SRGz: building an optical cross-match model for the X-ray SRG/eROSITA sources using the Lockman Hole data

TL;DR

This paper introduces a probabilistic optical cross-matching model for SRG/eROSITA X-ray sources using DESI Legacy Imaging Surveys data, achieving high precision and recall, and calibrating positional errors in the Lockman Hole.

Contribution

The paper presents a novel probabilistic model that combines positional and photometric data for accurate optical cross-matching of X-ray sources, integrated into the SRGz system.

Findings

Achieved approximately 94% precision and recall in source matching.

Calibrated the positional error for SRG/eROSITA sources in the Lockman Hole.

The model will be incorporated into the SRGz data analysis system.

Abstract

We present a probabilistic model built for the optical cross-match between the SRG/eROSITA X-ray sources and photometric data from the DESI Legacy Imaging Surveys. The model relies both on positional and photometric information on optical objects nearby X-ray sources and allows performing selection with precision and recall $\approx94$% (for $F_{\rm X,0.5-2}>10^{-14}$ erg/s/cm$^2$). With this model, we calibrated positional error of the SRG/eROSITA sources detected in the Lockman Hole: $\sigma_{\rm corr} = 0.87\sqrt{ \sigma_{\rm det}^{2.53} + 1.12^2}$. The model will become a part of the SRGz system for data analysis of the X-ray data obtained from the all-sky SRG/eROSITA survey.