Cosmology with Spectroscopic Surveys: Past, Present and Futurebenasque.org/2019uco/talks_contr/059_Gilmarin_Benasque_2019.pdf · • We can separate the effect of cosmological distortions

Cosmology with Spectroscopic Surveys: Past, Present and Future

Héctor Gil-MarínJunior leader 'la Caixa' Fellow

Institut de Ciencies del Cosmos, Universitat de Barcelona

LSS

CMB

Hector Gil-Marín Cosmology with Spectroscopic Surveys

Baryon Acoustic Oscillations

Before recombination After recombination

CMBGalaxies

Baryons+DM photons

Baryon Acoustic Oscillations

p+ + e- H

Wave propagation frozen

at z~1100

• Physics in early Universe

• Accelerated expansion at late times

Vp

!"!= x! !V! + x! ⊥V⊥

ztrue

vp

vp

LOS

vy

vy

vx

vx

f (z) = Ωm (z)γ

logarithmic growth of structure


• Universe assumed isotropic and homogeneous • RSD: Enhancement / reduction of the clustering along the

line-of-sight (LOS) direction due to peculiar velocities (Kaiser 1987)

Redshift Space Distortions

Induces anisotropy but does not shift the BAO

~ fσ8

Redshift Space Distortions


Alcock-Paczynski effect

Clustering of Tracers: What do we measure?

Ωfiducial ≠ Ωtrue

Galaxy Catalogue

Distortions along and across the line of sight

This effect distorts the homogeneity and isotropy of BAO as a function of Ωtrue & Ωfiducial



• AP effect: Anisotropy induced by transforming redshifts into coming distances assuming a wrong cosmology

Alcock-Paczynski & Redshift Space Distortions

ΔθΔr

Δr

Δr!(z1, z2;Ωm ) =cdz '

H0 Ωm (1+ z ')3 +1−Ωm

z1

z2∫ ≈ cΔzH (z ,Ωm )

Radial distance Angular diameter distance

Δr⊥ (θ1,θ2;z,Ωm ) = Δθ cdz 'H (z ',Ωm )0

z

∫



• AP effect: Anisotropy induced by transforming redshifts into coming distances assuming a wrong cosmology

Alcock-Paczynski & Redshift Space Distortions

α

BAO shift, but no extra anisotropy

Relative BAO shift along and across the line-of-sight +

induced anisotropy~ (DA2/H)1/3 / rs

~ DAH



Pow

er

wave number

observed range

• The sound horizon scale is well determined by CMB measurements (helps to calibrate)

• We can separate the effect of cosmological distortions (AP) from other effects such as RSD

BAO as standard ruler

Redshift Space Distortions vs. Alcock-Paczynski effect





• RSD enhance the Power at a given scale

• AP shift the power of a scale to another scale

Pow

er

wave number

observed range

RSD

AP



Pow

er

wave number

observed range













Pow

er

wave number

observed range







Pow

er

wave number

observed range

Difficult to disentangle: RSD & AP







Pow

er

wave number

observed range






Pow

er

wave number

observed range






Pow

er

wave number

observed range






Pow

er

wave number

observed range • BAO is the most distinct feature in ξ(s) and P(k) to look at.

• Position is robust under potential systematics:

• Non-linear effects are >1% • BAO position not affected by bias

or Kaiser boost • However, the feature is damped by

non-linear velocity bulks (can be solved using reconstruction)



Cosmology with BOSS and eBOSS

• Apache Point Observatory, 2.5-meter • Spectroscopic Galaxy Survey • 2009 - 2014 BOSS• 2014-2019 eBOSS• BOSS LRGs 0.15<z<0.75 (CMASS, LOWZ) • eBOSS LRGs 0.6<z<1.1 • eBOSS ELGs 0.6<z<1.1 • eBOSS quasars 0.8<z<2.2 • + Ly-α spectra (Andreu’s talk)

Main Goal: Measure BAO peak position with 1% accuracy Other cosmology studies: growth of structure, mod-GR, neutrino mass, primordial non-Gaussianity, etc.

Cosmology with BOSS & eBOSS


Area = 9376 deg2 Area = ~4500 deg2 0.2<z<0.5

0.4<z<0.60.5<z<0.75 0.8<z<2.2~106 LRG targets

~3.3·105 quasar targets

Cosmology with BOSS & eBOSSDR16 QSO & LRG

0.6<z<1.0~2.2·105 new LRG targets



Modelling the power spectrum: full shape vs. BAO

There are two main kind of complementary analyses:

1. BAO analysis: Based on the position of the BAO-peak

2. Full Shape analysis (aka RSD): Based on the PS full shape and amplitude signal


BAO analysis• Fit broadband with polynomial fit & BAO

template on oscillations

• Constrain on DA(z)/rs and H(z)rs through the BAO-feature only

• Damping terms for BAO due to bulk flows



Pbao(k,α 0,2 ) = Psm (k) 1+ Olin (k /α 0,2 )−1⎡⎣ ⎤⎦e−12k2Σnl

2⎧⎨⎩

⎫⎬⎭

Psm (k) = B2Plin,sm (k)+ A1k + A2 +

A3k+ ...

α! =α 0

−32α 2

52

α⊥ =α 0

94α 2

−54

P(µ2 ) ≡ P(0) + 2

5P(2) P(0), P(2) → P(0), P(μ2)

BroadbandBAO template

Reconstruction• Enhance BAO peak by

un-doing the non-linear bulk flows

• Assumptions on gravity Ωm and bias of tracer

• ‘Gaussianization’ of the galaxy field



∇ ⋅ Ψ +fb

∇ ⋅ (Ψ ⋅ r) r = −δg

b

Reconstruction



0.9 0.92 0.94 0.96 0.98

1 1.02 1.04 1.06

0 0.10 0.20 0.30

P(µ2 ) /

P sm

k [hMpc-1]

0.92 0.94 0.96 0.98

1 1.02 1.04 1.06 1.08

P(0) /P

sm

0

500

1000

1500

2000kP

(x)

Reconstruction



0.9 0.92 0.94 0.96 0.98

1 1.02 1.04 1.06

0 0.10 0.20 0.30

P(µ2 ) /

P sm

k [hMpc-1]

0.92 0.94 0.96 0.98

1 1.02 1.04 1.06 1.08

P(0) /P

sm

-200 0

200 400 600 800

1000 1200 1400

kP(x

)

Full Shape (RSD)• Constrain the growth of structure, fσ8(z), DA(z)/rs and H(z)rs

through the shape and amplitude of a range of scales. • It requires a full modelling of the amplitude and shape of

the power spectrum multipoles



• Non-linear dark matter PS shape Perturbation Theory 2-loop

• Galaxy bias, Non-linear & non-local

• RSD TNS-model

Power Spectrum Correlation Function

0.2<z<0.50.4<z<0.60.5<z<0.75

3overlapping z-bins

Main BOSS results




0.2<z<0.50.4<z<0.60.5<z<0.75

3overlapping z-bins

• Good Agreement with Planck+GR • First time 1% precision BAO measurement


Main BOSS results

-4-3-2-1 0 1 2 3 4

0 0.10 0.20 0.30 0.40

ΔP

/ σP

k [hMpc-1]

-400

-200

0

200

400

600

800

1000kP

(k) [

Mpc

/h]2

DR14Q 0.8<z<2.2

MonopoleQuadrupole

Hexadecapole

HGM et al. 2018



Main eBOSS results


BOSS gal. DR12BOSS Ly-α. DR12

eBOSS quasars DR14

0.85 0.9

0.95 1

1.05 1.1

1.15

0.1 0.5 1.0 1.5 2.0 2.5

(DA/

r s) /

(DA/

r s)Pl

anck

z

0.8 0.85

0.9 0.95

1 1.05

1.1 1.15

1.2

Hr s

/ (H

r s)Pl

anck

0.3 0.4

0.5

0.6 0.7

0.8fσ

8(z)

MGS DR7 SDSS-IIBOSS LRGs DR12 SDSS-IIIBOSS Ly-α DR12 SDSS-III

eBOSS quasars DR14 SDSS-IV


GR & ΛCDM assumed

0

0.2

0.4

0.6

0.8

1

60 62 64 66 68 70 72 74 76

Ωm

0

H0 (rs/rsfid)

H0 = 73.0 ±1.8

0

0.2

0.4

0.6

0.8

1

1.2

60 62 64 66 68 70 72 74 76

ΩΛ

H0 (rs/rsfid)

DA(z), H(z), fσ8(z) from BOSS galaxies / eBOSS quasars

0

0.2

0.4

0.6

0.8

1

1.2

0 0.2 0.4 0.6 0.8 1

flat ΛCDM

ΩΛ

Ωm0

BOSS galaxieseBOSS quasars + BOSS galaxies

eBOSS quasars + BOSS galaxies + Planck

relax flatness



H0 = 73.0 ±1.8

-2

-1

0

1

2

0.2 0.25 0.3 0.35 0.4 0.45

GR

flat ΛCDM assumed

γ

Ωm

eBOSS quasars + BOSS galaxieseBOSS quasars + Planck

eBOSS quasars + BOSS galaxies + Planck

0.2

0.25

0.3

0.35

0.4

0.45

0.5

60 62 64 66 68 70 72 74 76

Ωm

0

H0(rs/rsfid)

-2

-1

0

1

2

60 62 64 66 68 70 72 74 76

γ

H0(rs/rsfid)

flatness & ΛCDM assumed

DA(z), H(z), fσ8(z) from BOSS galaxies / eBOSS quasars

relax GR



Conclusions

• LSS as a complementary source of information to CMB

• BOSS & eBOSS have demonstrated over the last 8yr that BAO and RSD are robust techniques for estimating cosmological parameters from LSS

• Final DR16 eBOSS results December 2019

• So far no tension with ΛCDM+Planck

• LSS + CMB/BBN favours a low H0 value

αV modify the k vector in the monopoleαε generates an anisotropy (distort symmetric 3D-features along and across the LOS)

k! =α!k! 'k⊥ =α⊥k⊥ '

α! =rsfidH fid

rsHαV = α⊥

2α!3

αε =α⊥ α!

new variables Isotropic dilation

AP-effectα⊥ = DA / rsDA

fid / rsfid

αVαV (z) =

DV (z)DV (z)

fid

DV (z) = (1+ z)2DA2 (z) cz

H (z)⎡⎣⎢

⎤⎦⎥

1/3

αε (z) =DA(z)H (z)

[DA(z)H (z)]fid

× r fids /rs


Backup Slides

1

2

3

4

5

6

0.01 0.1

Nor

mal

ized

Pow

er

k

~ 1αV

~ 1αV

αV~ (DA2/H)1/3 / rs

Monopole

Quadrupole


Backup Slides

P(µ2 ) ≡ P(0) + 2

5P(2)

1

2

3

4

5

6

7

8

0.01 0.1

Nor

mal

ized

Pow

er

k

αε~ DAH

Quadrupole

Monopole

μ2-moment


Backup Slides

Alam et al. 2016


Backup Slides

BAO post-recon RSD Full Shape pre-recon Consensus Planck

0.2<

z<0.

50.

4<z<

0.6

0.5<

z<0.

75

Alam et al. 2016Hector Gil-Marín Cosmology with Spectroscopic Surveys

Backup Slides

Alam et al. 2016

Tension with local H0 measurement reduces with Neff>3


Backup Slides

HGM et al. 2018


Backup Slides

HGM et al. 2018


Backup Slides

Documents

Cosmology with Spectroscopic Surveys: Past, Present and Futurebenasque.org/2019uco/talks_contr/059_Gilmarin_Benasque_2019.pdf · • We can separate the effect of cosmological distortions