A grand-design spiral galaxy 1.5 billion years after the Big Bang with JWST

TL;DR

The paper addresses when grand-design spiral galaxies first emerged and how they form in the early Universe. Using an integrated approach that combines GALFIT bulge-disk modeling, BAGPIPES and Prospector SPS fits, and CASGM morphology metrics on JWST/ UNCOVER data, it identifies Alaknanda as a $z\sim4$ grand-design spiral with a massive disk. Alaknanda has $\log(M_\star/M_\odot)\sim10.2$, SFR $\approx 63\,M_\odot\,\mathrm{yr}^{-1}$, a disk radius of order $10$ kpc, and an age of a few $10^8$ years, with beads-on-a-string star formation along the two arms. This finding demonstrates that massive Spiral disks and long-lived spiral patterns can exist as early as 1.5 Gyr after the Big Bang, informing formation scenarios and motivating future kinematic follow-up with NIRSpec IFU and ALMA.

Abstract

We report the discovery of Alaknanda, a large ($\sim10$ kpc diameter), massive ($\log(M_\star/M_\odot)\sim10.2$), candidate grand-design spiral galaxy with photometric redshift $z_{phot}\sim4.05$ in the UNCOVER and Medium band, Mega Science surveys with JWST. This is among the highest redshift spiral galaxies discovered with JWST. Our morphological analysis using GALFIT reveals that this galaxy is a well-formed disk, with two symmetric spiral arms that are clearly visible in the GALFIT residual. In the rest-frame near-UV and far-UV, we clearly see the beads-on-a-string pattern of star formation; in the rest-frame visible bands, each string appears as an arm. Spectral energy distribution modeling using the BAGPIPES and Prospector codes is strongly constrained by detections and flux measurements in 21 JWST and HST filters. From the BAGPIPES modeling, the stellar mass-weighted age is $\sim 199$ Myr, implying 50\% of the stars in the galaxy formed after $z\sim4.6$. This is a highly star-forming galaxy with a star formation rate (SFR) of $\sim 63 \, M_\odot \, \text{yr}^{-1}$. We detect flux excesses in the F250M and F335M filters due to the presence of H-$α$+[NII] and [OIII]+H-$β$ emission line complexes respectively. Detection of a spiral galaxy at $z \sim 4$ indicates that massive and large spiral galaxies and disks were already in place merely 1.5 billion years after the Big Bang. Future observations with NIRSpec IFU and ALMA will be able to probe the kinematics of the galactic disk, throwing light on the possible origin of the spiral arms in this galaxy.

Figures (5)

• Figure 1: We show the grayscale cutouts of Alaknanda in all JWST/NIRCam broadband filters in the top two panels. The red horizontal bar at the bottom right of each cutout shows a 2 kpc scale at the redshift of the galaxy. We also provide the central rest frame wavelength (in nm) for each filter at the bottom left of each cutout. The F250M and F335M filters contain the [OIII]+H-$\beta$ and H-$\alpha$+[NII] emission line complexes respectively. In the last two columns of the middle row, we show the continuum subtracted images in the F250M and F335M filters that show the regions with line emission in the disk. In these two panels, we show the rest frame wavelengths (in nm) of the [OIII] doublet and H-$\beta$ in the F250M cutout and H-$\alpha$ and [NII] doublet in the F335M cutout, at the top. The bottom row shows RGB composite images of the galaxy in sequence: RGB composite with F200W, F150W, F090W SW filters and RGB image with F444W, F356W and F277W LW filters. The last two images in the bottom row show the GALFIT residual in F277W filter obtained after subtracting the bulge + disk model derived by GALFIT and the outline of the spiral arms of the galaxy (yellow curve) superposed on the GALFIT residual. The red ticks on the x and y axes point to the location of the possible satellite spheroid galaxy at $z=3.973671$. All cutouts are $2.5 \times 2.5$ arcsec in size and North is on top and East is to the left.
• Figure 2: The left panel shows a plot of the Gini coefficient against the M20 parameter. The blue and orange lines separating merger, disk, and E/S0 galaxies in the left panel have been taken from Lotz_2008b. The right panel shows the log of the asymmetry against concentration. The magenta line separating mergers and disks has been taken from Conselice_2003, where A = 0.35. The other two lines in red and orange color separating the population of disk, intermediate, and elliptical galaxies have been taken from Bershady_2000. Exact values of the parameters for our galaxy (denoted as star symbol) are marked in both panels.
• Figure 3: In the top panel, we show the fitted model (as an orange curve), the model fluxes in each filter (as orange points) and the observed photometry with error bars (as blue points) for the delayed SFH model. Along the X-axis, red-tick marks indicate the positions, from left to right, of the Lyman break, the Balmer break, [OIII] + H-$\beta$, and H-$\alpha$ + [NII] line complexes. In the lower panels, we show the 1-d posteriors of output SPS parameters such as stellar mass, SFR, sSFR, age, formed mass (M$_{del}$), metallicity ($\log(Z_{del}/Z_\odot$), $\tau$, dust attenuation (A$_{v}$) and redshift (labeled on the x-axis) with three black vertical dashed lines indicating the values of the 16th, 50th and 84th percentiles of the output posterior probability distribution of each parameter. Age$_{MW}$ is the mass-weighted age of the galaxy, and Age$_{del}$ is the amount of time that has passed since the onset of star formation in the galaxy, calculated from the modeled delayed SFH.
• Figure 4: In the upper panel of this figure, we show the delayed exponential star formation history fitted by BAGPIPES to the observed photometry of the galaxy. In the lower panel, we show the corner plots for the output parameters.
• Figure 5: The best fit SED model from Prospector derived from UNCOVER DR4 SPS output files. Observed flux, flux uncertainties and model fluxes in all the filters are plotted. Flux uncertainties are taken from UNCOVER DR4 SPS data release, which have a 5% error floor imposed on the catalog flux uncertainties due to potential calibration uncertainties with JWST photometry. Wavelength is plotted in log scale in units of Ångstrom.