Home

Context Navigation

source: palm/trunk/SOURCE/data_output_2d.f90 @ 3990

Last change on this file since 3990 was 3987, checked in by kanani, 6 years ago
clean up location, debug and error messages
Property svn:keywords set to `Id`
File size: 96.9 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 1997-2019 Leibniz Universitaet Hannover
18	!------------------------------------------------------------------------------!
19	!
20	! Current revisions:
21	! ------------------
22	!
23	!
24	! Former revisions:
25	! -----------------
26	! $Id: data_output_2d.f90 3987 2019-05-22 09:52:13Z suehring $
27	! Introduce alternative switch for debug output during timestepping
28	!
29	! 3943 2019-05-02 09:50:41Z maronga
30	! Added output of qsws for green roofs.
31	!
32	! 3885 2019-04-11 11:29:34Z kanani
33	! Changes related to global restructuring of location messages and introduction
34	! of additional debug messages
35	!
36	! 3766 2019-02-26 16:23:41Z raasch
37	! unused variables removed
38	!
39	! 3655 2019-01-07 16:51:22Z knoop
40	! Bugfix: use time_since_reference_point instead of simulated_time (relevant
41	! when using wall/soil spinup)
42	!
43	! 3637 2018-12-20 01:51:36Z knoop
44	! Implementation of the PALM module interface
45	!
46	! 3597 2018-12-04 08:40:18Z maronga
47	! Added theta_2m
48	!
49	! 3589 2018-11-30 15:09:51Z suehring
50	! Move the control parameter "salsa" from salsa_mod to control_parameters
51	! (M. Kurppa)
52	!
53	! 3582 2018-11-29 19:16:36Z suehring
54	! Remove fill_value from bio_data_output_2d call
55	! dom_dwd_user, Schrempf:
56	! Clean up of biometeorology calls,
57	! remove uv exposure model code, this is now part of biometeorology_mod.
58	!
59	! 3554 2018-11-22 11:24:52Z gronemeier
60	! - add variable description
61	! - rename variable 'if' into 'ivar'
62	! - removed namelist LOCAL
63	! - removed variable rtext
64	!
65	! 3525 2018-11-14 16:06:14Z kanani
66	! Changes related to clean-up of biometeorology (dom_dwd_user)
67	!
68	! 3467 2018-10-30 19:05:21Z suehring
69	! Implementation of a new aerosol module salsa.
70	!
71	! 3448 2018-10-29 18:14:31Z kanani
72	! Add biometeorology
73	!
74	! 3421 2018-10-24 18:39:32Z gronemeier
75	! Renamed output variables
76	!
77	! 3419 2018-10-24 17:27:31Z gronemeier
78	! minor formatting (kanani)
79	! chem_data_output_2d subroutine added (basit)
80	!
81	! 3294 2018-10-01 02:37:10Z raasch
82	! changes concerning modularization of ocean option
83	!
84	! 3274 2018-09-24 15:42:55Z knoop
85	! Modularization of all bulk cloud physics code components
86	!
87	! 3241 2018-09-12 15:02:00Z raasch
88	! unused variables removed
89	!
90	! 3176 2018-07-26 17:12:48Z suehring
91	! Remove output of latent heat flux at urban-surfaces and set fill values
92	! instead
93	!
94	! 3052 2018-05-31 06:11:20Z raasch
95	! Do not open FORTRAN binary files in case of parallel netCDF I/O
96	!
97	! 3049 2018-05-29 13:52:36Z Giersch
98	! Error messages revised
99	!
100	! 3045 2018-05-28 07:55:41Z Giersch
101	! Error messages revised
102	!
103	! 3014 2018-05-09 08:42:38Z maronga
104	! Added nzb_do and nzt_do for some modules for 2d output
105	!
106	! 3004 2018-04-27 12:33:25Z Giersch
107	! precipitation_rate removed, case prr*_xy removed, to_be_resorted have to point
108	! to ql_vp_av and not to ql_vp, allocation checks implemented (averaged data
109	! will be assigned to fill values if no allocation happened so far)
110	!
111	! 2963 2018-04-12 14:47:44Z suehring
112	! Introduce index for vegetation/wall, pavement/green-wall and water/window
113	! surfaces, for clearer access of surface fraction, albedo, emissivity, etc. .
114	!
115	! 2817 2018-02-19 16:32:21Z knoop
116	! Preliminary gust module interface implemented
117	!
118	! 2805 2018-02-14 17:00:09Z suehring
119	! Consider also default-type surfaces for surface temperature output.
120	!
121	! 2797 2018-02-08 13:24:35Z suehring
122	! Enable output of ground-heat flux also at urban surfaces.
123	!
124	! 2743 2018-01-12 16:03:39Z suehring
125	! In case of natural- and urban-type surfaces output surfaces fluxes in W/m2.
126	!
127	! 2742 2018-01-12 14:59:47Z suehring
128	! Enable output of surface temperature
129	!
130	! 2735 2018-01-11 12:01:27Z suehring
131	! output of r_a moved from land-surface to consider also urban-type surfaces
132	!
133	! 2718 2018-01-02 08:49:38Z maronga
134	! Corrected "Former revisions" section
135	!
136	! 2696 2017-12-14 17:12:51Z kanani
137	! Change in file header (GPL part)
138	! Implementation of uv exposure model (FK)
139	! Implementation of turbulence_closure_mod (TG)
140	! Set fill values at topography grid points or e.g. non-natural-type surface
141	! in case of LSM output (MS)
142	!
143	! 2512 2017-10-04 08:26:59Z raasch
144	! upper bounds of cross section output changed from nx+1,ny+1 to nx,ny
145	! no output of ghost layer data
146	!
147	! 2292 2017-06-20 09:51:42Z schwenkel
148	! Implementation of new microphysic scheme: cloud_scheme = 'morrison'
149	! includes two more prognostic equations for cloud drop concentration (nc)
150	! and cloud water content (qc).
151	!
152	! 2277 2017-06-12 10:47:51Z kanani
153	! Removed unused variables do2d_xy_n, do2d_xz_n, do2d_yz_n
154	!
155	! 2233 2017-05-30 18:08:54Z suehring
156	!
157	! 2232 2017-05-30 17:47:52Z suehring
158	! Adjustments to new surface concept
159	!
160	!
161	! 2190 2017-03-21 12:16:43Z raasch
162	! bugfix for r2031: string rho replaced by rho_ocean
163	!
164	! 2031 2016-10-21 15:11:58Z knoop
165	! renamed variable rho to rho_ocean and rho_av to rho_ocean_av
166	!
167	! 2000 2016-08-20 18:09:15Z knoop
168	! Forced header and separation lines into 80 columns
169	!
170	! 1980 2016-07-29 15:51:57Z suehring
171	! Bugfix, in order to steer user-defined output, setting flag found explicitly
172	! to .F.
173	!
174	! 1976 2016-07-27 13:28:04Z maronga
175	! Output of radiation quantities is now done directly in the respective module
176	!
177	! 1972 2016-07-26 07:52:02Z maronga
178	! Output of land surface quantities is now done directly in the respective
179	! module
180	!
181	! 1960 2016-07-12 16:34:24Z suehring
182	! Scalar surface flux added
183	! Rename INTEGER variable s into s_ind, as s is already assigned to scalar
184	!
185	! 1849 2016-04-08 11:33:18Z hoffmann
186	! precipitation_amount, precipitation_rate, prr moved to arrays_3d
187	!
188	! 1822 2016-04-07 07:49:42Z hoffmann
189	! Output of bulk cloud physics simplified.
190	!
191	! 1788 2016-03-10 11:01:04Z maronga
192	! Added output of z0q
193	!
194	! 1783 2016-03-06 18:36:17Z raasch
195	! name change of netcdf routines and module + related changes
196	!
197	! 1745 2016-02-05 13:06:51Z gronemeier
198	! Bugfix: test if time axis limit exceeds moved to point after call of check_open
199	!
200	! 1703 2015-11-02 12:38:44Z raasch
201	! bugfix for output of single (*) xy-sections in case of parallel netcdf I/O
202	!
203	! 1701 2015-11-02 07:43:04Z maronga
204	! Bugfix in output of RRTGM data
205	!
206	! 1691 2015-10-26 16:17:44Z maronga
207	! Added output of Obukhov length (ol) and radiative heating rates for RRTMG.
208	! Formatting corrections.
209	!
210	! 1682 2015-10-07 23:56:08Z knoop
211	! Code annotations made doxygen readable
212	!
213	! 1585 2015-04-30 07:05:52Z maronga
214	! Added support for RRTMG
215	!
216	! 1555 2015-03-04 17:44:27Z maronga
217	! Added output of r_a and r_s
218	!
219	! 1551 2015-03-03 14:18:16Z maronga
220	! Added suppport for land surface model and radiation model output. In the course
221	! of this action, the limits for vertical loops have been changed (from nzb and
222	! nzt+1 to nzb_do and nzt_do, respectively in order to allow soil model output).
223	! Moreover, a new vertical grid zs was introduced.
224	!
225	! 1359 2014-04-11 17:15:14Z hoffmann
226	! New particle structure integrated.
227	!
228	! 1353 2014-04-08 15:21:23Z heinze
229	! REAL constants provided with KIND-attribute
230	!
231	! 1327 2014-03-21 11:00:16Z raasch
232	! parts concerning iso2d output removed,
233	! -netcdf output queries
234	!
235	! 1320 2014-03-20 08:40:49Z raasch
236	! ONLY-attribute added to USE-statements,
237	! kind-parameters added to all INTEGER and REAL declaration statements,
238	! kinds are defined in new module kinds,
239	! revision history before 2012 removed,
240	! comment fields (!:) to be used for variable explanations added to
241	! all variable declaration statements
242	!
243	! 1318 2014-03-17 13:35:16Z raasch
244	! barrier argument removed from cpu_log.
245	! module interfaces removed
246	!
247	! 1311 2014-03-14 12:13:39Z heinze
248	! bugfix: close #if defined( __netcdf )
249	!
250	! 1308 2014-03-13 14:58:42Z fricke
251	! +local_2d_sections, local_2d_sections_l, ns
252	! Check, if the limit of the time dimension is exceeded for parallel output
253	! To increase the performance for parallel output, the following is done:
254	! - Update of time axis is only done by PE0
255	! - Cross sections are first stored on a local array and are written
256	! collectively to the output file by all PEs.
257	!
258	! 1115 2013-03-26 18:16:16Z hoffmann
259	! ql is calculated by calc_liquid_water_content
260	!
261	! 1076 2012-12-05 08:30:18Z hoffmann
262	! Bugfix in output of ql
263	!
264	! 1065 2012-11-22 17:42:36Z hoffmann
265	! Bugfix: Output of cross sections of ql
266	!
267	! 1053 2012-11-13 17:11:03Z hoffmann
268	! +qr, nr, qc and cross sections
269	!
270	! 1036 2012-10-22 13:43:42Z raasch
271	! code put under GPL (PALM 3.9)
272	!
273	! 1031 2012-10-19 14:35:30Z raasch
274	! netCDF4 without parallel file support implemented
275	!
276	! 1007 2012-09-19 14:30:36Z franke
277	! Bugfix: missing calculation of ql_vp added
278	!
279	! 978 2012-08-09 08:28:32Z fricke
280	! +z0h
281	!
282	! Revision 1.1 1997/08/11 06:24:09 raasch
283	! Initial revision
284	!
285	!
286	! Description:
287	! ------------
288	!> Data output of cross-sections in netCDF format or binary format
289	!> to be later converted to NetCDF by helper routine combine_plot_fields.
290	!> Attention: The position of the sectional planes is still not always computed
291	!> --------- correctly. (zu is used always)!
292	!------------------------------------------------------------------------------!
293	SUBROUTINE data_output_2d( mode, av )
294
295
296	USE arrays_3d, &
297	ONLY: dzw, d_exner, e, heatflux_output_conversion, p, pt, q, ql, ql_c, ql_v, s, tend, u, &
298	v, vpt, w, waterflux_output_conversion, zu, zw
299
300	USE averaging
301
302	USE basic_constants_and_equations_mod, &
303	ONLY: c_p, lv_d_cp, l_v
304
305	USE bulk_cloud_model_mod, &
306	ONLY: bulk_cloud_model
307
308	USE control_parameters, &
309	ONLY: data_output_2d_on_each_pe, &
310	data_output_xy, data_output_xz, data_output_yz, &
311	debug_output_timestep, &
312	do2d, &
313	do2d_xy_last_time, do2d_xy_time_count, &
314	do2d_xz_last_time, do2d_xz_time_count, &
315	do2d_yz_last_time, do2d_yz_time_count, &
316	ibc_uv_b, io_blocks, io_group, message_string, &
317	ntdim_2d_xy, ntdim_2d_xz, ntdim_2d_yz, &
318	psolver, section, &
319	time_since_reference_point
320
321	USE cpulog, &
322	ONLY: cpu_log, log_point
323
324	USE indices, &
325	ONLY: nbgp, nx, nxl, nxlg, nxr, nxrg, ny, nyn, nyng, nys, nysg, &
326	nzb, nzt, wall_flags_0
327
328	USE kinds
329
330	USE land_surface_model_mod, &
331	ONLY: zs
332
333	USE module_interface, &
334	ONLY: module_interface_data_output_2d
335
336	#if defined( __netcdf )
337	USE NETCDF
338	#endif
339
340	USE netcdf_interface, &
341	ONLY: fill_value, id_set_xy, id_set_xz, id_set_yz, id_var_do2d, &
342	id_var_time_xy, id_var_time_xz, id_var_time_yz, nc_stat, &
343	netcdf_data_format, netcdf_handle_error
344
345	USE particle_attributes, &
346	ONLY: grid_particles, number_of_particles, particle_advection_start, &
347	particles, prt_count
348
349	USE pegrid
350
351	USE surface_mod, &
352	ONLY: ind_pav_green, ind_veg_wall, ind_wat_win, surf_def_h, &
353	surf_lsm_h, surf_usm_h
354
355	USE turbulence_closure_mod, &
356	ONLY: tcm_data_output_2d
357
358
359	IMPLICIT NONE
360
361	CHARACTER (LEN=2) :: do2d_mode !< output mode of variable ('xy', 'xz', 'yz')
362	CHARACTER (LEN=2) :: mode !< mode with which the routine is called ('xy', 'xz', 'yz')
363	CHARACTER (LEN=4) :: grid !< string defining the vertical grid
364
365	INTEGER(iwp) :: av !< flag for (non-)average output
366	INTEGER(iwp) :: ngp !< number of grid points of an output slice
367	INTEGER(iwp) :: file_id !< id of output files
368	INTEGER(iwp) :: flag_nr !< number of masking flag
369	INTEGER(iwp) :: i !< loop index
370	INTEGER(iwp) :: iis !< vertical index of a xy slice in array 'local_2d_sections'
371	INTEGER(iwp) :: is !< slice index
372	INTEGER(iwp) :: ivar !< variable index
373	INTEGER(iwp) :: j !< loop index
374	INTEGER(iwp) :: k !< loop index
375	INTEGER(iwp) :: l !< loop index
376	INTEGER(iwp) :: layer_xy !< vertical index of a xy slice in array 'local_pf'
377	INTEGER(iwp) :: m !< loop index
378	INTEGER(iwp) :: n !< loop index
379	INTEGER(iwp) :: nis !< number of vertical slices to be written via parallel NetCDF output
380	INTEGER(iwp) :: ns !< number of output slices
381	INTEGER(iwp) :: nzb_do !< lower limit of the data field (usually nzb)
382	INTEGER(iwp) :: nzt_do !< upper limit of the data field (usually nzt+1)
383	INTEGER(iwp) :: s_ind !< index of slice types (xy=1, xz=2, yz=3)
384	INTEGER(iwp) :: sender !< PE id of sending PE
385	INTEGER(iwp) :: ind(4) !< index limits (lower/upper bounds) of array 'local_2d'
386
387	LOGICAL :: found !< true if output variable was found
388	LOGICAL :: resorted !< true if variable is resorted
389	LOGICAL :: two_d !< true if variable is only two dimensional
390
391	REAL(wp) :: mean_r !< mean particle radius
392	REAL(wp) :: s_r2 !< sum( particle-radius**2 )
393	REAL(wp) :: s_r3 !< sum( particle-radius**3 )
394
395	REAL(wp), DIMENSION(:), ALLOCATABLE :: level_z !< z levels for output array
396	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: local_2d !< local 2-dimensional array containing output values
397	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: local_2d_l !< local 2-dimensional array containing output values
398
399	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_pf !< output array
400	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_2d_sections !< local array containing values at all slices
401	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_2d_sections_l !< local array containing values at all slices
402
403	#if defined( __parallel )
404	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: total_2d !< same as local_2d
405	#endif
406	REAL(wp), DIMENSION(:,:,:), POINTER :: to_be_resorted !< points to array which shall be output
407
408
409	IF ( debug_output_timestep ) CALL debug_message( 'data_output_2d', 'start' )
410	!
411	!-- Immediate return, if no output is requested (no respective sections
412	!-- found in parameter data_output)
413	IF ( mode == 'xy' .AND. .NOT. data_output_xy(av) ) RETURN
414	IF ( mode == 'xz' .AND. .NOT. data_output_xz(av) ) RETURN
415	IF ( mode == 'yz' .AND. .NOT. data_output_yz(av) ) RETURN
416
417	CALL cpu_log (log_point(3),'data_output_2d','start')
418
419	two_d = .FALSE. ! local variable to distinguish between output of pure 2D
420	! arrays and cross-sections of 3D arrays.
421
422	!
423	!-- Depending on the orientation of the cross-section, the respective output
424	!-- files have to be opened.
425	SELECT CASE ( mode )
426
427	CASE ( 'xy' )
428	s_ind = 1
429	ALLOCATE( level_z(nzb:nzt+1), local_2d(nxl:nxr,nys:nyn) )
430
431	IF ( netcdf_data_format > 4 ) THEN
432	ns = 1
433	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
434	ns = ns + 1
435	ENDDO
436	ns = ns - 1
437	ALLOCATE( local_2d_sections(nxl:nxr,nys:nyn,1:ns) )
438	local_2d_sections = 0.0_wp
439	ENDIF
440
441	!
442	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
443	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
444	CALL check_open( 101+av*10 )
445	ENDIF
446	IF ( data_output_2d_on_each_pe .AND. netcdf_data_format < 5 ) THEN
447	CALL check_open( 21 )
448	ELSE
449	IF ( myid == 0 ) THEN
450	#if defined( __parallel )
451	ALLOCATE( total_2d(0:nx,0:ny) )
452	#endif
453	ENDIF
454	ENDIF
455
456	CASE ( 'xz' )
457	s_ind = 2
458	ALLOCATE( local_2d(nxl:nxr,nzb:nzt+1) )
459
460	IF ( netcdf_data_format > 4 ) THEN
461	ns = 1
462	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
463	ns = ns + 1
464	ENDDO
465	ns = ns - 1
466	ALLOCATE( local_2d_sections(nxl:nxr,1:ns,nzb:nzt+1) )
467	ALLOCATE( local_2d_sections_l(nxl:nxr,1:ns,nzb:nzt+1) )
468	local_2d_sections = 0.0_wp; local_2d_sections_l = 0.0_wp
469	ENDIF
470
471	!
472	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
473	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
474	CALL check_open( 102+av*10 )
475	ENDIF
476
477	IF ( data_output_2d_on_each_pe .AND. netcdf_data_format < 5 ) THEN
478	CALL check_open( 22 )
479	ELSE
480	IF ( myid == 0 ) THEN
481	#if defined( __parallel )
482	ALLOCATE( total_2d(0:nx,nzb:nzt+1) )
483	#endif
484	ENDIF
485	ENDIF
486
487	CASE ( 'yz' )
488	s_ind = 3
489	ALLOCATE( local_2d(nys:nyn,nzb:nzt+1) )
490
491	IF ( netcdf_data_format > 4 ) THEN
492	ns = 1
493	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
494	ns = ns + 1
495	ENDDO
496	ns = ns - 1
497	ALLOCATE( local_2d_sections(1:ns,nys:nyn,nzb:nzt+1) )
498	ALLOCATE( local_2d_sections_l(1:ns,nys:nyn,nzb:nzt+1) )
499	local_2d_sections = 0.0_wp; local_2d_sections_l = 0.0_wp
500	ENDIF
501
502	!
503	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
504	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
505	CALL check_open( 103+av*10 )
506	ENDIF
507
508	IF ( data_output_2d_on_each_pe .AND. netcdf_data_format < 5 ) THEN
509	CALL check_open( 23 )
510	ELSE
511	IF ( myid == 0 ) THEN
512	#if defined( __parallel )
513	ALLOCATE( total_2d(0:ny,nzb:nzt+1) )
514	#endif
515	ENDIF
516	ENDIF
517
518	CASE DEFAULT
519	message_string = 'unknown cross-section: ' // TRIM( mode )
520	CALL message( 'data_output_2d', 'PA0180', 1, 2, 0, 6, 0 )
521
522	END SELECT
523
524	!
525	!-- For parallel netcdf output the time axis must be limited. Return, if this
526	!-- limit is exceeded. This could be the case, if the simulated time exceeds
527	!-- the given end time by the length of the given output interval.
528	IF ( netcdf_data_format > 4 ) THEN
529	IF ( mode == 'xy' .AND. do2d_xy_time_count(av) + 1 > &
530	ntdim_2d_xy(av) ) THEN
531	WRITE ( message_string, * ) 'Output of xy cross-sections is not ', &
532	'given at t=', time_since_reference_point, 's because the', &
533	' maximum number of output time levels is exceeded.'
534	CALL message( 'data_output_2d', 'PA0384', 0, 1, 0, 6, 0 )
535	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
536	RETURN
537	ENDIF
538	IF ( mode == 'xz' .AND. do2d_xz_time_count(av) + 1 > &
539	ntdim_2d_xz(av) ) THEN
540	WRITE ( message_string, * ) 'Output of xz cross-sections is not ', &
541	'given at t=', time_since_reference_point, 's because the', &
542	' maximum number of output time levels is exceeded.'
543	CALL message( 'data_output_2d', 'PA0385', 0, 1, 0, 6, 0 )
544	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
545	RETURN
546	ENDIF
547	IF ( mode == 'yz' .AND. do2d_yz_time_count(av) + 1 > &
548	ntdim_2d_yz(av) ) THEN
549	WRITE ( message_string, * ) 'Output of yz cross-sections is not ', &
550	'given at t=', time_since_reference_point, 's because the', &
551	' maximum number of output time levels is exceeded.'
552	CALL message( 'data_output_2d', 'PA0386', 0, 1, 0, 6, 0 )
553	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
554	RETURN
555	ENDIF
556	ENDIF
557
558	!
559	!-- Allocate a temporary array for resorting (kji -> ijk).
560	ALLOCATE( local_pf(nxl:nxr,nys:nyn,nzb:nzt+1) )
561	local_pf = 0.0
562
563	!
564	!-- Loop of all variables to be written.
565	!-- Output dimensions chosen
566	ivar = 1
567	l = MAX( 2, LEN_TRIM( do2d(av,ivar) ) )
568	do2d_mode = do2d(av,ivar)(l-1:l)
569
570	DO WHILE ( do2d(av,ivar)(1:1) /= ' ' )
571
572	IF ( do2d_mode == mode ) THEN
573	!
574	!-- Set flag to steer output of radiation, land-surface, or user-defined
575	!-- quantities
576	found = .FALSE.
577
578	nzb_do = nzb
579	nzt_do = nzt+1
580	!
581	!-- Before each output, set array local_pf to fill value
582	local_pf = fill_value
583	!
584	!-- Set masking flag for topography for not resorted arrays
585	flag_nr = 0
586
587	!
588	!-- Store the array chosen on the temporary array.
589	resorted = .FALSE.
590	SELECT CASE ( TRIM( do2d(av,ivar) ) )
591	CASE ( 'e_xy', 'e_xz', 'e_yz' )
592	IF ( av == 0 ) THEN
593	to_be_resorted => e
594	ELSE
595	IF ( .NOT. ALLOCATED( e_av ) ) THEN
596	ALLOCATE( e_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
597	e_av = REAL( fill_value, KIND = wp )
598	ENDIF
599	to_be_resorted => e_av
600	ENDIF
601	IF ( mode == 'xy' ) level_z = zu
602
603	CASE ( 'thetal_xy', 'thetal_xz', 'thetal_yz' )
604	IF ( av == 0 ) THEN
605	to_be_resorted => pt
606	ELSE
607	IF ( .NOT. ALLOCATED( lpt_av ) ) THEN
608	ALLOCATE( lpt_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
609	lpt_av = REAL( fill_value, KIND = wp )
610	ENDIF
611	to_be_resorted => lpt_av
612	ENDIF
613	IF ( mode == 'xy' ) level_z = zu
614
615	CASE ( 'lwp*_xy' ) ! 2d-array
616	IF ( av == 0 ) THEN
617	DO i = nxl, nxr
618	DO j = nys, nyn
619	local_pf(i,j,nzb+1) = SUM( ql(nzb:nzt,j,i) * &
620	dzw(1:nzt+1) )
621	ENDDO
622	ENDDO
623	ELSE
624	IF ( .NOT. ALLOCATED( lwp_av ) ) THEN
625	ALLOCATE( lwp_av(nysg:nyng,nxlg:nxrg) )
626	lwp_av = REAL( fill_value, KIND = wp )
627	ENDIF
628	DO i = nxl, nxr
629	DO j = nys, nyn
630	local_pf(i,j,nzb+1) = lwp_av(j,i)
631	ENDDO
632	ENDDO
633	ENDIF
634	resorted = .TRUE.
635	two_d = .TRUE.
636	level_z(nzb+1) = zu(nzb+1)
637
638	CASE ( 'ghf*_xy' ) ! 2d-array
639	IF ( av == 0 ) THEN
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%ghf(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%frac(ind_veg_wall,m) * &
649	surf_usm_h%wghf_eb(m) + &
650	surf_usm_h%frac(ind_pav_green,m) * &
651	surf_usm_h%wghf_eb_green(m) + &
652	surf_usm_h%frac(ind_wat_win,m) * &
653	surf_usm_h%wghf_eb_window(m)
654	ENDDO
655	ELSE
656	IF ( .NOT. ALLOCATED( ghf_av ) ) THEN
657	ALLOCATE( ghf_av(nysg:nyng,nxlg:nxrg) )
658	ghf_av = REAL( fill_value, KIND = wp )
659	ENDIF
660	DO i = nxl, nxr
661	DO j = nys, nyn
662	local_pf(i,j,nzb+1) = ghf_av(j,i)
663	ENDDO
664	ENDDO
665	ENDIF
666
667	resorted = .TRUE.
668	two_d = .TRUE.
669	level_z(nzb+1) = zu(nzb+1)
670
671	CASE ( 'ol*_xy' ) ! 2d-array
672	IF ( av == 0 ) THEN
673	DO m = 1, surf_def_h(0)%ns
674	i = surf_def_h(0)%i(m)
675	j = surf_def_h(0)%j(m)
676	local_pf(i,j,nzb+1) = surf_def_h(0)%ol(m)
677	ENDDO
678	DO m = 1, surf_lsm_h%ns
679	i = surf_lsm_h%i(m)
680	j = surf_lsm_h%j(m)
681	local_pf(i,j,nzb+1) = surf_lsm_h%ol(m)
682	ENDDO
683	DO m = 1, surf_usm_h%ns
684	i = surf_usm_h%i(m)
685	j = surf_usm_h%j(m)
686	local_pf(i,j,nzb+1) = surf_usm_h%ol(m)
687	ENDDO
688	ELSE
689	IF ( .NOT. ALLOCATED( ol_av ) ) THEN
690	ALLOCATE( ol_av(nysg:nyng,nxlg:nxrg) )
691	ol_av = REAL( fill_value, KIND = wp )
692	ENDIF
693	DO i = nxl, nxr
694	DO j = nys, nyn
695	local_pf(i,j,nzb+1) = ol_av(j,i)
696	ENDDO
697	ENDDO
698	ENDIF
699	resorted = .TRUE.
700	two_d = .TRUE.
701	level_z(nzb+1) = zu(nzb+1)
702
703	CASE ( 'p_xy', 'p_xz', 'p_yz' )
704	IF ( av == 0 ) THEN
705	IF ( psolver /= 'sor' ) CALL exchange_horiz( p, nbgp )
706	to_be_resorted => p
707	ELSE
708	IF ( .NOT. ALLOCATED( p_av ) ) THEN
709	ALLOCATE( p_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
710	p_av = REAL( fill_value, KIND = wp )
711	ENDIF
712	IF ( psolver /= 'sor' ) CALL exchange_horiz( p_av, nbgp )
713	to_be_resorted => p_av
714	ENDIF
715	IF ( mode == 'xy' ) level_z = zu
716
717	CASE ( 'pc_xy', 'pc_xz', 'pc_yz' ) ! particle concentration
718	IF ( av == 0 ) THEN
719	IF ( time_since_reference_point >= particle_advection_start ) THEN
720	tend = prt_count
721	! CALL exchange_horiz( tend, nbgp )
722	ELSE
723	tend = 0.0_wp
724	ENDIF
725	DO i = nxl, nxr
726	DO j = nys, nyn
727	DO k = nzb, nzt+1
728	local_pf(i,j,k) = tend(k,j,i)
729	ENDDO
730	ENDDO
731	ENDDO
732	resorted = .TRUE.
733	ELSE
734	IF ( .NOT. ALLOCATED( pc_av ) ) THEN
735	ALLOCATE( pc_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
736	pc_av = REAL( fill_value, KIND = wp )
737	ENDIF
738	! CALL exchange_horiz( pc_av, nbgp )
739	to_be_resorted => pc_av
740	ENDIF
741
742	CASE ( 'pr_xy', 'pr_xz', 'pr_yz' ) ! mean particle radius (effective radius)
743	IF ( av == 0 ) THEN
744	IF ( time_since_reference_point >= particle_advection_start ) THEN
745	DO i = nxl, nxr
746	DO j = nys, nyn
747	DO k = nzb, nzt+1
748	number_of_particles = prt_count(k,j,i)
749	IF (number_of_particles <= 0) CYCLE
750	particles => grid_particles(k,j,i)%particles(1:number_of_particles)
751	s_r2 = 0.0_wp
752	s_r3 = 0.0_wp
753	DO n = 1, number_of_particles
754	IF ( particles(n)%particle_mask ) THEN
755	s_r2 = s_r2 + particles(n)%radius*2 &
756	particles(n)%weight_factor
757	s_r3 = s_r3 + particles(n)%radius*3 &
758	particles(n)%weight_factor
759	ENDIF
760	ENDDO
761	IF ( s_r2 > 0.0_wp ) THEN
762	mean_r = s_r3 / s_r2
763	ELSE
764	mean_r = 0.0_wp
765	ENDIF
766	tend(k,j,i) = mean_r
767	ENDDO
768	ENDDO
769	ENDDO
770	! CALL exchange_horiz( tend, nbgp )
771	ELSE
772	tend = 0.0_wp
773	ENDIF
774	DO i = nxl, nxr
775	DO j = nys, nyn
776	DO k = nzb, nzt+1
777	local_pf(i,j,k) = tend(k,j,i)
778	ENDDO
779	ENDDO
780	ENDDO
781	resorted = .TRUE.
782	ELSE
783	IF ( .NOT. ALLOCATED( pr_av ) ) THEN
784	ALLOCATE( pr_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
785	pr_av = REAL( fill_value, KIND = wp )
786	ENDIF
787	! CALL exchange_horiz( pr_av, nbgp )
788	to_be_resorted => pr_av
789	ENDIF
790
791	CASE ( 'theta_xy', 'theta_xz', 'theta_yz' )
792	IF ( av == 0 ) THEN
793	IF ( .NOT. bulk_cloud_model ) THEN
794	to_be_resorted => pt
795	ELSE
796	DO i = nxl, nxr
797	DO j = nys, nyn
798	DO k = nzb, nzt+1
799	local_pf(i,j,k) = pt(k,j,i) + lv_d_cp * &
800	d_exner(k) * &
801	ql(k,j,i)
802	ENDDO
803	ENDDO
804	ENDDO
805	resorted = .TRUE.
806	ENDIF
807	ELSE
808	IF ( .NOT. ALLOCATED( pt_av ) ) THEN
809	ALLOCATE( pt_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
810	pt_av = REAL( fill_value, KIND = wp )
811	ENDIF
812	to_be_resorted => pt_av
813	ENDIF
814	IF ( mode == 'xy' ) level_z = zu
815
816	CASE ( 'q_xy', 'q_xz', 'q_yz' )
817	IF ( av == 0 ) THEN
818	to_be_resorted => q
819	ELSE
820	IF ( .NOT. ALLOCATED( q_av ) ) THEN
821	ALLOCATE( q_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
822	q_av = REAL( fill_value, KIND = wp )
823	ENDIF
824	to_be_resorted => q_av
825	ENDIF
826	IF ( mode == 'xy' ) level_z = zu
827
828	CASE ( 'ql_xy', 'ql_xz', 'ql_yz' )
829	IF ( av == 0 ) THEN
830	to_be_resorted => ql
831	ELSE
832	IF ( .NOT. ALLOCATED( ql_av ) ) THEN
833	ALLOCATE( ql_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
834	ql_av = REAL( fill_value, KIND = wp )
835	ENDIF
836	to_be_resorted => ql_av
837	ENDIF
838	IF ( mode == 'xy' ) level_z = zu
839
840	CASE ( 'ql_c_xy', 'ql_c_xz', 'ql_c_yz' )
841	IF ( av == 0 ) THEN
842	to_be_resorted => ql_c
843	ELSE
844	IF ( .NOT. ALLOCATED( ql_c_av ) ) THEN
845	ALLOCATE( ql_c_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
846	ql_c_av = REAL( fill_value, KIND = wp )
847	ENDIF
848	to_be_resorted => ql_c_av
849	ENDIF
850	IF ( mode == 'xy' ) level_z = zu
851
852	CASE ( 'ql_v_xy', 'ql_v_xz', 'ql_v_yz' )
853	IF ( av == 0 ) THEN
854	to_be_resorted => ql_v
855	ELSE
856	IF ( .NOT. ALLOCATED( ql_v_av ) ) THEN
857	ALLOCATE( ql_v_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
858	ql_v_av = REAL( fill_value, KIND = wp )
859	ENDIF
860	to_be_resorted => ql_v_av
861	ENDIF
862	IF ( mode == 'xy' ) level_z = zu
863
864	CASE ( 'ql_vp_xy', 'ql_vp_xz', 'ql_vp_yz' )
865	IF ( av == 0 ) THEN
866	IF ( time_since_reference_point >= particle_advection_start ) THEN
867	DO i = nxl, nxr
868	DO j = nys, nyn
869	DO k = nzb, nzt+1
870	number_of_particles = prt_count(k,j,i)
871	IF (number_of_particles <= 0) CYCLE
872	particles => grid_particles(k,j,i)%particles(1:number_of_particles)
873	DO n = 1, number_of_particles
874	IF ( particles(n)%particle_mask ) THEN
875	tend(k,j,i) = tend(k,j,i) + &
876	particles(n)%weight_factor / &
877	prt_count(k,j,i)
878	ENDIF
879	ENDDO
880	ENDDO
881	ENDDO
882	ENDDO
883	! CALL exchange_horiz( tend, nbgp )
884	ELSE
885	tend = 0.0_wp
886	ENDIF
887	DO i = nxl, nxr
888	DO j = nys, nyn
889	DO k = nzb, nzt+1
890	local_pf(i,j,k) = tend(k,j,i)
891	ENDDO
892	ENDDO
893	ENDDO
894	resorted = .TRUE.
895	ELSE
896	IF ( .NOT. ALLOCATED( ql_vp_av ) ) THEN
897	ALLOCATE( ql_vp_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
898	ql_vp_av = REAL( fill_value, KIND = wp )
899	ENDIF
900	! CALL exchange_horiz( ql_vp_av, nbgp )
901	to_be_resorted => ql_vp_av
902	ENDIF
903	IF ( mode == 'xy' ) level_z = zu
904
905	CASE ( 'qsws*_xy' ) ! 2d-array
906	IF ( av == 0 ) THEN
907	local_pf(:,:,nzb+1) = REAL( fill_value, KIND = wp )
908	!
909	!-- In case of default surfaces, clean-up flux by density.
910	!-- In case of land-surfaces, convert fluxes into
911	!-- dynamic units
912	DO m = 1, surf_def_h(0)%ns
913	i = surf_def_h(0)%i(m)
914	j = surf_def_h(0)%j(m)
915	k = surf_def_h(0)%k(m)
916	local_pf(i,j,nzb+1) = surf_def_h(0)%qsws(m) * &
917	waterflux_output_conversion(k)
918	ENDDO
919	DO m = 1, surf_lsm_h%ns
920	i = surf_lsm_h%i(m)
921	j = surf_lsm_h%j(m)
922	k = surf_lsm_h%k(m)
923	local_pf(i,j,nzb+1) = surf_lsm_h%qsws(m) * l_v
924	ENDDO
925	DO m = 1, surf_usm_h%ns
926	i = surf_usm_h%i(m)
927	j = surf_usm_h%j(m)
928	k = surf_usm_h%k(m)
929	local_pf(i,j,nzb+1) = surf_usm_h%qsws(m) * l_v
930	ENDDO
931	ELSE
932	IF ( .NOT. ALLOCATED( qsws_av ) ) THEN
933	ALLOCATE( qsws_av(nysg:nyng,nxlg:nxrg) )
934	qsws_av = REAL( fill_value, KIND = wp )
935	ENDIF
936	DO i = nxl, nxr
937	DO j = nys, nyn
938	local_pf(i,j,nzb+1) = qsws_av(j,i)
939	ENDDO
940	ENDDO
941	ENDIF
942	resorted = .TRUE.
943	two_d = .TRUE.
944	level_z(nzb+1) = zu(nzb+1)
945
946	CASE ( 'qv_xy', 'qv_xz', 'qv_yz' )
947	IF ( av == 0 ) THEN
948	DO i = nxl, nxr
949	DO j = nys, nyn
950	DO k = nzb, nzt+1
951	local_pf(i,j,k) = q(k,j,i) - ql(k,j,i)
952	ENDDO
953	ENDDO
954	ENDDO
955	resorted = .TRUE.
956	ELSE
957	IF ( .NOT. ALLOCATED( qv_av ) ) THEN
958	ALLOCATE( qv_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
959	qv_av = REAL( fill_value, KIND = wp )
960	ENDIF
961	to_be_resorted => qv_av
962	ENDIF
963	IF ( mode == 'xy' ) level_z = zu
964
965	CASE ( 'r_a*_xy' ) ! 2d-array
966	IF ( av == 0 ) THEN
967	DO m = 1, surf_lsm_h%ns
968	i = surf_lsm_h%i(m)
969	j = surf_lsm_h%j(m)
970	local_pf(i,j,nzb+1) = surf_lsm_h%r_a(m)
971	ENDDO
972
973	DO m = 1, surf_usm_h%ns
974	i = surf_usm_h%i(m)
975	j = surf_usm_h%j(m)
976	local_pf(i,j,nzb+1) = &
977	( surf_usm_h%frac(ind_veg_wall,m) * &
978	surf_usm_h%r_a(m) + &
979	surf_usm_h%frac(ind_pav_green,m) * &
980	surf_usm_h%r_a_green(m) + &
981	surf_usm_h%frac(ind_wat_win,m) * &
982	surf_usm_h%r_a_window(m) )
983	ENDDO
984	ELSE
985	IF ( .NOT. ALLOCATED( r_a_av ) ) THEN
986	ALLOCATE( r_a_av(nysg:nyng,nxlg:nxrg) )
987	r_a_av = REAL( fill_value, KIND = wp )
988	ENDIF
989	DO i = nxl, nxr
990	DO j = nys, nyn
991	local_pf(i,j,nzb+1) = r_a_av(j,i)
992	ENDDO
993	ENDDO
994	ENDIF
995	resorted = .TRUE.
996	two_d = .TRUE.
997	level_z(nzb+1) = zu(nzb+1)
998
999	CASE ( 's_xy', 's_xz', 's_yz' )
1000	IF ( av == 0 ) THEN
1001	to_be_resorted => s
1002	ELSE
1003	IF ( .NOT. ALLOCATED( s_av ) ) THEN
1004	ALLOCATE( s_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1005	s_av = REAL( fill_value, KIND = wp )
1006	ENDIF
1007	to_be_resorted => s_av
1008	ENDIF
1009
1010	CASE ( 'shf*_xy' ) ! 2d-array
1011	IF ( av == 0 ) THEN
1012	!
1013	!-- In case of default surfaces, clean-up flux by density.
1014	!-- In case of land- and urban-surfaces, convert fluxes into
1015	!-- dynamic units.
1016	DO m = 1, surf_def_h(0)%ns
1017	i = surf_def_h(0)%i(m)
1018	j = surf_def_h(0)%j(m)
1019	k = surf_def_h(0)%k(m)
1020	local_pf(i,j,nzb+1) = surf_def_h(0)%shf(m) * &
1021	heatflux_output_conversion(k)
1022	ENDDO
1023	DO m = 1, surf_lsm_h%ns
1024	i = surf_lsm_h%i(m)
1025	j = surf_lsm_h%j(m)
1026	k = surf_lsm_h%k(m)
1027	local_pf(i,j,nzb+1) = surf_lsm_h%shf(m) * c_p
1028	ENDDO
1029	DO m = 1, surf_usm_h%ns
1030	i = surf_usm_h%i(m)
1031	j = surf_usm_h%j(m)
1032	k = surf_usm_h%k(m)
1033	local_pf(i,j,nzb+1) = surf_usm_h%shf(m) * c_p
1034	ENDDO
1035	ELSE
1036	IF ( .NOT. ALLOCATED( shf_av ) ) THEN
1037	ALLOCATE( shf_av(nysg:nyng,nxlg:nxrg) )
1038	shf_av = REAL( fill_value, KIND = wp )
1039	ENDIF
1040	DO i = nxl, nxr
1041	DO j = nys, nyn
1042	local_pf(i,j,nzb+1) = shf_av(j,i)
1043	ENDDO
1044	ENDDO
1045	ENDIF
1046	resorted = .TRUE.
1047	two_d = .TRUE.
1048	level_z(nzb+1) = zu(nzb+1)
1049
1050	CASE ( 'ssws*_xy' ) ! 2d-array
1051	IF ( av == 0 ) THEN
1052	DO m = 1, surf_def_h(0)%ns
1053	i = surf_def_h(0)%i(m)
1054	j = surf_def_h(0)%j(m)
1055	local_pf(i,j,nzb+1) = surf_def_h(0)%ssws(m)
1056	ENDDO
1057	DO m = 1, surf_lsm_h%ns
1058	i = surf_lsm_h%i(m)
1059	j = surf_lsm_h%j(m)
1060	local_pf(i,j,nzb+1) = surf_lsm_h%ssws(m)
1061	ENDDO
1062	DO m = 1, surf_usm_h%ns
1063	i = surf_usm_h%i(m)
1064	j = surf_usm_h%j(m)
1065	local_pf(i,j,nzb+1) = surf_usm_h%ssws(m)
1066	ENDDO
1067	ELSE
1068	IF ( .NOT. ALLOCATED( ssws_av ) ) THEN
1069	ALLOCATE( ssws_av(nysg:nyng,nxlg:nxrg) )
1070	ssws_av = REAL( fill_value, KIND = wp )
1071	ENDIF
1072	DO i = nxl, nxr
1073	DO j = nys, nyn
1074	local_pf(i,j,nzb+1) = ssws_av(j,i)
1075	ENDDO
1076	ENDDO
1077	ENDIF
1078	resorted = .TRUE.
1079	two_d = .TRUE.
1080	level_z(nzb+1) = zu(nzb+1)
1081
1082	CASE ( 't*_xy' ) ! 2d-array
1083	IF ( av == 0 ) THEN
1084	DO m = 1, surf_def_h(0)%ns
1085	i = surf_def_h(0)%i(m)
1086	j = surf_def_h(0)%j(m)
1087	local_pf(i,j,nzb+1) = surf_def_h(0)%ts(m)
1088	ENDDO
1089	DO m = 1, surf_lsm_h%ns
1090	i = surf_lsm_h%i(m)
1091	j = surf_lsm_h%j(m)
1092	local_pf(i,j,nzb+1) = surf_lsm_h%ts(m)
1093	ENDDO
1094	DO m = 1, surf_usm_h%ns
1095	i = surf_usm_h%i(m)
1096	j = surf_usm_h%j(m)
1097	local_pf(i,j,nzb+1) = surf_usm_h%ts(m)
1098	ENDDO
1099	ELSE
1100	IF ( .NOT. ALLOCATED( ts_av ) ) THEN
1101	ALLOCATE( ts_av(nysg:nyng,nxlg:nxrg) )
1102	ts_av = REAL( fill_value, KIND = wp )
1103	ENDIF
1104	DO i = nxl, nxr
1105	DO j = nys, nyn
1106	local_pf(i,j,nzb+1) = ts_av(j,i)
1107	ENDDO
1108	ENDDO
1109	ENDIF
1110	resorted = .TRUE.
1111	two_d = .TRUE.
1112	level_z(nzb+1) = zu(nzb+1)
1113
1114	CASE ( 'tsurf*_xy' ) ! 2d-array
1115	IF ( av == 0 ) THEN
1116	DO m = 1, surf_def_h(0)%ns
1117	i = surf_def_h(0)%i(m)
1118	j = surf_def_h(0)%j(m)
1119	local_pf(i,j,nzb+1) = surf_def_h(0)%pt_surface(m)
1120	ENDDO
1121
1122	DO m = 1, surf_lsm_h%ns
1123	i = surf_lsm_h%i(m)
1124	j = surf_lsm_h%j(m)
1125	local_pf(i,j,nzb+1) = surf_lsm_h%pt_surface(m)
1126	ENDDO
1127
1128	DO m = 1, surf_usm_h%ns
1129	i = surf_usm_h%i(m)
1130	j = surf_usm_h%j(m)
1131	local_pf(i,j,nzb+1) = surf_usm_h%pt_surface(m)
1132	ENDDO
1133
1134	ELSE
1135	IF ( .NOT. ALLOCATED( tsurf_av ) ) THEN
1136	ALLOCATE( tsurf_av(nysg:nyng,nxlg:nxrg) )
1137	tsurf_av = REAL( fill_value, KIND = wp )
1138	ENDIF
1139	DO i = nxl, nxr
1140	DO j = nys, nyn
1141	local_pf(i,j,nzb+1) = tsurf_av(j,i)
1142	ENDDO
1143	ENDDO
1144	ENDIF
1145	resorted = .TRUE.
1146	two_d = .TRUE.
1147	level_z(nzb+1) = zu(nzb+1)
1148
1149	CASE ( 'u_xy', 'u_xz', 'u_yz' )
1150	flag_nr = 1
1151	IF ( av == 0 ) THEN
1152	to_be_resorted => u
1153	ELSE
1154	IF ( .NOT. ALLOCATED( u_av ) ) THEN
1155	ALLOCATE( u_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1156	u_av = REAL( fill_value, KIND = wp )
1157	ENDIF
1158	to_be_resorted => u_av
1159	ENDIF
1160	IF ( mode == 'xy' ) level_z = zu
1161	!
1162	!-- Substitute the values generated by "mirror" boundary condition
1163	!-- at the bottom boundary by the real surface values.
1164	IF ( do2d(av,ivar) == 'u_xz' .OR. do2d(av,ivar) == 'u_yz' ) THEN
1165	IF ( ibc_uv_b == 0 ) local_pf(:,:,nzb) = 0.0_wp
1166	ENDIF
1167
1168	CASE ( 'us*_xy' ) ! 2d-array
1169	IF ( av == 0 ) THEN
1170	DO m = 1, surf_def_h(0)%ns
1171	i = surf_def_h(0)%i(m)
1172	j = surf_def_h(0)%j(m)
1173	local_pf(i,j,nzb+1) = surf_def_h(0)%us(m)
1174	ENDDO
1175	DO m = 1, surf_lsm_h%ns
1176	i = surf_lsm_h%i(m)
1177	j = surf_lsm_h%j(m)
1178	local_pf(i,j,nzb+1) = surf_lsm_h%us(m)
1179	ENDDO
1180	DO m = 1, surf_usm_h%ns
1181	i = surf_usm_h%i(m)
1182	j = surf_usm_h%j(m)
1183	local_pf(i,j,nzb+1) = surf_usm_h%us(m)
1184	ENDDO
1185	ELSE
1186	IF ( .NOT. ALLOCATED( us_av ) ) THEN
1187	ALLOCATE( us_av(nysg:nyng,nxlg:nxrg) )
1188	us_av = REAL( fill_value, KIND = wp )
1189	ENDIF
1190	DO i = nxl, nxr
1191	DO j = nys, nyn
1192	local_pf(i,j,nzb+1) = us_av(j,i)
1193	ENDDO
1194	ENDDO
1195	ENDIF
1196	resorted = .TRUE.
1197	two_d = .TRUE.
1198	level_z(nzb+1) = zu(nzb+1)
1199
1200	CASE ( 'v_xy', 'v_xz', 'v_yz' )
1201	flag_nr = 2
1202	IF ( av == 0 ) THEN
1203	to_be_resorted => v
1204	ELSE
1205	IF ( .NOT. ALLOCATED( v_av ) ) THEN
1206	ALLOCATE( v_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1207	v_av = REAL( fill_value, KIND = wp )
1208	ENDIF
1209	to_be_resorted => v_av
1210	ENDIF
1211	IF ( mode == 'xy' ) level_z = zu
1212	!
1213	!-- Substitute the values generated by "mirror" boundary condition
1214	!-- at the bottom boundary by the real surface values.
1215	IF ( do2d(av,ivar) == 'v_xz' .OR. do2d(av,ivar) == 'v_yz' ) THEN
1216	IF ( ibc_uv_b == 0 ) local_pf(:,:,nzb) = 0.0_wp
1217	ENDIF
1218
1219	CASE ( 'thetav_xy', 'thetav_xz', 'thetav_yz' )
1220	IF ( av == 0 ) THEN
1221	to_be_resorted => vpt
1222	ELSE
1223	IF ( .NOT. ALLOCATED( vpt_av ) ) THEN
1224	ALLOCATE( vpt_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1225	vpt_av = REAL( fill_value, KIND = wp )
1226	ENDIF
1227	to_be_resorted => vpt_av
1228	ENDIF
1229	IF ( mode == 'xy' ) level_z = zu
1230
1231	CASE ( 'theta_2m*_xy' ) ! 2d-array
1232	IF ( av == 0 ) THEN
1233	DO m = 1, surf_def_h(0)%ns
1234	i = surf_def_h(0)%i(m)
1235	j = surf_def_h(0)%j(m)
1236	local_pf(i,j,nzb+1) = surf_def_h(0)%pt_2m(m)
1237	ENDDO
1238	DO m = 1, surf_lsm_h%ns
1239	i = surf_lsm_h%i(m)
1240	j = surf_lsm_h%j(m)
1241	local_pf(i,j,nzb+1) = surf_lsm_h%pt_2m(m)
1242	ENDDO
1243	DO m = 1, surf_usm_h%ns
1244	i = surf_usm_h%i(m)
1245	j = surf_usm_h%j(m)
1246	local_pf(i,j,nzb+1) = surf_usm_h%pt_2m(m)
1247	ENDDO
1248	ELSE
1249	IF ( .NOT. ALLOCATED( pt_2m_av ) ) THEN
1250	ALLOCATE( pt_2m_av(nysg:nyng,nxlg:nxrg) )
1251	pt_2m_av = REAL( fill_value, KIND = wp )
1252	ENDIF
1253	DO i = nxl, nxr
1254	DO j = nys, nyn
1255	local_pf(i,j,nzb+1) = pt_2m_av(j,i)
1256	ENDDO
1257	ENDDO
1258	ENDIF
1259	resorted = .TRUE.
1260	two_d = .TRUE.
1261	level_z(nzb+1) = zu(nzb+1)
1262
1263
1264
1265	CASE ( 'w_xy', 'w_xz', 'w_yz' )
1266	flag_nr = 3
1267	IF ( av == 0 ) THEN
1268	to_be_resorted => w
1269	ELSE
1270	IF ( .NOT. ALLOCATED( w_av ) ) THEN
1271	ALLOCATE( w_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1272	w_av = REAL( fill_value, KIND = wp )
1273	ENDIF
1274	to_be_resorted => w_av
1275	ENDIF
1276	IF ( mode == 'xy' ) level_z = zw
1277
1278	CASE ( 'z0*_xy' ) ! 2d-array
1279	IF ( av == 0 ) THEN
1280	DO m = 1, surf_def_h(0)%ns
1281	i = surf_def_h(0)%i(m)
1282	j = surf_def_h(0)%j(m)
1283	local_pf(i,j,nzb+1) = surf_def_h(0)%z0(m)
1284	ENDDO
1285	DO m = 1, surf_lsm_h%ns
1286	i = surf_lsm_h%i(m)
1287	j = surf_lsm_h%j(m)
1288	local_pf(i,j,nzb+1) = surf_lsm_h%z0(m)
1289	ENDDO
1290	DO m = 1, surf_usm_h%ns
1291	i = surf_usm_h%i(m)
1292	j = surf_usm_h%j(m)
1293	local_pf(i,j,nzb+1) = surf_usm_h%z0(m)
1294	ENDDO
1295	ELSE
1296	IF ( .NOT. ALLOCATED( z0_av ) ) THEN
1297	ALLOCATE( z0_av(nysg:nyng,nxlg:nxrg) )
1298	z0_av = REAL( fill_value, KIND = wp )
1299	ENDIF
1300	DO i = nxl, nxr
1301	DO j = nys, nyn
1302	local_pf(i,j,nzb+1) = z0_av(j,i)
1303	ENDDO
1304	ENDDO
1305	ENDIF
1306	resorted = .TRUE.
1307	two_d = .TRUE.
1308	level_z(nzb+1) = zu(nzb+1)
1309
1310	CASE ( 'z0h*_xy' ) ! 2d-array
1311	IF ( av == 0 ) THEN
1312	DO m = 1, surf_def_h(0)%ns
1313	i = surf_def_h(0)%i(m)
1314	j = surf_def_h(0)%j(m)
1315	local_pf(i,j,nzb+1) = surf_def_h(0)%z0h(m)
1316	ENDDO
1317	DO m = 1, surf_lsm_h%ns
1318	i = surf_lsm_h%i(m)
1319	j = surf_lsm_h%j(m)
1320	local_pf(i,j,nzb+1) = surf_lsm_h%z0h(m)
1321	ENDDO
1322	DO m = 1, surf_usm_h%ns
1323	i = surf_usm_h%i(m)
1324	j = surf_usm_h%j(m)
1325	local_pf(i,j,nzb+1) = surf_usm_h%z0h(m)
1326	ENDDO
1327	ELSE
1328	IF ( .NOT. ALLOCATED( z0h_av ) ) THEN
1329	ALLOCATE( z0h_av(nysg:nyng,nxlg:nxrg) )
1330	z0h_av = REAL( fill_value, KIND = wp )
1331	ENDIF
1332	DO i = nxl, nxr
1333	DO j = nys, nyn
1334	local_pf(i,j,nzb+1) = z0h_av(j,i)
1335	ENDDO
1336	ENDDO
1337	ENDIF
1338	resorted = .TRUE.
1339	two_d = .TRUE.
1340	level_z(nzb+1) = zu(nzb+1)
1341
1342	CASE ( 'z0q*_xy' ) ! 2d-array
1343	IF ( av == 0 ) THEN
1344	DO m = 1, surf_def_h(0)%ns
1345	i = surf_def_h(0)%i(m)
1346	j = surf_def_h(0)%j(m)
1347	local_pf(i,j,nzb+1) = surf_def_h(0)%z0q(m)
1348	ENDDO
1349	DO m = 1, surf_lsm_h%ns
1350	i = surf_lsm_h%i(m)
1351	j = surf_lsm_h%j(m)
1352	local_pf(i,j,nzb+1) = surf_lsm_h%z0q(m)
1353	ENDDO
1354	DO m = 1, surf_usm_h%ns
1355	i = surf_usm_h%i(m)
1356	j = surf_usm_h%j(m)
1357	local_pf(i,j,nzb+1) = surf_usm_h%z0q(m)
1358	ENDDO
1359	ELSE
1360	IF ( .NOT. ALLOCATED( z0q_av ) ) THEN
1361	ALLOCATE( z0q_av(nysg:nyng,nxlg:nxrg) )
1362	z0q_av = REAL( fill_value, KIND = wp )
1363	ENDIF
1364	DO i = nxl, nxr
1365	DO j = nys, nyn
1366	local_pf(i,j,nzb+1) = z0q_av(j,i)
1367	ENDDO
1368	ENDDO
1369	ENDIF
1370	resorted = .TRUE.
1371	two_d = .TRUE.
1372	level_z(nzb+1) = zu(nzb+1)
1373
1374	CASE DEFAULT
1375
1376	!
1377	!-- Quantities of other modules
1378	IF ( .NOT. found ) THEN
1379	CALL module_interface_data_output_2d( &
1380	av, do2d(av,ivar), found, grid, mode, &
1381	local_pf, two_d, nzb_do, nzt_do, &
1382	fill_value &
1383	)
1384	ENDIF
1385
1386	resorted = .TRUE.
1387
1388	IF ( grid == 'zu' ) THEN
1389	IF ( mode == 'xy' ) level_z = zu
1390	ELSEIF ( grid == 'zw' ) THEN
1391	IF ( mode == 'xy' ) level_z = zw
1392	ELSEIF ( grid == 'zu1' ) THEN
1393	IF ( mode == 'xy' ) level_z(nzb+1) = zu(nzb+1)
1394	ELSEIF ( grid == 'zs' ) THEN
1395	IF ( mode == 'xy' ) level_z = zs
1396	ENDIF
1397
1398	IF ( .NOT. found ) THEN
1399	message_string = 'no output provided for: ' // &
1400	TRIM( do2d(av,ivar) )
1401	CALL message( 'data_output_2d', 'PA0181', 0, 0, 0, 6, 0 )
1402	ENDIF
1403
1404	END SELECT
1405
1406	!
1407	!-- Resort the array to be output, if not done above. Flag topography
1408	!-- grid points with fill values, using the corresponding maksing flag.
1409	IF ( .NOT. resorted ) THEN
1410	DO i = nxl, nxr
1411	DO j = nys, nyn
1412	DO k = nzb_do, nzt_do
1413	local_pf(i,j,k) = MERGE( to_be_resorted(k,j,i), &
1414	REAL( fill_value, KIND = wp ), &
1415	BTEST( wall_flags_0(k,j,i), &
1416	flag_nr ) )
1417	ENDDO
1418	ENDDO
1419	ENDDO
1420	ENDIF
1421
1422	!
1423	!-- Output of the individual cross-sections, depending on the cross-
1424	!-- section mode chosen.
1425	is = 1
1426	loop1: DO WHILE ( section(is,s_ind) /= -9999 .OR. two_d )
1427
1428	SELECT CASE ( mode )
1429
1430	CASE ( 'xy' )
1431	!
1432	!-- Determine the cross section index
1433	IF ( two_d ) THEN
1434	layer_xy = nzb+1
1435	ELSE
1436	layer_xy = section(is,s_ind)
1437	ENDIF
1438
1439	!
1440	!-- Exit the loop for layers beyond the data output domain
1441	!-- (used for soil model)
1442	IF ( layer_xy > nzt_do ) THEN
1443	EXIT loop1
1444	ENDIF
1445
1446	!
1447	!-- Update the netCDF xy cross section time axis.
1448	!-- In case of parallel output, this is only done by PE0
1449	!-- to increase the performance.
1450	IF ( time_since_reference_point /= do2d_xy_last_time(av) ) THEN
1451	do2d_xy_time_count(av) = do2d_xy_time_count(av) + 1
1452	do2d_xy_last_time(av) = time_since_reference_point
1453	IF ( myid == 0 ) THEN
1454	IF ( .NOT. data_output_2d_on_each_pe &
1455	.OR. netcdf_data_format > 4 ) &
1456	THEN
1457	#if defined( __netcdf )
1458	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1459	id_var_time_xy(av), &
1460	(/ time_since_reference_point /), &
1461	start = (/ do2d_xy_time_count(av) /), &
1462	count = (/ 1 /) )
1463	CALL netcdf_handle_error( 'data_output_2d', 53 )
1464	#endif
1465	ENDIF
1466	ENDIF
1467	ENDIF
1468	!
1469	!-- If required, carry out averaging along z
1470	IF ( section(is,s_ind) == -1 .AND. .NOT. two_d ) THEN
1471
1472	local_2d = 0.0_wp
1473	!
1474	!-- Carry out the averaging (all data are on the PE)
1475	DO k = nzb_do, nzt_do
1476	DO j = nys, nyn
1477	DO i = nxl, nxr
1478	local_2d(i,j) = local_2d(i,j) + local_pf(i,j,k)
1479	ENDDO
1480	ENDDO
1481	ENDDO
1482
1483	local_2d = local_2d / ( nzt_do - nzb_do + 1.0_wp)
1484
1485	ELSE
1486	!
1487	!-- Just store the respective section on the local array
1488	local_2d = local_pf(:,:,layer_xy)
1489
1490	ENDIF
1491
1492	#if defined( __parallel )
1493	IF ( netcdf_data_format > 4 ) THEN
1494	!
1495	!-- Parallel output in netCDF4/HDF5 format.
1496	IF ( two_d ) THEN
1497	iis = 1
1498	ELSE
1499	iis = is
1500	ENDIF
1501
1502	#if defined( __netcdf )
1503	!
1504	!-- For parallel output, all cross sections are first stored
1505	!-- here on a local array and will be written to the output
1506	!-- file afterwards to increase the performance.
1507	DO i = nxl, nxr
1508	DO j = nys, nyn
1509	local_2d_sections(i,j,iis) = local_2d(i,j)
1510	ENDDO
1511	ENDDO
1512	#endif
1513	ELSE
1514
1515	IF ( data_output_2d_on_each_pe ) THEN
1516	!
1517	!-- Output of partial arrays on each PE
1518	#if defined( __netcdf )
1519	IF ( myid == 0 ) THEN
1520	WRITE ( 21 ) time_since_reference_point, &
1521	do2d_xy_time_count(av), av
1522	ENDIF
1523	#endif
1524	DO i = 0, io_blocks-1
1525	IF ( i == io_group ) THEN
1526	WRITE ( 21 ) nxl, nxr, nys, nyn, nys, nyn
1527	WRITE ( 21 ) local_2d
1528	ENDIF
1529	#if defined( __parallel )
1530	CALL MPI_BARRIER( comm2d, ierr )
1531	#endif
1532	ENDDO
1533
1534	ELSE
1535	!
1536	!-- PE0 receives partial arrays from all processors and
1537	!-- then outputs them. Here a barrier has to be set,
1538	!-- because otherwise "-MPI- FATAL: Remote protocol queue
1539	!-- full" may occur.
1540	CALL MPI_BARRIER( comm2d, ierr )
1541
1542	ngp = ( nxr-nxl+1 ) * ( nyn-nys+1 )
1543	IF ( myid == 0 ) THEN
1544	!
1545	!-- Local array can be relocated directly.
1546	total_2d(nxl:nxr,nys:nyn) = local_2d
1547	!
1548	!-- Receive data from all other PEs.
1549	DO n = 1, numprocs-1
1550	!
1551	!-- Receive index limits first, then array.
1552	!-- Index limits are received in arbitrary order from
1553	!-- the PEs.
1554	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
1555	MPI_ANY_SOURCE, 0, comm2d, &
1556	status, ierr )
1557	sender = status(MPI_SOURCE)
1558	DEALLOCATE( local_2d )
1559	ALLOCATE( local_2d(ind(1):ind(2),ind(3):ind(4)) )
1560	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
1561	MPI_REAL, sender, 1, comm2d, &
1562	status, ierr )
1563	total_2d(ind(1):ind(2),ind(3):ind(4)) = local_2d
1564	ENDDO
1565	!
1566	!-- Relocate the local array for the next loop increment
1567	DEALLOCATE( local_2d )
1568	ALLOCATE( local_2d(nxl:nxr,nys:nyn) )
1569
1570	#if defined( __netcdf )
1571	IF ( two_d ) THEN
1572	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1573	id_var_do2d(av,ivar), &
1574	total_2d(0:nx,0:ny), &
1575	start = (/ 1, 1, 1, do2d_xy_time_count(av) /), &
1576	count = (/ nx+1, ny+1, 1, 1 /) )
1577	ELSE
1578	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1579	id_var_do2d(av,ivar), &
1580	total_2d(0:nx,0:ny), &
1581	start = (/ 1, 1, is, do2d_xy_time_count(av) /), &
1582	count = (/ nx+1, ny+1, 1, 1 /) )
1583	ENDIF
1584	CALL netcdf_handle_error( 'data_output_2d', 54 )
1585	#endif
1586
1587	ELSE
1588	!
1589	!-- First send the local index limits to PE0
1590	ind(1) = nxl; ind(2) = nxr
1591	ind(3) = nys; ind(4) = nyn
1592	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
1593	comm2d, ierr )
1594	!
1595	!-- Send data to PE0
1596	CALL MPI_SEND( local_2d(nxl,nys), ngp, &
1597	MPI_REAL, 0, 1, comm2d, ierr )
1598	ENDIF
1599	!
1600	!-- A barrier has to be set, because otherwise some PEs may
1601	!-- proceed too fast so that PE0 may receive wrong data on
1602	!-- tag 0
1603	CALL MPI_BARRIER( comm2d, ierr )
1604	ENDIF
1605
1606	ENDIF
1607	#else
1608	#if defined( __netcdf )
1609	IF ( two_d ) THEN
1610	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1611	id_var_do2d(av,ivar), &
1612	local_2d(nxl:nxr,nys:nyn), &
1613	start = (/ 1, 1, 1, do2d_xy_time_count(av) /), &
1614	count = (/ nx+1, ny+1, 1, 1 /) )
1615	ELSE
1616	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1617	id_var_do2d(av,ivar), &
1618	local_2d(nxl:nxr,nys:nyn), &
1619	start = (/ 1, 1, is, do2d_xy_time_count(av) /), &
1620	count = (/ nx+1, ny+1, 1, 1 /) )
1621	ENDIF
1622	CALL netcdf_handle_error( 'data_output_2d', 447 )
1623	#endif
1624	#endif
1625
1626	!
1627	!-- For 2D-arrays (e.g. u*) only one cross-section is available.
1628	!-- Hence exit loop of output levels.
1629	IF ( two_d ) THEN
1630	IF ( netcdf_data_format < 5 ) two_d = .FALSE.
1631	EXIT loop1
1632	ENDIF
1633
1634	CASE ( 'xz' )
1635	!
1636	!-- Update the netCDF xz cross section time axis.
1637	!-- In case of parallel output, this is only done by PE0
1638	!-- to increase the performance.
1639	IF ( time_since_reference_point /= do2d_xz_last_time(av) ) THEN
1640	do2d_xz_time_count(av) = do2d_xz_time_count(av) + 1
1641	do2d_xz_last_time(av) = time_since_reference_point
1642	IF ( myid == 0 ) THEN
1643	IF ( .NOT. data_output_2d_on_each_pe &
1644	.OR. netcdf_data_format > 4 ) &
1645	THEN
1646	#if defined( __netcdf )
1647	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
1648	id_var_time_xz(av), &
1649	(/ time_since_reference_point /), &
1650	start = (/ do2d_xz_time_count(av) /), &
1651	count = (/ 1 /) )
1652	CALL netcdf_handle_error( 'data_output_2d', 56 )
1653	#endif
1654	ENDIF
1655	ENDIF
1656	ENDIF
1657
1658	!
1659	!-- If required, carry out averaging along y
1660	IF ( section(is,s_ind) == -1 ) THEN
1661
1662	ALLOCATE( local_2d_l(nxl:nxr,nzb_do:nzt_do) )
1663	local_2d_l = 0.0_wp
1664	ngp = ( nxr-nxl + 1 ) * ( nzt_do-nzb_do + 1 )
1665	!
1666	!-- First local averaging on the PE
1667	DO k = nzb_do, nzt_do
1668	DO j = nys, nyn
1669	DO i = nxl, nxr
1670	local_2d_l(i,k) = local_2d_l(i,k) + &
1671	local_pf(i,j,k)
1672	ENDDO
1673	ENDDO
1674	ENDDO
1675	#if defined( __parallel )
1676	!
1677	!-- Now do the averaging over all PEs along y
1678	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
1679	CALL MPI_ALLREDUCE( local_2d_l(nxl,nzb_do), &
1680	local_2d(nxl,nzb_do), ngp, MPI_REAL, &
1681	MPI_SUM, comm1dy, ierr )
1682	#else
1683	local_2d = local_2d_l
1684	#endif
1685	local_2d = local_2d / ( ny + 1.0_wp )
1686
1687	DEALLOCATE( local_2d_l )
1688
1689	ELSE
1690	!
1691	!-- Just store the respective section on the local array
1692	!-- (but only if it is available on this PE!)
1693	IF ( section(is,s_ind) >= nys .AND. section(is,s_ind) <= nyn ) &
1694	THEN
1695	local_2d = local_pf(:,section(is,s_ind),nzb_do:nzt_do)
1696	ENDIF
1697
1698	ENDIF
1699
1700	#if defined( __parallel )
1701	IF ( netcdf_data_format > 4 ) THEN
1702	!
1703	!-- Output in netCDF4/HDF5 format.
1704	!-- Output only on those PEs where the respective cross
1705	!-- sections reside. Cross sections averaged along y are
1706	!-- output on the respective first PE along y (myidy=0).
1707	IF ( ( section(is,s_ind) >= nys .AND. &
1708	section(is,s_ind) <= nyn ) .OR. &
1709	( section(is,s_ind) == -1 .AND. myidy == 0 ) ) THEN
1710	#if defined( __netcdf )
1711	!
1712	!-- For parallel output, all cross sections are first
1713	!-- stored here on a local array and will be written to the
1714	!-- output file afterwards to increase the performance.
1715	DO i = nxl, nxr
1716	DO k = nzb_do, nzt_do
1717	local_2d_sections_l(i,is,k) = local_2d(i,k)
1718	ENDDO
1719	ENDDO
1720	#endif
1721	ENDIF
1722
1723	ELSE
1724
1725	IF ( data_output_2d_on_each_pe ) THEN
1726	!
1727	!-- Output of partial arrays on each PE. If the cross
1728	!-- section does not reside on the PE, output special
1729	!-- index values.
1730	#if defined( __netcdf )
1731	IF ( myid == 0 ) THEN
1732	WRITE ( 22 ) time_since_reference_point, &
1733	do2d_xz_time_count(av), av
1734	ENDIF
1735	#endif
1736	DO i = 0, io_blocks-1
1737	IF ( i == io_group ) THEN
1738	IF ( ( section(is,s_ind) >= nys .AND. &
1739	section(is,s_ind) <= nyn ) .OR. &
1740	( section(is,s_ind) == -1 .AND. &
1741	nys-1 == -1 ) ) &
1742	THEN
1743	WRITE (22) nxl, nxr, nzb_do, nzt_do, nzb, nzt+1
1744	WRITE (22) local_2d
1745	ELSE
1746	WRITE (22) -1, -1, -1, -1, -1, -1
1747	ENDIF
1748	ENDIF
1749	#if defined( __parallel )
1750	CALL MPI_BARRIER( comm2d, ierr )
1751	#endif
1752	ENDDO
1753
1754	ELSE
1755	!
1756	!-- PE0 receives partial arrays from all processors of the
1757	!-- respective cross section and outputs them. Here a
1758	!-- barrier has to be set, because otherwise
1759	!-- "-MPI- FATAL: Remote protocol queue full" may occur.
1760	CALL MPI_BARRIER( comm2d, ierr )
1761
1762	ngp = ( nxr-nxl + 1 ) * ( nzt_do-nzb_do + 1 )
1763	IF ( myid == 0 ) THEN
1764	!
1765	!-- Local array can be relocated directly.
1766	IF ( ( section(is,s_ind) >= nys .AND. &
1767	section(is,s_ind) <= nyn ) .OR. &
1768	( section(is,s_ind) == -1 .AND. &
1769	nys-1 == -1 ) ) THEN
1770	total_2d(nxl:nxr,nzb_do:nzt_do) = local_2d
1771	ENDIF
1772	!
1773	!-- Receive data from all other PEs.
1774	DO n = 1, numprocs-1
1775	!
1776	!-- Receive index limits first, then array.
1777	!-- Index limits are received in arbitrary order from
1778	!-- the PEs.
1779	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
1780	MPI_ANY_SOURCE, 0, comm2d, &
1781	status, ierr )
1782	!
1783	!-- Not all PEs have data for XZ-cross-section.
1784	IF ( ind(1) /= -9999 ) THEN
1785	sender = status(MPI_SOURCE)
1786	DEALLOCATE( local_2d )
1787	ALLOCATE( local_2d(ind(1):ind(2), &
1788	ind(3):ind(4)) )
1789	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
1790	MPI_REAL, sender, 1, comm2d, &
1791	status, ierr )
1792	total_2d(ind(1):ind(2),ind(3):ind(4)) = &
1793	local_2d
1794	ENDIF
1795	ENDDO
1796	!
1797	!-- Relocate the local array for the next loop increment
1798	DEALLOCATE( local_2d )
1799	ALLOCATE( local_2d(nxl:nxr,nzb_do:nzt_do) )
1800
1801	#if defined( __netcdf )
1802	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
1803	id_var_do2d(av,ivar), &
1804	total_2d(0:nx,nzb_do:nzt_do), &
1805	start = (/ 1, is, 1, do2d_xz_time_count(av) /), &
1806	count = (/ nx+1, 1, nzt_do-nzb_do+1, 1 /) )
1807	CALL netcdf_handle_error( 'data_output_2d', 58 )
1808	#endif
1809
1810	ELSE
1811	!
1812	!-- If the cross section resides on the PE, send the
1813	!-- local index limits, otherwise send -9999 to PE0.
1814	IF ( ( section(is,s_ind) >= nys .AND. &
1815	section(is,s_ind) <= nyn ) .OR. &
1816	( section(is,s_ind) == -1 .AND. nys-1 == -1 ) ) &
1817	THEN
1818	ind(1) = nxl; ind(2) = nxr
1819	ind(3) = nzb_do; ind(4) = nzt_do
1820	ELSE
1821	ind(1) = -9999; ind(2) = -9999
1822	ind(3) = -9999; ind(4) = -9999
1823	ENDIF
1824	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
1825	comm2d, ierr )
1826	!
1827	!-- If applicable, send data to PE0.
1828	IF ( ind(1) /= -9999 ) THEN
1829	CALL MPI_SEND( local_2d(nxl,nzb_do), ngp, &
1830	MPI_REAL, 0, 1, comm2d, ierr )
1831	ENDIF
1832	ENDIF
1833	!
1834	!-- A barrier has to be set, because otherwise some PEs may
1835	!-- proceed too fast so that PE0 may receive wrong data on
1836	!-- tag 0
1837	CALL MPI_BARRIER( comm2d, ierr )
1838	ENDIF
1839
1840	ENDIF
1841	#else
1842	#if defined( __netcdf )
1843	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
1844	id_var_do2d(av,ivar), &
1845	local_2d(nxl:nxr,nzb_do:nzt_do), &
1846	start = (/ 1, is, 1, do2d_xz_time_count(av) /), &
1847	count = (/ nx+1, 1, nzt_do-nzb_do+1, 1 /) )
1848	CALL netcdf_handle_error( 'data_output_2d', 451 )
1849	#endif
1850	#endif
1851
1852	CASE ( 'yz' )
1853	!
1854	!-- Update the netCDF yz cross section time axis.
1855	!-- In case of parallel output, this is only done by PE0
1856	!-- to increase the performance.
1857	IF ( time_since_reference_point /= do2d_yz_last_time(av) ) THEN
1858	do2d_yz_time_count(av) = do2d_yz_time_count(av) + 1
1859	do2d_yz_last_time(av) = time_since_reference_point
1860	IF ( myid == 0 ) THEN
1861	IF ( .NOT. data_output_2d_on_each_pe &
1862	.OR. netcdf_data_format > 4 ) &
1863	THEN
1864	#if defined( __netcdf )
1865	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
1866	id_var_time_yz(av), &
1867	(/ time_since_reference_point /), &
1868	start = (/ do2d_yz_time_count(av) /), &
1869	count = (/ 1 /) )
1870	CALL netcdf_handle_error( 'data_output_2d', 59 )
1871	#endif
1872	ENDIF
1873	ENDIF
1874	ENDIF
1875
1876	!
1877	!-- If required, carry out averaging along x
1878	IF ( section(is,s_ind) == -1 ) THEN
1879
1880	ALLOCATE( local_2d_l(nys:nyn,nzb_do:nzt_do) )
1881	local_2d_l = 0.0_wp
1882	ngp = ( nyn-nys+1 ) * ( nzt_do-nzb_do+1 )
1883	!
1884	!-- First local averaging on the PE
1885	DO k = nzb_do, nzt_do
1886	DO j = nys, nyn
1887	DO i = nxl, nxr
1888	local_2d_l(j,k) = local_2d_l(j,k) + &
1889	local_pf(i,j,k)
1890	ENDDO
1891	ENDDO
1892	ENDDO
1893	#if defined( __parallel )
1894	!
1895	!-- Now do the averaging over all PEs along x
1896	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
1897	CALL MPI_ALLREDUCE( local_2d_l(nys,nzb_do), &
1898	local_2d(nys,nzb_do), ngp, MPI_REAL, &
1899	MPI_SUM, comm1dx, ierr )
1900	#else
1901	local_2d = local_2d_l
1902	#endif
1903	local_2d = local_2d / ( nx + 1.0_wp )
1904
1905	DEALLOCATE( local_2d_l )
1906
1907	ELSE
1908	!
1909	!-- Just store the respective section on the local array
1910	!-- (but only if it is available on this PE!)
1911	IF ( section(is,s_ind) >= nxl .AND. section(is,s_ind) <= nxr ) &
1912	THEN
1913	local_2d = local_pf(section(is,s_ind),:,nzb_do:nzt_do)
1914	ENDIF
1915
1916	ENDIF
1917
1918	#if defined( __parallel )
1919	IF ( netcdf_data_format > 4 ) THEN
1920	!
1921	!-- Output in netCDF4/HDF5 format.
1922	!-- Output only on those PEs where the respective cross
1923	!-- sections reside. Cross sections averaged along x are
1924	!-- output on the respective first PE along x (myidx=0).
1925	IF ( ( section(is,s_ind) >= nxl .AND. &
1926	section(is,s_ind) <= nxr ) .OR. &
1927	( section(is,s_ind) == -1 .AND. myidx == 0 ) ) THEN
1928	#if defined( __netcdf )
1929	!
1930	!-- For parallel output, all cross sections are first
1931	!-- stored here on a local array and will be written to the
1932	!-- output file afterwards to increase the performance.
1933	DO j = nys, nyn
1934	DO k = nzb_do, nzt_do
1935	local_2d_sections_l(is,j,k) = local_2d(j,k)
1936	ENDDO
1937	ENDDO
1938	#endif
1939	ENDIF
1940
1941	ELSE
1942
1943	IF ( data_output_2d_on_each_pe ) THEN
1944	!
1945	!-- Output of partial arrays on each PE. If the cross
1946	!-- section does not reside on the PE, output special
1947	!-- index values.
1948	#if defined( __netcdf )
1949	IF ( myid == 0 ) THEN
1950	WRITE ( 23 ) time_since_reference_point, &
1951	do2d_yz_time_count(av), av
1952	ENDIF
1953	#endif
1954	DO i = 0, io_blocks-1
1955	IF ( i == io_group ) THEN
1956	IF ( ( section(is,s_ind) >= nxl .AND. &
1957	section(is,s_ind) <= nxr ) .OR. &
1958	( section(is,s_ind) == -1 .AND. &
1959	nxl-1 == -1 ) ) &
1960	THEN
1961	WRITE (23) nys, nyn, nzb_do, nzt_do, nzb, nzt+1
1962	WRITE (23) local_2d
1963	ELSE
1964	WRITE (23) -1, -1, -1, -1, -1, -1
1965	ENDIF
1966	ENDIF
1967	#if defined( __parallel )
1968	CALL MPI_BARRIER( comm2d, ierr )
1969	#endif
1970	ENDDO
1971
1972	ELSE
1973	!
1974	!-- PE0 receives partial arrays from all processors of the
1975	!-- respective cross section and outputs them. Here a
1976	!-- barrier has to be set, because otherwise
1977	!-- "-MPI- FATAL: Remote protocol queue full" may occur.
1978	CALL MPI_BARRIER( comm2d, ierr )
1979
1980	ngp = ( nyn-nys+1 ) * ( nzt_do-nzb_do+1 )
1981	IF ( myid == 0 ) THEN
1982	!
1983	!-- Local array can be relocated directly.
1984	IF ( ( section(is,s_ind) >= nxl .AND. &
1985	section(is,s_ind) <= nxr ) .OR. &
1986	( section(is,s_ind) == -1 .AND. nxl-1 == -1 ) ) &
1987	THEN
1988	total_2d(nys:nyn,nzb_do:nzt_do) = local_2d
1989	ENDIF
1990	!
1991	!-- Receive data from all other PEs.
1992	DO n = 1, numprocs-1
1993	!
1994	!-- Receive index limits first, then array.
1995	!-- Index limits are received in arbitrary order from
1996	!-- the PEs.
1997	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
1998	MPI_ANY_SOURCE, 0, comm2d, &
1999	status, ierr )
2000	!
2001	!-- Not all PEs have data for YZ-cross-section.
2002	IF ( ind(1) /= -9999 ) THEN
2003	sender = status(MPI_SOURCE)
2004	DEALLOCATE( local_2d )
2005	ALLOCATE( local_2d(ind(1):ind(2), &
2006	ind(3):ind(4)) )
2007	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
2008	MPI_REAL, sender, 1, comm2d, &
2009	status, ierr )
2010	total_2d(ind(1):ind(2),ind(3):ind(4)) = &
2011	local_2d
2012	ENDIF
2013	ENDDO
2014	!
2015	!-- Relocate the local array for the next loop increment
2016	DEALLOCATE( local_2d )
2017	ALLOCATE( local_2d(nys:nyn,nzb_do:nzt_do) )
2018
2019	#if defined( __netcdf )
2020	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2021	id_var_do2d(av,ivar), &
2022	total_2d(0:ny,nzb_do:nzt_do), &
2023	start = (/ is, 1, 1, do2d_yz_time_count(av) /), &
2024	count = (/ 1, ny+1, nzt_do-nzb_do+1, 1 /) )
2025	CALL netcdf_handle_error( 'data_output_2d', 61 )
2026	#endif
2027
2028	ELSE
2029	!
2030	!-- If the cross section resides on the PE, send the
2031	!-- local index limits, otherwise send -9999 to PE0.
2032	IF ( ( section(is,s_ind) >= nxl .AND. &
2033	section(is,s_ind) <= nxr ) .OR. &
2034	( section(is,s_ind) == -1 .AND. nxl-1 == -1 ) ) &
2035	THEN
2036	ind(1) = nys; ind(2) = nyn
2037	ind(3) = nzb_do; ind(4) = nzt_do
2038	ELSE
2039	ind(1) = -9999; ind(2) = -9999
2040	ind(3) = -9999; ind(4) = -9999
2041	ENDIF
2042	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
2043	comm2d, ierr )
2044	!
2045	!-- If applicable, send data to PE0.
2046	IF ( ind(1) /= -9999 ) THEN
2047	CALL MPI_SEND( local_2d(nys,nzb_do), ngp, &
2048	MPI_REAL, 0, 1, comm2d, ierr )
2049	ENDIF
2050	ENDIF
2051	!
2052	!-- A barrier has to be set, because otherwise some PEs may
2053	!-- proceed too fast so that PE0 may receive wrong data on
2054	!-- tag 0
2055	CALL MPI_BARRIER( comm2d, ierr )
2056	ENDIF
2057
2058	ENDIF
2059	#else
2060	#if defined( __netcdf )
2061	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2062	id_var_do2d(av,ivar), &
2063	local_2d(nys:nyn,nzb_do:nzt_do), &
2064	start = (/ is, 1, 1, do2d_xz_time_count(av) /), &
2065	count = (/ 1, ny+1, nzt_do-nzb_do+1, 1 /) )
2066	CALL netcdf_handle_error( 'data_output_2d', 452 )
2067	#endif
2068	#endif
2069
2070	END SELECT
2071
2072	is = is + 1
2073	ENDDO loop1
2074
2075	!
2076	!-- For parallel output, all data were collected before on a local array
2077	!-- and are written now to the netcdf file. This must be done to increase
2078	!-- the performance of the parallel output.
2079	#if defined( __netcdf )
2080	IF ( netcdf_data_format > 4 ) THEN
2081
2082	SELECT CASE ( mode )
2083
2084	CASE ( 'xy' )
2085	IF ( two_d ) THEN
2086	nis = 1
2087	two_d = .FALSE.
2088	ELSE
2089	nis = ns
2090	ENDIF
2091	!
2092	!-- Do not output redundant ghost point data except for the
2093	!-- boundaries of the total domain.
2094	! IF ( nxr == nx .AND. nyn /= ny ) THEN
2095	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2096	! id_var_do2d(av,ivar), &
2097	! local_2d_sections(nxl:nxr+1, &
2098	! nys:nyn,1:nis), &
2099	! start = (/ nxl+1, nys+1, 1, &
2100	! do2d_xy_time_count(av) /), &
2101	! count = (/ nxr-nxl+2, &
2102	! nyn-nys+1, nis, 1 &
2103	! /) )
2104	! ELSEIF ( nxr /= nx .AND. nyn == ny ) THEN
2105	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2106	! id_var_do2d(av,ivar), &
2107	! local_2d_sections(nxl:nxr, &
2108	! nys:nyn+1,1:nis), &
2109	! start = (/ nxl+1, nys+1, 1, &
2110	! do2d_xy_time_count(av) /), &
2111	! count = (/ nxr-nxl+1, &
2112	! nyn-nys+2, nis, 1 &
2113	! /) )
2114	! ELSEIF ( nxr == nx .AND. nyn == ny ) THEN
2115	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2116	! id_var_do2d(av,ivar), &
2117	! local_2d_sections(nxl:nxr+1, &
2118	! nys:nyn+1,1:nis), &
2119	! start = (/ nxl+1, nys+1, 1, &
2120	! do2d_xy_time_count(av) /), &
2121	! count = (/ nxr-nxl+2, &
2122	! nyn-nys+2, nis, 1 &
2123	! /) )
2124	! ELSE
2125	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2126	id_var_do2d(av,ivar), &
2127	local_2d_sections(nxl:nxr, &
2128	nys:nyn,1:nis), &
2129	start = (/ nxl+1, nys+1, 1, &
2130	do2d_xy_time_count(av) /), &
2131	count = (/ nxr-nxl+1, &
2132	nyn-nys+1, nis, 1 &
2133	/) )
2134	! ENDIF
2135
2136	CALL netcdf_handle_error( 'data_output_2d', 55 )
2137
2138	CASE ( 'xz' )
2139	!
2140	!-- First, all PEs get the information of all cross-sections.
2141	!-- Then the data are written to the output file by all PEs
2142	!-- while NF90_COLLECTIVE is set in subroutine
2143	!-- define_netcdf_header. Although redundant information are
2144	!-- written to the output file in that case, the performance
2145	!-- is significantly better compared to the case where only
2146	!-- the first row of PEs in x-direction (myidx = 0) is given
2147	!-- the output while NF90_INDEPENDENT is set.
2148	IF ( npey /= 1 ) THEN
2149
2150	#if defined( __parallel )
2151	!
2152	!-- Distribute data over all PEs along y
2153	ngp = ( nxr-nxl+1 ) * ( nzt_do-nzb_do+1 ) * ns
2154	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
2155	CALL MPI_ALLREDUCE( local_2d_sections_l(nxl,1,nzb_do), &
2156	local_2d_sections(nxl,1,nzb_do), &
2157	ngp, MPI_REAL, MPI_SUM, comm1dy, &
2158	ierr )
2159	#else
2160	local_2d_sections = local_2d_sections_l
2161	#endif
2162	ENDIF
2163	!
2164	!-- Do not output redundant ghost point data except for the
2165	!-- boundaries of the total domain.
2166	! IF ( nxr == nx ) THEN
2167	! nc_stat = NF90_PUT_VAR( id_set_xz(av), &
2168	! id_var_do2d(av,ivar), &
2169	! local_2d_sections(nxl:nxr+1,1:ns, &
2170	! nzb_do:nzt_do), &
2171	! start = (/ nxl+1, 1, 1, &
2172	! do2d_xz_time_count(av) /), &
2173	! count = (/ nxr-nxl+2, ns, nzt_do-nzb_do+1, &
2174	! 1 /) )
2175	! ELSE
2176	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
2177	id_var_do2d(av,ivar), &
2178	local_2d_sections(nxl:nxr,1:ns, &
2179	nzb_do:nzt_do), &
2180	start = (/ nxl+1, 1, 1, &
2181	do2d_xz_time_count(av) /), &
2182	count = (/ nxr-nxl+1, ns, nzt_do-nzb_do+1, &
2183	1 /) )
2184	! ENDIF
2185
2186	CALL netcdf_handle_error( 'data_output_2d', 57 )
2187
2188	CASE ( 'yz' )
2189	!
2190	!-- First, all PEs get the information of all cross-sections.
2191	!-- Then the data are written to the output file by all PEs
2192	!-- while NF90_COLLECTIVE is set in subroutine
2193	!-- define_netcdf_header. Although redundant information are
2194	!-- written to the output file in that case, the performance
2195	!-- is significantly better compared to the case where only
2196	!-- the first row of PEs in y-direction (myidy = 0) is given
2197	!-- the output while NF90_INDEPENDENT is set.
2198	IF ( npex /= 1 ) THEN
2199
2200	#if defined( __parallel )
2201	!
2202	!-- Distribute data over all PEs along x
2203	ngp = ( nyn-nys+1 ) * ( nzt-nzb + 2 ) * ns
2204	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
2205	CALL MPI_ALLREDUCE( local_2d_sections_l(1,nys,nzb_do), &
2206	local_2d_sections(1,nys,nzb_do), &
2207	ngp, MPI_REAL, MPI_SUM, comm1dx, &
2208	ierr )
2209	#else
2210	local_2d_sections = local_2d_sections_l
2211	#endif
2212	ENDIF
2213	!
2214	!-- Do not output redundant ghost point data except for the
2215	!-- boundaries of the total domain.
2216	! IF ( nyn == ny ) THEN
2217	! nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2218	! id_var_do2d(av,ivar), &
2219	! local_2d_sections(1:ns, &
2220	! nys:nyn+1,nzb_do:nzt_do), &
2221	! start = (/ 1, nys+1, 1, &
2222	! do2d_yz_time_count(av) /), &
2223	! count = (/ ns, nyn-nys+2, &
2224	! nzt_do-nzb_do+1, 1 /) )
2225	! ELSE
2226	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2227	id_var_do2d(av,ivar), &
2228	local_2d_sections(1:ns,nys:nyn, &
2229	nzb_do:nzt_do), &
2230	start = (/ 1, nys+1, 1, &
2231	do2d_yz_time_count(av) /), &
2232	count = (/ ns, nyn-nys+1, &
2233	nzt_do-nzb_do+1, 1 /) )
2234	! ENDIF
2235
2236	CALL netcdf_handle_error( 'data_output_2d', 60 )
2237
2238	CASE DEFAULT
2239	message_string = 'unknown cross-section: ' // TRIM( mode )
2240	CALL message( 'data_output_2d', 'PA0180', 1, 2, 0, 6, 0 )
2241
2242	END SELECT
2243
2244	ENDIF
2245	#endif
2246	ENDIF
2247
2248	ivar = ivar + 1
2249	l = MAX( 2, LEN_TRIM( do2d(av,ivar) ) )
2250	do2d_mode = do2d(av,ivar)(l-1:l)
2251
2252	ENDDO
2253
2254	!
2255	!-- Deallocate temporary arrays.
2256	IF ( ALLOCATED( level_z ) ) DEALLOCATE( level_z )
2257	IF ( netcdf_data_format > 4 ) THEN
2258	DEALLOCATE( local_pf, local_2d, local_2d_sections )
2259	IF( mode == 'xz' .OR. mode == 'yz' ) DEALLOCATE( local_2d_sections_l )
2260	ENDIF
2261	#if defined( __parallel )
2262	IF ( .NOT. data_output_2d_on_each_pe .AND. myid == 0 ) THEN
2263	DEALLOCATE( total_2d )
2264	ENDIF
2265	#endif
2266
2267	!
2268	!-- Close plot output file.
2269	file_id = 20 + s_ind
2270
2271	IF ( data_output_2d_on_each_pe ) THEN
2272	DO i = 0, io_blocks-1
2273	IF ( i == io_group ) THEN
2274	CALL close_file( file_id )
2275	ENDIF
2276	#if defined( __parallel )
2277	CALL MPI_BARRIER( comm2d, ierr )
2278	#endif
2279	ENDDO
2280	ELSE
2281	IF ( myid == 0 ) CALL close_file( file_id )
2282	ENDIF
2283
2284	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
2285
2286	IF ( debug_output_timestep ) CALL debug_message( 'data_output_2d', 'end' )
2287
2288
2289	END SUBROUTINE data_output_2d

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

Download in other formats:

| Impressum | ©Leibniz Universität Hannover |