Void BAO measurements on quasars from eBOSS

TL;DR

This paper investigates the use of voids in quasar distributions from eBOSS to enhance BAO measurements, finding potential improvements in mock data but limited impact on actual SDSS data.

Contribution

It introduces a novel void-based approach to improve BAO measurements using quasar data, demonstrating its effectiveness in mock surveys and analyzing its limitations with real data.

Findings

Void-based BAO analysis shows a 70% improvement in mocks.

No significant improvement found in SDSS data due to cosmic variance.

Forecasts suggest potential benefits for future surveys like DESI.

Abstract

We present the clustering of voids based on the quasar (QSO) sample of the extended Baryon Oscillation Spectroscopic Survey Data Release 16 in configuration space. We define voids as overlapping empty circumspheres computed by Delaunay tetrahedra spanned by quartets of quasars, allowing for an estimate of the depth of underdense regions. To maximise the BAO signal-to-noise ratio, we consider only voids with radii larger than 36$h^{-1}$Mpc. Our analysis shows a negative BAO peak in the cross-correlation of QSOs and voids. The joint BAO measurement of the QSO auto-correlation and the corresponding cross-correlation with voids shows an improvement in 70$\%$ of the QSO mocks with an average improvement of $\sim5\%$. However, on the SDSS data, we find no improvement compatible with cosmic variance. For both mocks and data, adding voids does not introduce any bias. We find under the flat $\Lambda$CDM assumption, a distance joint measurement on data at the effective redshift $z_{\rm eff}=1.48$ of $D_V(z_{\rm eff})=26.297\pm0.547$. A forecast of a DESI-like survey with 1000 boxes with a similar effective volume recovers the same results as for light-cone mocks with an average of 4.8$\%$ improvement in 68$\%$ of the boxes.