# Multifrequency filter search for high redshift sources and lensing systems in Herschel-ATLAS

## TL;DR

This paper introduces a multifrequency filtering method to identify high-redshift Herschel sources, improving detection sensitivity and estimating photometric redshifts, leading to a catalog of robust high-redshift candidates and potential lensing systems.

## Contribution

The paper presents a novel multifrequency filtering technique for high-redshift source detection and photometric redshift estimation in Herschel data, validated against known sources.

## Key findings

- Identified 370 bright high-redshift candidates.
- Confirmed 237 faint high-redshift sources near detection limit.
- Found potential lensing systems through cross-correlation with galaxy catalogs.

## Abstract

We present a new catalog of high-redshift candidate \textit{Herschel} sources. Our sample is obtained after applying a multifrequency filtering method ("matched multifilter"), which is designed to improve the signal-to-noise ratio (S/N) of faint extragalactic point sources. The method is tested against already-detected sources from the \textit{Herschel} Astrophysical Terahertz Large Area Survey (H-ATLAS) and used to search for new high-redshift candidates. The multifilter technique also produces an estimation of the photometric redshift of the sources. When compared with a sample of sources with known spectroscopic redshift, the photometric redshift returned from the multifilter is unbiased in the redshift range 0.8 < z < 4.3. Using simulated data we reproduced the same unbiased result in roughly the same redshift range and determined the error (and bias above $z\approx4$) in the photometric redshifts. Based on the multifilter technique, and a selection based on color, flux, and agreement of fit between the observed photometry and assumed SED, we find 370 robust candidates to be relatively bright high-redshift sources. A second sample with 237 objects focuses on the faint end at high-redshift. These 237 sources were previously near the H-ATLAS detection limit but are now confirmed with our technique as high significance detections. Finally, we look for possible lensed Herschel sources by cross-correlating the first sample of 370 objects with two different catalogs of known low-redshift objects, the redMaPPer Galaxy Cluster Catalog and a catalog of galaxies with spectroscopic redshift from the Sloan Digital Sky Survey Data Release 14. Our search renders a number of candidates to be lensed systems from the SDSS cross-correlation but none from the redMaPPeR confirming the more likely galactic nature of the lenses.