Measuring photometric redshifts for high-redshift radio source surveys

TL;DR

This study evaluates machine learning methods for estimating photometric redshifts of high-redshift radio sources, demonstrating that kNN performs best with a significant fraction of high-redshift sources correctly identified.

Contribution

It compiles a homogenized radio-selected dataset from northern and southern hemispheres and compares ML algorithms for redshift estimation, highlighting kNN's superior performance.

Findings

kNN has the lowest percentage of catastrophic outliers.

Sources up to redshift z=3 can be estimated effectively.

Approximately 76% of high-redshift sources are correctly identified.

Abstract

With the advent of deep, all-sky radio surveys, the need for ancillary data to make the most of the new, high-quality radio data from surveys like the Evolutionary Map of the Universe (EMU), GLEAM-X, VLASS and LoTSS is growing rapidly. Radio surveys produce significant numbers of Active Galactic Nuclei (AGNs), and have a significantly higher average redshift when compared with optical and infrared all-sky surveys. Thus, traditional methods of estimating redshift are challenged, with spectroscopic surveys not reaching the redshift depth of radio surveys, and AGNs making it difficult for template fitting methods to accurately model the source. Machine Learning (ML) methods have been used, but efforts have typically been directed towards optically selected samples, or samples at significantly lower redshift than expected from upcoming radio surveys. This work compiles and homogenises a radio-selected dataset from both the northern hemisphere (making use of SDSS optical photometry), and southern hemisphere (making use of Dark Energy Survey optical photometry). We then test commonly used ML algorithms such as k-Nearest Neighbours (kNN), Random Forest, ANNz and GPz on this monolithic radio-selected sample. We show that kNN has the lowest percentage of catastrophic outliers, providing the best match for the majority of science cases in the EMU survey. We note that the wider redshift range of the combined dataset used allows for estimation of sources up to z = 3 before random scatter begins to dominate. When binning the data into redshift bins and treating the problem as a classification problem, we are able to correctly identify $\approx$76% of the highest redshift sources - sources at redshift z $>$ 2.51 - as being in either the highest bin (z $>$ 2.51), or second highest (z = 2.25).