Home

Context Navigation

source: palm/trunk/SOURCE/data_output_2d.f90 @ 3049

Last change on this file since 3049 was 3049, checked in by Giersch, 6 years ago
Revision history corrected
Property svn:keywords set to `Id`
File size: 97.9 KB

Line
1	!> @file data_output_2d.f90
2	!------------------------------------------------------------------------------!
3	! This file is part of the PALM model system.
4	!
5	! PALM is free software: you can redistribute it and/or modify it under the
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.
9	!
10	! PALM is distributed in the hope that it will be useful, but WITHOUT ANY
11	! WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
12	! A PARTICULAR PURPOSE. See the GNU General Public License for more details.
13	!
14	! You should have received a copy of the GNU General Public License along with
15	! PALM. If not, see <http://www.gnu.org/licenses/>.
16	!
17	! Copyright 1997-2018 Leibniz Universitaet Hannover
18	!------------------------------------------------------------------------------!
19	!
20	! Current revisions:
21	! -----------------
22	!
23	!
24	! Former revisions:
25	! -----------------
26	! $Id: data_output_2d.f90 3049 2018-05-29 13:52:36Z Giersch $
27	! Error messages revised
28	!
29	! 3045 2018-05-28 07:55:41Z Giersch
30	! Error messages revised
31	!
32	! 3014 2018-05-09 08:42:38Z maronga
33	! Added nzb_do and nzt_do for some modules for 2d output
34	!
35	! 3004 2018-04-27 12:33:25Z Giersch
36	! precipitation_rate removed, case prr*_xy removed, to_be_resorted have to point
37	! to ql_vp_av and not to ql_vp, allocation checks implemented (averaged data
38	! will be assigned to fill values if no allocation happened so far)
39	!
40	! 2963 2018-04-12 14:47:44Z suehring
41	! Introduce index for vegetation/wall, pavement/green-wall and water/window
42	! surfaces, for clearer access of surface fraction, albedo, emissivity, etc. .
43	!
44	! 2817 2018-02-19 16:32:21Z knoop
45	! Preliminary gust module interface implemented
46	!
47	! 2805 2018-02-14 17:00:09Z suehring
48	! Consider also default-type surfaces for surface temperature output.
49	!
50	! 2797 2018-02-08 13:24:35Z suehring
51	! Enable output of ground-heat flux also at urban surfaces.
52	!
53	! 2743 2018-01-12 16:03:39Z suehring
54	! In case of natural- and urban-type surfaces output surfaces fluxes in W/m2.
55	!
56	! 2742 2018-01-12 14:59:47Z suehring
57	! Enable output of surface temperature
58	!
59	! 2735 2018-01-11 12:01:27Z suehring
60	! output of r_a moved from land-surface to consider also urban-type surfaces
61	!
62	! 2718 2018-01-02 08:49:38Z maronga
63	! Corrected "Former revisions" section
64	!
65	! 2696 2017-12-14 17:12:51Z kanani
66	! Change in file header (GPL part)
67	! Implementation of uv exposure model (FK)
68	! Implementation of turbulence_closure_mod (TG)
69	! Set fill values at topography grid points or e.g. non-natural-type surface
70	! in case of LSM output (MS)
71	!
72	! 2512 2017-10-04 08:26:59Z raasch
73	! upper bounds of cross section output changed from nx+1,ny+1 to nx,ny
74	! no output of ghost layer data
75	!
76	! 2292 2017-06-20 09:51:42Z schwenkel
77	! Implementation of new microphysic scheme: cloud_scheme = 'morrison'
78	! includes two more prognostic equations for cloud drop concentration (nc)
79	! and cloud water content (qc).
80	!
81	! 2277 2017-06-12 10:47:51Z kanani
82	! Removed unused variables do2d_xy_n, do2d_xz_n, do2d_yz_n
83	!
84	! 2233 2017-05-30 18:08:54Z suehring
85	!
86	! 2232 2017-05-30 17:47:52Z suehring
87	! Adjustments to new surface concept
88	!
89	!
90	! 2190 2017-03-21 12:16:43Z raasch
91	! bugfix for r2031: string rho replaced by rho_ocean
92	!
93	! 2031 2016-10-21 15:11:58Z knoop
94	! renamed variable rho to rho_ocean and rho_av to rho_ocean_av
95	!
96	! 2000 2016-08-20 18:09:15Z knoop
97	! Forced header and separation lines into 80 columns
98	!
99	! 1980 2016-07-29 15:51:57Z suehring
100	! Bugfix, in order to steer user-defined output, setting flag found explicitly
101	! to .F.
102	!
103	! 1976 2016-07-27 13:28:04Z maronga
104	! Output of radiation quantities is now done directly in the respective module
105	!
106	! 1972 2016-07-26 07:52:02Z maronga
107	! Output of land surface quantities is now done directly in the respective
108	! module
109	!
110	! 1960 2016-07-12 16:34:24Z suehring
111	! Scalar surface flux added
112	! Rename INTEGER variable s into s_ind, as s is already assigned to scalar
113	!
114	! 1849 2016-04-08 11:33:18Z hoffmann
115	! precipitation_amount, precipitation_rate, prr moved to arrays_3d
116	!
117	! 1822 2016-04-07 07:49:42Z hoffmann
118	! Output of bulk cloud physics simplified.
119	!
120	! 1788 2016-03-10 11:01:04Z maronga
121	! Added output of z0q
122	!
123	! 1783 2016-03-06 18:36:17Z raasch
124	! name change of netcdf routines and module + related changes
125	!
126	! 1745 2016-02-05 13:06:51Z gronemeier
127	! Bugfix: test if time axis limit exceeds moved to point after call of check_open
128	!
129	! 1703 2015-11-02 12:38:44Z raasch
130	! bugfix for output of single (*) xy-sections in case of parallel netcdf I/O
131	!
132	! 1701 2015-11-02 07:43:04Z maronga
133	! Bugfix in output of RRTGM data
134	!
135	! 1691 2015-10-26 16:17:44Z maronga
136	! Added output of Obukhov length (ol) and radiative heating rates for RRTMG.
137	! Formatting corrections.
138	!
139	! 1682 2015-10-07 23:56:08Z knoop
140	! Code annotations made doxygen readable
141	!
142	! 1585 2015-04-30 07:05:52Z maronga
143	! Added support for RRTMG
144	!
145	! 1555 2015-03-04 17:44:27Z maronga
146	! Added output of r_a and r_s
147	!
148	! 1551 2015-03-03 14:18:16Z maronga
149	! Added suppport for land surface model and radiation model output. In the course
150	! of this action, the limits for vertical loops have been changed (from nzb and
151	! nzt+1 to nzb_do and nzt_do, respectively in order to allow soil model output).
152	! Moreover, a new vertical grid zs was introduced.
153	!
154	! 1359 2014-04-11 17:15:14Z hoffmann
155	! New particle structure integrated.
156	!
157	! 1353 2014-04-08 15:21:23Z heinze
158	! REAL constants provided with KIND-attribute
159	!
160	! 1327 2014-03-21 11:00:16Z raasch
161	! parts concerning iso2d output removed,
162	! -netcdf output queries
163	!
164	! 1320 2014-03-20 08:40:49Z raasch
165	! ONLY-attribute added to USE-statements,
166	! kind-parameters added to all INTEGER and REAL declaration statements,
167	! kinds are defined in new module kinds,
168	! revision history before 2012 removed,
169	! comment fields (!:) to be used for variable explanations added to
170	! all variable declaration statements
171	!
172	! 1318 2014-03-17 13:35:16Z raasch
173	! barrier argument removed from cpu_log.
174	! module interfaces removed
175	!
176	! 1311 2014-03-14 12:13:39Z heinze
177	! bugfix: close #if defined( __netcdf )
178	!
179	! 1308 2014-03-13 14:58:42Z fricke
180	! +local_2d_sections, local_2d_sections_l, ns
181	! Check, if the limit of the time dimension is exceeded for parallel output
182	! To increase the performance for parallel output, the following is done:
183	! - Update of time axis is only done by PE0
184	! - Cross sections are first stored on a local array and are written
185	! collectively to the output file by all PEs.
186	!
187	! 1115 2013-03-26 18:16:16Z hoffmann
188	! ql is calculated by calc_liquid_water_content
189	!
190	! 1076 2012-12-05 08:30:18Z hoffmann
191	! Bugfix in output of ql
192	!
193	! 1065 2012-11-22 17:42:36Z hoffmann
194	! Bugfix: Output of cross sections of ql
195	!
196	! 1053 2012-11-13 17:11:03Z hoffmann
197	! +qr, nr, qc and cross sections
198	!
199	! 1036 2012-10-22 13:43:42Z raasch
200	! code put under GPL (PALM 3.9)
201	!
202	! 1031 2012-10-19 14:35:30Z raasch
203	! netCDF4 without parallel file support implemented
204	!
205	! 1007 2012-09-19 14:30:36Z franke
206	! Bugfix: missing calculation of ql_vp added
207	!
208	! 978 2012-08-09 08:28:32Z fricke
209	! +z0h
210	!
211	! Revision 1.1 1997/08/11 06:24:09 raasch
212	! Initial revision
213	!
214	!
215	! Description:
216	! ------------
217	!> Data output of cross-sections in netCDF format or binary format
218	!> to be later converted to NetCDF by helper routine combine_plot_fields.
219	!> Attention: The position of the sectional planes is still not always computed
220	!> --------- correctly. (zu is used always)!
221	!------------------------------------------------------------------------------!
222	SUBROUTINE data_output_2d( mode, av )
223
224
225	USE arrays_3d, &
226	ONLY: dzw, e, heatflux_output_conversion, nc, nr, p, pt, &
227	precipitation_amount, prr, q, qc, ql, ql_c, ql_v, ql_vp, qr, &
228	rho_ocean, s, sa, tend, u, v, vpt, w, zu, zw, &
229	waterflux_output_conversion
230
231	USE averaging
232
233	USE cloud_parameters, &
234	ONLY: cp, hyrho, l_d_cp, l_v, pt_d_t
235
236	USE control_parameters, &
237	ONLY: cloud_physics, data_output_2d_on_each_pe, data_output_xy, &
238	data_output_xz, data_output_yz, do2d, &
239	do2d_xy_last_time, do2d_xy_time_count, &
240	do2d_xz_last_time, do2d_xz_time_count, &
241	do2d_yz_last_time, do2d_yz_time_count, &
242	ibc_uv_b, io_blocks, io_group, land_surface, message_string, &
243	ntdim_2d_xy, ntdim_2d_xz, ntdim_2d_yz, &
244	psolver, section, simulated_time, simulated_time_chr, &
245	time_since_reference_point, uv_exposure
246
247	USE cpulog, &
248	ONLY: cpu_log, log_point
249
250	USE grid_variables, &
251	ONLY: dx, dy
252
253	USE gust_mod, &
254	ONLY: gust_data_output_2d, gust_module_enabled
255
256	USE indices, &
257	ONLY: nbgp, nx, nxl, nxlg, nxr, nxrg, ny, nyn, nyng, nys, nysg, nz, &
258	nzb, nzt, wall_flags_0
259
260	USE kinds
261
262	USE land_surface_model_mod, &
263	ONLY: lsm_data_output_2d, zs
264
265	#if defined( __netcdf )
266	USE NETCDF
267	#endif
268
269	USE netcdf_interface, &
270	ONLY: fill_value, id_set_xy, id_set_xz, id_set_yz, id_var_do2d, &
271	id_var_time_xy, id_var_time_xz, id_var_time_yz, nc_stat, &
272	netcdf_data_format, netcdf_handle_error
273
274	USE particle_attributes, &
275	ONLY: grid_particles, number_of_particles, particle_advection_start, &
276	particles, prt_count
277
278	USE pegrid
279
280	USE radiation_model_mod, &
281	ONLY: radiation, radiation_data_output_2d
282
283	USE surface_mod, &
284	ONLY: ind_pav_green, ind_veg_wall, ind_wat_win, surf_def_h, &
285	surf_lsm_h, surf_usm_h
286
287	USE turbulence_closure_mod, &
288	ONLY: tcm_data_output_2d
289
290	USE uv_exposure_model_mod, &
291	ONLY: uvem_data_output_2d
292
293
294	IMPLICIT NONE
295
296	CHARACTER (LEN=2) :: do2d_mode !<
297	CHARACTER (LEN=2) :: mode !<
298	CHARACTER (LEN=4) :: grid !<
299	CHARACTER (LEN=25) :: section_chr !<
300	CHARACTER (LEN=50) :: rtext !<
301
302	INTEGER(iwp) :: av !<
303	INTEGER(iwp) :: ngp !<
304	INTEGER(iwp) :: file_id !<
305	INTEGER(iwp) :: flag_nr !< number of masking flag
306	INTEGER(iwp) :: i !<
307	INTEGER(iwp) :: if !<
308	INTEGER(iwp) :: is !<
309	INTEGER(iwp) :: iis !<
310	INTEGER(iwp) :: j !<
311	INTEGER(iwp) :: k !<
312	INTEGER(iwp) :: l !<
313	INTEGER(iwp) :: layer_xy !<
314	INTEGER(iwp) :: m !<
315	INTEGER(iwp) :: n !<
316	INTEGER(iwp) :: nis !<
317	INTEGER(iwp) :: ns !<
318	INTEGER(iwp) :: nzb_do !< lower limit of the data field (usually nzb)
319	INTEGER(iwp) :: nzt_do !< upper limit of the data field (usually nzt+1)
320	INTEGER(iwp) :: psi !<
321	INTEGER(iwp) :: s_ind !<
322	INTEGER(iwp) :: sender !<
323	INTEGER(iwp) :: ind(4) !<
324
325	LOGICAL :: found !<
326	LOGICAL :: resorted !<
327	LOGICAL :: two_d !<
328
329	REAL(wp) :: mean_r !<
330	REAL(wp) :: s_r2 !<
331	REAL(wp) :: s_r3 !<
332
333	REAL(wp), DIMENSION(:), ALLOCATABLE :: level_z !<
334	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: local_2d !<
335	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: local_2d_l !<
336
337	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_pf !<
338	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_2d_sections !<
339	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_2d_sections_l !<
340
341	#if defined( __parallel )
342	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: total_2d !<
343	#endif
344	REAL(wp), DIMENSION(:,:,:), POINTER :: to_be_resorted !<
345
346	NAMELIST /LOCAL/ rtext
347
348	!
349	!-- Immediate return, if no output is requested (no respective sections
350	!-- found in parameter data_output)
351	IF ( mode == 'xy' .AND. .NOT. data_output_xy(av) ) RETURN
352	IF ( mode == 'xz' .AND. .NOT. data_output_xz(av) ) RETURN
353	IF ( mode == 'yz' .AND. .NOT. data_output_yz(av) ) RETURN
354
355	CALL cpu_log (log_point(3),'data_output_2d','start')
356
357	two_d = .FALSE. ! local variable to distinguish between output of pure 2D
358	! arrays and cross-sections of 3D arrays.
359
360	!
361	!-- Depending on the orientation of the cross-section, the respective output
362	!-- files have to be opened.
363	SELECT CASE ( mode )
364
365	CASE ( 'xy' )
366	s_ind = 1
367	ALLOCATE( level_z(nzb:nzt+1), local_2d(nxl:nxr,nys:nyn) )
368
369	IF ( netcdf_data_format > 4 ) THEN
370	ns = 1
371	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
372	ns = ns + 1
373	ENDDO
374	ns = ns - 1
375	ALLOCATE( local_2d_sections(nxl:nxr,nys:nyn,1:ns) )
376	local_2d_sections = 0.0_wp
377	ENDIF
378
379	!
380	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
381	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
382	CALL check_open( 101+av*10 )
383	ENDIF
384	IF ( data_output_2d_on_each_pe ) THEN
385	CALL check_open( 21 )
386	ELSE
387	IF ( myid == 0 ) THEN
388	#if defined( __parallel )
389	ALLOCATE( total_2d(0:nx,0:ny) )
390	#endif
391	ENDIF
392	ENDIF
393
394	CASE ( 'xz' )
395	s_ind = 2
396	ALLOCATE( local_2d(nxl:nxr,nzb:nzt+1) )
397
398	IF ( netcdf_data_format > 4 ) THEN
399	ns = 1
400	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
401	ns = ns + 1
402	ENDDO
403	ns = ns - 1
404	ALLOCATE( local_2d_sections(nxl:nxr,1:ns,nzb:nzt+1) )
405	ALLOCATE( local_2d_sections_l(nxl:nxr,1:ns,nzb:nzt+1) )
406	local_2d_sections = 0.0_wp; local_2d_sections_l = 0.0_wp
407	ENDIF
408
409	!
410	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
411	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
412	CALL check_open( 102+av*10 )
413	ENDIF
414
415	IF ( data_output_2d_on_each_pe ) THEN
416	CALL check_open( 22 )
417	ELSE
418	IF ( myid == 0 ) THEN
419	#if defined( __parallel )
420	ALLOCATE( total_2d(0:nx,nzb:nzt+1) )
421	#endif
422	ENDIF
423	ENDIF
424
425	CASE ( 'yz' )
426	s_ind = 3
427	ALLOCATE( local_2d(nys:nyn,nzb:nzt+1) )
428
429	IF ( netcdf_data_format > 4 ) THEN
430	ns = 1
431	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
432	ns = ns + 1
433	ENDDO
434	ns = ns - 1
435	ALLOCATE( local_2d_sections(1:ns,nys:nyn,nzb:nzt+1) )
436	ALLOCATE( local_2d_sections_l(1:ns,nys:nyn,nzb:nzt+1) )
437	local_2d_sections = 0.0_wp; local_2d_sections_l = 0.0_wp
438	ENDIF
439
440	!
441	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
442	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
443	CALL check_open( 103+av*10 )
444	ENDIF
445
446	IF ( data_output_2d_on_each_pe ) THEN
447	CALL check_open( 23 )
448	ELSE
449	IF ( myid == 0 ) THEN
450	#if defined( __parallel )
451	ALLOCATE( total_2d(0:ny,nzb:nzt+1) )
452	#endif
453	ENDIF
454	ENDIF
455
456	CASE DEFAULT
457	message_string = 'unknown cross-section: ' // TRIM( mode )
458	CALL message( 'data_output_2d', 'PA0180', 1, 2, 0, 6, 0 )
459
460	END SELECT
461
462	!
463	!-- For parallel netcdf output the time axis must be limited. Return, if this
464	!-- limit is exceeded. This could be the case, if the simulated time exceeds
465	!-- the given end time by the length of the given output interval.
466	IF ( netcdf_data_format > 4 ) THEN
467	IF ( mode == 'xy' .AND. do2d_xy_time_count(av) + 1 > &
468	ntdim_2d_xy(av) ) THEN
469	WRITE ( message_string, * ) 'Output of xy cross-sections is not ', &
470	'given at t=', simulated_time, '&because the', &
471	' maximum number of output time levels is exceeded.'
472	CALL message( 'data_output_2d', 'PA0384', 0, 1, 0, 6, 0 )
473	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
474	RETURN
475	ENDIF
476	IF ( mode == 'xz' .AND. do2d_xz_time_count(av) + 1 > &
477	ntdim_2d_xz(av) ) THEN
478	WRITE ( message_string, * ) 'Output of xz cross-sections is not ', &
479	'given at t=', simulated_time, '&because the', &
480	' maximum number of output time levels is exceeded.'
481	CALL message( 'data_output_2d', 'PA0385', 0, 1, 0, 6, 0 )
482	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
483	RETURN
484	ENDIF
485	IF ( mode == 'yz' .AND. do2d_yz_time_count(av) + 1 > &
486	ntdim_2d_yz(av) ) THEN
487	WRITE ( message_string, * ) 'Output of yz cross-sections is not ', &
488	'given at t=', simulated_time, '&because the', &
489	' maximum number of output time levels is exceeded.'
490	CALL message( 'data_output_2d', 'PA0386', 0, 1, 0, 6, 0 )
491	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
492	RETURN
493	ENDIF
494	ENDIF
495
496	!
497	!-- Allocate a temporary array for resorting (kji -> ijk).
498	ALLOCATE( local_pf(nxl:nxr,nys:nyn,nzb:nzt+1) )
499	local_pf = 0.0
500
501	!
502	!-- Loop of all variables to be written.
503	!-- Output dimensions chosen
504	if = 1
505	l = MAX( 2, LEN_TRIM( do2d(av,if) ) )
506	do2d_mode = do2d(av,if)(l-1:l)
507
508	DO WHILE ( do2d(av,if)(1:1) /= ' ' )
509
510	IF ( do2d_mode == mode ) THEN
511	!
512	!-- Set flag to steer output of radiation, land-surface, or user-defined
513	!-- quantities
514	found = .FALSE.
515
516	nzb_do = nzb
517	nzt_do = nzt+1
518	!
519	!-- Before each output, set array local_pf to fill value
520	local_pf = fill_value
521	!
522	!-- Set masking flag for topography for not resorted arrays
523	flag_nr = 0
524
525	!
526	!-- Store the array chosen on the temporary array.
527	resorted = .FALSE.
528	SELECT CASE ( TRIM( do2d(av,if) ) )
529	CASE ( 'e_xy', 'e_xz', 'e_yz' )
530	IF ( av == 0 ) THEN
531	to_be_resorted => e
532	ELSE
533	IF ( .NOT. ALLOCATED( e_av ) ) THEN
534	ALLOCATE( e_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
535	e_av = REAL( fill_value, KIND = wp )
536	ENDIF
537	to_be_resorted => e_av
538	ENDIF
539	IF ( mode == 'xy' ) level_z = zu
540
541	CASE ( 'lpt_xy', 'lpt_xz', 'lpt_yz' )
542	IF ( av == 0 ) THEN
543	to_be_resorted => pt
544	ELSE
545	IF ( .NOT. ALLOCATED( lpt_av ) ) THEN
546	ALLOCATE( lpt_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
547	lpt_av = REAL( fill_value, KIND = wp )
548	ENDIF
549	to_be_resorted => lpt_av
550	ENDIF
551	IF ( mode == 'xy' ) level_z = zu
552
553	CASE ( 'lwp*_xy' ) ! 2d-array
554	IF ( av == 0 ) THEN
555	DO i = nxl, nxr
556	DO j = nys, nyn
557	local_pf(i,j,nzb+1) = SUM( ql(nzb:nzt,j,i) * &
558	dzw(1:nzt+1) )
559	ENDDO
560	ENDDO
561	ELSE
562	IF ( .NOT. ALLOCATED( lwp_av ) ) THEN
563	ALLOCATE( lwp_av(nysg:nyng,nxlg:nxrg) )
564	lwp_av = REAL( fill_value, KIND = wp )
565	ENDIF
566	DO i = nxl, nxr
567	DO j = nys, nyn
568	local_pf(i,j,nzb+1) = lwp_av(j,i)
569	ENDDO
570	ENDDO
571	ENDIF
572	resorted = .TRUE.
573	two_d = .TRUE.
574	level_z(nzb+1) = zu(nzb+1)
575
576	CASE ( 'nc_xy', 'nc_xz', 'nc_yz' )
577	IF ( av == 0 ) THEN
578	to_be_resorted => nc
579	ELSE
580	IF ( .NOT. ALLOCATED( nc_av ) ) THEN
581	ALLOCATE( nc_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
582	nc_av = REAL( fill_value, KIND = wp )
583	ENDIF
584	to_be_resorted => nc_av
585	ENDIF
586	IF ( mode == 'xy' ) level_z = zu
587
588	CASE ( 'nr_xy', 'nr_xz', 'nr_yz' )
589	IF ( av == 0 ) THEN
590	to_be_resorted => nr
591	ELSE
592	IF ( .NOT. ALLOCATED( nr_av ) ) THEN
593	ALLOCATE( nr_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
594	nr_av = REAL( fill_value, KIND = wp )
595	ENDIF
596	to_be_resorted => nr_av
597	ENDIF
598	IF ( mode == 'xy' ) level_z = zu
599
600	CASE ( 'ghf*_xy' ) ! 2d-array
601	IF ( av == 0 ) THEN
602	DO m = 1, surf_lsm_h%ns
603	i = surf_lsm_h%i(m)
604	j = surf_lsm_h%j(m)
605	local_pf(i,j,nzb+1) = surf_lsm_h%ghf(m)
606	ENDDO
607	DO m = 1, surf_usm_h%ns
608	i = surf_usm_h%i(m)
609	j = surf_usm_h%j(m)
610	local_pf(i,j,nzb+1) = surf_usm_h%frac(ind_veg_wall,m) * &
611	surf_usm_h%wghf_eb(m) + &
612	surf_usm_h%frac(ind_pav_green,m) * &
613	surf_usm_h%wghf_eb_green(m) + &
614	surf_usm_h%frac(ind_wat_win,m) * &
615	surf_usm_h%wghf_eb_window(m)
616	ENDDO
617	ELSE
618	IF ( .NOT. ALLOCATED( ghf_av ) ) THEN
619	ALLOCATE( ghf_av(nysg:nyng,nxlg:nxrg) )
620	ghf_av = REAL( fill_value, KIND = wp )
621	ENDIF
622	DO i = nxl, nxr
623	DO j = nys, nyn
624	local_pf(i,j,nzb+1) = ghf_av(j,i)
625	ENDDO
626	ENDDO
627	ENDIF
628
629	resorted = .TRUE.
630	two_d = .TRUE.
631	level_z(nzb+1) = zu(nzb+1)
632
633	CASE ( 'ol*_xy' ) ! 2d-array
634	IF ( av == 0 ) THEN
635	DO m = 1, surf_def_h(0)%ns
636	i = surf_def_h(0)%i(m)
637	j = surf_def_h(0)%j(m)
638	local_pf(i,j,nzb+1) = surf_def_h(0)%ol(m)
639	ENDDO
640	DO m = 1, surf_lsm_h%ns
641	i = surf_lsm_h%i(m)
642	j = surf_lsm_h%j(m)
643	local_pf(i,j,nzb+1) = surf_lsm_h%ol(m)
644	ENDDO
645	DO m = 1, surf_usm_h%ns
646	i = surf_usm_h%i(m)
647	j = surf_usm_h%j(m)
648	local_pf(i,j,nzb+1) = surf_usm_h%ol(m)
649	ENDDO
650	ELSE
651	IF ( .NOT. ALLOCATED( ol_av ) ) THEN
652	ALLOCATE( ol_av(nysg:nyng,nxlg:nxrg) )
653	ol_av = REAL( fill_value, KIND = wp )
654	ENDIF
655	DO i = nxl, nxr
656	DO j = nys, nyn
657	local_pf(i,j,nzb+1) = ol_av(j,i)
658	ENDDO
659	ENDDO
660	ENDIF
661	resorted = .TRUE.
662	two_d = .TRUE.
663	level_z(nzb+1) = zu(nzb+1)
664
665	CASE ( 'p_xy', 'p_xz', 'p_yz' )
666	IF ( av == 0 ) THEN
667	IF ( psolver /= 'sor' ) CALL exchange_horiz( p, nbgp )
668	to_be_resorted => p
669	ELSE
670	IF ( .NOT. ALLOCATED( p_av ) ) THEN
671	ALLOCATE( p_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
672	p_av = REAL( fill_value, KIND = wp )
673	ENDIF
674	IF ( psolver /= 'sor' ) CALL exchange_horiz( p_av, nbgp )
675	to_be_resorted => p_av
676	ENDIF
677	IF ( mode == 'xy' ) level_z = zu
678
679	CASE ( 'pc_xy', 'pc_xz', 'pc_yz' ) ! particle concentration
680	IF ( av == 0 ) THEN
681	IF ( simulated_time >= particle_advection_start ) THEN
682	tend = prt_count
683	! CALL exchange_horiz( tend, nbgp )
684	ELSE
685	tend = 0.0_wp
686	ENDIF
687	DO i = nxl, nxr
688	DO j = nys, nyn
689	DO k = nzb, nzt+1
690	local_pf(i,j,k) = tend(k,j,i)
691	ENDDO
692	ENDDO
693	ENDDO
694	resorted = .TRUE.
695	ELSE
696	IF ( .NOT. ALLOCATED( pc_av ) ) THEN
697	ALLOCATE( pc_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
698	pc_av = REAL( fill_value, KIND = wp )
699	ENDIF
700	! CALL exchange_horiz( pc_av, nbgp )
701	to_be_resorted => pc_av
702	ENDIF
703
704	CASE ( 'pr_xy', 'pr_xz', 'pr_yz' ) ! mean particle radius (effective radius)
705	IF ( av == 0 ) THEN
706	IF ( simulated_time >= particle_advection_start ) THEN
707	DO i = nxl, nxr
708	DO j = nys, nyn
709	DO k = nzb, nzt+1
710	number_of_particles = prt_count(k,j,i)
711	IF (number_of_particles <= 0) CYCLE
712	particles => grid_particles(k,j,i)%particles(1:number_of_particles)
713	s_r2 = 0.0_wp
714	s_r3 = 0.0_wp
715	DO n = 1, number_of_particles
716	IF ( particles(n)%particle_mask ) THEN
717	s_r2 = s_r2 + particles(n)%radius*2 &
718	particles(n)%weight_factor
719	s_r3 = s_r3 + particles(n)%radius*3 &
720	particles(n)%weight_factor
721	ENDIF
722	ENDDO
723	IF ( s_r2 > 0.0_wp ) THEN
724	mean_r = s_r3 / s_r2
725	ELSE
726	mean_r = 0.0_wp
727	ENDIF
728	tend(k,j,i) = mean_r
729	ENDDO
730	ENDDO
731	ENDDO
732	! CALL exchange_horiz( tend, nbgp )
733	ELSE
734	tend = 0.0_wp
735	ENDIF
736	DO i = nxl, nxr
737	DO j = nys, nyn
738	DO k = nzb, nzt+1
739	local_pf(i,j,k) = tend(k,j,i)
740	ENDDO
741	ENDDO
742	ENDDO
743	resorted = .TRUE.
744	ELSE
745	IF ( .NOT. ALLOCATED( pr_av ) ) THEN
746	ALLOCATE( pr_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
747	pr_av = REAL( fill_value, KIND = wp )
748	ENDIF
749	! CALL exchange_horiz( pr_av, nbgp )
750	to_be_resorted => pr_av
751	ENDIF
752
753	CASE ( 'pra*_xy' ) ! 2d-array / integral quantity => no av
754	! CALL exchange_horiz_2d( precipitation_amount )
755	DO i = nxl, nxr
756	DO j = nys, nyn
757	local_pf(i,j,nzb+1) = precipitation_amount(j,i)
758	ENDDO
759	ENDDO
760	precipitation_amount = 0.0_wp ! reset for next integ. interval
761	resorted = .TRUE.
762	two_d = .TRUE.
763	level_z(nzb+1) = zu(nzb+1)
764
765	CASE ( 'prr_xy', 'prr_xz', 'prr_yz' )
766	IF ( av == 0 ) THEN
767	! CALL exchange_horiz( prr, nbgp )
768	DO i = nxl, nxr
769	DO j = nys, nyn
770	DO k = nzb, nzt+1
771	local_pf(i,j,k) = prr(k,j,i) * hyrho(nzb+1)
772	ENDDO
773	ENDDO
774	ENDDO
775	ELSE
776	IF ( .NOT. ALLOCATED( prr_av ) ) THEN
777	ALLOCATE( prr_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
778	prr_av = REAL( fill_value, KIND = wp )
779	ENDIF
780	! CALL exchange_horiz( prr_av, nbgp )
781	DO i = nxl, nxr
782	DO j = nys, nyn
783	DO k = nzb, nzt+1
784	local_pf(i,j,k) = prr_av(k,j,i) * hyrho(nzb+1)
785	ENDDO
786	ENDDO
787	ENDDO
788	ENDIF
789	resorted = .TRUE.
790	IF ( mode == 'xy' ) level_z = zu
791
792	CASE ( 'pt_xy', 'pt_xz', 'pt_yz' )
793	IF ( av == 0 ) THEN
794	IF ( .NOT. cloud_physics ) THEN
795	to_be_resorted => pt
796	ELSE
797	DO i = nxl, nxr
798	DO j = nys, nyn
799	DO k = nzb, nzt+1
800	local_pf(i,j,k) = pt(k,j,i) + l_d_cp * &
801	pt_d_t(k) * &
802	ql(k,j,i)
803	ENDDO
804	ENDDO
805	ENDDO
806	resorted = .TRUE.
807	ENDIF
808	ELSE
809	IF ( .NOT. ALLOCATED( pt_av ) ) THEN
810	ALLOCATE( pt_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
811	pt_av = REAL( fill_value, KIND = wp )
812	ENDIF
813	to_be_resorted => pt_av
814	ENDIF
815	IF ( mode == 'xy' ) level_z = zu
816
817	CASE ( 'q_xy', 'q_xz', 'q_yz' )
818	IF ( av == 0 ) THEN
819	to_be_resorted => q
820	ELSE
821	IF ( .NOT. ALLOCATED( q_av ) ) THEN
822	ALLOCATE( q_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
823	q_av = REAL( fill_value, KIND = wp )
824	ENDIF
825	to_be_resorted => q_av
826	ENDIF
827	IF ( mode == 'xy' ) level_z = zu
828
829	CASE ( 'qc_xy', 'qc_xz', 'qc_yz' )
830	IF ( av == 0 ) THEN
831	to_be_resorted => qc
832	ELSE
833	IF ( .NOT. ALLOCATED( qc_av ) ) THEN
834	ALLOCATE( qc_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
835	qc_av = REAL( fill_value, KIND = wp )
836	ENDIF
837	to_be_resorted => qc_av
838	ENDIF
839	IF ( mode == 'xy' ) level_z = zu
840
841	CASE ( 'ql_xy', 'ql_xz', 'ql_yz' )
842	IF ( av == 0 ) THEN
843	to_be_resorted => ql
844	ELSE
845	IF ( .NOT. ALLOCATED( ql_av ) ) THEN
846	ALLOCATE( ql_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
847	ql_av = REAL( fill_value, KIND = wp )
848	ENDIF
849	to_be_resorted => ql_av
850	ENDIF
851	IF ( mode == 'xy' ) level_z = zu
852
853	CASE ( 'ql_c_xy', 'ql_c_xz', 'ql_c_yz' )
854	IF ( av == 0 ) THEN
855	to_be_resorted => ql_c
856	ELSE
857	IF ( .NOT. ALLOCATED( ql_c_av ) ) THEN
858	ALLOCATE( ql_c_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
859	ql_c_av = REAL( fill_value, KIND = wp )
860	ENDIF
861	to_be_resorted => ql_c_av
862	ENDIF
863	IF ( mode == 'xy' ) level_z = zu
864
865	CASE ( 'ql_v_xy', 'ql_v_xz', 'ql_v_yz' )
866	IF ( av == 0 ) THEN
867	to_be_resorted => ql_v
868	ELSE
869	IF ( .NOT. ALLOCATED( ql_v_av ) ) THEN
870	ALLOCATE( ql_v_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
871	ql_v_av = REAL( fill_value, KIND = wp )
872	ENDIF
873	to_be_resorted => ql_v_av
874	ENDIF
875	IF ( mode == 'xy' ) level_z = zu
876
877	CASE ( 'ql_vp_xy', 'ql_vp_xz', 'ql_vp_yz' )
878	IF ( av == 0 ) THEN
879	IF ( simulated_time >= particle_advection_start ) THEN
880	DO i = nxl, nxr
881	DO j = nys, nyn
882	DO k = nzb, nzt+1
883	number_of_particles = prt_count(k,j,i)
884	IF (number_of_particles <= 0) CYCLE
885	particles => grid_particles(k,j,i)%particles(1:number_of_particles)
886	DO n = 1, number_of_particles
887	IF ( particles(n)%particle_mask ) THEN
888	tend(k,j,i) = tend(k,j,i) + &
889	particles(n)%weight_factor / &
890	prt_count(k,j,i)
891	ENDIF
892	ENDDO
893	ENDDO
894	ENDDO
895	ENDDO
896	! CALL exchange_horiz( tend, nbgp )
897	ELSE
898	tend = 0.0_wp
899	ENDIF
900	DO i = nxl, nxr
901	DO j = nys, nyn
902	DO k = nzb, nzt+1
903	local_pf(i,j,k) = tend(k,j,i)
904	ENDDO
905	ENDDO
906	ENDDO
907	resorted = .TRUE.
908	ELSE
909	IF ( .NOT. ALLOCATED( ql_vp_av ) ) THEN
910	ALLOCATE( ql_vp_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
911	ql_vp_av = REAL( fill_value, KIND = wp )
912	ENDIF
913	! CALL exchange_horiz( ql_vp_av, nbgp )
914	to_be_resorted => ql_vp_av
915	ENDIF
916	IF ( mode == 'xy' ) level_z = zu
917
918	CASE ( 'qr_xy', 'qr_xz', 'qr_yz' )
919	IF ( av == 0 ) THEN
920	to_be_resorted => qr
921	ELSE
922	IF ( .NOT. ALLOCATED( qr_av ) ) THEN
923	ALLOCATE( qr_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
924	qr_av = REAL( fill_value, KIND = wp )
925	ENDIF
926	to_be_resorted => qr_av
927	ENDIF
928	IF ( mode == 'xy' ) level_z = zu
929
930	CASE ( 'qsws*_xy' ) ! 2d-array
931	IF ( av == 0 ) THEN
932	!
933	!-- In case of default surfaces, clean-up flux by density.
934	!-- In case of land- and urban-surfaces, convert fluxes into
935	!-- dynamic units
936	DO m = 1, surf_def_h(0)%ns
937	i = surf_def_h(0)%i(m)
938	j = surf_def_h(0)%j(m)
939	k = surf_def_h(0)%k(m)
940	local_pf(i,j,nzb+1) = surf_def_h(0)%qsws(m) * &
941	waterflux_output_conversion(k)
942	ENDDO
943	DO m = 1, surf_lsm_h%ns
944	i = surf_lsm_h%i(m)
945	j = surf_lsm_h%j(m)
946	k = surf_lsm_h%k(m)
947	local_pf(i,j,nzb+1) = surf_lsm_h%qsws(m) * l_v
948	ENDDO
949	DO m = 1, surf_usm_h%ns
950	i = surf_usm_h%i(m)
951	j = surf_usm_h%j(m)
952	k = surf_usm_h%k(m)
953	local_pf(i,j,nzb+1) = surf_usm_h%qsws(m) * l_v
954	ENDDO
955	ELSE
956	IF ( .NOT. ALLOCATED( qsws_av ) ) THEN
957	ALLOCATE( qsws_av(nysg:nyng,nxlg:nxrg) )
958	qsws_av = REAL( fill_value, KIND = wp )
959	ENDIF
960	DO i = nxl, nxr
961	DO j = nys, nyn
962	local_pf(i,j,nzb+1) = qsws_av(j,i)
963	ENDDO
964	ENDDO
965	ENDIF
966	resorted = .TRUE.
967	two_d = .TRUE.
968	level_z(nzb+1) = zu(nzb+1)
969
970	CASE ( 'qv_xy', 'qv_xz', 'qv_yz' )
971	IF ( av == 0 ) THEN
972	DO i = nxl, nxr
973	DO j = nys, nyn
974	DO k = nzb, nzt+1
975	local_pf(i,j,k) = q(k,j,i) - ql(k,j,i)
976	ENDDO
977	ENDDO
978	ENDDO
979	resorted = .TRUE.
980	ELSE
981	IF ( .NOT. ALLOCATED( qv_av ) ) THEN
982	ALLOCATE( qv_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
983	qv_av = REAL( fill_value, KIND = wp )
984	ENDIF
985	to_be_resorted => qv_av
986	ENDIF
987	IF ( mode == 'xy' ) level_z = zu
988
989	CASE ( 'r_a*_xy' ) ! 2d-array
990	IF ( av == 0 ) THEN
991	DO m = 1, surf_lsm_h%ns
992	i = surf_lsm_h%i(m)
993	j = surf_lsm_h%j(m)
994	local_pf(i,j,nzb+1) = surf_lsm_h%r_a(m)
995	ENDDO
996
997	DO m = 1, surf_usm_h%ns
998	i = surf_usm_h%i(m)
999	j = surf_usm_h%j(m)
1000	local_pf(i,j,nzb+1) = &
1001	( surf_usm_h%frac(ind_veg_wall,m) * &
1002	surf_usm_h%r_a(m) + &
1003	surf_usm_h%frac(ind_pav_green,m) * &
1004	surf_usm_h%r_a_green(m) + &
1005	surf_usm_h%frac(ind_wat_win,m) * &
1006	surf_usm_h%r_a_window(m) )
1007	ENDDO
1008	ELSE
1009	IF ( .NOT. ALLOCATED( r_a_av ) ) THEN
1010	ALLOCATE( r_a_av(nysg:nyng,nxlg:nxrg) )
1011	r_a_av = REAL( fill_value, KIND = wp )
1012	ENDIF
1013	DO i = nxl, nxr
1014	DO j = nys, nyn
1015	local_pf(i,j,nzb+1) = r_a_av(j,i)
1016	ENDDO
1017	ENDDO
1018	ENDIF
1019	resorted = .TRUE.
1020	two_d = .TRUE.
1021	level_z(nzb+1) = zu(nzb+1)
1022
1023	CASE ( 'rho_ocean_xy', 'rho_ocean_xz', 'rho_ocean_yz' )
1024	IF ( av == 0 ) THEN
1025	to_be_resorted => rho_ocean
1026	ELSE
1027	IF ( .NOT. ALLOCATED( rho_ocean_av ) ) THEN
1028	ALLOCATE( rho_ocean_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1029	rho_ocean_av = REAL( fill_value, KIND = wp )
1030	ENDIF
1031	to_be_resorted => rho_ocean_av
1032	ENDIF
1033
1034	CASE ( 's_xy', 's_xz', 's_yz' )
1035	IF ( av == 0 ) THEN
1036	to_be_resorted => s
1037	ELSE
1038	IF ( .NOT. ALLOCATED( s_av ) ) THEN
1039	ALLOCATE( s_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1040	s_av = REAL( fill_value, KIND = wp )
1041	ENDIF
1042	to_be_resorted => s_av
1043	ENDIF
1044
1045	CASE ( 'sa_xy', 'sa_xz', 'sa_yz' )
1046	IF ( av == 0 ) THEN
1047	to_be_resorted => sa
1048	ELSE
1049	IF ( .NOT. ALLOCATED( sa_av ) ) THEN
1050	ALLOCATE( sa_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1051	sa_av = REAL( fill_value, KIND = wp )
1052	ENDIF
1053	to_be_resorted => sa_av
1054	ENDIF
1055
1056	CASE ( 'shf*_xy' ) ! 2d-array
1057	IF ( av == 0 ) THEN
1058	!
1059	!-- In case of default surfaces, clean-up flux by density.
1060	!-- In case of land- and urban-surfaces, convert fluxes into
1061	!-- dynamic units.
1062	DO m = 1, surf_def_h(0)%ns
1063	i = surf_def_h(0)%i(m)
1064	j = surf_def_h(0)%j(m)
1065	k = surf_def_h(0)%k(m)
1066	local_pf(i,j,nzb+1) = surf_def_h(0)%shf(m) * &
1067	heatflux_output_conversion(k)
1068	ENDDO
1069	DO m = 1, surf_lsm_h%ns
1070	i = surf_lsm_h%i(m)
1071	j = surf_lsm_h%j(m)
1072	k = surf_lsm_h%k(m)
1073	local_pf(i,j,nzb+1) = surf_lsm_h%shf(m) * cp
1074	ENDDO
1075	DO m = 1, surf_usm_h%ns
1076	i = surf_usm_h%i(m)
1077	j = surf_usm_h%j(m)
1078	k = surf_usm_h%k(m)
1079	local_pf(i,j,nzb+1) = surf_usm_h%shf(m) * cp
1080	ENDDO
1081	ELSE
1082	IF ( .NOT. ALLOCATED( shf_av ) ) THEN
1083	ALLOCATE( shf_av(nysg:nyng,nxlg:nxrg) )
1084	shf_av = REAL( fill_value, KIND = wp )
1085	ENDIF
1086	DO i = nxl, nxr
1087	DO j = nys, nyn
1088	local_pf(i,j,nzb+1) = shf_av(j,i)
1089	ENDDO
1090	ENDDO
1091	ENDIF
1092	resorted = .TRUE.
1093	two_d = .TRUE.
1094	level_z(nzb+1) = zu(nzb+1)
1095
1096	CASE ( 'ssws*_xy' ) ! 2d-array
1097	IF ( av == 0 ) THEN
1098	DO m = 1, surf_def_h(0)%ns
1099	i = surf_def_h(0)%i(m)
1100	j = surf_def_h(0)%j(m)
1101	local_pf(i,j,nzb+1) = surf_def_h(0)%ssws(m)
1102	ENDDO
1103	DO m = 1, surf_lsm_h%ns
1104	i = surf_lsm_h%i(m)
1105	j = surf_lsm_h%j(m)
1106	local_pf(i,j,nzb+1) = surf_lsm_h%ssws(m)
1107	ENDDO
1108	DO m = 1, surf_usm_h%ns
1109	i = surf_usm_h%i(m)
1110	j = surf_usm_h%j(m)
1111	local_pf(i,j,nzb+1) = surf_usm_h%ssws(m)
1112	ENDDO
1113	ELSE
1114	IF ( .NOT. ALLOCATED( ssws_av ) ) THEN
1115	ALLOCATE( ssws_av(nysg:nyng,nxlg:nxrg) )
1116	ssws_av = REAL( fill_value, KIND = wp )
1117	ENDIF
1118	DO i = nxl, nxr
1119	DO j = nys, nyn
1120	local_pf(i,j,nzb+1) = ssws_av(j,i)
1121	ENDDO
1122	ENDDO
1123	ENDIF
1124	resorted = .TRUE.
1125	two_d = .TRUE.
1126	level_z(nzb+1) = zu(nzb+1)
1127
1128	CASE ( 't*_xy' ) ! 2d-array
1129	IF ( av == 0 ) THEN
1130	DO m = 1, surf_def_h(0)%ns
1131	i = surf_def_h(0)%i(m)
1132	j = surf_def_h(0)%j(m)
1133	local_pf(i,j,nzb+1) = surf_def_h(0)%ts(m)
1134	ENDDO
1135	DO m = 1, surf_lsm_h%ns
1136	i = surf_lsm_h%i(m)
1137	j = surf_lsm_h%j(m)
1138	local_pf(i,j,nzb+1) = surf_lsm_h%ts(m)
1139	ENDDO
1140	DO m = 1, surf_usm_h%ns
1141	i = surf_usm_h%i(m)
1142	j = surf_usm_h%j(m)
1143	local_pf(i,j,nzb+1) = surf_usm_h%ts(m)
1144	ENDDO
1145	ELSE
1146	IF ( .NOT. ALLOCATED( ts_av ) ) THEN
1147	ALLOCATE( ts_av(nysg:nyng,nxlg:nxrg) )
1148	ts_av = REAL( fill_value, KIND = wp )
1149	ENDIF
1150	DO i = nxl, nxr
1151	DO j = nys, nyn
1152	local_pf(i,j,nzb+1) = ts_av(j,i)
1153	ENDDO
1154	ENDDO
1155	ENDIF
1156	resorted = .TRUE.
1157	two_d = .TRUE.
1158	level_z(nzb+1) = zu(nzb+1)
1159
1160	CASE ( 'tsurf*_xy' ) ! 2d-array
1161	IF ( av == 0 ) THEN
1162	DO m = 1, surf_def_h(0)%ns
1163	i = surf_def_h(0)%i(m)
1164	j = surf_def_h(0)%j(m)
1165	local_pf(i,j,nzb+1) = surf_def_h(0)%pt_surface(m)
1166	ENDDO
1167
1168	DO m = 1, surf_lsm_h%ns
1169	i = surf_lsm_h%i(m)
1170	j = surf_lsm_h%j(m)
1171	local_pf(i,j,nzb+1) = surf_lsm_h%pt_surface(m)
1172	ENDDO
1173
1174	DO m = 1, surf_usm_h%ns
1175	i = surf_usm_h%i(m)
1176	j = surf_usm_h%j(m)
1177	local_pf(i,j,nzb+1) = surf_usm_h%pt_surface(m)
1178	ENDDO
1179
1180	ELSE
1181	IF ( .NOT. ALLOCATED( tsurf_av ) ) THEN
1182	ALLOCATE( tsurf_av(nysg:nyng,nxlg:nxrg) )
1183	tsurf_av = REAL( fill_value, KIND = wp )
1184	ENDIF
1185	DO i = nxl, nxr
1186	DO j = nys, nyn
1187	local_pf(i,j,nzb+1) = tsurf_av(j,i)
1188	ENDDO
1189	ENDDO
1190	ENDIF
1191	resorted = .TRUE.
1192	two_d = .TRUE.
1193	level_z(nzb+1) = zu(nzb+1)
1194
1195	CASE ( 'u_xy', 'u_xz', 'u_yz' )
1196	flag_nr = 1
1197	IF ( av == 0 ) THEN
1198	to_be_resorted => u
1199	ELSE
1200	IF ( .NOT. ALLOCATED( u_av ) ) THEN
1201	ALLOCATE( u_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1202	u_av = REAL( fill_value, KIND = wp )
1203	ENDIF
1204	to_be_resorted => u_av
1205	ENDIF
1206	IF ( mode == 'xy' ) level_z = zu
1207	!
1208	!-- Substitute the values generated by "mirror" boundary condition
1209	!-- at the bottom boundary by the real surface values.
1210	IF ( do2d(av,if) == 'u_xz' .OR. do2d(av,if) == 'u_yz' ) THEN
1211	IF ( ibc_uv_b == 0 ) local_pf(:,:,nzb) = 0.0_wp
1212	ENDIF
1213
1214	CASE ( 'u*_xy' ) ! 2d-array
1215	IF ( av == 0 ) THEN
1216	DO m = 1, surf_def_h(0)%ns
1217	i = surf_def_h(0)%i(m)
1218	j = surf_def_h(0)%j(m)
1219	local_pf(i,j,nzb+1) = surf_def_h(0)%us(m)
1220	ENDDO
1221	DO m = 1, surf_lsm_h%ns
1222	i = surf_lsm_h%i(m)
1223	j = surf_lsm_h%j(m)
1224	local_pf(i,j,nzb+1) = surf_lsm_h%us(m)
1225	ENDDO
1226	DO m = 1, surf_usm_h%ns
1227	i = surf_usm_h%i(m)
1228	j = surf_usm_h%j(m)
1229	local_pf(i,j,nzb+1) = surf_usm_h%us(m)
1230	ENDDO
1231	ELSE
1232	IF ( .NOT. ALLOCATED( us_av ) ) THEN
1233	ALLOCATE( us_av(nysg:nyng,nxlg:nxrg) )
1234	us_av = REAL( fill_value, KIND = wp )
1235	ENDIF
1236	DO i = nxl, nxr
1237	DO j = nys, nyn
1238	local_pf(i,j,nzb+1) = us_av(j,i)
1239	ENDDO
1240	ENDDO
1241	ENDIF
1242	resorted = .TRUE.
1243	two_d = .TRUE.
1244	level_z(nzb+1) = zu(nzb+1)
1245
1246	CASE ( 'v_xy', 'v_xz', 'v_yz' )
1247	flag_nr = 2
1248	IF ( av == 0 ) THEN
1249	to_be_resorted => v
1250	ELSE
1251	IF ( .NOT. ALLOCATED( v_av ) ) THEN
1252	ALLOCATE( v_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1253	v_av = REAL( fill_value, KIND = wp )
1254	ENDIF
1255	to_be_resorted => v_av
1256	ENDIF
1257	IF ( mode == 'xy' ) level_z = zu
1258	!
1259	!-- Substitute the values generated by "mirror" boundary condition
1260	!-- at the bottom boundary by the real surface values.
1261	IF ( do2d(av,if) == 'v_xz' .OR. do2d(av,if) == 'v_yz' ) THEN
1262	IF ( ibc_uv_b == 0 ) local_pf(:,:,nzb) = 0.0_wp
1263	ENDIF
1264
1265	CASE ( 'vpt_xy', 'vpt_xz', 'vpt_yz' )
1266	IF ( av == 0 ) THEN
1267	to_be_resorted => vpt
1268	ELSE
1269	IF ( .NOT. ALLOCATED( vpt_av ) ) THEN
1270	ALLOCATE( vpt_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1271	vpt_av = REAL( fill_value, KIND = wp )
1272	ENDIF
1273	to_be_resorted => vpt_av
1274	ENDIF
1275	IF ( mode == 'xy' ) level_z = zu
1276
1277	CASE ( 'w_xy', 'w_xz', 'w_yz' )
1278	flag_nr = 3
1279	IF ( av == 0 ) THEN
1280	to_be_resorted => w
1281	ELSE
1282	IF ( .NOT. ALLOCATED( w_av ) ) THEN
1283	ALLOCATE( w_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1284	w_av = REAL( fill_value, KIND = wp )
1285	ENDIF
1286	to_be_resorted => w_av
1287	ENDIF
1288	IF ( mode == 'xy' ) level_z = zw
1289
1290	CASE ( 'z0*_xy' ) ! 2d-array
1291	IF ( av == 0 ) THEN
1292	DO m = 1, surf_def_h(0)%ns
1293	i = surf_def_h(0)%i(m)
1294	j = surf_def_h(0)%j(m)
1295	local_pf(i,j,nzb+1) = surf_def_h(0)%z0(m)
1296	ENDDO
1297	DO m = 1, surf_lsm_h%ns
1298	i = surf_lsm_h%i(m)
1299	j = surf_lsm_h%j(m)
1300	local_pf(i,j,nzb+1) = surf_lsm_h%z0(m)
1301	ENDDO
1302	DO m = 1, surf_usm_h%ns
1303	i = surf_usm_h%i(m)
1304	j = surf_usm_h%j(m)
1305	local_pf(i,j,nzb+1) = surf_usm_h%z0(m)
1306	ENDDO
1307	ELSE
1308	IF ( .NOT. ALLOCATED( z0_av ) ) THEN
1309	ALLOCATE( z0_av(nysg:nyng,nxlg:nxrg) )
1310	z0_av = REAL( fill_value, KIND = wp )
1311	ENDIF
1312	DO i = nxl, nxr
1313	DO j = nys, nyn
1314	local_pf(i,j,nzb+1) = z0_av(j,i)
1315	ENDDO
1316	ENDDO
1317	ENDIF
1318	resorted = .TRUE.
1319	two_d = .TRUE.
1320	level_z(nzb+1) = zu(nzb+1)
1321
1322	CASE ( 'z0h*_xy' ) ! 2d-array
1323	IF ( av == 0 ) THEN
1324	DO m = 1, surf_def_h(0)%ns
1325	i = surf_def_h(0)%i(m)
1326	j = surf_def_h(0)%j(m)
1327	local_pf(i,j,nzb+1) = surf_def_h(0)%z0h(m)
1328	ENDDO
1329	DO m = 1, surf_lsm_h%ns
1330	i = surf_lsm_h%i(m)
1331	j = surf_lsm_h%j(m)
1332	local_pf(i,j,nzb+1) = surf_lsm_h%z0h(m)
1333	ENDDO
1334	DO m = 1, surf_usm_h%ns
1335	i = surf_usm_h%i(m)
1336	j = surf_usm_h%j(m)
1337	local_pf(i,j,nzb+1) = surf_usm_h%z0h(m)
1338	ENDDO
1339	ELSE
1340	IF ( .NOT. ALLOCATED( z0h_av ) ) THEN
1341	ALLOCATE( z0h_av(nysg:nyng,nxlg:nxrg) )
1342	z0h_av = REAL( fill_value, KIND = wp )
1343	ENDIF
1344	DO i = nxl, nxr
1345	DO j = nys, nyn
1346	local_pf(i,j,nzb+1) = z0h_av(j,i)
1347	ENDDO
1348	ENDDO
1349	ENDIF
1350	resorted = .TRUE.
1351	two_d = .TRUE.
1352	level_z(nzb+1) = zu(nzb+1)
1353
1354	CASE ( 'z0q*_xy' ) ! 2d-array
1355	IF ( av == 0 ) THEN
1356	DO m = 1, surf_def_h(0)%ns
1357	i = surf_def_h(0)%i(m)
1358	j = surf_def_h(0)%j(m)
1359	local_pf(i,j,nzb+1) = surf_def_h(0)%z0q(m)
1360	ENDDO
1361	DO m = 1, surf_lsm_h%ns
1362	i = surf_lsm_h%i(m)
1363	j = surf_lsm_h%j(m)
1364	local_pf(i,j,nzb+1) = surf_lsm_h%z0q(m)
1365	ENDDO
1366	DO m = 1, surf_usm_h%ns
1367	i = surf_usm_h%i(m)
1368	j = surf_usm_h%j(m)
1369	local_pf(i,j,nzb+1) = surf_usm_h%z0q(m)
1370	ENDDO
1371	ELSE
1372	IF ( .NOT. ALLOCATED( z0q_av ) ) THEN
1373	ALLOCATE( z0q_av(nysg:nyng,nxlg:nxrg) )
1374	z0q_av = REAL( fill_value, KIND = wp )
1375	ENDIF
1376	DO i = nxl, nxr
1377	DO j = nys, nyn
1378	local_pf(i,j,nzb+1) = z0q_av(j,i)
1379	ENDDO
1380	ENDDO
1381	ENDIF
1382	resorted = .TRUE.
1383	two_d = .TRUE.
1384	level_z(nzb+1) = zu(nzb+1)
1385
1386	CASE DEFAULT
1387
1388	!
1389	!-- Land surface model quantity
1390	IF ( land_surface ) THEN
1391	CALL lsm_data_output_2d( av, do2d(av,if), found, grid, mode,&
1392	local_pf, two_d, nzb_do, nzt_do )
1393	ENDIF
1394
1395	!
1396	!-- Turbulence closure variables
1397	IF ( .NOT. found ) THEN
1398	CALL tcm_data_output_2d( av, do2d(av,if), found, grid, mode,&
1399	local_pf, two_d, nzb_do, nzt_do )
1400	ENDIF
1401
1402	!
1403	!-- Radiation quantity
1404	IF ( .NOT. found .AND. radiation ) THEN
1405	CALL radiation_data_output_2d( av, do2d(av,if), found, grid,&
1406	mode, local_pf, two_d, &
1407	nzb_do, nzt_do )
1408	ENDIF
1409
1410	!
1411	!-- Gust module quantities
1412	IF ( .NOT. found .AND. gust_module_enabled ) THEN
1413	CALL gust_data_output_2d( av, do2d(av,if), found, grid, &
1414	local_pf, two_d, nzb_do, nzt_do )
1415	ENDIF
1416
1417	!
1418	!-- UV exposure model quantity
1419	IF ( uv_exposure ) THEN
1420	CALL uvem_data_output_2d( av, do2d(av,if), found, grid, mode,&
1421	local_pf, two_d, nzb_do, nzt_do )
1422	ENDIF
1423
1424	!
1425	!-- User defined quantity
1426	IF ( .NOT. found ) THEN
1427	CALL user_data_output_2d( av, do2d(av,if), found, grid, &
1428	local_pf, two_d, nzb_do, nzt_do )
1429	ENDIF
1430
1431	resorted = .TRUE.
1432
1433	IF ( grid == 'zu' ) THEN
1434	IF ( mode == 'xy' ) level_z = zu
1435	ELSEIF ( grid == 'zw' ) THEN
1436	IF ( mode == 'xy' ) level_z = zw
1437	ELSEIF ( grid == 'zu1' ) THEN
1438	IF ( mode == 'xy' ) level_z(nzb+1) = zu(nzb+1)
1439	ELSEIF ( grid == 'zs' ) THEN
1440	IF ( mode == 'xy' ) level_z = zs
1441	ENDIF
1442
1443	IF ( .NOT. found ) THEN
1444	message_string = 'no output provided for: ' // &
1445	TRIM( do2d(av,if) )
1446	CALL message( 'data_output_2d', 'PA0181', 0, 0, 0, 6, 0 )
1447	ENDIF
1448
1449	END SELECT
1450
1451	!
1452	!-- Resort the array to be output, if not done above. Flag topography
1453	!-- grid points with fill values, using the corresponding maksing flag.
1454	IF ( .NOT. resorted ) THEN
1455	DO i = nxl, nxr
1456	DO j = nys, nyn
1457	DO k = nzb_do, nzt_do
1458	local_pf(i,j,k) = MERGE( to_be_resorted(k,j,i), &
1459	REAL( fill_value, KIND = wp ), &
1460	BTEST( wall_flags_0(k,j,i), &
1461	flag_nr ) )
1462	ENDDO
1463	ENDDO
1464	ENDDO
1465	ENDIF
1466
1467	!
1468	!-- Output of the individual cross-sections, depending on the cross-
1469	!-- section mode chosen.
1470	is = 1
1471	loop1: DO WHILE ( section(is,s_ind) /= -9999 .OR. two_d )
1472
1473	SELECT CASE ( mode )
1474
1475	CASE ( 'xy' )
1476	!
1477	!-- Determine the cross section index
1478	IF ( two_d ) THEN
1479	layer_xy = nzb+1
1480	ELSE
1481	layer_xy = section(is,s_ind)
1482	ENDIF
1483
1484	!
1485	!-- Exit the loop for layers beyond the data output domain
1486	!-- (used for soil model)
1487	IF ( layer_xy > nzt_do ) THEN
1488	EXIT loop1
1489	ENDIF
1490
1491	!
1492	!-- Update the netCDF xy cross section time axis.
1493	!-- In case of parallel output, this is only done by PE0
1494	!-- to increase the performance.
1495	IF ( simulated_time /= do2d_xy_last_time(av) ) THEN
1496	do2d_xy_time_count(av) = do2d_xy_time_count(av) + 1
1497	do2d_xy_last_time(av) = simulated_time
1498	IF ( myid == 0 ) THEN
1499	IF ( .NOT. data_output_2d_on_each_pe &
1500	.OR. netcdf_data_format > 4 ) &
1501	THEN
1502	#if defined( __netcdf )
1503	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1504	id_var_time_xy(av), &
1505	(/ time_since_reference_point /), &
1506	start = (/ do2d_xy_time_count(av) /), &
1507	count = (/ 1 /) )
1508	CALL netcdf_handle_error( 'data_output_2d', 53 )
1509	#endif
1510	ENDIF
1511	ENDIF
1512	ENDIF
1513	!
1514	!-- If required, carry out averaging along z
1515	IF ( section(is,s_ind) == -1 .AND. .NOT. two_d ) THEN
1516
1517	local_2d = 0.0_wp
1518	!
1519	!-- Carry out the averaging (all data are on the PE)
1520	DO k = nzb_do, nzt_do
1521	DO j = nys, nyn
1522	DO i = nxl, nxr
1523	local_2d(i,j) = local_2d(i,j) + local_pf(i,j,k)
1524	ENDDO
1525	ENDDO
1526	ENDDO
1527
1528	local_2d = local_2d / ( nzt_do - nzb_do + 1.0_wp)
1529
1530	ELSE
1531	!
1532	!-- Just store the respective section on the local array
1533	local_2d = local_pf(:,:,layer_xy)
1534
1535	ENDIF
1536
1537	#if defined( __parallel )
1538	IF ( netcdf_data_format > 4 ) THEN
1539	!
1540	!-- Parallel output in netCDF4/HDF5 format.
1541	IF ( two_d ) THEN
1542	iis = 1
1543	ELSE
1544	iis = is
1545	ENDIF
1546
1547	#if defined( __netcdf )
1548	!
1549	!-- For parallel output, all cross sections are first stored
1550	!-- here on a local array and will be written to the output
1551	!-- file afterwards to increase the performance.
1552	DO i = nxl, nxr
1553	DO j = nys, nyn
1554	local_2d_sections(i,j,iis) = local_2d(i,j)
1555	ENDDO
1556	ENDDO
1557	#endif
1558	ELSE
1559
1560	IF ( data_output_2d_on_each_pe ) THEN
1561	!
1562	!-- Output of partial arrays on each PE
1563	#if defined( __netcdf )
1564	IF ( myid == 0 ) THEN
1565	WRITE ( 21 ) time_since_reference_point, &
1566	do2d_xy_time_count(av), av
1567	ENDIF
1568	#endif
1569	DO i = 0, io_blocks-1
1570	IF ( i == io_group ) THEN
1571	WRITE ( 21 ) nxl, nxr, nys, nyn, nys, nyn
1572	WRITE ( 21 ) local_2d
1573	ENDIF
1574	#if defined( __parallel )
1575	CALL MPI_BARRIER( comm2d, ierr )
1576	#endif
1577	ENDDO
1578
1579	ELSE
1580	!
1581	!-- PE0 receives partial arrays from all processors and
1582	!-- then outputs them. Here a barrier has to be set,
1583	!-- because otherwise "-MPI- FATAL: Remote protocol queue
1584	!-- full" may occur.
1585	CALL MPI_BARRIER( comm2d, ierr )
1586
1587	ngp = ( nxr-nxl+1 ) * ( nyn-nys+1 )
1588	IF ( myid == 0 ) THEN
1589	!
1590	!-- Local array can be relocated directly.
1591	total_2d(nxl:nxr,nys:nyn) = local_2d
1592	!
1593	!-- Receive data from all other PEs.
1594	DO n = 1, numprocs-1
1595	!
1596	!-- Receive index limits first, then array.
1597	!-- Index limits are received in arbitrary order from
1598	!-- the PEs.
1599	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
1600	MPI_ANY_SOURCE, 0, comm2d, &
1601	status, ierr )
1602	sender = status(MPI_SOURCE)
1603	DEALLOCATE( local_2d )
1604	ALLOCATE( local_2d(ind(1):ind(2),ind(3):ind(4)) )
1605	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
1606	MPI_REAL, sender, 1, comm2d, &
1607	status, ierr )
1608	total_2d(ind(1):ind(2),ind(3):ind(4)) = local_2d
1609	ENDDO
1610	!
1611	!-- Relocate the local array for the next loop increment
1612	DEALLOCATE( local_2d )
1613	ALLOCATE( local_2d(nxl:nxr,nys:nyn) )
1614
1615	#if defined( __netcdf )
1616	IF ( two_d ) THEN
1617	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1618	id_var_do2d(av,if), &
1619	total_2d(0:nx,0:ny), &
1620	start = (/ 1, 1, 1, do2d_xy_time_count(av) /), &
1621	count = (/ nx+1, ny+1, 1, 1 /) )
1622	ELSE
1623	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1624	id_var_do2d(av,if), &
1625	total_2d(0:nx,0:ny), &
1626	start = (/ 1, 1, is, do2d_xy_time_count(av) /), &
1627	count = (/ nx+1, ny+1, 1, 1 /) )
1628	ENDIF
1629	CALL netcdf_handle_error( 'data_output_2d', 54 )
1630	#endif
1631
1632	ELSE
1633	!
1634	!-- First send the local index limits to PE0
1635	ind(1) = nxl; ind(2) = nxr
1636	ind(3) = nys; ind(4) = nyn
1637	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
1638	comm2d, ierr )
1639	!
1640	!-- Send data to PE0
1641	CALL MPI_SEND( local_2d(nxl,nys), ngp, &
1642	MPI_REAL, 0, 1, comm2d, ierr )
1643	ENDIF
1644	!
1645	!-- A barrier has to be set, because otherwise some PEs may
1646	!-- proceed too fast so that PE0 may receive wrong data on
1647	!-- tag 0
1648	CALL MPI_BARRIER( comm2d, ierr )
1649	ENDIF
1650
1651	ENDIF
1652	#else
1653	#if defined( __netcdf )
1654	IF ( two_d ) THEN
1655	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1656	id_var_do2d(av,if), &
1657	local_2d(nxl:nxr,nys:nyn), &
1658	start = (/ 1, 1, 1, do2d_xy_time_count(av) /), &
1659	count = (/ nx+1, ny+1, 1, 1 /) )
1660	ELSE
1661	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1662	id_var_do2d(av,if), &
1663	local_2d(nxl:nxr,nys:nyn), &
1664	start = (/ 1, 1, is, do2d_xy_time_count(av) /), &
1665	count = (/ nx+1, ny+1, 1, 1 /) )
1666	ENDIF
1667	CALL netcdf_handle_error( 'data_output_2d', 447 )
1668	#endif
1669	#endif
1670
1671	!
1672	!-- For 2D-arrays (e.g. u*) only one cross-section is available.
1673	!-- Hence exit loop of output levels.
1674	IF ( two_d ) THEN
1675	IF ( netcdf_data_format < 5 ) two_d = .FALSE.
1676	EXIT loop1
1677	ENDIF
1678
1679	CASE ( 'xz' )
1680	!
1681	!-- Update the netCDF xz cross section time axis.
1682	!-- In case of parallel output, this is only done by PE0
1683	!-- to increase the performance.
1684	IF ( simulated_time /= do2d_xz_last_time(av) ) THEN
1685	do2d_xz_time_count(av) = do2d_xz_time_count(av) + 1
1686	do2d_xz_last_time(av) = simulated_time
1687	IF ( myid == 0 ) THEN
1688	IF ( .NOT. data_output_2d_on_each_pe &
1689	.OR. netcdf_data_format > 4 ) &
1690	THEN
1691	#if defined( __netcdf )
1692	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
1693	id_var_time_xz(av), &
1694	(/ time_since_reference_point /), &
1695	start = (/ do2d_xz_time_count(av) /), &
1696	count = (/ 1 /) )
1697	CALL netcdf_handle_error( 'data_output_2d', 56 )
1698	#endif
1699	ENDIF
1700	ENDIF
1701	ENDIF
1702
1703	!
1704	!-- If required, carry out averaging along y
1705	IF ( section(is,s_ind) == -1 ) THEN
1706
1707	ALLOCATE( local_2d_l(nxl:nxr,nzb_do:nzt_do) )
1708	local_2d_l = 0.0_wp
1709	ngp = ( nxr-nxl + 1 ) * ( nzt_do-nzb_do + 1 )
1710	!
1711	!-- First local averaging on the PE
1712	DO k = nzb_do, nzt_do
1713	DO j = nys, nyn
1714	DO i = nxl, nxr
1715	local_2d_l(i,k) = local_2d_l(i,k) + &
1716	local_pf(i,j,k)
1717	ENDDO
1718	ENDDO
1719	ENDDO
1720	#if defined( __parallel )
1721	!
1722	!-- Now do the averaging over all PEs along y
1723	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
1724	CALL MPI_ALLREDUCE( local_2d_l(nxl,nzb_do), &
1725	local_2d(nxl,nzb_do), ngp, MPI_REAL, &
1726	MPI_SUM, comm1dy, ierr )
1727	#else
1728	local_2d = local_2d_l
1729	#endif
1730	local_2d = local_2d / ( ny + 1.0_wp )
1731
1732	DEALLOCATE( local_2d_l )
1733
1734	ELSE
1735	!
1736	!-- Just store the respective section on the local array
1737	!-- (but only if it is available on this PE!)
1738	IF ( section(is,s_ind) >= nys .AND. section(is,s_ind) <= nyn ) &
1739	THEN
1740	local_2d = local_pf(:,section(is,s_ind),nzb_do:nzt_do)
1741	ENDIF
1742
1743	ENDIF
1744
1745	#if defined( __parallel )
1746	IF ( netcdf_data_format > 4 ) THEN
1747	!
1748	!-- Output in netCDF4/HDF5 format.
1749	!-- Output only on those PEs where the respective cross
1750	!-- sections reside. Cross sections averaged along y are
1751	!-- output on the respective first PE along y (myidy=0).
1752	IF ( ( section(is,s_ind) >= nys .AND. &
1753	section(is,s_ind) <= nyn ) .OR. &
1754	( section(is,s_ind) == -1 .AND. myidy == 0 ) ) THEN
1755	#if defined( __netcdf )
1756	!
1757	!-- For parallel output, all cross sections are first
1758	!-- stored here on a local array and will be written to the
1759	!-- output file afterwards to increase the performance.
1760	DO i = nxl, nxr
1761	DO k = nzb_do, nzt_do
1762	local_2d_sections_l(i,is,k) = local_2d(i,k)
1763	ENDDO
1764	ENDDO
1765	#endif
1766	ENDIF
1767
1768	ELSE
1769
1770	IF ( data_output_2d_on_each_pe ) THEN
1771	!
1772	!-- Output of partial arrays on each PE. If the cross
1773	!-- section does not reside on the PE, output special
1774	!-- index values.
1775	#if defined( __netcdf )
1776	IF ( myid == 0 ) THEN
1777	WRITE ( 22 ) time_since_reference_point, &
1778	do2d_xz_time_count(av), av
1779	ENDIF
1780	#endif
1781	DO i = 0, io_blocks-1
1782	IF ( i == io_group ) THEN
1783	IF ( ( section(is,s_ind) >= nys .AND. &
1784	section(is,s_ind) <= nyn ) .OR. &
1785	( section(is,s_ind) == -1 .AND. &
1786	nys-1 == -1 ) ) &
1787	THEN
1788	WRITE (22) nxl, nxr, nzb_do, nzt_do, nzb, nzt+1
1789	WRITE (22) local_2d
1790	ELSE
1791	WRITE (22) -1, -1, -1, -1, -1, -1
1792	ENDIF
1793	ENDIF
1794	#if defined( __parallel )
1795	CALL MPI_BARRIER( comm2d, ierr )
1796	#endif
1797	ENDDO
1798
1799	ELSE
1800	!
1801	!-- PE0 receives partial arrays from all processors of the
1802	!-- respective cross section and outputs them. Here a
1803	!-- barrier has to be set, because otherwise
1804	!-- "-MPI- FATAL: Remote protocol queue full" may occur.
1805	CALL MPI_BARRIER( comm2d, ierr )
1806
1807	ngp = ( nxr-nxl + 1 ) * ( nzt_do-nzb_do + 1 )
1808	IF ( myid == 0 ) THEN
1809	!
1810	!-- Local array can be relocated directly.
1811	IF ( ( section(is,s_ind) >= nys .AND. &
1812	section(is,s_ind) <= nyn ) .OR. &
1813	( section(is,s_ind) == -1 .AND. &
1814	nys-1 == -1 ) ) THEN
1815	total_2d(nxl:nxr,nzb_do:nzt_do) = local_2d
1816	ENDIF
1817	!
1818	!-- Receive data from all other PEs.
1819	DO n = 1, numprocs-1
1820	!
1821	!-- Receive index limits first, then array.
1822	!-- Index limits are received in arbitrary order from
1823	!-- the PEs.
1824	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
1825	MPI_ANY_SOURCE, 0, comm2d, &
1826	status, ierr )
1827	!
1828	!-- Not all PEs have data for XZ-cross-section.
1829	IF ( ind(1) /= -9999 ) THEN
1830	sender = status(MPI_SOURCE)
1831	DEALLOCATE( local_2d )
1832	ALLOCATE( local_2d(ind(1):ind(2), &
1833	ind(3):ind(4)) )
1834	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
1835	MPI_REAL, sender, 1, comm2d, &
1836	status, ierr )
1837	total_2d(ind(1):ind(2),ind(3):ind(4)) = &
1838	local_2d
1839	ENDIF
1840	ENDDO
1841	!
1842	!-- Relocate the local array for the next loop increment
1843	DEALLOCATE( local_2d )
1844	ALLOCATE( local_2d(nxl:nxr,nzb_do:nzt_do) )
1845
1846	#if defined( __netcdf )
1847	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
1848	id_var_do2d(av,if), &
1849	total_2d(0:nx,nzb_do:nzt_do), &
1850	start = (/ 1, is, 1, do2d_xz_time_count(av) /), &
1851	count = (/ nx+1, 1, nzt_do-nzb_do+1, 1 /) )
1852	CALL netcdf_handle_error( 'data_output_2d', 58 )
1853	#endif
1854
1855	ELSE
1856	!
1857	!-- If the cross section resides on the PE, send the
1858	!-- local index limits, otherwise send -9999 to PE0.
1859	IF ( ( section(is,s_ind) >= nys .AND. &
1860	section(is,s_ind) <= nyn ) .OR. &
1861	( section(is,s_ind) == -1 .AND. nys-1 == -1 ) ) &
1862	THEN
1863	ind(1) = nxl; ind(2) = nxr
1864	ind(3) = nzb_do; ind(4) = nzt_do
1865	ELSE
1866	ind(1) = -9999; ind(2) = -9999
1867	ind(3) = -9999; ind(4) = -9999
1868	ENDIF
1869	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
1870	comm2d, ierr )
1871	!
1872	!-- If applicable, send data to PE0.
1873	IF ( ind(1) /= -9999 ) THEN
1874	CALL MPI_SEND( local_2d(nxl,nzb_do), ngp, &
1875	MPI_REAL, 0, 1, comm2d, ierr )
1876	ENDIF
1877	ENDIF
1878	!
1879	!-- A barrier has to be set, because otherwise some PEs may
1880	!-- proceed too fast so that PE0 may receive wrong data on
1881	!-- tag 0
1882	CALL MPI_BARRIER( comm2d, ierr )
1883	ENDIF
1884
1885	ENDIF
1886	#else
1887	#if defined( __netcdf )
1888	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
1889	id_var_do2d(av,if), &
1890	local_2d(nxl:nxr,nzb_do:nzt_do), &
1891	start = (/ 1, is, 1, do2d_xz_time_count(av) /), &
1892	count = (/ nx+1, 1, nzt_do-nzb_do+1, 1 /) )
1893	CALL netcdf_handle_error( 'data_output_2d', 451 )
1894	#endif
1895	#endif
1896
1897	CASE ( 'yz' )
1898	!
1899	!-- Update the netCDF yz cross section time axis.
1900	!-- In case of parallel output, this is only done by PE0
1901	!-- to increase the performance.
1902	IF ( simulated_time /= do2d_yz_last_time(av) ) THEN
1903	do2d_yz_time_count(av) = do2d_yz_time_count(av) + 1
1904	do2d_yz_last_time(av) = simulated_time
1905	IF ( myid == 0 ) THEN
1906	IF ( .NOT. data_output_2d_on_each_pe &
1907	.OR. netcdf_data_format > 4 ) &
1908	THEN
1909	#if defined( __netcdf )
1910	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
1911	id_var_time_yz(av), &
1912	(/ time_since_reference_point /), &
1913	start = (/ do2d_yz_time_count(av) /), &
1914	count = (/ 1 /) )
1915	CALL netcdf_handle_error( 'data_output_2d', 59 )
1916	#endif
1917	ENDIF
1918	ENDIF
1919	ENDIF
1920
1921	!
1922	!-- If required, carry out averaging along x
1923	IF ( section(is,s_ind) == -1 ) THEN
1924
1925	ALLOCATE( local_2d_l(nys:nyn,nzb_do:nzt_do) )
1926	local_2d_l = 0.0_wp
1927	ngp = ( nyn-nys+1 ) * ( nzt_do-nzb_do+1 )
1928	!
1929	!-- First local averaging on the PE
1930	DO k = nzb_do, nzt_do
1931	DO j = nys, nyn
1932	DO i = nxl, nxr
1933	local_2d_l(j,k) = local_2d_l(j,k) + &
1934	local_pf(i,j,k)
1935	ENDDO
1936	ENDDO
1937	ENDDO
1938	#if defined( __parallel )
1939	!
1940	!-- Now do the averaging over all PEs along x
1941	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
1942	CALL MPI_ALLREDUCE( local_2d_l(nys,nzb_do), &
1943	local_2d(nys,nzb_do), ngp, MPI_REAL, &
1944	MPI_SUM, comm1dx, ierr )
1945	#else
1946	local_2d = local_2d_l
1947	#endif
1948	local_2d = local_2d / ( nx + 1.0_wp )
1949
1950	DEALLOCATE( local_2d_l )
1951
1952	ELSE
1953	!
1954	!-- Just store the respective section on the local array
1955	!-- (but only if it is available on this PE!)
1956	IF ( section(is,s_ind) >= nxl .AND. section(is,s_ind) <= nxr ) &
1957	THEN
1958	local_2d = local_pf(section(is,s_ind),:,nzb_do:nzt_do)
1959	ENDIF
1960
1961	ENDIF
1962
1963	#if defined( __parallel )
1964	IF ( netcdf_data_format > 4 ) THEN
1965	!
1966	!-- Output in netCDF4/HDF5 format.
1967	!-- Output only on those PEs where the respective cross
1968	!-- sections reside. Cross sections averaged along x are
1969	!-- output on the respective first PE along x (myidx=0).
1970	IF ( ( section(is,s_ind) >= nxl .AND. &
1971	section(is,s_ind) <= nxr ) .OR. &
1972	( section(is,s_ind) == -1 .AND. myidx == 0 ) ) THEN
1973	#if defined( __netcdf )
1974	!
1975	!-- For parallel output, all cross sections are first
1976	!-- stored here on a local array and will be written to the
1977	!-- output file afterwards to increase the performance.
1978	DO j = nys, nyn
1979	DO k = nzb_do, nzt_do
1980	local_2d_sections_l(is,j,k) = local_2d(j,k)
1981	ENDDO
1982	ENDDO
1983	#endif
1984	ENDIF
1985
1986	ELSE
1987
1988	IF ( data_output_2d_on_each_pe ) THEN
1989	!
1990	!-- Output of partial arrays on each PE. If the cross
1991	!-- section does not reside on the PE, output special
1992	!-- index values.
1993	#if defined( __netcdf )
1994	IF ( myid == 0 ) THEN
1995	WRITE ( 23 ) time_since_reference_point, &
1996	do2d_yz_time_count(av), av
1997	ENDIF
1998	#endif
1999	DO i = 0, io_blocks-1
2000	IF ( i == io_group ) THEN
2001	IF ( ( section(is,s_ind) >= nxl .AND. &
2002	section(is,s_ind) <= nxr ) .OR. &
2003	( section(is,s_ind) == -1 .AND. &
2004	nxl-1 == -1 ) ) &
2005	THEN
2006	WRITE (23) nys, nyn, nzb_do, nzt_do, nzb, nzt+1
2007	WRITE (23) local_2d
2008	ELSE
2009	WRITE (23) -1, -1, -1, -1, -1, -1
2010	ENDIF
2011	ENDIF
2012	#if defined( __parallel )
2013	CALL MPI_BARRIER( comm2d, ierr )
2014	#endif
2015	ENDDO
2016
2017	ELSE
2018	!
2019	!-- PE0 receives partial arrays from all processors of the
2020	!-- respective cross section and outputs them. Here a
2021	!-- barrier has to be set, because otherwise
2022	!-- "-MPI- FATAL: Remote protocol queue full" may occur.
2023	CALL MPI_BARRIER( comm2d, ierr )
2024
2025	ngp = ( nyn-nys+1 ) * ( nzt_do-nzb_do+1 )
2026	IF ( myid == 0 ) THEN
2027	!
2028	!-- Local array can be relocated directly.
2029	IF ( ( section(is,s_ind) >= nxl .AND. &
2030	section(is,s_ind) <= nxr ) .OR. &
2031	( section(is,s_ind) == -1 .AND. nxl-1 == -1 ) ) &
2032	THEN
2033	total_2d(nys:nyn,nzb_do:nzt_do) = local_2d
2034	ENDIF
2035	!
2036	!-- Receive data from all other PEs.
2037	DO n = 1, numprocs-1
2038	!
2039	!-- Receive index limits first, then array.
2040	!-- Index limits are received in arbitrary order from
2041	!-- the PEs.
2042	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
2043	MPI_ANY_SOURCE, 0, comm2d, &
2044	status, ierr )
2045	!
2046	!-- Not all PEs have data for YZ-cross-section.
2047	IF ( ind(1) /= -9999 ) THEN
2048	sender = status(MPI_SOURCE)
2049	DEALLOCATE( local_2d )
2050	ALLOCATE( local_2d(ind(1):ind(2), &
2051	ind(3):ind(4)) )
2052	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
2053	MPI_REAL, sender, 1, comm2d, &
2054	status, ierr )
2055	total_2d(ind(1):ind(2),ind(3):ind(4)) = &
2056	local_2d
2057	ENDIF
2058	ENDDO
2059	!
2060	!-- Relocate the local array for the next loop increment
2061	DEALLOCATE( local_2d )
2062	ALLOCATE( local_2d(nys:nyn,nzb_do:nzt_do) )
2063
2064	#if defined( __netcdf )
2065	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2066	id_var_do2d(av,if), &
2067	total_2d(0:ny,nzb_do:nzt_do), &
2068	start = (/ is, 1, 1, do2d_yz_time_count(av) /), &
2069	count = (/ 1, ny+1, nzt_do-nzb_do+1, 1 /) )
2070	CALL netcdf_handle_error( 'data_output_2d', 61 )
2071	#endif
2072
2073	ELSE
2074	!
2075	!-- If the cross section resides on the PE, send the
2076	!-- local index limits, otherwise send -9999 to PE0.
2077	IF ( ( section(is,s_ind) >= nxl .AND. &
2078	section(is,s_ind) <= nxr ) .OR. &
2079	( section(is,s_ind) == -1 .AND. nxl-1 == -1 ) ) &
2080	THEN
2081	ind(1) = nys; ind(2) = nyn
2082	ind(3) = nzb_do; ind(4) = nzt_do
2083	ELSE
2084	ind(1) = -9999; ind(2) = -9999
2085	ind(3) = -9999; ind(4) = -9999
2086	ENDIF
2087	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
2088	comm2d, ierr )
2089	!
2090	!-- If applicable, send data to PE0.
2091	IF ( ind(1) /= -9999 ) THEN
2092	CALL MPI_SEND( local_2d(nys,nzb_do), ngp, &
2093	MPI_REAL, 0, 1, comm2d, ierr )
2094	ENDIF
2095	ENDIF
2096	!
2097	!-- A barrier has to be set, because otherwise some PEs may
2098	!-- proceed too fast so that PE0 may receive wrong data on
2099	!-- tag 0
2100	CALL MPI_BARRIER( comm2d, ierr )
2101	ENDIF
2102
2103	ENDIF
2104	#else
2105	#if defined( __netcdf )
2106	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2107	id_var_do2d(av,if), &
2108	local_2d(nys:nyn,nzb_do:nzt_do), &
2109	start = (/ is, 1, 1, do2d_xz_time_count(av) /), &
2110	count = (/ 1, ny+1, nzt_do-nzb_do+1, 1 /) )
2111	CALL netcdf_handle_error( 'data_output_2d', 452 )
2112	#endif
2113	#endif
2114
2115	END SELECT
2116
2117	is = is + 1
2118	ENDDO loop1
2119
2120	!
2121	!-- For parallel output, all data were collected before on a local array
2122	!-- and are written now to the netcdf file. This must be done to increase
2123	!-- the performance of the parallel output.
2124	#if defined( __netcdf )
2125	IF ( netcdf_data_format > 4 ) THEN
2126
2127	SELECT CASE ( mode )
2128
2129	CASE ( 'xy' )
2130	IF ( two_d ) THEN
2131	nis = 1
2132	two_d = .FALSE.
2133	ELSE
2134	nis = ns
2135	ENDIF
2136	!
2137	!-- Do not output redundant ghost point data except for the
2138	!-- boundaries of the total domain.
2139	! IF ( nxr == nx .AND. nyn /= ny ) THEN
2140	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2141	! id_var_do2d(av,if), &
2142	! local_2d_sections(nxl:nxr+1, &
2143	! nys:nyn,1:nis), &
2144	! start = (/ nxl+1, nys+1, 1, &
2145	! do2d_xy_time_count(av) /), &
2146	! count = (/ nxr-nxl+2, &
2147	! nyn-nys+1, nis, 1 &
2148	! /) )
2149	! ELSEIF ( nxr /= nx .AND. nyn == ny ) THEN
2150	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2151	! id_var_do2d(av,if), &
2152	! local_2d_sections(nxl:nxr, &
2153	! nys:nyn+1,1:nis), &
2154	! start = (/ nxl+1, nys+1, 1, &
2155	! do2d_xy_time_count(av) /), &
2156	! count = (/ nxr-nxl+1, &
2157	! nyn-nys+2, nis, 1 &
2158	! /) )
2159	! ELSEIF ( nxr == nx .AND. nyn == ny ) THEN
2160	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2161	! id_var_do2d(av,if), &
2162	! local_2d_sections(nxl:nxr+1, &
2163	! nys:nyn+1,1:nis), &
2164	! start = (/ nxl+1, nys+1, 1, &
2165	! do2d_xy_time_count(av) /), &
2166	! count = (/ nxr-nxl+2, &
2167	! nyn-nys+2, nis, 1 &
2168	! /) )
2169	! ELSE
2170	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2171	id_var_do2d(av,if), &
2172	local_2d_sections(nxl:nxr, &
2173	nys:nyn,1:nis), &
2174	start = (/ nxl+1, nys+1, 1, &
2175	do2d_xy_time_count(av) /), &
2176	count = (/ nxr-nxl+1, &
2177	nyn-nys+1, nis, 1 &
2178	/) )
2179	! ENDIF
2180
2181	CALL netcdf_handle_error( 'data_output_2d', 55 )
2182
2183	CASE ( 'xz' )
2184	!
2185	!-- First, all PEs get the information of all cross-sections.
2186	!-- Then the data are written to the output file by all PEs
2187	!-- while NF90_COLLECTIVE is set in subroutine
2188	!-- define_netcdf_header. Although redundant information are
2189	!-- written to the output file in that case, the performance
2190	!-- is significantly better compared to the case where only
2191	!-- the first row of PEs in x-direction (myidx = 0) is given
2192	!-- the output while NF90_INDEPENDENT is set.
2193	IF ( npey /= 1 ) THEN
2194
2195	#if defined( __parallel )
2196	!
2197	!-- Distribute data over all PEs along y
2198	ngp = ( nxr-nxl+1 ) * ( nzt_do-nzb_do+1 ) * ns
2199	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
2200	CALL MPI_ALLREDUCE( local_2d_sections_l(nxl,1,nzb_do), &
2201	local_2d_sections(nxl,1,nzb_do), &
2202	ngp, MPI_REAL, MPI_SUM, comm1dy, &
2203	ierr )
2204	#else
2205	local_2d_sections = local_2d_sections_l
2206	#endif
2207	ENDIF
2208	!
2209	!-- Do not output redundant ghost point data except for the
2210	!-- boundaries of the total domain.
2211	! IF ( nxr == nx ) THEN
2212	! nc_stat = NF90_PUT_VAR( id_set_xz(av), &
2213	! id_var_do2d(av,if), &
2214	! local_2d_sections(nxl:nxr+1,1:ns, &
2215	! nzb_do:nzt_do), &
2216	! start = (/ nxl+1, 1, 1, &
2217	! do2d_xz_time_count(av) /), &
2218	! count = (/ nxr-nxl+2, ns, nzt_do-nzb_do+1, &
2219	! 1 /) )
2220	! ELSE
2221	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
2222	id_var_do2d(av,if), &
2223	local_2d_sections(nxl:nxr,1:ns, &
2224	nzb_do:nzt_do), &
2225	start = (/ nxl+1, 1, 1, &
2226	do2d_xz_time_count(av) /), &
2227	count = (/ nxr-nxl+1, ns, nzt_do-nzb_do+1, &
2228	1 /) )
2229	! ENDIF
2230
2231	CALL netcdf_handle_error( 'data_output_2d', 57 )
2232
2233	CASE ( 'yz' )
2234	!
2235	!-- First, all PEs get the information of all cross-sections.
2236	!-- Then the data are written to the output file by all PEs
2237	!-- while NF90_COLLECTIVE is set in subroutine
2238	!-- define_netcdf_header. Although redundant information are
2239	!-- written to the output file in that case, the performance
2240	!-- is significantly better compared to the case where only
2241	!-- the first row of PEs in y-direction (myidy = 0) is given
2242	!-- the output while NF90_INDEPENDENT is set.
2243	IF ( npex /= 1 ) THEN
2244
2245	#if defined( __parallel )
2246	!
2247	!-- Distribute data over all PEs along x
2248	ngp = ( nyn-nys+1 ) * ( nzt-nzb + 2 ) * ns
2249	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
2250	CALL MPI_ALLREDUCE( local_2d_sections_l(1,nys,nzb_do), &
2251	local_2d_sections(1,nys,nzb_do), &
2252	ngp, MPI_REAL, MPI_SUM, comm1dx, &
2253	ierr )
2254	#else
2255	local_2d_sections = local_2d_sections_l
2256	#endif
2257	ENDIF
2258	!
2259	!-- Do not output redundant ghost point data except for the
2260	!-- boundaries of the total domain.
2261	! IF ( nyn == ny ) THEN
2262	! nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2263	! id_var_do2d(av,if), &
2264	! local_2d_sections(1:ns, &
2265	! nys:nyn+1,nzb_do:nzt_do), &
2266	! start = (/ 1, nys+1, 1, &
2267	! do2d_yz_time_count(av) /), &
2268	! count = (/ ns, nyn-nys+2, &
2269	! nzt_do-nzb_do+1, 1 /) )
2270	! ELSE
2271	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2272	id_var_do2d(av,if), &
2273	local_2d_sections(1:ns,nys:nyn, &
2274	nzb_do:nzt_do), &
2275	start = (/ 1, nys+1, 1, &
2276	do2d_yz_time_count(av) /), &
2277	count = (/ ns, nyn-nys+1, &
2278	nzt_do-nzb_do+1, 1 /) )
2279	! ENDIF
2280
2281	CALL netcdf_handle_error( 'data_output_2d', 60 )
2282
2283	CASE DEFAULT
2284	message_string = 'unknown cross-section: ' // TRIM( mode )
2285	CALL message( 'data_output_2d', 'PA0180', 1, 2, 0, 6, 0 )
2286
2287	END SELECT
2288
2289	ENDIF
2290	#endif
2291	ENDIF
2292
2293	if = if + 1
2294	l = MAX( 2, LEN_TRIM( do2d(av,if) ) )
2295	do2d_mode = do2d(av,if)(l-1:l)
2296
2297	ENDDO
2298
2299	!
2300	!-- Deallocate temporary arrays.
2301	IF ( ALLOCATED( level_z ) ) DEALLOCATE( level_z )
2302	IF ( netcdf_data_format > 4 ) THEN
2303	DEALLOCATE( local_pf, local_2d, local_2d_sections )
2304	IF( mode == 'xz' .OR. mode == 'yz' ) DEALLOCATE( local_2d_sections_l )
2305	ENDIF
2306	#if defined( __parallel )
2307	IF ( .NOT. data_output_2d_on_each_pe .AND. myid == 0 ) THEN
2308	DEALLOCATE( total_2d )
2309	ENDIF
2310	#endif
2311
2312	!
2313	!-- Close plot output file.
2314	file_id = 20 + s_ind
2315
2316	IF ( data_output_2d_on_each_pe ) THEN
2317	DO i = 0, io_blocks-1
2318	IF ( i == io_group ) THEN
2319	CALL close_file( file_id )
2320	ENDIF
2321	#if defined( __parallel )
2322	CALL MPI_BARRIER( comm2d, ierr )
2323	#endif
2324	ENDDO
2325	ELSE
2326	IF ( myid == 0 ) CALL close_file( file_id )
2327	ENDIF
2328
2329	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
2330
2331	END SUBROUTINE data_output_2d

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

Download in other formats:

| Impressum | ©Leibniz Universität Hannover |