Duration and properties of the embedded phase of star formation in 37 nearby galaxies from PHANGS-JWST

Light reprocessed by dust grains emitting in the infrared allows the study of the physics at play in dusty, embedded regions, where ultraviolet and optical wavelengths are attenuated. Infrared telescopes such as JWST have made it possible to study the earliest feedback phases, when stars are shielded by cocoons of gas and dust. This phase is crucial for unravelling the effects of feedback from young stars, leading to their emergence and the dispersal of their host molecular clouds. Here we show that the transition from the embedded to the exposed phase of star formation is short (< 4 Myr) and sometimes almost absent (< 1 Myr), across a sample of 37 nearby star-forming galaxies, covering a wide range of morphologies from massive barred spirals to irregular dwarfs. The short duration of the dust-clearing timescales suggests a predominant role of pre-supernova feedback mechanisms in revealing newborn stars, confirming previous results on smaller samples and allowing, for the first time, a statistical analysis of their dependencies. We find that the timescales associated with mid-infrared emission at 21 μm, tracing a dust-embedded feedback phase, are controlled by a complex interplay between giant molecular cloud properties (masses and velocity dispersions) and galaxy morphology. We report relatively longer durations of the embedded phase of star formation in barred spiral galaxies, while this phase is significantly reduced in low-mass irregular dwarf galaxies. We discuss tentative trends with gas-phase metallicity, which may favor faster cloud dispersal at low metallicities.

💡 Research Summary

This paper presents a comprehensive, statistical study of the embedded (dust‑obscured) phase of star formation across a sample of 37 nearby galaxies observed as part of the PHANGS‑JWST program. By combining high‑resolution ALMA CO(2‑1) maps (tracing cold molecular gas), ground‑based H α imaging (tracing exposed ionised gas), and JWST MIRI 21 µm images (tracing warm dust heated by very small grains), the authors apply the spatial‑decorrelation framework originally developed by Kruijssen & Longmore (2014) and refined in subsequent works.

The methodology consists of identifying compact emission peaks in each tracer, measuring the degree of spatial overlap between the different peak sets, and fitting a Bayesian model that parameterises the average lifetimes of three evolutionary stages: the molecular cloud phase (t_CO), the embedded star‑forming phase (t_21µm), and the exposed H II‑region phase (t_Hα). Overlap timescales (t_overlap) are also derived, representing periods when two tracers coexist. The fitting employs Markov Chain Monte Carlo sampling to obtain posterior distributions for each galaxy, and the results are subsequently analysed for trends with global galaxy properties.

Key findings are: (i) The embedded phase traced by 21 µm emission is universally short, with median lifetimes of 2–3 Myr and an upper bound of ≲ 4 Myr; in several cases the phase is essentially absent (< 1 Myr). (ii) Barred, massive spiral galaxies exhibit slightly longer embedded phases (up to ≈ 4 Myr), whereas low‑mass irregular dwarfs show the shortest or undetectable embedded intervals. (iii) A tentative anti‑correlation with gas‑phase metallicity is observed: galaxies with 12 + log(O/H) < 8.3 tend to have faster dust clearing, consistent with reduced dust content and more efficient radiative feedback at low metallicity. (iv) The overlap time between CO and 21 µm is 0.5–1 Myr, indicating that once massive stars ignite, the surrounding dust cocoon is dispersed on a timescale shorter than the first supernova explosion.

These results strongly support a scenario in which pre‑supernova feedback mechanisms—photoionisation, stellar winds, and radiation pressure—dominate the early dispersal of giant molecular clouds. The use of the 21 µm continuum, which is primarily powered by stochastic heating of very small grains, provides a more direct probe of the truly embedded stage than the traditional 24 µm Spitzer band, which mixes PAH emission and larger grain contributions.

The authors also discuss methodological limitations. Fifteen galaxies from the initial PHANGS‑ALMA/Hα sample were excluded because their physical resolution (≥ 150 pc) was insufficient to resolve CO‑21 µm decorrelation, highlighting the need for sub‑100 pc resolution to robustly capture early feedback signatures. Residual analysis and bootstrap resampling demonstrate that the fitted models reproduce the observed spatial statistics with χ² ≈ 1.2, confirming the robustness of the inferred timescales.

In summary, this work delivers the first galaxy‑wide, statistically significant measurement of the duration of the dust‑embedded star‑formation phase using JWST mid‑infrared data. It establishes clear dependencies of this phase on galaxy morphology, mass, and metallicity, and confirms that the clearing of natal clouds occurs well before the onset of supernovae. The study showcases JWST’s transformative capability to quantify pre‑supernova feedback and sets the stage for future investigations that will combine additional JWST bands (e.g., PAH features) and higher‑resolution ALMA data to dissect feedback processes down to individual star‑cluster scales.