bao analysis from the dr14 qso samplebenasque.org/2017uco/talks_contr/014_bao_dr14_benasque.pdf ·...

IntroductionBAO measurement

RSD issues

BAO analysis from the DR14 QSO sample

Héctor Gil-Maŕın (on behalf of the eBOSS QC WG)

Laboratoire de Physique Nucleaire et de Hautes Energies (LPNHE)Institut Lagrange de Paris (ILP)

Understanding Cosmological Observations @ Benasque1st Aug 2017

Understanding Cosmological Observations @ Benasque, 1st Aug 2017BAO from the DR14 QSO sample


RSD issues

Public DR14 data (yesterday!) arXiv:1707.09322

We’re very happy to report that the fourteenth data release ofSDSS is now live!!! Go have a look at www.sdss.org/dr14, and tellall your friends and colleagues that this awesome data set is now

available for them to play with!



RSD issuesThe eBOSS survey

Introduction: the eBOSS survey

Apache Point Observatory (APO) 2.5-m telescope.SDSS-III project. 2009-2014 BOSS: Baryon OscillationSpectroscopic SurveySDSS-IV project. 2014-2019 eBOSS: extended BaryonOscillation Spectroscopic Survey3 galaxy clustering programs (ELG, LRG, quasars) + Ly-α

new selection algorithms to identifyredshift of galaxies

BOSS had 99% success rateidentifying redshifts of LRGs

first eBOSS tests showed 70% ofsuccess rate on LRGs using sameBOSS algorithms!





eBOSS survey: ELG, LRG and QSO samples

LRG 0.6 < z < 1.0; zeff = 0.71.

ELG 0.6 < z < 1.1; zeff = 0.85.

QSO 0.8 < z < 2.2; zeff = 1.5.

Ly-α zeff = 2.33

[Credit : Anand Raichoor]





0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

z

0.2

0.3

0.4

0.5

0.6

0.7fσ

8(z

)assuming Planck ΛCDM cosmology

eBOSS year 6Planck ΛCDM

BOSS DR126dFGSSDSS MGS

GAMAWiggleZVipers




Low Density of Quasars!

Shot noise dominated covariance matrices(traditional) Reconstruction algorithms do not provide muchsignal gain





QSO DR14 area 2044 deg2.

quasar range: 0.8 < z < 2.2(zeff ' 1.5)Number quasars: 147,000

Low density: 2× 10−5 [Mpc/h]33 disconnected areas: 2 SGC &NGC.





Alphabetical paper Ata et al. 2017submitted to the JournalarXiv:1705.06373

First BAO measurement in0.8 ≤ z ≤ 2.2DV = 3855± 170 rd/rd ,fid Mpc atz=1.52 (4.4% precision)

∼ 2.5σ BAO significanceDR14 Area: 2044→4.4% precisionDR16 Area: 5300→ 2.7% precisioneBOSS is working!



RSD issues

Alcock-Paczynski effectMethodologyMeasurementSystematic Tests

Alcock-Paczynski effect

By assuming a wrong cosmological model (Ωm) we change theline-of-sight clustering respect to the angular clustering, creating ameasurable anisotropy: constrains H(z) and DA(z) (or acombination of both).

dcomov(z) =

∫ z0

cdz ′

H(z ′; Ωm)

The BAO scale is determined by the comoving sound horizon atreconvination scale (standard ruler)

rs =1

H0Ω1/2m

∫ a∗0

dacs

(a + aeq)1/2



RSD issues


Baryonic Acoustic Oscillations

k‖ → α‖k‖ k⊥ → α⊥k⊥

α‖ =Hfid(z)

H(z)α⊥ =

DA(z)

DfidA (z)

[Anderson et al. 2014,BOSS DR11]

Surveys measure angles andredshifts, and these are affectedby the assumed fiducial model.

This changes theapparent/observed position ofthe BAO peak in the powerspectrum differently in theradial and angular direction

Isotropic Correlation Function /Power Spectrum sensitive to

DV (z) =

[(1 + z)2D2A(z)

cz

H(z)

]1/3Understanding Cosmological Observations @ Benasque, 1st Aug 2017BAO from the DR14 QSO sample


RSD issues


Methodology

We perform two complementary and independent analyses onthe same dataset using the i) Power Spectrum and ii)Correlation Function.

Both observables should contain the same amount ofinformation, but in practice are affected differently by noiseand systematic effects.

We use mocks (1000 EZ mocks) and (400 QPM mocks) toestimate the covariance matrices and perform systematic tests.

We model the broadband shape of the PS/CFphenomenologically and the BAO as linear+damping (Σnl)

P(k , α) = Psm(k){

1 + [Olin(k/α)− 1] e−12

Σ2nlk2}

smooth PS: Psm(k) ≡ B2P linnw(k) + A1k + A2 + A3/kUnderstanding Cosmological Observations @ Benasque, 1st Aug 2017BAO from the DR14 QSO sample


RSD issues


Measurement

DR14 QSO isotropic Power spectrum and Correlation function

0.05 0.10 0.15 0.20 0.25 0.30k (hMpc−1)

200

400

600

800

1000

1200

kP

0(k

)(h−

2M

pc2

)

SGC, χ2/dof =22.1/27NGC, χ2/dof =33.2/27

25 50 75 100 125 150 175 200s (h−1Mpc)

−80

−60

−40

−20

0

20

40

60

80

100

s2ξ 0

(s)

(h−

2M

pc2

)

SGC, χ2/dof =28.3/24NGC, χ2/dof =21.0/24

Actual data + diagonal errors from EZ mocks.Dashed lines mean of the EZ-mocks



RSD issues


Measurement

DR14 QSO isotropic Power spectrum and Correlation function

0.05 0.10 0.15 0.20

k(hMpc−1)

−1.5

−1.0

−0.5

0.0

0.5

1.0

1.5

10(P−P

smoot

h)/P

smoot

h

25 50 75 100 125 150 175

s (h−1Mpc)

−4

−2

0

2

4

103(ξ

[s]−

ξ sm

oot

h)

Actual data + diagonal errors from EZ mocks.Solid lines best-fit model



RSD issues


Tests on Mocks

0.85 0.90 0.95 1.00 1.05 1.10 1.15

Correlation Function α

0.85

0.90

0.95

1.00

1.05

1.10

1.15

Pow

erS

pec

tru

mα

DR14

Mocks

0.02 0.03 0.04 0.05 0.06 0.07 0.08

Correlation Function σ(α)

0.02

0.03

0.04

0.05

0.06

0.07

0.08

Pow

erS

pec

tru

mσ

(α)

Correlation coefficient between PS and CF ρ = 0.97.Data is a very typical case of mocks (both in measurement and

error).



RSD issues


Tests on Mocks

NL effects shift the BAO peak to higher α.∼ 0.1% effect on measured α. Negligible on data!



RSD issues


Tests on data

Both P(k) and ξ(s) measurementsare very consistent!

0.8 0.9 1.0 1.1 1.2αBAO

0

2

4

6

8

10

12

14

16

∆χ

2

1σ

2σ

3σ

ξ(s)P (k)P (k) + ξ(s)



RSD issues


Cosmology

Fully consistent with LCDM + Planck.

Ly-α 2.5 measurement dominates at high z .

Further analyses will focus on redshift weighting schemes.

0.0 0.5 1.0 1.5 2.0 2.5

Redshift

0.95

1.00

1.05

Dis

tan

ce/D

ista

nce

(Pla

nck

ΛC

DM

)

6dFGSBOSS DR12

SDSS MGS

WiggleZDR14 quasars

BOSS Lyα



RSD issues


BAO Summary

Summary,

We have a robust BAO isotropic measurement at z ' 1.5.with 4.4% precision: DV (1.52) = 3855± 170 rd/rd ,fid Mpc.Very consistent with mocks and between P and ξ observables.

Fully consistent with LCDM+Planck

First BAO science result from WG, but not last one. Morecomplex BAO analyses will be done in the forthcomingmonths (weighting z evolution).

Future quasar releases (DR16) will focus on the anisotropicquasar BAO → Constrain DA and H.Further information on DV can be extracted from RSDanalysis (several papers to be released this Fall)



RSD issues

The main difficulty when performing RSD analyses is controllingthe observational systematic effects.Such effects have minor effect on the position of the BAO (giventhe current errorbars), but they turn to be more important for RSDanalyses,

Failure rate as a function of the plate position.

Spectroscopic errors.

Estimate of the quasar redshift using different algorithms.



RSD issues

Failure rate in NGC

0 5 10 15 20 25 30 35 40 45

0

5

10

15

20

25

30

35

40

45

0.75

0.8

0.85

0.9

0.95

1

∼ 10% failure rate



RSD issues

Failure rate in NGC

0 5 10 15 20 25 30 35 40 45

0

5

10

15

20

25

30

35

40

45

0.86

0.88

0.9

0.92

0.94

0.96

0.98

1

∼ 3.5% failure rate



RSD issues

Why do the quasars always fail at the same plate positions?

[Credit: P. Zarrouk]



RSD issues

Even with ∼ 3.5% failure rate, applying the nearest angularneighbour reduces the amplitude of the monopole and enhancesthe quadrupole signal significantly.As a consequence, f σ8 is affected by ∼ 0.7σ.BAO not affected

∆αiso Sα ∆f σ8 Sf σ8 NdetEZ 〈x〉i −1.64± 0.13 - −1.43± 0.16 - -EZ 〈xi 〉 −1.6± 4.4 4.8 −1.4± 4.9 5.5 979

EZ 〈x〉i wcol −1.90± 0.13 - 2.24± 0.17 - -EZ 〈xi 〉 wcol −1.7± 4.4 4.6 2.4± 5.4 5.4 959

(units of the table in 10−2).Better correction than just up-weight the nearest neighbour isneeded!



RSD issues

Quasar Clustering Working Group ReportCurrent RSD & BAO projects DR14 QSO sample

Gil-Marin et al. Fourier Space Multipoles RSD

Hou et al. Configuration Space Wedges RSD

Ruggeri et al. Fourier Space Multipoles with z-weights RSD

Wang et al. Fourier Space Multipoles with z-weights BAO

Zarrouk et al. Configuration Space Multipoles RSD

Zhao et al. Fourier Space Multipoles with z-weights RSD

Zhu et al. Configuration Space Multipoles with z-weightsBAO

WG is currently focused on

Producing reliable mocks for covariance matrix

Building appropriate weighting scheme for correcting redshiftfailures and close pairs.



RSD issues

Thank you!


IntroductionThe eBOSS survey

BAO measurementAlcock-Paczynski effectMethodologyMeasurementSystematic Tests

RSD issues

bao analysis from the dr14 qso samplebenasque.org/2017uco/talks_contr/014_bao_dr14_benasque.pdf ·...

Documents