Home

Context Navigation

source: palm/trunk/SOURCE/data_output_2d.f90 @ 3924

Last change on this file since 3924 was 3885, checked in by kanani, 6 years ago
restructure/add location/debug messages
Property svn:keywords set to `Id`
File size: 96.5 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 3885 2019-04-11 11:29:34Z monakurppa $
27	! Changes related to global restructuring of location messages and introduction
28	! of additional debug messages
29	!
30	! 3766 2019-02-26 16:23:41Z raasch
31	! unused variables removed
32	!
33	! 3655 2019-01-07 16:51:22Z knoop
34	! Bugfix: use time_since_reference_point instead of simulated_time (relevant
35	! when using wall/soil spinup)
36	!
37	! 3637 2018-12-20 01:51:36Z knoop
38	! Implementation of the PALM module interface
39	!
40	! 3597 2018-12-04 08:40:18Z maronga
41	! Added theta_2m
42	!
43	! 3589 2018-11-30 15:09:51Z suehring
44	! Move the control parameter "salsa" from salsa_mod to control_parameters
45	! (M. Kurppa)
46	!
47	! 3582 2018-11-29 19:16:36Z suehring
48	! Remove fill_value from bio_data_output_2d call
49	! dom_dwd_user, Schrempf:
50	! Clean up of biometeorology calls,
51	! remove uv exposure model code, this is now part of biometeorology_mod.
52	!
53	! 3554 2018-11-22 11:24:52Z gronemeier
54	! - add variable description
55	! - rename variable 'if' into 'ivar'
56	! - removed namelist LOCAL
57	! - removed variable rtext
58	!
59	! 3525 2018-11-14 16:06:14Z kanani
60	! Changes related to clean-up of biometeorology (dom_dwd_user)
61	!
62	! 3467 2018-10-30 19:05:21Z suehring
63	! Implementation of a new aerosol module salsa.
64	!
65	! 3448 2018-10-29 18:14:31Z kanani
66	! Add biometeorology
67	!
68	! 3421 2018-10-24 18:39:32Z gronemeier
69	! Renamed output variables
70	!
71	! 3419 2018-10-24 17:27:31Z gronemeier
72	! minor formatting (kanani)
73	! chem_data_output_2d subroutine added (basit)
74	!
75	! 3294 2018-10-01 02:37:10Z raasch
76	! changes concerning modularization of ocean option
77	!
78	! 3274 2018-09-24 15:42:55Z knoop
79	! Modularization of all bulk cloud physics code components
80	!
81	! 3241 2018-09-12 15:02:00Z raasch
82	! unused variables removed
83	!
84	! 3176 2018-07-26 17:12:48Z suehring
85	! Remove output of latent heat flux at urban-surfaces and set fill values
86	! instead
87	!
88	! 3052 2018-05-31 06:11:20Z raasch
89	! Do not open FORTRAN binary files in case of parallel netCDF I/O
90	!
91	! 3049 2018-05-29 13:52:36Z Giersch
92	! Error messages revised
93	!
94	! 3045 2018-05-28 07:55:41Z Giersch
95	! Error messages revised
96	!
97	! 3014 2018-05-09 08:42:38Z maronga
98	! Added nzb_do and nzt_do for some modules for 2d output
99	!
100	! 3004 2018-04-27 12:33:25Z Giersch
101	! precipitation_rate removed, case prr*_xy removed, to_be_resorted have to point
102	! to ql_vp_av and not to ql_vp, allocation checks implemented (averaged data
103	! will be assigned to fill values if no allocation happened so far)
104	!
105	! 2963 2018-04-12 14:47:44Z suehring
106	! Introduce index for vegetation/wall, pavement/green-wall and water/window
107	! surfaces, for clearer access of surface fraction, albedo, emissivity, etc. .
108	!
109	! 2817 2018-02-19 16:32:21Z knoop
110	! Preliminary gust module interface implemented
111	!
112	! 2805 2018-02-14 17:00:09Z suehring
113	! Consider also default-type surfaces for surface temperature output.
114	!
115	! 2797 2018-02-08 13:24:35Z suehring
116	! Enable output of ground-heat flux also at urban surfaces.
117	!
118	! 2743 2018-01-12 16:03:39Z suehring
119	! In case of natural- and urban-type surfaces output surfaces fluxes in W/m2.
120	!
121	! 2742 2018-01-12 14:59:47Z suehring
122	! Enable output of surface temperature
123	!
124	! 2735 2018-01-11 12:01:27Z suehring
125	! output of r_a moved from land-surface to consider also urban-type surfaces
126	!
127	! 2718 2018-01-02 08:49:38Z maronga
128	! Corrected "Former revisions" section
129	!
130	! 2696 2017-12-14 17:12:51Z kanani
131	! Change in file header (GPL part)
132	! Implementation of uv exposure model (FK)
133	! Implementation of turbulence_closure_mod (TG)
134	! Set fill values at topography grid points or e.g. non-natural-type surface
135	! in case of LSM output (MS)
136	!
137	! 2512 2017-10-04 08:26:59Z raasch
138	! upper bounds of cross section output changed from nx+1,ny+1 to nx,ny
139	! no output of ghost layer data
140	!
141	! 2292 2017-06-20 09:51:42Z schwenkel
142	! Implementation of new microphysic scheme: cloud_scheme = 'morrison'
143	! includes two more prognostic equations for cloud drop concentration (nc)
144	! and cloud water content (qc).
145	!
146	! 2277 2017-06-12 10:47:51Z kanani
147	! Removed unused variables do2d_xy_n, do2d_xz_n, do2d_yz_n
148	!
149	! 2233 2017-05-30 18:08:54Z suehring
150	!
151	! 2232 2017-05-30 17:47:52Z suehring
152	! Adjustments to new surface concept
153	!
154	!
155	! 2190 2017-03-21 12:16:43Z raasch
156	! bugfix for r2031: string rho replaced by rho_ocean
157	!
158	! 2031 2016-10-21 15:11:58Z knoop
159	! renamed variable rho to rho_ocean and rho_av to rho_ocean_av
160	!
161	! 2000 2016-08-20 18:09:15Z knoop
162	! Forced header and separation lines into 80 columns
163	!
164	! 1980 2016-07-29 15:51:57Z suehring
165	! Bugfix, in order to steer user-defined output, setting flag found explicitly
166	! to .F.
167	!
168	! 1976 2016-07-27 13:28:04Z maronga
169	! Output of radiation quantities is now done directly in the respective module
170	!
171	! 1972 2016-07-26 07:52:02Z maronga
172	! Output of land surface quantities is now done directly in the respective
173	! module
174	!
175	! 1960 2016-07-12 16:34:24Z suehring
176	! Scalar surface flux added
177	! Rename INTEGER variable s into s_ind, as s is already assigned to scalar
178	!
179	! 1849 2016-04-08 11:33:18Z hoffmann
180	! precipitation_amount, precipitation_rate, prr moved to arrays_3d
181	!
182	! 1822 2016-04-07 07:49:42Z hoffmann
183	! Output of bulk cloud physics simplified.
184	!
185	! 1788 2016-03-10 11:01:04Z maronga
186	! Added output of z0q
187	!
188	! 1783 2016-03-06 18:36:17Z raasch
189	! name change of netcdf routines and module + related changes
190	!
191	! 1745 2016-02-05 13:06:51Z gronemeier
192	! Bugfix: test if time axis limit exceeds moved to point after call of check_open
193	!
194	! 1703 2015-11-02 12:38:44Z raasch
195	! bugfix for output of single (*) xy-sections in case of parallel netcdf I/O
196	!
197	! 1701 2015-11-02 07:43:04Z maronga
198	! Bugfix in output of RRTGM data
199	!
200	! 1691 2015-10-26 16:17:44Z maronga
201	! Added output of Obukhov length (ol) and radiative heating rates for RRTMG.
202	! Formatting corrections.
203	!
204	! 1682 2015-10-07 23:56:08Z knoop
205	! Code annotations made doxygen readable
206	!
207	! 1585 2015-04-30 07:05:52Z maronga
208	! Added support for RRTMG
209	!
210	! 1555 2015-03-04 17:44:27Z maronga
211	! Added output of r_a and r_s
212	!
213	! 1551 2015-03-03 14:18:16Z maronga
214	! Added suppport for land surface model and radiation model output. In the course
215	! of this action, the limits for vertical loops have been changed (from nzb and
216	! nzt+1 to nzb_do and nzt_do, respectively in order to allow soil model output).
217	! Moreover, a new vertical grid zs was introduced.
218	!
219	! 1359 2014-04-11 17:15:14Z hoffmann
220	! New particle structure integrated.
221	!
222	! 1353 2014-04-08 15:21:23Z heinze
223	! REAL constants provided with KIND-attribute
224	!
225	! 1327 2014-03-21 11:00:16Z raasch
226	! parts concerning iso2d output removed,
227	! -netcdf output queries
228	!
229	! 1320 2014-03-20 08:40:49Z raasch
230	! ONLY-attribute added to USE-statements,
231	! kind-parameters added to all INTEGER and REAL declaration statements,
232	! kinds are defined in new module kinds,
233	! revision history before 2012 removed,
234	! comment fields (!:) to be used for variable explanations added to
235	! all variable declaration statements
236	!
237	! 1318 2014-03-17 13:35:16Z raasch
238	! barrier argument removed from cpu_log.
239	! module interfaces removed
240	!
241	! 1311 2014-03-14 12:13:39Z heinze
242	! bugfix: close #if defined( __netcdf )
243	!
244	! 1308 2014-03-13 14:58:42Z fricke
245	! +local_2d_sections, local_2d_sections_l, ns
246	! Check, if the limit of the time dimension is exceeded for parallel output
247	! To increase the performance for parallel output, the following is done:
248	! - Update of time axis is only done by PE0
249	! - Cross sections are first stored on a local array and are written
250	! collectively to the output file by all PEs.
251	!
252	! 1115 2013-03-26 18:16:16Z hoffmann
253	! ql is calculated by calc_liquid_water_content
254	!
255	! 1076 2012-12-05 08:30:18Z hoffmann
256	! Bugfix in output of ql
257	!
258	! 1065 2012-11-22 17:42:36Z hoffmann
259	! Bugfix: Output of cross sections of ql
260	!
261	! 1053 2012-11-13 17:11:03Z hoffmann
262	! +qr, nr, qc and cross sections
263	!
264	! 1036 2012-10-22 13:43:42Z raasch
265	! code put under GPL (PALM 3.9)
266	!
267	! 1031 2012-10-19 14:35:30Z raasch
268	! netCDF4 without parallel file support implemented
269	!
270	! 1007 2012-09-19 14:30:36Z franke
271	! Bugfix: missing calculation of ql_vp added
272	!
273	! 978 2012-08-09 08:28:32Z fricke
274	! +z0h
275	!
276	! Revision 1.1 1997/08/11 06:24:09 raasch
277	! Initial revision
278	!
279	!
280	! Description:
281	! ------------
282	!> Data output of cross-sections in netCDF format or binary format
283	!> to be later converted to NetCDF by helper routine combine_plot_fields.
284	!> Attention: The position of the sectional planes is still not always computed
285	!> --------- correctly. (zu is used always)!
286	!------------------------------------------------------------------------------!
287	SUBROUTINE data_output_2d( mode, av )
288
289
290	USE arrays_3d, &
291	ONLY: dzw, d_exner, e, heatflux_output_conversion, p, pt, q, ql, ql_c, ql_v, s, tend, u, &
292	v, vpt, w, waterflux_output_conversion, zu, zw
293
294	USE averaging
295
296	USE basic_constants_and_equations_mod, &
297	ONLY: c_p, lv_d_cp, l_v
298
299	USE bulk_cloud_model_mod, &
300	ONLY: bulk_cloud_model
301
302	USE control_parameters, &
303	ONLY: data_output_2d_on_each_pe, &
304	data_output_xy, data_output_xz, data_output_yz, &
305	debug_output, debug_string, &
306	do2d, &
307	do2d_xy_last_time, do2d_xy_time_count, &
308	do2d_xz_last_time, do2d_xz_time_count, &
309	do2d_yz_last_time, do2d_yz_time_count, &
310	ibc_uv_b, io_blocks, io_group, message_string, &
311	ntdim_2d_xy, ntdim_2d_xz, ntdim_2d_yz, &
312	psolver, section, &
313	time_since_reference_point
314
315	USE cpulog, &
316	ONLY: cpu_log, log_point
317
318	USE indices, &
319	ONLY: nbgp, nx, nxl, nxlg, nxr, nxrg, ny, nyn, nyng, nys, nysg, &
320	nzb, nzt, wall_flags_0
321
322	USE kinds
323
324	USE land_surface_model_mod, &
325	ONLY: zs
326
327	USE module_interface, &
328	ONLY: module_interface_data_output_2d
329
330	#if defined( __netcdf )
331	USE NETCDF
332	#endif
333
334	USE netcdf_interface, &
335	ONLY: fill_value, id_set_xy, id_set_xz, id_set_yz, id_var_do2d, &
336	id_var_time_xy, id_var_time_xz, id_var_time_yz, nc_stat, &
337	netcdf_data_format, netcdf_handle_error
338
339	USE particle_attributes, &
340	ONLY: grid_particles, number_of_particles, particle_advection_start, &
341	particles, prt_count
342
343	USE pegrid
344
345	USE surface_mod, &
346	ONLY: ind_pav_green, ind_veg_wall, ind_wat_win, surf_def_h, &
347	surf_lsm_h, surf_usm_h
348
349	USE turbulence_closure_mod, &
350	ONLY: tcm_data_output_2d
351
352
353	IMPLICIT NONE
354
355	CHARACTER (LEN=2) :: do2d_mode !< output mode of variable ('xy', 'xz', 'yz')
356	CHARACTER (LEN=2) :: mode !< mode with which the routine is called ('xy', 'xz', 'yz')
357	CHARACTER (LEN=4) :: grid !< string defining the vertical grid
358
359	INTEGER(iwp) :: av !< flag for (non-)average output
360	INTEGER(iwp) :: ngp !< number of grid points of an output slice
361	INTEGER(iwp) :: file_id !< id of output files
362	INTEGER(iwp) :: flag_nr !< number of masking flag
363	INTEGER(iwp) :: i !< loop index
364	INTEGER(iwp) :: iis !< vertical index of a xy slice in array 'local_2d_sections'
365	INTEGER(iwp) :: is !< slice index
366	INTEGER(iwp) :: ivar !< variable index
367	INTEGER(iwp) :: j !< loop index
368	INTEGER(iwp) :: k !< loop index
369	INTEGER(iwp) :: l !< loop index
370	INTEGER(iwp) :: layer_xy !< vertical index of a xy slice in array 'local_pf'
371	INTEGER(iwp) :: m !< loop index
372	INTEGER(iwp) :: n !< loop index
373	INTEGER(iwp) :: nis !< number of vertical slices to be written via parallel NetCDF output
374	INTEGER(iwp) :: ns !< number of output slices
375	INTEGER(iwp) :: nzb_do !< lower limit of the data field (usually nzb)
376	INTEGER(iwp) :: nzt_do !< upper limit of the data field (usually nzt+1)
377	INTEGER(iwp) :: s_ind !< index of slice types (xy=1, xz=2, yz=3)
378	INTEGER(iwp) :: sender !< PE id of sending PE
379	INTEGER(iwp) :: ind(4) !< index limits (lower/upper bounds) of array 'local_2d'
380
381	LOGICAL :: found !< true if output variable was found
382	LOGICAL :: resorted !< true if variable is resorted
383	LOGICAL :: two_d !< true if variable is only two dimensional
384
385	REAL(wp) :: mean_r !< mean particle radius
386	REAL(wp) :: s_r2 !< sum( particle-radius**2 )
387	REAL(wp) :: s_r3 !< sum( particle-radius**3 )
388
389	REAL(wp), DIMENSION(:), ALLOCATABLE :: level_z !< z levels for output array
390	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: local_2d !< local 2-dimensional array containing output values
391	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: local_2d_l !< local 2-dimensional array containing output values
392
393	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_pf !< output array
394	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_2d_sections !< local array containing values at all slices
395	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_2d_sections_l !< local array containing values at all slices
396
397	#if defined( __parallel )
398	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: total_2d !< same as local_2d
399	#endif
400	REAL(wp), DIMENSION(:,:,:), POINTER :: to_be_resorted !< points to array which shall be output
401
402
403	IF ( debug_output ) THEN
404	WRITE( debug_string, * ) 'data_output_2d'
405	CALL debug_message( debug_string, 'start' )
406	ENDIF
407	!
408	!-- Immediate return, if no output is requested (no respective sections
409	!-- found in parameter data_output)
410	IF ( mode == 'xy' .AND. .NOT. data_output_xy(av) ) RETURN
411	IF ( mode == 'xz' .AND. .NOT. data_output_xz(av) ) RETURN
412	IF ( mode == 'yz' .AND. .NOT. data_output_yz(av) ) RETURN
413
414	CALL cpu_log (log_point(3),'data_output_2d','start')
415
416	two_d = .FALSE. ! local variable to distinguish between output of pure 2D
417	! arrays and cross-sections of 3D arrays.
418
419	!
420	!-- Depending on the orientation of the cross-section, the respective output
421	!-- files have to be opened.
422	SELECT CASE ( mode )
423
424	CASE ( 'xy' )
425	s_ind = 1
426	ALLOCATE( level_z(nzb:nzt+1), local_2d(nxl:nxr,nys:nyn) )
427
428	IF ( netcdf_data_format > 4 ) THEN
429	ns = 1
430	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
431	ns = ns + 1
432	ENDDO
433	ns = ns - 1
434	ALLOCATE( local_2d_sections(nxl:nxr,nys:nyn,1:ns) )
435	local_2d_sections = 0.0_wp
436	ENDIF
437
438	!
439	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
440	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
441	CALL check_open( 101+av*10 )
442	ENDIF
443	IF ( data_output_2d_on_each_pe .AND. netcdf_data_format < 5 ) THEN
444	CALL check_open( 21 )
445	ELSE
446	IF ( myid == 0 ) THEN
447	#if defined( __parallel )
448	ALLOCATE( total_2d(0:nx,0:ny) )
449	#endif
450	ENDIF
451	ENDIF
452
453	CASE ( 'xz' )
454	s_ind = 2
455	ALLOCATE( local_2d(nxl:nxr,nzb:nzt+1) )
456
457	IF ( netcdf_data_format > 4 ) THEN
458	ns = 1
459	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
460	ns = ns + 1
461	ENDDO
462	ns = ns - 1
463	ALLOCATE( local_2d_sections(nxl:nxr,1:ns,nzb:nzt+1) )
464	ALLOCATE( local_2d_sections_l(nxl:nxr,1:ns,nzb:nzt+1) )
465	local_2d_sections = 0.0_wp; local_2d_sections_l = 0.0_wp
466	ENDIF
467
468	!
469	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
470	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
471	CALL check_open( 102+av*10 )
472	ENDIF
473
474	IF ( data_output_2d_on_each_pe .AND. netcdf_data_format < 5 ) THEN
475	CALL check_open( 22 )
476	ELSE
477	IF ( myid == 0 ) THEN
478	#if defined( __parallel )
479	ALLOCATE( total_2d(0:nx,nzb:nzt+1) )
480	#endif
481	ENDIF
482	ENDIF
483
484	CASE ( 'yz' )
485	s_ind = 3
486	ALLOCATE( local_2d(nys:nyn,nzb:nzt+1) )
487
488	IF ( netcdf_data_format > 4 ) THEN
489	ns = 1
490	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
491	ns = ns + 1
492	ENDDO
493	ns = ns - 1
494	ALLOCATE( local_2d_sections(1:ns,nys:nyn,nzb:nzt+1) )
495	ALLOCATE( local_2d_sections_l(1:ns,nys:nyn,nzb:nzt+1) )
496	local_2d_sections = 0.0_wp; local_2d_sections_l = 0.0_wp
497	ENDIF
498
499	!
500	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
501	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
502	CALL check_open( 103+av*10 )
503	ENDIF
504
505	IF ( data_output_2d_on_each_pe .AND. netcdf_data_format < 5 ) THEN
506	CALL check_open( 23 )
507	ELSE
508	IF ( myid == 0 ) THEN
509	#if defined( __parallel )
510	ALLOCATE( total_2d(0:ny,nzb:nzt+1) )
511	#endif
512	ENDIF
513	ENDIF
514
515	CASE DEFAULT
516	message_string = 'unknown cross-section: ' // TRIM( mode )
517	CALL message( 'data_output_2d', 'PA0180', 1, 2, 0, 6, 0 )
518
519	END SELECT
520
521	!
522	!-- For parallel netcdf output the time axis must be limited. Return, if this
523	!-- limit is exceeded. This could be the case, if the simulated time exceeds
524	!-- the given end time by the length of the given output interval.
525	IF ( netcdf_data_format > 4 ) THEN
526	IF ( mode == 'xy' .AND. do2d_xy_time_count(av) + 1 > &
527	ntdim_2d_xy(av) ) THEN
528	WRITE ( message_string, * ) 'Output of xy cross-sections is not ', &
529	'given at t=', time_since_reference_point, 's because the', &
530	' maximum number of output time levels is exceeded.'
531	CALL message( 'data_output_2d', 'PA0384', 0, 1, 0, 6, 0 )
532	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
533	RETURN
534	ENDIF
535	IF ( mode == 'xz' .AND. do2d_xz_time_count(av) + 1 > &
536	ntdim_2d_xz(av) ) THEN
537	WRITE ( message_string, * ) 'Output of xz cross-sections is not ', &
538	'given at t=', time_since_reference_point, 's because the', &
539	' maximum number of output time levels is exceeded.'
540	CALL message( 'data_output_2d', 'PA0385', 0, 1, 0, 6, 0 )
541	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
542	RETURN
543	ENDIF
544	IF ( mode == 'yz' .AND. do2d_yz_time_count(av) + 1 > &
545	ntdim_2d_yz(av) ) THEN
546	WRITE ( message_string, * ) 'Output of yz cross-sections is not ', &
547	'given at t=', time_since_reference_point, 's because the', &
548	' maximum number of output time levels is exceeded.'
549	CALL message( 'data_output_2d', 'PA0386', 0, 1, 0, 6, 0 )
550	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
551	RETURN
552	ENDIF
553	ENDIF
554
555	!
556	!-- Allocate a temporary array for resorting (kji -> ijk).
557	ALLOCATE( local_pf(nxl:nxr,nys:nyn,nzb:nzt+1) )
558	local_pf = 0.0
559
560	!
561	!-- Loop of all variables to be written.
562	!-- Output dimensions chosen
563	ivar = 1
564	l = MAX( 2, LEN_TRIM( do2d(av,ivar) ) )
565	do2d_mode = do2d(av,ivar)(l-1:l)
566
567	DO WHILE ( do2d(av,ivar)(1:1) /= ' ' )
568
569	IF ( do2d_mode == mode ) THEN
570	!
571	!-- Set flag to steer output of radiation, land-surface, or user-defined
572	!-- quantities
573	found = .FALSE.
574
575	nzb_do = nzb
576	nzt_do = nzt+1
577	!
578	!-- Before each output, set array local_pf to fill value
579	local_pf = fill_value
580	!
581	!-- Set masking flag for topography for not resorted arrays
582	flag_nr = 0
583
584	!
585	!-- Store the array chosen on the temporary array.
586	resorted = .FALSE.
587	SELECT CASE ( TRIM( do2d(av,ivar) ) )
588	CASE ( 'e_xy', 'e_xz', 'e_yz' )
589	IF ( av == 0 ) THEN
590	to_be_resorted => e
591	ELSE
592	IF ( .NOT. ALLOCATED( e_av ) ) THEN
593	ALLOCATE( e_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
594	e_av = REAL( fill_value, KIND = wp )
595	ENDIF
596	to_be_resorted => e_av
597	ENDIF
598	IF ( mode == 'xy' ) level_z = zu
599
600	CASE ( 'thetal_xy', 'thetal_xz', 'thetal_yz' )
601	IF ( av == 0 ) THEN
602	to_be_resorted => pt
603	ELSE
604	IF ( .NOT. ALLOCATED( lpt_av ) ) THEN
605	ALLOCATE( lpt_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
606	lpt_av = REAL( fill_value, KIND = wp )
607	ENDIF
608	to_be_resorted => lpt_av
609	ENDIF
610	IF ( mode == 'xy' ) level_z = zu
611
612	CASE ( 'lwp*_xy' ) ! 2d-array
613	IF ( av == 0 ) THEN
614	DO i = nxl, nxr
615	DO j = nys, nyn
616	local_pf(i,j,nzb+1) = SUM( ql(nzb:nzt,j,i) * &
617	dzw(1:nzt+1) )
618	ENDDO
619	ENDDO
620	ELSE
621	IF ( .NOT. ALLOCATED( lwp_av ) ) THEN
622	ALLOCATE( lwp_av(nysg:nyng,nxlg:nxrg) )
623	lwp_av = REAL( fill_value, KIND = wp )
624	ENDIF
625	DO i = nxl, nxr
626	DO j = nys, nyn
627	local_pf(i,j,nzb+1) = lwp_av(j,i)
628	ENDDO
629	ENDDO
630	ENDIF
631	resorted = .TRUE.
632	two_d = .TRUE.
633	level_z(nzb+1) = zu(nzb+1)
634
635	CASE ( 'ghf*_xy' ) ! 2d-array
636	IF ( av == 0 ) THEN
637	DO m = 1, surf_lsm_h%ns
638	i = surf_lsm_h%i(m)
639	j = surf_lsm_h%j(m)
640	local_pf(i,j,nzb+1) = surf_lsm_h%ghf(m)
641	ENDDO
642	DO m = 1, surf_usm_h%ns
643	i = surf_usm_h%i(m)
644	j = surf_usm_h%j(m)
645	local_pf(i,j,nzb+1) = surf_usm_h%frac(ind_veg_wall,m) * &
646	surf_usm_h%wghf_eb(m) + &
647	surf_usm_h%frac(ind_pav_green,m) * &
648	surf_usm_h%wghf_eb_green(m) + &
649	surf_usm_h%frac(ind_wat_win,m) * &
650	surf_usm_h%wghf_eb_window(m)
651	ENDDO
652	ELSE
653	IF ( .NOT. ALLOCATED( ghf_av ) ) THEN
654	ALLOCATE( ghf_av(nysg:nyng,nxlg:nxrg) )
655	ghf_av = REAL( fill_value, KIND = wp )
656	ENDIF
657	DO i = nxl, nxr
658	DO j = nys, nyn
659	local_pf(i,j,nzb+1) = ghf_av(j,i)
660	ENDDO
661	ENDDO
662	ENDIF
663
664	resorted = .TRUE.
665	two_d = .TRUE.
666	level_z(nzb+1) = zu(nzb+1)
667
668	CASE ( 'ol*_xy' ) ! 2d-array
669	IF ( av == 0 ) THEN
670	DO m = 1, surf_def_h(0)%ns
671	i = surf_def_h(0)%i(m)
672	j = surf_def_h(0)%j(m)
673	local_pf(i,j,nzb+1) = surf_def_h(0)%ol(m)
674	ENDDO
675	DO m = 1, surf_lsm_h%ns
676	i = surf_lsm_h%i(m)
677	j = surf_lsm_h%j(m)
678	local_pf(i,j,nzb+1) = surf_lsm_h%ol(m)
679	ENDDO
680	DO m = 1, surf_usm_h%ns
681	i = surf_usm_h%i(m)
682	j = surf_usm_h%j(m)
683	local_pf(i,j,nzb+1) = surf_usm_h%ol(m)
684	ENDDO
685	ELSE
686	IF ( .NOT. ALLOCATED( ol_av ) ) THEN
687	ALLOCATE( ol_av(nysg:nyng,nxlg:nxrg) )
688	ol_av = REAL( fill_value, KIND = wp )
689	ENDIF
690	DO i = nxl, nxr
691	DO j = nys, nyn
692	local_pf(i,j,nzb+1) = ol_av(j,i)
693	ENDDO
694	ENDDO
695	ENDIF
696	resorted = .TRUE.
697	two_d = .TRUE.
698	level_z(nzb+1) = zu(nzb+1)
699
700	CASE ( 'p_xy', 'p_xz', 'p_yz' )
701	IF ( av == 0 ) THEN
702	IF ( psolver /= 'sor' ) CALL exchange_horiz( p, nbgp )
703	to_be_resorted => p
704	ELSE
705	IF ( .NOT. ALLOCATED( p_av ) ) THEN
706	ALLOCATE( p_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
707	p_av = REAL( fill_value, KIND = wp )
708	ENDIF
709	IF ( psolver /= 'sor' ) CALL exchange_horiz( p_av, nbgp )
710	to_be_resorted => p_av
711	ENDIF
712	IF ( mode == 'xy' ) level_z = zu
713
714	CASE ( 'pc_xy', 'pc_xz', 'pc_yz' ) ! particle concentration
715	IF ( av == 0 ) THEN
716	IF ( time_since_reference_point >= particle_advection_start ) THEN
717	tend = prt_count
718	! CALL exchange_horiz( tend, nbgp )
719	ELSE
720	tend = 0.0_wp
721	ENDIF
722	DO i = nxl, nxr
723	DO j = nys, nyn
724	DO k = nzb, nzt+1
725	local_pf(i,j,k) = tend(k,j,i)
726	ENDDO
727	ENDDO
728	ENDDO
729	resorted = .TRUE.
730	ELSE
731	IF ( .NOT. ALLOCATED( pc_av ) ) THEN
732	ALLOCATE( pc_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
733	pc_av = REAL( fill_value, KIND = wp )
734	ENDIF
735	! CALL exchange_horiz( pc_av, nbgp )
736	to_be_resorted => pc_av
737	ENDIF
738
739	CASE ( 'pr_xy', 'pr_xz', 'pr_yz' ) ! mean particle radius (effective radius)
740	IF ( av == 0 ) THEN
741	IF ( time_since_reference_point >= particle_advection_start ) THEN
742	DO i = nxl, nxr
743	DO j = nys, nyn
744	DO k = nzb, nzt+1
745	number_of_particles = prt_count(k,j,i)
746	IF (number_of_particles <= 0) CYCLE
747	particles => grid_particles(k,j,i)%particles(1:number_of_particles)
748	s_r2 = 0.0_wp
749	s_r3 = 0.0_wp
750	DO n = 1, number_of_particles
751	IF ( particles(n)%particle_mask ) THEN
752	s_r2 = s_r2 + particles(n)%radius*2 &
753	particles(n)%weight_factor
754	s_r3 = s_r3 + particles(n)%radius*3 &
755	particles(n)%weight_factor
756	ENDIF
757	ENDDO
758	IF ( s_r2 > 0.0_wp ) THEN
759	mean_r = s_r3 / s_r2
760	ELSE
761	mean_r = 0.0_wp
762	ENDIF
763	tend(k,j,i) = mean_r
764	ENDDO
765	ENDDO
766	ENDDO
767	! CALL exchange_horiz( tend, nbgp )
768	ELSE
769	tend = 0.0_wp
770	ENDIF
771	DO i = nxl, nxr
772	DO j = nys, nyn
773	DO k = nzb, nzt+1
774	local_pf(i,j,k) = tend(k,j,i)
775	ENDDO
776	ENDDO
777	ENDDO
778	resorted = .TRUE.
779	ELSE
780	IF ( .NOT. ALLOCATED( pr_av ) ) THEN
781	ALLOCATE( pr_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
782	pr_av = REAL( fill_value, KIND = wp )
783	ENDIF
784	! CALL exchange_horiz( pr_av, nbgp )
785	to_be_resorted => pr_av
786	ENDIF
787
788	CASE ( 'theta_xy', 'theta_xz', 'theta_yz' )
789	IF ( av == 0 ) THEN
790	IF ( .NOT. bulk_cloud_model ) THEN
791	to_be_resorted => pt
792	ELSE
793	DO i = nxl, nxr
794	DO j = nys, nyn
795	DO k = nzb, nzt+1
796	local_pf(i,j,k) = pt(k,j,i) + lv_d_cp * &
797	d_exner(k) * &
798	ql(k,j,i)
799	ENDDO
800	ENDDO
801	ENDDO
802	resorted = .TRUE.
803	ENDIF
804	ELSE
805	IF ( .NOT. ALLOCATED( pt_av ) ) THEN
806	ALLOCATE( pt_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
807	pt_av = REAL( fill_value, KIND = wp )
808	ENDIF
809	to_be_resorted => pt_av
810	ENDIF
811	IF ( mode == 'xy' ) level_z = zu
812
813	CASE ( 'q_xy', 'q_xz', 'q_yz' )
814	IF ( av == 0 ) THEN
815	to_be_resorted => q
816	ELSE
817	IF ( .NOT. ALLOCATED( q_av ) ) THEN
818	ALLOCATE( q_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
819	q_av = REAL( fill_value, KIND = wp )
820	ENDIF
821	to_be_resorted => q_av
822	ENDIF
823	IF ( mode == 'xy' ) level_z = zu
824
825	CASE ( 'ql_xy', 'ql_xz', 'ql_yz' )
826	IF ( av == 0 ) THEN
827	to_be_resorted => ql
828	ELSE
829	IF ( .NOT. ALLOCATED( ql_av ) ) THEN
830	ALLOCATE( ql_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
831	ql_av = REAL( fill_value, KIND = wp )
832	ENDIF
833	to_be_resorted => ql_av
834	ENDIF
835	IF ( mode == 'xy' ) level_z = zu
836
837	CASE ( 'ql_c_xy', 'ql_c_xz', 'ql_c_yz' )
838	IF ( av == 0 ) THEN
839	to_be_resorted => ql_c
840	ELSE
841	IF ( .NOT. ALLOCATED( ql_c_av ) ) THEN
842	ALLOCATE( ql_c_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
843	ql_c_av = REAL( fill_value, KIND = wp )
844	ENDIF
845	to_be_resorted => ql_c_av
846	ENDIF
847	IF ( mode == 'xy' ) level_z = zu
848
849	CASE ( 'ql_v_xy', 'ql_v_xz', 'ql_v_yz' )
850	IF ( av == 0 ) THEN
851	to_be_resorted => ql_v
852	ELSE
853	IF ( .NOT. ALLOCATED( ql_v_av ) ) THEN
854	ALLOCATE( ql_v_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
855	ql_v_av = REAL( fill_value, KIND = wp )
856	ENDIF
857	to_be_resorted => ql_v_av
858	ENDIF
859	IF ( mode == 'xy' ) level_z = zu
860
861	CASE ( 'ql_vp_xy', 'ql_vp_xz', 'ql_vp_yz' )
862	IF ( av == 0 ) THEN
863	IF ( time_since_reference_point >= particle_advection_start ) THEN
864	DO i = nxl, nxr
865	DO j = nys, nyn
866	DO k = nzb, nzt+1
867	number_of_particles = prt_count(k,j,i)
868	IF (number_of_particles <= 0) CYCLE
869	particles => grid_particles(k,j,i)%particles(1:number_of_particles)
870	DO n = 1, number_of_particles
871	IF ( particles(n)%particle_mask ) THEN
872	tend(k,j,i) = tend(k,j,i) + &
873	particles(n)%weight_factor / &
874	prt_count(k,j,i)
875	ENDIF
876	ENDDO
877	ENDDO
878	ENDDO
879	ENDDO
880	! CALL exchange_horiz( tend, nbgp )
881	ELSE
882	tend = 0.0_wp
883	ENDIF
884	DO i = nxl, nxr
885	DO j = nys, nyn
886	DO k = nzb, nzt+1
887	local_pf(i,j,k) = tend(k,j,i)
888	ENDDO
889	ENDDO
890	ENDDO
891	resorted = .TRUE.
892	ELSE
893	IF ( .NOT. ALLOCATED( ql_vp_av ) ) THEN
894	ALLOCATE( ql_vp_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
895	ql_vp_av = REAL( fill_value, KIND = wp )
896	ENDIF
897	! CALL exchange_horiz( ql_vp_av, nbgp )
898	to_be_resorted => ql_vp_av
899	ENDIF
900	IF ( mode == 'xy' ) level_z = zu
901
902	CASE ( 'qsws*_xy' ) ! 2d-array
903	IF ( av == 0 ) THEN
904	local_pf(:,:,nzb+1) = REAL( fill_value, KIND = wp )
905	!
906	!-- In case of default surfaces, clean-up flux by density.
907	!-- In case of land-surfaces, convert fluxes into
908	!-- dynamic units
909	DO m = 1, surf_def_h(0)%ns
910	i = surf_def_h(0)%i(m)
911	j = surf_def_h(0)%j(m)
912	k = surf_def_h(0)%k(m)
913	local_pf(i,j,nzb+1) = surf_def_h(0)%qsws(m) * &
914	waterflux_output_conversion(k)
915	ENDDO
916	DO m = 1, surf_lsm_h%ns
917	i = surf_lsm_h%i(m)
918	j = surf_lsm_h%j(m)
919	k = surf_lsm_h%k(m)
920	local_pf(i,j,nzb+1) = surf_lsm_h%qsws(m) * l_v
921	ENDDO
922	ELSE
923	IF ( .NOT. ALLOCATED( qsws_av ) ) THEN
924	ALLOCATE( qsws_av(nysg:nyng,nxlg:nxrg) )
925	qsws_av = REAL( fill_value, KIND = wp )
926	ENDIF
927	DO i = nxl, nxr
928	DO j = nys, nyn
929	local_pf(i,j,nzb+1) = qsws_av(j,i)
930	ENDDO
931	ENDDO
932	ENDIF
933	resorted = .TRUE.
934	two_d = .TRUE.
935	level_z(nzb+1) = zu(nzb+1)
936
937	CASE ( 'qv_xy', 'qv_xz', 'qv_yz' )
938	IF ( av == 0 ) THEN
939	DO i = nxl, nxr
940	DO j = nys, nyn
941	DO k = nzb, nzt+1
942	local_pf(i,j,k) = q(k,j,i) - ql(k,j,i)
943	ENDDO
944	ENDDO
945	ENDDO
946	resorted = .TRUE.
947	ELSE
948	IF ( .NOT. ALLOCATED( qv_av ) ) THEN
949	ALLOCATE( qv_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
950	qv_av = REAL( fill_value, KIND = wp )
951	ENDIF
952	to_be_resorted => qv_av
953	ENDIF
954	IF ( mode == 'xy' ) level_z = zu
955
956	CASE ( 'r_a*_xy' ) ! 2d-array
957	IF ( av == 0 ) THEN
958	DO m = 1, surf_lsm_h%ns
959	i = surf_lsm_h%i(m)
960	j = surf_lsm_h%j(m)
961	local_pf(i,j,nzb+1) = surf_lsm_h%r_a(m)
962	ENDDO
963
964	DO m = 1, surf_usm_h%ns
965	i = surf_usm_h%i(m)
966	j = surf_usm_h%j(m)
967	local_pf(i,j,nzb+1) = &
968	( surf_usm_h%frac(ind_veg_wall,m) * &
969	surf_usm_h%r_a(m) + &
970	surf_usm_h%frac(ind_pav_green,m) * &
971	surf_usm_h%r_a_green(m) + &
972	surf_usm_h%frac(ind_wat_win,m) * &
973	surf_usm_h%r_a_window(m) )
974	ENDDO
975	ELSE
976	IF ( .NOT. ALLOCATED( r_a_av ) ) THEN
977	ALLOCATE( r_a_av(nysg:nyng,nxlg:nxrg) )
978	r_a_av = REAL( fill_value, KIND = wp )
979	ENDIF
980	DO i = nxl, nxr
981	DO j = nys, nyn
982	local_pf(i,j,nzb+1) = r_a_av(j,i)
983	ENDDO
984	ENDDO
985	ENDIF
986	resorted = .TRUE.
987	two_d = .TRUE.
988	level_z(nzb+1) = zu(nzb+1)
989
990	CASE ( 's_xy', 's_xz', 's_yz' )
991	IF ( av == 0 ) THEN
992	to_be_resorted => s
993	ELSE
994	IF ( .NOT. ALLOCATED( s_av ) ) THEN
995	ALLOCATE( s_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
996	s_av = REAL( fill_value, KIND = wp )
997	ENDIF
998	to_be_resorted => s_av
999	ENDIF
1000
1001	CASE ( 'shf*_xy' ) ! 2d-array
1002	IF ( av == 0 ) THEN
1003	!
1004	!-- In case of default surfaces, clean-up flux by density.
1005	!-- In case of land- and urban-surfaces, convert fluxes into
1006	!-- dynamic units.
1007	DO m = 1, surf_def_h(0)%ns
1008	i = surf_def_h(0)%i(m)
1009	j = surf_def_h(0)%j(m)
1010	k = surf_def_h(0)%k(m)
1011	local_pf(i,j,nzb+1) = surf_def_h(0)%shf(m) * &
1012	heatflux_output_conversion(k)
1013	ENDDO
1014	DO m = 1, surf_lsm_h%ns
1015	i = surf_lsm_h%i(m)
1016	j = surf_lsm_h%j(m)
1017	k = surf_lsm_h%k(m)
1018	local_pf(i,j,nzb+1) = surf_lsm_h%shf(m) * c_p
1019	ENDDO
1020	DO m = 1, surf_usm_h%ns
1021	i = surf_usm_h%i(m)
1022	j = surf_usm_h%j(m)
1023	k = surf_usm_h%k(m)
1024	local_pf(i,j,nzb+1) = surf_usm_h%shf(m) * c_p
1025	ENDDO
1026	ELSE
1027	IF ( .NOT. ALLOCATED( shf_av ) ) THEN
1028	ALLOCATE( shf_av(nysg:nyng,nxlg:nxrg) )
1029	shf_av = REAL( fill_value, KIND = wp )
1030	ENDIF
1031	DO i = nxl, nxr
1032	DO j = nys, nyn
1033	local_pf(i,j,nzb+1) = shf_av(j,i)
1034	ENDDO
1035	ENDDO
1036	ENDIF
1037	resorted = .TRUE.
1038	two_d = .TRUE.
1039	level_z(nzb+1) = zu(nzb+1)
1040
1041	CASE ( 'ssws*_xy' ) ! 2d-array
1042	IF ( av == 0 ) THEN
1043	DO m = 1, surf_def_h(0)%ns
1044	i = surf_def_h(0)%i(m)
1045	j = surf_def_h(0)%j(m)
1046	local_pf(i,j,nzb+1) = surf_def_h(0)%ssws(m)
1047	ENDDO
1048	DO m = 1, surf_lsm_h%ns
1049	i = surf_lsm_h%i(m)
1050	j = surf_lsm_h%j(m)
1051	local_pf(i,j,nzb+1) = surf_lsm_h%ssws(m)
1052	ENDDO
1053	DO m = 1, surf_usm_h%ns
1054	i = surf_usm_h%i(m)
1055	j = surf_usm_h%j(m)
1056	local_pf(i,j,nzb+1) = surf_usm_h%ssws(m)
1057	ENDDO
1058	ELSE
1059	IF ( .NOT. ALLOCATED( ssws_av ) ) THEN
1060	ALLOCATE( ssws_av(nysg:nyng,nxlg:nxrg) )
1061	ssws_av = REAL( fill_value, KIND = wp )
1062	ENDIF
1063	DO i = nxl, nxr
1064	DO j = nys, nyn
1065	local_pf(i,j,nzb+1) = ssws_av(j,i)
1066	ENDDO
1067	ENDDO
1068	ENDIF
1069	resorted = .TRUE.
1070	two_d = .TRUE.
1071	level_z(nzb+1) = zu(nzb+1)
1072
1073	CASE ( 't*_xy' ) ! 2d-array
1074	IF ( av == 0 ) THEN
1075	DO m = 1, surf_def_h(0)%ns
1076	i = surf_def_h(0)%i(m)
1077	j = surf_def_h(0)%j(m)
1078	local_pf(i,j,nzb+1) = surf_def_h(0)%ts(m)
1079	ENDDO
1080	DO m = 1, surf_lsm_h%ns
1081	i = surf_lsm_h%i(m)
1082	j = surf_lsm_h%j(m)
1083	local_pf(i,j,nzb+1) = surf_lsm_h%ts(m)
1084	ENDDO
1085	DO m = 1, surf_usm_h%ns
1086	i = surf_usm_h%i(m)
1087	j = surf_usm_h%j(m)
1088	local_pf(i,j,nzb+1) = surf_usm_h%ts(m)
1089	ENDDO
1090	ELSE
1091	IF ( .NOT. ALLOCATED( ts_av ) ) THEN
1092	ALLOCATE( ts_av(nysg:nyng,nxlg:nxrg) )
1093	ts_av = REAL( fill_value, KIND = wp )
1094	ENDIF
1095	DO i = nxl, nxr
1096	DO j = nys, nyn
1097	local_pf(i,j,nzb+1) = ts_av(j,i)
1098	ENDDO
1099	ENDDO
1100	ENDIF
1101	resorted = .TRUE.
1102	two_d = .TRUE.
1103	level_z(nzb+1) = zu(nzb+1)
1104
1105	CASE ( 'tsurf*_xy' ) ! 2d-array
1106	IF ( av == 0 ) THEN
1107	DO m = 1, surf_def_h(0)%ns
1108	i = surf_def_h(0)%i(m)
1109	j = surf_def_h(0)%j(m)
1110	local_pf(i,j,nzb+1) = surf_def_h(0)%pt_surface(m)
1111	ENDDO
1112
1113	DO m = 1, surf_lsm_h%ns
1114	i = surf_lsm_h%i(m)
1115	j = surf_lsm_h%j(m)
1116	local_pf(i,j,nzb+1) = surf_lsm_h%pt_surface(m)
1117	ENDDO
1118
1119	DO m = 1, surf_usm_h%ns
1120	i = surf_usm_h%i(m)
1121	j = surf_usm_h%j(m)
1122	local_pf(i,j,nzb+1) = surf_usm_h%pt_surface(m)
1123	ENDDO
1124
1125	ELSE
1126	IF ( .NOT. ALLOCATED( tsurf_av ) ) THEN
1127	ALLOCATE( tsurf_av(nysg:nyng,nxlg:nxrg) )
1128	tsurf_av = REAL( fill_value, KIND = wp )
1129	ENDIF
1130	DO i = nxl, nxr
1131	DO j = nys, nyn
1132	local_pf(i,j,nzb+1) = tsurf_av(j,i)
1133	ENDDO
1134	ENDDO
1135	ENDIF
1136	resorted = .TRUE.
1137	two_d = .TRUE.
1138	level_z(nzb+1) = zu(nzb+1)
1139
1140	CASE ( 'u_xy', 'u_xz', 'u_yz' )
1141	flag_nr = 1
1142	IF ( av == 0 ) THEN
1143	to_be_resorted => u
1144	ELSE
1145	IF ( .NOT. ALLOCATED( u_av ) ) THEN
1146	ALLOCATE( u_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1147	u_av = REAL( fill_value, KIND = wp )
1148	ENDIF
1149	to_be_resorted => u_av
1150	ENDIF
1151	IF ( mode == 'xy' ) level_z = zu
1152	!
1153	!-- Substitute the values generated by "mirror" boundary condition
1154	!-- at the bottom boundary by the real surface values.
1155	IF ( do2d(av,ivar) == 'u_xz' .OR. do2d(av,ivar) == 'u_yz' ) THEN
1156	IF ( ibc_uv_b == 0 ) local_pf(:,:,nzb) = 0.0_wp
1157	ENDIF
1158
1159	CASE ( 'us*_xy' ) ! 2d-array
1160	IF ( av == 0 ) THEN
1161	DO m = 1, surf_def_h(0)%ns
1162	i = surf_def_h(0)%i(m)
1163	j = surf_def_h(0)%j(m)
1164	local_pf(i,j,nzb+1) = surf_def_h(0)%us(m)
1165	ENDDO
1166	DO m = 1, surf_lsm_h%ns
1167	i = surf_lsm_h%i(m)
1168	j = surf_lsm_h%j(m)
1169	local_pf(i,j,nzb+1) = surf_lsm_h%us(m)
1170	ENDDO
1171	DO m = 1, surf_usm_h%ns
1172	i = surf_usm_h%i(m)
1173	j = surf_usm_h%j(m)
1174	local_pf(i,j,nzb+1) = surf_usm_h%us(m)
1175	ENDDO
1176	ELSE
1177	IF ( .NOT. ALLOCATED( us_av ) ) THEN
1178	ALLOCATE( us_av(nysg:nyng,nxlg:nxrg) )
1179	us_av = REAL( fill_value, KIND = wp )
1180	ENDIF
1181	DO i = nxl, nxr
1182	DO j = nys, nyn
1183	local_pf(i,j,nzb+1) = us_av(j,i)
1184	ENDDO
1185	ENDDO
1186	ENDIF
1187	resorted = .TRUE.
1188	two_d = .TRUE.
1189	level_z(nzb+1) = zu(nzb+1)
1190
1191	CASE ( 'v_xy', 'v_xz', 'v_yz' )
1192	flag_nr = 2
1193	IF ( av == 0 ) THEN
1194	to_be_resorted => v
1195	ELSE
1196	IF ( .NOT. ALLOCATED( v_av ) ) THEN
1197	ALLOCATE( v_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1198	v_av = REAL( fill_value, KIND = wp )
1199	ENDIF
1200	to_be_resorted => v_av
1201	ENDIF
1202	IF ( mode == 'xy' ) level_z = zu
1203	!
1204	!-- Substitute the values generated by "mirror" boundary condition
1205	!-- at the bottom boundary by the real surface values.
1206	IF ( do2d(av,ivar) == 'v_xz' .OR. do2d(av,ivar) == 'v_yz' ) THEN
1207	IF ( ibc_uv_b == 0 ) local_pf(:,:,nzb) = 0.0_wp
1208	ENDIF
1209
1210	CASE ( 'thetav_xy', 'thetav_xz', 'thetav_yz' )
1211	IF ( av == 0 ) THEN
1212	to_be_resorted => vpt
1213	ELSE
1214	IF ( .NOT. ALLOCATED( vpt_av ) ) THEN
1215	ALLOCATE( vpt_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1216	vpt_av = REAL( fill_value, KIND = wp )
1217	ENDIF
1218	to_be_resorted => vpt_av
1219	ENDIF
1220	IF ( mode == 'xy' ) level_z = zu
1221
1222	CASE ( 'theta_2m*_xy' ) ! 2d-array
1223	IF ( av == 0 ) THEN
1224	DO m = 1, surf_def_h(0)%ns
1225	i = surf_def_h(0)%i(m)
1226	j = surf_def_h(0)%j(m)
1227	local_pf(i,j,nzb+1) = surf_def_h(0)%pt_2m(m)
1228	ENDDO
1229	DO m = 1, surf_lsm_h%ns
1230	i = surf_lsm_h%i(m)
1231	j = surf_lsm_h%j(m)
1232	local_pf(i,j,nzb+1) = surf_lsm_h%pt_2m(m)
1233	ENDDO
1234	DO m = 1, surf_usm_h%ns
1235	i = surf_usm_h%i(m)
1236	j = surf_usm_h%j(m)
1237	local_pf(i,j,nzb+1) = surf_usm_h%pt_2m(m)
1238	ENDDO
1239	ELSE
1240	IF ( .NOT. ALLOCATED( pt_2m_av ) ) THEN
1241	ALLOCATE( pt_2m_av(nysg:nyng,nxlg:nxrg) )
1242	pt_2m_av = REAL( fill_value, KIND = wp )
1243	ENDIF
1244	DO i = nxl, nxr
1245	DO j = nys, nyn
1246	local_pf(i,j,nzb+1) = pt_2m_av(j,i)
1247	ENDDO
1248	ENDDO
1249	ENDIF
1250	resorted = .TRUE.
1251	two_d = .TRUE.
1252	level_z(nzb+1) = zu(nzb+1)
1253
1254
1255
1256	CASE ( 'w_xy', 'w_xz', 'w_yz' )
1257	flag_nr = 3
1258	IF ( av == 0 ) THEN
1259	to_be_resorted => w
1260	ELSE
1261	IF ( .NOT. ALLOCATED( w_av ) ) THEN
1262	ALLOCATE( w_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1263	w_av = REAL( fill_value, KIND = wp )
1264	ENDIF
1265	to_be_resorted => w_av
1266	ENDIF
1267	IF ( mode == 'xy' ) level_z = zw
1268
1269	CASE ( 'z0*_xy' ) ! 2d-array
1270	IF ( av == 0 ) THEN
1271	DO m = 1, surf_def_h(0)%ns
1272	i = surf_def_h(0)%i(m)
1273	j = surf_def_h(0)%j(m)
1274	local_pf(i,j,nzb+1) = surf_def_h(0)%z0(m)
1275	ENDDO
1276	DO m = 1, surf_lsm_h%ns
1277	i = surf_lsm_h%i(m)
1278	j = surf_lsm_h%j(m)
1279	local_pf(i,j,nzb+1) = surf_lsm_h%z0(m)
1280	ENDDO
1281	DO m = 1, surf_usm_h%ns
1282	i = surf_usm_h%i(m)
1283	j = surf_usm_h%j(m)
1284	local_pf(i,j,nzb+1) = surf_usm_h%z0(m)
1285	ENDDO
1286	ELSE
1287	IF ( .NOT. ALLOCATED( z0_av ) ) THEN
1288	ALLOCATE( z0_av(nysg:nyng,nxlg:nxrg) )
1289	z0_av = REAL( fill_value, KIND = wp )
1290	ENDIF
1291	DO i = nxl, nxr
1292	DO j = nys, nyn
1293	local_pf(i,j,nzb+1) = z0_av(j,i)
1294	ENDDO
1295	ENDDO
1296	ENDIF
1297	resorted = .TRUE.
1298	two_d = .TRUE.
1299	level_z(nzb+1) = zu(nzb+1)
1300
1301	CASE ( 'z0h*_xy' ) ! 2d-array
1302	IF ( av == 0 ) THEN
1303	DO m = 1, surf_def_h(0)%ns
1304	i = surf_def_h(0)%i(m)
1305	j = surf_def_h(0)%j(m)
1306	local_pf(i,j,nzb+1) = surf_def_h(0)%z0h(m)
1307	ENDDO
1308	DO m = 1, surf_lsm_h%ns
1309	i = surf_lsm_h%i(m)
1310	j = surf_lsm_h%j(m)
1311	local_pf(i,j,nzb+1) = surf_lsm_h%z0h(m)
1312	ENDDO
1313	DO m = 1, surf_usm_h%ns
1314	i = surf_usm_h%i(m)
1315	j = surf_usm_h%j(m)
1316	local_pf(i,j,nzb+1) = surf_usm_h%z0h(m)
1317	ENDDO
1318	ELSE
1319	IF ( .NOT. ALLOCATED( z0h_av ) ) THEN
1320	ALLOCATE( z0h_av(nysg:nyng,nxlg:nxrg) )
1321	z0h_av = REAL( fill_value, KIND = wp )
1322	ENDIF
1323	DO i = nxl, nxr
1324	DO j = nys, nyn
1325	local_pf(i,j,nzb+1) = z0h_av(j,i)
1326	ENDDO
1327	ENDDO
1328	ENDIF
1329	resorted = .TRUE.
1330	two_d = .TRUE.
1331	level_z(nzb+1) = zu(nzb+1)
1332
1333	CASE ( 'z0q*_xy' ) ! 2d-array
1334	IF ( av == 0 ) THEN
1335	DO m = 1, surf_def_h(0)%ns
1336	i = surf_def_h(0)%i(m)
1337	j = surf_def_h(0)%j(m)
1338	local_pf(i,j,nzb+1) = surf_def_h(0)%z0q(m)
1339	ENDDO
1340	DO m = 1, surf_lsm_h%ns
1341	i = surf_lsm_h%i(m)
1342	j = surf_lsm_h%j(m)
1343	local_pf(i,j,nzb+1) = surf_lsm_h%z0q(m)
1344	ENDDO
1345	DO m = 1, surf_usm_h%ns
1346	i = surf_usm_h%i(m)
1347	j = surf_usm_h%j(m)
1348	local_pf(i,j,nzb+1) = surf_usm_h%z0q(m)
1349	ENDDO
1350	ELSE
1351	IF ( .NOT. ALLOCATED( z0q_av ) ) THEN
1352	ALLOCATE( z0q_av(nysg:nyng,nxlg:nxrg) )
1353	z0q_av = REAL( fill_value, KIND = wp )
1354	ENDIF
1355	DO i = nxl, nxr
1356	DO j = nys, nyn
1357	local_pf(i,j,nzb+1) = z0q_av(j,i)
1358	ENDDO
1359	ENDDO
1360	ENDIF
1361	resorted = .TRUE.
1362	two_d = .TRUE.
1363	level_z(nzb+1) = zu(nzb+1)
1364
1365	CASE DEFAULT
1366
1367	!
1368	!-- Quantities of other modules
1369	IF ( .NOT. found ) THEN
1370	CALL module_interface_data_output_2d( &
1371	av, do2d(av,ivar), found, grid, mode, &
1372	local_pf, two_d, nzb_do, nzt_do, &
1373	fill_value &
1374	)
1375	ENDIF
1376
1377	resorted = .TRUE.
1378
1379	IF ( grid == 'zu' ) THEN
1380	IF ( mode == 'xy' ) level_z = zu
1381	ELSEIF ( grid == 'zw' ) THEN
1382	IF ( mode == 'xy' ) level_z = zw
1383	ELSEIF ( grid == 'zu1' ) THEN
1384	IF ( mode == 'xy' ) level_z(nzb+1) = zu(nzb+1)
1385	ELSEIF ( grid == 'zs' ) THEN
1386	IF ( mode == 'xy' ) level_z = zs
1387	ENDIF
1388
1389	IF ( .NOT. found ) THEN
1390	message_string = 'no output provided for: ' // &
1391	TRIM( do2d(av,ivar) )
1392	CALL message( 'data_output_2d', 'PA0181', 0, 0, 0, 6, 0 )
1393	ENDIF
1394
1395	END SELECT
1396
1397	!
1398	!-- Resort the array to be output, if not done above. Flag topography
1399	!-- grid points with fill values, using the corresponding maksing flag.
1400	IF ( .NOT. resorted ) THEN
1401	DO i = nxl, nxr
1402	DO j = nys, nyn
1403	DO k = nzb_do, nzt_do
1404	local_pf(i,j,k) = MERGE( to_be_resorted(k,j,i), &
1405	REAL( fill_value, KIND = wp ), &
1406	BTEST( wall_flags_0(k,j,i), &
1407	flag_nr ) )
1408	ENDDO
1409	ENDDO
1410	ENDDO
1411	ENDIF
1412
1413	!
1414	!-- Output of the individual cross-sections, depending on the cross-
1415	!-- section mode chosen.
1416	is = 1
1417	loop1: DO WHILE ( section(is,s_ind) /= -9999 .OR. two_d )
1418
1419	SELECT CASE ( mode )
1420
1421	CASE ( 'xy' )
1422	!
1423	!-- Determine the cross section index
1424	IF ( two_d ) THEN
1425	layer_xy = nzb+1
1426	ELSE
1427	layer_xy = section(is,s_ind)
1428	ENDIF
1429
1430	!
1431	!-- Exit the loop for layers beyond the data output domain
1432	!-- (used for soil model)
1433	IF ( layer_xy > nzt_do ) THEN
1434	EXIT loop1
1435	ENDIF
1436
1437	!
1438	!-- Update the netCDF xy cross section time axis.
1439	!-- In case of parallel output, this is only done by PE0
1440	!-- to increase the performance.
1441	IF ( time_since_reference_point /= do2d_xy_last_time(av) ) THEN
1442	do2d_xy_time_count(av) = do2d_xy_time_count(av) + 1
1443	do2d_xy_last_time(av) = time_since_reference_point
1444	IF ( myid == 0 ) THEN
1445	IF ( .NOT. data_output_2d_on_each_pe &
1446	.OR. netcdf_data_format > 4 ) &
1447	THEN
1448	#if defined( __netcdf )
1449	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1450	id_var_time_xy(av), &
1451	(/ time_since_reference_point /), &
1452	start = (/ do2d_xy_time_count(av) /), &
1453	count = (/ 1 /) )
1454	CALL netcdf_handle_error( 'data_output_2d', 53 )
1455	#endif
1456	ENDIF
1457	ENDIF
1458	ENDIF
1459	!
1460	!-- If required, carry out averaging along z
1461	IF ( section(is,s_ind) == -1 .AND. .NOT. two_d ) THEN
1462
1463	local_2d = 0.0_wp
1464	!
1465	!-- Carry out the averaging (all data are on the PE)
1466	DO k = nzb_do, nzt_do
1467	DO j = nys, nyn
1468	DO i = nxl, nxr
1469	local_2d(i,j) = local_2d(i,j) + local_pf(i,j,k)
1470	ENDDO
1471	ENDDO
1472	ENDDO
1473
1474	local_2d = local_2d / ( nzt_do - nzb_do + 1.0_wp)
1475
1476	ELSE
1477	!
1478	!-- Just store the respective section on the local array
1479	local_2d = local_pf(:,:,layer_xy)
1480
1481	ENDIF
1482
1483	#if defined( __parallel )
1484	IF ( netcdf_data_format > 4 ) THEN
1485	!
1486	!-- Parallel output in netCDF4/HDF5 format.
1487	IF ( two_d ) THEN
1488	iis = 1
1489	ELSE
1490	iis = is
1491	ENDIF
1492
1493	#if defined( __netcdf )
1494	!
1495	!-- For parallel output, all cross sections are first stored
1496	!-- here on a local array and will be written to the output
1497	!-- file afterwards to increase the performance.
1498	DO i = nxl, nxr
1499	DO j = nys, nyn
1500	local_2d_sections(i,j,iis) = local_2d(i,j)
1501	ENDDO
1502	ENDDO
1503	#endif
1504	ELSE
1505
1506	IF ( data_output_2d_on_each_pe ) THEN
1507	!
1508	!-- Output of partial arrays on each PE
1509	#if defined( __netcdf )
1510	IF ( myid == 0 ) THEN
1511	WRITE ( 21 ) time_since_reference_point, &
1512	do2d_xy_time_count(av), av
1513	ENDIF
1514	#endif
1515	DO i = 0, io_blocks-1
1516	IF ( i == io_group ) THEN
1517	WRITE ( 21 ) nxl, nxr, nys, nyn, nys, nyn
1518	WRITE ( 21 ) local_2d
1519	ENDIF
1520	#if defined( __parallel )
1521	CALL MPI_BARRIER( comm2d, ierr )
1522	#endif
1523	ENDDO
1524
1525	ELSE
1526	!
1527	!-- PE0 receives partial arrays from all processors and
1528	!-- then outputs them. Here a barrier has to be set,
1529	!-- because otherwise "-MPI- FATAL: Remote protocol queue
1530	!-- full" may occur.
1531	CALL MPI_BARRIER( comm2d, ierr )
1532
1533	ngp = ( nxr-nxl+1 ) * ( nyn-nys+1 )
1534	IF ( myid == 0 ) THEN
1535	!
1536	!-- Local array can be relocated directly.
1537	total_2d(nxl:nxr,nys:nyn) = local_2d
1538	!
1539	!-- Receive data from all other PEs.
1540	DO n = 1, numprocs-1
1541	!
1542	!-- Receive index limits first, then array.
1543	!-- Index limits are received in arbitrary order from
1544	!-- the PEs.
1545	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
1546	MPI_ANY_SOURCE, 0, comm2d, &
1547	status, ierr )
1548	sender = status(MPI_SOURCE)
1549	DEALLOCATE( local_2d )
1550	ALLOCATE( local_2d(ind(1):ind(2),ind(3):ind(4)) )
1551	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
1552	MPI_REAL, sender, 1, comm2d, &
1553	status, ierr )
1554	total_2d(ind(1):ind(2),ind(3):ind(4)) = local_2d
1555	ENDDO
1556	!
1557	!-- Relocate the local array for the next loop increment
1558	DEALLOCATE( local_2d )
1559	ALLOCATE( local_2d(nxl:nxr,nys:nyn) )
1560
1561	#if defined( __netcdf )
1562	IF ( two_d ) THEN
1563	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1564	id_var_do2d(av,ivar), &
1565	total_2d(0:nx,0:ny), &
1566	start = (/ 1, 1, 1, do2d_xy_time_count(av) /), &
1567	count = (/ nx+1, ny+1, 1, 1 /) )
1568	ELSE
1569	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1570	id_var_do2d(av,ivar), &
1571	total_2d(0:nx,0:ny), &
1572	start = (/ 1, 1, is, do2d_xy_time_count(av) /), &
1573	count = (/ nx+1, ny+1, 1, 1 /) )
1574	ENDIF
1575	CALL netcdf_handle_error( 'data_output_2d', 54 )
1576	#endif
1577
1578	ELSE
1579	!
1580	!-- First send the local index limits to PE0
1581	ind(1) = nxl; ind(2) = nxr
1582	ind(3) = nys; ind(4) = nyn
1583	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
1584	comm2d, ierr )
1585	!
1586	!-- Send data to PE0
1587	CALL MPI_SEND( local_2d(nxl,nys), ngp, &
1588	MPI_REAL, 0, 1, comm2d, ierr )
1589	ENDIF
1590	!
1591	!-- A barrier has to be set, because otherwise some PEs may
1592	!-- proceed too fast so that PE0 may receive wrong data on
1593	!-- tag 0
1594	CALL MPI_BARRIER( comm2d, ierr )
1595	ENDIF
1596
1597	ENDIF
1598	#else
1599	#if defined( __netcdf )
1600	IF ( two_d ) THEN
1601	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1602	id_var_do2d(av,ivar), &
1603	local_2d(nxl:nxr,nys:nyn), &
1604	start = (/ 1, 1, 1, do2d_xy_time_count(av) /), &
1605	count = (/ nx+1, ny+1, 1, 1 /) )
1606	ELSE
1607	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1608	id_var_do2d(av,ivar), &
1609	local_2d(nxl:nxr,nys:nyn), &
1610	start = (/ 1, 1, is, do2d_xy_time_count(av) /), &
1611	count = (/ nx+1, ny+1, 1, 1 /) )
1612	ENDIF
1613	CALL netcdf_handle_error( 'data_output_2d', 447 )
1614	#endif
1615	#endif
1616
1617	!
1618	!-- For 2D-arrays (e.g. u*) only one cross-section is available.
1619	!-- Hence exit loop of output levels.
1620	IF ( two_d ) THEN
1621	IF ( netcdf_data_format < 5 ) two_d = .FALSE.
1622	EXIT loop1
1623	ENDIF
1624
1625	CASE ( 'xz' )
1626	!
1627	!-- Update the netCDF xz cross section time axis.
1628	!-- In case of parallel output, this is only done by PE0
1629	!-- to increase the performance.
1630	IF ( time_since_reference_point /= do2d_xz_last_time(av) ) THEN
1631	do2d_xz_time_count(av) = do2d_xz_time_count(av) + 1
1632	do2d_xz_last_time(av) = time_since_reference_point
1633	IF ( myid == 0 ) THEN
1634	IF ( .NOT. data_output_2d_on_each_pe &
1635	.OR. netcdf_data_format > 4 ) &
1636	THEN
1637	#if defined( __netcdf )
1638	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
1639	id_var_time_xz(av), &
1640	(/ time_since_reference_point /), &
1641	start = (/ do2d_xz_time_count(av) /), &
1642	count = (/ 1 /) )
1643	CALL netcdf_handle_error( 'data_output_2d', 56 )
1644	#endif
1645	ENDIF
1646	ENDIF
1647	ENDIF
1648
1649	!
1650	!-- If required, carry out averaging along y
1651	IF ( section(is,s_ind) == -1 ) THEN
1652
1653	ALLOCATE( local_2d_l(nxl:nxr,nzb_do:nzt_do) )
1654	local_2d_l = 0.0_wp
1655	ngp = ( nxr-nxl + 1 ) * ( nzt_do-nzb_do + 1 )
1656	!
1657	!-- First local averaging on the PE
1658	DO k = nzb_do, nzt_do
1659	DO j = nys, nyn
1660	DO i = nxl, nxr
1661	local_2d_l(i,k) = local_2d_l(i,k) + &
1662	local_pf(i,j,k)
1663	ENDDO
1664	ENDDO
1665	ENDDO
1666	#if defined( __parallel )
1667	!
1668	!-- Now do the averaging over all PEs along y
1669	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
1670	CALL MPI_ALLREDUCE( local_2d_l(nxl,nzb_do), &
1671	local_2d(nxl,nzb_do), ngp, MPI_REAL, &
1672	MPI_SUM, comm1dy, ierr )
1673	#else
1674	local_2d = local_2d_l
1675	#endif
1676	local_2d = local_2d / ( ny + 1.0_wp )
1677
1678	DEALLOCATE( local_2d_l )
1679
1680	ELSE
1681	!
1682	!-- Just store the respective section on the local array
1683	!-- (but only if it is available on this PE!)
1684	IF ( section(is,s_ind) >= nys .AND. section(is,s_ind) <= nyn ) &
1685	THEN
1686	local_2d = local_pf(:,section(is,s_ind),nzb_do:nzt_do)
1687	ENDIF
1688
1689	ENDIF
1690
1691	#if defined( __parallel )
1692	IF ( netcdf_data_format > 4 ) THEN
1693	!
1694	!-- Output in netCDF4/HDF5 format.
1695	!-- Output only on those PEs where the respective cross
1696	!-- sections reside. Cross sections averaged along y are
1697	!-- output on the respective first PE along y (myidy=0).
1698	IF ( ( section(is,s_ind) >= nys .AND. &
1699	section(is,s_ind) <= nyn ) .OR. &
1700	( section(is,s_ind) == -1 .AND. myidy == 0 ) ) THEN
1701	#if defined( __netcdf )
1702	!
1703	!-- For parallel output, all cross sections are first
1704	!-- stored here on a local array and will be written to the
1705	!-- output file afterwards to increase the performance.
1706	DO i = nxl, nxr
1707	DO k = nzb_do, nzt_do
1708	local_2d_sections_l(i,is,k) = local_2d(i,k)
1709	ENDDO
1710	ENDDO
1711	#endif
1712	ENDIF
1713
1714	ELSE
1715
1716	IF ( data_output_2d_on_each_pe ) THEN
1717	!
1718	!-- Output of partial arrays on each PE. If the cross
1719	!-- section does not reside on the PE, output special
1720	!-- index values.
1721	#if defined( __netcdf )
1722	IF ( myid == 0 ) THEN
1723	WRITE ( 22 ) time_since_reference_point, &
1724	do2d_xz_time_count(av), av
1725	ENDIF
1726	#endif
1727	DO i = 0, io_blocks-1
1728	IF ( i == io_group ) THEN
1729	IF ( ( section(is,s_ind) >= nys .AND. &
1730	section(is,s_ind) <= nyn ) .OR. &
1731	( section(is,s_ind) == -1 .AND. &
1732	nys-1 == -1 ) ) &
1733	THEN
1734	WRITE (22) nxl, nxr, nzb_do, nzt_do, nzb, nzt+1
1735	WRITE (22) local_2d
1736	ELSE
1737	WRITE (22) -1, -1, -1, -1, -1, -1
1738	ENDIF
1739	ENDIF
1740	#if defined( __parallel )
1741	CALL MPI_BARRIER( comm2d, ierr )
1742	#endif
1743	ENDDO
1744
1745	ELSE
1746	!
1747	!-- PE0 receives partial arrays from all processors of the
1748	!-- respective cross section and outputs them. Here a
1749	!-- barrier has to be set, because otherwise
1750	!-- "-MPI- FATAL: Remote protocol queue full" may occur.
1751	CALL MPI_BARRIER( comm2d, ierr )
1752
1753	ngp = ( nxr-nxl + 1 ) * ( nzt_do-nzb_do + 1 )
1754	IF ( myid == 0 ) THEN
1755	!
1756	!-- Local array can be relocated directly.
1757	IF ( ( section(is,s_ind) >= nys .AND. &
1758	section(is,s_ind) <= nyn ) .OR. &
1759	( section(is,s_ind) == -1 .AND. &
1760	nys-1 == -1 ) ) THEN
1761	total_2d(nxl:nxr,nzb_do:nzt_do) = local_2d
1762	ENDIF
1763	!
1764	!-- Receive data from all other PEs.
1765	DO n = 1, numprocs-1
1766	!
1767	!-- Receive index limits first, then array.
1768	!-- Index limits are received in arbitrary order from
1769	!-- the PEs.
1770	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
1771	MPI_ANY_SOURCE, 0, comm2d, &
1772	status, ierr )
1773	!
1774	!-- Not all PEs have data for XZ-cross-section.
1775	IF ( ind(1) /= -9999 ) THEN
1776	sender = status(MPI_SOURCE)
1777	DEALLOCATE( local_2d )
1778	ALLOCATE( local_2d(ind(1):ind(2), &
1779	ind(3):ind(4)) )
1780	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
1781	MPI_REAL, sender, 1, comm2d, &
1782	status, ierr )
1783	total_2d(ind(1):ind(2),ind(3):ind(4)) = &
1784	local_2d
1785	ENDIF
1786	ENDDO
1787	!
1788	!-- Relocate the local array for the next loop increment
1789	DEALLOCATE( local_2d )
1790	ALLOCATE( local_2d(nxl:nxr,nzb_do:nzt_do) )
1791
1792	#if defined( __netcdf )
1793	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
1794	id_var_do2d(av,ivar), &
1795	total_2d(0:nx,nzb_do:nzt_do), &
1796	start = (/ 1, is, 1, do2d_xz_time_count(av) /), &
1797	count = (/ nx+1, 1, nzt_do-nzb_do+1, 1 /) )
1798	CALL netcdf_handle_error( 'data_output_2d', 58 )
1799	#endif
1800
1801	ELSE
1802	!
1803	!-- If the cross section resides on the PE, send the
1804	!-- local index limits, otherwise send -9999 to PE0.
1805	IF ( ( section(is,s_ind) >= nys .AND. &
1806	section(is,s_ind) <= nyn ) .OR. &
1807	( section(is,s_ind) == -1 .AND. nys-1 == -1 ) ) &
1808	THEN
1809	ind(1) = nxl; ind(2) = nxr
1810	ind(3) = nzb_do; ind(4) = nzt_do
1811	ELSE
1812	ind(1) = -9999; ind(2) = -9999
1813	ind(3) = -9999; ind(4) = -9999
1814	ENDIF
1815	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
1816	comm2d, ierr )
1817	!
1818	!-- If applicable, send data to PE0.
1819	IF ( ind(1) /= -9999 ) THEN
1820	CALL MPI_SEND( local_2d(nxl,nzb_do), ngp, &
1821	MPI_REAL, 0, 1, comm2d, ierr )
1822	ENDIF
1823	ENDIF
1824	!
1825	!-- A barrier has to be set, because otherwise some PEs may
1826	!-- proceed too fast so that PE0 may receive wrong data on
1827	!-- tag 0
1828	CALL MPI_BARRIER( comm2d, ierr )
1829	ENDIF
1830
1831	ENDIF
1832	#else
1833	#if defined( __netcdf )
1834	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
1835	id_var_do2d(av,ivar), &
1836	local_2d(nxl:nxr,nzb_do:nzt_do), &
1837	start = (/ 1, is, 1, do2d_xz_time_count(av) /), &
1838	count = (/ nx+1, 1, nzt_do-nzb_do+1, 1 /) )
1839	CALL netcdf_handle_error( 'data_output_2d', 451 )
1840	#endif
1841	#endif
1842
1843	CASE ( 'yz' )
1844	!
1845	!-- Update the netCDF yz cross section time axis.
1846	!-- In case of parallel output, this is only done by PE0
1847	!-- to increase the performance.
1848	IF ( time_since_reference_point /= do2d_yz_last_time(av) ) THEN
1849	do2d_yz_time_count(av) = do2d_yz_time_count(av) + 1
1850	do2d_yz_last_time(av) = time_since_reference_point
1851	IF ( myid == 0 ) THEN
1852	IF ( .NOT. data_output_2d_on_each_pe &
1853	.OR. netcdf_data_format > 4 ) &
1854	THEN
1855	#if defined( __netcdf )
1856	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
1857	id_var_time_yz(av), &
1858	(/ time_since_reference_point /), &
1859	start = (/ do2d_yz_time_count(av) /), &
1860	count = (/ 1 /) )
1861	CALL netcdf_handle_error( 'data_output_2d', 59 )
1862	#endif
1863	ENDIF
1864	ENDIF
1865	ENDIF
1866
1867	!
1868	!-- If required, carry out averaging along x
1869	IF ( section(is,s_ind) == -1 ) THEN
1870
1871	ALLOCATE( local_2d_l(nys:nyn,nzb_do:nzt_do) )
1872	local_2d_l = 0.0_wp
1873	ngp = ( nyn-nys+1 ) * ( nzt_do-nzb_do+1 )
1874	!
1875	!-- First local averaging on the PE
1876	DO k = nzb_do, nzt_do
1877	DO j = nys, nyn
1878	DO i = nxl, nxr
1879	local_2d_l(j,k) = local_2d_l(j,k) + &
1880	local_pf(i,j,k)
1881	ENDDO
1882	ENDDO
1883	ENDDO
1884	#if defined( __parallel )
1885	!
1886	!-- Now do the averaging over all PEs along x
1887	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
1888	CALL MPI_ALLREDUCE( local_2d_l(nys,nzb_do), &
1889	local_2d(nys,nzb_do), ngp, MPI_REAL, &
1890	MPI_SUM, comm1dx, ierr )
1891	#else
1892	local_2d = local_2d_l
1893	#endif
1894	local_2d = local_2d / ( nx + 1.0_wp )
1895
1896	DEALLOCATE( local_2d_l )
1897
1898	ELSE
1899	!
1900	!-- Just store the respective section on the local array
1901	!-- (but only if it is available on this PE!)
1902	IF ( section(is,s_ind) >= nxl .AND. section(is,s_ind) <= nxr ) &
1903	THEN
1904	local_2d = local_pf(section(is,s_ind),:,nzb_do:nzt_do)
1905	ENDIF
1906
1907	ENDIF
1908
1909	#if defined( __parallel )
1910	IF ( netcdf_data_format > 4 ) THEN
1911	!
1912	!-- Output in netCDF4/HDF5 format.
1913	!-- Output only on those PEs where the respective cross
1914	!-- sections reside. Cross sections averaged along x are
1915	!-- output on the respective first PE along x (myidx=0).
1916	IF ( ( section(is,s_ind) >= nxl .AND. &
1917	section(is,s_ind) <= nxr ) .OR. &
1918	( section(is,s_ind) == -1 .AND. myidx == 0 ) ) THEN
1919	#if defined( __netcdf )
1920	!
1921	!-- For parallel output, all cross sections are first
1922	!-- stored here on a local array and will be written to the
1923	!-- output file afterwards to increase the performance.
1924	DO j = nys, nyn
1925	DO k = nzb_do, nzt_do
1926	local_2d_sections_l(is,j,k) = local_2d(j,k)
1927	ENDDO
1928	ENDDO
1929	#endif
1930	ENDIF
1931
1932	ELSE
1933
1934	IF ( data_output_2d_on_each_pe ) THEN
1935	!
1936	!-- Output of partial arrays on each PE. If the cross
1937	!-- section does not reside on the PE, output special
1938	!-- index values.
1939	#if defined( __netcdf )
1940	IF ( myid == 0 ) THEN
1941	WRITE ( 23 ) time_since_reference_point, &
1942	do2d_yz_time_count(av), av
1943	ENDIF
1944	#endif
1945	DO i = 0, io_blocks-1
1946	IF ( i == io_group ) THEN
1947	IF ( ( section(is,s_ind) >= nxl .AND. &
1948	section(is,s_ind) <= nxr ) .OR. &
1949	( section(is,s_ind) == -1 .AND. &
1950	nxl-1 == -1 ) ) &
1951	THEN
1952	WRITE (23) nys, nyn, nzb_do, nzt_do, nzb, nzt+1
1953	WRITE (23) local_2d
1954	ELSE
1955	WRITE (23) -1, -1, -1, -1, -1, -1
1956	ENDIF
1957	ENDIF
1958	#if defined( __parallel )
1959	CALL MPI_BARRIER( comm2d, ierr )
1960	#endif
1961	ENDDO
1962
1963	ELSE
1964	!
1965	!-- PE0 receives partial arrays from all processors of the
1966	!-- respective cross section and outputs them. Here a
1967	!-- barrier has to be set, because otherwise
1968	!-- "-MPI- FATAL: Remote protocol queue full" may occur.
1969	CALL MPI_BARRIER( comm2d, ierr )
1970
1971	ngp = ( nyn-nys+1 ) * ( nzt_do-nzb_do+1 )
1972	IF ( myid == 0 ) THEN
1973	!
1974	!-- Local array can be relocated directly.
1975	IF ( ( section(is,s_ind) >= nxl .AND. &
1976	section(is,s_ind) <= nxr ) .OR. &
1977	( section(is,s_ind) == -1 .AND. nxl-1 == -1 ) ) &
1978	THEN
1979	total_2d(nys:nyn,nzb_do:nzt_do) = local_2d
1980	ENDIF
1981	!
1982	!-- Receive data from all other PEs.
1983	DO n = 1, numprocs-1
1984	!
1985	!-- Receive index limits first, then array.
1986	!-- Index limits are received in arbitrary order from
1987	!-- the PEs.
1988	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
1989	MPI_ANY_SOURCE, 0, comm2d, &
1990	status, ierr )
1991	!
1992	!-- Not all PEs have data for YZ-cross-section.
1993	IF ( ind(1) /= -9999 ) THEN
1994	sender = status(MPI_SOURCE)
1995	DEALLOCATE( local_2d )
1996	ALLOCATE( local_2d(ind(1):ind(2), &
1997	ind(3):ind(4)) )
1998	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
1999	MPI_REAL, sender, 1, comm2d, &
2000	status, ierr )
2001	total_2d(ind(1):ind(2),ind(3):ind(4)) = &
2002	local_2d
2003	ENDIF
2004	ENDDO
2005	!
2006	!-- Relocate the local array for the next loop increment
2007	DEALLOCATE( local_2d )
2008	ALLOCATE( local_2d(nys:nyn,nzb_do:nzt_do) )
2009
2010	#if defined( __netcdf )
2011	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2012	id_var_do2d(av,ivar), &
2013	total_2d(0:ny,nzb_do:nzt_do), &
2014	start = (/ is, 1, 1, do2d_yz_time_count(av) /), &
2015	count = (/ 1, ny+1, nzt_do-nzb_do+1, 1 /) )
2016	CALL netcdf_handle_error( 'data_output_2d', 61 )
2017	#endif
2018
2019	ELSE
2020	!
2021	!-- If the cross section resides on the PE, send the
2022	!-- local index limits, otherwise send -9999 to PE0.
2023	IF ( ( section(is,s_ind) >= nxl .AND. &
2024	section(is,s_ind) <= nxr ) .OR. &
2025	( section(is,s_ind) == -1 .AND. nxl-1 == -1 ) ) &
2026	THEN
2027	ind(1) = nys; ind(2) = nyn
2028	ind(3) = nzb_do; ind(4) = nzt_do
2029	ELSE
2030	ind(1) = -9999; ind(2) = -9999
2031	ind(3) = -9999; ind(4) = -9999
2032	ENDIF
2033	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
2034	comm2d, ierr )
2035	!
2036	!-- If applicable, send data to PE0.
2037	IF ( ind(1) /= -9999 ) THEN
2038	CALL MPI_SEND( local_2d(nys,nzb_do), ngp, &
2039	MPI_REAL, 0, 1, comm2d, ierr )
2040	ENDIF
2041	ENDIF
2042	!
2043	!-- A barrier has to be set, because otherwise some PEs may
2044	!-- proceed too fast so that PE0 may receive wrong data on
2045	!-- tag 0
2046	CALL MPI_BARRIER( comm2d, ierr )
2047	ENDIF
2048
2049	ENDIF
2050	#else
2051	#if defined( __netcdf )
2052	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2053	id_var_do2d(av,ivar), &
2054	local_2d(nys:nyn,nzb_do:nzt_do), &
2055	start = (/ is, 1, 1, do2d_xz_time_count(av) /), &
2056	count = (/ 1, ny+1, nzt_do-nzb_do+1, 1 /) )
2057	CALL netcdf_handle_error( 'data_output_2d', 452 )
2058	#endif
2059	#endif
2060
2061	END SELECT
2062
2063	is = is + 1
2064	ENDDO loop1
2065
2066	!
2067	!-- For parallel output, all data were collected before on a local array
2068	!-- and are written now to the netcdf file. This must be done to increase
2069	!-- the performance of the parallel output.
2070	#if defined( __netcdf )
2071	IF ( netcdf_data_format > 4 ) THEN
2072
2073	SELECT CASE ( mode )
2074
2075	CASE ( 'xy' )
2076	IF ( two_d ) THEN
2077	nis = 1
2078	two_d = .FALSE.
2079	ELSE
2080	nis = ns
2081	ENDIF
2082	!
2083	!-- Do not output redundant ghost point data except for the
2084	!-- boundaries of the total domain.
2085	! IF ( nxr == nx .AND. nyn /= ny ) THEN
2086	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2087	! id_var_do2d(av,ivar), &
2088	! local_2d_sections(nxl:nxr+1, &
2089	! nys:nyn,1:nis), &
2090	! start = (/ nxl+1, nys+1, 1, &
2091	! do2d_xy_time_count(av) /), &
2092	! count = (/ nxr-nxl+2, &
2093	! nyn-nys+1, nis, 1 &
2094	! /) )
2095	! ELSEIF ( nxr /= nx .AND. nyn == ny ) THEN
2096	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2097	! id_var_do2d(av,ivar), &
2098	! local_2d_sections(nxl:nxr, &
2099	! nys:nyn+1,1:nis), &
2100	! start = (/ nxl+1, nys+1, 1, &
2101	! do2d_xy_time_count(av) /), &
2102	! count = (/ nxr-nxl+1, &
2103	! nyn-nys+2, nis, 1 &
2104	! /) )
2105	! ELSEIF ( nxr == nx .AND. nyn == ny ) THEN
2106	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2107	! id_var_do2d(av,ivar), &
2108	! local_2d_sections(nxl:nxr+1, &
2109	! nys:nyn+1,1:nis), &
2110	! start = (/ nxl+1, nys+1, 1, &
2111	! do2d_xy_time_count(av) /), &
2112	! count = (/ nxr-nxl+2, &
2113	! nyn-nys+2, nis, 1 &
2114	! /) )
2115	! ELSE
2116	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2117	id_var_do2d(av,ivar), &
2118	local_2d_sections(nxl:nxr, &
2119	nys:nyn,1:nis), &
2120	start = (/ nxl+1, nys+1, 1, &
2121	do2d_xy_time_count(av) /), &
2122	count = (/ nxr-nxl+1, &
2123	nyn-nys+1, nis, 1 &
2124	/) )
2125	! ENDIF
2126
2127	CALL netcdf_handle_error( 'data_output_2d', 55 )
2128
2129	CASE ( 'xz' )
2130	!
2131	!-- First, all PEs get the information of all cross-sections.
2132	!-- Then the data are written to the output file by all PEs
2133	!-- while NF90_COLLECTIVE is set in subroutine
2134	!-- define_netcdf_header. Although redundant information are
2135	!-- written to the output file in that case, the performance
2136	!-- is significantly better compared to the case where only
2137	!-- the first row of PEs in x-direction (myidx = 0) is given
2138	!-- the output while NF90_INDEPENDENT is set.
2139	IF ( npey /= 1 ) THEN
2140
2141	#if defined( __parallel )
2142	!
2143	!-- Distribute data over all PEs along y
2144	ngp = ( nxr-nxl+1 ) * ( nzt_do-nzb_do+1 ) * ns
2145	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
2146	CALL MPI_ALLREDUCE( local_2d_sections_l(nxl,1,nzb_do), &
2147	local_2d_sections(nxl,1,nzb_do), &
2148	ngp, MPI_REAL, MPI_SUM, comm1dy, &
2149	ierr )
2150	#else
2151	local_2d_sections = local_2d_sections_l
2152	#endif
2153	ENDIF
2154	!
2155	!-- Do not output redundant ghost point data except for the
2156	!-- boundaries of the total domain.
2157	! IF ( nxr == nx ) THEN
2158	! nc_stat = NF90_PUT_VAR( id_set_xz(av), &
2159	! id_var_do2d(av,ivar), &
2160	! local_2d_sections(nxl:nxr+1,1:ns, &
2161	! nzb_do:nzt_do), &
2162	! start = (/ nxl+1, 1, 1, &
2163	! do2d_xz_time_count(av) /), &
2164	! count = (/ nxr-nxl+2, ns, nzt_do-nzb_do+1, &
2165	! 1 /) )
2166	! ELSE
2167	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
2168	id_var_do2d(av,ivar), &
2169	local_2d_sections(nxl:nxr,1:ns, &
2170	nzb_do:nzt_do), &
2171	start = (/ nxl+1, 1, 1, &
2172	do2d_xz_time_count(av) /), &
2173	count = (/ nxr-nxl+1, ns, nzt_do-nzb_do+1, &
2174	1 /) )
2175	! ENDIF
2176
2177	CALL netcdf_handle_error( 'data_output_2d', 57 )
2178
2179	CASE ( 'yz' )
2180	!
2181	!-- First, all PEs get the information of all cross-sections.
2182	!-- Then the data are written to the output file by all PEs
2183	!-- while NF90_COLLECTIVE is set in subroutine
2184	!-- define_netcdf_header. Although redundant information are
2185	!-- written to the output file in that case, the performance
2186	!-- is significantly better compared to the case where only
2187	!-- the first row of PEs in y-direction (myidy = 0) is given
2188	!-- the output while NF90_INDEPENDENT is set.
2189	IF ( npex /= 1 ) THEN
2190
2191	#if defined( __parallel )
2192	!
2193	!-- Distribute data over all PEs along x
2194	ngp = ( nyn-nys+1 ) * ( nzt-nzb + 2 ) * ns
2195	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
2196	CALL MPI_ALLREDUCE( local_2d_sections_l(1,nys,nzb_do), &
2197	local_2d_sections(1,nys,nzb_do), &
2198	ngp, MPI_REAL, MPI_SUM, comm1dx, &
2199	ierr )
2200	#else
2201	local_2d_sections = local_2d_sections_l
2202	#endif
2203	ENDIF
2204	!
2205	!-- Do not output redundant ghost point data except for the
2206	!-- boundaries of the total domain.
2207	! IF ( nyn == ny ) THEN
2208	! nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2209	! id_var_do2d(av,ivar), &
2210	! local_2d_sections(1:ns, &
2211	! nys:nyn+1,nzb_do:nzt_do), &
2212	! start = (/ 1, nys+1, 1, &
2213	! do2d_yz_time_count(av) /), &
2214	! count = (/ ns, nyn-nys+2, &
2215	! nzt_do-nzb_do+1, 1 /) )
2216	! ELSE
2217	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2218	id_var_do2d(av,ivar), &
2219	local_2d_sections(1:ns,nys:nyn, &
2220	nzb_do:nzt_do), &
2221	start = (/ 1, nys+1, 1, &
2222	do2d_yz_time_count(av) /), &
2223	count = (/ ns, nyn-nys+1, &
2224	nzt_do-nzb_do+1, 1 /) )
2225	! ENDIF
2226
2227	CALL netcdf_handle_error( 'data_output_2d', 60 )
2228
2229	CASE DEFAULT
2230	message_string = 'unknown cross-section: ' // TRIM( mode )
2231	CALL message( 'data_output_2d', 'PA0180', 1, 2, 0, 6, 0 )
2232
2233	END SELECT
2234
2235	ENDIF
2236	#endif
2237	ENDIF
2238
2239	ivar = ivar + 1
2240	l = MAX( 2, LEN_TRIM( do2d(av,ivar) ) )
2241	do2d_mode = do2d(av,ivar)(l-1:l)
2242
2243	ENDDO
2244
2245	!
2246	!-- Deallocate temporary arrays.
2247	IF ( ALLOCATED( level_z ) ) DEALLOCATE( level_z )
2248	IF ( netcdf_data_format > 4 ) THEN
2249	DEALLOCATE( local_pf, local_2d, local_2d_sections )
2250	IF( mode == 'xz' .OR. mode == 'yz' ) DEALLOCATE( local_2d_sections_l )
2251	ENDIF
2252	#if defined( __parallel )
2253	IF ( .NOT. data_output_2d_on_each_pe .AND. myid == 0 ) THEN
2254	DEALLOCATE( total_2d )
2255	ENDIF
2256	#endif
2257
2258	!
2259	!-- Close plot output file.
2260	file_id = 20 + s_ind
2261
2262	IF ( data_output_2d_on_each_pe ) THEN
2263	DO i = 0, io_blocks-1
2264	IF ( i == io_group ) THEN
2265	CALL close_file( file_id )
2266	ENDIF
2267	#if defined( __parallel )
2268	CALL MPI_BARRIER( comm2d, ierr )
2269	#endif
2270	ENDDO
2271	ELSE
2272	IF ( myid == 0 ) CALL close_file( file_id )
2273	ENDIF
2274
2275	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
2276
2277	IF ( debug_output ) THEN
2278	WRITE( debug_string, * ) 'data_output_2d'
2279	CALL debug_message( debug_string, 'end' )
2280	ENDIF
2281
2282	END SUBROUTINE data_output_2d

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

Download in other formats:

| Impressum | ©Leibniz Universität Hannover |