When interstellar comet 3I/ATLAS was first examined in detail last summer, its gas cloud was swimming in carbon dioxide, roughly 7.6 times more CO2 than water vapor. That made it unlike almost any comet astronomers had seen in our own solar system. Now, after the object swung within 1.4 astronomical units of the Sun and headed back toward deep space, high-resolution spectroscopy from the Subaru Telescope on Maunakea tells a different story: water appears to have overtaken carbon dioxide as the dominant gas streaming off the nucleus.
The reversal, reported in a March 2026 preprint led by researchers using Subaru’s High Dispersion Spectrograph, adds a striking time dimension to the chemistry of only the third confirmed interstellar object ever detected. It also raises a pointed question: how much can a single pass near a star reshape ices that formed billions of years ago around an entirely different sun?
The CO2-rich baseline before perihelion
The composition benchmark for 3I/ATLAS was set on August 6, 2025, when the James Webb Space Telescope’s NIRSpec instrument captured near-infrared spectra of the inbound comet at 3.32 au from the Sun. That dataset, led by Martin Cordiner and collaborators and archived as a preprint, revealed a coma overwhelmingly dominated by carbon dioxide. The CO2-to-water mixing ratio of 7.6 (plus or minus 0.3) dwarfed typical values for solar system comets, which rarely exceed 0.3. Detections of carbon monoxide, carbonyl sulfide, water ice, and dust rounded out the portrait of an object whose birthplace clearly favored CO2 ice formation or long-term retention.
For context, the only previous interstellar comet observed in comparable detail was 2I/Borisov, which passed through the inner solar system in late 2019. Borisov was notably rich in carbon monoxide relative to water but did not display the extreme CO2 dominance seen in 3I/ATLAS. The first interstellar object, 1I/’Oumuamua, showed no detectable coma at all, leaving its volatile inventory a mystery. So 3I/ATLAS offered something genuinely new: a chance to watch alien ices respond to our Sun’s heat in real time.
A blackout, then a changed comet
3I/ATLAS reached perihelion around October 30, 2025, passing roughly 1.4 au from the Sun. For several weeks surrounding closest approach, the comet sat too near the Sun in the sky for ground-based or space-based telescopes to observe safely. Whatever happened to the nucleus during that period of peak heating went unseen.
By early December 2025, the comet reappeared in darker skies, and observatories moved quickly. JWST’s MIRI instrument collected mid-infrared spectra on December 15-16 and again on December 27, at heliocentric distances of 2.20 and 2.54 au respectively. Those observations, detailed in a January 2026 preprint, detected both water bands and carbon dioxide features, confirming that both volatiles remained active well after perihelion. The spectra also revealed an atomic nickel emission line at 7.507 micrometers, the first such detection for an interstellar body in the mid-infrared. Nickel emission had previously been identified in solar system comets (a discovery first reported in Nature in 2021), but its appearance in 3I/ATLAS at 2.20 au sits outside the range where standard thermal models easily explain it.
Then came the Subaru observations. On January 7, 2026, with the comet at 2.87 au and heading outbound, the High Dispersion Spectrograph captured forbidden oxygen emission lines at 557.7, 630.0, and 636.4 nanometers. These lines are workhorses of cometary science: the ratios between them act as chemical fingerprints, revealing whether the oxygen atoms originated primarily from the breakup of water molecules or carbon dioxide molecules. The team’s analysis, presented in a March 2026 preprint, found line ratios consistent with water vapor now dominating the coma, a dramatic departure from the pre-perihelion CO2 ratio of 7.6.
Supporting observations filled in the broader activity picture. NASA’s TESS spacecraft conducted a dedicated observation run in mid-January 2026, tracking the comet’s brightness and rotation. Earlier, NASA’s SPHEREx mission followed the comet’s brightening during a December 2025 infrared campaign, picking up dust and volatile signatures as solar heating drove renewed outgassing. Earth’s closest approach to 3I/ATLAS was roughly 1.8 au, keeping the object within range of major observatories throughout the post-perihelion arc.
What the data do and do not prove
The strongest evidence for a composition shift comes from two independent datasets that bracket perihelion. The JWST/NIRSpec pre-perihelion spectrum provides a firm CO2/H2O baseline at 3.32 au inbound. The Subaru/HDS post-perihelion spectrum provides forbidden oxygen line ratios that, through well-established photochemistry, constrain the same ratio at a far lower value near 3 au outbound. Because these are independent instruments operating at different wavelengths and different epochs, their agreement on the direction of change is hard to dismiss as a systematic artifact.
But important gaps remain. No single instrument has yet delivered a direct, simultaneous measurement of both CO2 and H2O production rates in the post-perihelion coma. The Subaru oxygen line ratios are powerful diagnostics, yet translating them into absolute mixing ratios requires photochemical modeling that depends on assumptions about gas velocity, nucleus size, and the dust-to-gas ratio. Small changes in those parameters can shift the inferred CO2/H2O value, so the current estimates carry systematic uncertainties that are difficult to pin down.
The JWST/MIRI post-perihelion spectra occupy a middle tier of evidence for the mixing ratio question. They clearly confirm that water remains abundant and that CO2 has not vanished, but the published analyses stop short of reporting a single updated ratio comparable to the pre-perihelion 7.6 figure. Whether the post-perihelion coma is mildly water-dominated or overwhelmingly so remains an open question.
All three key papers are preprints that have not yet completed peer review, which is standard for fast-moving cometary science but worth noting. The core measurements, JWST spectra and Subaru forbidden oxygen lines, rely on well-tested techniques, so the findings are unlikely to be overturned wholesale. The precise numbers, however, could shift as referees and independent teams scrutinize the modeling assumptions.
Two competing explanations for the flip
One plausible scenario is straightforward thermal processing. Solar heating during perihelion passage preferentially sublimated near-surface CO2 reserves, which have a lower sublimation temperature than water ice. Once the easily accessible carbon dioxide vented away, water-rich layers deeper in the nucleus began to dominate the outgassing, naturally driving the observed transition.
An alternative is surface erosion without deep depletion. In this picture, the outer CO2-rich mantle was stripped away by solar heating, exposing a more water-rich stratum underneath. The bulk interior of the nucleus might still be heavily loaded with carbon dioxide, but the active patches currently facing the Sun sample a different layer. If this is correct, the comet’s overall composition has not fundamentally changed; only the face it presents to our telescopes has.
Distinguishing between these scenarios will require continued spectroscopy as 3I/ATLAS recedes beyond 3 au, where pre-perihelion observations at similar distances can serve as a direct comparison baseline. If CO2 resurges as different regions of the nucleus rotate into sunlight, that would favor the surface-erosion model. If water dominance persists, deeper thermal processing is the more likely explanation. Spatially resolved imaging, if achievable before the comet fades, could also reveal whether activity is confined to localized jets or spread more uniformly across the surface.
Why an interstellar chemistry shift matters
Solar system comets change composition as they orbit the Sun, sometimes dramatically. But those objects formed here; their ices reflect the chemistry of our own protoplanetary disk. 3I/ATLAS formed around another star entirely, carrying a volatile inventory shaped by conditions astronomers can only guess at. Watching our Sun rewrite that alien chemistry in a matter of months offers a natural experiment that has no parallel in solar system science.
If a single perihelion passage can strip away a CO2-rich outer layer and expose water-dominated material beneath, it suggests that the surface compositions of interstellar comets detected far from the Sun may not represent their deeper, more pristine reservoirs. That has implications for how astronomers interpret future interstellar visitors: a snapshot taken at large heliocentric distance might capture a veneer rather than the bulk truth.
The nickel detection adds a separate thread. If atomic nickel proves to be a common feature of interstellar cometary comae, it could point to broadly similar metal-incorporation pathways in protoplanetary disks across the galaxy, a hint that the raw ingredients of planetary systems are more universal than parochial.
For now, 3I/ATLAS is still within reach of the world’s largest telescopes, fading but not yet gone. Every additional spectrum taken as it retreats will tighten the constraints on how much our Sun changed an object that spent most of its existence in the cold between stars. The comet arrived as a CO2-rich stranger. It appears to be leaving with a water-dominated face, reshaped by a star it will never orbit again.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
