CAStArchive/COMBO17.html at main · Astroinformatics/CAStArchive · GitHub

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
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<!--AstroStatistics Home page-->
  <title>CASt: COMBO-17 dataset</title>
<!-- InstanceEndEditable -->
  <meta http-equiv="Content-Type" content="text/html;">
  <meta name="keywords"
 content="Center for Astrostatistics, CASt,
    Statcodes, SCMA, VoStat">
  <link href="assets/cast.css" rel="stylesheet" type="text/css">
</head>
<body style="background-color: rgb(255, 248, 229);" leftmargin="0"
 topmargin="0">
<!--body style="background-color: #FFF8E5"-->
<p><img src="assets/HmPg_Ban.gif" alt="Astrostatistics Image" usemap="#Map"
 href="http://www.stat.psu.edu" title=""
 style="border: 0px solid ; width: 700px; height: 107px;" hspace="0"
 vspace="0"> <map name="Map">
<area shape="rect" coords="3,12,126,98" href="http://www.psu.edu"
 target="_blank" alt="Penn State University">
<area shape="rect" coords="360,0,508,17"
 href="http://www.science.psu.edu" alt="Eberly College of Science">
<area shape="rect" coords="508,0,700,98"
 href="http://astrostatistics.psu.edu" alt="Center for Astrostatistics">
<area shape="rect" coords="127,18,508,98"
 href="http://astrostatistics.psu.edu" alt="Center for Astrostatistics">
</map>
</p>
<!-- InstanceBeginEditable name="head" -->
<p>&nbsp;<!--center--><a href="index.html">Home</a></p>
<p> <img src="http://astrostatistics.psu.edu/rainbow.gif" border="0"
 height="4" vspace="2" width="700"><!-- InstanceBeginEditable name="PageTitle" -->
<!-- InstanceEndEditable --><!-- InstanceBeginEditable name="body" --></p>
<table border="0" cellspacing="0" width="700">
  <tbody>
    <tr>
      <td style="vertical-align: top;">&nbsp; </td>
      <td>
      <h2 style="text-align: center;">COMBO-17 galaxy dataset</h2>
      <h3>The CASt dataset</h3>
      <a href="COMBO17.dat" target="_blank">COMBO17.dat</a><br>
      <a href="COMBO17.csv" target="_blank">COMBO17.csv</a><br>
      <br>
      <p><span style="font-weight: bold;">Astronomical background </span></p>
      <p>Galaxies are fundamental structures in the Universe.&nbsp; Our
Sun lives in the Milky Way Galaxy we can see as a patchy band of light
across the sky.&nbsp; The components of a typical galaxy are: a vast
number of stars (total mass ~10<sup>6</sup>-10<sup>11</sup> M<sub>o</sub>
where M<sub>o</sub> is the unit of a solar mass), a complex
interstellar medium of gas and dust from which stars form (typically
1-100% of the stellar component mass), a single supermassive black hole
at the center (typically &lt;1% of the stellar component mass), and a
poorly understood component called Dark Matter with mass ~5-10-times
all the other components combined. </p>
      <p>Over the ~14 billion years since the Big Bang,&nbsp; the rate
at which
galaxies convert interstellar matter into stars has not been constant,
and thus the brightness and color of galaxies change with cosmic
time.&nbsp; This phenomenon has several names in the astronomical
community: the history of star formation in the Universe, chemical
evolution of galaxies, or simply galaxy evolution.&nbsp; A major effort
over several decades has been made to quantify and understand galaxy
evolution using telescopes at all wavelengths. </p>
      <p>The traditional tool for such studies has been optical
spectroscopy
which easily reveals signatures of star formation in nearby
galaxies.&nbsp; However, to study star formation in the galaxies
recently emerged after the Big Bang, we must examine extremely faint
galaxies which are too faint for spectroscopy, even using the biggest
available telescopes.&nbsp; A feasible alternative is to obtain images
of faint galaxies at random locations in the sky in narrow spectral
bands, and thereby construct crude spectra.&nbsp; First, statistical
analysis of such multiband photometric datasets are used to classify
galaxies, stars and quasars.&nbsp; Second, for the galaxies,
multivariate regression is made to develop photometric estimates of
redshift, which is a measure both of distance from us and age since the
Big Bang.&nbsp; Third, one can examine galaxy colors as a function of
redshift (after various corrections are made) to study the evolution of
star formation. The present dataset is taken after these first two
steps are complete.</p>
      <br>
      <p><span style="font-weight: bold;">Dataset</span></p>
      <p><a href="http://arxiv.org/abs/astro-ph/0403666">Wolf et al.
(2004)</a>
provide the first public catalog of a large dataset
(63,501 objects) with brightness measurements in 17 bands in the
visible band.&nbsp; (Note that the Sloan Digital Sky Survey provides a
much larger dataset of 10<sup>8</sup> objects with measurements in 5
bands.)&nbsp; We provide here a subset of their catalog with 65 columns
of information on 3,462 galaxies.&nbsp; These are objects in the
Chandra Deep Field South field which Wolf and colleagues have
classified as `Galaxies'.&nbsp; The column headings are formally
described in their Table 3, and the columns we provide are summarized
here with brief commentary:</p>
      <p>&nbsp;&nbsp;&nbsp; Col 1:&nbsp; Nr, object number</p>
      <p>&nbsp;&nbsp;&nbsp; Col 2-3: Total R (red band) magnitude and
its
error.&nbsp; This was the band at which the basic catalog was
constructed.&nbsp; Magnitudes are inverted logarithmic measures of
brightness.&nbsp; A&nbsp; galaxy with R=21 is 100-times brighter than
one with R=26.&nbsp; The error is the standard deviation derived from
detailed knowledge of the measurement process.&nbsp; This dataset is an
excellent example of astronomical datasets where each variable is
accompanied by heteroscedastic measurement errors of known
variances.&nbsp; </p>
      <p>&nbsp;&nbsp;&nbsp; Col 4-5:&nbsp; ApDRmag is the difference
between the
total and aperture magnitude in the R band.&nbsp; This is a rough
measure of the size of the galaxy in the image where ApDRmag=0
corresponds to a point source.&nbsp; Negative values are not physically
meaningful.&nbsp; mu_max is the central surface brightness of the
object in the R band.&nbsp; The difference between Rmag and mu_max
should also be an indicator of galaxy size.<br>
&nbsp;&nbsp; Col 6-9: Mcz and MCzml are two redshift estimates.&nbsp;
Mcz is the preferred value.&nbsp; e.Mcz is its estimated error, and
chi2red is the reduced chi-squared value of the least-squares fit of
the 17-band magnitudes to the best-fit template galaxy spectrum.&nbsp;
Galaxies with large e.Mcz or chi2red might be omitted as unreliable.</p>
      <p>&nbsp;&nbsp; Col 10-29: These give the absolute magnitudes
(i.e.
intrinsic luminosities) of the galaxy in 10 bands, with their
measurement errors.&nbsp; They are based on the measured magnitudes and
the redshifts, and represent the intrinsic luminosities of the
galaxies; a galaxy with M=-15 is 100-times less luminous than one with
M=-20. &nbsp;&nbsp; These magnitudes are not all independent of each
others, but the are important for representing intrinsic properties of
the galaxies.&nbsp; Below is one of several redshift-stratified plots
of the B-band absolute magnitude (abscissa) against the difference of
magnitude (i.e. ratio of luminosities) between the 2800A ultraviolet
and blue band, which is a sensitive indicator of star formation.&nbsp;
A redshift-dependent bimodal distribution is seen.&nbsp; </p>
      <div style="text-align: center;"><img
 alt="Galaxy color distribution" src="COMBO17.jpg"
 style="width: 532px; height: 490px;"></div>
      <p>&nbsp;&nbsp; Col 30-55:&nbsp;
Observed
brightnesses in 13 bands in sequence from 420 nm in the ultraviolet to
915 nm in the far red.&nbsp; These are given in linear variables with
units of photon flux densities, photons/m<sup>2</sup>/s/nm.&nbsp;
Again, each measurement is accompanied by a measurement error which can
be used to distinguish measurement from intrinsic dispersions in the
distributions.</p>
      <p>&nbsp;&nbsp; Col 56-65: Observed brightnesses in 5 traditional
broad
spectral bands, UBVRI.&nbsp; These are largely redundant with the 13
bands in the previous columns.&nbsp; </p>
      <br>
      <p><span style="font-weight: bold;">Statistical exercises </span></p>
      <ul>
        <li>Examine basic characteristics of the survey which are not
of
scientific interest.&nbsp; These include the absence of high-redshift
(i.e. distant) high-absolute-magnitude (i.e. faint) galaxies; the
dropoff in flux with redshift; the dropoff in image size with redshift.<br>
        </li>
        <li>Reproduce the plot above, segregate the two populations of
redder
and bluer galaxies. and characterize each population in the 13-band
space.</li>
        <li>Study these two populations as a function of redshift to
investigate the evolution of star formation. <br>
        </li>
      </ul>
      </td>
    </tr>
  </tbody>
</table>
</body>
</html>