An X-Ray Census of Active Galactic Nuclei in the Virgo and Fornax Clusters of Galaxies with SRG/eROSITA

We present a uniform and sensitive X-ray census of active galactic nuclei (AGNs) in the two nearest galaxy clusters, Virgo and Fornax, utilizing the newly released X-ray source catalogs from the first all-sky scan of Spectrum-Roentgen-Gamma/eROSITA. A total of 50 and 10 X-ray sources are found positionally coincident with the nuclei of member galaxies in Virgo and Fornax, respectively, down to a 0.2–2.3 keV luminosity of ∼1039 erg s−1 and reaching out to a projected distance well beyond the virial radius of both clusters. The majority of the nuclear X-ray sources are newly identified. There is weak evidence that the nuclear X-ray sources are preferentially found in late-type hosts. Several hosts are dwarf galaxies with a stellar mass below ∼109 M ⊙. We find that contamination by nonnuclear X-ray emission can be neglected in most cases, indicating the dominance of a genuine AGN. In the meantime, no nuclear X-ray source exhibits a luminosity higher than a few times 1041 erg s−1, which might be owing to a steep intrinsic luminosity function. The X-ray AGN occupation rate is only ∼3% in both clusters, apparently much lower than that in field galaxies inferred from previous X-ray studies. Both aspects suggest that the cluster environment effectively suppresses AGN activity. The findings of this census have important implications for the interplay between galaxies and their central massive black holes in cluster environments.


INTRODUCTION
Active galactic nuclei (AGNs), the manifestation of accreting super-massive black holes (SMBHs) at the center of most galaxies (Padovani et al. 2017), play an indispensable and increasingly crucial role in our understanding of the formation and growth of these exotic objects, as well as their interplay and co-evolution with the host galaxies (Fabian 2012;Kormendy & Ho 2013).
From the very beginning (Schmidt 1963), AGN studies, in particular the demography of AGNs, have benefitted from two lines of research: towards greater distances and towards larger samples.The recently launched James Webb Space Telescope with its unprecehoumc@pku.edu.cnhuzhensong@smail.nju.edu.cnlizy@nju.edu.cndented sensitivity in the infrared window is now unveiling some of the first-generation AGNs at the early universe (Goulding et al. 2023).On the other hand, widefield, even all-sky, surveys of AGNs in the low-redshift universe continue to offer new insights and surprises by revealing a vastly large number of AGNs otherwise lurking at the center of normal and dwarf galaxies.Most of these low-redshift AGNs turn out to have low luminosities (Ho 2008), which trace weakly accreting SMBHs fed by a radiatively inefficient, hot accretion flow (Yuan & Narayan 2014).Growing attention has been paid by recent observational and theoretical studies about the importance of such weakly accreting SMBHs in our understanding of the SMBH-host galaxy co-evolution (e.g.Weinberger et al. 2017;Yoon et al. 2019;Shi et al. 2021).Alternatively, a small fraction of these low-luminosity AGNs could be smaller black holes accreting at a rate closer to the Eddington limit.Such dwarf MBHs or intermediate-mass black holes (IMBHs) are the long-sought population that bridges the stellar-mass black holes and SMBHs (Greene et al. 2020).
X-rays, which trace the hot plasma in and around the accretion flow onto the SMBH, provide arguably the most direct evidence for AGNs.This has been demonstrated, in particular, by the Chandra Deep Field surveys of distant AGNs up to a redshift z ≳ 6 (Giacconi et al. 2002;Luo et al. 2017;Xue 2017).The superb sensitivity and angular resolution of Chandra has also enabled routine detections of nuclear X-ray sources down to the Eddington limit of solar-mass objects (∼ 10 38 erg s −1 ) in nearby galaxies (e.g.Gallo et al. 2008;She et al. 2017), enabling the study of specific AGNs as well as AGN demography across diverse galactic environments.
The environment is believed to play a crucial role in galaxy evolution.In particular, in galaxy clusters, where individual galaxies travel at a typical velocity of ∼ 10 3 km s −1 within the hot intra-cluster medium (ICM) and frequently interact with other galaxies, environmental effects such as ram pressure stripping (Boselli et al. 2022) and tidal interaction continue to shape the member galaxies and their interstellar medium (ISM).It is generally expected that the long-term outcome of such effects would be to reduce the amount of the ISM, quench star formation and suppress the SMBH accretion.
Observational evidence for starving SMBHs in the nearest galaxy cluster, Virgo, was obtained by the AGN Multi-wavelength Survey of Early-Type Galaxies in the Virgo Cluster (AMUSE-Virgo; Gallo et al. 2008Gallo et al. , 2010)), which probed nuclear X-ray sources in 100 earlytype galaxies (ETGs; including elliptical, lenticular, and dwarf elliptical/lenticular), finding an occupation fraction of 24% − 34% down to an X-ray luminosity of few times 10 38 erg s −1 .A similar occupation fraction of X-ray AGNs is found for a sample of ETGs in the second nearest cluster, Fornax (Lee et al. 2019), and the third nearest cluster, Antlia (Hu et al. 2023), down to a comparable detection limit.On the other hand, among the AMUSE-Field sample of 103 nearby field and group ETGs, collectively designed as a 'field' comparison for AMUSE-Virgo, Miller et al. (2012) reported a higher Xray AGN occupation fraction of 45% ± 7% and a higher nuclear X-ray luminosity at a given black hole mass than the Virgo ETG sample.Soria et al. (2022) and Graham et al. (2021) have gathered a sample of 75 latetype galaxies (LTGs, including normal disks and dwarf irregulars) in Virgo, finding more than half hosting a nuclear X-ray source, again down to a detection limit of few times 10 38 erg s −1 .
These studies, all based on sensitive Chandra observations, provide an important insight on how the AGN activity may be related to the intrinsic properties of host galaxies continuously being modified by the environmental effects.Nevertheless, the sample galaxies covered by these programs are just a small fraction of the identified member galaxies in the respective clusters, and perhaps more unsatisfactorily, are biased to large galaxies and/or galaxies in the cluster core due to the design of the individual programs.
Thanks to the advent of the SRG/eROSITA all-sky survey (Sunyaev et al. 2021), a uniform and sensitive soft-X-ray census of AGNs in the two nearest galaxy clusters is within reach for the first time.Although the point source sensitivity of eROSITA is no better than that of Chandra due to its moderate point-spread function (with a half-energy width [HEW] ∼ 30 ′′ ; Merloni et al. 2024), a sensitivity limit down to ∼ 10 39 erg s −1 is readily achievable, sufficient to detect low-luminosity AGNs expected to be prevalent in these two clusters.We take up such a task in this work, utilizing the freshly released eROSITA data from its first all-sky scan (eRASS1).
The remainder of this paper is structured as follows.The catalogues of Virgo and Fornax galaxies are described in Section 2. The identification of nuclear X-ray sources from eRASS1 are described in Section 3. Statistical analyses of the nuclear X-ray sources are presented in Section 4, focusing on the X-ray AGN occupation fraction as a function of various galaxy properties.A brief summary and some immediate implications of our findings are addressed in Section 5.

MEMBER GALAXIES OF VIRGO AND FORNAX
To identify X-ray AGNs in the two clusters, we employ the Extended Virgo Cluster Catalog (EVCC; Kim et al. 2014) and the Fornax Cluster Catalog (FCC;Ferguson 1989), which provide the most up-to-date, uniformly classified member galaxies of the respective clusters.
The EVCC contains a total of 1589 spectroscopically identified galaxies, based primarily on the Sloan Digital Sky Survey (SDSS) and supplemented by spectroscopic data from the literature.The footprint of EVCC is more than 5 times that of the photography-based Virgo Cluster Catalog (Binggeli et al. 1985) and reaches out to a projected distance 3.5 times the virial radius of Virgo (R 200 ≈ 1.0 Mpc; Simionescu et al. 2017).As detailed in Kim et al. (2014), within this footprint, nearly 100% of the SDSS photometric galaxies with a r-band magnitude r ≲ 14 mag have a spectroscopic redshift, and the 50% completeness level for galaxies having a radial The size of each symbol is scaled with stellar mass (approximated by BT mag for the Fornax galaxies).
velocity < 3000 km s −1 (i.e., the criterion for a Virgo member) is at r ∼ 16.5 mag.The latter value translates to an absolute magnitude of M r ≈ −14.1 for the nominal distance of Virgo (16.5 Mpc; Mei et al. 2007), roughly corresponding to a stellar mass of ≲ 10 8 M ⊙ .Generally speaking, few (S)MBHs are expected to exist in galaxies with a similar stellar mass or lower.Therefore, the EVCC should contain the majority, if not all, of member galaxies within which an X-ray AGN could be present.
The FCC, on the other hand, is a photography-based catalog containing 340 likely members in a footprint of 6 • × 6 • (2.0 Mpc × 2.0 Mpc; at the nominal distance of 20.0 Mpc of Fornax; Blakeslee et al. 2009), reaching out to ∼ 1.5 times the virial radius (R 200 ≈ 0.7 Mpc; Drinkwater et al. 2001).The FCC has a completeness limit of B T ∼ 18 mag (an absolute magnitude of M BT ∼ −13.5), which corresponds to a stellar mass ∼ 10 8 M ⊙ .This means that the FCC should also include all member galaxies likely to host an AGN, with the caveat that the membership of most galaxies was not on a spectroscopic basis.

IDENTIFICATION OF NUCLEAR X-RAY SOURCES
To identify a putative X-ray AGN, we cross-match the galactic nuclei, whose sky coordinates are provided by the EVCC and FCC, with the newly released allsky survey X-ray source catalogs from eRASS1 (Merloni et al. 2024), which include a soft-band (0.2-2.3 keV) cat-alog and a hard-band (2.3-5 keV) catalog.We adopt a matching radius of 10 ′′ , corresponding to the 99 percentile of the position error of the soft-band sources, which spans a linear scale of ∼0.8 (∼1.0 kpc) at the distance of Virgo (Fornax).
We estimate the number of potential random matches by artificially shifting the centroid of all X-ray sources within the EVCC footprint by ±3 ′ in either R.A. or Decl., an amount sufficiently large compared to the default matching radius but also sufficiently small compared to the angular range of the member galaxies.This exercise results in a random match of 1.0 sources aver- aged over the four directions, which may account for only 2% of the actual matches.
In the case of Fornax, originally 11 nuclear Xray sources were found.After consulting with the NASA/IPAC Extragalactic Database (NED)1 , we find that one galaxy (FCC 129) has a large redshift that disputes its Fornax membership, while the other 10 galaxies have a redshift compatible with Fornax.Therefore, we retain a total of 10 nuclear sources (Figure 1), which are all detected in the soft band, with L 0.2−2.3ranging between 9.6 × 10 38 erg s −1 − 2.1 × 10 41 erg s −1 .We note that the eRASS1 source detection completeness limit is ∼ 1.2 × 10 39 erg s −1 within the footprint of FCC (Figure 2).Only one galaxy (FCC 121 = NGC 1365) is detected in the hard band with a 2.3-5 keV luminosity of 6.6 × 10 41 erg s −1 .We find 0.5 random matches for the Fornax galaxies.
Table 1 lists the host galaxy name, centroid coordinates, X-ray luminosity of the matched eRASS1 source, and stellar mass.The stellar mass of the Virgo galaxies is calculated based on the SDSS g-and r-band photometry given by the EVCC and the color-magnitude relation of Bell et al. (2003).Since only a single band photometry is provided by the FCC, we instead employ the WISE (Wright et al. 2010) W1 and W2 band images to calcu-late the stellar mass, following the calibration of Jarrett et al. (2019).We have adopted the nominal distances of Virgo and Fornax when deriving the X-ray luminosity and stellar mass.
Comparison with Gallo et al. (2010).The AMUSE-Virgo program detected 32 nuclear X-ray sources out of 100 ETGs in Virgo down to a limiting luminosity of 3.7 × 10 38 erg s −1 over 0.5-7 keV (corresponds to 3.2 × 10 38 erg s −1 in the 0.2-2.3keV by assuming an absorbed power-law model with the photon index of 2 and column density N H = 2.5 × 10 20 cm −2 adopted in Gallo et al. 2010).We find 3 eRASS1 detections among these 32 targets (EVCC 353, 681, 2211), while among the 68 non-detected ETGs, two (EVCC 626, 884) have an eRASS1 match.The non-detection of the remaining targets can be mainly attributed to a lack of sensitivity, because most of them had an X-ray luminosity below 10 39 erg s −1 according to Gallo et al. (2010).Particularly worth mentioning is M 87, the nuclear source of which has an X-ray luminosity ∼ 10 41 erg s −1 but is still undetected by eRASS1.This is likely due to the heavy contamination by the surrounding diffuse hot gas in this brightest cluster galaxy.
Comparison with Soria et al. (2022).Soria et al. (2022) reported that a total of 39 nuclear X-ray sources are present among the 75 LTGs in Virgo with a typical detection limit of ∼ 3 × 10 38 erg s −1 and a completeness limit of ∼ 10 39 erg s −1 in the 0.3-10 keV band, but with-out quoting the exact host galaxies.As a byproduct of our recent study of diffuse hot gas in these LTGs (Hou et al. 2024), we have independently detected 35 nuclear X-ray sources.Among them, 13 share the same host galaxy as the eRASS1 nuclear sources.One additional galaxy, EVCC 595, has an eRASS1 detection.
Comparison with Lee et al. (2019).Lee et al. ( 2019) detected a total of 11 out of 29 ETGs in Fornax down to a limiting luminosity of 5×10 38 erg s −1 over 0.3-10 keV, all having an X-ray luminosity below ∼ 2 × 10 39 erg s −1 .We find only one eRASS1 match (with FCC 153) among these 11 targets and one additional eRASS1 match (with FCC 147) among the 18 non-detected galaxies.is too faint to be reliably measured in the WISE images, hence its stellar mass a rough estimate from its B T mag.

Potential contamination by host galaxy
Due to the moderate resolution of eROSITA, the detected nuclear X-ray sources might be contaminated by non-nuclear X-ray emission of the host galaxy.In particular, low mass X-ray binaries (LMXBs), the amount and total X-ray luminosity of which scale with the stellar mass, is expected to produce a 0.2-2.3keV luminosity of L LMXB ≈ 7.9 × 10 38 erg s −1 (M * /10 10 M ⊙ ), where M * is the stellar mass enclosed within the eRASS1 photometry aperture (∼ 30 ′′ ).The above scaling relation is derived from the empirical L(2-10 keV)-M * relation of Lehmer et al. (2010), and we have assumed for LMXBs an intrinsic power-law spectrum with a canonical photon-index of 1.7.Based on the EVCC, we find that the majority of matched galaxies have an optical size (r Kron ) above 30 ′′ (i.e., well resolved by eROSITA); only a few galaxies having the lowest stellar masses (≲ 10 9 M ⊙ ) are entirely covered by the eRASS1 source aperture.Therefore, we conclude that statistically LMXBs have only a small contribution to the detected nuclear X-ray emission.
A small fraction of the matched galaxies could be actively forming stars, which would also contribute to the observed X-ray emission in the form of high-mass X-ray binaries and/or diffuse hot gas.The 0.2-2.3keV luminosity of this star formation-related component is estimated as (Lehmer et al. 2016): L SF ≈ 5.6 × 10 39 erg s −1 (SFR/M ⊙ yr −1 ), for which a factor of 1.4 is multiplied to the original scaling relation to roughly account for a different photon energy range (0.5-2 keV).Since no explicit information about the star formation rate (SFR) is provided by the EVCC, we employ the WISE W3-band image to estimate the SFR, following Jarrett et al. (2019).We find that all but two of the 50 host galaxies in Virgo have an SFR within the eRASS1 source aperture too low to have a significant contribution to the observed nuclear X-ray emission.The two exceptions are EVCC 359 and EVCC 892 which have a central SFR ∼ 1.7 M ⊙ yr −1 and ∼ 0.4 M ⊙ yr −1 , sufficient to explain the observed nuclear X-ray emission.Therefore, we conclude that statistically star-forming activities also have only a minor contribution to the detected nuclear X-ray emission.
By evaluating the stellar mass and SFR of the 10 Fornax host galaxies, we find that host galaxy contamination can be neglected in all but three galaxies.FCC 22, FCC 29 and FCC 62 have a substantial SFR to account for the observed X-ray emission.Moreover, it is still possible that a small fraction of the eRASS1 nuclear sources are X-ray binaries rather than a genuine AGN, because the detection limit of eRASS1 is still compatible with the Eddington limit of stellar-mass black holes.We note that the same caveat holds for the aforementioned Chandra programs, which all have an even lower detection limit.

Host galaxies and occupation fraction
Following the convention, we collectively assign ellipticals, lenticulars and dwarf ellipticals/lenticulars as ETGs, and the remaining morphological types as LTGs.
For Virgo, when the EVCC morphological type is 'edgeon', we consult with NED to determine the morphological type if available; for those without an explicit morphological type, we take them as ETGs, introducing only a ∼ 2.5% ambiguity.The ETG fraction of Virgo is thus ∼49%.A much higher fraction of ∼82% is found in Fornax according to the FCC, with the caveat that the FCC classification was based on old imaging data probably less sensitive than SDSS.It turns out that 17 of the 50 Virgo hosts are ETGs, while 4 of the 10 Fornax hosts are ETGs, indicating a similar percentage (34% ± 10% vs. 40% ± 24%) in the two clusters.Quoted errors are of Poisson and at 1σ confidence level.
The percentage of X-ray detected nuclei, or the occupation rate of an X-ray AGN, is also quite similar between Virgo and Fornax (50/1589 ≈ 3.1% ± 0.4% vs. 10/340 ≈ 2.9% ± 0.9%).The occupation rate becomes ≈ 2.6% ± 0.4% and ≈ 2.6% ± 0.9%, if we drop 9 sources in Virgo and 1 source in Fornax, whose X-ray luminosities are below the detection completeness limit (Section 3).If instead we restrict on host galaxies with log(M * /M ⊙ ) > 7.8, which is the lowest mass among the 50 Virgo hosts and also is roughly the completeness limit of the EVCC, the percentage becomes 4.2 ± 0.6% (50/1203).
We further evaluate the occupation rate by dividing the Virgo galaxies into subgroups, according to the stellar mass, the projected distance from the center of M 87, and the morphological type (ETG or LTG), respectively.Due to the small number of X-ray nuclei in Fornax, we do not perform a similar analysis.
As expected, while the majority of EVCC galaxies are actually dwarfs, only four X-ray nuclei are detected among galaxies with 7.8 < log(M * /M ⊙ ) < 8.65 (4/602 ≈ 0.7±0.3%),compared to 46 in the more massive group (46/601 ≈ 7.7 ± 1.1%).On the other hand, a comparable occupation rate (3.6 ± 0.7% vs. 2.6 ± 0.6%) is found between galaxies located inside and outside ∼ 1.7R 200 of Virgo.Intuitively, those galaxies on the first infall to the cluster core may still contain a significant amount of ISM to fuel the central SMBH, but this effect could be offset by the on-average heavier SMBHs in galaxies already in or passing the cluster core due to frequent galaxy merger therein.A full spectroscopic survey of the nuclei of the EVCC galaxies would be desired to provide a robust and uniform measurement of black hole mass and hence the Eddington ratio.Finally, a marginally significant (∼ 2σ) difference in the occupation rate is seen between the ETGs and LTGs (2.2 ± 0.5% vs. 4.0 ± 0.7%).This is consistent with the naive expectation that LTGs contain more available fuel for the SMBH.

SUMMARY AND DISCUSSION
In this work, we have utilized the newly released X-ray source catalogs from the first all-sky scan of eROSITA to probe putative X-ray AGNs in the Virgo and Fornax clusters, taking advantage of its full coverage of all known member galaxies.
A total of 50 and 10 X-ray sources are found to be positionally coincident with the nuclei of member galaxies in Virgo and Fornax, respectively.A large fraction of these nuclear X-ray sources are identified for the first time.Despite the moderate resolution of eROSITA, we find that in most cases host galaxy contamination is negligible to the observed X-ray luminosity, indicating the dominance of a genuine X-ray AGN.The majority of these nuclear X-ray sources are found in late-type hosts.
The overall occupation rate of the eRASS1 nuclear sources is remarkably low, ∼ 3% in both clusters.This appears to be much lower than found in nearby galaxies not residing in clusters.For instance, at least 13% of the AMUSE-Field ETGs have an X-ray luminosity ≳ 2 × 10 39 erg s −1 (Miller et al. 2012), i.e., above the eRASS1 completeness limit.In a Chandra survey of Xray AGNs in 719 galaxies within a distance of 50 Mpc (most are outside Virgo and Fornax), She et al. (2017) found 314 X-ray nuclei, among which ∼ 50% have an X-ray luminosity above the eRASS1 completeness limit.Although not free of selection effect, these two studies imply that the occupation rate of an X-ray AGN in field galaxies is likely higher than 10%.Hence, the relatively low occupation rate in Virgo and Fornax strongly suggests that AGN activity is generally suppressed in the cluster environment.
A related and equally remarkable feature about the eRASS1 nuclear X-ray sources is the paucity of luminous objects.Indeed the highest 0.2-2.3keV luminosity found is only ∼ 3 × 10 41 erg s −1 , which would correspond to ∼1% Eddington even for a black hole mass as low as 10 5 M ⊙ .For comparison, the maximum luminosity of the X-ray nuclei detected by She et al. (2017) well exceeds this value, regardless of the host galaxy morphological type.A potential issue here is that the Virgo and Fornax nuclear X-ray sources are essentially detected in soft X-rays, which might be subject to circumnuclear absorption.However, it is unlikely that a large number of obscured AGNs are present in present-day clusters, in which gas-rich galaxies are not a significant population.The few detections of nuclear sources in the eRASS1 2.3-5 keV catalog supports this notion.Alternatively, as found by Birchall et al. (2020Birchall et al. ( , 2022)), the steep X-ray AGN luminosity function naturally implies a rarity of luminous objects in volume-limited samples.Regardless, the first complete census of AGNs in Virgo and Fornax in the soft X-rays would provide useful constraints for cosmological simulations of galaxy cluster formation (Nelson et al. 2023).
A small but non-negligible fraction of the eRASS1 nuclear X-ray sources are associated with dwarf galaxies, which may host a dwarf MBH.High-resolution Chandra observations are desired to determine whether these sources are genuine AGNs.Moveover, some of the eRASS1 nuclear X-ray sources have been previously detected or undetected by Chandra observations (Section 3).The subsequent scans of eROSITA would shed light on the flux variability of these sources.

Figure 1 .
Figure1.The sky distribution of member galaxies in Virgo (left) and Fornax (right).Member galaxies in the Extended Virgo Cluster Catalog and Fornax Cluster Catalog are indicated by blue symbols.The nuclear X-ray sources detected by eRASS1 are marked by red symbols color-coded by the 0.2-2.3keV luminosity.ETGs and LTGs are represented by circles and triangles.The size of each symbol is scaled with stellar mass (approximated by BT mag for the Fornax galaxies).

Figure 2 .
Figure 2. X-ray flux and luminosity distributions of all the eRASS1 point sources in the Virgo (left)and Fornax (right) footprint.Soft band and hard band sources are shown in blue and orange histograms.The peak of the distributions indicates the detection completeness limit.
Galaxy Name as in the EVCC and FCC catalogs.(2) NGC name of the galaxy.(3)-(4) Right ascension and declination at equinox J2000 of the galactic nuclei.(5) The morphological type of the galaxy.(6) X-ray luminosity and 1σ error in the 0.2-2.3keV energy band.* Non-detected.(7) Stellar mass of the galaxy.For Virgo galaxies, this is derived from the g − r massluminosity relation, with a typical uncertainty of ∼ 0.1 dex.For Fornax galaxies, this is derived from WISE W1-and W2-band images, with a typical uncertainty of ∼ 0.1 − 0.2 dex.FCC 222 has a larger uncertainty of 1 dex due to its low luminosity.FCC 57

Table 1 .
Nuclear X-ray sources and host galaxy properties