An Empirical Analysis of Baryon Acoustic
Oscillations in Galaxy and Quasar Clustering

Abstract

We present our analysis of the baryon acoustic oscillation (BAO) signal in the
correlation functions of LOWZ and CMASS galaxy samples from Data Release 12 of the
SDSS BOSS survey. We draw a comparison between our results and the findings of Cuesta
et al. (2016) who analysed the BAO feature in the same datasets. Upon using subsets of
the data to obtain an empirical estimate of the uncertainties on the correlation functions,
we find our results to be in general agreement with the uncertainties presented by Cuesta
et al. (2016), obtained from simulated mocks. We detect the BAO peak at ~4 sigma and ~7 sigma
in the correlation function of the LOWZ and CMASS samples respectively. We demonstrate
that when fitting the correlation functions with our fiducial LCDM model, the A(s) nuisance
fitting parameters play a significant role in providing a good fit. Based on a F-ratio test, we
find that in our primary fitting range, the simple LCDM model without the nuisance fitting
terms is rejected in favour of the full model as indicated by the small p-values of p = 0.018
and p = 0.00084, for the CMASS and LOWZ samples respectively. We demonstrate that the
significance of the detection of the BAO peak is the quantity most sensitive to the choice of
the fitting range. Based on isotropic fitting to the correlation functions, we obtain a distance
of DV(z=0.32)rd,fid/rd=1226±32 Mpc and DV(z=0.57)rd,fid/rd=1988±24 Mpc,
a 2.6 per cent and 1.2 percent measurement respectively, assuming a fiducial sound horizon
rd,fid=147.10 Mpc. We extend our BAO analysis to higher redshifts by performing fitting
to the mean of the correlation functions obtained by Chehade et al. (2016) for the 2QDESp,
SDSS DR5 UNIFORM, 2QZ and 2SLAQ quasar samples. This gives a distance constraint
of DV(z=1.49)rd,fid/rd=3583±249 Mpc (assuming rd,fid=147.70 Mpc), a 6.9 per cent
measurement to z=1.49, the mean redshift of the combined QSO sample.