Three hypotheses can explain gas-rich quenched galaxies
Observations have found quenched galaxies that still hold substantial molecular gas reservoirs, which conflicts with the traditional picture of gas-poor, red and dead galaxies. This paper uses the Simba cosmological simulation to test three possible origins for those systems: quiescence is temporary and the galaxy will rejuvenate, molecular gas masses are overestimated because the CO-to-H2 conversion factor is underestimated, or star formation is ongoing but heavily obscured by dust.
The study examines galaxies with stellar mass above 3×10^9 solar masses from redshift z=0 to z=2. Within that sample, the authors report that all three mechanisms are plausible rather than selecting a single dominant cause. That framing matters because each hypothesis implies a different physical state for a galaxy that looks quenched: a temporary pause, a measurement problem, or hidden star formation.
Dust can hide ongoing star formation in quenched systems
Many galaxies that appear quiescent and molecular gas-rich still host dust-obscured star formation. In the simulation, visual optical depths increase on average from τ_V ∼ 0.5 to 2.3 between z=0 and z=2. This means that at higher redshift, dust can more effectively hide the emission that would otherwise reveal active star formation.
For readers comparing galaxy classifications, this result warns that a quenched label based on optical or rest-frame ultraviolet light may not mean star formation has fully stopped. The paper's own interpretation is that dusty cores with obscured star formation, surrounded by dusty outskirts, can make a galaxy's photometry resemble a post-starburst system even when star formation continues.
A Milky Way CO conversion factor can overestimate gas
Assuming a Milky Way-like CO-to-H2 conversion factor can overestimate molecular gas masses by a factor of about 5 in non-star-forming galaxies, compared with about 1.45 in star-forming galaxies. The contrast is the key number: the same conversion assumption produces a much larger error in quenched systems than in star-forming ones.
This result does not mean the observed molecular gas reservoirs are entirely spurious. It means that in non-star-forming galaxies, the inferred gas mass depends strongly on a conversion factor calibrated for a different physical regime. The paper treats this as one of three plausible mechanisms, not as a universal correction that eliminates the reservoirs.
Temporary quiescence lasts 100 Myr to 2 Gyr
Temporary quiescence can occur in gas-rich transitioning galaxies on timescales ranging from 100 Myr to 2 Gyr. The paper defines a rejuvenation candidate as a galaxy with stellar mass above 3×10^9 solar masses that is transitioning or quenched for more than one snapshot, and then rises back above the star-forming threshold for multiple snapshots.
This mechanism is most prominent at high redshift and is relatively rare overall. The authors also report that the fraction of massive transitioning or quenched galaxies that will rejuvenate is highest at high redshift and decreases monotonically toward z=0, while the median number of rejuvenation events rises toward the present day before turning over near z∼0.4. The turnover at low redshift is partly because galaxies identified in the last few gigayears have not had enough time to complete a roughly 2 Gyr rejuvenation process.
Rejuvenating galaxies are gas-rich and transitioning
Galaxies that rejuvenate are significantly more star-forming and gas-rich than those that do not, across z=0 to z=3.5. The majority of rejuvenating galaxies have a molecular gas-to-stellar mass ratio above 0.1, and they are predominantly transitioning rather than fully quenched, while those that will not rejuvenate are primarily quenched.
The paper links rejuvenation primarily to galaxy interactions such as mergers, which are most common for massive galaxies near z∼3. An example galaxy selected at z≈1.5 shows gradual evolution at high redshift followed by a dramatic increase in molecular gas mass and stellar mass around z≈0.6, driven by a major merging event. This mechanism is therefore tied to a specific physical trigger, not simply to the presence of gas.
Simba cannot reproduce bursty star formation
The simulation's subgrid treatment of the interstellar medium imposes a resolution-dependent temperature floor to prevent artificial fragmentation below the mass resolution. This effective equation of state stabilizes dense gas against collapse on unresolved scales and smooths star-formation variability in time, so Simba is not designed to include bursty star formation.
That limitation matters for interpreting the post-starburst comparison. The paper reports that SQuiGGLE galaxies have significantly stronger Balmer-line absorption than Simba galaxies, which is consistent with SQuiGGLE being more burst-dominated. The inability to produce bursty star formation is a property of the simulation method, and it qualifies how directly the simulated quenched population maps onto observed post-starburst systems.
Photometric selection can miss dusty impostor systems
The paper's UBV-selected post-starburst-like sample at z=0.5 includes systems that are also star-forming, with median gas depletion timescales of 798 Myr, while those that are also quenched have significantly longer median depletion timescales of about 2 Gyr. The authors connect the dust-rich UBV-selected systems to the category of impostor post-starburst galaxies, meaning photometrically selected systems that are actually star-forming.
The practical implication is that photometric selection alone can include systems with dust-obscured ongoing star formation. The paper argues that additional selection criteria focused on the current level of star formation are needed to avoid such impostors. This is a statement about sample definition, not about any single galaxy's fate.
The three mechanisms still need to be disentangled
The paper's central conclusion is that temporary quiescence, CO conversion factor overestimation, and dust-obscured star formation are all plausible contributors to molecular gas-rich quenched systems across z=0 to z=2. The authors state that further work is needed to disentangle their relative contributions across cosmic time.
For readers assessing a reported molecular gas reservoir in a quenched massive galaxy, the most decision-relevant split is therefore not whether the galaxy is quenched, but which of the three mechanisms is operating. A high inferred gas mass in a non-star-forming galaxy is most sensitive to the CO-to-H2 conversion assumption, a red but gas-rich system at higher redshift is more likely to hide dust-obscured star formation, and a gas-rich transitioning system may simply be temporarily quiescent. The paper's priority is to treat these as competing explanations that must be separated rather than as a single settled answer.
Sources used in this answer
Unveiling the Origins of the Molecular Gas Reservoirs in Recently Quenched Massive Galaxies
Using Simba, the paper finds that temporary quiescence, an underestimated CO-to-H2 conversion factor, and dust-obscured star formation are all plausible explanations for molecular gas-rich quenched massive galaxies from z=0 to z=2.