James Webb Space Telescope Feed Post


EarlyReleases
Date: 1/26/2024

The MBH - M* relation up to z ~ 2 through decomposition of COSMOS-Web NIRCam images


Comparison of different methods to evaluate empirical models of the PSF. Columns (a) to (c) show images of the top-5 stacked PSF, the top-75% stacked PSF, and the PSFEx-PSF for an example galaxy, CID-62 (z = 1.92), in F115W, F150W, F277W, and F444W, from top to bottom. FWHM along the semi-major axis measured from the 2D Gaussian fits shown in the top-left corner of each image. The white bars in the lower left indicate a scale of 1''. Columns (d) and (e) show the distribution of the FWHM (semi-major axis) and ellipticity b/a. Gray histograms display the distribution of each single PSF in the library (Section 3.1.1). The values for the top-5 stacked PSF, the top-75% stacked PSF, PSFEx-PSF are marked by the blue, green, and red vertical lines, respectively. Abstract: Our knowledge of relations between supermassive black holes and their host galaxies at z?1 is still limited, even though being actively sought out to z~6. Here, we use the high resolution and sensitivity of JWST to measure the host galaxy properties for 61 X-ray-selected type-I AGNs at 0. 7 < z < 2.5 with rest-frame optical/near-infrared imaging from COSMOS-Web and PRIMER. Black hole masses (log(MBH/M?)~7.5-9.5) are available from previous spectroscopic campaigns. We extract the host galaxy components from four NIRCam broadband images and the HST/ACS F814W image by applying a 2D image decomposition technique. We detect the host galaxy for ~90% of the sample after subtracting the unresolved AGN emission. With host photometry free of AGN emission, we determine the stellar mass of the host galaxies to be log(M*/M?)~10-11.5 through SED fitting and measure the evolution of the mass relation between SMBHs and their host galaxies. Considering selection biases and measurement uncertainties, we find that the MBH/M* ratio evolves as (1+z)0.37+0.35-0.60 thus remains essentially constant or exhibits mild evolution up to z~2.5. We also see an amount of scatter (sµ=0.28±0.13) is similar to the local relation and consistent with low-z studies; this appears to not rule out non-causal cosmic assembly where mergers contribute to the statistical averaging towards the local relation. We highlight improvements to come with larger samples from JWST and, particularly, Euclid, which will exceed the statistical power of wide and deep surveys such as Subaru Hyper Suprime-Cam.