A spectroscopically confirmed, strongly lensed, metal-poor Type II supernova at z = 5.13

TL;DR

This work reports the first spectroscopically confirmed, strongly lensed Type II supernova at $z=5.133$, SN Eos, discovered in JWST imaging of a galaxy cluster and identified as a multiply imaged transient aided by lensing magnification. JWST/NIRSpec spectroscopy, complemented by archival HST and VLT data, reveals a metal-poor host environment with Fe II features consistent with $Z\lesssim 0.1 Z_\odot$, while HST UV detections constrain the explosion epoch. Hydrodynamic modeling with low-metallicity red supergiant progenitors and a modest, confined CSM reproduces the UV and optical light curves, suggesting a $\sim16 M_\odot$ progenitor with a $\sim0.2 M_\odot$ CSM and an explosion energy of about $1.5\times10^{51}$ erg. The discovery demonstrates the power of strong lensing to access the faint end of the high-redshift galaxy population and to place direct constraints on early stellar physics and chemical enrichment, including the use of SNe II as metallicity probes at $z>5$.

Abstract

Observing supernovae (SNe) in the early Universe (z > 3) provides a window into how both galaxies and individual stars have evolved over cosmic time, yet a detailed study of high-redshift stars and SNe has remained difficult due to their extreme distances and cosmological redshifting. To overcome the former, searches for gravitationally lensed sources allow for the discovery of magnified SNe that appear as multiple images - further providing the opportunity for efficient follow-up. Here we present the discovery of "SN Eos": a strongly lensed, multiply-imaged, SN II at a spectroscopic redshift of z = 5.133 +/- 0.001. SN Eos exploded in a Lyman-α emitting galaxy when the Universe was only ~1 billion years old, shortly after it reionized and became transparent to ultraviolet radiation. A year prior to our discovery in JWST data, archival HST imaging of SN Eos reveals rest-frame far ultraviolet (~1,300Å) emission, indicative of shock breakout or interaction with circumstellar material in the first few (rest-frame) days after explosion. The JWST spectroscopy of SN Eos, now the farthest spectroscopically confirmed SN ever discovered, shows that SN Eos's progenitor star likely formed in a metal-poor environment (<= 0.1 Z_{\odot}), providing the first direct evidence of massive star formation in the metal-poor, early Universe. SN Eos would not have been detectable without the extreme lensing magnification of the system, highlighting the potential of such discoveries to eventually place constraints on the faint end of the cosmic star-formation rate density in the very early Universe.

Figures (6)

• Figure 1: JWST discovery image of the MACS 1931.8-2635 galaxy cluster containing SN Eos. The RGB color channels are as follows: blue [F115W + F150W], green [F200W + F277W], and red [F356W + F444W]. The magenta shaded area corresponds to a region where the model-predicted magnification is $\mu > 100$ at $z=5.133\pm0.001$. The upper left inset shows the two detected images of SN Eos, 101.1 & 101.2, mirrored by the critical curve, which represents the thin region of infinite magnification in the lens system for the source redshift. The proximity of these images to the critical curve results in a high magnification, predicted to be $\sim30$ at the image 101.1/101.2 positions. In the main image (outside the inset), we display the predicted positions and time delays, $\tau$, of images 101.3, 101.4, and 101.5. The angular scales are labeled on the N–E orientation arrows.
• Figure 2: Comparison of NIRSec PRISM spectra taken of SN Eos (top-left) to local SNe IIP (bottom-left). Top Left: In red and blue, spectra of individual lensed images of SN Eos are overlaid, demonstrating that each lens image shows the same SN at a similar phase. The inverse-variance weighted combined spectrum is shown in black and used in comparisons below. Bottom Left: Comparisons of our combined spectrum to the optical spectra of local SNe IIP, normalized to their continua, convolved and resampled with the JWST NIRSpec PRISM dispersion function. Spectra overlaid in blue are a sample of known, well-studied SNe II with metallicities consistent with $Z \lesssim0.1~Z_{\odot}$: SN 2017ivv Gutierrez2020_2017ivv, SN 2023ufx tucker_ufx_2024Ravi2025_ufx, and SN 2015bs anderson_2015bs_2018. The overlaid spectrum in red corresponds to the best-matching local SN II, SN 1992H at $\sim1~Z_{\odot}$ (see Sec. \ref{['sec:spectral_diffs']} for discussion of matching procedure). These comparisons demonstrate that SNe IIP display many similarities despite having a range of metallicities, and even for SNe IIP with similar metallicities there is still an intrinsic diversity in spectral features. This is consistent with findings from Anderson2016_SNeIIP_metallicity_probes which claim that metallicity alone is not strongly correlated with all spectral observables. Right: A detailed view of the Fe II complex for the same SNe, with SN Eos in gray. Each local SN has its native spectral resolution (purple and orange) and NIRSpec PRISM-convolved resolution (blue and red) overplotted onto SN Eos.
• Figure 3: VLT/MUSE observations of the SN Eos host galaxy. Left: the moment-zero map centered on the narrow wavelength range of $\lambda_{\rm obs}=7452$ Å to 7458 Å, generated from MUSE data cube. Multiple images of the system (image 101.1, 101.2, 101.3, and 101.4) are detected. Right: VLT/MUSE spectra extracted at the positions of multiple images. The Ly$\alpha$ lines are detected at $\lambda_{\rm obs}=7455$ Å in all images, with a measured redshift of $z= 5.133 \pm 0.001$. The best-fit single Gaussian profiles are shown in red.
• Figure 4: The evolution of pEW Fe II $\lambda5018$ Å measured in local SNe IIP as a function of SN phase. In teal are SNe data from Anderson2016_SNeIIP_metallicity_probes with errors, and overlaid in red are spectral models as a function of metallicity from Dessart2013_SNeIIP_properties. SN Eos is shown as a purple square, with SN 2023ufx as orange squares, with the inverted triangle as a pEW upper limit tucker_ufx_2024. A single measurement for SN 2015bs is plotted as a blue square anderson_2015bs_2018. Our measurement, along with the inferred phase, places SN Eos in a likely parameter space that is $Z\lesssim0.1\ Z_\odot$.
• Figure 5: Right: HST and JWST photometry for lens image 101.2, the higher signal-to-noise image, in case of a phase difference between 101.1 and 101.2 (see Table \ref{['tab:detections']}). Overplotted are in-band light curves from the modeling in Section \ref{['sec:eos_modeling']}, placing our JWST observations at the end of an SN IIP's plateau. The HST data were obtained as $4-5$ epochs over an observer-frame week, and are assumed to be near-explosion. The gray region marks the assumed window for $t_{0}$, the time of explosion, that we constrain to $\lesssim5$ rest-frame days from the first HST detection (see main text). Left: A zoomed-in view of the same HST F110W and F814W photometry, showing the variability of SN Eos on a rest-frame $\sim1$ day timescale. The model prediction, containing a moderate amount of CSM material, is in rough agreement with the later luminosity and evolution of this UV emission, however, we note that the first $2$ epochs in F814W and F1101W are in excess of this simple CSM model suggesting that this flux could be due to more complicated effects, e.g., asymetric structure in the confined CSM (see Section \ref{['sec:modeling']}).