Mining for Protoclusters at z ∼ 4 from Photometric Data Sets with Deep Learning

Protoclusters are high- z overdense regions that will evolve into clusters of galaxies by z = 0, making them ideal for studying galaxy evolution expected to be accelerated by environmental effects. However, it has been challenging to identify protoclusters beyond z = 3 only by photometry due to larg...

Full description

Saved in:
Bibliographic Details
Main Authors: Yoshihiro Takeda, Nobunari Kashikawa, Kei Ito, Jun Toshikawa, Rieko Momose, Kent Fujiwara, Yongming Liang, Rikako Ishimoto, Takehiro Yoshioka, Junya Arita, Mariko Kubo, Hisakazu Uchiyama
Format: Article
Language:English
Published: IOP Publishing 2024-01-01
Series:The Astrophysical Journal
Subjects:
Online Access:https://doi.org/10.3847/1538-4357/ad8a67
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850138760137670656
author Yoshihiro Takeda
Nobunari Kashikawa
Kei Ito
Jun Toshikawa
Rieko Momose
Kent Fujiwara
Yongming Liang
Rikako Ishimoto
Takehiro Yoshioka
Junya Arita
Mariko Kubo
Hisakazu Uchiyama
author_facet Yoshihiro Takeda
Nobunari Kashikawa
Kei Ito
Jun Toshikawa
Rieko Momose
Kent Fujiwara
Yongming Liang
Rikako Ishimoto
Takehiro Yoshioka
Junya Arita
Mariko Kubo
Hisakazu Uchiyama
author_sort Yoshihiro Takeda
collection DOAJ
description Protoclusters are high- z overdense regions that will evolve into clusters of galaxies by z = 0, making them ideal for studying galaxy evolution expected to be accelerated by environmental effects. However, it has been challenging to identify protoclusters beyond z = 3 only by photometry due to large redshift uncertainties hindering statistical study. To tackle the issue, we develop a new deep-learning-based protocluster detection model, PCFNet, which considers a protocluster as a point cloud. To detect protoclusters at z ∼ 4 using only optical broadband photometry, we train and evaluate PCFNet with mock g -dropout galaxies based on the N -body simulation with the semianalytic model. We use the sky distribution, i -band magnitude, ( g − i ) color, and the redshift probability density function surrounding a target galaxy on the sky. PCFNet detects 5 times more protocluster member candidates while maintaining high purity (recall = 7.5% ± 0.2%, precision = 44% ± 1%) than conventional methods. Moreover, PCFNet is able to detect more progenitors ( ${M}_{\mathrm{halo}}^{z=0}={10}^{14-14.5}\,{M}_{\odot }$ ) that are less massive than supermassive clusters like the Coma cluster. We apply PCFNet to the observational photometric data set of the Hyper Suprime-Cam Strategic Survey Program Deep/UltraDeep layer (∼17 deg ^2 ) and detect 121 protocluster candidates at z ∼ 4. We find that the rest-UV luminosities of our protocluster member candidates are brighter than those of field galaxies, which is consistent with previous studies. Additionally, the quenching of satellite galaxies depends on both the core galaxy’s halo mass at z ∼ 4 and accumulated mass until z = 0 in the simulation. PCFNet is very flexible and can find protoclusters at other redshifts or in future extensive surveys by Euclid, Legacy Survey of Space and Time, and Roman.
format Article
id doaj-art-837fd0f36a004c4eabca4443ac9a15c3
institution OA Journals
issn 1538-4357
language English
publishDate 2024-01-01
publisher IOP Publishing
record_format Article
series The Astrophysical Journal
spelling doaj-art-837fd0f36a004c4eabca4443ac9a15c32025-08-20T02:30:31ZengIOP PublishingThe Astrophysical Journal1538-43572024-01-0197718110.3847/1538-4357/ad8a67Mining for Protoclusters at z ∼ 4 from Photometric Data Sets with Deep LearningYoshihiro Takeda0https://orcid.org/0000-0001-7154-3756Nobunari Kashikawa1https://orcid.org/0000-0003-3954-4219Kei Ito2https://orcid.org/0000-0002-9453-0381Jun Toshikawa3https://orcid.org/0000-0001-5394-242XRieko Momose4https://orcid.org/0000-0002-8857-2905Kent Fujiwara5https://orcid.org/0000-0002-2205-6115Yongming Liang6https://orcid.org/0000-0002-2725-302XRikako Ishimoto7https://orcid.org/0000-0002-2134-2902Takehiro Yoshioka8https://orcid.org/0000-0002-3800-0554Junya Arita9https://orcid.org/0009-0007-0864-7094Mariko Kubo10https://orcid.org/0000-0002-7598-5292Hisakazu Uchiyama11https://orcid.org/0000-0002-0673-0632Department of Astronomy, School of Science, The University of Tokyo , 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan ; y.takeda@astron.s.u-tokyo.ac.jpDepartment of Astronomy, School of Science, The University of Tokyo , 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan ; y.takeda@astron.s.u-tokyo.ac.jp; Center for the Early Universe, The University of Tokyo , 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, JapanDepartment of Astronomy, School of Science, The University of Tokyo , 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan ; y.takeda@astron.s.u-tokyo.ac.jp; Cosmic Dawn Center (DAWN) , Denmark; DTU Space, Technical University of Denmark , Elektrovej 327, DK2800 Kgs. Lyngby, DenmarkNishi-Harima Astronomical Observatory, Center for Astronomy, University of Hyogo , 407-2, Nishigaichi, Sayo, Hyogo 679-5313, JapanObservatories of the Carnegie Institution for Science , 813 Santa Barbara Street, Pasadena, CA 91101, USALY Corporation , Tokyo, JapanInstitute for Cosmic Ray Research, The University of Tokyo , 5-1-5 Kashiwanoha, Kashiwa, Chiba 277-8582, Japan; National Astronomical Observatory of Japan , 2-21-1 Osawa, Mitaka, Tokyo 181-8588, JapanDepartment of Astronomy, School of Science, The University of Tokyo , 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan ; y.takeda@astron.s.u-tokyo.ac.jpDepartment of Astronomy, School of Science, The University of Tokyo , 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan ; y.takeda@astron.s.u-tokyo.ac.jpDepartment of Astronomy, School of Science, The University of Tokyo , 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan ; y.takeda@astron.s.u-tokyo.ac.jpAstronomical Institute, Tohoku University , 6-3, Aramaki Aoba, Aoba-ku, Sendai, Miyagi 980-8578, JapanNational Astronomical Observatory of Japan , 2-21-1 Osawa, Mitaka, Tokyo 181-8588, JapanProtoclusters are high- z overdense regions that will evolve into clusters of galaxies by z = 0, making them ideal for studying galaxy evolution expected to be accelerated by environmental effects. However, it has been challenging to identify protoclusters beyond z = 3 only by photometry due to large redshift uncertainties hindering statistical study. To tackle the issue, we develop a new deep-learning-based protocluster detection model, PCFNet, which considers a protocluster as a point cloud. To detect protoclusters at z ∼ 4 using only optical broadband photometry, we train and evaluate PCFNet with mock g -dropout galaxies based on the N -body simulation with the semianalytic model. We use the sky distribution, i -band magnitude, ( g − i ) color, and the redshift probability density function surrounding a target galaxy on the sky. PCFNet detects 5 times more protocluster member candidates while maintaining high purity (recall = 7.5% ± 0.2%, precision = 44% ± 1%) than conventional methods. Moreover, PCFNet is able to detect more progenitors ( ${M}_{\mathrm{halo}}^{z=0}={10}^{14-14.5}\,{M}_{\odot }$ ) that are less massive than supermassive clusters like the Coma cluster. We apply PCFNet to the observational photometric data set of the Hyper Suprime-Cam Strategic Survey Program Deep/UltraDeep layer (∼17 deg ^2 ) and detect 121 protocluster candidates at z ∼ 4. We find that the rest-UV luminosities of our protocluster member candidates are brighter than those of field galaxies, which is consistent with previous studies. Additionally, the quenching of satellite galaxies depends on both the core galaxy’s halo mass at z ∼ 4 and accumulated mass until z = 0 in the simulation. PCFNet is very flexible and can find protoclusters at other redshifts or in future extensive surveys by Euclid, Legacy Survey of Space and Time, and Roman.https://doi.org/10.3847/1538-4357/ad8a67ProtoclustersLyman-break galaxiesGalaxy environmentsHigh-redshift galaxy clusters
spellingShingle Yoshihiro Takeda
Nobunari Kashikawa
Kei Ito
Jun Toshikawa
Rieko Momose
Kent Fujiwara
Yongming Liang
Rikako Ishimoto
Takehiro Yoshioka
Junya Arita
Mariko Kubo
Hisakazu Uchiyama
Mining for Protoclusters at z ∼ 4 from Photometric Data Sets with Deep Learning
The Astrophysical Journal
Protoclusters
Lyman-break galaxies
Galaxy environments
High-redshift galaxy clusters
title Mining for Protoclusters at z ∼ 4 from Photometric Data Sets with Deep Learning
title_full Mining for Protoclusters at z ∼ 4 from Photometric Data Sets with Deep Learning
title_fullStr Mining for Protoclusters at z ∼ 4 from Photometric Data Sets with Deep Learning
title_full_unstemmed Mining for Protoclusters at z ∼ 4 from Photometric Data Sets with Deep Learning
title_short Mining for Protoclusters at z ∼ 4 from Photometric Data Sets with Deep Learning
title_sort mining for protoclusters at z ∼ 4 from photometric data sets with deep learning
topic Protoclusters
Lyman-break galaxies
Galaxy environments
High-redshift galaxy clusters
url https://doi.org/10.3847/1538-4357/ad8a67
work_keys_str_mv AT yoshihirotakeda miningforprotoclustersatz4fromphotometricdatasetswithdeeplearning
AT nobunarikashikawa miningforprotoclustersatz4fromphotometricdatasetswithdeeplearning
AT keiito miningforprotoclustersatz4fromphotometricdatasetswithdeeplearning
AT juntoshikawa miningforprotoclustersatz4fromphotometricdatasetswithdeeplearning
AT riekomomose miningforprotoclustersatz4fromphotometricdatasetswithdeeplearning
AT kentfujiwara miningforprotoclustersatz4fromphotometricdatasetswithdeeplearning
AT yongmingliang miningforprotoclustersatz4fromphotometricdatasetswithdeeplearning
AT rikakoishimoto miningforprotoclustersatz4fromphotometricdatasetswithdeeplearning
AT takehiroyoshioka miningforprotoclustersatz4fromphotometricdatasetswithdeeplearning
AT junyaarita miningforprotoclustersatz4fromphotometricdatasetswithdeeplearning
AT marikokubo miningforprotoclustersatz4fromphotometricdatasetswithdeeplearning
AT hisakazuuchiyama miningforprotoclustersatz4fromphotometricdatasetswithdeeplearning