A submillimetre galaxy at z=4.76 in the LABOCA survey of the Extended Chandra Deep Field South

We report on the identification of the highest redshift submm-selected source currently known: LESSJ033229.4-275619. This source was detected in the Large Apex BOlometer CAmera (LABOCA) Extended Chandra Deep Field South (ECDFS) Submillimetre Survey (LESS), a sensitive 870-um survey (1.2-mJy rms) of the full 30’x30’ ECDFS with the LABOCA camera on the Atacama Pathfinder EXperiment (APEX) telescope. The submm emission is identified with a radio counterpart for which optical spectroscopy provides a redshift of z=4.76. We show that the bolometric emission is dominated by a starburst with a star formation rate of 1000 Msun/yr, although we also identify a moderate luminosity Active Galactic Nucleus (AGN) in this galaxy. Thus it has characteristics similar to those of z2 submm galaxies (SMGs), with a mix of starburst and obscured AGN signatures. This demonstrates that ultraluminous starburst activity is not just restricted to the hosts of the most luminous (and hence rare) QSOs at z5, but was also occurring in less extreme galaxies at a time when the Universe was less than 10% of its current age. Assuming that we are seeing the major phase of star formation in this galaxy, then we demonstrate that it would be identified as a luminous distant red galaxy at z~3 and that the current estimate of the space density of z>4 SMGs is only sufficient to produce 10% of the luminous red galaxy population at these early times. However, this leaves open the possibility that some of these galaxies formed through less intense, but more extended star formation events. If the progenitors of all of the luminous red galaxies at z3 go through an ultraluminous starburst at z>4 then the required volume density of z>4 SMGs will exceed that predicted by current galaxy formation models by more than an order of magnitude.

💡 Research Summary

The paper presents the discovery and multi‑wavelength characterization of LESS J033229.4‑275619, the highest‑redshift sub‑millimetre‑selected galaxy (SMG) known at the time of writing (z = 4.76). The source was identified in the LABOCA ECDFS Sub‑mm Survey (LESS), a 870‑µm map of the full 30′ × 30′ Extended Chandra Deep Field South with an rms of ≈1.2 mJy beam⁻¹. A robust radio counterpart at 1.4 GHz (≈ 70 µJy) was found within the LABOCA beam, allowing precise astrometry and subsequent optical/near‑IR spectroscopy with VLT/FORS2. The spectrum shows a strong Ly α emission line and several interstellar absorption features, yielding a secure redshift of z = 4.76.

The authors compile photometry from the radio (VLA 1.4 GHz), sub‑mm (LABOCA 870 µm), far‑IR (Spitzer/MIPS 24 µm, Herschel if available), mid‑IR (IRAC), and optical/NIR (HST/ACS, VLT/ISAAC). Fitting a modified black‑body (β = 1.5) to the far‑IR/sub‑mm points gives a dust temperature of ≈ 35 K and a total infrared luminosity L_IR(8–1000 µm) ≈ 6 × 10¹² L_⊙. Using the standard Kennicutt conversion (adjusted to a Chabrier IMF) this corresponds to a star‑formation rate (SFR) of ≈ 1000 M_⊙ yr⁻¹, placing the galaxy firmly in the ultraluminous infrared galaxy (ULIRG) regime.

X‑ray data from the 1 Ms Chandra exposure show only a marginal detection, consistent with a modestly obscured active galactic nucleus (AGN) of L_X ≈ 10⁴³ erg s⁻¹. Mid‑IR colours also hint at an AGN contribution, but the bulk of the bolometric output is clearly powered by star formation. The authors therefore classify the object as a “composite” SMG, analogous to the majority of z ≈ 2 SMGs that show both starburst and AGN signatures.

The paper discusses the implications for galaxy formation at early epochs. The inferred stellar mass (≈ 10¹¹ M_⊙) and SFR imply that, if the starburst lasts ≈ 100 Myr, the galaxy could assemble a massive stellar component comparable to the luminous distant red galaxies (DRGs) observed at z ≈ 3. The measured space density of z > 4 SMGs in the LESS field (≈ 10⁻⁶ Mpc⁻³) is only sufficient to account for ~10 % of the DRG population, suggesting either that many massive galaxies at z ≈ 3 formed via less extreme, more extended star‑formation episodes, or that current SMG surveys miss a substantial fraction of high‑z dusty systems.

Finally, the authors compare the observed number density of such high‑z SMGs with predictions from semi‑analytic models (e.g., Baugh et al. 2005, Somerville et al. 2008). The models under‑predict the abundance by more than an order of magnitude, indicating that either the physical prescriptions for early, rapid star formation and dust production need revision, or that observational selection effects (e.g., surface‑brightness dimming, radio identification bias) are more severe than assumed.

In summary, the discovery of a z = 4.76 SMG demonstrates that massive, dust‑enshrouded starbursts were already in place less than 1.3 Gyr after the Big Bang, with properties remarkably similar to the better‑studied SMG population at z ≈ 2. This pushes back the epoch of intense, obscured star formation and provides a crucial benchmark for models of early galaxy assembly, the buildup of stellar mass, and the co‑evolution of starbursts and AGN in the first few billion years of cosmic history.