J/A+A/691/A185 GRG constraints from ML and Bayesian inf. (Mostert+, 2024)
Constraining the giant radio galaxy population with machine
learning-accelerated detection and Bayesian inference.
Mostert R.I.J., Oei M.S.S.L., Barkus B., Alegre L., Hardcastle M.J.,
Duncan K.J., Rottgering H.J.A., van Weeren R.J., Horton M.
<Astron. Astrophys. 691, A185 (2024)>
=2024A&A...691A.185M 2024A&A...691A.185M (SIMBAD/NED BibCode)
ADC_Keywords: Galaxies, radio ; Active gal. nuclei ; Surveys ; Radio sources
Keywords: methods: data analysis - catalogues - surveys - galaxies: active -
cosmology: observations - radio continuum: galaxies
Abstract:
Large-scale sky surveys at low frequencies, like the LOFAR Two-metre
Sky Survey (LoTSS), allow for the detection and characterisation of
unprecedented numbers of giant radio galaxies (GRGs, or 'giants').
This, in turn, enables us to study giants in a cosmological context. A
tantalising prospect of such studies is a measurement of the
contribution of giants to cosmic magnetogenesis. However, finding
large GRG samples requires the creation of radio-optical catalogues
for well-resolved radio sources and a suitable statistical framework
to infer intrinsic GRG population properties.
By automating the creation of radio--optical catalogues, we aim to
expand significantly the census of known giants. With the resulting
sample and a forward model mindful of selection effects, we aim to
constrain their intrinsic length distribution, number density, and
lobe volume-filling fraction (VFF) in the Cosmic Web.
We combine five existing codes into a single machine learning-driven
(ML) pipeline that automates radio source component association and
optical host identification for well-resolved radio sources. We create
a radio--optical catalogue for the entire LoTSS Data Release 2 (DR2)
footprint and subsequently select all sources that qualify as possible
giants. We combine the list of ML pipeline GRG candidates with an
existing list of LoTSS DR2 crowd-sourced GRG candidates and visually
confirm or reject all members of the merged sample. To infer intrinsic
GRG properties from GRG observations, we develop further a
population-based forward model and constrain its parameters using
Bayesian inference.
Roughly half of all radio sources that our ML pipeline identifies as
giants (of at least l(p,GRG) 0.7Mpc long) indeed turn out to be upon
visual inspection, whereas the success rate is one in eleven for the
previous best giant-finding ML technique in the literature. We confirm
5647 previously unknown giants from the crowd-sourced LoTSS DR2
catalogue and 2597 previously unknown giants from the ML pipeline.
Our confirmations and discoveries bring the total number of known
giants to at least 11,585. Our forward model for the intrinsic GRG
population is able to provide a good fit to the data. Our posterior
indicates that the projected lengths of giants are consistent with a
curved power law probability density function whose initial tail index
xi(l(p,GRG))=-2.8±0.2 changes by Δxi=-2.4±0.3 over the
interval up to lp=5Mpc. We predict a comoving GRG number density
nGRG=13±10(100Mpc)-3, close to a current estimate of the number
density of luminous non-giant radio galaxies. With the projected
length distribution, number density, and additional assumptions, we
derive a current-day GRG lobe VFF V(GRG-CW)(z=0)=1.1±0.9x10-5 in
clusters and filaments of the Cosmic Web.
We have created a state-of-the-art ML-accelerated pipeline for finding
giants, whose complex morphologies, arcminute extents, and
radio-emitting surroundings pose challenges. Our data analysis
suggests that giants are more common than previously thought. More
work is needed to make estimates of the GRG lobe VFF reliable, but the
first results indicate that it is possible that magnetic fields
originating from giants permeate significant (∼10%) fractions of
today's Cosmic Web.
Description:
Table of newly discovered GRG and existing GRG as described
in section 4.8 of the paper.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
catalog.dat 157 11585 Newly discovered GRG and existing GRG
--------------------------------------------------------------------------------
See also:
J/A+A/635/A5 : LoTSS giant radio galaxies. I. (Dabhade+, 2020)
J/MNRAS/497/5383 : GMRT 610-MHz survey of ELAIS N1 (Ishwara-Chandra+, 2020)
J/other/Galax/9.99 : Giant Radio Galaxies in RACS (Andernach+, 2021)
J/A+A/660/A59 : SAGAN III. New insights into giant radio quasars
(Mahato+, 2022)
J/MNRAS/515/2032 : LoTSS Bootes Deep Field Giant Radio Galaxies
(Simonte+, 2022)
J/A+A/672/A163 : Properties of the 2060 giant radio galaxies (Oei+, 2023)
J/A+A/678/A151 : LoTSS DR2 optical IDs (Hardcastle+, 2023)
Byte-by-byte Description of file: catalog.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 22 A22 --- Name Gian radio galaxy name, ILTJHHMMSS.ss+DDMMSS.s
if known, else empty
(giantradiogalaxy_name)
24- 38 F15.11 deg RAdeg Host right ascension (J2000) (hostRA(deg))
40- 54 F15.11 deg DEdeg Host declination (J2000) (hostDEC(deg))
56- 62 F7.5 Mpc LengthMpc Giant radio galaxy proper length estimate
(projectedproperlength_(Mpc))
64- 70 F7.5 Mpc e_LengthMpc ?=- Giant radio galaxy proper length estimate
error (projectedproperlengtherror(Mpc))
72- 79 F8.4 arcmin LengthAng Giant radio galaxy angular length
(angularlength(arcmin))
81- 87 F7.5 --- z Redshift (redshift_(1))
93- 99 F7.5 --- e_z ?=- Redshift error (redshifterror(1))
101-128 A28 --- Ref Publication used (publication_used) (1)
130-157 A28 --- FistDisc First discovered by (firstdiscoveredby) (1)
--------------------------------------------------------------------------------
Note (1): References as follows:
Andernach et al. 2021 = 2021Galax...9...99A 2021Galax...9...99A, Cat. J/other/Galax/9.99
Bassani et al. 2021 = 2021MNRAS.500.3111B 2021MNRAS.500.3111B
Dabhade et al. 2020 = 2020A&A...635A...5D 2020A&A...635A...5D, Cat. J/A+A/635/A5
Delhaize et al. 2021 = 2021MNRAS.501.3833D 2021MNRAS.501.3833D
Gurkan et al. 2022 = 2022MNRAS.512.6104G 2022MNRAS.512.6104G
Ishwara-Chandra et al. 2020 = 2020MNRAS.497.5383I 2020MNRAS.497.5383I, Cat. J/MNRAS/497/5383
ML (this work) = this work
Mahato et al. 2021 = 2022A&A...660A..59M 2022A&A...660A..59M, Cat. J/A+A/660/A59
Masini et al. 2021 = 2021A&A...650A..51M 2021A&A...650A..51M
Oei et al. 2023 = 2021A&A...650A..51M 2021A&A...650A..51M
RGZ (Hardcastle et al. 2023) = 2023A&A...678A.151H 2023A&A...678A.151H, Cat. J/A+A/678/A151
Simonte et al. 2022 = 2022MNRAS.515.2032S 2022MNRAS.515.2032S, Cat. J/MNRAS/515/2032
Tang et al. 2020 = 2020MNRAS.499...68T 2020MNRAS.499...68T
--------------------------------------------------------------------------------
Acknowledgements:
Rafael Inayat Jacobus Mostert, mostert(at)strw.leidenuniv.nl
Martijn Simon Soen Liong Oei, oei(at)caltech.edu, oei(at)strw.leidenuniv.nl
(End) Patricia Vannier [CDS] 11-Mar-2024