GA-NIFS: A smouldering disk galaxy undergoing ordered rotation at z=4.26

TL;DR

This work uses high-resolution JWST/NIRSpec IFU data, complemented by ALMA [CII] observations, to study the z=4.26 disk galaxy DLA0817g1. By applying spatially resolved line fitting and 3D tilted-ring modelling with ^3DBarolo, the authors show nearly identical rotation in H$ m ext{A}$ and [CII], but a higher velocity dispersion in H$ m ext{A}$, reflecting multi-phase ISM kinematics. They derive a gas-phase metallicity of about $0.7\,Z_\odot$, a low ionisation parameter, and a modest dust attenuation, painting a picture of a mature, 'smouldering' disk with substantial past star formation and a sizeable molecular gas reservoir. A mass decomposition indicates the dynamical mass is $\log_{10}(M_{\rm dyn}/M_\odot)\approx10.9$, with molecular gas comprising a significant but not dominant fraction, leaving room for a stellar component and possibly a dark matter halo. Overall, DLA0817g1 emerges as a representative high-z disk with ordered rotation, offering crucial insights into the interplay of gas, metallicity, star formation, and dynamics during early galaxy evolution.

Abstract

Rotating galaxies with relaxed gaseous disks have been discovered across cosmic time, from the local Universe to high redshift (z>4). But few such sources have been confirmed at z>4, making them a precious sample to examine what conditions result in such ordered kinematics in an early, more chaotic Universe. One of the best examples of this sample is the galaxy DLA0817g1 (z=4.2603), which shows remarkably clear rotation in ALMA [CII] data. We present recent JWST/NIRSpec IFU data (R~2700) of DLA0817g1, which we combine with archival ALMA [CII] observations to place constraints on its ISM conditions and morpho-kinematics. From a combination of line ratios, we find a high gas-phase metallicity (~0.7 solar), high fraction of obscured star formation, low ionisation (compared to other high-redshift galaxies observed with JWST), and no significant evidence for AGN (based on the WHAN diagnostic). Dynamical modelling with 3DBarolo reveal nearly identical rotation in Halpha and [CII], but with a higher velocity dispersion in the former. Using our metallicity estimate and previous CO and [CII] detections, we derive a new estimate of the molecular gas mass, relieving a previous strain in the mass budget. Altogether, we suggest that this is a 'smouldering' galaxy, where past star formation resulted in significant chemical enrichment (i.e., Zgas and Mdust), but the current activity is low (i.e., lower ionisation parameter and electron temperature). These new observations have opened a window into questions regarding the interplay of gas, metallicity, star formation, and kinematics in a prototypical early disk galaxy.

Figures (14)

• Figure 1: Best fit line intensity maps of $\rm H\alpha$, [NII]$\lambda6584$, and [CII], where values are given per pixel. In the rightmost panel, we instead compare the $\rm H\alpha$ (red) and [CII] (cyan) emission using contours of $[0.25,0.50,0.90]\times$ the peak value of each. In each panel, we show a physical scale of 2 kpc. The PSF is represented as a red circle or ellipse to the southeast (north is up and east is to the left). Our adopted aperture is shown as a white ellipse.
• Figure 2: Best fit $v_{50}$ (left column) and $w_{80}$ (right column) maps for $\rm H\alpha$ (top row) and [CII] (lower row). In each panel, we show a physical scale of 2 kpc. The PSF is represented as a red circle or ellipse to the southeast (north is up and east is to the left). Our adopted aperture is shown as a white ellipse.
• Figure 3: Results of fitting integrated spectrum of $\rm DLA0817g1$. The top panel shows a full spectrum (pink line with $1\sigma$ uncertainty shown as shaded region) along with the best-fit model (green). Panels a-e are zoomed-in views for each line, showing the individual best-fit Gaussian components (light blue) and the centroid wavelength of each of the 2-Gaussian fits (red and blue lines above each spectrum). The lower portion of each panel depicts the residual spectra for a single-Gaussian fit (purple) and out two-Gaussian fit (brown).
• Figure 4: Preliminary test to constrain $E(B-V)$ from our observed [NII]$\lambda\lambda 6548,6584$ and [SIII]$\lambda\lambda 9069,9532$ ratios. For each panel, we show the observed ratio and its $1\sigma$ uncertainty as a black line and shaded region, respectively. The dust-reddened intrinsic value of each ratio for a given $E(B-V)$ is shown as a red dashed line.
• Figure 5: Results of WHAN diagnostics. The top panel show the distribution of spaxel values in the $\log_{10}(EW_{\rm H\alpha}/AA)$-$\log_{10}({\rm N2})$ space, with the demarcation lines and classifications of cidf10 included. The circles and blue diamond show the values found through a fit to the spaxel-by-spaxel spectra and the integrated spectrum, respectively. For comparison, we also show the location of a $z=4.6348$ AGN host galaxy studied with JWST/NIRSpec (red point - Ulema; deug25). In the bottom panel, we show the spatial distribution of $\log_{10}({\rm N2})$ (i.e., the x-axis of the upper panel), with spaxels that would be classified as 'strong AGN' marked with black stars (and all other coloured spaxels classified as 'SF'). The PSF is represented as a red circle to the southeast (north is up and east is to the left), and we include a physical scale of 2 kpc. Our adopted aperture is shown as the white ellipse.