Science

Galaxy formation in the first half of the epoch of reionization

Background

The James Webb Space Telescope (JWST) has found more than a hundred candidate galaxies at redshifts z > 8, when the Universe was less than about 600 million years old, and has confirmed some of them spectroscopically out to z ≈ 14. The bright end of their ultraviolet luminosity function evolves surprisingly little with redshift (Harikane et al. 2022), which means that massive galaxies had to grow very efficiently within a short time. This places a stringent constraint on models of cosmic structure formation.

To understand what drives this rapid growth, we need spectroscopy that answers two questions:

• How common are bright, massive galaxies in the early epoch of reionization (their number density)?
• What physical conditions control their growth (metallicity, ionization state and gas density)?

Far-infrared nebular emission lines

The far-infrared fine-structure line of doubly ionized oxygen, [O III] 88 µm, was proposed as a powerful tracer for identifying the redshifts of the most distant galaxies (Inoue et al. 2014). ALMA observations of this line revealed extremely high [O III]/[C II] ratios, indicating highly ionized gas filling these young galaxies (Inoue et al. 2016), and held the record for the most distant spectroscopic redshift obtained with ALMA until 2022 (Hashimoto et al. 2018). Even in the JWST era, [O III] 88 µm at z > 9 is as detectable as the rest-frame ultraviolet lines observed with JWST/NIRSpec.

However, ALMA in the southern hemisphere cannot easily observe high-declination fields. Many bright galaxies at z > 8 found in northern deep fields, such as GOODS-North and CEERS, remain unexplored in the far-infrared lines (Figure 1).

Figure 1. UV absolute magnitude versus spectroscopic redshift of galaxies at z > 6. Green and blue circles are galaxies with and without ALMA detections; red crosses are galaxies in northern fields. Adapted from Harikane et al. 2025, ApJ, 980, 138 (colour coding by M. Hagimoto).

Key science goals

With FINER on the 50-m LMT in Mexico (latitude 19°N), we aim to:

• identify the spectroscopic redshifts of bright galaxies at z ≈ 8–15 in the northern sky using [O III] 88 µm and [C II] 158 µm, including galaxies confirmed by JWST and photometric candidates from wide-field surveys such as Euclid and Roman;
• detect [O III] 88 µm from a bright Lyman break galaxy in a single night of observation;
• measure the number density of bright galaxies in the first half of the epoch of reionization and diagnose their interstellar medium.

How FINER helps

A typical photometric redshift uncertainty of Δz ≈ 1 at z ≈ 10 requires a spectral coverage of more than 30 GHz to find the line. FINER covers about 38 GHz in a single tuning, whereas the current ALMA needs four or more tunings for the same range (Figure 2). Combined with 40% of ALMA’s collecting area and a high-altitude site (4,600 m), this offers the most sensitive spectral-scanning capability among millimeter and submillimeter telescopes in the northern hemisphere (Figure 3).

Figure 2. The [O III] 88 µm line of the z = 8.31 galaxy MACS0416_Y1 was found after four ALMA tunings (Tamura et al. 2019). The orange bars show the frequency range FINER covers in a single tuning (19 GHz × 2 sidebands).

Figure 3. Expected [O III] 88 µm and [C II] 158 µm flux densities of a bright Lyman break galaxy (M_UV = −21.5) at z = 8–15, compared with the minimum 5σ-detectable flux density of LMT-FINER in 10 hours (gray). The bars in the inset compare the instantaneous IF bandwidths of FINER and the current ALMA (Tamura et al. 2024).

Further science cases

FINER also offers unique capabilities for a broad range of science, including:

• the Solar System (planetary atmospheres and comets),
• the interstellar medium and astrochemistry,
• star formation,
• nearby galaxies,
• distant dusty star-forming galaxies, including spectroscopic follow-up of galaxies detected with TolTEC.

See the FINER proposal (PDF) for details.