_5 & |
|---|
| 3240 | + 7.0 * ibit13 * adv_mom_3 & |
|---|
| 3241 | + ibit12 * adv_mom_1 & |
|---|
| 3242 | ) * & |
|---|
| 3243 | ( u(k,j,i) + u(k,j-1,i) ) & |
|---|
| 3244 | - ( 8.0 * ibit14 * adv_mom_5 & |
|---|
| 3245 | + ibit13 * adv_mom_3 & |
|---|
| 3246 | ) * & |
|---|
| 3247 | ( u(k,j+1,i) + u(k,j-2,i) ) & |
|---|
| 3248 | + ( ibit14 * adv_mom_5 & |
|---|
| 3249 | ) * & |
|---|
| 3250 | ( u(k,j+2,i) + u(k,j-3,i) ) & |
|---|
| 3251 | ) |
|---|
| 3252 | |
|---|
| 3253 | diss_s = - ABS ( v_comp_s ) * ( & |
|---|
| 3254 | ( 10.0 * ibit14 * adv_mom_5 & |
|---|
| 3255 | + 3.0 * ibit13 * adv_mom_3 & |
|---|
| 3256 | + ibit12 * adv_mom_1 & |
|---|
| 3257 | ) * & |
|---|
| 3258 | ( u(k,j,i) - u(k,j-1,i) ) & |
|---|
| 3259 | - ( 5.0 * ibit14 * adv_mom_5 & |
|---|
| 3260 | + ibit13 * adv_mom_3 & |
|---|
| 3261 | ) * & |
|---|
| 3262 | ( u(k,j+1,i) - u(k,j-2,i) ) & |
|---|
| 3263 | + ( ibit14 * adv_mom_5 & |
|---|
| 3264 | ) * & |
|---|
| 3265 | ( u(k,j+2,i) - u(k,j-3,i) ) & |
|---|
| 3266 | ) |
|---|
| 3267 | |
|---|
| 3268 | |
|---|
| 3269 | v_comp = v(k,j+1,i) + v(k,j+1,i-1) - gv |
|---|
| 3270 | flux_n = v_comp * ( & |
|---|
| 3271 | ( 37.0 * ibit14 * adv_mom_5 & |
|---|
| 3272 | + 7.0 * ibit13 * adv_mom_3 & |
|---|
| 3273 | + ibit12 * adv_mom_1 & |
|---|
| 3274 | ) * & |
|---|
| 3275 | ( u(k,j+1,i) + u(k,j,i) ) & |
|---|
| 3276 | - ( 8.0 * ibit14 * adv_mom_5 & |
|---|
| 3277 | + ibit13 * adv_mom_3 & |
|---|
| 3278 | ) * & |
|---|
| 3279 | ( u(k,j+2,i) + u(k,j-1,i) ) & |
|---|
| 3280 | + ( ibit14 * adv_mom_5 & |
|---|
| 3281 | ) * & |
|---|
| 3282 | ( u(k,j+3,i) + u(k,j-2,i) ) & |
|---|
| 3283 | ) |
|---|
| 3284 | |
|---|
| 3285 | diss_n = - ABS ( v_comp ) * ( & |
|---|
| 3286 | ( 10.0 * ibit14 * adv_mom_5 & |
|---|
| 3287 | + 3.0 * ibit13 * adv_mom_3 & |
|---|
| 3288 | + ibit12 * adv_mom_1 & |
|---|
| 3289 | ) * & |
|---|
| 3290 | ( u(k,j+1,i) - u(k,j,i) ) & |
|---|
| 3291 | - ( 5.0 * ibit14 * adv_mom_5 & |
|---|
| 3292 | + ibit13 * adv_mom_3 & |
|---|
| 3293 | ) * & |
|---|
| 3294 | ( u(k,j+2,i) - u(k,j-1,i) ) & |
|---|
| 3295 | + ( ibit14 * adv_mom_5 & |
|---|
| 3296 | ) * & |
|---|
| 3297 | ( u(k,j+3,i) - u(k,j-2,i) ) & |
|---|
| 3298 | ) |
|---|
| 3299 | |
|---|
| 3300 | ibit17 = IBITS(wall_flags_0(k-1,j,i),17,1) |
|---|
| 3301 | ibit16 = IBITS(wall_flags_0(k-1,j,i),16,1) |
|---|
| 3302 | ibit15 = IBITS(wall_flags_0(k-1,j,i),15,1) |
|---|
| 3303 | |
|---|
| 3304 | k_pp = k + 2 * ibit17 |
|---|
| 3305 | k_mm = k - 2 * ( ibit16 + ibit17 ) |
|---|
| 3306 | k_mmm = k - 3 * ibit17 |
|---|
| 3307 | |
|---|
| 3308 | w_comp = w(k-1,j,i) + w(k-1,j,i-1) |
|---|
| 3309 | flux_d = w_comp * ( & |
|---|
| 3310 | ( 37.0 * ibit17 * adv_mom_5 & |
|---|
| 3311 | + 7.0 * ibit16 * adv_mom_3 & |
|---|
| 3312 | + ibit15 * adv_mom_1 & |
|---|
| 3313 | ) * & |
|---|
| 3314 | ( u(k,j,i) + u(k-1,j,i) ) & |
|---|
| 3315 | - ( 8.0 * ibit17 * adv_mom_5 & |
|---|
| 3316 | + ibit16 * adv_mom_3 & |
|---|
| 3317 | ) * & |
|---|
| 3318 | ( u(k+1,j,i) + u(k_mm,j,i) ) & |
|---|
| 3319 | + ( ibit17 * adv_mom_5 & |
|---|
| 3320 | ) * & |
|---|
| 3321 | ( u(k_pp,j,i) + u(k_mmm,j,i) ) & |
|---|
| 3322 | ) |
|---|
| 3323 | |
|---|
| 3324 | diss_d = - ABS( w_comp ) * ( & |
|---|
| 3325 | ( 10.0 * ibit17 * adv_mom_5 & |
|---|
| 3326 | + 3.0 * ibit16 * adv_mom_3 & |
|---|
| 3327 | + ibit15 * adv_mom_1 & |
|---|
| 3328 | ) * & |
|---|
| 3329 | ( u(k,j,i) - u(k-1,j,i) ) & |
|---|
| 3330 | - ( 5.0 * ibit17 * adv_mom_5 & |
|---|
| 3331 | + ibit16 * adv_mom_3 & |
|---|
| 3332 | ) * & |
|---|
| 3333 | ( u(k+1,j,i) - u(k_mm,j,i) ) & |
|---|
| 3334 | + ( ibit17 * adv_mom_5 & |
|---|
| 3335 | ) * & |
|---|
| 3336 | ( u(k_pp,j,i) - u(k_mmm,j,i) ) & |
|---|
| 3337 | ) |
|---|
| 3338 | ! |
|---|
| 3339 | !-- k index has to be modified near bottom and top, else array |
|---|
| 3340 | !-- subscripts will be exceeded. |
|---|
| 3341 | ibit17 = IBITS(wall_flags_0(k,j,i),17,1) |
|---|
| 3342 | ibit16 = IBITS(wall_flags_0(k,j,i),16,1) |
|---|
| 3343 | ibit15 = IBITS(wall_flags_0(k,j,i),15,1) |
|---|
| 3344 | |
|---|
| 3345 | k_ppp = k + 3 * ibit17 |
|---|
| 3346 | k_pp = k + 2 * ( 1 - ibit15 ) |
|---|
| 3347 | k_mm = k - 2 * ibit17 |
|---|
| 3348 | |
|---|
| 3349 | w_comp = w(k,j,i) + w(k,j,i-1) |
|---|
| 3350 | flux_t = w_comp * ( & |
|---|
| 3351 | ( 37.0 * ibit17 * adv_mom_5 & |
|---|
| 3352 | + 7.0 * ibit16 * adv_mom_3 & |
|---|
| 3353 | + ibit15 * adv_mom_1 & |
|---|
| 3354 | ) * & |
|---|
| 3355 | ( u(k+1,j,i) + u(k,j,i) ) & |
|---|
| 3356 | - ( 8.0 * ibit17 * adv_mom_5 & |
|---|
| 3357 | + ibit16 * adv_mom_3 & |
|---|
| 3358 | ) * & |
|---|
| 3359 | ( u(k_pp,j,i) + u(k-1,j,i) ) & |
|---|
| 3360 | + ( ibit17 * adv_mom_5 & |
|---|
| 3361 | ) * & |
|---|
| 3362 | ( u(k_ppp,j,i) + u(k_mm,j,i) ) & |
|---|
| 3363 | ) |
|---|
| 3364 | |
|---|
| 3365 | diss_t = - ABS( w_comp ) * ( & |
|---|
| 3366 | ( 10.0 * ibit17 * adv_mom_5 & |
|---|
| 3367 | + 3.0 * ibit16 * adv_mom_3 & |
|---|
| 3368 | + ibit15 * adv_mom_1 & |
|---|
| 3369 | ) * & |
|---|
| 3370 | ( u(k+1,j,i) - u(k,j,i) ) & |
|---|
| 3371 | - ( 5.0 * ibit17 * adv_mom_5 & |
|---|
| 3372 | + ibit16 * adv_mom_3 & |
|---|
| 3373 | ) * & |
|---|
| 3374 | ( u(k_pp,j,i) - u(k-1,j,i) ) & |
|---|
| 3375 | + ( ibit17 * adv_mom_5 & |
|---|
| 3376 | ) * & |
|---|
| 3377 | ( u(k_ppp,j,i) - u(k_mm,j,i) ) & |
|---|
| 3378 | ) |
|---|
| 3379 | ! |
|---|
| 3380 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 3381 | !-- correction is needed to overcome numerical instabilities caused |
|---|
| 3382 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 3383 | div = ( ( u_comp - ( u(k,j,i) + u(k,j,i-1) ) ) * ddx & |
|---|
| 3384 | + ( v_comp + gv - ( v(k,j,i) + v(k,j,i-1 ) ) ) * ddy & |
|---|
| 3385 | + ( w_comp - ( w(k-1,j,i) + w(k-1,j,i-1) ) ) & |
|---|
| 3386 | * ddzw(k) & |
|---|
| 3387 | ) * 0.5 |
|---|
| 3388 | |
|---|
| 3389 | tend(k,j,i) = - ( & |
|---|
| 3390 | ( flux_r + diss_r - flux_l - diss_l ) * ddx & |
|---|
| 3391 | + ( flux_n + diss_n - flux_s - diss_s ) * ddy & |
|---|
| 3392 | + ( flux_t + diss_t - flux_d - diss_d ) * ddzw(k) & |
|---|
| 3393 | ) + div * u(k,j,i) |
|---|
| 3394 | |
|---|
| 3395 | !++ |
|---|
| 3396 | !-- Statistical Evaluation of u'u'. The factor has to be applied |
|---|
| 3397 | !-- for right evaluation when gallilei_trans = .T. . |
|---|
| 3398 | ! sums_us2_ws_l(k,tn) = sums_us2_ws_l(k,tn) & |
|---|
| 3399 | ! + ( flux_r * & |
|---|
| 3400 | ! ( u_comp - 2.0 * hom(k,1,1,0) ) & |
|---|
| 3401 | ! / ( u_comp - gu + 1.0E-20 ) & |
|---|
| 3402 | ! + diss_r * & |
|---|
| 3403 | ! ABS( u_comp - 2.0 * hom(k,1,1,0) ) & |
|---|
| 3404 | ! / ( ABS( u_comp - gu ) + 1.0E-20 ) ) & |
|---|
| 3405 | ! * weight_substep(intermediate_timestep_count) |
|---|
| 3406 | ! |
|---|
| 3407 | !-- Statistical Evaluation of w'u'. |
|---|
| 3408 | ! sums_wsus_ws_l(k,tn) = sums_wsus_ws_l(k,tn) & |
|---|
| 3409 | ! + ( flux_t + diss_t ) & |
|---|
| 3410 | ! * weight_substep(intermediate_timestep_count) |
|---|
| 3411 | ENDDO |
|---|
| 3412 | ENDDO |
|---|
| 3413 | ENDDO |
|---|
| 3414 | !$acc end kernels |
|---|
| 3415 | |
|---|
| 3416 | !++ |
|---|
| 3417 | ! sums_us2_ws_l(nzb,tn) = sums_us2_ws_l(nzb+1,tn) |
|---|
| 3418 | |
|---|
| 3419 | END SUBROUTINE advec_u_ws_acc |
|---|
| 3420 | |
|---|
| 3421 | |
|---|
| 3422 | !------------------------------------------------------------------------------! |
|---|
| 3423 | ! Advection of v - Call for all grid points |
|---|
| 3424 | !------------------------------------------------------------------------------! |
|---|
| 3425 | SUBROUTINE advec_v_ws |
|---|
| 3426 | |
|---|
| 3427 | USE arrays_3d |
|---|
| 3428 | USE constants |
|---|
| 3429 | USE control_parameters |
|---|
| 3430 | USE grid_variables |
|---|
| 3431 | USE indices |
|---|
| 3432 | USE statistics |
|---|
| 3433 | |
|---|
| 3434 | IMPLICIT NONE |
|---|
| 3435 | |
|---|
| 3436 | |
|---|
| 3437 | INTEGER :: i, ibit18, ibit19, ibit20, ibit21, ibit22, ibit23, ibit24, & |
|---|
| 3438 | ibit25, ibit26, j, k, k_mm, k_pp, k_ppp, tn = 0 |
|---|
| 3439 | REAL :: diss_d, div, flux_d, gu, gv, u_comp, w_comp |
|---|
| 3440 | REAL, DIMENSION(nzb+1:nzt) :: swap_diss_y_local_v, swap_flux_y_local_v |
|---|
| 3441 | REAL, DIMENSION(nzb+1:nzt,nys:nyn) :: swap_diss_x_local_v, & |
|---|
| 3442 | swap_flux_x_local_v |
|---|
| 3443 | REAL, DIMENSION(nzb:nzt) :: diss_n, diss_r, diss_t, flux_n, flux_r, & |
|---|
| 3444 | flux_t, v_comp |
|---|
| 3445 | |
|---|
| 3446 | gu = 2.0 * u_gtrans |
|---|
| 3447 | gv = 2.0 * v_gtrans |
|---|
| 3448 | ! |
|---|
| 3449 | !-- First compute the whole left boundary of the processor domain |
|---|
| 3450 | i = nxl |
|---|
| 3451 | DO j = nysv, nyn |
|---|
| 3452 | DO k = nzb+1, nzb_max |
|---|
| 3453 | |
|---|
| 3454 | ibit20 = IBITS(wall_flags_0(k,j,i),20,1) |
|---|
| 3455 | ibit19 = IBITS(wall_flags_0(k,j,i),19,1) |
|---|
| 3456 | ibit18 = IBITS(wall_flags_0(k,j,i),18,1) |
|---|
| 3457 | |
|---|
| 3458 | u_comp = u(k,j-1,i) + u(k,j,i) - gu |
|---|
| 3459 | swap_flux_x_local_v(k,j) = u_comp * ( & |
|---|
| 3460 | ( 37.0 * ibit20 * adv_mom_5 & |
|---|
| 3461 | + 7.0 * ibit19 * adv_mom_3 & |
|---|
| 3462 | + ibit18 * adv_mom_1 & |
|---|
| 3463 | ) * & |
|---|
| 3464 | ( v(k,j,i) + v(k,j,i-1) ) & |
|---|
| 3465 | - ( 8.0 * ibit20 * adv_mom_5 & |
|---|
| 3466 | + ibit19 * adv_mom_3 & |
|---|
| 3467 | ) * & |
|---|
| 3468 | ( v(k,j,i+1) + v(k,j,i-2) ) & |
|---|
| 3469 | + ( ibit20 * adv_mom_5 & |
|---|
| 3470 | ) * & |
|---|
| 3471 | ( v(k,j,i+2) + v(k,j,i-3) ) & |
|---|
| 3472 | ) |
|---|
| 3473 | |
|---|
| 3474 | swap_diss_x_local_v(k,j) = - ABS( u_comp ) * ( & |
|---|
| 3475 | ( 10.0 * ibit20 * adv_mom_5 & |
|---|
| 3476 | + 3.0 * ibit19 * adv_mom_3 & |
|---|
| 3477 | + ibit18 * adv_mom_1 & |
|---|
| 3478 | ) * & |
|---|
| 3479 | ( v(k,j,i) - v(k,j,i-1) ) & |
|---|
| 3480 | - ( 5.0 * ibit20 * adv_mom_5 & |
|---|
| 3481 | + ibit19 * adv_mom_3 & |
|---|
| 3482 | ) * & |
|---|
| 3483 | ( v(k,j,i+1) - v(k,j,i-2) ) & |
|---|
| 3484 | + ( ibit20 * adv_mom_5 & |
|---|
| 3485 | ) * & |
|---|
| 3486 | ( v(k,j,i+2) - v(k,j,i-3) ) & |
|---|
| 3487 | ) |
|---|
| 3488 | |
|---|
| 3489 | ENDDO |
|---|
| 3490 | |
|---|
| 3491 | DO k = nzb_max+1, nzt |
|---|
| 3492 | |
|---|
| 3493 | u_comp = u(k,j-1,i) + u(k,j,i) - gu |
|---|
| 3494 | swap_flux_x_local_v(k,j) = u_comp * ( & |
|---|
| 3495 | 37.0 * ( v(k,j,i) + v(k,j,i-1) ) & |
|---|
| 3496 | - 8.0 * ( v(k,j,i+1) + v(k,j,i-2) ) & |
|---|
| 3497 | + ( v(k,j,i+2) + v(k,j,i-3) ) ) * adv_mom_5 |
|---|
| 3498 | swap_diss_x_local_v(k,j) = - ABS( u_comp ) * ( & |
|---|
| 3499 | 10.0 * ( v(k,j,i) - v(k,j,i-1) ) & |
|---|
| 3500 | - 5.0 * ( v(k,j,i+1) - v(k,j,i-2) ) & |
|---|
| 3501 | + ( v(k,j,i+2) - v(k,j,i-3) ) ) * adv_mom_5 |
|---|
| 3502 | |
|---|
| 3503 | ENDDO |
|---|
| 3504 | |
|---|
| 3505 | ENDDO |
|---|
| 3506 | |
|---|
| 3507 | DO i = nxl, nxr |
|---|
| 3508 | |
|---|
| 3509 | j = nysv |
|---|
| 3510 | DO k = nzb+1, nzb_max |
|---|
| 3511 | |
|---|
| 3512 | ibit23 = IBITS(wall_flags_0(k,j,i),23,1) |
|---|
| 3513 | ibit22 = IBITS(wall_flags_0(k,j,i),22,1) |
|---|
| 3514 | ibit21 = IBITS(wall_flags_0(k,j,i),21,1) |
|---|
| 3515 | |
|---|
| 3516 | v_comp(k) = v(k,j,i) + v(k,j-1,i) - gv |
|---|
| 3517 | swap_flux_y_local_v(k) = v_comp(k) * ( & |
|---|
| 3518 | ( 37.0 * ibit23 * adv_mom_5 & |
|---|
| 3519 | + 7.0 * ibit22 * adv_mom_3 & |
|---|
| 3520 | + ibit21 * adv_mom_1 & |
|---|
| 3521 | ) * & |
|---|
| 3522 | ( v(k,j,i) + v(k,j-1,i) ) & |
|---|
| 3523 | - ( 8.0 * ibit23 * adv_mom_5 & |
|---|
| 3524 | + ibit22 * adv_mom_3 & |
|---|
| 3525 | ) * & |
|---|
| 3526 | ( v(k,j+1,i) + v(k,j-2,i) ) & |
|---|
| 3527 | + ( ibit23 * adv_mom_5 & |
|---|
| 3528 | ) * & |
|---|
| 3529 | ( v(k,j+2,i) + v(k,j-3,i) ) & |
|---|
| 3530 | ) |
|---|
| 3531 | |
|---|
| 3532 | swap_diss_y_local_v(k) = - ABS( v_comp(k) ) * ( & |
|---|
| 3533 | ( 10.0 * ibit23 * adv_mom_5 & |
|---|
| 3534 | + 3.0 * ibit22 * adv_mom_3 & |
|---|
| 3535 | + ibit21 * adv_mom_1 & |
|---|
| 3536 | ) * & |
|---|
| 3537 | ( v(k,j,i) - v(k,j-1,i) ) & |
|---|
| 3538 | - ( 5.0 * ibit23 * adv_mom_5 & |
|---|
| 3539 | + ibit22 * adv_mom_3 & |
|---|
| 3540 | ) * & |
|---|
| 3541 | ( v(k,j+1,i) - v(k,j-2,i) ) & |
|---|
| 3542 | + ( ibit23 * adv_mom_5 & |
|---|
| 3543 | ) * & |
|---|
| 3544 | ( v(k,j+2,i) - v(k,j-3,i) ) & |
|---|
| 3545 | ) |
|---|
| 3546 | |
|---|
| 3547 | ENDDO |
|---|
| 3548 | |
|---|
| 3549 | DO k = nzb_max+1, nzt |
|---|
| 3550 | |
|---|
| 3551 | v_comp(k) = v(k,j,i) + v(k,j-1,i) - gv |
|---|
| 3552 | swap_flux_y_local_v(k) = v_comp(k) * ( & |
|---|
| 3553 | 37.0 * ( v(k,j,i) + v(k,j-1,i) ) & |
|---|
| 3554 | - 8.0 * ( v(k,j+1,i) + v(k,j-2,i) ) & |
|---|
| 3555 | + ( v(k,j+2,i) + v(k,j-3,i) ) ) * adv_mom_5 |
|---|
| 3556 | swap_diss_y_local_v(k) = - ABS( v_comp(k) ) * ( & |
|---|
| 3557 | 10.0 * ( v(k,j,i) - v(k,j-1,i) ) & |
|---|
| 3558 | - 5.0 * ( v(k,j+1,i) - v(k,j-2,i) ) & |
|---|
| 3559 | + ( v(k,j+2,i) - v(k,j-3,i) ) ) * adv_mom_5 |
|---|
| 3560 | |
|---|
| 3561 | ENDDO |
|---|
| 3562 | |
|---|
| 3563 | DO j = nysv, nyn |
|---|
| 3564 | |
|---|
| 3565 | flux_t(0) = 0.0 |
|---|
| 3566 | diss_t(0) = 0.0 |
|---|
| 3567 | flux_d = 0.0 |
|---|
| 3568 | diss_d = 0.0 |
|---|
| 3569 | |
|---|
| 3570 | DO k = nzb+1, nzb_max |
|---|
| 3571 | |
|---|
| 3572 | ibit20 = IBITS(wall_flags_0(k,j,i),20,1) |
|---|
| 3573 | ibit19 = IBITS(wall_flags_0(k,j,i),19,1) |
|---|
| 3574 | ibit18 = IBITS(wall_flags_0(k,j,i),18,1) |
|---|
| 3575 | |
|---|
| 3576 | u_comp = u(k,j-1,i+1) + u(k,j,i+1) - gu |
|---|
| 3577 | flux_r(k) = u_comp * ( & |
|---|
| 3578 | ( 37.0 * ibit20 * adv_mom_5 & |
|---|
| 3579 | + 7.0 * ibit19 * adv_mom_3 & |
|---|
| 3580 | + ibit18 * adv_mom_1 & |
|---|
| 3581 | ) * & |
|---|
| 3582 | ( v(k,j,i+1) + v(k,j,i) ) & |
|---|
| 3583 | - ( 8.0 * ibit20 * adv_mom_5 & |
|---|
| 3584 | + ibit19 * adv_mom_3 & |
|---|
| 3585 | ) * & |
|---|
| 3586 | ( v(k,j,i+2) + v(k,j,i-1) ) & |
|---|
| 3587 | + ( ibit20 * adv_mom_5 & |
|---|
| 3588 | ) * & |
|---|
| 3589 | ( v(k,j,i+3) + v(k,j,i-2) ) & |
|---|
| 3590 | ) |
|---|
| 3591 | |
|---|
| 3592 | diss_r(k) = - ABS( u_comp ) * ( & |
|---|
| 3593 | ( 10.0 * ibit20 * adv_mom_5 & |
|---|
| 3594 | + 3.0 * ibit19 * adv_mom_3 & |
|---|
| 3595 | + ibit18 * adv_mom_1 & |
|---|
| 3596 | ) * & |
|---|
| 3597 | ( v(k,j,i+1) - v(k,j,i) ) & |
|---|
| 3598 | - ( 5.0 * ibit20 * adv_mom_5 & |
|---|
| 3599 | + ibit19 * adv_mom_3 & |
|---|
| 3600 | ) * & |
|---|
| 3601 | ( v(k,j,i+2) - v(k,j,i-1) ) & |
|---|
| 3602 | + ( ibit20 * adv_mom_5 & |
|---|
| 3603 | ) * & |
|---|
| 3604 | ( v(k,j,i+3) - v(k,j,i-2) ) & |
|---|
| 3605 | ) |
|---|
| 3606 | |
|---|
| 3607 | ibit23 = IBITS(wall_flags_0(k,j,i),23,1) |
|---|
| 3608 | ibit22 = IBITS(wall_flags_0(k,j,i),22,1) |
|---|
| 3609 | ibit21 = IBITS(wall_flags_0(k,j,i),21,1) |
|---|
| 3610 | |
|---|
| 3611 | v_comp(k) = v(k,j+1,i) + v(k,j,i) |
|---|
| 3612 | flux_n(k) = ( v_comp(k) - gv ) * ( & |
|---|
| 3613 | ( 37.0 * ibit23 * adv_mom_5 & |
|---|
| 3614 | + 7.0 * ibit22 * adv_mom_3 & |
|---|
| 3615 | + ibit21 * adv_mom_1 & |
|---|
| 3616 | ) * & |
|---|
| 3617 | ( v(k,j+1,i) + v(k,j,i) ) & |
|---|
| 3618 | - ( 8.0 * ibit23 * adv_mom_5 & |
|---|
| 3619 | + ibit22 * adv_mom_3 & |
|---|
| 3620 | ) * & |
|---|
| 3621 | ( v(k,j+2,i) + v(k,j-1,i) ) & |
|---|
| 3622 | + ( ibit23 * adv_mom_5 & |
|---|
| 3623 | ) * & |
|---|
| 3624 | ( v(k,j+3,i) + v(k,j-2,i) ) & |
|---|
| 3625 | ) |
|---|
| 3626 | |
|---|
| 3627 | diss_n(k) = - ABS( v_comp(k) - gv ) * ( & |
|---|
| 3628 | ( 10.0 * ibit23 * adv_mom_5 & |
|---|
| 3629 | + 3.0 * ibit22 * adv_mom_3 & |
|---|
| 3630 | + ibit21 * adv_mom_1 & |
|---|
| 3631 | ) * & |
|---|
| 3632 | ( v(k,j+1,i) - v(k,j,i) ) & |
|---|
| 3633 | - ( 5.0 * ibit23 * adv_mom_5 & |
|---|
| 3634 | + ibit22 * adv_mom_3 & |
|---|
| 3635 | ) * & |
|---|
| 3636 | ( v(k,j+2,i) - v(k,j-1,i) ) & |
|---|
| 3637 | + ( ibit23 * adv_mom_5 & |
|---|
| 3638 | ) * & |
|---|
| 3639 | ( v(k,j+3,i) - v(k,j-2,i) ) & |
|---|
| 3640 | ) |
|---|
| 3641 | ! |
|---|
| 3642 | !-- k index has to be modified near bottom and top, else array |
|---|
| 3643 | !-- subscripts will be exceeded. |
|---|
| 3644 | ibit26 = IBITS(wall_flags_0(k,j,i),26,1) |
|---|
| 3645 | ibit25 = IBITS(wall_flags_0(k,j,i),25,1) |
|---|
| 3646 | ibit24 = IBITS(wall_flags_0(k,j,i),24,1) |
|---|
| 3647 | |
|---|
| 3648 | k_ppp = k + 3 * ibit26 |
|---|
| 3649 | k_pp = k + 2 * ( 1 - ibit24 ) |
|---|
| 3650 | k_mm = k - 2 * ibit26 |
|---|
| 3651 | |
|---|
| 3652 | w_comp = w(k,j-1,i) + w(k,j,i) |
|---|
| 3653 | flux_t(k) = w_comp * ( & |
|---|
| 3654 | ( 37.0 * ibit26 * adv_mom_5 & |
|---|
| 3655 | + 7.0 * ibit25 * adv_mom_3 & |
|---|
| 3656 | + ibit24 * adv_mom_1 & |
|---|
| 3657 | ) * & |
|---|
| 3658 | ( v(k+1,j,i) + v(k,j,i) ) & |
|---|
| 3659 | - ( 8.0 * ibit26 * adv_mom_5 & |
|---|
| 3660 | + ibit25 * adv_mom_3 & |
|---|
| 3661 | ) * & |
|---|
| 3662 | ( v(k_pp,j,i) + v(k-1,j,i) ) & |
|---|
| 3663 | + ( ibit26 * adv_mom_5 & |
|---|
| 3664 | ) * & |
|---|
| 3665 | ( v(k_ppp,j,i) + v(k_mm,j,i) ) & |
|---|
| 3666 | ) |
|---|
| 3667 | |
|---|
| 3668 | diss_t(k) = - ABS( w_comp ) * ( & |
|---|
| 3669 | ( 10.0 * ibit26 * adv_mom_5 & |
|---|
| 3670 | + 3.0 * ibit25 * adv_mom_3 & |
|---|
| 3671 | + ibit24 * adv_mom_1 & |
|---|
| 3672 | ) * & |
|---|
| 3673 | ( v(k+1,j,i) - v(k,j,i) ) & |
|---|
| 3674 | - ( 5.0 * ibit26 * adv_mom_5 & |
|---|
| 3675 | + ibit25 * adv_mom_3 & |
|---|
| 3676 | ) * & |
|---|
| 3677 | ( v(k_pp,j,i) - v(k-1,j,i) ) & |
|---|
| 3678 | + ( ibit26 * adv_mom_5 & |
|---|
| 3679 | ) * & |
|---|
| 3680 | ( v(k_ppp,j,i) - v(k_mm,j,i) ) & |
|---|
| 3681 | ) |
|---|
| 3682 | ! |
|---|
| 3683 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 3684 | !-- correction is needed to overcome numerical instabilities caused |
|---|
| 3685 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 3686 | div = ( ( u_comp + gu - ( u(k,j-1,i) + u(k,j,i) ) ) * ddx & |
|---|
| 3687 | + ( v_comp(k) - ( v(k,j,i) + v(k,j-1,i) ) ) * ddy & |
|---|
| 3688 | + ( w_comp - ( w(k-1,j-1,i) + w(k-1,j,i) ) & |
|---|
| 3689 | ) * ddzw(k) & |
|---|
| 3690 | ) * 0.5 |
|---|
| 3691 | |
|---|
| 3692 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 3693 | ( flux_r(k) + diss_r(k) & |
|---|
| 3694 | - swap_flux_x_local_v(k,j) - swap_diss_x_local_v(k,j) & |
|---|
| 3695 | ) * ddx & |
|---|
| 3696 | + ( flux_n(k) + diss_n(k) & |
|---|
| 3697 | - swap_flux_y_local_v(k) - swap_diss_y_local_v(k) & |
|---|
| 3698 | ) * ddy & |
|---|
| 3699 | + ( flux_t(k) + diss_t(k) & |
|---|
| 3700 | - flux_d - diss_d & |
|---|
| 3701 | ) * ddzw(k) & |
|---|
| 3702 | ) + v(k,j,i) * div |
|---|
| 3703 | |
|---|
| 3704 | swap_flux_x_local_v(k,j) = flux_r(k) |
|---|
| 3705 | swap_diss_x_local_v(k,j) = diss_r(k) |
|---|
| 3706 | swap_flux_y_local_v(k) = flux_n(k) |
|---|
| 3707 | swap_diss_y_local_v(k) = diss_n(k) |
|---|
| 3708 | flux_d = flux_t(k) |
|---|
| 3709 | diss_d = diss_t(k) |
|---|
| 3710 | |
|---|
| 3711 | ! |
|---|
| 3712 | !-- Statistical Evaluation of v'v'. The factor has to be applied |
|---|
| 3713 | !-- for right evaluation when gallilei_trans = .T. . |
|---|
| 3714 | sums_vs2_ws_l(k,tn) = sums_vs2_ws_l(k,tn) & |
|---|
| 3715 | + ( flux_n(k) & |
|---|
| 3716 | * ( v_comp(k) - 2.0 * hom(k,1,2,0) ) & |
|---|
| 3717 | / ( v_comp(k) - gv + 1.0E-20 ) & |
|---|
| 3718 | + diss_n(k) & |
|---|
| 3719 | * ABS( v_comp(k) - 2.0 * hom(k,1,2,0) ) & |
|---|
| 3720 | / ( ABS( v_comp(k) - gv ) +1.0E-20 ) ) & |
|---|
| 3721 | * weight_substep(intermediate_timestep_count) |
|---|
| 3722 | ! |
|---|
| 3723 | !-- Statistical Evaluation of w'v'. |
|---|
| 3724 | sums_wsvs_ws_l(k,tn) = sums_wsvs_ws_l(k,tn) & |
|---|
| 3725 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 3726 | * weight_substep(intermediate_timestep_count) |
|---|
| 3727 | |
|---|
| 3728 | ENDDO |
|---|
| 3729 | |
|---|
| 3730 | DO k = nzb_max+1, nzt |
|---|
| 3731 | |
|---|
| 3732 | u_comp = u(k,j-1,i+1) + u(k,j,i+1) - gu |
|---|
| 3733 | flux_r(k) = u_comp * ( & |
|---|
| 3734 | 37.0 * ( v(k,j,i+1) + v(k,j,i) ) & |
|---|
| 3735 | - 8.0 * ( v(k,j,i+2) + v(k,j,i-1) ) & |
|---|
| 3736 | + ( v(k,j,i+3) + v(k,j,i-2) ) ) * adv_mom_5 |
|---|
| 3737 | |
|---|
| 3738 | diss_r(k) = - ABS( u_comp ) * ( & |
|---|
| 3739 | 10.0 * ( v(k,j,i+1) - v(k,j,i) ) & |
|---|
| 3740 | - 5.0 * ( v(k,j,i+2) - v(k,j,i-1) ) & |
|---|
| 3741 | + ( v(k,j,i+3) - v(k,j,i-2) ) ) * adv_mom_5 |
|---|
| 3742 | |
|---|
| 3743 | |
|---|
| 3744 | v_comp(k) = v(k,j+1,i) + v(k,j,i) |
|---|
| 3745 | flux_n(k) = ( v_comp(k) - gv ) * ( & |
|---|
| 3746 | 37.0 * ( v(k,j+1,i) + v(k,j,i) ) & |
|---|
| 3747 | - 8.0 * ( v(k,j+2,i) + v(k,j-1,i) ) & |
|---|
| 3748 | + ( v(k,j+3,i) + v(k,j-2,i) ) ) * adv_mom_5 |
|---|
| 3749 | |
|---|
| 3750 | diss_n(k) = - ABS( v_comp(k) - gv ) * ( & |
|---|
| 3751 | 10.0 * ( v(k,j+1,i) - v(k,j,i) ) & |
|---|
| 3752 | - 5.0 * ( v(k,j+2,i) - v(k,j-1,i) ) & |
|---|
| 3753 | + ( v(k,j+3,i) - v(k,j-2,i) ) ) * adv_mom_5 |
|---|
| 3754 | ! |
|---|
| 3755 | !-- k index has to be modified near bottom and top, else array |
|---|
| 3756 | !-- subscripts will be exceeded. |
|---|
| 3757 | ibit26 = IBITS(wall_flags_0(k,j,i),26,1) |
|---|
| 3758 | ibit25 = IBITS(wall_flags_0(k,j,i),25,1) |
|---|
| 3759 | ibit24 = IBITS(wall_flags_0(k,j,i),24,1) |
|---|
| 3760 | |
|---|
| 3761 | k_ppp = k + 3 * ibit26 |
|---|
| 3762 | k_pp = k + 2 * ( 1 - ibit24 ) |
|---|
| 3763 | k_mm = k - 2 * ibit26 |
|---|
| 3764 | |
|---|
| 3765 | w_comp = w(k,j-1,i) + w(k,j,i) |
|---|
| 3766 | flux_t(k) = w_comp * ( & |
|---|
| 3767 | ( 37.0 * ibit26 * adv_mom_5 & |
|---|
| 3768 | + 7.0 * ibit25 * adv_mom_3 & |
|---|
| 3769 | + ibit24 * adv_mom_1 & |
|---|
| 3770 | ) * & |
|---|
| 3771 | ( v(k+1,j,i) + v(k,j,i) ) & |
|---|
| 3772 | - ( 8.0 * ibit26 * adv_mom_5 & |
|---|
| 3773 | + ibit25 * adv_mom_3 & |
|---|
| 3774 | ) * & |
|---|
| 3775 | ( v(k_pp,j,i) + v(k-1,j,i) ) & |
|---|
| 3776 | + ( ibit26 * adv_mom_5 & |
|---|
| 3777 | ) * & |
|---|
| 3778 | ( v(k_ppp,j,i) + v(k_mm,j,i) ) & |
|---|
| 3779 | ) |
|---|
| 3780 | |
|---|
| 3781 | diss_t(k) = - ABS( w_comp ) * ( & |
|---|
| 3782 | ( 10.0 * ibit26 * adv_mom_5 & |
|---|
| 3783 | + 3.0 * ibit25 * adv_mom_3 & |
|---|
| 3784 | + ibit24 * adv_mom_1 & |
|---|
| 3785 | ) * & |
|---|
| 3786 | ( v(k+1,j,i) - v(k,j,i) ) & |
|---|
| 3787 | - ( 5.0 * ibit26 * adv_mom_5 & |
|---|
| 3788 | + ibit25 * adv_mom_3 & |
|---|
| 3789 | ) * & |
|---|
| 3790 | ( v(k_pp,j,i) - v(k-1,j,i) ) & |
|---|
| 3791 | + ( ibit26 * adv_mom_5 & |
|---|
| 3792 | ) * & |
|---|
| 3793 | ( v(k_ppp,j,i) - v(k_mm,j,i) ) & |
|---|
| 3794 | ) |
|---|
| 3795 | ! |
|---|
| 3796 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 3797 | !-- correction is needed to overcome numerical instabilities caused |
|---|
| 3798 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 3799 | div = ( ( u_comp + gu - ( u(k,j-1,i) + u(k,j,i) ) ) * ddx & |
|---|
| 3800 | + ( v_comp(k) - ( v(k,j,i) + v(k,j-1,i) ) ) * ddy & |
|---|
| 3801 | + ( w_comp - ( w(k-1,j-1,i) + w(k-1,j,i) ) ) & |
|---|
| 3802 | * ddzw(k) & |
|---|
| 3803 | ) * 0.5 |
|---|
| 3804 | |
|---|
| 3805 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 3806 | ( flux_r(k) + diss_r(k) & |
|---|
| 3807 | - swap_flux_x_local_v(k,j) - swap_diss_x_local_v(k,j) & |
|---|
| 3808 | ) * ddx & |
|---|
| 3809 | + ( flux_n(k) + diss_n(k) & |
|---|
| 3810 | - swap_flux_y_local_v(k) - swap_diss_y_local_v(k) & |
|---|
| 3811 | ) * ddy & |
|---|
| 3812 | + ( flux_t(k) + diss_t(k) & |
|---|
| 3813 | - flux_d - diss_d & |
|---|
| 3814 | ) * ddzw(k) & |
|---|
| 3815 | ) + v(k,j,i) * div |
|---|
| 3816 | |
|---|
| 3817 | swap_flux_x_local_v(k,j) = flux_r(k) |
|---|
| 3818 | swap_diss_x_local_v(k,j) = diss_r(k) |
|---|
| 3819 | swap_flux_y_local_v(k) = flux_n(k) |
|---|
| 3820 | swap_diss_y_local_v(k) = diss_n(k) |
|---|
| 3821 | flux_d = flux_t(k) |
|---|
| 3822 | diss_d = diss_t(k) |
|---|
| 3823 | |
|---|
| 3824 | ! |
|---|
| 3825 | !-- Statistical Evaluation of v'v'. The factor has to be applied |
|---|
| 3826 | !-- for right evaluation when gallilei_trans = .T. . |
|---|
| 3827 | sums_vs2_ws_l(k,tn) = sums_vs2_ws_l(k,tn) & |
|---|
| 3828 | + ( flux_n(k) & |
|---|
| 3829 | * ( v_comp(k) - 2.0 * hom(k,1,2,0) ) & |
|---|
| 3830 | / ( v_comp(k) - gv + 1.0E-20 ) & |
|---|
| 3831 | + diss_n(k) & |
|---|
| 3832 | * ABS( v_comp(k) - 2.0 * hom(k,1,2,0) ) & |
|---|
| 3833 | / ( ABS( v_comp(k) - gv ) +1.0E-20 ) ) & |
|---|
| 3834 | * weight_substep(intermediate_timestep_count) |
|---|
| 3835 | ! |
|---|
| 3836 | !-- Statistical Evaluation of w'v'. |
|---|
| 3837 | sums_wsvs_ws_l(k,tn) = sums_wsvs_ws_l(k,tn) & |
|---|
| 3838 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 3839 | * weight_substep(intermediate_timestep_count) |
|---|
| 3840 | |
|---|
| 3841 | ENDDO |
|---|
| 3842 | ENDDO |
|---|
| 3843 | ENDDO |
|---|
| 3844 | sums_vs2_ws_l(nzb,tn) = sums_vs2_ws_l(nzb+1,tn) |
|---|
| 3845 | |
|---|
| 3846 | |
|---|
| 3847 | END SUBROUTINE advec_v_ws |
|---|
| 3848 | |
|---|
| 3849 | |
|---|
| 3850 | !------------------------------------------------------------------------------! |
|---|
| 3851 | ! Advection of v - Call for all grid points - accelerator version |
|---|
| 3852 | !------------------------------------------------------------------------------! |
|---|
| 3853 | SUBROUTINE advec_v_ws_acc |
|---|
| 3854 | |
|---|
| 3855 | USE arrays_3d |
|---|
| 3856 | USE constants |
|---|
| 3857 | USE control_parameters |
|---|
| 3858 | USE grid_variables |
|---|
| 3859 | USE indices |
|---|
| 3860 | USE statistics |
|---|
| 3861 | |
|---|
| 3862 | IMPLICIT NONE |
|---|
| 3863 | |
|---|
| 3864 | |
|---|
| 3865 | INTEGER :: i, ibit18, ibit19, ibit20, ibit21, ibit22, ibit23, ibit24, & |
|---|
| 3866 | ibit25, ibit26, j, k, k_mm, k_mmm, k_pp, k_ppp, tn = 0 |
|---|
| 3867 | |
|---|
| 3868 | REAL :: diss_d, diss_l, diss_n, diss_r, diss_s, diss_t, div, & |
|---|
| 3869 | flux_d, flux_l, flux_n, flux_r, flux_s, flux_t, gu, gv, & |
|---|
| 3870 | u_comp, u_comp_l, v_comp, v_comp_s, w_comp |
|---|
| 3871 | |
|---|
| 3872 | gu = 2.0 * u_gtrans |
|---|
| 3873 | gv = 2.0 * v_gtrans |
|---|
| 3874 | |
|---|
| 3875 | ! |
|---|
| 3876 | !-- Computation of fluxes and tendency terms |
|---|
| 3877 | !$acc kernels present( ddzw, tend, u, v, w, wall_flags_0, wall_flags_00 ) |
|---|
| 3878 | !$acc loop |
|---|
| 3879 | DO i = i_left, i_right |
|---|
| 3880 | DO j = j_south, j_north |
|---|
| 3881 | !$acc loop vector( 32 ) |
|---|
| 3882 | DO k = nzb+1, nzt |
|---|
| 3883 | |
|---|
| 3884 | ibit20 = IBITS(wall_flags_0(k,j,i),20,1) |
|---|
| 3885 | ibit19 = IBITS(wall_flags_0(k,j,i),19,1) |
|---|
| 3886 | ibit18 = IBITS(wall_flags_0(k,j,i),18,1) |
|---|
| 3887 | |
|---|
| 3888 | u_comp_l = u(k,j-1,i) + u(k,j,i) - gu |
|---|
| 3889 | flux_l = u_comp_l * ( & |
|---|
| 3890 | ( 37.0 * ibit20 * adv_mom_5 & |
|---|
| 3891 | + 7.0 * ibit19 * adv_mom_3 & |
|---|
| 3892 | + ibit18 * adv_mom_1 & |
|---|
| 3893 | ) * & |
|---|
| 3894 | ( v(k,j,i) + v(k,j,i-1) ) & |
|---|
| 3895 | - ( 8.0 * ibit20 * adv_mom_5 & |
|---|
| 3896 | + ibit19 * adv_mom_3 & |
|---|
| 3897 | ) * & |
|---|
| 3898 | ( v(k,j,i+1) + v(k,j,i-2) ) & |
|---|
| 3899 | + ( ibit20 * adv_mom_5 & |
|---|
| 3900 | ) * & |
|---|
| 3901 | ( v(k,j,i+2) + v(k,j,i-3) ) & |
|---|
| 3902 | ) |
|---|
| 3903 | |
|---|
| 3904 | diss_l = - ABS( u_comp_l ) * ( & |
|---|
| 3905 | ( 10.0 * ibit20 * adv_mom_5 & |
|---|
| 3906 | + 3.0 * ibit19 * adv_mom_3 & |
|---|
| 3907 | + ibit18 * adv_mom_1 & |
|---|
| 3908 | ) * & |
|---|
| 3909 | ( v(k,j,i) - v(k,j,i-1) ) & |
|---|
| 3910 | - ( 5.0 * ibit20 * adv_mom_5 & |
|---|
| 3911 | + ibit19 * adv_mom_3 & |
|---|
| 3912 | ) * & |
|---|
| 3913 | ( v(k,j,i+1) - v(k,j,i-2) ) & |
|---|
| 3914 | + ( ibit20 * adv_mom_5 & |
|---|
| 3915 | ) * & |
|---|
| 3916 | ( v(k,j,i+2) - v(k,j,i-3) ) & |
|---|
| 3917 | ) |
|---|
| 3918 | |
|---|
| 3919 | u_comp = u(k,j-1,i+1) + u(k,j,i+1) - gu |
|---|
| 3920 | flux_r = u_comp * ( & |
|---|
| 3921 | ( 37.0 * ibit20 * adv_mom_5 & |
|---|
| 3922 | + 7.0 * ibit19 * adv_mom_3 & |
|---|
| 3923 | + ibit18 * adv_mom_1 & |
|---|
| 3924 | ) * & |
|---|
| 3925 | ( v(k,j,i+1) + v(k,j,i) ) & |
|---|
| 3926 | - ( 8.0 * ibit20 * adv_mom_5 & |
|---|
| 3927 | + ibit19 * adv_mom_3 & |
|---|
| 3928 | ) * & |
|---|
| 3929 | ( v(k,j,i+2) + v(k,j,i-1) ) & |
|---|
| 3930 | + ( ibit20 * adv_mom_5 & |
|---|
| 3931 | ) * & |
|---|
| 3932 | ( v(k,j,i+3) + v(k,j,i-2) ) & |
|---|
| 3933 | ) |
|---|
| 3934 | |
|---|
| 3935 | diss_r = - ABS( u_comp ) * ( & |
|---|
| 3936 | ( 10.0 * ibit20 * adv_mom_5 & |
|---|
| 3937 | + 3.0 * ibit19 * adv_mom_3 & |
|---|
| 3938 | + ibit18 * adv_mom_1 & |
|---|
| 3939 | ) * & |
|---|
| 3940 | ( v(k,j,i+1) - v(k,j,i) ) & |
|---|
| 3941 | - ( 5.0 * ibit20 * adv_mom_5 & |
|---|
| 3942 | + ibit19 * adv_mom_3 & |
|---|
| 3943 | ) * & |
|---|
| 3944 | ( v(k,j,i+2) - v(k,j,i-1) ) & |
|---|
| 3945 | + ( ibit20 * adv_mom_5 & |
|---|
| 3946 | ) * & |
|---|
| 3947 | ( v(k,j,i+3) - v(k,j,i-2) ) & |
|---|
| 3948 | ) |
|---|
| 3949 | |
|---|
| 3950 | ibit23 = IBITS(wall_flags_0(k,j,i),23,1) |
|---|
| 3951 | ibit22 = IBITS(wall_flags_0(k,j,i),22,1) |
|---|
| 3952 | ibit21 = IBITS(wall_flags_0(k,j,i),21,1) |
|---|
| 3953 | |
|---|
| 3954 | |
|---|
| 3955 | v_comp_s = v(k,j,i) + v(k,j-1,i) - gv |
|---|
| 3956 | flux_s = v_comp_s * ( & |
|---|
| 3957 | ( 37.0 * ibit23 * adv_mom_5 & |
|---|
| 3958 | + 7.0 * ibit22 * adv_mom_3 & |
|---|
| 3959 | + ibit21 * adv_mom_1 & |
|---|
| 3960 | ) * & |
|---|
| 3961 | ( v(k,j,i) + v(k,j-1,i) ) & |
|---|
| 3962 | - ( 8.0 * ibit23 * adv_mom_5 & |
|---|
| 3963 | + ibit22 * adv_mom_3 & |
|---|
| 3964 | ) * & |
|---|
| 3965 | ( v(k,j+1,i) + v(k,j-2,i) ) & |
|---|
| 3966 | + ( ibit23 * adv_mom_5 & |
|---|
| 3967 | ) * & |
|---|
| 3968 | ( v(k,j+2,i) + v(k,j-3,i) ) & |
|---|
| 3969 | ) |
|---|
| 3970 | |
|---|
| 3971 | diss_s = - ABS( v_comp_s ) * ( & |
|---|
| 3972 | ( 10.0 * ibit23 * adv_mom_5 & |
|---|
| 3973 | + 3.0 * ibit22 * adv_mom_3 & |
|---|
| 3974 | + ibit21 * adv_mom_1 & |
|---|
| 3975 | ) * & |
|---|
| 3976 | ( v(k,j,i) - v(k,j-1,i) ) & |
|---|
| 3977 | - ( 5.0 * ibit23 * adv_mom_5 & |
|---|
| 3978 | + ibit22 * adv_mom_3 & |
|---|
| 3979 | ) * & |
|---|
| 3980 | ( v(k,j+1,i) - v(k,j-2,i) ) & |
|---|
| 3981 | + ( ibit23 * adv_mom_5 & |
|---|
| 3982 | ) * & |
|---|
| 3983 | ( v(k,j+2,i) - v(k,j-3,i) ) & |
|---|
| 3984 | ) |
|---|
| 3985 | |
|---|
| 3986 | v_comp = v(k,j+1,i) + v(k,j,i) |
|---|
| 3987 | flux_n = ( v_comp - gv ) * ( & |
|---|
| 3988 | ( 37.0 * ibit23 * adv_mom_5 & |
|---|
| 3989 | + 7.0 * ibit22 * adv_mom_3 & |
|---|
| 3990 | + ibit21 * adv_mom_1 & |
|---|
| 3991 | ) * & |
|---|
| 3992 | ( v(k,j+1,i) + v(k,j,i) ) & |
|---|
| 3993 | - ( 8.0 * ibit23 * adv_mom_5 & |
|---|
| 3994 | + ibit22 * adv_mom_3 & |
|---|
| 3995 | ) * & |
|---|
| 3996 | ( v(k,j+2,i) + v(k,j-1,i) ) & |
|---|
| 3997 | + ( ibit23 * adv_mom_5 & |
|---|
| 3998 | ) * & |
|---|
| 3999 | ( v(k,j+3,i) + v(k,j-2,i) ) & |
|---|
| 4000 | ) |
|---|
| 4001 | |
|---|
| 4002 | diss_n = - ABS( v_comp - gv ) * ( & |
|---|
| 4003 | ( 10.0 * ibit23 * adv_mom_5 & |
|---|
| 4004 | + 3.0 * ibit22 * adv_mom_3 & |
|---|
| 4005 | + ibit21 * adv_mom_1 & |
|---|
| 4006 | ) * & |
|---|
| 4007 | ( v(k,j+1,i) - v(k,j,i) ) & |
|---|
| 4008 | - ( 5.0 * ibit23 * adv_mom_5 & |
|---|
| 4009 | + ibit22 * adv_mom_3 & |
|---|
| 4010 | ) * & |
|---|
| 4011 | ( v(k,j+2,i) - v(k,j-1,i) ) & |
|---|
| 4012 | + ( ibit23 * adv_mom_5 & |
|---|
| 4013 | ) * & |
|---|
| 4014 | ( v(k,j+3,i) - v(k,j-2,i) ) & |
|---|
| 4015 | ) |
|---|
| 4016 | |
|---|
| 4017 | ibit26 = IBITS(wall_flags_0(k-1,j,i),26,1) |
|---|
| 4018 | ibit25 = IBITS(wall_flags_0(k-1,j,i),25,1) |
|---|
| 4019 | ibit24 = IBITS(wall_flags_0(k-1,j,i),24,1) |
|---|
| 4020 | |
|---|
| 4021 | k_pp = k + 2 * ibit26 |
|---|
| 4022 | k_mm = k - 2 * ( ibit25 + ibit26 ) |
|---|
| 4023 | k_mmm = k - 3 * ibit26 |
|---|
| 4024 | |
|---|
| 4025 | w_comp = w(k-1,j-1,i) + w(k-1,j,i) |
|---|
| 4026 | flux_d = w_comp * ( & |
|---|
| 4027 | ( 37.0 * ibit26 * adv_mom_5 & |
|---|
| 4028 | + 7.0 * ibit25 * adv_mom_3 & |
|---|
| 4029 | + ibit24 * adv_mom_1 & |
|---|
| 4030 | ) * & |
|---|
| 4031 | ( v(k,j,i) + v(k-1,j,i) ) & |
|---|
| 4032 | - ( 8.0 * ibit26 * adv_mom_5 & |
|---|
| 4033 | + ibit25 * adv_mom_3 & |
|---|
| 4034 | ) * & |
|---|
| 4035 | ( v(k+1,j,i) + v(k_mm,j,i) ) & |
|---|
| 4036 | + ( ibit26 * adv_mom_5 & |
|---|
| 4037 | ) * & |
|---|
| 4038 | ( v(k_pp,j,i) + v(k_mmm,j,i) ) & |
|---|
| 4039 | ) |
|---|
| 4040 | |
|---|
| 4041 | diss_d = - ABS( w_comp ) * ( & |
|---|
| 4042 | ( 10.0 * ibit26 * adv_mom_5 & |
|---|
| 4043 | + 3.0 * ibit25 * adv_mom_3 & |
|---|
| 4044 | + ibit24 * adv_mom_1 & |
|---|
| 4045 | ) * & |
|---|
| 4046 | ( v(k,j,i) - v(k-1,j,i) ) & |
|---|
| 4047 | - ( 5.0 * ibit26 * adv_mom_5 & |
|---|
| 4048 | + ibit25 * adv_mom_3 & |
|---|
| 4049 | ) * & |
|---|
| 4050 | ( v(k+1,j,i) - v(k_mm,j,i) ) & |
|---|
| 4051 | + ( ibit26 * adv_mom_5 & |
|---|
| 4052 | ) * & |
|---|
| 4053 | ( v(k_pp,j,i) - v(k_mmm,j,i) ) & |
|---|
| 4054 | ) |
|---|
| 4055 | ! |
|---|
| 4056 | !-- k index has to be modified near bottom and top, else array |
|---|
| 4057 | !-- subscripts will be exceeded. |
|---|
| 4058 | ibit26 = IBITS(wall_flags_0(k,j,i),26,1) |
|---|
| 4059 | ibit25 = IBITS(wall_flags_0(k,j,i),25,1) |
|---|
| 4060 | ibit24 = IBITS(wall_flags_0(k,j,i),24,1) |
|---|
| 4061 | |
|---|
| 4062 | k_ppp = k + 3 * ibit26 |
|---|
| 4063 | k_pp = k + 2 * ( 1 - ibit24 ) |
|---|
| 4064 | k_mm = k - 2 * ibit26 |
|---|
| 4065 | |
|---|
| 4066 | w_comp = w(k,j-1,i) + w(k,j,i) |
|---|
| 4067 | flux_t = w_comp * ( & |
|---|
| 4068 | ( 37.0 * ibit26 * adv_mom_5 & |
|---|
| 4069 | + 7.0 * ibit25 * adv_mom_3 & |
|---|
| 4070 | + ibit24 * adv_mom_1 & |
|---|
| 4071 | ) * & |
|---|
| 4072 | ( v(k+1,j,i) + v(k,j,i) ) & |
|---|
| 4073 | - ( 8.0 * ibit26 * adv_mom_5 & |
|---|
| 4074 | + ibit25 * adv_mom_3 & |
|---|
| 4075 | ) * & |
|---|
| 4076 | ( v(k_pp,j,i) + v(k-1,j,i) ) & |
|---|
| 4077 | + ( ibit26 * adv_mom_5 & |
|---|
| 4078 | ) * & |
|---|
| 4079 | ( v(k_ppp,j,i) + v(k_mm,j,i) ) & |
|---|
| 4080 | ) |
|---|
| 4081 | |
|---|
| 4082 | diss_t = - ABS( w_comp ) * ( & |
|---|
| 4083 | ( 10.0 * ibit26 * adv_mom_5 & |
|---|
| 4084 | + 3.0 * ibit25 * adv_mom_3 & |
|---|
| 4085 | + ibit24 * adv_mom_1 & |
|---|
| 4086 | ) * & |
|---|
| 4087 | ( v(k+1,j,i) - v(k,j,i) ) & |
|---|
| 4088 | - ( 5.0 * ibit26 * adv_mom_5 & |
|---|
| 4089 | + ibit25 * adv_mom_3 & |
|---|
| 4090 | ) * & |
|---|
| 4091 | ( v(k_pp,j,i) - v(k-1,j,i) ) & |
|---|
| 4092 | + ( ibit26 * adv_mom_5 & |
|---|
| 4093 | ) * & |
|---|
| 4094 | ( v(k_ppp,j,i) - v(k_mm,j,i) ) & |
|---|
| 4095 | ) |
|---|
| 4096 | ! |
|---|
| 4097 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 4098 | !-- correction is needed to overcome numerical instabilities caused |
|---|
| 4099 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 4100 | div = ( ( u_comp + gu - ( u(k,j-1,i) + u(k,j,i) ) ) * ddx & |
|---|
| 4101 | + ( v_comp - ( v(k,j,i) + v(k,j-1,i) ) ) * ddy & |
|---|
| 4102 | + ( w_comp - ( w(k-1,j-1,i) + w(k-1,j,i) ) & |
|---|
| 4103 | ) * ddzw(k) & |
|---|
| 4104 | ) * 0.5 |
|---|
| 4105 | |
|---|
| 4106 | tend(k,j,i) = - ( & |
|---|
| 4107 | ( flux_r + diss_r - flux_l - diss_l ) * ddx & |
|---|
| 4108 | + ( flux_n + diss_n - flux_s - diss_s ) * ddy & |
|---|
| 4109 | + ( flux_t + diss_t - flux_d - diss_d ) * ddzw(k) & |
|---|
| 4110 | ) + div * v(k,j,i) |
|---|
| 4111 | |
|---|
| 4112 | |
|---|
| 4113 | !++ |
|---|
| 4114 | !-- Statistical Evaluation of v'v'. The factor has to be applied |
|---|
| 4115 | !-- for right evaluation when gallilei_trans = .T. . |
|---|
| 4116 | ! sums_vs2_ws_l(k,tn) = sums_vs2_ws_l(k,tn) & |
|---|
| 4117 | ! + ( flux_n & |
|---|
| 4118 | ! * ( v_comp - 2.0 * hom(k,1,2,0) ) & |
|---|
| 4119 | ! / ( v_comp - gv + 1.0E-20 ) & |
|---|
| 4120 | ! + diss_n & |
|---|
| 4121 | ! * ABS( v_comp - 2.0 * hom(k,1,2,0) ) & |
|---|
| 4122 | ! / ( ABS( v_comp - gv ) +1.0E-20 ) ) & |
|---|
| 4123 | ! * weight_substep(intermediate_timestep_count) |
|---|
| 4124 | ! |
|---|
| 4125 | !-- Statistical Evaluation of w'v'. |
|---|
| 4126 | ! sums_wsvs_ws_l(k,tn) = sums_wsvs_ws_l(k,tn) & |
|---|
| 4127 | ! + ( flux_t + diss_t ) & |
|---|
| 4128 | ! * weight_substep(intermediate_timestep_count) |
|---|
| 4129 | |
|---|
| 4130 | ENDDO |
|---|
| 4131 | ENDDO |
|---|
| 4132 | ENDDO |
|---|
| 4133 | !$acc end kernels |
|---|
| 4134 | |
|---|
| 4135 | !++ |
|---|
| 4136 | ! sums_vs2_ws_l(nzb,tn) = sums_vs2_ws_l(nzb+1,tn) |
|---|
| 4137 | |
|---|
| 4138 | END SUBROUTINE advec_v_ws_acc |
|---|
| 4139 | |
|---|
| 4140 | |
|---|
| 4141 | !------------------------------------------------------------------------------! |
|---|
| 4142 | ! Advection of w - Call for all grid points |
|---|
| 4143 | !------------------------------------------------------------------------------! |
|---|
| 4144 | SUBROUTINE advec_w_ws |
|---|
| 4145 | |
|---|
| 4146 | USE arrays_3d |
|---|
| 4147 | USE constants |
|---|
| 4148 | USE control_parameters |
|---|
| 4149 | USE grid_variables |
|---|
| 4150 | USE indices |
|---|
| 4151 | USE statistics |
|---|
| 4152 | |
|---|
| 4153 | IMPLICIT NONE |
|---|
| 4154 | |
|---|
| 4155 | INTEGER :: i, ibit27, ibit28, ibit29, ibit30, ibit31, ibit32, ibit33, & |
|---|
| 4156 | ibit34, ibit35, j, k, k_mm, k_pp, k_ppp, tn = 0 |
|---|
| 4157 | REAL :: diss_d, div, flux_d, gu, gv, u_comp, v_comp, w_comp |
|---|
| 4158 | REAL, DIMENSION(nzb:nzt) :: diss_t, flux_t |
|---|
| 4159 | REAL, DIMENSION(nzb+1:nzt) :: diss_n, diss_r, flux_n, flux_r, & |
|---|
| 4160 | swap_diss_y_local_w, & |
|---|
| 4161 | swap_flux_y_local_w |
|---|
| 4162 | REAL, DIMENSION(nzb+1:nzt,nys:nyn) :: swap_diss_x_local_w, & |
|---|
| 4163 | swap_flux_x_local_w |
|---|
| 4164 | |
|---|
| 4165 | gu = 2.0 * u_gtrans |
|---|
| 4166 | gv = 2.0 * v_gtrans |
|---|
| 4167 | ! |
|---|
| 4168 | !-- compute the whole left boundary of the processor domain |
|---|
| 4169 | i = nxl |
|---|
| 4170 | DO j = nys, nyn |
|---|
| 4171 | DO k = nzb+1, nzb_max |
|---|
| 4172 | |
|---|
| 4173 | ibit29 = IBITS(wall_flags_0(k,j,i),29,1) |
|---|
| 4174 | ibit28 = IBITS(wall_flags_0(k,j,i),28,1) |
|---|
| 4175 | ibit27 = IBITS(wall_flags_0(k,j,i),27,1) |
|---|
| 4176 | |
|---|
| 4177 | u_comp = u(k+1,j,i) + u(k,j,i) - gu |
|---|
| 4178 | swap_flux_x_local_w(k,j) = u_comp * ( & |
|---|
| 4179 | ( 37.0 * ibit29 * adv_mom_5 & |
|---|
| 4180 | + 7.0 * ibit28 * adv_mom_3 & |
|---|
| 4181 | + ibit27 * adv_mom_1 & |
|---|
| 4182 | ) * & |
|---|
| 4183 | ( w(k,j,i) + w(k,j,i-1) ) & |
|---|
| 4184 | - ( 8.0 * ibit29 * adv_mom_5 & |
|---|
| 4185 | + ibit28 * adv_mom_3 & |
|---|
| 4186 | ) * & |
|---|
| 4187 | ( w(k,j,i+1) + w(k,j,i-2) ) & |
|---|
| 4188 | + ( ibit29 * adv_mom_5 & |
|---|
| 4189 | ) * & |
|---|
| 4190 | ( w(k,j,i+2) + w(k,j,i-3) ) & |
|---|
| 4191 | ) |
|---|
| 4192 | |
|---|
| 4193 | swap_diss_x_local_w(k,j) = - ABS( u_comp ) * ( & |
|---|
| 4194 | ( 10.0 * ibit29 * adv_mom_5 & |
|---|
| 4195 | + 3.0 * ibit28 * adv_mom_3 & |
|---|
| 4196 | + ibit27 * adv_mom_1 & |
|---|
| 4197 | ) * & |
|---|
| 4198 | ( w(k,j,i) - w(k,j,i-1) ) & |
|---|
| 4199 | - ( 5.0 * ibit29 * adv_mom_5 & |
|---|
| 4200 | + ibit28 * adv_mom_3 & |
|---|
| 4201 | ) * & |
|---|
| 4202 | ( w(k,j,i+1) - w(k,j,i-2) ) & |
|---|
| 4203 | + ( ibit29 * adv_mom_5 & |
|---|
| 4204 | ) * & |
|---|
| 4205 | ( w(k,j,i+2) - w(k,j,i-3) ) & |
|---|
| 4206 | ) |
|---|
| 4207 | |
|---|
| 4208 | ENDDO |
|---|
| 4209 | |
|---|
| 4210 | DO k = nzb_max+1, nzt |
|---|
| 4211 | |
|---|
| 4212 | u_comp = u(k+1,j,i) + u(k,j,i) - gu |
|---|
| 4213 | swap_flux_x_local_w(k,j) = u_comp * ( & |
|---|
| 4214 | 37.0 * ( w(k,j,i) + w(k,j,i-1) ) & |
|---|
| 4215 | - 8.0 * ( w(k,j,i+1) + w(k,j,i-2) ) & |
|---|
| 4216 | + ( w(k,j,i+2) + w(k,j,i-3) ) ) * adv_mom_5 |
|---|
| 4217 | swap_diss_x_local_w(k,j) = - ABS( u_comp ) * ( & |
|---|
| 4218 | 10.0 * ( w(k,j,i) - w(k,j,i-1) ) & |
|---|
| 4219 | - 5.0 * ( w(k,j,i+1) - w(k,j,i-2) ) & |
|---|
| 4220 | + ( w(k,j,i+2) - w(k,j,i-3) ) ) * adv_mom_5 |
|---|
| 4221 | |
|---|
| 4222 | ENDDO |
|---|
| 4223 | |
|---|
| 4224 | ENDDO |
|---|
| 4225 | |
|---|
| 4226 | DO i = nxl, nxr |
|---|
| 4227 | |
|---|
| 4228 | j = nys |
|---|
| 4229 | DO k = nzb+1, nzb_max |
|---|
| 4230 | |
|---|
| 4231 | ibit32 = IBITS(wall_flags_00(k,j,i),0,1) |
|---|
| 4232 | ibit31 = IBITS(wall_flags_0(k,j,i),31,1) |
|---|
| 4233 | ibit30 = IBITS(wall_flags_0(k,j,i),30,1) |
|---|
| 4234 | |
|---|
| 4235 | v_comp = v(k+1,j,i) + v(k,j,i) - gv |
|---|
| 4236 | swap_flux_y_local_w(k) = v_comp * ( & |
|---|
| 4237 | ( 37.0 * ibit32 * adv_mom_5 & |
|---|
| 4238 | + 7.0 * ibit31 * adv_mom_3 & |
|---|
| 4239 | + ibit30 * adv_mom_1 & |
|---|
| 4240 | ) * & |
|---|
| 4241 | ( w(k,j,i) + w(k,j-1,i) ) & |
|---|
| 4242 | - ( 8.0 * ibit32 * adv_mom_5 & |
|---|
| 4243 | + ibit31 * adv_mom_3 & |
|---|
| 4244 | ) * & |
|---|
| 4245 | ( w(k,j+1,i) + w(k,j-2,i) ) & |
|---|
| 4246 | + ( ibit32 * adv_mom_5 & |
|---|
| 4247 | ) * & |
|---|
| 4248 | ( w(k,j+2,i) + w(k,j-3,i) ) & |
|---|
| 4249 | ) |
|---|
| 4250 | |
|---|
| 4251 | swap_diss_y_local_w(k) = - ABS( v_comp ) * ( & |
|---|
| 4252 | ( 10.0 * ibit32 * adv_mom_5 & |
|---|
| 4253 | + 3.0 * ibit31 * adv_mom_3 & |
|---|
| 4254 | + ibit30 * adv_mom_1 & |
|---|
| 4255 | ) * & |
|---|
| 4256 | ( w(k,j,i) - w(k,j-1,i) ) & |
|---|
| 4257 | - ( 5.0 * ibit32 * adv_mom_5 & |
|---|
| 4258 | + ibit31 * adv_mom_3 & |
|---|
| 4259 | ) * & |
|---|
| 4260 | ( w(k,j+1,i) - w(k,j-2,i) ) & |
|---|
| 4261 | + ( ibit32 * adv_mom_5 & |
|---|
| 4262 | ) * & |
|---|
| 4263 | ( w(k,j+2,i) - w(k,j-3,i) ) & |
|---|
| 4264 | ) |
|---|
| 4265 | |
|---|
| 4266 | ENDDO |
|---|
| 4267 | |
|---|
| 4268 | DO k = nzb_max+1, nzt |
|---|
| 4269 | |
|---|
| 4270 | v_comp = v(k+1,j,i) + v(k,j,i) - gv |
|---|
| 4271 | swap_flux_y_local_w(k) = v_comp * ( & |
|---|
| 4272 | 37.0 * ( w(k,j,i) + w(k,j-1,i) ) & |
|---|
| 4273 | - 8.0 * ( w(k,j+1,i) +w(k,j-2,i) ) & |
|---|
| 4274 | + ( w(k,j+2,i) + w(k,j-3,i) ) ) * adv_mom_5 |
|---|
| 4275 | swap_diss_y_local_w(k) = - ABS( v_comp ) * ( & |
|---|
| 4276 | 10.0 * ( w(k,j,i) - w(k,j-1,i) ) & |
|---|
| 4277 | - 5.0 * ( w(k,j+1,i) - w(k,j-2,i) ) & |
|---|
| 4278 | + ( w(k,j+2,i) - w(k,j-3,i) ) ) * adv_mom_5 |
|---|
| 4279 | |
|---|
| 4280 | ENDDO |
|---|
| 4281 | |
|---|
| 4282 | DO j = nys, nyn |
|---|
| 4283 | |
|---|
| 4284 | ! |
|---|
| 4285 | !-- The lower flux has to be calculated explicetely for the tendency |
|---|
| 4286 | !-- at the first w-level. For topography wall this is done implicitely |
|---|
| 4287 | !-- by wall_flags_0. |
|---|
| 4288 | k = nzb + 1 |
|---|
| 4289 | w_comp = w(k,j,i) + w(k-1,j,i) |
|---|
| 4290 | flux_t(0) = w_comp * ( w(k,j,i) + w(k-1,j,i) ) * adv_mom_1 |
|---|
| 4291 | diss_t(0) = -ABS(w_comp) * ( w(k,j,i) - w(k-1,j,i) ) * adv_mom_1 |
|---|
| 4292 | flux_d = flux_t(0) |
|---|
| 4293 | diss_d = diss_t(0) |
|---|
| 4294 | |
|---|
| 4295 | DO k = nzb+1, nzb_max |
|---|
| 4296 | |
|---|
| 4297 | ibit29 = IBITS(wall_flags_0(k,j,i),29,1) |
|---|
| 4298 | ibit28 = IBITS(wall_flags_0(k,j,i),28,1) |
|---|
| 4299 | ibit27 = IBITS(wall_flags_0(k,j,i),27,1) |
|---|
| 4300 | |
|---|
| 4301 | u_comp = u(k+1,j,i+1) + u(k,j,i+1) - gu |
|---|
| 4302 | flux_r(k) = u_comp * ( & |
|---|
| 4303 | ( 37.0 * ibit29 * adv_mom_5 & |
|---|
| 4304 | + 7.0 * ibit28 * adv_mom_3 & |
|---|
| 4305 | + ibit27 * adv_mom_1 & |
|---|
| 4306 | ) * & |
|---|
| 4307 | ( w(k,j,i+1) + w(k,j,i) ) & |
|---|
| 4308 | - ( 8.0 * ibit29 * adv_mom_5 & |
|---|
| 4309 | + ibit28 * adv_mom_3 & |
|---|
| 4310 | ) * & |
|---|
| 4311 | ( w(k,j,i+2) + w(k,j,i-1) ) & |
|---|
| 4312 | + ( ibit29 * adv_mom_5 & |
|---|
| 4313 | ) * & |
|---|
| 4314 | ( w(k,j,i+3) + w(k,j,i-2) ) & |
|---|
| 4315 | ) |
|---|
| 4316 | |
|---|
| 4317 | diss_r(k) = - ABS( u_comp ) * ( & |
|---|
| 4318 | ( 10.0 * ibit29 * adv_mom_5 & |
|---|
| 4319 | + 3.0 * ibit28 * adv_mom_3 & |
|---|
| 4320 | + ibit27 * adv_mom_1 & |
|---|
| 4321 | ) * & |
|---|
| 4322 | ( w(k,j,i+1) - w(k,j,i) ) & |
|---|
| 4323 | - ( 5.0 * ibit29 * adv_mom_5 & |
|---|
| 4324 | + ibit28 * adv_mom_3 & |
|---|
| 4325 | ) * & |
|---|
| 4326 | ( w(k,j,i+2) - w(k,j,i-1) ) & |
|---|
| 4327 | + ( ibit29 * adv_mom_5 & |
|---|
| 4328 | ) * & |
|---|
| 4329 | ( w(k,j,i+3) - w(k,j,i-2) ) & |
|---|
| 4330 | ) |
|---|
| 4331 | |
|---|
| 4332 | ibit32 = IBITS(wall_flags_00(k,j,i),0,1) |
|---|
| 4333 | ibit31 = IBITS(wall_flags_0(k,j,i),31,1) |
|---|
| 4334 | ibit30 = IBITS(wall_flags_0(k,j,i),30,1) |
|---|
| 4335 | |
|---|
| 4336 | v_comp = v(k+1,j+1,i) + v(k,j+1,i) - gv |
|---|
| 4337 | flux_n(k) = v_comp * ( & |
|---|
| 4338 | ( 37.0 * ibit32 * adv_mom_5 & |
|---|
| 4339 | + 7.0 * ibit31 * adv_mom_3 & |
|---|
| 4340 | + ibit30 * adv_mom_1 & |
|---|
| 4341 | ) * & |
|---|
| 4342 | ( w(k,j+1,i) + w(k,j,i) ) & |
|---|
| 4343 | - ( 8.0 * ibit32 * adv_mom_5 & |
|---|
| 4344 | + ibit31 * adv_mom_3 & |
|---|
| 4345 | ) * & |
|---|
| 4346 | ( w(k,j+2,i) + w(k,j-1,i) ) & |
|---|
| 4347 | + ( ibit32 * adv_mom_5 & |
|---|
| 4348 | ) * & |
|---|
| 4349 | ( w(k,j+3,i) + w(k,j-2,i) ) & |
|---|
| 4350 | ) |
|---|
| 4351 | |
|---|
| 4352 | diss_n(k) = - ABS( v_comp ) * ( & |
|---|
| 4353 | ( 10.0 * ibit32 * adv_mom_5 & |
|---|
| 4354 | + 3.0 * ibit31 * adv_mom_3 & |
|---|
| 4355 | + ibit30 * adv_mom_1 & |
|---|
| 4356 | ) * & |
|---|
| 4357 | ( w(k,j+1,i) - w(k,j,i) ) & |
|---|
| 4358 | - ( 5.0 * ibit32 * adv_mom_5 & |
|---|
| 4359 | + ibit31 * adv_mom_3 & |
|---|
| 4360 | ) * & |
|---|
| 4361 | ( w(k,j+2,i) - w(k,j-1,i) ) & |
|---|
| 4362 | + ( ibit32 * adv_mom_5 & |
|---|
| 4363 | ) * & |
|---|
| 4364 | ( w(k,j+3,i) - w(k,j-2,i) ) & |
|---|
| 4365 | ) |
|---|
| 4366 | ! |
|---|
| 4367 | !-- k index has to be modified near bottom and top, else array |
|---|
| 4368 | !-- subscripts will be exceeded. |
|---|
| 4369 | ibit35 = IBITS(wall_flags_00(k,j,i),3,1) |
|---|
| 4370 | ibit34 = IBITS(wall_flags_00(k,j,i),2,1) |
|---|
| 4371 | ibit33 = IBITS(wall_flags_00(k,j,i),1,1) |
|---|
| 4372 | |
|---|
| 4373 | k_ppp = k + 3 * ibit35 |
|---|
| 4374 | k_pp = k + 2 * ( 1 - ibit33 ) |
|---|
| 4375 | k_mm = k - 2 * ibit35 |
|---|
| 4376 | |
|---|
| 4377 | w_comp = w(k+1,j,i) + w(k,j,i) |
|---|
| 4378 | flux_t(k) = w_comp * ( & |
|---|
| 4379 | ( 37.0 * ibit35 * adv_mom_5 & |
|---|
| 4380 | + 7.0 * ibit34 * adv_mom_3 & |
|---|
| 4381 | + ibit33 * adv_mom_1 & |
|---|
| 4382 | ) * & |
|---|
| 4383 | ( w(k+1,j,i) + w(k,j,i) ) & |
|---|
| 4384 | - ( 8.0 * ibit35 * adv_mom_5 & |
|---|
| 4385 | + ibit34 * adv_mom_3 & |
|---|
| 4386 | ) * & |
|---|
| 4387 | ( w(k_pp,j,i) + w(k-1,j,i) ) & |
|---|
| 4388 | + ( ibit35 * adv_mom_5 & |
|---|
| 4389 | ) * & |
|---|
| 4390 | ( w(k_ppp,j,i) + w(k_mm,j,i) ) & |
|---|
| 4391 | ) |
|---|
| 4392 | |
|---|
| 4393 | diss_t(k) = - ABS( w_comp ) * ( & |
|---|
| 4394 | ( 10.0 * ibit35 * adv_mom_5 & |
|---|
| 4395 | + 3.0 * ibit34 * adv_mom_3 & |
|---|
| 4396 | + ibit33 * adv_mom_1 & |
|---|
| 4397 | ) * & |
|---|
| 4398 | ( w(k+1,j,i) - w(k,j,i) ) & |
|---|
| 4399 | - ( 5.0 * ibit35 * adv_mom_5 & |
|---|
| 4400 | + ibit34 * adv_mom_3 & |
|---|
| 4401 | ) * & |
|---|
| 4402 | ( w(k_pp,j,i) - w(k-1,j,i) ) & |
|---|
| 4403 | + ( ibit35 * adv_mom_5 & |
|---|
| 4404 | ) * & |
|---|
| 4405 | ( w(k_ppp,j,i) - w(k_mm,j,i) ) & |
|---|
| 4406 | ) |
|---|
| 4407 | ! |
|---|
| 4408 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 4409 | !-- correction is needed to overcome numerical instabilities caused |
|---|
| 4410 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 4411 | div = ( ( u_comp + gu - ( u(k+1,j,i) + u(k,j,i) ) ) * ddx & |
|---|
| 4412 | + ( v_comp + gv - ( v(k+1,j,i) + v(k,j,i) ) ) * ddy & |
|---|
| 4413 | + ( w_comp - ( w(k,j,i) + w(k-1,j,i) ) ) & |
|---|
| 4414 | * ddzu(k+1) & |
|---|
| 4415 | ) * 0.5 |
|---|
| 4416 | |
|---|
| 4417 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 4418 | ( flux_r(k) + diss_r(k) & |
|---|
| 4419 | - swap_flux_x_local_w(k,j) - swap_diss_x_local_w(k,j) & |
|---|
| 4420 | ) * ddx & |
|---|
| 4421 | + ( flux_n(k) + diss_n(k) & |
|---|
| 4422 | - swap_flux_y_local_w(k) - swap_diss_y_local_w(k) & |
|---|
| 4423 | ) * ddy & |
|---|
| 4424 | + ( flux_t(k) + diss_t(k) & |
|---|
| 4425 | - flux_d - diss_d & |
|---|
| 4426 | ) * ddzu(k+1) & |
|---|
| 4427 | ) + div * w(k,j,i) |
|---|
| 4428 | |
|---|
| 4429 | swap_flux_x_local_w(k,j) = flux_r(k) |
|---|
| 4430 | swap_diss_x_local_w(k,j) = diss_r(k) |
|---|
| 4431 | swap_flux_y_local_w(k) = flux_n(k) |
|---|
| 4432 | swap_diss_y_local_w(k) = diss_n(k) |
|---|
| 4433 | flux_d = flux_t(k) |
|---|
| 4434 | diss_d = diss_t(k) |
|---|
| 4435 | |
|---|
| 4436 | sums_ws2_ws_l(k,tn) = sums_ws2_ws_l(k,tn) & |
|---|
| 4437 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 4438 | * weight_substep(intermediate_timestep_count) |
|---|
| 4439 | |
|---|
| 4440 | ENDDO |
|---|
| 4441 | |
|---|
| 4442 | DO k = nzb_max+1, nzt |
|---|
| 4443 | |
|---|
| 4444 | u_comp = u(k+1,j,i+1) + u(k,j,i+1) - gu |
|---|
| 4445 | flux_r(k) = u_comp * ( & |
|---|
| 4446 | 37.0 * ( w(k,j,i+1) + w(k,j,i) ) & |
|---|
| 4447 | - 8.0 * ( w(k,j,i+2) + w(k,j,i-1) ) & |
|---|
| 4448 | + ( w(k,j,i+3) + w(k,j,i-2) ) ) * adv_mom_5 |
|---|
| 4449 | |
|---|
| 4450 | diss_r(k) = - ABS( u_comp ) * ( & |
|---|
| 4451 | 10.0 * ( w(k,j,i+1) - w(k,j,i) ) & |
|---|
| 4452 | - 5.0 * ( w(k,j,i+2) - w(k,j,i-1) ) & |
|---|
| 4453 | + ( w(k,j,i+3) - w(k,j,i-2) ) ) * adv_mom_5 |
|---|
| 4454 | |
|---|
| 4455 | v_comp = v(k+1,j+1,i) + v(k,j+1,i) - gv |
|---|
| 4456 | flux_n(k) = v_comp * ( & |
|---|
| 4457 | 37.0 * ( w(k,j+1,i) + w(k,j,i) ) & |
|---|
| 4458 | - 8.0 * ( w(k,j+2,i) + w(k,j-1,i) ) & |
|---|
| 4459 | + ( w(k,j+3,i) + w(k,j-2,i) ) ) * adv_mom_5 |
|---|
| 4460 | |
|---|
| 4461 | diss_n(k) = - ABS( v_comp ) * ( & |
|---|
| 4462 | 10.0 * ( w(k,j+1,i) - w(k,j,i) ) & |
|---|
| 4463 | - 5.0 * ( w(k,j+2,i) - w(k,j-1,i) ) & |
|---|
| 4464 | + ( w(k,j+3,i) - w(k,j-2,i) ) ) * adv_mom_5 |
|---|
| 4465 | ! |
|---|
| 4466 | !-- k index has to be modified near bottom and top, else array |
|---|
| 4467 | !-- subscripts will be exceeded. |
|---|
| 4468 | ibit35 = IBITS(wall_flags_00(k,j,i),3,1) |
|---|
| 4469 | ibit34 = IBITS(wall_flags_00(k,j,i),2,1) |
|---|
| 4470 | ibit33 = IBITS(wall_flags_00(k,j,i),1,1) |
|---|
| 4471 | |
|---|
| 4472 | k_ppp = k + 3 * ibit35 |
|---|
| 4473 | k_pp = k + 2 * ( 1 - ibit33 ) |
|---|
| 4474 | k_mm = k - 2 * ibit35 |
|---|
| 4475 | |
|---|
| 4476 | w_comp = w(k+1,j,i) + w(k,j,i) |
|---|
| 4477 | flux_t(k) = w_comp * ( & |
|---|
| 4478 | ( 37.0 * ibit35 * adv_mom_5 & |
|---|
| 4479 | + 7.0 * ibit34 * adv_mom_3 & |
|---|
| 4480 | + ibit33 * adv_mom_1 & |
|---|
| 4481 | ) * & |
|---|
| 4482 | ( w(k+1,j,i) + w(k,j,i) ) & |
|---|
| 4483 | - ( 8.0 * ibit35 * adv_mom_5 & |
|---|
| 4484 | + ibit34 * adv_mom_3 & |
|---|
| 4485 | ) * & |
|---|
| 4486 | ( w(k_pp,j,i) + w(k-1,j,i) ) & |
|---|
| 4487 | + ( ibit35 * adv_mom_5 & |
|---|
| 4488 | ) * & |
|---|
| 4489 | ( w(k_ppp,j,i) + w(k_mm,j,i) ) & |
|---|
| 4490 | ) |
|---|
| 4491 | |
|---|
| 4492 | diss_t(k) = - ABS( w_comp ) * ( & |
|---|
| 4493 | ( 10.0 * ibit35 * adv_mom_5 & |
|---|
| 4494 | + 3.0 * ibit34 * adv_mom_3 & |
|---|
| 4495 | + ibit33 * adv_mom_1 & |
|---|
| 4496 | ) * & |
|---|
| 4497 | ( w(k+1,j,i) - w(k,j,i) ) & |
|---|
| 4498 | - ( 5.0 * ibit35 * adv_mom_5 & |
|---|
| 4499 | + ibit34 * adv_mom_3 & |
|---|
| 4500 | ) * & |
|---|
| 4501 | ( w(k_pp,j,i) - w(k-1,j,i) ) & |
|---|
| 4502 | + ( ibit35 * adv_mom_5 & |
|---|
| 4503 | ) * & |
|---|
| 4504 | ( w(k_ppp,j,i) - w(k_mm,j,i) ) & |
|---|
| 4505 | ) |
|---|
| 4506 | ! |
|---|
| 4507 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 4508 | !-- correction is needed to overcome numerical instabilities caused |
|---|
| 4509 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 4510 | div = ( ( u_comp + gu - ( u(k+1,j,i) + u(k,j,i) ) ) * ddx & |
|---|
| 4511 | + ( v_comp + gv - ( v(k+1,j,i) + v(k,j,i) ) ) * ddy & |
|---|
| 4512 | + ( w_comp - ( w(k,j,i) + w(k-1,j,i) ) ) & |
|---|
| 4513 | * ddzu(k+1) & |
|---|
| 4514 | ) * 0.5 |
|---|
| 4515 | |
|---|
| 4516 | tend(k,j,i) = tend(k,j,i) - ( & |
|---|
| 4517 | ( flux_r(k) + diss_r(k) & |
|---|
| 4518 | - swap_flux_x_local_w(k,j) - swap_diss_x_local_w(k,j) & |
|---|
| 4519 | ) * ddx & |
|---|
| 4520 | + ( flux_n(k) + diss_n(k) & |
|---|
| 4521 | - swap_flux_y_local_w(k) - swap_diss_y_local_w(k) & |
|---|
| 4522 | ) * ddy & |
|---|
| 4523 | + ( flux_t(k) + diss_t(k) & |
|---|
| 4524 | - flux_d - diss_d & |
|---|
| 4525 | ) * ddzu(k+1) & |
|---|
| 4526 | ) + div * w(k,j,i) |
|---|
| 4527 | |
|---|
| 4528 | swap_flux_x_local_w(k,j) = flux_r(k) |
|---|
| 4529 | swap_diss_x_local_w(k,j) = diss_r(k) |
|---|
| 4530 | swap_flux_y_local_w(k) = flux_n(k) |
|---|
| 4531 | swap_diss_y_local_w(k) = diss_n(k) |
|---|
| 4532 | flux_d = flux_t(k) |
|---|
| 4533 | diss_d = diss_t(k) |
|---|
| 4534 | |
|---|
| 4535 | sums_ws2_ws_l(k,tn) = sums_ws2_ws_l(k,tn) & |
|---|
| 4536 | + ( flux_t(k) + diss_t(k) ) & |
|---|
| 4537 | * weight_substep(intermediate_timestep_count) |
|---|
| 4538 | |
|---|
| 4539 | ENDDO |
|---|
| 4540 | ENDDO |
|---|
| 4541 | ENDDO |
|---|
| 4542 | |
|---|
| 4543 | END SUBROUTINE advec_w_ws |
|---|
| 4544 | |
|---|
| 4545 | |
|---|
| 4546 | !------------------------------------------------------------------------------! |
|---|
| 4547 | ! Advection of w - Call for all grid points - accelerator version |
|---|
| 4548 | !------------------------------------------------------------------------------! |
|---|
| 4549 | SUBROUTINE advec_w_ws_acc |
|---|
| 4550 | |
|---|
| 4551 | USE arrays_3d |
|---|
| 4552 | USE constants |
|---|
| 4553 | USE control_parameters |
|---|
| 4554 | USE grid_variables |
|---|
| 4555 | USE indices |
|---|
| 4556 | USE statistics |
|---|
| 4557 | |
|---|
| 4558 | IMPLICIT NONE |
|---|
| 4559 | |
|---|
| 4560 | INTEGER :: i, ibit27, ibit28, ibit29, ibit30, ibit31, ibit32, ibit33, & |
|---|
| 4561 | ibit34, ibit35, j, k, k_mmm, k_mm, k_pp, k_ppp, tn = 0 |
|---|
| 4562 | |
|---|
| 4563 | REAL :: diss_d, diss_l, diss_n, diss_r, diss_s, diss_t, div, & |
|---|
| 4564 | flux_d, flux_l, flux_n, flux_r, flux_s, flux_t, gu, gv, & |
|---|
| 4565 | u_comp, u_comp_l, v_comp, v_comp_s, w_comp |
|---|
| 4566 | |
|---|
| 4567 | gu = 2.0 * u_gtrans |
|---|
| 4568 | gv = 2.0 * v_gtrans |
|---|
| 4569 | |
|---|
| 4570 | |
|---|
| 4571 | ! |
|---|
| 4572 | !-- Computation of fluxes and tendency terms |
|---|
| 4573 | !$acc kernels present( ddzu, tend, u, v, w, wall_flags_0, wall_flags_00 ) |
|---|
| 4574 | !$acc loop |
|---|
| 4575 | DO i = i_left, i_right |
|---|
| 4576 | DO j = j_south, j_north |
|---|
| 4577 | !$acc loop vector( 32 ) |
|---|
| 4578 | DO k = nzb+1, nzt |
|---|
| 4579 | |
|---|
| 4580 | ibit27 = IBITS(wall_flags_0(k,j,i),27,1) |
|---|
| 4581 | ibit28 = IBITS(wall_flags_0(k,j,i),28,1) |
|---|
| 4582 | ibit29 = IBITS(wall_flags_0(k,j,i),29,1) |
|---|
| 4583 | |
|---|
| 4584 | u_comp_l = u(k+1,j,i) + u(k,j,i) - gu |
|---|
| 4585 | flux_l = u_comp_l * ( & |
|---|
| 4586 | ( 37.0 * ibit29 * adv_mom_5 & |
|---|
| 4587 | + 7.0 * ibit28 * adv_mom_3 & |
|---|
| 4588 | + ibit27 * adv_mom_1 & |
|---|
| 4589 | ) * & |
|---|
| 4590 | ( w(k,j,i) + w(k,j,i-1) ) & |
|---|
| 4591 | - ( 8.0 * ibit29 * adv_mom_5 & |
|---|
| 4592 | + ibit28 * adv_mom_3 & |
|---|
| 4593 | ) * & |
|---|
| 4594 | ( w(k,j,i+1) + w(k,j,i-2) ) & |
|---|
| 4595 | + ( ibit29 * adv_mom_5 & |
|---|
| 4596 | ) * & |
|---|
| 4597 | ( w(k,j,i+2) + w(k,j,i-3) ) & |
|---|
| 4598 | ) |
|---|
| 4599 | |
|---|
| 4600 | diss_l = - ABS( u_comp_l ) * ( & |
|---|
| 4601 | ( 10.0 * ibit29 * adv_mom_5 & |
|---|
| 4602 | + 3.0 * ibit28 * adv_mom_3 & |
|---|
| 4603 | + ibit27 * adv_mom_1 & |
|---|
| 4604 | ) * & |
|---|
| 4605 | ( w(k,j,i) - w(k,j,i-1) ) & |
|---|
| 4606 | - ( 5.0 * ibit29 * adv_mom_5 & |
|---|
| 4607 | + ibit28 * adv_mom_3 & |
|---|
| 4608 | ) * & |
|---|
| 4609 | ( w(k,j,i+1) - w(k,j,i-2) ) & |
|---|
| 4610 | + ( ibit29 * adv_mom_5 & |
|---|
| 4611 | ) * & |
|---|
| 4612 | ( w(k,j,i+2) - w(k,j,i-3) ) & |
|---|
| 4613 | ) |
|---|
| 4614 | |
|---|
| 4615 | u_comp = u(k+1,j,i+1) + u(k,j,i+1) - gu |
|---|
| 4616 | flux_r = u_comp * ( & |
|---|
| 4617 | ( 37.0 * ibit29 * adv_mom_5 & |
|---|
| 4618 | + 7.0 * ibit28 * adv_mom_3 & |
|---|
| 4619 | + ibit27 * adv_mom_1 & |
|---|
| 4620 | ) * & |
|---|
| 4621 | ( w(k,j,i+1) + w(k,j,i) ) & |
|---|
| 4622 | - ( 8.0 * ibit29 * adv_mom_5 & |
|---|
| 4623 | + ibit28 * adv_mom_3 & |
|---|
| 4624 | ) * & |
|---|
| 4625 | ( w(k,j,i+2) + w(k,j,i-1) ) & |
|---|
| 4626 | + ( ibit29 * adv_mom_5 & |
|---|
| 4627 | ) * & |
|---|
| 4628 | ( w(k,j,i+3) + w(k,j,i-2) ) & |
|---|
| 4629 | ) |
|---|
| 4630 | |
|---|
| 4631 | diss_r = - ABS( u_comp ) * ( & |
|---|
| 4632 | ( 10.0 * ibit29 * adv_mom_5 & |
|---|
| 4633 | + 3.0 * ibit28 * adv_mom_3 & |
|---|
| 4634 | + ibit27 * adv_mom_1 & |
|---|
| 4635 | ) * & |
|---|
| 4636 | ( w(k,j,i+1) - w(k,j,i) ) & |
|---|
| 4637 | - ( 5.0 * ibit29 * adv_mom_5 & |
|---|
| 4638 | + ibit28 * adv_mom_3 & |
|---|
| 4639 | ) * & |
|---|
| 4640 | ( w(k,j,i+2) - w(k,j,i-1) ) & |
|---|
| 4641 | + ( ibit29 * adv_mom_5 & |
|---|
| 4642 | ) * & |
|---|
| 4643 | ( w(k,j,i+3) - w(k,j,i-2) ) & |
|---|
| 4644 | ) |
|---|
| 4645 | ibit32 = IBITS(wall_flags_00(k,j,i),0,1) |
|---|
| 4646 | ibit31 = IBITS(wall_flags_0(k,j,i),31,1) |
|---|
| 4647 | ibit30 = IBITS(wall_flags_0(k,j,i),30,1) |
|---|
| 4648 | |
|---|
| 4649 | v_comp_s = v(k+1,j,i) + v(k,j,i) - gv |
|---|
| 4650 | flux_s = v_comp_s * ( & |
|---|
| 4651 | ( 37.0 * ibit32 * adv_mom_5 & |
|---|
| 4652 | + 7.0 * ibit31 * adv_mom_3 & |
|---|
| 4653 | + ibit30 * adv_mom_1 & |
|---|
| 4654 | ) * & |
|---|
| 4655 | ( w(k,j,i) + w(k,j-1,i) ) & |
|---|
| 4656 | - ( 8.0 * ibit32 * adv_mom_5 & |
|---|
| 4657 | + ibit31 * adv_mom_3 & |
|---|
| 4658 | ) * & |
|---|
| 4659 | ( w(k,j+1,i) + w(k,j-2,i) ) & |
|---|
| 4660 | + ( ibit32 * adv_mom_5 & |
|---|
| 4661 | ) * & |
|---|
| 4662 | ( w(k,j+2,i) + w(k,j-3,i) ) & |
|---|
| 4663 | ) |
|---|
| 4664 | |
|---|
| 4665 | diss_s = - ABS( v_comp_s ) * ( & |
|---|
| 4666 | ( 10.0 * ibit32 * adv_mom_5 & |
|---|
| 4667 | + 3.0 * ibit31 * adv_mom_3 & |
|---|
| 4668 | + ibit30 * adv_mom_1 & |
|---|
| 4669 | ) * & |
|---|
| 4670 | ( w(k,j,i) - w(k,j-1,i) ) & |
|---|
| 4671 | - ( 5.0 * ibit32 * adv_mom_5 & |
|---|
| 4672 | + ibit31 * adv_mom_3 & |
|---|
| 4673 | ) * & |
|---|
| 4674 | ( w(k,j+1,i) - w(k,j-2,i) ) & |
|---|
| 4675 | + ( ibit32 * adv_mom_5 & |
|---|
| 4676 | ) * & |
|---|
| 4677 | ( w(k,j+2,i) - w(k,j-3,i) ) & |
|---|
| 4678 | ) |
|---|
| 4679 | |
|---|
| 4680 | v_comp = v(k+1,j+1,i) + v(k,j+1,i) - gv |
|---|
| 4681 | flux_n = v_comp * ( & |
|---|
| 4682 | ( 37.0 * ibit32 * adv_mom_5 & |
|---|
| 4683 | + 7.0 * ibit31 * adv_mom_3 & |
|---|
| 4684 | + ibit30 * adv_mom_1 & |
|---|
| 4685 | ) * & |
|---|
| 4686 | ( w(k,j+1,i) + w(k,j,i) ) & |
|---|
| 4687 | - ( 8.0 * ibit32 * adv_mom_5 & |
|---|
| 4688 | + ibit31 * adv_mom_3 & |
|---|
| 4689 | ) * & |
|---|
| 4690 | ( w(k,j+2,i) + w(k,j-1,i) ) & |
|---|
| 4691 | + ( ibit32 * adv_mom_5 & |
|---|
| 4692 | ) * & |
|---|
| 4693 | ( w(k,j+3,i) + w(k,j-2,i) ) & |
|---|
| 4694 | ) |
|---|
| 4695 | |
|---|
| 4696 | diss_n = - ABS( v_comp ) * ( & |
|---|
| 4697 | ( 10.0 * ibit32 * adv_mom_5 & |
|---|
| 4698 | + 3.0 * ibit31 * adv_mom_3 & |
|---|
| 4699 | + ibit30 * adv_mom_1 & |
|---|
| 4700 | ) * & |
|---|
| 4701 | ( w(k,j+1,i) - w(k,j,i) ) & |
|---|
| 4702 | - ( 5.0 * ibit32 * adv_mom_5 & |
|---|
| 4703 | + ibit31 * adv_mom_3 & |
|---|
| 4704 | ) * & |
|---|
| 4705 | ( w(k,j+2,i) - w(k,j-1,i) ) & |
|---|
| 4706 | + ( ibit32 * adv_mom_5 & |
|---|
| 4707 | ) * & |
|---|
| 4708 | ( w(k,j+3,i) - w(k,j-2,i) ) & |
|---|
| 4709 | ) |
|---|
| 4710 | |
|---|
| 4711 | ibit35 = IBITS(wall_flags_00(k-1,j,i),3,1) |
|---|
| 4712 | ibit34 = IBITS(wall_flags_00(k-1,j,i),2,1) |
|---|
| 4713 | ibit33 = IBITS(wall_flags_00(k-1,j,i),1,1) |
|---|
| 4714 | |
|---|
| 4715 | k_pp = k + 2 * ibit35 |
|---|
| 4716 | k_mm = k - 2 * ( ibit34 + ibit35 ) |
|---|
| 4717 | k_mmm = k - 3 * ibit35 |
|---|
| 4718 | |
|---|
| 4719 | w_comp = w(k,j,i) + w(k-1,j,i) |
|---|
| 4720 | flux_d = w_comp * ( & |
|---|
| 4721 | ( 37.0 * ibit35 * adv_mom_5 & |
|---|
| 4722 | + 7.0 * ibit34 * adv_mom_3 & |
|---|
| 4723 | + ibit33 * adv_mom_1 & |
|---|
| 4724 | ) * & |
|---|
| 4725 | ( w(k,j,i) + w(k-1,j,i) ) & |
|---|
| 4726 | - ( 8.0 * ibit35 * adv_mom_5 & |
|---|
| 4727 | + ibit34 * adv_mom_3 & |
|---|
| 4728 | ) * & |
|---|
| 4729 | ( w(k+1,j,i) + w(k_mm,j,i) ) & |
|---|
| 4730 | + ( ibit35 * adv_mom_5 & |
|---|
| 4731 | ) * & |
|---|
| 4732 | ( w(k_pp,j,i) + w(k_mmm,j,i) ) & |
|---|
| 4733 | ) |
|---|
| 4734 | |
|---|
| 4735 | diss_d = - ABS( w_comp ) * ( & |
|---|
| 4736 | ( 10.0 * ibit35 * adv_mom_5 & |
|---|
| 4737 | + 3.0 * ibit34 * adv_mom_3 & |
|---|
| 4738 | + ibit33 * adv_mom_1 & |
|---|
| 4739 | ) * & |
|---|
| 4740 | ( w(k,j,i) - w(k-1,j,i) ) & |
|---|
| 4741 | - ( 5.0 * ibit35 * adv_mom_5 & |
|---|
| 4742 | + ibit34 * adv_mom_3 & |
|---|
| 4743 | ) * & |
|---|
| 4744 | ( w(k+1,j,i) - w(k_mm,j,i) ) & |
|---|
| 4745 | + ( ibit35 * adv_mom_5 & |
|---|
| 4746 | ) * & |
|---|
| 4747 | ( w(k_pp,j,i) - w(k_mmm,j,i) ) & |
|---|
| 4748 | ) |
|---|
| 4749 | |
|---|
| 4750 | ! |
|---|
| 4751 | !-- k index has to be modified near bottom and top, else array |
|---|
| 4752 | !-- subscripts will be exceeded. |
|---|
| 4753 | ibit35 = IBITS(wall_flags_00(k,j,i),3,1) |
|---|
| 4754 | ibit34 = IBITS(wall_flags_00(k,j,i),2,1) |
|---|
| 4755 | ibit33 = IBITS(wall_flags_00(k,j,i),1,1) |
|---|
| 4756 | |
|---|
| 4757 | k_ppp = k + 3 * ibit35 |
|---|
| 4758 | k_pp = k + 2 * ( 1 - ibit33 ) |
|---|
| 4759 | k_mm = k - 2 * ibit35 |
|---|
| 4760 | |
|---|
| 4761 | w_comp = w(k+1,j,i) + w(k,j,i) |
|---|
| 4762 | flux_t = w_comp * ( & |
|---|
| 4763 | ( 37.0 * ibit35 * adv_mom_5 & |
|---|
| 4764 | + 7.0 * ibit34 * adv_mom_3 & |
|---|
| 4765 | + ibit33 * adv_mom_1 & |
|---|
| 4766 | ) * & |
|---|
| 4767 | ( w(k+1,j,i) + w(k,j,i) ) & |
|---|
| 4768 | - ( 8.0 * ibit35 * adv_mom_5 & |
|---|
| 4769 | + ibit34 * adv_mom_3 & |
|---|
| 4770 | ) * & |
|---|
| 4771 | ( w(k_pp,j,i) + w(k-1,j,i) ) & |
|---|
| 4772 | + ( ibit35 * adv_mom_5 & |
|---|
| 4773 | ) * & |
|---|
| 4774 | ( w(k_ppp,j,i) + w(k_mm,j,i) ) & |
|---|
| 4775 | ) |
|---|
| 4776 | |
|---|
| 4777 | diss_t = - ABS( w_comp ) * ( & |
|---|
| 4778 | ( 10.0 * ibit35 * adv_mom_5 & |
|---|
| 4779 | + 3.0 * ibit34 * adv_mom_3 & |
|---|
| 4780 | + ibit33 * adv_mom_1 & |
|---|
| 4781 | ) * & |
|---|
| 4782 | ( w(k+1,j,i) - w(k,j,i) ) & |
|---|
| 4783 | - ( 5.0 * ibit35 * adv_mom_5 & |
|---|
| 4784 | + ibit34 * adv_mom_3 & |
|---|
| 4785 | ) * & |
|---|
| 4786 | ( w(k_pp,j,i) - w(k-1,j,i) ) & |
|---|
| 4787 | + ( ibit35 * adv_mom_5 & |
|---|
| 4788 | ) * & |
|---|
| 4789 | ( w(k_ppp,j,i) - w(k_mm,j,i) ) & |
|---|
| 4790 | ) |
|---|
| 4791 | ! |
|---|
| 4792 | !-- Calculate the divergence of the velocity field. A respective |
|---|
| 4793 | !-- correction is needed to overcome numerical instabilities caused |
|---|
| 4794 | !-- by a not sufficient reduction of divergences near topography. |
|---|
| 4795 | div = ( ( u_comp + gu - ( u(k+1,j,i) + u(k,j,i) ) ) * ddx & |
|---|
| 4796 | + ( v_comp + gv - ( v(k+1,j,i) + v(k,j,i) ) ) * ddy & |
|---|
| 4797 | + ( w_comp - ( w(k,j,i) + w(k-1,j,i) ) ) & |
|---|
| 4798 | * ddzu(k+1) & |
|---|
| 4799 | ) * 0.5 |
|---|
| 4800 | |
|---|
| 4801 | tend(k,j,i) = - ( & |
|---|
| 4802 | ( flux_r + diss_r - flux_l - diss_l ) * ddx & |
|---|
| 4803 | + ( flux_n + diss_n - flux_s - diss_s ) * ddy & |
|---|
| 4804 | + ( flux_t + diss_t - flux_d - diss_d ) * ddzu(k+1) & |
|---|
| 4805 | ) + div * w(k,j,i) |
|---|
| 4806 | |
|---|
| 4807 | |
|---|
| 4808 | !++ |
|---|
| 4809 | !-- Statistical Evaluation of w'w'. |
|---|
| 4810 | ! sums_ws2_ws_l(k,tn) = sums_ws2_ws_l(k,tn) & |
|---|
| 4811 | ! + ( flux_t + diss_t ) & |
|---|
| 4812 | ! * weight_substep(intermediate_timestep_count) |
|---|
| 4813 | |
|---|
| 4814 | ENDDO |
|---|
| 4815 | ENDDO |
|---|
| 4816 | ENDDO |
|---|
| 4817 | !$acc end kernels |
|---|
| 4818 | |
|---|
| 4819 | END SUBROUTINE advec_w_ws_acc |
|---|
| 4820 | |
|---|
| 4821 | END MODULE advec_ws |
|---|
