Estimating photometric redshifts for X-ray sources in the X-ATLAS field using machine-learning techniques
We present photometric redshifts for 1031 X-ray sources in the X-ATLAS field using the machine-learning technique TPZ. X-ATLAS covers 7.1 deg observed with XMM-Newton within the Science Demonstration Phase of the H-ATLAS field, making it one of the largest contiguous areas of the sky with both XMM-Newton and Herschel coverage. All of the sources have available SDSS photometry, while 810 additionally have mid-IR and/or near-IR photometry. A spectroscopic sample of 5157 sources primarily in the XMM/XXL field, but also from several X-ray surveys and the SDSS DR13 redshift catalogue, was used to train the algorithm. Our analysis reveals that the algorithm performs best when the sources are split, based on their optical morphology, into point-like and extended sources. Optical photometry alone is not enough to estimate accurate photometric redshifts, but the results greatly improve when at least mid-IR photometry is added in the training process. In particular, our measurements show that the estimated photometric redshifts for the X-ray sources of the training sample have a normalized absolute median deviation, nmad - 0.06, and a percentage of outliers, - = 10-14%, depending upon whether the sources are extended or point like. Our final catalogue contains photometric redshifts for 933 out of the 1031 X-ray sources with a median redshift of 0.9. ; The research leading to these results has received funding from the European Union's Horizon 2020 Programme under the AHEAD project (grant agreement No. 654215). G.M. acknowledges financial support from the AHEAD project, which is funded by the European Union as Research and Innovation Action under Grant No: 654215. F.J.C. and A.C.R. acknowledge financial support through grant AYA2015-64346-C2-1-P (MINECO/FEDER). A.C.R. also acknowledges financial support by the European Space Agency (ESA) under the PRODEX program. ; Peer Reviewed