The Timescales of Embedded Star Formation as Observed in STARFORGE astropy ( Astropy Collaboration et al. 2013 ; Astropy Collaboration et al. 2018 ; Astropy Collaboration et al. 2022 ) , Jupyter ( Kluyver et al. 2016 ) , matplotlib ( Hunter 2007 ) , numpy ( Harris et al. 2020 ) , python ( Van Rossum

Tobin M. Wainer, Julianne J. Dalcanton, Michael Y. Grudić, Stella S. R. Offner, Adam Smercina, Benjamin F. Williams, L. Clifton Johnson, J. Peltonen, Eric W. Koch, Kartik R. Neralwar · 2025

Read The Timescales of Embedded Star Formation as Observed in STARFORGE astropy ( Astropy Collaboration et al. 2013 ; Astropy Collaboration et al. 2018 ; Astropy Collaboration et al. 2022 ) , Jupyter ( Kluyver et al. 2016 ) , matplotlib ( Hunter 2007 ) , numpy ( Harris et al. 2020 ) , python ( Van Rossum by Tobin M. Wainer, Julianne J. Dalcanton, Michael Y. Grudić, Stella S. R. Offner, Adam Smercina, Benjamin F. Williams, L. Clifton Johnson, J. Peltonen, Eric W. Koch, Kartik R. Neralwar (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-18322.txt.

Introduction

Stars are born inside dusty clouds, then break free as their own light and winds push gas away. Tobin M. Wainer and nine co-authors ask how long that hidden phase lasts. Their 2025 paper uses the STARFORGE simulations, which model whole star-forming clouds. They track how each star moves from buried to exposed.

For single stars, the step from buried to exposed is fast, done within about 1.3 million years of peak mass. Heavy stars drive the change: they keep feeding while hidden, until their output balances the inflow and then wins. Once the heavy stars show, their local feedback clears gas nearby. Because heavy stars give most of the light, their quick reveal sets the cloud's brightening. The timing scales with how fast the cloud itself would collapse. Dust and gas glow peak while the heavy stars are still close to their birth gas.

Catalog record

CREATORTobin M. Wainer, Julianne J. Dalcanton, Michael Y. Grudić, Stella S. R. Offner, Adam Smercina, Benjamin F. Williams, L. Clifton Johnson, J. Peltonen, Eric W. Koch, Kartik R. Neralwar
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE114 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-18322

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

The Timescales of Embedded Star Formation as Observed in STARFORGE astropy ( Astropy Collaboration et al. 2013 ; Astropy Collaboration et al. 2018 ; Astropy Collaboration et al. 2022 ) , Jupyter ( Kluyver et al. 2016 ) , matplotlib ( Hunter 2007 ) , numpy ( Harris et al. 2020 ) , python ( Van Rossum & Drake 2009 ) , scipy ( Virtanen et al. 2020 ; Gommers et al. 2024 ) , h5py ( Collette 2013 ; Collette et al. 2023 ) , skirt ( Baes et al. 2011 ; Camps & Baes 2015 ) , openai ( OpenAI 2023 ) Tobin M. Wainer, Julianne J. Dalcanton, Michael Y. Grudić, Stella S. R. Offner, Adam Smercina, Benjamin F. Williams, L. Clifton Johnson, J. Peltonen, Eric W. Koch, Kartik R. Neralwar Abstract Star formation occurs within dusty molecular clouds that are then disrupted by stellar feedback. However, the timing and physical mechanisms that govern the transition from deeply embedded to exposed stars remain uncertain. Using the STARFORGE simulations, we analyze the evolution of “embeddedness”, identifying what drives emergence. We find the transition from embedded to exposed is fast for individual stars, within 1.3 Myr after the star reaches its maximum mass. This rapid transition is dominated by massive stars, which accrete while remaining highly obscured until their feedback eventually balances, then overcomes, the local accretion. For these massive stars, their maximum mass is reached simultaneously with their emergence. Once these stars are revealed, their localized, pre-supernova feedback then impacts the cloud, driving gas clearance. Because massive stars dominate the luminosity, their fast, local evolution dominates the light emergence from the dust. We calculate the dependence of these processes on the mass of the cloud and find that emergence always depends on when massive stars form, which scales with the cloud’s free-fall time. We also examine the evolution of dust emission and H α \alpha luminosity. We find that dust dominates the luminosity for roughly 2 Myrs before stellar luminosity becomes more luminous. These results suggest that deeply embedded star-forming clusters tend to be rare compared to those partially exposed. Thus, because the initial embedding of the most luminous stars is highly local, the emergence of stars is a faster, earlier, more local event than the overall disruption of the cloud by gas expulsion. I Introduction It is widely accepted that stars are born during the gravitational collapse of turbulent molecular clouds (e.g., Lada & Lada 2003 , for review) . These stars eventually emerge from the cloud, through a variety of feedback processes that clear the surrounding gas, leaving a largely un-obscured view of the newly-formed stars ( Krause et al. 2020 ; Pineda et al. 2023 ) . Early on, however, accreting protostars remain enshrouded in their dusty natal clouds. These young stars are typically shielded by heavy layers of dust, making them largely invisible at optical wavelengths. At longer wavelengths however, they can be detected – either directly as individual stars, or, indirectly through their heating of the surrounding dust. In this early phase, these collections of enshrouded stars and protostars are known as embedded clusters and are some of the earliest windows onto to the star formation process ( Ascenso 2018 ) . This embedded stage concludes when gas is expelled from the region, exposing the stars, and consequently destroying the GMC (See reviews Krumholz 2014 ; Chevance et al. 2020 ; Schinnerer & Leroy 2024 , and references within) . The primary mechanisms for this gas removal are through accretion-driven bipolar jets and outflows, radiation, and stellar winds from protostars ( Grudić et al. 2022 ; Guszejnov et al. 2022 ) . Two primary parameters are thought to determine the evolution of an emerging embedded cluster: the star formation efficiency and the timescales of gas dispersal ( Lada & Lada 2003 ) . Specifically, gas removal, and the timescales on which it occurs, is critical to star formation models. Once material is removed from the system, star formation is quenched, not only regulating star formation locally, but on a galactic scale ( Krumholz et al. 2005 ; Krumholz et al. 2019 , e.g.,) . Despite its importance, the underlying mechanism which govern how, and when, gas is cleared – and the associated observability of the embedded phase – remains poorly constrained. There remain ongoing conversations about the timescales over which this gas removal takes place, how long the newly formed stars remain embedded observationally, and what physical processes drive the clearance and its timescales ( McLeod et al. 2021 ; Chevance et al. 2022 ; McQuaid et al. 2024 , e.g.,) . Efforts to constrain embedded cluster lifetimes have historically relied on associating young stellar clusters with their parent molecular clouds. Early studies found that clusters younger than \sim 5 Myr were often still embedded, while older ones were not, suggesting gas is typically cleared on \sim 5 Myr timescales ( Leisawitz et al. 1989 ; Lada & Lada 2003 , e.g.,) . More recent work extended this method to external galaxies using HST to detect young clusters and using ALMA observations to trace molecular gas or H \alpha to trace ionized gas ( Whitmore et al. 2011 ; Grasha et al. 2018 ; Hannon et al. 2022 , e.g.,) . These larger samples often yield shorter inferred timescales of 1–2 Myr across diverse galactic environments. However, a central limitation is that optical and UV studies are blind to the most deeply embedded phases. The stars must already be clearing their surroundings to be observable at these wavelengths, meaning that these methods trace when clusters become visible, not when star formation starts or gas dispersal begins. The resulting timescales are therefore biased toward later, more exposed stages. Infrared (IR) observations fill part of this gap. Dust emission in the mid- and far-IR can trace embedded clusters before their stellar populations emerge in the optical or UV. When individual stars are resolved, their ages provide constraints on how long dust and gas remain detectable ( Getman et al. 2014 , e.g.,) . Using this approach, studies with Spitzer and Herschel generally recover longer embedded lifetimes of \sim 3-5 Myr ( Morales et al. 2013 ; Battersby et al. 2017 ; Kim et al. 2021 ; Chevance et al. 2022 , e.g.,) , reflecting their sensitivity to earlier stages of cluster formation. However, IR observations face their own challenges. Dust emission is not exclusive to embedded clusters; contamination from field stars and unrelated star-forming regions can lead to false positives ( Ascenso 2018 ) . Without high-resolution imaging and multi-wavelength context, it’s difficult to determine whether a dusty region truly corresponds to an embedded system ( Johnson et al. 2022 ) . The capabilities of JWST are beginning to resolve this ambiguity. For the first time, we can spatially disentangle embedded structures in external galaxies, simultaneously tracing both stellar content and dust emission at sub-parsec resolution ( Lee et al. 2023 ; Rodríguez et al. 2023 ; Rodríguez et al. 2025 , e.g.,) . In particular, polycyclic aromatic hydrocarbons (PAHs) serve as a powerful tracer of cluster emergence, illuminating photodissociation regions (PDRs) as UV radiation escapes young stars ( Rodríguez et al. 2023 ; Whitmore et

Read the complete text