J/A+A/692/A260 PICZL. Image-based photo-z for AGN (Roster+, 2024)
PICZL: Image-based photometric redshifts for AGN.
Roster W., Salvato M., Krippendorf S., Saxena A., Shirley R., Buchner J.,
Wolf J., Dwelly T., Bauer F.E., Aird J., Ricci C., Assef R.J.,
Anderson S.F., Liu X., Merloni A., Weller J., Nandra K.
<Astron. Astrophys. 692, A260 (2024)>
=2024A&A...692A.260R 2024A&A...692A.260R (SIMBAD/NED BibCode)
ADC_Keywords: Active gal. nuclei ; Redshifts ; Optical ; X-ray sources
Keywords: methods: statistical - techniques: photometric - galaxies: active -
quasars: supermassive black holes
Abstract:
Computing reliable photometric redshifts (photo-z) for active galactic
nuclei (AGN) is a challenging task, primarily due to the complex
interplay between the unresolved relative emissions associated with
the supermassive black hole and its host galaxy. Spectral energy
distribution (SED) fitting methods, while effective for galaxies and
AGN in pencil-beam surveys, face limitations in wide or all-sky
surveys with fewer bands available, lacking the ability to accurately
capture the AGN contribution to the SED, hindering reliable redshift
estimation. This limitation is affecting the many 10s of millions of
AGN detected in existing datasets, e.g., those AGN clearly singled out
and identified by SRG/eROSITA.
Our goal is to enhance photometric redshift performance for AGN in
all-sky surveys while simultaneously simplifying the approach by
avoiding the need to merge multiple data sets. Instead, we employ
readily available data products from the 10th Data Release of the
Imaging Legacy Survey for the Dark Energy Spectroscopic Instrument,
which covers >20000deg2 of extragalactic sky with deep imaging and
catalog-based photometry in the grizW1-W4 bands. We fully utilize the
spatial flux distribution in the vicinity of each source to produce
reliable photo-z.
We introduce PICZL, a machine-learning algorithm leveraging an
ensemble of convolutional neural networks. Utilizing a cross-channel
approach, the algorithm integrates distinct SED features from images
with those obtained from catalog-level data. Full probability
distributions are achieved via the integration of Gaussian mixture
models.
On a validation sample of 8098 AGN, PICZL achieves an accuracy
σNMAD of 4.5% with an outlier fraction η of 5.6%. These
results significantly outperform previous attempts to compute accurate
photo-z for AGN using machine learning. We highlight that the
model's performance depends on many variables, predominantly the
depth of the data and associated photometric error. A thorough
evaluation of these dependencies is presented in the paper.
Our streamlined methodology maintains consistent performance across
the entire survey area, when accounting for differing data quality.
The same approach can be adopted for future deep photometric surveys
such as LSST and Euclid, showcasing its potential for wide-scale
realization. With this paper, we release an updated photo-z (including
errors) for the XMM-SERVS W-CDF-S, ELAIS-S1 and LSS fields.
Description:
Photometric redshift catalog of all XMM-SERVS sources across the three
fields (ELAIS-S1, W-CDF-S, and LSS) matched to the DESI Legacy Survey
(LS DR10). The table includes a unique SERVS identifier, LS10 unique
FULLID, optical coordinates (RA and DEC), spectroscopic redshifts from
SERVS (if available), and photometric redshifts from PICZL,
accompanied by 1σ and 3σ uncertainties.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
photo-z.dat 121 3623 Photo-z catalog
--------------------------------------------------------------------------------
See also:
VII/292 : DESI Legacy Imaging Surveys DR8 photometric redshifts
(Duncan, 2022)
J/MNRAS/478/2132 : XMM-LSS field. New XMM-Newton point-source cat.
(Cheng+, 2018)
J/ApJS/256/21 : XMM-SERVS survey: X-ray sources for W-CDF-S + ELAIS-S1
(Ni+, 2021)
Byte-by-byte Description of file: photo-z.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 8 A8 --- XID Unique source ID assigned to each X-ray source
in the original papers (XID) (1)
10- 12 A3 --- Survey [ES1 LSS WCD] Original survey where the source
was detected (SURVEY) (2)
14- 31 A18 --- LS10 Unique LS source ID (DESI Legacy Survey,
LS DR10) assigned to optical
counterpart (LS10_FULLID) (3)
33- 50 F18.15 deg RAdeg Right Ascension (J2000) of the LS10
optical counterpart (LS10_RA)
52- 71 F20.16 deg DEdeg Declination (J2000) of the LS10
optical counterpart (LS10_DEC)
73- 91 F19.15 --- zsp ?=-99 Spectroscopic redshift from original
catalog (SPECZ)
93- 97 F5.3 --- zph Photometric redshift from PICZL (PHZ_PICZL)
99-103 F5.3 --- b1zph PICZL Photometric redshift minimum at 1 sigma
(PHZPICZLl68)
105-109 F5.3 --- B1zph PICZL Photometric redshift maximum at 1 sigma
(PHZPICZLu68)
111-115 F5.3 --- b3zph PICZL Photometric redshift minimum at 3 sigma
(PHZPICZLl99)
117-121 F5.3 --- B3zph PICZL Photometric redshift maximum at 3 sigma
(PHZPICZLu99)
--------------------------------------------------------------------------------
Note (1): see Chen et al. 2018MNRAS.478.2132C 2018MNRAS.478.2132C, Cat. J/MNRAS/478/2132;
Ni et al. 2021ApJS..256...21N 2021ApJS..256...21N, Cat. J/ApJS/256/21.
Note (2): Surveys as follows:
ES1 = European Large-Area ISO Survey-South 1, ELAIS-S1,
Ni et al. 2021ApJS..256...21N 2021ApJS..256...21N, Cat. J/ApJS/256/21
LSS = XMM-Large Scale Structure, lSS,
Chen et al. 2018MNRAS.478.2132C 2018MNRAS.478.2132C, Cat. J/MNRAS/478/2132
WCD = Wide Chandra Deep Field-South, XMM-SERVS W-CDF-S,
Ni et al. 2021ApJS..256...21N 2021ApJS..256...21N, Cat. J/ApJS/256/21
Note (3): It is created by concatenating the LS columns RELEASE, BRICKID
and OBJID.
--------------------------------------------------------------------------------
Acknowledgements:
Willian Rosterm, wroster(at)mpe.mpg.de
(End) Patricia Vannier [CDS] 13-Nov-2024