📝 SummarXiv

Abstract

We present new ALMA [OIII]$_{88}$ observations of eight previously [CII]$_{158}$-detected galaxies at $6.8 \lesssim z \lesssim 7.7$. Six of our targets -- the primary sample -- are massive, UV-luminous galaxies drawn from the REBELS survey, while the remaining two are UV-fainter galaxies that were previously serendipitously detected through their luminous [CII] lines in the REBELS fields. We detect [OIII]$_{88}$ emission in all eight galaxies at $6.2 - 17.7\sigma$ significance, and find them to be consistent with the local dwarf galaxy relation between $L_\mathrm{[OIII]}$ and star formation rate. Our sample spans [OIII]/[CII] $\approx 1.9 - 9.6$, which is typical for the high-redshift galaxy population. Five of the primary targets benefit from JWST/NIRSpec observations, enabling a direct comparison of the [OIII]/[CII] ratio against rest-optical ISM diagnostics. We supplement our high-redshift sample with eleven $z\approx6-14$ galaxies in the literature for which similar ALMA and JWST observations are available, and furthermore compare to the [OIII]/[CII] ratios measured for local dwarf galaxies. We find that, at fixed metallicity and ionization parameter, $z>6$ galaxies show elevated [OIII]/[CII] ratios compared to local dwarfs. Instead, we find that a large [OIII]$_{4959,5007}$+H$\beta$ equivalent width -- a proxy for burstiness -- is the main driver of the high [OIII]/[CII] ratios seen in the early Universe, which is primarily due to [CII] being suppressed in bursty galaxies. Given the apparent validity of the [OIII]$_{88}$-SFR relation across most of cosmic time, as well as the abundance of young, bursty galaxies at high redshift, [OIII]$_{88}$ is set to remain a powerful ISM tracer at the cosmic dawn.