Precision cosmology as a laboratory for particle physics (or, - - PowerPoint PPT Presentation

Precision cosmology as a laboratory for particle physics

(or, Evidence for a 4th neutrino?)

Yvonne Y. Y. Wong RWTH Aachen

SpacePart12, CERN November 5 – 7, 2012

Probes of inhomogeneities

CMB temperature & polarisation anisotropies Large-scale matter distribution

Distance vs redshift

Type 1a supernovae Baryon Acoustic Oscillations feature in the galaxy distribution

Galaxy distribution Cluster abundance Lyman-α forest Cosmic shear

> 0.2 deg: COBE, WMAP, Planck < 0.2 deg: DASI, CBI, ACBAR, Boomerang, VSA, QuaD, QUIET, BICEP, ACT, SPT, etc.

Local Hubble expansion rate (HST calibration of SNIa abs. magnitude)

SDSS, ROSAT/Chandra, CFHTLS, etc.

The concordance flat ΛCDM model...

13.4 billion years ago (at photon decoupling) Composition today

• The simplest model consistent with observations.

Massless Neutrinos (3 families) Plus flat spatial geometry+initial conditions from single-field inflation Cosmological constant

The concordance flat ΛCDM model...

13.4 billion years ago (at photon decoupling) Composition today

Massless Neutrinos (3 families)

ν-to-γ energy density ratio fixed by Standard Model physics

T CMB=2.725±0.001K

Photon energy density fixed by CMB temperature & spectrum measurements:

• The simplest model consistent with observations.

• Neutrino decoupling at T ~ O(1) MeV.
• After e+e- annihilation (T ~ 0.2 MeV):

– Temperature: – Number density per flavour: – Energy density per (massless)

flavour:

Neutrino energy density (standard picture)...

T ν=( 4 11)

1/3

T γ

nν= 6 4 ζ(3) π

2 T ν 3= 3

11 nγ ρν=7 8 π2 15 T ν

4=7

8( 4 11)

4/3

ργ

Photon temperature, number density, & energy density Fixed by weak interactions and the assumption of equilibrium statistics Assuming instantaneous decoupling

3ρν ργ ∼0.68

This ratio is put into the ΛCDM model by hand and should be tested!

Plan...

• Evidence for extra neutrino energy density from precision cosmological
• bservations
• What could this extra energy density be?
• What future observations can do

• 1. Evidence for extra neutrinos from

precision cosmology...

• Treating the neutrino energy density

as a free parameter, recent

• bservations prefer Neff > 3 at 2σ+.

Dunkley et al. [Atacama Cosmology Telescope] 2010

WMAP+ACT WMAP+ACT+H0+BAO WMAP

Standard value

Evidence for extra neutrinos...

WMAP ACT SDSS (BAO) HST CMB TT WMAP ACT, SPT ℓ<1000 ℓ>1000

• Allowing the neutrino energy density

to be a free parameter, recent

• bservations prefer Neff > 3 at 2σ+.

Evidence for extra neutrinos...

WMAP SDSS (BAO) HST SPT Keisler et al. [South Pole Telescope] 2011 Standard value CMB TT WMAP ACT, SPT ℓ<1000 ℓ>1000

W-7=WMAP-7

= 2-2.5σ evidence

Abazajian et al., “Light sterile neutrinos: a white paper”, 2012

Λ CDM+N eff

More complex cosmological models Simplest cosmological model

• Main effect of Neff is to delay

matter-radiation equality.

• Looks easy to detect... but

we use the same data to measure at least 6 other parameters:

• Plenty of degeneracies!

How does it work...

(ωb ,ωm ,h , As, ns ,τ)

Figure courtesy of J. Hamann baryon density matter density Hubble parameter primordial fluctuation amplitude & spectral index

• ptical depth

to reionisation

What the CMB really probes: equality redshift...

Exact degeneracy between the physical matter density ωm and Neff.

1+ zeq=ωm ωr ≃ ωm ωγ 1 1+ 0.2271 N eff

• Ratio of 3rd and 1st peaks sensitive

to the redshift of matter-radiation equality via the early ISW effect.

Fixed: zeq

Figure courtesy of J. Hamann

What the CMB really probes: sound horizon...

• Peak positions depend on:

Fixed: zeq

Figure courtesy of J. Hamann

Fixed: zeq, ωb, θs

θs= rs D A

Sound horizon at decoupling Angular distance to the last scattering surface

θs∝ (ωm h

−2) −1/ 2

∫

a * 1

da

√ωm h

−2a −3+ (1−ωmh −2)

Fixed zeq, ωb Exact degeneracy between ωm and the Hubble parameter h. Flat ΛCDM

What the CMB really probes: anisotropic stress...

• Apparent (i.e., not physical) partial

degeneracies with primordial fluctuation amplitude As and spectral index ns.

Figure courtesy of J. Hamann

Fixed: zeq, ωb, θs Fixed: zeq, ωb, θs, As(l=200)

• However, free-streaming particles

generate anisotropic stress.

• First real signature of Neff is in the

3rd acoustic peak!

Komatsu et al. [WMAP5] 2008

• Measurement of the third acoustic peak (since WMAP 5 years) gives

lower limit on Neff from WMAP alone.

• Upper limit requires combination of WMAP with other observations to

break the remaining Neff–ωm–h parameter degeneracies.

– Pinning down either ωm or h will do!

WMAP only +BAO+SN+HST

• Multipoles l > 1000.
• Probe photon diffusion scale:

Breaking degeneracies with the CMB damping tail...

θd θs = rd rs ∝ωm

1/4

Hou, Keisler, Knox et al. 2011

Fixed by WMAP: zeq, ωb, θs, As(l=200)

CMB acoustic peaks (WMAP) & damping tail (ACT, SPT) together breaks the Neff–ωm–h degeneracy and measures Neff > 3 at 1.5σ.

ACT since 2010 SPT since 2011 Also Planck Sound horizon from acoustic peaks

• Adding BAO & local Hubble expansion rate measurements reduces the

Neff–ωm–h degeneracy and pushes the detection significance to 2σ+.

The role of non-CMB observations...

• BAO = baryon acoustic oscillations;

sensitive to ωm and h.

Percival et al. 2010 Baryon acoustic oscillations in the matter power spectrum

• 2. Is it really an extra neutrino?

Spacetime metric Nuclei/ electrons Photons Massless neutrinos Cold dark matter

Cosmic microwave background

EM interaction

Galaxies, Hydrogen clouds, etc. Observables

Particle content of the concordance ΛCDM model...

Spacetime metric Nuclei/ electrons Photons Massless neutrinos Cold dark matter

Cosmic microwave background

EM interaction

Non-interacting Interacting Non-relativistic (no vel. dispersion) Relativistic

Galaxies, Hydrogen clouds, etc. Observables

Particle content of the concordance ΛCDM model...

Extra neutrinos = extra relativistic, non-interacting stuff (no requirement

• n the quantum

numbers or statistics)

“Extra neutrinos” don't have to be real neutrinos...

• Any particle species whose production is associated with some thermal

process and that decoupled while relativistic at relatively late times [T< O(100) MeV] will behave (more or less) like a neutrino as far as cosmological observations are concerned.

Neutrino temperature per definition Three SM neutrinos Other light stuff

T ν=( 4 11)

1/3

T γ

∑i ρν,i+ρX =N eff(

7 8 π

2

15 T ν

4)

=(3.046+Δ N eff )ρν

(0)

Corrections due to non-instantaneous decoupling, finite temperature effects in the EM plasma, and flavour oscillations

• Peccei-Quinn scale fa < 108 GeV.
• Axion decoupling occurs after QCD

phase transition (T < 150 MeV).

– Dominant processes:

Relative number density Axion number density today (relative to 1 ν)

Example 1: Hot QCD axions...

Axion temperature today (relative to ν temperature) Relative temperature

ma [eV] f a/GeV=6×10

7

6×10

5

Hannestad, Mirizzi & Raffelt 2005

π+π ↔π+a a+N ↔ N +π

• Can contribute up to

Δ N eff <0.57

Pseudoscalar = 1 dof, Bose-Einstein statistics; Axions are a little colder than neutrinos

CERN Axion Solar Telescope (CAST)

L = 9.26 m B = 9 T (Decomissioned LHC test magnet) Aune et al. [CAST] 2011

f a[GeV]

6×106 6×107 6×108

The parameter region for hot axions is probed also by CAST which looks for axions from the sun.

Sun

a γ γvirtual e, Ze

Example 2: light sterile neutrinos...

LSND Short-baseline reactors

Mention et al. 2011

MiniBooNE Each anomaly can be individually explained in terms of active-sterile

• scillations with:

νe appearance νe appearance νe disappearance

ΔmSBL

2

∼1→10eV

2

sin22θSBL∼10−3

Probα→β=sin

22θsin 2(

Δ m

2 L

4 E ) Probα→ α=1−Probα→β

Two-flavour oscillations:

Baseline Energy Mass splitting Not coupled to the Z

Hannestad, Tamborra & Tram 2012

ms> mα ms< mα

• Experimentally preferred Δm2 and

mixing favour the production and thermalisation of sterile neutrinos in the early universe via να↔νs

• scillations + να scattering.

→ Can easily produce → Sterile states have the same temperature as the SM neutrinos.

Δ N eff∼1

Can we tell which particle it is?

• In the completely general case, no.
• But, if the particle is a thermal relic and has a mass so that it becomes a

nonrelativistic particle species today, then in principle

– Particle mass – Temperature – Quantum statistics (Fermi-Dirac or Bose-Einstein)

can be determined from a combination of cosmological observations.

– Not possible with current observations

• 3. Coming up next...

Planck and Neff...

Hamann, Hannestad, Lesgourgues, Rampf & Y3W 2010

• If Neff is as large as 4, it

will be settled almost immediately by Planck (launched May 14, 2009; public data release early 2013).

1σ sensitivities WMAP Planck (expected) Acoustic peaks and damping tail

• Euclid will improve Planck's sensitivity to

Neff by a factor of ~4 [σ(Neff) ~ 0.055].

2 Euclid spacecraft concepts

Further down the road: Euclid and Neff...

Hamann, Hannestad & Y3W 2012 Planck+Euclid galaxies Planck+Euclid cosmic shear Planck+Euclid galaxies+ shear

• Current precision cosmological data show a preference for extra

relativistic energy density (beyond 3 neutrinos) at 2-2.5σ significance.

– Excess roughly equivalent to one additional species of neutrinos.

• Possible forms of extra relativistic energy density include hot QCD axions

and light sterile neutrinos.

– And many more...

• Planck with tell (at least part of the story).

Summary...