J/A+A/698/A276 KiDS-Bright DR4 galaxy sample zph (John William+, 2025)
Hybrid-z: Enhancing the Kilo-Degree Survey bright galaxy sample photometric
redshifts with deep learning.
John William A., Jalan P., Bilicki M., Hellwing W.A., Thuruthipilly H.,
Nakoneczny S.J.
<Astron. Astrophys. 698, A276 (2025)>
=2025A&A...698A.276J 2025A&A...698A.276J (SIMBAD/NED BibCode)
ADC_Keywords: Galaxy catalogs ; Redshifts ; Optical
Keywords: techniques: miscellaneous - catalogs - surveys -
galaxies: distances and redshifts - galaxies: photometry -
cosmology: observations
Abstract:
We employ deep learning (DL) to improve photometric redshifts
(photo-zs) in the Kilo-Degree Survey Data Release 4 Bright galaxy
sample (KiDS-Bright DR4). This dataset, used as foreground for KiDS
lensing and clustering studies, is flux-limited to r<20mag with mean
z=0.23 and covers 1000deg2. Its photo-zs were previously derived
with artificial neural networks from the ANNz2 package trained on the
Galaxy And Mass Assembly (GAMA) spectroscopy. Here we considerably
improve over these previous redshift estimations by building a DL
model, Hybrid-z, which combines an Inception-based convolutional
neural network operating on four-band KiDS images with an artificial
neural network using nine-band magnitudes from KiDS+VIKING. The
Hybrid-z framework provides state-of-the-art photo-zs for KiDS-Bright,
with negligible mean residuals of O(10-4) and scatter at the level of
0.014(1+z) - representing a reduction by 20% over the previous
nine-band derivations with ANNz2. Our photo-zs are robust and stable
independently of galaxy magnitude, redshift, and color. We check our
photo-z model performance on test data drawn from GAMA, as well as
from other KiDS-overlapping wide-angle spectroscopic surveys, namely
SDSS, 2dFLenS and 2dFGRS. We find stable behavior and consistent
improvement over ANNz2 throughout. We finally apply Hybrid-z trained
on GAMA to the entire KiDS-Bright DR4 sample of 1.2 million galaxies.
Description:
This catalog presents photometric redshift estimates for over 1.2
million galaxies in the KiDS-Bright sample (r<20mag), derived using
a deep-learning framework called Hybrid-z. The method combines
convolutional neural networks (CNNs) that operate on ugri images from
KiDS DR4, with a fully-connected network (ONN) that processes 9-band
GAAP magnitudes (ugriZYJHKs) from KiDS+VIKING.
This catalog includes coordinates, calibrated r-band magnitudes,
bitmask flags, and the photometric redshift output for each source.
The 'masked' column provides a clean selection criterion (0 = clean, 1
= masked). These redshifts are suitable for large-scale structure and
weak lensing analyses.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
catalog.dat 88 1235020 Photometric redshift estimates for over 1.2
million galaxies in the KiDS-Bright sample
(r<20mag)
--------------------------------------------------------------------------------
See also:
II/383 : The ESO KiDS-DR5 multi-band source catalog (Wright+ 2024)
Byte-by-byte Description of file: catalog.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 29 A29 --- ID Source identifier from KiDS DR4 catalog
(ESO ID), KiDSDR4 JHHMMSS.sss+DDMMSS.ss
31- 40 F10.6 deg RAdeg Right ascension (J2000)
42- 51 F10.6 deg DEdeg Declination (J2000)
53- 61 F9.6 mag rmag Zero-point calibrated and extinction-corrected
Kron-like elliptical aperture magnitude in
the r band
63- 70 F8.6 mag e_rmag RMS error for MAG_AUTO
72- 76 I5 --- Mask KiDS 9-band bitmask value
78 I1 --- masked [0/1] Quality flag: 0 = clean, 1 = masked
80- 88 F9.7 --- zphHybridz Photometric redshift from Hybrid-z model
--------------------------------------------------------------------------------
Acknowledgements:
Anjitha John William, anjithajm(at)cft.edu.pl
Center for Theoretical Physics, Polish Academy of Sciences, Warsaw, Poland
(End) Anjitha John William [Poland], Patricia Vannier [CDS] 11-May-2025