Identifying Circumgalactic Medium Absorption in QSO Spectra: A Bayesian Approach

TL;DR

This paper introduces a Bayesian method to identify galaxy-absorber pairs in QSO spectra using only photometric data, enabling efficient study of circumgalactic medium properties at low redshift.

Contribution

The study develops a novel Bayesian approach combining SDSS photometry with CGM properties to match absorbers to galaxies, achieving high accuracy and broad applicability.

Findings

Identified ~630 candidate galaxy-absorber pairs.

Reproduced known pairs with 72% accuracy.

Median impact parameter of pairs is ~430 kpc.

Abstract

We present a study of candidate galaxy-absorber pairs for 43 low redshift QSO sightlines ($0.06 < z < 0.85$) observed with the {\it Hubble Space Telescope}/Cosmic Origins Spectrograph that lie within the footprint of the Sloan Digital Sky Survey with a statistical approach to match absorbers with galaxies near the QSO lines of sight using only the SDSS Data Release 12 photometric data for the galaxies, including estimates of their redshifts. Our Bayesian methods combine the SDSS photometric information with measured properties of the circumgalactic medium to find the most probable galaxy match, if any, for each absorber in the line of sight QSO spectrum. We find $\sim$630 candidate galaxy-absorber pairs using two different statistics. The methods are able to reproduce pairs reported in the targeted spectroscopic studies upon which we base the statistics at a rate of 72\%. The properties of the galaxies comprising the candidate pairs have median redshift, luminosity, and stellar mass, all estimated from the photometric data, $z=0.13$, $L=0.1L^*$, and $\log(M_*/M_{Sun}) = 9.7$. The median impact parameter of the candidate pairs is $\sim$430~kpc, or $\sim 3.5$ times the galaxy virial radius. The results are broadly consistent with the high \lya\ covering fraction out to this radius found in previous studies. This method of matching absorbers and galaxies can be used to prioritize targets for spectroscopic studies, and we present specific examples of promising systems for such follow-up.