Home

Context Navigation

source: palm/trunk/SOURCE/data_output_2d.f90 @ 2934

Last change on this file since 2934 was 2817, checked in by knoop, 7 years ago
Preliminary gust module interface implemented
Property svn:keywords set to `Id`
File size: 90.0 KB

Rev	Line
[1682]	1	!> @file data_output_2d.f90
[2000]	2	!------------------------------------------------------------------------------!
[2696]	3	! This file is part of the PALM model system.
[1036]	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	!
[2718]	17	! Copyright 1997-2018 Leibniz Universitaet Hannover
[2000]	18	!------------------------------------------------------------------------------!
[1036]	19	!
[254]	20	! Current revisions:
[1]	21	! -----------------
[1961]	22	!
[2233]	23	!
[1552]	24	! Former revisions:
	25	! -----------------
	26	! $Id: data_output_2d.f90 2817 2018-02-19 16:32:21Z suehring $
[2817]	27	! Preliminary gust module interface implemented
	28	!
	29	! 2805 2018-02-14 17:00:09Z suehring
[2798]	30	! Consider also default-type surfaces for surface temperature output.
	31	!
	32	! 2797 2018-02-08 13:24:35Z suehring
[2797]	33	! Enable output of ground-heat flux also at urban surfaces.
	34	!
	35	! 2743 2018-01-12 16:03:39Z suehring
[2743]	36	! In case of natural- and urban-type surfaces output surfaces fluxes in W/m2.
	37	!
	38	! 2742 2018-01-12 14:59:47Z suehring
[2742]	39	! Enable output of surface temperature
	40	!
	41	! 2735 2018-01-11 12:01:27Z suehring
[2735]	42	! output of r_a moved from land-surface to consider also urban-type surfaces
	43	!
	44	! 2718 2018-01-02 08:49:38Z maronga
[2716]	45	! Corrected "Former revisions" section
	46	!
	47	! 2696 2017-12-14 17:12:51Z kanani
	48	! Change in file header (GPL part)
[2696]	49	! Implementation of uv exposure model (FK)
	50	! Implementation of turbulence_closure_mod (TG)
	51	! Set fill values at topography grid points or e.g. non-natural-type surface
	52	! in case of LSM output (MS)
	53	!
	54	! 2512 2017-10-04 08:26:59Z raasch
[2512]	55	! upper bounds of cross section output changed from nx+1,ny+1 to nx,ny
	56	! no output of ghost layer data
	57	!
	58	! 2292 2017-06-20 09:51:42Z schwenkel
[2292]	59	! Implementation of new microphysic scheme: cloud_scheme = 'morrison'
	60	! includes two more prognostic equations for cloud drop concentration (nc)
	61	! and cloud water content (qc).
	62	!
	63	! 2277 2017-06-12 10:47:51Z kanani
[2277]	64	! Removed unused variables do2d_xy_n, do2d_xz_n, do2d_yz_n
	65	!
	66	! 2233 2017-05-30 18:08:54Z suehring
[1552]	67	!
[2233]	68	! 2232 2017-05-30 17:47:52Z suehring
	69	! Adjustments to new surface concept
	70	!
	71	!
[2191]	72	! 2190 2017-03-21 12:16:43Z raasch
	73	! bugfix for r2031: string rho replaced by rho_ocean
	74	!
[2032]	75	! 2031 2016-10-21 15:11:58Z knoop
	76	! renamed variable rho to rho_ocean and rho_av to rho_ocean_av
	77	!
[2001]	78	! 2000 2016-08-20 18:09:15Z knoop
	79	! Forced header and separation lines into 80 columns
	80	!
[1981]	81	! 1980 2016-07-29 15:51:57Z suehring
	82	! Bugfix, in order to steer user-defined output, setting flag found explicitly
	83	! to .F.
	84	!
[1977]	85	! 1976 2016-07-27 13:28:04Z maronga
	86	! Output of radiation quantities is now done directly in the respective module
	87	!
[1973]	88	! 1972 2016-07-26 07:52:02Z maronga
[1976]	89	! Output of land surface quantities is now done directly in the respective
	90	! module
[1973]	91	!
[1961]	92	! 1960 2016-07-12 16:34:24Z suehring
	93	! Scalar surface flux added
	94	! Rename INTEGER variable s into s_ind, as s is already assigned to scalar
	95	!
[1851]	96	! 1849 2016-04-08 11:33:18Z hoffmann
	97	! precipitation_amount, precipitation_rate, prr moved to arrays_3d
	98	!
[1823]	99	! 1822 2016-04-07 07:49:42Z hoffmann
	100	! Output of bulk cloud physics simplified.
	101	!
[1789]	102	! 1788 2016-03-10 11:01:04Z maronga
	103	! Added output of z0q
	104	!
[1784]	105	! 1783 2016-03-06 18:36:17Z raasch
	106	! name change of netcdf routines and module + related changes
	107	!
[1746]	108	! 1745 2016-02-05 13:06:51Z gronemeier
	109	! Bugfix: test if time axis limit exceeds moved to point after call of check_open
	110	!
[1704]	111	! 1703 2015-11-02 12:38:44Z raasch
	112	! bugfix for output of single (*) xy-sections in case of parallel netcdf I/O
	113	!
[1702]	114	! 1701 2015-11-02 07:43:04Z maronga
	115	! Bugfix in output of RRTGM data
	116	!
[1692]	117	! 1691 2015-10-26 16:17:44Z maronga
	118	! Added output of Obukhov length (ol) and radiative heating rates for RRTMG.
	119	! Formatting corrections.
	120	!
[1683]	121	! 1682 2015-10-07 23:56:08Z knoop
	122	! Code annotations made doxygen readable
	123	!
[1586]	124	! 1585 2015-04-30 07:05:52Z maronga
	125	! Added support for RRTMG
	126	!
[1556]	127	! 1555 2015-03-04 17:44:27Z maronga
	128	! Added output of r_a and r_s
	129	!
[1552]	130	! 1551 2015-03-03 14:18:16Z maronga
[1551]	131	! Added suppport for land surface model and radiation model output. In the course
	132	! of this action, the limits for vertical loops have been changed (from nzb and
	133	! nzt+1 to nzb_do and nzt_do, respectively in order to allow soil model output).
	134	! Moreover, a new vertical grid zs was introduced.
[1329]	135	!
[1360]	136	! 1359 2014-04-11 17:15:14Z hoffmann
	137	! New particle structure integrated.
	138	!
[1354]	139	! 1353 2014-04-08 15:21:23Z heinze
	140	! REAL constants provided with KIND-attribute
	141	!
[1329]	142	! 1327 2014-03-21 11:00:16Z raasch
	143	! parts concerning iso2d output removed,
	144	! -netcdf output queries
	145	!
[1321]	146	! 1320 2014-03-20 08:40:49Z raasch
[1320]	147	! ONLY-attribute added to USE-statements,
	148	! kind-parameters added to all INTEGER and REAL declaration statements,
	149	! kinds are defined in new module kinds,
	150	! revision history before 2012 removed,
	151	! comment fields (!:) to be used for variable explanations added to
	152	! all variable declaration statements
[1309]	153	!
[1319]	154	! 1318 2014-03-17 13:35:16Z raasch
	155	! barrier argument removed from cpu_log.
	156	! module interfaces removed
	157	!
[1312]	158	! 1311 2014-03-14 12:13:39Z heinze
	159	! bugfix: close #if defined( __netcdf )
	160	!
[1309]	161	! 1308 2014-03-13 14:58:42Z fricke
[1308]	162	! +local_2d_sections, local_2d_sections_l, ns
	163	! Check, if the limit of the time dimension is exceeded for parallel output
	164	! To increase the performance for parallel output, the following is done:
	165	! - Update of time axis is only done by PE0
	166	! - Cross sections are first stored on a local array and are written
	167	! collectively to the output file by all PEs.
[674]	168	!
[1116]	169	! 1115 2013-03-26 18:16:16Z hoffmann
	170	! ql is calculated by calc_liquid_water_content
	171	!
[1077]	172	! 1076 2012-12-05 08:30:18Z hoffmann
	173	! Bugfix in output of ql
	174	!
[1066]	175	! 1065 2012-11-22 17:42:36Z hoffmann
	176	! Bugfix: Output of cross sections of ql
	177	!
[1054]	178	! 1053 2012-11-13 17:11:03Z hoffmann
	179	! +qr, nr, qc and cross sections
	180	!
[1037]	181	! 1036 2012-10-22 13:43:42Z raasch
	182	! code put under GPL (PALM 3.9)
	183	!
[1035]	184	! 1031 2012-10-19 14:35:30Z raasch
	185	! netCDF4 without parallel file support implemented
	186	!
[1008]	187	! 1007 2012-09-19 14:30:36Z franke
	188	! Bugfix: missing calculation of ql_vp added
	189	!
[979]	190	! 978 2012-08-09 08:28:32Z fricke
	191	! +z0h
	192	!
[1]	193	! Revision 1.1 1997/08/11 06:24:09 raasch
	194	! Initial revision
	195	!
	196	!
	197	! Description:
	198	! ------------
[2512]	199	!> Data output of cross-sections in netCDF format or binary format
	200	!> to be later converted to NetCDF by helper routine combine_plot_fields.
[1682]	201	!> Attention: The position of the sectional planes is still not always computed
	202	!> --------- correctly. (zu is used always)!
[1]	203	!------------------------------------------------------------------------------!
[1682]	204	SUBROUTINE data_output_2d( mode, av )
	205
[1]	206
[1320]	207	USE arrays_3d, &
[2743]	208	ONLY: dzw, e, heatflux_output_conversion, nc, nr, p, pt, &
	209	precipitation_amount, precipitation_rate, &
[2232]	210	prr, q, qc, ql, ql_c, ql_v, ql_vp, qr, rho_ocean, s, sa, &
[2743]	211	tend, u, v, vpt, w, zu, zw, waterflux_output_conversion
[1320]	212
[1]	213	USE averaging
[1320]	214
	215	USE cloud_parameters, &
[2743]	216	ONLY: cp, hyrho, l_d_cp, l_v, pt_d_t
[1320]	217
	218	USE control_parameters, &
	219	ONLY: cloud_physics, data_output_2d_on_each_pe, data_output_xy, &
	220	data_output_xz, data_output_yz, do2d, &
[2277]	221	do2d_xy_last_time, do2d_xy_time_count, &
	222	do2d_xz_last_time, do2d_xz_time_count, &
	223	do2d_yz_last_time, do2d_yz_time_count, &
[2232]	224	ibc_uv_b, io_blocks, io_group, land_surface, message_string, &
[1822]	225	ntdim_2d_xy, ntdim_2d_xz, ntdim_2d_yz, &
	226	psolver, section, simulated_time, simulated_time_chr, &
[2696]	227	time_since_reference_point, uv_exposure
[1320]	228
	229	USE cpulog, &
	230	ONLY: cpu_log, log_point
	231
	232	USE grid_variables, &
	233	ONLY: dx, dy
[2817]	234
	235	USE gust_mod, &
	236	ONLY: gust_data_output_2d, gust_module_enabled
[1320]	237
	238	USE indices, &
[2696]	239	ONLY: nbgp, nx, nxl, nxr, ny, nyn, nys, nz, nzb, nzt, wall_flags_0
[1320]	240
	241	USE kinds
[1551]	242
	243	USE land_surface_model_mod, &
[2232]	244	ONLY: lsm_data_output_2d, zs
[1551]	245
[1783]	246	#if defined( __netcdf )
	247	USE NETCDF
	248	#endif
[1320]	249
[1783]	250	USE netcdf_interface, &
[2696]	251	ONLY: fill_value, id_set_xy, id_set_xz, id_set_yz, id_var_do2d, &
	252	id_var_time_xy, id_var_time_xz, id_var_time_yz, nc_stat, &
	253	netcdf_data_format, netcdf_handle_error
[1783]	254
[1320]	255	USE particle_attributes, &
[1359]	256	ONLY: grid_particles, number_of_particles, particle_advection_start, &
	257	particles, prt_count
[1320]	258
[1]	259	USE pegrid
	260
[1551]	261	USE radiation_model_mod, &
[1976]	262	ONLY: radiation, radiation_data_output_2d
[1551]	263
[2232]	264	USE surface_mod, &
	265	ONLY: surf_def_h, surf_lsm_h, surf_usm_h
	266
[2696]	267	USE turbulence_closure_mod, &
	268	ONLY: tcm_data_output_2d
	269
	270	USE uv_exposure_model_mod, &
	271	ONLY: uvem_data_output_2d
	272
	273
[1]	274	IMPLICIT NONE
	275
[1682]	276	CHARACTER (LEN=2) :: do2d_mode !<
	277	CHARACTER (LEN=2) :: mode !<
	278	CHARACTER (LEN=4) :: grid !<
	279	CHARACTER (LEN=25) :: section_chr !<
	280	CHARACTER (LEN=50) :: rtext !<
[1320]	281
[1682]	282	INTEGER(iwp) :: av !<
	283	INTEGER(iwp) :: ngp !<
	284	INTEGER(iwp) :: file_id !<
[2696]	285	INTEGER(iwp) :: flag_nr !< number of masking flag
[1682]	286	INTEGER(iwp) :: i !<
	287	INTEGER(iwp) :: if !<
	288	INTEGER(iwp) :: is !<
	289	INTEGER(iwp) :: iis !<
	290	INTEGER(iwp) :: j !<
	291	INTEGER(iwp) :: k !<
	292	INTEGER(iwp) :: l !<
	293	INTEGER(iwp) :: layer_xy !<
[2232]	294	INTEGER(iwp) :: m !<
[1682]	295	INTEGER(iwp) :: n !<
[1703]	296	INTEGER(iwp) :: nis !<
[1682]	297	INTEGER(iwp) :: ns !<
	298	INTEGER(iwp) :: nzb_do !< lower limit of the data field (usually nzb)
	299	INTEGER(iwp) :: nzt_do !< upper limit of the data field (usually nzt+1)
	300	INTEGER(iwp) :: psi !<
[1960]	301	INTEGER(iwp) :: s_ind !<
[1682]	302	INTEGER(iwp) :: sender !<
	303	INTEGER(iwp) :: ind(4) !<
[1320]	304
[1682]	305	LOGICAL :: found !<
	306	LOGICAL :: resorted !<
	307	LOGICAL :: two_d !<
[1320]	308
[1682]	309	REAL(wp) :: mean_r !<
	310	REAL(wp) :: s_r2 !<
	311	REAL(wp) :: s_r3 !<
[1320]	312
[1682]	313	REAL(wp), DIMENSION(:), ALLOCATABLE :: level_z !<
	314	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: local_2d !<
	315	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: local_2d_l !<
[2232]	316
[1682]	317	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_pf !<
	318	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_2d_sections !<
	319	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_2d_sections_l !<
[1359]	320
[1]	321	#if defined( __parallel )
[1682]	322	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: total_2d !<
[1]	323	#endif
[1682]	324	REAL(wp), DIMENSION(:,:,:), POINTER :: to_be_resorted !<
[1]	325
	326	NAMELIST /LOCAL/ rtext
	327
	328	!
	329	!-- Immediate return, if no output is requested (no respective sections
	330	!-- found in parameter data_output)
	331	IF ( mode == 'xy' .AND. .NOT. data_output_xy(av) ) RETURN
	332	IF ( mode == 'xz' .AND. .NOT. data_output_xz(av) ) RETURN
	333	IF ( mode == 'yz' .AND. .NOT. data_output_yz(av) ) RETURN
	334
[1308]	335	CALL cpu_log (log_point(3),'data_output_2d','start')
	336
[1]	337	two_d = .FALSE. ! local variable to distinguish between output of pure 2D
	338	! arrays and cross-sections of 3D arrays.
	339
	340	!
	341	!-- Depending on the orientation of the cross-section, the respective output
	342	!-- files have to be opened.
	343	SELECT CASE ( mode )
	344
	345	CASE ( 'xy' )
[1960]	346	s_ind = 1
[2512]	347	ALLOCATE( level_z(nzb:nzt+1), local_2d(nxl:nxr,nys:nyn) )
[1]	348
[1308]	349	IF ( netcdf_data_format > 4 ) THEN
	350	ns = 1
[1960]	351	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
[1308]	352	ns = ns + 1
	353	ENDDO
	354	ns = ns - 1
[2512]	355	ALLOCATE( local_2d_sections(nxl:nxr,nys:nyn,1:ns) )
[1353]	356	local_2d_sections = 0.0_wp
[1308]	357	ENDIF
	358
[493]	359	!
[1031]	360	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
[1327]	361	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
[493]	362	CALL check_open( 101+av*10 )
	363	ENDIF
[1]	364	IF ( data_output_2d_on_each_pe ) THEN
	365	CALL check_open( 21 )
	366	ELSE
	367	IF ( myid == 0 ) THEN
	368	#if defined( __parallel )
[2512]	369	ALLOCATE( total_2d(0:nx,0:ny) )
[1]	370	#endif
	371	ENDIF
	372	ENDIF
	373
	374	CASE ( 'xz' )
[1960]	375	s_ind = 2
[2512]	376	ALLOCATE( local_2d(nxl:nxr,nzb:nzt+1) )
[1]	377
[1308]	378	IF ( netcdf_data_format > 4 ) THEN
	379	ns = 1
[1960]	380	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
[1308]	381	ns = ns + 1
	382	ENDDO
	383	ns = ns - 1
[2512]	384	ALLOCATE( local_2d_sections(nxl:nxr,1:ns,nzb:nzt+1) )
	385	ALLOCATE( local_2d_sections_l(nxl:nxr,1:ns,nzb:nzt+1) )
[1353]	386	local_2d_sections = 0.0_wp; local_2d_sections_l = 0.0_wp
[1308]	387	ENDIF
	388
[493]	389	!
[1031]	390	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
[1327]	391	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
[493]	392	CALL check_open( 102+av*10 )
	393	ENDIF
[1]	394
	395	IF ( data_output_2d_on_each_pe ) THEN
	396	CALL check_open( 22 )
	397	ELSE
	398	IF ( myid == 0 ) THEN
	399	#if defined( __parallel )
[2512]	400	ALLOCATE( total_2d(0:nx,nzb:nzt+1) )
[1]	401	#endif
	402	ENDIF
	403	ENDIF
	404
	405	CASE ( 'yz' )
[1960]	406	s_ind = 3
[2512]	407	ALLOCATE( local_2d(nys:nyn,nzb:nzt+1) )
[1]	408
[1308]	409	IF ( netcdf_data_format > 4 ) THEN
	410	ns = 1
[1960]	411	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
[1308]	412	ns = ns + 1
	413	ENDDO
	414	ns = ns - 1
[2512]	415	ALLOCATE( local_2d_sections(1:ns,nys:nyn,nzb:nzt+1) )
	416	ALLOCATE( local_2d_sections_l(1:ns,nys:nyn,nzb:nzt+1) )
[1353]	417	local_2d_sections = 0.0_wp; local_2d_sections_l = 0.0_wp
[1308]	418	ENDIF
	419
[493]	420	!
[1031]	421	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
[1327]	422	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
[493]	423	CALL check_open( 103+av*10 )
	424	ENDIF
[1]	425
	426	IF ( data_output_2d_on_each_pe ) THEN
	427	CALL check_open( 23 )
	428	ELSE
	429	IF ( myid == 0 ) THEN
	430	#if defined( __parallel )
[2512]	431	ALLOCATE( total_2d(0:ny,nzb:nzt+1) )
[1]	432	#endif
	433	ENDIF
	434	ENDIF
	435
	436	CASE DEFAULT
[254]	437	message_string = 'unknown cross-section: ' // TRIM( mode )
	438	CALL message( 'data_output_2d', 'PA0180', 1, 2, 0, 6, 0 )
[1]	439
	440	END SELECT
	441
	442	!
[1745]	443	!-- For parallel netcdf output the time axis must be limited. Return, if this
	444	!-- limit is exceeded. This could be the case, if the simulated time exceeds
	445	!-- the given end time by the length of the given output interval.
	446	IF ( netcdf_data_format > 4 ) THEN
	447	IF ( mode == 'xy' .AND. do2d_xy_time_count(av) + 1 > &
	448	ntdim_2d_xy(av) ) THEN
	449	WRITE ( message_string, * ) 'Output of xy cross-sections is not ', &
	450	'given at t=', simulated_time, '&because the', &
	451	' maximum number of output time levels is exceeded.'
	452	CALL message( 'data_output_2d', 'PA0384', 0, 1, 0, 6, 0 )
	453	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
	454	RETURN
	455	ENDIF
	456	IF ( mode == 'xz' .AND. do2d_xz_time_count(av) + 1 > &
	457	ntdim_2d_xz(av) ) THEN
	458	WRITE ( message_string, * ) 'Output of xz cross-sections is not ', &
	459	'given at t=', simulated_time, '&because the', &
	460	' maximum number of output time levels is exceeded.'
	461	CALL message( 'data_output_2d', 'PA0385', 0, 1, 0, 6, 0 )
	462	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
	463	RETURN
	464	ENDIF
	465	IF ( mode == 'yz' .AND. do2d_yz_time_count(av) + 1 > &
	466	ntdim_2d_yz(av) ) THEN
	467	WRITE ( message_string, * ) 'Output of yz cross-sections is not ', &
	468	'given at t=', simulated_time, '&because the', &
	469	' maximum number of output time levels is exceeded.'
	470	CALL message( 'data_output_2d', 'PA0386', 0, 1, 0, 6, 0 )
	471	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
	472	RETURN
	473	ENDIF
	474	ENDIF
	475
	476	!
[1]	477	!-- Allocate a temporary array for resorting (kji -> ijk).
[2512]	478	ALLOCATE( local_pf(nxl:nxr,nys:nyn,nzb:nzt+1) )
[2232]	479	local_pf = 0.0
[1]	480
	481	!
	482	!-- Loop of all variables to be written.
	483	!-- Output dimensions chosen
	484	if = 1
	485	l = MAX( 2, LEN_TRIM( do2d(av,if) ) )
	486	do2d_mode = do2d(av,if)(l-1:l)
	487
	488	DO WHILE ( do2d(av,if)(1:1) /= ' ' )
	489
	490	IF ( do2d_mode == mode ) THEN
[1980]	491	!
	492	!-- Set flag to steer output of radiation, land-surface, or user-defined
	493	!-- quantities
	494	found = .FALSE.
[1551]	495
	496	nzb_do = nzb
	497	nzt_do = nzt+1
[1]	498	!
[2696]	499	!-- Before each output, set array local_pf to fill value
	500	local_pf = fill_value
	501	!
	502	!-- Set masking flag for topography for not resorted arrays
	503	flag_nr = 0
	504
	505	!
[1]	506	!-- Store the array chosen on the temporary array.
	507	resorted = .FALSE.
	508	SELECT CASE ( TRIM( do2d(av,if) ) )
	509	CASE ( 'e_xy', 'e_xz', 'e_yz' )
	510	IF ( av == 0 ) THEN
	511	to_be_resorted => e
	512	ELSE
	513	to_be_resorted => e_av
	514	ENDIF
	515	IF ( mode == 'xy' ) level_z = zu
	516
[771]	517	CASE ( 'lpt_xy', 'lpt_xz', 'lpt_yz' )
	518	IF ( av == 0 ) THEN
	519	to_be_resorted => pt
	520	ELSE
	521	to_be_resorted => lpt_av
	522	ENDIF
	523	IF ( mode == 'xy' ) level_z = zu
	524
[1]	525	CASE ( 'lwp*_xy' ) ! 2d-array
	526	IF ( av == 0 ) THEN
[2512]	527	DO i = nxl, nxr
	528	DO j = nys, nyn
[1320]	529	local_pf(i,j,nzb+1) = SUM( ql(nzb:nzt,j,i) * &
[1]	530	dzw(1:nzt+1) )
	531	ENDDO
	532	ENDDO
	533	ELSE
[2512]	534	DO i = nxl, nxr
	535	DO j = nys, nyn
[1]	536	local_pf(i,j,nzb+1) = lwp_av(j,i)
	537	ENDDO
	538	ENDDO
	539	ENDIF
	540	resorted = .TRUE.
	541	two_d = .TRUE.
	542	level_z(nzb+1) = zu(nzb+1)
	543
[2292]	544	CASE ( 'nc_xy', 'nc_xz', 'nc_yz' )
	545	IF ( av == 0 ) THEN
	546	to_be_resorted => nc
	547	ELSE
	548	to_be_resorted => nc_av
	549	ENDIF
	550	IF ( mode == 'xy' ) level_z = zu
	551
[1053]	552	CASE ( 'nr_xy', 'nr_xz', 'nr_yz' )
	553	IF ( av == 0 ) THEN
	554	to_be_resorted => nr
	555	ELSE
	556	to_be_resorted => nr_av
	557	ENDIF
	558	IF ( mode == 'xy' ) level_z = zu
	559
[2797]	560	CASE ( 'ghf*_xy' ) ! 2d-array
	561	IF ( av == 0 ) THEN
	562	DO m = 1, surf_lsm_h%ns
	563	i = surf_lsm_h%i(m)
	564	j = surf_lsm_h%j(m)
	565	local_pf(i,j,nzb+1) = surf_lsm_h%ghf(m)
	566	ENDDO
	567	DO m = 1, surf_usm_h%ns
	568	i = surf_usm_h%i(m)
	569	j = surf_usm_h%j(m)
	570	local_pf(i,j,nzb+1) = surf_usm_h%frac(0,m) * &
	571	surf_usm_h%wghf_eb(m) + &
	572	surf_usm_h%frac(1,m) * &
	573	surf_usm_h%wghf_eb_green(m) + &
	574	surf_usm_h%frac(2,m) * &
	575	surf_usm_h%wghf_eb_window(m)
	576	ENDDO
	577	ELSE
	578	DO i = nxl, nxr
	579	DO j = nys, nyn
	580	local_pf(i,j,nzb+1) = ghf_av(j,i)
	581	ENDDO
	582	ENDDO
	583	ENDIF
	584
	585	resorted = .TRUE.
	586	two_d = .TRUE.
	587	level_z(nzb+1) = zu(nzb+1)
	588
[1691]	589	CASE ( 'ol*_xy' ) ! 2d-array
	590	IF ( av == 0 ) THEN
[2232]	591	DO m = 1, surf_def_h(0)%ns
	592	i = surf_def_h(0)%i(m)
	593	j = surf_def_h(0)%j(m)
[2512]	594	local_pf(i,j,nzb+1) = surf_def_h(0)%ol(m)
[2232]	595	ENDDO
	596	DO m = 1, surf_lsm_h%ns
	597	i = surf_lsm_h%i(m)
	598	j = surf_lsm_h%j(m)
[2512]	599	local_pf(i,j,nzb+1) = surf_lsm_h%ol(m)
[2232]	600	ENDDO
	601	DO m = 1, surf_usm_h%ns
	602	i = surf_usm_h%i(m)
	603	j = surf_usm_h%j(m)
[2512]	604	local_pf(i,j,nzb+1) = surf_usm_h%ol(m)
[2232]	605	ENDDO
[1691]	606	ELSE
[2512]	607	DO i = nxl, nxr
	608	DO j = nys, nyn
[1691]	609	local_pf(i,j,nzb+1) = ol_av(j,i)
	610	ENDDO
	611	ENDDO
	612	ENDIF
	613	resorted = .TRUE.
	614	two_d = .TRUE.
	615	level_z(nzb+1) = zu(nzb+1)
	616
[1]	617	CASE ( 'p_xy', 'p_xz', 'p_yz' )
	618	IF ( av == 0 ) THEN
[729]	619	IF ( psolver /= 'sor' ) CALL exchange_horiz( p, nbgp )
[1]	620	to_be_resorted => p
	621	ELSE
[729]	622	IF ( psolver /= 'sor' ) CALL exchange_horiz( p_av, nbgp )
[1]	623	to_be_resorted => p_av
	624	ENDIF
	625	IF ( mode == 'xy' ) level_z = zu
	626
	627	CASE ( 'pc_xy', 'pc_xz', 'pc_yz' ) ! particle concentration
	628	IF ( av == 0 ) THEN
[215]	629	IF ( simulated_time >= particle_advection_start ) THEN
	630	tend = prt_count
[2512]	631	! CALL exchange_horiz( tend, nbgp )
[215]	632	ELSE
[1353]	633	tend = 0.0_wp
[215]	634	ENDIF
[2512]	635	DO i = nxl, nxr
	636	DO j = nys, nyn
[1]	637	DO k = nzb, nzt+1
	638	local_pf(i,j,k) = tend(k,j,i)
	639	ENDDO
	640	ENDDO
	641	ENDDO
	642	resorted = .TRUE.
	643	ELSE
[2512]	644	! CALL exchange_horiz( pc_av, nbgp )
[1]	645	to_be_resorted => pc_av
	646	ENDIF
	647
[1359]	648	CASE ( 'pr_xy', 'pr_xz', 'pr_yz' ) ! mean particle radius (effective radius)
[1]	649	IF ( av == 0 ) THEN
[215]	650	IF ( simulated_time >= particle_advection_start ) THEN
	651	DO i = nxl, nxr
	652	DO j = nys, nyn
	653	DO k = nzb, nzt+1
[1359]	654	number_of_particles = prt_count(k,j,i)
	655	IF (number_of_particles <= 0) CYCLE
	656	particles => grid_particles(k,j,i)%particles(1:number_of_particles)
	657	s_r2 = 0.0_wp
[1353]	658	s_r3 = 0.0_wp
[1359]	659	DO n = 1, number_of_particles
	660	IF ( particles(n)%particle_mask ) THEN
	661	s_r2 = s_r2 + particles(n)%radius*2 &
	662	particles(n)%weight_factor
	663	s_r3 = s_r3 + particles(n)%radius*3 &
	664	particles(n)%weight_factor
	665	ENDIF
[215]	666	ENDDO
[1359]	667	IF ( s_r2 > 0.0_wp ) THEN
	668	mean_r = s_r3 / s_r2
[215]	669	ELSE
[1353]	670	mean_r = 0.0_wp
[215]	671	ENDIF
	672	tend(k,j,i) = mean_r
[1]	673	ENDDO
	674	ENDDO
	675	ENDDO
[2512]	676	! CALL exchange_horiz( tend, nbgp )
[215]	677	ELSE
[1353]	678	tend = 0.0_wp
[1359]	679	ENDIF
[2512]	680	DO i = nxl, nxr
	681	DO j = nys, nyn
[1]	682	DO k = nzb, nzt+1
	683	local_pf(i,j,k) = tend(k,j,i)
	684	ENDDO
	685	ENDDO
	686	ENDDO
	687	resorted = .TRUE.
	688	ELSE
[2512]	689	! CALL exchange_horiz( pr_av, nbgp )
[1]	690	to_be_resorted => pr_av
	691	ENDIF
	692
[72]	693	CASE ( 'pra*_xy' ) ! 2d-array / integral quantity => no av
[2512]	694	! CALL exchange_horiz_2d( precipitation_amount )
	695	DO i = nxl, nxr
	696	DO j = nys, nyn
[72]	697	local_pf(i,j,nzb+1) = precipitation_amount(j,i)
	698	ENDDO
	699	ENDDO
[1353]	700	precipitation_amount = 0.0_wp ! reset for next integ. interval
[72]	701	resorted = .TRUE.
	702	two_d = .TRUE.
	703	level_z(nzb+1) = zu(nzb+1)
	704
	705	CASE ( 'prr*_xy' ) ! 2d-array
[1822]	706	IF ( av == 0 ) THEN
[2512]	707	! CALL exchange_horiz_2d( prr(nzb+1,:,:) )
	708	DO i = nxl, nxr
	709	DO j = nys, nyn
[1822]	710	local_pf(i,j,nzb+1) = prr(nzb+1,j,i) * hyrho(nzb+1)
[1053]	711	ENDDO
[1822]	712	ENDDO
[1053]	713	ELSE
[2512]	714	! CALL exchange_horiz_2d( prr_av(nzb+1,:,:) )
	715	DO i = nxl, nxr
	716	DO j = nys, nyn
[1822]	717	local_pf(i,j,nzb+1) = prr_av(nzb+1,j,i) * hyrho(nzb+1)
[1053]	718	ENDDO
[1822]	719	ENDDO
[1053]	720	ENDIF
	721	resorted = .TRUE.
	722	two_d = .TRUE.
	723	level_z(nzb+1) = zu(nzb+1)
	724
	725	CASE ( 'prr_xy', 'prr_xz', 'prr_yz' )
[72]	726	IF ( av == 0 ) THEN
[2512]	727	! CALL exchange_horiz( prr, nbgp )
	728	DO i = nxl, nxr
	729	DO j = nys, nyn
[1053]	730	DO k = nzb, nzt+1
[1822]	731	local_pf(i,j,k) = prr(k,j,i) * hyrho(nzb+1)
[1053]	732	ENDDO
[72]	733	ENDDO
	734	ENDDO
	735	ELSE
[2512]	736	! CALL exchange_horiz( prr_av, nbgp )
	737	DO i = nxl, nxr
	738	DO j = nys, nyn
[1053]	739	DO k = nzb, nzt+1
[1822]	740	local_pf(i,j,k) = prr_av(k,j,i) * hyrho(nzb+1)
[1053]	741	ENDDO
[72]	742	ENDDO
	743	ENDDO
	744	ENDIF
	745	resorted = .TRUE.
[1053]	746	IF ( mode == 'xy' ) level_z = zu
[72]	747
[1]	748	CASE ( 'pt_xy', 'pt_xz', 'pt_yz' )
	749	IF ( av == 0 ) THEN
	750	IF ( .NOT. cloud_physics ) THEN
	751	to_be_resorted => pt
	752	ELSE
[2512]	753	DO i = nxl, nxr
	754	DO j = nys, nyn
[1]	755	DO k = nzb, nzt+1
[1320]	756	local_pf(i,j,k) = pt(k,j,i) + l_d_cp * &
	757	pt_d_t(k) * &
[1]	758	ql(k,j,i)
	759	ENDDO
	760	ENDDO
	761	ENDDO
	762	resorted = .TRUE.
	763	ENDIF
	764	ELSE
	765	to_be_resorted => pt_av
	766	ENDIF
	767	IF ( mode == 'xy' ) level_z = zu
	768
	769	CASE ( 'q_xy', 'q_xz', 'q_yz' )
	770	IF ( av == 0 ) THEN
	771	to_be_resorted => q
	772	ELSE
	773	to_be_resorted => q_av
	774	ENDIF
	775	IF ( mode == 'xy' ) level_z = zu
	776
[1053]	777	CASE ( 'qc_xy', 'qc_xz', 'qc_yz' )
[1]	778	IF ( av == 0 ) THEN
[1115]	779	to_be_resorted => qc
[1]	780	ELSE
[1115]	781	to_be_resorted => qc_av
[1]	782	ENDIF
	783	IF ( mode == 'xy' ) level_z = zu
	784
[1053]	785	CASE ( 'ql_xy', 'ql_xz', 'ql_yz' )
	786	IF ( av == 0 ) THEN
[1115]	787	to_be_resorted => ql
[1053]	788	ELSE
[1115]	789	to_be_resorted => ql_av
[1053]	790	ENDIF
	791	IF ( mode == 'xy' ) level_z = zu
	792
[1]	793	CASE ( 'ql_c_xy', 'ql_c_xz', 'ql_c_yz' )
	794	IF ( av == 0 ) THEN
	795	to_be_resorted => ql_c
	796	ELSE
	797	to_be_resorted => ql_c_av
	798	ENDIF
	799	IF ( mode == 'xy' ) level_z = zu
	800
	801	CASE ( 'ql_v_xy', 'ql_v_xz', 'ql_v_yz' )
	802	IF ( av == 0 ) THEN
	803	to_be_resorted => ql_v
	804	ELSE
	805	to_be_resorted => ql_v_av
	806	ENDIF
	807	IF ( mode == 'xy' ) level_z = zu
	808
	809	CASE ( 'ql_vp_xy', 'ql_vp_xz', 'ql_vp_yz' )
	810	IF ( av == 0 ) THEN
[1007]	811	IF ( simulated_time >= particle_advection_start ) THEN
	812	DO i = nxl, nxr
	813	DO j = nys, nyn
	814	DO k = nzb, nzt+1
[1359]	815	number_of_particles = prt_count(k,j,i)
	816	IF (number_of_particles <= 0) CYCLE
	817	particles => grid_particles(k,j,i)%particles(1:number_of_particles)
	818	DO n = 1, number_of_particles
	819	IF ( particles(n)%particle_mask ) THEN
	820	tend(k,j,i) = tend(k,j,i) + &
	821	particles(n)%weight_factor / &
	822	prt_count(k,j,i)
	823	ENDIF
[1007]	824	ENDDO
	825	ENDDO
	826	ENDDO
	827	ENDDO
[2512]	828	! CALL exchange_horiz( tend, nbgp )
[1007]	829	ELSE
[1353]	830	tend = 0.0_wp
[1359]	831	ENDIF
[2512]	832	DO i = nxl, nxr
	833	DO j = nys, nyn
[1007]	834	DO k = nzb, nzt+1
	835	local_pf(i,j,k) = tend(k,j,i)
	836	ENDDO
	837	ENDDO
	838	ENDDO
	839	resorted = .TRUE.
	840	ELSE
[2512]	841	! CALL exchange_horiz( ql_vp_av, nbgp )
[1]	842	to_be_resorted => ql_vp
	843	ENDIF
	844	IF ( mode == 'xy' ) level_z = zu
	845
[1053]	846	CASE ( 'qr_xy', 'qr_xz', 'qr_yz' )
	847	IF ( av == 0 ) THEN
	848	to_be_resorted => qr
	849	ELSE
	850	to_be_resorted => qr_av
	851	ENDIF
	852	IF ( mode == 'xy' ) level_z = zu
	853
[354]	854	CASE ( 'qsws*_xy' ) ! 2d-array
	855	IF ( av == 0 ) THEN
[2743]	856	!
	857	!-- In case of default surfaces, clean-up flux by density.
	858	!-- In case of land- and urban-surfaces, convert fluxes into
	859	!-- dynamic units
[2232]	860	DO m = 1, surf_def_h(0)%ns
	861	i = surf_def_h(0)%i(m)
	862	j = surf_def_h(0)%j(m)
[2743]	863	k = surf_def_h(0)%k(m)
	864	local_pf(i,j,nzb+1) = surf_def_h(0)%qsws(m) * &
	865	waterflux_output_conversion(k)
[2232]	866	ENDDO
	867	DO m = 1, surf_lsm_h%ns
	868	i = surf_lsm_h%i(m)
	869	j = surf_lsm_h%j(m)
[2743]	870	k = surf_lsm_h%k(m)
	871	local_pf(i,j,nzb+1) = surf_lsm_h%qsws(m) * l_v
[2232]	872	ENDDO
	873	DO m = 1, surf_usm_h%ns
	874	i = surf_usm_h%i(m)
	875	j = surf_usm_h%j(m)
[2743]	876	k = surf_usm_h%k(m)
	877	local_pf(i,j,nzb+1) = surf_usm_h%qsws(m) * l_v
[2232]	878	ENDDO
[354]	879	ELSE
[2512]	880	DO i = nxl, nxr
	881	DO j = nys, nyn
[354]	882	local_pf(i,j,nzb+1) = qsws_av(j,i)
	883	ENDDO
	884	ENDDO
	885	ENDIF
	886	resorted = .TRUE.
	887	two_d = .TRUE.
	888	level_z(nzb+1) = zu(nzb+1)
	889
[1]	890	CASE ( 'qv_xy', 'qv_xz', 'qv_yz' )
	891	IF ( av == 0 ) THEN
[2512]	892	DO i = nxl, nxr
	893	DO j = nys, nyn
[1]	894	DO k = nzb, nzt+1
	895	local_pf(i,j,k) = q(k,j,i) - ql(k,j,i)
	896	ENDDO
	897	ENDDO
	898	ENDDO
	899	resorted = .TRUE.
	900	ELSE
	901	to_be_resorted => qv_av
	902	ENDIF
	903	IF ( mode == 'xy' ) level_z = zu
	904
[2735]	905	CASE ( 'r_a*_xy' ) ! 2d-array
	906	IF ( av == 0 ) THEN
	907	DO m = 1, surf_lsm_h%ns
	908	i = surf_lsm_h%i(m)
	909	j = surf_lsm_h%j(m)
	910	local_pf(i,j,nzb+1) = surf_lsm_h%r_a(m)
	911	ENDDO
[1551]	912
[2735]	913	DO m = 1, surf_usm_h%ns
	914	i = surf_usm_h%i(m)
	915	j = surf_usm_h%j(m)
	916	local_pf(i,j,nzb+1) = &
	917	( surf_usm_h%frac(0,m) * surf_usm_h%r_a(m) + &
	918	surf_usm_h%frac(1,m) * surf_usm_h%r_a_green(m) + &
	919	surf_usm_h%frac(2,m) * surf_usm_h%r_a_window(m) )
	920	ENDDO
	921	ELSE
	922	DO i = nxl, nxr
	923	DO j = nys, nyn
	924	local_pf(i,j,nzb+1) = r_a_av(j,i)
	925	ENDDO
	926	ENDDO
	927	ENDIF
	928	resorted = .TRUE.
	929	two_d = .TRUE.
	930	level_z(nzb+1) = zu(nzb+1)
	931
[2190]	932	CASE ( 'rho_ocean_xy', 'rho_ocean_xz', 'rho_ocean_yz' )
[96]	933	IF ( av == 0 ) THEN
[2031]	934	to_be_resorted => rho_ocean
[96]	935	ELSE
[2031]	936	to_be_resorted => rho_ocean_av
[96]	937	ENDIF
	938
[1]	939	CASE ( 's_xy', 's_xz', 's_yz' )
	940	IF ( av == 0 ) THEN
[1960]	941	to_be_resorted => s
[1]	942	ELSE
[355]	943	to_be_resorted => s_av
[1]	944	ENDIF
	945
[96]	946	CASE ( 'sa_xy', 'sa_xz', 'sa_yz' )
	947	IF ( av == 0 ) THEN
	948	to_be_resorted => sa
	949	ELSE
	950	to_be_resorted => sa_av
	951	ENDIF
	952
[354]	953	CASE ( 'shf*_xy' ) ! 2d-array
	954	IF ( av == 0 ) THEN
[2743]	955	!
	956	!-- In case of default surfaces, clean-up flux by density.
	957	!-- In case of land- and urban-surfaces, convert fluxes into
	958	!-- dynamic units.
[2232]	959	DO m = 1, surf_def_h(0)%ns
	960	i = surf_def_h(0)%i(m)
	961	j = surf_def_h(0)%j(m)
[2743]	962	k = surf_def_h(0)%k(m)
	963	local_pf(i,j,nzb+1) = surf_def_h(0)%shf(m) * &
	964	heatflux_output_conversion(k)
[2232]	965	ENDDO
	966	DO m = 1, surf_lsm_h%ns
	967	i = surf_lsm_h%i(m)
	968	j = surf_lsm_h%j(m)
[2743]	969	k = surf_lsm_h%k(m)
	970	local_pf(i,j,nzb+1) = surf_lsm_h%shf(m) * cp
[2232]	971	ENDDO
	972	DO m = 1, surf_usm_h%ns
	973	i = surf_usm_h%i(m)
	974	j = surf_usm_h%j(m)
[2743]	975	k = surf_usm_h%k(m)
	976	local_pf(i,j,nzb+1) = surf_usm_h%shf(m) * cp
[2232]	977	ENDDO
[354]	978	ELSE
[2512]	979	DO i = nxl, nxr
	980	DO j = nys, nyn
[354]	981	local_pf(i,j,nzb+1) = shf_av(j,i)
	982	ENDDO
	983	ENDDO
	984	ENDIF
	985	resorted = .TRUE.
	986	two_d = .TRUE.
	987	level_z(nzb+1) = zu(nzb+1)
[1960]	988
	989	CASE ( 'ssws*_xy' ) ! 2d-array
	990	IF ( av == 0 ) THEN
[2232]	991	DO m = 1, surf_def_h(0)%ns
	992	i = surf_def_h(0)%i(m)
	993	j = surf_def_h(0)%j(m)
[2512]	994	local_pf(i,j,nzb+1) = surf_def_h(0)%ssws(m)
[2232]	995	ENDDO
	996	DO m = 1, surf_lsm_h%ns
	997	i = surf_lsm_h%i(m)
	998	j = surf_lsm_h%j(m)
[2512]	999	local_pf(i,j,nzb+1) = surf_lsm_h%ssws(m)
[2232]	1000	ENDDO
	1001	DO m = 1, surf_usm_h%ns
	1002	i = surf_usm_h%i(m)
	1003	j = surf_usm_h%j(m)
[2512]	1004	local_pf(i,j,nzb+1) = surf_usm_h%ssws(m)
[2232]	1005	ENDDO
[1960]	1006	ELSE
[2512]	1007	DO i = nxl, nxr
	1008	DO j = nys, nyn
[1960]	1009	local_pf(i,j,nzb+1) = ssws_av(j,i)
	1010	ENDDO
	1011	ENDDO
	1012	ENDIF
	1013	resorted = .TRUE.
	1014	two_d = .TRUE.
	1015	level_z(nzb+1) = zu(nzb+1)
[1551]	1016
[1]	1017	CASE ( 't*_xy' ) ! 2d-array
	1018	IF ( av == 0 ) THEN
[2232]	1019	DO m = 1, surf_def_h(0)%ns
	1020	i = surf_def_h(0)%i(m)
	1021	j = surf_def_h(0)%j(m)
[2512]	1022	local_pf(i,j,nzb+1) = surf_def_h(0)%ts(m)
[2232]	1023	ENDDO
	1024	DO m = 1, surf_lsm_h%ns
	1025	i = surf_lsm_h%i(m)
	1026	j = surf_lsm_h%j(m)
[2512]	1027	local_pf(i,j,nzb+1) = surf_lsm_h%ts(m)
[2232]	1028	ENDDO
	1029	DO m = 1, surf_usm_h%ns
	1030	i = surf_usm_h%i(m)
	1031	j = surf_usm_h%j(m)
[2512]	1032	local_pf(i,j,nzb+1) = surf_usm_h%ts(m)
[2232]	1033	ENDDO
[1]	1034	ELSE
[2512]	1035	DO i = nxl, nxr
	1036	DO j = nys, nyn
[1]	1037	local_pf(i,j,nzb+1) = ts_av(j,i)
	1038	ENDDO
	1039	ENDDO
	1040	ENDIF
	1041	resorted = .TRUE.
	1042	two_d = .TRUE.
	1043	level_z(nzb+1) = zu(nzb+1)
	1044
[2742]	1045	CASE ( 'tsurf*_xy' ) ! 2d-array
	1046	IF ( av == 0 ) THEN
[2798]	1047	DO m = 1, surf_def_h(0)%ns
	1048	i = surf_def_h(0)%i(m)
	1049	j = surf_def_h(0)%j(m)
	1050	local_pf(i,j,nzb+1) = surf_def_h(0)%pt_surface(m)
	1051	ENDDO
	1052
[2742]	1053	DO m = 1, surf_lsm_h%ns
	1054	i = surf_lsm_h%i(m)
	1055	j = surf_lsm_h%j(m)
	1056	local_pf(i,j,nzb+1) = surf_lsm_h%pt_surface(m)
	1057	ENDDO
	1058
	1059	DO m = 1, surf_usm_h%ns
	1060	i = surf_usm_h%i(m)
	1061	j = surf_usm_h%j(m)
	1062	local_pf(i,j,nzb+1) = surf_usm_h%pt_surface(m)
	1063	ENDDO
	1064
	1065	ELSE
	1066	DO i = nxl, nxr
	1067	DO j = nys, nyn
	1068	local_pf(i,j,nzb+1) = tsurf_av(j,i)
	1069	ENDDO
	1070	ENDDO
	1071	ENDIF
	1072	resorted = .TRUE.
	1073	two_d = .TRUE.
	1074	level_z(nzb+1) = zu(nzb+1)
	1075
[1]	1076	CASE ( 'u_xy', 'u_xz', 'u_yz' )
[2696]	1077	flag_nr = 1
[1]	1078	IF ( av == 0 ) THEN
	1079	to_be_resorted => u
	1080	ELSE
	1081	to_be_resorted => u_av
	1082	ENDIF
	1083	IF ( mode == 'xy' ) level_z = zu
	1084	!
	1085	!-- Substitute the values generated by "mirror" boundary condition
	1086	!-- at the bottom boundary by the real surface values.
	1087	IF ( do2d(av,if) == 'u_xz' .OR. do2d(av,if) == 'u_yz' ) THEN
[1353]	1088	IF ( ibc_uv_b == 0 ) local_pf(:,:,nzb) = 0.0_wp
[1]	1089	ENDIF
	1090
	1091	CASE ( 'u*_xy' ) ! 2d-array
	1092	IF ( av == 0 ) THEN
[2232]	1093	DO m = 1, surf_def_h(0)%ns
	1094	i = surf_def_h(0)%i(m)
	1095	j = surf_def_h(0)%j(m)
[2512]	1096	local_pf(i,j,nzb+1) = surf_def_h(0)%us(m)
[2232]	1097	ENDDO
	1098	DO m = 1, surf_lsm_h%ns
	1099	i = surf_lsm_h%i(m)
	1100	j = surf_lsm_h%j(m)
[2512]	1101	local_pf(i,j,nzb+1) = surf_lsm_h%us(m)
[2232]	1102	ENDDO
	1103	DO m = 1, surf_usm_h%ns
	1104	i = surf_usm_h%i(m)
	1105	j = surf_usm_h%j(m)
[2512]	1106	local_pf(i,j,nzb+1) = surf_usm_h%us(m)
[2232]	1107	ENDDO
[1]	1108	ELSE
[2512]	1109	DO i = nxl, nxr
	1110	DO j = nys, nyn
[1]	1111	local_pf(i,j,nzb+1) = us_av(j,i)
	1112	ENDDO
	1113	ENDDO
	1114	ENDIF
	1115	resorted = .TRUE.
	1116	two_d = .TRUE.
	1117	level_z(nzb+1) = zu(nzb+1)
	1118
	1119	CASE ( 'v_xy', 'v_xz', 'v_yz' )
[2696]	1120	flag_nr = 2
[1]	1121	IF ( av == 0 ) THEN
	1122	to_be_resorted => v
	1123	ELSE
	1124	to_be_resorted => v_av
	1125	ENDIF
	1126	IF ( mode == 'xy' ) level_z = zu
	1127	!
	1128	!-- Substitute the values generated by "mirror" boundary condition
	1129	!-- at the bottom boundary by the real surface values.
	1130	IF ( do2d(av,if) == 'v_xz' .OR. do2d(av,if) == 'v_yz' ) THEN
[1353]	1131	IF ( ibc_uv_b == 0 ) local_pf(:,:,nzb) = 0.0_wp
[1]	1132	ENDIF
	1133
	1134	CASE ( 'vpt_xy', 'vpt_xz', 'vpt_yz' )
	1135	IF ( av == 0 ) THEN
	1136	to_be_resorted => vpt
	1137	ELSE
	1138	to_be_resorted => vpt_av
	1139	ENDIF
	1140	IF ( mode == 'xy' ) level_z = zu
	1141
	1142	CASE ( 'w_xy', 'w_xz', 'w_yz' )
[2696]	1143	flag_nr = 3
[1]	1144	IF ( av == 0 ) THEN
	1145	to_be_resorted => w
	1146	ELSE
	1147	to_be_resorted => w_av
	1148	ENDIF
	1149	IF ( mode == 'xy' ) level_z = zw
	1150
[72]	1151	CASE ( 'z0*_xy' ) ! 2d-array
	1152	IF ( av == 0 ) THEN
[2232]	1153	DO m = 1, surf_def_h(0)%ns
	1154	i = surf_def_h(0)%i(m)
	1155	j = surf_def_h(0)%j(m)
[2512]	1156	local_pf(i,j,nzb+1) = surf_def_h(0)%z0(m)
[2232]	1157	ENDDO
	1158	DO m = 1, surf_lsm_h%ns
	1159	i = surf_lsm_h%i(m)
	1160	j = surf_lsm_h%j(m)
[2512]	1161	local_pf(i,j,nzb+1) = surf_lsm_h%z0(m)
[2232]	1162	ENDDO
	1163	DO m = 1, surf_usm_h%ns
	1164	i = surf_usm_h%i(m)
	1165	j = surf_usm_h%j(m)
[2512]	1166	local_pf(i,j,nzb+1) = surf_usm_h%z0(m)
[2232]	1167	ENDDO
[72]	1168	ELSE
[2512]	1169	DO i = nxl, nxr
	1170	DO j = nys, nyn
[72]	1171	local_pf(i,j,nzb+1) = z0_av(j,i)
	1172	ENDDO
	1173	ENDDO
	1174	ENDIF
	1175	resorted = .TRUE.
	1176	two_d = .TRUE.
	1177	level_z(nzb+1) = zu(nzb+1)
	1178
[978]	1179	CASE ( 'z0h*_xy' ) ! 2d-array
	1180	IF ( av == 0 ) THEN
[2232]	1181	DO m = 1, surf_def_h(0)%ns
	1182	i = surf_def_h(0)%i(m)
	1183	j = surf_def_h(0)%j(m)
[2512]	1184	local_pf(i,j,nzb+1) = surf_def_h(0)%z0h(m)
[2232]	1185	ENDDO
	1186	DO m = 1, surf_lsm_h%ns
	1187	i = surf_lsm_h%i(m)
	1188	j = surf_lsm_h%j(m)
[2512]	1189	local_pf(i,j,nzb+1) = surf_lsm_h%z0h(m)
[2232]	1190	ENDDO
	1191	DO m = 1, surf_usm_h%ns
	1192	i = surf_usm_h%i(m)
	1193	j = surf_usm_h%j(m)
[2512]	1194	local_pf(i,j,nzb+1) = surf_usm_h%z0h(m)
[2232]	1195	ENDDO
[978]	1196	ELSE
[2512]	1197	DO i = nxl, nxr
	1198	DO j = nys, nyn
[978]	1199	local_pf(i,j,nzb+1) = z0h_av(j,i)
	1200	ENDDO
	1201	ENDDO
	1202	ENDIF
	1203	resorted = .TRUE.
	1204	two_d = .TRUE.
	1205	level_z(nzb+1) = zu(nzb+1)
	1206
[1788]	1207	CASE ( 'z0q*_xy' ) ! 2d-array
	1208	IF ( av == 0 ) THEN
[2232]	1209	DO m = 1, surf_def_h(0)%ns
	1210	i = surf_def_h(0)%i(m)
	1211	j = surf_def_h(0)%j(m)
[2512]	1212	local_pf(i,j,nzb+1) = surf_def_h(0)%z0q(m)
[2232]	1213	ENDDO
	1214	DO m = 1, surf_lsm_h%ns
	1215	i = surf_lsm_h%i(m)
	1216	j = surf_lsm_h%j(m)
[2512]	1217	local_pf(i,j,nzb+1) = surf_lsm_h%z0q(m)
[2232]	1218	ENDDO
	1219	DO m = 1, surf_usm_h%ns
	1220	i = surf_usm_h%i(m)
	1221	j = surf_usm_h%j(m)
[2512]	1222	local_pf(i,j,nzb+1) = surf_usm_h%z0q(m)
[2232]	1223	ENDDO
[1788]	1224	ELSE
[2512]	1225	DO i = nxl, nxr
	1226	DO j = nys, nyn
[1788]	1227	local_pf(i,j,nzb+1) = z0q_av(j,i)
	1228	ENDDO
	1229	ENDDO
	1230	ENDIF
	1231	resorted = .TRUE.
	1232	two_d = .TRUE.
	1233	level_z(nzb+1) = zu(nzb+1)
	1234
[1]	1235	CASE DEFAULT
[1972]	1236
[1]	1237	!
[1972]	1238	!-- Land surface model quantity
	1239	IF ( land_surface ) THEN
	1240	CALL lsm_data_output_2d( av, do2d(av,if), found, grid, mode,&
	1241	local_pf, two_d, nzb_do, nzt_do )
	1242	ENDIF
	1243
	1244	!
[2696]	1245	!-- Turbulence closure variables
	1246	IF ( .NOT. found ) THEN
	1247	CALL tcm_data_output_2d( av, do2d(av,if), found, grid, mode,&
	1248	local_pf, two_d, nzb_do, nzt_do )
	1249	ENDIF
	1250
	1251	!
[1976]	1252	!-- Radiation quantity
	1253	IF ( .NOT. found .AND. radiation ) THEN
	1254	CALL radiation_data_output_2d( av, do2d(av,if), found, grid,&
	1255	mode, local_pf, two_d )
	1256	ENDIF
	1257
	1258	!
[2817]	1259	!-- Gust module quantities
	1260	IF ( .NOT. found .AND. gust_module_enabled ) THEN
	1261	CALL gust_data_output_2d( av, do2d(av,if), found, grid, &
	1262	local_pf, two_d, nzb_do, nzt_do )
	1263	ENDIF
	1264
	1265	!
[2696]	1266	!-- UV exposure model quantity
	1267	IF ( uv_exposure ) THEN
	1268	CALL uvem_data_output_2d( av, do2d(av,if), found, grid, mode,&
	1269	local_pf, two_d, nzb_do, nzt_do )
	1270	ENDIF
	1271
	1272	!
[1]	1273	!-- User defined quantity
[1972]	1274	IF ( .NOT. found ) THEN
	1275	CALL user_data_output_2d( av, do2d(av,if), found, grid, &
	1276	local_pf, two_d, nzb_do, nzt_do )
	1277	ENDIF
	1278
[1]	1279	resorted = .TRUE.
	1280
	1281	IF ( grid == 'zu' ) THEN
	1282	IF ( mode == 'xy' ) level_z = zu
	1283	ELSEIF ( grid == 'zw' ) THEN
	1284	IF ( mode == 'xy' ) level_z = zw
[343]	1285	ELSEIF ( grid == 'zu1' ) THEN
	1286	IF ( mode == 'xy' ) level_z(nzb+1) = zu(nzb+1)
[1551]	1287	ELSEIF ( grid == 'zs' ) THEN
	1288	IF ( mode == 'xy' ) level_z = zs
[1]	1289	ENDIF
	1290
	1291	IF ( .NOT. found ) THEN
[1320]	1292	message_string = 'no output provided for: ' // &
[274]	1293	TRIM( do2d(av,if) )
[254]	1294	CALL message( 'data_output_2d', 'PA0181', 0, 0, 0, 6, 0 )
[1]	1295	ENDIF
	1296
	1297	END SELECT
	1298
	1299	!
[2696]	1300	!-- Resort the array to be output, if not done above. Flag topography
	1301	!-- grid points with fill values, using the corresponding maksing flag.
[1]	1302	IF ( .NOT. resorted ) THEN
[2512]	1303	DO i = nxl, nxr
	1304	DO j = nys, nyn
[1551]	1305	DO k = nzb_do, nzt_do
[2696]	1306	local_pf(i,j,k) = MERGE( to_be_resorted(k,j,i), &
	1307	REAL( fill_value, KIND = wp ), &
	1308	BTEST( wall_flags_0(k,j,i), &
	1309	flag_nr ) )
[1]	1310	ENDDO
	1311	ENDDO
	1312	ENDDO
	1313	ENDIF
	1314
	1315	!
	1316	!-- Output of the individual cross-sections, depending on the cross-
	1317	!-- section mode chosen.
	1318	is = 1
[1960]	1319	loop1: DO WHILE ( section(is,s_ind) /= -9999 .OR. two_d )
[1]	1320
	1321	SELECT CASE ( mode )
	1322
	1323	CASE ( 'xy' )
	1324	!
	1325	!-- Determine the cross section index
	1326	IF ( two_d ) THEN
	1327	layer_xy = nzb+1
	1328	ELSE
[1960]	1329	layer_xy = section(is,s_ind)
[1]	1330	ENDIF
	1331
	1332	!
[1551]	1333	!-- Exit the loop for layers beyond the data output domain
	1334	!-- (used for soil model)
[1691]	1335	IF ( layer_xy > nzt_do ) THEN
[1551]	1336	EXIT loop1
	1337	ENDIF
	1338
	1339	!
[1308]	1340	!-- Update the netCDF xy cross section time axis.
	1341	!-- In case of parallel output, this is only done by PE0
	1342	!-- to increase the performance.
	1343	IF ( simulated_time /= do2d_xy_last_time(av) ) THEN
	1344	do2d_xy_time_count(av) = do2d_xy_time_count(av) + 1
	1345	do2d_xy_last_time(av) = simulated_time
	1346	IF ( myid == 0 ) THEN
[1327]	1347	IF ( .NOT. data_output_2d_on_each_pe &
	1348	.OR. netcdf_data_format > 4 ) &
[493]	1349	THEN
[1]	1350	#if defined( __netcdf )
	1351	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
	1352	id_var_time_xy(av), &
[291]	1353	(/ time_since_reference_point /), &
[1]	1354	start = (/ do2d_xy_time_count(av) /), &
	1355	count = (/ 1 /) )
[1783]	1356	CALL netcdf_handle_error( 'data_output_2d', 53 )
[1]	1357	#endif
	1358	ENDIF
	1359	ENDIF
	1360	ENDIF
	1361	!
	1362	!-- If required, carry out averaging along z
[1960]	1363	IF ( section(is,s_ind) == -1 .AND. .NOT. two_d ) THEN
[1]	1364
[1353]	1365	local_2d = 0.0_wp
[1]	1366	!
	1367	!-- Carry out the averaging (all data are on the PE)
[1551]	1368	DO k = nzb_do, nzt_do
[2512]	1369	DO j = nys, nyn
	1370	DO i = nxl, nxr
[1]	1371	local_2d(i,j) = local_2d(i,j) + local_pf(i,j,k)
	1372	ENDDO
	1373	ENDDO
	1374	ENDDO
	1375
[1551]	1376	local_2d = local_2d / ( nzt_do - nzb_do + 1.0_wp)
[1]	1377
	1378	ELSE
	1379	!
	1380	!-- Just store the respective section on the local array
	1381	local_2d = local_pf(:,:,layer_xy)
	1382
	1383	ENDIF
	1384
	1385	#if defined( __parallel )
[1327]	1386	IF ( netcdf_data_format > 4 ) THEN
[1]	1387	!
[1031]	1388	!-- Parallel output in netCDF4/HDF5 format.
[493]	1389	IF ( two_d ) THEN
	1390	iis = 1
	1391	ELSE
	1392	iis = is
	1393	ENDIF
	1394
[1]	1395	#if defined( __netcdf )
[1308]	1396	!
	1397	!-- For parallel output, all cross sections are first stored
	1398	!-- here on a local array and will be written to the output
	1399	!-- file afterwards to increase the performance.
[2512]	1400	DO i = nxl, nxr
	1401	DO j = nys, nyn
[1308]	1402	local_2d_sections(i,j,iis) = local_2d(i,j)
	1403	ENDDO
	1404	ENDDO
[1]	1405	#endif
[493]	1406	ELSE
[1]	1407
[493]	1408	IF ( data_output_2d_on_each_pe ) THEN
[1]	1409	!
[493]	1410	!-- Output of partial arrays on each PE
	1411	#if defined( __netcdf )
[1327]	1412	IF ( myid == 0 ) THEN
[1320]	1413	WRITE ( 21 ) time_since_reference_point, &
[493]	1414	do2d_xy_time_count(av), av
	1415	ENDIF
	1416	#endif
[759]	1417	DO i = 0, io_blocks-1
	1418	IF ( i == io_group ) THEN
[2512]	1419	WRITE ( 21 ) nxl, nxr, nys, nyn, nys, nyn
[759]	1420	WRITE ( 21 ) local_2d
	1421	ENDIF
	1422	#if defined( __parallel )
	1423	CALL MPI_BARRIER( comm2d, ierr )
	1424	#endif
	1425	ENDDO
[559]	1426
[493]	1427	ELSE
[1]	1428	!
[493]	1429	!-- PE0 receives partial arrays from all processors and
	1430	!-- then outputs them. Here a barrier has to be set,
	1431	!-- because otherwise "-MPI- FATAL: Remote protocol queue
	1432	!-- full" may occur.
	1433	CALL MPI_BARRIER( comm2d, ierr )
	1434
[2512]	1435	ngp = ( nxr-nxl+1 ) * ( nyn-nys+1 )
[493]	1436	IF ( myid == 0 ) THEN
[1]	1437	!
[493]	1438	!-- Local array can be relocated directly.
[2512]	1439	total_2d(nxl:nxr,nys:nyn) = local_2d
[1]	1440	!
[493]	1441	!-- Receive data from all other PEs.
	1442	DO n = 1, numprocs-1
[1]	1443	!
[493]	1444	!-- Receive index limits first, then array.
	1445	!-- Index limits are received in arbitrary order from
	1446	!-- the PEs.
[1320]	1447	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
	1448	MPI_ANY_SOURCE, 0, comm2d, &
[493]	1449	status, ierr )
	1450	sender = status(MPI_SOURCE)
	1451	DEALLOCATE( local_2d )
	1452	ALLOCATE( local_2d(ind(1):ind(2),ind(3):ind(4)) )
[1320]	1453	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
	1454	MPI_REAL, sender, 1, comm2d, &
[493]	1455	status, ierr )
	1456	total_2d(ind(1):ind(2),ind(3):ind(4)) = local_2d
	1457	ENDDO
[1]	1458	!
[493]	1459	!-- Relocate the local array for the next loop increment
	1460	DEALLOCATE( local_2d )
[2512]	1461	ALLOCATE( local_2d(nxl:nxr,nys:nyn) )
[1]	1462
	1463	#if defined( __netcdf )
[1327]	1464	IF ( two_d ) THEN
	1465	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
	1466	id_var_do2d(av,if), &
[2512]	1467	total_2d(0:nx,0:ny), &
[1327]	1468	start = (/ 1, 1, 1, do2d_xy_time_count(av) /), &
[2512]	1469	count = (/ nx+1, ny+1, 1, 1 /) )
[1327]	1470	ELSE
	1471	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
	1472	id_var_do2d(av,if), &
[2512]	1473	total_2d(0:nx,0:ny), &
[1327]	1474	start = (/ 1, 1, is, do2d_xy_time_count(av) /), &
[2512]	1475	count = (/ nx+1, ny+1, 1, 1 /) )
[1]	1476	ENDIF
[1783]	1477	CALL netcdf_handle_error( 'data_output_2d', 54 )
[1]	1478	#endif
	1479
[493]	1480	ELSE
[1]	1481	!
[493]	1482	!-- First send the local index limits to PE0
[2512]	1483	ind(1) = nxl; ind(2) = nxr
	1484	ind(3) = nys; ind(4) = nyn
[1320]	1485	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
[493]	1486	comm2d, ierr )
[1]	1487	!
[493]	1488	!-- Send data to PE0
[2512]	1489	CALL MPI_SEND( local_2d(nxl,nys), ngp, &
[493]	1490	MPI_REAL, 0, 1, comm2d, ierr )
	1491	ENDIF
	1492	!
	1493	!-- A barrier has to be set, because otherwise some PEs may
	1494	!-- proceed too fast so that PE0 may receive wrong data on
	1495	!-- tag 0
	1496	CALL MPI_BARRIER( comm2d, ierr )
[1]	1497	ENDIF
[493]	1498
[1]	1499	ENDIF
	1500	#else
	1501	#if defined( __netcdf )
[1327]	1502	IF ( two_d ) THEN
	1503	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
	1504	id_var_do2d(av,if), &
[2512]	1505	local_2d(nxl:nxr,nys:nyn), &
[1327]	1506	start = (/ 1, 1, 1, do2d_xy_time_count(av) /), &
[2512]	1507	count = (/ nx+1, ny+1, 1, 1 /) )
[1327]	1508	ELSE
	1509	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
	1510	id_var_do2d(av,if), &
[2512]	1511	local_2d(nxl:nxr,nys:nyn), &
[1327]	1512	start = (/ 1, 1, is, do2d_xy_time_count(av) /), &
[2512]	1513	count = (/ nx+1, ny+1, 1, 1 /) )
[1]	1514	ENDIF
[1783]	1515	CALL netcdf_handle_error( 'data_output_2d', 447 )
[1]	1516	#endif
	1517	#endif
[2277]	1518
[1]	1519	!
	1520	!-- For 2D-arrays (e.g. u*) only one cross-section is available.
	1521	!-- Hence exit loop of output levels.
	1522	IF ( two_d ) THEN
[1703]	1523	IF ( netcdf_data_format < 5 ) two_d = .FALSE.
[1]	1524	EXIT loop1
	1525	ENDIF
	1526
	1527	CASE ( 'xz' )
	1528	!
[1308]	1529	!-- Update the netCDF xz cross section time axis.
	1530	!-- In case of parallel output, this is only done by PE0
	1531	!-- to increase the performance.
	1532	IF ( simulated_time /= do2d_xz_last_time(av) ) THEN
	1533	do2d_xz_time_count(av) = do2d_xz_time_count(av) + 1
	1534	do2d_xz_last_time(av) = simulated_time
	1535	IF ( myid == 0 ) THEN
[1327]	1536	IF ( .NOT. data_output_2d_on_each_pe &
	1537	.OR. netcdf_data_format > 4 ) &
[493]	1538	THEN
[1]	1539	#if defined( __netcdf )
	1540	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
	1541	id_var_time_xz(av), &
[291]	1542	(/ time_since_reference_point /), &
[1]	1543	start = (/ do2d_xz_time_count(av) /), &
	1544	count = (/ 1 /) )
[1783]	1545	CALL netcdf_handle_error( 'data_output_2d', 56 )
[1]	1546	#endif
	1547	ENDIF
	1548	ENDIF
	1549	ENDIF
[667]	1550
[1]	1551	!
	1552	!-- If required, carry out averaging along y
[1960]	1553	IF ( section(is,s_ind) == -1 ) THEN
[1]	1554
[2512]	1555	ALLOCATE( local_2d_l(nxl:nxr,nzb_do:nzt_do) )
[1353]	1556	local_2d_l = 0.0_wp
[2512]	1557	ngp = ( nxr-nxl + 1 ) * ( nzt_do-nzb_do + 1 )
[1]	1558	!
	1559	!-- First local averaging on the PE
[1551]	1560	DO k = nzb_do, nzt_do
[1]	1561	DO j = nys, nyn
[2512]	1562	DO i = nxl, nxr
[1320]	1563	local_2d_l(i,k) = local_2d_l(i,k) + &
[1]	1564	local_pf(i,j,k)
	1565	ENDDO
	1566	ENDDO
	1567	ENDDO
	1568	#if defined( __parallel )
	1569	!
	1570	!-- Now do the averaging over all PEs along y
[622]	1571	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
[2512]	1572	CALL MPI_ALLREDUCE( local_2d_l(nxl,nzb_do), &
	1573	local_2d(nxl,nzb_do), ngp, MPI_REAL, &
[1]	1574	MPI_SUM, comm1dy, ierr )
	1575	#else
	1576	local_2d = local_2d_l
	1577	#endif
[1353]	1578	local_2d = local_2d / ( ny + 1.0_wp )
[1]	1579
	1580	DEALLOCATE( local_2d_l )
	1581
	1582	ELSE
	1583	!
	1584	!-- Just store the respective section on the local array
	1585	!-- (but only if it is available on this PE!)
[1960]	1586	IF ( section(is,s_ind) >= nys .AND. section(is,s_ind) <= nyn ) &
[1]	1587	THEN
[1960]	1588	local_2d = local_pf(:,section(is,s_ind),nzb_do:nzt_do)
[1]	1589	ENDIF
	1590
	1591	ENDIF
	1592
	1593	#if defined( __parallel )
[1327]	1594	IF ( netcdf_data_format > 4 ) THEN
[1]	1595	!
[1031]	1596	!-- Output in netCDF4/HDF5 format.
[493]	1597	!-- Output only on those PEs where the respective cross
	1598	!-- sections reside. Cross sections averaged along y are
	1599	!-- output on the respective first PE along y (myidy=0).
[1960]	1600	IF ( ( section(is,s_ind) >= nys .AND. &
	1601	section(is,s_ind) <= nyn ) .OR. &
	1602	( section(is,s_ind) == -1 .AND. myidy == 0 ) ) THEN
[1]	1603	#if defined( __netcdf )
[493]	1604	!
[1308]	1605	!-- For parallel output, all cross sections are first
	1606	!-- stored here on a local array and will be written to the
	1607	!-- output file afterwards to increase the performance.
[2512]	1608	DO i = nxl, nxr
[1551]	1609	DO k = nzb_do, nzt_do
[1308]	1610	local_2d_sections_l(i,is,k) = local_2d(i,k)
	1611	ENDDO
	1612	ENDDO
[1]	1613	#endif
	1614	ENDIF
	1615
	1616	ELSE
	1617
[493]	1618	IF ( data_output_2d_on_each_pe ) THEN
[1]	1619	!
[493]	1620	!-- Output of partial arrays on each PE. If the cross
	1621	!-- section does not reside on the PE, output special
	1622	!-- index values.
	1623	#if defined( __netcdf )
[1327]	1624	IF ( myid == 0 ) THEN
[1320]	1625	WRITE ( 22 ) time_since_reference_point, &
[493]	1626	do2d_xz_time_count(av), av
	1627	ENDIF
	1628	#endif
[759]	1629	DO i = 0, io_blocks-1
	1630	IF ( i == io_group ) THEN
[1960]	1631	IF ( ( section(is,s_ind) >= nys .AND. &
	1632	section(is,s_ind) <= nyn ) .OR. &
	1633	( section(is,s_ind) == -1 .AND. &
[1320]	1634	nys-1 == -1 ) ) &
[759]	1635	THEN
[2512]	1636	WRITE (22) nxl, nxr, nzb_do, nzt_do, nzb, nzt+1
[759]	1637	WRITE (22) local_2d
	1638	ELSE
[1551]	1639	WRITE (22) -1, -1, -1, -1, -1, -1
[759]	1640	ENDIF
	1641	ENDIF
	1642	#if defined( __parallel )
	1643	CALL MPI_BARRIER( comm2d, ierr )
	1644	#endif
	1645	ENDDO
[493]	1646
	1647	ELSE
[1]	1648	!
[493]	1649	!-- PE0 receives partial arrays from all processors of the
	1650	!-- respective cross section and outputs them. Here a
	1651	!-- barrier has to be set, because otherwise
	1652	!-- "-MPI- FATAL: Remote protocol queue full" may occur.
	1653	CALL MPI_BARRIER( comm2d, ierr )
	1654
[2512]	1655	ngp = ( nxr-nxl + 1 ) * ( nzt_do-nzb_do + 1 )
[493]	1656	IF ( myid == 0 ) THEN
[1]	1657	!
[493]	1658	!-- Local array can be relocated directly.
[1960]	1659	IF ( ( section(is,s_ind) >= nys .AND. &
	1660	section(is,s_ind) <= nyn ) .OR. &
	1661	( section(is,s_ind) == -1 .AND. &
	1662	nys-1 == -1 ) ) THEN
[2512]	1663	total_2d(nxl:nxr,nzb_do:nzt_do) = local_2d
[493]	1664	ENDIF
[1]	1665	!
[493]	1666	!-- Receive data from all other PEs.
	1667	DO n = 1, numprocs-1
	1668	!
	1669	!-- Receive index limits first, then array.
	1670	!-- Index limits are received in arbitrary order from
	1671	!-- the PEs.
[1320]	1672	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
	1673	MPI_ANY_SOURCE, 0, comm2d, &
[1]	1674	status, ierr )
[493]	1675	!
	1676	!-- Not all PEs have data for XZ-cross-section.
	1677	IF ( ind(1) /= -9999 ) THEN
	1678	sender = status(MPI_SOURCE)
	1679	DEALLOCATE( local_2d )
[1320]	1680	ALLOCATE( local_2d(ind(1):ind(2), &
[493]	1681	ind(3):ind(4)) )
	1682	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
	1683	MPI_REAL, sender, 1, comm2d, &
	1684	status, ierr )
[1320]	1685	total_2d(ind(1):ind(2),ind(3):ind(4)) = &
[493]	1686	local_2d
	1687	ENDIF
	1688	ENDDO
	1689	!
	1690	!-- Relocate the local array for the next loop increment
	1691	DEALLOCATE( local_2d )
[2512]	1692	ALLOCATE( local_2d(nxl:nxr,nzb_do:nzt_do) )
[1]	1693
	1694	#if defined( __netcdf )
[2512]	1695	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
	1696	id_var_do2d(av,if), &
	1697	total_2d(0:nx,nzb_do:nzt_do), &
	1698	start = (/ 1, is, 1, do2d_xz_time_count(av) /), &
	1699	count = (/ nx+1, 1, nzt_do-nzb_do+1, 1 /) )
[1783]	1700	CALL netcdf_handle_error( 'data_output_2d', 58 )
[1]	1701	#endif
	1702
[493]	1703	ELSE
[1]	1704	!
[493]	1705	!-- If the cross section resides on the PE, send the
	1706	!-- local index limits, otherwise send -9999 to PE0.
[1960]	1707	IF ( ( section(is,s_ind) >= nys .AND. &
	1708	section(is,s_ind) <= nyn ) .OR. &
	1709	( section(is,s_ind) == -1 .AND. nys-1 == -1 ) ) &
[493]	1710	THEN
[2512]	1711	ind(1) = nxl; ind(2) = nxr
[1551]	1712	ind(3) = nzb_do; ind(4) = nzt_do
[493]	1713	ELSE
	1714	ind(1) = -9999; ind(2) = -9999
	1715	ind(3) = -9999; ind(4) = -9999
	1716	ENDIF
[1320]	1717	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
[493]	1718	comm2d, ierr )
	1719	!
	1720	!-- If applicable, send data to PE0.
	1721	IF ( ind(1) /= -9999 ) THEN
[2512]	1722	CALL MPI_SEND( local_2d(nxl,nzb_do), ngp, &
[493]	1723	MPI_REAL, 0, 1, comm2d, ierr )
	1724	ENDIF
[1]	1725	ENDIF
	1726	!
[493]	1727	!-- A barrier has to be set, because otherwise some PEs may
	1728	!-- proceed too fast so that PE0 may receive wrong data on
	1729	!-- tag 0
	1730	CALL MPI_BARRIER( comm2d, ierr )
[1]	1731	ENDIF
[493]	1732
[1]	1733	ENDIF
	1734	#else
	1735	#if defined( __netcdf )
[1327]	1736	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
	1737	id_var_do2d(av,if), &
[2512]	1738	local_2d(nxl:nxr,nzb_do:nzt_do), &
[1327]	1739	start = (/ 1, is, 1, do2d_xz_time_count(av) /), &
[2512]	1740	count = (/ nx+1, 1, nzt_do-nzb_do+1, 1 /) )
[1783]	1741	CALL netcdf_handle_error( 'data_output_2d', 451 )
[1]	1742	#endif
	1743	#endif
	1744
	1745	CASE ( 'yz' )
	1746	!
[1308]	1747	!-- Update the netCDF yz cross section time axis.
	1748	!-- In case of parallel output, this is only done by PE0
	1749	!-- to increase the performance.
	1750	IF ( simulated_time /= do2d_yz_last_time(av) ) THEN
	1751	do2d_yz_time_count(av) = do2d_yz_time_count(av) + 1
	1752	do2d_yz_last_time(av) = simulated_time
	1753	IF ( myid == 0 ) THEN
[1327]	1754	IF ( .NOT. data_output_2d_on_each_pe &
	1755	.OR. netcdf_data_format > 4 ) &
[493]	1756	THEN
[1]	1757	#if defined( __netcdf )
	1758	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
	1759	id_var_time_yz(av), &
[291]	1760	(/ time_since_reference_point /), &
[1]	1761	start = (/ do2d_yz_time_count(av) /), &
	1762	count = (/ 1 /) )
[1783]	1763	CALL netcdf_handle_error( 'data_output_2d', 59 )
[1]	1764	#endif
	1765	ENDIF
	1766	ENDIF
[1308]	1767	ENDIF
[493]	1768
[1]	1769	!
	1770	!-- If required, carry out averaging along x
[1960]	1771	IF ( section(is,s_ind) == -1 ) THEN
[1]	1772
[2512]	1773	ALLOCATE( local_2d_l(nys:nyn,nzb_do:nzt_do) )
[1353]	1774	local_2d_l = 0.0_wp
[2512]	1775	ngp = ( nyn-nys+1 ) * ( nzt_do-nzb_do+1 )
[1]	1776	!
	1777	!-- First local averaging on the PE
[1551]	1778	DO k = nzb_do, nzt_do
[2512]	1779	DO j = nys, nyn
[1]	1780	DO i = nxl, nxr
[1320]	1781	local_2d_l(j,k) = local_2d_l(j,k) + &
[1]	1782	local_pf(i,j,k)
	1783	ENDDO
	1784	ENDDO
	1785	ENDDO
	1786	#if defined( __parallel )
	1787	!
	1788	!-- Now do the averaging over all PEs along x
[622]	1789	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
[2512]	1790	CALL MPI_ALLREDUCE( local_2d_l(nys,nzb_do), &
	1791	local_2d(nys,nzb_do), ngp, MPI_REAL, &
[1]	1792	MPI_SUM, comm1dx, ierr )
	1793	#else
	1794	local_2d = local_2d_l
	1795	#endif
[1353]	1796	local_2d = local_2d / ( nx + 1.0_wp )
[1]	1797
	1798	DEALLOCATE( local_2d_l )
	1799
	1800	ELSE
	1801	!
	1802	!-- Just store the respective section on the local array
	1803	!-- (but only if it is available on this PE!)
[1960]	1804	IF ( section(is,s_ind) >= nxl .AND. section(is,s_ind) <= nxr ) &
[1]	1805	THEN
[1960]	1806	local_2d = local_pf(section(is,s_ind),:,nzb_do:nzt_do)
[1]	1807	ENDIF
	1808
	1809	ENDIF
	1810
	1811	#if defined( __parallel )
[1327]	1812	IF ( netcdf_data_format > 4 ) THEN
[1]	1813	!
[1031]	1814	!-- Output in netCDF4/HDF5 format.
[493]	1815	!-- Output only on those PEs where the respective cross
	1816	!-- sections reside. Cross sections averaged along x are
	1817	!-- output on the respective first PE along x (myidx=0).
[1960]	1818	IF ( ( section(is,s_ind) >= nxl .AND. &
	1819	section(is,s_ind) <= nxr ) .OR. &
	1820	( section(is,s_ind) == -1 .AND. myidx == 0 ) ) THEN
[1]	1821	#if defined( __netcdf )
[493]	1822	!
[1308]	1823	!-- For parallel output, all cross sections are first
	1824	!-- stored here on a local array and will be written to the
	1825	!-- output file afterwards to increase the performance.
[2512]	1826	DO j = nys, nyn
[1551]	1827	DO k = nzb_do, nzt_do
[1308]	1828	local_2d_sections_l(is,j,k) = local_2d(j,k)
	1829	ENDDO
	1830	ENDDO
[1]	1831	#endif
	1832	ENDIF
	1833
	1834	ELSE
	1835
[493]	1836	IF ( data_output_2d_on_each_pe ) THEN
[1]	1837	!
[493]	1838	!-- Output of partial arrays on each PE. If the cross
	1839	!-- section does not reside on the PE, output special
	1840	!-- index values.
	1841	#if defined( __netcdf )
[1327]	1842	IF ( myid == 0 ) THEN
[1320]	1843	WRITE ( 23 ) time_since_reference_point, &
[493]	1844	do2d_yz_time_count(av), av
	1845	ENDIF
	1846	#endif
[759]	1847	DO i = 0, io_blocks-1
	1848	IF ( i == io_group ) THEN
[1960]	1849	IF ( ( section(is,s_ind) >= nxl .AND. &
	1850	section(is,s_ind) <= nxr ) .OR. &
	1851	( section(is,s_ind) == -1 .AND. &
[1320]	1852	nxl-1 == -1 ) ) &
[759]	1853	THEN
[2512]	1854	WRITE (23) nys, nyn, nzb_do, nzt_do, nzb, nzt+1
[759]	1855	WRITE (23) local_2d
	1856	ELSE
[1551]	1857	WRITE (23) -1, -1, -1, -1, -1, -1
[759]	1858	ENDIF
	1859	ENDIF
	1860	#if defined( __parallel )
	1861	CALL MPI_BARRIER( comm2d, ierr )
	1862	#endif
	1863	ENDDO
[493]	1864
	1865	ELSE
[1]	1866	!
[493]	1867	!-- PE0 receives partial arrays from all processors of the
	1868	!-- respective cross section and outputs them. Here a
	1869	!-- barrier has to be set, because otherwise
	1870	!-- "-MPI- FATAL: Remote protocol queue full" may occur.
	1871	CALL MPI_BARRIER( comm2d, ierr )
	1872
[2512]	1873	ngp = ( nyn-nys+1 ) * ( nzt_do-nzb_do+1 )
[493]	1874	IF ( myid == 0 ) THEN
[1]	1875	!
[493]	1876	!-- Local array can be relocated directly.
[1960]	1877	IF ( ( section(is,s_ind) >= nxl .AND. &
	1878	section(is,s_ind) <= nxr ) .OR. &
	1879	( section(is,s_ind) == -1 .AND. nxl-1 == -1 ) ) &
[493]	1880	THEN
[2512]	1881	total_2d(nys:nyn,nzb_do:nzt_do) = local_2d
[493]	1882	ENDIF
[1]	1883	!
[493]	1884	!-- Receive data from all other PEs.
	1885	DO n = 1, numprocs-1
	1886	!
	1887	!-- Receive index limits first, then array.
	1888	!-- Index limits are received in arbitrary order from
	1889	!-- the PEs.
[1320]	1890	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
	1891	MPI_ANY_SOURCE, 0, comm2d, &
[1]	1892	status, ierr )
[493]	1893	!
	1894	!-- Not all PEs have data for YZ-cross-section.
	1895	IF ( ind(1) /= -9999 ) THEN
	1896	sender = status(MPI_SOURCE)
	1897	DEALLOCATE( local_2d )
[1320]	1898	ALLOCATE( local_2d(ind(1):ind(2), &
[493]	1899	ind(3):ind(4)) )
	1900	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
	1901	MPI_REAL, sender, 1, comm2d, &
	1902	status, ierr )
[1320]	1903	total_2d(ind(1):ind(2),ind(3):ind(4)) = &
[493]	1904	local_2d
	1905	ENDIF
	1906	ENDDO
	1907	!
	1908	!-- Relocate the local array for the next loop increment
	1909	DEALLOCATE( local_2d )
[2512]	1910	ALLOCATE( local_2d(nys:nyn,nzb_do:nzt_do) )
[1]	1911
	1912	#if defined( __netcdf )
[2512]	1913	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
	1914	id_var_do2d(av,if), &
	1915	total_2d(0:ny,nzb_do:nzt_do), &
	1916	start = (/ is, 1, 1, do2d_yz_time_count(av) /), &
	1917	count = (/ 1, ny+1, nzt_do-nzb_do+1, 1 /) )
[1783]	1918	CALL netcdf_handle_error( 'data_output_2d', 61 )
[1]	1919	#endif
	1920
[493]	1921	ELSE
[1]	1922	!
[493]	1923	!-- If the cross section resides on the PE, send the
	1924	!-- local index limits, otherwise send -9999 to PE0.
[1960]	1925	IF ( ( section(is,s_ind) >= nxl .AND. &
	1926	section(is,s_ind) <= nxr ) .OR. &
	1927	( section(is,s_ind) == -1 .AND. nxl-1 == -1 ) ) &
[493]	1928	THEN
[2512]	1929	ind(1) = nys; ind(2) = nyn
[1551]	1930	ind(3) = nzb_do; ind(4) = nzt_do
[493]	1931	ELSE
	1932	ind(1) = -9999; ind(2) = -9999
	1933	ind(3) = -9999; ind(4) = -9999
	1934	ENDIF
[1320]	1935	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
[493]	1936	comm2d, ierr )
	1937	!
	1938	!-- If applicable, send data to PE0.
	1939	IF ( ind(1) /= -9999 ) THEN
[2512]	1940	CALL MPI_SEND( local_2d(nys,nzb_do), ngp, &
[493]	1941	MPI_REAL, 0, 1, comm2d, ierr )
	1942	ENDIF
[1]	1943	ENDIF
	1944	!
[493]	1945	!-- A barrier has to be set, because otherwise some PEs may
	1946	!-- proceed too fast so that PE0 may receive wrong data on
	1947	!-- tag 0
	1948	CALL MPI_BARRIER( comm2d, ierr )
[1]	1949	ENDIF
[493]	1950
[1]	1951	ENDIF
	1952	#else
	1953	#if defined( __netcdf )
[1327]	1954	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
	1955	id_var_do2d(av,if), &
[2512]	1956	local_2d(nys:nyn,nzb_do:nzt_do), &
[1327]	1957	start = (/ is, 1, 1, do2d_xz_time_count(av) /), &
[2512]	1958	count = (/ 1, ny+1, nzt_do-nzb_do+1, 1 /) )
[1783]	1959	CALL netcdf_handle_error( 'data_output_2d', 452 )
[1]	1960	#endif
	1961	#endif
	1962
	1963	END SELECT
	1964
	1965	is = is + 1
	1966	ENDDO loop1
	1967
[1308]	1968	!
	1969	!-- For parallel output, all data were collected before on a local array
	1970	!-- and are written now to the netcdf file. This must be done to increase
	1971	!-- the performance of the parallel output.
	1972	#if defined( __netcdf )
[1327]	1973	IF ( netcdf_data_format > 4 ) THEN
[1308]	1974
	1975	SELECT CASE ( mode )
	1976
	1977	CASE ( 'xy' )
	1978	IF ( two_d ) THEN
[1703]	1979	nis = 1
	1980	two_d = .FALSE.
[1308]	1981	ELSE
[1703]	1982	nis = ns
[1308]	1983	ENDIF
	1984	!
	1985	!-- Do not output redundant ghost point data except for the
	1986	!-- boundaries of the total domain.
[2512]	1987	! IF ( nxr == nx .AND. nyn /= ny ) THEN
	1988	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
	1989	! id_var_do2d(av,if), &
	1990	! local_2d_sections(nxl:nxr+1, &
	1991	! nys:nyn,1:nis), &
	1992	! start = (/ nxl+1, nys+1, 1, &
	1993	! do2d_xy_time_count(av) /), &
	1994	! count = (/ nxr-nxl+2, &
	1995	! nyn-nys+1, nis, 1 &
	1996	! /) )
	1997	! ELSEIF ( nxr /= nx .AND. nyn == ny ) THEN
	1998	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
	1999	! id_var_do2d(av,if), &
	2000	! local_2d_sections(nxl:nxr, &
	2001	! nys:nyn+1,1:nis), &
	2002	! start = (/ nxl+1, nys+1, 1, &
	2003	! do2d_xy_time_count(av) /), &
	2004	! count = (/ nxr-nxl+1, &
	2005	! nyn-nys+2, nis, 1 &
	2006	! /) )
	2007	! ELSEIF ( nxr == nx .AND. nyn == ny ) THEN
	2008	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
	2009	! id_var_do2d(av,if), &
	2010	! local_2d_sections(nxl:nxr+1, &
	2011	! nys:nyn+1,1:nis), &
	2012	! start = (/ nxl+1, nys+1, 1, &
	2013	! do2d_xy_time_count(av) /), &
	2014	! count = (/ nxr-nxl+2, &
	2015	! nyn-nys+2, nis, 1 &
	2016	! /) )
	2017	! ELSE
[1308]	2018	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
	2019	id_var_do2d(av,if), &
	2020	local_2d_sections(nxl:nxr, &
[1703]	2021	nys:nyn,1:nis), &
[1308]	2022	start = (/ nxl+1, nys+1, 1, &
	2023	do2d_xy_time_count(av) /), &
	2024	count = (/ nxr-nxl+1, &
[1703]	2025	nyn-nys+1, nis, 1 &
[1308]	2026	/) )
[2512]	2027	! ENDIF
[1308]	2028
[1783]	2029	CALL netcdf_handle_error( 'data_output_2d', 55 )
[1308]	2030
	2031	CASE ( 'xz' )
	2032	!
	2033	!-- First, all PEs get the information of all cross-sections.
	2034	!-- Then the data are written to the output file by all PEs
	2035	!-- while NF90_COLLECTIVE is set in subroutine
	2036	!-- define_netcdf_header. Although redundant information are
	2037	!-- written to the output file in that case, the performance
	2038	!-- is significantly better compared to the case where only
	2039	!-- the first row of PEs in x-direction (myidx = 0) is given
	2040	!-- the output while NF90_INDEPENDENT is set.
	2041	IF ( npey /= 1 ) THEN
	2042
	2043	#if defined( __parallel )
	2044	!
	2045	!-- Distribute data over all PEs along y
[2512]	2046	ngp = ( nxr-nxl+1 ) * ( nzt_do-nzb_do+1 ) * ns
[1308]	2047	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
[2512]	2048	CALL MPI_ALLREDUCE( local_2d_sections_l(nxl,1,nzb_do), &
	2049	local_2d_sections(nxl,1,nzb_do), &
[1308]	2050	ngp, MPI_REAL, MPI_SUM, comm1dy, &
	2051	ierr )
	2052	#else
	2053	local_2d_sections = local_2d_sections_l
	2054	#endif
	2055	ENDIF
	2056	!
	2057	!-- Do not output redundant ghost point data except for the
	2058	!-- boundaries of the total domain.
[2512]	2059	! IF ( nxr == nx ) THEN
	2060	! nc_stat = NF90_PUT_VAR( id_set_xz(av), &
	2061	! id_var_do2d(av,if), &
	2062	! local_2d_sections(nxl:nxr+1,1:ns, &
	2063	! nzb_do:nzt_do), &
	2064	! start = (/ nxl+1, 1, 1, &
	2065	! do2d_xz_time_count(av) /), &
	2066	! count = (/ nxr-nxl+2, ns, nzt_do-nzb_do+1, &
	2067	! 1 /) )
	2068	! ELSE
[1308]	2069	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
	2070	id_var_do2d(av,if), &
	2071	local_2d_sections(nxl:nxr,1:ns, &
[1551]	2072	nzb_do:nzt_do), &
[1308]	2073	start = (/ nxl+1, 1, 1, &
	2074	do2d_xz_time_count(av) /), &
[1551]	2075	count = (/ nxr-nxl+1, ns, nzt_do-nzb_do+1, &
[1308]	2076	1 /) )
[2512]	2077	! ENDIF
[1308]	2078
[1783]	2079	CALL netcdf_handle_error( 'data_output_2d', 57 )
[1308]	2080
	2081	CASE ( 'yz' )
	2082	!
	2083	!-- First, all PEs get the information of all cross-sections.
	2084	!-- Then the data are written to the output file by all PEs
	2085	!-- while NF90_COLLECTIVE is set in subroutine
	2086	!-- define_netcdf_header. Although redundant information are
	2087	!-- written to the output file in that case, the performance
	2088	!-- is significantly better compared to the case where only
	2089	!-- the first row of PEs in y-direction (myidy = 0) is given
	2090	!-- the output while NF90_INDEPENDENT is set.
	2091	IF ( npex /= 1 ) THEN
	2092
	2093	#if defined( __parallel )
	2094	!
	2095	!-- Distribute data over all PEs along x
[2512]	2096	ngp = ( nyn-nys+1 ) * ( nzt-nzb + 2 ) * ns
[1308]	2097	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
[2512]	2098	CALL MPI_ALLREDUCE( local_2d_sections_l(1,nys,nzb_do), &
	2099	local_2d_sections(1,nys,nzb_do), &
[1308]	2100	ngp, MPI_REAL, MPI_SUM, comm1dx, &
	2101	ierr )
	2102	#else
	2103	local_2d_sections = local_2d_sections_l
	2104	#endif
	2105	ENDIF
	2106	!
	2107	!-- Do not output redundant ghost point data except for the
	2108	!-- boundaries of the total domain.
[2512]	2109	! IF ( nyn == ny ) THEN
	2110	! nc_stat = NF90_PUT_VAR( id_set_yz(av), &
	2111	! id_var_do2d(av,if), &
	2112	! local_2d_sections(1:ns, &
	2113	! nys:nyn+1,nzb_do:nzt_do), &
	2114	! start = (/ 1, nys+1, 1, &
	2115	! do2d_yz_time_count(av) /), &
	2116	! count = (/ ns, nyn-nys+2, &
	2117	! nzt_do-nzb_do+1, 1 /) )
	2118	! ELSE
[1308]	2119	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
	2120	id_var_do2d(av,if), &
	2121	local_2d_sections(1:ns,nys:nyn, &
[1551]	2122	nzb_do:nzt_do), &
[1308]	2123	start = (/ 1, nys+1, 1, &
	2124	do2d_yz_time_count(av) /), &
	2125	count = (/ ns, nyn-nys+1, &
[1551]	2126	nzt_do-nzb_do+1, 1 /) )
[2512]	2127	! ENDIF
[1308]	2128
[1783]	2129	CALL netcdf_handle_error( 'data_output_2d', 60 )
[1308]	2130
	2131	CASE DEFAULT
	2132	message_string = 'unknown cross-section: ' // TRIM( mode )
	2133	CALL message( 'data_output_2d', 'PA0180', 1, 2, 0, 6, 0 )
	2134
	2135	END SELECT
	2136
	2137	ENDIF
[1311]	2138	#endif
[1]	2139	ENDIF
	2140
	2141	if = if + 1
	2142	l = MAX( 2, LEN_TRIM( do2d(av,if) ) )
	2143	do2d_mode = do2d(av,if)(l-1:l)
	2144
	2145	ENDDO
	2146
	2147	!
	2148	!-- Deallocate temporary arrays.
	2149	IF ( ALLOCATED( level_z ) ) DEALLOCATE( level_z )
[1308]	2150	IF ( netcdf_data_format > 4 ) THEN
	2151	DEALLOCATE( local_pf, local_2d, local_2d_sections )
	2152	IF( mode == 'xz' .OR. mode == 'yz' ) DEALLOCATE( local_2d_sections_l )
	2153	ENDIF
[1]	2154	#if defined( __parallel )
	2155	IF ( .NOT. data_output_2d_on_each_pe .AND. myid == 0 ) THEN
	2156	DEALLOCATE( total_2d )
	2157	ENDIF
	2158	#endif
	2159
	2160	!
	2161	!-- Close plot output file.
[1960]	2162	file_id = 20 + s_ind
[1]	2163
	2164	IF ( data_output_2d_on_each_pe ) THEN
[759]	2165	DO i = 0, io_blocks-1
	2166	IF ( i == io_group ) THEN
	2167	CALL close_file( file_id )
	2168	ENDIF
	2169	#if defined( __parallel )
	2170	CALL MPI_BARRIER( comm2d, ierr )
	2171	#endif
	2172	ENDDO
[1]	2173	ELSE
	2174	IF ( myid == 0 ) CALL close_file( file_id )
	2175	ENDIF
	2176
[1318]	2177	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
[1]	2178
	2179	END SUBROUTINE data_output_2d

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

Download in other formats:

| Impressum | ©Leibniz Universität Hannover |