Home

Context Navigation

source: palm/trunk/SOURCE/pres.f90 @ 2232

Last change on this file since 2232 was 2232, checked in by suehring, 7 years ago
Adjustments according new topography and surface-modelling concept implemented
Property svn:keywords set to `Id`
File size: 30.5 KB

Rev	Line
[1682]	1	!> @file pres.f90
[2000]	2	!------------------------------------------------------------------------------!
[1036]	3	! This file is part of PALM.
	4	!
[2000]	5	! PALM is free software: you can redistribute it and/or modify it under the
	6	! terms of the GNU General Public License as published by the Free Software
	7	! Foundation, either version 3 of the License, or (at your option) any later
	8	! version.
[1036]	9	!
	10	! PALM is distributed in the hope that it will be useful, but WITHOUT ANY
	11	! WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
	12	! A PARTICULAR PURPOSE. See the GNU General Public License for more details.
	13	!
	14	! You should have received a copy of the GNU General Public License along with
	15	! PALM. If not, see <http://www.gnu.org/licenses/>.
	16	!
[2101]	17	! Copyright 1997-2017 Leibniz Universitaet Hannover
[2000]	18	!------------------------------------------------------------------------------!
[1036]	19	!
[484]	20	! Current revisions:
[1212]	21	! ------------------
[2232]	22	! Adjustments to new topography and surface concept
[1932]	23	!
[1919]	24	! Former revisions:
	25	! -----------------
	26	! $Id: pres.f90 2232 2017-05-30 17:47:52Z suehring $
	27	!
[2119]	28	! 2118 2017-01-17 16:38:49Z raasch
	29	! OpenACC directives and related code removed
	30	!
[2074]	31	! 2073 2016-11-30 14:34:05Z raasch
	32	! openmp bugfix for calculation of new divergence
	33	!
[2038]	34	! 2037 2016-10-26 11:15:40Z knoop
	35	! Anelastic approximation implemented
	36	!
[2001]	37	! 2000 2016-08-20 18:09:15Z knoop
	38	! Forced header and separation lines into 80 columns
	39	!
[1933]	40	! 1932 2016-06-10 12:09:21Z suehring
	41	! Initial version of purely vertical nesting introduced.
	42	!
[1932]	43	! 1931 2016-06-10 12:06:59Z suehring
	44	! Rename multigrid into multigrid_noopt and multigrid_fast into multigrid
	45	!
[1930]	46	! 1929 2016-06-09 16:25:25Z suehring
	47	! Bugfix: weight_substep for initial call, replace by local variable
	48	!
[1919]	49	! 1918 2016-05-27 14:35:57Z raasch
[1918]	50	! sum of divergence is also calculated when pres is called before the initial
	51	! first time step,
	52	! stearing is modified by using intermediate_timestep_count = 0 in order to
	53	! determine if pres is called before the first initial timestep,
	54	! bugfix: in case of Neumann conditions for pressure both at bottom and top,
	55	! mean vertical velocity is also removed for the first time step
	56	! bugfix for calculating divergences
[1321]	57	!
[1909]	58	! 1908 2016-05-25 17:22:32Z suehring
	59	! New divergence for output into RUN_CONTROL file is calculated only at last
	60	! Runge-Kutta step
	61	!
[1846]	62	! 1845 2016-04-08 08:29:13Z raasch
	63	! nzb_2d replace by nzb_u\|v_inner
	64	!
[1800]	65	! 1799 2016-04-05 08:35:55Z gronemeier
	66	! Bugfix: excluded third dimension from horizontal volume flow calculation
	67	!
[1763]	68	! 1762 2016-02-25 12:31:13Z hellstea
	69	! Introduction of nested domain feature
	70	!
[1683]	71	! 1682 2015-10-07 23:56:08Z knoop
	72	! Code annotations made doxygen readable
	73	!
[1576]	74	! 1575 2015-03-27 09:56:27Z raasch
	75	! poismg_fast + respective module added, adjustments for psolver-queries
	76	!
[1343]	77	! 1342 2014-03-26 17:04:47Z kanani
	78	! REAL constants defined as wp-kind
	79	!
[1321]	80	! 1320 2014-03-20 08:40:49Z raasch
[1320]	81	! ONLY-attribute added to USE-statements,
	82	! kind-parameters added to all INTEGER and REAL declaration statements,
	83	! kinds are defined in new module kinds,
	84	! old module precision_kind is removed,
	85	! revision history before 2012 removed,
	86	! comment fields (!:) to be used for variable explanations added to
	87	! all variable declaration statements
[708]	88	!
[1319]	89	! 1318 2014-03-17 13:35:16Z raasch
	90	! module interfaces removed
	91	!
[1307]	92	! 1306 2014-03-13 14:30:59Z raasch
	93	! second argument removed from routine poisfft
	94	!
[1258]	95	! 1257 2013-11-08 15:18:40Z raasch
	96	! openacc loop and loop vector clauses removed, independent clauses added,
	97	! end parallel replaced by end parallel loop
	98	!
[1222]	99	! 1221 2013-09-10 08:59:13Z raasch
	100	! openACC porting of reduction operations, loops for calculating d are
	101	! using the rflags_s_inner multiply flag instead of the nzb_s_inner loop index
	102	!
[1213]	103	! 1212 2013-08-15 08:46:27Z raasch
	104	! call of poisfft_hybrid removed
	105	!
[1118]	106	! 1117 2013-03-27 11:15:36Z suehring
	107	! Bugfix in OpenMP parallelization.
	108	!
[1114]	109	! 1113 2013-03-10 02:48:14Z raasch
	110	! GPU-porting of several loops, some loops rearranged
	111	!
[1112]	112	! 1111 2013-03-08 23:54:10Z
	113	! openACC statements added,
	114	! ibc_p_b = 2 removed
	115	!
[1093]	116	! 1092 2013-02-02 11:24:22Z raasch
	117	! unused variables removed
	118	!
[1037]	119	! 1036 2012-10-22 13:43:42Z raasch
	120	! code put under GPL (PALM 3.9)
	121	!
[1004]	122	! 1003 2012-09-14 14:35:53Z raasch
	123	! adjustment of array tend for cases with unequal subdomain sizes removed
	124	!
[1]	125	! Revision 1.1 1997/07/24 11:24:44 raasch
	126	! Initial revision
	127	!
	128	!
	129	! Description:
	130	! ------------
[1682]	131	!> Compute the divergence of the provisional velocity field. Solve the Poisson
	132	!> equation for the perturbation pressure. Compute the final velocities using
	133	!> this perturbation pressure. Compute the remaining divergence.
[1]	134	!------------------------------------------------------------------------------!
[1682]	135	SUBROUTINE pres
	136
[1]	137
[1320]	138	USE arrays_3d, &
[2037]	139	ONLY: d, ddzu, ddzu_pres, ddzw, dzw, p, p_loc, rho_air, rho_air_zw, &
	140	tend, u, v, w
[1320]	141
	142	USE control_parameters, &
	143	ONLY: bc_lr_cyc, bc_ns_cyc, conserve_volume_flow, dt_3d, &
	144	gathered_size, ibc_p_b, ibc_p_t, intermediate_timestep_count, &
[1908]	145	intermediate_timestep_count_max, mg_switch_to_pe0_level, &
[2118]	146	nest_domain, outflow_l, outflow_n, outflow_r, &
[1918]	147	outflow_s, psolver, subdomain_size, topography, volume_flow, &
	148	volume_flow_area, volume_flow_initial
[1320]	149
	150	USE cpulog, &
	151	ONLY: cpu_log, log_point, log_point_s
	152
	153	USE grid_variables, &
	154	ONLY: ddx, ddy
	155
	156	USE indices, &
	157	ONLY: nbgp, ngp_2dh_outer, nx, nxl, nxlg, nxl_mg, nxr, nxrg, nxr_mg, &
[2232]	158	ny, nys, nysg, nys_mg, nyn, nyng, nyn_mg, nzb, nzt, nzt_mg, &
	159	wall_flags_0
[1320]	160
	161	USE kinds
	162
[1]	163	USE pegrid
[1933]	164
	165	USE pmc_interface, &
	166	ONLY: nesting_mode
[1]	167
[1320]	168	USE poisfft_mod, &
	169	ONLY: poisfft
	170
[1575]	171	USE poismg_mod
	172
[1320]	173	USE statistics, &
	174	ONLY: statistic_regions, sums_divnew_l, sums_divold_l, weight_pres, &
	175	weight_substep
	176
[2232]	177	USE surface_mod, &
	178	ONLY : bc_h
	179
[1]	180	IMPLICIT NONE
	181
[1682]	182	INTEGER(iwp) :: i !<
	183	INTEGER(iwp) :: j !<
	184	INTEGER(iwp) :: k !<
[2232]	185	INTEGER(iwp) :: m !<
[1]	186
[1682]	187	REAL(wp) :: ddt_3d !<
[1918]	188	REAL(wp) :: d_weight_pres !<
[1682]	189	REAL(wp) :: localsum !<
	190	REAL(wp) :: threadsum !<
[1918]	191	REAL(wp) :: weight_pres_l !<
[1929]	192	REAL(wp) :: weight_substep_l !<
[1]	193
[1762]	194	REAL(wp), DIMENSION(1:3) :: volume_flow_l !<
	195	REAL(wp), DIMENSION(1:3) :: volume_flow_offset !<
[1682]	196	REAL(wp), DIMENSION(1:nzt) :: w_l !<
	197	REAL(wp), DIMENSION(1:nzt) :: w_l_l !<
[1]	198
[1933]	199	LOGICAL :: nest_domain_nvn !<
[1]	200
[1933]	201
[1]	202	CALL cpu_log( log_point(8), 'pres', 'start' )
	203
[85]	204
[1918]	205	!
	206	!-- Calculate quantities to be used locally
[1342]	207	ddt_3d = 1.0_wp / dt_3d
[1918]	208	IF ( intermediate_timestep_count == 0 ) THEN
	209	!
	210	!-- If pres is called before initial time step
[1929]	211	weight_pres_l = 1.0_wp
	212	d_weight_pres = 1.0_wp
	213	weight_substep_l = 1.0_wp
[1918]	214	ELSE
[1929]	215	weight_pres_l = weight_pres(intermediate_timestep_count)
	216	d_weight_pres = 1.0_wp / weight_pres(intermediate_timestep_count)
	217	weight_substep_l = weight_substep(intermediate_timestep_count)
[1918]	218	ENDIF
[85]	219
[1]	220	!
[707]	221	!-- Multigrid method expects array d to have one ghost layer.
	222	!--
[1575]	223	IF ( psolver(1:9) == 'multigrid' ) THEN
[667]	224
[1]	225	DEALLOCATE( d )
[667]	226	ALLOCATE( d(nzb:nzt+1,nys-1:nyn+1,nxl-1:nxr+1) )
[707]	227
	228	!
	229	!-- Since p is later used to hold the weighted average of the substeps, it
	230	!-- cannot be used in the iterative solver. Therefore, its initial value is
	231	!-- stored on p_loc, which is then iteratively advanced in every substep.
[1762]	232	IF ( intermediate_timestep_count <= 1 ) THEN
[707]	233	DO i = nxl-1, nxr+1
	234	DO j = nys-1, nyn+1
	235	DO k = nzb, nzt+1
	236	p_loc(k,j,i) = p(k,j,i)
	237	ENDDO
	238	ENDDO
	239	ENDDO
[667]	240	ENDIF
	241
[1918]	242	ELSEIF ( psolver == 'sor' .AND. intermediate_timestep_count <= 1 ) THEN
[707]	243
	244	!
	245	!-- Since p is later used to hold the weighted average of the substeps, it
	246	!-- cannot be used in the iterative solver. Therefore, its initial value is
	247	!-- stored on p_loc, which is then iteratively advanced in every substep.
	248	p_loc = p
	249
[1]	250	ENDIF
	251
	252	!
[75]	253	!-- Conserve the volume flow at the outflow in case of non-cyclic lateral
	254	!-- boundary conditions
[106]	255	!-- WARNING: so far, this conservation does not work at the left/south
	256	!-- boundary if the topography at the inflow differs from that at the
	257	!-- outflow! For this case, volume_flow_area needs adjustment!
	258	!
	259	!-- Left/right
[709]	260	IF ( conserve_volume_flow .AND. ( outflow_l .OR. outflow_r ) ) THEN
[680]	261
[1342]	262	volume_flow(1) = 0.0_wp
	263	volume_flow_l(1) = 0.0_wp
[106]	264
	265	IF ( outflow_l ) THEN
	266	i = 0
	267	ELSEIF ( outflow_r ) THEN
	268	i = nx+1
	269	ENDIF
	270
	271	DO j = nys, nyn
	272	!
	273	!-- Sum up the volume flow through the south/north boundary
[2232]	274	DO k = nzb+1, nzt
	275	volume_flow_l(1) = volume_flow_l(1) + u(k,j,i) * dzw(k) &
	276	* MERGE( 1.0_wp, 0.0_wp, &
	277	BTEST( wall_flags_0(k,j,i), 1 ) &
	278	)
[106]	279	ENDDO
	280	ENDDO
	281
	282	#if defined( __parallel )
[680]	283	IF ( collective_wait ) CALL MPI_BARRIER( comm1dy, ierr )
[106]	284	CALL MPI_ALLREDUCE( volume_flow_l(1), volume_flow(1), 1, MPI_REAL, &
	285	MPI_SUM, comm1dy, ierr )
	286	#else
	287	volume_flow = volume_flow_l
	288	#endif
[709]	289	volume_flow_offset(1) = ( volume_flow_initial(1) - volume_flow(1) ) &
[106]	290	/ volume_flow_area(1)
	291
[667]	292	DO j = nysg, nyng
[2232]	293	DO k = nzb+1, nzt
	294	u(k,j,i) = u(k,j,i) + volume_flow_offset(1) &
	295	* MERGE( 1.0_wp, 0.0_wp, &
	296	BTEST( wall_flags_0(k,j,i), 1 ) &
	297	)
[106]	298	ENDDO
	299	ENDDO
	300
	301	ENDIF
	302
	303	!
	304	!-- South/north
[709]	305	IF ( conserve_volume_flow .AND. ( outflow_n .OR. outflow_s ) ) THEN
[106]	306
[1342]	307	volume_flow(2) = 0.0_wp
	308	volume_flow_l(2) = 0.0_wp
[75]	309
[106]	310	IF ( outflow_s ) THEN
	311	j = 0
	312	ELSEIF ( outflow_n ) THEN
[75]	313	j = ny+1
[106]	314	ENDIF
	315
	316	DO i = nxl, nxr
[75]	317	!
[106]	318	!-- Sum up the volume flow through the south/north boundary
[2232]	319	DO k = nzb+1, nzt
	320	volume_flow_l(2) = volume_flow_l(2) + v(k,j,i) * dzw(k) &
	321	* MERGE( 1.0_wp, 0.0_wp, &
	322	BTEST( wall_flags_0(k,j,i), 2 ) &
	323	)
[75]	324	ENDDO
[106]	325	ENDDO
	326
[75]	327	#if defined( __parallel )
[680]	328	IF ( collective_wait ) CALL MPI_BARRIER( comm1dx, ierr )
[75]	329	CALL MPI_ALLREDUCE( volume_flow_l(2), volume_flow(2), 1, MPI_REAL, &
	330	MPI_SUM, comm1dx, ierr )
	331	#else
	332	volume_flow = volume_flow_l
	333	#endif
	334	volume_flow_offset(2) = ( volume_flow_initial(2) - volume_flow(2) ) &
[106]	335	/ volume_flow_area(2)
[75]	336
[667]	337	DO i = nxlg, nxrg
[2232]	338	DO k = nzb+1, nzt
	339	v(k,j,i) = v(k,j,i) + volume_flow_offset(2) &
	340	* MERGE( 1.0_wp, 0.0_wp, &
	341	BTEST( wall_flags_0(k,j,i), 2 ) &
	342	)
[75]	343	ENDDO
[106]	344	ENDDO
[75]	345
	346	ENDIF
	347
[76]	348	!
[1762]	349	!-- Remove mean vertical velocity in case that Neumann conditions are
[1933]	350	!-- used both at bottom and top boundary, and if not a nested domain in a
	351	!-- normal nesting run. In case of vertical nesting, this must be done.
	352	!-- Therefore an auxiliary logical variable nest_domain_nvn is used here, and
	353	!-- nvn stands for non-vertical nesting.
[1918]	354	!-- This cannot be done before the first initial time step because ngp_2dh_outer
	355	!-- is not yet known then.
[1933]	356	nest_domain_nvn = nest_domain
	357	IF ( nest_domain .AND. nesting_mode == 'vertical' ) THEN
	358	nest_domain_nvn = .FALSE.
	359	ENDIF
	360
	361	IF ( ibc_p_b == 1 .AND. ibc_p_t == 1 .AND. &
	362	.NOT. nest_domain_nvn .AND. intermediate_timestep_count /= 0 ) &
[1918]	363	THEN
	364	w_l = 0.0_wp; w_l_l = 0.0_wp
	365	DO i = nxl, nxr
	366	DO j = nys, nyn
[2232]	367	DO k = nzb+1, nzt
	368	w_l_l(k) = w_l_l(k) + w(k,j,i) &
	369	* MERGE( 1.0_wp, 0.0_wp, &
	370	BTEST( wall_flags_0(k,j,i), 3 ) &
	371	)
[76]	372	ENDDO
	373	ENDDO
[1918]	374	ENDDO
[76]	375	#if defined( __parallel )
[1918]	376	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
	377	CALL MPI_ALLREDUCE( w_l_l(1), w_l(1), nzt, MPI_REAL, MPI_SUM, &
	378	comm2d, ierr )
[76]	379	#else
[1918]	380	w_l = w_l_l
[76]	381	#endif
[1918]	382	DO k = 1, nzt
	383	w_l(k) = w_l(k) / ngp_2dh_outer(k,0)
	384	ENDDO
	385	DO i = nxlg, nxrg
	386	DO j = nysg, nyng
[2232]	387	DO k = nzb+1, nzt
	388	w(k,j,i) = w(k,j,i) - w_l(k) &
	389	* MERGE( 1.0_wp, 0.0_wp, &
	390	BTEST( wall_flags_0(k,j,i), 3 ) &
	391	)
[76]	392	ENDDO
	393	ENDDO
[1918]	394	ENDDO
[76]	395	ENDIF
[75]	396
	397	!
[1]	398	!-- Compute the divergence of the provisional velocity field.
	399	CALL cpu_log( log_point_s(1), 'divergence', 'start' )
	400
[1575]	401	IF ( psolver(1:9) == 'multigrid' ) THEN
[2232]	402	!$OMP PARALLEL DO SCHEDULE( STATIC ) PRIVATE (i,j,k)
[1]	403	DO i = nxl-1, nxr+1
	404	DO j = nys-1, nyn+1
	405	DO k = nzb, nzt+1
[1342]	406	d(k,j,i) = 0.0_wp
[1]	407	ENDDO
	408	ENDDO
	409	ENDDO
	410	ELSE
[2232]	411	!$OMP PARALLEL DO SCHEDULE( STATIC ) PRIVATE (i,j,k)
[1003]	412	DO i = nxl, nxr
	413	DO j = nys, nyn
	414	DO k = nzb+1, nzt
[1342]	415	d(k,j,i) = 0.0_wp
[1]	416	ENDDO
	417	ENDDO
	418	ENDDO
	419	ENDIF
	420
[1342]	421	localsum = 0.0_wp
	422	threadsum = 0.0_wp
[1]	423
	424	#if defined( __ibm )
	425	!$OMP PARALLEL PRIVATE (i,j,k) FIRSTPRIVATE(threadsum) REDUCTION(+:localsum)
	426	!$OMP DO SCHEDULE( STATIC )
	427	DO i = nxl, nxr
	428	DO j = nys, nyn
[2232]	429	DO k = nzb+1, nzt
[2037]	430	d(k,j,i) = ( ( u(k,j,i+1) - u(k,j,i) ) * rho_air(k) * ddx + &
	431	( v(k,j+1,i) - v(k,j,i) ) * rho_air(k) * ddy + &
	432	( w(k,j,i) * rho_air_zw(k) - &
	433	w(k-1,j,i) * rho_air_zw(k-1) ) * ddzw(k) &
[2232]	434	) * ddt_3d * d_weight_pres &
	435	* MERGE( 1.0_wp, 0.0_wp, &
	436	BTEST( wall_flags_0(k,j,i), 0 ) &
	437	)
[1]	438	ENDDO
	439	!
	440	!-- Compute possible PE-sum of divergences for flow_statistics
[2232]	441	DO k = nzb+1, nzt
	442	threadsum = threadsum + ABS( d(k,j,i) ) &
	443	* MERGE( 1.0_wp, 0.0_wp, &
	444	BTEST( wall_flags_0(k,j,i), 0 ) &
	445	)
[1]	446	ENDDO
	447
	448	ENDDO
	449	ENDDO
	450
[1918]	451	IF ( intermediate_timestep_count == intermediate_timestep_count_max .OR. &
	452	intermediate_timestep_count == 0 ) THEN
	453	localsum = localsum + threadsum * dt_3d * weight_pres_l
[1908]	454	ENDIF
[1]	455	!$OMP END PARALLEL
	456	#else
	457
[1111]	458	!$OMP PARALLEL PRIVATE (i,j,k)
	459	!$OMP DO SCHEDULE( STATIC )
	460	DO i = nxl, nxr
	461	DO j = nys, nyn
	462	DO k = 1, nzt
[2037]	463	d(k,j,i) = ( ( u(k,j,i+1) - u(k,j,i) ) * rho_air(k) * ddx + &
	464	( v(k,j+1,i) - v(k,j,i) ) * rho_air(k) * ddy + &
	465	( w(k,j,i) * rho_air_zw(k) - &
	466	w(k-1,j,i) * rho_air_zw(k-1) ) * ddzw(k) &
[2232]	467	) * ddt_3d * d_weight_pres &
	468	* MERGE( 1.0_wp, 0.0_wp, &
	469	BTEST( wall_flags_0(k,j,i), 0 ) &
	470	)
[1]	471	ENDDO
	472	ENDDO
[1111]	473	ENDDO
	474	!$OMP END PARALLEL
[1]	475
	476	!
[1908]	477	!-- Compute possible PE-sum of divergences for flow_statistics. Carry out
	478	!-- computation only at last Runge-Kutta substep.
[1918]	479	IF ( intermediate_timestep_count == intermediate_timestep_count_max .OR. &
	480	intermediate_timestep_count == 0 ) THEN
[1908]	481	!$OMP PARALLEL PRIVATE (i,j,k) FIRSTPRIVATE(threadsum) REDUCTION(+:localsum)
	482	!$OMP DO SCHEDULE( STATIC )
	483	DO i = nxl, nxr
	484	DO j = nys, nyn
	485	DO k = nzb+1, nzt
	486	threadsum = threadsum + ABS( d(k,j,i) )
	487	ENDDO
[1]	488	ENDDO
	489	ENDDO
[1918]	490	localsum = localsum + threadsum * dt_3d * weight_pres_l
[1908]	491	!$OMP END PARALLEL
	492	ENDIF
[1]	493	#endif
	494
	495	!
	496	!-- For completeness, set the divergence sum of all statistic regions to those
	497	!-- of the total domain
[1918]	498	IF ( intermediate_timestep_count == intermediate_timestep_count_max .OR. &
	499	intermediate_timestep_count == 0 ) THEN
[1908]	500	sums_divold_l(0:statistic_regions) = localsum
[1918]	501	ENDIF
[1]	502
	503	CALL cpu_log( log_point_s(1), 'divergence', 'stop' )
	504
	505	!
	506	!-- Compute the pressure perturbation solving the Poisson equation
[1575]	507	IF ( psolver == 'poisfft' ) THEN
[1]	508
	509	!
	510	!-- Solve Poisson equation via FFT and solution of tridiagonal matrices
[1575]	511	CALL poisfft( d )
[1212]	512
[1]	513	!
	514	!-- Store computed perturbation pressure and set boundary condition in
	515	!-- z-direction
[2232]	516	!$OMP PARALLEL DO PRIVATE (i,j,k)
[1]	517	DO i = nxl, nxr
	518	DO j = nys, nyn
	519	DO k = nzb+1, nzt
	520	tend(k,j,i) = d(k,j,i)
	521	ENDDO
	522	ENDDO
	523	ENDDO
	524
	525	!
	526	!-- Bottom boundary:
	527	!-- This condition is only required for internal output. The pressure
	528	!-- gradient (dp(nzb+1)-dp(nzb))/dz is not used anywhere else.
	529	IF ( ibc_p_b == 1 ) THEN
	530	!
[2232]	531	!-- Neumann (dp/dz = 0). Using surfae data type, first for non-natural
	532	!-- surfaces, then for natural and urban surfaces
	533	!-- Upward facing
	534	!$OMP PARALLEL DO PRIVATE( i, j, k )
	535	DO m = 1, bc_h(0)%ns
	536	i = bc_h(0)%i(m)
	537	j = bc_h(0)%j(m)
	538	k = bc_h(0)%k(m)
	539	tend(k-1,j,i) = tend(k,j,i)
[1]	540	ENDDO
[2232]	541	!
	542	!-- Downward facing
	543	!$OMP PARALLEL DO PRIVATE( i, j, k )
	544	DO m = 1, bc_h(1)%ns
	545	i = bc_h(1)%i(m)
	546	j = bc_h(1)%j(m)
	547	k = bc_h(1)%k(m)
	548	tend(k+1,j,i) = tend(k,j,i)
	549	ENDDO
[1]	550
	551	ELSE
	552	!
[2232]	553	!-- Dirichlet. Using surface data type, first for non-natural
	554	!-- surfaces, then for natural and urban surfaces
	555	!-- Upward facing
	556	!$OMP PARALLEL DO PRIVATE( i, j, k )
	557	DO m = 1, bc_h(0)%ns
	558	i = bc_h(0)%i(m)
	559	j = bc_h(0)%j(m)
	560	k = bc_h(0)%k(m)
	561	tend(k-1,j,i) = 0.0_wp
[1]	562	ENDDO
[2232]	563	!
	564	!-- Downward facing
	565	!$OMP PARALLEL DO PRIVATE( i, j, k )
	566	DO m = 1, bc_h(1)%ns
	567	i = bc_h(1)%i(m)
	568	j = bc_h(1)%j(m)
	569	k = bc_h(1)%k(m)
	570	tend(k+1,j,i) = 0.0_wp
	571	ENDDO
[1]	572
	573	ENDIF
	574
	575	!
	576	!-- Top boundary
	577	IF ( ibc_p_t == 1 ) THEN
	578	!
	579	!-- Neumann
[2232]	580	!$OMP PARALLEL DO PRIVATE (i,j,k)
[667]	581	DO i = nxlg, nxrg
	582	DO j = nysg, nyng
[1]	583	tend(nzt+1,j,i) = tend(nzt,j,i)
	584	ENDDO
	585	ENDDO
	586
	587	ELSE
	588	!
	589	!-- Dirichlet
[2232]	590	!$OMP PARALLEL DO PRIVATE (i,j,k)
[667]	591	DO i = nxlg, nxrg
	592	DO j = nysg, nyng
[1342]	593	tend(nzt+1,j,i) = 0.0_wp
[1]	594	ENDDO
	595	ENDDO
	596
	597	ENDIF
	598
	599	!
	600	!-- Exchange boundaries for p
[667]	601	CALL exchange_horiz( tend, nbgp )
[1]	602
	603	ELSEIF ( psolver == 'sor' ) THEN
	604
	605	!
	606	!-- Solve Poisson equation for perturbation pressure using SOR-Red/Black
	607	!-- scheme
[707]	608	CALL sor( d, ddzu_pres, ddzw, p_loc )
	609	tend = p_loc
[1]	610
[1575]	611	ELSEIF ( psolver(1:9) == 'multigrid' ) THEN
[1]	612
	613	!
	614	!-- Solve Poisson equation for perturbation pressure using Multigrid scheme,
[667]	615	!-- array tend is used to store the residuals, logical exchange_mg is used
	616	!-- to discern data exchange in multigrid ( 1 ghostpoint ) and normal grid
	617	!-- ( nbgp ghost points ).
[778]	618
	619	!-- If the number of grid points of the gathered grid, which is collected
	620	!-- on PE0, is larger than the number of grid points of an PE, than array
	621	!-- tend will be enlarged.
	622	IF ( gathered_size > subdomain_size ) THEN
	623	DEALLOCATE( tend )
	624	ALLOCATE( tend(nzb:nzt_mg(mg_switch_to_pe0_level)+1,nys_mg( &
	625	mg_switch_to_pe0_level)-1:nyn_mg(mg_switch_to_pe0_level)+1,&
	626	nxl_mg(mg_switch_to_pe0_level)-1:nxr_mg( &
	627	mg_switch_to_pe0_level)+1) )
	628	ENDIF
	629
[1575]	630	IF ( psolver == 'multigrid' ) THEN
	631	CALL poismg( tend )
	632	ELSE
[1931]	633	CALL poismg_noopt( tend )
[1575]	634	ENDIF
[707]	635
[778]	636	IF ( gathered_size > subdomain_size ) THEN
	637	DEALLOCATE( tend )
	638	ALLOCATE( tend(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
	639	ENDIF
	640
[1]	641	!
	642	!-- Restore perturbation pressure on tend because this array is used
	643	!-- further below to correct the velocity fields
[707]	644	DO i = nxl-1, nxr+1
	645	DO j = nys-1, nyn+1
	646	DO k = nzb, nzt+1
	647	tend(k,j,i) = p_loc(k,j,i)
	648	ENDDO
	649	ENDDO
	650	ENDDO
[667]	651
[1]	652	ENDIF
	653
	654	!
[707]	655	!-- Store perturbation pressure on array p, used for pressure data output.
	656	!-- Ghost layers are added in the output routines (except sor-method: see below)
[1918]	657	IF ( intermediate_timestep_count <= 1 ) THEN
[707]	658	!$OMP PARALLEL PRIVATE (i,j,k)
	659	!$OMP DO
	660	DO i = nxl-1, nxr+1
	661	DO j = nys-1, nyn+1
	662	DO k = nzb, nzt+1
	663	p(k,j,i) = tend(k,j,i) * &
[1929]	664	weight_substep_l
[673]	665	ENDDO
[1]	666	ENDDO
[707]	667	ENDDO
	668	!$OMP END PARALLEL
	669
[1762]	670	ELSEIF ( intermediate_timestep_count > 1 ) THEN
[707]	671	!$OMP PARALLEL PRIVATE (i,j,k)
	672	!$OMP DO
	673	DO i = nxl-1, nxr+1
	674	DO j = nys-1, nyn+1
	675	DO k = nzb, nzt+1
	676	p(k,j,i) = p(k,j,i) + tend(k,j,i) * &
[1929]	677	weight_substep_l
[673]	678	ENDDO
	679	ENDDO
[707]	680	ENDDO
	681	!$OMP END PARALLEL
	682
	683	ENDIF
[673]	684
[707]	685	!
	686	!-- SOR-method needs ghost layers for the next timestep
	687	IF ( psolver == 'sor' ) CALL exchange_horiz( p, nbgp )
[682]	688
[1]	689	!
	690	!-- Correction of the provisional velocities with the current perturbation
	691	!-- pressure just computed
[709]	692	IF ( conserve_volume_flow .AND. ( bc_lr_cyc .OR. bc_ns_cyc ) ) THEN
[1342]	693	volume_flow_l(1) = 0.0_wp
	694	volume_flow_l(2) = 0.0_wp
[1]	695	ENDIF
[707]	696
[1]	697	!$OMP PARALLEL PRIVATE (i,j,k)
	698	!$OMP DO
[673]	699	DO i = nxl, nxr
[1]	700	DO j = nys, nyn
[2118]	701
[2232]	702	DO k = nzb+1, nzt
	703	w(k,j,i) = w(k,j,i) - dt_3d * &
	704	( tend(k+1,j,i) - tend(k,j,i) ) * ddzu(k+1) &
	705	* weight_pres_l &
	706	* MERGE( 1.0_wp, 0.0_wp, &
	707	BTEST( wall_flags_0(k,j,i), 3 ) &
	708	)
[1]	709	ENDDO
[2118]	710
[2232]	711	DO k = nzb+1, nzt
	712	u(k,j,i) = u(k,j,i) - dt_3d * &
	713	( tend(k,j,i) - tend(k,j,i-1) ) * ddx &
	714	* weight_pres_l &
	715	* MERGE( 1.0_wp, 0.0_wp, &
	716	BTEST( wall_flags_0(k,j,i), 1 ) &
	717	)
[1]	718	ENDDO
[2118]	719
[2232]	720	DO k = nzb+1, nzt
	721	v(k,j,i) = v(k,j,i) - dt_3d * &
	722	( tend(k,j,i) - tend(k,j-1,i) ) * ddy &
	723	* weight_pres_l &
	724	* MERGE( 1.0_wp, 0.0_wp, &
	725	BTEST( wall_flags_0(k,j,i), 2 ) &
	726	)
[673]	727	ENDDO
[1]	728
	729	ENDDO
	730	ENDDO
	731	!$OMP END PARALLEL
[1113]	732
	733	!
	734	!-- Sum up the volume flow through the right and north boundary
[2232]	735	IF ( conserve_volume_flow .AND. bc_lr_cyc .AND. bc_ns_cyc .AND. &
[1113]	736	nxr == nx ) THEN
	737
	738	!$OMP PARALLEL PRIVATE (j,k)
	739	!$OMP DO
	740	DO j = nys, nyn
	741	!$OMP CRITICAL
[2232]	742	DO k = nzb+1, nzt
	743	volume_flow_l(1) = volume_flow_l(1) + u(k,j,nxr) * dzw(k) &
	744	* MERGE( 1.0_wp, 0.0_wp, &
	745	BTEST( wall_flags_0(k,j,nxr), 1 )&
	746	)
[1113]	747	ENDDO
	748	!$OMP END CRITICAL
	749	ENDDO
	750	!$OMP END PARALLEL
	751
	752	ENDIF
	753
[2232]	754	IF ( conserve_volume_flow .AND. bc_ns_cyc .AND. bc_lr_cyc .AND. &
[1113]	755	nyn == ny ) THEN
	756
	757	!$OMP PARALLEL PRIVATE (i,k)
	758	!$OMP DO
	759	DO i = nxl, nxr
	760	!$OMP CRITICAL
[2232]	761	DO k = nzb+1, nzt
	762	volume_flow_l(2) = volume_flow_l(2) + v(k,nyn,i) * dzw(k) &
	763	* MERGE( 1.0_wp, 0.0_wp, &
	764	BTEST( wall_flags_0(k,nyn,i), 2 )&
	765	)
[1113]	766	ENDDO
	767	!$OMP END CRITICAL
	768	ENDDO
	769	!$OMP END PARALLEL
	770
	771	ENDIF
[673]	772
[1]	773	!
	774	!-- Conserve the volume flow
[707]	775	IF ( conserve_volume_flow .AND. ( bc_lr_cyc .AND. bc_ns_cyc ) ) THEN
[1]	776
	777	#if defined( __parallel )
[622]	778	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
[1]	779	CALL MPI_ALLREDUCE( volume_flow_l(1), volume_flow(1), 2, MPI_REAL, &
	780	MPI_SUM, comm2d, ierr )
	781	#else
	782	volume_flow = volume_flow_l
	783	#endif
	784
[1799]	785	volume_flow_offset(1:2) = ( volume_flow_initial(1:2) - volume_flow(1:2) ) / &
	786	volume_flow_area(1:2)
[1]	787
	788	!$OMP PARALLEL PRIVATE (i,j,k)
	789	!$OMP DO
	790	DO i = nxl, nxr
	791	DO j = nys, nyn
[2232]	792	DO k = nzb+1, nzt
	793	u(k,j,i) = u(k,j,i) + volume_flow_offset(1) &
	794	* MERGE( 1.0_wp, 0.0_wp, &
	795	BTEST( wall_flags_0(k,j,i), 1 ) &
	796	)
[719]	797	ENDDO
[2232]	798	DO k = nzb+1, nzt
	799	v(k,j,i) = v(k,j,i) + volume_flow_offset(2) &
	800	* MERGE( 1.0_wp, 0.0_wp, &
	801	BTEST( wall_flags_0(k,j,i), 2 ) &
	802	)
[667]	803	ENDDO
[1]	804	ENDDO
	805	ENDDO
[667]	806
[1]	807	!$OMP END PARALLEL
	808
	809	ENDIF
	810
	811	!
	812	!-- Exchange of boundaries for the velocities
[667]	813	CALL exchange_horiz( u, nbgp )
	814	CALL exchange_horiz( v, nbgp )
	815	CALL exchange_horiz( w, nbgp )
[1]	816
	817	!
	818	!-- Compute the divergence of the corrected velocity field,
[1908]	819	!-- A possible PE-sum is computed in flow_statistics. Carry out computation
	820	!-- only at last Runge-Kutta step.
[1918]	821	IF ( intermediate_timestep_count == intermediate_timestep_count_max .OR. &
	822	intermediate_timestep_count == 0 ) THEN
[1908]	823	CALL cpu_log( log_point_s(1), 'divergence', 'start' )
	824	sums_divnew_l = 0.0_wp
[1]	825
	826	!
[1908]	827	!-- d must be reset to zero because it can contain nonzero values below the
	828	!-- topography
	829	IF ( topography /= 'flat' ) d = 0.0_wp
[1]	830
[1908]	831	localsum = 0.0_wp
	832	threadsum = 0.0_wp
[1]	833
[1908]	834	!$OMP PARALLEL PRIVATE (i,j,k) FIRSTPRIVATE(threadsum) REDUCTION(+:localsum)
[2073]	835	#if defined( __ibm )
[1908]	836	!$OMP DO SCHEDULE( STATIC )
	837	DO i = nxl, nxr
	838	DO j = nys, nyn
[2232]	839	DO k = nzb+1, nzt
	840	d(k,j,i) = ( ( u(k,j,i+1) - u(k,j,i) ) * rho_air(k) * ddx + &
	841	( v(k,j+1,i) - v(k,j,i) ) * rho_air(k) * ddy + &
	842	( w(k,j,i) * rho_air_zw(k) - &
	843	w(k-1,j,i) * rho_air_zw(k-1) ) * ddzw(k) &
	844	) * MERGE( 1.0_wp, 0.0_wp, &
	845	BTEST( wall_flags_0(k,j,i), 0 ) &
	846	)
[1908]	847	ENDDO
	848	DO k = nzb+1, nzt
	849	threadsum = threadsum + ABS( d(k,j,i) )
	850	ENDDO
[1]	851	ENDDO
	852	ENDDO
	853	#else
[2073]	854	!$OMP DO SCHEDULE( STATIC )
[1908]	855	DO i = nxl, nxr
	856	DO j = nys, nyn
[2232]	857	DO k = nzb+1, nzt
[2037]	858	d(k,j,i) = ( ( u(k,j,i+1) - u(k,j,i) ) * rho_air(k) * ddx + &
	859	( v(k,j+1,i) - v(k,j,i) ) * rho_air(k) * ddy + &
	860	( w(k,j,i) * rho_air_zw(k) - &
	861	w(k-1,j,i) * rho_air_zw(k-1) ) * ddzw(k) &
[2232]	862	) * MERGE( 1.0_wp, 0.0_wp, &
	863	BTEST( wall_flags_0(k,j,i), 0 ) &
	864	)
[1908]	865	ENDDO
[1113]	866	ENDDO
	867	ENDDO
	868	!
[1908]	869	!-- Compute possible PE-sum of divergences for flow_statistics
[2073]	870	!$OMP DO SCHEDULE( STATIC )
[1908]	871	DO i = nxl, nxr
	872	DO j = nys, nyn
	873	DO k = nzb+1, nzt
	874	threadsum = threadsum + ABS( d(k,j,i) )
	875	ENDDO
[1]	876	ENDDO
	877	ENDDO
	878	#endif
[667]	879
[1908]	880	localsum = localsum + threadsum
	881	!$OMP END PARALLEL
[1]	882
	883	!
[1908]	884	!-- For completeness, set the divergence sum of all statistic regions to those
	885	!-- of the total domain
	886	sums_divnew_l(0:statistic_regions) = localsum
[1]	887
[1908]	888	CALL cpu_log( log_point_s(1), 'divergence', 'stop' )
[1]	889
[1908]	890	ENDIF
	891
[1]	892	CALL cpu_log( log_point(8), 'pres', 'stop' )
	893
	894
	895	END SUBROUTINE pres

Note: See TracBrowser for help on using the repository browser.

Download in other formats:

| Impressum | ©Leibniz Universität Hannover |