1 |
#!/usr/bin/env python |
---|
2 |
# Last modified: Time-stamp: <2012-06-28 14:44:14 haines> |
---|
3 |
""" |
---|
4 |
how to parse data, and assert what data and info goes into |
---|
5 |
creating and updating monthly netcdf files |
---|
6 |
|
---|
7 |
parse data ctd data collected on Seabird CTD -- SBE37 |
---|
8 |
derive salinity, depth, and density using seawater.csiro toolbox |
---|
9 |
|
---|
10 |
parser : sample date and time, wtemp, cond, press, (derive) depth, salin, dens |
---|
11 |
|
---|
12 |
creator : lat, lon, z, time, wtemp, cond, press, depth, salin, dens |
---|
13 |
updator : time, wtemp, cond, press, depth, salin, dens |
---|
14 |
|
---|
15 |
Examples |
---|
16 |
-------- |
---|
17 |
>>> sensor_info = sensor_info['ctd1'] |
---|
18 |
>>> (parse, create, update) = import_processors(sensor_info['process_module']) |
---|
19 |
|
---|
20 |
>>> data = parse(platform_info, sensor_info, lines) |
---|
21 |
>>> create(platform_info, sensor_info, data) or |
---|
22 |
>>> update(platform_info, sensor_info, data) |
---|
23 |
|
---|
24 |
Testing |
---|
25 |
------- |
---|
26 |
from raw2proc import * |
---|
27 |
cn = 'b1_config_20111112' |
---|
28 |
sensor_info = get_config(cn+'.sensor_info') |
---|
29 |
sensor_info = sensor_info['ctd1'] |
---|
30 |
platform_info = get_config(cn+'.platform_info') |
---|
31 |
(parse, create, update) = import_processors(sensor_info['process_module']) |
---|
32 |
|
---|
33 |
filename = '/seacoos/data/nccoos/level0/b1/ctd1/store/B1_CTD1_2011_11_12.asc' |
---|
34 |
lines = load_data(filename) |
---|
35 |
sensor_info['fn'] = filename |
---|
36 |
data = parse(platform_info, sensor_info, lines) |
---|
37 |
|
---|
38 |
create(platform_info, sensor_info, data) |
---|
39 |
update(platform_info, sensor_info, data) |
---|
40 |
|
---|
41 |
""" |
---|
42 |
from raw2proc import * |
---|
43 |
from procutil import * |
---|
44 |
from ncutil import * |
---|
45 |
|
---|
46 |
now_dt = datetime.utcnow() |
---|
47 |
now_dt.replace(microsecond=0) |
---|
48 |
|
---|
49 |
def parser(platform_info, sensor_info, lines): |
---|
50 |
""" |
---|
51 |
Header comments start with '*' |
---|
52 |
Last line of comments *END* |
---|
53 |
|
---|
54 |
* Sea-Bird SBE37 Data File: |
---|
55 |
* FileName = C:\Documents and Settings\haines\Desktop\nc-wind 2012 CTD Recovery\B1_CTD1_3085_2012-04-07.asc |
---|
56 |
... |
---|
57 |
* S> |
---|
58 |
*END* |
---|
59 |
start time = 12 Nov 2011 15:28:43 |
---|
60 |
sample interval = 360 seconds |
---|
61 |
start sample number = 1 |
---|
62 |
17.4036, 4.35264, 3.521, 12 Nov 2011, 15:28:43 |
---|
63 |
17.4289, 4.35624, 3.593, 12 Nov 2011, 15:34:44 |
---|
64 |
17.4110, 4.35376, 3.600, 12 Nov 2011, 15:40:44 |
---|
65 |
17.4106, 4.35395, 3.618, 12 Nov 2011, 15:46:44 |
---|
66 |
17.3798, 4.34961, 3.515, 12 Nov 2011, 15:52:44 |
---|
67 |
17.3861, 4.35033, 3.708, 12 Nov 2011, 15:58:44 |
---|
68 |
17.4136, 4.35348, 3.488, 12 Nov 2011, 16:04:44 |
---|
69 |
17.4269, 4.35530, 3.616, 12 Nov 2011, 16:10:44 |
---|
70 |
17.4421, 4.35679, 3.612, 12 Nov 2011, 16:16:44 |
---|
71 |
17.4417, 4.35679, 3.537, 12 Nov 2011, 16:22:44 |
---|
72 |
... EOF |
---|
73 |
|
---|
74 |
""" |
---|
75 |
|
---|
76 |
import numpy |
---|
77 |
from datetime import datetime |
---|
78 |
from time import strptime |
---|
79 |
|
---|
80 |
# get sample datetime from filename |
---|
81 |
fn = sensor_info['fn'] |
---|
82 |
sample_dt_start = filt_datetime(fn) |
---|
83 |
|
---|
84 |
# tease out the header info and where it ends |
---|
85 |
# **** may want more info extracted later for data attributes (e.g. sensor coeff) |
---|
86 |
serial_number, end_idx, sample_interval_str = (None, None, None) |
---|
87 |
for idx, k in enumerate(lines[0:100]): |
---|
88 |
# serial number |
---|
89 |
m = re.search(r'^\*.*(SERIAL NO\.)\s+(\d*)', k) |
---|
90 |
if m: serial_number = m.group(2) |
---|
91 |
m = re.search(r'^\*\w+(sample interval)\s*=\s*(.*)', k) |
---|
92 |
if m: sample_interval_str = m.group(2) |
---|
93 |
m = re.search(r'^(\*END\*).*', k) |
---|
94 |
if m: end_idx = idx |
---|
95 |
# check that serial_info serial_number, sample_interval matches |
---|
96 |
|
---|
97 |
# split data from header info and get how many samples (start count 3 lines past *END*) |
---|
98 |
if end_idx: lines = lines[end_idx+3:] |
---|
99 |
nsamp = len(lines) |
---|
100 |
|
---|
101 |
N = nsamp |
---|
102 |
data = { |
---|
103 |
'dt' : numpy.array(numpy.ones((N,), dtype=object)*numpy.nan), |
---|
104 |
'time' : numpy.array(numpy.ones((N,), dtype=long)*numpy.nan), |
---|
105 |
'wtemp' : numpy.array(numpy.ones((N,), dtype=float)*numpy.nan), |
---|
106 |
'cond' : numpy.array(numpy.ones((N,), dtype=float)*numpy.nan), |
---|
107 |
'press' : numpy.array(numpy.ones((N,), dtype=float)*numpy.nan), |
---|
108 |
'salin' : numpy.array(numpy.ones((N,), dtype=float)*numpy.nan), |
---|
109 |
'density' : numpy.array(numpy.ones((N,), dtype=float)*numpy.nan), |
---|
110 |
'depth' : numpy.array(numpy.ones((N,), dtype=float)*numpy.nan), |
---|
111 |
} |
---|
112 |
|
---|
113 |
# sample count |
---|
114 |
i = 0 |
---|
115 |
|
---|
116 |
for line in lines: |
---|
117 |
# if line has weird ascii chars -- skip it and iterate to next line |
---|
118 |
csi = [] |
---|
119 |
# split line |
---|
120 |
sw = re.split(',', line) |
---|
121 |
if len(sw)<=0: |
---|
122 |
print ' ... skipping line %d -- %s' % (i,line) |
---|
123 |
continue |
---|
124 |
|
---|
125 |
# replace "NAN" |
---|
126 |
for index, s in enumerate(sw): |
---|
127 |
m = re.search(NAN_RE_STR, s) |
---|
128 |
if m: |
---|
129 |
sw[index] = '-99999' |
---|
130 |
|
---|
131 |
# parse date-time, and all other float and integers |
---|
132 |
for s in sw[0:3]: |
---|
133 |
m = re.search(REAL_RE_STR, s) |
---|
134 |
if m: |
---|
135 |
csi.append(float(m.groups()[0])) |
---|
136 |
|
---|
137 |
if len(sw)>=5: |
---|
138 |
dstr = sw[3]+' '+sw[4] |
---|
139 |
# print dstr |
---|
140 |
m = re.search('\s*(\d{2})\s*(\w{2,3})\s*(\d{4})\s*(\d{2}):(\d{2}):(\d{2}).*', dstr) |
---|
141 |
else: |
---|
142 |
print ' ... skipping line %d -- %s ' % (i,line) |
---|
143 |
continue |
---|
144 |
|
---|
145 |
if m: |
---|
146 |
dstr = '%s %s %s %s:%s:%s' % m.groups() |
---|
147 |
else: |
---|
148 |
print ' ... skipping line %d -- %s ' % (i,line) |
---|
149 |
continue |
---|
150 |
|
---|
151 |
if sensor_info['utc_offset']: |
---|
152 |
sample_dt = scanf_datetime(dstr, fmt='%d %b %Y %H:%M:%S') + \ |
---|
153 |
timedelta(hours=sensor_info['utc_offset']) |
---|
154 |
else: |
---|
155 |
sample_dt = scanf_datetime(dstr, fmt='%d %b %Y %H:%M:%S') |
---|
156 |
|
---|
157 |
# ***** TO DO: may need to adjust any offset in CTD sample time to UTC clock |
---|
158 |
# This requires knowing what UTC time is at a CTD sample |
---|
159 |
data['dt'][i] = sample_dt # sample datetime |
---|
160 |
data['time'][i] = dt2es(sample_dt) # sample time in epoch seconds |
---|
161 |
|
---|
162 |
if len(csi)==3: |
---|
163 |
# |
---|
164 |
# (pg 31 SBE IMP Microcat User Manual) |
---|
165 |
# "#iiFORMAT=1 (default) Output converted to data |
---|
166 |
# date format dd mmm yyyy, |
---|
167 |
# conductivity = S/m, |
---|
168 |
# temperature precedes conductivity" |
---|
169 |
data['wtemp'][i] = csi[0] # water temperature (C) |
---|
170 |
data['cond'][i] = csi[1] # specific conductivity (S/m) |
---|
171 |
data['press'][i] = csi[2] # pressure decibars |
---|
172 |
i=i+1 |
---|
173 |
else: |
---|
174 |
print ' ... skipping line %d -- %s ' % (i,line) |
---|
175 |
continue |
---|
176 |
|
---|
177 |
# if re.search |
---|
178 |
# for line |
---|
179 |
|
---|
180 |
# check that no data[dt] is set to Nan or anything but datetime |
---|
181 |
# keep only data that has a resolved datetime |
---|
182 |
keep = numpy.array([type(datetime(1970,1,1)) == type(dt) for dt in data['dt'][:]]) |
---|
183 |
if keep.any(): |
---|
184 |
for param in data.keys(): |
---|
185 |
data[param] = data[param][keep] |
---|
186 |
|
---|
187 |
# Quality Control steps for temp, depth, and cond |
---|
188 |
# (1) within range |
---|
189 |
# (2) if not pumped |
---|
190 |
good = (5<data['wtemp']) & (data['wtemp']<30) |
---|
191 |
bad = ~good |
---|
192 |
data['wtemp'][bad] = numpy.nan |
---|
193 |
|
---|
194 |
good = (2<data['cond']) & (data['cond']<7) |
---|
195 |
bad = ~good |
---|
196 |
data['cond'][bad] = numpy.nan |
---|
197 |
|
---|
198 |
# press range depends on deployment depth and instrument transducer rating |
---|
199 |
|
---|
200 |
# calculate depth, salinity and density |
---|
201 |
import seawater.csiro |
---|
202 |
import seawater.constants |
---|
203 |
|
---|
204 |
# seawater.constants.C3515 is units of mS/cm |
---|
205 |
# data['cond'] is units of S/m |
---|
206 |
# You have: mS cm-1 |
---|
207 |
# You want: S m-1 |
---|
208 |
# <S m-1> = <mS cm-1>*0.1 |
---|
209 |
# <S m-1> = <mS cm-1>/10 |
---|
210 |
|
---|
211 |
data['depth'] = -1*seawater.csiro.depth(data['press'], platform_info['lat']) # meters |
---|
212 |
data['salin'] = seawater.csiro.salt(10*data['cond']/seawater.constants.C3515, data['wtemp'], data['press']) # psu |
---|
213 |
data['density'] = seawater.csiro.dens(data['salin'], data['wtemp'], data['press']) # kg/m^3 |
---|
214 |
|
---|
215 |
return data |
---|
216 |
|
---|
217 |
def creator(platform_info, sensor_info, data): |
---|
218 |
# |
---|
219 |
# |
---|
220 |
# subset data only to month being processed (see raw2proc.process()) |
---|
221 |
i = data['in'] |
---|
222 |
|
---|
223 |
title_str = sensor_info['description']+' at '+ platform_info['location'] |
---|
224 |
global_atts = { |
---|
225 |
'title' : title_str, |
---|
226 |
'institution' : platform_info['institution'], |
---|
227 |
'institution_url' : platform_info['institution_url'], |
---|
228 |
'institution_dods_url' : platform_info['institution_dods_url'], |
---|
229 |
'metadata_url' : platform_info['metadata_url'], |
---|
230 |
'references' : platform_info['references'], |
---|
231 |
'contact' : platform_info['contact'], |
---|
232 |
# |
---|
233 |
'source' : platform_info['source']+' '+sensor_info['source'], |
---|
234 |
'history' : 'raw2proc using ' + sensor_info['process_module'], |
---|
235 |
'comment' : 'File created using pycdf'+pycdfVersion()+' and numpy '+pycdfArrayPkg(), |
---|
236 |
# conventions |
---|
237 |
'Conventions' : platform_info['conventions'], |
---|
238 |
# SEACOOS CDL codes |
---|
239 |
'format_category_code' : platform_info['format_category_code'], |
---|
240 |
'institution_code' : platform_info['institution_code'], |
---|
241 |
'platform_code' : platform_info['id'], |
---|
242 |
'package_code' : sensor_info['id'], |
---|
243 |
# institution specific |
---|
244 |
'project' : platform_info['project'], |
---|
245 |
'project_url' : platform_info['project_url'], |
---|
246 |
# timeframe of data contained in file yyyy-mm-dd HH:MM:SS |
---|
247 |
# first date in monthly file |
---|
248 |
'start_date' : data['dt'][i][0].strftime("%Y-%m-%d %H:%M:%S"), |
---|
249 |
# last date in monthly file |
---|
250 |
'end_date' : data['dt'][i][-1].strftime("%Y-%m-%d %H:%M:%S"), |
---|
251 |
'release_date' : now_dt.strftime("%Y-%m-%d %H:%M:%S"), |
---|
252 |
# |
---|
253 |
'creation_date' : now_dt.strftime("%Y-%m-%d %H:%M:%S"), |
---|
254 |
'modification_date' : now_dt.strftime("%Y-%m-%d %H:%M:%S"), |
---|
255 |
'process_level' : 'level1', |
---|
256 |
# |
---|
257 |
# must type match to data (e.g. fillvalue is real if data is real) |
---|
258 |
'_FillValue' : -99999., |
---|
259 |
} |
---|
260 |
|
---|
261 |
var_atts = { |
---|
262 |
# coordinate variables |
---|
263 |
'time' : {'short_name': 'time', |
---|
264 |
'long_name': 'Time', |
---|
265 |
'standard_name': 'time', |
---|
266 |
'units': 'seconds since 1970-1-1 00:00:00 -0', # UTC |
---|
267 |
'axis': 'T', |
---|
268 |
}, |
---|
269 |
'lat' : {'short_name': 'lat', |
---|
270 |
'long_name': 'Latitude', |
---|
271 |
'standard_name': 'latitude', |
---|
272 |
'reference':'geographic coordinates', |
---|
273 |
'units': 'degrees_north', |
---|
274 |
'valid_range':(-90.,90.), |
---|
275 |
'axis': 'Y', |
---|
276 |
}, |
---|
277 |
'lon' : {'short_name': 'lon', |
---|
278 |
'long_name': 'Longitude', |
---|
279 |
'standard_name': 'longitude', |
---|
280 |
'reference':'geographic coordinates', |
---|
281 |
'units': 'degrees_east', |
---|
282 |
'valid_range':(-180.,180.), |
---|
283 |
'axis': 'Y', |
---|
284 |
}, |
---|
285 |
'z' : {'short_name': 'z', |
---|
286 |
'long_name': 'Depth', |
---|
287 |
'standard_name': 'depth', |
---|
288 |
'reference':'zero at sea-surface', |
---|
289 |
'positive' : 'up', |
---|
290 |
'units': 'm', |
---|
291 |
'axis': 'Z', |
---|
292 |
}, |
---|
293 |
# data variables |
---|
294 |
'wtemp': {'short_name': 'wtemp', |
---|
295 |
'long_name': 'Water Temperature', |
---|
296 |
'standard_name': 'water_temperature', |
---|
297 |
'units': 'degrees_Celsius', |
---|
298 |
}, |
---|
299 |
'cond': {'short_name': 'cond', |
---|
300 |
'long_name': 'Conductivity', |
---|
301 |
'standard_name': 'conductivity', |
---|
302 |
'units': 'S m-1', |
---|
303 |
}, |
---|
304 |
'press': {'short_name': 'press', |
---|
305 |
'long_name': 'Pressure', |
---|
306 |
'standard_name': 'water_pressure', |
---|
307 |
'units': 'decibar', |
---|
308 |
}, |
---|
309 |
'depth': {'short_name': 'depth', |
---|
310 |
'long_name': 'Depth', |
---|
311 |
'standard_name': 'depth', |
---|
312 |
'reference':'zero at sea-surface', |
---|
313 |
'positive' : 'up', |
---|
314 |
'units': 'm', |
---|
315 |
'comment': 'Derived using seawater.csiro.depth(press,lat)', |
---|
316 |
}, |
---|
317 |
'salin': {'short_name': 'salin', |
---|
318 |
'long_name': 'Salinity', |
---|
319 |
'standard_name': 'salinity', |
---|
320 |
'units': 'psu', |
---|
321 |
'comment': 'Derived using seawater.csiro.salt(cond/C3515,wtemp,press)', |
---|
322 |
}, |
---|
323 |
'density': {'short_name': 'density', |
---|
324 |
'long_name': 'Density', |
---|
325 |
'standard_name': 'density', |
---|
326 |
'units': 'kg m-3', |
---|
327 |
'comment': 'Derived using seawater.csiro.dens0(salin,wtemp,press)', |
---|
328 |
}, |
---|
329 |
} |
---|
330 |
|
---|
331 |
# dimension names use tuple so order of initialization is maintained |
---|
332 |
dim_inits = ( |
---|
333 |
('ntime', NC.UNLIMITED), |
---|
334 |
('nlat', 1), |
---|
335 |
('nlon', 1), |
---|
336 |
('nz', 1), |
---|
337 |
) |
---|
338 |
|
---|
339 |
# using tuple of tuples so order of initialization is maintained |
---|
340 |
# using dict for attributes order of init not important |
---|
341 |
# use dimension names not values |
---|
342 |
# (varName, varType, (dimName1, [dimName2], ...)) |
---|
343 |
var_inits = ( |
---|
344 |
# coordinate variables |
---|
345 |
('time', NC.INT, ('ntime',)), |
---|
346 |
('lat', NC.FLOAT, ('nlat',)), |
---|
347 |
('lon', NC.FLOAT, ('nlon',)), |
---|
348 |
('z', NC.FLOAT, ('nz',)), |
---|
349 |
# data variables |
---|
350 |
('wtemp', NC.FLOAT, ('ntime',)), |
---|
351 |
('cond', NC.FLOAT, ('ntime',)), |
---|
352 |
('press', NC.FLOAT, ('ntime',)), |
---|
353 |
# derived variables |
---|
354 |
('depth', NC.FLOAT, ('ntime',)), |
---|
355 |
('salin', NC.FLOAT, ('ntime',)), |
---|
356 |
('density', NC.FLOAT, ('ntime',)), |
---|
357 |
) |
---|
358 |
|
---|
359 |
# var data |
---|
360 |
var_data = ( |
---|
361 |
('lat', platform_info['lat']), |
---|
362 |
('lon', platform_info['lon']), |
---|
363 |
('z', sensor_info['nominal_depth']), |
---|
364 |
# |
---|
365 |
('time', data['time'][i]), |
---|
366 |
# |
---|
367 |
('wtemp', data['wtemp'][i]), |
---|
368 |
('cond', data['cond'][i]), |
---|
369 |
('press', data['press'][i]), |
---|
370 |
# derived variables |
---|
371 |
('depth', data['depth'][i]), |
---|
372 |
('salin', data['salin'][i]), |
---|
373 |
('density', data['density'][i]), |
---|
374 |
) |
---|
375 |
|
---|
376 |
return (global_atts, var_atts, dim_inits, var_inits, var_data) |
---|
377 |
|
---|
378 |
def updater(platform_info, sensor_info, data): |
---|
379 |
# |
---|
380 |
|
---|
381 |
# subset data only to month being processed (see raw2proc.process()) |
---|
382 |
i = data['in'] |
---|
383 |
|
---|
384 |
global_atts = { |
---|
385 |
# update times of data contained in file (yyyy-mm-dd HH:MM:SS) |
---|
386 |
# last date in monthly file |
---|
387 |
'end_date' : data['dt'][i][-1].strftime("%Y-%m-%d %H:%M:%S"), |
---|
388 |
'release_date' : now_dt.strftime("%Y-%m-%d %H:%M:%S"), |
---|
389 |
# |
---|
390 |
'modification_date' : now_dt.strftime("%Y-%m-%d %H:%M:%S"), |
---|
391 |
} |
---|
392 |
|
---|
393 |
# data variables |
---|
394 |
# update any variable attributes like range, min, max |
---|
395 |
var_atts = {} |
---|
396 |
# var_atts = { |
---|
397 |
# 'wtemp': {'max': max(data.u), |
---|
398 |
# 'min': min(data.v), |
---|
399 |
# }, |
---|
400 |
# 'cond': {'max': max(data.u), |
---|
401 |
# 'min': min(data.v), |
---|
402 |
# }, |
---|
403 |
# } |
---|
404 |
|
---|
405 |
# data |
---|
406 |
var_data = ( |
---|
407 |
('time', data['time'][i]), |
---|
408 |
('wtemp', data['wtemp'][i]), |
---|
409 |
('cond', data['cond'][i]), |
---|
410 |
('press', data['press'][i]), |
---|
411 |
# derived variables |
---|
412 |
('depth', data['depth'][i]), |
---|
413 |
('salin', data['salin'][i]), |
---|
414 |
('density', data['density'][i]), |
---|
415 |
) |
---|
416 |
|
---|
417 |
return (global_atts, var_atts, var_data) |
---|
418 |
# |
---|