SLOAN DIGITAL SKY SURVEY

1
Probing Large-scale Structure with The Sloan
Digital Sky Survey
Josh Frieman
AAAS, Seattle, February, 2004
2
(No Transcript)
3
Progress in Sky Surveys
The advent of telescopes led astronomers to find
catalog non-stellar objects nebulae and
clusters (Messier, Dunlop, Herschels,
Dreyers NGC, ) Advances in telescope optics
and photography led to deeper catalogs
covering larger portions of the sky (e.g.,
Schmidt telescopes with large FOV) Recent
advances in detector technology (CCDs) and
computing power have ushered in the new age of
digital sky surveys. 1780 Messier 110
objects 2005 SDSS 108 objects
1700s 1800s
1900s
2000s
4
Why Survey the Universe Now?
By determining how the matter of the Universe
is distributed in space, we can help address
some basic questions in cosmology
5
Why Survey the Universe Now?
By determining how the matter of the Universe
is distributed in space, we can help address
some basic questions in cosmology How
did galaxies and large-scale structures form?
6
Why Survey the Universe Now?
By determining how the matter of the Universe
is distributed in space, we can help address
some basic questions in cosmology How
did galaxies and large-scale structures
form? What is the Universe made of?
Dark Matter Dark Energy
7
Why Survey the Universe Now?
By determining how the matter of the Universe
is distributed in space, we can help address
some basic questions in cosmology How
did galaxies and large-scale structures
form? What is the Universe made of? What
happened in the earliest moments of the Big Bang?
Inflation
8
The Structure Formation Cookbook
1. Initial Conditions A Theory for the Origin of
Density Perturbations in the Early
Universe Quantum fluctuations from
Inflation primordial spectrum P kns 2.
Cooking with Gravity Growing Perturbations to
Form Structure Set the Oven to Cold (or
Hot or Warm) Dark Matter Season with a
few Baryons and add Dark Energy 3. Let Cool for
13 Billion years 4. Tweak (1) and (2) until it
tastes like the observed Universe.


9
Early
Evolution of Structure in a Simulated Cold
Dark Matter Universe The Cosmic
Web Galaxies and Clusters form in sheets and
filaments Similar to the structures seen in
galaxy surveys
Today
10
Cold Dark Matter Models Power Spectrum of the
Mass Density ?(x) ? ?(x) ??(x)?
??(x)? ?(k) ? d3x eikx ?(x) ??(k1)?(k2)?
(2?)3 P(k1)?3(k1k2)
P k
keq ?mh
P k3
SCDM
?CDM Open CDM
Non-linear
Linear
h/Mpc
11
Superclusters and Large-scale Structure
Filaments, Walls, and Voids of Galaxies
Center for Astrophysics Redshift Survey (1986)
300 Million Light-years
You Are Here
Watermelon Slice 6 degrees thick containing
1060 galaxies position of each galaxy
represented by a single dot Radial coordinate is
redshift (much easier to measure than distance)
12
Superclusters and Large-scale Structure
Filaments, Walls, and Voids of Galaxies
Center for Astrophysics Redshift Survey (1986)
300 Million Light-years
You Are Here
Watermelon Slice 6 degrees thick containing
1060 galaxies position of each galaxy
represented by a single dot
13
Superclusters and Large-scale Structure
Filaments, Walls, and Voids of Galaxies
Coma Cluster of Galaxies Finger of God
300 Million Light-years
You Are Here
Watermelon Slice 6 degrees thick containing
1060 galaxies position of each galaxy
represented by a single dot
14
SDSS
CfA
15
Two Kinds of Galaxy Surveys
Photometric imaging ? 2D sky maps positions,
brightnesses
(and colors if more than one
band) Spectroscopic redshifts ? distances
(via Hubbles Law)
3D maps
16
UK Schmidt Imaging Survey (photographic plates)
17
2MASS Infrared Survey Extended Source
Distribution
18
221283 galaxies completed 2002
19
SLOAN DIGITAL SKY SURVEY
http//www.sdss.org
GOAL MAP THE UNIVERSE IN 3 DIMENSIONS
OVER A LARGE VOLUME

• Photometric Survey 108 5-band CCD images

• Spectroscopic Survey 106 galaxy and 105 QSO
  redshifts

University of Chicago
Fermilab Princeton University New Mexico
State
Max-Planck A and IA
Johns Hopkins University
Institute for Advanced Study
U.S. Naval Observatory University of
Washington
Japan Participation Group
Funding Sloan Foundation, NSF, DOE, NASA,
member institutions, Japan Ministry of Education
Los Alamos National Lab
University of Pittsburgh
20
Scott Burles David Schlegel Mark Subbarao
21
SDSS Nuts Bolts

• 2.5m Dedicated Telescope
• Ritchey-Chretien design with 3 deg corrected
  FOV,
• sited at Apache Point Observatory (NM)
• Large multi-CCD Camera
• Filters ugriz (3540-9250 A) for
  star/galaxy/QSO selection and
• photometric redshift estimates
• 30 primary 2048x2048 chips (0.4 arcsec/pixel)
  astrometric chips
• Drift-scan mode 55 sec exposures ? limiting
  magnitude r 23
• Multi-fiber spectrographs
• 2 double fiber-fed spectrographs covering
  3900-9200 A
• 640 fibers on the sky, using pre-drilled
  plug plates
• Obtain redshifts for 106 galaxies with
  rlt17.7, 105 QSOs with glt19.7,
• and 105 luminous red galaxies
• Data processing 10s of Terabytes of raw imaging
  data, processed
• promptly at
  Fermilab for follow-up spectroscopy

22
SDSS 2.5 meter Telescope
23
Apache Point Observatory Southern New Mexico
24
SDSS Imaging Camera Top to bottom g
z u i r Drift Scan Mode
25
Photometric Camera filter response with and
w/o atmospheric extinction of 1.2 airmasses
26
g
r
i
27
g
r
i
28
Galaxy photometric redshift estimates
Predicted redshift from 5-band SDSS Photometry
?z 0.05
Connolly, etal Csabai, etal
Spectroscopic measured redshift
29
M101
30
M109
31
Galactic Mergers Acquisitions
32
Perseus cluster
33
Spectroscopic Plates for Redshift Survey
640 fibers per plate
34
(No Transcript)
35
SDSS Data
April 2000 Survey begins June 2005 Survey
nominally finishes (may be extended) Data so
far gt5,000 total square degrees of 5-band
imaging gt500,000 spectra
(G,Q,S redshifts) Samples recently analyzed
3,400 sq. deg. imaging with
photometric redshifts 205,000
main galaxy (spectroscopic) redshifts
First Data Release
www.sdss.org/dr1 2,100 unique
sq. deg. imaging (53 million
objects) 186,250
spectra/redshifts DR2 early 2004
36
Which Universe is ours?
More Cold Dark Matter
Less Cold Dark Matter (Open)
Cold Dark Matter with ?
37
Precision Cosmology Statistical accuracy of
Power spectrum from
Completed SDSS Redshift
Survey 900,000 galaxies in Main galaxy
sample (flux-limited) 100,000 Luminous Red
Galaxies (color/photo-z selected)
?h0.2 ?CDM
(assumed biased)
?h0.5 SCDM
Cold Dark Matter Models
38
Probing Neutrino Mass and Baryon Density
Wiggles Due to Non-zero Baryon Density
SDSS WMAP will constrain sum of stable
neutrino masses as low as 0.5 eV
39
Galaxy Clustering varies with Galaxy Type
Luminosity How are each of them related to the
underlying Mass distribution? Bias ( relation
of Mass to light) depends upon Galaxy Color
Luminosity Caveat for inference of
Cosmological Parameters from LSS
40
(No Transcript)
41
(No Transcript)
42
Lum funcs sel funcs by Michael Blanton (NYU)
Redshift Survey Divide Galaxies
by Intrinsic Luminosity Volume- Limited (complete
) subsamples
Bright
Faint
43
Galaxy Clustering as a function of Galaxy
Luminosity
bright
faint
Zehavi, etal
Tegmark, etal
44
Angular Correlation Function Luminosity
Dependence
bright

• -22 gt M gt -23
• -21 gt M gt -22
• -20 gt M gt -21

Less bright
Budavari, etal
45
Clustering Strength vs. Luminosity Same trend
seen in Photometric and Spectroscopic samples

• Photometric
• Spectroscopic

46
Correct For Luminosity Bias
47
SDSS Galaxy Power Spectrum Full Sample
48
SDSS Galaxy Power Spectrum Full
Sample corrected for Luminosity Bias
49
SDSSGalaxyPowerSpectrumFull samplewith
best-fitCDM modelsTegmark, etal
?mh0.20 ?0.02 ?80.93?0.03
for L galaxies
ns0.93?0.03 WMAP ns0.99?0.04 WMAP only
50
(No Transcript)
51
Cosmic Microwave Background Wilkinson Microwave

Anisotropy Probe (WMAP)
SDSS galaxies today
Universe at 400,000 years Combine these
two to constrain Cosmology, e.g., the amount and
properties of the Dark Matter
52
WMAP Temperature Angular Power Spectra
53
CMB
Clusters
LSS
Tegmark Zaldarriaga
54
95 Constraints Neutrino masses ?m? lt 1.7
eV Priors flat w 1 ns const r 0
?mh2
55
Constraints on Dark Energy Priors w
1 ns const r 0 m? 0
Flat
56
CMB
Clusters
LSS
Lensing
Lya
Tegmark Zaldarriaga, astro-ph/0207047 updates
57
Large-scale Galaxy Distribution what have we
learned?

• Pattern of large-scale structure Microwave
  background
• observations tell us
• The Universe comprises
• 5 Ordinary Matter (atoms, )
• 25 Dark Matter (exotic particles)
• 70 Dark Energy (even more exotic)

58
Large-scale Galaxy Distribution what have we
learned?

• Pattern of large-scale structure Microwave
  background
• observations tell us
• The Universe comprises
• 5 Ordinary Matter (atoms, )
• 25 Dark Matter (exotic particles)
• 70 Dark Energy (even more exotic)
• 2. Structure formed by gravity probably acting
  upon
• quantum vacuum fluctuations formed in
  the first
• 0.0000000000000000000000000000000001
  second
• after the Big Bang.

59
Precision Cosmology with Large-scale Structure?
Requires a more nuanced treatment of Bias, i.e.,
of how luminous galaxies of different types
are related to the underlying mass
distribution We must jointly constrain
cosmological and bias model parameters
60
Halo Occupation Distribution
Halo Occupation Model for Bias

• All galaxies live in dark matter halos.
• Galaxy content of a halo is statistically
  independent of the
• halos larger scale environment. Depends
  only on halo mass.

Assume
The bias of a certain galaxy class (type,
luminosity, etc) is fully defined by

• The probability distribution P(NM) that a halo
  of mass M contains N galaxies
• ltNgtM
  P(NltNgt)
• The relation between the galaxy and dark matter
  spatial distribution within halos

61
Galaxy Clustering
Cosmological Model O, P(k), etc.
Galaxy Formation Gas cooling, Star formation,
Feedback, Mergers, etc.
Halo Occupation Distribution P(NM) Spatial bias
within halos
Dark Halo Population n(M), ?(rM), v(rM), ?(r)
Galaxy-Mass Correlations
62
Predicted Halo Occupation Moments
Berlind, Weinberg, etal
63
Two-point Correlations
Large scales All pairs come from separate halos
Small scales All pairs from same halo
64
HOD Model fit to Clustering of Bright
SDSS Galaxies Scale-dependent Bias
galaxies
2-halo
1-halo
mass
NM?
M1
65
Toward an Understanding of Bias SDSS Galaxy-galax
y Correlations vs. Luminosity Zehavi,
Zheng, etal
66
SDSS Clustering by Color Zehavi, Zheng
67
Integrated Sachs-Wolfe Effect

• An analog of the Sachs-Wolfe effect
• Growth of structure imprints a signature on
  temperature of
• the CMB photons (integrated over line of
  sight).
• Time-dependent gravitational potential due to
  Dark Energy
• dominance at late times
• Correlation between CMB temperature and
• angular distribution of galaxies (on large
  scales)
• Probe of Dark Energy and bias

68
SDSS-WMAP Cross-correlation Independent
confirmation of Dark Energy Luminous
Red Galaxies (trace deep Potential
wells) Scranton etal Fosalba Gaztanaga APM
mK
69
Gravitational Lensing
See the same effects that occur in more familiar
optical circumstances magnification and
distortion (shear)
Objects farther from the line of sight are
distorted less.
Lensing conserves surface brightness bigger
image ?? magnified
70
Helen Frieman (late 1999)
71
Gravitational Lensing Helen behind a Black Hole
72
Cluster of Galaxies
giant arcs are galaxies behind the cluster,
gravitationally lensed by it
73
Weak Lensing of Faint Galaxies distortion of
shapes
Background Source shape
74
Weak Lensing of Faint Galaxies distortion of
shapes
Foreground galaxy
Background Source shape
Note the effect has been greatly exaggerated here
75
Lensing of real (elliptically shaped) galaxies
Foreground galaxy
Background Source shape
Must co-add signal from a large number of
foreground galaxies
76
December 14, 1999
SDSS Galaxy- Galaxy Lensing
77
Toward an Understanding of Bias SDSS Galaxy-mass
vs. Galaxy-galaxy Correlations weak
lensing Sheldon, etal
?gg
78
Seeing Double SDSS Gravitationally Lensed Quasar
SDSS image in 5 filters
Two images of the same Quasar
Follow-up image with Magellan 6-meter telescope
Inada, etal
79
Seeing Triple another Gravitationally Lensed
Quasar
SDSS image in 5 bands
Magellan follow-up image
80
Seeing Triple another Gravitationally Lensed
Quasar
SDSS image in 5 bands
Quasar images
Foreground lens galaxy
81
New Gravitational Lens SDSS 09035028
ARC 3.5m image D. Johnston,
G. Richards Keck spectrum
PCA-decomposed by UC undergrad E. Johnson
82
Wide Separation Lens System
13
SDSS images
Oguri, Inada, etal
83
Subaru image
84
Cosmological Constraints from Strong Lens
Statistics
SN Ia
Lenses CLASS Radio Survey Re-calibrate lens
galaxy parameters using SDSS
velocity-dispersion function for Early type
galaxies Mitchell, etal Sheth, etal
Tonry etal
85
Newberg, Yanny, etal
86
Mass (Solar)
Temperature
O B A F G K M L T
20 10 5 2 1 0.7 0.4 0.1 0.05
50000 K 20000 K 10000 K 7500 K 5500 K 3500
K 2000 K 1200 K 900 K
u g r i z
87
Stars in the field of Globular Cluster Pal 5
Horizontal Branch
A
Red Giant Branch
g'
Blue Stragglers
F Turn off
Main Sequence
g'-r'
88
F turnoff stars on the celestial equator from SDSS
New structures
C
Debris From Sagittarius Dwarf Galaxy
A
89
Sagittarius Tidal Stream a dwarf
galaxy being ripped apart by the
Milky Way
Ring
New view of the Galaxy
90
Simulated Milky Way Halo with dwarf
galaxy satellites Are these rings the tip of
the iceberg? Can they survive in clumpy
halos? Moore, etal
91
The Future
New Surveys with even more powerful
telescopes Virtual Observatory Deskchair
Astronomy
92
Large Synoptic Survey Telescope

• Proposed 8.5m ground based telescope with 7
  square degree field of view
• 5000 Gbytes/night of data
• Real-time analysis
• Celestial Cinematography
• Also PANSTARRS,

93
Astronomy is Facing a Major Data
Avalanche Multi-Terabyte Sky Surveys and
Archives (Soon Multi-Petabyte), Billions of
detected Sources, Hundreds of measured attributes
per source
94
A Virtual Observatory A complete, distributed,
web-based research environment for astronomy with
massive and complex data sets In the US the
National Virtual Observatory (NVO) (see
http//www.us-vo.org) Globally International
V.O. Alliance (IVOA)
95
NGC 253 Visible
96
NGC 253 Infrared
97
Accessing SDSS Public Data

• Data is being released in stages, semi-annually
• Access via the SkyServer website
• http//skyserver.fnal.gov
• excellent resource for education/outreach
• General information about the Survey
• http//www.sdss.org

98
Movie