Impact of stochastic star-formation histories and dust on selecting quiescent galaxies with JWST photometry

This study quantifies how stochastic star-formation histories and dust influence the photometric selection and physical characterization of quiescent galaxy candidates using JWST/MIRI data, challenging the assumption that these galaxies are uniformly dust-free.

The Gist

Imagine the universe as a giant, bustling city. In this city, there are two main types of neighborhoods:

1. The "Party Districts" (Star-Forming Galaxies): These are loud, chaotic, and full of activity. New stars are being born constantly, like new clubs opening every night. They are bright, colorful, and energetic.
2. The "Quiet Retirement Communities" (Quiescent Galaxies): These are the calm, older neighborhoods. The stars here are old, the "parties" have stopped, and the population is mostly retired. They are dimmer, redder, and very peaceful.

Astronomers want to study these "Quiet Retirement Communities" to understand how galaxies grow up and stop having new stars. But there's a problem: It's hard to tell them apart from a distance.

Sometimes, a "Party District" is just covered in a thick fog (dust). From far away, that fog makes the bright party lights look dim and red, tricking us into thinking it's a quiet retirement home. This is the "Dust-Age Degeneracy": Is the galaxy quiet because it's old, or is it just dusty?

The New Tool: JWST's "Super-Eyes"

Enter the James Webb Space Telescope (JWST). Think of JWST as a pair of super-glasses that can see through the fog. Specifically, it has a special camera called MIRI that looks at the universe in "Mid-Infrared" light. While our eyes (and older telescopes) see visible light, MIRI sees the heat radiating from the dust itself.

This paper is like a detective story where the authors ask: "If we use these new super-glasses and different theories about how galaxies grow up, do we find more or fewer 'Quiet' galaxies? And are we identifying them correctly?"

The Detective's Toolkit: Three Different Theories

To figure out a galaxy's history, astronomers have to guess its "Star-Formation History" (SFH). It's like trying to guess a person's life story just by looking at their photo. The authors tested three different ways to guess this story:

1. The "Smooth Slope" (Delayed Model): Imagine a galaxy that had a big party early in life, then slowly, smoothly, the music got quieter and quieter until it stopped. This is the traditional, simple way of thinking.
2. The "Random Walk" (Non-Parametric Model): Imagine a galaxy that had a wild party, then a quiet week, then a sudden burst of energy, then silence again. This model allows for messy, unpredictable changes in the galaxy's life.
3. The "Gas Tank" (Regulator Model): Imagine a galaxy as a car with a gas tank. The amount of gas (gas reservoir) controls how fast it can drive (star formation). If the tank is full, it speeds up; if it's empty, it slows down. This model is based on the physics of how gas flows in and out of galaxies.

The Investigation: What Did They Find?

The authors took about 13,000 galaxies from a deep field of space (the CEERS survey) and ran them through their detective toolkit. Here is what they discovered:

1. The Fog Matters (MIRI is Crucial)
When they looked at the galaxies without the MIRI camera (just the "foggy" view), they got the wrong answers. They thought some galaxies were making new stars when they weren't, and they overestimated how much dust was hiding them.

• The Analogy: It's like trying to guess the temperature of a room by looking at a thermometer through a thick blanket. You can't tell if the room is hot or if the blanket is just insulating the heat.
• The Result: Adding the MIRI data (seeing through the blanket) changed the results significantly. It helped them find more true "Quiet" galaxies because they could finally distinguish between a dusty party and a quiet retirement home.

2. The Theory Changes the Count
Depending on which "Life Story" theory they used, the number of galaxies they classified as "Quiet" changed wildly!

• The "Random Walk" theory found the most quiet galaxies.
• The "Gas Tank" theory found the fewest.
• The Takeaway: The rules we use to define "quiet" matter a lot. If you change the rules, you change the population.

3. The "Dusty Retirees"
One of the most surprising findings was that many of these "Quiet" galaxies are actually dusty.

• The Analogy: We used to think that when a galaxy retires, it cleans up its house and throws away all the dust. But this paper shows that many "retired" galaxies are still messy! They have a lot of dust left over, or they are even making new dust from old dying stars.
• The Result: About 13% of the "Quiet" galaxies they found are actually quite dusty. This challenges our old ideas about how galaxies age.

4. Size Matters
They found a clear link between size and dust. The biggest, most massive "Quiet" galaxies were the dustiest. It's like the largest retirement communities having the most gardens (dust) still growing in them.

The Bottom Line

This paper is a reality check for astronomers. It tells us that:

• We need the new telescope (JWST/MIRI): Without it, we are guessing in the dark and misidentifying galaxies.
• We need to be careful with our theories: The way we model a galaxy's history changes the results. There is no single "correct" answer yet; we have to compare different models.
• Galaxies are complex: Even the "quiet" ones are still messy, dusty, and evolving in ways we didn't expect.

In short, the universe is more complicated than we thought, but with our new "super-glasses," we are finally starting to see the real picture behind the fog.

Technical Summary

Here is a detailed technical summary of the paper "Impact of stochastic star-formation histories and dust information on selecting quiescent galaxies with JWST photometry" by Lisiecki et al. (2026).

1. Problem Statement

The identification and characterization of quiescent galaxies (QGs) at high redshifts ($z \gtrsim 2$) using the James Webb Space Telescope (JWST) face significant challenges due to:

• Dust-Age Degeneracy: Distinguishing between old stellar populations and young, dust-reddened populations is difficult with optical/Near-IR (NIR) photometry alone.
• Star-Formation History (SFH) Assumptions: Traditional parametric SFH models (e.g., exponentially declining) may fail to capture complex, stochastic, or bursty star-formation histories common in early galaxies.
• Selection Bias: The number of identified QG candidates (QGCs) and their derived physical properties (stellar mass $M_\star$, dust attenuation $A_V$, star formation rate SFR) vary significantly depending on the chosen SFH model and the inclusion of Mid-Infrared (MIR) data.
• Quenching Mechanisms: There is a lack of consensus on how long dust persists after quenching and whether massive QGs retain significant dust reservoirs.

2. Methodology

The authors utilized data from the CEERS (Cosmic Evolution Early Release Science) survey, specifically focusing on the MIRI footprint ($\sim 28.28$ arcmin$^2$).

• Sample: A parent catalog of $\sim 13,000$ sources was reduced to a mass-complete sample of 5,021 galaxies ($M_\star \ge 10^8 M_\odot$) at $0.1 \le z \le 6$. The sample includes sources with direct MIRI detections and those with upper limits.
• SED Fitting: The study employed CIGALE (Code Investigating GALaxy Emission) to perform Spectral Energy Distribution (SED) fitting.
• SFH Models: Three distinct SFH prescriptions were implemented and compared:
  1. DelayedBQ: A flexible parametric model (delayed-$\tau$ + instantaneous burst/quench).
  2. NonParametric: A stochastic model where SFR changes randomly between time steps following a Student-t distribution (Leja et al. 2019a).
  3. Regulator (Extended Regulator): A physically motivated stochastic model based on gas reservoir control, inflow/outflow rates, and GMC physics (Tacchella et al. 2020; Iyer et al. 2024).
• Experimental Design:
  • MIRI vs. No-MIRI Runs: Each SFH model was run twice: once including JWST/MIRI photometry and dust emission models, and once excluding them (NIR-only).
  • Selection Criteria: QGCs were selected using three methods:
    1. Main Sequence (MS) Offset: $\log(\text{SFR}/\text{SFR}_{\text{MS}}) < -0.6$.
    2. Specific SFR (sSFR): $\text{sSFR} < 0.2/\tau(z)$ (Pacifici et al. 2016).
    3. UVJ Diagram: Rest-frame color-color selection.
• Validation: Stability was tested via 5 independent iterations per stochastic run. A mock analysis using a SIMBA cosmological simulation galaxy was used to validate the recovery of quenching timescales.

3. Key Contributions

• Implementation of Stochastic SFHs in CIGALE: The authors successfully integrated the NonParametric and Extended Regulator stochastic SFH modules into CIGALE, allowing for a direct comparison of stochastic vs. parametric approaches on real JWST data.
• Quantification of MIRI Impact: They systematically isolated the impact of MIRI data on breaking the dust-age degeneracy, demonstrating its necessity for accurate $A_V$ and SFR estimation.
• Systematic Bias Analysis: The study provides a comprehensive map of how different SFH models and selection criteria alter the statistical properties of QG samples, revealing that SFH choice is a dominant source of uncertainty, often exceeding the impact of selection criteria.

4. Key Results

A. Impact on Physical Properties

• Stellar Mass ($M_\star$) and Age: The inclusion of MIRI data has a negligible effect on derived $M_\star$ ($<0.09$ dex) and ages ($<0.1$ dex). These properties remain consistent across all SFH models.
• SFR and Dust Attenuation ($A_V$): MIRI data is crucial for constraining SFR and $A_V$.
  • Without MIRI, SFRs are systematically overestimated (up to $0.65$ dex) due to poor dust constraints.
  • Without MIRI, $A_V$ shows a bias toward higher values (up to $0.25$ mag).
  • Including MIRI reduces the median $A_V$ and breaks the SFR-$A_V$ degeneracy.

B. Impact on QGC Selection

• SFH Model Dependence: The choice of SFH model drastically changes the number of selected QGCs.
  • NonParametric selects the largest samples (up to $\sim 3.5\times$ more than Regulator).
  • Regulator selects the smallest samples due to residual star formation inherent in the gas-regulator physics.
  • The difference in sample size driven by SFH choice is larger than the difference caused by the selection criterion (MS vs. sSFR).
• MIRI Impact on Selection: Including MIRI data increases the number of selected QGCs by a factor of 1.4 to 2, regardless of the SFH or selection criterion used.
• UVJ Limitations: A significant fraction of QGCs (up to 69% for NonParametric models) selected via sSFR or MS criteria are not recovered by the UVJ color diagram. This suggests UVJ is insufficient for high-redshift or dusty QGs.

C. Dust and Quenching Evolution

• Dusty Quiescent Galaxies: Approximately 13% of selected QGCs exhibit significant attenuation ($A_V > 0.5$), supporting recent findings of dust-rich QGs.
• Mass-Attenuation Correlation: A strong correlation exists between $A_V$ and $M_\star$ at $z \le 2.5$. Massive galaxies ($M_\star \sim 10^{11} M_\odot$) are $\sim 1.5-4\times$ more attenuated than low-mass galaxies ($M_\star \sim 10^9 M_\odot$).
• Quenching Timescales:
  • The quenching moment ($\tau_q$) can be recovered with $\sim 45\%$ precision.
  • The quenching duration ($T_q$) is unstable and does not converge well in photometry-only studies (scatter $\sim 67\%$).
  • Post-Quenching Dust: Substantial dust attenuation is sustained for the first $\sim 1$ Gyr after quenching. Mature QGCs ($\Delta T > 1$ Gyr) show diverse dust properties, with some retaining high $A_V$, suggesting mechanisms for dust re-growth or accumulation (e.g., AGB stars) rather than rapid destruction.

5. Significance and Conclusions

• JWST/MIRI is Essential: The study confirms that MIRI data is fundamental for realistic dust attenuation estimates and for distinguishing between quiescent and dusty star-forming galaxies. Relying solely on optical/NIR photometry leads to significant biases in SFR and $A_V$.
• SFH Model Sensitivity: The physical interpretation of QG populations is highly sensitive to the assumed SFH. The NonParametric model favors early mass assembly and faster quenching, while the Regulator model implies more gradual processes. Researchers must account for this systematic uncertainty when comparing samples.
• Re-evaluating QG Selection: The UVJ diagram is insufficient for selecting the full population of high-redshift QGs, particularly those with high dust content or young stellar populations. Multi-criteria selection (sSFR/MS + MIRI constraints) is required.
• Dust in Quiescence: The persistence of dust in mature QGs challenges simple "dust-free" quenching scenarios, implying complex ISM evolution where dust can be replenished or preserved long after star formation ceases.

Final Recommendation: Future studies of high-redshift galaxy evolution should utilize full optical-to-MIR photometry and consider stochastic SFH models to accurately characterize the quenching pathways and dust properties of the early universe's quiescent galaxies.