Photometric redshifts for supernovae Ia in the Supernova Legacy Survey

TL;DR

This paper introduces a photometric redshift estimation method for Type Ia supernovae using SALT2 light curve fitting, achieving high precision and low catastrophic error rates, surpassing host galaxy photometric redshifts.

Contribution

The study demonstrates a novel application of SALT2 light curve fitting to accurately determine supernova redshifts from photometry alone, with implications for large-scale surveys.

Findings

Achieved a redshift precision of 0.022 up to z=1.0

Catastrophic error rate of 1.4%

Outperforms host galaxy photometric redshifts in accuracy

Abstract

We present a method using the SALT2 light curve fitter to determine the redshift of Type Ia supernovae in the Supernova Legacy Survey (SNLS) based on their photometry in g', r', i' and z'. On 289 supernovae of the first three years of SNLS data, we obtain a precision $\sigma_{\Delta z/(1+z)} = 0.022$ on average up to a redshift of 1.0, with a higher precision of 0.016 for z<0.45 and a lower one of 0.025 for z>0.45. The rate of events with $|\Delta z|/(1+z)>0.15$ (catastrophic errors) is 1.4%. Both the precision and the rate of catastrophic errors are better than what can be currently obtained using host galaxy photometric redshifts. Photometric redshifts of this precision may be useful for future experiments which aim to discover up to millions of supernovae Ia but without spectroscopy for most of them.