On ‘cosmology-cluster physics’ degeneracies and cluster surveys (Applications of self-calibration) Subha Majumdar Canadian Institute for Theoretical Astrophysics Distant Clusters of Galaxies Ringberg Castle Workshop, Oct 24-28, 2005 Graham Cox, Joe Mohr, Howard Yee, Mike Gladders, Henk Hoekstra, Jose Diego
15
Embed
On ‘cosmology-cluster physics’ degeneracies and cluster surveys (Applications of self-calibration)
On ‘cosmology-cluster physics’ degeneracies and cluster surveys (Applications of self-calibration). Subha Majumdar C anadian I nstitute for T heoretical A strophysics. Graham Cox, Joe Mohr, Howard Yee, Mike Gladders, Henk Hoekstra, Jose Diego. Distant Clusters of Galaxies - PowerPoint PPT Presentation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
On ‘cosmology-cluster physics’ degeneracies and cluster
surveys(Applications of self-calibration)
Subha MajumdarCanadian Institute for Theoretical Astrophysics
Distant Clusters of GalaxiesRingberg Castle Workshop, Oct 24-28, 2005
Graham Cox, Joe Mohr, Howard Yee, Mike Gladders, Henk Hoekstra, Jose Diego
2) abundance evolution --- growth function : (H(z))
density fluctuation: 8, ns
3) cluster structure and evolution.
f(M) contains the connections between observables and mass (which connects to theory/simulations).
Distant Clusters of Galaxies
Ringberg Workshop 2005
Mass-observable relations...
)1()()( 3/2 zzEMAfS SZ
Option 1: Existing cluster catalogs give us an idea of these scalings.However, much difference between observational and simulation results.
A biased (incorrect) scaling between mass-observable may give us tight constraints but wrong cosmological parameter estimates. (Pratt’s talk: plot of sigma8 as fn of M-T normalization)
Option 2: Self-Calibration (Majumdar & Mohr 2003,2004; Hu 2003)
Take cluster parameters (Amp, ) as completely unknown.If there are enough clusters in a sample then it is possible to determinecosmology and cluster scaling parameters from the same data. Itappears that all large yield surveys have this capability!!
Crucial Assumption: scaling relations exist (tractable & preferably simple)Smaller assumption: simulations give tight mass fn (dn/dM)
some universality of cluster DM structure (NFW)
It is also possible to determine unknown evolution () from the same databut at the price of having weaker constraints.
Distant Clusters of Galaxies
Ringberg Workshop 2005
More on self-calibration (unashamedly copied from one of Joe Mohr’s astro-ph submissions )
The implications for self-calibration are quite broad. It is more than calibratingthe mass-observable relation from observations!
Uncertainties in absolute calibration (perhaps due to uncertainties in effective areaof the telescope) will self-calibrate out.
A redshift or mass dependence in AGN contamination will self-calibrate out mainlyin (for Xray surveys).
Angular filtering in SZ surveys will introduce redshift dependent errors, but that would be taken care by .
Systematic photometric redshift uncertainties will self-calibrate out (again in )
Systematic redshift dependence of completeness in Bgc (as proxy for mass) will self-calibrate out as a non-standard evolution .
Even, redshift dependence of theoretical halo-mass dependence will self-calibrateout as one can solve for a scaling between simulation defined halo mass and cluster observable.
The functional form of the sacling can be checked with observed and predictedmass functions.
Distant Clusters of Galaxies
Ringberg Workshop 2005
‘Self-Calibration’ Techniques:0. Just let dn/dz information self-calibrate the survey with simple scaling reln.
However, extra informations help!
1.Limited mass follow-up (using XRay temp/weak lensing) (Majumdar & Mohr 2003,2004, Majumdar 2005)2.Using shape of mass-function in redshift slices (Hu 2003)3.Using the cluster power spectrum and P(k) oscillations (Majumdar & Mohr 2004, Hu & Haiman 2004, Huetsi 2005)4.Adding information from counts-in-cell (Lima & Hu 2004, 2005)5.Time or flux slicing of survey: using shape of dndz
(Majumdar 2005)6. For SZ surveys, adding SZ rms distortions to number counts
(Diego & Majumdar, 2004)
Scatter: As long as scatter is ~25%, self-cal is possible but weaker constraints(Levine etal 2002, Lima & Hu 2005, Cox&Majumdar, in prep)
Bias : of more concern, must be ~10% (Lima & Hu 2005)
Distant Clusters of Galaxies
Ringberg Workshop 2005
Doing cosmology AND cluster physics with actual data: RCS1
RCS: the resultsM = 0.34 +/- 0.064 (0.29 +/- 0.07)
ConclusionsSelf-calibrating cluster surveys are possible with large
yield, thus reducing mass-observable uncertainties. Way to do cosmology and cluster physics at the same time.
It is important to understand cosmology-gas physics degeneracies to get accurate constraints. Additional information (like limited independent mass followup) can break these degeneracies. We need
observed lensing masses. We need simulations connecting lensing masses with other observables.
First results from RCS shows that it is now possible to do cosmology and cluster physics with cluster dndz. Moreover, agreement with RCS results with other probes comes as an important endorsement for self-calibration. These results also endorse the promise of doing precision cosmology with upcoming surveys (whether in Optical, SZ or Xray)