Cosmology with Cluster Surveys

1
Cosmology with Cluster Surveys

• Subha Majumdar
• Canadian Institute of Theoretical Astrophysics
• Toronto

along with Joe Mohr, Martin White Jose Diego
International Conference on Gravitation and
Cosmology, Kochi, 5th January 2004
2
Overview

• Introduction to Galaxy Clusters
• Introduction to the Surveys
• Studying Dark Energy with Galaxy Cluster Surveys,
  the stress in on the eqn. of state w of dark
  energy
• (This was the parameter that was not
  constrained by WMAPSDSS!
• But knowledge of w is fundamental to
  our understanding of dark energy)
• Something new Self Calibration in cluster
  surveys and precision cosmology

3
What Are Galaxy Clusters?

• Galaxy clusters are the most massive, collapsed
  structures in the universe. They contain
  galaxies, hot, ionized gas (107-8K) and dark
  matter.
• In typical structure formation scenarios, low
  mass clusters emerge in significant numbers at
  z2-3
• Clusters are good probes, because they are
  massive and easy to detect through their

• X-ray emission

• Sunyaev-Zeldovich Effect

• Gravitational lensing

4
The Basics of SZ Effect
5
Whats nice about SZE?

• 1) Ofcourse, the distinct spectral signature
• 2) Measures the total thermal content of the
  cluster
• 3) More or less redshift independent
• 4) Less susceptible to messy cluster
  substructure, core
• physics (prop to density and not density
  squared as in XRays)

6
Example Local Abundance/Mass Function
Look at the large error bars!! We want to do
much better.
Things will change drastically with future
surveys. Especially as a probe of dark energy eqn
of state w
7
Upcoming Cluster Surveys

• SZ-surveys
• Planck 2008(?) 7,000/8,000
  30,000/40,000
• SPT 2005(?) 20,000-30,000
• ACT late 2004 few thousands
• APEX middle 2004 few thousands

X-Ray surveys DUET Unsuccesful
20,000-30,000 DUO
very ve report, late 2004
10,000 XMM-LSS 2004 ,
1000(?)
Cluster surveys will be a thrust area for
some time to come!
8
From observations to detecting clusters
an example
Diego SM 04
9
The Cluster Redshift Distribution
One can get dN/dz if we can get the redshifts of
the detected clusters
Cluster redshift distribution probes
1) volume-redshift relation
2) abundance evolution
3) cluster structure and evolution.
10
Sensitivity of Cluster Redshift Distribution to
Dark Energy Equation of State

• Increasing w keeping WE fixed
• hasthe following effects

It decreases volume surveyed
It decreases growth rate of perturbations
Fig courtesy Joe Mohr
11
Potential for different methods to constrain w
Complimentary Highly Competetive
Levine, Shultz White 2002
12
Dual Nature of Galaxy Clusters

• Its critical- for almost any analysis- to keep
  in mind that clusters are young objects and yet
  as a population they exhibit striking regularity
• Statistical studies of (x-ray flux limited
  samples of) galaxy clusters reveal that more than
  half exhibit merger signatures.
• Theres lot of substructures.
• Statistical studies of (x-ray flux limited
  samples of) galaxy clusters reveal regularity

Existence of simple flux-mass relation
13
Self-Calibration in SZ surveys

• Two surveys, the South Pole Telescope Survey and
  the Planck all sky survey (yielding gt 20000
  clusters), contain enough information to
  constrain the interesting cosmological parameters
  and solve for the structure of galaxy clusters
  simultaneously!

• Assumptions required
• Hierarchichal structure formation is correct
• A mass-X-ray luminosity relation exists (or a
  mass-SZE luminosity relation exists)
• Crude redshift estimates are available for each
  cluster detected in the survey

Survey Wm Wtot w s8 h n Wb Norm Slope
Priors flat 0.07 0.050 0.004
Planck 0.017 - 0.075 0.013 0.053 0.041 0.004 24 0.004
SPT 0.024 - 0.062 0.013 0.047 0.048 0.004 21 0.005
SM J. Mohr 2003
14
Self-Calibration Continued

• A Caveat What if there is evolution ??

Or is it?
Everything is lost !
Self Calibration can be regained (by adding
complimentary info) ! 1. Need to do
follow-up mass estimates of a few clusters
2. Cluster power spectrum, P(k)
Not impressive by themselves, when combined with
dN/dz they do wonders
15
An Example Planck All Sky Cluster Survey
Nclusters 22000
SM J. Mohr, 2004
16
How well do we get w ?
1s error around w-1, normalized over 9 other
parameters
Survey On its own P(k) 100 clusters follow-up Both
SPT 0.18 0.16 0.06 0.035
Planck 0.39 0.10 0.12 0.041
Note WMAPext gives 0.11
To pin down dark energy we must have cluster
surveys (CMBSNe)
17
So what did I say,

• Upcoming large yield cluster surveys would unveil
  a new
• era of doing cosmology with clusters.
• It has the promise to become 4th pillar of
  precision cosmology
• along with CMB, SNe and weak lensing.
• As an example, we have shown that these surveys
  provide us an
• opportunity to probe the enigma of dark energy
  with high precision
• (the other great way
  is to do weak lensing tomography)
• Most importantly, weve seen that even with
  uncertainties in cluster
• physics, self-calibration in these surveys
  makes clusters an important
• tool in precision cosmology ( w to few
  percents, certainly competetive or maybe
• even better than what can be done with CMB or SNe
  alone. )