1 | !> @file lagrangian_particle_model_mod.f90 |
---|
2 | !--------------------------------------------------------------------------------------------------! |
---|
3 | ! This file is part of the PALM model system. |
---|
4 | ! |
---|
5 | ! PALM is free software: you can redistribute it and/or modify it under the terms of the GNU General |
---|
6 | ! Public License as published by the Free Software Foundation, either version 3 of the License, or |
---|
7 | ! (at your option) any later version. |
---|
8 | ! |
---|
9 | ! PALM is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the |
---|
10 | ! implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General |
---|
11 | ! Public License for more details. |
---|
12 | ! |
---|
13 | ! You should have received a copy of the GNU General Public License along with PALM. If not, see |
---|
14 | ! <http://www.gnu.org/licenses/>. |
---|
15 | ! |
---|
16 | ! Copyright 1997-2020 Leibniz Universitaet Hannover |
---|
17 | !--------------------------------------------------------------------------------------------------! |
---|
18 | ! |
---|
19 | ! Current revisions: |
---|
20 | ! ------------------ |
---|
21 | ! |
---|
22 | ! |
---|
23 | ! Former revisions: |
---|
24 | ! ----------------- |
---|
25 | ! $Id: lagrangian_particle_model_mod.f90 4731 2020-10-07 13:25:11Z pavelkrc $ |
---|
26 | ! Move exchange_horiz from time_integration to modules |
---|
27 | ! |
---|
28 | ! 4673 2020-09-10 07:56:36Z schwenkel |
---|
29 | ! bugfix in case of mpi-restarts |
---|
30 | ! |
---|
31 | ! 4671 2020-09-09 20:27:58Z pavelkrc |
---|
32 | ! Implementation of downward facing USM and LSM surfaces |
---|
33 | ! |
---|
34 | ! 4648 2020-08-25 07:52:08Z raasch |
---|
35 | ! file re-formatted to follow the PALM coding standard |
---|
36 | ! |
---|
37 | ! 4629 2020-07-29 09:37:56Z raasch |
---|
38 | ! support for MPI Fortran77 interface (mpif.h) removed |
---|
39 | ! |
---|
40 | ! 4628 2020-07-29 07:23:03Z raasch |
---|
41 | ! extensions required for MPI-I/O of particle data to restart files |
---|
42 | ! |
---|
43 | ! 4616 2020-07-21 10:09:46Z schwenkel |
---|
44 | ! Bugfix in case of strechting: k-calculation limited lower bound of 1 |
---|
45 | ! |
---|
46 | ! 4589 2020-07-06 12:34:09Z suehring |
---|
47 | ! remove unused variables |
---|
48 | ! |
---|
49 | ! 4588 2020-07-06 11:06:02Z suehring |
---|
50 | ! Simplify particle-speed interpolation in logarithmic layer |
---|
51 | ! |
---|
52 | ! 4585 2020-06-30 15:05:20Z suehring |
---|
53 | ! Limit logarithmically interpolated particle speed to the velocity component at the first |
---|
54 | ! prognostic grid point (since no stability corrected interpolation is employed the particle speed |
---|
55 | ! could be overestimated in unstable conditions). |
---|
56 | ! |
---|
57 | ! 4546 2020-05-24 12:16:41Z raasch |
---|
58 | ! Variables iran and iran_part completely removed, added I/O of parallel random numbers to restart |
---|
59 | ! file |
---|
60 | ! |
---|
61 | ! 4545 2020-05-22 13:17:57Z schwenkel |
---|
62 | ! Using parallel random generator, thus preventing dependency of PE number |
---|
63 | ! |
---|
64 | ! 4535 2020-05-15 12:07:23Z raasch |
---|
65 | ! bugfix for restart data format query |
---|
66 | ! |
---|
67 | ! 4520 2020-05-06 08:57:19Z schwenkel |
---|
68 | ! Add error number |
---|
69 | ! |
---|
70 | ! 4517 2020-05-03 14:29:30Z raasch |
---|
71 | ! restart data handling with MPI-IO added |
---|
72 | ! |
---|
73 | ! 4471 2020-03-24 12:08:06Z schwenkel |
---|
74 | ! Bugfix in lpm_droplet_interactions_ptq |
---|
75 | ! |
---|
76 | ! 4457 2020-03-11 14:20:43Z raasch |
---|
77 | ! use statement for exchange horiz added |
---|
78 | ! |
---|
79 | ! 4444 2020-03-05 15:59:50Z raasch |
---|
80 | ! bugfix: cpp-directives for serial mode added |
---|
81 | ! |
---|
82 | ! 4430 2020-02-27 18:02:20Z suehring |
---|
83 | ! - Bugfix in logarithmic interpolation of near-ground particle speed (density was not considered). |
---|
84 | ! - Revise CFL-check when SGS particle speeds are considered. |
---|
85 | ! - In nested case with SGS particle speeds in the child domain, do not give warning that particles |
---|
86 | ! are on domain boundaries. At the end of the particle time integration these will be transferred |
---|
87 | ! to the parent domain anyhow. |
---|
88 | ! |
---|
89 | ! 4360 2020-01-07 11:25:50Z suehring |
---|
90 | ! Introduction of wall_flags_total_0, which currently sets bits based on static topography |
---|
91 | ! information used in wall_flags_static_0 |
---|
92 | ! |
---|
93 | ! 4336 2019-12-13 10:12:05Z raasch |
---|
94 | ! bugfix: wrong header output of particle group features (density ratio) in case of restarts |
---|
95 | ! corrected |
---|
96 | ! |
---|
97 | ! 4329 2019-12-10 15:46:36Z motisi |
---|
98 | ! Renamed wall_flags_0 to wall_flags_static_0 |
---|
99 | ! |
---|
100 | ! 4282 2019-10-29 16:18:46Z schwenkel |
---|
101 | ! Bugfix of particle timeseries in case of more than one particle group |
---|
102 | ! |
---|
103 | ! 4277 2019-10-28 16:53:23Z schwenkel |
---|
104 | ! Bugfix: Added first_call_lpm in use statement |
---|
105 | ! |
---|
106 | ! 4276 2019-10-28 16:03:29Z schwenkel |
---|
107 | ! Modularize lpm: Move conditions in time intergration to module |
---|
108 | ! |
---|
109 | ! 4275 2019-10-28 15:34:55Z schwenkel |
---|
110 | ! Change call of simple predictor corrector method, i.e. two divergence free velocitiy fields are |
---|
111 | ! now used. |
---|
112 | ! |
---|
113 | ! 4232 2019-09-20 09:34:22Z knoop |
---|
114 | ! Removed INCLUDE "mpif.h", as it is not needed because of USE pegrid |
---|
115 | ! |
---|
116 | ! 4195 2019-08-28 13:44:27Z schwenkel |
---|
117 | ! Bugfix for simple_corrector interpolation method in case of ocean runs and output particle |
---|
118 | ! advection interpolation method into header |
---|
119 | ! |
---|
120 | ! 4182 2019-08-22 15:20:23Z scharf |
---|
121 | ! Corrected "Former revisions" section |
---|
122 | ! |
---|
123 | ! 4168 2019-08-16 13:50:17Z suehring |
---|
124 | ! Replace function get_topography_top_index by topo_top_ind |
---|
125 | ! |
---|
126 | ! 4145 2019-08-06 09:55:22Z schwenkel |
---|
127 | ! Some reformatting |
---|
128 | ! |
---|
129 | ! 4144 2019-08-06 09:11:47Z raasch |
---|
130 | ! relational operators .EQ., .NE., etc. replaced by ==, /=, etc. |
---|
131 | ! |
---|
132 | ! 4143 2019-08-05 15:14:53Z schwenkel |
---|
133 | ! Rename variable and change select case to if statement |
---|
134 | ! |
---|
135 | ! 4122 2019-07-26 13:11:56Z schwenkel |
---|
136 | ! Implement reset method as bottom boundary condition |
---|
137 | ! |
---|
138 | ! 4121 2019-07-26 10:01:22Z schwenkel |
---|
139 | ! Implementation of an simple method for interpolating the velocities to particle position |
---|
140 | ! |
---|
141 | ! 4114 2019-07-23 14:09:27Z schwenkel |
---|
142 | ! Bugfix: Added working precision for if statement |
---|
143 | ! |
---|
144 | ! 4054 2019-06-27 07:42:18Z raasch |
---|
145 | ! bugfix for calculating the minimum particle time step |
---|
146 | ! |
---|
147 | ! 4044 2019-06-19 12:28:27Z schwenkel |
---|
148 | ! Bugfix in case of grid strecting: corrected calculation of k-Index |
---|
149 | ! |
---|
150 | ! 4043 2019-06-18 16:59:00Z schwenkel |
---|
151 | ! Remove min_nr_particle, Add lpm_droplet_interactions_ptq into module |
---|
152 | ! |
---|
153 | ! 4028 2019-06-13 12:21:37Z schwenkel |
---|
154 | ! Further modularization of particle code components |
---|
155 | ! |
---|
156 | ! 4020 2019-06-06 14:57:48Z schwenkel |
---|
157 | ! Removing submodules |
---|
158 | ! |
---|
159 | ! 4018 2019-06-06 13:41:50Z eckhard |
---|
160 | ! Bugfix for former revision |
---|
161 | ! |
---|
162 | ! 4017 2019-06-06 12:16:46Z schwenkel |
---|
163 | ! Modularization of all lagrangian particle model code components |
---|
164 | ! |
---|
165 | ! 3655 2019-01-07 16:51:22Z knoop |
---|
166 | ! bugfix to guarantee correct particle releases in case that the release interval is smaller than |
---|
167 | ! the model timestep |
---|
168 | ! |
---|
169 | ! Revision 1.1 1999/11/25 16:16:06 raasch |
---|
170 | ! Initial revision |
---|
171 | ! |
---|
172 | ! |
---|
173 | ! Description: |
---|
174 | ! ------------ |
---|
175 | !> The embedded LPM allows for studying transport and dispersion processes within turbulent flows. |
---|
176 | !> This model is including passive particles that do not show any feedback on the turbulent flow. |
---|
177 | !> Further also particles with inertia and cloud droplets can be simulated explicitly. |
---|
178 | !> |
---|
179 | !> @todo test lcm |
---|
180 | !> implement simple interpolation method for subgrid scale velocites |
---|
181 | !> @note <Enter notes on the module> |
---|
182 | !> @bug <Enter bug on the module> |
---|
183 | !--------------------------------------------------------------------------------------------------! |
---|
184 | MODULE lagrangian_particle_model_mod |
---|
185 | |
---|
186 | USE, INTRINSIC :: ISO_C_BINDING |
---|
187 | |
---|
188 | USE arrays_3d, & |
---|
189 | ONLY: d_exner, de_dx, de_dy, de_dz, diss, dzw, e, exner, hyp, km, pt, q, ql, ql_1, ql_2, & |
---|
190 | ql_c, ql_v, ql_vp, u, v, w, zu, zw |
---|
191 | |
---|
192 | USE averaging, & |
---|
193 | ONLY: pc_av, pr_av, ql_c_av, ql_v_av, ql_vp_av |
---|
194 | |
---|
195 | USE basic_constants_and_equations_mod, & |
---|
196 | ONLY: g, kappa, l_v, lv_d_cp, magnus, molecular_weight_of_solute, & |
---|
197 | molecular_weight_of_water, pi, r_v, rd_d_rv, rho_l, rho_s, vanthoff |
---|
198 | |
---|
199 | USE control_parameters, & |
---|
200 | ONLY: bc_dirichlet_l, bc_dirichlet_n, bc_dirichlet_r, bc_dirichlet_s, & |
---|
201 | child_domain, cloud_droplets, constant_flux_layer, current_timestep_number, & |
---|
202 | debug_output, dopts_time_count, dt_3d, dt_3d_reached, dt_3d_reached_l, dt_dopts, dz,& |
---|
203 | dz_stretch_level, dz_stretch_level_start, first_call_lpm, humidity, & |
---|
204 | initializing_actions, intermediate_timestep_count, intermediate_timestep_count_max, & |
---|
205 | message_string, molecular_viscosity, ocean_mode, particle_maximum_age, & |
---|
206 | restart_data_format_input, restart_data_format_output, rho_surface, simulated_time, & |
---|
207 | time_since_reference_point, topography, u_gtrans, v_gtrans |
---|
208 | |
---|
209 | USE cpulog, & |
---|
210 | ONLY: cpu_log, log_point, log_point_s |
---|
211 | |
---|
212 | USE indices, & |
---|
213 | ONLY: nbgp, ngp_2dh_outer, nx, nxl, nxlg, nxrg, nxr, ny, nyn, nys, nyng, nysg, nz, nzb, & |
---|
214 | nzb_max, nzt, topo_top_ind, wall_flags_total_0 |
---|
215 | |
---|
216 | USE kinds |
---|
217 | |
---|
218 | USE pegrid |
---|
219 | |
---|
220 | USE particle_attributes |
---|
221 | |
---|
222 | #if defined( __parallel ) |
---|
223 | USE pmc_particle_interface, & |
---|
224 | ONLY: pmcp_c_get_particle_from_parent, pmcp_c_send_particle_to_parent, pmcp_g_init, & |
---|
225 | pmcp_g_print_number_of_particles, pmcp_p_delete_particles_in_fine_grid_area, & |
---|
226 | pmcp_p_empty_particle_win, pmcp_p_fill_particle_win |
---|
227 | #endif |
---|
228 | |
---|
229 | USE pmc_interface, & |
---|
230 | ONLY: nested_run |
---|
231 | |
---|
232 | USE grid_variables, & |
---|
233 | ONLY: ddx, dx, ddy, dy |
---|
234 | |
---|
235 | USE netcdf_interface, & |
---|
236 | ONLY: dopts_num, id_set_pts, id_var_dopts, id_var_time_pts, nc_stat, netcdf_data_format, & |
---|
237 | netcdf_deflate, netcdf_handle_error |
---|
238 | |
---|
239 | USE random_generator_parallel, & |
---|
240 | ONLY: init_parallel_random_generator, & |
---|
241 | id_random_array, & |
---|
242 | random_dummy, & |
---|
243 | random_number_parallel, & |
---|
244 | random_number_parallel_gauss, & |
---|
245 | random_seed_parallel |
---|
246 | |
---|
247 | USE restart_data_mpi_io_mod, & |
---|
248 | ONLY: rd_mpi_io_check_array, & |
---|
249 | rd_mpi_io_check_open, & |
---|
250 | rd_mpi_io_close, & |
---|
251 | rd_mpi_io_open, & |
---|
252 | rd_mpi_io_particle_filetypes, & |
---|
253 | rrd_mpi_io, & |
---|
254 | rrd_mpi_io_global_array, & |
---|
255 | rrd_mpi_io_particles, & |
---|
256 | wrd_mpi_io, & |
---|
257 | wrd_mpi_io_global_array, & |
---|
258 | wrd_mpi_io_particles |
---|
259 | |
---|
260 | USE statistics, & |
---|
261 | ONLY: hom |
---|
262 | |
---|
263 | USE surface_mod, & |
---|
264 | ONLY: bc_h, & |
---|
265 | surf_def_h, & |
---|
266 | surf_lsm_h, & |
---|
267 | surf_usm_h |
---|
268 | |
---|
269 | !-- Next lines are in preparation for the output of particle data |
---|
270 | ! USE data_output_particle_mod, & |
---|
271 | ! ONLY: dop_active, dop_init, dop_output_tseries |
---|
272 | |
---|
273 | #if defined( __parallel ) |
---|
274 | USE MPI |
---|
275 | #endif |
---|
276 | |
---|
277 | #if defined( __netcdf ) |
---|
278 | USE NETCDF |
---|
279 | #endif |
---|
280 | |
---|
281 | IMPLICIT NONE |
---|
282 | |
---|
283 | INTEGER(iwp), PARAMETER :: nr_2_direction_move = 10000 !< |
---|
284 | INTEGER(iwp), PARAMETER :: phase_init = 1 !< |
---|
285 | INTEGER(iwp), PARAMETER, PUBLIC :: phase_release = 2 !< |
---|
286 | |
---|
287 | REAL(wp), PARAMETER :: c_0 = 3.0_wp !< parameter for lagrangian timescale |
---|
288 | |
---|
289 | CHARACTER(LEN=15) :: aero_species = 'nacl' !< aerosol species |
---|
290 | CHARACTER(LEN=15) :: aero_type = 'maritime' !< aerosol type |
---|
291 | CHARACTER(LEN=15) :: bc_par_b = 'reflect' !< bottom boundary condition |
---|
292 | CHARACTER(LEN=15) :: bc_par_lr = 'cyclic' !< left/right boundary condition |
---|
293 | CHARACTER(LEN=15) :: bc_par_ns = 'cyclic' !< north/south boundary condition |
---|
294 | CHARACTER(LEN=15) :: bc_par_t = 'absorb' !< top boundary condition |
---|
295 | CHARACTER(LEN=15) :: collision_kernel = 'none' !< collision kernel |
---|
296 | |
---|
297 | CHARACTER(LEN=5) :: splitting_function = 'gamma' !< function for calculation critical weighting factor |
---|
298 | CHARACTER(LEN=5) :: splitting_mode = 'const' !< splitting mode |
---|
299 | |
---|
300 | CHARACTER(LEN=25) :: particle_advection_interpolation = 'trilinear' !< interpolation method for calculatin the particle |
---|
301 | |
---|
302 | INTEGER(iwp) :: deleted_particles = 0 !< number of deleted particles per time step |
---|
303 | INTEGER(iwp) :: i_splitting_mode !< dummy for splitting mode |
---|
304 | INTEGER(iwp) :: isf !< dummy for splitting function |
---|
305 | INTEGER(iwp) :: max_number_particles_per_gridbox = 100 !< namelist parameter (see documentation) |
---|
306 | INTEGER(iwp) :: number_particles_per_gridbox = -1 !< namelist parameter (see documentation) |
---|
307 | INTEGER(iwp) :: number_of_sublayers = 20 !< number of sublayers for particle velocities betwenn surface |
---|
308 | !< and first grid level |
---|
309 | INTEGER(iwp) :: offset_ocean_nzt = 0 !< in case of oceans runs, the vertical index calculations need |
---|
310 | !< an offset |
---|
311 | INTEGER(iwp) :: offset_ocean_nzt_m1 = 0 !< in case of oceans runs, the vertical index calculations need |
---|
312 | !< an offset |
---|
313 | INTEGER(iwp) :: particles_per_point = 1 !< namelist parameter (see documentation) |
---|
314 | INTEGER(iwp) :: radius_classes = 20 !< namelist parameter (see documentation) |
---|
315 | INTEGER(iwp) :: splitting_factor = 2 !< namelist parameter (see documentation) |
---|
316 | INTEGER(iwp) :: splitting_factor_max = 5 !< namelist parameter (see documentation) |
---|
317 | INTEGER(iwp) :: step_dealloc = 100 !< namelist parameter (see documentation) |
---|
318 | INTEGER(iwp) :: total_number_of_particles !< total number of particles in the whole model domain |
---|
319 | INTEGER(iwp) :: trlp_count_recv_sum !< parameter for particle exchange of PEs |
---|
320 | INTEGER(iwp) :: trlp_count_sum !< parameter for particle exchange of PEs |
---|
321 | INTEGER(iwp) :: trrp_count_recv_sum !< parameter for particle exchange of PEs |
---|
322 | INTEGER(iwp) :: trrp_count_sum !< parameter for particle exchange of PEs |
---|
323 | INTEGER(iwp) :: trsp_count_recv_sum !< parameter for particle exchange of PEs |
---|
324 | INTEGER(iwp) :: trsp_count_sum !< parameter for particle exchange of PEs |
---|
325 | INTEGER(iwp) :: trnp_count_recv_sum !< parameter for particle exchange of PEs |
---|
326 | INTEGER(iwp) :: trnp_count_sum !< parameter for particle exchange of PEs |
---|
327 | |
---|
328 | INTEGER(isp), DIMENSION(:,:,:), ALLOCATABLE :: seq_random_array_particles !< sequence of random array for particle |
---|
329 | |
---|
330 | LOGICAL :: curvature_solution_effects = .FALSE. !< namelist parameter (see documentation) |
---|
331 | LOGICAL :: deallocate_memory = .TRUE. !< namelist parameter (see documentation) |
---|
332 | LOGICAL :: hall_kernel = .FALSE. !< flag for collision kernel |
---|
333 | LOGICAL :: interpolation_simple_corrector = .FALSE. !< flag for simple particle advection interpolation with corrector step |
---|
334 | LOGICAL :: interpolation_simple_predictor = .FALSE. !< flag for simple particle advection interpolation with predictor step |
---|
335 | LOGICAL :: interpolation_trilinear = .FALSE. !< flag for trilinear particle advection interpolation |
---|
336 | LOGICAL :: lagrangian_particle_model = .FALSE. !< namelist parameter (see documentation) |
---|
337 | LOGICAL :: merging = .FALSE. !< namelist parameter (see documentation) |
---|
338 | LOGICAL :: random_start_position = .FALSE. !< namelist parameter (see documentation) |
---|
339 | LOGICAL :: read_particles_from_restartfile = .TRUE. !< namelist parameter (see documentation) |
---|
340 | LOGICAL :: seed_follows_topography = .FALSE. !< namelist parameter (see documentation) |
---|
341 | LOGICAL :: splitting = .FALSE. !< namelist parameter (see documentation) |
---|
342 | LOGICAL :: use_kernel_tables = .FALSE. !< parameter, which turns on the use of precalculated collision kernels |
---|
343 | LOGICAL :: write_particle_statistics = .FALSE. !< namelist parameter (see documentation) |
---|
344 | |
---|
345 | LOGICAL, DIMENSION(max_number_of_particle_groups) :: vertical_particle_advection = .TRUE. !< Switch for vertical particle |
---|
346 | !< transport |
---|
347 | |
---|
348 | REAL(wp) :: aero_weight = 1.0_wp !< namelist parameter (see documentation) |
---|
349 | REAL(wp) :: dt_min_part = 0.0002_wp !< minimum particle time step when SGS velocities are used (s) |
---|
350 | REAL(wp) :: dt_prel = 9999999.9_wp !< namelist parameter (see documentation) |
---|
351 | REAL(wp) :: dt_write_particle_data = 9999999.9_wp !< namelist parameter (see documentation) |
---|
352 | REAL(wp) :: epsilon_collision !< |
---|
353 | REAL(wp) :: end_time_prel = 9999999.9_wp !< namelist parameter (see documentation) |
---|
354 | REAL(wp) :: initial_weighting_factor = 1.0_wp !< namelist parameter (see documentation) |
---|
355 | REAL(wp) :: last_particle_release_time = 0.0_wp !< last time of particle release |
---|
356 | REAL(wp) :: log_sigma(3) = 1.0_wp !< namelist parameter (see documentation) |
---|
357 | REAL(wp) :: na(3) = 0.0_wp !< namelist parameter (see documentation) |
---|
358 | REAL(wp) :: number_concentration = -1.0_wp !< namelist parameter (see documentation) |
---|
359 | REAL(wp) :: radius_merge = 1.0E-7_wp !< namelist parameter (see documentation) |
---|
360 | REAL(wp) :: radius_split = 40.0E-6_wp !< namelist parameter (see documentation) |
---|
361 | REAL(wp) :: rclass_lbound !< |
---|
362 | REAL(wp) :: rclass_ubound !< |
---|
363 | REAL(wp) :: rm(3) = 1.0E-6_wp !< namelist parameter (see documentation) |
---|
364 | REAL(wp) :: sgs_wf_part !< parameter for sgs |
---|
365 | REAL(wp) :: time_write_particle_data = 0.0_wp !< write particle data at current time on file |
---|
366 | REAL(wp) :: urms !< |
---|
367 | REAL(wp) :: weight_factor_merge = -1.0_wp !< namelist parameter (see documentation) |
---|
368 | REAL(wp) :: weight_factor_split = -1.0_wp !< namelist parameter (see documentation) |
---|
369 | REAL(wp) :: z0_av_global !< horizontal mean value of z0 |
---|
370 | |
---|
371 | REAL(wp), DIMENSION(max_number_of_particle_groups) :: density_ratio = 9999999.9_wp !< namelist parameter (see documentation) |
---|
372 | REAL(wp), DIMENSION(max_number_of_particle_groups) :: pdx = 9999999.9_wp !< namelist parameter (see documentation) |
---|
373 | REAL(wp), DIMENSION(max_number_of_particle_groups) :: pdy = 9999999.9_wp !< namelist parameter (see documentation) |
---|
374 | REAL(wp), DIMENSION(max_number_of_particle_groups) :: pdz = 9999999.9_wp !< namelist parameter (see documentation) |
---|
375 | REAL(wp), DIMENSION(max_number_of_particle_groups) :: psb = 9999999.9_wp !< namelist parameter (see documentation) |
---|
376 | REAL(wp), DIMENSION(max_number_of_particle_groups) :: psl = 9999999.9_wp !< namelist parameter (see documentation) |
---|
377 | REAL(wp), DIMENSION(max_number_of_particle_groups) :: psn = 9999999.9_wp !< namelist parameter (see documentation) |
---|
378 | REAL(wp), DIMENSION(max_number_of_particle_groups) :: psr = 9999999.9_wp !< namelist parameter (see documentation) |
---|
379 | REAL(wp), DIMENSION(max_number_of_particle_groups) :: pss = 9999999.9_wp !< namelist parameter (see documentation) |
---|
380 | REAL(wp), DIMENSION(max_number_of_particle_groups) :: pst = 9999999.9_wp !< namelist parameter (see documentation). |
---|
381 | REAL(wp), DIMENSION(max_number_of_particle_groups) :: radius = 9999999.9_wp !< namelist parameter (see documentation) |
---|
382 | |
---|
383 | REAL(wp), DIMENSION(:), ALLOCATABLE :: log_z_z0 !< Precalculate LOG(z/z0) |
---|
384 | |
---|
385 | #if defined( __parallel ) |
---|
386 | INTEGER(iwp) :: nr_move_north !< |
---|
387 | INTEGER(iwp) :: nr_move_south !< |
---|
388 | |
---|
389 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: move_also_north |
---|
390 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: move_also_south |
---|
391 | #endif |
---|
392 | |
---|
393 | REAL(wp), DIMENSION(:), ALLOCATABLE :: epsclass !< dissipation rate class |
---|
394 | REAL(wp), DIMENSION(:), ALLOCATABLE :: radclass !< radius class |
---|
395 | REAL(wp), DIMENSION(:), ALLOCATABLE :: winf !< |
---|
396 | |
---|
397 | REAL(wp), DIMENSION(:,:), ALLOCATABLE :: ec !< |
---|
398 | REAL(wp), DIMENSION(:,:), ALLOCATABLE :: ecf !< |
---|
399 | REAL(wp), DIMENSION(:,:), ALLOCATABLE :: gck !< |
---|
400 | REAL(wp), DIMENSION(:,:), ALLOCATABLE :: hkernel !< |
---|
401 | REAL(wp), DIMENSION(:,:), ALLOCATABLE :: hwratio !< |
---|
402 | |
---|
403 | REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: ckernel !< |
---|
404 | REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: u_t !< u value of old timelevel t |
---|
405 | REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: v_t !< v value of old timelevel t |
---|
406 | REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: w_t !< w value of old timelevel t |
---|
407 | |
---|
408 | SAVE |
---|
409 | |
---|
410 | PRIVATE |
---|
411 | |
---|
412 | PUBLIC lpm_parin, & |
---|
413 | lpm_header, & |
---|
414 | lpm_init_arrays, & |
---|
415 | lpm_init, & |
---|
416 | lpm_actions, & |
---|
417 | lpm_data_output_ptseries, & |
---|
418 | lpm_exchange_horiz_bounds, & |
---|
419 | lpm_interaction_droplets_ptq, & |
---|
420 | lpm_rrd_local_particles, & |
---|
421 | lpm_wrd_local, & |
---|
422 | lpm_rrd_global, & |
---|
423 | lpm_wrd_global, & |
---|
424 | lpm_rrd_local, & |
---|
425 | lpm_check_parameters |
---|
426 | |
---|
427 | PUBLIC lagrangian_particle_model |
---|
428 | |
---|
429 | INTERFACE lpm_check_parameters |
---|
430 | MODULE PROCEDURE lpm_check_parameters |
---|
431 | END INTERFACE lpm_check_parameters |
---|
432 | |
---|
433 | INTERFACE lpm_parin |
---|
434 | MODULE PROCEDURE lpm_parin |
---|
435 | END INTERFACE lpm_parin |
---|
436 | |
---|
437 | INTERFACE lpm_header |
---|
438 | MODULE PROCEDURE lpm_header |
---|
439 | END INTERFACE lpm_header |
---|
440 | |
---|
441 | INTERFACE lpm_init_arrays |
---|
442 | MODULE PROCEDURE lpm_init_arrays |
---|
443 | END INTERFACE lpm_init_arrays |
---|
444 | |
---|
445 | INTERFACE lpm_init |
---|
446 | MODULE PROCEDURE lpm_init |
---|
447 | END INTERFACE lpm_init |
---|
448 | |
---|
449 | INTERFACE lpm_actions |
---|
450 | MODULE PROCEDURE lpm_actions |
---|
451 | END INTERFACE lpm_actions |
---|
452 | |
---|
453 | INTERFACE lpm_data_output_ptseries |
---|
454 | MODULE PROCEDURE lpm_data_output_ptseries |
---|
455 | END INTERFACE |
---|
456 | |
---|
457 | INTERFACE lpm_rrd_local_particles |
---|
458 | MODULE PROCEDURE lpm_rrd_local_particles |
---|
459 | END INTERFACE lpm_rrd_local_particles |
---|
460 | |
---|
461 | INTERFACE lpm_rrd_global |
---|
462 | MODULE PROCEDURE lpm_rrd_global_ftn |
---|
463 | MODULE PROCEDURE lpm_rrd_global_mpi |
---|
464 | END INTERFACE lpm_rrd_global |
---|
465 | |
---|
466 | INTERFACE lpm_rrd_local |
---|
467 | MODULE PROCEDURE lpm_rrd_local_ftn |
---|
468 | MODULE PROCEDURE lpm_rrd_local_mpi |
---|
469 | END INTERFACE lpm_rrd_local |
---|
470 | |
---|
471 | INTERFACE lpm_wrd_local |
---|
472 | MODULE PROCEDURE lpm_wrd_local |
---|
473 | END INTERFACE lpm_wrd_local |
---|
474 | |
---|
475 | INTERFACE lpm_wrd_global |
---|
476 | MODULE PROCEDURE lpm_wrd_global |
---|
477 | END INTERFACE lpm_wrd_global |
---|
478 | |
---|
479 | INTERFACE lpm_advec |
---|
480 | MODULE PROCEDURE lpm_advec |
---|
481 | END INTERFACE lpm_advec |
---|
482 | |
---|
483 | INTERFACE lpm_calc_liquid_water_content |
---|
484 | MODULE PROCEDURE lpm_calc_liquid_water_content |
---|
485 | END INTERFACE |
---|
486 | |
---|
487 | INTERFACE lpm_interaction_droplets_ptq |
---|
488 | MODULE PROCEDURE lpm_interaction_droplets_ptq |
---|
489 | MODULE PROCEDURE lpm_interaction_droplets_ptq_ij |
---|
490 | END INTERFACE lpm_interaction_droplets_ptq |
---|
491 | |
---|
492 | INTERFACE lpm_boundary_conds |
---|
493 | MODULE PROCEDURE lpm_boundary_conds |
---|
494 | END INTERFACE lpm_boundary_conds |
---|
495 | |
---|
496 | INTERFACE lpm_droplet_condensation |
---|
497 | MODULE PROCEDURE lpm_droplet_condensation |
---|
498 | END INTERFACE |
---|
499 | |
---|
500 | INTERFACE lpm_droplet_collision |
---|
501 | MODULE PROCEDURE lpm_droplet_collision |
---|
502 | END INTERFACE lpm_droplet_collision |
---|
503 | |
---|
504 | INTERFACE lpm_init_kernels |
---|
505 | MODULE PROCEDURE lpm_init_kernels |
---|
506 | END INTERFACE lpm_init_kernels |
---|
507 | |
---|
508 | INTERFACE lpm_splitting |
---|
509 | MODULE PROCEDURE lpm_splitting |
---|
510 | END INTERFACE lpm_splitting |
---|
511 | |
---|
512 | INTERFACE lpm_merging |
---|
513 | MODULE PROCEDURE lpm_merging |
---|
514 | END INTERFACE lpm_merging |
---|
515 | |
---|
516 | INTERFACE lpm_exchange_horiz |
---|
517 | MODULE PROCEDURE lpm_exchange_horiz |
---|
518 | END INTERFACE lpm_exchange_horiz |
---|
519 | |
---|
520 | INTERFACE lpm_exchange_horiz_bounds |
---|
521 | MODULE PROCEDURE lpm_exchange_horiz_bounds |
---|
522 | END INTERFACE lpm_exchange_horiz_bounds |
---|
523 | |
---|
524 | INTERFACE lpm_move_particle |
---|
525 | MODULE PROCEDURE lpm_move_particle |
---|
526 | END INTERFACE lpm_move_particle |
---|
527 | |
---|
528 | INTERFACE realloc_particles_array |
---|
529 | MODULE PROCEDURE realloc_particles_array |
---|
530 | END INTERFACE realloc_particles_array |
---|
531 | |
---|
532 | INTERFACE dealloc_particles_array |
---|
533 | MODULE PROCEDURE dealloc_particles_array |
---|
534 | END INTERFACE dealloc_particles_array |
---|
535 | |
---|
536 | INTERFACE lpm_sort_and_delete |
---|
537 | MODULE PROCEDURE lpm_sort_and_delete |
---|
538 | END INTERFACE lpm_sort_and_delete |
---|
539 | |
---|
540 | INTERFACE lpm_sort_timeloop_done |
---|
541 | MODULE PROCEDURE lpm_sort_timeloop_done |
---|
542 | END INTERFACE lpm_sort_timeloop_done |
---|
543 | |
---|
544 | INTERFACE lpm_pack |
---|
545 | MODULE PROCEDURE lpm_pack |
---|
546 | END INTERFACE lpm_pack |
---|
547 | |
---|
548 | CONTAINS |
---|
549 | |
---|
550 | |
---|
551 | !--------------------------------------------------------------------------------------------------! |
---|
552 | ! Description: |
---|
553 | ! ------------ |
---|
554 | !> Parin for &particle_parameters for the Lagrangian particle model |
---|
555 | !--------------------------------------------------------------------------------------------------! |
---|
556 | SUBROUTINE lpm_parin |
---|
557 | |
---|
558 | CHARACTER (LEN=80) :: line !< |
---|
559 | |
---|
560 | NAMELIST /particles_par/ & |
---|
561 | aero_species, & |
---|
562 | aero_type, & |
---|
563 | aero_weight, & |
---|
564 | alloc_factor, & |
---|
565 | bc_par_b, & |
---|
566 | bc_par_lr, & |
---|
567 | bc_par_ns, & |
---|
568 | bc_par_t, & |
---|
569 | collision_kernel, & |
---|
570 | curvature_solution_effects, & |
---|
571 | deallocate_memory, & |
---|
572 | density_ratio, & |
---|
573 | dissipation_classes, & |
---|
574 | dt_dopts, & |
---|
575 | dt_min_part, & |
---|
576 | dt_prel, & |
---|
577 | dt_write_particle_data, & |
---|
578 | end_time_prel, & |
---|
579 | initial_weighting_factor, & |
---|
580 | log_sigma, & |
---|
581 | max_number_particles_per_gridbox, & |
---|
582 | merging, & |
---|
583 | na, & |
---|
584 | number_concentration, & |
---|
585 | number_of_particle_groups, & |
---|
586 | number_particles_per_gridbox, & |
---|
587 | particles_per_point, & |
---|
588 | particle_advection_start, & |
---|
589 | particle_advection_interpolation, & |
---|
590 | particle_maximum_age, & |
---|
591 | pdx, & |
---|
592 | pdy, & |
---|
593 | pdz, & |
---|
594 | psb, & |
---|
595 | psl, & |
---|
596 | psn, & |
---|
597 | psr, & |
---|
598 | pss, & |
---|
599 | pst, & |
---|
600 | radius, & |
---|
601 | radius_classes, & |
---|
602 | radius_merge, & |
---|
603 | radius_split, & |
---|
604 | random_start_position, & |
---|
605 | read_particles_from_restartfile, & |
---|
606 | rm, & |
---|
607 | seed_follows_topography, & |
---|
608 | splitting, & |
---|
609 | splitting_factor, & |
---|
610 | splitting_factor_max, & |
---|
611 | splitting_function, & |
---|
612 | splitting_mode, & |
---|
613 | step_dealloc, & |
---|
614 | use_sgs_for_particles, & |
---|
615 | vertical_particle_advection, & |
---|
616 | weight_factor_merge, & |
---|
617 | weight_factor_split, & |
---|
618 | write_particle_statistics |
---|
619 | |
---|
620 | NAMELIST /particle_parameters/ & |
---|
621 | aero_species, & |
---|
622 | aero_type, & |
---|
623 | aero_weight, & |
---|
624 | alloc_factor, & |
---|
625 | bc_par_b, & |
---|
626 | bc_par_lr, & |
---|
627 | bc_par_ns, & |
---|
628 | bc_par_t, & |
---|
629 | collision_kernel, & |
---|
630 | curvature_solution_effects, & |
---|
631 | deallocate_memory, & |
---|
632 | density_ratio, & |
---|
633 | dissipation_classes, & |
---|
634 | dt_dopts, & |
---|
635 | dt_min_part, & |
---|
636 | dt_prel, & |
---|
637 | dt_write_particle_data, & |
---|
638 | end_time_prel, & |
---|
639 | initial_weighting_factor, & |
---|
640 | log_sigma, & |
---|
641 | max_number_particles_per_gridbox, & |
---|
642 | merging, & |
---|
643 | na, & |
---|
644 | number_concentration, & |
---|
645 | number_of_output_particles, & |
---|
646 | number_of_particle_groups, & |
---|
647 | number_particles_per_gridbox, & |
---|
648 | oversize, & |
---|
649 | particles_per_point, & |
---|
650 | particle_advection_start, & |
---|
651 | particle_advection_interpolation, & |
---|
652 | particle_maximum_age, & |
---|
653 | part_output, & |
---|
654 | part_inc, & |
---|
655 | part_percent, & |
---|
656 | pdx, & |
---|
657 | pdy, & |
---|
658 | pdz, & |
---|
659 | psb, & |
---|
660 | psl, & |
---|
661 | psn, & |
---|
662 | psr, & |
---|
663 | pss, & |
---|
664 | pst, & |
---|
665 | radius, & |
---|
666 | radius_classes, & |
---|
667 | radius_merge, & |
---|
668 | radius_split, & |
---|
669 | random_start_position, & |
---|
670 | read_particles_from_restartfile, & |
---|
671 | rm, & |
---|
672 | seed_follows_topography, & |
---|
673 | splitting, & |
---|
674 | splitting_factor, & |
---|
675 | splitting_factor_max, & |
---|
676 | splitting_function, & |
---|
677 | splitting_mode, & |
---|
678 | step_dealloc, & |
---|
679 | unlimited_dimension, & |
---|
680 | use_sgs_for_particles, & |
---|
681 | vertical_particle_advection, & |
---|
682 | weight_factor_merge, & |
---|
683 | weight_factor_split, & |
---|
684 | write_particle_statistics |
---|
685 | |
---|
686 | ! |
---|
687 | !-- Position the namelist-file at the beginning (it was already opened in parin), search for the |
---|
688 | !-- namelist-group of the package and position the file at this line. Do the same for each |
---|
689 | !-- optionally used package. |
---|
690 | line = ' ' |
---|
691 | |
---|
692 | ! |
---|
693 | !-- Try to find particles package |
---|
694 | REWIND ( 11 ) |
---|
695 | line = ' ' |
---|
696 | DO WHILE ( INDEX( line, '&particle_parameters' ) == 0 ) |
---|
697 | READ ( 11, '(A)', END=12 ) line |
---|
698 | ENDDO |
---|
699 | BACKSPACE ( 11 ) |
---|
700 | ! |
---|
701 | !-- Read user-defined namelist |
---|
702 | READ ( 11, particle_parameters, ERR = 10 ) |
---|
703 | ! |
---|
704 | !-- Set flag that indicates that particles are switched on |
---|
705 | particle_advection = .TRUE. |
---|
706 | |
---|
707 | GOTO 14 |
---|
708 | |
---|
709 | 10 BACKSPACE( 11 ) |
---|
710 | READ( 11 , '(A)') line |
---|
711 | CALL parin_fail_message( 'particle_parameters', line ) |
---|
712 | ! |
---|
713 | !-- Try to find particles package (old namelist) |
---|
714 | 12 REWIND ( 11 ) |
---|
715 | line = ' ' |
---|
716 | DO WHILE ( INDEX( line, '&particles_par' ) == 0 ) |
---|
717 | READ ( 11, '(A)', END=14 ) line |
---|
718 | ENDDO |
---|
719 | BACKSPACE ( 11 ) |
---|
720 | ! |
---|
721 | !-- Read user-defined namelist |
---|
722 | READ ( 11, particles_par, ERR = 13, END = 14 ) |
---|
723 | |
---|
724 | message_string = 'namelist particles_par is deprecated and will be ' // & |
---|
725 | 'removed in near future. Please use namelist ' // & |
---|
726 | 'particle_parameters instead' |
---|
727 | CALL message( 'package_parin', 'PA0487', 0, 1, 0, 6, 0 ) |
---|
728 | |
---|
729 | ! |
---|
730 | !-- Set flag that indicates that particles are switched on |
---|
731 | particle_advection = .TRUE. |
---|
732 | |
---|
733 | GOTO 14 |
---|
734 | |
---|
735 | 13 BACKSPACE( 11 ) |
---|
736 | READ( 11 , '(A)') line |
---|
737 | CALL parin_fail_message( 'particles_par', line ) |
---|
738 | |
---|
739 | 14 CONTINUE |
---|
740 | |
---|
741 | END SUBROUTINE lpm_parin |
---|
742 | |
---|
743 | !--------------------------------------------------------------------------------------------------! |
---|
744 | ! Description: |
---|
745 | ! ------------ |
---|
746 | !> Writes used particle attributes in header file. |
---|
747 | !--------------------------------------------------------------------------------------------------! |
---|
748 | SUBROUTINE lpm_header ( io ) |
---|
749 | |
---|
750 | CHARACTER (LEN=40) :: output_format !< netcdf format |
---|
751 | |
---|
752 | INTEGER(iwp) :: i !< |
---|
753 | INTEGER(iwp), INTENT(IN) :: io !< Unit of the output file |
---|
754 | |
---|
755 | |
---|
756 | IF ( humidity .AND. cloud_droplets ) THEN |
---|
757 | WRITE ( io, 433 ) |
---|
758 | IF ( curvature_solution_effects ) WRITE ( io, 434 ) |
---|
759 | IF ( collision_kernel /= 'none' ) THEN |
---|
760 | WRITE ( io, 435 ) TRIM( collision_kernel ) |
---|
761 | IF ( collision_kernel(6:9) == 'fast' ) THEN |
---|
762 | WRITE ( io, 436 ) radius_classes, dissipation_classes |
---|
763 | ENDIF |
---|
764 | ELSE |
---|
765 | WRITE ( io, 437 ) |
---|
766 | ENDIF |
---|
767 | ENDIF |
---|
768 | |
---|
769 | IF ( particle_advection ) THEN |
---|
770 | ! |
---|
771 | !-- Particle attributes |
---|
772 | WRITE ( io, 480 ) particle_advection_start, TRIM(particle_advection_interpolation), & |
---|
773 | dt_prel, bc_par_lr, bc_par_ns, bc_par_b, bc_par_t, particle_maximum_age, & |
---|
774 | end_time_prel |
---|
775 | IF ( use_sgs_for_particles ) WRITE ( io, 488 ) dt_min_part |
---|
776 | IF ( random_start_position ) WRITE ( io, 481 ) |
---|
777 | IF ( seed_follows_topography ) WRITE ( io, 496 ) |
---|
778 | IF ( particles_per_point > 1 ) WRITE ( io, 489 ) particles_per_point |
---|
779 | WRITE ( io, 495 ) total_number_of_particles |
---|
780 | IF ( dt_write_particle_data /= 9999999.9_wp ) THEN |
---|
781 | WRITE ( io, 485 ) dt_write_particle_data |
---|
782 | IF ( netcdf_data_format > 1 ) THEN |
---|
783 | output_format = 'netcdf (64 bit offset) and binary' |
---|
784 | ELSE |
---|
785 | output_format = 'netcdf and binary' |
---|
786 | ENDIF |
---|
787 | IF ( netcdf_deflate == 0 ) THEN |
---|
788 | WRITE ( io, 344 ) output_format |
---|
789 | ELSE |
---|
790 | WRITE ( io, 354 ) TRIM( output_format ), netcdf_deflate |
---|
791 | ENDIF |
---|
792 | ENDIF |
---|
793 | IF ( dt_dopts /= 9999999.9_wp ) WRITE ( io, 494 ) dt_dopts |
---|
794 | IF ( write_particle_statistics ) WRITE ( io, 486 ) |
---|
795 | |
---|
796 | WRITE ( io, 487 ) number_of_particle_groups |
---|
797 | |
---|
798 | DO i = 1, number_of_particle_groups |
---|
799 | WRITE ( io, 490 ) i, radius(i) |
---|
800 | IF ( density_ratio(i) /= 0.0_wp ) THEN |
---|
801 | WRITE ( io, 491 ) density_ratio(i) |
---|
802 | ELSE |
---|
803 | WRITE ( io, 492 ) |
---|
804 | ENDIF |
---|
805 | WRITE ( io, 493 ) psl(i), psr(i), pss(i), psn(i), psb(i), pst(i), pdx(i), pdy(i), pdz(i) |
---|
806 | IF ( .NOT. vertical_particle_advection(i) ) WRITE ( io, 482 ) |
---|
807 | ENDDO |
---|
808 | |
---|
809 | ENDIF |
---|
810 | |
---|
811 | 344 FORMAT (' Output format: ',A/) |
---|
812 | 354 FORMAT (' Output format: ',A, ' compressed with level: ',I1/) |
---|
813 | |
---|
814 | 433 FORMAT (' Cloud droplets treated explicitly using the Lagrangian part','icle model') |
---|
815 | 434 FORMAT (' Curvature and solution effecs are considered for growth of', & |
---|
816 | ' droplets < 1.0E-6 m') |
---|
817 | 435 FORMAT (' Droplet collision is handled by ',A,'-kernel') |
---|
818 | 436 FORMAT (' Fast kernel with fixed radius- and dissipation classes ','are used'/ & |
---|
819 | ' number of radius classes: ',I3,' interval ','[1.0E-6,2.0E-4] m'/ & |
---|
820 | ' number of dissipation classes: ',I2,' interval ','[0,1000] cm**2/s**3') |
---|
821 | 437 FORMAT (' Droplet collision is switched off') |
---|
822 | |
---|
823 | 480 FORMAT (' Particles:'/ & |
---|
824 | ' ---------'// & |
---|
825 | ' Particle advection is active (switched on at t = ', F7.1,' s)'/ & |
---|
826 | ' Interpolation of particle velocities is done by using ', A,' method'/ & |
---|
827 | ' Start of new particle generations every ',F6.1,' s'/ & |
---|
828 | ' Boundary conditions: left/right: ', A, ' north/south: ', A/ & |
---|
829 | ' bottom: ', A, ' top: ', A/ & |
---|
830 | ' Maximum particle age: ',F9.1,' s'/ & |
---|
831 | ' Advection stopped at t = ',F9.1,' s'/) |
---|
832 | 481 FORMAT (' Particles have random start positions'/) |
---|
833 | 482 FORMAT (' Particles are advected only horizontally'/) |
---|
834 | 485 FORMAT (' Particle data are written on file every ', F9.1, ' s') |
---|
835 | 486 FORMAT (' Particle statistics are written on file'/) |
---|
836 | 487 FORMAT (' Number of particle groups: ',I2/) |
---|
837 | 488 FORMAT (' SGS velocity components are used for particle advection'/ & |
---|
838 | ' minimum timestep for advection:', F8.5/) |
---|
839 | 489 FORMAT (' Number of particles simultaneously released at each ','point: ', I5/) |
---|
840 | 490 FORMAT (' Particle group ',I2,':'/ & |
---|
841 | ' Particle radius: ',E10.3, 'm') |
---|
842 | 491 FORMAT (' Particle inertia is activated'/ & |
---|
843 | ' density_ratio (rho_fluid/rho_particle) =',F6.3/) |
---|
844 | 492 FORMAT (' Particles are advected only passively (no inertia)'/) |
---|
845 | 493 FORMAT (' Boundaries of particle source: x:',F8.1,' - ',F8.1,' m'/ & |
---|
846 | ' y:',F8.1,' - ',F8.1,' m'/ & |
---|
847 | ' z:',F8.1,' - ',F8.1,' m'/ & |
---|
848 | ' Particle distances: dx = ',F8.1,' m dy = ',F8.1,' m dz = ',F8.1,' m'/) |
---|
849 | 494 FORMAT (' Output of particle time series in NetCDF format every ',F8.2,' s'/) |
---|
850 | 495 FORMAT (' Number of particles in total domain: ',I10/) |
---|
851 | 496 FORMAT (' Initial vertical particle positions are interpreted ', & |
---|
852 | 'as relative to the given topography') |
---|
853 | |
---|
854 | END SUBROUTINE lpm_header |
---|
855 | |
---|
856 | !--------------------------------------------------------------------------------------------------! |
---|
857 | ! Description: |
---|
858 | ! ------------ |
---|
859 | !> Writes used particle attributes in header file. |
---|
860 | !--------------------------------------------------------------------------------------------------! |
---|
861 | SUBROUTINE lpm_check_parameters |
---|
862 | |
---|
863 | ! |
---|
864 | !-- Collision kernels: |
---|
865 | SELECT CASE ( TRIM( collision_kernel ) ) |
---|
866 | |
---|
867 | CASE ( 'hall', 'hall_fast' ) |
---|
868 | hall_kernel = .TRUE. |
---|
869 | |
---|
870 | CASE ( 'wang', 'wang_fast' ) |
---|
871 | wang_kernel = .TRUE. |
---|
872 | |
---|
873 | CASE ( 'none' ) |
---|
874 | |
---|
875 | |
---|
876 | CASE DEFAULT |
---|
877 | message_string = 'unknown collision kernel: collision_kernel = "' // & |
---|
878 | TRIM( collision_kernel ) // '"' |
---|
879 | CALL message( 'lpm_check_parameters', 'PA0350', 1, 2, 0, 6, 0 ) |
---|
880 | |
---|
881 | END SELECT |
---|
882 | IF ( collision_kernel(6:9) == 'fast' ) use_kernel_tables = .TRUE. |
---|
883 | |
---|
884 | ! |
---|
885 | !-- Subgrid scale velocites with the simple interpolation method for resolved velocites is not |
---|
886 | !-- implemented for passive particles. However, for cloud it can be combined as the sgs-velocites |
---|
887 | !-- for active particles are calculated differently, i.e. no subboxes are needed. |
---|
888 | IF ( .NOT. TRIM( particle_advection_interpolation ) == 'trilinear' .AND. & |
---|
889 | use_sgs_for_particles .AND. .NOT. cloud_droplets ) THEN |
---|
890 | message_string = 'subrgrid scale velocities in combination with ' // & |
---|
891 | 'simple interpolation method is not ' // & |
---|
892 | 'implemented' |
---|
893 | CALL message( 'lpm_check_parameters', 'PA0659', 1, 2, 0, 6, 0 ) |
---|
894 | ENDIF |
---|
895 | |
---|
896 | IF ( nested_run .AND. cloud_droplets ) THEN |
---|
897 | message_string = 'nested runs in combination with cloud droplets ' // & |
---|
898 | 'is not implemented' |
---|
899 | CALL message( 'lpm_check_parameters', 'PA0687', 1, 2, 0, 6, 0 ) |
---|
900 | ENDIF |
---|
901 | |
---|
902 | |
---|
903 | END SUBROUTINE lpm_check_parameters |
---|
904 | |
---|
905 | !--------------------------------------------------------------------------------------------------! |
---|
906 | ! Description: |
---|
907 | ! ------------ |
---|
908 | !> Initialize arrays for lpm |
---|
909 | !--------------------------------------------------------------------------------------------------! |
---|
910 | SUBROUTINE lpm_init_arrays |
---|
911 | |
---|
912 | IF ( cloud_droplets ) THEN |
---|
913 | ! |
---|
914 | !-- Liquid water content, change in liquid water content |
---|
915 | ALLOCATE ( ql_1(nzb:nzt+1,nysg:nyng,nxlg:nxrg), & |
---|
916 | ql_2(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
917 | !-- Real volume of particles (with weighting), volume of particles |
---|
918 | ALLOCATE ( ql_v(nzb:nzt+1,nysg:nyng,nxlg:nxrg), & |
---|
919 | ql_vp(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
920 | ENDIF |
---|
921 | |
---|
922 | |
---|
923 | ALLOCATE( u_t(nzb:nzt+1,nysg:nyng,nxlg:nxrg), & |
---|
924 | v_t(nzb:nzt+1,nysg:nyng,nxlg:nxrg), & |
---|
925 | w_t(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
926 | ! |
---|
927 | !-- Initialize values with current time step |
---|
928 | u_t = u |
---|
929 | v_t = v |
---|
930 | w_t = w |
---|
931 | ! |
---|
932 | !-- Initial assignment of the pointers |
---|
933 | IF ( cloud_droplets ) THEN |
---|
934 | ql => ql_1 |
---|
935 | ql_c => ql_2 |
---|
936 | ENDIF |
---|
937 | |
---|
938 | END SUBROUTINE lpm_init_arrays |
---|
939 | |
---|
940 | !--------------------------------------------------------------------------------------------------! |
---|
941 | ! Description: |
---|
942 | ! ------------ |
---|
943 | !> Initialize Lagrangian particle model |
---|
944 | !--------------------------------------------------------------------------------------------------! |
---|
945 | SUBROUTINE lpm_init |
---|
946 | |
---|
947 | INTEGER(iwp) :: i !< |
---|
948 | INTEGER(iwp) :: j !< |
---|
949 | INTEGER(iwp) :: k !< |
---|
950 | |
---|
951 | REAL(wp) :: div !< |
---|
952 | REAL(wp) :: height_int !< |
---|
953 | REAL(wp) :: height_p !< |
---|
954 | REAL(wp) :: z_p !< |
---|
955 | REAL(wp) :: z0_av_local !< |
---|
956 | |
---|
957 | ! |
---|
958 | !-- In case of oceans runs, the vertical index calculations need an offset, because otherwise the k |
---|
959 | !-- indices will become negative |
---|
960 | IF ( ocean_mode ) THEN |
---|
961 | offset_ocean_nzt = nzt |
---|
962 | offset_ocean_nzt_m1 = nzt - 1 |
---|
963 | ENDIF |
---|
964 | |
---|
965 | ! |
---|
966 | !-- Define block offsets for dividing a gridcell in 8 sub cells. |
---|
967 | !-- See documentation for List of subgrid boxes. |
---|
968 | !-- See pack_and_sort in lpm_pack_arrays.f90 for assignment of the subgrid boxes. |
---|
969 | block_offset(0) = block_offset_def ( 0, 0, 0) |
---|
970 | block_offset(1) = block_offset_def ( 0, 0,-1) |
---|
971 | block_offset(2) = block_offset_def ( 0,-1, 0) |
---|
972 | block_offset(3) = block_offset_def ( 0,-1,-1) |
---|
973 | block_offset(4) = block_offset_def (-1, 0, 0) |
---|
974 | block_offset(5) = block_offset_def (-1, 0,-1) |
---|
975 | block_offset(6) = block_offset_def (-1,-1, 0) |
---|
976 | block_offset(7) = block_offset_def (-1,-1,-1) |
---|
977 | ! |
---|
978 | !-- Check the number of particle groups. |
---|
979 | IF ( number_of_particle_groups > max_number_of_particle_groups ) THEN |
---|
980 | WRITE( message_string, * ) 'max_number_of_particle_groups =', & |
---|
981 | max_number_of_particle_groups , & |
---|
982 | '&number_of_particle_groups reset to ', & |
---|
983 | max_number_of_particle_groups |
---|
984 | CALL message( 'lpm_init', 'PA0213', 0, 1, 0, 6, 0 ) |
---|
985 | number_of_particle_groups = max_number_of_particle_groups |
---|
986 | ENDIF |
---|
987 | ! |
---|
988 | !-- Check if downward-facing walls exist. This case, reflection boundary conditions (as well as |
---|
989 | !-- subgrid-scale velocities) may do not work properly (not realized so far). |
---|
990 | IF ( surf_def_h(1)%ns >= 1 ) THEN |
---|
991 | WRITE( message_string, * ) 'Overhanging topography do not work '// & |
---|
992 | 'with particles' |
---|
993 | CALL message( 'lpm_init', 'PA0212', 0, 1, 0, 6, 0 ) |
---|
994 | |
---|
995 | ENDIF |
---|
996 | |
---|
997 | ! |
---|
998 | !-- Set default start positions, if necessary |
---|
999 | IF ( psl(1) == 9999999.9_wp ) psl(1) = 0.0_wp |
---|
1000 | IF ( psr(1) == 9999999.9_wp ) psr(1) = ( nx +1 ) * dx |
---|
1001 | IF ( pss(1) == 9999999.9_wp ) pss(1) = 0.0_wp |
---|
1002 | IF ( psn(1) == 9999999.9_wp ) psn(1) = ( ny +1 ) * dy |
---|
1003 | IF ( psb(1) == 9999999.9_wp ) psb(1) = zu(nz/2) |
---|
1004 | IF ( pst(1) == 9999999.9_wp ) pst(1) = psb(1) |
---|
1005 | |
---|
1006 | IF ( pdx(1) == 9999999.9_wp .OR. pdx(1) == 0.0_wp ) pdx(1) = dx |
---|
1007 | IF ( pdy(1) == 9999999.9_wp .OR. pdy(1) == 0.0_wp ) pdy(1) = dy |
---|
1008 | IF ( pdz(1) == 9999999.9_wp .OR. pdz(1) == 0.0_wp ) pdz(1) = zu(2) - zu(1) |
---|
1009 | |
---|
1010 | ! |
---|
1011 | !-- If number_particles_per_gridbox is set, the parametres pdx, pdy and pdz are calculated |
---|
1012 | !-- diagnostically. Therfore an isotropic distribution is prescribed. |
---|
1013 | IF ( number_particles_per_gridbox /= -1 .AND. & |
---|
1014 | number_particles_per_gridbox >= 1 ) THEN |
---|
1015 | pdx(1) = (( dx * dy * ( zu(2) - zu(1) ) ) / & |
---|
1016 | REAL(number_particles_per_gridbox))**0.3333333_wp |
---|
1017 | ! |
---|
1018 | !-- Ensure a smooth value (two significant digits) of distance between particles (pdx, pdy, pdz). |
---|
1019 | div = 1000.0_wp |
---|
1020 | DO WHILE ( pdx(1) < div ) |
---|
1021 | div = div / 10.0_wp |
---|
1022 | ENDDO |
---|
1023 | pdx(1) = NINT( pdx(1) * 100.0_wp / div ) * div / 100.0_wp |
---|
1024 | pdy(1) = pdx(1) |
---|
1025 | pdz(1) = pdx(1) |
---|
1026 | |
---|
1027 | ENDIF |
---|
1028 | |
---|
1029 | DO j = 2, number_of_particle_groups |
---|
1030 | IF ( psl(j) == 9999999.9_wp ) psl(j) = psl(j-1) |
---|
1031 | IF ( psr(j) == 9999999.9_wp ) psr(j) = psr(j-1) |
---|
1032 | IF ( pss(j) == 9999999.9_wp ) pss(j) = pss(j-1) |
---|
1033 | IF ( psn(j) == 9999999.9_wp ) psn(j) = psn(j-1) |
---|
1034 | IF ( psb(j) == 9999999.9_wp ) psb(j) = psb(j-1) |
---|
1035 | IF ( pst(j) == 9999999.9_wp ) pst(j) = pst(j-1) |
---|
1036 | IF ( pdx(j) == 9999999.9_wp .OR. pdx(j) == 0.0_wp ) pdx(j) = pdx(j-1) |
---|
1037 | IF ( pdy(j) == 9999999.9_wp .OR. pdy(j) == 0.0_wp ) pdy(j) = pdy(j-1) |
---|
1038 | IF ( pdz(j) == 9999999.9_wp .OR. pdz(j) == 0.0_wp ) pdz(j) = pdz(j-1) |
---|
1039 | ENDDO |
---|
1040 | |
---|
1041 | ! |
---|
1042 | !-- Allocate arrays required for calculating particle SGS velocities. |
---|
1043 | !-- Initialize prefactor required for stoachastic Weil equation. |
---|
1044 | IF ( use_sgs_for_particles .AND. .NOT. cloud_droplets ) THEN |
---|
1045 | ALLOCATE( de_dx(nzb:nzt+1,nysg:nyng,nxlg:nxrg), & |
---|
1046 | de_dy(nzb:nzt+1,nysg:nyng,nxlg:nxrg), & |
---|
1047 | de_dz(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
1048 | |
---|
1049 | de_dx = 0.0_wp |
---|
1050 | de_dy = 0.0_wp |
---|
1051 | de_dz = 0.0_wp |
---|
1052 | |
---|
1053 | sgs_wf_part = 1.0_wp / 3.0_wp |
---|
1054 | ENDIF |
---|
1055 | |
---|
1056 | ! |
---|
1057 | !-- Allocate array required for logarithmic vertical interpolation of horizontal particle velocities |
---|
1058 | !-- between the surface and the first vertical grid level. In order to avoid repeated CPU |
---|
1059 | !-- cost-intensive CALLS of intrinsic FORTRAN procedure LOG(z/z0), LOG(z/z0) is precalculated for |
---|
1060 | !-- several heights. Splitting into 20 sublayers turned out to be sufficient. |
---|
1061 | !-- To obtain exact height levels of particles, linear interpolation is applied (see lpm_advec.f90). |
---|
1062 | IF ( constant_flux_layer ) THEN |
---|
1063 | |
---|
1064 | ALLOCATE ( log_z_z0(0:number_of_sublayers) ) |
---|
1065 | z_p = zu(nzb+1) - zw(nzb) |
---|
1066 | |
---|
1067 | ! |
---|
1068 | !-- Calculate horizontal mean value of z0 used for logartihmic interpolation. Note: this is not |
---|
1069 | !-- exact for heterogeneous z0. |
---|
1070 | !-- However, sensitivity studies showed that the effect is negligible. |
---|
1071 | z0_av_local = SUM( surf_def_h(0)%z0 ) + SUM( surf_lsm_h(0)%z0 ) + SUM( surf_usm_h(0)%z0 ) |
---|
1072 | z0_av_global = 0.0_wp |
---|
1073 | |
---|
1074 | #if defined( __parallel ) |
---|
1075 | CALL MPI_ALLREDUCE( z0_av_local, z0_av_global, 1, MPI_REAL, MPI_SUM, comm2d, ierr ) |
---|
1076 | #else |
---|
1077 | z0_av_global = z0_av_local |
---|
1078 | #endif |
---|
1079 | |
---|
1080 | z0_av_global = z0_av_global / ( ( ny + 1 ) * ( nx + 1 ) ) |
---|
1081 | ! |
---|
1082 | !-- Horizontal wind speed is zero below and at z0 |
---|
1083 | log_z_z0(0) = 0.0_wp |
---|
1084 | ! |
---|
1085 | !-- Calculate vertical depth of the sublayers |
---|
1086 | height_int = ( z_p - z0_av_global ) / REAL( number_of_sublayers, KIND=wp ) |
---|
1087 | ! |
---|
1088 | !-- Precalculate LOG(z/z0) |
---|
1089 | height_p = z0_av_global |
---|
1090 | DO k = 1, number_of_sublayers |
---|
1091 | |
---|
1092 | height_p = height_p + height_int |
---|
1093 | log_z_z0(k) = LOG( height_p / z0_av_global ) |
---|
1094 | |
---|
1095 | ENDDO |
---|
1096 | |
---|
1097 | ENDIF |
---|
1098 | |
---|
1099 | ! |
---|
1100 | !-- Check which particle interpolation method should be used |
---|
1101 | IF ( TRIM( particle_advection_interpolation ) == 'trilinear' ) THEN |
---|
1102 | interpolation_simple_corrector = .FALSE. |
---|
1103 | interpolation_simple_predictor = .FALSE. |
---|
1104 | interpolation_trilinear = .TRUE. |
---|
1105 | ELSEIF ( TRIM( particle_advection_interpolation ) == 'simple_corrector' ) THEN |
---|
1106 | interpolation_simple_corrector = .TRUE. |
---|
1107 | interpolation_simple_predictor = .FALSE. |
---|
1108 | interpolation_trilinear = .FALSE. |
---|
1109 | ELSEIF ( TRIM( particle_advection_interpolation ) == 'simple_predictor' ) THEN |
---|
1110 | interpolation_simple_corrector = .FALSE. |
---|
1111 | interpolation_simple_predictor = .TRUE. |
---|
1112 | interpolation_trilinear = .FALSE. |
---|
1113 | ENDIF |
---|
1114 | |
---|
1115 | ! |
---|
1116 | !-- Check boundary condition and set internal variables |
---|
1117 | SELECT CASE ( bc_par_b ) |
---|
1118 | |
---|
1119 | CASE ( 'absorb' ) |
---|
1120 | ibc_par_b = 1 |
---|
1121 | |
---|
1122 | CASE ( 'reflect' ) |
---|
1123 | ibc_par_b = 2 |
---|
1124 | |
---|
1125 | CASE ( 'reset' ) |
---|
1126 | ibc_par_b = 3 |
---|
1127 | |
---|
1128 | CASE DEFAULT |
---|
1129 | WRITE( message_string, * ) 'unknown boundary condition ', & |
---|
1130 | 'bc_par_b = "', TRIM( bc_par_b ), '"' |
---|
1131 | CALL message( 'lpm_init', 'PA0217', 1, 2, 0, 6, 0 ) |
---|
1132 | |
---|
1133 | END SELECT |
---|
1134 | SELECT CASE ( bc_par_t ) |
---|
1135 | |
---|
1136 | CASE ( 'absorb' ) |
---|
1137 | ibc_par_t = 1 |
---|
1138 | |
---|
1139 | CASE ( 'reflect' ) |
---|
1140 | ibc_par_t = 2 |
---|
1141 | |
---|
1142 | CASE ( 'nested' ) |
---|
1143 | ibc_par_t = 3 |
---|
1144 | |
---|
1145 | CASE DEFAULT |
---|
1146 | WRITE( message_string, * ) 'unknown boundary condition ', & |
---|
1147 | 'bc_par_t = "', TRIM( bc_par_t ), '"' |
---|
1148 | CALL message( 'lpm_init', 'PA0218', 1, 2, 0, 6, 0 ) |
---|
1149 | |
---|
1150 | END SELECT |
---|
1151 | SELECT CASE ( bc_par_lr ) |
---|
1152 | |
---|
1153 | CASE ( 'cyclic' ) |
---|
1154 | ibc_par_lr = 0 |
---|
1155 | |
---|
1156 | CASE ( 'absorb' ) |
---|
1157 | ibc_par_lr = 1 |
---|
1158 | |
---|
1159 | CASE ( 'reflect' ) |
---|
1160 | ibc_par_lr = 2 |
---|
1161 | |
---|
1162 | CASE ( 'nested' ) |
---|
1163 | ibc_par_lr = 3 |
---|
1164 | |
---|
1165 | CASE DEFAULT |
---|
1166 | WRITE( message_string, * ) 'unknown boundary condition ', & |
---|
1167 | 'bc_par_lr = "', TRIM( bc_par_lr ), '"' |
---|
1168 | CALL message( 'lpm_init', 'PA0219', 1, 2, 0, 6, 0 ) |
---|
1169 | |
---|
1170 | END SELECT |
---|
1171 | SELECT CASE ( bc_par_ns ) |
---|
1172 | |
---|
1173 | CASE ( 'cyclic' ) |
---|
1174 | ibc_par_ns = 0 |
---|
1175 | |
---|
1176 | CASE ( 'absorb' ) |
---|
1177 | ibc_par_ns = 1 |
---|
1178 | |
---|
1179 | CASE ( 'reflect' ) |
---|
1180 | ibc_par_ns = 2 |
---|
1181 | |
---|
1182 | CASE ( 'nested' ) |
---|
1183 | ibc_par_ns = 3 |
---|
1184 | |
---|
1185 | CASE DEFAULT |
---|
1186 | WRITE( message_string, * ) 'unknown boundary condition ', & |
---|
1187 | 'bc_par_ns = "', TRIM( bc_par_ns ), '"' |
---|
1188 | CALL message( 'lpm_init', 'PA0220', 1, 2, 0, 6, 0 ) |
---|
1189 | |
---|
1190 | END SELECT |
---|
1191 | SELECT CASE ( splitting_mode ) |
---|
1192 | |
---|
1193 | CASE ( 'const' ) |
---|
1194 | i_splitting_mode = 1 |
---|
1195 | |
---|
1196 | CASE ( 'cl_av' ) |
---|
1197 | i_splitting_mode = 2 |
---|
1198 | |
---|
1199 | CASE ( 'gb_av' ) |
---|
1200 | i_splitting_mode = 3 |
---|
1201 | |
---|
1202 | CASE DEFAULT |
---|
1203 | WRITE( message_string, * ) 'unknown splitting_mode = "', TRIM( splitting_mode ), '"' |
---|
1204 | CALL message( 'lpm_init', 'PA0146', 1, 2, 0, 6, 0 ) |
---|
1205 | |
---|
1206 | END SELECT |
---|
1207 | SELECT CASE ( splitting_function ) |
---|
1208 | |
---|
1209 | CASE ( 'gamma' ) |
---|
1210 | isf = 1 |
---|
1211 | |
---|
1212 | CASE ( 'log' ) |
---|
1213 | isf = 2 |
---|
1214 | |
---|
1215 | CASE ( 'exp' ) |
---|
1216 | isf = 3 |
---|
1217 | |
---|
1218 | CASE DEFAULT |
---|
1219 | WRITE( message_string, * ) 'unknown splitting function = "', & |
---|
1220 | TRIM( splitting_function ), '"' |
---|
1221 | CALL message( 'lpm_init', 'PA0147', 1, 2, 0, 6, 0 ) |
---|
1222 | |
---|
1223 | END SELECT |
---|
1224 | ! |
---|
1225 | !-- Initialize collision kernels |
---|
1226 | IF ( collision_kernel /= 'none' ) CALL lpm_init_kernels |
---|
1227 | |
---|
1228 | ! |
---|
1229 | !-- For the first model run of a possible job chain initialize the particles, otherwise read the |
---|
1230 | !-- particle data from restart file. |
---|
1231 | IF ( TRIM( initializing_actions ) == 'read_restart_data' & |
---|
1232 | .AND. read_particles_from_restartfile ) THEN |
---|
1233 | CALL lpm_rrd_local_particles |
---|
1234 | ELSE |
---|
1235 | ! |
---|
1236 | !-- Allocate particle arrays and set attributes of the initial set of particles, which can be |
---|
1237 | !-- also periodically released at later times. |
---|
1238 | ALLOCATE( prt_count(nzb:nzt+1,nysg:nyng,nxlg:nxrg), & |
---|
1239 | grid_particles(nzb+1:nzt,nys:nyn,nxl:nxr) ) |
---|
1240 | |
---|
1241 | number_of_particles = 0 |
---|
1242 | prt_count = 0 |
---|
1243 | ! |
---|
1244 | !-- Initialize counter for particle IDs |
---|
1245 | grid_particles%id_counter = 1 |
---|
1246 | ! |
---|
1247 | !-- Initialize all particles with dummy values (otherwise errors may occur within restart runs). |
---|
1248 | !-- The reason for this is still not clear and may be presumably caused by errors in the |
---|
1249 | !-- respective user-interface. |
---|
1250 | zero_particle = particle_type( 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, & |
---|
1251 | 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, & |
---|
1252 | 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, & |
---|
1253 | 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, & |
---|
1254 | 0, 0, 0_idp, .FALSE., -1, -1 ) |
---|
1255 | |
---|
1256 | particle_groups = particle_groups_type( 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp ) |
---|
1257 | ! |
---|
1258 | !-- Set values for the density ratio and radius for all particle groups, if necessary. |
---|
1259 | IF ( density_ratio(1) == 9999999.9_wp ) density_ratio(1) = 0.0_wp |
---|
1260 | IF ( radius(1) == 9999999.9_wp ) radius(1) = 0.0_wp |
---|
1261 | DO i = 2, number_of_particle_groups |
---|
1262 | IF ( density_ratio(i) == 9999999.9_wp ) THEN |
---|
1263 | density_ratio(i) = density_ratio(i-1) |
---|
1264 | ENDIF |
---|
1265 | IF ( radius(i) == 9999999.9_wp ) radius(i) = radius(i-1) |
---|
1266 | ENDDO |
---|
1267 | |
---|
1268 | DO i = 1, number_of_particle_groups |
---|
1269 | IF ( density_ratio(i) /= 0.0_wp .AND. radius(i) == 0 ) THEN |
---|
1270 | WRITE( message_string, * ) 'particle group #', i, ' has a', & |
---|
1271 | 'density ratio /= 0 but radius = 0' |
---|
1272 | CALL message( 'lpm_init', 'PA0215', 1, 2, 0, 6, 0 ) |
---|
1273 | ENDIF |
---|
1274 | particle_groups(i)%density_ratio = density_ratio(i) |
---|
1275 | particle_groups(i)%radius = radius(i) |
---|
1276 | ENDDO |
---|
1277 | |
---|
1278 | ! |
---|
1279 | !-- Initialize parallel random number sequence seed for particles. |
---|
1280 | !-- This is done separately here, as thus particle random numbers do not affect the random |
---|
1281 | !-- numbers used for the flow field (e.g. for generating flow disturbances). |
---|
1282 | ALLOCATE ( seq_random_array_particles(5,nys:nyn,nxl:nxr) ) |
---|
1283 | seq_random_array_particles = 0 |
---|
1284 | |
---|
1285 | !-- Initializing with random_seed_parallel for every vertical gridpoint column. |
---|
1286 | random_dummy = 0 |
---|
1287 | DO i = nxl, nxr |
---|
1288 | DO j = nys, nyn |
---|
1289 | CALL random_seed_parallel (random_sequence=id_random_array(j, i)) |
---|
1290 | CALL random_number_parallel (random_dummy) |
---|
1291 | CALL random_seed_parallel (get=seq_random_array_particles(:, j, i)) |
---|
1292 | ENDDO |
---|
1293 | ENDDO |
---|
1294 | ! |
---|
1295 | !-- Create the particle set, and set the initial particles |
---|
1296 | CALL lpm_create_particle( phase_init ) |
---|
1297 | last_particle_release_time = particle_advection_start |
---|
1298 | ! |
---|
1299 | !-- User modification of initial particles |
---|
1300 | CALL user_lpm_init |
---|
1301 | ! |
---|
1302 | !-- Open file for statistical informations about particle conditions |
---|
1303 | IF ( write_particle_statistics ) THEN |
---|
1304 | CALL check_open( 80 ) |
---|
1305 | WRITE ( 80, 8000 ) current_timestep_number, simulated_time, number_of_particles |
---|
1306 | CALL close_file( 80 ) |
---|
1307 | ENDIF |
---|
1308 | |
---|
1309 | ENDIF |
---|
1310 | |
---|
1311 | #if defined( __parallel ) |
---|
1312 | IF ( nested_run ) CALL pmcp_g_init |
---|
1313 | #endif |
---|
1314 | |
---|
1315 | ! |
---|
1316 | !-- next line is in preparation for particle data output |
---|
1317 | ! CALL dop_init |
---|
1318 | |
---|
1319 | ! |
---|
1320 | !-- To avoid programm abort, assign particles array to the local version of |
---|
1321 | !-- first grid cell |
---|
1322 | number_of_particles = prt_count(nzb+1,nys,nxl) |
---|
1323 | particles => grid_particles(nzb+1,nys,nxl)%particles(1:number_of_particles) |
---|
1324 | ! |
---|
1325 | !-- Formats |
---|
1326 | 8000 FORMAT (I6,1X,F7.2,4X,I10,71X,I10) |
---|
1327 | |
---|
1328 | END SUBROUTINE lpm_init |
---|
1329 | |
---|
1330 | !--------------------------------------------------------------------------------------------------! |
---|
1331 | ! Description: |
---|
1332 | ! ------------ |
---|
1333 | !> Create Lagrangian particles |
---|
1334 | !--------------------------------------------------------------------------------------------------! |
---|
1335 | SUBROUTINE lpm_create_particle (phase) |
---|
1336 | |
---|
1337 | INTEGER(iwp) :: alloc_size !< relative increase of allocated memory for particles |
---|
1338 | INTEGER(iwp) :: i !< loop variable ( particle groups ) |
---|
1339 | INTEGER(iwp) :: ip !< index variable along x |
---|
1340 | INTEGER(iwp) :: j !< loop variable ( particles per point ) |
---|
1341 | INTEGER(iwp) :: jp !< index variable along y |
---|
1342 | INTEGER(iwp) :: k !< index variable along z |
---|
1343 | INTEGER(iwp) :: k_surf !< index of surface grid point |
---|
1344 | INTEGER(iwp) :: kp !< index variable along z |
---|
1345 | INTEGER(iwp) :: loop_stride !< loop variable for initialization |
---|
1346 | INTEGER(iwp) :: n !< loop variable ( number of particles ) |
---|
1347 | INTEGER(iwp) :: new_size !< new size of allocated memory for particles |
---|
1348 | |
---|
1349 | INTEGER(iwp), INTENT(IN) :: phase !< mode of inititialization |
---|
1350 | |
---|
1351 | INTEGER(iwp), DIMENSION(nzb:nzt+1,nysg:nyng,nxlg:nxrg) :: local_count !< start address of new particle |
---|
1352 | INTEGER(iwp), DIMENSION(nzb:nzt+1,nysg:nyng,nxlg:nxrg) :: local_start !< start address of new particle |
---|
1353 | |
---|
1354 | LOGICAL :: first_stride !< flag for initialization |
---|
1355 | |
---|
1356 | REAL(wp) :: pos_x !< increment for particle position in x |
---|
1357 | REAL(wp) :: pos_y !< increment for particle position in y |
---|
1358 | REAL(wp) :: pos_z !< increment for particle position in z |
---|
1359 | REAL(wp) :: rand_contr !< dummy argument for random position |
---|
1360 | |
---|
1361 | TYPE(particle_type), TARGET :: tmp_particle !< temporary particle used for initialization |
---|
1362 | |
---|
1363 | |
---|
1364 | ! |
---|
1365 | !-- Calculate particle positions and store particle attributes, if particle is situated on this PE. |
---|
1366 | DO loop_stride = 1, 2 |
---|
1367 | first_stride = (loop_stride == 1) |
---|
1368 | IF ( first_stride ) THEN |
---|
1369 | local_count = 0 ! count number of particles |
---|
1370 | ELSE |
---|
1371 | local_count = prt_count ! Start address of new particles |
---|
1372 | ENDIF |
---|
1373 | |
---|
1374 | ! |
---|
1375 | !-- Calculate initial_weighting_factor diagnostically |
---|
1376 | IF ( number_concentration /= -1.0_wp .AND. number_concentration > 0.0_wp ) THEN |
---|
1377 | initial_weighting_factor = number_concentration * pdx(1) * pdy(1) * pdz(1) |
---|
1378 | END IF |
---|
1379 | |
---|
1380 | n = 0 |
---|
1381 | DO i = 1, number_of_particle_groups |
---|
1382 | pos_z = psb(i) |
---|
1383 | DO WHILE ( pos_z <= pst(i) ) |
---|
1384 | IF ( pos_z >= zw(0) .AND. pos_z < zw(nzt) ) THEN |
---|
1385 | pos_y = pss(i) |
---|
1386 | DO WHILE ( pos_y <= psn(i) ) |
---|
1387 | IF ( pos_y >= nys * dy .AND. pos_y < ( nyn + 1 ) * dy ) THEN |
---|
1388 | pos_x = psl(i) |
---|
1389 | xloop: DO WHILE ( pos_x <= psr(i) ) |
---|
1390 | IF ( pos_x >= nxl * dx .AND. pos_x < ( nxr + 1) * dx ) THEN |
---|
1391 | DO j = 1, particles_per_point |
---|
1392 | n = n + 1 |
---|
1393 | tmp_particle%x = pos_x |
---|
1394 | tmp_particle%y = pos_y |
---|
1395 | tmp_particle%z = pos_z |
---|
1396 | tmp_particle%age = 0.0_wp |
---|
1397 | tmp_particle%age_m = 0.0_wp |
---|
1398 | tmp_particle%dt_sum = 0.0_wp |
---|
1399 | tmp_particle%e_m = 0.0_wp |
---|
1400 | tmp_particle%rvar1 = 0.0_wp |
---|
1401 | tmp_particle%rvar2 = 0.0_wp |
---|
1402 | tmp_particle%rvar3 = 0.0_wp |
---|
1403 | tmp_particle%speed_x = 0.0_wp |
---|
1404 | tmp_particle%speed_y = 0.0_wp |
---|
1405 | tmp_particle%speed_z = 0.0_wp |
---|
1406 | tmp_particle%origin_x = pos_x |
---|
1407 | tmp_particle%origin_y = pos_y |
---|
1408 | tmp_particle%origin_z = pos_z |
---|
1409 | IF ( curvature_solution_effects ) THEN |
---|
1410 | tmp_particle%aux1 = 0.0_wp ! dry aerosol radius |
---|
1411 | tmp_particle%aux2 = dt_3d ! last Rosenbrock timestep |
---|
1412 | ELSE |
---|
1413 | tmp_particle%aux1 = 0.0_wp ! free to use |
---|
1414 | tmp_particle%aux2 = 0.0_wp ! free to use |
---|
1415 | ENDIF |
---|
1416 | tmp_particle%radius = particle_groups(i)%radius |
---|
1417 | tmp_particle%weight_factor = initial_weighting_factor |
---|
1418 | tmp_particle%class = 1 |
---|
1419 | tmp_particle%group = i |
---|
1420 | tmp_particle%id = 0_idp |
---|
1421 | tmp_particle%particle_mask = .TRUE. |
---|
1422 | tmp_particle%block_nr = -1 |
---|
1423 | ! |
---|
1424 | !-- Determine the grid indices of the particle position |
---|
1425 | ip = INT( tmp_particle%x * ddx ) |
---|
1426 | jp = INT( tmp_particle%y * ddy ) |
---|
1427 | ! |
---|
1428 | !-- In case of stretching the actual k index is found iteratively |
---|
1429 | IF ( dz_stretch_level /= -9999999.9_wp .OR. & |
---|
1430 | dz_stretch_level_start(1) /= -9999999.9_wp ) THEN |
---|
1431 | kp = MAX( MINLOC( ABS( tmp_particle%z - zu ), DIM = 1 ) - 1, 1 ) |
---|
1432 | ELSE |
---|
1433 | kp = INT( tmp_particle%z / dz(1) + 1 + offset_ocean_nzt ) |
---|
1434 | ENDIF |
---|
1435 | ! |
---|
1436 | !-- Determine surface level. Therefore, check for upward-facing wall on |
---|
1437 | !-- w-grid. |
---|
1438 | k_surf = topo_top_ind(jp,ip,3) |
---|
1439 | IF ( seed_follows_topography ) THEN |
---|
1440 | ! |
---|
1441 | !-- Particle height is given relative to topography |
---|
1442 | kp = kp + k_surf |
---|
1443 | tmp_particle%z = tmp_particle%z + zw(k_surf) |
---|
1444 | ! |
---|
1445 | !-- Skip particle release if particle position is above model top, or |
---|
1446 | !-- within topography in case of overhanging structures. |
---|
1447 | IF ( kp > nzt .OR. & |
---|
1448 | .NOT. BTEST( wall_flags_total_0(kp,jp,ip), 0 ) ) THEN |
---|
1449 | pos_x = pos_x + pdx(i) |
---|
1450 | CYCLE xloop |
---|
1451 | ENDIF |
---|
1452 | ! |
---|
1453 | !-- Skip particle release if particle position is below surface, or |
---|
1454 | !-- within topography in case of overhanging structures. |
---|
1455 | ELSEIF ( .NOT. seed_follows_topography .AND. & |
---|
1456 | tmp_particle%z <= zw(k_surf) .OR. & |
---|
1457 | .NOT. BTEST( wall_flags_total_0(kp,jp,ip), 0 ) ) THEN |
---|
1458 | pos_x = pos_x + pdx(i) |
---|
1459 | CYCLE xloop |
---|
1460 | ENDIF |
---|
1461 | |
---|
1462 | local_count(kp,jp,ip) = local_count(kp,jp,ip) + 1 |
---|
1463 | |
---|
1464 | IF ( .NOT. first_stride ) THEN |
---|
1465 | IF ( ip < nxl .OR. jp < nys .OR. kp < nzb+1 ) THEN |
---|
1466 | write(6,*) 'xl ',ip,jp,kp,nxl,nys,nzb+1 |
---|
1467 | ENDIF |
---|
1468 | IF ( ip > nxr .OR. jp > nyn .OR. kp > nzt ) THEN |
---|
1469 | write(6,*) 'xu ',ip,jp,kp,nxr,nyn,nzt |
---|
1470 | ENDIF |
---|
1471 | grid_particles(kp,jp,ip)%particles(local_count(kp,jp,ip)) = & |
---|
1472 | tmp_particle |
---|
1473 | ENDIF |
---|
1474 | ENDDO |
---|
1475 | ENDIF |
---|
1476 | pos_x = pos_x + pdx(i) |
---|
1477 | ENDDO xloop |
---|
1478 | ENDIF |
---|
1479 | pos_y = pos_y + pdy(i) |
---|
1480 | ENDDO |
---|
1481 | ENDIF |
---|
1482 | |
---|
1483 | pos_z = pos_z + pdz(i) |
---|
1484 | ENDDO |
---|
1485 | ENDDO |
---|
1486 | |
---|
1487 | IF ( first_stride ) THEN |
---|
1488 | DO ip = nxl, nxr |
---|
1489 | DO jp = nys, nyn |
---|
1490 | DO kp = nzb+1, nzt |
---|
1491 | IF ( phase == phase_init ) THEN |
---|
1492 | IF ( local_count(kp,jp,ip) > 0 ) THEN |
---|
1493 | alloc_size = MAX( INT( local_count(kp,jp,ip) * & |
---|
1494 | ( 1.0_wp + alloc_factor / 100.0_wp ) ), 1 ) |
---|
1495 | ELSE |
---|
1496 | alloc_size = 1 |
---|
1497 | ENDIF |
---|
1498 | ALLOCATE(grid_particles(kp,jp,ip)%particles(1:alloc_size)) |
---|
1499 | DO n = 1, alloc_size |
---|
1500 | grid_particles(kp,jp,ip)%particles(n) = zero_particle |
---|
1501 | ENDDO |
---|
1502 | ELSEIF ( phase == phase_release ) THEN |
---|
1503 | IF ( local_count(kp,jp,ip) > 0 ) THEN |
---|
1504 | new_size = local_count(kp,jp,ip) + prt_count(kp,jp,ip) |
---|
1505 | alloc_size = MAX( INT( new_size * ( 1.0_wp + & |
---|
1506 | alloc_factor / 100.0_wp ) ), 1 ) |
---|
1507 | IF( alloc_size > SIZE( grid_particles(kp,jp,ip)%particles) ) THEN |
---|
1508 | CALL realloc_particles_array( ip, jp, kp, alloc_size ) |
---|
1509 | ENDIF |
---|
1510 | ENDIF |
---|
1511 | ENDIF |
---|
1512 | ENDDO |
---|
1513 | ENDDO |
---|
1514 | ENDDO |
---|
1515 | ENDIF |
---|
1516 | |
---|
1517 | ENDDO |
---|
1518 | |
---|
1519 | local_start = prt_count+1 |
---|
1520 | prt_count = local_count |
---|
1521 | ! |
---|
1522 | !-- Calculate particle IDs |
---|
1523 | DO ip = nxl, nxr |
---|
1524 | DO jp = nys, nyn |
---|
1525 | DO kp = nzb+1, nzt |
---|
1526 | number_of_particles = prt_count(kp,jp,ip) |
---|
1527 | IF ( number_of_particles <= 0 ) CYCLE |
---|
1528 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
1529 | |
---|
1530 | DO n = local_start(kp,jp,ip), number_of_particles !only new particles |
---|
1531 | |
---|
1532 | particles(n)%id = 10000_idp**3 * grid_particles(kp,jp,ip)%id_counter + & |
---|
1533 | 10000_idp**2 * kp + 10000_idp * jp + ip |
---|
1534 | ! |
---|
1535 | !-- Count the number of particles that have been released before |
---|
1536 | grid_particles(kp,jp,ip)%id_counter = grid_particles(kp,jp,ip)%id_counter + 1 |
---|
1537 | |
---|
1538 | ENDDO |
---|
1539 | |
---|
1540 | ENDDO |
---|
1541 | ENDDO |
---|
1542 | ENDDO |
---|
1543 | ! |
---|
1544 | !-- Initialize aerosol background spectrum |
---|
1545 | IF ( curvature_solution_effects ) THEN |
---|
1546 | CALL lpm_init_aerosols( local_start ) |
---|
1547 | ENDIF |
---|
1548 | ! |
---|
1549 | !-- Add random fluctuation to particle positions. |
---|
1550 | IF ( random_start_position ) THEN |
---|
1551 | DO ip = nxl, nxr |
---|
1552 | DO jp = nys, nyn |
---|
1553 | ! |
---|
1554 | !-- Put the random seeds at grid point jp, ip |
---|
1555 | CALL random_seed_parallel( put=seq_random_array_particles(:,jp,ip) ) |
---|
1556 | DO kp = nzb+1, nzt |
---|
1557 | number_of_particles = prt_count(kp,jp,ip) |
---|
1558 | IF ( number_of_particles <= 0 ) CYCLE |
---|
1559 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
1560 | ! |
---|
1561 | !-- Move only new particles. Moreover, limit random fluctuation in order to prevent that |
---|
1562 | !-- particles move more than one grid box, which would lead to problems concerning |
---|
1563 | !-- particle exchange between processors in case pdx/pdy are larger than dx/dy, |
---|
1564 | !-- respectively. |
---|
1565 | DO n = local_start(kp,jp,ip), number_of_particles |
---|
1566 | IF ( psl(particles(n)%group) /= psr(particles(n)%group) ) THEN |
---|
1567 | CALL random_number_parallel( random_dummy ) |
---|
1568 | rand_contr = ( random_dummy - 0.5_wp ) * pdx(particles(n)%group) |
---|
1569 | particles(n)%x = particles(n)%x + & |
---|
1570 | MERGE( rand_contr, SIGN( dx, rand_contr ), & |
---|
1571 | ABS( rand_contr ) < dx & |
---|
1572 | ) |
---|
1573 | ENDIF |
---|
1574 | IF ( pss(particles(n)%group) /= psn(particles(n)%group) ) THEN |
---|
1575 | CALL random_number_parallel( random_dummy ) |
---|
1576 | rand_contr = ( random_dummy - 0.5_wp ) * pdy(particles(n)%group) |
---|
1577 | particles(n)%y = particles(n)%y + & |
---|
1578 | MERGE( rand_contr, SIGN( dy, rand_contr ), & |
---|
1579 | ABS( rand_contr ) < dy & |
---|
1580 | ) |
---|
1581 | ENDIF |
---|
1582 | IF ( psb(particles(n)%group) /= pst(particles(n)%group) ) THEN |
---|
1583 | CALL random_number_parallel( random_dummy ) |
---|
1584 | rand_contr = ( random_dummy - 0.5_wp ) * pdz(particles(n)%group) |
---|
1585 | particles(n)%z = particles(n)%z + & |
---|
1586 | MERGE( rand_contr, SIGN( dzw(kp), rand_contr ), & |
---|
1587 | ABS( rand_contr ) < dzw(kp) & |
---|
1588 | ) |
---|
1589 | ENDIF |
---|
1590 | ENDDO |
---|
1591 | ! |
---|
1592 | !-- Identify particles located outside the model domain and reflect or absorb them if |
---|
1593 | !-- necessary. |
---|
1594 | CALL lpm_boundary_conds( 'bottom/top', i, j, k ) |
---|
1595 | ! |
---|
1596 | !-- Furthermore, remove particles located in topography. Note, as |
---|
1597 | !-- the particle speed is still zero at this point, wall |
---|
1598 | !-- reflection boundary conditions will not work in this case. |
---|
1599 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
1600 | DO n = local_start(kp,jp,ip), number_of_particles |
---|
1601 | i = particles(n)%x * ddx |
---|
1602 | j = particles(n)%y * ddy |
---|
1603 | k = particles(n)%z / dz(1) + 1 + offset_ocean_nzt |
---|
1604 | DO WHILE( zw(k) < particles(n)%z ) |
---|
1605 | k = k + 1 |
---|
1606 | ENDDO |
---|
1607 | DO WHILE( zw(k-1) > particles(n)%z ) |
---|
1608 | k = k - 1 |
---|
1609 | ENDDO |
---|
1610 | ! |
---|
1611 | !-- Check if particle is within topography |
---|
1612 | IF ( .NOT. BTEST( wall_flags_total_0(k,j,i), 0 ) ) THEN |
---|
1613 | particles(n)%particle_mask = .FALSE. |
---|
1614 | deleted_particles = deleted_particles + 1 |
---|
1615 | ENDIF |
---|
1616 | |
---|
1617 | ENDDO |
---|
1618 | ENDDO |
---|
1619 | ! |
---|
1620 | !-- Get the new random seeds from last call at grid point jp, ip |
---|
1621 | CALL random_seed_parallel( get=seq_random_array_particles(:,jp,ip) ) |
---|
1622 | ENDDO |
---|
1623 | ENDDO |
---|
1624 | ! |
---|
1625 | !-- Exchange particles between grid cells and processors |
---|
1626 | CALL lpm_move_particle |
---|
1627 | CALL lpm_exchange_horiz |
---|
1628 | |
---|
1629 | ENDIF |
---|
1630 | ! |
---|
1631 | !-- In case of random_start_position, delete particles identified by lpm_exchange_horiz and |
---|
1632 | !-- lpm_boundary_conds. Then sort particles into blocks, which is needed for a fast interpolation of |
---|
1633 | !-- the LES fields on the particle position. |
---|
1634 | CALL lpm_sort_and_delete |
---|
1635 | ! |
---|
1636 | !-- Determine the current number of particles |
---|
1637 | DO ip = nxl, nxr |
---|
1638 | DO jp = nys, nyn |
---|
1639 | DO kp = nzb+1, nzt |
---|
1640 | number_of_particles = number_of_particles + prt_count(kp,jp,ip) |
---|
1641 | ENDDO |
---|
1642 | ENDDO |
---|
1643 | ENDDO |
---|
1644 | ! |
---|
1645 | !-- Calculate the number of particles of the total domain |
---|
1646 | #if defined( __parallel ) |
---|
1647 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
1648 | CALL MPI_ALLREDUCE( number_of_particles, total_number_of_particles, 1, MPI_INTEGER, MPI_SUM, & |
---|
1649 | comm2d, ierr ) |
---|
1650 | #else |
---|
1651 | total_number_of_particles = number_of_particles |
---|
1652 | #endif |
---|
1653 | |
---|
1654 | RETURN |
---|
1655 | |
---|
1656 | END SUBROUTINE lpm_create_particle |
---|
1657 | |
---|
1658 | |
---|
1659 | !--------------------------------------------------------------------------------------------------! |
---|
1660 | ! Description: |
---|
1661 | ! ------------ |
---|
1662 | !> This routine initializes the particles as aerosols with physio-chemical properties. |
---|
1663 | !--------------------------------------------------------------------------------------------------! |
---|
1664 | SUBROUTINE lpm_init_aerosols(local_start) |
---|
1665 | |
---|
1666 | INTEGER(iwp) :: ip !< |
---|
1667 | INTEGER(iwp) :: jp !< |
---|
1668 | INTEGER(iwp) :: kp !< |
---|
1669 | INTEGER(iwp) :: n !< |
---|
1670 | |
---|
1671 | INTEGER(iwp), DIMENSION(nzb:nzt+1,nysg:nyng,nxlg:nxrg), INTENT(IN) :: local_start !< |
---|
1672 | |
---|
1673 | REAL(wp) :: afactor !< curvature effects |
---|
1674 | REAL(wp) :: bfactor !< solute effects |
---|
1675 | REAL(wp) :: dlogr !< logarithmic width of radius bin |
---|
1676 | REAL(wp) :: e_a !< vapor pressure |
---|
1677 | REAL(wp) :: e_s !< saturation vapor pressure |
---|
1678 | REAL(wp) :: rmax = 10.0e-6_wp !< maximum aerosol radius |
---|
1679 | REAL(wp) :: rmin = 0.005e-6_wp !< minimum aerosol radius |
---|
1680 | REAL(wp) :: r_l !< left radius of bin |
---|
1681 | REAL(wp) :: r_mid !< mean radius of bin |
---|
1682 | REAL(wp) :: r_r !< right radius of bin |
---|
1683 | REAL(wp) :: sigma !< surface tension |
---|
1684 | REAL(wp) :: t_int !< temperature |
---|
1685 | |
---|
1686 | |
---|
1687 | ! |
---|
1688 | !-- Set constants for different aerosol species |
---|
1689 | IF ( TRIM( aero_species ) == 'nacl' ) THEN |
---|
1690 | molecular_weight_of_solute = 0.05844_wp |
---|
1691 | rho_s = 2165.0_wp |
---|
1692 | vanthoff = 2.0_wp |
---|
1693 | ELSEIF ( TRIM( aero_species ) == 'c3h4o4' ) THEN |
---|
1694 | molecular_weight_of_solute = 0.10406_wp |
---|
1695 | rho_s = 1600.0_wp |
---|
1696 | vanthoff = 1.37_wp |
---|
1697 | ELSEIF ( TRIM( aero_species ) == 'nh4o3' ) THEN |
---|
1698 | molecular_weight_of_solute = 0.08004_wp |
---|
1699 | rho_s = 1720.0_wp |
---|
1700 | vanthoff = 2.31_wp |
---|
1701 | ELSE |
---|
1702 | WRITE( message_string, * ) 'unknown aerosol species ', & |
---|
1703 | 'aero_species = "', TRIM( aero_species ), '"' |
---|
1704 | CALL message( 'lpm_init', 'PA0470', 1, 2, 0, 6, 0 ) |
---|
1705 | ENDIF |
---|
1706 | ! |
---|
1707 | !-- The following typical aerosol spectra are taken from Jaenicke (1993): |
---|
1708 | !-- Tropospheric aerosols. Published in Aerosol-Cloud-Climate Interactions. |
---|
1709 | IF ( TRIM( aero_type ) == 'polar' ) THEN |
---|
1710 | na = (/ 2.17e1, 1.86e-1, 3.04e-4 /) * 1.0E6_wp |
---|
1711 | rm = (/ 0.0689, 0.375, 4.29 /) * 1.0E-6_wp |
---|
1712 | log_sigma = (/ 0.245, 0.300, 0.291 /) |
---|
1713 | ELSEIF ( TRIM( aero_type ) == 'background' ) THEN |
---|
1714 | na = (/ 1.29e2, 5.97e1, 6.35e1 /) * 1.0E6_wp |
---|
1715 | rm = (/ 0.0036, 0.127, 0.259 /) * 1.0E-6_wp |
---|
1716 | log_sigma = (/ 0.645, 0.253, 0.425 /) |
---|
1717 | ELSEIF ( TRIM( aero_type ) == 'maritime' ) THEN |
---|
1718 | na = (/ 1.33e2, 6.66e1, 3.06e0 /) * 1.0E6_wp |
---|
1719 | rm = (/ 0.0039, 0.133, 0.29 /) * 1.0E-6_wp |
---|
1720 | log_sigma = (/ 0.657, 0.210, 0.396 /) |
---|
1721 | ELSEIF ( TRIM( aero_type ) == 'continental' ) THEN |
---|
1722 | na = (/ 3.20e3, 2.90e3, 3.00e-1 /) * 1.0E6_wp |
---|
1723 | rm = (/ 0.01, 0.058, 0.9 /) * 1.0E-6_wp |
---|
1724 | log_sigma = (/ 0.161, 0.217, 0.380 /) |
---|
1725 | ELSEIF ( TRIM( aero_type ) == 'desert' ) THEN |
---|
1726 | na = (/ 7.26e2, 1.14e3, 1.78e-1 /) * 1.0E6_wp |
---|
1727 | rm = (/ 0.001, 0.0188, 10.8 /) * 1.0E-6_wp |
---|
1728 | log_sigma = (/ 0.247, 0.770, 0.438 /) |
---|
1729 | ELSEIF ( TRIM( aero_type ) == 'rural' ) THEN |
---|
1730 | na = (/ 6.65e3, 1.47e2, 1.99e3 /) * 1.0E6_wp |
---|
1731 | rm = (/ 0.00739, 0.0269, 0.0419 /) * 1.0E-6_wp |
---|
1732 | log_sigma = (/ 0.225, 0.557, 0.266 /) |
---|
1733 | ELSEIF ( TRIM( aero_type ) == 'urban' ) THEN |
---|
1734 | na = (/ 9.93e4, 1.11e3, 3.64e4 /) * 1.0E6_wp |
---|
1735 | rm = (/ 0.00651, 0.00714, 0.0248 /) * 1.0E-6_wp |
---|
1736 | log_sigma = (/ 0.245, 0.666, 0.337 /) |
---|
1737 | ELSEIF ( TRIM( aero_type ) == 'user' ) THEN |
---|
1738 | CONTINUE |
---|
1739 | ELSE |
---|
1740 | WRITE( message_string, * ) 'unknown aerosol type ', & |
---|
1741 | 'aero_type = "', TRIM( aero_type ), '"' |
---|
1742 | CALL message( 'lpm_init', 'PA0459', 1, 2, 0, 6, 0 ) |
---|
1743 | ENDIF |
---|
1744 | |
---|
1745 | DO ip = nxl, nxr |
---|
1746 | DO jp = nys, nyn |
---|
1747 | ! |
---|
1748 | !-- Put the random seeds at grid point jp, ip |
---|
1749 | CALL random_seed_parallel( put=seq_random_array_particles(:,jp,ip) ) |
---|
1750 | DO kp = nzb+1, nzt |
---|
1751 | |
---|
1752 | number_of_particles = prt_count(kp,jp,ip) |
---|
1753 | IF ( number_of_particles <= 0 ) CYCLE |
---|
1754 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
1755 | |
---|
1756 | dlogr = ( LOG10( rmax ) - LOG10( rmin ) ) / & |
---|
1757 | ( number_of_particles - local_start(kp,jp,ip) + 1 ) |
---|
1758 | ! |
---|
1759 | !-- Initialize the aerosols with a predefined spectral distribution of the dry radius |
---|
1760 | !-- (logarithmically increasing bins) and a varying weighting factor. |
---|
1761 | DO n = local_start(kp,jp,ip), number_of_particles !only new particles |
---|
1762 | |
---|
1763 | r_l = 10.0**( LOG10( rmin ) + (n-1) * dlogr ) |
---|
1764 | r_r = 10.0**( LOG10( rmin ) + n * dlogr ) |
---|
1765 | r_mid = SQRT( r_l * r_r ) |
---|
1766 | |
---|
1767 | particles(n)%aux1 = r_mid |
---|
1768 | particles(n)%weight_factor = & |
---|
1769 | ( na(1) / ( SQRT( 2.0_wp * pi ) * log_sigma(1) ) * & |
---|
1770 | EXP( - LOG10( r_mid / rm(1) )**2 / ( 2.0_wp * log_sigma(1)**2 ) ) + & |
---|
1771 | na(2) / ( SQRT( 2.0_wp * pi ) * log_sigma(2) ) * & |
---|
1772 | EXP( - LOG10( r_mid / rm(2) )**2 / ( 2.0_wp * log_sigma(2)**2 ) ) + & |
---|
1773 | na(3) / ( SQRT( 2.0_wp * pi ) * log_sigma(3) ) * & |
---|
1774 | EXP( - LOG10( r_mid / rm(3) )**2 / ( 2.0_wp * log_sigma(3)**2 ) ) & |
---|
1775 | ) * ( LOG10( r_r ) - LOG10( r_l ) ) * ( dx * dy * dzw(kp) ) |
---|
1776 | |
---|
1777 | ! |
---|
1778 | !-- Multiply weight_factor with the namelist parameter aero_weight to increase or |
---|
1779 | !-- decrease the number of simulated aerosols |
---|
1780 | particles(n)%weight_factor = particles(n)%weight_factor * aero_weight |
---|
1781 | ! |
---|
1782 | !-- Create random numver with parallel number generator |
---|
1783 | CALL random_number_parallel( random_dummy ) |
---|
1784 | IF ( particles(n)%weight_factor - FLOOR( particles(n)%weight_factor, KIND=wp ) & |
---|
1785 | > random_dummy ) THEN |
---|
1786 | particles(n)%weight_factor = FLOOR( particles(n)%weight_factor, KIND=wp ) & |
---|
1787 | + 1.0_wp |
---|
1788 | ELSE |
---|
1789 | particles(n)%weight_factor = FLOOR( particles(n)%weight_factor, KIND=wp ) |
---|
1790 | ENDIF |
---|
1791 | ! |
---|
1792 | !-- Unnecessary particles will be deleted |
---|
1793 | IF ( particles(n)%weight_factor <= 0.0_wp ) particles(n)%particle_mask = .FALSE. |
---|
1794 | |
---|
1795 | ENDDO |
---|
1796 | ! |
---|
1797 | !-- Set particle radius to equilibrium radius based on the environmental supersaturation |
---|
1798 | !-- (Khvorostyanov and Curry, 2007, JGR). This avoids the sometimes lengthy growth toward |
---|
1799 | !-- their equilibrium radius within the simulation. |
---|
1800 | t_int = pt(kp,jp,ip) * exner(kp) |
---|
1801 | |
---|
1802 | e_s = magnus( t_int ) |
---|
1803 | e_a = q(kp,jp,ip) * hyp(kp) / ( q(kp,jp,ip) + rd_d_rv ) |
---|
1804 | |
---|
1805 | sigma = 0.0761_wp - 0.000155_wp * ( t_int - 273.15_wp ) |
---|
1806 | afactor = 2.0_wp * sigma / ( rho_l * r_v * t_int ) |
---|
1807 | |
---|
1808 | bfactor = vanthoff * molecular_weight_of_water * & |
---|
1809 | rho_s / ( molecular_weight_of_solute * rho_l ) |
---|
1810 | ! |
---|
1811 | !-- The formula is only valid for subsaturated environments. For supersaturations higher |
---|
1812 | !-- than -5 %, the supersaturation is set to -5%. |
---|
1813 | IF ( e_a / e_s >= 0.95_wp ) e_a = 0.95_wp * e_s |
---|
1814 | |
---|
1815 | DO n = local_start(kp,jp,ip), number_of_particles !only new particles |
---|
1816 | ! |
---|
1817 | !-- For details on this equation, see Eq. (14) of Khvorostyanov and |
---|
1818 | !-- Curry (2007, JGR) |
---|
1819 | particles(n)%radius = bfactor**0.3333333_wp * & |
---|
1820 | particles(n)%aux1 / ( 1.0_wp - e_a / e_s )**0.3333333_wp / & |
---|
1821 | ( 1.0_wp + ( afactor / ( 3.0_wp * bfactor**0.3333333_wp * & |
---|
1822 | particles(n)%aux1 ) ) / & |
---|
1823 | ( 1.0_wp - e_a / e_s )**0.6666666_wp & |
---|
1824 | ) |
---|
1825 | |
---|
1826 | ENDDO |
---|
1827 | |
---|
1828 | ENDDO |
---|
1829 | ! |
---|
1830 | !-- Get the new random seeds from last call at grid point j |
---|
1831 | CALL random_seed_parallel( get=seq_random_array_particles(:,jp,ip) ) |
---|
1832 | ENDDO |
---|
1833 | ENDDO |
---|
1834 | |
---|
1835 | END SUBROUTINE lpm_init_aerosols |
---|
1836 | |
---|
1837 | |
---|
1838 | !--------------------------------------------------------------------------------------------------! |
---|
1839 | ! Description: |
---|
1840 | ! ------------ |
---|
1841 | !> Calculates quantities required for considering the SGS velocity fluctuations in the particle |
---|
1842 | !> transport by a stochastic approach. The respective quantities are: SGS-TKE gradients and |
---|
1843 | !> horizontally averaged profiles of the SGS TKE and the resolved-scale velocity variances. |
---|
1844 | !--------------------------------------------------------------------------------------------------! |
---|
1845 | SUBROUTINE lpm_init_sgs_tke |
---|
1846 | |
---|
1847 | USE exchange_horiz_mod, & |
---|
1848 | ONLY: exchange_horiz |
---|
1849 | |
---|
1850 | USE statistics, & |
---|
1851 | ONLY: flow_statistics_called, hom, sums, sums_l |
---|
1852 | |
---|
1853 | INTEGER(iwp) :: i !< index variable along x |
---|
1854 | INTEGER(iwp) :: j !< index variable along y |
---|
1855 | INTEGER(iwp) :: k !< index variable along z |
---|
1856 | INTEGER(iwp) :: m !< running index for the surface elements |
---|
1857 | |
---|
1858 | REAL(wp) :: flag1 !< flag to mask topography |
---|
1859 | |
---|
1860 | ! |
---|
1861 | !-- TKE gradient along x and y |
---|
1862 | DO i = nxl, nxr |
---|
1863 | DO j = nys, nyn |
---|
1864 | DO k = nzb, nzt+1 |
---|
1865 | |
---|
1866 | IF ( .NOT. BTEST( wall_flags_total_0(k,j,i-1), 0 ) .AND. & |
---|
1867 | BTEST( wall_flags_total_0(k,j,i), 0 ) .AND. & |
---|
1868 | BTEST( wall_flags_total_0(k,j,i+1), 0 ) ) & |
---|
1869 | THEN |
---|
1870 | de_dx(k,j,i) = 2.0_wp * sgs_wf_part * ( e(k,j,i+1) - e(k,j,i) ) * ddx |
---|
1871 | ELSEIF ( BTEST( wall_flags_total_0(k,j,i-1), 0 ) .AND. & |
---|
1872 | BTEST( wall_flags_total_0(k,j,i), 0 ) .AND. & |
---|
1873 | .NOT. BTEST( wall_flags_total_0(k,j,i+1), 0 ) ) & |
---|
1874 | THEN |
---|
1875 | de_dx(k,j,i) = 2.0_wp * sgs_wf_part * ( e(k,j,i) - e(k,j,i-1) ) * ddx |
---|
1876 | ELSEIF ( .NOT. BTEST( wall_flags_total_0(k,j,i), 22 ) .AND. & |
---|
1877 | .NOT. BTEST( wall_flags_total_0(k,j,i+1), 22 ) ) & |
---|
1878 | THEN |
---|
1879 | de_dx(k,j,i) = 0.0_wp |
---|
1880 | ELSEIF ( .NOT. BTEST( wall_flags_total_0(k,j,i-1), 22 ) .AND. & |
---|
1881 | .NOT. BTEST( wall_flags_total_0(k,j,i), 22 ) ) & |
---|
1882 | THEN |
---|
1883 | de_dx(k,j,i) = 0.0_wp |
---|
1884 | ELSE |
---|
1885 | de_dx(k,j,i) = sgs_wf_part * ( e(k,j,i+1) - e(k,j,i-1) ) * ddx |
---|
1886 | ENDIF |
---|
1887 | |
---|
1888 | IF ( .NOT. BTEST( wall_flags_total_0(k,j-1,i), 0 ) .AND. & |
---|
1889 | BTEST( wall_flags_total_0(k,j,i), 0 ) .AND. & |
---|
1890 | BTEST( wall_flags_total_0(k,j+1,i), 0 ) ) & |
---|
1891 | THEN |
---|
1892 | de_dy(k,j,i) = 2.0_wp * sgs_wf_part * ( e(k,j+1,i) - e(k,j,i) ) * ddy |
---|
1893 | ELSEIF ( BTEST( wall_flags_total_0(k,j-1,i), 0 ) .AND. & |
---|
1894 | BTEST( wall_flags_total_0(k,j,i), 0 ) .AND. & |
---|
1895 | .NOT. BTEST( wall_flags_total_0(k,j+1,i), 0 ) ) & |
---|
1896 | THEN |
---|
1897 | de_dy(k,j,i) = 2.0_wp * sgs_wf_part * ( e(k,j,i) - e(k,j-1,i) ) * ddy |
---|
1898 | ELSEIF ( .NOT. BTEST( wall_flags_total_0(k,j,i), 22 ) .AND. & |
---|
1899 | .NOT. BTEST( wall_flags_total_0(k,j+1,i), 22 ) ) & |
---|
1900 | THEN |
---|
1901 | de_dy(k,j,i) = 0.0_wp |
---|
1902 | ELSEIF ( .NOT. BTEST( wall_flags_total_0(k,j-1,i), 22 ) .AND. & |
---|
1903 | .NOT. BTEST( wall_flags_total_0(k,j,i), 22 ) ) & |
---|
1904 | THEN |
---|
1905 | de_dy(k,j,i) = 0.0_wp |
---|
1906 | ELSE |
---|
1907 | de_dy(k,j,i) = sgs_wf_part * ( e(k,j+1,i) - e(k,j-1,i) ) * ddy |
---|
1908 | ENDIF |
---|
1909 | |
---|
1910 | ENDDO |
---|
1911 | ENDDO |
---|
1912 | ENDDO |
---|
1913 | |
---|
1914 | ! |
---|
1915 | !-- TKE gradient along z at topograhy and including bottom and top boundary conditions |
---|
1916 | DO i = nxl, nxr |
---|
1917 | DO j = nys, nyn |
---|
1918 | DO k = nzb+1, nzt-1 |
---|
1919 | ! |
---|
1920 | !-- Flag to mask topography |
---|
1921 | flag1 = MERGE( 1.0_wp, 0.0_wp, BTEST( wall_flags_total_0(k,j,i), 0 ) ) |
---|
1922 | |
---|
1923 | de_dz(k,j,i) = 2.0_wp * sgs_wf_part * & |
---|
1924 | ( e(k+1,j,i) - e(k-1,j,i) ) / ( zu(k+1) - zu(k-1) ) * flag1 |
---|
1925 | ENDDO |
---|
1926 | ! |
---|
1927 | !-- upward-facing surfaces |
---|
1928 | DO m = bc_h(0)%start_index(j,i), bc_h(0)%end_index(j,i) |
---|
1929 | k = bc_h(0)%k(m) |
---|
1930 | de_dz(k,j,i) = 2.0_wp * sgs_wf_part * ( e(k+1,j,i) - e(k,j,i) ) / ( zu(k+1) - zu(k) ) |
---|
1931 | ENDDO |
---|
1932 | ! |
---|
1933 | !-- downward-facing surfaces |
---|
1934 | DO m = bc_h(1)%start_index(j,i), bc_h(1)%end_index(j,i) |
---|
1935 | k = bc_h(1)%k(m) |
---|
1936 | de_dz(k,j,i) = 2.0_wp * sgs_wf_part * ( e(k,j,i) - e(k-1,j,i) ) / ( zu(k) - zu(k-1) ) |
---|
1937 | ENDDO |
---|
1938 | |
---|
1939 | de_dz(nzb,j,i) = 0.0_wp |
---|
1940 | de_dz(nzt,j,i) = 0.0_wp |
---|
1941 | de_dz(nzt+1,j,i) = 0.0_wp |
---|
1942 | ENDDO |
---|
1943 | ENDDO |
---|
1944 | ! |
---|
1945 | !-- Ghost point exchange |
---|
1946 | CALL exchange_horiz( de_dx, nbgp ) |
---|
1947 | CALL exchange_horiz( de_dy, nbgp ) |
---|
1948 | CALL exchange_horiz( de_dz, nbgp ) |
---|
1949 | CALL exchange_horiz( diss, nbgp ) |
---|
1950 | ! |
---|
1951 | !-- Set boundary conditions at non-periodic boundaries. Note, at non-period boundaries zero-gradient |
---|
1952 | !-- boundary conditions are set for the subgrid TKE. |
---|
1953 | !-- Thus, TKE gradients normal to the respective lateral boundaries are zero, |
---|
1954 | !-- while tangetial TKE gradients then must be the same as within the prognostic domain. |
---|
1955 | IF ( bc_dirichlet_l ) THEN |
---|
1956 | de_dx(:,:,-1) = 0.0_wp |
---|
1957 | de_dy(:,:,-1) = de_dy(:,:,0) |
---|
1958 | de_dz(:,:,-1) = de_dz(:,:,0) |
---|
1959 | ENDIF |
---|
1960 | IF ( bc_dirichlet_r ) THEN |
---|
1961 | de_dx(:,:,nxr+1) = 0.0_wp |
---|
1962 | de_dy(:,:,nxr+1) = de_dy(:,:,nxr) |
---|
1963 | de_dz(:,:,nxr+1) = de_dz(:,:,nxr) |
---|
1964 | ENDIF |
---|
1965 | IF ( bc_dirichlet_n ) THEN |
---|
1966 | de_dx(:,nyn+1,:) = de_dx(:,nyn,:) |
---|
1967 | de_dy(:,nyn+1,:) = 0.0_wp |
---|
1968 | de_dz(:,nyn+1,:) = de_dz(:,nyn,:) |
---|
1969 | ENDIF |
---|
1970 | IF ( bc_dirichlet_s ) THEN |
---|
1971 | de_dx(:,nys-1,:) = de_dx(:,nys,:) |
---|
1972 | de_dy(:,nys-1,:) = 0.0_wp |
---|
1973 | de_dz(:,nys-1,:) = de_dz(:,nys,:) |
---|
1974 | ENDIF |
---|
1975 | ! |
---|
1976 | !-- Calculate the horizontally averaged profiles of SGS TKE and resolved velocity variances (they |
---|
1977 | !-- may have been already calculated in routine flow_statistics). |
---|
1978 | IF ( .NOT. flow_statistics_called ) THEN |
---|
1979 | |
---|
1980 | ! |
---|
1981 | !-- First calculate horizontally averaged profiles of the horizontal velocities. |
---|
1982 | sums_l(:,1,0) = 0.0_wp |
---|
1983 | sums_l(:,2,0) = 0.0_wp |
---|
1984 | |
---|
1985 | DO i = nxl, nxr |
---|
1986 | DO j = nys, nyn |
---|
1987 | DO k = nzb, nzt+1 |
---|
1988 | ! |
---|
1989 | !-- Flag indicating vicinity of wall |
---|
1990 | flag1 = MERGE( 1.0_wp, 0.0_wp, BTEST( wall_flags_total_0(k,j,i), 24 ) ) |
---|
1991 | |
---|
1992 | sums_l(k,1,0) = sums_l(k,1,0) + u(k,j,i) * flag1 |
---|
1993 | sums_l(k,2,0) = sums_l(k,2,0) + v(k,j,i) * flag1 |
---|
1994 | ENDDO |
---|
1995 | ENDDO |
---|
1996 | ENDDO |
---|
1997 | |
---|
1998 | #if defined( __parallel ) |
---|
1999 | ! |
---|
2000 | !-- Compute total sum from local sums |
---|
2001 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
2002 | CALL MPI_ALLREDUCE( sums_l(nzb,1,0), sums(nzb,1), nzt+2-nzb, MPI_REAL, MPI_SUM, comm2d, & |
---|
2003 | ierr ) |
---|
2004 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
2005 | CALL MPI_ALLREDUCE( sums_l(nzb,2,0), sums(nzb,2), nzt+2-nzb, MPI_REAL, MPI_SUM, comm2d, & |
---|
2006 | ierr ) |
---|
2007 | #else |
---|
2008 | sums(:,1) = sums_l(:,1,0) |
---|
2009 | sums(:,2) = sums_l(:,2,0) |
---|
2010 | #endif |
---|
2011 | |
---|
2012 | ! |
---|
2013 | !-- Final values are obtained by division by the total number of grid points used for the |
---|
2014 | !-- summation. |
---|
2015 | hom(:,1,1,0) = sums(:,1) / ngp_2dh_outer(:,0) ! u |
---|
2016 | hom(:,1,2,0) = sums(:,2) / ngp_2dh_outer(:,0) ! v |
---|
2017 | |
---|
2018 | ! |
---|
2019 | !-- Now calculate the profiles of SGS TKE and the resolved-scale velocity variances |
---|
2020 | sums_l(:,8,0) = 0.0_wp |
---|
2021 | sums_l(:,30,0) = 0.0_wp |
---|
2022 | sums_l(:,31,0) = 0.0_wp |
---|
2023 | sums_l(:,32,0) = 0.0_wp |
---|
2024 | DO i = nxl, nxr |
---|
2025 | DO j = nys, nyn |
---|
2026 | DO k = nzb, nzt+1 |
---|
2027 | ! |
---|
2028 | !-- Flag indicating vicinity of wall |
---|
2029 | flag1 = MERGE( 1.0_wp, 0.0_wp, BTEST( wall_flags_total_0(k,j,i), 24 ) ) |
---|
2030 | |
---|
2031 | sums_l(k,8,0) = sums_l(k,8,0) + e(k,j,i) * flag1 |
---|
2032 | sums_l(k,30,0) = sums_l(k,30,0) + ( u(k,j,i) - hom(k,1,1,0) )**2 * flag1 |
---|
2033 | sums_l(k,31,0) = sums_l(k,31,0) + ( v(k,j,i) - hom(k,1,2,0) )**2 * flag1 |
---|
2034 | sums_l(k,32,0) = sums_l(k,32,0) + w(k,j,i)**2 * flag1 |
---|
2035 | ENDDO |
---|
2036 | ENDDO |
---|
2037 | ENDDO |
---|
2038 | |
---|
2039 | #if defined( __parallel ) |
---|
2040 | ! |
---|
2041 | !-- Compute total sum from local sums |
---|
2042 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
2043 | CALL MPI_ALLREDUCE( sums_l(nzb,8,0), sums(nzb,8), nzt+2-nzb, MPI_REAL, MPI_SUM, comm2d, & |
---|
2044 | ierr ) |
---|
2045 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
2046 | CALL MPI_ALLREDUCE( sums_l(nzb,30,0), sums(nzb,30), nzt+2-nzb, MPI_REAL, MPI_SUM, comm2d, & |
---|
2047 | ierr ) |
---|
2048 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
2049 | CALL MPI_ALLREDUCE( sums_l(nzb,31,0), sums(nzb,31), nzt+2-nzb, MPI_REAL, MPI_SUM, comm2d, & |
---|
2050 | ierr ) |
---|
2051 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
2052 | CALL MPI_ALLREDUCE( sums_l(nzb,32,0), sums(nzb,32), nzt+2-nzb, MPI_REAL, MPI_SUM, comm2d, & |
---|
2053 | ierr ) |
---|
2054 | |
---|
2055 | #else |
---|
2056 | sums(:,8) = sums_l(:,8,0) |
---|
2057 | sums(:,30) = sums_l(:,30,0) |
---|
2058 | sums(:,31) = sums_l(:,31,0) |
---|
2059 | sums(:,32) = sums_l(:,32,0) |
---|
2060 | #endif |
---|
2061 | |
---|
2062 | ! |
---|
2063 | !-- Final values are obtained by division by the total number of grid points used for the |
---|
2064 | !-- summation. |
---|
2065 | hom(:,1,8,0) = sums(:,8) / ngp_2dh_outer(:,0) ! e |
---|
2066 | hom(:,1,30,0) = sums(:,30) / ngp_2dh_outer(:,0) ! u*2 |
---|
2067 | hom(:,1,31,0) = sums(:,31) / ngp_2dh_outer(:,0) ! v*2 |
---|
2068 | hom(:,1,32,0) = sums(:,32) / ngp_2dh_outer(:,0) ! w*2 |
---|
2069 | |
---|
2070 | ENDIF |
---|
2071 | |
---|
2072 | END SUBROUTINE lpm_init_sgs_tke |
---|
2073 | |
---|
2074 | |
---|
2075 | !--------------------------------------------------------------------------------------------------! |
---|
2076 | ! Description: |
---|
2077 | ! ------------ |
---|
2078 | !> Sobroutine control lpm actions, i.e. all actions during one time step. |
---|
2079 | !--------------------------------------------------------------------------------------------------! |
---|
2080 | SUBROUTINE lpm_actions( location ) |
---|
2081 | |
---|
2082 | USE exchange_horiz_mod, & |
---|
2083 | ONLY: exchange_horiz |
---|
2084 | |
---|
2085 | CHARACTER (LEN=*), INTENT(IN) :: location !< call location string |
---|
2086 | |
---|
2087 | INTEGER(iwp) :: i !< |
---|
2088 | INTEGER(iwp) :: ie !< |
---|
2089 | INTEGER(iwp) :: is !< |
---|
2090 | INTEGER(iwp) :: j !< |
---|
2091 | INTEGER(iwp) :: je !< |
---|
2092 | INTEGER(iwp) :: js !< |
---|
2093 | INTEGER(iwp), SAVE :: lpm_count = 0 !< |
---|
2094 | INTEGER(iwp) :: k !< |
---|
2095 | INTEGER(iwp) :: ke !< |
---|
2096 | INTEGER(iwp) :: ks !< |
---|
2097 | INTEGER(iwp) :: m !< |
---|
2098 | INTEGER(iwp), SAVE :: steps = 0 !< |
---|
2099 | |
---|
2100 | LOGICAL :: first_loop_stride !< |
---|
2101 | |
---|
2102 | |
---|
2103 | SELECT CASE ( location ) |
---|
2104 | |
---|
2105 | CASE ( 'after_pressure_solver' ) |
---|
2106 | ! |
---|
2107 | !-- The particle model is executed if particle advection start is reached and only at the end |
---|
2108 | !-- of the intermediate time step loop. |
---|
2109 | IF ( time_since_reference_point >= particle_advection_start & |
---|
2110 | .AND. intermediate_timestep_count == intermediate_timestep_count_max ) & |
---|
2111 | THEN |
---|
2112 | CALL cpu_log( log_point(25), 'lpm', 'start' ) |
---|
2113 | ! |
---|
2114 | !-- Write particle data at current time on file. |
---|
2115 | !-- This has to be done here, before particles are further processed, because they may be |
---|
2116 | !-- deleted within this timestep (in case that dt_write_particle_data = dt_prel = |
---|
2117 | !-- particle_maximum_age). |
---|
2118 | time_write_particle_data = time_write_particle_data + dt_3d |
---|
2119 | IF ( time_write_particle_data >= dt_write_particle_data ) THEN |
---|
2120 | |
---|
2121 | CALL lpm_data_output_particles |
---|
2122 | ! |
---|
2123 | !-- The MOD function allows for changes in the output interval with restart runs. |
---|
2124 | time_write_particle_data = MOD( time_write_particle_data, & |
---|
2125 | MAX( dt_write_particle_data, dt_3d ) ) |
---|
2126 | ENDIF |
---|
2127 | |
---|
2128 | ! |
---|
2129 | !-- Initialize arrays for marking those particles to be deleted after the (sub-) timestep. |
---|
2130 | deleted_particles = 0 |
---|
2131 | |
---|
2132 | ! |
---|
2133 | !-- Initialize variables used for accumulating the number of particles exchanged between |
---|
2134 | !-- the subdomains during all sub-timesteps (if sgs velocities are included). These data |
---|
2135 | !-- are output further below on the particle statistics file. |
---|
2136 | trlp_count_sum = 0 |
---|
2137 | trlp_count_recv_sum = 0 |
---|
2138 | trrp_count_sum = 0 |
---|
2139 | trrp_count_recv_sum = 0 |
---|
2140 | trsp_count_sum = 0 |
---|
2141 | trsp_count_recv_sum = 0 |
---|
2142 | trnp_count_sum = 0 |
---|
2143 | trnp_count_recv_sum = 0 |
---|
2144 | ! |
---|
2145 | !-- Calculate exponential term used in case of particle inertia for each |
---|
2146 | !-- of the particle groups |
---|
2147 | DO m = 1, number_of_particle_groups |
---|
2148 | IF ( particle_groups(m)%density_ratio /= 0.0_wp ) THEN |
---|
2149 | particle_groups(m)%exp_arg = 4.5_wp * particle_groups(m)%density_ratio * & |
---|
2150 | molecular_viscosity / & |
---|
2151 | ( particle_groups(m)%radius )**2 |
---|
2152 | |
---|
2153 | particle_groups(m)%exp_term = EXP( -particle_groups(m)%exp_arg * dt_3d ) |
---|
2154 | ENDIF |
---|
2155 | ENDDO |
---|
2156 | ! |
---|
2157 | !-- If necessary, release new set of particles |
---|
2158 | IF ( ( simulated_time - last_particle_release_time ) >= dt_prel .AND. & |
---|
2159 | end_time_prel > simulated_time ) THEN |
---|
2160 | DO WHILE ( ( simulated_time - last_particle_release_time ) >= dt_prel ) |
---|
2161 | CALL lpm_create_particle( phase_release ) |
---|
2162 | last_particle_release_time = last_particle_release_time + dt_prel |
---|
2163 | ENDDO |
---|
2164 | ENDIF |
---|
2165 | ! |
---|
2166 | !-- Reset summation arrays |
---|
2167 | IF ( cloud_droplets ) THEN |
---|
2168 | ql_c = 0.0_wp |
---|
2169 | ql_v = 0.0_wp |
---|
2170 | ql_vp = 0.0_wp |
---|
2171 | ENDIF |
---|
2172 | |
---|
2173 | first_loop_stride = .TRUE. |
---|
2174 | grid_particles(:,:,:)%time_loop_done = .TRUE. |
---|
2175 | ! |
---|
2176 | !-- Timestep loop for particle advection. |
---|
2177 | !-- This loop has to be repeated until the advection time of every particle (within the |
---|
2178 | !-- total domain!) has reached the LES timestep (dt_3d). |
---|
2179 | !-- In case of including the SGS velocities, the particle timestep may be smaller than the |
---|
2180 | !-- LES timestep (because of the Lagrangian timescale restriction) and particles may |
---|
2181 | !-- require to undergo several particle timesteps, before the LES timestep is reached. |
---|
2182 | !-- Because the number of these particle timesteps to be carried out is unknown at first, |
---|
2183 | !-- these steps are carried out in the following infinite loop with exit condition. |
---|
2184 | DO |
---|
2185 | CALL cpu_log( log_point_s(44), 'lpm_advec', 'start' ) |
---|
2186 | CALL cpu_log( log_point_s(44), 'lpm_advec', 'pause' ) |
---|
2187 | |
---|
2188 | ! |
---|
2189 | !-- If particle advection includes SGS velocity components, calculate the required SGS |
---|
2190 | !-- quantities (i.e. gradients of the TKE, as well as horizontally averaged profiles of |
---|
2191 | !-- the SGS TKE and the resolved-scale velocity variances) |
---|
2192 | IF ( use_sgs_for_particles .AND. .NOT. cloud_droplets ) THEN |
---|
2193 | CALL lpm_init_sgs_tke |
---|
2194 | ENDIF |
---|
2195 | ! |
---|
2196 | !-- In case SGS-particle speed is considered, particles may carry out several particle |
---|
2197 | !-- timesteps. In order to prevent unnecessary treatment of particles that already |
---|
2198 | !-- reached the final time level, particles are sorted into contiguous blocks of |
---|
2199 | !-- finished and not-finished particles, in addition to their already sorting |
---|
2200 | !-- according to their sub-boxes. |
---|
2201 | IF ( .NOT. first_loop_stride .AND. use_sgs_for_particles ) & |
---|
2202 | CALL lpm_sort_timeloop_done |
---|
2203 | DO i = nxl, nxr |
---|
2204 | DO j = nys, nyn |
---|
2205 | ! |
---|
2206 | !-- Put the random seeds at grid point j, i |
---|
2207 | CALL random_seed_parallel( put=seq_random_array_particles(:,j,i) ) |
---|
2208 | |
---|
2209 | DO k = nzb+1, nzt |
---|
2210 | |
---|
2211 | number_of_particles = prt_count(k,j,i) |
---|
2212 | ! |
---|
2213 | !-- If grid cell gets empty, flag must be true |
---|
2214 | IF ( number_of_particles <= 0 ) THEN |
---|
2215 | grid_particles(k,j,i)%time_loop_done = .TRUE. |
---|
2216 | CYCLE |
---|
2217 | ENDIF |
---|
2218 | |
---|
2219 | IF ( .NOT. first_loop_stride .AND. & |
---|
2220 | grid_particles(k,j,i)%time_loop_done ) CYCLE |
---|
2221 | |
---|
2222 | particles => grid_particles(k,j,i)%particles(1:number_of_particles) |
---|
2223 | |
---|
2224 | particles(1:number_of_particles)%particle_mask = .TRUE. |
---|
2225 | ! |
---|
2226 | !-- Initialize the variable storing the total time that a particle has advanced |
---|
2227 | !-- within the timestep procedure. |
---|
2228 | IF ( first_loop_stride ) THEN |
---|
2229 | particles(1:number_of_particles)%dt_sum = 0.0_wp |
---|
2230 | ENDIF |
---|
2231 | ! |
---|
2232 | !-- Particle (droplet) growth by condensation/evaporation and collision |
---|
2233 | IF ( cloud_droplets .AND. first_loop_stride) THEN |
---|
2234 | ! |
---|
2235 | !-- Droplet growth by condensation / evaporation |
---|
2236 | CALL lpm_droplet_condensation(i,j,k) |
---|
2237 | ! |
---|
2238 | !-- Particle growth by collision |
---|
2239 | IF ( collision_kernel /= 'none' ) THEN |
---|
2240 | CALL lpm_droplet_collision(i,j,k) |
---|
2241 | ENDIF |
---|
2242 | |
---|
2243 | ENDIF |
---|
2244 | ! |
---|
2245 | !-- Initialize the switch used for the loop exit condition checked at the end |
---|
2246 | !-- of this loop. If at least one particle has failed to reach the LES |
---|
2247 | !-- timestep, this switch will be set false in lpm_advec. |
---|
2248 | dt_3d_reached_l = .TRUE. |
---|
2249 | |
---|
2250 | ! |
---|
2251 | !-- Particle advection |
---|
2252 | CALL lpm_advec( i, j, k ) |
---|
2253 | ! |
---|
2254 | !-- Particle reflection from walls. Only applied if the particles are in the |
---|
2255 | !-- vertical range of the topography. (Here, some optimization is still |
---|
2256 | !-- possible.) |
---|
2257 | IF ( topography /= 'flat' .AND. k < nzb_max + 2 ) THEN |
---|
2258 | CALL lpm_boundary_conds( 'walls', i, j, k ) |
---|
2259 | ENDIF |
---|
2260 | ! |
---|
2261 | !-- User-defined actions after the calculation of the new particle position |
---|
2262 | CALL user_lpm_advec( i, j, k ) |
---|
2263 | ! |
---|
2264 | !-- Apply boundary conditions to those particles that have crossed the top or |
---|
2265 | !-- bottom boundary and delete those particles, which are older than allowed |
---|
2266 | CALL lpm_boundary_conds( 'bottom/top', i, j, k ) |
---|
2267 | ! |
---|
2268 | !--- If not all particles of the actual grid cell have reached the LES timestep, |
---|
2269 | !-- this cell has to do another loop iteration. Due to the fact that particles |
---|
2270 | !-- can move into neighboring grid cells, these neighbor cells also have to |
---|
2271 | !-- perform another loop iteration. |
---|
2272 | !-- Please note, this realization does not work properly if particles move into |
---|
2273 | !-- another subdomain. |
---|
2274 | IF ( .NOT. dt_3d_reached_l ) THEN |
---|
2275 | ks = MAX(nzb+1,k-1) |
---|
2276 | ke = MIN(nzt,k+1) |
---|
2277 | js = MAX(nys,j-1) |
---|
2278 | je = MIN(nyn,j+1) |
---|
2279 | is = MAX(nxl,i-1) |
---|
2280 | ie = MIN(nxr,i+1) |
---|
2281 | grid_particles(ks:ke,js:je,is:ie)%time_loop_done = .FALSE. |
---|
2282 | ELSE |
---|
2283 | grid_particles(k,j,i)%time_loop_done = .TRUE. |
---|
2284 | ENDIF |
---|
2285 | |
---|
2286 | ENDDO |
---|
2287 | ! |
---|
2288 | !-- Get the new random seeds from last call at grid point jp, ip |
---|
2289 | CALL random_seed_parallel( get=seq_random_array_particles(:,j,i) ) |
---|
2290 | |
---|
2291 | ENDDO |
---|
2292 | ENDDO |
---|
2293 | steps = steps + 1 |
---|
2294 | dt_3d_reached_l = ALL(grid_particles(:,:,:)%time_loop_done) |
---|
2295 | ! |
---|
2296 | !-- Find out, if all particles on every PE have completed the LES timestep and set the |
---|
2297 | !-- switch corespondingly |
---|
2298 | #if defined( __parallel ) |
---|
2299 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
2300 | CALL MPI_ALLREDUCE( dt_3d_reached_l, dt_3d_reached, 1, MPI_LOGICAL, MPI_LAND, & |
---|
2301 | comm2d, ierr ) |
---|
2302 | #else |
---|
2303 | dt_3d_reached = dt_3d_reached_l |
---|
2304 | #endif |
---|
2305 | CALL cpu_log( log_point_s(44), 'lpm_advec', 'stop' ) |
---|
2306 | |
---|
2307 | ! |
---|
2308 | !-- Apply splitting and merging algorithm |
---|
2309 | IF ( cloud_droplets ) THEN |
---|
2310 | IF ( splitting ) THEN |
---|
2311 | CALL lpm_splitting |
---|
2312 | ENDIF |
---|
2313 | IF ( merging ) THEN |
---|
2314 | CALL lpm_merging |
---|
2315 | ENDIF |
---|
2316 | ENDIF |
---|
2317 | ! |
---|
2318 | !-- Move Particles local to PE to a different grid cell |
---|
2319 | CALL lpm_move_particle |
---|
2320 | ! |
---|
2321 | !-- Horizontal boundary conditions including exchange between subdmains |
---|
2322 | CALL lpm_exchange_horiz |
---|
2323 | |
---|
2324 | ! |
---|
2325 | !-- IF .FALSE., lpm_sort_and_delete is done inside pcmp |
---|
2326 | IF ( .NOT. dt_3d_reached .OR. .NOT. nested_run ) THEN |
---|
2327 | ! |
---|
2328 | !-- Pack particles (eliminate those marked for deletion), determine new number of |
---|
2329 | !-- particles |
---|
2330 | CALL lpm_sort_and_delete |
---|
2331 | |
---|
2332 | !-- Initialize variables for the next (sub-) timestep, i.e., for marking those |
---|
2333 | !-- particles to be deleted after the timestep |
---|
2334 | deleted_particles = 0 |
---|
2335 | ENDIF |
---|
2336 | |
---|
2337 | IF ( dt_3d_reached ) EXIT |
---|
2338 | |
---|
2339 | first_loop_stride = .FALSE. |
---|
2340 | ENDDO ! timestep loop |
---|
2341 | |
---|
2342 | #if defined( __parallel ) |
---|
2343 | ! |
---|
2344 | !-- In case of nested runs do the transfer of particles after every full model time step |
---|
2345 | IF ( nested_run ) THEN |
---|
2346 | CALL particles_from_parent_to_child |
---|
2347 | CALL particles_from_child_to_parent |
---|
2348 | CALL pmcp_p_delete_particles_in_fine_grid_area |
---|
2349 | |
---|
2350 | CALL lpm_sort_and_delete |
---|
2351 | |
---|
2352 | deleted_particles = 0 |
---|
2353 | ENDIF |
---|
2354 | #endif |
---|
2355 | |
---|
2356 | ! |
---|
2357 | !-- Calculate the new liquid water content for each grid box |
---|
2358 | IF ( cloud_droplets ) CALL lpm_calc_liquid_water_content |
---|
2359 | |
---|
2360 | ! |
---|
2361 | !-- At the end all arrays are exchanged |
---|
2362 | IF ( cloud_droplets ) THEN |
---|
2363 | CALL exchange_horiz( ql, nbgp ) |
---|
2364 | CALL exchange_horiz( ql_c, nbgp ) |
---|
2365 | CALL exchange_horiz( ql_v, nbgp ) |
---|
2366 | CALL exchange_horiz( ql_vp, nbgp ) |
---|
2367 | ENDIF |
---|
2368 | |
---|
2369 | ! |
---|
2370 | !-- Deallocate unused memory |
---|
2371 | IF ( deallocate_memory .AND. lpm_count == step_dealloc ) THEN |
---|
2372 | CALL dealloc_particles_array |
---|
2373 | lpm_count = 0 |
---|
2374 | ELSEIF ( deallocate_memory ) THEN |
---|
2375 | lpm_count = lpm_count + 1 |
---|
2376 | ENDIF |
---|
2377 | |
---|
2378 | ! |
---|
2379 | !-- Write particle statistics (in particular the number of particles exchanged between the |
---|
2380 | !-- subdomains) on file |
---|
2381 | IF ( write_particle_statistics ) CALL lpm_write_exchange_statistics |
---|
2382 | ! |
---|
2383 | !-- Execute Interactions of condnesation and evaporation to humidity and temperature field |
---|
2384 | IF ( cloud_droplets ) THEN |
---|
2385 | CALL lpm_interaction_droplets_ptq |
---|
2386 | CALL exchange_horiz( pt, nbgp ) |
---|
2387 | CALL exchange_horiz( q, nbgp ) |
---|
2388 | ENDIF |
---|
2389 | |
---|
2390 | CALL cpu_log( log_point(25), 'lpm', 'stop' ) |
---|
2391 | |
---|
2392 | ! ! |
---|
2393 | ! !-- Output of particle time series |
---|
2394 | ! IF ( particle_advection ) THEN |
---|
2395 | ! IF ( time_dopts >= dt_dopts .OR. & |
---|
2396 | ! ( time_since_reference_point >= particle_advection_start .AND. & |
---|
2397 | ! first_call_lpm ) ) THEN |
---|
2398 | ! CALL lpm_data_output_ptseries |
---|
2399 | ! time_dopts = MOD( time_dopts, MAX( dt_dopts, dt_3d ) ) |
---|
2400 | ! ENDIF |
---|
2401 | ! ENDIF |
---|
2402 | |
---|
2403 | ! |
---|
2404 | !-- Set this switch to .false. @todo: maybe find better solution. |
---|
2405 | first_call_lpm = .FALSE. |
---|
2406 | ENDIF! ENDIF statement of lpm_actions('after_pressure_solver') |
---|
2407 | |
---|
2408 | CASE ( 'after_integration' ) |
---|
2409 | ! |
---|
2410 | !-- Call at the end of timestep routine to save particle velocities fields for the next |
---|
2411 | !-- timestep |
---|
2412 | CALL lpm_swap_timelevel_for_particle_advection |
---|
2413 | |
---|
2414 | CASE DEFAULT |
---|
2415 | CONTINUE |
---|
2416 | |
---|
2417 | END SELECT |
---|
2418 | |
---|
2419 | END SUBROUTINE lpm_actions |
---|
2420 | |
---|
2421 | |
---|
2422 | #if defined( __parallel ) |
---|
2423 | !--------------------------------------------------------------------------------------------------! |
---|
2424 | ! Description: |
---|
2425 | ! ------------ |
---|
2426 | ! |
---|
2427 | !--------------------------------------------------------------------------------------------------! |
---|
2428 | SUBROUTINE particles_from_parent_to_child |
---|
2429 | |
---|
2430 | CALL pmcp_c_get_particle_from_parent ! Child actions |
---|
2431 | CALL pmcp_p_fill_particle_win ! Parent actions |
---|
2432 | |
---|
2433 | RETURN |
---|
2434 | |
---|
2435 | END SUBROUTINE particles_from_parent_to_child |
---|
2436 | |
---|
2437 | |
---|
2438 | !--------------------------------------------------------------------------------------------------! |
---|
2439 | ! Description: |
---|
2440 | ! ------------ |
---|
2441 | ! |
---|
2442 | !--------------------------------------------------------------------------------------------------! |
---|
2443 | SUBROUTINE particles_from_child_to_parent |
---|
2444 | |
---|
2445 | CALL pmcp_c_send_particle_to_parent ! Child actions |
---|
2446 | CALL pmcp_p_empty_particle_win ! Parent actions |
---|
2447 | |
---|
2448 | RETURN |
---|
2449 | |
---|
2450 | END SUBROUTINE particles_from_child_to_parent |
---|
2451 | #endif |
---|
2452 | |
---|
2453 | !--------------------------------------------------------------------------------------------------! |
---|
2454 | ! Description: |
---|
2455 | ! ------------ |
---|
2456 | !> This routine write exchange statistics of the lpm in a ascii file. |
---|
2457 | !--------------------------------------------------------------------------------------------------! |
---|
2458 | SUBROUTINE lpm_write_exchange_statistics |
---|
2459 | |
---|
2460 | INTEGER(iwp) :: ip !< |
---|
2461 | INTEGER(iwp) :: jp !< |
---|
2462 | INTEGER(iwp) :: kp !< |
---|
2463 | INTEGER(iwp) :: tot_number_of_particles !< |
---|
2464 | |
---|
2465 | ! |
---|
2466 | !-- Determine the current number of particles |
---|
2467 | number_of_particles = 0 |
---|
2468 | DO ip = nxl, nxr |
---|
2469 | DO jp = nys, nyn |
---|
2470 | DO kp = nzb+1, nzt |
---|
2471 | number_of_particles = number_of_particles + prt_count(kp,jp,ip) |
---|
2472 | ENDDO |
---|
2473 | ENDDO |
---|
2474 | ENDDO |
---|
2475 | |
---|
2476 | CALL check_open( 80 ) |
---|
2477 | #if defined( __parallel ) |
---|
2478 | WRITE ( 80, 8000 ) current_timestep_number+1, simulated_time+dt_3d, number_of_particles, & |
---|
2479 | pleft, trlp_count_sum, trlp_count_recv_sum, pright, trrp_count_sum, & |
---|
2480 | trrp_count_recv_sum, psouth, trsp_count_sum, trsp_count_recv_sum, pnorth, & |
---|
2481 | trnp_count_sum, trnp_count_recv_sum |
---|
2482 | #else |
---|
2483 | WRITE ( 80, 8000 ) current_timestep_number+1, simulated_time+dt_3d, number_of_particles |
---|
2484 | #endif |
---|
2485 | CALL close_file( 80 ) |
---|
2486 | |
---|
2487 | IF ( number_of_particles > 0 ) THEN |
---|
2488 | WRITE(9,*) 'number_of_particles ', number_of_particles, current_timestep_number + 1, & |
---|
2489 | simulated_time + dt_3d |
---|
2490 | ENDIF |
---|
2491 | |
---|
2492 | #if defined( __parallel ) |
---|
2493 | CALL MPI_ALLREDUCE( number_of_particles, tot_number_of_particles, 1, MPI_INTEGER, MPI_SUM, & |
---|
2494 | comm2d, ierr ) |
---|
2495 | #else |
---|
2496 | tot_number_of_particles = number_of_particles |
---|
2497 | #endif |
---|
2498 | |
---|
2499 | #if defined( __parallel ) |
---|
2500 | IF ( nested_run ) THEN |
---|
2501 | CALL pmcp_g_print_number_of_particles( simulated_time + dt_3d, tot_number_of_particles) |
---|
2502 | ENDIF |
---|
2503 | #endif |
---|
2504 | |
---|
2505 | ! |
---|
2506 | !-- Formats |
---|
2507 | 8000 FORMAT (I6,1X,F7.2,4X,I10,5X,4(I3,1X,I4,'/',I4,2X),6X,I10) |
---|
2508 | |
---|
2509 | |
---|
2510 | END SUBROUTINE lpm_write_exchange_statistics |
---|
2511 | |
---|
2512 | |
---|
2513 | !--------------------------------------------------------------------------------------------------! |
---|
2514 | ! Description: |
---|
2515 | ! ------------ |
---|
2516 | !> Write particle data in FORTRAN binary and/or netCDF format |
---|
2517 | !--------------------------------------------------------------------------------------------------! |
---|
2518 | SUBROUTINE lpm_data_output_particles |
---|
2519 | |
---|
2520 | INTEGER(iwp) :: ip !< |
---|
2521 | INTEGER(iwp) :: jp !< |
---|
2522 | INTEGER(iwp) :: kp !< |
---|
2523 | |
---|
2524 | CALL cpu_log( log_point_s(40), 'lpm_data_output', 'start' ) |
---|
2525 | |
---|
2526 | ! |
---|
2527 | !-- Attention: change version number for unit 85 (in routine check_open) whenever the output format |
---|
2528 | !-- for this unit is changed! |
---|
2529 | CALL check_open( 85 ) |
---|
2530 | |
---|
2531 | WRITE ( 85 ) simulated_time |
---|
2532 | WRITE ( 85 ) prt_count |
---|
2533 | |
---|
2534 | DO ip = nxl, nxr |
---|
2535 | DO jp = nys, nyn |
---|
2536 | DO kp = nzb+1, nzt |
---|
2537 | number_of_particles = prt_count(kp,jp,ip) |
---|
2538 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
2539 | IF ( number_of_particles <= 0 ) CYCLE |
---|
2540 | WRITE ( 85 ) particles |
---|
2541 | ENDDO |
---|
2542 | ENDDO |
---|
2543 | ENDDO |
---|
2544 | |
---|
2545 | CALL close_file( 85 ) |
---|
2546 | |
---|
2547 | |
---|
2548 | #if defined( __netcdf ) |
---|
2549 | ! ! |
---|
2550 | ! !-- Output in netCDF format |
---|
2551 | ! CALL check_open( 108 ) |
---|
2552 | ! |
---|
2553 | ! ! |
---|
2554 | ! !-- Update the NetCDF time axis |
---|
2555 | ! prt_time_count = prt_time_count + 1 |
---|
2556 | ! |
---|
2557 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_time_prt, & |
---|
2558 | ! (/ simulated_time /), & |
---|
2559 | ! start = (/ prt_time_count /), count = (/ 1 /) ) |
---|
2560 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 1 ) |
---|
2561 | ! |
---|
2562 | ! ! |
---|
2563 | ! !-- Output the real number of particles used |
---|
2564 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_rnop_prt, & |
---|
2565 | ! (/ number_of_particles /), & |
---|
2566 | ! start = (/ prt_time_count /), count = (/ 1 /) ) |
---|
2567 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 2 ) |
---|
2568 | ! |
---|
2569 | ! ! |
---|
2570 | ! !-- Output all particle attributes |
---|
2571 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(1), particles%age, & |
---|
2572 | ! start = (/ 1, prt_time_count /), & |
---|
2573 | ! count = (/ maximum_number_of_particles /) ) |
---|
2574 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 3 ) |
---|
2575 | ! |
---|
2576 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(2), particles%user, & |
---|
2577 | ! start = (/ 1, prt_time_count /), & |
---|
2578 | ! count = (/ maximum_number_of_particles /) ) |
---|
2579 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 4 ) |
---|
2580 | ! |
---|
2581 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(3), particles%origin_x, & |
---|
2582 | ! start = (/ 1, prt_time_count /), & |
---|
2583 | ! count = (/ maximum_number_of_particles /) ) |
---|
2584 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 5 ) |
---|
2585 | ! |
---|
2586 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(4), particles%origin_y, & |
---|
2587 | ! start = (/ 1, prt_time_count /), & |
---|
2588 | ! count = (/ maximum_number_of_particles /) ) |
---|
2589 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 6 ) |
---|
2590 | ! |
---|
2591 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(5), particles%origin_z, & |
---|
2592 | ! start = (/ 1, prt_time_count /), & |
---|
2593 | ! count = (/ maximum_number_of_particles /) ) |
---|
2594 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 7 ) |
---|
2595 | ! |
---|
2596 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(6), particles%radius, & |
---|
2597 | ! start = (/ 1, prt_time_count /), & |
---|
2598 | ! count = (/ maximum_number_of_particles /) ) |
---|
2599 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 8 ) |
---|
2600 | ! |
---|
2601 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(7), particles%speed_x, & |
---|
2602 | ! start = (/ 1, prt_time_count /), & |
---|
2603 | ! count = (/ maximum_number_of_particles /) ) |
---|
2604 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 9 ) |
---|
2605 | ! |
---|
2606 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(8), particles%speed_y, & |
---|
2607 | ! start = (/ 1, prt_time_count /), & |
---|
2608 | ! count = (/ maximum_number_of_particles /) ) |
---|
2609 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 10 ) |
---|
2610 | ! |
---|
2611 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(9), particles%speed_z, & |
---|
2612 | ! start = (/ 1, prt_time_count /), & |
---|
2613 | ! count = (/ maximum_number_of_particles /) ) |
---|
2614 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 11 ) |
---|
2615 | ! |
---|
2616 | ! nc_stat = NF90_PUT_VAR( id_set_prt,id_var_prt(10), & |
---|
2617 | ! particles%weight_factor, & |
---|
2618 | ! start = (/ 1, prt_time_count /), & |
---|
2619 | ! count = (/ maximum_number_of_particles /) ) |
---|
2620 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 12 ) |
---|
2621 | ! |
---|
2622 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(11), particles%x, & |
---|
2623 | ! start = (/ 1, prt_time_count /), & |
---|
2624 | ! count = (/ maximum_number_of_particles /) ) |
---|
2625 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 13 ) |
---|
2626 | ! |
---|
2627 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(12), particles%y, & |
---|
2628 | ! start = (/ 1, prt_time_count /), & |
---|
2629 | ! count = (/ maximum_number_of_particles /) ) |
---|
2630 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 14 ) |
---|
2631 | ! |
---|
2632 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(13), particles%z, & |
---|
2633 | ! start = (/ 1, prt_time_count /), & |
---|
2634 | ! count = (/ maximum_number_of_particles /) ) |
---|
2635 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 15 ) |
---|
2636 | ! |
---|
2637 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(14), particles%class, & |
---|
2638 | ! start = (/ 1, prt_time_count /), & |
---|
2639 | ! count = (/ maximum_number_of_particles /) ) |
---|
2640 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 16 ) |
---|
2641 | ! |
---|
2642 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(15), particles%group, & |
---|
2643 | ! start = (/ 1, prt_time_count /), & |
---|
2644 | ! count = (/ maximum_number_of_particles /) ) |
---|
2645 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 17 ) |
---|
2646 | ! |
---|
2647 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(16), & |
---|
2648 | ! particles%id2, & |
---|
2649 | ! start = (/ 1, prt_time_count /), & |
---|
2650 | ! count = (/ maximum_number_of_particles /) ) |
---|
2651 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 18 ) |
---|
2652 | ! |
---|
2653 | ! nc_stat = NF90_PUT_VAR( id_set_prt, id_var_prt(17), particles%id1, & |
---|
2654 | ! start = (/ 1, prt_time_count /), & |
---|
2655 | ! count = (/ maximum_number_of_particles /) ) |
---|
2656 | ! CALL netcdf_handle_error( 'lpm_data_output_particles', 19 ) |
---|
2657 | ! |
---|
2658 | #endif |
---|
2659 | |
---|
2660 | CALL cpu_log( log_point_s(40), 'lpm_data_output', 'stop' ) |
---|
2661 | |
---|
2662 | END SUBROUTINE lpm_data_output_particles |
---|
2663 | |
---|
2664 | !--------------------------------------------------------------------------------------------------! |
---|
2665 | ! Description: |
---|
2666 | ! ------------ |
---|
2667 | !> This routine calculates and provide particle timeseries output. |
---|
2668 | !--------------------------------------------------------------------------------------------------! |
---|
2669 | SUBROUTINE lpm_data_output_ptseries |
---|
2670 | |
---|
2671 | INTEGER(iwp) :: i !< |
---|
2672 | INTEGER(iwp) :: inum !< |
---|
2673 | INTEGER(iwp) :: j !< |
---|
2674 | INTEGER(iwp) :: jg !< |
---|
2675 | INTEGER(iwp) :: k !< |
---|
2676 | INTEGER(iwp) :: n !< |
---|
2677 | |
---|
2678 | REAL(wp), DIMENSION(:,:), ALLOCATABLE :: pts_value !< |
---|
2679 | REAL(wp), DIMENSION(:,:), ALLOCATABLE :: pts_value_l !< |
---|
2680 | |
---|
2681 | |
---|
2682 | CALL cpu_log( log_point(36), 'data_output_ptseries', 'start' ) |
---|
2683 | |
---|
2684 | IF ( myid == 0 ) THEN |
---|
2685 | ! |
---|
2686 | !-- Open file for time series output in NetCDF format |
---|
2687 | dopts_time_count = dopts_time_count + 1 |
---|
2688 | CALL check_open( 109 ) |
---|
2689 | #if defined( __netcdf ) |
---|
2690 | ! |
---|
2691 | !-- Update the particle time series time axis |
---|
2692 | nc_stat = NF90_PUT_VAR( id_set_pts, id_var_time_pts, (/ time_since_reference_point /), & |
---|
2693 | start = (/ dopts_time_count /), count = (/ 1 /) ) |
---|
2694 | CALL netcdf_handle_error( 'data_output_ptseries', 391 ) |
---|
2695 | #endif |
---|
2696 | |
---|
2697 | ENDIF |
---|
2698 | |
---|
2699 | ALLOCATE( pts_value(0:number_of_particle_groups,dopts_num), & |
---|
2700 | pts_value_l(0:number_of_particle_groups,dopts_num) ) |
---|
2701 | |
---|
2702 | pts_value_l = 0.0_wp |
---|
2703 | pts_value_l(:,16) = 9999999.9_wp ! for calculation of minimum radius |
---|
2704 | |
---|
2705 | ! |
---|
2706 | !-- Calculate or collect the particle time series quantities for all particles and seperately for |
---|
2707 | !-- each particle group (if there is more than one group) |
---|
2708 | DO i = nxl, nxr |
---|
2709 | DO j = nys, nyn |
---|
2710 | DO k = nzb, nzt |
---|
2711 | number_of_particles = prt_count(k,j,i) |
---|
2712 | IF (number_of_particles <= 0) CYCLE |
---|
2713 | particles => grid_particles(k,j,i)%particles(1:number_of_particles) |
---|
2714 | DO n = 1, number_of_particles |
---|
2715 | |
---|
2716 | IF ( particles(n)%particle_mask ) THEN ! Restrict analysis to active particles |
---|
2717 | |
---|
2718 | pts_value_l(0,1) = pts_value_l(0,1) + 1.0_wp ! total # of particles |
---|
2719 | pts_value_l(0,2) = pts_value_l(0,2) + & |
---|
2720 | ( particles(n)%x - particles(n)%origin_x ) ! mean x |
---|
2721 | pts_value_l(0,3) = pts_value_l(0,3) + & |
---|
2722 | ( particles(n)%y - particles(n)%origin_y ) ! mean y |
---|
2723 | pts_value_l(0,4) = pts_value_l(0,4) + & |
---|
2724 | ( particles(n)%z - particles(n)%origin_z ) ! mean z |
---|
2725 | pts_value_l(0,5) = pts_value_l(0,5) + particles(n)%z ! mean z (absolute) |
---|
2726 | pts_value_l(0,6) = pts_value_l(0,6) + particles(n)%speed_x ! mean u |
---|
2727 | pts_value_l(0,7) = pts_value_l(0,7) + particles(n)%speed_y ! mean v |
---|
2728 | pts_value_l(0,8) = pts_value_l(0,8) + particles(n)%speed_z ! mean w |
---|
2729 | pts_value_l(0,9) = pts_value_l(0,9) + particles(n)%rvar1 ! mean sgsu |
---|
2730 | pts_value_l(0,10) = pts_value_l(0,10) + particles(n)%rvar2 ! mean sgsv |
---|
2731 | pts_value_l(0,11) = pts_value_l(0,11) + particles(n)%rvar3 ! mean sgsw |
---|
2732 | IF ( particles(n)%speed_z > 0.0_wp ) THEN |
---|
2733 | pts_value_l(0,12) = pts_value_l(0,12) + 1.0_wp ! # of upward moving prts |
---|
2734 | pts_value_l(0,13) = pts_value_l(0,13) + particles(n)%speed_z ! mean w upw. |
---|
2735 | ELSE |
---|
2736 | pts_value_l(0,14) = pts_value_l(0,14) + particles(n)%speed_z ! mean w down |
---|
2737 | ENDIF |
---|
2738 | pts_value_l(0,15) = pts_value_l(0,15) + particles(n)%radius ! mean rad |
---|
2739 | pts_value_l(0,16) = MIN( pts_value_l(0,16), particles(n)%radius ) ! minrad |
---|
2740 | pts_value_l(0,17) = MAX( pts_value_l(0,17), particles(n)%radius ) ! maxrad |
---|
2741 | pts_value_l(0,18) = pts_value_l(0,18) + 1.0_wp |
---|
2742 | pts_value_l(0,19) = pts_value_l(0,18) + 1.0_wp |
---|
2743 | ! |
---|
2744 | !-- Repeat the same for the respective particle group |
---|
2745 | IF ( number_of_particle_groups > 1 ) THEN |
---|
2746 | jg = particles(n)%group |
---|
2747 | |
---|
2748 | pts_value_l(jg,1) = pts_value_l(jg,1) + 1.0_wp |
---|
2749 | pts_value_l(jg,2) = pts_value_l(jg,2) + & |
---|
2750 | ( particles(n)%x - particles(n)%origin_x ) |
---|
2751 | pts_value_l(jg,3) = pts_value_l(jg,3) + & |
---|
2752 | ( particles(n)%y - particles(n)%origin_y ) |
---|
2753 | pts_value_l(jg,4) = pts_value_l(jg,4) + & |
---|
2754 | ( particles(n)%z - particles(n)%origin_z ) |
---|
2755 | pts_value_l(jg,5) = pts_value_l(jg,5) + particles(n)%z |
---|
2756 | pts_value_l(jg,6) = pts_value_l(jg,6) + particles(n)%speed_x |
---|
2757 | pts_value_l(jg,7) = pts_value_l(jg,7) + particles(n)%speed_y |
---|
2758 | pts_value_l(jg,8) = pts_value_l(jg,8) + particles(n)%speed_z |
---|
2759 | pts_value_l(jg,9) = pts_value_l(jg,9) + particles(n)%rvar1 |
---|
2760 | pts_value_l(jg,10) = pts_value_l(jg,10) + particles(n)%rvar2 |
---|
2761 | pts_value_l(jg,11) = pts_value_l(jg,11) + particles(n)%rvar3 |
---|
2762 | IF ( particles(n)%speed_z > 0.0_wp ) THEN |
---|
2763 | pts_value_l(jg,12) = pts_value_l(jg,12) + 1.0_wp |
---|
2764 | pts_value_l(jg,13) = pts_value_l(jg,13) + particles(n)%speed_z |
---|
2765 | ELSE |
---|
2766 | pts_value_l(jg,14) = pts_value_l(jg,14) + particles(n)%speed_z |
---|
2767 | ENDIF |
---|
2768 | pts_value_l(jg,15) = pts_value_l(jg,15) + particles(n)%radius |
---|
2769 | pts_value_l(jg,16) = MIN( pts_value_l(jg,16), particles(n)%radius ) |
---|
2770 | pts_value_l(jg,17) = MAX( pts_value_l(jg,17), particles(n)%radius ) |
---|
2771 | pts_value_l(jg,18) = pts_value_l(jg,18) + 1.0_wp |
---|
2772 | pts_value_l(jg,19) = pts_value_l(jg,19) + 1.0_wp |
---|
2773 | ENDIF |
---|
2774 | |
---|
2775 | ENDIF |
---|
2776 | |
---|
2777 | ENDDO |
---|
2778 | |
---|
2779 | ENDDO |
---|
2780 | ENDDO |
---|
2781 | ENDDO |
---|
2782 | |
---|
2783 | |
---|
2784 | #if defined( __parallel ) |
---|
2785 | ! |
---|
2786 | !-- Sum values of the subdomains |
---|
2787 | inum = number_of_particle_groups + 1 |
---|
2788 | |
---|
2789 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
2790 | CALL MPI_ALLREDUCE( pts_value_l(0,1), pts_value(0,1), 15*inum, MPI_REAL, MPI_SUM, comm2d, ierr ) |
---|
2791 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
2792 | CALL MPI_ALLREDUCE( pts_value_l(0,16), pts_value(0,16), inum, MPI_REAL, MPI_MIN, comm2d, ierr ) |
---|
2793 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
2794 | CALL MPI_ALLREDUCE( pts_value_l(0,17), pts_value(0,17), inum, MPI_REAL, MPI_MAX, comm2d, ierr ) |
---|
2795 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
2796 | CALL MPI_ALLREDUCE( pts_value_l(0,18), pts_value(0,18), inum, MPI_REAL, MPI_MAX, comm2d, ierr ) |
---|
2797 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
2798 | CALL MPI_ALLREDUCE( pts_value_l(0,19), pts_value(0,19), inum, MPI_REAL, MPI_MIN, comm2d, ierr ) |
---|
2799 | #else |
---|
2800 | pts_value(:,1:19) = pts_value_l(:,1:19) |
---|
2801 | #endif |
---|
2802 | |
---|
2803 | ! |
---|
2804 | !-- Normalize the above calculated quantities (except min/max values) with the total number of |
---|
2805 | !-- particles |
---|
2806 | IF ( number_of_particle_groups > 1 ) THEN |
---|
2807 | inum = number_of_particle_groups |
---|
2808 | ELSE |
---|
2809 | inum = 0 |
---|
2810 | ENDIF |
---|
2811 | |
---|
2812 | DO j = 0, inum |
---|
2813 | |
---|
2814 | IF ( pts_value(j,1) > 0.0_wp ) THEN |
---|
2815 | |
---|
2816 | pts_value(j,2:15) = pts_value(j,2:15) / pts_value(j,1) |
---|
2817 | IF ( pts_value(j,12) > 0.0_wp .AND. pts_value(j,12) < 1.0_wp ) THEN |
---|
2818 | pts_value(j,13) = pts_value(j,13) / pts_value(j,12) |
---|
2819 | pts_value(j,14) = pts_value(j,14) / ( 1.0_wp - pts_value(j,12) ) |
---|
2820 | ELSEIF ( pts_value(j,12) == 0.0_wp ) THEN |
---|
2821 | pts_value(j,13) = -1.0_wp |
---|
2822 | ELSE |
---|
2823 | pts_value(j,14) = -1.0_wp |
---|
2824 | ENDIF |
---|
2825 | |
---|
2826 | ENDIF |
---|
2827 | |
---|
2828 | ENDDO |
---|
2829 | |
---|
2830 | ! |
---|
2831 | !-- Calculate higher order moments of particle time series quantities, seperately for each particle |
---|
2832 | !-- group (if there is more than one group) |
---|
2833 | DO i = nxl, nxr |
---|
2834 | DO j = nys, nyn |
---|
2835 | DO k = nzb, nzt |
---|
2836 | number_of_particles = prt_count(k,j,i) |
---|
2837 | IF (number_of_particles <= 0) CYCLE |
---|
2838 | particles => grid_particles(k,j,i)%particles(1:number_of_particles) |
---|
2839 | DO n = 1, number_of_particles |
---|
2840 | |
---|
2841 | pts_value_l(0,20) = pts_value_l(0,20) + ( particles(n)%x - & |
---|
2842 | particles(n)%origin_x - pts_value(0,2) )**2 ! x*2 |
---|
2843 | pts_value_l(0,21) = pts_value_l(0,21) + ( particles(n)%y - & |
---|
2844 | particles(n)%origin_y - pts_value(0,3) )**2 ! y*2 |
---|
2845 | pts_value_l(0,22) = pts_value_l(0,22) + ( particles(n)%z - & |
---|
2846 | particles(n)%origin_z - pts_value(0,4) )**2 ! z*2 |
---|
2847 | pts_value_l(0,23) = pts_value_l(0,23) + ( particles(n)%speed_x - & |
---|
2848 | pts_value(0,6) )**2 ! u*2 |
---|
2849 | pts_value_l(0,24) = pts_value_l(0,24) + ( particles(n)%speed_y - & |
---|
2850 | pts_value(0,7) )**2 ! v*2 |
---|
2851 | pts_value_l(0,25) = pts_value_l(0,25) + ( particles(n)%speed_z - & |
---|
2852 | pts_value(0,8) )**2 ! w*2 |
---|
2853 | pts_value_l(0,26) = pts_value_l(0,26) + ( particles(n)%rvar1 - & |
---|
2854 | pts_value(0,9) )**2 ! u"2 |
---|
2855 | pts_value_l(0,27) = pts_value_l(0,27) + ( particles(n)%rvar2 - & |
---|
2856 | pts_value(0,10) )**2 ! v"2 |
---|
2857 | pts_value_l(0,28) = pts_value_l(0,28) + ( particles(n)%rvar3 - & |
---|
2858 | pts_value(0,11) )**2 ! w"2 |
---|
2859 | ! |
---|
2860 | !-- Repeat the same for the respective particle group |
---|
2861 | IF ( number_of_particle_groups > 1 ) THEN |
---|
2862 | jg = particles(n)%group |
---|
2863 | |
---|
2864 | pts_value_l(jg,20) = pts_value_l(jg,20) + ( particles(n)%x - & |
---|
2865 | particles(n)%origin_x - pts_value(jg,2) )**2 |
---|
2866 | pts_value_l(jg,21) = pts_value_l(jg,21) + ( particles(n)%y - & |
---|
2867 | particles(n)%origin_y - pts_value(jg,3) )**2 |
---|
2868 | pts_value_l(jg,22) = pts_value_l(jg,22) + ( particles(n)%z - & |
---|
2869 | particles(n)%origin_z - pts_value(jg,4) )**2 |
---|
2870 | pts_value_l(jg,23) = pts_value_l(jg,23) + ( particles(n)%speed_x - & |
---|
2871 | pts_value(jg,6) )**2 |
---|
2872 | pts_value_l(jg,24) = pts_value_l(jg,24) + ( particles(n)%speed_y - & |
---|
2873 | pts_value(jg,7) )**2 |
---|
2874 | pts_value_l(jg,25) = pts_value_l(jg,25) + ( particles(n)%speed_z - & |
---|
2875 | pts_value(jg,8) )**2 |
---|
2876 | pts_value_l(jg,26) = pts_value_l(jg,26) + ( particles(n)%rvar1 - & |
---|
2877 | pts_value(jg,9) )**2 |
---|
2878 | pts_value_l(jg,27) = pts_value_l(jg,27) + ( particles(n)%rvar2 - & |
---|
2879 | pts_value(jg,10) )**2 |
---|
2880 | pts_value_l(jg,28) = pts_value_l(jg,28) + ( particles(n)%rvar3 - & |
---|
2881 | pts_value(jg,11) )**2 |
---|
2882 | ENDIF |
---|
2883 | |
---|
2884 | ENDDO |
---|
2885 | ENDDO |
---|
2886 | ENDDO |
---|
2887 | ENDDO |
---|
2888 | |
---|
2889 | pts_value_l(0,29) = ( number_of_particles - pts_value(0,1) / numprocs )**2 |
---|
2890 | ! variance of particle numbers |
---|
2891 | IF ( number_of_particle_groups > 1 ) THEN |
---|
2892 | DO j = 1, number_of_particle_groups |
---|
2893 | pts_value_l(j,29) = ( pts_value_l(j,1) - pts_value(j,1) / numprocs )**2 |
---|
2894 | ENDDO |
---|
2895 | ENDIF |
---|
2896 | |
---|
2897 | #if defined( __parallel ) |
---|
2898 | ! |
---|
2899 | !-- Sum values of the subdomains |
---|
2900 | inum = number_of_particle_groups + 1 |
---|
2901 | |
---|
2902 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
2903 | CALL MPI_ALLREDUCE( pts_value_l(0,20), pts_value(0,20), inum*10, MPI_REAL, MPI_SUM, comm2d, & |
---|
2904 | ierr ) |
---|
2905 | #else |
---|
2906 | pts_value(:,20:29) = pts_value_l(:,20:29) |
---|
2907 | #endif |
---|
2908 | |
---|
2909 | ! |
---|
2910 | !-- Normalize the above calculated quantities with the total number of particles |
---|
2911 | IF ( number_of_particle_groups > 1 ) THEN |
---|
2912 | inum = number_of_particle_groups |
---|
2913 | ELSE |
---|
2914 | inum = 0 |
---|
2915 | ENDIF |
---|
2916 | |
---|
2917 | DO j = 0, inum |
---|
2918 | |
---|
2919 | IF ( pts_value(j,1) > 0.0_wp ) THEN |
---|
2920 | pts_value(j,20:28) = pts_value(j,20:28) / pts_value(j,1) |
---|
2921 | ENDIF |
---|
2922 | pts_value(j,29) = pts_value(j,29) / numprocs |
---|
2923 | |
---|
2924 | ENDDO |
---|
2925 | |
---|
2926 | #if defined( __netcdf ) |
---|
2927 | ! |
---|
2928 | !-- Output particle time series quantities in NetCDF format |
---|
2929 | IF ( myid == 0 ) THEN |
---|
2930 | DO j = 0, inum |
---|
2931 | DO i = 1, dopts_num |
---|
2932 | nc_stat = NF90_PUT_VAR( id_set_pts, id_var_dopts(i,j), & |
---|
2933 | (/ pts_value(j,i) /), & |
---|
2934 | start = (/ dopts_time_count /), & |
---|
2935 | count = (/ 1 /) ) |
---|
2936 | CALL netcdf_handle_error( 'data_output_ptseries', 392 ) |
---|
2937 | ENDDO |
---|
2938 | ENDDO |
---|
2939 | ENDIF |
---|
2940 | #endif |
---|
2941 | |
---|
2942 | DEALLOCATE( pts_value, pts_value_l ) |
---|
2943 | |
---|
2944 | CALL cpu_log( log_point(36), 'data_output_ptseries', 'stop' ) |
---|
2945 | |
---|
2946 | END SUBROUTINE lpm_data_output_ptseries |
---|
2947 | |
---|
2948 | |
---|
2949 | !--------------------------------------------------------------------------------------------------! |
---|
2950 | ! Description: |
---|
2951 | ! ------------ |
---|
2952 | !> This routine reads the respective restart data for the lpm. |
---|
2953 | !--------------------------------------------------------------------------------------------------! |
---|
2954 | SUBROUTINE lpm_rrd_local_particles |
---|
2955 | |
---|
2956 | CHARACTER(LEN=10) :: particle_binary_version !< |
---|
2957 | CHARACTER(LEN=10) :: version_on_file !< |
---|
2958 | |
---|
2959 | CHARACTER(LEN=20) :: save_restart_data_format_input !< |
---|
2960 | |
---|
2961 | INTEGER(iwp) :: alloc_size !< |
---|
2962 | INTEGER(iwp) :: ip !< |
---|
2963 | INTEGER(iwp) :: jp !< |
---|
2964 | INTEGER(iwp) :: kp !< |
---|
2965 | |
---|
2966 | INTEGER(idp), ALLOCATABLE, DIMENSION(:,:,:) :: prt_global_index !< |
---|
2967 | |
---|
2968 | LOGICAL :: save_debug_output !< |
---|
2969 | |
---|
2970 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: tmp_particles !< |
---|
2971 | |
---|
2972 | IF ( TRIM( restart_data_format_input ) == 'fortran_binary' ) THEN |
---|
2973 | |
---|
2974 | ! |
---|
2975 | !-- Read particle data from previous model run. |
---|
2976 | !-- First open the input unit. |
---|
2977 | IF ( myid_char == '' ) THEN |
---|
2978 | OPEN ( 90, FILE='PARTICLE_RESTART_DATA_IN'//myid_char, FORM='UNFORMATTED' ) |
---|
2979 | ELSE |
---|
2980 | OPEN ( 90, FILE='PARTICLE_RESTART_DATA_IN/'//myid_char, FORM='UNFORMATTED' ) |
---|
2981 | ENDIF |
---|
2982 | |
---|
2983 | ! |
---|
2984 | !-- First compare the version numbers |
---|
2985 | READ ( 90 ) version_on_file |
---|
2986 | particle_binary_version = '4.0' |
---|
2987 | IF ( TRIM( version_on_file ) /= TRIM( particle_binary_version ) ) THEN |
---|
2988 | message_string = 'version mismatch concerning data from prior ' // & |
---|
2989 | 'run &version on file = "' // & |
---|
2990 | TRIM( version_on_file ) // & |
---|
2991 | '&version in program = "' // & |
---|
2992 | TRIM( particle_binary_version ) // '"' |
---|
2993 | CALL message( 'lpm_read_restart_file', 'PA0214', 1, 2, 0, 6, 0 ) |
---|
2994 | ENDIF |
---|
2995 | |
---|
2996 | ! |
---|
2997 | !-- If less particles are stored on the restart file than prescribed by 1, the remainder is |
---|
2998 | !-- initialized by zero_particle to avoid errors. |
---|
2999 | zero_particle = particle_type( 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, & |
---|
3000 | 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, & |
---|
3001 | 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, & |
---|
3002 | 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, 0.0_wp, & |
---|
3003 | 0, 0, 0_idp, .FALSE., -1, -1 ) |
---|
3004 | ! |
---|
3005 | !-- Read some particle parameters and the size of the particle arrays, allocate them and read |
---|
3006 | !-- their contents. |
---|
3007 | READ ( 90 ) bc_par_b, bc_par_lr, bc_par_ns, bc_par_t, last_particle_release_time, & |
---|
3008 | number_of_particle_groups, particle_groups, time_write_particle_data |
---|
3009 | |
---|
3010 | ALLOCATE( prt_count(nzb:nzt+1,nysg:nyng,nxlg:nxrg), & |
---|
3011 | grid_particles(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3012 | |
---|
3013 | READ ( 90 ) prt_count |
---|
3014 | |
---|
3015 | DO ip = nxl, nxr |
---|
3016 | DO jp = nys, nyn |
---|
3017 | DO kp = nzb+1, nzt |
---|
3018 | |
---|
3019 | number_of_particles = prt_count(kp,jp,ip) |
---|
3020 | IF ( number_of_particles > 0 ) THEN |
---|
3021 | alloc_size = MAX( INT( number_of_particles * & |
---|
3022 | ( 1.0_wp + alloc_factor / 100.0_wp ) ), & |
---|
3023 | 1 ) |
---|
3024 | ELSE |
---|
3025 | alloc_size = 1 |
---|
3026 | ENDIF |
---|
3027 | |
---|
3028 | ALLOCATE( grid_particles(kp,jp,ip)%particles(1:alloc_size) ) |
---|
3029 | |
---|
3030 | IF ( number_of_particles > 0 ) THEN |
---|
3031 | ALLOCATE( tmp_particles(1:number_of_particles) ) |
---|
3032 | READ ( 90 ) tmp_particles |
---|
3033 | grid_particles(kp,jp,ip)%particles(1:number_of_particles) = tmp_particles |
---|
3034 | DEALLOCATE( tmp_particles ) |
---|
3035 | IF ( number_of_particles < alloc_size ) THEN |
---|
3036 | grid_particles(kp,jp,ip)%particles(number_of_particles+1:alloc_size) & |
---|
3037 | = zero_particle |
---|
3038 | ENDIF |
---|
3039 | ELSE |
---|
3040 | grid_particles(kp,jp,ip)%particles(1:alloc_size) = zero_particle |
---|
3041 | ENDIF |
---|
3042 | |
---|
3043 | ENDDO |
---|
3044 | ENDDO |
---|
3045 | ENDDO |
---|
3046 | |
---|
3047 | CLOSE ( 90 ) |
---|
3048 | |
---|
3049 | ELSEIF ( restart_data_format_input(1:3) == 'mpi' ) THEN |
---|
3050 | |
---|
3051 | WRITE ( 9, * ) 'Here is MPI-IO praticle input ', rd_mpi_io_check_open() |
---|
3052 | FLUSH(9) |
---|
3053 | |
---|
3054 | ALLOCATE( prt_count(nzb:nzt+1,nysg:nyng,nxlg:nxrg), & |
---|
3055 | grid_particles(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3056 | |
---|
3057 | ALLOCATE( prt_global_index(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3058 | ! |
---|
3059 | !-- Open restart file for read, if not already open, and do not allow usage of shared-memory I/O |
---|
3060 | IF ( .NOT. rd_mpi_io_check_open() ) THEN |
---|
3061 | save_restart_data_format_input = restart_data_format_input |
---|
3062 | restart_data_format_input = 'mpi' |
---|
3063 | CALL rd_mpi_io_open( 'READ', 'BININ' ) |
---|
3064 | restart_data_format_input = save_restart_data_format_input |
---|
3065 | ENDIF |
---|
3066 | |
---|
3067 | CALL rd_mpi_io_particle_filetypes |
---|
3068 | |
---|
3069 | prt_count = 0 |
---|
3070 | CALL rrd_mpi_io( 'prt_count', prt_count ) |
---|
3071 | CALL rrd_mpi_io( 'prt_global_index', prt_global_index ) |
---|
3072 | |
---|
3073 | ! |
---|
3074 | !-- Allocate particles arrays |
---|
3075 | DO ip = nxl, nxr |
---|
3076 | DO jp = nys, nyn |
---|
3077 | DO kp = nzb+1, nzt |
---|
3078 | |
---|
3079 | number_of_particles = prt_count(kp,jp,ip) |
---|
3080 | IF ( number_of_particles > 0 ) THEN |
---|
3081 | alloc_size = MAX( INT( number_of_particles * & |
---|
3082 | ( 1.0_wp + alloc_factor / 100.0_wp ) ), & |
---|
3083 | 1 ) |
---|
3084 | ELSE |
---|
3085 | alloc_size = 1 |
---|
3086 | ENDIF |
---|
3087 | |
---|
3088 | ALLOCATE( grid_particles(kp,jp,ip)%particles(1:alloc_size) ) |
---|
3089 | |
---|
3090 | ENDDO |
---|
3091 | ENDDO |
---|
3092 | ENDDO |
---|
3093 | |
---|
3094 | ! |
---|
3095 | !-- Now read particle data from restart file |
---|
3096 | CALL rrd_mpi_io_particles( 'particles', prt_global_index ) |
---|
3097 | |
---|
3098 | IF ( .NOT. rd_mpi_io_check_open() ) THEN |
---|
3099 | ! |
---|
3100 | !-- Do not print header a second time to the debug file |
---|
3101 | save_debug_output = debug_output |
---|
3102 | debug_output = .FALSE. |
---|
3103 | CALL rd_mpi_io_close() |
---|
3104 | debug_output = save_debug_output |
---|
3105 | ENDIF |
---|
3106 | |
---|
3107 | DEALLOCATE( prt_global_index ) |
---|
3108 | |
---|
3109 | ENDIF |
---|
3110 | ! |
---|
3111 | !-- Must be called to sort particles into blocks, which is needed for a fast interpolation of the |
---|
3112 | !-- LES fields on the particle position. |
---|
3113 | CALL lpm_sort_and_delete |
---|
3114 | |
---|
3115 | END SUBROUTINE lpm_rrd_local_particles |
---|
3116 | |
---|
3117 | |
---|
3118 | !--------------------------------------------------------------------------------------------------! |
---|
3119 | ! Description: |
---|
3120 | ! ------------ |
---|
3121 | !> Read module-specific local restart data arrays (Fortran binary format). |
---|
3122 | !--------------------------------------------------------------------------------------------------! |
---|
3123 | SUBROUTINE lpm_rrd_local_ftn( k, nxlf, nxlc, nxl_on_file, nxrf, nxrc, nxr_on_file, nynf, nync, & |
---|
3124 | nyn_on_file, nysf, nysc, nys_on_file, tmp_3d, found ) |
---|
3125 | |
---|
3126 | |
---|
3127 | USE control_parameters, & |
---|
3128 | ONLY: length, restart_string |
---|
3129 | |
---|
3130 | INTEGER(iwp) :: k !< |
---|
3131 | INTEGER(iwp) :: nxlc !< |
---|
3132 | INTEGER(iwp) :: nxlf !< |
---|
3133 | INTEGER(iwp) :: nxl_on_file !< |
---|
3134 | INTEGER(iwp) :: nxrc !< |
---|
3135 | INTEGER(iwp) :: nxrf !< |
---|
3136 | INTEGER(iwp) :: nxr_on_file !< |
---|
3137 | INTEGER(iwp) :: nync !< |
---|
3138 | INTEGER(iwp) :: nynf !< |
---|
3139 | INTEGER(iwp) :: nyn_on_file !< |
---|
3140 | INTEGER(iwp) :: nysc !< |
---|
3141 | INTEGER(iwp) :: nysf !< |
---|
3142 | INTEGER(iwp) :: nys_on_file !< |
---|
3143 | |
---|
3144 | INTEGER(isp), DIMENSION(:,:,:), ALLOCATABLE :: tmp_2d_seq_random_particles !< temporary array for storing random generator |
---|
3145 | !< data for the lpm |
---|
3146 | |
---|
3147 | LOGICAL, INTENT(OUT) :: found |
---|
3148 | |
---|
3149 | REAL(wp), DIMENSION(nzb:nzt+1,nys_on_file-nbgp:nyn_on_file+nbgp,nxl_on_file-nbgp:nxr_on_file+nbgp) :: tmp_3d !< |
---|
3150 | |
---|
3151 | |
---|
3152 | found = .TRUE. |
---|
3153 | |
---|
3154 | SELECT CASE ( restart_string(1:length) ) |
---|
3155 | |
---|
3156 | CASE ( 'pc_av' ) |
---|
3157 | IF ( .NOT. ALLOCATED( pc_av ) ) THEN |
---|
3158 | ALLOCATE( pc_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3159 | ENDIF |
---|
3160 | IF ( k == 1 ) READ ( 13 ) tmp_3d |
---|
3161 | pc_av(:,nysc-nbgp:nync+nbgp,nxlc-nbgp:nxrc+nbgp) = & |
---|
3162 | tmp_3d(:,nysf-nbgp:nynf+nbgp,nxlf-nbgp:nxrf+nbgp) |
---|
3163 | |
---|
3164 | CASE ( 'pr_av' ) |
---|
3165 | IF ( .NOT. ALLOCATED( pr_av ) ) THEN |
---|
3166 | ALLOCATE( pr_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3167 | ENDIF |
---|
3168 | IF ( k == 1 ) READ ( 13 ) tmp_3d |
---|
3169 | pr_av(:,nysc-nbgp:nync+nbgp,nxlc-nbgp:nxrc+nbgp) = & |
---|
3170 | tmp_3d(:,nysf-nbgp:nynf+nbgp,nxlf-nbgp:nxrf+nbgp) |
---|
3171 | |
---|
3172 | CASE ( 'ql_c_av' ) |
---|
3173 | IF ( .NOT. ALLOCATED( ql_c_av ) ) THEN |
---|
3174 | ALLOCATE( ql_c_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3175 | ENDIF |
---|
3176 | IF ( k == 1 ) READ ( 13 ) tmp_3d |
---|
3177 | ql_c_av(:,nysc-nbgp:nync+nbgp,nxlc-nbgp:nxrc+nbgp) = & |
---|
3178 | tmp_3d(:,nysf-nbgp:nynf+nbgp,nxlf-nbgp:nxrf+nbgp) |
---|
3179 | |
---|
3180 | CASE ( 'ql_v_av' ) |
---|
3181 | IF ( .NOT. ALLOCATED( ql_v_av ) ) THEN |
---|
3182 | ALLOCATE( ql_v_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3183 | ENDIF |
---|
3184 | IF ( k == 1 ) READ ( 13 ) tmp_3d |
---|
3185 | ql_v_av(:,nysc-nbgp:nync+nbgp,nxlc-nbgp:nxrc+nbgp) = & |
---|
3186 | tmp_3d(:,nysf-nbgp:nynf+nbgp,nxlf-nbgp:nxrf+nbgp) |
---|
3187 | |
---|
3188 | CASE ( 'ql_vp_av' ) |
---|
3189 | IF ( .NOT. ALLOCATED( ql_vp_av ) ) THEN |
---|
3190 | ALLOCATE( ql_vp_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3191 | ENDIF |
---|
3192 | IF ( k == 1 ) READ ( 13 ) tmp_3d |
---|
3193 | ql_vp_av(:,nysc-nbgp:nync+nbgp,nxlc-nbgp:nxrc+nbgp) = & |
---|
3194 | tmp_3d(:,nysf-nbgp:nynf+nbgp,nxlf-nbgp:nxrf+nbgp) |
---|
3195 | |
---|
3196 | CASE ( 'seq_random_array_particles' ) |
---|
3197 | ALLOCATE( tmp_2d_seq_random_particles(5,nys_on_file:nyn_on_file,nxl_on_file:nxr_on_file) ) |
---|
3198 | IF ( .NOT. ALLOCATED( seq_random_array_particles ) ) THEN |
---|
3199 | ALLOCATE( seq_random_array_particles(5,nys:nyn,nxl:nxr) ) |
---|
3200 | ENDIF |
---|
3201 | IF ( k == 1 ) READ ( 13 ) tmp_2d_seq_random_particles |
---|
3202 | seq_random_array_particles(:,nysc:nync,nxlc:nxrc) = & |
---|
3203 | tmp_2d_seq_random_particles(:,nysf:nynf,nxlf:nxrf) |
---|
3204 | DEALLOCATE( tmp_2d_seq_random_particles ) |
---|
3205 | |
---|
3206 | CASE DEFAULT |
---|
3207 | |
---|
3208 | found = .FALSE. |
---|
3209 | |
---|
3210 | END SELECT |
---|
3211 | |
---|
3212 | END SUBROUTINE lpm_rrd_local_ftn |
---|
3213 | |
---|
3214 | |
---|
3215 | !--------------------------------------------------------------------------------------------------! |
---|
3216 | ! Description: |
---|
3217 | ! ------------ |
---|
3218 | !> Read module-specific local restart data arrays (MPI-IO). |
---|
3219 | !--------------------------------------------------------------------------------------------------! |
---|
3220 | SUBROUTINE lpm_rrd_local_mpi |
---|
3221 | |
---|
3222 | IMPLICIT NONE |
---|
3223 | |
---|
3224 | CHARACTER (LEN=32) :: tmp_name !< temporary variable |
---|
3225 | |
---|
3226 | INTEGER(iwp) :: i !< loop index |
---|
3227 | |
---|
3228 | LOGICAL :: array_found !< |
---|
3229 | |
---|
3230 | CALL rd_mpi_io_check_array( 'pc_av' , found = array_found ) |
---|
3231 | IF ( array_found ) THEN |
---|
3232 | IF ( .NOT. ALLOCATED( pc_av ) ) ALLOCATE( pc_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3233 | CALL rrd_mpi_io( 'pc_av', pc_av ) |
---|
3234 | ENDIF |
---|
3235 | |
---|
3236 | CALL rd_mpi_io_check_array( 'pr_av' , found = array_found ) |
---|
3237 | IF ( array_found ) THEN |
---|
3238 | IF ( .NOT. ALLOCATED( pr_av ) ) ALLOCATE( pr_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3239 | CALL rrd_mpi_io( 'pr_av', pr_av ) |
---|
3240 | ENDIF |
---|
3241 | |
---|
3242 | CALL rd_mpi_io_check_array( 'ql_c_av' , found = array_found ) |
---|
3243 | IF ( array_found ) THEN |
---|
3244 | IF ( .NOT. ALLOCATED( ql_c_av ) ) ALLOCATE( ql_c_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3245 | CALL rrd_mpi_io( 'ql_c_av', ql_c_av ) |
---|
3246 | ENDIF |
---|
3247 | |
---|
3248 | CALL rd_mpi_io_check_array( 'ql_v_av' , found = array_found ) |
---|
3249 | IF ( array_found ) THEN |
---|
3250 | IF ( .NOT. ALLOCATED( ql_v_av ) ) ALLOCATE( ql_v_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3251 | CALL rrd_mpi_io( 'ql_v_av', ql_v_av ) |
---|
3252 | ENDIF |
---|
3253 | |
---|
3254 | CALL rd_mpi_io_check_array( 'ql_vp_av' , found = array_found ) |
---|
3255 | IF ( array_found ) THEN |
---|
3256 | IF ( .NOT. ALLOCATED( ql_vp_av ) ) ALLOCATE( ql_vp_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3257 | CALL rrd_mpi_io( 'ql_vp_av', ql_vp_av ) |
---|
3258 | ENDIF |
---|
3259 | |
---|
3260 | CALL rd_mpi_io_check_array( 'seq_random_array_particles01' , found = array_found ) |
---|
3261 | IF ( array_found ) THEN |
---|
3262 | IF ( .NOT. ALLOCATED( seq_random_array_particles ) ) THEN |
---|
3263 | ALLOCATE( seq_random_array_particles(5,nys:nyn,nxl:nxr) ) |
---|
3264 | ENDIF |
---|
3265 | DO i = 1, SIZE( seq_random_array_particles, 1 ) |
---|
3266 | WRITE( tmp_name, '(A,I2.2)' ) 'seq_random_array_particles', i |
---|
3267 | CALL rrd_mpi_io( TRIM(tmp_name), seq_random_array_particles(i,:,:) ) |
---|
3268 | ENDDO |
---|
3269 | ENDIF |
---|
3270 | |
---|
3271 | END SUBROUTINE lpm_rrd_local_mpi |
---|
3272 | |
---|
3273 | |
---|
3274 | !--------------------------------------------------------------------------------------------------! |
---|
3275 | ! Description: |
---|
3276 | ! ------------ |
---|
3277 | !> This routine writes the respective restart data for the lpm. |
---|
3278 | !--------------------------------------------------------------------------------------------------! |
---|
3279 | SUBROUTINE lpm_wrd_local |
---|
3280 | |
---|
3281 | CHARACTER (LEN=10) :: particle_binary_version !< |
---|
3282 | CHARACTER (LEN=32) :: tmp_name !< temporary variable |
---|
3283 | |
---|
3284 | INTEGER(iwp) :: i !< loop index |
---|
3285 | INTEGER(iwp) :: ip !< |
---|
3286 | INTEGER(iwp) :: j !< loop index |
---|
3287 | INTEGER(iwp) :: jp !< |
---|
3288 | INTEGER(iwp) :: k !< loop index |
---|
3289 | INTEGER(iwp) :: kp !< |
---|
3290 | |
---|
3291 | #if defined( __parallel ) |
---|
3292 | INTEGER :: ierr !< |
---|
3293 | #endif |
---|
3294 | INTEGER(iwp) :: start_index !< |
---|
3295 | INTEGER(iwp) :: start_index_on_pe !< |
---|
3296 | |
---|
3297 | INTEGER(iwp), DIMENSION(0:numprocs-1) :: nr_particles_global |
---|
3298 | INTEGER(iwp), DIMENSION(0:numprocs-1) :: nr_particles_local |
---|
3299 | |
---|
3300 | INTEGER(idp), ALLOCATABLE, DIMENSION(:,:,:) :: prt_global_index |
---|
3301 | |
---|
3302 | |
---|
3303 | IF ( TRIM( restart_data_format_output ) == 'fortran_binary' ) THEN |
---|
3304 | ! |
---|
3305 | !-- First open the output unit. |
---|
3306 | IF ( myid_char == '' ) THEN |
---|
3307 | OPEN ( 90, FILE='PARTICLE_RESTART_DATA_OUT'//myid_char, FORM='UNFORMATTED') |
---|
3308 | ELSE |
---|
3309 | IF ( myid == 0 ) CALL local_system( 'mkdir PARTICLE_RESTART_DATA_OUT' ) |
---|
3310 | #if defined( __parallel ) |
---|
3311 | ! |
---|
3312 | !-- Set a barrier in order to allow that thereafter all other processors in the directory |
---|
3313 | !-- created by PE0 can open their file |
---|
3314 | CALL MPI_BARRIER( comm2d, ierr ) |
---|
3315 | #endif |
---|
3316 | OPEN ( 90, FILE='PARTICLE_RESTART_DATA_OUT/'//myid_char, FORM='UNFORMATTED' ) |
---|
3317 | ENDIF |
---|
3318 | |
---|
3319 | ! |
---|
3320 | !-- Write the version number of the binary format. |
---|
3321 | !-- Attention: After changes to the following output commands the version number of the variable |
---|
3322 | !-- --------- particle_binary_version must be changed! Also, the version number and the list of |
---|
3323 | !-- arrays to be read in lpm_read_restart_file must be adjusted accordingly. |
---|
3324 | particle_binary_version = '4.0' |
---|
3325 | WRITE ( 90 ) particle_binary_version |
---|
3326 | |
---|
3327 | ! |
---|
3328 | !-- Write some particle parameters, the size of the particle arrays |
---|
3329 | WRITE ( 90 ) bc_par_b, bc_par_lr, bc_par_ns, bc_par_t, last_particle_release_time, & |
---|
3330 | number_of_particle_groups, particle_groups, time_write_particle_data |
---|
3331 | |
---|
3332 | WRITE ( 90 ) prt_count |
---|
3333 | |
---|
3334 | DO ip = nxl, nxr |
---|
3335 | DO jp = nys, nyn |
---|
3336 | DO kp = nzb+1, nzt |
---|
3337 | number_of_particles = prt_count(kp,jp,ip) |
---|
3338 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
3339 | IF ( number_of_particles <= 0 ) CYCLE |
---|
3340 | WRITE ( 90 ) particles |
---|
3341 | ENDDO |
---|
3342 | ENDDO |
---|
3343 | ENDDO |
---|
3344 | |
---|
3345 | CLOSE ( 90 ) |
---|
3346 | |
---|
3347 | #if defined( __parallel ) |
---|
3348 | CALL MPI_BARRIER( comm2d, ierr ) |
---|
3349 | #endif |
---|
3350 | |
---|
3351 | |
---|
3352 | IF ( ALLOCATED( seq_random_array_particles ) ) THEN |
---|
3353 | CALL wrd_write_string( 'seq_random_array_particles' ) |
---|
3354 | WRITE ( 14 ) seq_random_array_particles |
---|
3355 | ENDIF |
---|
3356 | |
---|
3357 | |
---|
3358 | ELSEIF ( restart_data_format_output(1:3) == 'mpi' ) THEN |
---|
3359 | |
---|
3360 | |
---|
3361 | IF ( ALLOCATED( seq_random_array_particles ) ) THEN |
---|
3362 | DO i = 1, SIZE( seq_random_array_particles, 1 ) |
---|
3363 | WRITE( tmp_name, '(A,I2.2)' ) 'seq_random_array_particles', i |
---|
3364 | CALL wrd_mpi_io( TRIM( tmp_name ), seq_random_array_particles(i,:,:) ) |
---|
3365 | ENDDO |
---|
3366 | ENDIF |
---|
3367 | |
---|
3368 | ALLOCATE( prt_global_index(nzb:nzt+1,nysg:nyng,nxlg:nxrg) ) |
---|
3369 | |
---|
3370 | #if defined( __parallel ) |
---|
3371 | !-- TODO: needs to be replaced by standard PALM message |
---|
3372 | IF ( TRIM( restart_data_format_output ) == 'mpi_shared_memory' ) THEN |
---|
3373 | WRITE( 9, * ) 'mpi_shared_memory is NOT implemented yet for particle IO' |
---|
3374 | FLUSH( 9 ) |
---|
3375 | CALL MPI_ABORT( MPI_COMM_WORLD, 1, ierr ) |
---|
3376 | ENDIF |
---|
3377 | #endif |
---|
3378 | |
---|
3379 | CALL rd_mpi_io_particle_filetypes |
---|
3380 | |
---|
3381 | nr_particles_local = 0 |
---|
3382 | nr_particles_local(myid) = SUM( prt_count ) |
---|
3383 | |
---|
3384 | #if defined( __parallel ) |
---|
3385 | CALL MPI_ALLREDUCE( nr_particles_local, nr_particles_global, numprocs, MPI_INTEGER, MPI_SUM,& |
---|
3386 | comm2d, ierr ) |
---|
3387 | #else |
---|
3388 | nr_particles_global = nr_particles_local |
---|
3389 | #endif |
---|
3390 | |
---|
3391 | start_index_on_pe = 0 |
---|
3392 | IF ( myid > 0 ) THEN |
---|
3393 | DO i = 1, myid |
---|
3394 | start_index_on_pe = start_index_on_pe + nr_particles_global(i-1) |
---|
3395 | ENDDO |
---|
3396 | ENDIF |
---|
3397 | |
---|
3398 | CALL wrd_mpi_io( 'prt_count', prt_count ) |
---|
3399 | |
---|
3400 | start_index = start_index_on_pe |
---|
3401 | DO i = nxl, nxr |
---|
3402 | DO j = nys, nyn |
---|
3403 | DO k = nzb, nzt+1 |
---|
3404 | prt_global_index(k,j,i) = start_index |
---|
3405 | start_index = start_index + prt_count(k,j,i) |
---|
3406 | ENDDO |
---|
3407 | ENDDO |
---|
3408 | ENDDO |
---|
3409 | |
---|
3410 | CALL wrd_mpi_io( 'prt_global_index', prt_global_index ) |
---|
3411 | CALL wrd_mpi_io_particles( 'particles', prt_global_index ) |
---|
3412 | |
---|
3413 | DEALLOCATE( prt_global_index ) |
---|
3414 | |
---|
3415 | ENDIF |
---|
3416 | |
---|
3417 | END SUBROUTINE lpm_wrd_local |
---|
3418 | |
---|
3419 | |
---|
3420 | !--------------------------------------------------------------------------------------------------! |
---|
3421 | ! Description: |
---|
3422 | ! ------------ |
---|
3423 | !> This routine writes the respective restart data for the lpm. |
---|
3424 | !--------------------------------------------------------------------------------------------------! |
---|
3425 | SUBROUTINE lpm_wrd_global |
---|
3426 | |
---|
3427 | #if defined( __parallel ) |
---|
3428 | INTEGER :: ierr !< |
---|
3429 | #endif |
---|
3430 | |
---|
3431 | REAL(wp), DIMENSION(4,max_number_of_particle_groups) :: particle_groups_array !< |
---|
3432 | |
---|
3433 | |
---|
3434 | IF ( TRIM( restart_data_format_output ) == 'fortran_binary' ) THEN |
---|
3435 | |
---|
3436 | CALL wrd_write_string( 'curvature_solution_effects' ) |
---|
3437 | WRITE ( 14 ) curvature_solution_effects |
---|
3438 | |
---|
3439 | CALL wrd_write_string( 'interpolation_simple_corrector' ) |
---|
3440 | WRITE ( 14 ) interpolation_simple_corrector |
---|
3441 | |
---|
3442 | CALL wrd_write_string( 'interpolation_simple_predictor' ) |
---|
3443 | WRITE ( 14 ) interpolation_simple_predictor |
---|
3444 | |
---|
3445 | CALL wrd_write_string( 'interpolation_trilinear' ) |
---|
3446 | WRITE ( 14 ) interpolation_trilinear |
---|
3447 | |
---|
3448 | ELSEIF ( restart_data_format_output(1:3) == 'mpi' ) THEN |
---|
3449 | |
---|
3450 | CALL wrd_mpi_io( 'curvature_solution_effects', curvature_solution_effects ) |
---|
3451 | CALL wrd_mpi_io( 'interpolation_simple_corrector', interpolation_simple_corrector ) |
---|
3452 | CALL wrd_mpi_io( 'interpolation_simple_predictor', interpolation_simple_predictor ) |
---|
3453 | CALL wrd_mpi_io( 'interpolation_trilinear', interpolation_trilinear ) |
---|
3454 | ! |
---|
3455 | !-- Write some global particle parameters |
---|
3456 | !-- In case of Fortran binary format, these variables are written to unit 90 |
---|
3457 | CALL wrd_mpi_io( 'bc_par_b', bc_par_b ) |
---|
3458 | CALL wrd_mpi_io( 'bc_par_lr', bc_par_lr ) |
---|
3459 | CALL wrd_mpi_io( 'bc_par_ns', bc_par_ns ) |
---|
3460 | CALL wrd_mpi_io( 'bc_par_t', bc_par_t ) |
---|
3461 | CALL wrd_mpi_io( 'last_particle_release_time', last_particle_release_time ) |
---|
3462 | CALL wrd_mpi_io( 'number_of_particle_groups', number_of_particle_groups ) |
---|
3463 | CALL wrd_mpi_io( 'time_write_particle_data', time_write_particle_data ) |
---|
3464 | |
---|
3465 | ! |
---|
3466 | !-- Write particle_group informations via 2D array to avoid another overlay in |
---|
3467 | !-- restart_data_mpi_io_mod. |
---|
3468 | !-- TODO: convert the following to a standard PALM message |
---|
3469 | IF( STORAGE_SIZE( particle_groups(1) ) / (wp*8) /= 4 ) THEN |
---|
3470 | WRITE( 9, * ) 'size of structure particle_groups_type has changed ' |
---|
3471 | FLUSH( 9 ) |
---|
3472 | #if defined( __parallel ) |
---|
3473 | CALL MPI_ABORT( MPI_COMM_WORLD, 1, ierr ) |
---|
3474 | #else |
---|
3475 | STOP 'error' |
---|
3476 | #endif |
---|
3477 | ENDIF |
---|
3478 | |
---|
3479 | particle_groups_array(1,:) = particle_groups(:)%density_ratio |
---|
3480 | particle_groups_array(2,:) = particle_groups(:)%radius |
---|
3481 | particle_groups_array(3,:) = particle_groups(:)%exp_arg |
---|
3482 | particle_groups_array(4,:) = particle_groups(:)%exp_term |
---|
3483 | |
---|
3484 | CALL wrd_mpi_io_global_array( 'particle_groups', particle_groups_array ) |
---|
3485 | |
---|
3486 | ENDIF |
---|
3487 | |
---|
3488 | END SUBROUTINE lpm_wrd_global |
---|
3489 | |
---|
3490 | |
---|
3491 | !--------------------------------------------------------------------------------------------------! |
---|
3492 | ! Description: |
---|
3493 | ! ------------ |
---|
3494 | !> Read module-specific global restart data (Fortran binary format). |
---|
3495 | !--------------------------------------------------------------------------------------------------! |
---|
3496 | SUBROUTINE lpm_rrd_global_ftn( found ) |
---|
3497 | |
---|
3498 | USE control_parameters, & |
---|
3499 | ONLY: length, restart_string |
---|
3500 | |
---|
3501 | LOGICAL, INTENT(OUT) :: found |
---|
3502 | |
---|
3503 | found = .TRUE. |
---|
3504 | |
---|
3505 | SELECT CASE ( restart_string(1:length) ) |
---|
3506 | |
---|
3507 | CASE ( 'curvature_solution_effects' ) |
---|
3508 | READ ( 13 ) curvature_solution_effects |
---|
3509 | |
---|
3510 | CASE ( 'interpolation_simple_corrector' ) |
---|
3511 | READ ( 13 ) interpolation_simple_corrector |
---|
3512 | |
---|
3513 | CASE ( 'interpolation_simple_predictor' ) |
---|
3514 | READ ( 13 ) interpolation_simple_predictor |
---|
3515 | |
---|
3516 | CASE ( 'interpolation_trilinear' ) |
---|
3517 | READ ( 13 ) interpolation_trilinear |
---|
3518 | |
---|
3519 | CASE DEFAULT |
---|
3520 | |
---|
3521 | found = .FALSE. |
---|
3522 | |
---|
3523 | END SELECT |
---|
3524 | |
---|
3525 | END SUBROUTINE lpm_rrd_global_ftn |
---|
3526 | |
---|
3527 | |
---|
3528 | !--------------------------------------------------------------------------------------------------! |
---|
3529 | ! Description: |
---|
3530 | ! ------------ |
---|
3531 | !> Read module-specific global restart data (MPI-IO). |
---|
3532 | !--------------------------------------------------------------------------------------------------! |
---|
3533 | SUBROUTINE lpm_rrd_global_mpi |
---|
3534 | |
---|
3535 | #if defined( __parallel ) |
---|
3536 | INTEGER :: ierr !< |
---|
3537 | #endif |
---|
3538 | |
---|
3539 | REAL(wp), DIMENSION(4,max_number_of_particle_groups) :: particle_groups_array !< |
---|
3540 | |
---|
3541 | |
---|
3542 | CALL rrd_mpi_io( 'curvature_solution_effects', curvature_solution_effects ) |
---|
3543 | CALL rrd_mpi_io( 'interpolation_simple_corrector', interpolation_simple_corrector ) |
---|
3544 | CALL rrd_mpi_io( 'interpolation_simple_predictor', interpolation_simple_predictor ) |
---|
3545 | CALL rrd_mpi_io( 'interpolation_trilinear', interpolation_trilinear ) |
---|
3546 | ! |
---|
3547 | !-- Read some particle parameters. |
---|
3548 | !-- In case of Fortran binary format, these variables are read from unit 90. |
---|
3549 | CALL rrd_mpi_io( 'bc_par_b', bc_par_b ) |
---|
3550 | CALL rrd_mpi_io( 'bc_par_lr', bc_par_lr ) |
---|
3551 | CALL rrd_mpi_io( 'bc_par_ns', bc_par_ns ) |
---|
3552 | CALL rrd_mpi_io( 'bc_par_t', bc_par_t ) |
---|
3553 | CALL rrd_mpi_io( 'last_particle_release_time', last_particle_release_time ) |
---|
3554 | CALL rrd_mpi_io( 'number_of_particle_groups', number_of_particle_groups ) |
---|
3555 | CALL rrd_mpi_io( 'time_write_particle_data', time_write_particle_data ) |
---|
3556 | |
---|
3557 | ! |
---|
3558 | !-- Read particle group information via 2d-array to avoid another overlay in |
---|
3559 | !-- restart_data_mpi_io_mod. |
---|
3560 | !-- TODO: convert the following to a standard PALM message |
---|
3561 | IF ( STORAGE_SIZE( particle_groups(1) ) / (wp*8) /= 4 ) THEN |
---|
3562 | WRITE( 9, * ) 'size of structure particle_groups_type has changed ' |
---|
3563 | FLUSH( 9 ) |
---|
3564 | #if defined( __parallel ) |
---|
3565 | CALL MPI_ABORT( MPI_COMM_WORLD, 1, ierr ) |
---|
3566 | #else |
---|
3567 | STOP 'error' |
---|
3568 | #endif |
---|
3569 | ENDIF |
---|
3570 | |
---|
3571 | CALL rrd_mpi_io_global_array( 'particle_groups', particle_groups_array ) |
---|
3572 | |
---|
3573 | particle_groups(:)%density_ratio = particle_groups_array(1,:) |
---|
3574 | particle_groups(:)%radius = particle_groups_array(2,:) |
---|
3575 | particle_groups(:)%exp_arg = particle_groups_array(3,:) |
---|
3576 | particle_groups(:)%exp_term = particle_groups_array(4,:) |
---|
3577 | |
---|
3578 | END SUBROUTINE lpm_rrd_global_mpi |
---|
3579 | |
---|
3580 | |
---|
3581 | !--------------------------------------------------------------------------------------------------! |
---|
3582 | ! Description: |
---|
3583 | ! ------------ |
---|
3584 | !> This is a submodule of the lagrangian particle model. It contains all dynamic processes of the |
---|
3585 | !> lpm. This includes the advection (resolved and sub-grid scale) as well as the boundary conditions |
---|
3586 | !> of particles. As a next step this submodule should be excluded as an own file. |
---|
3587 | !--------------------------------------------------------------------------------------------------! |
---|
3588 | SUBROUTINE lpm_advec (ip,jp,kp) |
---|
3589 | |
---|
3590 | REAL(wp), PARAMETER :: a_rog = 9.65_wp !< parameter for fall velocity |
---|
3591 | REAL(wp), PARAMETER :: b_rog = 10.43_wp !< parameter for fall velocity |
---|
3592 | REAL(wp), PARAMETER :: c_rog = 0.6_wp !< parameter for fall velocity |
---|
3593 | REAL(wp), PARAMETER :: k_cap_rog = 4.0_wp !< parameter for fall velocity |
---|
3594 | REAL(wp), PARAMETER :: k_low_rog = 12.0_wp !< parameter for fall velocity |
---|
3595 | REAL(wp), PARAMETER :: d0_rog = 0.745_wp !< separation diameter |
---|
3596 | |
---|
3597 | INTEGER(iwp) :: i !< index variable along x |
---|
3598 | INTEGER(iwp) :: i_next !< index variable along x |
---|
3599 | INTEGER(iwp) :: ip !< index variable along x |
---|
3600 | INTEGER(iwp) :: iteration_steps = 1 !< amount of iterations steps for corrector step |
---|
3601 | INTEGER(iwp) :: j !< index variable along y |
---|
3602 | INTEGER(iwp) :: j_next !< index variable along y |
---|
3603 | INTEGER(iwp) :: jp !< index variable along y |
---|
3604 | INTEGER(iwp) :: k !< index variable along z |
---|
3605 | INTEGER(iwp) :: k_wall !< vertical index of topography top |
---|
3606 | INTEGER(iwp) :: kp !< index variable along z |
---|
3607 | INTEGER(iwp) :: k_next !< index variable along z |
---|
3608 | INTEGER(iwp) :: kw !< index variable along z |
---|
3609 | INTEGER(iwp) :: kkw !< index variable along z |
---|
3610 | INTEGER(iwp) :: n !< loop variable over all particles in a grid box |
---|
3611 | INTEGER(iwp) :: nn !< loop variable over iterations steps |
---|
3612 | INTEGER(iwp) :: nb !< block number particles are sorted in |
---|
3613 | INTEGER(iwp) :: particle_end !< end index for partilce loop |
---|
3614 | INTEGER(iwp) :: particle_start !< start index for particle loop |
---|
3615 | INTEGER(iwp) :: subbox_end !< end index for loop over subboxes in particle advection |
---|
3616 | INTEGER(iwp) :: subbox_start !< start index for loop over subboxes in particle advection |
---|
3617 | |
---|
3618 | INTEGER(iwp), DIMENSION(0:7) :: end_index !< start particle index for current block |
---|
3619 | INTEGER(iwp), DIMENSION(0:7) :: start_index !< start particle index for current block |
---|
3620 | |
---|
3621 | LOGICAL :: subbox_at_wall !< flag to see if the current subgridbox is adjacent to a wall |
---|
3622 | |
---|
3623 | REAL(wp) :: aa !< dummy argument for horizontal particle interpolation |
---|
3624 | REAL(wp) :: alpha !< interpolation facor for x-direction |
---|
3625 | REAL(wp) :: bb !< dummy argument for horizontal particle interpolation |
---|
3626 | REAL(wp) :: beta !< interpolation facor for y-direction |
---|
3627 | REAL(wp) :: cc !< dummy argument for horizontal particle interpolation |
---|
3628 | REAL(wp) :: d_z_p_z0 !< inverse of interpolation length for logarithmic interpolation |
---|
3629 | REAL(wp) :: dd !< dummy argument for horizontal particle interpolation |
---|
3630 | REAL(wp) :: de_dx_int_l !< x/y-interpolated TKE gradient (x) at particle position at lower vertical level |
---|
3631 | REAL(wp) :: de_dx_int_u !< x/y-interpolated TKE gradient (x) at particle position at upper vertical level |
---|
3632 | REAL(wp) :: de_dy_int_l !< x/y-interpolated TKE gradient (y) at particle position at lower vertical level |
---|
3633 | REAL(wp) :: de_dy_int_u !< x/y-interpolated TKE gradient (y) at particle position at upper vertical level |
---|
3634 | REAL(wp) :: de_dt !< temporal derivative of TKE experienced by the particle |
---|
3635 | REAL(wp) :: de_dt_min !< lower level for temporal TKE derivative |
---|
3636 | REAL(wp) :: de_dz_int_l !< x/y-interpolated TKE gradient (z) at particle position at lower vertical level |
---|
3637 | REAL(wp) :: de_dz_int_u !< x/y-interpolated TKE gradient (z) at particle position at upper vertical level |
---|
3638 | REAL(wp) :: diameter !< diamter of droplet |
---|
3639 | REAL(wp) :: diss_int_l !< x/y-interpolated dissipation at particle position at lower vertical level |
---|
3640 | REAL(wp) :: diss_int_u !< x/y-interpolated dissipation at particle position at upper vertical level |
---|
3641 | REAL(wp) :: dt_particle_m !< previous particle time step |
---|
3642 | REAL(wp) :: dz_temp !< dummy for the vertical grid spacing |
---|
3643 | REAL(wp) :: e_int_l !< x/y-interpolated TKE at particle position at lower vertical level |
---|
3644 | REAL(wp) :: e_int_u !< x/y-interpolated TKE at particle position at upper vertical level |
---|
3645 | REAL(wp) :: e_mean_int !< horizontal mean TKE at particle height |
---|
3646 | REAL(wp) :: exp_arg !< argument in the exponent - particle radius |
---|
3647 | REAL(wp) :: exp_term !< exponent term |
---|
3648 | REAL(wp) :: gamma !< interpolation facor for z-direction |
---|
3649 | REAL(wp) :: gg !< dummy argument for horizontal particle interpolation |
---|
3650 | REAL(wp) :: height_p !< dummy argument for logarithmic interpolation |
---|
3651 | REAL(wp) :: log_z_z0_int !< logarithmus used for surface_layer interpolation |
---|
3652 | REAL(wp) :: rl !< Lagrangian autocorrelation coefficient |
---|
3653 | REAL(wp) :: rg1 !< Gaussian distributed random number |
---|
3654 | REAL(wp) :: rg2 !< Gaussian distributed random number |
---|
3655 | REAL(wp) :: rg3 !< Gaussian distributed random number |
---|
3656 | REAL(wp) :: sigma !< velocity standard deviation |
---|
3657 | REAL(wp) :: u_int_l !< x/y-interpolated u-component at particle position at lower vertical level |
---|
3658 | REAL(wp) :: u_int_u !< x/y-interpolated u-component at particle position at upper vertical level |
---|
3659 | REAL(wp) :: unext !< calculated particle u-velocity of corrector step |
---|
3660 | REAL(wp) :: us_int !< friction velocity at particle grid box |
---|
3661 | REAL(wp) :: v_int_l !< x/y-interpolated v-component at particle position at lower vertical level |
---|
3662 | REAL(wp) :: v_int_u !< x/y-interpolated v-component at particle position at upper vertical level |
---|
3663 | REAL(wp) :: vnext !< calculated particle v-velocity of corrector step |
---|
3664 | REAL(wp) :: vv_int !< dummy to compute interpolated mean SGS TKE, used to scale SGS advection |
---|
3665 | REAL(wp) :: w_int_l !< x/y-interpolated w-component at particle position at lower vertical level |
---|
3666 | REAL(wp) :: w_int_u !< x/y-interpolated w-component at particle position at upper vertical level |
---|
3667 | REAL(wp) :: wnext !< calculated particle w-velocity of corrector step |
---|
3668 | REAL(wp) :: w_s !< terminal velocity of droplets |
---|
3669 | REAL(wp) :: x !< dummy argument for horizontal particle interpolation |
---|
3670 | REAL(wp) :: xp !< calculated particle position in x of predictor step |
---|
3671 | REAL(wp) :: y !< dummy argument for horizontal particle interpolation |
---|
3672 | REAL(wp) :: yp !< calculated particle position in y of predictor step |
---|
3673 | REAL(wp) :: z_p !< surface layer height (0.5 dz) |
---|
3674 | REAL(wp) :: zp !< calculated particle position in z of predictor step |
---|
3675 | |
---|
3676 | REAL(wp), DIMENSION(number_of_particles) :: de_dx_int !< horizontal TKE gradient along x at particle position |
---|
3677 | REAL(wp), DIMENSION(number_of_particles) :: de_dy_int !< horizontal TKE gradient along y at particle position |
---|
3678 | REAL(wp), DIMENSION(number_of_particles) :: de_dz_int !< horizontal TKE gradient along z at particle position |
---|
3679 | REAL(wp), DIMENSION(number_of_particles) :: dens_ratio !< ratio between the density of the fluid and the density of the |
---|
3680 | !< particles |
---|
3681 | REAL(wp), DIMENSION(number_of_particles) :: diss_int !< dissipation at particle position |
---|
3682 | REAL(wp), DIMENSION(number_of_particles) :: dt_gap !< remaining time until particle time integration reaches LES time |
---|
3683 | REAL(wp), DIMENSION(number_of_particles) :: dt_particle !< particle time step |
---|
3684 | REAL(wp), DIMENSION(number_of_particles) :: e_int !< TKE at particle position |
---|
3685 | REAL(wp), DIMENSION(number_of_particles) :: fs_int !< weighting factor for subgrid-scale particle speed |
---|
3686 | REAL(wp), DIMENSION(number_of_particles) :: lagr_timescale !< Lagrangian timescale |
---|
3687 | REAL(wp), DIMENSION(number_of_particles) :: rvar1_temp !< SGS particle velocity - u-component |
---|
3688 | REAL(wp), DIMENSION(number_of_particles) :: rvar2_temp !< SGS particle velocity - v-component |
---|
3689 | REAL(wp), DIMENSION(number_of_particles) :: rvar3_temp !< SGS particle velocity - w-component |
---|
3690 | REAL(wp), DIMENSION(number_of_particles) :: term_1_2 !< flag to communicate whether a particle is near topography or not |
---|
3691 | REAL(wp), DIMENSION(number_of_particles) :: u_int !< u-component of particle speed |
---|
3692 | REAL(wp), DIMENSION(number_of_particles) :: v_int !< v-component of particle speed |
---|
3693 | REAL(wp), DIMENSION(number_of_particles) :: w_int !< w-component of particle speed |
---|
3694 | REAL(wp), DIMENSION(number_of_particles) :: xv !< x-position |
---|
3695 | REAL(wp), DIMENSION(number_of_particles) :: yv !< y-position |
---|
3696 | REAL(wp), DIMENSION(number_of_particles) :: zv !< z-position |
---|
3697 | |
---|
3698 | REAL(wp), DIMENSION(number_of_particles, 3) :: rg !< vector of Gaussian distributed random numbers |
---|
3699 | |
---|
3700 | CALL cpu_log( log_point_s(44), 'lpm_advec', 'continue' ) |
---|
3701 | ! |
---|
3702 | !-- Determine height of Prandtl layer and distance between Prandtl-layer height and horizontal mean |
---|
3703 | !-- roughness height, which are required for vertical logarithmic interpolation of horizontal |
---|
3704 | !-- particle speeds (for particles below first vertical grid level). |
---|
3705 | z_p = zu(nzb+1) - zw(nzb) |
---|
3706 | d_z_p_z0 = 1.0_wp / ( z_p - z0_av_global ) |
---|
3707 | |
---|
3708 | xv = particles(1:number_of_particles)%x |
---|
3709 | yv = particles(1:number_of_particles)%y |
---|
3710 | zv = particles(1:number_of_particles)%z |
---|
3711 | dt_particle = dt_3d |
---|
3712 | |
---|
3713 | ! |
---|
3714 | !-- This case uses a simple interpolation method for the particle velocites, and applying a |
---|
3715 | !-- predictor-corrector method. @note the current time divergence free time step is denoted with |
---|
3716 | !-- u_t etc.; the velocities of the time level of t+1 wit u,v, and w, as the model is called after |
---|
3717 | !-- swap timelevel |
---|
3718 | !-- @attention: for the corrector step the velocities of t(n+1) are required. |
---|
3719 | !-- Therefore the particle code is executed at the end of the time intermediate timestep routine. |
---|
3720 | !-- This interpolation method is described in more detail in Grabowski et al., 2018 (GMD). |
---|
3721 | IF ( interpolation_simple_corrector ) THEN |
---|
3722 | ! |
---|
3723 | !-- Predictor step |
---|
3724 | kkw = kp - 1 |
---|
3725 | DO n = 1, number_of_particles |
---|
3726 | |
---|
3727 | alpha = MAX( MIN( ( particles(n)%x - ip * dx ) * ddx, 1.0_wp ), 0.0_wp ) |
---|
3728 | u_int(n) = u_t(kp,jp,ip) * ( 1.0_wp - alpha ) + u_t(kp,jp,ip+1) * alpha |
---|
3729 | |
---|
3730 | beta = MAX( MIN( ( particles(n)%y - jp * dy ) * ddy, 1.0_wp ), 0.0_wp ) |
---|
3731 | v_int(n) = v_t(kp,jp,ip) * ( 1.0_wp - beta ) + v_t(kp,jp+1,ip) * beta |
---|
3732 | |
---|
3733 | gamma = MAX( MIN( ( particles(n)%z - zw(kkw) ) / ( zw(kkw+1) - zw(kkw) ), 1.0_wp ), & |
---|
3734 | 0.0_wp ) |
---|
3735 | w_int(n) = w_t(kkw,jp,ip) * ( 1.0_wp - gamma ) + w_t(kkw+1,jp,ip) * gamma |
---|
3736 | |
---|
3737 | ENDDO |
---|
3738 | ! |
---|
3739 | !-- Corrector step |
---|
3740 | DO n = 1, number_of_particles |
---|
3741 | |
---|
3742 | IF ( .NOT. particles(n)%particle_mask ) CYCLE |
---|
3743 | |
---|
3744 | DO nn = 1, iteration_steps |
---|
3745 | |
---|
3746 | ! |
---|
3747 | !-- Guess new position |
---|
3748 | xp = particles(n)%x + u_int(n) * dt_particle(n) |
---|
3749 | yp = particles(n)%y + v_int(n) * dt_particle(n) |
---|
3750 | zp = particles(n)%z + w_int(n) * dt_particle(n) |
---|
3751 | ! |
---|
3752 | !-- x direction |
---|
3753 | i_next = FLOOR( xp * ddx , KIND=iwp) |
---|
3754 | alpha = MAX( MIN( ( xp - i_next * dx ) * ddx, 1.0_wp ), 0.0_wp ) |
---|
3755 | ! |
---|
3756 | !-- y direction |
---|
3757 | j_next = FLOOR( yp * ddy ) |
---|
3758 | beta = MAX( MIN( ( yp - j_next * dy ) * ddy, 1.0_wp ), 0.0_wp ) |
---|
3759 | ! |
---|
3760 | !-- z_direction |
---|
3761 | k_next = MAX( MIN( FLOOR( zp / (zw(kkw+1)-zw(kkw)) + offset_ocean_nzt ), nzt ), 0) |
---|
3762 | gamma = MAX( MIN( ( zp - zw(k_next) ) / & |
---|
3763 | ( zw(k_next+1) - zw(k_next) ), 1.0_wp ), 0.0_wp ) |
---|
3764 | ! |
---|
3765 | !-- Calculate part of the corrector step |
---|
3766 | unext = u(k_next+1, j_next, i_next) * ( 1.0_wp - alpha ) + & |
---|
3767 | u(k_next+1, j_next, i_next+1) * alpha |
---|
3768 | |
---|
3769 | vnext = v(k_next+1, j_next, i_next) * ( 1.0_wp - beta ) + & |
---|
3770 | v(k_next+1, j_next+1, i_next ) * beta |
---|
3771 | |
---|
3772 | wnext = w(k_next, j_next, i_next) * ( 1.0_wp - gamma ) + & |
---|
3773 | w(k_next+1, j_next, i_next ) * gamma |
---|
3774 | |
---|
3775 | ! |
---|
3776 | !-- Calculate interpolated particle velocity with predictor corrector step. u_int, v_int |
---|
3777 | !-- and w_int describes the part of the predictor step. unext, vnext and wnext is the part |
---|
3778 | !-- of the corrector step. The resulting new position is set below. The implementation is |
---|
3779 | !-- based on Grabowski et al., 2018 (GMD). |
---|
3780 | u_int(n) = 0.5_wp * ( u_int(n) + unext ) |
---|
3781 | v_int(n) = 0.5_wp * ( v_int(n) + vnext ) |
---|
3782 | w_int(n) = 0.5_wp * ( w_int(n) + wnext ) |
---|
3783 | |
---|
3784 | ENDDO |
---|
3785 | ENDDO |
---|
3786 | ! |
---|
3787 | !-- This case uses a simple interpolation method for the particle velocites, and applying a |
---|
3788 | !-- predictor. |
---|
3789 | ELSEIF ( interpolation_simple_predictor ) THEN |
---|
3790 | ! |
---|
3791 | !-- The particle position for the w velociy is based on the value of kp and kp-1 |
---|
3792 | kkw = kp - 1 |
---|
3793 | DO n = 1, number_of_particles |
---|
3794 | IF ( .NOT. particles(n)%particle_mask ) CYCLE |
---|
3795 | |
---|
3796 | alpha = MAX( MIN( ( particles(n)%x - ip * dx ) * ddx, 1.0_wp ), 0.0_wp ) |
---|
3797 | u_int(n) = u(kp,jp,ip) * ( 1.0_wp - alpha ) + u(kp,jp,ip+1) * alpha |
---|
3798 | |
---|
3799 | beta = MAX( MIN( ( particles(n)%y - jp * dy ) * ddy, 1.0_wp ), 0.0_wp ) |
---|
3800 | v_int(n) = v(kp,jp,ip) * ( 1.0_wp - beta ) + v(kp,jp+1,ip) * beta |
---|
3801 | |
---|
3802 | gamma = MAX( MIN( ( particles(n)%z - zw(kkw) ) / ( zw(kkw+1) - zw(kkw) ), 1.0_wp ), & |
---|
3803 | 0.0_wp ) |
---|
3804 | w_int(n) = w(kkw,jp,ip) * ( 1.0_wp - gamma ) + w(kkw+1,jp,ip) * gamma |
---|
3805 | ENDDO |
---|
3806 | ! |
---|
3807 | !-- The trilinear interpolation. |
---|
3808 | ELSEIF ( interpolation_trilinear ) THEN |
---|
3809 | |
---|
3810 | start_index = grid_particles(kp,jp,ip)%start_index |
---|
3811 | end_index = grid_particles(kp,jp,ip)%end_index |
---|
3812 | |
---|
3813 | DO nb = 0, 7 |
---|
3814 | ! |
---|
3815 | !-- Interpolate u velocity-component |
---|
3816 | i = ip |
---|
3817 | j = jp + block_offset(nb)%j_off |
---|
3818 | k = kp + block_offset(nb)%k_off |
---|
3819 | |
---|
3820 | DO n = start_index(nb), end_index(nb) |
---|
3821 | ! |
---|
3822 | !-- Interpolation of the u velocity component onto particle position. |
---|
3823 | !-- Particles are interpolation bi-linearly in the horizontal and a linearly in the |
---|
3824 | !-- vertical. An exception is made for particles below the first vertical grid level in |
---|
3825 | !-- case of a prandtl layer. In this case the horizontal particle velocity components are |
---|
3826 | !-- determined using Monin-Obukhov relations (if branch). |
---|
3827 | !-- First, check if particle is located below first vertical grid level above topography |
---|
3828 | !-- (Prandtl-layer height). |
---|
3829 | !-- Determine vertical index of topography top |
---|
3830 | k_wall = topo_top_ind(jp,ip,0) |
---|
3831 | |
---|
3832 | IF ( constant_flux_layer .AND. zv(n) - zw(k_wall) < z_p ) THEN |
---|
3833 | ! |
---|
3834 | !-- Resolved-scale horizontal particle velocity is zero below z0. |
---|
3835 | IF ( zv(n) - zw(k_wall) < z0_av_global ) THEN |
---|
3836 | u_int(n) = 0.0_wp |
---|
3837 | ELSE |
---|
3838 | ! |
---|
3839 | !-- Determine the sublayer. Further used as index. |
---|
3840 | height_p = ( zv(n) - zw(k_wall) - z0_av_global ) & |
---|
3841 | * REAL( number_of_sublayers, KIND=wp ) & |
---|
3842 | * d_z_p_z0 |
---|
3843 | ! |
---|
3844 | !-- Calculate LOG(z/z0) for exact particle height. Therefore, |
---|
3845 | !-- interpolate linearly between precalculated logarithm. |
---|
3846 | log_z_z0_int = log_z_z0( INT( height_p ) ) + ( height_p - INT( height_p ) ) * & |
---|
3847 | ( log_z_z0( INT( height_p ) + 1 ) - log_z_z0( INT( height_p ) ) ) |
---|
3848 | ! |
---|
3849 | !-- Compute u*-portion for u-component based on mean roughness. |
---|
3850 | !-- Note, neutral solution is applied for all situations, e.g. also for unstable and |
---|
3851 | !-- stable situations. Even though this is not exact this saves a lot of CPU time |
---|
3852 | !-- since several calls of intrinsic FORTRAN procedures (LOG, ATAN) are avoided. This |
---|
3853 | !-- is justified as sensitivity studies revealed no significant effect of using the |
---|
3854 | !-- neutral solution also for un/stable situations. Based on the u* recalculate the |
---|
3855 | !-- velocity at height z_particle. Since the analytical solution only yields absolute |
---|
3856 | !-- values, include the sign using the intrinsic SIGN function. |
---|
3857 | us_int = kappa * 0.5_wp * ABS( u(k_wall+1,jp,ip) + u(k_wall+1,jp,ip+1) ) / & |
---|
3858 | log_z_z0(number_of_sublayers) |
---|
3859 | u_int(n) = SIGN( 1.0_wp, u(k_wall+1,jp,ip) + u(k_wall+1,jp,ip+1) ) * & |
---|
3860 | log_z_z0_int * us_int / kappa - u_gtrans |
---|
3861 | |
---|
3862 | ENDIF |
---|
3863 | ! |
---|
3864 | !-- Particle above the first grid level. Bi-linear interpolation in the horizontal and |
---|
3865 | !-- linear interpolation in the vertical direction. |
---|
3866 | ELSE |
---|
3867 | x = xv(n) - i * dx |
---|
3868 | y = yv(n) + ( 0.5_wp - j ) * dy |
---|
3869 | aa = x**2 + y**2 |
---|
3870 | bb = ( dx - x )**2 + y**2 |
---|
3871 | cc = x**2 + ( dy - y )**2 |
---|
3872 | dd = ( dx - x )**2 + ( dy - y )**2 |
---|
3873 | gg = aa + bb + cc + dd |
---|
3874 | |
---|
3875 | u_int_l = ( ( gg - aa ) * u(k,j,i) + ( gg - bb ) * u(k,j,i+1) & |
---|
3876 | + ( gg - cc ) * u(k,j+1,i) + ( gg - dd ) * u(k,j+1,i+1) ) & |
---|
3877 | / ( 3.0_wp * gg ) - u_gtrans |
---|
3878 | |
---|
3879 | IF ( k == nzt ) THEN |
---|
3880 | u_int(n) = u_int_l |
---|
3881 | ELSE |
---|
3882 | u_int_u = ( ( gg-aa ) * u(k+1,j,i) + ( gg-bb ) * u(k+1,j,i+1) & |
---|
3883 | + ( gg-cc ) * u(k+1,j+1,i) + ( gg-dd ) * u(k+1,j+1,i+1) ) & |
---|
3884 | / ( 3.0_wp * gg ) - u_gtrans |
---|
3885 | u_int(n) = u_int_l + ( zv(n) - zu(k) ) / dzw(k+1) * ( u_int_u - u_int_l ) |
---|
3886 | ENDIF |
---|
3887 | ENDIF |
---|
3888 | ENDDO |
---|
3889 | ! |
---|
3890 | !-- Same procedure for interpolation of the v velocity-component |
---|
3891 | i = ip + block_offset(nb)%i_off |
---|
3892 | j = jp |
---|
3893 | k = kp + block_offset(nb)%k_off |
---|
3894 | |
---|
3895 | DO n = start_index(nb), end_index(nb) |
---|
3896 | ! |
---|
3897 | !-- Determine vertical index of topography top |
---|
3898 | k_wall = topo_top_ind(jp,ip,0) |
---|
3899 | |
---|
3900 | IF ( constant_flux_layer .AND. zv(n) - zw(k_wall) < z_p ) THEN |
---|
3901 | IF ( zv(n) - zw(k_wall) < z0_av_global ) THEN |
---|
3902 | ! |
---|
3903 | !-- Resolved-scale horizontal particle velocity is zero below z0. |
---|
3904 | v_int(n) = 0.0_wp |
---|
3905 | ELSE |
---|
3906 | ! |
---|
3907 | !-- Determine the sublayer. Further used as index. |
---|
3908 | height_p = ( zv(n) - zw(k_wall) - z0_av_global ) & |
---|
3909 | * REAL( number_of_sublayers, KIND=wp ) & |
---|
3910 | * d_z_p_z0 |
---|
3911 | ! |
---|
3912 | !-- Calculate LOG(z/z0) for exact particle height. Therefore, interpolate linearly |
---|
3913 | !-- between precalculated logarithm. |
---|
3914 | log_z_z0_int = log_z_z0(INT(height_p)) & |
---|
3915 | + ( height_p - INT(height_p) ) & |
---|
3916 | * ( log_z_z0(INT(height_p)+1) - log_z_z0(INT(height_p)) & |
---|
3917 | ) |
---|
3918 | ! |
---|
3919 | !-- Compute u*-portion for v-component based on mean roughness. |
---|
3920 | !-- Note, neutral solution is applied for all situations, e.g. also for unstable and |
---|
3921 | !-- stable situations. Even though this is not exact this saves a lot of CPU time |
---|
3922 | !-- since several calls of intrinsic FORTRAN procedures (LOG, ATAN) are avoided, This |
---|
3923 | !-- is justified as sensitivity studies revealed no significant effect of using the |
---|
3924 | !-- neutral solution also for un/stable situations. Based on the u* recalculate the |
---|
3925 | !-- velocity at height z_particle. Since the analytical solution only yields absolute |
---|
3926 | !-- values, include the sign using the intrinsic SIGN function. |
---|
3927 | us_int = kappa * 0.5_wp * ABS( v(k_wall+1,jp,ip) + v(k_wall+1,jp+1,ip) ) / & |
---|
3928 | log_z_z0(number_of_sublayers) |
---|
3929 | v_int(n) = SIGN( 1.0_wp, v(k_wall+1,jp,ip) + v(k_wall+1,jp+1,ip) ) * & |
---|
3930 | log_z_z0_int * us_int / kappa - v_gtrans |
---|
3931 | |
---|
3932 | ENDIF |
---|
3933 | ELSE |
---|
3934 | x = xv(n) + ( 0.5_wp - i ) * dx |
---|
3935 | y = yv(n) - j * dy |
---|
3936 | aa = x**2 + y**2 |
---|
3937 | bb = ( dx - x )**2 + y**2 |
---|
3938 | cc = x**2 + ( dy - y )**2 |
---|
3939 | dd = ( dx - x )**2 + ( dy - y )**2 |
---|
3940 | gg = aa + bb + cc + dd |
---|
3941 | |
---|
3942 | v_int_l = ( ( gg - aa ) * v(k,j,i) + ( gg - bb ) * v(k,j,i+1) & |
---|
3943 | + ( gg - cc ) * v(k,j+1,i) + ( gg - dd ) * v(k,j+1,i+1) & |
---|
3944 | ) / ( 3.0_wp * gg ) - v_gtrans |
---|
3945 | |
---|
3946 | IF ( k == nzt ) THEN |
---|
3947 | v_int(n) = v_int_l |
---|
3948 | ELSE |
---|
3949 | v_int_u = ( ( gg-aa ) * v(k+1,j,i) + ( gg-bb ) * v(k+1,j,i+1) & |
---|
3950 | + ( gg-cc ) * v(k+1,j+1,i) + ( gg-dd ) * v(k+1,j+1,i+1) & |
---|
3951 | ) / ( 3.0_wp * gg ) - v_gtrans |
---|
3952 | v_int(n) = v_int_l + ( zv(n) - zu(k) ) / dzw(k+1) * ( v_int_u - v_int_l ) |
---|
3953 | ENDIF |
---|
3954 | ENDIF |
---|
3955 | ENDDO |
---|
3956 | ! |
---|
3957 | !-- Same procedure for interpolation of the w velocity-component |
---|
3958 | i = ip + block_offset(nb)%i_off |
---|
3959 | j = jp + block_offset(nb)%j_off |
---|
3960 | k = kp - 1 |
---|
3961 | |
---|
3962 | DO n = start_index(nb), end_index(nb) |
---|
3963 | IF ( vertical_particle_advection(particles(n)%group) ) THEN |
---|
3964 | x = xv(n) + ( 0.5_wp - i ) * dx |
---|
3965 | y = yv(n) + ( 0.5_wp - j ) * dy |
---|
3966 | aa = x**2 + y**2 |
---|
3967 | bb = ( dx - x )**2 + y**2 |
---|
3968 | cc = x**2 + ( dy - y )**2 |
---|
3969 | dd = ( dx - x )**2 + ( dy - y )**2 |
---|
3970 | gg = aa + bb + cc + dd |
---|
3971 | |
---|
3972 | w_int_l = ( ( gg - aa ) * w(k,j,i) + ( gg - bb ) * w(k,j,i+1) & |
---|
3973 | + ( gg - cc ) * w(k,j+1,i) + ( gg - dd ) * w(k,j+1,i+1) & |
---|
3974 | ) / ( 3.0_wp * gg ) |
---|
3975 | |
---|
3976 | IF ( k == nzt ) THEN |
---|
3977 | w_int(n) = w_int_l |
---|
3978 | ELSE |
---|
3979 | w_int_u = ( ( gg-aa ) * w(k+1,j,i) + & |
---|
3980 | ( gg-bb ) * w(k+1,j,i+1) + & |
---|
3981 | ( gg-cc ) * w(k+1,j+1,i) + & |
---|
3982 | ( gg-dd ) * w(k+1,j+1,i+1) & |
---|
3983 | ) / ( 3.0_wp * gg ) |
---|
3984 | w_int(n) = w_int_l + ( zv(n) - zw(k) ) / dzw(k+1) * ( w_int_u - w_int_l ) |
---|
3985 | ENDIF |
---|
3986 | ELSE |
---|
3987 | w_int(n) = 0.0_wp |
---|
3988 | ENDIF |
---|
3989 | ENDDO |
---|
3990 | ENDDO |
---|
3991 | ENDIF |
---|
3992 | |
---|
3993 | !-- Interpolate and calculate quantities needed for calculating the SGS velocities |
---|
3994 | IF ( use_sgs_for_particles .AND. .NOT. cloud_droplets ) THEN |
---|
3995 | |
---|
3996 | DO nb = 0,7 |
---|
3997 | |
---|
3998 | subbox_at_wall = .FALSE. |
---|
3999 | ! |
---|
4000 | !-- In case of topography check if subbox is adjacent to a wall |
---|
4001 | IF ( .NOT. topography == 'flat' ) THEN |
---|
4002 | i = ip + MERGE( -1_iwp , 1_iwp, BTEST( nb, 2 ) ) |
---|
4003 | j = jp + MERGE( -1_iwp , 1_iwp, BTEST( nb, 1 ) ) |
---|
4004 | k = kp + MERGE( -1_iwp , 1_iwp, BTEST( nb, 0 ) ) |
---|
4005 | IF ( .NOT. BTEST(wall_flags_total_0(k, jp, ip), 0) .OR. & |
---|
4006 | .NOT. BTEST(wall_flags_total_0(kp, j, ip), 0) .OR. & |
---|
4007 | .NOT. BTEST(wall_flags_total_0(kp, jp, i ), 0) ) & |
---|
4008 | THEN |
---|
4009 | subbox_at_wall = .TRUE. |
---|
4010 | ENDIF |
---|
4011 | ENDIF |
---|
4012 | IF ( subbox_at_wall ) THEN |
---|
4013 | e_int(start_index(nb):end_index(nb)) = e(kp,jp,ip) |
---|
4014 | diss_int(start_index(nb):end_index(nb)) = diss(kp,jp,ip) |
---|
4015 | de_dx_int(start_index(nb):end_index(nb)) = de_dx(kp,jp,ip) |
---|
4016 | de_dy_int(start_index(nb):end_index(nb)) = de_dy(kp,jp,ip) |
---|
4017 | de_dz_int(start_index(nb):end_index(nb)) = de_dz(kp,jp,ip) |
---|
4018 | ! |
---|
4019 | !-- Set flag for stochastic equation. |
---|
4020 | term_1_2(start_index(nb):end_index(nb)) = 0.0_wp |
---|
4021 | ELSE |
---|
4022 | i = ip + block_offset(nb)%i_off |
---|
4023 | j = jp + block_offset(nb)%j_off |
---|
4024 | k = kp + block_offset(nb)%k_off |
---|
4025 | |
---|
4026 | DO n = start_index(nb), end_index(nb) |
---|
4027 | ! |
---|
4028 | !-- Interpolate TKE |
---|
4029 | x = xv(n) + ( 0.5_wp - i ) * dx |
---|
4030 | y = yv(n) + ( 0.5_wp - j ) * dy |
---|
4031 | aa = x**2 + y**2 |
---|
4032 | bb = ( dx - x )**2 + y**2 |
---|
4033 | cc = x**2 + ( dy - y )**2 |
---|
4034 | dd = ( dx - x )**2 + ( dy - y )**2 |
---|
4035 | gg = aa + bb + cc + dd |
---|
4036 | |
---|
4037 | e_int_l = ( ( gg-aa ) * e(k,j,i) + ( gg-bb ) * e(k,j,i+1) & |
---|
4038 | + ( gg-cc ) * e(k,j+1,i) + ( gg-dd ) * e(k,j+1,i+1) & |
---|
4039 | ) / ( 3.0_wp * gg ) |
---|
4040 | |
---|
4041 | IF ( k+1 == nzt+1 ) THEN |
---|
4042 | e_int(n) = e_int_l |
---|
4043 | ELSE |
---|
4044 | e_int_u = ( ( gg - aa ) * e(k+1,j,i) + & |
---|
4045 | ( gg - bb ) * e(k+1,j,i+1) + & |
---|
4046 | ( gg - cc ) * e(k+1,j+1,i) + & |
---|
4047 | ( gg - dd ) * e(k+1,j+1,i+1) & |
---|
4048 | ) / ( 3.0_wp * gg ) |
---|
4049 | e_int(n) = e_int_l + ( zv(n) - zu(k) ) / dzw(k+1) * ( e_int_u - e_int_l ) |
---|
4050 | ENDIF |
---|
4051 | ! |
---|
4052 | !-- Needed to avoid NaN particle velocities (this might not be required any more) |
---|
4053 | IF ( e_int(n) <= 0.0_wp ) THEN |
---|
4054 | e_int(n) = 1.0E-20_wp |
---|
4055 | ENDIF |
---|
4056 | ! |
---|
4057 | !-- Interpolate the TKE gradient along x (adopt incides i,j,k and all position variables |
---|
4058 | !-- from above (TKE)) |
---|
4059 | de_dx_int_l = ( ( gg - aa ) * de_dx(k,j,i) + & |
---|
4060 | ( gg - bb ) * de_dx(k,j,i+1) + & |
---|
4061 | ( gg - cc ) * de_dx(k,j+1,i) + & |
---|
4062 | ( gg - dd ) * de_dx(k,j+1,i+1) & |
---|
4063 | ) / ( 3.0_wp * gg ) |
---|
4064 | |
---|
4065 | IF ( ( k+1 == nzt+1 ) .OR. ( k == nzb ) ) THEN |
---|
4066 | de_dx_int(n) = de_dx_int_l |
---|
4067 | ELSE |
---|
4068 | de_dx_int_u = ( ( gg - aa ) * de_dx(k+1,j,i) + & |
---|
4069 | ( gg - bb ) * de_dx(k+1,j,i+1) + & |
---|
4070 | ( gg - cc ) * de_dx(k+1,j+1,i) + & |
---|
4071 | ( gg - dd ) * de_dx(k+1,j+1,i+1) & |
---|
4072 | ) / ( 3.0_wp * gg ) |
---|
4073 | de_dx_int(n) = de_dx_int_l + ( zv(n) - zu(k) ) / dzw(k+1) * & |
---|
4074 | ( de_dx_int_u - de_dx_int_l ) |
---|
4075 | ENDIF |
---|
4076 | ! |
---|
4077 | !-- Interpolate the TKE gradient along y |
---|
4078 | de_dy_int_l = ( ( gg - aa ) * de_dy(k,j,i) + & |
---|
4079 | ( gg - bb ) * de_dy(k,j,i+1) + & |
---|
4080 | ( gg - cc ) * de_dy(k,j+1,i) + & |
---|
4081 | ( gg - dd ) * de_dy(k,j+1,i+1) & |
---|
4082 | ) / ( 3.0_wp * gg ) |
---|
4083 | IF ( ( k+1 == nzt+1 ) .OR. ( k == nzb ) ) THEN |
---|
4084 | de_dy_int(n) = de_dy_int_l |
---|
4085 | ELSE |
---|
4086 | de_dy_int_u = ( ( gg - aa ) * de_dy(k+1,j,i) + & |
---|
4087 | ( gg - bb ) * de_dy(k+1,j,i+1) + & |
---|
4088 | ( gg - cc ) * de_dy(k+1,j+1,i) + & |
---|
4089 | ( gg - dd ) * de_dy(k+1,j+1,i+1) & |
---|
4090 | ) / ( 3.0_wp * gg ) |
---|
4091 | de_dy_int(n) = de_dy_int_l + ( zv(n) - zu(k) ) / dzw(k+1) * & |
---|
4092 | ( de_dy_int_u - de_dy_int_l ) |
---|
4093 | ENDIF |
---|
4094 | |
---|
4095 | ! |
---|
4096 | !-- Interpolate the TKE gradient along z |
---|
4097 | IF ( zv(n) < 0.5_wp * dz(1) ) THEN |
---|
4098 | de_dz_int(n) = 0.0_wp |
---|
4099 | ELSE |
---|
4100 | de_dz_int_l = ( ( gg - aa ) * de_dz(k,j,i) + & |
---|
4101 | ( gg - bb ) * de_dz(k,j,i+1) + & |
---|
4102 | ( gg - cc ) * de_dz(k,j+1,i) + & |
---|
4103 | ( gg - dd ) * de_dz(k,j+1,i+1) & |
---|
4104 | ) / ( 3.0_wp * gg ) |
---|
4105 | |
---|
4106 | IF ( ( k+1 == nzt+1 ) .OR. ( k == nzb ) ) THEN |
---|
4107 | de_dz_int(n) = de_dz_int_l |
---|
4108 | ELSE |
---|
4109 | de_dz_int_u = ( ( gg - aa ) * de_dz(k+1,j,i) + & |
---|
4110 | ( gg - bb ) * de_dz(k+1,j,i+1) + & |
---|
4111 | ( gg - cc ) * de_dz(k+1,j+1,i) + & |
---|
4112 | ( gg - dd ) * de_dz(k+1,j+1,i+1) & |
---|
4113 | ) / ( 3.0_wp * gg ) |
---|
4114 | de_dz_int(n) = de_dz_int_l + ( zv(n) - zu(k) ) / dzw(k+1) * & |
---|
4115 | ( de_dz_int_u - de_dz_int_l ) |
---|
4116 | ENDIF |
---|
4117 | ENDIF |
---|
4118 | |
---|
4119 | ! |
---|
4120 | !-- Interpolate the dissipation of TKE |
---|
4121 | diss_int_l = ( ( gg - aa ) * diss(k,j,i) + & |
---|
4122 | ( gg - bb ) * diss(k,j,i+1) + & |
---|
4123 | ( gg - cc ) * diss(k,j+1,i) + & |
---|
4124 | ( gg - dd ) * diss(k,j+1,i+1) & |
---|
4125 | ) / ( 3.0_wp * gg ) |
---|
4126 | |
---|
4127 | IF ( k == nzt ) THEN |
---|
4128 | diss_int(n) = diss_int_l |
---|
4129 | ELSE |
---|
4130 | diss_int_u = ( ( gg - aa ) * diss(k+1,j,i) + & |
---|
4131 | ( gg - bb ) * diss(k+1,j,i+1) + & |
---|
4132 | ( gg - cc ) * diss(k+1,j+1,i) + & |
---|
4133 | ( gg - dd ) * diss(k+1,j+1,i+1) & |
---|
4134 | ) / ( 3.0_wp * gg ) |
---|
4135 | diss_int(n) = diss_int_l + ( zv(n) - zu(k) ) / dzw(k+1) * & |
---|
4136 | ( diss_int_u - diss_int_l ) |
---|
4137 | ENDIF |
---|
4138 | |
---|
4139 | ! |
---|
4140 | !-- Set flag for stochastic equation. |
---|
4141 | term_1_2(n) = 1.0_wp |
---|
4142 | ENDDO |
---|
4143 | ENDIF |
---|
4144 | ENDDO |
---|
4145 | |
---|
4146 | DO nb = 0,7 |
---|
4147 | i = ip + block_offset(nb)%i_off |
---|
4148 | j = jp + block_offset(nb)%j_off |
---|
4149 | k = kp + block_offset(nb)%k_off |
---|
4150 | |
---|
4151 | DO n = start_index(nb), end_index(nb) |
---|
4152 | ! |
---|
4153 | !-- Vertical interpolation of the horizontally averaged SGS TKE and resolved-scale velocity |
---|
4154 | !-- variances and use the interpolated values to calculate the coefficient fs, which is a |
---|
4155 | !-- measure of the ratio of the subgrid-scale turbulent kinetic energy to the total amount |
---|
4156 | !-- of turbulent kinetic energy. |
---|
4157 | IF ( k == 0 ) THEN |
---|
4158 | e_mean_int = hom(0,1,8,0) |
---|
4159 | ELSE |
---|
4160 | e_mean_int = hom(k,1,8,0) + ( hom(k+1,1,8,0) - hom(k,1,8,0) ) / & |
---|
4161 | ( zu(k+1) - zu(k) ) * & |
---|
4162 | ( zv(n) - zu(k) ) |
---|
4163 | ENDIF |
---|
4164 | |
---|
4165 | kw = kp - 1 |
---|
4166 | |
---|
4167 | IF ( k == 0 ) THEN |
---|
4168 | aa = hom(k+1,1,30,0) * ( zv(n) / & |
---|
4169 | ( 0.5_wp * ( zu(k+1) - zu(k) ) ) ) |
---|
4170 | bb = hom(k+1,1,31,0) * ( zv(n) / & |
---|
4171 | ( 0.5_wp * ( zu(k+1) - zu(k) ) ) ) |
---|
4172 | cc = hom(kw+1,1,32,0) * ( zv(n) / & |
---|
4173 | ( 1.0_wp * ( zw(kw+1) - zw(kw) ) ) ) |
---|
4174 | ELSE |
---|
4175 | aa = hom(k,1,30,0) + ( hom(k+1,1,30,0) - hom(k,1,30,0) ) * & |
---|
4176 | ( ( zv(n) - zu(k) ) / ( zu(k+1) - zu(k) ) ) |
---|
4177 | bb = hom(k,1,31,0) + ( hom(k+1,1,31,0) - hom(k,1,31,0) ) * & |
---|
4178 | ( ( zv(n) - zu(k) ) / ( zu(k+1) - zu(k) ) ) |
---|
4179 | cc = hom(kw,1,32,0) + ( hom(kw+1,1,32,0)-hom(kw,1,32,0) ) * & |
---|
4180 | ( ( zv(n) - zw(kw) ) / ( zw(kw+1)-zw(kw) ) ) |
---|
4181 | ENDIF |
---|
4182 | |
---|
4183 | vv_int = ( 1.0_wp / 3.0_wp ) * ( aa + bb + cc ) |
---|
4184 | ! |
---|
4185 | !-- Needed to avoid NaN particle velocities. The value of 1.0 is just an educated guess for |
---|
4186 | !-- the given case. |
---|
4187 | IF ( vv_int + ( 2.0_wp / 3.0_wp ) * e_mean_int == 0.0_wp ) THEN |
---|
4188 | fs_int(n) = 1.0_wp |
---|
4189 | ELSE |
---|
4190 | fs_int(n) = ( 2.0_wp / 3.0_wp ) * e_mean_int / & |
---|
4191 | ( vv_int + ( 2.0_wp / 3.0_wp ) * e_mean_int ) |
---|
4192 | ENDIF |
---|
4193 | |
---|
4194 | ENDDO |
---|
4195 | ENDDO |
---|
4196 | |
---|
4197 | DO nb = 0, 7 |
---|
4198 | DO n = start_index(nb), end_index(nb) |
---|
4199 | CALL random_number_parallel_gauss( random_dummy ) |
---|
4200 | rg(n,1) = random_dummy |
---|
4201 | CALL random_number_parallel_gauss( random_dummy ) |
---|
4202 | rg(n,2) = random_dummy |
---|
4203 | CALL random_number_parallel_gauss( random_dummy ) |
---|
4204 | rg(n,3) = random_dummy |
---|
4205 | ENDDO |
---|
4206 | ENDDO |
---|
4207 | |
---|
4208 | DO nb = 0, 7 |
---|
4209 | DO n = start_index(nb), end_index(nb) |
---|
4210 | |
---|
4211 | ! |
---|
4212 | !-- Calculate the Lagrangian timescale according to Weil et al. (2004). |
---|
4213 | lagr_timescale(n) = ( 4.0_wp * e_int(n) + 1E-20_wp ) / & |
---|
4214 | ( 3.0_wp * fs_int(n) * c_0 * diss_int(n) + 1E-20_wp ) |
---|
4215 | |
---|
4216 | ! |
---|
4217 | !-- Calculate the next particle timestep. dt_gap is the time needed to complete the current |
---|
4218 | !-- LES timestep. |
---|
4219 | dt_gap(n) = dt_3d - particles(n)%dt_sum |
---|
4220 | dt_particle(n) = MIN( dt_3d, 0.025_wp * lagr_timescale(n), dt_gap(n) ) |
---|
4221 | particles(n)%aux1 = lagr_timescale(n) |
---|
4222 | particles(n)%aux2 = dt_gap(n) |
---|
4223 | ! |
---|
4224 | !-- The particle timestep should not be too small in order to prevent the number of |
---|
4225 | !-- particle timesteps of getting too large |
---|
4226 | IF ( dt_particle(n) < dt_min_part ) THEN |
---|
4227 | IF ( dt_min_part < dt_gap(n) ) THEN |
---|
4228 | dt_particle(n) = dt_min_part |
---|
4229 | ELSE |
---|
4230 | dt_particle(n) = dt_gap(n) |
---|
4231 | ENDIF |
---|
4232 | ENDIF |
---|
4233 | |
---|
4234 | rvar1_temp(n) = particles(n)%rvar1 |
---|
4235 | rvar2_temp(n) = particles(n)%rvar2 |
---|
4236 | rvar3_temp(n) = particles(n)%rvar3 |
---|
4237 | ! |
---|
4238 | !-- Calculate the SGS velocity components |
---|
4239 | IF ( particles(n)%age == 0.0_wp ) THEN |
---|
4240 | ! |
---|
4241 | !-- For new particles the SGS components are derived from the SGS TKE. Limit the |
---|
4242 | !-- Gaussian random number to the interval [-5.0*sigma, 5.0*sigma] in order to prevent |
---|
4243 | !-- the SGS velocities from becoming unrealistically large. |
---|
4244 | rvar1_temp(n) = SQRT( 2.0_wp * sgs_wf_part * e_int(n) + 1E-20_wp ) * rg(n,1) |
---|
4245 | rvar2_temp(n) = SQRT( 2.0_wp * sgs_wf_part * e_int(n) + 1E-20_wp ) * rg(n,2) |
---|
4246 | rvar3_temp(n) = SQRT( 2.0_wp * sgs_wf_part * e_int(n) + 1E-20_wp ) * rg(n,3) |
---|
4247 | ELSE |
---|
4248 | ! |
---|
4249 | !-- Restriction of the size of the new timestep: compared to the |
---|
4250 | !-- previous timestep the increase must not exceed 200%. First, |
---|
4251 | !-- check if age > age_m, in order to prevent that particles get zero |
---|
4252 | !-- timestep. |
---|
4253 | dt_particle_m = MERGE( dt_particle(n), & |
---|
4254 | particles(n)%age - particles(n)%age_m, & |
---|
4255 | particles(n)%age - particles(n)%age_m < 1E-8_wp ) |
---|
4256 | IF ( dt_particle(n) > 2.0_wp * dt_particle_m ) THEN |
---|
4257 | dt_particle(n) = 2.0_wp * dt_particle_m |
---|
4258 | ENDIF |
---|
4259 | |
---|
4260 | !-- For old particles the SGS components are correlated with the values from the |
---|
4261 | !-- previous timestep. Random numbers have also to be limited (see above). |
---|
4262 | !-- As negative values for the subgrid TKE are not allowed, the change of the subgrid |
---|
4263 | !-- TKE with time cannot be smaller than -e_int(n)/dt_particle. This value is used as a |
---|
4264 | !-- lower boundary value for the change of TKE |
---|
4265 | de_dt_min = - e_int(n) / dt_particle(n) |
---|
4266 | |
---|
4267 | de_dt = ( e_int(n) - particles(n)%e_m ) / dt_particle_m |
---|
4268 | |
---|
4269 | IF ( de_dt < de_dt_min ) THEN |
---|
4270 | de_dt = de_dt_min |
---|
4271 | ENDIF |
---|
4272 | |
---|
4273 | CALL weil_stochastic_eq( rvar1_temp(n), fs_int(n), e_int(n), de_dx_int(n), de_dt, & |
---|
4274 | diss_int(n), dt_particle(n), rg(n,1), term_1_2(n) ) |
---|
4275 | |
---|
4276 | CALL weil_stochastic_eq( rvar2_temp(n), fs_int(n), e_int(n), de_dy_int(n), de_dt, & |
---|
4277 | diss_int(n), dt_particle(n), rg(n,2), term_1_2(n) ) |
---|
4278 | |
---|
4279 | CALL weil_stochastic_eq( rvar3_temp(n), fs_int(n), e_int(n), de_dz_int(n), de_dt, & |
---|
4280 | diss_int(n), dt_particle(n), rg(n,3), term_1_2(n) ) |
---|
4281 | |
---|
4282 | ENDIF |
---|
4283 | |
---|
4284 | ENDDO |
---|
4285 | ENDDO |
---|
4286 | ! |
---|
4287 | !-- Check if the added SGS velocities result in a violation of the CFL-criterion. If yes, limt |
---|
4288 | !-- the SGS particle speed to match the CFL criterion. Note, a re-calculation of the SGS particle |
---|
4289 | !-- speed with smaller timestep does not necessarily fulfill the CFL criterion as the new SGS |
---|
4290 | !-- speed can be even larger (due to the random term with scales with the square-root of |
---|
4291 | !-- dt_particle, for small dt the random contribution increases). |
---|
4292 | !-- Thus, we would need to re-calculate the SGS speeds as long as they would fulfill the |
---|
4293 | !-- requirements, which could become computationally expensive, |
---|
4294 | !-- Hence, we just limit them. |
---|
4295 | dz_temp = zw(kp)-zw(kp-1) |
---|
4296 | |
---|
4297 | DO nb = 0, 7 |
---|
4298 | DO n = start_index(nb), end_index(nb) |
---|
4299 | IF ( ABS( u_int(n) + rvar1_temp(n) ) > ( dx / dt_particle(n) ) .OR. & |
---|
4300 | ABS( v_int(n) + rvar2_temp(n) ) > ( dy / dt_particle(n) ) .OR. & |
---|
4301 | ABS( w_int(n) + rvar3_temp(n) ) > ( dz_temp / dt_particle(n) ) ) THEN |
---|
4302 | ! |
---|
4303 | !-- If total speed exceeds the allowed speed according to CFL |
---|
4304 | !-- criterion, limit the SGS speed to |
---|
4305 | !-- dx_i / dt_particle - u_resolved_i, considering a safty factor. |
---|
4306 | rvar1_temp(n) = MERGE( rvar1_temp(n), & |
---|
4307 | 0.9_wp * & |
---|
4308 | SIGN( dx / dt_particle(n) & |
---|
4309 | - ABS( u_int(n) ), rvar1_temp(n) ), & |
---|
4310 | ABS( u_int(n) + rvar1_temp(n) ) < & |
---|
4311 | ( dx / dt_particle(n) ) ) |
---|
4312 | rvar2_temp(n) = MERGE( rvar2_temp(n), & |
---|
4313 | 0.9_wp * & |
---|
4314 | SIGN( dy / dt_particle(n) & |
---|
4315 | - ABS( v_int(n) ), rvar2_temp(n) ), & |
---|
4316 | ABS( v_int(n) + rvar2_temp(n) ) < & |
---|
4317 | ( dy / dt_particle(n) ) ) |
---|
4318 | rvar3_temp(n) = MERGE( rvar3_temp(n), & |
---|
4319 | 0.9_wp * & |
---|
4320 | SIGN( zw(kp)-zw(kp-1) / dt_particle(n) & |
---|
4321 | - ABS( w_int(n) ), rvar3_temp(n) ), & |
---|
4322 | ABS( w_int(n) + rvar3_temp(n) ) < & |
---|
4323 | ( zw(kp)-zw(kp-1) / dt_particle(n) ) ) |
---|
4324 | ENDIF |
---|
4325 | ! |
---|
4326 | !-- Update particle velocites |
---|
4327 | particles(n)%rvar1 = rvar1_temp(n) |
---|
4328 | particles(n)%rvar2 = rvar2_temp(n) |
---|
4329 | particles(n)%rvar3 = rvar3_temp(n) |
---|
4330 | u_int(n) = u_int(n) + particles(n)%rvar1 |
---|
4331 | v_int(n) = v_int(n) + particles(n)%rvar2 |
---|
4332 | w_int(n) = w_int(n) + particles(n)%rvar3 |
---|
4333 | ! |
---|
4334 | !-- Store the SGS TKE of the current timelevel which is needed for for calculating the SGS |
---|
4335 | !-- particle velocities at the next timestep |
---|
4336 | particles(n)%e_m = e_int(n) |
---|
4337 | ENDDO |
---|
4338 | ENDDO |
---|
4339 | |
---|
4340 | ELSE |
---|
4341 | ! |
---|
4342 | !-- If no SGS velocities are used, only the particle timestep has to be set |
---|
4343 | dt_particle = dt_3d |
---|
4344 | |
---|
4345 | ENDIF |
---|
4346 | |
---|
4347 | dens_ratio = particle_groups(particles(1:number_of_particles)%group)%density_ratio |
---|
4348 | IF ( ANY( dens_ratio == 0.0_wp ) ) THEN |
---|
4349 | ! |
---|
4350 | !-- Decide whether the particle loop runs over the subboxes or only over 1, number_of_particles. |
---|
4351 | !-- This depends on the selected interpolation method. |
---|
4352 | !-- If particle interpolation method is not trilinear, then the sorting within subboxes is not |
---|
4353 | !-- required. However, therefore the index start_index(nb) and end_index(nb) are not defined and |
---|
4354 | !-- the loops are still over number_of_particles. @todo find a more generic way to write this |
---|
4355 | !-- loop or delete trilinear interpolation |
---|
4356 | IF ( interpolation_trilinear ) THEN |
---|
4357 | subbox_start = 0 |
---|
4358 | subbox_end = 7 |
---|
4359 | ELSE |
---|
4360 | subbox_start = 1 |
---|
4361 | subbox_end = 1 |
---|
4362 | ENDIF |
---|
4363 | ! |
---|
4364 | !-- loop over subboxes. In case of simple interpolation scheme no subboxes are introduced, as |
---|
4365 | !-- they are not required. Accordingly, this loop goes from 1 to 1. |
---|
4366 | DO nb = subbox_start, subbox_end |
---|
4367 | IF ( interpolation_trilinear ) THEN |
---|
4368 | particle_start = start_index(nb) |
---|
4369 | particle_end = end_index(nb) |
---|
4370 | ELSE |
---|
4371 | particle_start = 1 |
---|
4372 | particle_end = number_of_particles |
---|
4373 | ENDIF |
---|
4374 | ! |
---|
4375 | !-- Loop from particle start to particle end |
---|
4376 | DO n = particle_start, particle_end |
---|
4377 | |
---|
4378 | ! |
---|
4379 | !-- Particle advection |
---|
4380 | IF ( dens_ratio(n) == 0.0_wp ) THEN |
---|
4381 | ! |
---|
4382 | !-- Pure passive transport (without particle inertia) |
---|
4383 | particles(n)%x = xv(n) + u_int(n) * dt_particle(n) |
---|
4384 | particles(n)%y = yv(n) + v_int(n) * dt_particle(n) |
---|
4385 | particles(n)%z = zv(n) + w_int(n) * dt_particle(n) |
---|
4386 | |
---|
4387 | particles(n)%speed_x = u_int(n) |
---|
4388 | particles(n)%speed_y = v_int(n) |
---|
4389 | particles(n)%speed_z = w_int(n) |
---|
4390 | |
---|
4391 | ELSE |
---|
4392 | ! |
---|
4393 | !-- Transport of particles with inertia |
---|
4394 | particles(n)%x = particles(n)%x + particles(n)%speed_x * dt_particle(n) |
---|
4395 | particles(n)%y = particles(n)%y + particles(n)%speed_y * dt_particle(n) |
---|
4396 | particles(n)%z = particles(n)%z + particles(n)%speed_z * dt_particle(n) |
---|
4397 | |
---|
4398 | ! |
---|
4399 | !-- Update of the particle velocity |
---|
4400 | IF ( cloud_droplets ) THEN |
---|
4401 | ! |
---|
4402 | !-- Terminal velocity is computed for vertical direction (Rogers et |
---|
4403 | !-- al., 1993, J. Appl. Meteorol.) |
---|
4404 | diameter = particles(n)%radius * 2000.0_wp !diameter in mm |
---|
4405 | IF ( diameter <= d0_rog ) THEN |
---|
4406 | w_s = k_cap_rog * diameter * ( 1.0_wp - EXP( -k_low_rog * diameter ) ) |
---|
4407 | ELSE |
---|
4408 | w_s = a_rog - b_rog * EXP( -c_rog * diameter ) |
---|
4409 | ENDIF |
---|
4410 | |
---|
4411 | ! |
---|
4412 | !-- If selected, add random velocities following Soelch and Kaercher |
---|
4413 | !-- (2010, Q. J. R. Meteorol. Soc.) |
---|
4414 | IF ( use_sgs_for_particles ) THEN |
---|
4415 | lagr_timescale(n) = km(kp,jp,ip) / MAX( e(kp,jp,ip), 1.0E-20_wp ) |
---|
4416 | rl = EXP( -1.0_wp * dt_3d / MAX( lagr_timescale(n), 1.0E-20_wp ) ) |
---|
4417 | sigma = SQRT( e(kp,jp,ip) ) |
---|
4418 | ! |
---|
4419 | !-- Calculate random component of particle sgs velocity using parallel |
---|
4420 | !-- random generator |
---|
4421 | CALL random_number_parallel_gauss( random_dummy ) |
---|
4422 | rg1 = random_dummy |
---|
4423 | CALL random_number_parallel_gauss( random_dummy ) |
---|
4424 | rg2 = random_dummy |
---|
4425 | CALL random_number_parallel_gauss( random_dummy ) |
---|
4426 | rg3 = random_dummy |
---|
4427 | |
---|
4428 | particles(n)%rvar1 = rl * particles(n)%rvar1 + & |
---|
4429 | SQRT( 1.0_wp - rl**2 ) * sigma * rg1 |
---|
4430 | particles(n)%rvar2 = rl * particles(n)%rvar2 + & |
---|
4431 | SQRT( 1.0_wp - rl**2 ) * sigma * rg2 |
---|
4432 | particles(n)%rvar3 = rl * particles(n)%rvar3 + & |
---|
4433 | SQRT( 1.0_wp - rl**2 ) * sigma * rg3 |
---|
4434 | |
---|
4435 | particles(n)%speed_x = u_int(n) + particles(n)%rvar1 |
---|
4436 | particles(n)%speed_y = v_int(n) + particles(n)%rvar2 |
---|
4437 | particles(n)%speed_z = w_int(n) + particles(n)%rvar3 - w_s |
---|
4438 | ELSE |
---|
4439 | particles(n)%speed_x = u_int(n) |
---|
4440 | particles(n)%speed_y = v_int(n) |
---|
4441 | particles(n)%speed_z = w_int(n) - w_s |
---|
4442 | ENDIF |
---|
4443 | |
---|
4444 | ELSE |
---|
4445 | |
---|
4446 | IF ( use_sgs_for_particles ) THEN |
---|
4447 | exp_arg = particle_groups(particles(n)%group)%exp_arg |
---|
4448 | exp_term = EXP( -exp_arg * dt_particle(n) ) |
---|
4449 | ELSE |
---|
4450 | exp_arg = particle_groups(particles(n)%group)%exp_arg |
---|
4451 | exp_term = particle_groups(particles(n)%group)%exp_term |
---|
4452 | ENDIF |
---|
4453 | particles(n)%speed_x = particles(n)%speed_x * exp_term + & |
---|
4454 | u_int(n) * ( 1.0_wp - exp_term ) |
---|
4455 | particles(n)%speed_y = particles(n)%speed_y * exp_term + & |
---|
4456 | v_int(n) * ( 1.0_wp - exp_term ) |
---|
4457 | particles(n)%speed_z = particles(n)%speed_z * exp_term + & |
---|
4458 | ( w_int(n) - ( 1.0_wp - dens_ratio(n) ) * & |
---|
4459 | g / exp_arg ) * ( 1.0_wp - exp_term ) |
---|
4460 | ENDIF |
---|
4461 | |
---|
4462 | ENDIF |
---|
4463 | ENDDO |
---|
4464 | ENDDO |
---|
4465 | |
---|
4466 | ELSE |
---|
4467 | ! |
---|
4468 | !-- Decide whether the particle loop runs over the subboxes or only over 1, number_of_particles. |
---|
4469 | !-- This depends on the selected interpolation method. |
---|
4470 | IF ( interpolation_trilinear ) THEN |
---|
4471 | subbox_start = 0 |
---|
4472 | subbox_end = 7 |
---|
4473 | ELSE |
---|
4474 | subbox_start = 1 |
---|
4475 | subbox_end = 1 |
---|
4476 | ENDIF |
---|
4477 | !-- loop over subboxes. In case of simple interpolation scheme no subboxes are introduced, as |
---|
4478 | !-- they are not required. Accordingly, this loop goes from 1 to 1. |
---|
4479 | DO nb = subbox_start, subbox_end |
---|
4480 | IF ( interpolation_trilinear ) THEN |
---|
4481 | particle_start = start_index(nb) |
---|
4482 | particle_end = end_index(nb) |
---|
4483 | ELSE |
---|
4484 | particle_start = 1 |
---|
4485 | particle_end = number_of_particles |
---|
4486 | ENDIF |
---|
4487 | ! |
---|
4488 | !-- Loop from particle start to particle end |
---|
4489 | DO n = particle_start, particle_end |
---|
4490 | |
---|
4491 | ! |
---|
4492 | !-- Transport of particles with inertia |
---|
4493 | particles(n)%x = xv(n) + particles(n)%speed_x * dt_particle(n) |
---|
4494 | particles(n)%y = yv(n) + particles(n)%speed_y * dt_particle(n) |
---|
4495 | particles(n)%z = zv(n) + particles(n)%speed_z * dt_particle(n) |
---|
4496 | ! |
---|
4497 | !-- Update of the particle velocity |
---|
4498 | IF ( cloud_droplets ) THEN |
---|
4499 | ! |
---|
4500 | !-- Terminal velocity is computed for vertical direction (Rogers et al., 1993, |
---|
4501 | !-- J. Appl. Meteorol.) |
---|
4502 | diameter = particles(n)%radius * 2000.0_wp !diameter in mm |
---|
4503 | IF ( diameter <= d0_rog ) THEN |
---|
4504 | w_s = k_cap_rog * diameter * ( 1.0_wp - EXP( -k_low_rog * diameter ) ) |
---|
4505 | ELSE |
---|
4506 | w_s = a_rog - b_rog * EXP( -c_rog * diameter ) |
---|
4507 | ENDIF |
---|
4508 | |
---|
4509 | ! |
---|
4510 | !-- If selected, add random velocities following Soelch and Kaercher |
---|
4511 | !-- (2010, Q. J. R. Meteorol. Soc.) |
---|
4512 | IF ( use_sgs_for_particles ) THEN |
---|
4513 | lagr_timescale(n) = km(kp,jp,ip) / MAX( e(kp,jp,ip), 1.0E-20_wp ) |
---|
4514 | rl = EXP( -1.0_wp * dt_3d / MAX( lagr_timescale(n), 1.0E-20_wp ) ) |
---|
4515 | sigma = SQRT( e(kp,jp,ip) ) |
---|
4516 | |
---|
4517 | ! |
---|
4518 | !-- Calculate random component of particle sgs velocity using parallel random |
---|
4519 | !-- generator |
---|
4520 | CALL random_number_parallel_gauss( random_dummy ) |
---|
4521 | rg1 = random_dummy |
---|
4522 | CALL random_number_parallel_gauss( random_dummy ) |
---|
4523 | rg2 = random_dummy |
---|
4524 | CALL random_number_parallel_gauss( random_dummy ) |
---|
4525 | rg3 = random_dummy |
---|
4526 | |
---|
4527 | particles(n)%rvar1 = rl * particles(n)%rvar1 + & |
---|
4528 | SQRT( 1.0_wp - rl**2 ) * sigma * rg1 |
---|
4529 | particles(n)%rvar2 = rl * particles(n)%rvar2 + & |
---|
4530 | SQRT( 1.0_wp - rl**2 ) * sigma * rg2 |
---|
4531 | particles(n)%rvar3 = rl * particles(n)%rvar3 + & |
---|
4532 | SQRT( 1.0_wp - rl**2 ) * sigma * rg3 |
---|
4533 | |
---|
4534 | particles(n)%speed_x = u_int(n) + particles(n)%rvar1 |
---|
4535 | particles(n)%speed_y = v_int(n) + particles(n)%rvar2 |
---|
4536 | particles(n)%speed_z = w_int(n) + particles(n)%rvar3 - w_s |
---|
4537 | ELSE |
---|
4538 | particles(n)%speed_x = u_int(n) |
---|
4539 | particles(n)%speed_y = v_int(n) |
---|
4540 | particles(n)%speed_z = w_int(n) - w_s |
---|
4541 | ENDIF |
---|
4542 | |
---|
4543 | ELSE |
---|
4544 | |
---|
4545 | IF ( use_sgs_for_particles ) THEN |
---|
4546 | exp_arg = particle_groups(particles(n)%group)%exp_arg |
---|
4547 | exp_term = EXP( -exp_arg * dt_particle(n) ) |
---|
4548 | ELSE |
---|
4549 | exp_arg = particle_groups(particles(n)%group)%exp_arg |
---|
4550 | exp_term = particle_groups(particles(n)%group)%exp_term |
---|
4551 | ENDIF |
---|
4552 | particles(n)%speed_x = particles(n)%speed_x * exp_term + & |
---|
4553 | u_int(n) * ( 1.0_wp - exp_term ) |
---|
4554 | particles(n)%speed_y = particles(n)%speed_y * exp_term + & |
---|
4555 | v_int(n) * ( 1.0_wp - exp_term ) |
---|
4556 | particles(n)%speed_z = particles(n)%speed_z * exp_term + & |
---|
4557 | ( w_int(n) - ( 1.0_wp - dens_ratio(n) ) * g / & |
---|
4558 | exp_arg ) * ( 1.0_wp - exp_term ) |
---|
4559 | ENDIF |
---|
4560 | ENDDO |
---|
4561 | ENDDO |
---|
4562 | |
---|
4563 | ENDIF |
---|
4564 | |
---|
4565 | ! |
---|
4566 | !-- Store the old age of the particle ( needed to prevent that a particle crosses several PEs during |
---|
4567 | !-- one timestep, and for the evaluation of the subgrid particle velocity fluctuations ) |
---|
4568 | particles(1:number_of_particles)%age_m = particles(1:number_of_particles)%age |
---|
4569 | |
---|
4570 | ! |
---|
4571 | !-- loop over subboxes. In case of simple interpolation scheme no subboxes are introduced, as |
---|
4572 | !-- they are not required. Accordingly, this loop goes from 1 to 1. |
---|
4573 | ! |
---|
4574 | !-- Decide whether the particle loop runs over the subboxes or only over 1, number_of_particles. |
---|
4575 | !-- This depends on the selected interpolation method. |
---|
4576 | IF ( interpolation_trilinear ) THEN |
---|
4577 | subbox_start = 0 |
---|
4578 | subbox_end = 7 |
---|
4579 | ELSE |
---|
4580 | subbox_start = 1 |
---|
4581 | subbox_end = 1 |
---|
4582 | ENDIF |
---|
4583 | DO nb = subbox_start, subbox_end |
---|
4584 | IF ( interpolation_trilinear ) THEN |
---|
4585 | particle_start = start_index(nb) |
---|
4586 | particle_end = end_index(nb) |
---|
4587 | ELSE |
---|
4588 | particle_start = 1 |
---|
4589 | particle_end = number_of_particles |
---|
4590 | ENDIF |
---|
4591 | ! |
---|
4592 | !-- Loop from particle start to particle end and increment the particle age and the total time |
---|
4593 | !-- that the particle has advanced within the particle timestep procedure. |
---|
4594 | DO n = particle_start, particle_end |
---|
4595 | particles(n)%age = particles(n)%age + dt_particle(n) |
---|
4596 | particles(n)%dt_sum = particles(n)%dt_sum + dt_particle(n) |
---|
4597 | ENDDO |
---|
4598 | ! |
---|
4599 | !-- Particles that leave the child domain during the SGS-timestep loop |
---|
4600 | !-- must not continue timestepping until they are transferred to the |
---|
4601 | !-- parent. Hence, set their dt_sum to dt. |
---|
4602 | IF ( child_domain .AND. use_sgs_for_particles ) THEN |
---|
4603 | DO n = particle_start, particle_end |
---|
4604 | IF ( particles(n)%x < 0.0_wp .OR. & |
---|
4605 | particles(n)%y < 0.0_wp .OR. & |
---|
4606 | particles(n)%x > ( nx+1 ) * dx .OR. & |
---|
4607 | particles(n)%y < ( ny+1 ) * dy ) THEN |
---|
4608 | particles(n)%dt_sum = dt_3d |
---|
4609 | ENDIF |
---|
4610 | ENDDO |
---|
4611 | ENDIF |
---|
4612 | ! |
---|
4613 | !-- Check whether there is still a particle that has not yet completed the total LES timestep |
---|
4614 | DO n = particle_start, particle_end |
---|
4615 | IF ( ( dt_3d - particles(n)%dt_sum ) > 1E-8_wp ) dt_3d_reached_l = .FALSE. |
---|
4616 | ENDDO |
---|
4617 | ENDDO |
---|
4618 | |
---|
4619 | CALL cpu_log( log_point_s(44), 'lpm_advec', 'pause' ) |
---|
4620 | |
---|
4621 | |
---|
4622 | END SUBROUTINE lpm_advec |
---|
4623 | |
---|
4624 | |
---|
4625 | !--------------------------------------------------------------------------------------------------! |
---|
4626 | ! Description: |
---|
4627 | ! ------------ |
---|
4628 | !> Calculation of subgrid-scale particle speed using the stochastic model |
---|
4629 | !> of Weil et al. (2004, JAS, 61, 2877-2887). |
---|
4630 | !--------------------------------------------------------------------------------------------------! |
---|
4631 | SUBROUTINE weil_stochastic_eq( v_sgs, fs_n, e_n, dedxi_n, dedt_n, diss_n, dt_n, rg_n, fac ) |
---|
4632 | |
---|
4633 | REAL(wp) :: a1 !< dummy argument |
---|
4634 | REAL(wp) :: dedt_n !< time derivative of TKE at particle position |
---|
4635 | REAL(wp) :: dedxi_n !< horizontal derivative of TKE at particle position |
---|
4636 | REAL(wp) :: diss_n !< dissipation at particle position |
---|
4637 | REAL(wp) :: dt_n !< particle timestep |
---|
4638 | REAL(wp) :: e_n !< TKE at particle position |
---|
4639 | REAL(wp) :: fac !< flag to identify adjacent topography |
---|
4640 | REAL(wp) :: fs_n !< weighting factor to prevent that subgrid-scale particle speed becomes too large |
---|
4641 | REAL(wp) :: rg_n !< random number |
---|
4642 | REAL(wp) :: term1 !< memory term |
---|
4643 | REAL(wp) :: term2 !< drift correction term |
---|
4644 | REAL(wp) :: term3 !< random term |
---|
4645 | REAL(wp) :: v_sgs !< subgrid-scale velocity component |
---|
4646 | |
---|
4647 | !-- At first, limit TKE to a small non-zero number, in order to prevent the occurrence of extremely |
---|
4648 | !-- large SGS-velocities in case TKE is zero, (could occur at the simulation begin). |
---|
4649 | e_n = MAX( e_n, 1E-20_wp ) |
---|
4650 | ! |
---|
4651 | !-- Please note, terms 1 and 2 (drift and memory term, respectively) are multiplied by a flag to |
---|
4652 | !-- switch of both terms near topography. |
---|
4653 | !-- This is necessary, as both terms may cause a subgrid-scale velocity build up if particles are |
---|
4654 | !-- trapped in regions with very small TKE, e.g. in narrow street canyons resolved by only a few |
---|
4655 | !-- grid points. Hence, term 1 and term 2 are disabled if one of the adjacent grid points belongs to |
---|
4656 | !-- topography. |
---|
4657 | !-- Moreover, in this case, the previous subgrid-scale component is also set to zero. |
---|
4658 | |
---|
4659 | a1 = fs_n * c_0 * diss_n |
---|
4660 | ! |
---|
4661 | !-- Memory term |
---|
4662 | term1 = - a1 * v_sgs * dt_n / ( 4.0_wp * sgs_wf_part * e_n + 1E-20_wp ) * fac |
---|
4663 | ! |
---|
4664 | !-- Drift correction term |
---|
4665 | term2 = ( ( dedt_n * v_sgs / e_n ) + dedxi_n ) * 0.5_wp * dt_n * fac |
---|
4666 | ! |
---|
4667 | !-- Random term |
---|
4668 | term3 = SQRT( MAX( a1, 1E-20_wp ) ) * ( rg_n - 1.0_wp ) * SQRT( dt_n ) |
---|
4669 | ! |
---|
4670 | !-- In case one of the adjacent grid-boxes belongs to topograhy, the previous subgrid-scale velocity |
---|
4671 | !-- component is set to zero, in order to prevent a velocity build-up. |
---|
4672 | !-- This case, set also previous subgrid-scale component to zero. |
---|
4673 | v_sgs = v_sgs * fac + term1 + term2 + term3 |
---|
4674 | |
---|
4675 | END SUBROUTINE weil_stochastic_eq |
---|
4676 | |
---|
4677 | |
---|
4678 | !--------------------------------------------------------------------------------------------------! |
---|
4679 | ! Description: |
---|
4680 | ! ------------ |
---|
4681 | !> swap timelevel in case of particle advection interpolation 'simple-corrector' |
---|
4682 | !> This routine is called at the end of one timestep, the velocities are then used for the next |
---|
4683 | !> timestep |
---|
4684 | !--------------------------------------------------------------------------------------------------! |
---|
4685 | SUBROUTINE lpm_swap_timelevel_for_particle_advection |
---|
4686 | |
---|
4687 | ! |
---|
4688 | !-- Save the divergence free velocites of t+1 to use them at the end of the next time step |
---|
4689 | u_t = u |
---|
4690 | v_t = v |
---|
4691 | w_t = w |
---|
4692 | |
---|
4693 | END SUBROUTINE lpm_swap_timelevel_for_particle_advection |
---|
4694 | |
---|
4695 | |
---|
4696 | !--------------------------------------------------------------------------------------------------! |
---|
4697 | ! Description: |
---|
4698 | ! ------------ |
---|
4699 | !> Boundary conditions for the Lagrangian particles. |
---|
4700 | !> The routine consists of two different parts. One handles the bottom (flat) and top boundary. In |
---|
4701 | !> this part, also particles which exceeded their lifetime are deleted. |
---|
4702 | !> The other part handles the reflection of particles from vertical walls. |
---|
4703 | !> This part was developed by Jin Zhang during 2006-2007. |
---|
4704 | !> |
---|
4705 | !> To do: Code structure for finding the t_index values and for checking the reflection conditions |
---|
4706 | !> ------ is basically the same for all four cases, so it should be possible to further |
---|
4707 | !> simplify/shorten it. |
---|
4708 | !> |
---|
4709 | !> THE WALLS PART OF THIS ROUTINE HAS NOT BEEN TESTED FOR OCEAN RUNS SO FAR!!!! |
---|
4710 | !> (see offset_ocean_*) |
---|
4711 | !--------------------------------------------------------------------------------------------------! |
---|
4712 | SUBROUTINE lpm_boundary_conds( location_bc , i, j, k ) |
---|
4713 | |
---|
4714 | CHARACTER (LEN=*), INTENT(IN) :: location_bc !< general mode: boundary conditions at bottom/top of the model domain |
---|
4715 | !< or at vertical surfaces (buildings, terrain steps) |
---|
4716 | INTEGER(iwp), INTENT(IN) :: i !< grid index of particle box along x |
---|
4717 | INTEGER(iwp), INTENT(IN) :: j !< grid index of particle box along y |
---|
4718 | INTEGER(iwp), INTENT(IN) :: k !< grid index of particle box along z |
---|
4719 | |
---|
4720 | INTEGER(iwp) :: inc !< dummy for sorting algorithmus |
---|
4721 | INTEGER(iwp) :: ir !< dummy for sorting algorithmus |
---|
4722 | INTEGER(iwp) :: i1 !< grid index (x) of old particle position |
---|
4723 | INTEGER(iwp) :: i2 !< grid index (x) of current particle position |
---|
4724 | INTEGER(iwp) :: i3 !< grid index (x) of intermediate particle position |
---|
4725 | INTEGER(iwp) :: index_reset !< index reset height |
---|
4726 | INTEGER(iwp) :: jr !< dummy for sorting algorithmus |
---|
4727 | INTEGER(iwp) :: j1 !< grid index (y) of old particle position |
---|
4728 | INTEGER(iwp) :: j2 !< grid index (y) of current particle position |
---|
4729 | INTEGER(iwp) :: j3 !< grid index (y) of intermediate particle position |
---|
4730 | INTEGER(iwp) :: k1 !< grid index (z) of old particle position |
---|
4731 | INTEGER(iwp) :: k2 !< grid index (z) of current particle position |
---|
4732 | INTEGER(iwp) :: k3 !< grid index (z) of intermediate particle position |
---|
4733 | INTEGER(iwp) :: n !< particle number |
---|
4734 | INTEGER(iwp) :: particles_top !< maximum reset height |
---|
4735 | INTEGER(iwp) :: t_index !< running index for intermediate particle timesteps in reflection algorithmus |
---|
4736 | INTEGER(iwp) :: t_index_number !< number of intermediate particle timesteps in reflection algorithmus |
---|
4737 | INTEGER(iwp) :: tmp_x !< dummy for sorting algorithm |
---|
4738 | INTEGER(iwp) :: tmp_y !< dummy for sorting algorithm |
---|
4739 | INTEGER(iwp) :: tmp_z !< dummy for sorting algorithm |
---|
4740 | |
---|
4741 | INTEGER(iwp), DIMENSION(0:10) :: x_ind(0:10) = 0 !< index array (x) of intermediate particle positions |
---|
4742 | INTEGER(iwp), DIMENSION(0:10) :: y_ind(0:10) = 0 !< index array (y) of intermediate particle positions |
---|
4743 | INTEGER(iwp), DIMENSION(0:10) :: z_ind(0:10) = 0 !< index array (z) of intermediate particle positions |
---|
4744 | |
---|
4745 | LOGICAL :: cross_wall_x !< flag to check if particle reflection along x is necessary |
---|
4746 | LOGICAL :: cross_wall_y !< flag to check if particle reflection along y is necessary |
---|
4747 | LOGICAL :: cross_wall_z !< flag to check if particle reflection along z is necessary |
---|
4748 | LOGICAL :: reflect_x !< flag to check if particle is already reflected along x |
---|
4749 | LOGICAL :: reflect_y !< flag to check if particle is already reflected along y |
---|
4750 | LOGICAL :: reflect_z !< flag to check if particle is already reflected along z |
---|
4751 | LOGICAL :: tmp_reach_x !< dummy for sorting algorithmus |
---|
4752 | LOGICAL :: tmp_reach_y !< dummy for sorting algorithmus |
---|
4753 | LOGICAL :: tmp_reach_z !< dummy for sorting algorithmus |
---|
4754 | LOGICAL :: x_wall_reached !< flag to check if particle has already reached wall |
---|
4755 | LOGICAL :: y_wall_reached !< flag to check if particle has already reached wall |
---|
4756 | LOGICAL :: z_wall_reached !< flag to check if particle has already reached wall |
---|
4757 | |
---|
4758 | LOGICAL, DIMENSION(0:10) :: reach_x !< flag to check if particle is at a yz-wall |
---|
4759 | LOGICAL, DIMENSION(0:10) :: reach_y !< flag to check if particle is at a xz-wall |
---|
4760 | LOGICAL, DIMENSION(0:10) :: reach_z !< flag to check if particle is at a xy-wall |
---|
4761 | |
---|
4762 | REAL(wp) :: dt_particle !< particle timestep |
---|
4763 | REAL(wp) :: eps = 1E-10_wp !< security number to check if particle has reached a wall |
---|
4764 | REAL(wp) :: pos_x !< intermediate particle position (x) |
---|
4765 | REAL(wp) :: pos_x_old !< particle position (x) at previous particle timestep |
---|
4766 | REAL(wp) :: pos_y !< intermediate particle position (y) |
---|
4767 | REAL(wp) :: pos_y_old !< particle position (y) at previous particle timestep |
---|
4768 | REAL(wp) :: pos_z !< intermediate particle position (z) |
---|
4769 | REAL(wp) :: pos_z_old !< particle position (z) at previous particle timestep |
---|
4770 | REAL(wp) :: prt_x !< current particle position (x) |
---|
4771 | REAL(wp) :: prt_y !< current particle position (y) |
---|
4772 | REAL(wp) :: prt_z !< current particle position (z) |
---|
4773 | REAL(wp) :: reset_top !< location of wall in z |
---|
4774 | REAL(wp) :: t_old !< previous reflection time |
---|
4775 | REAL(wp) :: tmp_t !< dummy for sorting algorithmus |
---|
4776 | REAL(wp) :: xwall !< location of wall in x |
---|
4777 | REAL(wp) :: ywall !< location of wall in y |
---|
4778 | REAL(wp) :: zwall !< location of wall in z |
---|
4779 | |
---|
4780 | REAL(wp), DIMENSION(0:10) :: t !< reflection time |
---|
4781 | |
---|
4782 | SELECT CASE ( location_bc ) |
---|
4783 | |
---|
4784 | CASE ( 'bottom/top' ) |
---|
4785 | |
---|
4786 | ! |
---|
4787 | !-- Apply boundary conditions to those particles that have crossed the top or bottom boundary and |
---|
4788 | !-- delete those particles, which are older than allowed |
---|
4789 | DO n = 1, number_of_particles |
---|
4790 | |
---|
4791 | ! |
---|
4792 | !-- Stop if particles have moved further than the length of one PE subdomain (newly released |
---|
4793 | !-- particles have age = age_m!) |
---|
4794 | IF ( particles(n)%age /= particles(n)%age_m ) THEN |
---|
4795 | IF ( ABS(particles(n)%speed_x) > & |
---|
4796 | ((nxr-nxl+2)*dx)/(particles(n)%age-particles(n)%age_m) .OR. & |
---|
4797 | ABS(particles(n)%speed_y) > & |
---|
4798 | ((nyn-nys+2)*dy)/(particles(n)%age-particles(n)%age_m) ) THEN |
---|
4799 | |
---|
4800 | WRITE( message_string, * ) 'particle too fast. n = ', n |
---|
4801 | CALL message( 'lpm_boundary_conds', 'PA0148', 2, 2, -1, 6, 1 ) |
---|
4802 | ENDIF |
---|
4803 | ENDIF |
---|
4804 | |
---|
4805 | IF ( particles(n)%age > particle_maximum_age .AND. particles(n)%particle_mask ) THEN |
---|
4806 | particles(n)%particle_mask = .FALSE. |
---|
4807 | deleted_particles = deleted_particles + 1 |
---|
4808 | ENDIF |
---|
4809 | |
---|
4810 | IF ( particles(n)%z >= zw(nz) .AND. particles(n)%particle_mask ) THEN |
---|
4811 | IF ( ibc_par_t == 1 ) THEN |
---|
4812 | ! |
---|
4813 | !-- Particle absorption |
---|
4814 | particles(n)%particle_mask = .FALSE. |
---|
4815 | deleted_particles = deleted_particles + 1 |
---|
4816 | ELSEIF ( ibc_par_t == 2 ) THEN |
---|
4817 | ! |
---|
4818 | !-- Particle reflection |
---|
4819 | particles(n)%z = 2.0_wp * zw(nz) - particles(n)%z |
---|
4820 | particles(n)%speed_z = -particles(n)%speed_z |
---|
4821 | IF ( use_sgs_for_particles .AND. & |
---|
4822 | particles(n)%rvar3 > 0.0_wp ) THEN |
---|
4823 | particles(n)%rvar3 = -particles(n)%rvar3 |
---|
4824 | ENDIF |
---|
4825 | ENDIF |
---|
4826 | ENDIF |
---|
4827 | |
---|
4828 | IF ( particles(n)%z < zw(0) .AND. particles(n)%particle_mask ) THEN |
---|
4829 | IF ( ibc_par_b == 1 ) THEN |
---|
4830 | ! |
---|
4831 | !-- Particle absorption |
---|
4832 | particles(n)%particle_mask = .FALSE. |
---|
4833 | deleted_particles = deleted_particles + 1 |
---|
4834 | ELSEIF ( ibc_par_b == 2 ) THEN |
---|
4835 | ! |
---|
4836 | !-- Particle reflection |
---|
4837 | particles(n)%z = 2.0_wp * zw(0) - particles(n)%z |
---|
4838 | particles(n)%speed_z = -particles(n)%speed_z |
---|
4839 | IF ( use_sgs_for_particles .AND. & |
---|
4840 | particles(n)%rvar3 < 0.0_wp ) THEN |
---|
4841 | particles(n)%rvar3 = -particles(n)%rvar3 |
---|
4842 | ENDIF |
---|
4843 | ELSEIF ( ibc_par_b == 3 ) THEN |
---|
4844 | ! |
---|
4845 | !-- Find reset height. @note this works only in non-strechted cases |
---|
4846 | particles_top = INT( pst(1) / dz(1) ) |
---|
4847 | index_reset = MINLOC( prt_count(nzb+1:particles_top,j,i), DIM = 1 ) |
---|
4848 | reset_top = zu(index_reset) |
---|
4849 | CALL random_number_parallel( random_dummy ) |
---|
4850 | particles(n)%z = reset_top * ( 1.0 + ( random_dummy / 10.0_wp) ) |
---|
4851 | particles(n)%speed_z = 0.0_wp |
---|
4852 | IF ( curvature_solution_effects ) THEN |
---|
4853 | particles(n)%radius = particles(n)%aux1 |
---|
4854 | ELSE |
---|
4855 | particles(n)%radius = 1.0E-8 |
---|
4856 | ENDIF |
---|
4857 | ENDIF |
---|
4858 | ENDIF |
---|
4859 | ENDDO |
---|
4860 | |
---|
4861 | CASE ( 'walls' ) |
---|
4862 | |
---|
4863 | CALL cpu_log( log_point_s(48), 'lpm_wall_reflect', 'start' ) |
---|
4864 | |
---|
4865 | DO n = 1, number_of_particles |
---|
4866 | ! |
---|
4867 | !-- Recalculate particle timestep |
---|
4868 | dt_particle = particles(n)%age - particles(n)%age_m |
---|
4869 | ! |
---|
4870 | !-- Obtain x/y indices for current particle position |
---|
4871 | i2 = particles(n)%x * ddx |
---|
4872 | j2 = particles(n)%y * ddy |
---|
4873 | IF ( zw(k) < particles(n)%z ) k2 = k + 1 |
---|
4874 | IF ( zw(k) > particles(n)%z .AND. zw(k-1) < particles(n)%z ) k2 = k |
---|
4875 | IF ( zw(k-1) > particles(n)%z ) k2 = k - 1 |
---|
4876 | ! |
---|
4877 | !-- Save current particle positions |
---|
4878 | prt_x = particles(n)%x |
---|
4879 | prt_y = particles(n)%y |
---|
4880 | prt_z = particles(n)%z |
---|
4881 | ! |
---|
4882 | !-- Recalculate old particle positions |
---|
4883 | pos_x_old = particles(n)%x - particles(n)%speed_x * dt_particle |
---|
4884 | pos_y_old = particles(n)%y - particles(n)%speed_y * dt_particle |
---|
4885 | pos_z_old = particles(n)%z - particles(n)%speed_z * dt_particle |
---|
4886 | ! |
---|
4887 | !-- Obtain x/y indices for old particle positions |
---|
4888 | i1 = i |
---|
4889 | j1 = j |
---|
4890 | k1 = k |
---|
4891 | ! |
---|
4892 | !-- Determine horizontal as well as vertical walls at which particle can be potentially |
---|
4893 | !-- reflected. |
---|
4894 | !-- Start with walls aligned in yz layer. |
---|
4895 | !-- Wall to the right |
---|
4896 | IF ( prt_x > pos_x_old ) THEN |
---|
4897 | xwall = ( i1 + 1 ) * dx |
---|
4898 | ! |
---|
4899 | !-- Wall to the left |
---|
4900 | ELSE |
---|
4901 | xwall = i1 * dx |
---|
4902 | ENDIF |
---|
4903 | ! |
---|
4904 | !-- Walls aligned in xz layer |
---|
4905 | !-- Wall to the north |
---|
4906 | IF ( prt_y > pos_y_old ) THEN |
---|
4907 | ywall = ( j1 + 1 ) * dy |
---|
4908 | !-- Wall to the south |
---|
4909 | ELSE |
---|
4910 | ywall = j1 * dy |
---|
4911 | ENDIF |
---|
4912 | |
---|
4913 | IF ( prt_z > pos_z_old ) THEN |
---|
4914 | zwall = zw(k) |
---|
4915 | ELSE |
---|
4916 | zwall = zw(k-1) |
---|
4917 | ENDIF |
---|
4918 | ! |
---|
4919 | !-- Initialize flags to check if particle reflection is necessary |
---|
4920 | cross_wall_x = .FALSE. |
---|
4921 | cross_wall_y = .FALSE. |
---|
4922 | cross_wall_z = .FALSE. |
---|
4923 | ! |
---|
4924 | !-- Initialize flags to check if a wall is reached |
---|
4925 | reach_x = .FALSE. |
---|
4926 | reach_y = .FALSE. |
---|
4927 | reach_z = .FALSE. |
---|
4928 | ! |
---|
4929 | !-- Initialize flags to check if a particle was already reflected |
---|
4930 | reflect_x = .FALSE. |
---|
4931 | reflect_y = .FALSE. |
---|
4932 | reflect_z = .FALSE. |
---|
4933 | ! |
---|
4934 | !-- Initialize flags to check if a wall is already crossed. |
---|
4935 | !-- ( Required to obtain correct indices. ) |
---|
4936 | x_wall_reached = .FALSE. |
---|
4937 | y_wall_reached = .FALSE. |
---|
4938 | z_wall_reached = .FALSE. |
---|
4939 | ! |
---|
4940 | !-- Initialize time array |
---|
4941 | t = 0.0_wp |
---|
4942 | ! |
---|
4943 | !-- Check if particle can reach any wall. This case, calculate the fractional time needed to |
---|
4944 | !-- reach this wall. Store this fractional timestep in array t. Moreover, store indices for |
---|
4945 | !-- these grid boxes where the respective wall belongs to. |
---|
4946 | !-- Start with x-direction. |
---|
4947 | t_index = 1 |
---|
4948 | t(t_index) = ( xwall - pos_x_old ) & |
---|
4949 | / MERGE( MAX( prt_x - pos_x_old, 1E-30_wp ), & |
---|
4950 | MIN( prt_x - pos_x_old, -1E-30_wp ), & |
---|
4951 | prt_x > pos_x_old ) |
---|
4952 | x_ind(t_index) = i2 |
---|
4953 | y_ind(t_index) = j1 |
---|
4954 | z_ind(t_index) = k1 |
---|
4955 | reach_x(t_index) = .TRUE. |
---|
4956 | reach_y(t_index) = .FALSE. |
---|
4957 | reach_z(t_index) = .FALSE. |
---|
4958 | ! |
---|
4959 | !-- Store these values only if particle really reaches any wall. t must be in an interval |
---|
4960 | !-- between [0:1]. |
---|
4961 | IF ( t(t_index) <= 1.0_wp .AND. t(t_index) >= 0.0_wp ) THEN |
---|
4962 | t_index = t_index + 1 |
---|
4963 | cross_wall_x = .TRUE. |
---|
4964 | ENDIF |
---|
4965 | ! |
---|
4966 | !-- y-direction |
---|
4967 | t(t_index) = ( ywall - pos_y_old ) & |
---|
4968 | / MERGE( MAX( prt_y - pos_y_old, 1E-30_wp ), & |
---|
4969 | MIN( prt_y - pos_y_old, -1E-30_wp ), & |
---|
4970 | prt_y > pos_y_old ) |
---|
4971 | x_ind(t_index) = i1 |
---|
4972 | y_ind(t_index) = j2 |
---|
4973 | z_ind(t_index) = k1 |
---|
4974 | reach_x(t_index) = .FALSE. |
---|
4975 | reach_y(t_index) = .TRUE. |
---|
4976 | reach_z(t_index) = .FALSE. |
---|
4977 | IF ( t(t_index) <= 1.0_wp .AND. t(t_index) >= 0.0_wp ) THEN |
---|
4978 | t_index = t_index + 1 |
---|
4979 | cross_wall_y = .TRUE. |
---|
4980 | ENDIF |
---|
4981 | ! |
---|
4982 | !-- z-direction |
---|
4983 | t(t_index) = (zwall - pos_z_old ) & |
---|
4984 | / MERGE( MAX( prt_z - pos_z_old, 1E-30_wp ), & |
---|
4985 | MIN( prt_z - pos_z_old, -1E-30_wp ), & |
---|
4986 | prt_z > pos_z_old ) |
---|
4987 | |
---|
4988 | x_ind(t_index) = i1 |
---|
4989 | y_ind(t_index) = j1 |
---|
4990 | z_ind(t_index) = k2 |
---|
4991 | reach_x(t_index) = .FALSE. |
---|
4992 | reach_y(t_index) = .FALSE. |
---|
4993 | reach_z(t_index) = .TRUE. |
---|
4994 | IF( t(t_index) <= 1.0_wp .AND. t(t_index) >= 0.0_wp) THEN |
---|
4995 | t_index = t_index + 1 |
---|
4996 | cross_wall_z = .TRUE. |
---|
4997 | ENDIF |
---|
4998 | |
---|
4999 | t_index_number = t_index - 1 |
---|
5000 | ! |
---|
5001 | !-- Carry out reflection only if particle reaches any wall |
---|
5002 | IF ( cross_wall_x .OR. cross_wall_y .OR. cross_wall_z ) THEN |
---|
5003 | ! |
---|
5004 | !-- Sort fractional timesteps in ascending order. Also sort the corresponding indices and |
---|
5005 | !-- flag according to the time interval a particle reaches the respective wall. |
---|
5006 | inc = 1 |
---|
5007 | jr = 1 |
---|
5008 | DO WHILE ( inc <= t_index_number ) |
---|
5009 | inc = 3 * inc + 1 |
---|
5010 | ENDDO |
---|
5011 | |
---|
5012 | DO WHILE ( inc > 1 ) |
---|
5013 | inc = inc / 3 |
---|
5014 | DO ir = inc+1, t_index_number |
---|
5015 | tmp_t = t(ir) |
---|
5016 | tmp_x = x_ind(ir) |
---|
5017 | tmp_y = y_ind(ir) |
---|
5018 | tmp_z = z_ind(ir) |
---|
5019 | tmp_reach_x = reach_x(ir) |
---|
5020 | tmp_reach_y = reach_y(ir) |
---|
5021 | tmp_reach_z = reach_z(ir) |
---|
5022 | jr = ir |
---|
5023 | DO WHILE ( t(jr-inc) > tmp_t ) |
---|
5024 | t(jr) = t(jr-inc) |
---|
5025 | x_ind(jr) = x_ind(jr-inc) |
---|
5026 | y_ind(jr) = y_ind(jr-inc) |
---|
5027 | z_ind(jr) = z_ind(jr-inc) |
---|
5028 | reach_x(jr) = reach_x(jr-inc) |
---|
5029 | reach_y(jr) = reach_y(jr-inc) |
---|
5030 | reach_z(jr) = reach_z(jr-inc) |
---|
5031 | jr = jr - inc |
---|
5032 | IF ( jr <= inc ) EXIT |
---|
5033 | ENDDO |
---|
5034 | t(jr) = tmp_t |
---|
5035 | x_ind(jr) = tmp_x |
---|
5036 | y_ind(jr) = tmp_y |
---|
5037 | z_ind(jr) = tmp_z |
---|
5038 | reach_x(jr) = tmp_reach_x |
---|
5039 | reach_y(jr) = tmp_reach_y |
---|
5040 | reach_z(jr) = tmp_reach_z |
---|
5041 | ENDDO |
---|
5042 | ENDDO |
---|
5043 | ! |
---|
5044 | !-- Initialize temporary particle positions |
---|
5045 | pos_x = pos_x_old |
---|
5046 | pos_y = pos_y_old |
---|
5047 | pos_z = pos_z_old |
---|
5048 | ! |
---|
5049 | !-- Loop over all times a particle possibly moves into a new grid box |
---|
5050 | t_old = 0.0_wp |
---|
5051 | DO t_index = 1, t_index_number |
---|
5052 | ! |
---|
5053 | !-- Calculate intermediate particle position according to the timesteps a particle |
---|
5054 | !-- reaches any wall. |
---|
5055 | pos_x = pos_x + ( t(t_index) - t_old ) * dt_particle * particles(n)%speed_x |
---|
5056 | pos_y = pos_y + ( t(t_index) - t_old ) * dt_particle * particles(n)%speed_y |
---|
5057 | pos_z = pos_z + ( t(t_index) - t_old ) * dt_particle * particles(n)%speed_z |
---|
5058 | ! |
---|
5059 | !-- Obtain x/y grid indices for intermediate particle position from sorted index array |
---|
5060 | i3 = x_ind(t_index) |
---|
5061 | j3 = y_ind(t_index) |
---|
5062 | k3 = z_ind(t_index) |
---|
5063 | ! |
---|
5064 | !-- Check which wall is already reached |
---|
5065 | IF ( .NOT. x_wall_reached ) x_wall_reached = reach_x(t_index) |
---|
5066 | IF ( .NOT. y_wall_reached ) y_wall_reached = reach_y(t_index) |
---|
5067 | IF ( .NOT. z_wall_reached ) z_wall_reached = reach_z(t_index) |
---|
5068 | ! |
---|
5069 | !-- Check if a particle needs to be reflected at any yz-wall. If necessary, carry out |
---|
5070 | !-- reflection. Please note, a security constant is required, as the particle position |
---|
5071 | !-- does not necessarily exactly match the wall location due to rounding errors. |
---|
5072 | IF ( reach_x(t_index) .AND. & |
---|
5073 | ABS( pos_x - xwall ) < eps .AND. & |
---|
5074 | .NOT. BTEST(wall_flags_total_0(k3,j3,i3),0) .AND. & |
---|
5075 | .NOT. reflect_x ) THEN |
---|
5076 | ! |
---|
5077 | ! |
---|
5078 | !-- Reflection in x-direction. |
---|
5079 | !-- Ensure correct reflection by MIN/MAX functions, depending on direction of |
---|
5080 | !-- particle transport. |
---|
5081 | !-- Due to rounding errors pos_x does not exactly match the wall location, leading to |
---|
5082 | !-- erroneous reflection. |
---|
5083 | pos_x = MERGE( MIN( 2.0_wp * xwall - pos_x, xwall ), & |
---|
5084 | MAX( 2.0_wp * xwall - pos_x, xwall ), & |
---|
5085 | particles(n)%x > xwall ) |
---|
5086 | ! |
---|
5087 | !-- Change sign of particle speed |
---|
5088 | particles(n)%speed_x = - particles(n)%speed_x |
---|
5089 | ! |
---|
5090 | !-- Also change sign of subgrid-scale particle speed |
---|
5091 | particles(n)%rvar1 = - particles(n)%rvar1 |
---|
5092 | ! |
---|
5093 | !-- Set flag that reflection along x is already done |
---|
5094 | reflect_x = .TRUE. |
---|
5095 | ! |
---|
5096 | !-- As the particle does not cross any further yz-wall during this timestep, set |
---|
5097 | !-- further x-indices to the current one. |
---|
5098 | x_ind(t_index:t_index_number) = i1 |
---|
5099 | ! |
---|
5100 | !-- If particle already reached the wall but was not reflected, set further x-indices to |
---|
5101 | !-- the new one. |
---|
5102 | ELSEIF ( x_wall_reached .AND. .NOT. reflect_x ) THEN |
---|
5103 | x_ind(t_index:t_index_number) = i2 |
---|
5104 | ENDIF !particle reflection in x direction done |
---|
5105 | |
---|
5106 | ! |
---|
5107 | !-- Check if a particle needs to be reflected at any xz-wall. If necessary, carry out |
---|
5108 | !-- reflection. Please note, a security constant is required, as the particle position |
---|
5109 | !-- does not necessarily exactly match the wall location due to rounding errors. |
---|
5110 | IF ( reach_y(t_index) .AND. & |
---|
5111 | ABS( pos_y - ywall ) < eps .AND. & |
---|
5112 | .NOT. BTEST(wall_flags_total_0(k3,j3,i3),0) .AND. & |
---|
5113 | .NOT. reflect_y ) THEN |
---|
5114 | ! |
---|
5115 | ! |
---|
5116 | !-- Reflection in y-direction. |
---|
5117 | !-- Ensure correct reflection by MIN/MAX functions, depending on direction of |
---|
5118 | !-- particle transport. |
---|
5119 | !-- Due to rounding errors pos_y does not exactly match the wall location, leading to |
---|
5120 | !-- erroneous reflection. |
---|
5121 | pos_y = MERGE( MIN( 2.0_wp * ywall - pos_y, ywall ), & |
---|
5122 | MAX( 2.0_wp * ywall - pos_y, ywall ), & |
---|
5123 | particles(n)%y > ywall ) |
---|
5124 | ! |
---|
5125 | !-- Change sign of particle speed |
---|
5126 | particles(n)%speed_y = - particles(n)%speed_y |
---|
5127 | ! |
---|
5128 | !-- Also change sign of subgrid-scale particle speed |
---|
5129 | particles(n)%rvar2 = - particles(n)%rvar2 |
---|
5130 | ! |
---|
5131 | !-- Set flag that reflection along y is already done |
---|
5132 | reflect_y = .TRUE. |
---|
5133 | ! |
---|
5134 | !-- As the particle does not cross any further xz-wall during this timestep, set |
---|
5135 | !-- further y-indices to the current one. |
---|
5136 | y_ind(t_index:t_index_number) = j1 |
---|
5137 | ! |
---|
5138 | !-- If particle already reached the wall but was not reflected, set further y-indices to |
---|
5139 | !-- the new one. |
---|
5140 | ELSEIF ( y_wall_reached .AND. .NOT. reflect_y ) THEN |
---|
5141 | y_ind(t_index:t_index_number) = j2 |
---|
5142 | ENDIF !particle reflection in y direction done |
---|
5143 | |
---|
5144 | ! |
---|
5145 | !-- Check if a particle needs to be reflected at any xy-wall. If necessary, carry out |
---|
5146 | !-- reflection. Please note, a security constant is required, as the particle position |
---|
5147 | !-- does not necessarily exactly match the wall location due to rounding errors. |
---|
5148 | IF ( reach_z(t_index) .AND. & |
---|
5149 | ABS( pos_z - zwall ) < eps .AND. & |
---|
5150 | .NOT. BTEST(wall_flags_total_0(k3,j3,i3),0) .AND. & |
---|
5151 | .NOT. reflect_z ) THEN |
---|
5152 | ! |
---|
5153 | ! |
---|
5154 | !-- Reflection in z-direction. |
---|
5155 | !-- Ensure correct reflection by MIN/MAX functions, depending on direction of |
---|
5156 | !-- particle transport. |
---|
5157 | !-- Due to rounding errors pos_z does not exactly match the wall location, leading to |
---|
5158 | !-- erroneous reflection. |
---|
5159 | pos_z = MERGE( MIN( 2.0_wp * zwall - pos_z, zwall ), & |
---|
5160 | MAX( 2.0_wp * zwall - pos_z, zwall ), & |
---|
5161 | particles(n)%z > zwall ) |
---|
5162 | ! |
---|
5163 | !-- Change sign of particle speed |
---|
5164 | particles(n)%speed_z = - particles(n)%speed_z |
---|
5165 | ! |
---|
5166 | !-- Also change sign of subgrid-scale particle speed |
---|
5167 | particles(n)%rvar3 = - particles(n)%rvar3 |
---|
5168 | ! |
---|
5169 | !-- Set flag that reflection along z is already done |
---|
5170 | reflect_z = .TRUE. |
---|
5171 | ! |
---|
5172 | !-- As the particle does not cross any further xy-wall during this timestep, set |
---|
5173 | !-- further z-indices to the current one. |
---|
5174 | z_ind(t_index:t_index_number) = k1 |
---|
5175 | ! |
---|
5176 | !-- If particle already reached the wall but was not reflected, set further z-indices to |
---|
5177 | !-- the new one. |
---|
5178 | ELSEIF ( z_wall_reached .AND. .NOT. reflect_z ) THEN |
---|
5179 | z_ind(t_index:t_index_number) = k2 |
---|
5180 | ENDIF !particle reflection in z direction done |
---|
5181 | |
---|
5182 | ! |
---|
5183 | !-- Swap time |
---|
5184 | t_old = t(t_index) |
---|
5185 | |
---|
5186 | ENDDO |
---|
5187 | ! |
---|
5188 | !-- If a particle was reflected, calculate final position from last intermediate position. |
---|
5189 | IF ( reflect_x .OR. reflect_y .OR. reflect_z ) THEN |
---|
5190 | |
---|
5191 | particles(n)%x = pos_x + ( 1.0_wp - t_old ) * dt_particle * particles(n)%speed_x |
---|
5192 | particles(n)%y = pos_y + ( 1.0_wp - t_old ) * dt_particle * particles(n)%speed_y |
---|
5193 | particles(n)%z = pos_z + ( 1.0_wp - t_old ) * dt_particle * particles(n)%speed_z |
---|
5194 | |
---|
5195 | ENDIF |
---|
5196 | |
---|
5197 | ENDIF |
---|
5198 | |
---|
5199 | ENDDO |
---|
5200 | |
---|
5201 | CALL cpu_log( log_point_s(48), 'lpm_wall_reflect', 'stop' ) |
---|
5202 | |
---|
5203 | CASE DEFAULT |
---|
5204 | CONTINUE |
---|
5205 | |
---|
5206 | END SELECT |
---|
5207 | |
---|
5208 | END SUBROUTINE lpm_boundary_conds |
---|
5209 | |
---|
5210 | |
---|
5211 | !--------------------------------------------------------------------------------------------------! |
---|
5212 | ! Description: |
---|
5213 | ! ------------ |
---|
5214 | !> Calculates change in droplet radius by condensation/evaporation, using either an analytic formula |
---|
5215 | !> or by numerically integrating the radius growth equation including curvature and solution effects |
---|
5216 | !> using Rosenbrocks method (see Numerical recipes in FORTRAN, 2nd edition, p. 731). |
---|
5217 | !> The analytical formula and growth equation follow those given in |
---|
5218 | !> Rogers and Yau (A short course in cloud physics, 3rd edition, p. 102/103). |
---|
5219 | !--------------------------------------------------------------------------------------------------! |
---|
5220 | SUBROUTINE lpm_droplet_condensation (i,j,k) |
---|
5221 | |
---|
5222 | ! |
---|
5223 | !-- Parameters for Rosenbrock method (see Verwer et al., 1999) |
---|
5224 | REAL(wp), PARAMETER :: prec = 1.0E-3_wp !< precision of Rosenbrock solution |
---|
5225 | REAL(wp), PARAMETER :: q_increase = 1.5_wp !< increase factor in timestep |
---|
5226 | REAL(wp), PARAMETER :: q_decrease = 0.9_wp !< decrease factor in timestep |
---|
5227 | REAL(wp), PARAMETER :: gamma = 0.292893218814_wp !< = 1.0 - 1.0 / SQRT(2.0) |
---|
5228 | ! |
---|
5229 | !-- Parameters for terminal velocity |
---|
5230 | REAL(wp), PARAMETER :: a_rog = 9.65_wp !< parameter for fall velocity |
---|
5231 | REAL(wp), PARAMETER :: b_rog = 10.43_wp !< parameter for fall velocity |
---|
5232 | REAL(wp), PARAMETER :: c_rog = 0.6_wp !< parameter for fall velocity |
---|
5233 | REAL(wp), PARAMETER :: k_cap_rog = 4.0_wp !< parameter for fall velocity |
---|
5234 | REAL(wp), PARAMETER :: k_low_rog = 12.0_wp !< parameter for fall velocity |
---|
5235 | REAL(wp), PARAMETER :: d0_rog = 0.745_wp !< separation diameter |
---|
5236 | |
---|
5237 | INTEGER(iwp), INTENT(IN) :: i !< |
---|
5238 | INTEGER(iwp), INTENT(IN) :: j !< |
---|
5239 | INTEGER(iwp), INTENT(IN) :: k !< |
---|
5240 | INTEGER(iwp) :: n !< |
---|
5241 | |
---|
5242 | REAL(wp) :: afactor !< curvature effects |
---|
5243 | REAL(wp) :: arg !< |
---|
5244 | REAL(wp) :: bfactor !< solute effects |
---|
5245 | REAL(wp) :: ddenom !< |
---|
5246 | REAL(wp) :: delta_r !< |
---|
5247 | REAL(wp) :: diameter !< diameter of cloud droplets |
---|
5248 | REAL(wp) :: diff_coeff !< diffusivity for water vapor |
---|
5249 | REAL(wp) :: drdt !< |
---|
5250 | REAL(wp) :: dt_ros !< |
---|
5251 | REAL(wp) :: dt_ros_sum !< |
---|
5252 | REAL(wp) :: d2rdtdr !< |
---|
5253 | REAL(wp) :: e_a !< current vapor pressure |
---|
5254 | REAL(wp) :: e_s !< current saturation vapor pressure |
---|
5255 | REAL(wp) :: error !< local truncation error in Rosenbrock |
---|
5256 | REAL(wp) :: k1 !< |
---|
5257 | REAL(wp) :: k2 !< |
---|
5258 | REAL(wp) :: r_err !< First order estimate of Rosenbrock radius |
---|
5259 | REAL(wp) :: r_ros !< Rosenbrock radius |
---|
5260 | REAL(wp) :: r_ros_ini !< initial Rosenbrock radius |
---|
5261 | REAL(wp) :: r0 !< gas-kinetic lengthscale |
---|
5262 | REAL(wp) :: re_p !< particle Reynolds number |
---|
5263 | REAL(wp) :: sigma !< surface tension of water |
---|
5264 | REAL(wp) :: thermal_conductivity !< thermal conductivity for water |
---|
5265 | REAL(wp) :: t_int !< temperature |
---|
5266 | REAL(wp) :: w_s !< terminal velocity of droplets |
---|
5267 | |
---|
5268 | REAL(wp), DIMENSION(number_of_particles) :: new_r !< |
---|
5269 | REAL(wp), DIMENSION(number_of_particles) :: ventilation_effect !< |
---|
5270 | |
---|
5271 | CALL cpu_log( log_point_s(42), 'lpm_droplet_condens', 'start' ) |
---|
5272 | |
---|
5273 | ! |
---|
5274 | !-- Absolute temperature |
---|
5275 | t_int = pt(k,j,i) * exner(k) |
---|
5276 | ! |
---|
5277 | !-- Saturation vapor pressure (Eq. 10 in Bolton, 1980) |
---|
5278 | e_s = magnus( t_int ) |
---|
5279 | ! |
---|
5280 | !-- Current vapor pressure |
---|
5281 | e_a = q(k,j,i) * hyp(k) / ( q(k,j,i) + rd_d_rv ) |
---|
5282 | ! |
---|
5283 | !-- Thermal conductivity for water (from Rogers and Yau, Table 7.1) |
---|
5284 | thermal_conductivity = 7.94048E-05_wp * t_int + 0.00227011_wp |
---|
5285 | ! |
---|
5286 | !-- Moldecular diffusivity of water vapor in air (Hall und Pruppacher, 1976) |
---|
5287 | diff_coeff = 0.211E-4_wp * ( t_int / 273.15_wp )**1.94_wp * ( 101325.0_wp / hyp(k) ) |
---|
5288 | ! |
---|
5289 | !-- Lengthscale for gas-kinetic effects (from Mordy, 1959, p. 23): |
---|
5290 | r0 = diff_coeff / 0.036_wp * SQRT( 2.0_wp * pi / ( r_v * t_int ) ) |
---|
5291 | ! |
---|
5292 | !-- Calculate effects of heat conductivity and diffusion of water vapor on the |
---|
5293 | !-- diffusional growth process (usually known as 1.0 / (F_k + F_d) ) |
---|
5294 | ddenom = 1.0_wp / ( rho_l * r_v * t_int / ( e_s * diff_coeff ) + & |
---|
5295 | ( l_v / ( r_v * t_int ) - 1.0_wp ) * rho_l * & |
---|
5296 | l_v / ( thermal_conductivity * t_int ) & |
---|
5297 | ) |
---|
5298 | new_r = 0.0_wp |
---|
5299 | ! |
---|
5300 | !-- Determine ventilation effect on evaporation of large drops |
---|
5301 | DO n = 1, number_of_particles |
---|
5302 | |
---|
5303 | IF ( particles(n)%radius >= 4.0E-5_wp .AND. e_a / e_s < 1.0_wp ) THEN |
---|
5304 | ! |
---|
5305 | !-- Terminal velocity is computed for vertical direction (Rogers et al., |
---|
5306 | !-- 1993, J. Appl. Meteorol.) |
---|
5307 | diameter = particles(n)%radius * 2000.0_wp !diameter in mm |
---|
5308 | IF ( diameter <= d0_rog ) THEN |
---|
5309 | w_s = k_cap_rog * diameter * ( 1.0_wp - EXP( -k_low_rog * diameter ) ) |
---|
5310 | ELSE |
---|
5311 | w_s = a_rog - b_rog * EXP( -c_rog * diameter ) |
---|
5312 | ENDIF |
---|
5313 | ! |
---|
5314 | !-- Calculate droplet's Reynolds number |
---|
5315 | re_p = 2.0_wp * particles(n)%radius * w_s / molecular_viscosity |
---|
5316 | ! |
---|
5317 | !-- Ventilation coefficient (Rogers and Yau, 1989): |
---|
5318 | IF ( re_p > 2.5_wp ) THEN |
---|
5319 | ventilation_effect(n) = 0.78_wp + 0.28_wp * SQRT( re_p ) |
---|
5320 | ELSE |
---|
5321 | ventilation_effect(n) = 1.0_wp + 0.09_wp * re_p |
---|
5322 | ENDIF |
---|
5323 | ELSE |
---|
5324 | ! |
---|
5325 | !-- For small droplets or in supersaturated environments, the ventilation effect does not play |
---|
5326 | !-- a role. |
---|
5327 | ventilation_effect(n) = 1.0_wp |
---|
5328 | ENDIF |
---|
5329 | ENDDO |
---|
5330 | |
---|
5331 | IF( .NOT. curvature_solution_effects ) THEN |
---|
5332 | ! |
---|
5333 | !-- Use analytic model for diffusional growth including gas-kinetic effects (Mordy, 1959) but |
---|
5334 | !-- without the impact of aerosols. |
---|
5335 | DO n = 1, number_of_particles |
---|
5336 | arg = ( particles(n)%radius + r0 )**2 + 2.0_wp * dt_3d * ddenom * & |
---|
5337 | ventilation_effect(n) * & |
---|
5338 | ( e_a / e_s - 1.0_wp ) |
---|
5339 | arg = MAX( arg, ( 0.01E-6 + r0 )**2 ) |
---|
5340 | new_r(n) = SQRT( arg ) - r0 |
---|
5341 | ENDDO |
---|
5342 | |
---|
5343 | ELSE |
---|
5344 | ! |
---|
5345 | !-- Integrate the diffusional growth including gas-kinetic (Mordy, 1959), |
---|
5346 | !-- as well as curvature and solute effects (e.g., Köhler, 1936). |
---|
5347 | ! |
---|
5348 | !-- Curvature effect (afactor) with surface tension (sigma) by Straka (2009) |
---|
5349 | sigma = 0.0761_wp - 0.000155_wp * ( t_int - 273.15_wp ) |
---|
5350 | ! |
---|
5351 | !-- Solute effect (afactor) |
---|
5352 | afactor = 2.0_wp * sigma / ( rho_l * r_v * t_int ) |
---|
5353 | |
---|
5354 | DO n = 1, number_of_particles |
---|
5355 | ! |
---|
5356 | !-- Solute effect (bfactor) |
---|
5357 | bfactor = vanthoff * rho_s * particles(n)%aux1**3 * & |
---|
5358 | molecular_weight_of_water / ( rho_l * molecular_weight_of_solute ) |
---|
5359 | |
---|
5360 | dt_ros = particles(n)%aux2 ! use previously stored Rosenbrock timestep |
---|
5361 | dt_ros_sum = 0.0_wp |
---|
5362 | |
---|
5363 | r_ros = particles(n)%radius ! initialize Rosenbrock particle radius |
---|
5364 | r_ros_ini = r_ros |
---|
5365 | ! |
---|
5366 | !-- Integrate growth equation using a 2nd-order Rosenbrock method |
---|
5367 | !-- (see Verwer et al., 1999, Eq. (3.2)). The Rosenbrock method adjusts its with internal |
---|
5368 | !-- timestep to minimize the local truncation error. |
---|
5369 | DO WHILE ( dt_ros_sum < dt_3d ) |
---|
5370 | |
---|
5371 | dt_ros = MIN( dt_ros, dt_3d - dt_ros_sum ) |
---|
5372 | |
---|
5373 | DO |
---|
5374 | |
---|
5375 | drdt = ddenom * ventilation_effect(n) * ( e_a / e_s - 1.0_wp - & |
---|
5376 | afactor / r_ros + & |
---|
5377 | bfactor / r_ros**3 & |
---|
5378 | ) / ( r_ros + r0 ) |
---|
5379 | |
---|
5380 | d2rdtdr = -ddenom * ventilation_effect(n) * ( & |
---|
5381 | ( e_a / e_s - 1.0_wp ) * r_ros**4 - & |
---|
5382 | afactor * r0 * r_ros**2 - & |
---|
5383 | 2.0_wp * afactor * r_ros**3 + & |
---|
5384 | 3.0_wp * bfactor * r0 + & |
---|
5385 | 4.0_wp * bfactor * r_ros & |
---|
5386 | ) & |
---|
5387 | / ( r_ros**4 * ( r_ros + r0 )**2 ) |
---|
5388 | |
---|
5389 | k1 = drdt / ( 1.0_wp - gamma * dt_ros * d2rdtdr ) |
---|
5390 | |
---|
5391 | r_ros = MAX(r_ros_ini + k1 * dt_ros, particles(n)%aux1) |
---|
5392 | r_err = r_ros |
---|
5393 | |
---|
5394 | drdt = ddenom * ventilation_effect(n) * ( e_a / e_s - 1.0_wp - & |
---|
5395 | afactor / r_ros + & |
---|
5396 | bfactor / r_ros**3 & |
---|
5397 | ) / ( r_ros + r0 ) |
---|
5398 | |
---|
5399 | k2 = ( drdt - dt_ros * 2.0 * gamma * d2rdtdr * k1 ) / & |
---|
5400 | ( 1.0_wp - dt_ros * gamma * d2rdtdr ) |
---|
5401 | |
---|
5402 | r_ros = MAX(r_ros_ini + dt_ros * ( 1.5_wp * k1 + 0.5_wp * k2), particles(n)%aux1) |
---|
5403 | ! |
---|
5404 | !-- Check error of the solution, and reduce dt_ros if necessary. |
---|
5405 | error = ABS(r_err - r_ros) / r_ros |
---|
5406 | IF ( error > prec ) THEN |
---|
5407 | dt_ros = SQRT( q_decrease * prec / error ) * dt_ros |
---|
5408 | r_ros = r_ros_ini |
---|
5409 | ELSE |
---|
5410 | dt_ros_sum = dt_ros_sum + dt_ros |
---|
5411 | dt_ros = q_increase * dt_ros |
---|
5412 | r_ros_ini = r_ros |
---|
5413 | EXIT |
---|
5414 | ENDIF |
---|
5415 | |
---|
5416 | END DO |
---|
5417 | |
---|
5418 | END DO !Rosenbrock loop |
---|
5419 | ! |
---|
5420 | !-- Store new particle radius |
---|
5421 | new_r(n) = r_ros |
---|
5422 | ! |
---|
5423 | !-- Store internal time step value for next PALM step |
---|
5424 | particles(n)%aux2 = dt_ros |
---|
5425 | |
---|
5426 | ENDDO !Particle loop |
---|
5427 | |
---|
5428 | ENDIF |
---|
5429 | |
---|
5430 | DO n = 1, number_of_particles |
---|
5431 | ! |
---|
5432 | !-- Sum up the change in liquid water for the respective grid box for the computation of the |
---|
5433 | !-- release/depletion of water vapor and heat. |
---|
5434 | ql_c(k,j,i) = ql_c(k,j,i) + particles(n)%weight_factor * & |
---|
5435 | rho_l * 1.33333333_wp * pi * & |
---|
5436 | ( new_r(n)**3 - particles(n)%radius**3 ) / & |
---|
5437 | ( rho_surface * dx * dy * dzw(k) ) |
---|
5438 | ! |
---|
5439 | !-- Check if the increase in liqid water is not too big. If this is the case, the model timestep |
---|
5440 | !-- might be too long. |
---|
5441 | IF ( ql_c(k,j,i) > 100.0_wp ) THEN |
---|
5442 | WRITE( message_string, * ) 'k=',k,' j=',j,' i=',i, & |
---|
5443 | ' ql_c=',ql_c(k,j,i), '&part(',n,')%wf=', & |
---|
5444 | particles(n)%weight_factor,' delta_r=',delta_r |
---|
5445 | CALL message( 'lpm_droplet_condensation', 'PA0143', 2, 2, -1, 6, 1 ) |
---|
5446 | ENDIF |
---|
5447 | ! |
---|
5448 | !-- Check if the change in the droplet radius is not too big. If this is the case, the model |
---|
5449 | !-- timestep might be too long. |
---|
5450 | delta_r = new_r(n) - particles(n)%radius |
---|
5451 | IF ( delta_r < 0.0_wp .AND. new_r(n) < 0.0_wp ) THEN |
---|
5452 | WRITE( message_string, * ) '#1 k=',k,' j=',j,' i=',i, & |
---|
5453 | ' e_s=',e_s, ' e_a=',e_a,' t_int=',t_int, & |
---|
5454 | '&delta_r=',delta_r, & |
---|
5455 | ' particle_radius=',particles(n)%radius |
---|
5456 | CALL message( 'lpm_droplet_condensation', 'PA0144', 2, 2, -1, 6, 1 ) |
---|
5457 | ENDIF |
---|
5458 | ! |
---|
5459 | !-- Sum up the total volume of liquid water (needed below for re-calculating the weighting |
---|
5460 | !-- factors) |
---|
5461 | ql_v(k,j,i) = ql_v(k,j,i) + particles(n)%weight_factor * new_r(n)**3 |
---|
5462 | ! |
---|
5463 | !-- Determine radius class of the particle needed for collision |
---|
5464 | IF ( use_kernel_tables ) THEN |
---|
5465 | particles(n)%class = ( LOG( new_r(n) ) - rclass_lbound ) / & |
---|
5466 | ( rclass_ubound - rclass_lbound ) * radius_classes |
---|
5467 | particles(n)%class = MIN( particles(n)%class, radius_classes ) |
---|
5468 | particles(n)%class = MAX( particles(n)%class, 1 ) |
---|
5469 | ENDIF |
---|
5470 | ! |
---|
5471 | !-- Store new radius to particle features |
---|
5472 | particles(n)%radius = new_r(n) |
---|
5473 | |
---|
5474 | ENDDO |
---|
5475 | |
---|
5476 | CALL cpu_log( log_point_s(42), 'lpm_droplet_condens', 'stop' ) |
---|
5477 | |
---|
5478 | |
---|
5479 | END SUBROUTINE lpm_droplet_condensation |
---|
5480 | |
---|
5481 | |
---|
5482 | !--------------------------------------------------------------------------------------------------! |
---|
5483 | ! Description: |
---|
5484 | ! ------------ |
---|
5485 | !> Release of latent heat and change of mixing ratio due to condensation / evaporation of droplets. |
---|
5486 | !--------------------------------------------------------------------------------------------------! |
---|
5487 | SUBROUTINE lpm_interaction_droplets_ptq |
---|
5488 | |
---|
5489 | INTEGER(iwp) :: i !< running index x direction |
---|
5490 | INTEGER(iwp) :: j !< running index y direction |
---|
5491 | INTEGER(iwp) :: k !< running index z direction |
---|
5492 | |
---|
5493 | REAL(wp) :: flag !< flag to mask topography grid points |
---|
5494 | |
---|
5495 | DO i = nxl, nxr |
---|
5496 | DO j = nys, nyn |
---|
5497 | DO k = nzb+1, nzt |
---|
5498 | ! |
---|
5499 | !-- Predetermine flag to mask topography |
---|
5500 | flag = MERGE( 1.0_wp, 0.0_wp, BTEST( wall_flags_total_0(k,j,i), 0 ) ) |
---|
5501 | |
---|
5502 | q(k,j,i) = q(k,j,i) - ql_c(k,j,i) * flag |
---|
5503 | pt(k,j,i) = pt(k,j,i) + lv_d_cp * ql_c(k,j,i) * d_exner(k) * flag |
---|
5504 | ENDDO |
---|
5505 | ENDDO |
---|
5506 | ENDDO |
---|
5507 | |
---|
5508 | END SUBROUTINE lpm_interaction_droplets_ptq |
---|
5509 | |
---|
5510 | |
---|
5511 | !--------------------------------------------------------------------------------------------------! |
---|
5512 | ! Description: |
---|
5513 | ! ------------ |
---|
5514 | !> Release of latent heat and change of mixing ratio due to condensation / evaporation of droplets. |
---|
5515 | !> Call for grid point i,j |
---|
5516 | !--------------------------------------------------------------------------------------------------! |
---|
5517 | SUBROUTINE lpm_interaction_droplets_ptq_ij( i, j ) |
---|
5518 | |
---|
5519 | INTEGER(iwp) :: i !< running index x direction |
---|
5520 | INTEGER(iwp) :: j !< running index y direction |
---|
5521 | INTEGER(iwp) :: k !< running index z direction |
---|
5522 | |
---|
5523 | REAL(wp) :: flag !< flag to mask topography grid points |
---|
5524 | |
---|
5525 | |
---|
5526 | DO k = nzb+1, nzt |
---|
5527 | ! |
---|
5528 | !-- Predetermine flag to mask topography |
---|
5529 | flag = MERGE( 1.0_wp, 0.0_wp, BTEST( wall_flags_total_0(k,j,i), 0 ) ) |
---|
5530 | |
---|
5531 | q(k,j,i) = q(k,j,i) - ql_c(k,j,i) * flag |
---|
5532 | pt(k,j,i) = pt(k,j,i) + lv_d_cp * ql_c(k,j,i) * d_exner(k) * flag |
---|
5533 | ENDDO |
---|
5534 | |
---|
5535 | END SUBROUTINE lpm_interaction_droplets_ptq_ij |
---|
5536 | |
---|
5537 | |
---|
5538 | !--------------------------------------------------------------------------------------------------! |
---|
5539 | ! Description: |
---|
5540 | ! ------------ |
---|
5541 | !> Calculate the liquid water content for each grid box. |
---|
5542 | !--------------------------------------------------------------------------------------------------! |
---|
5543 | SUBROUTINE lpm_calc_liquid_water_content |
---|
5544 | |
---|
5545 | |
---|
5546 | INTEGER(iwp) :: i !< |
---|
5547 | INTEGER(iwp) :: j !< |
---|
5548 | INTEGER(iwp) :: k !< |
---|
5549 | INTEGER(iwp) :: n !< |
---|
5550 | |
---|
5551 | CALL cpu_log( log_point_s(45), 'lpm_calc_ql', 'start' ) |
---|
5552 | |
---|
5553 | ! |
---|
5554 | !-- Set water content initially to zero |
---|
5555 | ql = 0.0_wp; ql_v = 0.0_wp; ql_vp = 0.0_wp |
---|
5556 | |
---|
5557 | ! |
---|
5558 | !-- Calculate for each grid box |
---|
5559 | DO i = nxl, nxr |
---|
5560 | DO j = nys, nyn |
---|
5561 | DO k = nzb+1, nzt |
---|
5562 | number_of_particles = prt_count(k,j,i) |
---|
5563 | IF ( number_of_particles <= 0 ) CYCLE |
---|
5564 | particles => grid_particles(k,j,i)%particles(1:number_of_particles) |
---|
5565 | ! |
---|
5566 | !-- Calculate the total volume in the boxes (ql_v, weighting factor has to beincluded) |
---|
5567 | DO n = 1, prt_count(k,j,i) |
---|
5568 | ql_v(k,j,i) = ql_v(k,j,i) + particles(n)%weight_factor * particles(n)%radius**3 |
---|
5569 | ENDDO |
---|
5570 | ! |
---|
5571 | !-- Calculate the liquid water content |
---|
5572 | IF ( ql_v(k,j,i) /= 0.0_wp ) THEN |
---|
5573 | ql(k,j,i) = ql(k,j,i) + rho_l * 1.33333333_wp * pi * & |
---|
5574 | ql_v(k,j,i) / ( rho_surface * dx * dy * dzw(k) ) |
---|
5575 | IF ( ql(k,j,i) < 0.0_wp ) THEN |
---|
5576 | WRITE( message_string, * ) 'LWC out of range: ' , ql(k,j,i),i,j,k |
---|
5577 | CALL message( 'lpm_calc_liquid_water_content', 'PA0719', 2, 2, -1, 6, 1 ) |
---|
5578 | ENDIF |
---|
5579 | ELSE |
---|
5580 | ql(k,j,i) = 0.0_wp |
---|
5581 | ENDIF |
---|
5582 | ENDDO |
---|
5583 | ENDDO |
---|
5584 | ENDDO |
---|
5585 | |
---|
5586 | CALL cpu_log( log_point_s(45), 'lpm_calc_ql', 'stop' ) |
---|
5587 | |
---|
5588 | END SUBROUTINE lpm_calc_liquid_water_content |
---|
5589 | |
---|
5590 | |
---|
5591 | !--------------------------------------------------------------------------------------------------! |
---|
5592 | ! Description: |
---|
5593 | ! ------------ |
---|
5594 | !> Calculates change in droplet radius by collision. Droplet collision is calculated for each grid |
---|
5595 | !> box seperately. Collision is parameterized by using collision kernels. Two different kernels are |
---|
5596 | !> available: |
---|
5597 | !> Hall kernel: Kernel from Hall (1980, J. Atmos. Sci., 2486-2507), which considers collision due to |
---|
5598 | !> pure gravitational effects. |
---|
5599 | !> Wang kernel: Beside gravitational effects (treated with the Hall-kernel) also the effects of |
---|
5600 | !> turbulence on the collision are considered using parameterizations of Ayala et al. |
---|
5601 | !> (2008, New J. Phys., 10, 075015) and Wang and Grabowski (2009, Atmos. Sci. Lett., |
---|
5602 | !> 10, 1-8). This kernel includes three possible effects of turbulence: |
---|
5603 | !> the modification of the relative velocity between the droplets, |
---|
5604 | !> the effect of preferential concentration, |
---|
5605 | !> and the enhancement of collision efficiencies. |
---|
5606 | !--------------------------------------------------------------------------------------------------! |
---|
5607 | SUBROUTINE lpm_droplet_collision (i,j,k) |
---|
5608 | |
---|
5609 | INTEGER(iwp), INTENT(IN) :: i !< |
---|
5610 | INTEGER(iwp), INTENT(IN) :: j !< |
---|
5611 | INTEGER(iwp), INTENT(IN) :: k !< |
---|
5612 | |
---|
5613 | INTEGER(iwp) :: eclass !< |
---|
5614 | INTEGER(iwp) :: n !< |
---|
5615 | INTEGER(iwp) :: m !< |
---|
5616 | INTEGER(iwp) :: rclass_l !< |
---|
5617 | INTEGER(iwp) :: rclass_s !< |
---|
5618 | |
---|
5619 | REAL(wp) :: collection_probability !< probability for collection |
---|
5620 | REAL(wp) :: ddV !< inverse grid box volume |
---|
5621 | REAL(wp) :: epsilon_collision !< dissipation rate |
---|
5622 | REAL(wp) :: factor_volume_to_mass !< 4.0 / 3.0 * pi * rho_l |
---|
5623 | REAL(wp) :: xm !< droplet mass of super-droplet m |
---|
5624 | REAL(wp) :: xn !< droplet mass of super-droplet n |
---|
5625 | REAL(wp) :: xsm !< aerosol mass of super-droplet m |
---|
5626 | REAL(wp) :: xsn !< aerosol mass of super-droplet n |
---|
5627 | |
---|
5628 | REAL(wp), DIMENSION(:), ALLOCATABLE :: aero_mass !< total aerosol mass of super droplet |
---|
5629 | REAL(wp), DIMENSION(:), ALLOCATABLE :: mass !< total mass of super droplet |
---|
5630 | REAL(wp), DIMENSION(:), ALLOCATABLE :: weight !< weighting factor |
---|
5631 | |
---|
5632 | CALL cpu_log( log_point_s(43), 'lpm_droplet_coll', 'start' ) |
---|
5633 | |
---|
5634 | number_of_particles = prt_count(k,j,i) |
---|
5635 | factor_volume_to_mass = 4.0_wp / 3.0_wp * pi * rho_l |
---|
5636 | ddV = 1.0_wp / ( dx * dy * dzw(k) ) |
---|
5637 | ! |
---|
5638 | !-- Collision requires at least one super droplet inside the box |
---|
5639 | IF ( number_of_particles > 0 ) THEN |
---|
5640 | |
---|
5641 | IF ( use_kernel_tables ) THEN |
---|
5642 | ! |
---|
5643 | !-- Fast method with pre-calculated collection kernels for discrete radius- and |
---|
5644 | !-- dissipation-classes. |
---|
5645 | IF ( wang_kernel ) THEN |
---|
5646 | eclass = INT( diss(k,j,i) * 1.0E4_wp / 600.0_wp * dissipation_classes ) + 1 |
---|
5647 | epsilon_collision = diss(k,j,i) |
---|
5648 | ELSE |
---|
5649 | epsilon_collision = 0.0_wp |
---|
5650 | ENDIF |
---|
5651 | |
---|
5652 | IF ( hall_kernel .OR. epsilon_collision * 1.0E4_wp < 0.001_wp ) THEN |
---|
5653 | eclass = 0 ! Hall kernel is used |
---|
5654 | ELSE |
---|
5655 | eclass = MIN( dissipation_classes, eclass ) |
---|
5656 | ENDIF |
---|
5657 | |
---|
5658 | ELSE |
---|
5659 | ! |
---|
5660 | !-- Collection kernels are re-calculated for every new grid box. First, allocate memory for |
---|
5661 | !-- kernel table. |
---|
5662 | !-- Third dimension is 1, because table is re-calculated for every new dissipation value. |
---|
5663 | ALLOCATE( ckernel(1:number_of_particles,1:number_of_particles,1:1) ) |
---|
5664 | ! |
---|
5665 | !-- Now calculate collection kernel for this box. Note that the kernel is based on the |
---|
5666 | !-- previous time step |
---|
5667 | CALL recalculate_kernel( i, j, k ) |
---|
5668 | |
---|
5669 | ENDIF |
---|
5670 | ! |
---|
5671 | !-- Temporary fields for total mass of super-droplet, aerosol mass, and weighting factor are |
---|
5672 | !-- allocated. |
---|
5673 | ALLOCATE(mass(1:number_of_particles), weight(1:number_of_particles)) |
---|
5674 | IF ( curvature_solution_effects ) ALLOCATE(aero_mass(1:number_of_particles)) |
---|
5675 | |
---|
5676 | mass(1:number_of_particles) = particles(1:number_of_particles)%weight_factor * & |
---|
5677 | particles(1:number_of_particles)%radius**3 * & |
---|
5678 | factor_volume_to_mass |
---|
5679 | |
---|
5680 | weight(1:number_of_particles) = particles(1:number_of_particles)%weight_factor |
---|
5681 | |
---|
5682 | IF ( curvature_solution_effects ) THEN |
---|
5683 | aero_mass(1:number_of_particles) = particles(1:number_of_particles)%weight_factor * & |
---|
5684 | particles(1:number_of_particles)%aux1**3 * & |
---|
5685 | 4.0_wp / 3.0_wp * pi * rho_s |
---|
5686 | ENDIF |
---|
5687 | ! |
---|
5688 | !-- Calculate collision/coalescence |
---|
5689 | DO n = 1, number_of_particles |
---|
5690 | |
---|
5691 | DO m = n, number_of_particles |
---|
5692 | ! |
---|
5693 | !-- For collisions, the weighting factor of at least one super-droplet needs to be larger |
---|
5694 | !-- or equal to one. |
---|
5695 | IF ( MIN( weight(n), weight(m) ) < 1.0_wp ) CYCLE |
---|
5696 | ! |
---|
5697 | !-- Get mass of individual droplets (aerosols) |
---|
5698 | xn = mass(n) / weight(n) |
---|
5699 | xm = mass(m) / weight(m) |
---|
5700 | IF ( curvature_solution_effects ) THEN |
---|
5701 | xsn = aero_mass(n) / weight(n) |
---|
5702 | xsm = aero_mass(m) / weight(m) |
---|
5703 | ENDIF |
---|
5704 | ! |
---|
5705 | !-- Probability that the necessary collisions take place |
---|
5706 | IF ( use_kernel_tables ) THEN |
---|
5707 | rclass_l = particles(n)%class |
---|
5708 | rclass_s = particles(m)%class |
---|
5709 | |
---|
5710 | collection_probability = MAX( weight(n), weight(m) ) * & |
---|
5711 | ckernel(rclass_l,rclass_s,eclass) * ddV * dt_3d |
---|
5712 | ELSE |
---|
5713 | collection_probability = MAX( weight(n), weight(m) ) * & |
---|
5714 | ckernel(n,m,1) * ddV * dt_3d |
---|
5715 | ENDIF |
---|
5716 | ! |
---|
5717 | !-- Calculate the number of collections and consider multiple collections. |
---|
5718 | !-- (Accordingly, p_crit will be 0.0, 1.0, 2.0, ...) |
---|
5719 | CALL random_number_parallel( random_dummy ) |
---|
5720 | IF ( collection_probability - FLOOR(collection_probability) > random_dummy ) THEN |
---|
5721 | collection_probability = FLOOR(collection_probability) + 1.0_wp |
---|
5722 | ELSE |
---|
5723 | collection_probability = FLOOR(collection_probability) |
---|
5724 | ENDIF |
---|
5725 | |
---|
5726 | IF ( collection_probability > 0.0_wp ) THEN |
---|
5727 | ! |
---|
5728 | !-- Super-droplet n collects droplets of super-droplet m |
---|
5729 | IF ( weight(n) < weight(m) ) THEN |
---|
5730 | |
---|
5731 | mass(n) = mass(n) + weight(n) * xm * collection_probability |
---|
5732 | weight(m) = weight(m) - weight(n) * collection_probability |
---|
5733 | mass(m) = mass(m) - weight(n) * xm * collection_probability |
---|
5734 | IF ( curvature_solution_effects ) THEN |
---|
5735 | aero_mass(n) = aero_mass(n) + weight(n) * xsm * collection_probability |
---|
5736 | aero_mass(m) = aero_mass(m) - weight(n) * xsm * collection_probability |
---|
5737 | ENDIF |
---|
5738 | |
---|
5739 | ELSEIF ( weight(m) < weight(n) ) THEN |
---|
5740 | |
---|
5741 | mass(m) = mass(m) + weight(m) * xn * collection_probability |
---|
5742 | weight(n) = weight(n) - weight(m) * collection_probability |
---|
5743 | mass(n) = mass(n) - weight(m) * xn * collection_probability |
---|
5744 | IF ( curvature_solution_effects ) THEN |
---|
5745 | aero_mass(m) = aero_mass(m) + weight(m) * xsn * collection_probability |
---|
5746 | aero_mass(n) = aero_mass(n) - weight(m) * xsn * collection_probability |
---|
5747 | ENDIF |
---|
5748 | |
---|
5749 | ELSE |
---|
5750 | ! |
---|
5751 | !-- Collisions of particles of the same weighting factor. |
---|
5752 | !-- Particle n collects 1/2 weight(n) droplets of particle m, |
---|
5753 | !-- particle m collects 1/2 weight(m) droplets of particle n. |
---|
5754 | !-- The total mass mass changes accordingly. |
---|
5755 | !-- If n = m, the first half of the droplets coalesces with the second half of the |
---|
5756 | !-- droplets; mass is unchanged because xm = xn for n = m. |
---|
5757 | !-- |
---|
5758 | !-- Note: For m = n this equation is an approximation only valid for weight >> 1 |
---|
5759 | !-- (which is usually the case). The approximation is weight(n)-1 = weight(n). |
---|
5760 | mass(n) = mass(n) + 0.5_wp * weight(n) * ( xm - xn ) |
---|
5761 | mass(m) = mass(m) + 0.5_wp * weight(m) * ( xn - xm ) |
---|
5762 | IF ( curvature_solution_effects ) THEN |
---|
5763 | aero_mass(n) = aero_mass(n) + 0.5_wp * weight(n) * ( xsm - xsn ) |
---|
5764 | aero_mass(m) = aero_mass(m) + 0.5_wp * weight(m) * ( xsn - xsm ) |
---|
5765 | ENDIF |
---|
5766 | weight(n) = weight(n) - 0.5_wp * weight(m) |
---|
5767 | weight(m) = weight(n) |
---|
5768 | |
---|
5769 | ENDIF |
---|
5770 | |
---|
5771 | ENDIF |
---|
5772 | |
---|
5773 | ENDDO |
---|
5774 | |
---|
5775 | ql_vp(k,j,i) = ql_vp(k,j,i) + mass(n) / factor_volume_to_mass |
---|
5776 | |
---|
5777 | ENDDO |
---|
5778 | |
---|
5779 | IF ( ANY(weight < 0.0_wp) ) THEN |
---|
5780 | WRITE( message_string, * ) 'negative weighting factor' |
---|
5781 | CALL message( 'lpm_droplet_collision', 'PA0028', 2, 2, -1, 6, 1 ) |
---|
5782 | ENDIF |
---|
5783 | |
---|
5784 | particles(1:number_of_particles)%radius = ( mass(1:number_of_particles) / & |
---|
5785 | ( weight(1:number_of_particles) & |
---|
5786 | * factor_volume_to_mass & |
---|
5787 | ) & |
---|
5788 | )**0.33333333333333_wp |
---|
5789 | |
---|
5790 | IF ( curvature_solution_effects ) THEN |
---|
5791 | particles(1:number_of_particles)%aux1 = ( aero_mass(1:number_of_particles) / & |
---|
5792 | ( weight(1:number_of_particles) & |
---|
5793 | * 4.0_wp / 3.0_wp * pi * rho_s & |
---|
5794 | ) & |
---|
5795 | )**0.33333333333333_wp |
---|
5796 | ENDIF |
---|
5797 | |
---|
5798 | particles(1:number_of_particles)%weight_factor = weight(1:number_of_particles) |
---|
5799 | |
---|
5800 | DEALLOCATE( weight, mass ) |
---|
5801 | IF ( curvature_solution_effects ) DEALLOCATE( aero_mass ) |
---|
5802 | IF ( .NOT. use_kernel_tables ) DEALLOCATE( ckernel ) |
---|
5803 | |
---|
5804 | ! |
---|
5805 | !-- Check if LWC is conserved during collision process |
---|
5806 | IF ( ql_v(k,j,i) /= 0.0_wp ) THEN |
---|
5807 | IF ( ql_vp(k,j,i) / ql_v(k,j,i) >= 1.0001_wp .OR. & |
---|
5808 | ql_vp(k,j,i) / ql_v(k,j,i) <= 0.9999_wp ) THEN |
---|
5809 | WRITE( message_string, * ) ' LWC is not conserved during',' collision! ', & |
---|
5810 | ' LWC after condensation: ', ql_v(k,j,i), & |
---|
5811 | ' LWC after collision: ', ql_vp(k,j,i) |
---|
5812 | CALL message( 'lpm_droplet_collision', 'PA0040', 2, 2, -1, 6, 1 ) |
---|
5813 | ENDIF |
---|
5814 | ENDIF |
---|
5815 | |
---|
5816 | ENDIF |
---|
5817 | |
---|
5818 | CALL cpu_log( log_point_s(43), 'lpm_droplet_coll', 'stop' ) |
---|
5819 | |
---|
5820 | END SUBROUTINE lpm_droplet_collision |
---|
5821 | |
---|
5822 | !--------------------------------------------------------------------------------------------------! |
---|
5823 | ! Description: |
---|
5824 | ! ------------ |
---|
5825 | !> Initialization of the collision efficiency matrix with fixed radius and dissipation classes, |
---|
5826 | !> calculated at simulation start only. |
---|
5827 | !--------------------------------------------------------------------------------------------------! |
---|
5828 | SUBROUTINE lpm_init_kernels |
---|
5829 | |
---|
5830 | INTEGER(iwp) :: i !< |
---|
5831 | INTEGER(iwp) :: j !< |
---|
5832 | INTEGER(iwp) :: k !< |
---|
5833 | |
---|
5834 | ! |
---|
5835 | !-- Calculate collision efficiencies for fixed radius- and dissipation classes |
---|
5836 | IF ( collision_kernel(6:9) == 'fast' ) THEN |
---|
5837 | |
---|
5838 | ALLOCATE( ckernel(1:radius_classes,1:radius_classes,0:dissipation_classes), & |
---|
5839 | epsclass(1:dissipation_classes), & |
---|
5840 | radclass(1:radius_classes) ) |
---|
5841 | |
---|
5842 | ! |
---|
5843 | !-- Calculate the radius class bounds with logarithmic distances in the interval |
---|
5844 | !-- [1.0E-6, 1000.0E-6] m |
---|
5845 | rclass_lbound = LOG( 1.0E-6_wp ) |
---|
5846 | rclass_ubound = LOG( 1000.0E-6_wp ) |
---|
5847 | radclass(1) = EXP( rclass_lbound ) |
---|
5848 | DO i = 2, radius_classes |
---|
5849 | radclass(i) = EXP( rclass_lbound + & |
---|
5850 | ( rclass_ubound - rclass_lbound ) * & |
---|
5851 | ( i - 1.0_wp ) / ( radius_classes - 1.0_wp ) ) |
---|
5852 | ENDDO |
---|
5853 | |
---|
5854 | ! |
---|
5855 | !-- Set the class bounds for dissipation in interval [0.0, 600.0] cm**2/s**3 |
---|
5856 | DO i = 1, dissipation_classes |
---|
5857 | epsclass(i) = 0.06_wp * REAL( i, KIND=wp ) / dissipation_classes |
---|
5858 | ENDDO |
---|
5859 | ! |
---|
5860 | !-- Calculate collision efficiencies of the Wang/ayala kernel |
---|
5861 | ALLOCATE( ec(1:radius_classes,1:radius_classes), & |
---|
5862 | ecf(1:radius_classes,1:radius_classes), & |
---|
5863 | gck(1:radius_classes,1:radius_classes), & |
---|
5864 | winf(1:radius_classes) ) |
---|
5865 | |
---|
5866 | DO k = 1, dissipation_classes |
---|
5867 | |
---|
5868 | epsilon_collision = epsclass(k) |
---|
5869 | urms = 2.02_wp * ( epsilon_collision / 0.04_wp )**( 1.0_wp / 3.0_wp ) |
---|
5870 | |
---|
5871 | CALL turbsd |
---|
5872 | CALL turb_enhance_eff |
---|
5873 | CALL effic |
---|
5874 | |
---|
5875 | DO j = 1, radius_classes |
---|
5876 | DO i = 1, radius_classes |
---|
5877 | ckernel(i,j,k) = ec(i,j) * gck(i,j) * ecf(i,j) |
---|
5878 | ENDDO |
---|
5879 | ENDDO |
---|
5880 | |
---|
5881 | ENDDO |
---|
5882 | |
---|
5883 | ! |
---|
5884 | !-- Calculate collision efficiencies of the Hall kernel |
---|
5885 | ALLOCATE( hkernel(1:radius_classes,1:radius_classes), & |
---|
5886 | hwratio(1:radius_classes,1:radius_classes) ) |
---|
5887 | |
---|
5888 | CALL fallg |
---|
5889 | CALL effic |
---|
5890 | |
---|
5891 | DO j = 1, radius_classes |
---|
5892 | DO i = 1, radius_classes |
---|
5893 | hkernel(i,j) = pi * ( radclass(j) + radclass(i) )**2 & |
---|
5894 | * ec(i,j) * ABS( winf(j) - winf(i) ) |
---|
5895 | ckernel(i,j,0) = hkernel(i,j) ! hall kernel stored on index 0 |
---|
5896 | ENDDO |
---|
5897 | ENDDO |
---|
5898 | |
---|
5899 | ! |
---|
5900 | !-- Test output of efficiencies |
---|
5901 | IF ( j == -1 ) THEN |
---|
5902 | PRINT*, '*** Hall kernel' |
---|
5903 | WRITE ( *,'(5X,20(F4.0,1X))' ) ( radclass(i)*1.0E6_wp, i = 1,radius_classes ) |
---|
5904 | DO j = 1, radius_classes |
---|
5905 | WRITE ( *,'(F4.0,1X,20(F8.4,1X))' ) radclass(j), ( hkernel(i,j), i = 1,radius_classes ) |
---|
5906 | ENDDO |
---|
5907 | |
---|
5908 | DO k = 1, dissipation_classes |
---|
5909 | DO i = 1, radius_classes |
---|
5910 | DO j = 1, radius_classes |
---|
5911 | IF ( hkernel(i,j) == 0.0_wp ) THEN |
---|
5912 | hwratio(i,j) = 9999999.9_wp |
---|
5913 | ELSE |
---|
5914 | hwratio(i,j) = ckernel(i,j,k) / hkernel(i,j) |
---|
5915 | ENDIF |
---|
5916 | ENDDO |
---|
5917 | ENDDO |
---|
5918 | |
---|
5919 | PRINT*, '*** epsilon = ', epsclass(k) |
---|
5920 | WRITE ( *,'(5X,20(F4.0,1X))' ) ( radclass(i) * 1.0E6_wp, i = 1,radius_classes ) |
---|
5921 | DO j = 1, radius_classes |
---|
5922 | WRITE ( *,'(F4.0,1X,20(F8.4,1X))' ) radclass(j) * 1.0E6_wp, & |
---|
5923 | ( hwratio(i,j), i = 1,radius_classes ) |
---|
5924 | ENDDO |
---|
5925 | ENDDO |
---|
5926 | ENDIF |
---|
5927 | |
---|
5928 | DEALLOCATE( ec, ecf, epsclass, gck, hkernel, winf ) |
---|
5929 | |
---|
5930 | ENDIF |
---|
5931 | |
---|
5932 | END SUBROUTINE lpm_init_kernels |
---|
5933 | |
---|
5934 | !--------------------------------------------------------------------------------------------------! |
---|
5935 | ! Description: |
---|
5936 | ! ------------ |
---|
5937 | !> Calculation of collision kernels during each timestep and for each grid box |
---|
5938 | !--------------------------------------------------------------------------------------------------! |
---|
5939 | SUBROUTINE recalculate_kernel( i1, j1, k1 ) |
---|
5940 | |
---|
5941 | |
---|
5942 | INTEGER(iwp) :: i !< |
---|
5943 | INTEGER(iwp) :: i1 !< |
---|
5944 | INTEGER(iwp) :: j !< |
---|
5945 | INTEGER(iwp) :: j1 !< |
---|
5946 | INTEGER(iwp) :: k1 !< |
---|
5947 | |
---|
5948 | |
---|
5949 | number_of_particles = prt_count(k1,j1,i1) |
---|
5950 | radius_classes = number_of_particles ! necessary to use the same |
---|
5951 | ! subroutines as for |
---|
5952 | ! precalculated kernels |
---|
5953 | |
---|
5954 | ALLOCATE( ec(1:number_of_particles,1:number_of_particles), & |
---|
5955 | radclass(1:number_of_particles), winf(1:number_of_particles) ) |
---|
5956 | |
---|
5957 | ! |
---|
5958 | !-- Store particle radii on the radclass array |
---|
5959 | radclass(1:number_of_particles) = particles(1:number_of_particles)%radius |
---|
5960 | |
---|
5961 | IF ( wang_kernel ) THEN |
---|
5962 | epsilon_collision = diss(k1,j1,i1) ! dissipation rate in m**2/s**3 |
---|
5963 | ELSE |
---|
5964 | epsilon_collision = 0.0_wp |
---|
5965 | ENDIF |
---|
5966 | urms = 2.02_wp * ( epsilon_collision / 0.04_wp )**( 0.33333333333_wp ) |
---|
5967 | |
---|
5968 | IF ( wang_kernel .AND. epsilon_collision > 1.0E-7_wp ) THEN |
---|
5969 | ! |
---|
5970 | !-- Call routines to calculate efficiencies for the Wang kernel |
---|
5971 | ALLOCATE( gck(1:number_of_particles,1:number_of_particles), & |
---|
5972 | ecf(1:number_of_particles,1:number_of_particles) ) |
---|
5973 | |
---|
5974 | CALL turbsd |
---|
5975 | CALL turb_enhance_eff |
---|
5976 | CALL effic |
---|
5977 | |
---|
5978 | DO j = 1, number_of_particles |
---|
5979 | DO i = 1, number_of_particles |
---|
5980 | ckernel(1+i-1,1+j-1,1) = ec(i,j) * gck(i,j) * ecf(i,j) |
---|
5981 | ENDDO |
---|
5982 | ENDDO |
---|
5983 | |
---|
5984 | DEALLOCATE( gck, ecf ) |
---|
5985 | ELSE |
---|
5986 | ! |
---|
5987 | !-- Call routines to calculate efficiencies for the Hall kernel |
---|
5988 | CALL fallg |
---|
5989 | CALL effic |
---|
5990 | |
---|
5991 | DO j = 1, number_of_particles |
---|
5992 | DO i = 1, number_of_particles |
---|
5993 | ckernel(i,j,1) = pi * ( radclass(j) + radclass(i) )**2 & |
---|
5994 | * ec(i,j) * ABS( winf(j) - winf(i) ) |
---|
5995 | ENDDO |
---|
5996 | ENDDO |
---|
5997 | ENDIF |
---|
5998 | |
---|
5999 | DEALLOCATE( ec, radclass, winf ) |
---|
6000 | |
---|
6001 | END SUBROUTINE recalculate_kernel |
---|
6002 | |
---|
6003 | !--------------------------------------------------------------------------------------------------! |
---|
6004 | ! Description: |
---|
6005 | ! ------------ |
---|
6006 | !> Calculation of effects of turbulence on the geometric collision kernel (by including the |
---|
6007 | !> droplets' average radial relative velocities and their radial distribution function) following |
---|
6008 | !> the analytic model by Aayala et al. (2008, New J. Phys.). For details check the second part 2 of |
---|
6009 | !> the publication, page 37ff. |
---|
6010 | !> |
---|
6011 | !> Input parameters, which need to be replaced by PALM parameters: water density, air density |
---|
6012 | !--------------------------------------------------------------------------------------------------! |
---|
6013 | SUBROUTINE turbsd |
---|
6014 | |
---|
6015 | INTEGER(iwp) :: i !< |
---|
6016 | INTEGER(iwp) :: j !< |
---|
6017 | |
---|
6018 | REAL(wp) :: ao !< |
---|
6019 | REAL(wp) :: ao_gr !< |
---|
6020 | REAL(wp) :: bbb !< |
---|
6021 | REAL(wp) :: be !< |
---|
6022 | REAL(wp) :: b1 !< |
---|
6023 | REAL(wp) :: b2 !< |
---|
6024 | REAL(wp) :: ccc !< |
---|
6025 | REAL(wp) :: c1 !< |
---|
6026 | REAL(wp) :: c1_gr !< |
---|
6027 | REAL(wp) :: c2 !< |
---|
6028 | REAL(wp) :: d1 !< |
---|
6029 | REAL(wp) :: d2 !< |
---|
6030 | REAL(wp) :: eta !< |
---|
6031 | REAL(wp) :: e1 !< |
---|
6032 | REAL(wp) :: e2 !< |
---|
6033 | REAL(wp) :: fao_gr !< |
---|
6034 | REAL(wp) :: fr !< |
---|
6035 | REAL(wp) :: grfin !< |
---|
6036 | REAL(wp) :: lambda !< |
---|
6037 | REAL(wp) :: lambda_re !< |
---|
6038 | REAL(wp) :: lf !< |
---|
6039 | REAL(wp) :: rc !< |
---|
6040 | REAL(wp) :: rrp !< |
---|
6041 | REAL(wp) :: sst !< |
---|
6042 | REAL(wp) :: tauk !< |
---|
6043 | REAL(wp) :: tl !< |
---|
6044 | REAL(wp) :: t2 !< |
---|
6045 | REAL(wp) :: tt !< |
---|
6046 | REAL(wp) :: t1 !< |
---|
6047 | REAL(wp) :: vk !< |
---|
6048 | REAL(wp) :: vrms1xy !< |
---|
6049 | REAL(wp) :: vrms2xy !< |
---|
6050 | REAL(wp) :: v1 !< |
---|
6051 | REAL(wp) :: v1v2xy !< |
---|
6052 | REAL(wp) :: v1xysq !< |
---|
6053 | REAL(wp) :: v2 !< |
---|
6054 | REAL(wp) :: v2xysq !< |
---|
6055 | REAL(wp) :: wrfin !< |
---|
6056 | REAL(wp) :: wrgrav2 !< |
---|
6057 | REAL(wp) :: wrtur2xy !< |
---|
6058 | REAL(wp) :: xx !< |
---|
6059 | REAL(wp) :: yy !< |
---|
6060 | REAL(wp) :: z !< |
---|
6061 | |
---|
6062 | REAL(wp), DIMENSION(1:radius_classes) :: st !< Stokes number |
---|
6063 | REAL(wp), DIMENSION(1:radius_classes) :: tau !< inertial time scale |
---|
6064 | |
---|
6065 | lambda = urms * SQRT( 15.0_wp * molecular_viscosity / epsilon_collision ) |
---|
6066 | lambda_re = urms**2 * SQRT( 15.0_wp / epsilon_collision / molecular_viscosity ) |
---|
6067 | tl = urms**2 / epsilon_collision |
---|
6068 | lf = 0.5_wp * urms**3 / epsilon_collision |
---|
6069 | tauk = SQRT( molecular_viscosity / epsilon_collision ) |
---|
6070 | eta = ( molecular_viscosity**3 / epsilon_collision )**0.25_wp |
---|
6071 | vk = eta / tauk |
---|
6072 | |
---|
6073 | ao = ( 11.0_wp + 7.0_wp * lambda_re ) / ( 205.0_wp + lambda_re ) |
---|
6074 | tt = SQRT( 2.0_wp * lambda_re / ( SQRT( 15.0_wp ) * ao ) ) * tauk |
---|
6075 | |
---|
6076 | ! |
---|
6077 | !-- Get terminal velocity of droplets |
---|
6078 | CALL fallg |
---|
6079 | |
---|
6080 | DO i = 1, radius_classes |
---|
6081 | tau(i) = winf(i) / g ! inertial time scale |
---|
6082 | st(i) = tau(i) / tauk ! Stokes number |
---|
6083 | ENDDO |
---|
6084 | |
---|
6085 | ! |
---|
6086 | !-- Calculate average radial relative velocity at contact (wrfin) |
---|
6087 | z = tt / tl |
---|
6088 | be = SQRT( 2.0_wp ) * lambda / lf |
---|
6089 | bbb = SQRT( 1.0_wp - 2.0_wp * be**2 ) |
---|
6090 | d1 = ( 1.0_wp + bbb ) / ( 2.0_wp * bbb ) |
---|
6091 | e1 = lf * ( 1.0_wp + bbb ) * 0.5_wp |
---|
6092 | d2 = ( 1.0_wp - bbb ) * 0.5_wp / bbb |
---|
6093 | e2 = lf * ( 1.0_wp - bbb ) * 0.5_wp |
---|
6094 | ccc = SQRT( 1.0_wp - 2.0_wp * z**2 ) |
---|
6095 | b1 = ( 1.0_wp + ccc ) * 0.5_wp / ccc |
---|
6096 | c1 = tl * ( 1.0_wp + ccc ) * 0.5_wp |
---|
6097 | b2 = ( 1.0_wp - ccc ) * 0.5_wp / ccc |
---|
6098 | c2 = tl * ( 1.0_wp - ccc ) * 0.5_wp |
---|
6099 | |
---|
6100 | DO i = 1, radius_classes |
---|
6101 | |
---|
6102 | v1 = winf(i) |
---|
6103 | t1 = tau(i) |
---|
6104 | |
---|
6105 | DO j = 1, i |
---|
6106 | rrp = radclass(i) + radclass(j) |
---|
6107 | v2 = winf(j) |
---|
6108 | t2 = tau(j) |
---|
6109 | |
---|
6110 | v1xysq = b1 * d1 * phi_w(c1,e1,v1,t1) - b1 * d2 * phi_w(c1,e2,v1,t1) & |
---|
6111 | - b2 * d1 * phi_w(c2,e1,v1,t1) + b2 * d2 * phi_w(c2,e2,v1,t1) |
---|
6112 | v1xysq = v1xysq * urms**2 / t1 |
---|
6113 | vrms1xy = SQRT( v1xysq ) |
---|
6114 | |
---|
6115 | v2xysq = b1 * d1 * phi_w(c1,e1,v2,t2) - b1 * d2 * phi_w(c1,e2,v2,t2) & |
---|
6116 | - b2 * d1 * phi_w(c2,e1,v2,t2) + b2 * d2 * phi_w(c2,e2,v2,t2) |
---|
6117 | v2xysq = v2xysq * urms**2 / t2 |
---|
6118 | vrms2xy = SQRT( v2xysq ) |
---|
6119 | |
---|
6120 | IF ( winf(i) >= winf(j) ) THEN |
---|
6121 | v1 = winf(i) |
---|
6122 | t1 = tau(i) |
---|
6123 | v2 = winf(j) |
---|
6124 | t2 = tau(j) |
---|
6125 | ELSE |
---|
6126 | v1 = winf(j) |
---|
6127 | t1 = tau(j) |
---|
6128 | v2 = winf(i) |
---|
6129 | t2 = tau(i) |
---|
6130 | ENDIF |
---|
6131 | |
---|
6132 | v1v2xy = b1 * d1 * zhi(c1,e1,v1,t1,v2,t2) - & |
---|
6133 | b1 * d2 * zhi(c1,e2,v1,t1,v2,t2) - & |
---|
6134 | b2 * d1 * zhi(c2,e1,v1,t1,v2,t2) + & |
---|
6135 | b2 * d2* zhi(c2,e2,v1,t1,v2,t2) |
---|
6136 | fr = d1 * EXP( -rrp / e1 ) - d2 * EXP( -rrp / e2 ) |
---|
6137 | v1v2xy = v1v2xy * fr * urms**2 / tau(i) / tau(j) |
---|
6138 | wrtur2xy = vrms1xy**2 + vrms2xy**2 - 2.0_wp * v1v2xy |
---|
6139 | IF ( wrtur2xy < 0.0_wp ) wrtur2xy = 0.0_wp |
---|
6140 | wrgrav2 = pi / 8.0_wp * ( winf(j) - winf(i) )**2 |
---|
6141 | wrfin = SQRT( ( 2.0_wp / pi ) * ( wrtur2xy + wrgrav2) ) |
---|
6142 | |
---|
6143 | ! |
---|
6144 | !-- Calculate radial distribution function (grfin) |
---|
6145 | IF ( st(j) > st(i) ) THEN |
---|
6146 | sst = st(j) |
---|
6147 | ELSE |
---|
6148 | sst = st(i) |
---|
6149 | ENDIF |
---|
6150 | |
---|
6151 | xx = -0.1988_wp * sst**4 + 1.5275_wp * sst**3 - 4.2942_wp * sst**2 + 5.3406_wp * sst |
---|
6152 | IF ( xx < 0.0_wp ) xx = 0.0_wp |
---|
6153 | yy = 0.1886_wp * EXP( 20.306_wp / lambda_re ) |
---|
6154 | |
---|
6155 | c1_gr = xx / ( g / vk * tauk )**yy |
---|
6156 | |
---|
6157 | ao_gr = ao + ( pi / 8.0_wp) * ( g / vk * tauk )**2 |
---|
6158 | fao_gr = 20.115_wp * SQRT( ao_gr / lambda_re ) |
---|
6159 | rc = SQRT( fao_gr * ABS( st(j) - st(i) ) ) * eta |
---|
6160 | |
---|
6161 | grfin = ( ( eta**2 + rc**2 ) / ( rrp**2 + rc**2) )**( c1_gr*0.5_wp ) |
---|
6162 | IF ( grfin < 1.0_wp ) grfin = 1.0_wp |
---|
6163 | |
---|
6164 | ! |
---|
6165 | !-- Calculate general collection kernel (without the consideration of collection efficiencies) |
---|
6166 | gck(i,j) = 2.0_wp * pi * rrp**2 * wrfin * grfin |
---|
6167 | gck(j,i) = gck(i,j) |
---|
6168 | |
---|
6169 | ENDDO |
---|
6170 | ENDDO |
---|
6171 | |
---|
6172 | END SUBROUTINE turbsd |
---|
6173 | |
---|
6174 | REAL(wp) FUNCTION phi_w( a, b, vsett, tau0 ) |
---|
6175 | ! |
---|
6176 | !-- Function used in the Ayala et al. (2008) analytical model for turbulent effects on the |
---|
6177 | !-- collision kernel |
---|
6178 | |
---|
6179 | |
---|
6180 | REAL(wp) :: a !< |
---|
6181 | REAL(wp) :: aa1 !< |
---|
6182 | REAL(wp) :: b !< |
---|
6183 | REAL(wp) :: tau0 !< |
---|
6184 | REAL(wp) :: vsett !< |
---|
6185 | |
---|
6186 | aa1 = 1.0_wp / tau0 + 1.0_wp / a + vsett / b |
---|
6187 | phi_w = 1.0_wp / aa1 - 0.5_wp * vsett / b / aa1**2 |
---|
6188 | |
---|
6189 | END FUNCTION phi_w |
---|
6190 | |
---|
6191 | REAL(wp) FUNCTION zhi( a, b, vsett1, tau1, vsett2, tau2 ) |
---|
6192 | ! |
---|
6193 | !-- Function used in the Ayala et al. (2008) analytical model for turbulent effects on the collision |
---|
6194 | !-- kernel |
---|
6195 | |
---|
6196 | REAL(wp) :: a !< |
---|
6197 | REAL(wp) :: aa1 !< |
---|
6198 | REAL(wp) :: aa2 !< |
---|
6199 | REAL(wp) :: aa3 !< |
---|
6200 | REAL(wp) :: aa4 !< |
---|
6201 | REAL(wp) :: aa5 !< |
---|
6202 | REAL(wp) :: aa6 !< |
---|
6203 | REAL(wp) :: b !< |
---|
6204 | REAL(wp) :: tau1 !< |
---|
6205 | REAL(wp) :: tau2 !< |
---|
6206 | REAL(wp) :: vsett1 !< |
---|
6207 | REAL(wp) :: vsett2 !< |
---|
6208 | |
---|
6209 | aa1 = vsett2 / b - 1.0_wp / tau2 - 1.0_wp / a |
---|
6210 | aa2 = vsett1 / b + 1.0_wp / tau1 + 1.0_wp / a |
---|
6211 | aa3 = ( vsett1 - vsett2 ) / b + 1.0_wp / tau1 + 1.0_wp / tau2 |
---|
6212 | aa4 = ( vsett2 / b )**2 - ( 1.0_wp / tau2 + 1.0_wp / a )**2 |
---|
6213 | aa5 = vsett2 / b + 1.0_wp / tau2 + 1.0_wp / a |
---|
6214 | aa6 = 1.0_wp / tau1 - 1.0_wp / a + ( 1.0_wp / tau2 + 1.0_wp / a) * vsett1 / vsett2 |
---|
6215 | zhi = ( 1.0_wp / aa1 - 1.0_wp / aa2 ) * ( vsett1 - vsett2 ) * 0.5_wp / & |
---|
6216 | b / aa3**2 + ( 4.0_wp / aa4 - 1.0_wp / aa5**2 - 1.0_wp / aa1**2 ) & |
---|
6217 | * vsett2 * 0.5_wp / b /aa6 + ( 2.0_wp * ( b / aa2 - b / aa1 ) - & |
---|
6218 | vsett1 / aa2**2 + vsett2 / aa1**2 ) * 0.5_wp / b / aa3 |
---|
6219 | |
---|
6220 | END FUNCTION zhi |
---|
6221 | |
---|
6222 | |
---|
6223 | !--------------------------------------------------------------------------------------------------! |
---|
6224 | ! Description: |
---|
6225 | ! ------------ |
---|
6226 | !> Parameterization of terminal velocity following Rogers et al. (1993, J. Appl.Meteorol.) |
---|
6227 | !--------------------------------------------------------------------------------------------------! |
---|
6228 | SUBROUTINE fallg |
---|
6229 | |
---|
6230 | INTEGER(iwp) :: j !< |
---|
6231 | |
---|
6232 | REAL(wp), PARAMETER :: a_rog = 9.65_wp !< parameter |
---|
6233 | REAL(wp), PARAMETER :: b_rog = 10.43_wp !< parameter |
---|
6234 | REAL(wp), PARAMETER :: c_rog = 0.6_wp !< parameter |
---|
6235 | REAL(wp), PARAMETER :: d0_rog = 0.745_wp !< seperation diameter |
---|
6236 | REAL(wp), PARAMETER :: k_cap_rog = 4.0_wp !< parameter |
---|
6237 | REAL(wp), PARAMETER :: k_low_rog = 12.0_wp !< parameter |
---|
6238 | |
---|
6239 | REAL(wp) :: diameter !< droplet diameter in mm |
---|
6240 | |
---|
6241 | |
---|
6242 | DO j = 1, radius_classes |
---|
6243 | |
---|
6244 | diameter = radclass(j) * 2000.0_wp |
---|
6245 | |
---|
6246 | IF ( diameter <= d0_rog ) THEN |
---|
6247 | winf(j) = k_cap_rog * diameter * ( 1.0_wp - EXP( -k_low_rog * diameter ) ) |
---|
6248 | ELSE |
---|
6249 | winf(j) = a_rog - b_rog * EXP( -c_rog * diameter ) |
---|
6250 | ENDIF |
---|
6251 | |
---|
6252 | ENDDO |
---|
6253 | |
---|
6254 | END SUBROUTINE fallg |
---|
6255 | |
---|
6256 | |
---|
6257 | !--------------------------------------------------------------------------------------------------! |
---|
6258 | ! Description: |
---|
6259 | ! ------------ |
---|
6260 | !> Interpolation of collision efficiencies (Hall, 1980, J. Atmos. Sci.) |
---|
6261 | !--------------------------------------------------------------------------------------------------! |
---|
6262 | SUBROUTINE effic |
---|
6263 | |
---|
6264 | INTEGER(iwp) :: i !< |
---|
6265 | INTEGER(iwp) :: iq !< |
---|
6266 | INTEGER(iwp) :: ir !< |
---|
6267 | INTEGER(iwp) :: j !< |
---|
6268 | INTEGER(iwp) :: k !< |
---|
6269 | |
---|
6270 | INTEGER(iwp), DIMENSION(:), ALLOCATABLE :: ira !< |
---|
6271 | |
---|
6272 | LOGICAL, SAVE :: first = .TRUE. !< |
---|
6273 | |
---|
6274 | REAL(wp) :: ek !< |
---|
6275 | REAL(wp) :: particle_radius !< |
---|
6276 | REAL(wp) :: pp !< |
---|
6277 | REAL(wp) :: qq !< |
---|
6278 | REAL(wp) :: rq !< |
---|
6279 | |
---|
6280 | REAL(wp), DIMENSION(1:21), SAVE :: rat !< |
---|
6281 | |
---|
6282 | REAL(wp), DIMENSION(1:15), SAVE :: r0 !< |
---|
6283 | |
---|
6284 | REAL(wp), DIMENSION(1:15,1:21), SAVE :: ecoll !< |
---|
6285 | |
---|
6286 | ! |
---|
6287 | !-- Initial assignment of constants |
---|
6288 | IF ( first ) THEN |
---|
6289 | |
---|
6290 | first = .FALSE. |
---|
6291 | r0 = (/ 6.0_wp, 8.0_wp, 10.0_wp, 15.0_wp, 20.0_wp, 25.0_wp, & |
---|
6292 | 30.0_wp, 40.0_wp, 50.0_wp, 60.0_wp, 70.0_wp, 100.0_wp, & |
---|
6293 | 150.0_wp, 200.0_wp, 300.0_wp /) |
---|
6294 | |
---|
6295 | rat = (/ 0.00_wp, 0.05_wp, 0.10_wp, 0.15_wp, 0.20_wp, 0.25_wp, & |
---|
6296 | 0.30_wp, 0.35_wp, 0.40_wp, 0.45_wp, 0.50_wp, 0.55_wp, & |
---|
6297 | 0.60_wp, 0.65_wp, 0.70_wp, 0.75_wp, 0.80_wp, 0.85_wp, & |
---|
6298 | 0.90_wp, 0.95_wp, 1.00_wp /) |
---|
6299 | |
---|
6300 | ecoll(:,1) = (/ 0.001_wp, 0.001_wp, 0.001_wp, 0.001_wp, 0.001_wp, & |
---|
6301 | 0.001_wp, 0.001_wp, 0.001_wp, 0.001_wp, 0.001_wp, & |
---|
6302 | 0.001_wp, 0.001_wp, 0.001_wp, 0.001_wp, 0.001_wp /) |
---|
6303 | ecoll(:,2) = (/ 0.003_wp, 0.003_wp, 0.003_wp, 0.004_wp, 0.005_wp, & |
---|
6304 | 0.005_wp, 0.005_wp, 0.010_wp, 0.100_wp, 0.050_wp, & |
---|
6305 | 0.200_wp, 0.500_wp, 0.770_wp, 0.870_wp, 0.970_wp /) |
---|
6306 | ecoll(:,3) = (/ 0.007_wp, 0.007_wp, 0.007_wp, 0.008_wp, 0.009_wp, & |
---|
6307 | 0.010_wp, 0.010_wp, 0.070_wp, 0.400_wp, 0.430_wp, & |
---|
6308 | 0.580_wp, 0.790_wp, 0.930_wp, 0.960_wp, 1.000_wp /) |
---|
6309 | ecoll(:,4) = (/ 0.009_wp, 0.009_wp, 0.009_wp, 0.012_wp, 0.015_wp, & |
---|
6310 | 0.010_wp, 0.020_wp, 0.280_wp, 0.600_wp, 0.640_wp, & |
---|
6311 | 0.750_wp, 0.910_wp, 0.970_wp, 0.980_wp, 1.000_wp /) |
---|
6312 | ecoll(:,5) = (/ 0.014_wp, 0.014_wp, 0.014_wp, 0.015_wp, 0.016_wp, & |
---|
6313 | 0.030_wp, 0.060_wp, 0.500_wp, 0.700_wp, 0.770_wp, & |
---|
6314 | 0.840_wp, 0.950_wp, 0.970_wp, 1.000_wp, 1.000_wp /) |
---|
6315 | ecoll(:,6) = (/ 0.017_wp, 0.017_wp, 0.017_wp, 0.020_wp, 0.022_wp, & |
---|
6316 | 0.060_wp, 0.100_wp, 0.620_wp, 0.780_wp, 0.840_wp, & |
---|
6317 | 0.880_wp, 0.950_wp, 1.000_wp, 1.000_wp, 1.000_wp /) |
---|
6318 | ecoll(:,7) = (/ 0.030_wp, 0.030_wp, 0.024_wp, 0.022_wp, 0.032_wp, & |
---|
6319 | 0.062_wp, 0.200_wp, 0.680_wp, 0.830_wp, 0.870_wp, & |
---|
6320 | 0.900_wp, 0.950_wp, 1.000_wp, 1.000_wp, 1.000_wp /) |
---|
6321 | ecoll(:,8) = (/ 0.025_wp, 0.025_wp, 0.025_wp, 0.036_wp, 0.043_wp, & |
---|
6322 | 0.130_wp, 0.270_wp, 0.740_wp, 0.860_wp, 0.890_wp, & |
---|
6323 | 0.920_wp, 1.000_wp, 1.000_wp, 1.000_wp, 1.000_wp /) |
---|
6324 | ecoll(:,9) = (/ 0.027_wp, 0.027_wp, 0.027_wp, 0.040_wp, 0.052_wp, & |
---|
6325 | 0.200_wp, 0.400_wp, 0.780_wp, 0.880_wp, 0.900_wp, & |
---|
6326 | 0.940_wp, 1.000_wp, 1.000_wp, 1.000_wp, 1.000_wp /) |
---|
6327 | ecoll(:,10) = (/ 0.030_wp, 0.030_wp, 0.030_wp, 0.047_wp, 0.064_wp, & |
---|
6328 | 0.250_wp, 0.500_wp, 0.800_wp, 0.900_wp, 0.910_wp, & |
---|
6329 | 0.950_wp, 1.000_wp, 1.000_wp, 1.000_wp, 1.000_wp /) |
---|
6330 | ecoll(:,11) = (/ 0.040_wp, 0.040_wp, 0.033_wp, 0.037_wp, 0.068_wp, & |
---|
6331 | 0.240_wp, 0.550_wp, 0.800_wp, 0.900_wp, 0.910_wp, & |
---|
6332 | 0.950_wp, 1.000_wp, 1.000_wp, 1.000_wp, 1.000_wp /) |
---|
6333 | ecoll(:,12) = (/ 0.035_wp, 0.035_wp, 0.035_wp, 0.055_wp, 0.079_wp, & |
---|
6334 | 0.290_wp, 0.580_wp, 0.800_wp, 0.900_wp, 0.910_wp, & |
---|
6335 | 0.950_wp, 1.000_wp, 1.000_wp, 1.000_wp, 1.000_wp /) |
---|
6336 | ecoll(:,13) = (/ 0.037_wp, 0.037_wp, 0.037_wp, 0.062_wp, 0.082_wp, & |
---|
6337 | 0.290_wp, 0.590_wp, 0.780_wp, 0.900_wp, 0.910_wp, & |
---|
6338 | 0.950_wp, 1.000_wp, 1.000_wp, 1.000_wp, 1.000_wp /) |
---|
6339 | ecoll(:,14) = (/ 0.037_wp, 0.037_wp, 0.037_wp, 0.060_wp, 0.080_wp, & |
---|
6340 | 0.290_wp, 0.580_wp, 0.770_wp, 0.890_wp, 0.910_wp, & |
---|
6341 | 0.950_wp, 1.000_wp, 1.000_wp, 1.000_wp, 1.000_wp /) |
---|
6342 | ecoll(:,15) = (/ 0.037_wp, 0.037_wp, 0.037_wp, 0.041_wp, 0.075_wp, & |
---|
6343 | 0.250_wp, 0.540_wp, 0.760_wp, 0.880_wp, 0.920_wp, & |
---|
6344 | 0.950_wp, 1.000_wp, 1.000_wp, 1.000_wp, 1.000_wp /) |
---|
6345 | ecoll(:,16) = (/ 0.037_wp, 0.037_wp, 0.037_wp, 0.052_wp, 0.067_wp, & |
---|
6346 | 0.250_wp, 0.510_wp, 0.770_wp, 0.880_wp, 0.930_wp, & |
---|
6347 | 0.970_wp, 1.000_wp, 1.000_wp, 1.000_wp, 1.000_wp /) |
---|
6348 | ecoll(:,17) = (/ 0.037_wp, 0.037_wp, 0.037_wp, 0.047_wp, 0.057_wp, & |
---|
6349 | 0.250_wp, 0.490_wp, 0.770_wp, 0.890_wp, 0.950_wp, & |
---|
6350 | 1.000_wp, 1.000_wp, 1.000_wp, 1.000_wp, 1.000_wp /) |
---|
6351 | ecoll(:,18) = (/ 0.036_wp, 0.036_wp, 0.036_wp, 0.042_wp, 0.048_wp, & |
---|
6352 | 0.230_wp, 0.470_wp, 0.780_wp, 0.920_wp, 1.000_wp, & |
---|
6353 | 1.020_wp, 1.020_wp, 1.020_wp, 1.020_wp, 1.020_wp /) |
---|
6354 | ecoll(:,19) = (/ 0.040_wp, 0.040_wp, 0.035_wp, 0.033_wp, 0.040_wp, & |
---|
6355 | 0.112_wp, 0.450_wp, 0.790_wp, 1.010_wp, 1.030_wp, & |
---|
6356 | 1.040_wp, 1.040_wp, 1.040_wp, 1.040_wp, 1.040_wp /) |
---|
6357 | ecoll(:,20) = (/ 0.033_wp, 0.033_wp, 0.033_wp, 0.033_wp, 0.033_wp, & |
---|
6358 | 0.119_wp, 0.470_wp, 0.950_wp, 1.300_wp, 1.700_wp, & |
---|
6359 | 2.300_wp, 2.300_wp, 2.300_wp, 2.300_wp, 2.300_wp /) |
---|
6360 | ecoll(:,21) = (/ 0.027_wp, 0.027_wp, 0.027_wp, 0.027_wp, 0.027_wp, & |
---|
6361 | 0.125_wp, 0.520_wp, 1.400_wp, 2.300_wp, 3.000_wp, & |
---|
6362 | 4.000_wp, 4.000_wp, 4.000_wp, 4.000_wp, 4.000_wp /) |
---|
6363 | ENDIF |
---|
6364 | |
---|
6365 | ! |
---|
6366 | !-- Calculate the radius class index of particles with respect to array r. |
---|
6367 | !-- Radius has to be in microns. |
---|
6368 | ALLOCATE( ira(1:radius_classes) ) |
---|
6369 | DO j = 1, radius_classes |
---|
6370 | particle_radius = radclass(j) * 1.0E6_wp |
---|
6371 | DO k = 1, 15 |
---|
6372 | IF ( particle_radius < r0(k) ) THEN |
---|
6373 | ira(j) = k |
---|
6374 | EXIT |
---|
6375 | ENDIF |
---|
6376 | ENDDO |
---|
6377 | IF ( particle_radius >= r0(15) ) ira(j) = 16 |
---|
6378 | ENDDO |
---|
6379 | |
---|
6380 | ! |
---|
6381 | !-- Two-dimensional linear interpolation of the collision efficiency. |
---|
6382 | !-- Radius has to be in microns. |
---|
6383 | DO j = 1, radius_classes |
---|
6384 | DO i = 1, j |
---|
6385 | |
---|
6386 | ir = MAX( ira(i), ira(j) ) |
---|
6387 | rq = MIN( radclass(i) / radclass(j), radclass(j) / radclass(i) ) |
---|
6388 | iq = INT( rq * 20 ) + 1 |
---|
6389 | iq = MAX( iq , 2) |
---|
6390 | |
---|
6391 | IF ( ir < 16 ) THEN |
---|
6392 | IF ( ir >= 2 ) THEN |
---|
6393 | pp = ( ( MAX( radclass(j), radclass(i) ) * 1.0E6_wp ) - r0(ir-1) ) & |
---|
6394 | / ( r0(ir) - r0(ir-1) ) |
---|
6395 | qq = ( rq - rat(iq-1) ) / ( rat(iq) - rat(iq-1) ) |
---|
6396 | ec(j,i) = ( 1.0_wp - pp ) * ( 1.0_wp - qq ) * ecoll(ir-1,iq-1) & |
---|
6397 | + pp * ( 1.0_wp - qq ) * ecoll(ir,iq-1) & |
---|
6398 | + qq * ( 1.0_wp - pp ) * ecoll(ir-1,iq) & |
---|
6399 | + pp * qq * ecoll(ir,iq) |
---|
6400 | ELSE |
---|
6401 | qq = ( rq - rat(iq-1) ) / ( rat(iq) - rat(iq-1) ) |
---|
6402 | ec(j,i) = ( 1.0_wp - qq ) * ecoll(1,iq-1) + qq * ecoll(1,iq) |
---|
6403 | ENDIF |
---|
6404 | ELSE |
---|
6405 | qq = ( rq - rat(iq-1) ) / ( rat(iq) - rat(iq-1) ) |
---|
6406 | ek = ( 1.0_wp - qq ) * ecoll(15,iq-1) + qq * ecoll(15,iq) |
---|
6407 | ec(j,i) = MIN( ek, 1.0_wp ) |
---|
6408 | ENDIF |
---|
6409 | |
---|
6410 | IF ( ec(j,i) < 1.0E-20_wp ) ec(j,i) = 0.0_wp |
---|
6411 | |
---|
6412 | ec(i,j) = ec(j,i) |
---|
6413 | |
---|
6414 | ENDDO |
---|
6415 | ENDDO |
---|
6416 | |
---|
6417 | DEALLOCATE( ira ) |
---|
6418 | |
---|
6419 | END SUBROUTINE effic |
---|
6420 | |
---|
6421 | |
---|
6422 | !--------------------------------------------------------------------------------------------------! |
---|
6423 | ! Description: |
---|
6424 | ! ------------ |
---|
6425 | !> Interpolation of turbulent enhancement factor for collision efficencies following |
---|
6426 | !> Wang and Grabowski (2009, Atmos. Sci. Let.) |
---|
6427 | !--------------------------------------------------------------------------------------------------! |
---|
6428 | SUBROUTINE turb_enhance_eff |
---|
6429 | |
---|
6430 | INTEGER(iwp) :: i !< |
---|
6431 | INTEGER(iwp) :: iq !< |
---|
6432 | INTEGER(iwp) :: ir !< |
---|
6433 | INTEGER(iwp) :: j !< |
---|
6434 | INTEGER(iwp) :: k !< |
---|
6435 | INTEGER(iwp) :: kk !< |
---|
6436 | |
---|
6437 | INTEGER(iwp), DIMENSION(:), ALLOCATABLE :: ira !< |
---|
6438 | |
---|
6439 | LOGICAL, SAVE :: first = .TRUE. !< |
---|
6440 | |
---|
6441 | REAL(wp) :: particle_radius !< |
---|
6442 | REAL(wp) :: pp !< |
---|
6443 | REAL(wp) :: qq !< |
---|
6444 | REAL(wp) :: rq !< |
---|
6445 | REAL(wp) :: y1 !< |
---|
6446 | REAL(wp) :: y2 !< |
---|
6447 | REAL(wp) :: y3 !< |
---|
6448 | |
---|
6449 | REAL(wp), DIMENSION(1:11), SAVE :: rat !< |
---|
6450 | REAL(wp), DIMENSION(1:7), SAVE :: r0 !< |
---|
6451 | |
---|
6452 | REAL(wp), DIMENSION(1:7,1:11), SAVE :: ecoll_100 !< |
---|
6453 | REAL(wp), DIMENSION(1:7,1:11), SAVE :: ecoll_400 !< |
---|
6454 | |
---|
6455 | ! |
---|
6456 | !-- Initial assignment of constants |
---|
6457 | IF ( first ) THEN |
---|
6458 | |
---|
6459 | first = .FALSE. |
---|
6460 | |
---|
6461 | r0 = (/ 10.0_wp, 20.0_wp, 30.0_wp, 40.0_wp, 50.0_wp, 60.0_wp, 100.0_wp /) |
---|
6462 | |
---|
6463 | rat = (/ 0.0_wp, 0.1_wp, 0.2_wp, 0.3_wp, 0.4_wp, 0.5_wp, 0.6_wp, 0.7_wp, 0.8_wp, 0.9_wp, & |
---|
6464 | 1.0_wp /) |
---|
6465 | ! |
---|
6466 | !-- Tabulated turbulent enhancement factor at 100 cm**2/s**3 |
---|
6467 | ecoll_100(:,1) = (/ 1.74_wp, 1.74_wp, 1.773_wp, 1.49_wp, & |
---|
6468 | 1.207_wp, 1.207_wp, 1.0_wp /) |
---|
6469 | ecoll_100(:,2) = (/ 1.46_wp, 1.46_wp, 1.421_wp, 1.245_wp, & |
---|
6470 | 1.069_wp, 1.069_wp, 1.0_wp /) |
---|
6471 | ecoll_100(:,3) = (/ 1.32_wp, 1.32_wp, 1.245_wp, 1.123_wp, & |
---|
6472 | 1.000_wp, 1.000_wp, 1.0_wp /) |
---|
6473 | ecoll_100(:,4) = (/ 1.250_wp, 1.250_wp, 1.148_wp, 1.087_wp, & |
---|
6474 | 1.025_wp, 1.025_wp, 1.0_wp /) |
---|
6475 | ecoll_100(:,5) = (/ 1.186_wp, 1.186_wp, 1.066_wp, 1.060_wp, & |
---|
6476 | 1.056_wp, 1.056_wp, 1.0_wp /) |
---|
6477 | ecoll_100(:,6) = (/ 1.045_wp, 1.045_wp, 1.000_wp, 1.014_wp, & |
---|
6478 | 1.028_wp, 1.028_wp, 1.0_wp /) |
---|
6479 | ecoll_100(:,7) = (/ 1.070_wp, 1.070_wp, 1.030_wp, 1.038_wp, & |
---|
6480 | 1.046_wp, 1.046_wp, 1.0_wp /) |
---|
6481 | ecoll_100(:,8) = (/ 1.000_wp, 1.000_wp, 1.054_wp, 1.042_wp, & |
---|
6482 | 1.029_wp, 1.029_wp, 1.0_wp /) |
---|
6483 | ecoll_100(:,9) = (/ 1.223_wp, 1.223_wp, 1.117_wp, 1.069_wp, & |
---|
6484 | 1.021_wp, 1.021_wp, 1.0_wp /) |
---|
6485 | ecoll_100(:,10) = (/ 1.570_wp, 1.570_wp, 1.244_wp, 1.166_wp, & |
---|
6486 | 1.088_wp, 1.088_wp, 1.0_wp /) |
---|
6487 | ecoll_100(:,11) = (/ 20.3_wp, 20.3_wp, 14.6_wp, 8.61_wp, & |
---|
6488 | 2.60_wp, 2.60_wp, 1.0_wp /) |
---|
6489 | ! |
---|
6490 | !-- Tabulated turbulent enhancement factor at 400 cm**2/s**3 |
---|
6491 | ecoll_400(:,1) = (/ 4.976_wp, 4.976_wp, 3.593_wp, 2.519_wp, & |
---|
6492 | 1.445_wp, 1.445_wp, 1.0_wp /) |
---|
6493 | ecoll_400(:,2) = (/ 2.984_wp, 2.984_wp, 2.181_wp, 1.691_wp, & |
---|
6494 | 1.201_wp, 1.201_wp, 1.0_wp /) |
---|
6495 | ecoll_400(:,3) = (/ 1.988_wp, 1.988_wp, 1.475_wp, 1.313_wp, & |
---|
6496 | 1.150_wp, 1.150_wp, 1.0_wp /) |
---|
6497 | ecoll_400(:,4) = (/ 1.490_wp, 1.490_wp, 1.187_wp, 1.156_wp, & |
---|
6498 | 1.126_wp, 1.126_wp, 1.0_wp /) |
---|
6499 | ecoll_400(:,5) = (/ 1.249_wp, 1.249_wp, 1.088_wp, 1.090_wp, & |
---|
6500 | 1.092_wp, 1.092_wp, 1.0_wp /) |
---|
6501 | ecoll_400(:,6) = (/ 1.139_wp, 1.139_wp, 1.130_wp, 1.091_wp, & |
---|
6502 | 1.051_wp, 1.051_wp, 1.0_wp /) |
---|
6503 | ecoll_400(:,7) = (/ 1.220_wp, 1.220_wp, 1.190_wp, 1.138_wp, & |
---|
6504 | 1.086_wp, 1.086_wp, 1.0_wp /) |
---|
6505 | ecoll_400(:,8) = (/ 1.325_wp, 1.325_wp, 1.267_wp, 1.165_wp, & |
---|
6506 | 1.063_wp, 1.063_wp, 1.0_wp /) |
---|
6507 | ecoll_400(:,9) = (/ 1.716_wp, 1.716_wp, 1.345_wp, 1.223_wp, & |
---|
6508 | 1.100_wp, 1.100_wp, 1.0_wp /) |
---|
6509 | ecoll_400(:,10) = (/ 3.788_wp, 3.788_wp, 1.501_wp, 1.311_wp, & |
---|
6510 | 1.120_wp, 1.120_wp, 1.0_wp /) |
---|
6511 | ecoll_400(:,11) = (/ 36.52_wp, 36.52_wp, 19.16_wp, 22.80_wp, & |
---|
6512 | 26.0_wp, 26.0_wp, 1.0_wp /) |
---|
6513 | |
---|
6514 | ENDIF |
---|
6515 | |
---|
6516 | ! |
---|
6517 | !-- Calculate the radius class index of particles with respect to array r0. |
---|
6518 | !-- The droplet radius has to be given in microns. |
---|
6519 | ALLOCATE( ira(1:radius_classes) ) |
---|
6520 | |
---|
6521 | DO j = 1, radius_classes |
---|
6522 | particle_radius = radclass(j) * 1.0E6_wp |
---|
6523 | DO k = 1, 7 |
---|
6524 | IF ( particle_radius < r0(k) ) THEN |
---|
6525 | ira(j) = k |
---|
6526 | EXIT |
---|
6527 | ENDIF |
---|
6528 | ENDDO |
---|
6529 | IF ( particle_radius >= r0(7) ) ira(j) = 8 |
---|
6530 | ENDDO |
---|
6531 | |
---|
6532 | ! |
---|
6533 | !-- Two-dimensional linear interpolation of the turbulent enhancement factor. |
---|
6534 | !-- The droplet radius has to be given in microns. |
---|
6535 | DO j = 1, radius_classes |
---|
6536 | DO i = 1, j |
---|
6537 | |
---|
6538 | ir = MAX( ira(i), ira(j) ) |
---|
6539 | rq = MIN( radclass(i) / radclass(j), radclass(j) / radclass(i) ) |
---|
6540 | |
---|
6541 | DO kk = 2, 11 |
---|
6542 | IF ( rq <= rat(kk) ) THEN |
---|
6543 | iq = kk |
---|
6544 | EXIT |
---|
6545 | ENDIF |
---|
6546 | ENDDO |
---|
6547 | |
---|
6548 | y1 = 1.0_wp ! turbulent enhancement factor at 0 m**2/s**3 |
---|
6549 | |
---|
6550 | IF ( ir < 8 ) THEN |
---|
6551 | IF ( ir >= 2 ) THEN |
---|
6552 | pp = ( MAX( radclass(j), radclass(i) ) * 1.0E6_wp - r0(ir-1) ) & |
---|
6553 | / ( r0(ir) - r0(ir-1) ) |
---|
6554 | qq = ( rq - rat(iq-1) ) / ( rat(iq) - rat(iq-1) ) |
---|
6555 | y2 = ( 1.0_wp - pp ) * ( 1.0_wp - qq ) * ecoll_100(ir-1,iq-1) + & |
---|
6556 | pp * ( 1.0_wp - qq ) * ecoll_100(ir,iq-1) + & |
---|
6557 | qq * ( 1.0_wp - pp ) * ecoll_100(ir-1,iq) + & |
---|
6558 | pp * qq * ecoll_100(ir,iq) |
---|
6559 | y3 = ( 1.0-pp ) * ( 1.0_wp - qq ) * ecoll_400(ir-1,iq-1) + & |
---|
6560 | pp * ( 1.0_wp - qq ) * ecoll_400(ir,iq-1) + & |
---|
6561 | qq * ( 1.0_wp - pp ) * ecoll_400(ir-1,iq) + & |
---|
6562 | pp * qq * ecoll_400(ir,iq) |
---|
6563 | ELSE |
---|
6564 | qq = ( rq - rat(iq-1) ) / ( rat(iq) - rat(iq-1) ) |
---|
6565 | y2 = ( 1.0_wp - qq ) * ecoll_100(1,iq-1) + qq * ecoll_100(1,iq) |
---|
6566 | y3 = ( 1.0_wp - qq ) * ecoll_400(1,iq-1) + qq * ecoll_400(1,iq) |
---|
6567 | ENDIF |
---|
6568 | ELSE |
---|
6569 | qq = ( rq - rat(iq-1) ) / ( rat(iq) - rat(iq-1) ) |
---|
6570 | y2 = ( 1.0_wp - qq ) * ecoll_100(7,iq-1) + qq * ecoll_100(7,iq) |
---|
6571 | y3 = ( 1.0_wp - qq ) * ecoll_400(7,iq-1) + qq * ecoll_400(7,iq) |
---|
6572 | ENDIF |
---|
6573 | ! |
---|
6574 | !-- Linear interpolation of turbulent enhancement factor |
---|
6575 | IF ( epsilon_collision <= 0.01_wp ) THEN |
---|
6576 | ecf(j,i) = ( epsilon_collision - 0.01_wp ) / ( 0.0_wp - 0.01_wp ) * y1 & |
---|
6577 | + ( epsilon_collision - 0.0_wp ) / ( 0.01_wp - 0.0_wp ) * y2 |
---|
6578 | ELSEIF ( epsilon_collision <= 0.06_wp ) THEN |
---|
6579 | ecf(j,i) = ( epsilon_collision - 0.04_wp ) / ( 0.01_wp - 0.04_wp ) * y2 & |
---|
6580 | + ( epsilon_collision - 0.01_wp ) / ( 0.04_wp - 0.01_wp ) * y3 |
---|
6581 | ELSE |
---|
6582 | ecf(j,i) = ( 0.06_wp - 0.04_wp ) / ( 0.01_wp - 0.04_wp ) * y2 & |
---|
6583 | + ( 0.06_wp - 0.01_wp ) / ( 0.04_wp - 0.01_wp ) * y3 |
---|
6584 | ENDIF |
---|
6585 | |
---|
6586 | IF ( ecf(j,i) < 1.0_wp ) ecf(j,i) = 1.0_wp |
---|
6587 | |
---|
6588 | ecf(i,j) = ecf(j,i) |
---|
6589 | |
---|
6590 | ENDDO |
---|
6591 | ENDDO |
---|
6592 | |
---|
6593 | END SUBROUTINE turb_enhance_eff |
---|
6594 | |
---|
6595 | |
---|
6596 | !-------------------------------------------------------------------------------------------------! |
---|
6597 | ! Description: |
---|
6598 | ! ------------ |
---|
6599 | ! This routine is a part of the Lagrangian particle model. Super droplets which fulfill certain |
---|
6600 | ! criterion's (e.g. a big weighting factor and a large radius) can be split into several super |
---|
6601 | ! droplets with a reduced number of represented particles of every super droplet. This mechanism |
---|
6602 | ! ensures an improved representation of the right tail of the drop size distribution with a feasible |
---|
6603 | ! amount of computational costs. The limits of particle creation should be chosen carefully! The |
---|
6604 | ! idea of this algorithm is based on Unterstrasser and Soelch, 2014. |
---|
6605 | !--------------------------------------------------------------------------------------------------! |
---|
6606 | SUBROUTINE lpm_splitting |
---|
6607 | |
---|
6608 | INTEGER(iwp), PARAMETER :: n_max = 100 !< number of radii bin for splitting functions |
---|
6609 | |
---|
6610 | INTEGER(iwp) :: i !< |
---|
6611 | INTEGER(iwp) :: j !< |
---|
6612 | INTEGER(iwp) :: jpp !< |
---|
6613 | INTEGER(iwp) :: k !< |
---|
6614 | INTEGER(iwp) :: n !< |
---|
6615 | INTEGER(iwp) :: new_particles_gb !< counter of created particles within one grid box |
---|
6616 | INTEGER(iwp) :: new_size !< new particle array size |
---|
6617 | INTEGER(iwp) :: np !< |
---|
6618 | INTEGER(iwp) :: old_size !< old particle array size |
---|
6619 | |
---|
6620 | LOGICAL :: first_loop_stride_sp = .TRUE. !< flag to calculate constants only once |
---|
6621 | |
---|
6622 | REAL(wp) :: diameter !< diameter of droplet |
---|
6623 | REAL(wp) :: dlog !< factor for DSD calculation |
---|
6624 | REAL(wp) :: factor_volume_to_mass !< pre calculate factor volume to mass |
---|
6625 | REAL(wp) :: lambda !< slope parameter of gamma-distribution |
---|
6626 | REAL(wp) :: lwc !< liquid water content of grid box |
---|
6627 | REAL(wp) :: lwc_total !< average liquid water content of cloud |
---|
6628 | REAL(wp) :: m1 !< first moment of DSD |
---|
6629 | REAL(wp) :: m1_total !< average over all PEs of first moment of DSD |
---|
6630 | REAL(wp) :: m2 !< second moment of DSD |
---|
6631 | REAL(wp) :: m2_total !< average average over all PEs second moment of DSD |
---|
6632 | REAL(wp) :: m3 !< third moment of DSD |
---|
6633 | REAL(wp) :: m3_total !< average average over all PEs third moment of DSD |
---|
6634 | REAL(wp) :: mu !< spectral shape parameter of gamma distribution |
---|
6635 | REAL(wp) :: nrclgb !< number of cloudy grid boxes (ql >= 1.0E-5 kg/kg) |
---|
6636 | REAL(wp) :: nrclgb_total !< average over all PEs of number of cloudy grid boxes |
---|
6637 | REAL(wp) :: nr !< number concentration of cloud droplets |
---|
6638 | REAL(wp) :: nr_total !< average over all PEs of number of cloudy grid boxes |
---|
6639 | REAL(wp) :: nr0 !< intercept parameter of gamma distribution |
---|
6640 | REAL(wp) :: pirho_l !< pi * rho_l / 6.0 |
---|
6641 | REAL(wp) :: ql_crit = 1.0E-5_wp !< threshold lwc for cloudy grid cells (Siebesma et al 2003, JAS, 60) |
---|
6642 | REAL(wp) :: rm !< volume averaged mean radius |
---|
6643 | REAL(wp) :: rm_total !< average over all PEs of volume averaged mean radius |
---|
6644 | REAL(wp) :: r_min = 1.0E-6_wp !< minimum radius of approximated spectra |
---|
6645 | REAL(wp) :: r_max = 1.0E-3_wp !< maximum radius of approximated spectra |
---|
6646 | REAL(wp) :: sigma_log = 1.5_wp !< standard deviation of the LOG-distribution |
---|
6647 | REAL(wp) :: zeta !< Parameter for DSD calculation of Seifert |
---|
6648 | |
---|
6649 | REAL(wp), DIMENSION(0:n_max-1) :: an_spl !< size dependent critical weight factor |
---|
6650 | REAL(wp), DIMENSION(0:n_max-1) :: r_bin_mid !< mass weighted mean radius of a bin |
---|
6651 | REAL(wp), DIMENSION(0:n_max) :: r_bin !< boundaries of a radius bin |
---|
6652 | |
---|
6653 | TYPE(particle_type) :: tmp_particle !< temporary particle TYPE |
---|
6654 | |
---|
6655 | CALL cpu_log( log_point_s(80), 'lpm_splitting', 'start' ) |
---|
6656 | |
---|
6657 | IF ( first_loop_stride_sp ) THEN |
---|
6658 | IF ( i_splitting_mode == 2 .OR. i_splitting_mode == 3 ) THEN |
---|
6659 | dlog = ( LOG10(r_max) - LOG10(r_min) ) / ( n_max - 1 ) |
---|
6660 | DO i = 0, n_max-1 |
---|
6661 | r_bin(i) = 10.0_wp**( LOG10(r_min) + i * dlog - 0.5_wp * dlog ) |
---|
6662 | r_bin_mid(i) = 10.0_wp**( LOG10(r_min) + i * dlog ) |
---|
6663 | ENDDO |
---|
6664 | r_bin(n_max) = 10.0_wp**( LOG10(r_min) + n_max * dlog - 0.5_wp * dlog ) |
---|
6665 | ENDIF |
---|
6666 | factor_volume_to_mass = 4.0_wp / 3.0_wp * pi * rho_l |
---|
6667 | pirho_l = pi * rho_l / 6.0_wp |
---|
6668 | IF ( weight_factor_split == -1.0_wp ) THEN |
---|
6669 | weight_factor_split = 0.1_wp * initial_weighting_factor |
---|
6670 | ENDIF |
---|
6671 | ENDIF |
---|
6672 | |
---|
6673 | |
---|
6674 | IF ( i_splitting_mode == 1 ) THEN |
---|
6675 | |
---|
6676 | DO i = nxl, nxr |
---|
6677 | DO j = nys, nyn |
---|
6678 | DO k = nzb+1, nzt |
---|
6679 | |
---|
6680 | new_particles_gb = 0 |
---|
6681 | number_of_particles = prt_count(k,j,i) |
---|
6682 | IF ( number_of_particles <= 0 .OR. ql(k,j,i) < ql_crit ) CYCLE |
---|
6683 | particles => grid_particles(k,j,i)%particles(1:number_of_particles) |
---|
6684 | ! |
---|
6685 | !-- Start splitting operations. Each particle is checked if it fulfilled the splitting |
---|
6686 | !-- criterion's. In splitting mode 'const' a critical radius (radius_split) a critical |
---|
6687 | !-- weighting factor (weight_factor_split) and a splitting factor (splitting_factor) |
---|
6688 | !-- must be prescribed (see particle_parameters). Super droplets which have a larger |
---|
6689 | !-- radius and larger weighting factor are split into 'splitting_factor' super droplets. |
---|
6690 | !-- Therefore, the weighting factor of the super droplet and all created clones is |
---|
6691 | !-- reduced by the factor of 'splitting_factor'. |
---|
6692 | DO n = 1, number_of_particles |
---|
6693 | IF ( particles(n)%particle_mask .AND. & |
---|
6694 | particles(n)%radius >= radius_split .AND. & |
---|
6695 | particles(n)%weight_factor >= weight_factor_split ) & |
---|
6696 | THEN |
---|
6697 | ! |
---|
6698 | !-- Calculate the new number of particles. |
---|
6699 | new_size = prt_count(k,j,i) + splitting_factor - 1 |
---|
6700 | ! |
---|
6701 | !-- Cycle if maximum number of particles per grid box is greater than the allowed |
---|
6702 | !-- maximum number. |
---|
6703 | IF ( new_size >= max_number_particles_per_gridbox ) CYCLE |
---|
6704 | ! |
---|
6705 | !-- Reallocate particle array if necessary. |
---|
6706 | IF ( new_size > SIZE( particles ) ) THEN |
---|
6707 | CALL realloc_particles_array( i, j, k, new_size ) |
---|
6708 | ENDIF |
---|
6709 | old_size = prt_count(k,j,i) |
---|
6710 | ! |
---|
6711 | !-- Calculate new weighting factor. |
---|
6712 | particles(n)%weight_factor = particles(n)%weight_factor / splitting_factor |
---|
6713 | tmp_particle = particles(n) |
---|
6714 | ! |
---|
6715 | !-- Create splitting_factor-1 new particles. |
---|
6716 | DO jpp = 1, splitting_factor-1 |
---|
6717 | grid_particles(k,j,i)%particles(jpp+old_size) = tmp_particle |
---|
6718 | ENDDO |
---|
6719 | new_particles_gb = new_particles_gb + splitting_factor - 1 |
---|
6720 | ! |
---|
6721 | !-- Save the new number of super droplets for every grid box. |
---|
6722 | prt_count(k,j,i) = prt_count(k,j,i) + splitting_factor - 1 |
---|
6723 | ENDIF |
---|
6724 | ENDDO |
---|
6725 | |
---|
6726 | ENDDO |
---|
6727 | ENDDO |
---|
6728 | ENDDO |
---|
6729 | |
---|
6730 | ELSEIF ( i_splitting_mode == 2 ) THEN |
---|
6731 | ! |
---|
6732 | !-- Initialize summing variables. |
---|
6733 | lwc = 0.0_wp |
---|
6734 | lwc_total = 0.0_wp |
---|
6735 | m1 = 0.0_wp |
---|
6736 | m1_total = 0.0_wp |
---|
6737 | m2 = 0.0_wp |
---|
6738 | m2_total = 0.0_wp |
---|
6739 | m3 = 0.0_wp |
---|
6740 | m3_total = 0.0_wp |
---|
6741 | nr = 0.0_wp |
---|
6742 | nrclgb = 0.0_wp |
---|
6743 | nrclgb_total = 0.0_wp |
---|
6744 | nr_total = 0.0_wp |
---|
6745 | rm = 0.0_wp |
---|
6746 | rm_total = 0.0_wp |
---|
6747 | |
---|
6748 | DO i = nxl, nxr |
---|
6749 | DO j = nys, nyn |
---|
6750 | DO k = nzb+1, nzt |
---|
6751 | number_of_particles = prt_count(k,j,i) |
---|
6752 | IF ( number_of_particles <= 0 .OR. ql(k,j,i) < ql_crit ) CYCLE |
---|
6753 | particles => grid_particles(k,j,i)%particles(1:number_of_particles) |
---|
6754 | nrclgb = nrclgb + 1.0_wp |
---|
6755 | ! |
---|
6756 | !-- Calculate moments of DSD. |
---|
6757 | DO n = 1, number_of_particles |
---|
6758 | IF ( particles(n)%particle_mask .AND. particles(n)%radius >= r_min ) & |
---|
6759 | THEN |
---|
6760 | nr = nr + particles(n)%weight_factor |
---|
6761 | rm = rm + factor_volume_to_mass * & |
---|
6762 | particles(n)%radius**3 * & |
---|
6763 | particles(n)%weight_factor |
---|
6764 | IF ( isf == 1 ) THEN |
---|
6765 | diameter = particles(n)%radius * 2.0_wp |
---|
6766 | lwc = lwc + factor_volume_to_mass * & |
---|
6767 | particles(n)%radius**3 * & |
---|
6768 | particles(n)%weight_factor |
---|
6769 | m1 = m1 + particles(n)%weight_factor * diameter |
---|
6770 | m2 = m2 + particles(n)%weight_factor * diameter**2 |
---|
6771 | m3 = m3 + particles(n)%weight_factor * diameter**3 |
---|
6772 | ENDIF |
---|
6773 | ENDIF |
---|
6774 | ENDDO |
---|
6775 | ENDDO |
---|
6776 | ENDDO |
---|
6777 | ENDDO |
---|
6778 | |
---|
6779 | #if defined( __parallel ) |
---|
6780 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
6781 | CALL MPI_ALLREDUCE( nr, nr_total, 1 , MPI_REAL, MPI_SUM, comm2d, ierr ) |
---|
6782 | CALL MPI_ALLREDUCE( rm, rm_total, 1 , MPI_REAL, MPI_SUM, comm2d, ierr ) |
---|
6783 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
6784 | CALL MPI_ALLREDUCE( nrclgb, nrclgb_total, 1 , MPI_REAL, MPI_SUM, comm2d, ierr ) |
---|
6785 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
6786 | CALL MPI_ALLREDUCE( lwc, lwc_total, 1 , MPI_REAL, MPI_SUM, comm2d, ierr ) |
---|
6787 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
6788 | CALL MPI_ALLREDUCE( m1, m1_total, 1 , MPI_REAL, MPI_SUM, comm2d, ierr ) |
---|
6789 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
6790 | CALL MPI_ALLREDUCE( m2, m2_total, 1 , MPI_REAL, MPI_SUM, comm2d, ierr ) |
---|
6791 | IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr ) |
---|
6792 | CALL MPI_ALLREDUCE( m3, m3_total, 1 , MPI_REAL, MPI_SUM, comm2d, ierr ) |
---|
6793 | #endif |
---|
6794 | |
---|
6795 | ! |
---|
6796 | !-- Calculate number concentration and mean volume averaged radius. |
---|
6797 | nr_total = MERGE( nr_total / nrclgb_total, 0.0_wp, nrclgb_total > 0.0_wp ) |
---|
6798 | rm_total = MERGE( ( rm_total / ( nr_total * factor_volume_to_mass ) )**0.3333333_wp, 0.0_wp,& |
---|
6799 | nrclgb_total > 0.0_wp & |
---|
6800 | ) |
---|
6801 | ! |
---|
6802 | !-- Check which function should be used to approximate the DSD. |
---|
6803 | IF ( isf == 1 ) THEN |
---|
6804 | lwc_total = MERGE( lwc_total / nrclgb_total, 0.0_wp, nrclgb_total > 0.0_wp ) |
---|
6805 | m1_total = MERGE( m1_total / nrclgb_total, 0.0_wp, nrclgb_total > 0.0_wp ) |
---|
6806 | m2_total = MERGE( m2_total / nrclgb_total, 0.0_wp, nrclgb_total > 0.0_wp ) |
---|
6807 | m3_total = MERGE( m3_total / nrclgb_total, 0.0_wp, nrclgb_total > 0.0_wp ) |
---|
6808 | zeta = m1_total * m3_total / m2_total**2 |
---|
6809 | mu = MAX( ( ( 1.0_wp - zeta ) * 2.0_wp + 1.0_wp ) / ( zeta - 1.0_wp ), 0.0_wp ) |
---|
6810 | |
---|
6811 | lambda = ( pirho_l * nr_total / lwc_total * & |
---|
6812 | ( mu + 3.0_wp ) * ( mu + 2.0_wp ) * ( mu + 1.0_wp ) & |
---|
6813 | )**0.3333333_wp |
---|
6814 | nr0 = nr_total / gamma( mu + 1.0_wp ) * lambda**( mu + 1.0_wp ) |
---|
6815 | |
---|
6816 | DO n = 0, n_max-1 |
---|
6817 | diameter = r_bin_mid(n) * 2.0_wp |
---|
6818 | an_spl(n) = nr0 * diameter**mu * EXP( -lambda * diameter ) * & |
---|
6819 | ( r_bin(n+1) - r_bin(n) ) * 2.0_wp |
---|
6820 | ENDDO |
---|
6821 | ELSEIF ( isf == 2 ) THEN |
---|
6822 | DO n = 0, n_max-1 |
---|
6823 | an_spl(n) = nr_total / ( SQRT( 2.0_wp * pi ) * LOG(sigma_log) * r_bin_mid(n) ) * & |
---|
6824 | EXP( -( LOG( r_bin_mid(n) / rm_total )**2 ) / & |
---|
6825 | ( 2.0_wp * LOG(sigma_log)**2 ) & |
---|
6826 | ) * & |
---|
6827 | ( r_bin(n+1) - r_bin(n) ) |
---|
6828 | ENDDO |
---|
6829 | ELSEIF( isf == 3 ) THEN |
---|
6830 | DO n = 0, n_max-1 |
---|
6831 | an_spl(n) = 3.0_wp * nr_total * r_bin_mid(n)**2 / rm_total**3 * & |
---|
6832 | EXP( -( r_bin_mid(n)**3 / rm_total**3 ) ) * & |
---|
6833 | ( r_bin(n+1) - r_bin(n) ) |
---|
6834 | ENDDO |
---|
6835 | ENDIF |
---|
6836 | ! |
---|
6837 | !-- Criterion to avoid super droplets with a weighting factor < 1.0. |
---|
6838 | an_spl = MAX(an_spl, 1.0_wp) |
---|
6839 | |
---|
6840 | DO i = nxl, nxr |
---|
6841 | DO j = nys, nyn |
---|
6842 | DO k = nzb+1, nzt |
---|
6843 | number_of_particles = prt_count(k,j,i) |
---|
6844 | IF ( number_of_particles <= 0 .OR. ql(k,j,i) < ql_crit ) CYCLE |
---|
6845 | particles => grid_particles(k,j,i)%particles(1:number_of_particles) |
---|
6846 | new_particles_gb = 0 |
---|
6847 | ! |
---|
6848 | !-- Start splitting operations. Each particle is checked if it fulfilled the splitting |
---|
6849 | !-- criterion's. In splitting mode 'cl_av' a critical radius (radius_split) and a |
---|
6850 | !-- splitting function must be prescribed (see particles_par). The critical weighting |
---|
6851 | !-- factor is calculated while approximating a 'gamma', 'log' or 'exp'- drop size |
---|
6852 | !-- distribution. In this mode the DSD is calculated as an average over all cloudy grid |
---|
6853 | !-- boxes. Super droplets which have a larger radius and larger weighting factor are |
---|
6854 | !-- split into 'splitting_factor' super droplets. In this case the splitting factor is |
---|
6855 | !-- calculated of weighting factor of the super droplet and the approximated number |
---|
6856 | !-- concentration for droplet of such a size. Due to the splitting, the weighting factor |
---|
6857 | !-- of the super droplet and all created clones is reduced by the factor of |
---|
6858 | !-- 'splitting_facor'. |
---|
6859 | DO n = 1, number_of_particles |
---|
6860 | DO np = 0, n_max-1 |
---|
6861 | IF ( r_bin(np) >= radius_split .AND. & |
---|
6862 | particles(n)%particle_mask .AND. & |
---|
6863 | particles(n)%radius >= r_bin(np) .AND. & |
---|
6864 | particles(n)%radius < r_bin(np+1) .AND. & |
---|
6865 | particles(n)%weight_factor >= an_spl(np) ) & |
---|
6866 | THEN |
---|
6867 | ! |
---|
6868 | !-- Calculate splitting factor |
---|
6869 | splitting_factor = MIN( INT( particles(n)%weight_factor / & |
---|
6870 | an_spl(np) & |
---|
6871 | ), splitting_factor_max & |
---|
6872 | ) |
---|
6873 | IF ( splitting_factor < 2 ) CYCLE |
---|
6874 | ! |
---|
6875 | !-- Calculate the new number of particles. |
---|
6876 | new_size = prt_count(k,j,i) + splitting_factor - 1 |
---|
6877 | ! |
---|
6878 | !-- Cycle if maximum number of particles per grid box is greater than the |
---|
6879 | !-- allowed maximum number. |
---|
6880 | IF ( new_size >= max_number_particles_per_gridbox ) CYCLE |
---|
6881 | ! |
---|
6882 | !-- Reallocate particle array if necessary. |
---|
6883 | IF ( new_size > SIZE( particles ) ) THEN |
---|
6884 | CALL realloc_particles_array( i, j, k, new_size ) |
---|
6885 | ENDIF |
---|
6886 | old_size = prt_count(k,j,i) |
---|
6887 | new_particles_gb = new_particles_gb + splitting_factor - 1 |
---|
6888 | ! |
---|
6889 | !-- Calculate new weighting factor. |
---|
6890 | particles(n)%weight_factor = particles(n)%weight_factor / splitting_factor |
---|
6891 | tmp_particle = particles(n) |
---|
6892 | ! |
---|
6893 | !-- Create splitting_factor-1 new particles. |
---|
6894 | DO jpp = 1, splitting_factor-1 |
---|
6895 | grid_particles(k,j,i)%particles(jpp+old_size) = tmp_particle |
---|
6896 | ENDDO |
---|
6897 | ! |
---|
6898 | !-- Save the new number of super droplets. |
---|
6899 | prt_count(k,j,i) = prt_count(k,j,i) + splitting_factor - 1 |
---|
6900 | ENDIF |
---|
6901 | ENDDO |
---|
6902 | ENDDO |
---|
6903 | |
---|
6904 | ENDDO |
---|
6905 | ENDDO |
---|
6906 | ENDDO |
---|
6907 | |
---|
6908 | ELSEIF ( i_splitting_mode == 3 ) THEN |
---|
6909 | |
---|
6910 | DO i = nxl, nxr |
---|
6911 | DO j = nys, nyn |
---|
6912 | DO k = nzb+1, nzt |
---|
6913 | |
---|
6914 | ! |
---|
6915 | !-- Initialize summing variables. |
---|
6916 | lwc = 0.0_wp |
---|
6917 | m1 = 0.0_wp |
---|
6918 | m2 = 0.0_wp |
---|
6919 | m3 = 0.0_wp |
---|
6920 | nr = 0.0_wp |
---|
6921 | rm = 0.0_wp |
---|
6922 | |
---|
6923 | new_particles_gb = 0 |
---|
6924 | number_of_particles = prt_count(k,j,i) |
---|
6925 | IF ( number_of_particles <= 0 .OR. ql(k,j,i) < ql_crit ) CYCLE |
---|
6926 | particles => grid_particles(k,j,i)%particles |
---|
6927 | ! |
---|
6928 | !-- Calculate moments of DSD. |
---|
6929 | DO n = 1, number_of_particles |
---|
6930 | IF ( particles(n)%particle_mask .AND. particles(n)%radius >= r_min ) & |
---|
6931 | THEN |
---|
6932 | nr = nr + particles(n)%weight_factor |
---|
6933 | rm = rm + factor_volume_to_mass * & |
---|
6934 | particles(n)%radius**3 * & |
---|
6935 | particles(n)%weight_factor |
---|
6936 | IF ( isf == 1 ) THEN |
---|
6937 | diameter = particles(n)%radius * 2.0_wp |
---|
6938 | lwc = lwc + factor_volume_to_mass * & |
---|
6939 | particles(n)%radius**3 * & |
---|
6940 | particles(n)%weight_factor |
---|
6941 | m1 = m1 + particles(n)%weight_factor * diameter |
---|
6942 | m2 = m2 + particles(n)%weight_factor * diameter**2 |
---|
6943 | m3 = m3 + particles(n)%weight_factor * diameter**3 |
---|
6944 | ENDIF |
---|
6945 | ENDIF |
---|
6946 | ENDDO |
---|
6947 | |
---|
6948 | IF ( nr <= 0.0_wp .OR. rm <= 0.0_wp ) CYCLE |
---|
6949 | ! |
---|
6950 | !-- Calculate mean volume averaged radius. |
---|
6951 | rm = ( rm / ( nr * factor_volume_to_mass ) )**0.3333333_wp |
---|
6952 | ! |
---|
6953 | !-- Check which function should be used to approximate the DSD. |
---|
6954 | IF ( isf == 1 ) THEN |
---|
6955 | ! |
---|
6956 | !-- Gamma size distribution to calculate |
---|
6957 | !-- critical weight_factor (e.g. Marshall + Palmer, 1948). |
---|
6958 | zeta = m1 * m3 / m2**2 |
---|
6959 | mu = MAX( ( ( 1.0_wp - zeta ) * 2.0_wp + 1.0_wp ) / ( zeta - 1.0_wp ), 0.0_wp ) |
---|
6960 | lambda = ( pirho_l * nr / lwc * & |
---|
6961 | ( mu + 3.0_wp ) * ( mu + 2.0_wp ) * ( mu + 1.0_wp ) & |
---|
6962 | )**0.3333333_wp |
---|
6963 | nr0 = ( nr / (gamma( mu + 1.0_wp ) ) ) * & |
---|
6964 | lambda**( mu + 1.0_wp ) |
---|
6965 | |
---|
6966 | DO n = 0, n_max-1 |
---|
6967 | diameter = r_bin_mid(n) * 2.0_wp |
---|
6968 | an_spl(n) = nr0 * diameter**mu * & |
---|
6969 | EXP( -lambda * diameter ) * & |
---|
6970 | ( r_bin(n+1) - r_bin(n) ) * 2.0_wp |
---|
6971 | ENDDO |
---|
6972 | ELSEIF ( isf == 2 ) THEN |
---|
6973 | ! |
---|
6974 | !-- Lognormal size distribution to calculate critical |
---|
6975 | !-- weight_factor (e.g. Levin, 1971, Bradley + Stow, 1974). |
---|
6976 | DO n = 0, n_max-1 |
---|
6977 | an_spl(n) = nr / ( SQRT( 2.0_wp * pi ) * & |
---|
6978 | LOG(sigma_log) * r_bin_mid(n) & |
---|
6979 | ) * & |
---|
6980 | EXP( -( LOG( r_bin_mid(n) / rm )**2 ) / & |
---|
6981 | ( 2.0_wp * LOG(sigma_log)**2 ) & |
---|
6982 | ) * & |
---|
6983 | ( r_bin(n+1) - r_bin(n) ) |
---|
6984 | ENDDO |
---|
6985 | ELSEIF ( isf == 3 ) THEN |
---|
6986 | ! |
---|
6987 | !-- Exponential size distribution to calculate critical weight_factor |
---|
6988 | !-- (e.g. Berry + Reinhardt, 1974). |
---|
6989 | DO n = 0, n_max-1 |
---|
6990 | an_spl(n) = 3.0_wp * nr * r_bin_mid(n)**2 / rm**3 * & |
---|
6991 | EXP( - ( r_bin_mid(n)**3 / rm**3 ) ) * & |
---|
6992 | ( r_bin(n+1) - r_bin(n) ) |
---|
6993 | ENDDO |
---|
6994 | ENDIF |
---|
6995 | |
---|
6996 | ! |
---|
6997 | !-- Criterion to avoid super droplets with a weighting factor < 1.0. |
---|
6998 | an_spl = MAX(an_spl, 1.0_wp) |
---|
6999 | ! |
---|
7000 | !-- Start splitting operations. Each particle is checked if it fulfilled the splitting |
---|
7001 | !-- criterions. In splitting mode 'gb_av' a critical radius (radius_split) and a |
---|
7002 | !-- splitting function must be prescribed (see particles_par). The critical weighting |
---|
7003 | !-- factor is calculated while appoximating a 'gamma', 'log' or 'exp'-drop size |
---|
7004 | !-- distribution. In this mode a DSD is calculated for every cloudy grid box. Super |
---|
7005 | !-- droplets which have a larger radius and larger weighting factor are split into |
---|
7006 | !-- 'splitting_factor' super droplets. In this case the splitting factor is calculated |
---|
7007 | !-- by the weighting factor of the super droplet and the approximated number |
---|
7008 | !-- concentration for droplets of such size. Due to the splitting, the weighting factor |
---|
7009 | !-- of the super droplet and all created clones are reduced by the factor of |
---|
7010 | !-- 'splitting_facor'. |
---|
7011 | DO n = 1, number_of_particles |
---|
7012 | DO np = 0, n_max-1 |
---|
7013 | IF ( r_bin(np) >= radius_split .AND. & |
---|
7014 | particles(n)%particle_mask .AND. & |
---|
7015 | particles(n)%radius >= r_bin(np) .AND. & |
---|
7016 | particles(n)%radius < r_bin(np+1) .AND. & |
---|
7017 | particles(n)%weight_factor >= an_spl(np) ) & |
---|
7018 | THEN |
---|
7019 | ! |
---|
7020 | !-- Calculate splitting factor. |
---|
7021 | splitting_factor = MIN( INT( particles(n)%weight_factor / an_spl(np) ), & |
---|
7022 | splitting_factor_max & |
---|
7023 | ) |
---|
7024 | IF ( splitting_factor < 2 ) CYCLE |
---|
7025 | |
---|
7026 | ! |
---|
7027 | !-- Calculate the new number of particles. |
---|
7028 | new_size = prt_count(k,j,i) + splitting_factor - 1 |
---|
7029 | ! |
---|
7030 | !-- Cycle if maximum number of particles per grid box |
---|
7031 | !-- is greater than the allowed maximum number. |
---|
7032 | IF ( new_size >= max_number_particles_per_gridbox ) CYCLE |
---|
7033 | ! |
---|
7034 | !-- Reallocate particle array if necessary. |
---|
7035 | IF ( new_size > SIZE( particles ) ) THEN |
---|
7036 | CALL realloc_particles_array( i, j, k, new_size ) |
---|
7037 | ENDIF |
---|
7038 | ! |
---|
7039 | !-- Calculate new weighting factor. |
---|
7040 | particles(n)%weight_factor = particles(n)%weight_factor / splitting_factor |
---|
7041 | tmp_particle = particles(n) |
---|
7042 | old_size = prt_count(k,j,i) |
---|
7043 | ! |
---|
7044 | !-- Create splitting_factor-1 new particles. |
---|
7045 | DO jpp = 1, splitting_factor-1 |
---|
7046 | grid_particles(k,j,i)%particles( jpp + old_size ) = tmp_particle |
---|
7047 | ENDDO |
---|
7048 | ! |
---|
7049 | !-- Save the new number of droplets for every grid box. |
---|
7050 | prt_count(k,j,i) = prt_count(k,j,i) + splitting_factor - 1 |
---|
7051 | new_particles_gb = new_particles_gb + splitting_factor - 1 |
---|
7052 | ENDIF |
---|
7053 | ENDDO |
---|
7054 | ENDDO |
---|
7055 | ENDDO |
---|
7056 | ENDDO |
---|
7057 | ENDDO |
---|
7058 | ENDIF |
---|
7059 | |
---|
7060 | CALL cpu_log( log_point_s(80), 'lpm_splitting', 'stop' ) |
---|
7061 | |
---|
7062 | END SUBROUTINE lpm_splitting |
---|
7063 | |
---|
7064 | |
---|
7065 | !--------------------------------------------------------------------------------------------------! |
---|
7066 | ! Description: |
---|
7067 | ! ------------ |
---|
7068 | ! This routine is a part of the Lagrangian particle model. Two Super droplets which fulfill certain |
---|
7069 | ! criterions (e.g. a big weighting factor and a small radius) can be merged into one super droplet |
---|
7070 | ! with a increased number of represented particles of the super droplet. This mechanism ensures an |
---|
7071 | ! improved feasible amount of computational costs. The limits of particle creation should be chosen |
---|
7072 | ! carefully! The idea of this algorithm is based on Unterstrasser and Soelch, 2014. |
---|
7073 | !--------------------------------------------------------------------------------------------------! |
---|
7074 | SUBROUTINE lpm_merging |
---|
7075 | |
---|
7076 | INTEGER(iwp) :: i !< |
---|
7077 | INTEGER(iwp) :: j !< |
---|
7078 | INTEGER(iwp) :: k !< |
---|
7079 | INTEGER(iwp) :: n !< |
---|
7080 | INTEGER(iwp) :: merge_drp = 0 !< number of merged droplets |
---|
7081 | |
---|
7082 | |
---|
7083 | REAL(wp) :: ql_crit = 1.0E-5_wp !< threshold lwc for cloudy grid cells |
---|
7084 | !< (e.g. Siebesma et al 2003, JAS, 60) |
---|
7085 | |
---|
7086 | CALL cpu_log( log_point_s(81), 'lpm_merging', 'start' ) |
---|
7087 | |
---|
7088 | merge_drp = 0 |
---|
7089 | |
---|
7090 | IF ( weight_factor_merge == -1.0_wp ) THEN |
---|
7091 | weight_factor_merge = 0.5_wp * initial_weighting_factor |
---|
7092 | ENDIF |
---|
7093 | |
---|
7094 | DO i = nxl, nxr |
---|
7095 | DO j = nys, nyn |
---|
7096 | DO k = nzb+1, nzt |
---|
7097 | |
---|
7098 | number_of_particles = prt_count(k,j,i) |
---|
7099 | IF ( number_of_particles <= 0 .OR. ql(k,j,i) >= ql_crit ) CYCLE |
---|
7100 | particles => grid_particles(k,j,i)%particles(1:number_of_particles) |
---|
7101 | ! |
---|
7102 | !-- Start merging operations: This routine deletes super droplets with a small radius |
---|
7103 | !-- (radius <= radius_merge) and a low weighting factor (weight_factor <= |
---|
7104 | !-- weight_factor_merge). The number of represented particles will be added to the next |
---|
7105 | !-- particle of the particle array. Tests showed that this simplified method can be used |
---|
7106 | !-- because it will only take place outside of cloudy grid boxes where ql <= 1.0E-5 kg/kg. |
---|
7107 | !-- Therefore, especially former cloned and subsequent evaporated super droplets will be |
---|
7108 | !-- merged. |
---|
7109 | DO n = 1, number_of_particles-1 |
---|
7110 | IF ( particles(n)%particle_mask .AND. & |
---|
7111 | particles(n+1)%particle_mask .AND. & |
---|
7112 | particles(n)%radius <= radius_merge .AND. & |
---|
7113 | particles(n)%weight_factor <= weight_factor_merge ) & |
---|
7114 | THEN |
---|
7115 | particles(n+1)%weight_factor = particles(n+1)%weight_factor + & |
---|
7116 | ( particles(n)%radius**3 / & |
---|
7117 | particles(n+1)%radius**3 * & |
---|
7118 | particles(n)%weight_factor & |
---|
7119 | ) |
---|
7120 | particles(n)%particle_mask = .FALSE. |
---|
7121 | deleted_particles = deleted_particles + 1 |
---|
7122 | merge_drp = merge_drp + 1 |
---|
7123 | |
---|
7124 | ENDIF |
---|
7125 | ENDDO |
---|
7126 | ENDDO |
---|
7127 | ENDDO |
---|
7128 | ENDDO |
---|
7129 | |
---|
7130 | |
---|
7131 | CALL cpu_log( log_point_s(81), 'lpm_merging', 'stop' ) |
---|
7132 | |
---|
7133 | END SUBROUTINE lpm_merging |
---|
7134 | |
---|
7135 | |
---|
7136 | |
---|
7137 | |
---|
7138 | !--------------------------------------------------------------------------------------------------! |
---|
7139 | ! Description: |
---|
7140 | ! ------------ |
---|
7141 | !> Exchange between subdomains. |
---|
7142 | !> As soon as one particle has moved beyond the boundary of the domain, it is included in the |
---|
7143 | !> relevant transfer arrays and marked for subsequent deletion on this PE. |
---|
7144 | !> First sweep for crossings in x direction. Find out first the number of particles to be |
---|
7145 | !> transferred and allocate temporary arrays needed to store them. |
---|
7146 | !> For a one-dimensional decomposition along y, no transfer is necessary, because the particle |
---|
7147 | !> remains on the PE, but the particle coordinate has to be adjusted. |
---|
7148 | !--------------------------------------------------------------------------------------------------! |
---|
7149 | SUBROUTINE lpm_exchange_horiz |
---|
7150 | |
---|
7151 | INTEGER(iwp) :: ip !< index variable along x |
---|
7152 | INTEGER(iwp) :: jp !< index variable along y |
---|
7153 | INTEGER(iwp) :: kp !< index variable along z |
---|
7154 | INTEGER(iwp) :: n !< particle index variable |
---|
7155 | |
---|
7156 | #if defined( __parallel ) |
---|
7157 | INTEGER(iwp) :: i !< grid index (x) of particle positition |
---|
7158 | INTEGER(iwp) :: j !< grid index (y) of particle positition |
---|
7159 | INTEGER(iwp) :: par_size !< Particle size in bytes |
---|
7160 | INTEGER(iwp) :: trlp_count !< number of particles send to left PE |
---|
7161 | INTEGER(iwp) :: trlp_count_recv !< number of particles receive from right PE |
---|
7162 | INTEGER(iwp) :: trnp_count !< number of particles send to north PE |
---|
7163 | INTEGER(iwp) :: trnp_count_recv !< number of particles receive from south PE |
---|
7164 | INTEGER(iwp) :: trrp_count !< number of particles send to right PE |
---|
7165 | INTEGER(iwp) :: trrp_count_recv !< number of particles receive from left PE |
---|
7166 | INTEGER(iwp) :: trsp_count !< number of particles send to south PE |
---|
7167 | INTEGER(iwp) :: trsp_count_recv !< number of particles receive from north PE |
---|
7168 | |
---|
7169 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: rvlp !< particles received from right PE |
---|
7170 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: rvnp !< particles received from south PE |
---|
7171 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: rvrp !< particles received from left PE |
---|
7172 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: rvsp !< particles received from north PE |
---|
7173 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: trlp !< particles send to left PE |
---|
7174 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: trnp !< particles send to north PE |
---|
7175 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: trrp !< particles send to right PE |
---|
7176 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: trsp !< particles send to south PE |
---|
7177 | #endif |
---|
7178 | |
---|
7179 | CALL cpu_log( log_point_s(23), 'lpm_exchange_horiz', 'start' ) |
---|
7180 | |
---|
7181 | #if defined( __parallel ) |
---|
7182 | |
---|
7183 | ! |
---|
7184 | !-- Exchange between subdomains. |
---|
7185 | !-- As soon as one particle has moved beyond the boundary of the domain, it is included in the |
---|
7186 | !-- relevant transfer arrays and marked for subsequent deletion on this PE. |
---|
7187 | !-- First sweep for crossings in x direction. Find out first the number of particles to be |
---|
7188 | !-- transferred and allocate temporary arrays needed to store them. |
---|
7189 | !-- For a one-dimensional decomposition along y, no transfer is necessary, because the particle |
---|
7190 | !-- remains on the PE, but the particle coordinate has to be adjusted. |
---|
7191 | trlp_count = 0 |
---|
7192 | trrp_count = 0 |
---|
7193 | |
---|
7194 | trlp_count_recv = 0 |
---|
7195 | trrp_count_recv = 0 |
---|
7196 | |
---|
7197 | IF ( pdims(1) /= 1 ) THEN |
---|
7198 | ! |
---|
7199 | !-- First calculate the storage necessary for sending and receiving the data. Compute only first |
---|
7200 | !-- (nxl) and last (nxr) loop iterration. |
---|
7201 | DO ip = nxl, nxr, nxr - nxl |
---|
7202 | DO jp = nys, nyn |
---|
7203 | DO kp = nzb+1, nzt |
---|
7204 | |
---|
7205 | number_of_particles = prt_count(kp,jp,ip) |
---|
7206 | IF ( number_of_particles <= 0 ) CYCLE |
---|
7207 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
7208 | DO n = 1, number_of_particles |
---|
7209 | IF ( particles(n)%particle_mask ) THEN |
---|
7210 | i = particles(n)%x * ddx |
---|
7211 | ! |
---|
7212 | !-- Above calculation does not work for indices less than zero |
---|
7213 | IF ( particles(n)%x < 0.0_wp) i = -1 |
---|
7214 | |
---|
7215 | IF ( i < nxl ) THEN |
---|
7216 | trlp_count = trlp_count + 1 |
---|
7217 | ELSEIF ( i > nxr ) THEN |
---|
7218 | trrp_count = trrp_count + 1 |
---|
7219 | ENDIF |
---|
7220 | ENDIF |
---|
7221 | ENDDO |
---|
7222 | |
---|
7223 | ENDDO |
---|
7224 | ENDDO |
---|
7225 | ENDDO |
---|
7226 | |
---|
7227 | IF ( trlp_count == 0 ) trlp_count = 1 |
---|
7228 | IF ( trrp_count == 0 ) trrp_count = 1 |
---|
7229 | |
---|
7230 | ALLOCATE( trlp(trlp_count), trrp(trrp_count) ) |
---|
7231 | |
---|
7232 | trlp = zero_particle |
---|
7233 | trrp = zero_particle |
---|
7234 | |
---|
7235 | trlp_count = 0 |
---|
7236 | trrp_count = 0 |
---|
7237 | |
---|
7238 | ENDIF |
---|
7239 | ! |
---|
7240 | !-- Compute only first (nxl) and last (nxr) loop iterration |
---|
7241 | DO ip = nxl, nxr, nxr-nxl |
---|
7242 | DO jp = nys, nyn |
---|
7243 | DO kp = nzb+1, nzt |
---|
7244 | number_of_particles = prt_count(kp,jp,ip) |
---|
7245 | IF ( number_of_particles <= 0 ) CYCLE |
---|
7246 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
7247 | DO n = 1, number_of_particles |
---|
7248 | ! |
---|
7249 | !-- Only those particles that have not been marked as 'deleted' may be moved. |
---|
7250 | IF ( particles(n)%particle_mask ) THEN |
---|
7251 | |
---|
7252 | i = particles(n)%x * ddx |
---|
7253 | ! |
---|
7254 | !-- Above calculation does not work for indices less than zero |
---|
7255 | IF ( particles(n)%x < 0.0_wp ) i = -1 |
---|
7256 | |
---|
7257 | IF ( i < nxl ) THEN |
---|
7258 | IF ( i < 0 ) THEN |
---|
7259 | ! |
---|
7260 | !-- Apply boundary condition along x |
---|
7261 | IF ( ibc_par_lr == 0 ) THEN |
---|
7262 | ! |
---|
7263 | !-- Cyclic condition |
---|
7264 | IF ( pdims(1) == 1 ) THEN |
---|
7265 | particles(n)%x = ( nx + 1 ) * dx + particles(n)%x |
---|
7266 | particles(n)%origin_x = ( nx + 1 ) * dx + & |
---|
7267 | particles(n)%origin_x |
---|
7268 | ELSE |
---|
7269 | trlp_count = trlp_count + 1 |
---|
7270 | trlp(trlp_count) = particles(n) |
---|
7271 | trlp(trlp_count)%x = ( nx + 1 ) * dx + trlp(trlp_count)%x |
---|
7272 | trlp(trlp_count)%origin_x = trlp(trlp_count)%origin_x + & |
---|
7273 | ( nx + 1 ) * dx |
---|
7274 | particles(n)%particle_mask = .FALSE. |
---|
7275 | deleted_particles = deleted_particles + 1 |
---|
7276 | |
---|
7277 | IF ( trlp(trlp_count)%x >= (nx + 1)* dx - 1.0E-12_wp ) THEN |
---|
7278 | trlp(trlp_count)%x = trlp(trlp_count)%x - 1.0E-10_wp |
---|
7279 | !++ why is 1 subtracted in next statement??? |
---|
7280 | trlp(trlp_count)%origin_x = trlp(trlp_count)%origin_x - 1 |
---|
7281 | ENDIF |
---|
7282 | |
---|
7283 | ENDIF |
---|
7284 | |
---|
7285 | ELSEIF ( ibc_par_lr == 1 ) THEN |
---|
7286 | ! |
---|
7287 | !-- Particle absorption |
---|
7288 | particles(n)%particle_mask = .FALSE. |
---|
7289 | deleted_particles = deleted_particles + 1 |
---|
7290 | |
---|
7291 | ELSEIF ( ibc_par_lr == 2 ) THEN |
---|
7292 | ! |
---|
7293 | !-- Particle reflection |
---|
7294 | particles(n)%x = -particles(n)%x |
---|
7295 | particles(n)%speed_x = -particles(n)%speed_x |
---|
7296 | |
---|
7297 | ENDIF |
---|
7298 | ELSE |
---|
7299 | ! |
---|
7300 | !-- Store particle data in the transfer array, which will be send to the |
---|
7301 | !-- neighbouring PE |
---|
7302 | trlp_count = trlp_count + 1 |
---|
7303 | trlp(trlp_count) = particles(n) |
---|
7304 | particles(n)%particle_mask = .FALSE. |
---|
7305 | deleted_particles = deleted_particles + 1 |
---|
7306 | |
---|
7307 | ENDIF |
---|
7308 | |
---|
7309 | ELSEIF ( i > nxr ) THEN |
---|
7310 | IF ( i > nx ) THEN |
---|
7311 | ! |
---|
7312 | !-- Apply boundary condition along x |
---|
7313 | IF ( ibc_par_lr == 0 ) THEN |
---|
7314 | ! |
---|
7315 | !-- Cyclic condition |
---|
7316 | IF ( pdims(1) == 1 ) THEN |
---|
7317 | particles(n)%x = particles(n)%x - ( nx + 1 ) * dx |
---|
7318 | particles(n)%origin_x = particles(n)%origin_x - ( nx + 1 ) * dx |
---|
7319 | ELSE |
---|
7320 | trrp_count = trrp_count + 1 |
---|
7321 | trrp(trrp_count) = particles(n) |
---|
7322 | trrp(trrp_count)%x = trrp(trrp_count)%x - ( nx + 1 ) * dx |
---|
7323 | trrp(trrp_count)%origin_x = trrp(trrp_count)%origin_x - & |
---|
7324 | ( nx + 1 ) * dx |
---|
7325 | particles(n)%particle_mask = .FALSE. |
---|
7326 | deleted_particles = deleted_particles + 1 |
---|
7327 | |
---|
7328 | ENDIF |
---|
7329 | |
---|
7330 | ELSEIF ( ibc_par_lr == 1 ) THEN |
---|
7331 | ! |
---|
7332 | !-- Particle absorption |
---|
7333 | particles(n)%particle_mask = .FALSE. |
---|
7334 | deleted_particles = deleted_particles + 1 |
---|
7335 | |
---|
7336 | ELSEIF ( ibc_par_lr == 2 ) THEN |
---|
7337 | ! |
---|
7338 | !-- Particle reflection |
---|
7339 | particles(n)%x = 2 * ( nx * dx ) - particles(n)%x |
---|
7340 | particles(n)%speed_x = -particles(n)%speed_x |
---|
7341 | |
---|
7342 | ENDIF |
---|
7343 | ELSE |
---|
7344 | ! |
---|
7345 | !-- Store particle data in the transfer array, which will be send to the |
---|
7346 | !-- neighbouring PE |
---|
7347 | trrp_count = trrp_count + 1 |
---|
7348 | trrp(trrp_count) = particles(n) |
---|
7349 | particles(n)%particle_mask = .FALSE. |
---|
7350 | deleted_particles = deleted_particles + 1 |
---|
7351 | |
---|
7352 | ENDIF |
---|
7353 | |
---|
7354 | ENDIF |
---|
7355 | ENDIF |
---|
7356 | |
---|
7357 | ENDDO |
---|
7358 | ENDDO |
---|
7359 | ENDDO |
---|
7360 | ENDDO |
---|
7361 | |
---|
7362 | ! |
---|
7363 | !-- STORAGE_SIZE returns the storage size of argument A in bits. However , it |
---|
7364 | !-- is needed in bytes. The function C_SIZEOF which produces this value directly |
---|
7365 | !-- causes problems with gfortran. For this reason the use of C_SIZEOF is avoided |
---|
7366 | par_size = STORAGE_SIZE( trlp(1) ) / 8 |
---|
7367 | |
---|
7368 | |
---|
7369 | ! |
---|
7370 | !-- Allocate arrays required for north-south exchange, as these are used directly after particles |
---|
7371 | !-- are exchange along x-direction. |
---|
7372 | ALLOCATE( move_also_north(1:nr_2_direction_move) ) |
---|
7373 | ALLOCATE( move_also_south(1:nr_2_direction_move) ) |
---|
7374 | |
---|
7375 | nr_move_north = 0 |
---|
7376 | nr_move_south = 0 |
---|
7377 | ! |
---|
7378 | !-- Send left boundary, receive right boundary (but first exchange how many and check, if particle |
---|
7379 | !-- storage must be extended) |
---|
7380 | IF ( pdims(1) /= 1 ) THEN |
---|
7381 | |
---|
7382 | CALL MPI_SENDRECV( trlp_count, 1, MPI_INTEGER, pleft, 0, & |
---|
7383 | trrp_count_recv, 1, MPI_INTEGER, pright, 0, & |
---|
7384 | comm2d, status, ierr ) |
---|
7385 | |
---|
7386 | ALLOCATE(rvrp(MAX(1,trrp_count_recv))) |
---|
7387 | |
---|
7388 | CALL MPI_SENDRECV( trlp, MAX(1,trlp_count)*par_size, MPI_BYTE, pleft, 1, & |
---|
7389 | rvrp, MAX(1,trrp_count_recv)*par_size, MPI_BYTE, pright, 1, & |
---|
7390 | comm2d, status, ierr ) |
---|
7391 | |
---|
7392 | IF ( trrp_count_recv > 0 ) CALL lpm_add_particles_to_gridcell(rvrp(1:trrp_count_recv)) |
---|
7393 | |
---|
7394 | DEALLOCATE(rvrp) |
---|
7395 | |
---|
7396 | ! |
---|
7397 | !-- Send right boundary, receive left boundary |
---|
7398 | CALL MPI_SENDRECV( trrp_count, 1, MPI_INTEGER, pright, 0, & |
---|
7399 | trlp_count_recv, 1, MPI_INTEGER, pleft, 0, & |
---|
7400 | comm2d, status, ierr ) |
---|
7401 | |
---|
7402 | ALLOCATE(rvlp(MAX(1,trlp_count_recv))) |
---|
7403 | ! |
---|
7404 | !-- This MPI_SENDRECV should work even with odd mixture on 32 and 64 Bit variables in structure |
---|
7405 | !-- particle_type (due to the calculation of par_size) |
---|
7406 | CALL MPI_SENDRECV( trrp, MAX(1,trrp_count)*par_size, MPI_BYTE, pright, 1, & |
---|
7407 | rvlp, MAX(1,trlp_count_recv)*par_size, MPI_BYTE, pleft, 1, & |
---|
7408 | comm2d, status, ierr ) |
---|
7409 | |
---|
7410 | IF ( trlp_count_recv > 0 ) CALL lpm_add_particles_to_gridcell(rvlp(1:trlp_count_recv)) |
---|
7411 | |
---|
7412 | DEALLOCATE( rvlp ) |
---|
7413 | DEALLOCATE( trlp, trrp ) |
---|
7414 | |
---|
7415 | ENDIF |
---|
7416 | |
---|
7417 | ! |
---|
7418 | !-- Check whether particles have crossed the boundaries in y direction. Note that this case can also |
---|
7419 | !-- apply to particles that have just been received from the adjacent right or left PE. |
---|
7420 | !-- Find out first the number of particles to be transferred and allocate temporary arrays needed to |
---|
7421 | !-- store them. |
---|
7422 | !-- For a one-dimensional decomposition along y, no transfer is necessary, because the particle |
---|
7423 | !-- remains on the PE. |
---|
7424 | trsp_count = nr_move_south |
---|
7425 | trnp_count = nr_move_north |
---|
7426 | |
---|
7427 | trsp_count_recv = 0 |
---|
7428 | trnp_count_recv = 0 |
---|
7429 | |
---|
7430 | IF ( pdims(2) /= 1 ) THEN |
---|
7431 | ! |
---|
7432 | !-- First calculate the storage necessary for sending and receiving the data |
---|
7433 | DO ip = nxl, nxr |
---|
7434 | DO jp = nys, nyn, nyn-nys !compute only first (nys) and last (nyn) loop iterration |
---|
7435 | DO kp = nzb+1, nzt |
---|
7436 | number_of_particles = prt_count(kp,jp,ip) |
---|
7437 | IF ( number_of_particles <= 0 ) CYCLE |
---|
7438 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
7439 | DO n = 1, number_of_particles |
---|
7440 | IF ( particles(n)%particle_mask ) THEN |
---|
7441 | j = particles(n)%y * ddy |
---|
7442 | ! |
---|
7443 | !-- Above calculation does not work for indices less than zero |
---|
7444 | IF ( particles(n)%y < 0.0_wp) j = -1 |
---|
7445 | |
---|
7446 | IF ( j < nys ) THEN |
---|
7447 | trsp_count = trsp_count + 1 |
---|
7448 | ELSEIF ( j > nyn ) THEN |
---|
7449 | trnp_count = trnp_count + 1 |
---|
7450 | ENDIF |
---|
7451 | ENDIF |
---|
7452 | ENDDO |
---|
7453 | ENDDO |
---|
7454 | ENDDO |
---|
7455 | ENDDO |
---|
7456 | |
---|
7457 | IF ( trsp_count == 0 ) trsp_count = 1 |
---|
7458 | IF ( trnp_count == 0 ) trnp_count = 1 |
---|
7459 | |
---|
7460 | ALLOCATE( trsp(trsp_count), trnp(trnp_count) ) |
---|
7461 | |
---|
7462 | trsp = zero_particle |
---|
7463 | trnp = zero_particle |
---|
7464 | |
---|
7465 | trsp_count = nr_move_south |
---|
7466 | trnp_count = nr_move_north |
---|
7467 | |
---|
7468 | trsp(1:nr_move_south) = move_also_south(1:nr_move_south) |
---|
7469 | trnp(1:nr_move_north) = move_also_north(1:nr_move_north) |
---|
7470 | |
---|
7471 | ENDIF |
---|
7472 | |
---|
7473 | DO ip = nxl, nxr |
---|
7474 | DO jp = nys, nyn, nyn-nys ! compute only first (nys) and last (nyn) loop iterration |
---|
7475 | DO kp = nzb+1, nzt |
---|
7476 | number_of_particles = prt_count(kp,jp,ip) |
---|
7477 | IF ( number_of_particles <= 0 ) CYCLE |
---|
7478 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
7479 | DO n = 1, number_of_particles |
---|
7480 | ! |
---|
7481 | !-- Only those particles that have not been marked as 'deleted' may be moved. |
---|
7482 | IF ( particles(n)%particle_mask ) THEN |
---|
7483 | |
---|
7484 | j = particles(n)%y * ddy |
---|
7485 | ! |
---|
7486 | !-- Above calculation does not work for indices less than zero |
---|
7487 | IF ( particles(n)%y < 0.0_wp ) j = -1 |
---|
7488 | |
---|
7489 | IF ( j < nys ) THEN |
---|
7490 | IF ( j < 0 ) THEN |
---|
7491 | ! |
---|
7492 | !-- Apply boundary condition along y |
---|
7493 | IF ( ibc_par_ns == 0 ) THEN |
---|
7494 | ! |
---|
7495 | !-- Cyclic condition |
---|
7496 | IF ( pdims(2) == 1 ) THEN |
---|
7497 | particles(n)%y = ( ny + 1 ) * dy + particles(n)%y |
---|
7498 | particles(n)%origin_y = ( ny + 1 ) * dy + particles(n)%origin_y |
---|
7499 | ELSE |
---|
7500 | trsp_count = trsp_count + 1 |
---|
7501 | trsp(trsp_count) = particles(n) |
---|
7502 | trsp(trsp_count)%y = ( ny + 1 ) * dy + trsp(trsp_count)%y |
---|
7503 | trsp(trsp_count)%origin_y = trsp(trsp_count)%origin_y & |
---|
7504 | + ( ny + 1 ) * dy |
---|
7505 | particles(n)%particle_mask = .FALSE. |
---|
7506 | deleted_particles = deleted_particles + 1 |
---|
7507 | |
---|
7508 | IF ( trsp(trsp_count)%y >= (ny+1)* dy - 1.0E-12_wp ) THEN |
---|
7509 | trsp(trsp_count)%y = trsp(trsp_count)%y - 1.0E-10_wp |
---|
7510 | !++ why is 1 subtracted in next statement??? |
---|
7511 | trsp(trsp_count)%origin_y = trsp(trsp_count)%origin_y - 1 |
---|
7512 | ENDIF |
---|
7513 | |
---|
7514 | ENDIF |
---|
7515 | |
---|
7516 | ELSEIF ( ibc_par_ns == 1 ) THEN |
---|
7517 | ! |
---|
7518 | !-- Particle absorption |
---|
7519 | particles(n)%particle_mask = .FALSE. |
---|
7520 | deleted_particles = deleted_particles + 1 |
---|
7521 | |
---|
7522 | ELSEIF ( ibc_par_ns == 2 ) THEN |
---|
7523 | ! |
---|
7524 | !-- Particle reflection |
---|
7525 | particles(n)%y = -particles(n)%y |
---|
7526 | particles(n)%speed_y = -particles(n)%speed_y |
---|
7527 | |
---|
7528 | ENDIF |
---|
7529 | ELSE |
---|
7530 | ! |
---|
7531 | !-- Store particle data in the transfer array, which will be send to the |
---|
7532 | !-- neighbouring PE |
---|
7533 | trsp_count = trsp_count + 1 |
---|
7534 | trsp(trsp_count) = particles(n) |
---|
7535 | particles(n)%particle_mask = .FALSE. |
---|
7536 | deleted_particles = deleted_particles + 1 |
---|
7537 | |
---|
7538 | ENDIF |
---|
7539 | |
---|
7540 | ELSEIF ( j > nyn ) THEN |
---|
7541 | IF ( j > ny ) THEN |
---|
7542 | ! |
---|
7543 | !-- Apply boundary condition along y |
---|
7544 | IF ( ibc_par_ns == 0 ) THEN |
---|
7545 | ! |
---|
7546 | !-- Cyclic condition |
---|
7547 | IF ( pdims(2) == 1 ) THEN |
---|
7548 | particles(n)%y = particles(n)%y - ( ny + 1 ) * dy |
---|
7549 | particles(n)%origin_y = particles(n)%origin_y - ( ny + 1 ) * dy |
---|
7550 | ELSE |
---|
7551 | trnp_count = trnp_count + 1 |
---|
7552 | trnp(trnp_count) = particles(n) |
---|
7553 | trnp(trnp_count)%y = trnp(trnp_count)%y - ( ny + 1 ) * dy |
---|
7554 | trnp(trnp_count)%origin_y = & |
---|
7555 | trnp(trnp_count)%origin_y - ( ny + 1 ) * dy |
---|
7556 | particles(n)%particle_mask = .FALSE. |
---|
7557 | deleted_particles = deleted_particles + 1 |
---|
7558 | ENDIF |
---|
7559 | |
---|
7560 | ELSEIF ( ibc_par_ns == 1 ) THEN |
---|
7561 | ! |
---|
7562 | !-- Particle absorption |
---|
7563 | particles(n)%particle_mask = .FALSE. |
---|
7564 | deleted_particles = deleted_particles + 1 |
---|
7565 | |
---|
7566 | ELSEIF ( ibc_par_ns == 2 ) THEN |
---|
7567 | ! |
---|
7568 | !-- Particle reflection |
---|
7569 | particles(n)%y = 2 * ( ny * dy ) - particles(n)%y |
---|
7570 | particles(n)%speed_y = -particles(n)%speed_y |
---|
7571 | |
---|
7572 | ENDIF |
---|
7573 | ELSE |
---|
7574 | ! |
---|
7575 | !-- Store particle data in the transfer array, which will be send to the |
---|
7576 | !-- neighbouring PE |
---|
7577 | trnp_count = trnp_count + 1 |
---|
7578 | trnp(trnp_count) = particles(n) |
---|
7579 | particles(n)%particle_mask = .FALSE. |
---|
7580 | deleted_particles = deleted_particles + 1 |
---|
7581 | |
---|
7582 | ENDIF |
---|
7583 | |
---|
7584 | ENDIF |
---|
7585 | ENDIF |
---|
7586 | ENDDO |
---|
7587 | ENDDO |
---|
7588 | ENDDO |
---|
7589 | ENDDO |
---|
7590 | |
---|
7591 | ! |
---|
7592 | !-- Send front boundary, receive back boundary (but first exchange how many and check, if particle |
---|
7593 | !-- storage must be extended) |
---|
7594 | IF ( pdims(2) /= 1 ) THEN |
---|
7595 | |
---|
7596 | CALL MPI_SENDRECV( trsp_count, 1, MPI_INTEGER, psouth, 0, & |
---|
7597 | trnp_count_recv, 1, MPI_INTEGER, pnorth, 0, & |
---|
7598 | comm2d, status, ierr ) |
---|
7599 | |
---|
7600 | ALLOCATE(rvnp(MAX(1,trnp_count_recv))) |
---|
7601 | ! |
---|
7602 | !-- This MPI_SENDRECV should work even with odd mixture on 32 and 64 Bit variables in structure |
---|
7603 | !-- particle_type (due to the calculation of par_size) |
---|
7604 | CALL MPI_SENDRECV( trsp, trsp_count*par_size, MPI_BYTE, psouth, 1, & |
---|
7605 | rvnp, trnp_count_recv*par_size, MPI_BYTE, pnorth, 1, & |
---|
7606 | comm2d, status, ierr ) |
---|
7607 | |
---|
7608 | IF ( trnp_count_recv > 0 ) CALL lpm_add_particles_to_gridcell(rvnp(1:trnp_count_recv)) |
---|
7609 | |
---|
7610 | DEALLOCATE(rvnp) |
---|
7611 | |
---|
7612 | ! |
---|
7613 | !-- Send back boundary, receive front boundary |
---|
7614 | CALL MPI_SENDRECV( trnp_count, 1, MPI_INTEGER, pnorth, 0, & |
---|
7615 | trsp_count_recv, 1, MPI_INTEGER, psouth, 0, & |
---|
7616 | comm2d, status, ierr ) |
---|
7617 | |
---|
7618 | ALLOCATE(rvsp(MAX(1,trsp_count_recv))) |
---|
7619 | ! |
---|
7620 | !-- This MPI_SENDRECV should work even with odd mixture on 32 and 64 Bit variables in structure |
---|
7621 | !-- particle_type (due to the calculation of par_size) |
---|
7622 | CALL MPI_SENDRECV( trnp, trnp_count*par_size, MPI_BYTE, pnorth, 1, & |
---|
7623 | rvsp, trsp_count_recv*par_size, MPI_BYTE, psouth, 1, & |
---|
7624 | comm2d, status, ierr ) |
---|
7625 | |
---|
7626 | IF ( trsp_count_recv > 0 ) CALL lpm_add_particles_to_gridcell(rvsp(1:trsp_count_recv)) |
---|
7627 | |
---|
7628 | |
---|
7629 | DEALLOCATE(rvsp) |
---|
7630 | |
---|
7631 | number_of_particles = number_of_particles + trsp_count_recv |
---|
7632 | |
---|
7633 | DEALLOCATE( trsp, trnp ) |
---|
7634 | |
---|
7635 | ENDIF |
---|
7636 | |
---|
7637 | DEALLOCATE( move_also_north ) |
---|
7638 | DEALLOCATE( move_also_south ) |
---|
7639 | |
---|
7640 | #else |
---|
7641 | |
---|
7642 | DO ip = nxl, nxr, nxr-nxl |
---|
7643 | DO jp = nys, nyn |
---|
7644 | DO kp = nzb+1, nzt |
---|
7645 | number_of_particles = prt_count(kp,jp,ip) |
---|
7646 | IF ( number_of_particles <= 0 ) CYCLE |
---|
7647 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
7648 | DO n = 1, number_of_particles |
---|
7649 | ! |
---|
7650 | !-- Apply boundary conditions |
---|
7651 | |
---|
7652 | IF ( particles(n)%x < 0.0_wp ) THEN |
---|
7653 | |
---|
7654 | IF ( ibc_par_lr == 0 ) THEN |
---|
7655 | ! |
---|
7656 | !-- Cyclic boundary. Relevant coordinate has to be changed. |
---|
7657 | particles(n)%x = ( nx + 1 ) * dx + particles(n)%x |
---|
7658 | particles(n)%origin_x = ( nx + 1 ) * dx + particles(n)%origin_x |
---|
7659 | ELSEIF ( ibc_par_lr == 1 ) THEN |
---|
7660 | ! |
---|
7661 | !-- Particle absorption |
---|
7662 | particles(n)%particle_mask = .FALSE. |
---|
7663 | deleted_particles = deleted_particles + 1 |
---|
7664 | |
---|
7665 | ELSEIF ( ibc_par_lr == 2 ) THEN |
---|
7666 | ! |
---|
7667 | !-- Particle reflection |
---|
7668 | particles(n)%x = -dx - particles(n)%x |
---|
7669 | particles(n)%speed_x = -particles(n)%speed_x |
---|
7670 | ENDIF |
---|
7671 | |
---|
7672 | ELSEIF ( particles(n)%x >= ( nx + 1) * dx ) THEN |
---|
7673 | |
---|
7674 | IF ( ibc_par_lr == 0 ) THEN |
---|
7675 | ! |
---|
7676 | !-- Cyclic boundary. Relevant coordinate has to be changed. |
---|
7677 | particles(n)%x = particles(n)%x - ( nx + 1 ) * dx |
---|
7678 | particles(n)%origin_x = particles(n)%origin_x - ( nx + 1 ) * dx |
---|
7679 | |
---|
7680 | ELSEIF ( ibc_par_lr == 1 ) THEN |
---|
7681 | ! |
---|
7682 | !-- Particle absorption |
---|
7683 | particles(n)%particle_mask = .FALSE. |
---|
7684 | deleted_particles = deleted_particles + 1 |
---|
7685 | |
---|
7686 | ELSEIF ( ibc_par_lr == 2 ) THEN |
---|
7687 | ! |
---|
7688 | !-- Particle reflection |
---|
7689 | particles(n)%x = ( nx + 1 ) * dx - particles(n)%x |
---|
7690 | particles(n)%speed_x = -particles(n)%speed_x |
---|
7691 | ENDIF |
---|
7692 | |
---|
7693 | ENDIF |
---|
7694 | ENDDO |
---|
7695 | ENDDO |
---|
7696 | ENDDO |
---|
7697 | ENDDO |
---|
7698 | |
---|
7699 | DO ip = nxl, nxr |
---|
7700 | DO jp = nys, nyn, nyn-nys |
---|
7701 | DO kp = nzb+1, nzt |
---|
7702 | number_of_particles = prt_count(kp,jp,ip) |
---|
7703 | IF ( number_of_particles <= 0 ) CYCLE |
---|
7704 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
7705 | DO n = 1, number_of_particles |
---|
7706 | |
---|
7707 | IF ( particles(n)%y < 0.0_wp) THEN |
---|
7708 | |
---|
7709 | IF ( ibc_par_ns == 0 ) THEN |
---|
7710 | ! |
---|
7711 | !-- Cyclic boundary. Relevant coordinate has to be changed. |
---|
7712 | particles(n)%y = ( ny + 1 ) * dy + particles(n)%y |
---|
7713 | particles(n)%origin_y = ( ny + 1 ) * dy + particles(n)%origin_y |
---|
7714 | |
---|
7715 | ELSEIF ( ibc_par_ns == 1 ) THEN |
---|
7716 | ! |
---|
7717 | !-- Particle absorption |
---|
7718 | particles(n)%particle_mask = .FALSE. |
---|
7719 | deleted_particles = deleted_particles + 1 |
---|
7720 | |
---|
7721 | ELSEIF ( ibc_par_ns == 2 ) THEN |
---|
7722 | ! |
---|
7723 | !-- Particle reflection |
---|
7724 | particles(n)%y = -dy - particles(n)%y |
---|
7725 | particles(n)%speed_y = -particles(n)%speed_y |
---|
7726 | ENDIF |
---|
7727 | |
---|
7728 | ELSEIF ( particles(n)%y >= ( ny + 1) * dy ) THEN |
---|
7729 | |
---|
7730 | IF ( ibc_par_ns == 0 ) THEN |
---|
7731 | ! |
---|
7732 | !-- Cyclic boundary. Relevant coordinate has to be changed. |
---|
7733 | particles(n)%y = particles(n)%y - ( ny + 1 ) * dy |
---|
7734 | particles(n)%origin_y = particles(n)%origin_y - ( ny + 1 ) * dy |
---|
7735 | |
---|
7736 | ELSEIF ( ibc_par_ns == 1 ) THEN |
---|
7737 | ! |
---|
7738 | !-- Particle absorption |
---|
7739 | particles(n)%particle_mask = .FALSE. |
---|
7740 | deleted_particles = deleted_particles + 1 |
---|
7741 | |
---|
7742 | ELSEIF ( ibc_par_ns == 2 ) THEN |
---|
7743 | ! |
---|
7744 | !-- Particle reflection |
---|
7745 | particles(n)%y = ( ny + 1 ) * dy - particles(n)%y |
---|
7746 | particles(n)%speed_y = -particles(n)%speed_y |
---|
7747 | ENDIF |
---|
7748 | |
---|
7749 | ENDIF |
---|
7750 | |
---|
7751 | ENDDO |
---|
7752 | ENDDO |
---|
7753 | ENDDO |
---|
7754 | ENDDO |
---|
7755 | #endif |
---|
7756 | |
---|
7757 | ! |
---|
7758 | !-- Accumulate the number of particles transferred between the subdomains |
---|
7759 | #if defined( __parallel ) |
---|
7760 | trlp_count_sum = trlp_count_sum + trlp_count |
---|
7761 | trlp_count_recv_sum = trlp_count_recv_sum + trlp_count_recv |
---|
7762 | trrp_count_sum = trrp_count_sum + trrp_count |
---|
7763 | trrp_count_recv_sum = trrp_count_recv_sum + trrp_count_recv |
---|
7764 | trsp_count_sum = trsp_count_sum + trsp_count |
---|
7765 | trsp_count_recv_sum = trsp_count_recv_sum + trsp_count_recv |
---|
7766 | trnp_count_sum = trnp_count_sum + trnp_count |
---|
7767 | trnp_count_recv_sum = trnp_count_recv_sum + trnp_count_recv |
---|
7768 | #endif |
---|
7769 | |
---|
7770 | CALL cpu_log( log_point_s(23), 'lpm_exchange_horiz', 'stop' ) |
---|
7771 | |
---|
7772 | END SUBROUTINE lpm_exchange_horiz |
---|
7773 | |
---|
7774 | |
---|
7775 | !--------------------------------------------------------------------------------------------------! |
---|
7776 | ! Description: |
---|
7777 | ! ------------ |
---|
7778 | !> Exchange ghostpoints |
---|
7779 | !--------------------------------------------------------------------------------------------------! |
---|
7780 | SUBROUTINE lpm_exchange_horiz_bounds( location ) |
---|
7781 | |
---|
7782 | USE exchange_horiz_mod, & |
---|
7783 | ONLY: exchange_horiz |
---|
7784 | |
---|
7785 | CHARACTER (LEN=*), INTENT(IN) :: location !< call location string |
---|
7786 | |
---|
7787 | SELECT CASE ( location ) |
---|
7788 | |
---|
7789 | CASE ( 'before_prognostic_equation' ) |
---|
7790 | |
---|
7791 | CASE ( 'after_prognostic_equation' ) |
---|
7792 | |
---|
7793 | IF ( wang_kernel .OR. use_sgs_for_particles ) THEN |
---|
7794 | CALL exchange_horiz( diss, nbgp ) |
---|
7795 | ENDIF |
---|
7796 | |
---|
7797 | END SELECT |
---|
7798 | |
---|
7799 | END SUBROUTINE lpm_exchange_horiz_bounds |
---|
7800 | |
---|
7801 | |
---|
7802 | #if defined( __parallel ) |
---|
7803 | !--------------------------------------------------------------------------------------------------! |
---|
7804 | ! Description: |
---|
7805 | ! ------------ |
---|
7806 | !> If a particle moves from one processor to another, this subroutine moves the corresponding |
---|
7807 | !> elements from the particle arrays of the old grid cells to the particle arrays of the new grid |
---|
7808 | !> cells. |
---|
7809 | !--------------------------------------------------------------------------------------------------! |
---|
7810 | SUBROUTINE lpm_add_particles_to_gridcell (particle_array) |
---|
7811 | |
---|
7812 | IMPLICIT NONE |
---|
7813 | |
---|
7814 | INTEGER(iwp) :: ip !< grid index (x) of particle |
---|
7815 | INTEGER(iwp) :: jp !< grid index (x) of particle |
---|
7816 | INTEGER(iwp) :: kp !< grid index (x) of particle |
---|
7817 | INTEGER(iwp) :: n !< index variable of particle |
---|
7818 | INTEGER(iwp) :: pindex !< dummy argument for new number of particles per grid box |
---|
7819 | |
---|
7820 | LOGICAL :: pack_done !< |
---|
7821 | |
---|
7822 | TYPE(particle_type), DIMENSION(:), INTENT(IN) :: particle_array !< new particles in a grid box |
---|
7823 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: temp_ns !< temporary particle array for reallocation |
---|
7824 | |
---|
7825 | |
---|
7826 | pack_done = .FALSE. |
---|
7827 | |
---|
7828 | DO n = 1, SIZE( particle_array ) |
---|
7829 | |
---|
7830 | IF ( .NOT. particle_array(n)%particle_mask ) CYCLE |
---|
7831 | |
---|
7832 | ip = particle_array(n)%x * ddx |
---|
7833 | jp = particle_array(n)%y * ddy |
---|
7834 | ! |
---|
7835 | !-- In case of stretching the actual k index must be found |
---|
7836 | IF ( dz_stretch_level /= -9999999.9_wp .OR. dz_stretch_level_start(1) /= -9999999.9_wp ) & |
---|
7837 | THEN |
---|
7838 | kp = MAX( MINLOC( ABS( particle_array(n)%z - zu ), DIM = 1 ) - 1, 1 ) |
---|
7839 | ELSE |
---|
7840 | kp = INT( particle_array(n)%z / dz(1) + 1 + offset_ocean_nzt ) |
---|
7841 | ENDIF |
---|
7842 | |
---|
7843 | IF ( ip >= nxl .AND. ip <= nxr .AND. jp >= nys .AND. jp <= nyn .AND. & |
---|
7844 | kp >= nzb+1 .AND. kp <= nzt) THEN ! particle stays on processor |
---|
7845 | |
---|
7846 | number_of_particles = prt_count(kp,jp,ip) |
---|
7847 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
7848 | |
---|
7849 | pindex = prt_count(kp,jp,ip)+1 |
---|
7850 | IF( pindex > SIZE( grid_particles(kp,jp,ip)%particles ) ) THEN |
---|
7851 | IF ( pack_done ) THEN |
---|
7852 | CALL realloc_particles_array ( ip, jp, kp ) |
---|
7853 | ELSE |
---|
7854 | CALL lpm_pack |
---|
7855 | prt_count(kp,jp,ip) = number_of_particles |
---|
7856 | pindex = prt_count(kp,jp,ip)+1 |
---|
7857 | IF ( pindex > SIZE( grid_particles(kp,jp,ip)%particles ) ) THEN |
---|
7858 | CALL realloc_particles_array ( ip, jp, kp ) |
---|
7859 | ENDIF |
---|
7860 | pack_done = .TRUE. |
---|
7861 | ENDIF |
---|
7862 | ENDIF |
---|
7863 | grid_particles(kp,jp,ip)%particles(pindex) = particle_array(n) |
---|
7864 | prt_count(kp,jp,ip) = pindex |
---|
7865 | |
---|
7866 | ELSE |
---|
7867 | |
---|
7868 | IF ( jp <= nys - 1 ) THEN |
---|
7869 | |
---|
7870 | nr_move_south = nr_move_south+1 |
---|
7871 | ! |
---|
7872 | !-- Before particle information is swapped to exchange-array, check if enough memory is |
---|
7873 | !-- allocated. If required, reallocate exchange array. |
---|
7874 | IF ( nr_move_south > SIZE( move_also_south ) ) THEN |
---|
7875 | ! |
---|
7876 | !-- At first, allocate further temporary array to swap particle information. |
---|
7877 | ALLOCATE( temp_ns(SIZE( move_also_south )+nr_2_direction_move) ) |
---|
7878 | temp_ns(1:nr_move_south-1) = move_also_south(1:nr_move_south-1) |
---|
7879 | DEALLOCATE( move_also_south ) |
---|
7880 | ALLOCATE( move_also_south(SIZE( temp_ns )) ) |
---|
7881 | move_also_south(1:nr_move_south-1) = temp_ns(1:nr_move_south-1) |
---|
7882 | DEALLOCATE( temp_ns ) |
---|
7883 | |
---|
7884 | ENDIF |
---|
7885 | |
---|
7886 | move_also_south(nr_move_south) = particle_array(n) |
---|
7887 | |
---|
7888 | IF ( jp == -1 ) THEN |
---|
7889 | ! |
---|
7890 | !-- Apply boundary condition along y |
---|
7891 | IF ( ibc_par_ns == 0 ) THEN |
---|
7892 | move_also_south(nr_move_south)%y = & |
---|
7893 | move_also_south(nr_move_south)%y + ( ny + 1 ) * dy |
---|
7894 | move_also_south(nr_move_south)%origin_y = & |
---|
7895 | move_also_south(nr_move_south)%origin_y + ( ny + 1 ) * dy |
---|
7896 | ELSEIF ( ibc_par_ns == 1 ) THEN |
---|
7897 | ! |
---|
7898 | !-- Particle absorption |
---|
7899 | move_also_south(nr_move_south)%particle_mask = .FALSE. |
---|
7900 | deleted_particles = deleted_particles + 1 |
---|
7901 | |
---|
7902 | ELSEIF ( ibc_par_ns == 2 ) THEN |
---|
7903 | ! |
---|
7904 | !-- Particle reflection |
---|
7905 | move_also_south(nr_move_south)%y = -move_also_south(nr_move_south)%y |
---|
7906 | move_also_south(nr_move_south)%speed_y = -move_also_south(nr_move_south)%speed_y |
---|
7907 | |
---|
7908 | ENDIF |
---|
7909 | |
---|
7910 | ENDIF |
---|
7911 | |
---|
7912 | ELSEIF ( jp >= nyn+1 ) THEN |
---|
7913 | |
---|
7914 | nr_move_north = nr_move_north+1 |
---|
7915 | ! |
---|
7916 | !-- Before particle information is swapped to exchange-array, check if enough memory is |
---|
7917 | !-- allocated. If required, reallocate exchange array. |
---|
7918 | IF ( nr_move_north > SIZE( move_also_north ) ) THEN |
---|
7919 | ! |
---|
7920 | !-- At first, allocate further temporary array to swap particle information. |
---|
7921 | ALLOCATE( temp_ns(SIZE( move_also_north )+nr_2_direction_move) ) |
---|
7922 | temp_ns(1:nr_move_north-1) = move_also_south(1:nr_move_north-1) |
---|
7923 | DEALLOCATE( move_also_north ) |
---|
7924 | ALLOCATE( move_also_north(SIZE( temp_ns )) ) |
---|
7925 | move_also_north(1:nr_move_north-1) = temp_ns(1:nr_move_north-1) |
---|
7926 | DEALLOCATE( temp_ns ) |
---|
7927 | |
---|
7928 | ENDIF |
---|
7929 | |
---|
7930 | move_also_north(nr_move_north) = particle_array(n) |
---|
7931 | IF ( jp == ny+1 ) THEN |
---|
7932 | ! |
---|
7933 | !-- Apply boundary condition along y |
---|
7934 | IF ( ibc_par_ns == 0 ) THEN |
---|
7935 | |
---|
7936 | move_also_north(nr_move_north)%y = & |
---|
7937 | move_also_north(nr_move_north)%y - ( ny + 1 ) * dy |
---|
7938 | move_also_north(nr_move_north)%origin_y = & |
---|
7939 | move_also_north(nr_move_north)%origin_y - ( ny + 1 ) * dy |
---|
7940 | ELSEIF ( ibc_par_ns == 1 ) THEN |
---|
7941 | ! |
---|
7942 | !-- Particle absorption |
---|
7943 | move_also_north(nr_move_north)%particle_mask = .FALSE. |
---|
7944 | deleted_particles = deleted_particles + 1 |
---|
7945 | |
---|
7946 | ELSEIF ( ibc_par_ns == 2 ) THEN |
---|
7947 | ! |
---|
7948 | !-- Particle reflection |
---|
7949 | move_also_north(nr_move_north)%y = -move_also_north(nr_move_north)%y |
---|
7950 | move_also_north(nr_move_north)%speed_y = -move_also_north(nr_move_north)%speed_y |
---|
7951 | |
---|
7952 | ENDIF |
---|
7953 | |
---|
7954 | ENDIF |
---|
7955 | |
---|
7956 | ELSE |
---|
7957 | |
---|
7958 | IF ( .NOT. child_domain ) THEN |
---|
7959 | WRITE(0,'(a,8i7)') 'particle out of range ',myid,ip,jp,kp,nxl,nxr,nys,nyn |
---|
7960 | |
---|
7961 | ENDIF |
---|
7962 | |
---|
7963 | ENDIF |
---|
7964 | |
---|
7965 | ENDIF |
---|
7966 | |
---|
7967 | ENDDO |
---|
7968 | |
---|
7969 | END SUBROUTINE lpm_add_particles_to_gridcell |
---|
7970 | #endif |
---|
7971 | |
---|
7972 | |
---|
7973 | !--------------------------------------------------------------------------------------------------! |
---|
7974 | ! Description: |
---|
7975 | ! ------------ |
---|
7976 | !> If a particle moves from one grid cell to another (on the current processor!), this subroutine |
---|
7977 | !> moves the corresponding element from the particle array of the old grid cell to the particle |
---|
7978 | !> array of the new grid cell. |
---|
7979 | !--------------------------------------------------------------------------------------------------! |
---|
7980 | SUBROUTINE lpm_move_particle |
---|
7981 | |
---|
7982 | INTEGER(iwp) :: i !< grid index (x) of particle position |
---|
7983 | INTEGER(iwp) :: ip !< index variable along x |
---|
7984 | INTEGER(iwp) :: j !< grid index (y) of particle position |
---|
7985 | INTEGER(iwp) :: jp !< index variable along y |
---|
7986 | INTEGER(iwp) :: k !< grid index (z) of particle position |
---|
7987 | INTEGER(iwp) :: kp !< index variable along z |
---|
7988 | INTEGER(iwp) :: n !< index variable for particle array |
---|
7989 | INTEGER(iwp) :: np_before_move !< number of particles per grid box before moving |
---|
7990 | INTEGER(iwp) :: pindex !< dummy argument for number of new particle per grid box |
---|
7991 | |
---|
7992 | TYPE(particle_type), DIMENSION(:), POINTER :: particles_before_move !< particles before moving |
---|
7993 | |
---|
7994 | CALL cpu_log( log_point_s(41), 'lpm_move_particle', 'start' ) |
---|
7995 | CALL lpm_check_cfl |
---|
7996 | DO ip = nxl, nxr |
---|
7997 | DO jp = nys, nyn |
---|
7998 | DO kp = nzb+1, nzt |
---|
7999 | |
---|
8000 | np_before_move = prt_count(kp,jp,ip) |
---|
8001 | IF ( np_before_move <= 0 ) CYCLE |
---|
8002 | particles_before_move => grid_particles(kp,jp,ip)%particles(1:np_before_move) |
---|
8003 | |
---|
8004 | DO n = 1, np_before_move |
---|
8005 | i = particles_before_move(n)%x * ddx |
---|
8006 | j = particles_before_move(n)%y * ddy |
---|
8007 | k = kp |
---|
8008 | ! |
---|
8009 | !-- Find correct vertical particle grid box (necessary in case of grid stretching). |
---|
8010 | !-- Due to the CFL limitations only the neighbouring grid boxes are considered. |
---|
8011 | IF( zw(k) < particles_before_move(n)%z ) k = k + 1 |
---|
8012 | IF( zw(k-1) > particles_before_move(n)%z ) k = k - 1 |
---|
8013 | |
---|
8014 | !-- For lpm_exchange_horiz to work properly particles need to be moved to the outermost |
---|
8015 | !-- gridboxes of the respective processor. If the particle index is inside the processor |
---|
8016 | !-- the following lines will not change the index. |
---|
8017 | i = MIN ( i , nxr ) |
---|
8018 | i = MAX ( i , nxl ) |
---|
8019 | j = MIN ( j , nyn ) |
---|
8020 | j = MAX ( j , nys ) |
---|
8021 | |
---|
8022 | k = MIN ( k , nzt ) |
---|
8023 | k = MAX ( k , nzb+1 ) |
---|
8024 | |
---|
8025 | ! |
---|
8026 | !-- Check, if particle has moved to another grid cell. |
---|
8027 | IF ( i /= ip .OR. j /= jp .OR. k /= kp ) THEN |
---|
8028 | !! |
---|
8029 | !-- If the particle stays on the same processor, the particle will be added to the |
---|
8030 | !-- particle array of the new processor. |
---|
8031 | number_of_particles = prt_count(k,j,i) |
---|
8032 | particles => grid_particles(k,j,i)%particles(1:number_of_particles) |
---|
8033 | |
---|
8034 | pindex = prt_count(k,j,i)+1 |
---|
8035 | IF ( pindex > SIZE( grid_particles(k,j,i)%particles ) ) THEN |
---|
8036 | CALL realloc_particles_array( i, j, k ) |
---|
8037 | ENDIF |
---|
8038 | |
---|
8039 | grid_particles(k,j,i)%particles(pindex) = particles_before_move(n) |
---|
8040 | prt_count(k,j,i) = pindex |
---|
8041 | |
---|
8042 | particles_before_move(n)%particle_mask = .FALSE. |
---|
8043 | ENDIF |
---|
8044 | ENDDO |
---|
8045 | |
---|
8046 | ENDDO |
---|
8047 | ENDDO |
---|
8048 | ENDDO |
---|
8049 | |
---|
8050 | CALL cpu_log( log_point_s(41), 'lpm_move_particle', 'stop' ) |
---|
8051 | |
---|
8052 | RETURN |
---|
8053 | |
---|
8054 | END SUBROUTINE lpm_move_particle |
---|
8055 | |
---|
8056 | |
---|
8057 | !--------------------------------------------------------------------------------------------------! |
---|
8058 | ! Description: |
---|
8059 | ! ------------ |
---|
8060 | !> Check CFL-criterion for each particle. If one particle violated the criterion the particle will |
---|
8061 | !> be deleted and a warning message is given. |
---|
8062 | !--------------------------------------------------------------------------------------------------! |
---|
8063 | SUBROUTINE lpm_check_cfl |
---|
8064 | |
---|
8065 | IMPLICIT NONE |
---|
8066 | |
---|
8067 | INTEGER(iwp) :: i !< running index, x-direction |
---|
8068 | INTEGER(iwp) :: j !< running index, y-direction |
---|
8069 | INTEGER(iwp) :: k !< running index, z-direction |
---|
8070 | INTEGER(iwp) :: n !< running index, number of particles |
---|
8071 | |
---|
8072 | DO i = nxl, nxr |
---|
8073 | DO j = nys, nyn |
---|
8074 | DO k = nzb+1, nzt |
---|
8075 | number_of_particles = prt_count(k,j,i) |
---|
8076 | IF ( number_of_particles <= 0 ) CYCLE |
---|
8077 | particles => grid_particles(k,j,i)%particles(1:number_of_particles) |
---|
8078 | DO n = 1, number_of_particles |
---|
8079 | ! |
---|
8080 | !-- Note, check for CFL does not work at first particle timestep when both, age and |
---|
8081 | !-- age_m are zero. |
---|
8082 | IF ( particles(n)%age - particles(n)%age_m > 0.0_wp ) THEN |
---|
8083 | IF( ABS( particles(n)%speed_x ) > & |
---|
8084 | ( dx / ( particles(n)%age - particles(n)%age_m) ) .OR. & |
---|
8085 | ABS( particles(n)%speed_y ) > & |
---|
8086 | ( dy / ( particles(n)%age - particles(n)%age_m) ) .OR. & |
---|
8087 | ABS( particles(n)%speed_z ) > & |
---|
8088 | ( ( zw(k)-zw(k-1) ) / ( particles(n)%age - particles(n)%age_m) ) ) & |
---|
8089 | THEN |
---|
8090 | WRITE( message_string, * ) & |
---|
8091 | 'Particle violated CFL-criterion: &particle with id ', particles(n)%id, & |
---|
8092 | ' will be deleted!' |
---|
8093 | CALL message( 'lpm_check_cfl', 'PA0475', 0, 1, -1, 6, 0 ) |
---|
8094 | |
---|
8095 | particles(n)%particle_mask= .FALSE. |
---|
8096 | ENDIF |
---|
8097 | ENDIF |
---|
8098 | ENDDO |
---|
8099 | ENDDO |
---|
8100 | ENDDO |
---|
8101 | ENDDO |
---|
8102 | |
---|
8103 | END SUBROUTINE lpm_check_cfl |
---|
8104 | |
---|
8105 | |
---|
8106 | !--------------------------------------------------------------------------------------------------! |
---|
8107 | ! Description: |
---|
8108 | ! ------------ |
---|
8109 | !> If the allocated memory for the particle array does not suffice to add arriving particles from |
---|
8110 | !> neighbour grid cells, this subrouting reallocates the particle array to assure enough memory is |
---|
8111 | !> available. |
---|
8112 | !--------------------------------------------------------------------------------------------------! |
---|
8113 | SUBROUTINE realloc_particles_array ( i, j, k, size_in ) |
---|
8114 | |
---|
8115 | INTEGER(iwp), INTENT(IN) :: i !< |
---|
8116 | INTEGER(iwp), INTENT(IN) :: j !< |
---|
8117 | INTEGER(iwp), INTENT(IN) :: k !< |
---|
8118 | INTEGER(iwp), INTENT(IN), OPTIONAL :: size_in !< |
---|
8119 | |
---|
8120 | INTEGER(iwp) :: new_size !< |
---|
8121 | INTEGER(iwp) :: old_size !< |
---|
8122 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: tmp_particles_d !< |
---|
8123 | TYPE(particle_type), DIMENSION(500) :: tmp_particles_s !< |
---|
8124 | |
---|
8125 | old_size = SIZE( grid_particles(k,j,i)%particles ) |
---|
8126 | |
---|
8127 | IF ( PRESENT( size_in) ) THEN |
---|
8128 | new_size = size_in |
---|
8129 | ELSE |
---|
8130 | new_size = old_size * ( 1.0_wp + alloc_factor / 100.0_wp ) |
---|
8131 | ENDIF |
---|
8132 | |
---|
8133 | new_size = MAX( new_size, 1, old_size + 1 ) |
---|
8134 | |
---|
8135 | IF ( old_size <= 500 ) THEN |
---|
8136 | |
---|
8137 | tmp_particles_s(1:old_size) = grid_particles(k,j,i)%particles(1:old_size) |
---|
8138 | |
---|
8139 | DEALLOCATE(grid_particles(k,j,i)%particles) |
---|
8140 | ALLOCATE(grid_particles(k,j,i)%particles(new_size)) |
---|
8141 | |
---|
8142 | grid_particles(k,j,i)%particles(1:old_size) = tmp_particles_s(1:old_size) |
---|
8143 | grid_particles(k,j,i)%particles(old_size+1:new_size) = zero_particle |
---|
8144 | |
---|
8145 | ELSE |
---|
8146 | |
---|
8147 | ALLOCATE(tmp_particles_d(new_size)) |
---|
8148 | tmp_particles_d(1:old_size) = grid_particles(k,j,i)%particles |
---|
8149 | |
---|
8150 | DEALLOCATE(grid_particles(k,j,i)%particles) |
---|
8151 | ALLOCATE(grid_particles(k,j,i)%particles(new_size)) |
---|
8152 | |
---|
8153 | grid_particles(k,j,i)%particles(1:old_size) = tmp_particles_d(1:old_size) |
---|
8154 | grid_particles(k,j,i)%particles(old_size+1:new_size) = zero_particle |
---|
8155 | |
---|
8156 | DEALLOCATE(tmp_particles_d) |
---|
8157 | |
---|
8158 | ENDIF |
---|
8159 | particles => grid_particles(k,j,i)%particles(1:new_size) |
---|
8160 | |
---|
8161 | RETURN |
---|
8162 | |
---|
8163 | END SUBROUTINE realloc_particles_array |
---|
8164 | |
---|
8165 | |
---|
8166 | !--------------------------------------------------------------------------------------------------! |
---|
8167 | ! Description: |
---|
8168 | ! ------------ |
---|
8169 | !> Not needed but allocated space for particles is dealloced. |
---|
8170 | !--------------------------------------------------------------------------------------------------! |
---|
8171 | SUBROUTINE dealloc_particles_array |
---|
8172 | |
---|
8173 | |
---|
8174 | INTEGER(iwp) :: i !< |
---|
8175 | INTEGER(iwp) :: j !< |
---|
8176 | INTEGER(iwp) :: k !< |
---|
8177 | INTEGER(iwp) :: old_size !< |
---|
8178 | INTEGER(iwp) :: new_size !< |
---|
8179 | |
---|
8180 | LOGICAL :: dealloc |
---|
8181 | |
---|
8182 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: tmp_particles_d !< |
---|
8183 | TYPE(particle_type), DIMENSION(500) :: tmp_particles_s !< |
---|
8184 | |
---|
8185 | DO i = nxl, nxr |
---|
8186 | DO j = nys, nyn |
---|
8187 | DO k = nzb+1, nzt |
---|
8188 | ! |
---|
8189 | !-- Determine number of active particles |
---|
8190 | number_of_particles = prt_count(k,j,i) |
---|
8191 | ! |
---|
8192 | !-- Determine allocated memory size |
---|
8193 | old_size = SIZE( grid_particles(k,j,i)%particles ) |
---|
8194 | ! |
---|
8195 | !-- Check for large unused memory |
---|
8196 | dealloc = ( ( number_of_particles < 1 .AND. old_size > 1 ) .OR. & |
---|
8197 | ( number_of_particles > 1 .AND. & |
---|
8198 | old_size - number_of_particles * ( 1.0_wp + 0.01_wp * alloc_factor ) & |
---|
8199 | > 0.0_wp ) & |
---|
8200 | ) |
---|
8201 | |
---|
8202 | IF ( dealloc ) THEN |
---|
8203 | IF ( number_of_particles < 1 ) THEN |
---|
8204 | new_size = 1 |
---|
8205 | ELSE |
---|
8206 | new_size = INT( number_of_particles * ( 1.0_wp + 0.01_wp * alloc_factor ) ) |
---|
8207 | ENDIF |
---|
8208 | |
---|
8209 | IF ( number_of_particles <= 500 ) THEN |
---|
8210 | |
---|
8211 | tmp_particles_s(1:number_of_particles) = grid_particles(k,j,i)%particles(1:number_of_particles) |
---|
8212 | |
---|
8213 | DEALLOCATE(grid_particles(k,j,i)%particles) |
---|
8214 | ALLOCATE(grid_particles(k,j,i)%particles(new_size)) |
---|
8215 | |
---|
8216 | grid_particles(k,j,i)%particles(1:number_of_particles) = tmp_particles_s(1:number_of_particles) |
---|
8217 | grid_particles(k,j,i)%particles(number_of_particles+1:new_size) = zero_particle |
---|
8218 | |
---|
8219 | ELSE |
---|
8220 | |
---|
8221 | ALLOCATE(tmp_particles_d(number_of_particles)) |
---|
8222 | tmp_particles_d(1:number_of_particles) = grid_particles(k,j,i)%particles(1:number_of_particles) |
---|
8223 | |
---|
8224 | DEALLOCATE(grid_particles(k,j,i)%particles) |
---|
8225 | ALLOCATE(grid_particles(k,j,i)%particles(new_size)) |
---|
8226 | |
---|
8227 | grid_particles(k,j,i)%particles(1:number_of_particles) = tmp_particles_d(1:number_of_particles) |
---|
8228 | grid_particles(k,j,i)%particles(number_of_particles+1:new_size) = zero_particle |
---|
8229 | |
---|
8230 | DEALLOCATE(tmp_particles_d) |
---|
8231 | |
---|
8232 | ENDIF |
---|
8233 | |
---|
8234 | ENDIF |
---|
8235 | ENDDO |
---|
8236 | ENDDO |
---|
8237 | ENDDO |
---|
8238 | |
---|
8239 | END SUBROUTINE dealloc_particles_array |
---|
8240 | |
---|
8241 | |
---|
8242 | !--------------------------------------------------------------------------------------------------! |
---|
8243 | ! Description: |
---|
8244 | ! ----------- |
---|
8245 | !> Routine for the whole processor. |
---|
8246 | !> Sort all particles into the 8 respective subgrid boxes (in case of trilinear interpolation |
---|
8247 | !> method) and free space of particles which has been marked for deletion. |
---|
8248 | !--------------------------------------------------------------------------------------------------! |
---|
8249 | SUBROUTINE lpm_sort_and_delete |
---|
8250 | |
---|
8251 | INTEGER(iwp) :: i !< |
---|
8252 | INTEGER(iwp) :: ip !< |
---|
8253 | INTEGER(iwp) :: is !< |
---|
8254 | INTEGER(iwp) :: j !< |
---|
8255 | INTEGER(iwp) :: jp !< |
---|
8256 | INTEGER(iwp) :: kp !< |
---|
8257 | INTEGER(iwp) :: m !< |
---|
8258 | INTEGER(iwp) :: n !< |
---|
8259 | INTEGER(iwp) :: nn !< |
---|
8260 | INTEGER(iwp) :: sort_index !< |
---|
8261 | |
---|
8262 | INTEGER(iwp), DIMENSION(0:7) :: sort_count !< |
---|
8263 | |
---|
8264 | TYPE(particle_type), DIMENSION(:,:), ALLOCATABLE :: sort_particles !< |
---|
8265 | |
---|
8266 | CALL cpu_log( log_point_s(51), 'lpm_sort_and_delete', 'start' ) |
---|
8267 | IF ( interpolation_trilinear ) THEN |
---|
8268 | DO ip = nxl, nxr |
---|
8269 | DO jp = nys, nyn |
---|
8270 | DO kp = nzb+1, nzt |
---|
8271 | number_of_particles = prt_count(kp,jp,ip) |
---|
8272 | IF ( number_of_particles <= 0 ) CYCLE |
---|
8273 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
8274 | nn = 0 |
---|
8275 | sort_count = 0 |
---|
8276 | ALLOCATE( sort_particles(number_of_particles, 0:7) ) |
---|
8277 | |
---|
8278 | DO n = 1, number_of_particles |
---|
8279 | sort_index = 0 |
---|
8280 | |
---|
8281 | IF ( particles(n)%particle_mask ) THEN |
---|
8282 | nn = nn + 1 |
---|
8283 | ! |
---|
8284 | !-- Sorting particles with a binary scheme. |
---|
8285 | !-- sort_index=111_2=7_10 -> particle at the left,south,bottom subgridbox |
---|
8286 | !-- sort_index=000_2=0_10 -> particle at the right,north,top subgridbox |
---|
8287 | !-- For this the center of the gridbox is calculated. |
---|
8288 | i = (particles(n)%x + 0.5_wp * dx) * ddx |
---|
8289 | j = (particles(n)%y + 0.5_wp * dy) * ddy |
---|
8290 | |
---|
8291 | IF ( i == ip ) sort_index = sort_index + 4 |
---|
8292 | IF ( j == jp ) sort_index = sort_index + 2 |
---|
8293 | IF ( zu(kp) > particles(n)%z ) sort_index = sort_index + 1 |
---|
8294 | |
---|
8295 | sort_count(sort_index) = sort_count(sort_index) + 1 |
---|
8296 | m = sort_count(sort_index) |
---|
8297 | sort_particles(m,sort_index) = particles(n) |
---|
8298 | sort_particles(m,sort_index)%block_nr = sort_index |
---|
8299 | ENDIF |
---|
8300 | ENDDO |
---|
8301 | ! |
---|
8302 | !-- Delete and resort particles by overwritting and set the number_of_particles to |
---|
8303 | !-- the actual value. |
---|
8304 | nn = 0 |
---|
8305 | DO is = 0,7 |
---|
8306 | grid_particles(kp,jp,ip)%start_index(is) = nn + 1 |
---|
8307 | DO n = 1,sort_count(is) |
---|
8308 | nn = nn + 1 |
---|
8309 | particles(nn) = sort_particles(n,is) |
---|
8310 | ENDDO |
---|
8311 | grid_particles(kp,jp,ip)%end_index(is) = nn |
---|
8312 | ENDDO |
---|
8313 | |
---|
8314 | number_of_particles = nn |
---|
8315 | prt_count(kp,jp,ip) = number_of_particles |
---|
8316 | DEALLOCATE(sort_particles) |
---|
8317 | ENDDO |
---|
8318 | ENDDO |
---|
8319 | ENDDO |
---|
8320 | |
---|
8321 | !-- In case of the simple interpolation method the particles must not be sorted in subboxes. |
---|
8322 | !-- Particles marked for deletion however, must be deleted and number of particles must be |
---|
8323 | !-- recalculated as it is also done for the trilinear particle advection interpolation method. |
---|
8324 | ELSE |
---|
8325 | |
---|
8326 | DO ip = nxl, nxr |
---|
8327 | DO jp = nys, nyn |
---|
8328 | DO kp = nzb+1, nzt |
---|
8329 | |
---|
8330 | number_of_particles = prt_count(kp,jp,ip) |
---|
8331 | IF ( number_of_particles <= 0 ) CYCLE |
---|
8332 | particles => grid_particles(kp,jp,ip)%particles(1:number_of_particles) |
---|
8333 | ! |
---|
8334 | !-- Repack particles array, i.e. delete particles and recalculate number of particles |
---|
8335 | CALL lpm_pack |
---|
8336 | prt_count(kp,jp,ip) = number_of_particles |
---|
8337 | ENDDO |
---|
8338 | ENDDO |
---|
8339 | ENDDO |
---|
8340 | ENDIF |
---|
8341 | CALL cpu_log( log_point_s(51), 'lpm_sort_and_delete', 'stop' ) |
---|
8342 | |
---|
8343 | END SUBROUTINE lpm_sort_and_delete |
---|
8344 | |
---|
8345 | |
---|
8346 | !--------------------------------------------------------------------------------------------------! |
---|
8347 | ! Description: |
---|
8348 | ! ------------ |
---|
8349 | !> Move all particles not marked for deletion to lowest indices (packing) |
---|
8350 | !--------------------------------------------------------------------------------------------------! |
---|
8351 | SUBROUTINE lpm_pack |
---|
8352 | |
---|
8353 | INTEGER(iwp) :: n !< |
---|
8354 | INTEGER(iwp) :: nn !< |
---|
8355 | ! |
---|
8356 | !-- Find out elements marked for deletion and move data from highest index values to these free |
---|
8357 | !-- indices |
---|
8358 | nn = number_of_particles |
---|
8359 | |
---|
8360 | DO WHILE ( .NOT. particles(nn)%particle_mask ) |
---|
8361 | nn = nn-1 |
---|
8362 | IF ( nn == 0 ) EXIT |
---|
8363 | ENDDO |
---|
8364 | |
---|
8365 | IF ( nn > 0 ) THEN |
---|
8366 | DO n = 1, number_of_particles |
---|
8367 | IF ( .NOT. particles(n)%particle_mask ) THEN |
---|
8368 | particles(n) = particles(nn) |
---|
8369 | nn = nn - 1 |
---|
8370 | DO WHILE ( .NOT. particles(nn)%particle_mask ) |
---|
8371 | nn = nn-1 |
---|
8372 | IF ( n == nn ) EXIT |
---|
8373 | ENDDO |
---|
8374 | ENDIF |
---|
8375 | IF ( n == nn ) EXIT |
---|
8376 | ENDDO |
---|
8377 | ENDIF |
---|
8378 | |
---|
8379 | ! |
---|
8380 | !-- The number of deleted particles has been determined in routines lpm_boundary_conds, |
---|
8381 | !-- lpm_droplet_collision, and lpm_exchange_horiz |
---|
8382 | number_of_particles = nn |
---|
8383 | |
---|
8384 | END SUBROUTINE lpm_pack |
---|
8385 | |
---|
8386 | |
---|
8387 | !--------------------------------------------------------------------------------------------------! |
---|
8388 | ! Description: |
---|
8389 | ! ------------ |
---|
8390 | !> Sort particles in each sub-grid box into two groups: particles that already completed the LES |
---|
8391 | !> timestep, and particles that need further timestepping to complete the LES timestep. |
---|
8392 | !--------------------------------------------------------------------------------------------------! |
---|
8393 | SUBROUTINE lpm_sort_timeloop_done |
---|
8394 | |
---|
8395 | INTEGER(iwp) :: end_index !< particle end index for each sub-box |
---|
8396 | INTEGER(iwp) :: i !< index of particle grid box in x-direction |
---|
8397 | INTEGER(iwp) :: j !< index of particle grid box in y-direction |
---|
8398 | INTEGER(iwp) :: k !< index of particle grid box in z-direction |
---|
8399 | INTEGER(iwp) :: n !< running index for number of particles |
---|
8400 | INTEGER(iwp) :: nb !< index of subgrid boux |
---|
8401 | INTEGER(iwp) :: nf !< indices for particles in each sub-box that already finalized their substeps |
---|
8402 | INTEGER(iwp) :: nnf !< indices for particles in each sub-box that need further treatment |
---|
8403 | INTEGER(iwp) :: num_finalized !< number of particles in each sub-box that already finalized their substeps |
---|
8404 | INTEGER(iwp) :: start_index !< particle start index for each sub-box |
---|
8405 | |
---|
8406 | TYPE(particle_type), DIMENSION(:), ALLOCATABLE :: sort_particles !< temporary particle array |
---|
8407 | |
---|
8408 | DO i = nxl, nxr |
---|
8409 | DO j = nys, nyn |
---|
8410 | DO k = nzb+1, nzt |
---|
8411 | |
---|
8412 | number_of_particles = prt_count(k,j,i) |
---|
8413 | IF ( number_of_particles <= 0 ) CYCLE |
---|
8414 | particles => grid_particles(k,j,i)%particles(1:number_of_particles) |
---|
8415 | |
---|
8416 | DO nb = 0, 7 |
---|
8417 | ! |
---|
8418 | !-- Obtain start and end index for each subgrid box |
---|
8419 | start_index = grid_particles(k,j,i)%start_index(nb) |
---|
8420 | end_index = grid_particles(k,j,i)%end_index(nb) |
---|
8421 | ! |
---|
8422 | !-- Allocate temporary array used for sorting. |
---|
8423 | ALLOCATE( sort_particles(start_index:end_index) ) |
---|
8424 | ! |
---|
8425 | !-- Determine number of particles already completed the LES timestep, and write them |
---|
8426 | !-- into a temporary array. |
---|
8427 | nf = start_index |
---|
8428 | num_finalized = 0 |
---|
8429 | DO n = start_index, end_index |
---|
8430 | IF ( dt_3d - particles(n)%dt_sum < 1E-8_wp ) THEN |
---|
8431 | sort_particles(nf) = particles(n) |
---|
8432 | nf = nf + 1 |
---|
8433 | num_finalized = num_finalized + 1 |
---|
8434 | ENDIF |
---|
8435 | ENDDO |
---|
8436 | ! |
---|
8437 | !-- Determine number of particles that not completed the LES timestep, and write them |
---|
8438 | !-- into a temporary array. |
---|
8439 | nnf = nf |
---|
8440 | DO n = start_index, end_index |
---|
8441 | IF ( dt_3d - particles(n)%dt_sum > 1E-8_wp ) THEN |
---|
8442 | sort_particles(nnf) = particles(n) |
---|
8443 | nnf = nnf + 1 |
---|
8444 | ENDIF |
---|
8445 | ENDDO |
---|
8446 | ! |
---|
8447 | !-- Write back sorted particles |
---|
8448 | particles(start_index:end_index) = sort_particles(start_index:end_index) |
---|
8449 | ! |
---|
8450 | !-- Determine updated start_index, used to masked already |
---|
8451 | !-- completed particles. |
---|
8452 | grid_particles(k,j,i)%start_index(nb) = grid_particles(k,j,i)%start_index(nb) & |
---|
8453 | + num_finalized |
---|
8454 | ! |
---|
8455 | !-- Deallocate dummy array |
---|
8456 | DEALLOCATE ( sort_particles ) |
---|
8457 | ! |
---|
8458 | !-- Finally, if number of non-completed particles is non zero |
---|
8459 | !-- in any of the sub-boxes, set control flag appropriately. |
---|
8460 | IF ( nnf > nf ) grid_particles(k,j,i)%time_loop_done = .FALSE. |
---|
8461 | |
---|
8462 | ENDDO |
---|
8463 | ENDDO |
---|
8464 | ENDDO |
---|
8465 | ENDDO |
---|
8466 | |
---|
8467 | END SUBROUTINE lpm_sort_timeloop_done |
---|
8468 | |
---|
8469 | END MODULE lagrangian_particle_model_mod |
---|