Jupiter is mostly an atmosphere wrapped around a comparatively small heavy-element interior. That familiar description conceals a severe timing problem. Hydrogen and helium are plentiful around a newborn star, but only while its planet-forming disc survives. Once that gas has accreted on to the star or escaped in jets and winds, a growing planet cannot call it back.
A new survey with the James Webb Space Telescope has made that disappearing reservoir unusually visible. Naman S. Bajaj of the University of Arizona and 12 colleagues studied 72 young systems, using warm hydrogen and ionised neon as tracers of gas leaving their protoplanetary discs. The work, published in The Astronomical Journal, is one of the largest Webb studies of disc dispersal so far.
The result is not a film of 72 Jupiters acquiring atmospheres. It is a population study of their possible nurseries at different stages. Its relevance to giant planets comes from what those outflows remove: the same gas from which a Jupiter or Saturn must build most of its mass. This is one study, not settled consensus, but its scale makes the changing pattern of disc loss harder to treat as an oddity of one or two young stars.
Webb found escaping gas around 66 discs
The team reanalysed archival observations from Webb’s Mid-Infrared Instrument, or MIRI. The sample consisted of 72 inclined discs, viewed at angles greater than 40 degrees, and most were Class II systems. At that stage the central star is visible, yet a substantial disc remains and continues to feed it.
MIRI is useful here because warm gas emits at mid-infrared wavelengths that Earth’s atmosphere can make difficult or impossible to study from the ground. It is not simply collecting a picture. Its integral-field spectroscopy records a spectrum at each position across a small field, allowing researchers to ask where a particular emission line appears as well as whether it is present.
The survey concentrated on rotational lines from molecular hydrogen, written H2, and a line from singly ionised neon, [Ne II]. Extended emission from one or both appeared around 66 of the 72 discs. Within those detections, the team identified conical molecular-hydrogen winds in 46 systems and high-velocity neon jets in 40.
Those categories can overlap. Indeed, every system with a neon jet also showed a wind traced by molecular hydrogen or oxygen. Among the jet sources, 85 percent had an H2 wind; the remaining cases had a wind signature in neutral oxygen. A narrow jet and a broader disc wind are therefore connected parts of the same young system, not necessarily competing explanations for one feature.
A jet is not the same thing as a disc wind
Material in a young disc orbits because it carries angular momentum. For some of it to spiral inward and accrete on to the star, angular momentum has to be redistributed or removed. Magnetic fields threading the disc can help. Gas follows those field lines away from the disc, carrying angular momentum with it and allowing other material to move inward.
The result can take more than one form. Jets are fast and tightly collimated, emerging roughly along the star’s rotational axis. Disc winds are broader and can rise across a larger area. Both remove mass, but their geometry, speed and physical origin need not be identical. Photoevaporative winds add another route: ultraviolet and X-ray radiation heat exposed gas until it is no longer gravitationally bound.
The inclined orientation of the survey’s discs helps separate these structures on the sky. A conical H2 feature extending away from the disc can be distinguished from the narrower [Ne II] structure of a high-velocity jet. That selection also matters when interpreting the result. The 72 objects are not a random census of every young planetary system; they were chosen in part because their geometry made spatially resolved outflows practical to identify.
The outflow changes as accretion declines
The strongest trend was with the rate at which disc material falls on to the star. Detection fractions for both molecular-hydrogen winds and neon jets rose with accretion rate. The researchers did not find a comparable dependence on stellar mass or on inclination within the selected sample.
At moderate to high accretion rates, above roughly 10-8.5 solar masses per year, extended molecular winds and low-velocity oxygen emission were more common. Hotter H2 wind components disappeared faster as the accretion rate fell than the cooler component did. At lower accretion rates, the fast jets weakened and the remaining wind signatures were predominantly atomic.
That pattern is consistent with an early phase in which magnetically driven jets and molecular-plus-atomic winds are prominent. Later, as the disc thins and stellar radiation penetrates it more easily, atomic winds including photoevaporative flows become relatively more important. The SETI Institute’s account of the study describes the 72 systems as frames in a film. It is a useful image so long as one caveat remains attached: the team observed different systems, not the same disc ageing before the telescope.
Why a growing Jupiter cannot wait
In the core-accretion account of giant-planet formation, dust grains first become pebbles, planetesimals and eventually a heavy core. When that core becomes sufficiently massive, its gravity can bind gas efficiently and the envelope grows rapidly. The order is important. The core must be ready while hydrogen and helium still dominate the disc.
This is why the new wind census matters to a planet that is not directly visible in the observations. Jupiter-like worlds do not merely require enough total material at the beginning. They require the solid-building stage and the gas-accretion stage to fit inside the disc’s useful lifetime, generally its first several million years for the gas-rich phase around Sun-like stars.
Space Daily previously covered Webb’s atmospheric measurements of the four giant planets around HR 8799. Their sulphur chemistry favours growth around solid cores followed by gas capture, even for worlds several times Jupiter’s mass. The new disc study looks earlier in the sequence, at the reservoir from which such envelopes would have to be drawn.
A core that reaches the critical stage early can become a gas giant. One that reaches it after the disc has been stripped may remain a smaller planet with a much thinner atmosphere. The outcome is not controlled by time alone; disc mass, metallicity, temperature, orbital location and the rate of core growth all matter. But no combination of those factors allows late accretion from gas that has already left the system.
Winds both enable the disc and empty it
Calling this a contest between planets and winds is tempting but incomplete. Magnetically launched winds do not only destroy a disc. By carrying angular momentum away, they can enable gas to flow inward. The same process that helps organise an accreting system also steadily reduces its mass.
Jets are similarly double-edged as evidence. They are a visible sign of vigorous accretion, when the disc still contains substantial material, yet the jet itself carries matter out. Later, stellar radiation can reach more of the thinning disc and drive photoevaporation. Disc dispersal is therefore not a single switch. The dominant route changes as the system evolves.
That changing mixture had already been seen in individual objects. In 2024, Bajaj and several members of the present team used Webb to resolve gas leaving T Cha, an ageing planet-forming disc with a large dust gap. Space Daily’s report on the T Cha wind observation captured the promise and the limitation of a single target. It showed that Webb could image disc dispersal; it could not say how representative that system was. The 72-disc study supplies the broader comparison.
The survey does not yet measure the countdown
Emission lines reveal composition, temperature, ionisation and motion, but turning their brightness into a total mass-loss rate requires a physical model. The researchers’ next goal is to determine how much gas each kind of wind removes and where in the disc that gas begins its escape. Those measurements would connect the observed structures more directly to the closing window for planet formation.
There are further boundaries. The archive combines observations made for different programmes. A correlation with accretion rate is not by itself proof that every system follows one fixed sequence. The sample’s inclination cut makes outflows easier to map but limits how confidently its raw detection fractions can be applied to all discs.
Disc lifetimes are not identical either. Lower-mass stars can retain material for unusually long periods, as another Webb study of a roughly 30-million-year-old disc showed. That earlier Space Daily report found that some planet-forming discs can persist far longer than the usual few-million-year picture. Such exceptions do not erase the deadline; they show that the deadline depends on the star and its environment.
From isolated winds to an evolutionary anatomy
The advance here is not the discovery that young stars launch jets or that their discs lose gas. Both were known. It is the simultaneous comparison of molecular winds, atomic winds and fast jets across a large Webb sample, revealing how their relative prominence shifts as accretion weakens.
That anatomy also explains why a simple disc-lifetime number can hide the real formation problem. The reservoir is changing throughout those several million years. Gas is moving inward, flowing outward, being heated, cooling, changing chemical state and becoming more exposed to stellar radiation. A forming planet occupies that evolving system rather than waiting beside a static tank.
Rocky bodies may continue colliding and assembling long after the nebular gas has mostly vanished. A Jupiter-like world faces a different constraint. Its defining material is the part of the disc that leaves first. Webb has now shown, across 72 young systems, how the machinery of that loss appears to change as the nursery ages. The remaining task is to measure how fast the machinery empties it.