The 7-Year WMAP Observations: Cosmological Interpretation

The 7-Year WMAP

Observations: Cosmological Interpretation

Eiichiro Komatsu (Texas Cosmology Center, UT Austin) Cosmology Seminar, Perimeter Institute, April 27, 2010

1

WMAP will have collected 9 years of data by August

•

January 2010: The seven-year data release

June 2001:

WMAP launched!

February 2003:

The first-year data release

March 2006:

The three-year data release

March 2008:

The five-year data

release ²

WMAP 7-Year Papers

•

Jarosik et al., “Sky Maps, Systematic Errors, and Basic Results”

arXiv:1001.4744

•

Gold et al., “Galactic Foreground Emission” arXiv:1001.4555

•

Weiland et al., “Planets and Celestial Calibration Sources”

arXiv:1001.4731

•

Bennett et al., “Are There CMB Anomalies?” arXiv:1001.4758

•

Larson et al., “Power Spectra and WMAP-Derived Parameters”

arXiv:1001.4635

•

Komatsu et al., “Cosmological Interpretation” arXiv:1001.4538

3

WMAP 7-Year Science Team

•

C.L. Bennett

•

^{G. Hinshaw}

•

^{N. Jarosik}

•

^{S.S. Meyer}

•

^{L. Page}

•

D.N. Spergel

•

E.L. Wright

•

M.R. Greason

•

^{M. Halpern}

•

^{R.S. Hill}

•

^{A. Kogut}

•

^{M. Limon}

•

^{N. Odegard}

•

G.S. Tucker

•

J. L.Weiland

•

^E.Wollack

•

^{J. Dunkley}

•

^{B. Gold}

•

^{E. Komatsu}

•

^{D. Larson}

•

^{M.R. Nolta}

•

^{K.M. Smith}

•

^{C. Barnes}

•

^{R. Bean}

•

^{O. Dore}

•

H.V. Peiris

•

^{L. Verde}

4

7-year Science Highlights

•

First detection (>3σ) of the effect of primordial helium on the temperature power spectrum.

•

The primordial tilt is less than one at >3σ:

•

^ns=0.96

±0.01 (68%CL)

• Improved limits on neutrino parameters:

• ^{∑ m}

^ν

<0.58eV (95%CL); N

eff

=4.3±0.9 (68%CL)

• First direct confirmation of the predicted

polarization pattern around temperature spots.

• Measurement of the SZ effect: missing pressure?

⁵

7-year Temperature C l

6

(T emperatur e Fluctuation)

2

=180 deg/θ

Zooming into the 3rd peak...

7

(T emperatur e Fluctuation)

2

=180 deg/θ

High-l Temperature C l :

Improvement from 5-year

=180 deg/θ 8

(T emperatur e Fluctuation)

2

WMAP and ACT

10 50 100 500 1000 1500 2000 3000 9

Multipole moment l 0

1000 2000 3000 4000 5000 6000

l(l+1)C lTT /2! [µK2 ]

WMAP 7yr ACT 148 GHz

=180 deg/θ

(T emperatur e Fluctuation)

2

Detection of Primordial Helium

10

(T emperatur e Fluctuation)

2

=180 deg/θ

Effect of helium on C l TT

•

We measure the baryon number density, nb, from the 1st- to-2nd peak ratio.

•

As helium recombined at z~1800, there were fewer

electrons at the decoupling epoch (z=1090): ne=(1–Yp)nb.

•

More helium = Fewer electrons = Longer photon mean free path 1/(σ_Tne) = Enhanced Silk damping

•

This effect might be degenerate with Ωbh² or ns...

11

WMAP + higher-l CMB = Detection of Helium

•

The combination of WMAP and high-l CMB data

(ACBAR and QUaD) is powerful enough to isolate the

effect of helium: Yp = 0.33 ± 0.08 (68%CL) ¹² Baryon Density Tilting

Why this can be useful

•

The helium abundance has been measured from Sun and ionized regions (HII regions); however, as helium can be produced in the stellar core, one has to

extrapolate the measured Yp to the zero-metallicity values.

•

In other words, the traditional methods give a robust upper limit on Yp: Yp<0.3.

•

The CMB data give us a robust lower limit on Yp.

13

0.23<Y p <0.3 (68%CL)

•

Planck is expected to yield ΔYp~0.01 (68%CL; Ichikawa

et al. 2008). ₁₄

Baryon Density Tilting

Another “3rd peak science”:

Number of Relativistic Species

15

from 3rd peak from external data

Neff=4.3±0.9

7-year TE Correlation

10 50 100 500 1000 Multipole moment l

-1.0 -0.5 0.0 0.5 1.0 1.5 2.0

(l+1)C lTE /2! [µK2 ]

16

Let’s talk about CMB polarization.

=180 deg/θ

(TxP olarization Fluctuation)

Improvements from 5-year

•

For 5-year, we used Q

and V bands to measure the high-l TE and TB. For 7-year, we also include

the W-band data.

•

TE: 21σ detection!

(It was 13σ in 5 year.)

•

TB is expected to vanish in a parity-conserving

universe, and it is

consistent with zero. ₁₇

What Are We Seeing Here?

10 50 100 500 1000 Multipole moment l

-1.0 -0.5 0.0 0.5 1.0 1.5 2.0

(l+1)C lTE /2! [µK2 ]

18

I don’t know about you, but I have been

struggling to explain what the TE correlation is.

Actually, I have been struggling to explain what the CMB polarization is in the first place. How

can we solve this problem?

(TxP olarization Fluctuation)

=180 deg/θ

CMB Polarization On the Sky

• Solution: Leave Fourier space.

Go back to real space.

¹⁹

CMB Polarization is a Real-space Stuff

•

CMB Polarization is created by a local temperature

quadrupole anisotropy. ²⁰

Wayne Hu

Principle

•

Polarization direction is parallel to “hot.”

•

This is the so-called “E-mode” polarization.

21

Q<0; U=0 North

East

Hot Hot

Cold Cold

Stokes Q and U

(and KKS’s Q r and U r )

•

As (E-mode) polarization is either radial or

tangential around

temperature spots, it is convenient to define Qr

and Ur as:

Kamionkowski, Kosowsky and Stebbins (1997)

Q r <0 U r =0

22

CMB Polarization on Large Angular Scales (>2 deg)

•

How does the photon-baryon plasma move?

Matter Density

ΔT

Polarization

ΔT/T = (Newton’s Gravitation Potential)/3

23

Potential

CMB Polarization Tells Us How Plasma Moves at z=1090

•

Plasma falling into the gravitational

potential well = Radial polarization pattern Matter

Density

ΔT

Polarization

ΔT/T = (Newton’s Gravitation Potential)/3

24

Potential

Zaldarriaga & Harari (1995)

Quadrupole From

Velocity Gradient (Large Scale)

25

Potential Φ

Acceleration

a=–∂Φ

a>0 =0

Velocity

Velocity in the rest

frame of electron e^– e^–

Polarization

Radial None

ΔT Sachs-Wolfe: ΔT/T=Φ/3

Stuff flowing in

Velocity gradient

The left electron sees colder photons along the plane wave

Quadrupole From

Velocity Gradient (Small Scale)

26

Potential Φ

Acceleration

a=–∂Φ–∂P

a>0

Velocity

Velocity in the rest

frame of electron e^– e^–

Polarization

Radial

ΔT Compression increases

temperature Stuff flowing in

Velocity gradient

<0

Pressure gradient slows down the flow

Tangential

Hence, TE Correlation (Coulson et al. 1994)

• ^{C TQr ( θ ) = – ∫ dlnl [l 2 C l TE /(2 π )] J 2 (l θ )}

θA=

(sound horizon)/dA

27

–^∂Φ≈∂P

Peak Theory and Stacking Analysis

• Peak theory gives:

•

Stack polarization images around temperature hot

and cold spots.

•

Outside of the Galaxy mask (not shown), there are 12387 hot spots and 12628 cold spots.

[Note the l² term! 28

(Desjacques 2008)]

Analogy to Weak Lensing

•

If you are familiar with weak lensing, this statistic is equivalent to the tangential shear:

29

However, all the formulae given in the literature use a scale-independent

bias, b1. This formula must be modified to include the k² term.

Tangential shear,

<γt>, is positive for this example.

Temperatur e hot spots ar e stack ed

30

Radial Tang.

Low peaks: enhanced small-scale correlation

High peaks: basically the same as C^TQ(θ)

stuff is

flowing in

Two-dimensional View

•

All hot and cold spots are stacked (the threshold peak height, ΔT/σ, is zero)

•

“Compression phase” at θ=1.2 deg and

“slow-down phase” at θ=0.6 deg are predicted to be there and we observe them!

•

The overall significance level: 8σ

•

Striking confirmation of the physics of

CMB and the dominance of adiabatic

& scalar perturbation.

31

How About U r ?

•

^Ur is produced by the TB correlation, which is expected to vanish in a parity-conserving universe.

•

^{The Ur} map is consistent with noise.

32

Probing Parity Violation

•

Cosmological parity violation (“birefringence,” Carroll 1998; Lue et al. 1999) may rotate the polarization plane by an angle Δα, and convert E modes to B modes:

•

Non-detection of Ur gives Δα=1±3 deg (68%CL)

•

The full analysis using ClTB (as well as ClEB) gives

•

^{Δα = –1.1 ± 1.3}

(statistical) ± 1.5(systematic) deg.

33

Probing Inflation (Power Spectrum)

•

Joint constraint on the

primordial tilt, ns, and the tensor-to-scalar ratio, r.

•

Not so different from the 5-year limit.

•

r < 0.24 (95%CL; w/o SN)

•

r < 0.20 (95%CL; w/ SN)

34

Probing Inflation (Bispectrum)

•

No detection of 3-point functions of primordial curvature perturbations. The 95% CL limits are:

•

^{–10 < fNLlocal < 74}

•

–214 < fNLequilateral < 266

•

–410 < fNLorthogonal < 6

•

The WMAP data are consistent with the prediction of simple single-inflation inflation models:

•

^{1–ns≈r≈fNLlocal, fNL}equilateral = 0 = fNLorthogonal.

35

If this means anything to you...

Senatore et al.

36

Sunyaev–Zel’dovich Effect

•

^{ΔT/Tcmb = gν y}

37

Zel’dovich & Sunyaev (1969); Sunyaev & Zel’dovich (1972)

observer Hot gas with the

electron temperature of Te >> Tcmb

y = (optical depth of gas) kBTe/(mec²)

= [σT/(mec²)]∫nekBTe d(los)

= [σT/(mec²)]∫(electron pressure)d(los)

gν=–2 (ν=0); –1.91, –1.81 and –1.56 at ν=41, 61 and 94 GHz

Coma Cluster (z=0.023)

•

“Optimal V and W band” analysis can separate SZ and CMB. The SZ effect toward Coma is detected at 3.6σ.

61GHz

94GHz ^{ggνν=–1.81=–1.56} We find that the

CMB fluctuation in the direction of

Coma is ≈ –100uK.

(This is a new result!) ycoma(0)=(7±2)x10^–5

(68%CL)

(determined from X-ray)

38

A Question

•

Are we detecting the expected amount of electron pressure, Pe, in the SZ effect?

•

Expected from X-ray observations.

•

Expected from theory.

39

Three approaches

•

1. Compare the WMAP data and X-ray data on individual clusters individually.

•

But, not all clusters are detected individually by the WMAP data...

•

2. Compare the stacked WMAP data and the expected average X-ray profiles.

•

Systematic effects in the stacking analysis must be carefully studied.

•

3. Compare the stacked WMAP data and the expected profiles from theory (hydro simulation etc).

40

Arnaud et al. Profile

•

A fitting formula for the average electron pressure

profile as a function of the cluster mass (M500), derived from 33 nearby (z<0.2) clusters.

41

Arnaud et al. Profile

•

A significant

scatter exists at R<0.2R500, but a

good convergence in the outer part.

X-ray data

sim.

42

Coma Data vs Arnaud ^•

^Mestimated from the ^{500=6.6x1014h–1Msun is} mass-temperature

relation (Vikhlinin et al.)

•

^{TXcoma =8.4keV.}

•

Arnaud et al.’s profile

overestimates both the direct X-ray data and

WMAP data by the same factor (0.64)!

•

To reconcile them, Tx^coma=6.5keV is

required, but that is way too low.

The X-ray data (XMM) are provided by A. Finoguenov.

43

Well...

•

That’s just one cluster. What about the other clusters?

•

We measure the SZ effect of a very nice sample of

well-studied nearby clusters compiled by Vikhlinin et al.

44

New results! (Prelim.)

•

Normalization for the 5 very high-mass clusters (M500>6x10¹⁴h^–1Msun) is 0.72±0.13.

•

Normalization for the 12 high-mass clusters

(M500>4x10¹⁴h^–1Msun) is 0.68±0.11.

•

Arnaud et al. profile systematically

overestimates the

electron pressure! ⁴⁵

How about the X-ray data?

•

The apple-to-apple comparison, i.e., the direct

comparison between the WMAP data and the X-ray data (Vikhlinin et al.) is underway. Stay tuned!

46

Statistical Detection of SZ

•

Coma is bright enough to be detected by WMAP.

•

Some clusters are bright enough to be detected

individually by WMAP, but the number is still limited.

•

By stacking the pixels at the locations of known clusters of galaxies (detected in X-ray), we detected the SZ

effect at 8σ.

•

Many statistical detections reported in the literature:

47

ROSAT Cluster Catalog

z≤0.1; 0.1<z≤0.2; 0.2<z≤0.45 Radius = 5θ₅₀₀

Virgo Coma

•

742 clusters in |b|>20 deg (before Galaxy mask)

•

400, 228 & 114 clusters in z≤0.1, 0.1<z≤0.2 & 0.2<z≤0.45.⁴⁸

Mass Distribution

•

^M500~(virial mass)/1.6 Most of the signals

come from

M500>0.8x10¹⁴h^–1Msun

Angular Profiles

•

(Top) Significant detection of the SZ effect.

•

(Middle) Repeating the same analysis on the random locations on the sky does not reveal any noticeable bias.

•

(Bottom) Comparison to the

expectations. The observed SZ ~ 0.5–0.7 times the expectations.

Why?

50

Small-scale CMB Data

•

The SPT measured the secondary anisotropy from

(possibly) SZ. The power spectrum amplitude is ASZ=0.4–0.6 times the expectations. Why?

point source thermal SZ

kinetic SZ

51

SPT ACT

Lueker et al. Fowler et al.

point source thermal SZ

Lower A SZ : Two Possibilities

•

The SZ power spectrum is sensitive to the number of clusters (i.e., σ₈) and the pressure of individual clusters.

•

Lower SZ power spectrum can imply:

•

^σ8 is 0.77 (rather than 0.8): ∑mν~0.2eV?

•

Gas pressure per cluster is lower than expected x [gas pressure]

WMAP measurement favors this possibility. ⁵²

Gory Details and Systematic Error Checks

•

What are the “expectations”?

•

Empirical pressure profiles derived from X-ray observations (Arnaud et al. 2009)

•

Theoretical pressure profiles derived from

hydrodynamical simulations (Nagai et al. 2007)

•

Theoretical pressure profiles derived from simple analytical modeling of the intracluster medium

(Komatsu & Seljak 2001; 2002)

•

All of these agree with each other reasonably well.

53

•

The central part of the clusters cannot be

resolved by WMAP’s beam. ₅₄

WMAP solid: X-ray

others:

KS

r500~0.5(virial radius)

Size-Luminosity Relations

•

To calculate the expected pressure profile for each cluster, we need to know the size of the cluster, r500.

•

This needs to be derived from the observed properties of X-ray clusters.

•

The best quantity is the gas mass times

temperature, but this is available only for a small subset of clusters.

•

^{We use r500–LX} relation (Boehringer et al.):

55

Uncertainty in this relation

is the major source of sys. error.

Missing P in Low Mass Clusters?

•

One picture has emerged:

•

^{“High LX}” clusters [M500>4x10¹⁴ h^–1Msun] can be brought into agreement with the expectations by playing with the r500–LX relation.

•

^{“Low LX}” clusters reveal a significant missing pressure. ⁵⁶

Comparison with Melin et al.

•

That low-mass

clusters have lower normalization than

high-mass clusters is also seen by a

different group using a different method.

•

While our overall normalization is

much lower than theirs, the relative normalization is in an agreement.

“High LX”

“Low LX”

57

This is consistent with the lower-than-expected C l SZ

•

At l>3000, the dominant contributions to the SZ power spectrum come from low-mass clusters (M500<4x10¹⁴h^–1Msun).

58

Summary

•

Significant improvements in the high-l temperature data, and the polarization data at all multipoles.

•

High-l temperature: ns<1, detection of helium, improved limits on neutrino properties.

•

Polarization: polarization on the sky!

•

Polarization-only limit on r: r<0.93 (95%CL).

•

All data included: r<0.24 (95%CL; w/o SN)

•

^{Δα = –1.1 ± 1.3}

(statistical) ± 1.5(systematic) deg.

59

Puzzle?

•

SZ effect: Coma’s radial profile is measured, several massive clusters are detected, and the statistical

detection reaches 8σ.

•

Evidence for lower-than-expected gas pressure.

•

The X-ray data are fine? We just need to revise the existing models.

60