Herschel-ATLAS: The far-infrared--radio correlation at z<0.5

TL;DR

This study investigates the far-infrared--radio correlation at redshifts 0<z<0.5 using Herschel-ATLAS data, finding no significant evolution and suggesting that previous observed variations may be due to observational resolution effects.

Contribution

It provides the first comprehensive analysis of the far-infrared--radio correlation evolution over 0<z<0.5 with Herschel-ATLAS data, challenging previous claims of evolution.

Findings

No evidence for evolution in the far-infrared--radio correlation up to z<0.5.

The median q_IR value is 2.40±0.12, consistent with local universe measurements.

Extended emission resolution effects may explain previous observed increases in q_IR at low redshift.

Abstract

We use data from the Herschel-ATLAS to investigate the evolution of the far-infrared--radio correlation over the redshift range 0<z<0.5. Using the total far-infrared luminosity of all >5sigma sources in the Herschel-ATLAS Science Demonstration Field and cross-matching these data with radio data from the Faint Images of the Radio Sky at Twenty-Centimetres (FIRST) survey and the NRAO VLA Northern Sky Survey (NVSS), we obtain 104 radio counterparts to the Herschel sources. With these data we find no evidence for evolution in the far-infrared--radio correlation over the redshift range 0<z<0.5, where the median value for the ratio between far-infrared and radio luminosity, $q_{\rm IR}$, over this range is $q_{\rm IR} = 2.40\pm 0.12$ (and a mean of $q_{\rm IR}=2.52 \pm 0.03$ accounting for the lower limits), consistent with both the local value determined from {\em IRAS} and values derived from surveys targeting the high-redshift Universe. By comparing the radio fluxes of our sample measured from both FIRST and NVSS we show that previous results suggesting an increase in the value of $q_{\rm IR}$ from high to low redshift may be the result of resolving out extended emission of the low-redshift sources with relatively high-resolution interferometric data, although AGN contamination could still play a significant role. We also find tentative evidence that the longer wavelength, cooler dust is heated by an evolved stellar population which does not trace the star-formation rate as closely as the shorter wavelength $\ltsim 250~\mu$m emission or the radio emission, supporting suggestions based on detailed models of individual galaxies.