Whittle : EXTRAGALACTIC ASTRONOMY
3. DATASETS & GLOBAL PARAMETERS
Extragalactic research often relies on
making use of preexisting data. Knowing whats out there, how to access it,
what it means, and what its limitations are, is an important aspect of
your astronomical "competence". This topic aims to give a brief overview of
the growing databases and catalogues which you may need in the future.
(2) WEB Literature Resources
(3) Major WEB Resources
Rather than describe these in detail, it is best to simply try them out
to see what's available.
- NASA/IPAC Extragalactic
This has a remarkably wide coverage and set of tools : object search;
images; SEDs; catalogues; literature.
Check out the "knowledge base - level 5"
which gives routes into many topics via review articles and books.
- Sloan Digital Sky Survey (SDSS)
This provides a wide range of tools and projects to explore the vast SDSS database (currently, at DR-9 (data-release-9).
- Astronomical Data Center (ADC)
A NASA site for astronomical data and catalogues from the literature.
- Centre de Donees astronomique de
European center for astronomical data and bibliography. Routes to a
variety of other sites.
Gives basic data on objects, with references to all papers which include
- Digital Sky Survey (DSS)
Easy way to view (and/or download) images from the many Schmidt sky surveys.
created by HEASARC, allows you to access images of a region of sky from
many of the all sky multiwavelength surveys.
Ability to view the sky and superimpose many images and source lists
at different wavelengths, and access information and references on these
Great way to find catalogues of interest and download them
(4) Optical Galaxy Catalogs
(a) Important, Currently Used Catalogs
"Third Reference Catalogue of Bright Galaxies" 1991 deVaucouleurs
"Uppsala General Catalog of Galaxies" by Nilsen using PSS (1973)
Data on ~13000 galaxies north of -2.5deg, diameter 1 arcmin or more, or brighter
than 14.5. Morphology and descriptions good. Commonly used. A related
ESO/Uppsala survey of the ESO(B) plates extends the UGC to the south (1982, Lauberts), with densitometry by Lauberts and Valentijn (1989).
"Revised Shapley Ames" (1987) by Sandage and Tammann
Based on the original 1932 Harvard catalog. Positions, morphology, magnitudes,
velocities for ~1200 galaxies brighter than m~13.2 (complete to m=12, but
only 50% at m=12.7). Images illustrate the luminosity classes. Related
catalogs/atlases from Sandage include the Hubble Atlas (1961), the
Carnegie Atlas of Galaxies (1994), the Atlas of Galaxies Useful for
Measuring the Cosmic Distance Scale (NASA 1988). These all have wonderful
large format images.
PGC : "Principal Galaxy Catalogue" 1989 Paturel et al
A gargantuan list of ~73000 galaxies with cross-references, morphologies, sizes,
magnitudes and velocities.
The RC3 uses the PGC number as primary ID.
(b) Previously Important or Less Used Catalogs
NGC : "New(!) General Catalog" originally from Dreyer (1880s), revised
1973 (RNGC) by Sulentic and Tifft
Contains star clusters, galaxies, HII regions, planetary nebulae. The
catalog itself is not much used now, though the names obviously are. Closely
associated is the IC (Index Catalog) of Dreyer which added ~7000 objects. Both
are combined in the NGC 2000 catalog.
The "NGC/IC Project" to clarify all historical NGC & IC information is here: o-link
A compilation of modern NGC data, with links to DSS and SDSS images, is here: o-link
MCG : "Morphological Catalog of Galaxies" 1964, Vorontsov-Velyaminov and Arhipova (Moscow)
Coded descriptions and rough magnitudes for ~29000 galaxies above m=15 and
north of -33. Not much used except for names if no others apply. Careful
with names : MCG 8-11-11 is a different galaxy from MCG -8-11-11.
"Catalog of Galaxies and Clusters of Galaxies" Zwicky (1961-1968) generated from PSS
N hemisphere. Positions and relatively reliable magnitudes (using defocussed
images) for ~31000 galaxies. Nearly complete to mpg=15.5; limit
is 15.7. Also identifies 10000 galaxy clusters. Long used to define galaxy samples (magnitude limited) and for reasonable photometry.
"Atlas of Peculiar Galaxies" Arps famous 1966 atlas
NBG : "Nearby Bright Galaxies" 1988 by Tully.
A catalog specifically devoted to the three dimensional distribution of
Further lists of the major catalogs are given :
(5) All Sky Surveys
It is often important to combine information from several wavebands
One way to do this is to make use of all-sky surveys.
Below is a partial list of the some of the more commonly used ones (see o-link for a full list).
Many can be accessed directly through:
- STScI's MAST (Multimission Archive at STScI): o-link
- NASA's SkyView: o-link
- FIRST (Faint Images of the Radio Sky at Twenty centimeters; 1994-96): o-link
20cm VLA survey of North Galactic Cap (10,000 square degrees).
1 mJy limit and 5 arcsec resolution and 1.8 arcsec pixels. ~106 sources.
- NVSS (Northern VLA Sky Survey; 1994-96): o-link
1.4 Ghz (20cm) VLA survey north of -40 dec (82% of all-sky).
45 arcsec resolution, 15 arcsec pixels, complete to 2.5 mJy, 1.8 million sources
- IRAS (Infrared Astronomical Satellite; 1983): o-link
IRIS (Improved Reprocessing of the IRAS Survey; 2005): o-link
12, 25, 60, 100 micron all-sky images with resolution of ~4 arcmin. 75,000 starburst galaxies found.
- WISE (Wide-field Infrared Survey Explorer; 2010): o-link
4 bands from 3 - 22 microns; 6 - 12 arcsec resolution; 100 × sensitivity of IRAS
Launched Dec 2009, expected data release March 2012
- 2MASS (2 micron All-Sky Survey; 1997-2002): o-link
J,H,K survey of the whole sky. Resolution ~1 arcsec, with 1 arcsec pixels.
- DSS (Digitized Sky Survey; 1945-55 North, 1980-90 South): o-link
Original Palomar and UK Schmidt sky survey plates scanned at STScI in the early 1990s.
Several plate colors/emulsions (E V J R N); most at 1.7 arcsec/pixel.
- Mellinger 3-color allsky image; 2005: o-link
Alex Mellinger mosaic of 3000 SBIG CCD frames covering 70 fields of 40×27 degrees.
This is primarily for visual aesthetic, the above link is to a flash pan/zoom
- SDSS (Sloan Digital Sky Survey; 2000-08): o-link
1/4 sky; north galactic cap; 5 color imaging; full optical spectroscopy. See Section 8
- GALEX (Galaxy Evolution Explorer; 2003-07): o-link
Two bands: 1350-1780 A (FUV) and 1770-2730 (NUV) with resolutions 4.3 & 5.3 arcsec with 1.5 arcsec/pix.
- EUVE (Extreme Ultraviolet Explorer; 1991-93): o-link
All-sky survey in 4 UV bands: 80, 170, 400, 550 Angstroms. 2 arcmin resolution; 1.5 arcmin pixels.
- RASS (ROSAT All Sky Survey; 1991): o-link
3 soft X-ray bands: 0.25 0.75 1.5 keV; ~2 degree resolution, 45 arcsec/pix; 60,000 sources found.
- RXTE (Rosse X-ray Timing Explorer; 1996-2002): o-link
2 Xray bands: 3-8 kev and 8-20 keV. Resolution ~1 degree; uneven exposure times.
- CGRO (Compton Gamma Ray Observatory; 1991-94): o-link
Two Instruments: COMPTEL: 1-30 MeV resolution 3 deg; EGRET: 1MeV-10GeV, resolution 2-5 deg.
- Fermi (Fermi Gamma Ray Space Telescope; 2010): o-link
Next generation beyond CGRO; 20 MeV - 100 GeV; Whole sky every 3 hours.
(6) Selection Effects in Surveys and Catalogs
The principal criteria which exclude galaxies from surveys and/or catalogs are :
(a) Malmquist Bias
- Flux (apparent magnitude) limited samples always over-represent high luminosity objects.
These can be detected to larger distances and therefore a larger volume is included in the sample.
Hence, the observed distribution of lumiosities is artificially skewed to high luminosity.
- This figure [image] (from Keel) illustrates
the effect for the simple case of
(a) uniform spatial distribution, and
A line of constant (limiting) magnitude is shown.
(b) Gaussian distribution of absolute magnitude
Only objects above the line are included in the sample.
- Similar kinds of bias result from other selection effects (eg diameter,
SB), and in general are very widespread in astronomy.
- In general "selection functions" must be applid to correct the bias.
These can be difficult and uncertain.
To illustrate, Trimble (1994, PASP 108 1073) remarks : "Any large gathering
of observational cosmologists today will include at least one person who
thinks that someone else in the room does not understand the Malmquist
(7) Global Parameters
There are a number of standard global parameters which galaxy catalogues often
It can be quite complex to convert/reduce to a standard system (eg the RC3 introduction is long and detailed!)
Lets briefly review these, using the RC3 as example.
(b) Angular Sizes
- D25 : angular diameter to 25th B mag/ss ispohote (major axis)
this is ~few % of sky and is about the size you would guess from the PSS
- D0 : same as D25 but corrected for galactic extinction
and galaxy inclination (ie corrected to "face on")
- Ae : "Effective" (circular) aperture diameter which would enclose
half the total light (see below)
- R25 : major/minor axis ratio at the 25th B mag/ss isophote
Note that conversion to inclination is not simply inverse sine (several
prescriptions are used)
- BT : Total photoelectric B magnitude, derived either
from aperture photometry or surface photometry (CCD or calibrated photographic).
Aperture data requires fitting to
"growth curves" [image] allowing extrapolation to
"infinite" aperture, or "Total" magnitude.
- BTo : Total B magnitude, corrected for three effects
- galactic absorption, Ag,
- internal absorption (to face on), Ai,
- redshift: "K correction", since redshift brings a bluer part of the spectrum into the B filter
this correction depends on both z and Hubble type (since different Hubble types have different spectra).
- mB : Total photographic B magnitude. Since many sources of
magnitudes (eg Zwicky's 31000 CGCG galaxies) are photographic, RC3 includes
them but separately from the photoelectric magnitudes. They are less accurate than BT, and need to be reduced to the standard system.
- Ag : Galactic extinction in B band. After a complex history,
RC3 uses the reddening maps of Burstein &
Heiles (1978) derived from HI maps and faint galaxy counts. Since then,
improved (by factor 2) maps have been derived from HI maps and FIR maps
of COBE/DIRBE and IRAS.
A NED calculator gives the galactic absorption at any location (o-link), and shown here: [image].
- Ai : Internal Extinction in the B band - ie extinction due to
dust in the galaxy itself. This is much less well known, but is clearly a
function of (a) the Hubble type and (b) the inclination. After a complex
history, RC3 gives a prescription.
Note that RC3 estimates
extinction to "face on", while RSA estimates extinction to
"no dust" (and therefore much higher values, particularly for later Hubble types)
- (U-B)T; (B-V)T; (U-B)e; (B-V)e
: are colors for Total galaxy light, and effective (half light)
apertures. They are also corrected for galactic and internal extinction and
redshift, yielding (U-B)To etc.
(f) HI Kinematics and Fluxes
- W20; W50 :
The width of the integrated 21cm
HI line profile, measured at the 20% and 50% height levels [image]. This
measures the (projected) galaxy rotation velocity, and is an
important parameter in, for example, the Tully-Fisher relation.
- m21 : HI "magnitude" is a measure of the integrated
HI 21cm line flux. This is corrected for internal self-absorption
(A21) and a redshift (1+z) to yield m21o.
The quantity m21o - BTo then
represents an HI/LB ratio, which is a standard
galaxies (and increases along the Hubble sequence).
- Vopt; V21 : Optical and 21cm systemic
heliocentric velocities (always given as cz). Usually, these agree well.
- VGSR : velocity (cz, weighted mean of Vopt & V21)
referred to the Galactic Standard of Rest (the center of the galaxy).
This combines a transformation from HC to LSR (16.5 km/s towards l=53,
b=+25) with a further transformation to the galactic center (220 km/s
towards l=90, b=0). This can be used for distance estimates
using the Hubble constant.
- V3K : velocity referred to the frame of the CMB (microwave
background): a tranformation from HC by 369 km/s towards
l=264 b=+48. This may be better for distance estimates beyond about 5000 km/s.
Note, for cz within ~5000 km/s people sometimes use a "Virgocentric Infall"
model to correct both our and the galaxy's redshifts for large scale flows
before applying the Hubble law.
(8) Sloan Digital Sky Survey: SDSS
The SDSS is such a huge and remarkable survey that we should briefly review. [o-link]
(a) Survey Overview o-link
- Dedicated telescope: 2.5m f/5 at Apache Point Observatory (APO): [image]
- Specially designed camera: 1.5 square degrees, 120 Mpix, 30 × 20482 CCDs with u,g,r,i,z filters.
Driftscan method: clock out the CCDs at the same rate the telescope scans a great circle.
Each location gets 54 seconds of exposure (only!) in each filter.
- Two 320 fiber spectrographs use aluminum plates with holes plugged with 3 arcsec fibers.
Dichroics split spectrum into separate blue and red channels: 3800 - 9200A coverage with R ~ 2000
- Survey (SDSS-I & SDSS-II) took 8 years (2000 - 2008), and covered 1/4 of the sky (8400 sq deg)[image]
- Photometry of 230 million objects (stars & galaxies)
- Spectra of: 930,000 galaxies; 120,000 stars; 120,000 quasars
- Spectra Limits: galaxies r < 17.8; quasars i < 19.1
- Imaging limits: u,g,r ~ 22; i ~ 21.3, z ~ 20.3
(b) SDSS Parameters
The scale and nature of the SDSS demands a careful approach to parameter definitions.
Ultimately, there are many parameters and their errors (and flags) defined
Here are just a few (taken from here: o-link)
SDSS defines all magnitudes using an asinh (= ln [x + (x2 + 1)] ) function (see reference).
For S/N > 5 asinh magnitudes = normal magnitudes (i.e. -2.5 Log (flux), Pogson 1856)
They are therefore much better behaved at low flux levels (avoiding the divergence of Log 0).
For S/N < 5 asinh magnitudes are ~linear in flux.
Fit pure deVauc & exponential deVMag, expMag
Best fit linear combination of these cmodelMag
Petrosian magnitude (petroMag)
Define petroRad = rp by I(rp)/< I(< rp) > = 0.2 where I is surface brightness in flux units.
Then petroMag = flux inside 2rp
for exponential profile petroMag gets ~all the flux, for a deVauc profile it gets ~80%
Virtue: measures constant fraction of flux regardless of distance (or size), and is insensitive to sky noise.
Concentration: = petro90/petro50
where these are radii enclosing 90% and 50% of the petrosian flux
Goodness of fit Likelihoods: deV_L, exp_L, star_L
use f(deV_L) = deV_L / (deV_L + exp_L + star_L) > 0.5 to select early type (deVauc) profiles.
Ellipticities and PA:
Measured from second moments or 25th isophote shape.
There are also many flags to identify problems with the analysis of a given object.
Many emission & absorption line strengths, and continuum strengths
Redshifts, from both emission and absorption lines
Velocity dispersion from absorption lines (only early type, bulge dominated, spectra).
Classification of type (early, late, agn, etc) by comparing to a number of templates.
(c) Web Access and Analysis
The entire dataset is available online (o-link)
It can be interrogated in several ways; here are just two:
(d) Other large/deep galaxy surveys
The SDSS is not the only large deep survey. Here are some others:
- 2dFGRS: o-link
1500 square degrees in two stripes in N & S galactic polar regions.
Photometry for 380,000 bJ < 19.5 galaxies and redshifts for 232,000 galaxies.
- The Millenium Galaxy Catalogue: o-link
sub-region of SDSS & 2dFGRS: 10h 00m - 14h 45m × 35 arcmin.
Deep (750s) B images of 10,095 galaxies with B < 24 and redshifts for all B < 20.
- The Combo-17 Survey: o-link
1 degree square with deep (~3 hr on 2.2m) images in 17 filters:
classification & redshifts (photo-z) for 25,000 galaxies and 300 QSOs.
Galaxy & cluster evolution to z~1; Weak lensing; QSO evolution.
- HST:: there are several deep small-area surveys by HST:
HUDF: Hubble Ultra Deep Field: 11 sq arcmin; 400 orbits with ACS: o-link
GOODS: The Great Observatories Origins Deep Survey: 320 sq arcmin; HST, Spitzer, Chandra... o-link
COSMOS: The Cosmic Evolution Survey: 2 sq deg; 2 million gals; o-link
(e) Some Papers from SDSS and MGC
Some of the important papers emerging from the SDSS and MGC surveys are listed: here