-
Notifications
You must be signed in to change notification settings - Fork 1
/
am_fit.py
362 lines (297 loc) · 13.6 KB
/
am_fit.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
# coding: utf-8
import os
import glob
import math
import numpy as np
import numpy.fft as fft
import pylab as py
import array
import pyspeckit
from astropy.io import fits
import matplotlib.pyplot as plt
from astropy.table import Table
from scipy.ndimage.filters import median_filter
from kdist import *
from NH3FirstGuess import *
from parse_coords import *
"""
Variables: data* = spectral data extracted from .fits file
A* = array used to calculate nu*
nu* = frequency
v* = doppler shift velocity
guess = "guesses" of fit paramaters to pass through pyspeckit; derived via cross-correlation
spec* = Spectrum object created from pyspeckit
spectrum = dictionary to pass through NH3 fitting from pyspeckit
Note: * is the value of the transition; ie. 1 = (1,1), 2 = (2,2), etc.
Output: nh3dict = Dictionary of the entire spectrum
Histogram of fit parameters
t_par = table of fit parameters
t_int = table of integrated intensities
t_W11 = table of W11; with errors
t_errs = table of errors of fit parameters
creates fitted spectra plots and saves it all into a directory
Bivariate plots with fit parameters
(╯°□°）╯︵ ┻━┻ (╯°□°）╯︵ ┻━┻ (╯°□°）╯︵ ┻━┻ (╯°□°）╯︵ ┻━┻ (╯°□°）╯︵ ┻━┻
"""
# Whole set of data
fileNames = glob.glob('./nh3_all/*fits')
# Small section of data
#fileNames = glob.glob('./nh3_all/G049*.fits')
a = np.arange(len(fileNames))
objects = [((os.path.basename(fileNames[name])))[0:-9] for name in range(max(a))]
objects = sorted(set(objects))
# Tables; Note: We can extract an array from the astropy.Tables that are generated; ex. t_pars['TKIN']
t_int = Table(names=('FILENAME','W11','W22','W33','W44'),dtype=('S20','f5','f5','f5','f5'))
t_w11 = Table(names=('FILENAME','W11OBS','W11EMP','RMSOBS','RMSEMP','DIFF','PERCERR'),dtype=('S20','f5','f5','f5','f5','f5','f5'))
t_pars = Table(names=('FILENAME','TKIN','TEX','N','SIGV','V','F'),dtype=('S20','f5','f5','f5','f5','f5','f1'))
t_errs = Table(names=('FILENAME','TKINERR','TEXERR','NERR','SIGVERR','VERR','FERR'),dtype=('S20','f5','f5','f5','f5','f5','f1'))
t_dist = Table(names=('FILENAME','DIST','RGAL','G.LONG','G.LAT'),dtype=('S20','f5','f5','f5','f5'))
c = 2.99792458e8
voff_lines = np.array([19.8513,
19.3159,
7.88669,
7.46967,
7.35132,
0.460409,
0.322042,
-0.0751680,
-0.213003,
0.311034,
0.192266,
-0.132382,
-0.250923,
-7.23349,
-7.37280,
-7.81526,
-19.4117,
-19.5500])
tau_wts = np.array([0.0740740,
0.148148,
0.0925930,
0.166667,
0.0185190,
0.0370370,
0.0185190,
0.0185190,
0.0925930,
0.0333330,
0.300000,
0.466667,
0.0333330,
0.0925930,
0.0185190,
0.166667,
0.0740740,
0.148148])
deltanu = -1*voff_lines/((c/1000)*23.6944955e9)
vmin = -250
vmax = 250
for thisObject in objects:
spectrum = {}
fnameT = './nh3_figures/'+thisObject+'.png'
fnameT2 = './nh3_figures2/'+thisObject+'.png'
fnameT3 = './nh3_figures3/'+thisObject+'.png'
if os.path.exists('./nh3_all/'+thisObject+'.n11.fits'):
data1 = fits.getdata('./nh3_all/'+thisObject+'.n11.fits')
A1 = np.arange(len(data1['DATA'].T))
nu1 = data1['CDELT1']*(A1-data1['CRPIX1']+1)+data1['CRVAL1']
v1 = c*(1-nu1/data1['RESTFREQ'])
spec1 = pyspeckit.Spectrum(data=data1['DATA'].T.squeeze(),unit='K',xarr=v1,xarrkwargs={'unit':'m/s','refX':data1['RESTFREQ']/1E6,'refX_units':'MHz','xtype':'VLSR-RAD'})
spectrum['oneone'] = spec1
# Cross-correlation method to get first guess
linewidth = 0.5
chanwidth = (spec1.xarr[1]-spec1.xarr[0])/1e3
ftdata = fft.fft(spec1.data.filled(0))
tvals = fft.fftfreq(len(spec1.data))/chanwidth
deltafcns = np.zeros(spec1.data.shape,dtype=np.complex)
for idx, dv in enumerate(voff_lines):
deltafcns += tau_wts[idx]*(np.cos(2*np.pi*dv*tvals)+
1j*np.sin(2*np.pi*dv*tvals))*\
np.exp(-tvals**2*(linewidth/chanwidth)**2/(2))
ccor = np.real((fft.ifft(np.conj(ftdata)*deltafcns))[::-1])
vaxis = np.array(spec1.xarr.as_unit('km/s'))
subsetidx = (vaxis>vmin)*(vaxis<vmax)*np.isfinite(np.array(spec1.data))
peakIndex = np.argmax(ccor[subsetidx])
#pull out a 6 km/s slice around the peak
deltachan = np.abs(3.0 / chanwidth)
t = ((spec1.data.filled(0))[subsetidx])[(peakIndex-deltachan):(peakIndex+deltachan)]
v = (vaxis[subsetidx])[(peakIndex-deltachan):(peakIndex+deltachan)]
# Calculate line width.
sigv = np.sqrt(abs(np.sum(t*v**2)/np.sum(t)-(np.sum(t*v)/np.sum(t))**2))
if (np.isnan(sigv)):
sigv = 0.85840189 # mean of sigv from first set of data from ./nh3_all/
# Peak of cross correlation is the brightness.
v0 = np.float((vaxis[subsetidx])[peakIndex])
# Set the excitation temperature to be between CMB and 20 K
# and equal to the peak brightness + 2.73 if valid.
tex = np.min([np.max([spec1.data.filled(0)[peakIndex],0])+2.73,20])
guess = [20, # 20 K kinetic temperature
tex, #
15, # Log NH3
sigv, # velocity dispersion
v0,
0.5]
if os.path.exists('./nh3_all/'+thisObject+'.n22.fits'):
data2 = fits.getdata('./nh3_all/'+thisObject+'.n22.fits')
A2 = np.arange(len(data2['DATA'].T))
nu2 = data2['CDELT1']*(A2-data2['CRPIX1']+1)+data2['CRVAL1']
v2 = c*(1-nu2/data2['RESTFREQ'])
spec2 = pyspeckit.Spectrum(data=data2['DATA'].T.squeeze(),unit='K',xarr=v2,xarrkwargs={'unit':'m/s','refX':data2['RESTFREQ']/1E6,'refX_units':'MHz','xtype':'VLSR-RAD'})
spectrum['twotwo'] = spec2
if os.path.exists('./nh3_all/'+thisObject+'.n33.fits'):
data3 = fits.getdata('./nh3_all/'+thisObject+'.n33.fits')
A3 = np.arange(len(data3['DATA'].T))
nu3 = data3['CDELT1']*(A3-data3['CRPIX1']+1)+data3['CRVAL1']
v3 = c*(1-nu3/data3['RESTFREQ'])
spec3 = pyspeckit.Spectrum(data=data3['DATA'].T.squeeze(),unit='K',xarr=v3,xarrkwargs={'unit':'m/s','refX':data3['RESTFREQ']/1E6,'refX_units':'MHz','xtype':'VLSR-RAD'})
spectrum['threethree'] = spec3
if os.path.exists('./nh3_all/'+thisObject+'.n44.fits'):
data4 = fits.getdata('./nh3_all/'+thisObject+'.n44.fits')
A4 = np.arange(len(data4['DATA'].T))
nu4 = data4['CDELT1']*(A4-data4['CRPIX1']+1)+data4['CRVAL1']
v4 = c*(1-nu4/data4['RESTFREQ'])
spec4 = pyspeckit.Spectrum(data=data4['DATA'].T.squeeze(),unit='K',xarr=v4,xarrkwargs={'unit':'m/s','refX':data4['RESTFREQ']/1E6,'refX_units':'MHz','xtype':'VLSR-RAD'})
spectrum['fourfour'] = spec4
spdict1,spectra1 = pyspeckit.wrappers.fitnh3.fitnh3tkin(spectrum,dobaseline=False,guesses=guess,fixed=[False,False,False,False,False,True])
glong, glat = parse_coords(data1)
# Filters out good and bad fits
if -150 < spectra1.specfit.modelpars[4] < 150:
# Further filtering out bad fits with Tk < 8 and Tex < 3
if spectra1.specfit.modelpars[0] < 7.5:
plt.savefig(fnameT2.format(thisObject), dpi = 100, format='png')
plt.close()
elif spectra1.specfit.modelpars[1] < 2.9:
plt.savefig(fnameT2.format(thisObject), dpi = 100, format='png')
plt.close()
else:
# The good fits are stored in tables
spec_pars = [thisObject,spectra1.specfit.modelpars[0],spectra1.specfit.modelpars[1],spectra1.specfit.modelpars[2],spectra1.specfit.modelpars[3],spectra1.specfit.modelpars[4],spectra1.specfit.modelpars[5]]
spec_errs = spectra1.specfit.modelerrs
spec_errs.insert(0,thisObject)
# Distances and galactic coordinates
distance, rgal = kdist(np.atleast_1d(glong), np.atleast_1d(glat), np.atleast_1d(spectra1.specfit.modelpars[4]), rrgal = True)
d_row = [thisObject,distance,rgal,glong,glat]
# Error calculation for W11 between observational and empirical
W11_oarr = spec1.specfit.model
W11_obs = np.sum(W11_oarr)*(v1.max()-v1.min())/(len(v1)*1000)
W11_index = np.where(W11_oarr > 1e-6)
W11_emp = np.sum(spec1.data[W11_index])*(v1.max()-v1.min())/(len(v1)*1000)
W11_diff = W11_obs - W11_emp
W11_perc = abs(((W11_obs - W11_emp)*100)/W11_obs)
W11_oerr = np.nanstd(W11_oarr)
NoSignal = np.where(W11_oarr < 1e-6)
W11_eerr = np.nanstd(spec1.data[NoSignal])
W11_row = [thisObject,W11_obs,W11_emp,W11_oerr,W11_eerr,W11_diff,W11_perc]
# Integrated intensities for all transitions
w_int = [thisObject,np.sum(spec1.specfit.model)*(v1.max()-v1.min())/(len(v1)*1000),np.sum(spec2.specfit.model)*(v2.max()-v2.min())/(len(v2)*1000),None,None]
if os.path.exists('./nh3/'+thisObject+'.n33.fits'):
w_int[3] = np.sum(spec3.specfit.model)*(v3.max()-v3.min())/(len(v3)*1000)
if os.path.exists('./nh3/'+thisObject+'.n44.fits'):
w_int[4] = np.sum(spec4.specfit.model)*(v4.max()-v4.min())/(len(v4)*1000)
t_pars.add_row(spec_pars)
t_errs.add_row(spec_errs)
t_w11.add_row(W11_row)
t_int.add_row(w_int)
t_dist.add_row(d_row)
plt.savefig(fnameT.format(thisObject), dpi = 100, format='png')
plt.close()
else:
plt.savefig(fnameT2.format(thisObject), dpi = 100, format='png')
plt.close()
# Save tables after loop is done; note we get errors as we can't overwrite it
t_pars.write('./nh3_tables/nh3_pars.fits',format='fits')
t_errs.write('./nh3_tables/nh3_errs.fits',format='fits')
t_w11.write('./nh3_tables/nh3_w11.fits',format='fits')
t_int.write('./nh3_tables/nh3_int.fits',format='fits')
t_dist.write('./nh3_tables/nh3_dist.fits',format='fits')
"""
#plotting
pars = fits.getdata('./nh3_tables/final/nh3_pars.fits')
errs = fits.getdata('./nh3_tables/final/nh3_errs.fits')
dist = fits.getdata('./nh3_tables/final/nh3_dist.fits')
w11 = fits.getdata('./nh3_tables/final/nh3_w11.fits')
wint = fits.getdata('./nh3_tables/final/nh3_int.fits')
f, axarr = plt.subplots(2,sharex=True)
axarr[0].scatter(dist['RGAL']/1000,pars['N'])
axarr[0].set_ylabel('Column Density (log(N))')
#axarr[0].set_ylim([0,55])
axarr[0].set_xlim([0,15])
axarr[1].scatter(dist['RGAL']/1000,pars['TEX'])
axarr[1].set_ylabel('Excitation Temperature (K)')
axarr[1].set_xlabel('Galactocentric Distance (kpc)')
#axarr[1].set_ylim([-65,140])
axarr[1].set_xlim([0,15])
# Fit parameter histograms
plt.clf()
py.hist(t_pars['TKIN'],bins=100)
plt.xlabel('Kinetic Temperature (K)')
plt.ylabel('Numbers')
plt.title('Histogram of Kinetic Temperatures ($T_k$)')
plt.savefig('./ammonia_plots/histogram_tkin.png', format='png')
plt.close()
plt.clf()
py.hist(t_pars['TEX'],bins=100)
plt.xlabel('Excitation Temperature (K)')
plt.ylabel('Numbers')
plt.title('Histogram of Excitation Temperatures ($T_{ex}$)')
plt.savefig('./ammonia_plots/histogram_tex.png', format='png')
plt.close()
plt.clf()
py.hist(t_pars['N'],bins=100)
plt.xlabel('Column Density')
plt.ylabel('Numbers')
plt.title('Histogram of Column Density ($log(N)$)')
plt.savefig('./ammonia_plots/histogram_N.png', format='png')
plt.close()
plt.clf()
py.hist(t_pars['SIGV'],bins=100)
plt.xlabel('Line Width ($cm^{-2}$)')
plt.ylabel('Numbers')
plt.title('Histogram of Line Width ($\sigma$)')
plt.savefig('./ammonia_plots/histogram_sigma.png', format='png')
plt.close()
plt.clf()
py.hist(t_pars['V'],bins=100)
plt.xlabel('Line-of-Sight Velocity (km/s)')
plt.ylabel('Numbers')
plt.title('Histogram of Line-of-Sight Velocity ($v$)')
plt.savefig('./ammonia_plots/histogram_vlos.png', format='png')
plt.close()
# Scatter plots with fit parameters
plt.clf()
plt.scatter(t_pars['TKIN'],t_pars['SIGV'])
plt.xlabel('Kinetic Temperature (K)')
plt.ylabel('Line Width ($cm^{-2}$)')
plt.title('Kinetic Temperature ($T_k$) and Line Width ($\sigma$)')
plt.savefig('./ammonia_plots/tkin_vs_sigma.png', format='png')
plt.close()
plt.clf()
plt.scatter(t_pars['TKIN'],t_pars['TEX'])
plt.xlabel('Kinetic Temperature (K)')
plt.ylabel('Excitation Temperature (K)')
plt.title('Kinetic Temperature ($T_k$) vs Excitation Temperature ($T_{ex}$)')
plt.savefig('./ammonia_plots/tkin_tex.png', format='png')
plt.close()
plt.clf()
plt.scatter(t_pars['N'],t_pars['TKIN'])
plt.ylabel('Kinetic Temperature (K)')
plt.xlabel('Column Density (log(N))')
plt.title('Column Density ($log(N)$) vs Kinetic Temperature ($T_k$)')
plt.savefig('./ammonia_plots/N_vs_tkin.png', format='png')
plt.close()
# assume gamma = 1 cause its isothermal
kb = 1.3806488E-23
m = 2.82E-26
c_s = np.zeros(len(t_pars['TKIN']),dtype = np.float64)
Ma = np.zeros(len(t_pars['TKIN']),dtype = np.float64)
for i in range(0,len(t_pars['TKIN'])):
c_s[i] = math.sqrt(kb*t_pars['TKIN'][i]/m)
Ma[i] = t_pars['SIGV'][i]/(np.float(c_s[i])/1000)
plt.clf()
plt.scatter(Ma,t_pars['TKIN'])
plt.ylabel('Kinetic Temperature (K)')
plt.xlabel('Mach Number')
plt.title('Mach Number ($Ma$) vs Kinetic Temperature ($T_k$)')
plt.savefig('./ammonia_plots/Ma_vs_tkin.png', format='png')
plt.close()
"""