The role of mergers and rejuvenation in the buildup of the quiescent population at cosmic noon

We investigate the quenching of galaxies using a mock observational lightcone generated from the Semi-Analytic Model (SAM) L-Galaxies, closely matched to observations from the UKIDSS Ultra Deep Survey (UDS). The sample is used to study merging, rejuvenation, and visibility times for star-forming, quiescent, and post-starburst (PSB) galaxies, to assess the impact on the build-up of the passive galaxy mass functions. We find, for example, that a typical PSB ($M_\ast\sim10^{10}$,M$\odot$) at $z\approx1$ has a 15 per cent likelihood of merging and around a 25 per cent likelihood of rejuvenating within 1 Gyr of being identified. Applying these rates and timescales to the observational data, we estimate the fraction of quiescent galaxies that passed through a PSB phase. We find that $18 - 28$ per cent of the build-up in the massive end ($M\ast>10^{10}$,M$,\odot$) of the passive mass function at $1<z<2$ can be explained by PSBs, with the contribution declining to $\sim5$ per cent by $z \simeq 0.5$. Accounting for mergers and rejuvenation reduces the inferred PSB contribution by approximately a factor of two. At lower stellar masses ($M\ast < 10^{10}$,M$_\odot$), rapid quenching through a PSB phase explains a significantly larger fraction of the growth in the passive mass function. With a visibility time of $\sim$ 0.75 Gyr, we find that around $60-80$ per cent of low-mass passive galaxies underwent a PSB phase. Our findings provide further evidence that low- and high-mass galaxies follow different quenching pathways.

💡 Research Summary

This study investigates how galaxies transition from star‑forming to quiescent during the epoch known as “cosmic noon” (z ≈ 1–2) by combining a sophisticated semi‑analytic model (SAM) with deep observational data. The authors generate a mock light‑cone from the latest L‑Galaxies SAM (Henriques et al. 2015) that reproduces the UKIDSS Ultra‑Deep Survey (UDS) selection functions, photometric depths, and redshift coverage (0.5 < z < 3). Photometric redshifts are derived with EAzY using a suite of FSPS templates, achieving NMAD = 0.019 and an outlier fraction of ~3 %. Galaxies are classified into star‑forming (SF), passive, post‑starburst (PSB), and dusty categories using the Principal Component Analysis‑based “Super‑Colour” (SC) method, where SC1 traces mean stellar age and dust, SC2 traces recent (<1 Gyr) star formation, and SC3 helps break mass‑metallicity degeneracies.

The mock catalogue is then directly compared to the UDS stellar mass functions (SMFs) for each class. While the model reproduces the SF and passive SMFs at low redshift, it over‑predicts low‑mass PSBs and under‑predicts high‑mass SF galaxies at z > 2, indicating tensions in the treatment of rapid quenching at early times.

Crucially, the authors exploit the full merger trees of L‑Galaxies to follow every galaxy’s evolutionary history. They identify the moment a galaxy first satisfies the PSB SC criteria and track its fate for the next gigayear. Within this window they record (i) major mergers (mass ratio > 1:3) and (ii) rejuvenation events, defined as a rise in star‑formation rate above 0.1 M⊙ yr⁻¹ after a prior quenching episode. For a typical PSB of M* ≈ 10¹⁰ M⊙ at z ≈ 1, the probability of experiencing a merger within 1 Gyr is 15 %, while the probability of rejuvenation is 25 %. These probabilities increase toward lower stellar masses and lower redshifts.

The authors also estimate the “visibility time” of PSBs – the interval during which a galaxy remains within the SC‑defined PSB region. Using the mock, they find an average visibility of 0.75 Gyr (±0.1 Gyr). By combining this timescale with the observed number densities of PSBs, they infer the fraction of quiescent galaxies that have passed through a PSB phase. Between 1 < z < 2, PSBs can account for 18–28 % of the growth of the massive (M* > 10¹⁰ M⊙) passive population; this contribution falls to ~5 % by z ≈ 0.5. When the merger and rejuvenation probabilities are folded in, the effective PSB contribution drops by roughly a factor of two, to about 10–15 % at high mass.

In contrast, at lower masses (M* < 10¹⁰ M⊙) the PSB pathway dominates. The analysis suggests that 60–80 % of low‑mass passive galaxies have experienced a PSB phase, implying that rapid quenching is the primary channel for building up the low‑mass quiescent population.

The paper also discusses the treatment of “orphan” galaxies—systems whose dark‑matter subhaloes have fallen below the detection threshold. L‑Galaxies strips the hot gas of orphans and assigns them a dynamical merger timescale, causing them to become quiescent rapidly. This mechanism helps reproduce the observed passive fraction at moderate masses but may contribute to the excess of low‑mass passive galaxies seen in the mock.

Overall, the study provides a quantitative framework linking PSB statistics, merger/rejuvenation rates, and the evolution of the quiescent stellar mass function. It demonstrates a mass‑dependent quenching picture: massive galaxies tend to quench gradually, with only a modest PSB contribution that is further diluted by subsequent mergers or rejuvenation; low‑mass galaxies, however, quench abruptly via a PSB phase, with little subsequent alteration. The authors highlight that upcoming facilities such as JWST/NIRSpec, Euclid, and the Rubin Observatory will enable direct spectroscopic measurements of PSB lifetimes and merger/rejuvenation signatures, offering a path to refine SAM prescriptions—particularly the handling of orphans and environmental quenching—in order to reconcile remaining discrepancies at high redshift.