Abstract
High-redshift low-mass galaxies are believed to be the building blocks of
present-day galaxies. Their sheer prevalence, compared to their more massive
counterparts, makes them the most representative examples of the first
generation of galaxies. Furthermore, they likely played a pivotal role in
cosmic reionization between redshifts of $z=9$ and $z=6$. However, this
population has continued to elude comprehensive spectroscopic studies. As a
consequence, their role in comic reionization has remained unclear, owing to
the uncertainties surrounding their photometric redshifts and the lack of
constraints on both their ionizing photon production and escape fraction. Here
we report the first spectroscopic analysis of 8 ultra-faint galaxies during the
epoch of reionization with absolute magnitudes between M$_UV -17$ to
$-15$ mag (down to 0.005 $L^\star$). The combination of ultra-deep NIRSpec
(Near-Infrared Spectrograph) observations and the strong gravitational lensing
of Abell~2744 allows us to explore an uncharted territory of early galaxy
formation. Remarkably, some of these galaxies have stellar masses as low as
$7.6_-1.4^+1.7 10^5$ \msol, comparable to dwarf galaxies in the
local group, and extremely low metallicities of only 1 to 6\% $Z_ødot$. This
sample allows us to derive the first combined spectroscopic constraints on both
the prevalence of faint galaxies and their ionizing properties during the
Universe's first billion years. We find that faint galaxies have an ionizing
efficiency of log($\xi_ion$/ Hz erg$^-1$)=$25.80.05$, about a
factor of 4 larger than canonical values. This means that the total ionizing
photon budget produced by galaxies exceeds the reionization threshold, even for
modest values of $f_esc$ =5%. These findings provide robust evidence that
faint galaxies were the main drivers of cosmic reionization at $z\sim7$.
Users
Please
log in to take part in the discussion (add own reviews or comments).