The Fading Fountain: Evolution of Protoplanetary Disk Winds from JWST Data ✛
Context
Seventy-two protoplanetary disks, seventy-two embryos of planetary systems—and most are spewing gas streams like jets of a cosmic fountain. These outflows carry away angular momentum, allowing the Accretion disk to feed the star. Without this mechanism, star and planet formation would grind to a halt. Understanding how this fountain works across different stages of star formation is key to theories of exoplanet formation. Models predict that as accretion weakens, dense molecular winds should give way to tenuous atomic ones, but direct evidence was missing. Now, thanks to JWST, this evolution has been statistically traced.
Methods
The JWST/MIRI integral field spectrograph recorded spectra for each point of 72 inclined disks. Pixel-by-pixel emission maps were built for lines of hydrogen (from S(1) to S(7)) and ionized neon ([Ne II]). To identify extended structures, they compared with observations of calibration stars: if a spot was broader than a point source, it was considered elongated. Winds were recognized by their conical shape: they computed the brightness ratio inside a 100° opening-angle cone to outside and compared it with the same ratio for the continuum (thus removing PSF artifacts). For [Ne II], high-resolution spectroscopy was used to build velocity maps via the Doppler effect. A jet was defined as a flow with speed >30 km/s directed perpendicular to the disk plane.
Results
Of the 72 disks, 66 showed extended emission. Conical H₂ winds were found in 46 objects, and high-velocity [Ne II] jets in 40. In 34 systems, both are present. The fraction of disks with outflows strongly depends on accretion rate: above 10⁻⁸ M⊙/year, about 80% of disks launch powerful flows, while below, only 20%. Importantly, the hot H₂ lines (S(5) and S(7)) disappear faster than the cold S(1) as accretion drops. In other words, the cosmic fountain loses its hottest, innermost jets first. Comparison with ground-based [O I] observations showed that at high accretion, both atomic and molecular winds are always present, but at low accretion, only the atomic wind remains.
Implications
This is the first statistical proof of wind evolution from molecular to atomic. The picture fits perfectly with the magnetohydrodynamic model (Maxwell's equations for plasma): in early stages, dense molecular cocoons shield the star's hard radiation, but as accretion fades and jets weaken, X-rays and ultraviolet break through and drive photoevaporation—predominantly an atomic wind. Thus, the disk gradually sheds its gas envelope, transitioning from a turbulent fountain to a barely perceptible evaporation.
Future development
Future modeling of the observed winds within MHD and photoevaporative frameworks will allow estimates of their mass loss rates and contribution to disk dispersal. Observations at the resolution limit will be extremely interesting to directly measure the wind launching radius. Already, JWST is changing our understanding of how the cosmic fountain breathes around young stars.
Impact
These results will impact models of protoplanetary disk evolution and planet formation: winds alter the density profile, slow planet migration, and change the chemical composition of inner regions.
Next steps
The next step is to estimate mass loss rates for the 46 detected molecular winds and compare with theoretical predictions. Also planned is to compare these results with data on younger protostars (Class 0/I) to build a complete evolutionary sequence.
Key open problems
This work directly addresses the fundamental problem of protoplanetary disk evolution. The main unresolved question—the mechanism of angular momentum removal—receives observational support here: MHD winds indeed dominate at early stages and can provide the necessary angular momentum sink. Moreover, the data constrain the physics of photoevaporation and determine how quickly disks dissipate.
The knowledge machine's view
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
A survey of 72 protoplanetary disks with JWST/MIRI has revealed extended emission in H₂ and [Ne II] lines. Emission maps were constructed, and conical molecular winds and high-velocity atomic jets were identified. The fraction of disks with outflows is shown to decrease with decreasing accretion rate, with hot H₂ lines disappearing before cold ones, indicating an evolution from molecular to atomic winds as accretion wanes.
Where its strength is
For the first time, a statistical confirmation of the transition in outflow regimes is provided: from dense molecular winds at high accretion rates to predominantly atomic photoevaporation at late stages. This offers a solid observational foundation for magnetohydrodynamic models of disk evolution.
Where it could go next
- It might be interesting to compare the found wind evolution with data on turbulent magnetic field amplification in young objects: in our archive, there is an analysis of a study showing how turbulence in neutron stars generates magnetic fields. Perhaps a similar mechanism at early disk stages influences the launching of molecular winds.based on: Cosmic Kneading: Turbulent Amplification of Magnetic Fields in Newborn Neutron Stars arXiv:2607.21990v1
- The observed predominance of atomic winds at late stages echoes the idea that stellar winds can 'quench' star formation by dispersing residual gas. Our archive contains a rapid analysis of the work on winds that quench stars—this could provide context for understanding how a disk ultimately loses its gas.based on: The Wind That Snuffs Out Stars arXiv:2606.10271
- Since both molecular and atomic components were identified in the study, it is worth looking at a neighboring analysis where a molecular wind literally 'opens' the disk for observations. A similar approach might allow a more detailed study of the transition zone between regimes.based on: Molecular Wind Unveils the Disk arXiv:2607.25770
Worth trying
- Test the correlation between molecular wind morphology and disk turbulence indicators, using approaches from the work on turbulent magnetic field amplification.
- Investigate how the role of photoevaporation relative to magnetohydrodynamic winds changes at different stages, building on the 'wind that quenches stars' model.
- Construct high-resolution velocity maps for H₂ winds, similar to those made for [Ne II], to pinpoint where the molecular wind transitions to an atomic one.
Why this may matter
Understanding the evolution of protoplanetary disk winds is critical for planet formation theory: it determines how long a disk retains the gas needed to form gas giants and when the dispersal of the remnant nebula begins. The statistical approach paves the way for predicting the architecture of mature planetary systems.
Nearby works in our archive
- How Dust Disks Dance Around Newborn Starssimilarity 0.69
- Why a Star Explosion Is Like a Tug of Warsimilarity 0.68
- Young Stars: Gamma Rays from Jetssimilarity 0.67
- The Stellar Forge: A First Glimpse at the Birth of Cosmic Crystalssimilarity 0.65
- The Wind That Snuffs Out Starssimilarity 0.64
- A Spinning Top at the Galaxy’s Core Snuffs Out Starssimilarity 0.64
- Molecular Wind Unveils the Disksimilarity 0.64
- Cosmic Kneading: Turbulent Amplification of Magnetic Fields in Newborn Neutron Starssimilarity 0.63
The closest work in the archive sits at 0.69 by meaning — the usual level for our corpus.
Original abstract
as written by the authors arXiv:2607.21733v1
The evolution and dispersal of protoplanetary disks--governed by accretion, magnetically launched jets and winds, and photoevaporative winds--fundamentally shape planetary systems. Determining how these mass-loss processes co-evolve is crucial for constraining planet formation pathways. We analyze archival JWST/MIRI/IFU data of 72 inclined (i>40deg) mostly ClassII disks to identify and characterize spatially resolved jets and winds, focusing on [NeII] and H2 lines. Extended emission in H2 S(1), S(3), S(5), S(7) and/or [NeII] is detected toward 66 disks, revealing diverse morphologies. We develop a framework to identify conical H2 winds and high-velocity [NeII] jets perpendicular to the disk, detecting them toward 46 and 40 disks, respectively. All sources with [NeII] jets exhibit a corresponding wind traced in either H2 (85%) or [OI], establishing a connection between jets and winds. The detection fractions of [NeII]-jets and H2-winds correlate positively with mass accretion rate, with no dependence on disk inclination or stellar mass. Conversely, marginally resolved low-velocity [NeII] winds are found preferentially toward lower accretors. Among sources with H2 winds, detection of hotter winds traced by S(7) and S(5) declines more rapidly with decreasing accretion rate than the colder S(1) component. Comparison with high-resolution [OI]6300\text{\AA} spectroscopy reveals [OI] LVC and extended H2 wind detections preferentially toward moderate-to-high accretors (>~10^{-8.5}~Msun/yr), whereas lower accretors exhibit only [OI] and [NeII] winds. Together, these results indicate that atomic jets and atomic+molecular winds, consistent with an MHD disk-wind origin, dominate during early, actively accreting disk phases, while at lower accretion rates, jets weaken and winds become predominantly atomic.
In two disks, IRAS 04385 and WSB 52, both the jet and the molecular wind turned out to be "red"—that is, receding from us. This indicates an intrinsic launch asymmetry, possible only in models with non-ideal magnetohydrodynamics where the Hall effect creates a one-sided bias.
Key numbers
- fraction of disks with extended H₂ winds: 64% (46 out of 72)
- fraction of disks with [Ne II] jets: 56% (40 out of 72)
- accretion rate at the morphology boundary: ≈10⁻⁸ M⊙/year
- excitation temperature of the H₂ S(7) line: 7200 K (hotter than the Sun's surface)
- threshold velocity for a jet: >30 km/s