The WISSH QSOs project
Key Words.:galaxies: active – galaxies: nuclei – quasars: emission lines –quasars: general – quasars: supermassive black holes – techniques: imaging spectroscopy
We present an ALMA high-resolution () observation of the 840 m continuum and [CII] m line emission in the WISE-SDSS selected hyper-luminous (WISSH) QSO J10150020, at . Our analysis reveals an exceptional overdensity of [CII]-emitting companions with a very small ( km s) velocity shift with respect to the QSO redshift. We report the discovery of the closest companion observed so far in sub-mm observations of high-z QSOs, distant only 2.2 kpc and merging with J10150020, while two other [CII] emitters are found at 8 and 17 kpc from J10150020. Two strong continuum emitters are also detected at arcsec from the QSO. They are likely associated to the same overdense structure of J10150020, as they exceed by a factor of 100 the number of expected sources, considering the Log(N)Log(S) at 850 m. The host galaxy of J10150020 shows a SFR of about 100 M yr, while the total SFR of the QSO and its companion galaxies is a factor of higher, indicating that substantial stellar mass assembly at early epochs may have taken place in the QSO satellites. For J10150020 we compute a SMBH mass M. As we resolve the [CII] emission of the QSO, we can compute a dynamical mass of M. This translates into an extreme ratio /, i.e. two orders of magnitude smaller than what typically observed in local galaxies. The total stellar mass of the QSO host galaxy plus the [CII] emitters in the ALMA field of view already exceeds M at . These sources will likely merge and develop into a giant galaxy of M. By comparing the black hole accretion rate and the SFR, and assuming that local relations between the mass outflow rate and the QSO bolometric luminosity hold at the observed redshift, we conclude that J10150020 is potentially able to drive a massive molecular outflow affecting the SFR in the QSO host galaxy, similarly to what observed for a sample of 47 QSOs at . Under the assumption of constant or equal to the observed ones, we find that the growth timescale of the host galaxy of J10150020 is comparable or even shorter than the one inferred for the SMBH.
The most popular models of AGN-galaxy co-evolution (Di Matteo et al., 2005; Menci et al., 2008; Hopkins et al., 2008) include galaxy interactions (both major and minor mergers) and AGN-driven feedback (i.e. the injection of energy and entropy in the interstellar medium trough winds and shocks) among the majors processes driving this phenomenon. These two processes are highly correlated, e.g. galaxy interactions may destabilize the gas and make it available for fuelling both star formation (SF) and nuclear accretion, giving rise to the growth of the central SMBH through luminous AGN phases. The AGN fraction increases in IR-luminous, star forming sources (Nardini & Risaliti, 2011; Rosario et al., 2013), and a correlation between the AGN luminosity () and the SF luminosity is observed for a wide range of (Netzer, 2009; Lutz et al., 2010). In turn, the AGN can power winds (Fiore et al., 2017) hampering further SF and nuclear gas accretion (as well as AGN-driven winds).
A linked growth of the SMBH and its host galaxy is observationally supported by the well-known local scaling relations between the SMBH mass () and the physical properties of the host galaxy bulge (e.g. see Kormendy & Ho, 2013), such as the dynamical mass () or the velocity dispersion. Theoretically, such relations can be shaped by merger events (see Alexander & Hickox, 2012, and references therein), triggering at the same time bursts of nuclear and SF activity (Volonteri et al., 2015; Gabor et al., 2016; Anglés-Alcázar et al., 2017b). The - relation also indicates that the assembly of the giant galaxies can be probed by observing the QSOs with the most massive SMBHs shining at . This allows to get insights into key evolutionary phases not observable in the local universe and investigate the hotly-debated role of mergers and AGN feedback. This field of research has been revolutionized by ALMA, thanks to its unprecedented sensitivity and broad-band coverage as, the QSO emission typically outshines the host galaxy at all wavelengths below few tens of m. The most powerful observational tool to study the high-z QSOs host galaxies is the [CII] fine structure line at 157.75 m. [CII] is in fact the strongest line from the cool gas ( K) and, given its low ionisation potential of 11.3 eV, traces both the neutral and ionised medium. It is also a tracer of SF activity (Maiolino et al., 2005; Carniani et al., 2013). Trakhtenbrot et al. (2017) found a wide variety of host galaxy properties of hyper-luminous QSO at z5 in terms of possible SMBH fuelling mechanisms and SF activity, suggesting that galaxyâgalaxy interactions may not be a necessary condition for either of the two processes.
[CII] has been investigated even in the most distant QSOs. These objects typically reside in compact hosts where rotating disks are already in place and intense SF activity of tens to thousands of solar masses is ongoing (e.g. Wang et al., 2013, 2016; Cicone et al., 2015; Díaz-Santos et al., 2016; Venemans et al., 2016, 2017; Willott et al., 2017; Decarli et al., 2018). Recent studies of QSOs reveal that they are powered by SMBH at the massive end of the black hole mass function (Jiang et al., 2007; De Rosa et al., 2011, 2014; Venemans et al., 2015; Bañados et al., 2016) and that their hosts are among the brightest and most massive galaxies at these redshifts. According to local relations (e.g. Jiang et al., 2011), these sources are therefore expected to assemble stellar masses typical of giant galaxies at . In the - plane, most of high-z QSOs lie above the local relation, as they are characterised by very low stellar-to-BH mass ratios as low as 10. However, the number of sources with available estimates from single epoch relations is very limited to date, while the bulk of the is still derived from the QSO under the assumption of accretion at the Eddington limit. Furthermore, the majority of the high-z QSOs is still unresolved or only marginally resolved, thus affecting size and dynamical mass estimates with large uncertainty. Accordingly, high-resolution studies are of primary importance.
It is therefore crucial to study the SMBH and host galaxy growth at early epochs, i.e. when both processes are maximised. Accordingly, we have undertaken the WISE-SDSS selected hyper-luminous (WISSH) QSOs project to study the most powerful AGN in the Universe, which happen to shine at these cosmic epochs (Bischetti et al., 2017). Similarly to the QSOs studied so far, WISSH have erg s and are powered by accretion onto SMBH of M at rates close (or even higher than) the Eddington limit. Such huge luminosities at Eddington regimes are likely triggered by galaxy interactions (Menci et al., 2014; Valiante et al., 2014, 2016), and will drive powerful winds that may affect the whole host galaxy. Therefore, these QSOs are ideal targets to shed light on the AGN-galaxy feeding & feedback cycle. To this purpose, from multi-wavelength spectroscopy and photometry, we collected information about the AGN power and the multi-transition presence of nuclear and galaxy-scale winds (Bischetti et al., 2017; Vietri et al., 2018). We also built up the FIR-to-UV spectral energy distributions (SED) of 14 WISSH QSOs with Herschel photometry (Duras et al., 2017) to derive the SFR in their host galaxy.
From this WISSH-Herschel subsample, we selected the QSO J10150020 at for a pilot ALMA observing program aimed to characterise the host galaxy and environment properties of hyper-luminous QSOs. The high redshift and low declination guaranteed this target to well fit the ALMA band 7 and be observed with good sensitivity. Specifically, we present here the results from an high-resolution () ALMA observation of the 840 m and [CII] m line emission in J10150020.
Throughout this paper, we assumed a cosmology with = 67.3 km s Mpc, = 0.69 and = 0.31 (Planck Collaboration et al., 2016).
2 ALMA observation and data analysis
In this work we present the ALMA Cycle 3 observation (project 2016.1.00718.S) of the WISSH quasar SDSS J10150020 (celestial coordinates RA 10:15:49.00, Dec +00:20:20.03). The observation was carried out on 5 March 2016 for a 0.6 hours on-source integration time in C36-5 configuration, with a maximum projected baseline of 1396 m. We used the ALMA band 7 receiver and the frequency division mode of the ALMA correlator. This provided us with four spectral windows of 1.875 GHz width, for a total spectral coverage of 7.5 GHz, with a spectral resolution of 31.25 MHz. A first spectral window was centred at 352 GHz to cover the expected observed frequency of the [CII] emission line, given the SDSS DR10 redshift (Pâris et al., 2014). A second window was put adjacent to it on the lower frequency side, in case of a blueshift of the rest-frame UV lines with respect to the systemic emission of the source, i.e. the [CII] emission. The remaining two spectral windows were placed at higher frequencies to account for the continuum emission.
Visibilities were calibrated with the CASA 4.7.0 software (McMullin et al., 2007) in pipeline mode, by applying the default phase, bandpass and flux calibrators provided. Images were produced by using the CASA task clean with natural weights, a 0.03 arcsec pixel size and a 30 km s channel width, and a final beam size of . The ALMA field of view (FOV) of our observation, defined as the region in which the relative sensitivity is higher than 0.5, is a circular area with a radius of . The continuum map was obtained by averaging over all the spectral windows and excluding 1.2 GHz around the [CII] emission. The continuum emission in the spectral region of the [CII] line was first modelled by fitting the UV plane of the first two spectral windows by a first degree polynomial and then subtracted to the visibilities.
For all the sources detected with a signal-to-noise ratio higher than five in the ALMA FOV, continuum and [CII] flux densities were measured by fitting a 2D Gaussian model to the final map in the image plane (see Sect. 3.1). Furthermore, from spectral fitting of the [CII] emission we derived the parameters of the [CII] line profile (Sect. 3.2). For each source, we applied a spectral model including one or two Gaussian emission line components, based on the profile of the [CII] line.
3 Results from the ALMA observation
3.1 Continuum and [CII] emission maps
The WISSH quasar J10150020 is detected both in continuum and [CII] line emission. Furthermore, continuum and [CII] maps reveal multiple sources with an angular separation of less than 4 arcsec from the central QSO.
|(2) [Jy]||59564||125276||38861||¡ 120||¡ 120||¡ 125|
|(5) [Jy km s]||0.470.05||¡ 0.15||¡ 0.15||0.19||0.350.06||1.130.16|
|(8) FWHM [km s]||33938||6013||48045||40942|
|(9) [10 L]||2.90.3||¡ 0.92||¡ 0.92||1.20.1||2.20.4||7.01.0|
|(10) Angular sep ||3.5||3.5||0.33||1.2||2.4|
Fig. 1 gives an insight into these crowded surroundings. Specifically, top panels show the ALMA continuum map at 840 m, with a rms sensitivity of 0.04 mJy beam, revealing cold dust emission from the host galaxy of J10150020 and two additional sources both at a distance of arcsec (Cont1 and Cont2 hereafter). The continuum emission from the QSO is detected at significance with a flux density of Jy. It is resolved by the ALMA beam, with a deconvolved size of arcsec (see Table 1). Cont1 and Cont2 are detected at and significance with a flux density and Jy, respectively. They are also resolved, see Table 1. Cont1 and Cont2 are not detected in [CII] line emission and, therefore, we are not able to derive a spectral redshift for these sources.
The ALMA [CII] map also reveals multiple sources around J10150020. Fig. 1 (bottom middle panel) shows the presence at a significance, of a clearly distinct companion galaxy with an angular separation of only 0.33 (CompA hereafter), corresponding to a proper distance of 2.2 kpc at the QSO redshift.
Moving to larger distance from the QSO, two additional [CII] emitters are detected at 1.2(8.2 kpc) and 2.3(16 kpc) with a significance of and , respectively. The angular separations from J10150020 are 1.2 and 2.3, corresponding to 8.2 kpc and 16 kpc, respectively. Hereafter, these sources will be referred as CompB and CompC. All three [CII] emitters are not detected in continuum emission. The [CII] line emission of the QSO is detected at significance and has an integrated flux density of Jy km s, derived from 2D-Gaussian fitting. CompA, CompB, and CompC have a [CII] flux density of 0.19, 0.35 and 1.13 Jy km s, respectively. The [CII] emission associated with J10150020 and CompA is marginally resolved by the ALMA beam (see Table 1). CompB is unresolved, while CompC has an elongated morphology which clearly rules out a point source nature.
3.2 Spectra and velocity maps
We extracted the [CII] spectrum of J10150020, CompA, CompB and CompC from the ALMA cube in regions of 1.5 to 3 beam areas, according to the size of the source. The resulting continuum subtracted spectra are shown in Fig. 2. The [CII] line profile of J10150020 is well reproduced by a single Gaussian component centred at 351.5 GHz, corresponding to a [CII]-based redshift . This translates into a small velocity shift of about 400 km s between the [CII] emission and .
All the [CII] line-detected sources in the ALMA map have a very small velocity shift of ( km s) with respect to and, therefore, we consider them as companions of J10150020. For each source, we calculate the velocity shift as the difference between and the velocity that bisects the cumulative [CII] line flux. CompA, which is the closest companion, has an extremely narrow [CII] line profile with a FWHM km s centred at the same redshift of the QSO. CompB and CompC are instead characterised by a boxy-double peaked [CII] line, with FWHM km sand FWHM km s, respectively. These profiles may indicate multiple kinetic components or rotation. The velocity shift of CompB is also consistent with , while CompC shows a [CII] line characterised by a blueshift of 14530 km s.
Fig. 3 shows the velocity and velocity dispersion maps of J10150020 (top panel) and CompC (lower panel). As for the QSO, a small velocity gradient seems to be present in the north-south direction, from about to km s, while the [CII] line of CompA on the QSO left is too narrow to show a gradient at our spectral resolution and appears as a monochromatic spot. The velocity dispersion map of J10150020 clearly shows an increase in the south direction, with a peak dispersion of km swhich is offset with respect to the peak of the [CII] emission, indicated by the black cross in Fig. 3. An increased central velocity dispersion is usually found at the AGN location (e.g. Trakhtenbrot et al., 2017), while we likely observe a perturbation of the gas. This can be interpreted in terms of (i) the very close presence of CompA merging with the QSO, (ii) an additional ongoing, disk scale merger, (iii) a possible [CII] outflow component. Deeper observations with higher spatial resolution are needed to draw firm conclusions on the nature of this feature.
Two blobs with positive and negative velocity with respect to the QSO redshift are present in the velocity map of CompC. The velocity gradient is much larger, ranging from about km s to km s. We may interpret this as a rotating disk seen at high inclination (see also Sect. 3.1) with a diameter of kpc. Another possibility is that CompC consists of two interacting sources, characterised by slightly blueshifted and redshifted [CII] emission with respect to the of the QSO. This is in agreement with the velocity dispersion map, in which the red blob appears more perturbed than the blue one. However, the quality of the data does not allow us to clearly discriminate between these hypotheses.
4 Overdensity around J10150020
4.1 Multiple companions
The ALMA observation has shed light on the crowded surroundings of J10150020. Three sources, in addition to the QSO, are detected in [CII] line emission at an angular separation of (see Sect. 3.1), corresponding to a proper distance of kpc at the observed redshift. This high number of sources is quite surprising within the context of sub-mm observations of high-z QSOs, where a single companion at larger separation is usually observed, e.g. see Trakhtenbrot et al. (2017). Among the [CII] emitters in our high-resolution observation, we report the discovery of the closest companion observed so far around a high-z QSO. CompA is indeed located at , corresponding to only kpc from J10150020.
Furthermore, we detect two additional continuum emitting sources separated from J10150020, with comparable or even higher continuum flux density with respect to the hyper-luminous QSO. Cont1 and Cont2 lack [CII] line emission in the spectral band covered by our ALMA observation. The non-detection of [CII] emission from these two sources implies a comoving distance along the line of sight larger than 18.5 Mpc foreground and 25 Mpc background the QSO. In order to understand whether they are related to the QSO, we compute the expected number of field sources of any redshift within a region of around the central QSO. By using the Log(N)Log(S) at 850 m derived by Simpson et al. (2015), we should expect 0.02 field sources with a flux density mJy, i.e. the value observed for Cont2, the faintest continuum source in the ALMA map. The expected number of sources in the same region, with a flux comparable or larger than that of Cont1 is even smaller, i.e. . Indeed, assuming a Poissonian distribution with average number of successes equal to the expected number counts, the probability of having two detections is . Given the limited sky coverage (about one square degree) of the Simpson et al. (2015) survey, the observed counts may not be representative of the whole sky. However, as the observed number of continuum sources around the QSO is a factor of 100 larger than the expected value, we conclude that Cont1 and Cont2 are likely associated to the same overdensity traced by the [CII] emitters around J10150020. Recent deep sub-mm surveys carried out with ALMA (Carniani et al., 2015; Fujimoto et al., 2016; Aravena et al., 2016b) typically observed sources per ALMA pointing with similar flux to Cont2.
We do not detect any counterpart of the five companions in the SDSS, HST/ACS, UKIDSS and WISE images in correspondence of their coordinates. Stark et al. (2009) and Bouwens et al. (2015) measured typical densities of 0.01 galaxies with SFR100 M yr per ALMA band-7 pointing. Concerning number counts of [CII]-emitting galaxies at , Aravena et al. (2016a) predict about 0.06 galaxies per pointing. In conclusion, it is very unlikely that the presence of three [CII]- and two continuum-emitting companions in the field surrounding J10150020 is due to chance superposition of unrelated galaxies close to the line of sight. Accordingly, we are likely observing a very significant overdensity of star-forming galaxies around a powerful and massive QSO.
4.2 Star formation in and around the QSO host galaxy
The ALMA observation allows us to derive the 840 m continuum flux of J10150020 and, therefore, extend the coverage of the SED presented in Duras et al. (2017) to larger wavelengths. Furthermore, ALMA has highlighted the presence of two continuum emitters (Cont1 and Cont2) at an angular separation from J10150020 (see Sect. 3.1), with a 840 m continuum flux density comparable to the QSO one. These sources cannot be resolved as distinct objects in the Herschel images, given the SPIRE point spread function ranging from at 250 m to at 500 m. We therefore perform a new SED fit of J10150020 to remove the contamination of the continuum emitters and more accurately estimate the FIR luminosity () of the QSO. We add the ALMA data to the SDSS DR10 (Pâris et al., 2014), WISE (Wright et al., 2010) and Herschel/SPIRE (Pilbratt et al., 2010; Griffin et al., 2010) photometry, presented in Duras et al. (2017). We also include three additional near-IR photometric points from the UKIDSS large area survey (Warren et al., 2007). In order to quantify the possible contribution of Cont1 and Cont2 to the SPIRE fluxes, we build the following SED:
SED A, which includes photometric points from SDSS to ALMA 840 m. The latter takes into account the continuum emission from the QSO, Cont1 and Cont2 (Sect. 3.1).
SED B, which provides an estimate of the emission from the QSO by removing contamination from Cont1 and Cont2. Specifically, (i) we verify a negligible contribution from the two continuum emitters to the photometric points at wavelength m; (ii) we rescale the SPIRE fluxes by a factor (the latter represents the ratio between and the ALMA-based flux of the QSO listed in Table 1) and (iii) we consider as the flux density for the photometric point at 840 m.
The rest-frame SEDs A and B are shown in Fig. 4. Both are well modelled by the sum of an accretion disk plus torus emission component and a cold dust component in the far-IR. However, the SPIRE photometric points show an excess with respect to the best fit model in SED A. This feature is significantly reduced in SED B, indicating that the main contributors to this excess have been removed. Some residual contamination may still be present in particular at 500 m, because of continuum emitters accounted in the SPIRE fluxes lying outside of the ALMA FOV, i.e. at an angular separation larger than 8.5 which translates to a distance kpc at the QSO redshift. As expected, SED A provides a larger estimate of the FIR luminosity (Log(/erg s) = 46.33) than SED B (Log(/erg s) = 45.69). Accordingly, the SFR derived by following Kennicutt & Evans (2012) is reduced by a factor of 4, i.e. from about 940 to M yr. The difference with the SFR derived in Duras et al. (2017) is even higher, i.e. a factor of 6. This highlights the importance of sampling the FIR and sub-mm bands with high angular resolution in case of high and high- sources, for which a significantly enhanced merger rate is expected (Treister et al., 2012). Similar results have been indeed reported by Banerji et al. (2017) for heavily-reddened QSOs, pointing out the need of ALMA observations to uncover these structures around luminous QSOs.
We measure a bolometric luminosity erg s (the uncertainty is dominated by that on the QSO inclination). We therefore correct the SFR derived from SED B according to Duras et al. (2017), who found that in hyper-luminous QSOs with Log(/erg s) the AGN contributes to about 50% of the total FIR luminosity. Accordingly, the resulting SFR of J10150020 is M yr. Throughout the paper, we use quantities derived by SED B as representative of the physical properties of J10150020 and its host galaxy.
Our ALMA observation reveals that a significant fraction of the 840 m flux is not associated with the host galaxy of J10150020, but instead arises from the surrounding continuum emitters, that likely belong to the QSO overdensity (Sect. 4). We roughly characterise the spectral shape of Cont1, which is the strongest continuum emitting source in our ALMA map (see Fig. 1). Cont1 is detected only by ALMA, but SDSS to WISE non detections can be used to derive upper limits on its flux in all these bands. Specifically, we fit the star forming M82 and the starburst Arp220 galaxy templates (Polletta et al., 2007) to the Cont1 photometry, requiring them to match the ALMA point and leaving the redshift free to vary from to 5. A M82-like template does not match with the upper-limits at any redshift in this range. A 10 M yr star-forming galaxy would be in fact visible in the UKIDSS and WISE bands. On the contrary, a more extreme starburst, i.e. an Arp220-like template, is compatible with the SED of Cont1. This suggests that Cont1 is undergoing intense star formation activity of hundreds of M yr, in agreement with the FIR excess (corresponding to M yr) observed in SED A associated to the Herschel photometry. Such value is derived under the assumption that Cont1 and Cont2 sources belong to the same structure of J10150020, which is supported by the very unlikely possibility of having two field sources with such a close angular separation (see Sect. 4).
We can estimate the star-formation activity of the [CII] emitters revealed around J10150020 by using the relation from Sargsyan et al. (2014):
which gives SFR within a 50% uncertainty for low redshift star forming galaxies. For CompA, CompB and CompC we derived [CII] luminosities of L, L and L, respectively (see Table 1). According to Eq. 1, these values translate in SFR M yr, SFR M yr and SFR M yr. We note that, in case of J10150020, Eq. 1 would correspond to a SFR of only 30 M yr, indicating a discrepancy with the -based value. This is likely due to the huge radiative power of the AGN in hyper-luminous sources, whose ionizing effect reduces the [CII] emission.
By using the [CII] flux derived in Sect. 3.1 and Eq. (1) in Solomon & Vanden Bout (2005), we compute the [CII] luminosity of J10150020, L. This implies a Log(/) = 3.64, which is among the lowest values found for high-z QSO so far. This is shown in Fig. 5, where J10150020 (magenta star) is compared to a sample of QSOs from literature. Specifically, we collect and for 42 sources from the works of Wang et al. (2013, 2016); Venemans et al. (2016, 2017); Willott et al. (2013, 2015, 2017); Kimball et al. (2015); Díaz-Santos et al. (2016); Decarli et al. (2017, 2018), with relative uncertainties. Whether these are not available, we assume the average uncertainty within the sample. The / ratio for QSOs span about 1.5 dex and, although with large scatter, a negative trend is evident. We obtain the relation Log(/) = Log(/L) + , with and by fitting the data with an orthogonal linear regression accounting for errors on both axes. The slope is slightly steeper than the value of derived in Willott et al. (2017) for QSOs. This difference is due to the addition, in our sample, of the hyper-luminous sources from Trakhtenbrot et al. (2017) and Kimball et al. (2015), populating the high tail of the sample.
5 SMBH vs host galaxy properties
5.1 SMBH and dynamical masses
We perform a spectral analysis of the rest-frame UV region around the CIV emission line in the SDSS DR10 (Pâris et al., 2014) spectrum of J10150020, with the aim of estimating the SMBH mass based on the width of the CIV line profile and the luminosity at 1350 Å . The velocity shift of the CIV line with respect to the [CII]-based redshift of the QSO is modest ( km s). Therefore, we are likely observing the CIV emitting region at large inclination angle, as also indicated by the ALMA observation (see below in this section). According to Vietri et al. (2018), higher line-of-sight inclinations correspond to smaller distortions of the CIV line profile, while low inclinations are associated to very broad, asymmetric profiles due to outflowing gas. We can therefore consider the FWHM as a reliable tracer of in J10150020.
Specifically, we fit the spectral region between 1300 Å and 1700 Å with a model consisting of a power-law to reproduce the continuum emission, one Gaussian component to account for the BLR emission of CIV and a second Gaussian component to fit possible CIV wings associated with outflowing gas (see Fig. 6).
We find that the CIV profile is best-fitted by the combination of two Gaussians with dominant contribution from the BLR component. The best fit values are FWHM km sand erg s. These quantities are used to derive according to the single epoch relation from Vestergaard & Peterson (2006), and considering the empirical correction proposed by Coatman et al. (2017) for high-luminosity QSOs to take into account the blueshift km s affecting the CIV line profile:
with and . The resulting black-hole mass of J10150020 is M, where the uncertainties are dominated by the intrinsic 0.2 dex scatter in the -corrected relation from Coatman et al. (2017). This leads to an Eddington ratio of .
The measurements of the FWHM and size of the [CII] emission can be used to estimate the dynamical masses of both the QSO host galaxy and CompA, CompB and CompC. Under the assumption that the ISM is mainly distributed in an inclined, rotating disk, the dynamical mass can be expressed as:
where is the deconvolved major axis of the [CII]-emitting region, computed as 1.5 times the deconvolved major axis (Wang et al., 2013), is the inclination angle between the line of sight and the polar axis of the disk. The FWHM of the [CII] line is related to the circular velocity in the disk by the relation (Wang et al., 2013). In case of a resolved source the inclination of the disk can be estimated from the ratio between semi-minor and semi-major axes as . The resulting inclination values for J10150020 and the [CII] emitters resolved by the ALMA beam are listed in Table 2.
Significant uncertainties can affect this estimate of . Firstly, the approximation of a rotating disk might be incorrect as part of the emitting gas might be associated to non-circular motion, especially in interacting systems. Furthermore, in our high-resolution ALMA observation we might be losing the more extended, low surface brightness [CII] emitting regions around the source, thus underestimating . Moreover, the deconvolved size of the source and the semi-axes ratio are estimated from a 2D-Gaussian profile that might not well reproduce the surface-brightness distribution. Finally, the measured inclination might be altered by non-circular beam shapes in case of marginally-resolved sources. Further discussion on these issues can be found in Trakhtenbrot et al. (2017) and references therein. Keeping in mind all these limitations, we use Eq. 3 to estimate the dynamical mass of J10150020, i.e. Log. We note that assuming the galaxy to be supported by velocity dispersion would lead to a smaller by a factor of 3.
J10150020 is characterised by an extreme ratio of about of with respect to , as it is shown in Fig. 7. Our QSO is compared to the high-z QSOs with single epoch estimates from the sample introduced in Sect. 4. We also plot the relative uncertainties. Whether these are not available, we consider a 0.3 dex error on the (MgII-based) estimates, while we propagate the inclination error to the uncertainty on . For unresolved sources, we plot the value as lower limit on the true dynamical mass. The - relation from Jiang et al. (2011), derived from local galaxies in a wide range of M, is also shown for comparison. According to the local relation, the - ratio at the observed black hole mass should be about , translating into a host galaxy dynamical mass of M, suggesting that we are observing the cradle of what would be a giant galaxy at .
As we also resolve CompA and CompC in our ALMA observation, we are able to estimate their dynamical masses Log and Log (see Table 2). CompB is instead only marginally resolved on one axis in our observation, and an assumption on is therefore necessary. We use (where 0.18 is the minor axis of the ALMA beam) and Eq. 3 to derive a lower limit on its dynamical mass of M. Given their small distance from J10150020 all these companions are likely going to merge and eventually build up the mass of the QSO host galaxy. By combining the dynamical masses of the QSO and [CII] emitters we obtain a large value of M already at , as it is shown in Fig. 7.
5.2 SMBH and host galaxy growth
In the previous sections we have discussed the presence of multiple companions at very close distance ( kpc) from J10150020, which will likely contribute to the final mass of the QSO host galaxy. By comparing the mass accretion rate of the SMBH with the SFR, we can in principle understand how high-z QSOs have grown to reach their location in Fig. 7, under the assumption that we observe them when most of the black hole and galaxy mass is being assembled.
can be derived as , once assumed a standard accretion efficiency . The resulting value M yr for J10150020 is shown in Fig. 8a as a function of the SFR, compared to the 42 high-z QSOs sample introduced in Sect. 5.13. We also include five WISSH QSOs with available and estimates presented in Bischetti et al. (2017); Duras et al. (2017); Vietri et al. (2018). Whether not available in literature, we compute by using the bolometric correction from of Runnoe et al. (2012), assuming as uncertainty the 0.1 dex scatter found for this correction. We also compute the SFR from the (8 m) according to the relation reported in Kennicutt & Evans (2012), with an associated scatter of 0.2 dex, and dividing the SFR by a factor of two in case of QSOs with erg s, as suggested by Duras et al. (2017).
From the relation of Fiore et al. (2017) between mass outflow rate and AGN bolometric luminosity Log(Log(/erg s)32, we can define the locus of points in Fig. 8a with unitary mass loading factor /SFR which translates into a . J10150020 and most of the high-z QSOs clearly lie above this line, suggesting that they are potentially able to develop massive molecular outflows affecting the growth of their host galaxies. Future deep ALMA observations of the CO emission in these objects will be able to confirm this prediction.
From the non-detection of the five QSO companions in the WISE bands, we can compute an upper limit on the of a possible AGN contribution in these sources. Specifically, by using Mrk231 and NGC6420 templates (Polletta et al., 2007; Fiore et al., 2008), we derived a value of Log(/erg s) , corresponding to a M yr. In the companion galaxies, where the AGN effect is absent or very low, it is possible to have star formation activity at a comparable or even higher SFR (orange and red stars in Fig. 8a, see also Sect. 4.2) with respect to the QSO host galaxy itself. This suggests that a significant fraction of stellar mass may be assembled in the QSO satellites and then contribute at later times to the QSO host galaxy mass by mergers.
We compare the SMBH growth and stellar mass assembly timescales in J10150020. A basic estimate of the present growth timescale can be derived as the ratio , if a constant mass accretion rate equal to the observed value is assumed. Following the same approach, i.e. by assuming a constant SFR, one can estimate the present stellar mass growth timescale as . Few combined [CII] and CO observations of high-z QSOs are available so far (Wang et al., 2013; Venemans et al., 2017), providing a wide range of molecular gas fractions contributing to the total dynamical mass, from 10% to 80%, if a negligible dark matter content in the inner regions of the galaxy is assumed (Genzel et al., 2017). Therefore, the resulting should be considered as upper limits of the real stellar mass assembly timescales. Fig. 8b displays as a function of for J10150020 and the same high-z QSOs plotted in Fig. 7. ranges from about 30 Myr to 1 Gyr, while the black holes have reached their current mass in Myr. The two growth timescales appear comparable, bearing in mind the large uncertainties affecting these measurements, as most sources lie close to the relation. This suggests that the QSOs of Fig. 7 are moving in the - plane in parallel to the local relation. However, in most of the QSOs the / ratio is . This is likely due to the extreme accretion rates measured in these sources.
An alternative approach consists of estimating the exponential growth timescale of a black hole accreting at constant Eddington rate , where /erg s is the Eddington luminosity. Specifically, following Volonteri & Rees (2005):
where Gyr is the characteristic accretion timescale (Shapiro, 2005) and M is the initial mass of the black hole seed (see Fig 8c). We thus calculate the exponential stellar mass growth timescale of the host galaxy by assuming constant SFR/ ratio, equal to the observed value:
where is the initial dynamical mass. Specifically, we computed M from the minimum mass of a star forming dark matter halo (Finlator et al., 2011) by rescaling for the cosmological baryon fraction , since for the majority of the sources (including J10150020) we are not able to distinguish between gas and stellar mass. We find that is on average a factor of 10 larger than (see Fig. 8c), ranging from 300 Myr to 6 Gyr. This translates into a typical ratio between exponential black hole mass growth and exponential stellar mass growth of , as is comparable to . According to this scenario, the host galaxies of high-z QSOs have grown in a shorter timescale than their central SMBHs. A shorter BH growth timescale can be obtained by either assuming a more massive seed (e.g. M) or a super-Eddington accretion regime onto stellar mass seeds (100 M, e.g. Volonteri, 2010; Valiante et al., 2017).
According to this scenario, the host galaxies of high-z QSOs are growing faster than their central SMBHs. Finally, we note that adopting a Gyr, typical of galaxies (Stark et al., 2009; González et al., 2010), would shift most QSOs towards larger stellar , i.e. closer to the relation. This indicates that the bulk of the high-z QSOs sample considered here is undergoing a peculiar phase of intense SF activity.
In this work, we report on the analysis of the ALMA high-resolution observation of the 840 m continuum and [CII] m line emission of the WISSH quasar J10150020 and its surrounding (8.5 arcsec radius) field. This data allows us to characterise the host galaxy and environment properties of this hyper-luminous QSO at . Our main findings can be summarised as follows:
The ALMA observation reveals an exceptional overdensity of [CII]-emitting companions with a very small ( km s) velocity shift with respect to the QSO redshift. Specifically, we report the discovery of the closest companion observed so far in sub-mm observations of high-z QSOs, distant only 2.2 kpc and merging with the QSO, as indicated by the increased velocity dispersion in the host of J10150020 offset from the AGN location. The other two [CII] emitters are located at 8 and 17 kpc.
We also detect two continuum emitters (Cont1 and Cont2 in Table 1) within an angular separation of less than 3.5 arcsec, which are characterised by a 840 m continuum flux density comparable to that of the hyper-luminous QSO. These sources do not show line emission in the ALMA band but are likely physically associated with J10150020. In fact, they exceed by a factor of 100 the number of expected sources according to the 850 m Log(N)-Log(S) (see Sect. 4). Comparing the five companions detected around J10150020 with typical number density of galaxies from deep sub-mm surveys, clearly indicates the presence of a significant overdensity of star-forming galaxies around this QSO. We are observing the early phase of the formation of a giant, massive galaxy, assembled by merging of the ALMA-detected companions with a M SMBH at its centre.
We are able to accurately build up the SED of the emission from the QSO, by quantifying and removing the contribution to the 250840 m fluxes from Cont1 and Cont2 (see Sect. 4.2). The QSO host galaxy has a SFR of about 100 M yr, while the bulk of the SF activity takes place in Cont1 and Cont2, for which we derive a SFR700 M yr. The [CII] emitters contribute to additional 100 M yr. A significant fraction of the stellar mass assembly at earlier epochs may have therefore taken place in the companion galaxies, more than in the QSO host galaxy itself (e.g. Anglés-Alcázar et al., 2017a).
For J10150020 we measure a SMBH mass of M by using a single epoch relation based on the CIV emission line profile, which is a reliable tracer of the mass since it does not exhibit a strong blueshifted wing associated with non-virial motions (Coatman et al., 2017; Vietri et al., 2018, see Sect. 7). From the [CII] line profile and emitting region, we also compute the dynamical mass of the QSO host galaxy, M. This translates into an extreme / ratio of which is a factor of 100 smaller than what typically observed in local galaxies. According to the local relation from Jiang et al. (2011), such a SMBH mass should be hosted in giant galaxy with a stellar mass of M. Remarkably, the total stellar mass of QSO plus [CII]-emitting companions already exceeds M at
J10150020, as most of the 47 QSOs at from the sample described in Sect. 8, is potentially able to drive a massive molecular outflow affecting the SFR in the host galaxy, according to the relation from Fiore et al. (2017) between and (see Sect. 5.2). Future deep ALMA observations of the CO emission in these objects will be able to confirm this prediction. Fig 8b and 8c compare the SMBH vs galaxy growth timescales of high-z QSOs by assuming a constant (, SFR) and (, ), respectively. We find that the present growth rate of the host galaxy is comparable to the one inferred for the SMBH (Fig. 8b), while Fig. 8c suggests that the time necessary to reach the observed is shorter than the time required to the SMBH to accrete the observed mass.
Acknowledgements.This paper makes use of the following ALMA data: ADS/JAO.ALMA#2016.1.00718.S. ALMA is a partnership of ESO (representing its member states), NSF (USA) and NINS (Japan), together with NRC (Canada), MOST and ASIAA (Taiwan), and KASI (Republic of Korea), in cooperation with the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO and NAOJ. Funding for SDSS-III has been provided by the Alfred P. Sloan Foundation, the Participating Institutions, the National Science Foundation, and the U.S. Department of Energy Office of Science. The SDSS-III web site is http://www.sdss3.org/. SDSS-III is managed by the Astrophysical Research Consortium for the Participating Institutions of the SDSS-III Collaboration including the University of Arizona, the Brazilian Participation Group, Brookhaven National Laboratory, Carnegie Mellon University, University of Florida, the French Participation Group, the German Participation Group, Harvard University, the Instituto de Astrofisica de Canarias, the Michigan State/Notre Dame/JINA Participation Group, Johns Hopkins University, Lawrence Berkeley National Laboratory, Max Planck Institute for Astrophysics, Max Planck Institute for Extraterrestrial Physics, New Mexico State University, New York University, Ohio State University, Pennsylvania State University, University of Portsmouth, Princeton University, the Spanish Participation Group, University of Tokyo, University of Utah, Vanderbilt University, University of Virginia, University of Washington, and Yale University. The UKIDSS project is defined in Lawrence et al 2007. UKIDSS uses the UKIRT Wide Field Camera (WFCAM; Casali et al 2007) and a photometric system described in Hewett et al 2006. The pipeline processing and science archive are described in Irwin et al (2008) and Hambly et al (2008). We have used data from the 2nd data release, which is described in detail in Warren et al (2007). We thank E. Merlin for the assistance with the HST photometry and M. Ginolfi for the useful discussion on the detection of sources in interferometric images. C.P. acknowledges support from the Science and Technology Foundation (FCT, Portugal) through the Post-doctoral Fellowship SFRH/BPD/90559/2012, PEst-OE/FIS/UI2751/2014, and PTDC/FIS-AST/2194/2012. S.C. acknowledges financial support from the Science and Technology Facilities Council (STFC).
- Alexander & Hickox (2012) Alexander, D. M. & Hickox, R. C. 2012, New A Rev., 56, 93
- Anglés-Alcázar et al. (2017a) Anglés-Alcázar, D., Faucher-Giguère, C.-A., Kereš, D., et al. 2017a, MNRAS, 470, 4698
- Anglés-Alcázar et al. (2017b) Anglés-Alcázar, D., Faucher-Giguère, C.-A., Quataert, E., et al. 2017b, MNRAS, 472, L109
- Aravena et al. (2016a) Aravena, M., Decarli, R., Walter, F., et al. 2016a, ApJ, 833, 71
- Aravena et al. (2016b) Aravena, M., Decarli, R., Walter, F., et al. 2016b, ApJ, 833, 68
- Bañados et al. (2016) Bañados, E., Venemans, B. P., Decarli, R., et al. 2016, ApJS, 227, 11
- Banerji et al. (2017) Banerji, M., Carilli, C. L., Jones, G., et al. 2017, MNRAS, 465, 4390
- Bischetti et al. (2017) Bischetti, M., Piconcelli, E., Vietri, G., et al. 2017, A&A, 598, A122
- Bouwens et al. (2015) Bouwens, R. J., Illingworth, G. D., Oesch, P. A., et al. 2015, ApJ, 803, 34
- Carniani et al. (2015) Carniani, S., Maiolino, R., De Zotti, G., et al. 2015, A&A, 584, A78
- Carniani et al. (2013) Carniani, S., Marconi, A., Biggs, A., et al. 2013, A&A, 559, A29
- Cicone et al. (2015) Cicone, C., Maiolino, R., Gallerani, S., et al. 2015, A&A, 574, A14
- Coatman et al. (2017) Coatman, L., Hewett, P. C., Banerji, M., et al. 2017, MNRAS, 465, 2120
- De Rosa et al. (2011) De Rosa, G., Decarli, R., Walter, F., et al. 2011, ApJ, 739, 56
- De Rosa et al. (2014) De Rosa, G., Venemans, B. P., Decarli, R., et al. 2014, ApJ, 790, 145
- Decarli et al. (2017) Decarli, R., Walter, F., Venemans, B. P., et al. 2017, Nature, 545, 457
- Decarli et al. (2018) Decarli, R., Walter, F., Venemans, B. P., et al. 2018, ApJ, 854, 97
- Di Matteo et al. (2005) Di Matteo, T., Springel, V., & Hernquist, L. 2005, Nature, 433, 604
- Díaz-Santos et al. (2016) Díaz-Santos, T., Assef, R. J., Blain, A. W., et al. 2016, ApJ, 816, L6
- Duras et al. (2017) Duras, F., Bongiorno, A., Piconcelli, E., et al. 2017, A&A, 604, A67
- Finlator et al. (2011) Finlator, K., Oppenheimer, B. D., & Davé, R. 2011, MNRAS, 410, 1703
- Fiore et al. (2017) Fiore, F., Feruglio, C., Shankar, F., et al. 2017, A&A, 601, A143
- Fiore et al. (2008) Fiore, F., Grazian, A., Santini, P., et al. 2008, ApJ, 672, 94
- Fujimoto et al. (2016) Fujimoto, S., Ouchi, M., Ono, Y., et al. 2016, ApJS, 222, 1
- Gabor et al. (2016) Gabor, J. M., Capelo, P. R., Volonteri, M., et al. 2016, A&A, 592, A62
- Genzel et al. (2017) Genzel, R., Schreiber, N. M. F., Übler, H., et al. 2017, Nature, 543, 397
- González et al. (2010) González, V., Labbé, I., Bouwens, R. J., et al. 2010, ApJ, 713, 115
- Griffin et al. (2010) Griffin, M. J., Abergel, A., Abreu, A., et al. 2010, A&A, 518, L3
- Hopkins et al. (2008) Hopkins, P. F., Hernquist, L., Cox, T. J., & Kereš, D. 2008, ApJS, 175, 356
- Jiang et al. (2007) Jiang, L., Fan, X., Vestergaard, M., et al. 2007, AJ, 134, 1150
- Jiang et al. (2011) Jiang, Y.-F., Greene, J. E., & Ho, L. C. 2011, ApJ, 737, L45
- Kennicutt & Evans (2012) Kennicutt, R. C. & Evans, N. J. 2012, ARA&A, 50, 531
- Kimball et al. (2015) Kimball, A. E., Lacy, M., Lonsdale, C. J., & Macquart, J.-P. 2015, MNRAS, 452, 88
- Kormendy & Ho (2013) Kormendy, J. & Ho, L. C. 2013, ARA&A, 51, 511
- Lutz et al. (2010) Lutz, D., Mainieri, V., Rafferty, D., et al. 2010, ApJ, 712, 1287
- Maiolino et al. (2005) Maiolino, R., Cox, P., Caselli, P., et al. 2005, A&A, 440, L51
- McMullin et al. (2007) McMullin, J. P., Waters, B., Schiebel, D., Young, W., & Golap, K. 2007, in Astronomical Society of the Pacific Conference Series, Vol. 376, Astronomical Data Analysis Software and Systems XVI, ed. R. A. Shaw, F. Hill, & D. J. Bell, 127
- Menci et al. (2008) Menci, N., Fiore, F., Puccetti, S., & Cavaliere, A. 2008, ApJ, 686, 219
- Menci et al. (2014) Menci, N., Gatti, M., Fiore, F., & Lamastra, A. 2014, A&A, 569, A37
- Nardini & Risaliti (2011) Nardini, E. & Risaliti, G. 2011, MNRAS, 415, 619
- Netzer (2009) Netzer, H. 2009, MNRAS, 399, 1907
- Pâris et al. (2014) Pâris, I., Petitjean, P., Aubourg, É., et al. 2014, A&A, 563, A54
- Pilbratt et al. (2010) Pilbratt, G. L., Riedinger, J. R., Passvogel, T., et al. 2010, A&A, 518, L1
- Planck Collaboration et al. (2016) Planck Collaboration, Ade, P. A. R., Aghanim, N., et al. 2016, A&A, 594, A13
- Polletta et al. (2007) Polletta, M., Tajer, M., Maraschi, L., et al. 2007, ApJ, 663, 81
- Rosario et al. (2013) Rosario, D. J., Santini, P., Lutz, D., et al. 2013, ApJ, 771, 63
- Runnoe et al. (2012) Runnoe, J. C., Brotherton, M. S., & Shang, Z. 2012, MNRAS, 422, 478
- Sargsyan et al. (2014) Sargsyan, L., Samsonyan, A., Lebouteiller, V., et al. 2014, ApJ, 790, 15
- Shapiro (2005) Shapiro, S. L. 2005, ApJ, 620, 59
- Simpson et al. (2015) Simpson, J. M., Smail, I., Swinbank, A. M., et al. 2015, ApJ, 807, 128
- Solomon & Vanden Bout (2005) Solomon, P. M. & Vanden Bout, P. A. 2005, ARA&A, 43, 677
- Stark et al. (2009) Stark, D. P., Ellis, R. S., Bunker, A., et al. 2009, ApJ, 697, 1493
- Trakhtenbrot et al. (2017) Trakhtenbrot, B., Lira, P., Netzer, H., et al. 2017, ApJ, 836, 8
- Treister et al. (2012) Treister, E., Schawinski, K., Urry, C. M., & Simmons, B. D. 2012, ApJ, 758, L39
- Valiante et al. (2017) Valiante, R., Agarwal, B., Habouzit, M., & Pezzulli, E. 2017, PASA, 34, e031
- Valiante et al. (2014) Valiante, R., Schneider, R., Salvadori, S., & Gallerani, S. 2014, MNRAS, 444, 2442
- Valiante et al. (2016) Valiante, R., Schneider, R., & Volonteri, M. 2016, PASA, 33, e032
- Venemans et al. (2015) Venemans, B. P., Bañados, E., Decarli, R., et al. 2015, ApJ, 801, L11
- Venemans et al. (2017) Venemans, B. P., Walter, F., Decarli, R., et al. 2017, ApJ, 837, 146
- Venemans et al. (2017) Venemans, B. P., Walter, F., Decarli, R., et al. 2017, The Astrophysical Journal, 845, 154
- Venemans et al. (2016) Venemans, B. P., Walter, F., Zschaechner, L., et al. 2016, ApJ, 816, 37
- Vestergaard & Peterson (2006) Vestergaard, M. & Peterson, B. M. 2006, ApJ, 641, 689
- Vietri et al. (2018) Vietri, G., Piconcelli, E., Bischetti, M., et al. 2018, ArXiv e-prints [\eprint[arXiv]1802.03423]
- Volonteri (2010) Volonteri, M. 2010, A&A Rev., 18, 279
- Volonteri et al. (2015) Volonteri, M., Capelo, P. R., Netzer, H., et al. 2015, MNRAS, 449, 1470
- Volonteri & Rees (2005) Volonteri, M. & Rees, M. J. 2005, ApJ, 633, 624
- Wang et al. (2013) Wang, R., Wagg, J., Carilli, C. L., et al. 2013, ApJ, 773, 44
- Wang et al. (2016) Wang, R., Wu, X.-B., Neri, R., et al. 2016, ApJ, 830, 53
- Warren et al. (2007) Warren, S. J., Hambly, N. C., Dye, S., et al. 2007, MNRAS, 375, 213
- Willott et al. (2015) Willott, C. J., Bergeron, J., & Omont, A. 2015, ApJ, 801, 123
- Willott et al. (2017) Willott, C. J., Bergeron, J., & Omont, A. 2017, ApJ, 850, 108
- Willott et al. (2013) Willott, C. J., Omont, A., & Bergeron, J. 2013, ApJ, 770, 13
- Wright et al. (2010) Wright, E. L., Eisenhardt, P. R. M., Mainzer, A. K., et al. 2010, AJ, 140, 1868