JWST and Lensing Reveal Abundant Ultra-Faint Galaxies, Bolstering the Case They Powered Reionization

A new arXiv preprint reports that the number of ultra-faint galaxies seen by the James Webb Space Telescope continues to rise at redshifts 6 to 9 — when the universe was still in its first billion years — with no clear sign of a cutoff down to ultraviolet magnitudes of about minus 12, a result that strengthens the case that dim galaxies helped keep cosmic reionization going.

In the paper, posted April 26 as arXiv:2604.23823, lead author Hakim Atek and colleagues say the galaxy ultraviolet luminosity function at about redshift 7 has a faint-end slope of alpha equals minus 1.98, with an uncertainty of plus 0.06 and minus 0.05. They report “no clear evidence of a turnover” down to MUV = -12.3, where MUV is a standard measure of a galaxy’s ultraviolet brightness. The team also estimates a comoving ionizing emissivity at redshift 7 of about log(nion / s^-1 Mpc^-3) = 50.85, which they interpret as enough for faint galaxies to dominate the ionizing photon budget and maintain reionization even if intergalactic gas was fairly clumpy, using a clumping factor of C_HII = 5.

That matters because cosmic reionization was the period when ultraviolet light from the first galaxies reionized hydrogen spread through intergalactic space. A central question has been whether that job was done mainly by huge numbers of very faint galaxies, rather than by the much rarer bright galaxies or by other sources. The new work does not prove the case, but it pushes direct measurements deeper into the faint population and suggests those small systems could have supplied enough photons if the result holds up.

The study uses JWST observations from the GLIMPSE survey together with gravitational lensing, the natural magnification produced by a massive foreground galaxy cluster. In this case, the cluster is Abell S1063, at redshift 0.348, which bends and brightens the light from much more distant background galaxies. GLIMPSE — JWST program PID 3293, led by Atek and John Chisholm — uses that magnification to probe galaxies too faint to study directly in blank fields. The authors say they reach MUV = -12 magnitudes, about three magnitudes deeper than previous robust constraints.

The result is especially important for galaxy formation theory. Many models predict that star formation should flatten or shut down in the smallest dark matter halos, which would show up as a turnover in the ultraviolet luminosity function at faint magnitudes. Instead, the paper says its detections “effectively rule out a luminosity function truncation at MUV ≥ -15.” The authors also argue that several post-JWST models adjusted to explain the apparent excess of very bright galaxies at redshifts above 10 generally do not reproduce the observed evolution toward lower redshifts and much fainter magnitudes.

The claim lands in a part of the field with a complicated history. Hubble Frontier Fields studies in the late 2010s also argued for abundant very faint galaxies, but those results were heavily debated because lensing systematics were hard to control. The authors say their new ultraviolet luminosity function estimate includes a “rigorous end-to-end uncertainty framework,” folding in both statistical and systematic lensing uncertainties.

Even so, the caveats are substantial. The result still depends on lensing magnification models, the source-plane area is small enough that cosmic variance could matter, and the faintest photometric candidates still need more spectroscopic confirmation. Those limits mean the paper is best read as a strong new indication, not a final verdict, that faint galaxies were numerous enough to sustain reionization. The study, “A GLIMPSE of the 99%: a census of the faintest galaxies during the epoch reionization and its implications for galaxy formation models,” is a preprint posted to arXiv and has not yet been peer-reviewed.