UNCOVER/MegaScience Finds Uniform and Highly Bursty Star Formation at 3 < z < 9, consistent with the High-Redshift UV Luminosity Function

TL;DR

This work uses UNCOVER/MegaScience JWST/NIRCam 20-band photometry to empirically measure line-to-UV ratios and line EWs for H$\alpha$+[N II] and [O III]+H$\beta$ across $z\sim3$--$9$, enabling population-level inferences about star formation histories without strong priors. The authors construct mass- and flux-complete samples, perform photometric line/continuum measurements, and compare results to FSPS-based toy SFH models, revealing little evolution in the intrinsic scatter of $R_{\rm line}$ with redshift up to $z\sim7$ and favoring rising, long-duration, high-amplitude bursts. No-burst SFHs fail to reproduce the observed distributions, indicating burstiness is essential to the population. The study shows that bursty SFHs can boost UV luminosities by up to about 2 magnitudes at fixed mass, potentially explaining the abundance of UV-bright galaxies at $z>10$ without requiring an evolution in burstiness; this has substantial implications for interpreting the high-redshift UV luminosity function and the detectability of early galaxies.

Abstract

Star formation timescales are key to understanding fundamental physics like feedback mechanisms, as well as the abundance of bright galaxies at $z>10$. We investigate galaxy star formation histories (SFHs) and their evolution across $z\sim3$--9 by measuring the line-to-UV ratio (\rline) and line equivalent width (EW) of \hanii\ and \oiiihb\ directly from UNCOVER/MegaScience spectro-photometry without relying on a specific SFH or nebular line modeling. Our photometric measurements recover \rline\ and EW to $<10\%$ systematic accuracy compared to spectroscopy. This allows us to construct a large mass- (and flux-) complete sample and quantitatively examine how \rline\ evolves with redshift and stellar mass. We find that the intrinsic scatter in \rline\ does not significantly evolve with redshift across $3<z<7$, though it may increase at $z\gtrsim8$. We build population-level toy models using \texttt{fsps} to help interpret our observations, and find that scatter in \rline\ primarily reflects the amplitude of SFH fluctuations; this implies that our observed lack of evolution in the scatter of \rline\ is due to similar star formation burstiness from $z\sim3$ to $z\sim7$. Our observations are best reproduced by a set of SFHs with rising, long-duration, and large-amplitude bursts. Finally, we demonstrate that the toy model that best describes our $z\sim6$ data can boost UV brightness by up to $ΔM_{\rm UV}\sim-2.0\,{\rm mag}$ compared with a 200\,Myr constant SFH, and naturally produces a large number of galaxies at $z>10$. This suggests that no significant evolution in star formation burstiness is required to explain the abundance of UV-bright galaxies at high redshift.

Figures (21)

• Figure 1: (left) Sample selection for the mass-complete sample. The redshift gaps we avoided in the sample selection due to wavelength coverage are filled in gray. The background gray hexbins indicate all galaxies in the UNCOVER/MegaScience DR3 catalog. Our mass-limited samples satisfying $\log M_{\ast}[M_{\odot}]>8.0$ and some additional criteria (see text) at $z\sim3$, 5, 6, and 8 are shown in blue, green, orange, and red, respectively. The different distributions between the selected sample and the parent UNCOVER DR3 sample arise from the criteria for the HST/medium band coverage or known AGN contribution. (right) The relationship between the mass and rest-frame $\sim4000$Å flux. Galaxies included in both mass-complete samples are shown in filled markers. The background gray hexbins indicate galaxies in the UNCOVER/MegaScience DR3 catalog at $z>2.5$. Typical $5\sigma$ detection limits in the filters tracing $m_{\rm opt}$, depending on the sample redshift (see text), are shown in the dashed lines with corresponding colors. The flux-complete sample is constructed from a threshold sufficiently smaller than $5\sigma$ detection limit in the UNCOVER survey (28 mag). The mass-complete sample is mostly included in the flux-complete sample, enabling us to check whether the mass estimation from Prospector-$\beta$ affects the results significantly.
• Figure 2: Examples of our emcee fitting results for significant (left) and insignificant (right) line emitters. Best-fit power-law(+Gaussian) model and its uncertainties are shown in gray lines and shaded regions, respectively. Blue and red wavelength ranges indicate the rest-frame UV and optical wavelengths specified in the fitting. Observed photometries in broad+F410M and other medium bands are displayed in gray and black circles, respectively. Gray diamonds show the best-fit photometry from the 50th percentile of the likelihood, overlaid in purple and brown for filters containing H$\alpha$+[N ii] and [O iii]+H$\beta$, respectively. The inset panels at the bottom right in each panel show the posterior distribution of the H$\alpha$+[N ii] (brown) and [O iii]+H$\beta$ (purple) line amplitude in $\mu{\rm Jy}$ unit. The left top panels exhibit $2"\times2"$ thumbnails of the target galaxies.
• Figure 3: Comparison between photometric and spectroscopic measurement of EWs (top) and $R_{\rm line}$ (bottom). H$\alpha$+[N ii] and [O iii]+H$\beta$ measurements are shown in the left and right panels, respectively. Different colors and markers correspond to the redshift samples (Section \ref{['subsec:sample']}). Inserted panels indicate the normalized deviation between NIRCam and NIRSpec measurements, as written by $\Delta=({\rm NIRCam}-{\rm NIRSpec})/{\rm NIRSpec}$. Our photometric measurements agree well with the spectroscopic measurements.
• Figure 4: $\log\!R_{{\rm H}\alpha}$ (top panels) and $\log\!R_{\rm [OIII]}$ (bottom panels) as a function of stellar mass. From left to right, we show the $z\sim3$, 5, 6, and 8 subsamples. Small markers indicate individual galaxies, and large markers with black edges show the 16-50-84th percentiles of the stellar mass bins of $\log M_{\ast}[M_{\odot}]=8.0$--8.5 and 8.5--9.5. Gray shaded regions in the top panels and bottom panels show the range of $R_{{\rm H}\alpha}$ equilibilium value (1/85--1/60) and the range multiplied by two (see text), respectively. Here, we use the median value of the 16th-50th-84th percentile values among 500 posterior distributions for individual galaxies. There is no significant mass dependence of the scatter at the mass range of $\log M_{\ast}[M_{\odot}]>8.0$. $R_{{\rm H}\alpha}$ and $R_{{\rm [OIII]}}$ exhibit similar trends specifically at $z>4$, suggesting usefulness of the [O iii]+H$\beta$ lines as a tracer of the burstiness.
• Figure 5: Distribution of the $R_{{\rm H}\alpha}$ (left panel) and $R_{{\rm [OIII]}}$ (right panel) of the mass-complete samples as a function of redshift. The left and right sides of the representative redshift exhibit the Bayesian hierarchical model of low-$M_{\ast}$, high-$M_{\ast}$, respectively. The dashed lines on each side correspond to the observed distribution used for the Bayesian modeling.