Linear ISW The Imprint of Superstructures Coldspot Analysis

The Effect of Supervoids and Superclusters on the Cosmic Microwave Background

Istvn Szapudi1 , Ben Granett1 , & Mark Neyrinck2
1 Institute 2 Johns

for Astronomy, University of Hawaii Hopkins University

KEK Cosmophysics Workshop LLTB2009, Tsukuba, Oct 20-23, 2009

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Outline

Linear ISW

The Imprint of Superstructures

Coldspot Analysis

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Integrated Sachs-Wolf Effect

Photons passing through changing gravitational potentials are becoming slightly hotter or colder TISW where d d , d

d(x) (x) d = [(1 + z)D1 (z)] d (1 + z) d

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Linear ISW as a Signature of DE

In a at, matter dominated universe D1 = 1/(1 + z), There is a linear effect if M = 1, e.g., CDM If the universe is at (e.g., from CMB), linear ISW effect signals Dark Energy Caviat: there can be a non-linear effect as well

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Physical Picture

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

ISW Effect and Cross-Correlation

Galaxy catalogs and the CMB are correlated due to the ISW effect This is a tiny correlation compared to the CMB uctuations, but it has been detected in several galaxy catalogs I will explore that cross-correlation with several techniques: correlation function, matched lter of the reconstructed potential, and imprint of superstructures. We use the SDSS LRG data set, and the WMAP5 maps Nota that combining all available galaxy data sets gives up to 4 cross-correlation signal (Giannantonio etal, Ho etal)

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

SDSS DR6 LRG

7500 square degree 2SLAQ cuts 746962 objects 0.45 < z < 0.75 with median z = 0.52 median photo-z errors z 0.04 matches 2SLAQ LRG spectroscopic targets

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

The SDSS DR6 LRG catalog

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

WMAP 5-year data set

co-added Q,V,W ILC map MCMC map (joint t to temperature, polarization and foregrounds) KQ75 galactic foreground masks maps are smoothed to 1 FWHM resolution Healpix Nside = 64 maps, or 55 arcminute resolution

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Cross-correlation results with SpICE and MLHood

Agreement with Giannantonio etal, Ho etal, e.g., 2.1 from the MCMC map..

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Beyond Cross-correlation

Non-optimal weighting No redshift information was used Cosmic variance of galaxy data, even though we have access to a particular realization Correlation between two datasets can arise for several reasons other than ISW (e.g. systematics)

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Potential reconstruction
First we calculate local density using a Voronoi tessalation (adaptive smoothing) Then we some the potential contributed by each galaxy (r) = Gcr,0 m (1 + zmed ) c2 V (zi ) Vi |r ri |

Survey is projected to median redshift z = 0.52 Derivative of potential is calculated by reprojecting to the appropriateredshift using linear growth function Finally. raytracing through the linear potential derivative eld We expect that our TISW = (1/bg )TISW
I. Szapudi

(true)

(LRG)

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Potential map corresponding to the LRG catalog

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Max. Likelihood/Matched Filter

if Y = X plus some noise (here the CMB) Maximum likelihood gives
1 XC Y , = YC 1 Y

(1)

with variance 2 = (YC 1 Y )1 , where C is the CMB (noise) cov. matrix is related to the bias (and any numerical factors missed in the prediction) looks like an optimal sum over a two-point quantity

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Dependence on redshift cuts

I II III IV

redshift range 0.49-0.56 0.48-0.58 0.47-0.60 0.45-0.63

N /103 300 400 500 600

Amplitude uncorrected 3.12 1.63 2.51 1.25 1.85 0.97 0.95 0.66

Amplitude corrected 2.2 2.1 1.7 1.0

1.9 2.0 1.9 1.4

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Noise biases
From shot noise and redshift errors

We decided to use only sample II (0.48-0.58)

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Detection Signicance from Potential Map

Map Q Coadd V Coadd W Coadd Q FG reduced V FG reduced W FG reduced MCMC ILC Amplitude 2.96 1.71 3.33 1.71 3.01 1.71 3.43 1.71 3.52 1.71 3.20 1.71 3.75 1.71 4.33 1.67

1.7 1.9 1.8 2.0 2.1 1.9 2.2 2.6

Fully consistent with previous measurements

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Disappearing the cross-correlations

We conclude that the cross-correlation is due to linear ISW

I. Szapudi Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

About factor of 2 high compared to CDM

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Finding Superstructures (100Mpc scales)

The magic of Voboz

Voboz/Zobov algorithms to nd supervoids and superclusters int the LRG catalog Cutting out the highest signal-to-noise areas (simple weighting) Photometric redshift information is used Actual realization of the galaxy (DM) eld is used Linear use of the data Possibility of localizing the signal, especially if there is a non-linear component

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Zobov

Roughly, a catchment basin in the density eld. Uses parameter-free Voronoi tessellation for density No smoothing of densities; instead, assigns a probability (based on density contrast) that they arose from discreteness noise Inverse of a VOBOZ (Neyrinck, Gnedin & Hamilton 2005) cluster
I. Szapudi Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Aspen-Amsterdam Void-Finder Comparison

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Superstructures Found by VOBOZ/ZOBOV

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Superstructures Found by VOBOZ/ZOBOV

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Superstructures Found by VOBOZ/ZOBOV

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Supervoids and Superclusters

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Back to the Basics: Image-stacking

Granett, Neyrinck, & Szapudi 2008, ApjL, 68, L99-102

Two different Monte Carlos to determine signicance: agree within 2%

I. Szapudi Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Detection signicance
Robust against number, radius, color.

N 30 50 70 50 50 50 50 50

Radius 4.0 4.0 4.0 3.0 3.5 4.0 4.5 5.0

T K 11.1 9.6 5.4 8.4 9.3 9.6 9.2 7.8

T / 4.0 4.4 2.8 3.4 4.0 4.4 4.4 3.8

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Prole the Imprint of Superstructures

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Potential + Superstructures
Formally a 5.3 signal

Superstructures correspond to a different signal than cross-correlation

I. Szapudi Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Coldspot
Non-Gaussian area (Vielva etal 2004, Cruz etal 2005) unlikely at the 0.5% level (Cruz etal 2006) Zhang & Huterer *2009): skeptical view T 70 K 5 area Rudnick etal 2007: hint of a supervoid in the NVSS (debated by Smith & Huterer 2008) Inoue & Silk (2007): 200 h1 Mpc void with = 0.3 Many other ideas (e.g., SZ, textures, Cruz etal 2008 for review) We performed observations with CFHT Megacam to test a 100 h1 Mpcvoid at z < 1
I. Szapudi Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Megacam observations
0.8 deg2 elds within 5 of the Cold Spot

Table: Survey elds

ID A B C D E F G

Area (sqr deg) 0.80 0.82 0.81 0.81 0.77 0.88 0.79

95% Completeness (AB ma g) g r i z 22.7 21.6 22.4 20.9 22.8 22.8 22.5 21.0 23.2 23.4 22.4 20.8 22.6 23.2 22.4 20.5 22.4 23.4 22.4 21.1 23.9 23.3 22.1 21.5 23.1 23.0 22.1 21.1

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Coldspot Area

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Color Selection

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

CFHT Legacy elds

CFHTLS Data release 4 same instrument, deeper, with i lter different 6 elds total 9.8 2 Used for calibration and to determine expected counts

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Differential Galaxy Counts

Fields B and F with low counts

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Photo-z Training Set

Survey ZCOSMOS DEEP2 VVDS SDSS LRG

CFHTLS Field D2 D3 D1 D2,D3

z-Range (Median z) 0.121.22 (0.53) 0.201.03 (0.64) 0.141.17 (0.61) 0.310.38 (0.35)

N 819 158 134 14

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Photo-z precision with EAZY

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Selection Function

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Individual Density Maps

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Individual Structures
Field A, z = 0.6 a cluster, conrmed by visual Field B, z = 0.2, strong 60% drop in counts, a 2.8 uctuation Field F shows low counts overall, especially in the two low-z bins. Like to cause the NVSS feature Oscillation in D might be enhanced by anticorrelations, but the underdensity at z=0.6 appears to be real In general, there is more variation then cosmic variance would suggest: many elds deviate more than 2 in at least one bin Caviat: the lowest and highest z bin might be the most uncertain
I. Szapudi Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Bayesian Inversion of the Density with MCMC

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Marginalized Likelihood

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Combined Fields

a = 0.3 100 h1 Mpc void would correspond to a density drop of b = 0.45 this would show up at z = 0.6 as a drop of 15% We rule out a void of 200 h1 Mpc = 0.3 at 105 level in the high z bins at z = 0.4 it is ruled out only at 7 103 Finally the low redshift bin shows a signicant underdensity

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

The Low Redshift Bin (Combined Fields)

in terms of cosmic variance a deviation of 20% is only 2.3 It is seen in A,B,D,F, but not in the adjacent E We cannot conclusively detect a supervoid The data are consistent with a 160 h1 Mpc = 0.3 void that would do the trick We will have Pan-STARRS data in the future that will decide Note that the our previous supervoid measurement appear to indicate that the signal might be higher than our LCDM estimate

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Robustness and Cosmic Variance

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Summary
www.ifa.hawaii.edu/supervoids for more

Over 2 detection of the linear ISW from cross-correlation and (marginally better) potential maps from the LRGs Cross-correlation disappears when the best t ISW map is subtracted Signal from superstructures over 4 very robustly This appears to be in addition to the linear ISW signal Potential + Superstructures 5.3! The nature of the signal from superstructures is yet to be determined (astrophysical, non-linear, or...? )

I. Szapudi

Supervoids and Superclusters

Linear ISW The Imprint of Superstructures Coldspot Analysis

Summary
www.ifa.hawaii.edu/supervoids for more

Our Coldspot measurements do not support the existence of a 200 h1 Mpc supervoid on redhifts 0.3 < z < 0.9 and it is unlikely that one would hide at z > 1 A hint of a large low redshift void that could be a reason for the Coldspot, however, we cannot make a robust statement with the present data Pan-STARRS will conrm the cross correlations with overwhelming statistical signicance: 6 for linear ISW. Pan-STARRS will conrm the strength of the supervoid signal (10 K ) Pan-STARRS will be able to detect a low redshift supervoid in the COldspot region if it exists Theoretical investigations of the non-linear ISW are on-going.
I. Szapudi Supervoids and Superclusters