Finding the brightest cosmic beacons in the Southern Hemisphere

TL;DR

This paper introduces a new method using Canonical Correlation Analysis to identify bright high-redshift QSOs in the Southern Hemisphere, significantly increasing the available sample for cosmological studies.

Contribution

The study develops a novel candidate selection technique combining multiple databases, resulting in a high success rate and improved completeness for bright high-z QSOs in the south.

Findings

Identified 1476 QSO candidates over 12,400 sq. degrees.

Confirmed 56 new bright QSOs at z > 2.5 with ~80% success rate.

Achieved an estimated 90% completeness in the candidate sample.

Abstract

The study of absorptions along the lines of sight to bright high-$z$ QSOs is an invaluable cosmological tool that provides a wealth of information on the inter-/circum-galactic medium, Dark Matter, primordial elements, reionization, fundamental constants, and General Relativity. Unfortunately, the number of bright ($i \lesssim$ 18) QSOs at $z \gtrsim 2$ in the Southern hemisphere is much lower than in the North, due to the lack of wide multi-wavelength surveys at declination $\delta <$ 0$^\circ$, hampering the effectiveness of observations from southern observatories. In this work we present a new method based on Canonical Correlation Analysis to identify such objects, taking advantage of a number of available databases: Skymapper, Gaia DR2, WISE, 2MASS. Our QSO candidate sample lists 1476 sources with $i < 18$ over 12,400 square degrees in the southern hemisphere. With a preliminary campaign we observed spectroscopically 70 of them, confirming 56 new bright QSOs at $z > 2.5$, corresponding to a success rate of our method of $\sim$ 80\%. Furthermore, we estimate a completeness of $\sim$ 90\% of our sample at completion of our observation campaign. The new QSOs confirmed by this first and the forthcoming campaigns will be the targets of subsequent studies using higher resolution spectrographs, like ESPRESSO, UVES, and (in the long term) ELT/HIRES.