Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Héctor Gil-Marín (Institute Lagrange de Paris Fellow, LPNHE Sorbonne University) Statistical challenges for large-scale structure in the era of LSST Oxford, 20th April 2018 Based on HGM et al. 2018, arXiv:1801.02689
BAO & RSD Papers on DR14Q This talk RSD - Full Shape Gil-Marín et al. 18 (RSD in Fourier space) • f σ 8 (z eff ) H(z eff ) Hou et al. 18 (RSD in config. Space) • D A (z eff ) Zarrouk et al. 18 (RSD in conf. space) • Ruggeri et al. 18 (z-weighting RSD in Fourier Space) f σ 8 (0.8<z<2.2) • H(0.8<z<2.2) Zhao et al. 18 (z-weighting RSD in Fourier Space) D A (0.8<z<2.2) • D v (z eff ) Ata et al. 18 (BAO isotropic, Fourier & conf. space) • BAO Wang et al. 18 (z-weighting BAO in Fourier space) • D v ( 0.8<z<2.2 ) Zhu et al. 18 (z-weighting BAO in conf. space) • Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Hector Gil-Marin
DR14 footprint for the quasar sample eBOSS in a nutshell e xtended B aryon O scillation S pectroscopic S urvey Part of SDSS-IV collaboration • Spectroscopic survey: σ z ~0.001 • Apache Point Telescope 2.5m • 2014 - 2019 observing LRGs, ELGs, quasars • + Lya 1000fibres per plate (~7deg 2 ) • 1000 EZ & 400 QPM mocks for covariances • 2.2 NGC 0.8 < z < 2.2 • SGC 2 Wide redshift range • 1.8 n(z)x10 5 [hMpc -1 ] 3 148,659 quasars • 1.6 Low density of tracers 1.4 • 1.2 2 x 10 -5 h/Mpc 1 Low density variation • 0.8 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2112.9 deg 2 redshift Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Hector Gil-Marin
Three types of redshift estimates 1200 fid 1000 MgII kP (l) (k) [Mpc/h] 2 800 PCA 600 400 200 0 -200 4 (2) 2 ∆ P (2) / σ P 0 -2 -4 4 (0) 2 ∆ P (0) / σ P 0 -2 -4 0.02 0.10 0.20 0.30 k [hMpc -1 ] The z PL automated classification: Template-based PCA fit to CIV-line The z PCA automated classification: Template-based PCA fit to MgII-line The z MgII automated classification: Location of the MgII-line peak (when present). Standard redshift estimate z fid: Any of the above options depending on the particular object, which provides the lowest rate of catastrophic failures . Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Hector Gil-Marin
Potential observational Systematics Redshift efficiency pattern Success rate after visual inspections • Redshift Failures: i) Weight the NGC SGC nearest neighbour (NN), use in BOSS analysis. ii) Weight all observed galaxies by their position in the plate, 1 w spec ( x foc , y foc ) ∼ P sucess ( x foc , y foc ) • Collision Pairs: Traditional nearest lower at the edges neighbour weighting (NN) Fibres corresponding to edges of the spectrograph Imprint such effects on the mocks and check how these correction schemes perform Zarrouk et al. 18 Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Hector Gil-Marin
Potential observational Systematics Monopole Quadrupole 500 z f 1000 w foc w cp 400 w noz w cp 800 kP (0)(k) [Mpc/h] 2 kP (2)(k) [Mpc/h] 2 300 raw 600 200 400 100 200 0 0 150 60 100 40 50 20 k ∆ P (0) k ∆ P (2) 0 0 -20 -50 -40 -100 -60 -150 0.10 0.20 0.30 0.10 0.20 0.30 k [hMpc -1 ] k [hMpc -1 ] True signal (systematic effect not applied) Corrected: redshift failures (focal weight) + close pairs (NN) Corrected: redshift failures (NN) + close pairs (NN) Corrected: redshift failures (focal weight) [close pairs not applied] Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Hector Gil-Marin
Potential modelling Systematics • Use OuterRim N-body simulation at z=1.43 (Habib et al 2016) with different HOD prescriptions & w/ or wo/ (Gaussian) redshift smearing. HOD of quasars observational modelling 0.1 no satellite fraction no sys? 0.05 13% satellite fraction problems? ∆α perp 22% satellite fraction 0 -0.05 Model: 2-loop RPT + TNS @ z=1.43, 0.02<k[h/Mpc]<0.30 -0.1 0.1 f σ 8 , α para , α perp , b 1 σ 8 , b 2 σ 8 , σ fog , A noise 0.05 ∆α para 0 -0.05 +17% -0.1 0.1 +16% 0.05 +13% ∆ f σ 8 0 -0.05 Take highest deviation from obs & mod -0.1 add in quadrature both ‘obs’ and ‘sys’ errors EZ mocks QPM mocks OR w/o sm. OR w/ sm. … but not clear if the overall measurement would be shifted Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Hector Gil-Marin
Measurement and best-fitting model on DR14Q DR14Q 0.8<z<2.2 Monopole 1000 Quadrupole 800 Hexadecapole kP(k) [Mpc/h] 2 600 400 200 0 -200 -400 4 χ 2 =84/(84-7) 3 χ 2P0 =20/(28-7) 2 χ 2P2 =30/(28-6) 1 ∆ P / σ P χ 2P4 =35/(28-4) 0 -1 -2 -3 -4 0 0.10 0.20 0.30 0.40 k [hMpc -1 ] 0.02 < k [h/Mpc] < 0.30 Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Hector Gil-Marin
Comparing data with mocks EZ-Mocks DR14Q w z fid • Data looks like a typical DR14Q w z PCA realisation wrt the mocks DR14Q w z MgII • Redshift estimate does not 50 60 70 80 90 100 affect the best-fitting value 1.3 χ 2 1.2 significantly 1.1 α para 1.0 • However, does affect the tails 0.9 (errors) of the distribution 0.8 0.7 1.3 1.2 1.1 α perp 0.20 1.0 EZ-Mocks 0.9 0.16 DR14Q w z fid 0.8 DR14Q w z PCA 0.7 σ α para 0.12 DR14Q w z MgII 0.2 0.4 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 0.7 0.8 0.9 1.0 1.1 1.2 1.3 f σ 8 α para α perp 0.08 0.04 0.20 0.16 Mocks as pipeline validation • σ α perp 0.12 Potential systematic tests • 0.08 0.04 Unfortunately, mocks do not have spectra • 0.04 0.08 0.12 0.16 0.20 0.04 0.08 0.12 0.16 0.20 σ f σ 8 σ α para Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Hector Gil-Marin
Consistency results on the data Test the effect of adding/removing the hexadecapole Test the effect of redshift estimates on the cosmological parameters Hexadecapole helps to break H z PCA larger tails vs. f σ 8 and D A vs H degeneracies Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Hector Gil-Marin
Split the 0.8<z<2.2 in 3 overlapping z-bins • We individually fit the 3 redshift bins • The covariance among parameters is computed through the EZmocks DR14Q Monopole 1200 lowz 1000 midz lowz 0.8< z <1.5 kP (l) (k) [Mpc/h] 2 highz 800 midz 1.2< z <1.8 600 highz 1.5< z <2.2 400 200 4 3 2 ∆ P (l) / σ P 1 0 -1 -2 -3 -4 0.02 0.10 0.20 0.30 k [hMpc -1 ] Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Hector Gil-Marin
Split the 0.8<z<2.2 in 3 overlapping z-bins • We individually fit the 3 redshift bins • The covariance among parameters is computed through the EZmocks DR14Q Quadrupole 800 600 lowz 0.8< z <1.5 400 midz 1.2< z <1.8 200 highz 1.5< z <2.2 0 -200 -400 4 3 2 1 0 -1 -2 -3 -4 0.02 0.10 0.20 0.30 k [hMpc -1 ] Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Hector Gil-Marin
Split the 0.8<z<2.2 in 3 overlapping z-bins • We individually fit the 3 redshift bins • The covariance among parameters is computed through the EZmocks DR14Q Hexadecapole 400 200 lowz 0.8< z <1.5 0 midz 1.2< z <1.8 highz 1.5< z <2.2 -200 -400 -600 4 3 2 1 0 -1 -2 -3 -4 0.02 0.10 0.20 0.30 k [hMpc -1 ] Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Hector Gil-Marin
Split the 0.8<z<2.2 in 3 overlapping z-bins Redshift-space distortion analysis from the DR14 eBOSS quasar sample in Fourier space Hector Gil-Marin
Recommend
More recommend
