GA-NIFS and EIGER: A merging quasar host at z=7 with an overmassive black hole

TL;DR

This study uses JWST NIRSpec IFU spectroscopy and NIRCam imaging to dissect the $z=7.08$ quasar J1120+0641's host galaxy and its environment. The data reveal a major merger with a close companion, yielding dynamical masses of order $10^{10} M_\odot$ for both galaxies and a host stellar mass of $M_{*,\rm host} \approx 3\times10^9 M_\odot$, while the virial BH mass is $M_{\rm BH} \approx 1.9\times10^9 M_\odot$, producing an extreme $M_{\rm BH}/M_*\approx0.63$. This ratio is ~3 dex above local BH–stellar-mass relations, indicating an overmassive black hole during a major merger, and illustrating the power of combining NIRSpec IFU and NIRCam data to study early BH growth and host galaxy assembly. The results also align with ALMA observations showing the host dominates the far-IR emission, while the companion shows relatively little star formation, consistent with a merger-driven growth scenario. Overall, the work demonstrates JWST’s capability to map kinematics, measure BH masses, and constrain stellar populations in the most distant quasar hosts, advancing our understanding of black hole–galaxy co-evolution in the early universe.

Abstract

The James Webb Space Telescope is revolutionising our ability to understand the host galaxies and local environments of high-z quasars. Here we obtain a comprehensive understanding of the host galaxy of the z=7.08 quasar J1120+0641 by combining NIRSpec integral field spectroscopy with NIRCam photometry of the host continuum emission. Our emission-line maps reveal that this quasar host is undergoing a merger with a bright companion galaxy. The quasar host and the companion have similar dynamical masses of $\sim10^{10}M_\odot$, suggesting that this is a major galaxy interaction. Through detailed quasar subtraction and SED fitting using the NIRCam data, we obtained an estimate of the host stellar mass of $M_{\ast}=(3.0^{+2.5}_{-1.4})\times10^9M_\odot$, with $M_{*}=(2.7^{+0.5}_{-0.5})\times10^9M_\odot$ for the companion galaxy. Using the H$β$ Balmer line we estimated a virial black hole mass of $M_{\rm{BH}}=(1.9^{+2.9}_{-1.1})\times10^9 M_\odot$. Thus, J1120+0641 has an extreme black hole-stellar mass ratio of $M_{\rm{BH}}/M_\ast=0.63^{+0.54}_{-0.31}$, which is ~3 dex larger than expected by the local scaling relations between black hole and stellar mass. J1120+0641 is powered by an overmassive black hole with the highest reported black hole-stellar mass ratio in a quasar host that is currently undergoing a major merger. These new insights highlight the power of JWST for measuring and understanding these extreme first quasars.

Figures (11)

• Figure 1: Integrated spectrum of J1120+0641 from the IFU data (blue). The spectrum is integrated over an aperture of radius $0\overset{\prime\prime}{.}35$ centred on the peak of the quasar emission. An aperture flux correction of 1.23$\times$ has been applied. The top panel shows the spectrum integrated from the original background-subtracted data cube. The bottom panel shows the spectrum integrated from the continuum-subtracted data cube. The green shaded regions show the continuum windows used to model and subtract the continuum emission (Section \ref{['sec:Continuum']}). At this redshift, $z=7.08$, the [O$\;$iii] $\lambda\lambda4959,5007$ doublet falls just blueward of the detector gap (grey shaded region), and the broad H$\alpha$ line falls just off the red edge of detector. The purple curve shows the wavelength coverage of the NIRCam F356W filter used in the EIGER images. The yellow curve shows the Park2022 iron emission template, with redshift and normalisation taken from our best spectral model fit (Section \ref{['sec:BHfitting']}).
• Figure 2: Quasar-subtracted [O$\;$iii] $\lambda5007$ emission map from the NIRSpec IFU (grey image) compared to the [C$\;$ii] $158\mu$m emission map (orange contours) and the rest-frame 158$\mu$m FIR continuum emission map (purple contours) from ALMA (see Section \ref{['sec:ALMA']}). The [C$\;$ii] $158\mu$m contours are linearly spaced from 5$\sigma$ to 26$\sigma$, where $\sigma=16\mu$Jy. The FIR continuum contours are linearly spaced from 5$\sigma$ to 42$\sigma$, where $\sigma=7\mu$Jy/beam. The grey circle depicts the approximate PSF of the ALMA observations, with beam diameter $0\overset{\prime\prime}{.}3$. The green cross marks the peak of the quasar emission from the NIRSpec IFU data that has been astrometrically aligned to the NIRCam imaging, which is aligned to Gaia DR2 Yue2023. The purple circle marks the peak of the 158$\mu$m FIR continuum emission reported by Venemans2017a. The blue diamond marks the original quasar position from UKIDSS quoted by Mortlock2011.
• Figure 3: Emission-line regions surrounding J1120+0641 showing flux and kinematic maps after the subtraction of the quasar emission. The top-left and -right panels show the flux of the [O$\;$iii] $\lambda5007$ and H$\beta$ lines, respectively, from the integrated flux of the fitted Gaussian in each spaxel. The bottom-left and -right panels show our [O$\;$iii] $\lambda5007$ kinematic maps, depicting the non-parametric central velocity of the line ($v_{50}$) relative to the quasar host redshift of $z=7.0804\pm0.0028$ and the line width ($w_{80}$), respectively. Three emission-line regions are highlighted by coloured ellipses, and the crosses show the location of the quasar.
• Figure 4: Quasar-subtracted spectra integrated over the three spatial regions shown in Figure \ref{['fig:RegionMaps']} (opaque coloured lines) along with our best-fit Gaussian models for the H$\beta$ and [O$\;$iii] $\lambda\lambda4959,5007$ emission lines (black). The analogous spectra from the non-quasar-subtracted cube are also plotted for comparison (transparent coloured lines), showing the necessity of quasar subtraction to accurately measure the emission from Regions 1 and 2. All spectra have been continuum subtracted. The vertical lines mark the location of the H$\beta$ and [O$\;$iii] $\lambda\lambda4959,5007$ lines at the redshift of the quasar host galaxy, Region 1, as measured from the fit to this spectrum, $z=7.0804\pm0.0028$. For the host Region 1, we exclude the central $5\times5$ pixels surrounding the quasar peak as well as nearby spaxels with [O$\;$iii] $\lambda5007$ velocity offset $>300$ km/s, as these are highly corrupted by the quasar subtraction and introduce significant noise and artefacts. This means that we slightly underestimate the total flux in this region; however, the fluxes are significantly more reliable than if these most corrupted spaxels were included.
• Figure 5: Integrated quasar spectrum for J1120+0641 (black) showing the region around H$\beta$. The spectrum is integrated over a radius of $0\overset{\prime\prime}{.}35$, and no correction for the loss of flux in this aperture has been applied (although this is accounted for in the related calculations). The best model fit (red) is shown alongside the narrow-line components (green), the broader outflow components (blue), the iron emission template (pink), and the model for the BLR (orange). The lower panels show the residual of the model fit. Both models assume the iron emission follows the Park2022 template. The left panels show a model assuming that the H$\beta$ BLR is described by a BPL model. The right panels show a model assuming that the H$\beta$ BLR is described by two Gaussian profiles (the DG model). The shaded grey regions show where the detector gap falls, resulting in no coverage of those wavelengths.