The SDSS-IV extended Baryon Oscillation Spectroscopic Survey: Baryon Acoustic Oscillations at redshift of 0.72 with the DR14 Luminous Red Galaxy Sample

TL;DR

The study reports a 2.6% precision BAO distance measurement at $z_{ m eff}=0.72$ using eBOSS DR14 LRGs combined with the CMASS tail. It introduces a reconstruction-based BAO analysis with a new treatment of photometric systematics and redshift failures, validated on 1000 Quick Particle Mesh mocks to robustly estimate the covariance and eliminate biases. The isotropic BAO constraint, $D_V(z_{ m eff})/r_d$, is consistent with Planck ΛCDM and demonstrates the viability of extending BAO measurements to higher redshift with eBOSS data. The methods and software are designed for future DESI/Euclid analyses and generalize to full-shape studies including redshift-space distortions.

Abstract

The extended Baryon Oscillation Spectroscopic Survey (eBOSS) Data Release 14 sample includes 80,118 Luminous Red Galaxies. By combining these galaxies with the high-redshift tail of the BOSS galaxy sample, we form a sample of LRGs at an effective redshift $z=0.72$, covering an effective volume of 0.9~Gpc$^3$. We introduce new techniques to account for spurious fluctuations caused by targeting and by redshift failures which were validated on a set of mock catalogs. This analysis is sufficient to provide a $2.6$\% measurement of spherically averaged BAO, $D_V(z=0.72) = 2353^{+63}_{-61} (r_d/r_{d,\rm{fid}}) h^{-1}$Mpc, at 2.8$σ$ of significance. Together with the recent quasar-based BAO measurement at $z=1.5$, and forthcoming Emission Line Galaxy-based measurements, this measurement demonstrates that eBOSS is fulfilling its remit of extending the range of redshifts covered by such measurements, laying the ground work for forthcoming surveys such as the Dark Energy Spectroscopic Survey and Euclid.

Figures (15)

• Figure 21: The footprint of the DR14 LRG sample. The colors show the fiber completeness per region for the eBOSS sample only. Regions with fiber completeness below 0.5 were removed from the final sample.
• Figure 22: Density of LRGs as a function of redshift. Dashed vertical lines indicate the redshift range used in our clustering measurement. Here we see that a significant fraction of eBOSS LRGs have redshifts below 0.6, where CMASS are more numerous. We can remark the importance of the CMASS sample between redshift 0.6 and 0.7.
• Figure 33: Monopole (top panels) and quadrupole (bottom) of the correlation function for NGC (left) and SGC (right) before (red lines) and after correcting for targeting systematics. The cyan line shows the result when using the up-weighting scheme and the blue line shows the sub-sampling of randoms. Since our corrections for redshift failures are also applied on the random catalog, we employ hereafter the sub-sampling of randoms as our fiducial method to correct for targeting systematics.
• Figure 34: Average redshift efficiency as a function of median pixel signal-to-noise ratio in the $i$ (blue line) and $z$ (red line) SDSS bands. Dotted lines indicate the distribution of median S/N of the eBOSS LRG sample. The dashed line represents the average redshift efficiency of the whole sample of 90.5%.
• Figure 35: Average redshift efficiency as a function of physical position of the optical fiber in the focal plane. Left (resp. right) panel shows the NGC (resp. SGC) failure rates. The region where YFOCAL is negative correspond to spectrograph #1 while YFOCAL $>0$ corresponds to spectrograph #2.