Active Galactic Nucleus Properties of ∼1 Million Member Galaxies of Galaxy Groups and Clusters at z < 1.4 Based on the Subaru Hyper Suprime-Cam Survey

Herein, we present the statistical properties of active galactic nuclei (AGNs) for approximately 1 million member galaxies of galaxy groups and clusters with 0.1 < cluster redshift (z cl) < 1.4 selected using the Subaru Hyper Suprime-Cam, the so-called CAMIRA clusters. In this research, we focused on the AGN power fraction (f AGN), which is defined as the proportion of the contribution of AGNs to the total infrared (IR) luminosity, L IR (AGN)/L IR, and examined how f AGN depends on (i) z cl and (ii) the distance from the cluster center. We compiled multiwavelength data using the ultraviolet–mid-IR range. Moreover, we performed spectral energy distribution fits to determine f AGN using the CIGALE code with the SKIRTOR AGN model. We found that (i) the value of f AGN in the CAMIRA clusters is positively correlated with z cl, with the correlation slope being steeper than that for field galaxies, and (ii) f AGN exhibits a high value at the cluster outskirts. These results indicate that the emergence of the AGN population depends on the redshift and environment and that galaxy groups and clusters at high redshifts are important in AGN evolution. Additionally, we demonstrated that cluster–cluster mergers may enhance AGN activity at the outskirts of particularly massive galaxy clusters. Our findings are consistent with a related study on the CAMIRA clusters that was based on the AGN number fraction.


Introduction
Determination of the effects of active galactic nuclei (AGNs) on the formation and evolution of galaxy clusters and their member galaxies during the Universe's history is important.This is because (i) almost all galaxies contain supermassive black holes that can influence the host galaxies (e.g., Magorrian et al. 1998;Ferrarese & Merritt 2000;Woo et al. 2013) and (ii) AGNs may affect the dynamics and energetics of galaxy clusters (e.g., Fabian 2012, and references therein).Therefore, studies on these systems offer a unique opportunity to investigate the relation between AGNs and the host galaxies.
Notably, AGN number fraction is an important parameter for understanding the abovementioned effects.It is often defined as the number of AGNs within the member galaxies of a cluster.Many researchers have investigated the AGN number fraction for galaxy groups and clusters (e.g., Krick et al. 2009;Martini et al. 2009;Tomczak et al. 2011;Pentericci et al. 2013;Ehlert et al. 2014;Magliocchetti et al. 2018;Koulouridis & Bartalucci 2019;Koulouridis et al. 2024).In addition, this parameter has been explored using semianalytic galaxy Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence.Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
For example, for galaxy clusters, the AGN number fraction is found to increase with increasing redshift (e.g., Eastman et al. 2007;Martini et al. 2009;Pentericci et al. 2013;Mishra & Dai 2020;Bhargava et al. 2023).This is similar to the redshift evolution of blue galaxies in galaxy clusters (Butcher & Oemler 1984).Furthermore, some studies have reported that the AGN fraction depends on the environment-that is, the fraction is higher in denser environments than in the field (Manzer & De Robertis 2014; see also Manzer &De Robertis 2014 andSantos et al. 2021 for counterarguments).
Recently, based on numerous galaxy groups and clusters up to z ∼ 1.4, Hashiguchi et al. (2023) investigated the dependence of the AGN number fraction on the cluster redshift (z cl ) and distance from the cluster center.These galaxy groups and clusters were discovered using the Hyper Suprime-Cam (HSC; Miyazaki et al. 2018) Subaru Strategic Program (SSP; Aihara et al. 2018aAihara et al. , 2018bAihara et al. , 2019Aihara et al. , 2022)).The HSC-SSP is an opticalimaging survey covering approximately 1200 deg 2 with five broadband filters and approximately 30 deg 2 with five broadband and four narrowband filters (see Bosch et al. 2018;Coupon et al. 2018;Furusawa et al. 2018;Huang et al. 2018;Kawanomoto et al. 2018;Komiyama et al. 2018). 21 Hashiguchi et al. (2023) constructed an unbiased AGN sample by combining AGN selection methods with multiwavelength data.They reported that the AGN number fraction increases with increasing z cl ; its value was higher than that of field galaxies, regardless of z cl .Moreover, they reported that the AGN number fraction primarily contributed by radio-selected AGNs shows considerable excess in the cluster center, while that primarily contributed by infrared (IR)-selected AGNs exhibits a small excess at the cluster outskirts.
A nonnegligible issue concerning the AGN number fraction is its poor statistics owing to low AGN surface densities.Most previous studies have employed tens-hundreds of AGN samples, which tend to cause large Poisson errors when the AGN number fractions are distributed in subsamples, such as in redshift bins.In addition, optical/IR color(s) and luminosity have been utilized for identifying member galaxies that host an AGN.This always involves a trade-off between purity and completeness (e.g., Toba et al. 2015;Assef et al. 2018).Hence, color(s)-or luminosity-based AGN selection may miss weak AGNs or induce contamination from star-forming galaxies.
Herein, we revisit the dependence of the AGN fraction on z cl and distance from the cluster center from an AGN energy perspective.We define the ratio of the contribution of the AGN IR luminosity to the total IR luminosity-i.e., L IR (AGNs)/L IR -as the AGN IR "power fraction" f AGN .We are able to determine this quantity through analysis of the spectral energy distribution (SED) of each galaxy.The advantages of using the AGN power fraction are the following: (i) f AGN can be determined for each member galaxy, allowing discussion of the abovementioned issues, but with small statistical errors, and (ii) we can detect signatures even from weak AGNs that may be missed by color-and luminosity-based selections, as demonstrated by Pouliasis et al. (2020).Following Hashiguchi et al. (2023), we utilized the sample of galaxy groups and clusters discovered using the HSC-SSP.
The remainder of this paper is structured as follows.In Section 2, we describe the galaxy cluster sample and SED analysis used herein to calculate f AGN .The resultant dependence of the AGN power fraction on z cl and distance from the cluster center are presented in Section 3. In Section 4, we discuss the possible uncertainties in these results and their consistency with results based on the AGN number fraction.We also explain the enhancements in AGN activity at the cluster outskirts due to cluster-cluster mergers.Finally, we summarize the primary conclusions in Section 5.All information about galaxy clusters with member galaxy samples, such as coordinates, photometry, and derived physical quantities, is available in catalog form (Appendix A).We also provide the best-fit SED templates of the sources (Appendix B).Throughout this paper, we adopt the cosmology of a flat Universe with H 0 = 70 km s −1 Mpc −1 , Ω M = 0.28, and Ω Λ = 0.72. 22All magnitudes are given according to the AB system, unless specified otherwise.

CAMIRA Galaxy Cluster Catalog
We used the same sample of galaxy groups and clusters as used by Hashiguchi et al. (2023).We refer the reader to that paper for details but present the following brief summary.
We utilized the HSC-selected galaxy group and cluster catalog obtained by applying the CAMIRA (CAMIRA stands for the cluster-finding algorithm based on the multiband identification of red-sequence galaxies; Oguri 2014) to the HSC-SSP data.This algorithm essentially uses the r, i, and z colors for HSC sources with z AB < 24.0 to find the overdensity of galaxies (see Oguri et al. 2018 for additional details).We employed the latest version (s21a_v1) of the CAMIRA catalog with bright-star masks, which provides 27,037 galaxy groups and clusters with a richness of N mem > 10 over ∼1027 deg 2 .The z cl of the CAMIRA groups/clusters is distributed over the range 0.1 < z cl < 1.4; it could be efficiently determined, in contrast to the spectroscopic redshift (z spec ) of the brightest cluster galaxies (BCGs; Figure 7 in Oguri et al. 2018).Figure 1 presents the distributions of N mem and z cl in the CAMIRA clusters (see also Figure 1 in Hashiguchi et al. 2023).The CAMIRA catalog contains 1,052,529 member galaxies and is one of the largest of such catalogs to date.The distribution of the CAMIRA member galaxies on the sky is shown in Figure 2. Hereafter, galaxy groups and clusters are simply referred to as galaxy clusters. 23he redshifts of the member galaxies originate from z spec and photometric redshift (z phot ).We compiled the values of z spec from the literature, such as the Sloan Digital Sky Survey (SDSS; Yorket al. 2000) Data Release (DR) 15 (Aguado et al. 2019), GAMA (Driver et al. 2011) DR2 (Baldry et al. 2018), DEEP2/3 DR4 (Cooper et al. 2012;Newman et al. 2013), 3D-HST v4.1.5 (Skelton et al. 2014;Momcheva et al. 2016), PRIMUS DR1 (Coil et al. 2011;Cool et al. 2013), VIPERS DR2 (Guzzo et al. 2014;Scodeggio et al. 2018), andVVDS final DR (Le Fèvre et al. 2013; see also Tanaka et al. 2018;Nishizawa et al. 2020).Furthermore, we employed the Direct Empirical Photometric code (DEmP: Hsieh & Yee 2014) for member galaxies without z spec .DEmP is an empirical quadraticpolynomial photometric redshift fitting code that exhibits good performance with red-sequence galaxies (see Hsieh et al. 2005 and Hsieh & Yee 2014 for a full description of this redshift code).Hashiguchi et al. (2023) reported that the z phot of member galaxies could be efficiently estimated with a bias (δ z ) of 0.004, a scatter (σ z ) of 0.012, and an outlier rate ( f out ) of 0.008. 24ecause of the high quality involved, we do not herein consider the uncertainty in z phot , which exhibits a negligible impact on the final results.If a member galaxy has a value of z spec , it is used; otherwise, z phot is used.Hereafter, we refer to the redshift calculated in this manner as z mem .We narrowed down the member galaxy sample to sources with |z cl − z mem | 0.05 × (1 + z cl ) for selecting reliable member galaxies (Tanaka et al. 2018; see also Ando et al. 2023).This procedure left 877,642 member galaxies in the CAMIRA clusters, which were used for our subsequent analyses.

Multiwavelength Data Set
We compiled multiwavelength data from the ultraviolet (UV) to the mid-IR (MIR) range for CAMIRA member galaxies.As described in Section 2.4, this multiwavelength data set enables us to derive reliable AGN power fractions.

UV Data
For the far-UV (FUV) and near-UV (NUV) data, we utilized the revised catalog (GUVcat_AIS; 25 Bianchi et al. 2017) from the Galaxy Evolution Explorer (GALEX; Martin et al. 2005).This catalog contains 82,992,062 sources with 5σ limiting magnitudes of 19.9 and 20.8 for the FUV and NUV, respectively.
We extracted 79,750,759 sources with (i) GRANK 1 and (ii) fuv_artifact = 0 and fuv_flags = 0 or nuv_ artifact = 0 and nuv_flags = 0, where GRANK, f/ nuv_artifact, and f/nuv_flags are the primary-source, artifact, and extraction flags, respectively.A source with GRANK = 0 has no other sources within 2 5, while a source with GRANK = 1 is the best source comprising more than one source within this radius.In the GALEX pipeline, the program SExtractor (Bertin & Arnouts 1996) is used for the detection and photometry of sources in GALEX.Here, f/nuv_flags contains eight flag bits, with f/nuv_flags = 0 meaning that there are no warnings about the source extraction process.f/ nuv_artifact is the bitwise logical "OR," and f/ nuv_artifact = 0 implies that a source is unaffected by any artifacts.We refer the reader to Section 6.2 and Appendix A of a paper by Bianchi et al. (2017) and the SExtractor User Manual26 for more details.

u-band Data
Further, we utilized the u-band data from SDSS and the Kilo-Degree Survey (KiDS; de Jong et al. 2013).In particular, we used the SDSS PhotoPrimary table in DR17 (Abdurro'uf et al. 2022) and KiDS DR3 (de Jong et al. 2017), which contain 469,053,874 and 48,736,590 sources, respectively.The SDSS survey has a 5σ limiting u-band magnitude of approximately 22.0, while that of the KiDS survey is approximately 24.3.Because KiDS DR3 does not extend over the entire region covered by the HSC-SSP, we employed the SDSS u-band data for objects outside the KiDS footprint.To ensure reliable u-band fluxes for KiDS sources, we extracted sources with FLAGS_U = 0, where FLAGS_U is the extraction flag output by SExtractor (Section 2.2.1).If an object lies outside the KiDS footprint, we obtain its u-band flux densities based on the SDSS data; otherwise, we refer to the KiDS uband flux.We also checked the consistency between KiDS uband mag (u KiDS ) and SDSS u-band mag (u SDSS ).The weighted mean of u KiDS − u SDSS is 0.05 mag, indicating that the u-band photometry is consistent with the KiDS and SDSS data.

Optical Data
The HSC-SSP survey comprises three layers with different survey depths and areas: wide, deep, and ultradeep layers (Tables 3 and 4 in Aihara et al. 2018a).We used the HSC-SSP s21a wide-layer data obtained between 2014 March and 2021 January, which provide forced photometry in the g, r, i, z, and y bands with 5σ limiting magnitudes of 26.8, 26.4, 26.4, 25.5, and 24.7, respectively.These photometry data are included in the CAMIRA catalog.

Near-IR Data
We compiled near-IR (NIR) data based on the VISTA Kilodegree Infrared Galaxy Survey (VIKING; Edge et al. 2013) DR4,27 containing 94,819,861 sources.We used J-, H-, and K S -band data with 5σ limiting magnitudes of approximately J = 21 in Vega magnitude.The VIKING catalog contains the Vega magnitude of each source.We converted these Vega magnitudes into AB magnitudes by employing the offset values Δm (m AB = m Vega + Δm) for the J, H, and K S bands of 0.916, 1.366, and 1.827, respectively, following González-Fernández et al. (2018).Before cross-matching, we extracted 80,580,274 objects with primary_source = 1 and (jpperrbits < 256 or hpperrbits < 256 or kspperrbits < 256) to ensure clean photometry for uniquely detected objects, similar to Toba et al. (2019b).
Because the VIKING DR4 partially covers the HSC-SSP footprint, we employed data from the UKIRT Infrared Deep Sky Survey (UKIDSS; Lawrence et al. 2007) and Two Micron All Sky Survey (2MASS; Skrutskie et al. 2006).We utilized the UKIDSS Large Area Survey DR11plus obtained from the WSA-WFCAM Science Archive,28 containing 88,298,646 sources.The limiting magnitudes of UKIDSS are 20.2, 19.6, 18.8, and 18.2 Vega magnitudes in the Y, J, H, and K bands, respectively.The UKIDSS catalog lists the Vega magnitudes for each source; subsequently, they are converted into AB magnitudes using offset values Δm (m AB = m Vega + Δm) for the Y, J, H, and K bands of 0.634, 0.938, 1.379, and 1.900, respectively, following Hewett et al. (2006).Before cross-matching, we selected 77,225,762 objects with (priOrSec 0 or =frameSetID) and (jpperrbits < 256 or hpperrbits < 256 or kpperrbits < 256) to ensure clean photometry for uniquely detected objects.For the 2MASS data, we employed the AllWISE catalog (Section 2.2.5) that includes 2MASS photometry.If an object lies outside the VIKING footprint but inside the UKIDSS footprint, we obtain its NIR flux densities based on UKIDSS.If an object lies outside the VIKING footprint and even the UKIDSS footprint, we use 2MASS data.Otherwise, we always refer to the VIKING NIR data.
We evaluated the consistency among the VIKING, UKIDSS, and 2MASS NIR photometry data.For the J band, the weighted mean of J UKIDSS − J VIKING and J UKIDSS − J 2MASS is 0.06 and 0.08 mag, respectively.For the H band, the weighted mean of H UKIDSS − H VIKING and H UKIDSS − H 2MASS is 0.03 and 0.08 mag, respectively.For the K band, the weighted mean of K UKIDSS − Ks VIKING and K UKIDSS − Ks 2MASS is 0.02 and 0.11 mag, respectively.These results indicate that the NIR photometry used herein is broadly consistent with the VIKING, UKIDSS, and 2MASS data.

MIR Data
We obtained MIR data based on the Wide-field Infrared Survey Explorer (WISE; Wright et al. 2010).We utilized the unWISE (Lang 2014;Schlafly et al. 2019) and AllWISE catalogs (Cutri et al. 2021), which provide 2,214,734,224 and 747,634,026 sources, respectively.The 5σ detection limits in the AllWISE catalog are approximately 0.054, 0.071, 0.73, and 5 mJy at 3.4, 4.6, 12, and 22 μm, respectively.The detection limit in the unWISE catalog is deeper by a factor of 2 at 3.4 and 4.6 μm.Following Toba et al. (2017b), we extracted 741,753,366 sources from the AllWISE catalog, with (w1sat = 0 and w1cc_map = 0), (w2sat = 0 and w2cc_map = 0), (w3sat = 0 and w3cc_map = 0), or (w4sat = 0 and w4cc_map = 0) to obtain reliable photometry in each band.The unWISE and AllWISE catalogs contain the Vega magnitude of each source.We converted these Vega magnitudes into AB magnitudes using offset values Δm (m AB = m Vega + Δm) of 2.699, 3.339, 5.174, and 6.620 for 3.4, 4.6, 12, and 22 μm, respectively, following the Explanatory Supplement to the AllWISE Data Release Products. 29For flux densities in unWISE, we corrected possible contamination from surrounding sources by using frac_flux.Since unWISE 3.4 and 4.6 μm flux densities are deeper than those of ALLWISE, we basically used unWISE data for 3.4 and 4.6 μm data, while ALLWISE was used for 12 and 22 μm data.
To find a best-fit SED and estimate physical properties with their uncertainties, CIGALE adopted an analysis module, that is, the so-called pdf_analysis.This module computes the likelihood for all the possible combinations of parameters and generates the probability distribution function for each parameter and each object.Further, it scales the models by a factor (α) to obtain physically meaningful values before computing the likelihood.α is described as follows: where f i and m i are the observed and model flux densities, respectively; f j and m j are the observed and model extensive physical properties, respectively; and σ is the corresponding uncertainty.The only parameter that is subjected to fitting is the scale factor α. Finally, pdf_analysis computes the probability-weighted mean and standard deviation corresponding to the resultant value and its uncertainty for each parameter.An advantage of the methodology adopted in CIGALE is that the models must be computed only once for all the objects because a fixed grid of models is used (see Section 4.3 in Boquien et al. 2019 for a detailed explanation of this module).We adopted a delayed-SFH model by assuming a single starburst with an exponential decay (e.g., Ciesla et al. 2015Ciesla et al. , 2016)), where we parameterized the age of the main stellar population and e-folding time of the main stellar population (τ main ).We chose the single stellar population model of Bruzual & Charlot (2003) by assuming the initial mass function (IMF) of Chabrier (2003) and employed the nebular emission model reported by Inoue (2011) with the implementation of the new CLOUDY H II-region model (Villa-Vélez et al. 2021).For the attenuation of dust associated with the host galaxy, the models proposed by Calzetti et al. (2000) and Leitherer et al. (2002) were used.We parameterized the color excess of the nebular emission lines E(B − V ) lines , extended using the Leitherer et al. (2002) curve.Additionally, we modeled the reprocessed IR emission of dust absorbed from UV/optical stellar emission considering the dust absorption and emission templates of Dale et al. (2014).
We modeled the AGN emission as emission from an accretion disk and dust torus using SKIRTOR31 (Stalevski et al. 2016), a clumpy two-phase torus model produced within the framework of the three-dimensional radiative-transfer code SKIRT (Baes et al. 2011;Camps & Baes 2015).The AGN model in CIGALE comprises seven parameters: the optical depth of the torus at 9.7 μm (τ 9.7 ), a torus-density radial parameter (p), a torus-density angular parameter (q), the angle between the equatorial plane and torus edge (Δ), the ratio between the maximum and minimum torus radii (R R max min ), the viewing angle (θ), and the AGN power fraction ( f AGN ).32 R R max min and θ are fixed to be 30 and 40, respectively, to avoid degeneracy of AGN templates (see Yang et al. 2020).Because f AGN is the primary topic of this study, we parameterized it using fine intervals.For the polar dust emission, we assumed a graybody with a dust temperature T dust polar = 100 K and emissivity β = 1.6 (e.g., Casey 2012).We parameterized E(B − V ) for the polar dust considering the SMC extinction curve (Prevot et al. 1984).
Under the parameter settings summarized in Table 1, we fit the stellar, nebular, AGN, and star formation (SF) components to at most 15 photometric points (FUV, NUV, u, g, r, i, z, y, J, H, and K/K s bands and 3.4, 4.6, 12, and 22 μm) of 877,642 CAMIRA member galaxies.Figure 3 shows the normalized cumulative histogram of the detected bands, demonstrating that about 43% of the sources have more than eight detected bands for the SED fitting.Even if an object is not detected in a band, we utilize this information by setting 5σ upper limits following Toba et al. (2019b) to further constrain the SEDs (Section 3.1).We will discuss how the number of detected bands affects the resultant f AGN in Section 4.1.1.
Notably, the photometry adopted in each catalog is different.As the photometric flux densities are expected to trace the total flux densities, the influence of different photometry methods is likely to be small.Nevertheless, it is worth investigating if physical properties can actually be reliably estimated given the uncertainty of each photometry.This topic will be discussed in Section 4.1.1.We also note that the wavelength coverage of our data is limited to UV-MIR.This is because the energy output is expected to peak in the UV (if unprocessed) and IR (when dust-reprocessed); thus, the data set we used will be sufficient to estimate the AGN power fraction.However, since some CAMIRA member galaxies are detected by using other wavelengths such as X-ray and radio (Hashiguchi et al. 2023), it is worth investigating how the lack of these data, especially X-ray data, affects the resulting AGN power fraction.This topic will be discussed in Section 4.1.2.

SED Fitting Results
Figure 4 presents examples of SED fitting obtained with CIGALE.The normalized cumulative distribution of the reduced χ 2 is shown in Figure 5. 33 We confirmed that 822,666 out of 877,642 objects (approximately 94%) exhibited reduced χ 2 < 2.0, while 851,294 out of 877,642 objects (approximately 97%) demonstrated reduced χ 2 < 3.0.Those results suggest that the data were well fitted by CIGALE using a combination of the stellar, nebular, AGN, and SF components.Meanwhile, notably, approximately 74% of the sources have reduced χ 2 < 0.5, which may mean that the data are overfitted and that the estimated uncertainties may be underestimated.In Section 4.1.1,we will discuss how the AGN power fraction of sources with low reduced χ 2 are affected.Hereafter, we will focus on the subsample of 822,666 sources with reduced χ 2 < 2.0.We refer to these sources as the "CAMIRA AGN power sample (CAMIRA_AP)."In this work, we primarily focus on the AGN power fraction ( f AGN ) of CAMIRA_AP member galaxies.The other AGN properties estimated by CIGALE (i.e., AGN torus-related values) are also provided in Appendix C.

AGN Power Fraction Distribution
In this study, we focused on galaxies that are expected to be member galaxies of clusters by comparing z cl and z mem , as outlined in Section 2.1.Notably, the spectroscopic completeness of the CAMIRA_AP member galaxies is considerably low (28,966/822,666 ∼ 3.5%); therefore, we relied on z phot for most galaxies.Hence, foreground and background galaxies could possibly contaminate the member galaxies.The CAMIRA catalog assigns a weight factor (w mem34 ) to each member galaxy corresponding to its membership probability.This factor is obtained by applying the fast Fourier transform to the two-dimensional richness map containing properties of galaxies (e.g., color and stellar mass).A full description of w mem , ranging between 0 and 1, is provided by Oguri (2014).A galaxy with w mem closer to 1 is more likely to be a cluster member galaxy, as evident from its increased membership probability.
Therefore, we employed a membership-probability-weighted AGN power fraction to mitigate the influence of contamination from foreground and background galaxies, as introduced by Bufanda et al. (2017).Because the AGN power fraction ( f AGN = L IR (AGNs)/L IR ) can be calculated for each member galaxy and each galaxy is assigned a probability of being a cluster member (w mem ), the membership-probability-weighted AGN power fraction is defined here in a CAMIRA_AP cluster where f i AGN and w i mem are the AGN power fractions of the ith member galaxy in a CAMIRA_AP cluster and its corresponding membership weight factor, respectively.The uncertainty in f AGN cl is calculated as its standard deviation.As this work compares f AGN for galaxy clusters with that for field galaxies (Section 3.3), we define the AGN power fraction for field galaxies ( f AGN fd ).For this purpose, we selected 503,595 field galaxies using the HSC-SSP with the same magnitude and color cuts as those used for red-sequence galaxies (Oguri 2014;Oguri et al. 2018), although the field galaxies do not belong to a cluster.We collected multiwavelength data from the UV-MIR range in exactly the same manner as that discussed in Section 2.2 and estimated f AGN based on SED fitting using the parameter sets described in Section 2.4.The range of the redshift of the field galaxies (z fd ) is the same as that of the CAMIRA_AP member galaxies.For subsequent analysis, we extracted 426,149 field galaxies by adopting reduced χ 2 < 2.0 from the SED fitting, in the same manner as that adopted for cluster member galaxies (Section 3.1).Because a field galaxy, by definition, does not have w mem , we used only field galaxies with z spec and measured the AGN power fraction by weighting the uncertainties of f AGN rather than w mem .Hence, f AGN fd is formulated as follows: where f i AGN and s f i AGN are the AGN power fractions of the ith field galaxy and its uncertainty, respectively.The uncertainty in f AGN fd is calculated as its standard deviation.Figure 6 presents the distribution of f AGN cl in CAMIRA_AP clusters.The mean and standard deviations of f AGN cl are 0.45 and 0.07, respectively.We also divided the cluster sample into subsamples based on redshifts (z cl 0.4, 0.4 < z cl 0.8, 0.8 < z cl 1.2, and z cl > 1.2) following Hashiguchi et al. (2023).The numbers of clusters in the redshift bins are 5193, 10,819, 10,113, and 912, respectively.We examined the f AGN cl distribution of the clusters in each redshift bin (Figure 7).The peak value of the histogram for each panel gradually increases with increasing z cl , providing tentative evidence for the redshift dependence of the AGN power fraction (Section 3.3.1 for a quantitative discussion).However, the range of the distribution is relatively large regardless of the redshift, and this should be kept in mind in the subsequent discussion.
Figure 8 shows the distribution of f AGN fd of field galaxies in each redshift bin (z fd 0.4, 0.4 < z fd 0.8, 0.8 < z fd 1.2, and z fd > 1.2), and these redshift bins have 145,377,200,463,73,347, and 6962 field galaxies, respectively.The peak value of the histogram for each panel slightly increases with increasing  z fd ; this increase also provides tentative evidence for the redshift dependence of the AGN power fraction even for field galaxies (see Section 3.3.1 for a quantitative discussion).The mean and standard deviations of f AGN fd over the whole redshift range (0.1 < z fd < 1.4) are 0.39 and 0.13, respectively, which are lower than those for galaxy clusters, suggesting that the AGNs may tend to ignite in a dense environment.

Redshift and Clustercentric Radius Dependence of the AGN Power Fraction for Galaxy Clusters
We next consider the dependence of the AGN power fraction on (i) redshift and (ii) distance (R) from the cluster center (clustercentric radius) scaled using the virial radius (R 200 ), R/R 200 .Further, we compared the AGN power fraction for clusters with that for field galaxies.Here, we stacked the redshift and radial distributions of member galaxies in redshift or R/R 200 bins.We then calculated the AGN power fraction and its uncertainty in each bin by applying the w mem weights, similar to that described in Section 3.2 (Equation ( 2)).Those enabled us to compare the AGN power fraction for clusters to that for field galaxies, as described in Sections 3.3.1 and 3.3.2.Hereafter, we denote the AGN power fraction calculated in this manner as f AGN mem .

Redshift Dependence of the AGN Power Fraction
Figure 9 presents the AGN power fraction as a function of redshift.The mean relative error of the AGN power fraction across the redshift bin is approximately 0.2%. 35We verified that f AGN mem increases with increasing redshift, even from an AGN power fraction point of view, similarly to what is reported for the AGN number fraction by previous studies (e.g., Galametz et al. 2009;Martini et al. 2009;Haggard et al. 2010;Hashiguchi et al. 2023).We performed a linear regression to fit the z memf AGN mem relation considering the uncertainties in each redshift bin.Additionally, we calculated the correlation coefficient (r) based on a Bayesian regression method (Kelly 2007), which yields the correlation coefficient with its corresponding uncertainty (e.g., Toba et al. 2019aToba et al. , 2021b)).The resulting value r = 0.93 ± 0.11 indicates a strong positive correlation between redshift and AGN power fraction; in other words, a Butcher-Oemler-like (Butcher & Oemler 1984) effect for AGNs in galaxy clusters was confirmed, even from the perspective of the AGN power fraction.
A positive correlation was found between redshift and AGN number fraction, as has been reported for "field galaxies" by many researchers (e.g., Silverman et al. 2009;Haggard et al. 2010;Oi et al. 2021).Buat et al. (2015) also reported that the AGN power fraction increases with increasing redshift.To determine if this observed trend is specific to the AGNs in clusters or if it merely reflects the trend observed in field galaxies, we calculated the AGN fraction for field galaxies detected using the HSC-SSP.
Figure 9 shows the AGN power fraction for the field galaxies as a function of redshift.We here calculated the AGN power fraction and its uncertainty in each redshift bin by applying the s f AGN weights, similar to what is described in Equation (3).An increasing trend of AGN fraction toward high redshift is also confirmed for field galaxies, with a correlation coefficient of r = 0.86 ± 0.18.Moreover, we observe that the best-fit slope for clusters is steeper than that for the field, indicating accelerated AGN growth in cluster galaxies with increasing redshift compared to the field, as reported by previous studies (Eastman et al. 2007;Hashiguchi et al. 2023).The average excess ( f AGN mem / f AGN fd ) is about 1.2 at z > 0.6 (bottom panel of Figure 9).A rapidly increasing AGN number fraction at high redshifts has also been reported for MIR-selected AGNs (Tomczak et al. 2011;Hashiguchi et al. 2023; see Section 4.2 in Hashiguchi et al. 2023 for more details).These results suggest that a denser environment (galaxy clusters), particularly in a high-z Universe, tends to boost AGN activity.

Clustercentric Radius Dependence on AGN Power Fraction
Next, we examined the AGN power fraction as a function of the projected distance from the cluster center (clustercentric radius) scaled using R 200 , R/R 200 ; we determined the cluster centers using the centroids of the BCGs identified with the  Figure 10 shows the resultant f AGN mem as a function of R/R 200 .We found an excess of f AGN mem at the outskirts of galaxy clusters, which is in good agreement with the results of previous studies (e.g., Khabiboulline et al. 2014;Koulouridis & Bartalucci 2019).Hashiguchi et al. (2023) also reported that the MIRselected AGNs are dominant at the outskirts of CAMIRA clusters, while the radio-selected AGNs are condensed to the cluster center.Because the AGN power fraction in this research is based on the IR luminosity, the consistency of our result with that for the MIR AGNs reported by Hashiguchi et al. (2023; see also Section 4.1.3)is reasonable.

Mock Analysis
Given the photometric uncertainties, we checked the reliability of the f AGN derived using CIGALE.We performed a mock analysis, which is a procedure provided by CIGALE.This mock analysis was performed by creating a mock catalog.To build the mock catalog, the best-fit value for each object was considered; each quantity was then modified by adding a value taken from a Gaussian distribution with the same standard deviation as the observation uncertainty.Finally, the same method used in the original estimation was applied to obtain mock estimations.Through this analysis, the reliability of the obtained estimations for the physical parameters was estimated.A full description of this process can be found in a paper by Boquien et al. (2019; see also Ciesla et al. 2015Ciesla et al. , 2016;;Toba et al. 2019bToba et al. , 2020a;;Pouliasis et al. 2022).
Figure 11 shows the differences in the f AGN derived herein using CIGALE and those derived from the mock catalog as a function of redshift and clustercentric radius.The mean (with standard deviation) of Δf AGN is 0.06 ± 0.11, which is acceptable for this research.This result could also suggest that the derived AGN power fraction is not sensitive to the photometric uncertainty.We also proved that Δf AGN is not considerably dependent on the redshift and distance from the cluster center, indicating that f AGN is well determined with negligible systematic uncertainty.
In Sections 2.4 and 3.1, we reported that the number of detected bands and reduced χ 2 obtained through SED fitting have a large variation (Figures 3 and 5).We also tested how the number of detected bands and reduced χ 2 depend on Δf AGN .Figure 12 shows Δf AGN as a function of the number of detected bands and reduced χ 2 , demonstrating that Δf AGN does not substantially depend on these values.Therefore, we conclude that the limited number of detected bands and threshold of reduced χ 2 in this research did not cause systematic uncertainty in the power fraction.This result could also suggest that the influence of the difference in photometry for the SED fitting is likely to be small, as mentioned in Section 2.4.

Lack of X-Ray Information
Herein, UV-MIR data were employed to constrain the SED and estimate the AGN power fraction, as described in Section 2. However, some CAMIRA member galaxies are detected using other wavelengths, such as those of X-rays and radio waves (Hashiguchi et al. 2023).In particular, X-ray data may be crucial to constrain AGN properties such as f AGN , as reported by Yang et al. (2020).Since CIGALE is capable of handling X-ray data, as demonstrated by Yang et al. (2020Yang et al. ( , 2022)), we tested how the lack of X-ray data affects the resultant f AGN .For this purpose, we used 263 X-raydetected CAMIRA member galaxies reported by Hashiguchi et al. (2023), who used XMM-Newton data to identify X-ray sources.We conducted SED fitting to those 263 X-ray sources with and without X-ray flux and compared the calculated AGN power fractions.Figure 13 shows the differences in the AGN  power fraction with and without adding X-ray information to SED fitting with CIGALE as a function of X-ray luminosity (L X ) in the 2-10 keV range, where L X was calculated as shown in a study by Hashiguchi et al. (2023).We found that the weighted mean and its standard deviation of f AGN (w/ X-ray)-f AGN (wo/X-ray) is −0.10 ± 0.04.We also found that this offset depends on the X-ray luminosity.For a lowluminosity regime, f AGN tends to be overestimated if X-ray information is not used.Meanwhile, for a high-luminosity regime, f AGN tends to be underestimated if X-ray information is not used.These tendencies are in good agreement with those reported by Mountrichas et al. (2021).Thus, we should note the possible systematics of the AGN power fraction in this research.

Comparisons with the AGN Number Fractions
Hashiguchi et al. (2023) have identified 2688 AGNs based on multiwavelength data and determined the AGN number fraction for the same initial sample as in this study (i.e., CAMIRA clusters).They have also reported that 2536 CAMIRA clusters have at least one member galaxy that hosts an AGN.We therefore expected that (i) the member galaxies hosting AGNs in a study by Hashiguchi et al. (2023) will have systematically larger values for f AGN than those without AGNs and (ii) the AGN number fraction for each cluster will be correlated with f AGN cl .Figure 14 compares the distributions of the AGN power fraction (we obtained through SED fitting) for the member galaxies that do and do not host AGNs that were determined by Hashiguchi et al. (2023).We found that the f AGN of AGNs identified by Hashiguchi et al. (2023) is systematically higher than that of objects not identified as AGNs by Hashiguchi et al. (2023), which is supported with >99.9% significance by a twosided Kolmogorov-Smirnov (KS) test.The mean values of f AGN for member galaxies with and without AGNs are 0.72 and 0.45, respectively.Figure 15 compares the AGN number fraction (Hashiguchi et al. 2023) with the AGN power fraction ( f AGN cl ) for each cluster.We confirmed a positive correlation between the AGN number and power fractions, with a correlation coefficient of r∼0.5.This implies that, as expected, a cluster with a high AGN number fraction tends to have a high AGN power fraction.We note that about 5% of objects unclassified as AGNs by Hashiguchi et al. (2023) have a large AGN power fraction ( f AGN > 0.7).Those objects may be good candidates for heavily obscured AGNs that were missed by previous surveys.

Stellar Mass and Star Formation Rate Dependence
We observed a positive correlation between the AGN power fraction and redshift for both member galaxies in galaxy clusters and those in the field, as discussed in Section 3.3.1.However, some studies have reported that the AGN number fraction depends on the stellar mass (M * ) of the host galaxy (e.g., Kauffmann et al. 2003;Best et al. 2005;Koss et al. 2011;Bitsakis et al. 2015;Miraghaei 2020).Because our sample is flux-limited, more massive host galaxies at higher redshifts could exhibit more luminous AGNs, leading to a biased positive correlation between f AGN and redshift.This may also be the case for the star formation rate (SFR) because the stellar mass and SFR are often related, and this relation would evolve along with the redshift reported for star-forming galaxies (e.g., Noeske et al. 2007;Schreiber et al. 2015;Pearson et al. 2018).
Following Hashiguchi et al. (2023), we divided our sample into subsamples to investigate how the stellar mass and SFR of member galaxies influence the correlation between the AGN power fraction and redshift.We defined the subsamples using the following ranges of stellar mass: .We obtained M * and SFR from CIGALE outputs, as described in Section 2.4. Figure 16 depicts f AGN as a function of redshift for three subsamples with stellar mass and SFR.We found no significant differences in the correlation coefficients among the subsamples for stellar mass or SFR that were within the respective errors.Hence, we concluded that the observed positive correlation of z clf AGN mem was not affected by such biases.
Figure 12.Differences between the f AGN derived herein using CIGALE and the values derived using the mock catalog as a function of the number of detected bands (top) and reduced χ 2 (bottom).
Figure 13.Differences in the AGN power fraction derived using CIGALE with and without X-ray information as a function of X-ray luminosity in the 2-10 keV range.The red solid and dotted lines indicate the weighted mean and standard deviation, respectively, over the X-ray luminosity range of the sample.

Cluster Mass Dependence on AGN Power Fraction
We also investigated how the AGN power fraction depends on M 200 (or richness).As shown in Figure 17, the AGN power fraction decreases with increasing M 200 , with the correlation coefficient being r = −0.82± 0.25.This trend is in good agreement with the results of previous studies (e.g., Koulouridis et al. 2018;Noordeh et al. 2020;Hashiguchi et al. 2023).It was also observed in a study by Popesso & Biviano (2006), who reported an anticorrelation between the AGN number fraction and velocity dispersion of galaxy clusters. 36Because the velocity dispersion of a galaxy cluster can be translated into M 200 (e.g., Smith 1936), this provides additional support for our findings.These results suggest that a galaxy in a group environment is more likely to ignite an AGN than in a cluster environment, which is consistent with previous studies by Pentericci et al. (2013), Li et al. (2019), andHashiguchi et al. (2023).
In addition, we examined how the AGN power fraction depends on the clustercentric radius in different M 200 ranges (Figure 18).Notably, massive clusters tend to be responsible for the rapid increase in the AGN power fraction at the cluster outskirts, as reported in Section 3.3.2.We further discuss this point in Section 4.3, considering the morphologies of the galaxy clusters.

Cluster Morphology Dependence on AGN Power Fraction
We next consider how the emergence of AGNs depends on the substructure and morphology of galaxy clusters.Okabe et al. (2019) identified merging-cluster candidates based on the CAMIRA cluster catalog using a peak-finding method.This method involves counting the number of peaks above a redshift-dependent threshold based on Gaussian-smoothed maps of the number densities of member galaxies.A galaxy cluster with a single peak was classified as a "relaxed" cluster, while a cluster with more than two peaks was considered a "merging" cluster.Accordingly, they classified 2558 out of 27,037 CAMIRA clusters (approximately 9.5%) as merging clusters, with 150,684 member galaxies associated with them.Figure 19 presents the distribution of f AGN cl for relaxed and merging clusters.There was no significant difference between the two clusters, and we confirmed this with >99.9% significance using a two-sided KS test.This result suggests that cluster-cluster mergers may not necessarily trigger AGNs, as also reported by Silva et al. (2021) and Hashiguchi et al. (2023).The dynamical activity within a cluster may also activate SF activity in member galaxies (e.g., Miller & Owen 2003;Sobral et al. 2015;Stroe et al. 2015;Okabe et al. 2019).Stroe & Sobral (2021) recently reported that a large fraction of emission-line galaxies in merging clusters are powered by star formation rather than by AGNs.On the other hand, Noordeh et al. (2020) suggested that merging-cluster environments may contribute to the enhancement of AGN activity.Several enhancement mechanisms have been proposed for SF and AGN activity in merging clusters, including gas incorporation driven by ram pressure and galaxy-galaxy interactions (Treu et al. 2003, and references therein).The relative strengths of SF and AGNs may depend on the abovementioned dominant mechanism and the sequence of cluster-cluster mergers.Future statistical work that considers the merger stage of a cluster-cluster merger may provide a way to resolve this issue.
Figure 20 illustrates the clustercentric radius dependence of the AGN fraction for relaxed and merging clusters.We found that AGNs were enhanced in relaxed clusters and that clustercluster mergers may not lead to increased AGN activity in the cluster center.The member galaxies may be moving too quickly to interact with each other, particularly in the cluster center, even if a cluster-cluster merger has occurred.We also found that AGNs are more likely to be enhanced at the outskirts of merging clusters rather than relaxed ones.These findings are in good agreement with those obtained from the AGN number fraction (Hashiguchi et al. 2023).The suppression of AGN activity at the cluster center of the merging clusters was also reported by a cosmological simulation (Chadayammuri et al. 2021).Figure 21 illustrates the clustercentric radius dependence of the AGN fraction for relaxed and merging clusters in some redshift bins (z 0.4, 0.4 < z 1.0, and z > 1.0).The observed trend seen in Figure 20 seems to be established at z 1.0.Although the AGNs may be enhanced even in the cluster center in merging clusters at z > 1.0, we need more samples to confirm this possibility.Because the AGN activity in massive clusters may be enhanced in the outskirts of clusters (as mentioned in Section 4.2), merging clusters with larger M 200 values are more likely to experience increased AGN activity in their outskirts.Notably, the fraction of AGNs in galaxy-galaxy mergers increases with increasing luminosity (e.g., Treister et al. 2012;Glikman et al. 2015;Dietrich et al. 2018;Weigel et al. 2018).We tested how the AGN power fraction depends on AGN bolometric luminosity (L bol ).To estimate L bol , we integrated the best-fit SED template of the AGN component output by CIGALE over wavelengths longward of Lyα in the same manner as that performed by Toba et al. (2017a).
Figure 22 shows the AGN power fraction as a function of AGN bolometric luminosity for CAMIRA_AP member galaxies.We found that f AGN is strongly correlated with L bol with a correlation coefficient of r = 0.92 ± 0.13.This result supports the idea that luminous AGNs can be enhanced in a dense environment, thereby merging clusters.
We note that the optical center of a cluster does not always correspond to the peak galaxy density, and it may also significantly differ from the X-ray center in certain instances, as reported by various researchers, including Mahdavi et al. (2013), Oguri et al. (2018), andOta et al. (2023).This effect may be more severe when clusters merge.Although the fraction of merging clusters was small in the present study, this effect could impact the overall trend discussed in Section 3.3.2.It is therefore important to consider the potential uncertainty in R for merging clusters.

AGN Feedback to SFR in Galaxy Clusters
Finally, we discuss how the AGN power fraction depends on the SFR of CAMIRA_AP member galaxies and field galaxies.Figure depicts the AGN power fraction as a function of SFR for member galaxies of clusters and field galaxies.The AGN power fraction of field galaxies ( f AGN fd ) does not exhibit any remarkable SFR dependence.In contrast, the AGN power fraction of CAMIRA_AP member galaxies ( f AGN mem ) rapidly decreases when log SFR > 0. In other words, when the AGN power fraction is high, the SFR tends to be small, possibly implying that the AGN feedback quenches the SF activity in member galaxies of the cluster.This implication suggests that AGN feedback (such as gas heating and outflow by AGNs) is more effective in dense environments (see also Boselli et al. 2016;Maier et al. 2022;Peluso et al. 2022).

Summary
We have investigated how AGN activity depends on the environment-in particular, on z cl and the distance from the cluster center-from an AGN-energy-contribution point of   view.Following Hashiguchi et al. (2023), we utilized one of the largest optically selected galaxy cluster catalogs: the CAMIRA clusters selected with the Subaru HSC.For approximately 1 million member galaxies of CAMIRA clusters in the redshift range 0.1 < z cl < 1.4, we collected multiwavelength data from the UV-MIR range and performed SED fitting to determine the AGN power fractions.To mitigate against possible contamination from foreground and background galaxies, we introduced a membership-probabilityweighted AGN power fraction and determined how this value depends on z cl and clustercentric radius.Our primary findings are as follows.
1.In agreement with recent studies based on the AGN number fraction, we find that the AGN power fraction increases with increasing redshift for cluster members and field galaxies.In addition, the AGN power fraction for galaxy clusters increases more rapidly than for field galaxies (Section 3.3.1). 2. The AGN power fraction increases toward the outskirts of galaxy clusters, which is consistent with the results reported by Hashiguchi et al. (2023) based on the number fraction of IR-selected AGNs.In contrast, the AGN power fraction decreases with increasing M 200 , suggesting that AGN formation may be favored in galaxy groups (Sections 3.3.2and 4.2). 3.Although the centers of merging clusters may be somewhat uncertain, we find that cluster-cluster mergers may not be the primary trigger for AGN activity in member galaxies.However, a cluster-cluster merger may enhance AGN formation at the outskirts of a cluster, especially for massive galaxy clusters (Section 4.3).4. We have tentative evidence of AGN negative feedback in clusters, suggesting that AGNs could suppress the SF activity of member galaxies in denser environments (Section 4.4).
These results indicate that the emergence of an AGN population is influenced by its environment and redshift and that galaxy groups and clusters at high redshifts are perhaps crucial in AGN evolution.However, most CAMIRA cluster members have not yet been spectroscopically confirmed.Future spectroscopic studies using next-generation multiobject spectrographs-such as the Subaru Prime Focus Spectrograph (Takada et al. 2014  The best-fit SED derived from CIGALE is available for each galaxy on Zenodo, doi:10.5281/zenodo.11049732.Table 3 provides a brief summary of the contents.We encourage using a template of objects with reduced χ 2 < 2.0 for science (Section 3.1).

Figure 1 .
Figure 1.Distributions of cluster redshift (z cl ) and richness (N mem ) for the CAMIRA clusters, color-coded by the number of objects per pixel.

Figure 4 .
Figure4.Examples of SED fitting for CAMIRA member galaxies.The black points represent photometric data, while the gray solid line represents the best-fit SED; the contributions of the stellar, AGN, and SF components to the total SED are shown as blue, yellow, and red lines, respectively.The AGN power fraction increases from the top left to bottom right.

Figure 5 .
Figure5.Normalized cumulative histogram of reduced χ 2 obtained through SED fitting for CAMIRA member galaxies.The vertical line indicates the reduced χ 2 threshold to make a subsample.

Figure 6 .
Figure 6.Histogram of the membership-probability-weighted AGN power fraction ( f AGN cl ) for the CAMIRA_AP clusters.

Figure 7 .
Figure 7. Histograms of the AGN power fraction ( f AGN cl ) for the CAMIRA_AP clusters in each redshift bin.

Figure 8 .
Figure8.Histograms of the AGN power fraction ( f AGN fd ) for the field galaxies in each redshift bin.

Figure 9 .
Figure 9. Red circles and blue squares represent the AGN power fractions in the CAMIRA_AP clusters and field, respectively.The vertical error bars are calculated based on the standard deviation of the weighted mean.The mean relative error of the AGN power fraction across the redshift bin is approximately 0.2%.The solid lines with shaded regions represent the bestfit linear regressions with 1σ confidence intervals.The bottom panel shows the ratios of f AGN mem and f AGN fd .

Figure 10 .
Figure 10.AGN power fraction as a function of clustercentric radius scaled using the virial radius (R/R 200 ).The vertical error bars were calculated based on the standard deviation of the weighted mean.The mean relative error of the AGN power fraction across the R/R 200 bin is about 0.2%.The green shaded region shows an average AGN power fraction for the field.Figure11.Differences in the f AGN derived herein using CIGALE and those derived using the mock catalog as a function of redshift (top) and distance from the cluster center (bottom); both are color-coded according to the number of objects per pixel.

Figure 11 .
Figure 10.AGN power fraction as a function of clustercentric radius scaled using the virial radius (R/R 200 ).The vertical error bars were calculated based on the standard deviation of the weighted mean.The mean relative error of the AGN power fraction across the R/R 200 bin is about 0.2%.The green shaded region shows an average AGN power fraction for the field.Figure11.Differences in the f AGN derived herein using CIGALE and those derived using the mock catalog as a function of redshift (top) and distance from the cluster center (bottom); both are color-coded according to the number of objects per pixel.

Figure 15 .
Figure 15.Comparison of the AGN power fraction ( f AGN cl ) obtained herein for each cluster with the AGN number fraction reported by Hashiguchi et al. (2023).

Figure 17 .
Figure 17.AGN power fraction as a function of cluster mass (M 200 ).The vertical error bars represent the standard deviations of the weighted means.

Figure 18 .
Figure 18.AGN power fraction as a function of clustercentric radius scaled using the virial radius (R/R 200 ) for different cluster mass ranges (blue diamonds:  ( ) M M log cl 14.0, green squares: <

Figure 21 .
Figure 21.AGN power fraction as a function of clustercentric radius scaled using the virial radius (R/R 200 ) for relaxed (blue circles)> and merging (red squares) clusters in each redshift bin.

Figure 22 .
Figure 22.AGN power fraction as a function of bolometric luminosity.

Figure 23 .
Figure23.AGN power fraction as a function of SFR for member galaxies in CAMIRA_AP clusters (red circles) and field galaxies (blue squares).

Table 1
Parameter Values Used in SED Fitting with CIGALE ; see alsoGreene etal.2022)-will help to resolve this issue and provide more robust conclusions.Institut für Astronomie (MPIA Heidelberg), Max-Planck-Institut für Astrophysik (MPA Garching), Max-Planck-Institut für Extraterrestrische Physik (MPE), National Astronomical Observatories of China, New Mexico State University, New York University, the University of Notre Dame, Observatário Nacional/MCTI, The Ohio State University, Pennsylvania State University, Shanghai Astronomical Observatory, the United Kingdom Participation Group, Universidad Nacional Autónoma de México, the University of Arizona, the University of Colorado Boulder, the University of Oxford, the University of Portsmouth, the University of Utah, the University of Virginia, the University of Washington, the University of Wisconsin, Vanderbilt University, and Yale University.This publication has made use of data from the VIKING survey from VISTA at the ESO Paranal Observatory, program ID 179.A-2004.Data processing has been contributed by the VISTA Data Flow System at CASU, Cambridge, and WFAU, Edinburgh.This work is based in part on data obtained as part of the UKIRT Infrared Deep Sky Survey.This publication makes use of data products from the Two Micron All Sky Survey, which is a joint project of the University of Massachusetts and the Infrared Processing and Analysis Center/California Institute of Technology, funded by the National Aeronautics and Space Administration and the National Science Foundation.

Table 2
(Okabe et al. 2019) of 877,642 CAMIRA Member Galaxies at 0 < z cl < 1.4 True (multiple peaks), false (single peak)(Okabe et al. 2019)In this work, a galaxy cluster with a single peak is classified as a relaxed cluster, while a cluster with multiple peaks is considered a merging cluster (Section 4.3)