Expanding the High-z Supernova Frontier: "Wide-Area" JWST Discoveries from the First Two Years of COSMOS-Web

TL;DR

This study demonstrates JWST's ability to push supernova discoveries into the high-redshift regime by exploiting the wide-area overlap of COSMOS-Web and PRIMER, yielding 68 SNe up to $z<5$ in ~133 arcmin$^2$ with predominantly single-epoch data. It combines robust image-differencing, forced photometry, host redshift estimation, and single-epoch Bayesian classification (STARDUST2) to prioritize follow-up targets, exemplified by SN 2023aeab (Type IIP at $z\approx3.37$) and SN 2023aeax (Type Ia at $z\approx2.13$). Targeted JWST follow-up with NIRCam imaging and NIRSpec spectroscopy (DDT PID 6585) anchors redshifts and refines classifications, enabling early-universe SN population studies and rate measurements. The results highlight that wide-area, moderately deep JWST surveys can uncover younger, bluer, and potentially more extreme explosions than deeper, narrower surveys, and provide a framework for optimizing future JWST time-domain programs and their synergy with Roman. They also emphasize methodological requirements (e.g., photo-$z$ for $z>1$, multi-epoch coverage, and representative training data) to maximize the scientific return of high-$z$ transient science.

Abstract

Transient astronomy in the early Universe (z > 2) remains largely unexplored, lying beyond the rest-frame optical spectroscopic reach of most current observatories. Yet this regime promises transformative insights, with high-redshift transients providing direct access to the early Universe and enabling studies of how stellar populations and cosmology evolve over cosmic time. JWST is uniquely equipped to probe these redshifts efficiently in the rest-frame optical and near-IR. We present results from an initial pathfinder search, covering an area of ~133 arcmin^2 (~0.037 deg^2) independently imaged by the PRIMER and COSMOS-Web (hereafter COSMOS) extragalactic surveys. Although neither program was designed for time-domain astronomy, combining their data results in difference images separated by roughly one year, leading to the discovery of 68 supernovae (SNe) with host photometric redshifts reaching z < 5. For most SNe, only a single epoch is available, but the combination of host redshift, classification, color, and magnitude enables us to prioritize candidates for detailed photometric and spectroscopic follow-up. Among the most notable sources are a relatively bright, blue CCSN at z > 3 (SN 2023aeab) and a young, normal SN Ia at z > 2 (SN 2023aeax). The sample distribution highlights the increasing likelihood that a wide-area JWST program can uncover younger, bluer, and potentially more extreme explosions. While this pathfinder effort is limited in cadence and number of filters, it demonstrates the strong potential of a dedicated, well-planned time-domain survey with JWST to obtain the sample sizes and rate measurements needed to chart SN populations deep into the early Universe.

Figures (10)

• Figure 1: Total expected number of SNe derived from simulations of a corresponding 5-year survey: ( left) Type Ia SNe at redshifts $z>3$; ( right) Pop-III PISNe. For simplicity, the simulated surveys all have a 6-month cadence in four JWST/NIRcam bands (F115W, F150W, F277W, and F444W), all with the same limiting magnitude per epoch. The SN yields correspond to the number of SNe detected in at least one band at any phase. The red line traces the area and magnitude limits possible with 200 hours on JWST (including overheads). Increasing depth has limited benefits below about 28th magnitude, but the survey yields increase proportionally with survey area. The area and depths of several Cycle 1 JWST surveys are marked for comparison, but note these surveys are not uniform and all have different durations, filters, and cadences from the simulations. The open star marks the overlapping area of COSMOS-Web and PRIMER used in this paper (see Figure \ref{['fig:cosmosprimer']}).
• Figure 2: JWST/NIRCam observations of the COSMOS-Web and PRIMER surveys as gray- and blue-shaded areas, respectively, with roughly $\sim$0.037 deg$^2$ of overlapping area. Grey squares and cyan circles indicate the positions of transients presented in this study with redshifts of $z<1.5$ and $z>1.5$, respectively. Over-plotted are footprints of NIRSpec/MSA (red square) and NIRCam (black square) from follow-up observations obtained via a Director's Discretionary Time Proposal (PID 6585; PI Coulter), described in detail in Section \ref{['subsec:follow_up']}. NIRSpec MSA slitlets used in that program are shown with small black lines. SNe for which MSA spectra were obtained are outlined in yellow.
• Figure 3: Cutouts for SN 2023aeab and SN 2023aeax (SN#1 and SN#4 in Table \ref{['tab:targets']}, respectively), both discussed throughout the text. Shown are the ( top) discovery epoch from COSMOS-Web, ( middle) template from PRIMER, and ( bottom) difference image.
• Figure 4: Magnitude and photometric redshift distribution for the COSMOS sample of SNe (Table \ref{['tab:cosmos_classifications']}), sorted by classification. Classifications assigned by STARDUST2 for targets with at least two filters in a single epoch or two epochs (but see limitations in Section \ref{['sec:classification']}). The JADES sample decoursey25a is plotted for comparison (grey circles). Brightest magnitudes are chosen for plotting purposes because rest-frame photometric bands vary with redshift. However, there is no single standard rest-frame filter that we can use across all redshifts for all SNe. Shaded regions correspond to approximate sensitivity and redshift range for surveys on other telescopes, including HST CANDELS grogin11koekemoer11, CLASH postman12 and Roman's High-Latitude Time Domain Survey jha25.
• Figure 5: Color-magnitude diagram of theoretical models of SNe Ia (blue), SNe II (purple), and PISNe (red) in four redshift bins (where redshifts are photometric redshifts in Table \ref{['tab:cosmos_classifications']}). Type Ia and CCSN models are described in Section \ref{['sec:classification']}, while we use the 250 solar mass Red Supergiant (RSG) PISN models taken from kasen11. Models range in epoch and redshift, although they are limited to one observer-frame year and no extinction. The shaded tiers for each color correspond to 68% and 95% of the models for each subclass, but do not provide any information on likelihood of each model. These contours could be loosely interpreted as 1- and 2-$\sigma$ values, although the distributions are not gaussian (see accompanying histograms). Overplotted are empirical results from COSMOS/PRIMER (black squares; this work) and JADES decoursey25a. Histograms to the top and right of each contour provide absolute numbers. Note that the PISN histogram is multiplied by a factor of 100 for visualization purposes.