Context Navigation

nc_read.pro @ 27

Last change on this file since 27 was 27, checked in by kolasinski, 16 years ago
Conform to SAXO new version - Structures varcontient and contient do not exist anymore in read_ncdf_varget
File size: 22.5 KB

Line
1	;---------------------------
2	; Reading ORCA netcdf files
3	; EG 24/2/99
4	;---------------------------
5	FUNCTION nc_read, file_name, var_name, ncdf_db, TIME_1 = time_1, TIME_2 = time_2, ALL_DATA = all_data, NO_MEAN = no_mean
6
7	; arguments = file_name, varname
8	; ncdf_db=<location>:<path> or just <path>
9	; mot clef = iodir time_1 time_2
10
11	@common
12	@com_eg
13
14	;; print, 'key_yreverse : ', key_yreverse
15
16	; inits
17	IF debug_w THEN print, 'DEBUG: Entering nc_read...'
18	IF NOT keyword_set(TIME_1) THEN time_1 = 1
19	IF NOT keyword_set(TIME_2) THEN time_2 = time_1
20	;
21	; decide which subdomain of data to read
22	;
23
24	IF debug_w THEN print, 'keyword_set(ALL_DATA) : ', keyword_set(ALL_DATA)
25
26	IF keyword_set(ALL_DATA) THEN BEGIN
27	print, ' Reading whole domain'
28	firstx = 0
29	firsty = 0
30	firstz = 0
31	nx = jpi
32	ny = jpj
33	nz = jpk
34	lastx = jpi-1
35	lasty = jpj-1
36	lastz = jpk-1
37	;Rajout MK
38	;Trouver une meilleure place
39	ixminmesh = 0
40	iyminmesh = 0
41	ixmaxmesh = jpi-1
42	iymaxmesh = jpj-1
43	;Fin rajout
44	ENDIF ELSE BEGIN
45	grille,mask,glam,gphi,gdep,nx,ny,nz,firstx,firsty,firstz,lastx,lasty,lastz
46	mask = 1
47	ENDELSE
48
49	IF debug_w THEN print, 'nx,ny,nz,firstx,firsty,firstz,lastx,lasty,lastz'
50	IF debug_w THEN print, nx,ny,nz,firstx,firsty,firstz,lastx,lasty,lastz
51
52	IF debug_w THEN print, 'key_offset =', key_offset
53	IF debug_w THEN print, 'jpi, jpj, jpk =', jpi, jpj, jpk
54
55
56	; define reading boundaries
57
58	x_count = nx
59	y_count = ny
60	z_count = nz
61
62	IF debug_w THEN print, 'nx,ny,nz =', nx,ny,nz
63
64	; force izmaxmesh to lastz
65	jpk = lastz+1
66
67	; dealing with longitude periodicity
68	IF x_count NE jpi THEN BEGIN
69	key_shift_read = 0
70	ENDIF ELSE key_shift_read = key_shift
71	IF debug_w THEN print, 'key_shift_read=', key_shift_read
72
73	x_offset = key_offset[0]+ firstx+(key_shift_read-key_shift)
74	y_offset = key_offset[1]+ firsty
75	z_offset = key_offset[2]+ firstz
76	IF debug_w THEN print, 'offset =', x_offset, y_offset, z_offset
77
78	; pour trouver un fichier, tu as
79	; findfile (tres pratique) et dialog_pickfile (interactif)
80
81	; pour verifier si il y a une variable je fais
82
83	; inq=ncdf_inquire(cdfid)
84	; for varid=0,inq.nvars-1 do BEGIN
85	; varinq=ncdf_varinq(cdfid,varid)
86	; if varinq.name eq nom then goto, variabletrouvee
87	; endfor
88	; return, -1
89	; variabletrouvee:
90
91	; local directory
92
93	IF strpos(ncdf_db, ':') GE 1 THEN directory = (str_sep(ncdf_db, ':'))[1] $
94	ELSE directory = ncdf_db
95
96	; existence of file
97
98	list = findfile(directory+file_name, count = nb_file)
99
100	IF nb_file EQ 0 THEN BEGIN
101	print, ''
102	print, ' ***** file ', file_name, ' not found in ', $
103	directory
104	print, ''
105	field = {data: -1.0}
106	return, field
107	ENDIF
108
109	; ouverture fichier netCDF + contenu
110	cdfid=ncdf_open(directory+file_name)
111	inq=ncdf_inquire(cdfid)
112	; que contient la variable
113	varid = ncdf_varid(cdfid, var_name)
114
115	name_suff = ''
116
117	IF varid EQ -1 THEN BEGIN
118	print, ''
119	print, ' ***** field ', var_name, ' not found in file ', $
120	file_name
121	print, ''
122	field = {data: -1.0}
123	return, field
124	ENDIF
125	varinq=ncdf_varinq(cdfid, var_name)
126
127	; test sur la dimension
128	err_mess = ''
129	field_dim = n_elements(varinq.dim)
130
131	; get unlimited record variable
132	IF inq.recdim NE -1 THEN BEGIN
133	ncdf_diminq, cdfid, inq.recdim, name_time, nb_time
134	;;ncdf_varget, cdfid, inq.recdim, time_array
135	;;nb_time = (size(time_array))(1)
136	ENDIF ELSE BEGIN
137	; Look for a potential record dimension
138	IF debug_w THEN print, ' Warning : no unlimited record in netCDF file'
139	; Look for the names of all dimensions and all variables
140	list_dims = ncdf_listdims(cdfid)
141	list_vars = ncdf_listvars(cdfid)
142	; Propose all the names and make the user choose the right one
143	; for the record dimension
144	print, 'In the file ', file_name, ', the dimensions are named :'
145	print, list_dims
146	print, 'In the file ', file_name, ', the variables are named :'
147	print, list_vars
148	name_time = xquestion('Among the preceding names, which one could refer to the record dimension ?', /chkwid)
149	IF name_time NE '-' THEN BEGIN
150	dimidt = ncdf_dimid(cdfid, name_time)
151	ncdf_diminq, cdfid, dimidt, name_time, nb_time
152	print, 'You chose ', name_time, ' as a record dimension and its size is ', nb_time
153	inq.recdim = dimidt
154	ENDIF ELSE BEGIN
155	print, 'No record dimension considered in the file'
156	nb_time = 0
157	ENDELSE
158
159	IF varinq.dim[varinq.ndims-1] NE dimidt THEN STOP, $
160	'Post_it cannot handle variables whose record dimension is not the last one'
161
162	; ncdf_diminq,cdfid,(n_elements(varinq.dim)-1), name_time, nb_time
163	; dimidl = ncdf_dimid(cdfid, name_time)
164	; ncdf_diminq,cdfid,dimidl, name_time, nb_time
165	; varidl = ncdf_varid(cdfid, name_time)
166	; ncdf_varget, cdfid, varidl, time_array
167
168	ENDELSE
169
170	IF jpt GT nb_time THEN BEGIN
171	jpt = nb_time
172	print, 'Warning : the specified jpt does not correspond to the real time dimension in the file !'
173	print, 'New jpt =', jpt
174	ENDIF
175	; defs for @read_ncdf_varget
176
177	; key_stride = time_stride
178	key_stride = 1
179	if n_elements(key_stride) LE 2 then key_stride = [1, 1, 1]
180	key_stride = 1l > long(key_stride)
181	jpitotal = long(ixmaxmesh-ixminmesh+1)
182	key_shift = long(testvar(var = key_shift))
183
184	key = long(key_shift MOD jpitotal)
185	if key LT 0 then key = key+jpitotal
186	ixmin = ixminmesh-ixmindta
187	iymin = iyminmesh-iymindta
188	firsttps = time_1-1
189	lasttps = time_2-1
190
191	name = varid
192
193	IF debug_w THEN print, 'key=', key
194	IF debug_w THEN print, 'jpitotal=', jpitotal
195	IF debug_w THEN print, 'ixmindta,iymindta,izmindta =', ixmindta,iymindta,izmindta
196	IF debug_w THEN print, 'ixminmesh,iyminmesh=', ixminmesh,iyminmesh
197	IF debug_w THEN print, 'ixmaxmesh,iymaxmesh=', ixmaxmesh,iymaxmesh
198	IF debug_w THEN print, 'izminmesh,izmaxmesh=', izminmesh,izmaxmesh
199	IF debug_w THEN print, 'ixmin,iymin=', ixmin,iymin
200	IF debug_w THEN print, 'firsttps,lasttps=', firsttps, lasttps
201	IF debug_w THEN print, 'key_shift in nc_read=', key_shift
202	IF debug_w THEN print, 'key_yreverse, firsty, lasty',key_yreverse, firsty, lasty
203
204	CASE n_elements(varinq.dim) OF
205
206	;; fichier 2d : surface
207	2: BEGIN
208	print, ' Reading ', var_name, ' (2D data) from ', file_name
209	@read_ncdf_varget
210	lec_data = res
211	data_direc = 'xy'
212	niveau = 1
213	END
214
215	;; fichier 3d : 2 cas = 1/ 2d espace + temps 2/ 3d espace
216	3: BEGIN
217	;; si varinq.dim contient la dim infinie (no 3) -> temps
218	dim_3 = '3d'
219	IF nb_time GE 1 THEN dim_3 = '2d'
220	IF dim_3 EQ '2d' THEN BEGIN
221
222	IF time_2 EQ time_1 THEN BEGIN
223	print, ' Reading ', var_name, ' (2D data - index ', strcompress(string(time_1)),') from ', file_name
224	@read_ncdf_varget
225	lec_data = res
226	data_direc = 'xy'
227	ENDIF ELSE BEGIN
228	print, ' Reading ', var_name, ' (2D data time serie)', strcompress(string(time_1)),'-', strtrim(string(time_2), 2), ' [every ',strtrim(string(time_stride), 2), '] from ', file_name
229	IF debug_w THEN print, 'x_offset et y_offset et time_1 :', x_offset, y_offset, time_1
230	@read_ncdf_varget
231	lec_data = res
232	data_direc = 'xyt'
233	ENDELSE
234
235	ENDIF ELSE BEGIN
236	print, ' Reading ', var_name, ' (3D data)', ' from ', file_name
237	@read_ncdf_varget
238	lec_data = res
239	data_direc = 'xyz'
240	ENDELSE
241	END
242	;; fichier 4d : volume + temps
243	4: BEGIN
244	IF time_2 EQ time_1 THEN BEGIN
245	print, ' Reading ', var_name, ' (3D data - index ', strcompress(string(time_1)),') from ', file_name
246	; read vertical levels
247	IF mesh_type EQ 'atm' THEN BEGIN
248	ncdf_diminq,cdfid,2, name_level, nb_level
249
250	; get name of vertical level from metadata
251	file_grid_config = hom_def+'grid_config.def'
252	spawn, 'grep -i "\ '+meshlec_type+' " '+file_grid_config, line
253	line = strcompress(strtrim(line[0], 2))
254	length = strlen(line)
255
256	IF length EQ 0 THEN BEGIN
257	print, ' *** nc_read : define name_level from grid ', meshlec_type, $
258	' in file ', file_grid_config
259	name_level = ''
260	return, -1
261	ENDIF ELSE BEGIN
262	argvar = str_sep(line, ' ')
263	name_level = argvar[4]
264	ENDELSE
265	dimidl = ncdf_dimid(cdfid, name_level)
266	ncdf_diminq,cdfid,dimidl, name_level, nb_level
267
268	varidl = ncdf_varid(cdfid, name_level)
269	; make sure name_level is in hPa
270	ncdf_attget, cdfid, varidl, 'units', val_unit
271	IF string(val_unit) NE 'hPa' AND string(val_unit) NE 'millibar' AND string(val_unit) NE 'mbar' THEN BEGIN
272	print, ' vertical levels coordinate not obvious = ', name_level
273	print, ' ... using <levels> ...'
274	varidl = ncdf_varid(cdfid, 'levels')
275	ENDIF
276
277	ncdf_varget, cdfid, varidl, gdept
278	gdept = gdept
279	gdepw = gdept
280	e3t = shift(gdept, 1)-gdept
281	e3t[0] = e3t[1]
282	e3w = e3t
283	jpk = nb_level
284	tmask = reform((reform(tmask(, ), jpi*jpj)#replicate(1, jpk)), jpi, jpj, jpk)
285	ENDIF
286	IF debug_w THEN print, ' going into @read_ncdf_varget '
287	@read_ncdf_varget
288	lec_data = res
289	; lec_data = reform(lec_data,x_count, y_count, z_count)
290	data_direc = 'xyz'
291	IF debug_w THEN help, lec_data
292
293	; vertical average ?
294	IF vert_switch ge 1 AND NOT keyword_set(no_mean) THEN BEGIN
295	old_boite = [lon1, lon2, lat1, lat2, prof1, prof2]
296	print, ' Average in vertical domain ', vert_type, vert_mean
297	IF mesh_type EQ 'atm' THEN BEGIN
298	CASE atmos_msk OF
299	0: print, ' [atmos grid : take all points] ' ; take all points
300	1: BEGIN
301	; take ocean points only
302	idx = where(tmask EQ 0)
303	lec_data(idx) = 1.e20
304	print, ' [atmos grid : take ocean points only] '
305	END
306	2: BEGIN
307	; take land points only
308	idx = where(tmask GT 0)
309	lec_data(idx) = 1.e20
310	print, ' [atmos grid : take land points only] '
311	END
312	ENDCASE
313	CASE vert_type OF
314	'z': BEGIN ; depth range
315	IF time_1 EQ time_2 THEN BEGIN
316	zmean = moyenne(lec_data, 'z', boite = vert_mean, NAN =1.e20)
317	END ELSE zmean = grossemoyenne(lec_data, 'z', boite = vert_mean, NAN =1.e20)
318	IF vert_mean[0] EQ vert_mean[1] THEN BEGIN
319	name_suff = ' at '+strtrim(string(long(gdept(vert_mean(0)))), 2)+' hPa'
320	ENDIF ELSE BEGIN
321	name_suff = ' averaged in '+vert_type+'['+strtrim(string(long(vert_mean(0))), 2)+','+strtrim(string(long(vert_mean(1))), 2)+']'
322	ENDELSE
323	END
324	ELSE: BEGIN ; levels range
325	IF vert_mean[0] EQ vert_mean[1] THEN BEGIN
326	zmean = lec_data(, , vert_mean[0])
327	name_suff = ' at '+strtrim(string(long(gdept(vert_mean(0)))), 2)+' hPa'
328	ENDIF ELSE BEGIN
329	zmean = moyenne(lec_data, 'z', boite = vert_mean, /zindex, NAN =1.e20)
330	name_suff = ' averaged in '+vert_type+'['+strtrim(string(long(vert_mean(0))), 2)+','+strtrim(string(long(vert_mean(1))), 2)+']'
331	ENDELSE
332	END
333	ENDCASE
334	; tmask = reform(tmask(, , 0), jpi, jpj)
335	ENDIF ELSE BEGIN
336	; ocean = always mask
337	; idx = where(tmask EQ 0)
338	; lec_data(idx) = 1.e20
339	CASE vert_type OF
340	'z': BEGIN
341	IF vert_mean[0] EQ vert_mean[1] THEN BEGIN
342	zmean = lec_data ; right depth already selected
343	name_suff = ' at '+strtrim(string(long(gdept(vert_mean(0))+.1)), 2)+' m'
344	ENDIF ELSE BEGIN
345	zmean = moyenne(lec_data, 'z', boite = vert_mean)
346	name_suff = ' averaged in '+vert_type+'['+strtrim(string(long(vert_mean(0))), 2)+','+strtrim(string(long(vert_mean(1))), 2)+']'
347	ENDELSE
348	END
349	ELSE: BEGIN ; case level (zindex)
350	IF vert_mean[0] EQ vert_mean[1] THEN BEGIN
351	zmean = lec_data ; right depth already selected
352	name_suff = ' at '+strtrim(string(long(gdept(vert_mean(0))+.1)), 2)+' m'
353	ENDIF ELSE BEGIN
354	zmean = moyenne(lec_data, 'z', boite = vert_mean, /zindex)
355	name_suff = ' averaged in '+vert_type+'['+strtrim(string(long(vert_mean(0))), 2)+','+strtrim(string(long(vert_mean(1))), 2)+']'
356	ENDELSE
357	END
358	ENDCASE
359	ENDELSE
360	lec_data = zmean
361	domdef, old_boite
362	field_dim = field_dim - 1
363	data_direc = 'xy'
364	nzt = 1
365	firstz = 0
366	lastz = 0
367	ENDIF
368
369	ENDIF ELSE BEGIN
370	print, ' Reading ', var_name, ' (3D data time serie)', strcompress(string(time_1)),'-', strtrim(string(time_2), 2), ' from ', file_name
371	; read vertical levels
372	IF mesh_type EQ 'atm' THEN BEGIN
373	ncdf_diminq,cdfid,2, name_level, nb_level
374
375	; get name of vertical level from metadata
376	file_grid_config = hom_def+'grid_config.def'
377	spawn, 'grep -i "\ '+meshlec_type+' " '+file_grid_config, line
378	line = strcompress(strtrim(line[0], 2))
379	length = strlen(line)
380
381	IF length EQ 0 THEN BEGIN
382	print, ' *** nc_read : define name_level from grid ', meshlec_type, $
383	' in file ', file_grid_config
384	name_level = ''
385	return, -1
386	ENDIF ELSE BEGIN
387	argvar = str_sep(line, ' ')
388	name_level = argvar[1]
389	ENDELSE
390	dimidl = ncdf_dimid(cdfid, name_level)
391	ncdf_diminq,cdfid,dimidl, name_level, nb_level
392
393	varidl = ncdf_varid(cdfid, name_level)
394	; make sure name_level is in hPa
395	ncdf_attget, cdfid, varidl, 'units', val_unit
396	IF string(val_unit) NE 'hPa' AND string(val_unit) NE 'millibar' AND string(val_unit) NE 'mbar' THEN BEGIN
397	print, ' vertical levels coordinate not obvious'
398	print, ' ... using <levels> ...'
399	varidl = ncdf_varid(cdfid, 'levels')
400	ENDIF
401	ncdf_varget, cdfid, varidl, gdept
402	gdepw = gdept
403	e3t = shift(gdept, 1)-gdept
404	e3t[0] = e3t[1]
405	e3w = e3t
406	jpk = nb_level
407	tmask = reform((reform(tmask(, ), jpi*jpj)#replicate(1, jpk)), jpi, jpj, jpk)
408	ENDIF
409	@read_ncdf_varget
410	lec_data = res
411	data_direc = 'xyzt'
412
413	; vertical average ?
414	IF vert_switch ge 1 AND NOT keyword_set(no_mean) THEN BEGIN
415	old_boite = [lon1, lon2, lat1, lat2, prof1, prof2]
416	print, ' Average in vertical domain ', vert_type, vert_mean
417	IF mesh_type EQ 'atm' THEN BEGIN
418	CASE atmos_msk OF
419	0: print, ' [take all points] ' ; take all points
420	1: BEGIN
421	; take ocean points only
422	idx = where(tmask EQ 0)
423	lec_data(idx) = 1.e20
424	print, ' [take ocean points only] '
425	END
426	2: BEGIN
427	; take land points only
428	idx = where(tmask GT 0)
429	lec_data(idx) = 1.e20
430	print, ' [take land points only] '
431	END
432	ENDCASE
433	CASE vert_type OF
434	'z': BEGIN ; levels
435
436	IF vert_mean[0] EQ vert_mean[1] THEN BEGIN
437	name_suff = ' at '+strtrim(string(long(gdept(vert_mean(0)))), 2)+' hPa'
438	ENDIF ELSE BEGIN
439	zmean = grossemoyenne(lec_data, 'z', boite = vert_mean)
440	name_suff = ' averaged in '+vert_type+'['+strtrim(string(vert_mean(0)), 2)+','+strtrim(string(vert_mean(1)), 2)+']'
441	ENDELSE
442	END
443	ELSE: BEGIN ; levels
444	IF vert_mean[0] EQ vert_mean[1] THEN BEGIN
445	zmean = lec_data(, , vert_mean[0])
446	name_suff = ' at '+vert_type+' '+strtrim(string(long(vert_mean(0))), 2)+' ['+strtrim(string(gdept(vert_mean(0))), 2)+' hPa]'
447	ENDIF ELSE BEGIN
448	zmean = moyenne(lec_data, 'z', boite = vert_mean, /zindex)
449	name_suff = ' averaged in '+vert_type+'['+strtrim(string(long(vert_mean(0))), 2)+','+strtrim(string(long(vert_mean(1))), 2)+']'
450	ENDELSE
451	END
452	ENDCASE
453	; tmask = reform(tmask(, , 0), jpi, jpj)
454	ENDIF ELSE BEGIN
455	; ocean = always mask
456	; idx = where(tmask EQ 0)
457	; lec_data(idx) = 1.e20
458	CASE vert_type OF
459	'z': BEGIN
460	IF vert_mean[0] EQ vert_mean[1] THEN BEGIN
461	zmean = lec_data
462	name_suff = ' at '+strtrim(string(long(gdept(vert_mean(0))+.1)), 2)+' m'
463	ENDIF ELSE BEGIN
464	zmean = grossemoyenne(lec_data, 'z', boite = vert_mean)
465	name_suff = ' averaged in '+vert_type+'['+strtrim(string(long(vert_mean(0))), 2)+','+strtrim(string(long(vert_mean(1))), 2)+']'
466	ENDELSE
467	END
468	ELSE: BEGIN
469	IF vert_mean[0] EQ vert_mean[1] THEN BEGIN
470	zmean = lec_data
471	name_suff = ' at '+strtrim(string(long(gdept(vert_mean(0))+.1)), 2)+' m'
472	ENDIF ELSE BEGIN
473	zmean = moyenne(lec_data, 'z', boite = vert_mean, /zindex, NAN =1.e20)
474	name_suff = ' averaged in '+vert_type+'['+strtrim(string(long(vert_mean(0))), 2)+','+strtrim(string(long(vert_mean(1))), 2)+']'
475	ENDELSE
476	END
477	ENDCASE
478	ENDELSE
479	lec_data = zmean
480	domdef, old_boite
481	field_dim = field_dim - 1
482	data_direc = 'xyt'
483	nzt = 1
484	firstz = 0
485	lastz = 0
486	ENDIF
487	ENDELSE
488	END
489	;; erreur si dim > 4
490	ELSE: BEGIN
491	err_mess = ' *** nc_read : ERROR dimension > 4'
492	lec_data = -1.0
493	END
494	ENDCASE
495
496	; scaling ?
497	FOR i = 0, varinq.natts-1 DO BEGIN
498	att_txt = ncdf_attname(cdfid, varid, i)
499	IF att_txt EQ 'scale_factor' THEN BEGIN
500	ncdf_attget, cdfid, varid, att_txt, valscale
501	lec_data = lec_data*valscale
502	ENDIF
503	ENDFOR
504
505	; Attribut du champ
506
507	field = {name: '', data: lec_data, legend: '', units: '', origin: '', dim: 0, direc: data_direc}
508	field.name = var_name
509	field.origin = directory+file_name
510	;
511	; get long name
512	; result = '???'
513	FOR i = 0, varinq.natts-1 DO BEGIN
514	att_txt = ncdf_attname(cdfid, varid, i)
515	IF att_txt EQ 'long_name' OR att_txt EQ 'title' THEN ncdf_attget, cdfid, varid, att_txt, result
516	ENDFOR
517	field.legend = strtrim(string(result),1)+name_suff
518
519	; get unit
520	; if it exists
521	isunits = ncdf_attinq(cdfid, varid, 'units')
522	IF isunits.datatype NE 'UNKNOWN' THEN BEGIN
523	ncdf_attget, cdfid, varid, 'units', result
524	field.units = strtrim(string(result),1)
525	ENDIF ELSE BEGIN
526	print, 'No units for the variable ', field.name
527	print, ' Or units were forgotten in the file !'
528	ENDELSE
529	field.dim = field_dim
530
531
532	; get valmask (might need valmask = float(string(valmask))
533
534	valmask = 1.e20
535	FOR i = 0, varinq.natts-1 DO BEGIN
536	att_txt = ncdf_attname(cdfid, varid, i)
537	IF att_txt EQ 'missing_value' OR att_txt EQ 'mask value' OR att_txt EQ '_FillValue' THEN ncdf_attget, cdfid, varid, att_txt, valmask
538	ENDFOR
539
540	; set valmask to 1.e20
541
542	; ensure valmask is positive
543
544	IF valmask LT 0 THEN BEGIN
545	print, ' *** Warning valmask is negative - changing sign: ', valmask
546	idx_t = where (field.data EQ valmask)
547	IF idx_t(0) NE -1 THEN field.data(idx_t) = -valmask
548	valmask = -valmask
549	idx_t=0 ; free memory
550	ENDIF
551
552	; min/max
553
554	chardim = ' - dim = '
555	FOR i = 1, (size(field.data))[0] DO BEGIN
556	chardim = chardim+string((size(field.data))[i], format = '(I5)')
557	ENDFOR
558
559	index_test = (where (field.data NE valmask))
560	IF index_test(0) NE -1 THEN BEGIN
561	minf = min(field.data(where (field.data NE valmask)))
562	maxf = max(field.data(where (field.data NE valmask)))
563	ENDIF ELSE BEGIN
564	minf = min(field.data)
565	maxf = max(field.data)
566	ENDELSE
567
568	print, ' = ',field.legend, ' [min/max = ',minf , maxf,' ', field.units,' - masked values = ',valmask, chardim, ']'
569
570
571	ncdf_close, cdfid
572
573	;key_offset = key_offset_orig
574	;jpi = jpi_orig
575	;jpj = jpj_orig
576	;jpk = jpk_orig
577
578	IF keyword_set(key_yreverse) THEN print, ' key_yreverse active : ', key_yreverse
579
580	IF debug_w THEN print, 'DEBUG: Exiting nc_read...'
581
582	return, field
583
584
585	END

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

Download in other formats:

Original Format