Introduction
An interstellar object is a body that was not formed around the Sun and passes through the Solar System on an unbound trajectory. The defining observational signature is a heliocentric orbit with eccentricity greater than 1. Three such objects have been confirmed since 2017, each carrying an "I" (interstellar) number: 1I/'Oumuamua, 2I/Borisov and 3I/ATLAS. Interstellar comets offer direct samples of the volatiles in distant protoplanetary discs, material otherwise known only from unresolved observations of other stars. This report reviews what is established about each, using the current orbit solutions from the JPL Small-Body Database (SBDB) and the peer-reviewed literature.
1I/'Oumuamua: discovery and orbit
'Oumuamua was discovered on 19 October 2017 by the Pan-STARRS1 survey in Hawai'i. Follow-up astrometry quickly showed a hyperbolic orbit with eccentricity of about 1.2, an object approaching from the direction of the constellation Lyra with an asymptotic inbound velocity of about 26 km/s. The current JPL solution, based on 207 observations spanning 14 October 2017 to 2 January 2018, gives e = 1.201, a perihelion distance of 0.256 AU reached on 9 September 2017, and an inclination of 122.7 degrees. By the time of discovery it was already outbound.
Its physical properties were unlike either expectation for an interstellar comet. Deep imaging with large telescopes found no coma and no dust, and visible-wavelength spectra showed no identifiable gas emission. The brightness varied by a factor of about ten over a rotation of roughly 7.3 hours, implying an extremely elongated shape (an axis ratio of order 10:1 in the original Nature analysis, Meech et al. 2017) and a tumbling rotation state. Its surface colour is red, comparable to outer Solar System bodies. With no detectable activity, it was briefly classified as a comet and then reclassified as an asteroid.
The non-gravitational acceleration debate
Micheli et al. (2018) analysed 177 ground-based and 30 Hubble Space Telescope astrometric positions and found that no purely gravitational trajectory could fit the full arc. Adding a radial acceleration away from the Sun, scaling as r-2 (or r-1), resolved the residuals. At 1 AU the fitted term was (4.92 ± 0.16) × 10-6 m/s2, a detection at about 30 sigma.
The authors tested and rejected several non-cometary causes. Solar radiation pressure would require a bulk density three to four orders of magnitude below that of comparable Solar System asteroids. The Yarkovsky effect is too weak and acts mainly along-track. Drag-like forces would oppose the direction of motion, which the fit contradicts. A single impulsive velocity change fits worse than continuous acceleration, and no companion or fragment was seen. They concluded that comet-like outgassing was physically viable, while noting the tension: the required outgassing implied that 'Oumuamua would need to be depleted in CN by more than a factor of about 15 relative to water, and predicted dust that would only escape detection if the grains were larger than a few hundred microns.
That tension drove a range of alternative models. Bergner and Seligman (2023) summarise the problem: energetic arguments favour hyper-volatile species over water or CO2, but models based on pure H2, N2 or CO ice all face theoretical or observational inconsistencies, and non-outgassing explanations require fine-tuned formation or unrealistic numbers of progenitors. Their proposed resolution is the release of molecular hydrogen trapped in water ice, produced by cosmic-ray irradiation during the interstellar journey and freed as the amorphous ice was warmed near the Sun. In this picture 'Oumuamua began as an icy planetesimal broadly similar to Solar System comets. Because no gas was ever detected directly from 'Oumuamua, competing explanations remain under discussion.
2I/Borisov: an unmistakable comet
The second interstellar object was reported as a comet candidate on 30 August 2019 by Gennady Borisov, observing from MARGO, Nauchnij, in Crimea. Guzik et al. (2020) identified the hyperbolic orbit in public astrometry and derived an excess speed of about 32 km/s. The current JPL solution, from 964 observations spanning February 2019 to September 2020, gives e = 3.356, a perihelion of 2.007 AU on 8 December 2019 and an inclination of 44.1 degrees.
In contrast to 'Oumuamua, Borisov showed an extended coma and a faint, broad tail in follow-up images taken on 10 and 13 September 2019. Its colour (g'-r' = 0.66 ± 0.01) and dust morphology led Guzik et al. to conclude that, setting the orbit aside, it appeared indistinguishable from native Solar System comets. Hubble imaging (Jewitt et al. 2020) constrained the nucleus radius to between 0.2 and 0.5 km.
The composition was not entirely ordinary. Ultraviolet observations by Bodewits et al. (2020) showed that the coma contained more carbon monoxide than water, with a CO abundance of at least 173 percent relative to H2O, more than three times higher than previously measured for any comet in the inner (within 2.5 AU) Solar System. The authors describe this as a first glimpse of the ice content of another star's protoplanetary disc, one substantially different from the Sun's.
3I/ATLAS: the fastest visitor so far
3I/ATLAS (C/2025 N1) was first reported to the Minor Planet Center on 1 July 2025 by the ATLAS survey telescope in Chile, with pre-discovery images extending back to 14 June 2025. The initial characterisation (Seligman et al. 2025) gave e of about 6.1, a perihelion near 1.36 AU, a retrograde inclination of about 175 degrees and a hyperbolic velocity of about 58 km/s. The JPL solution dated 21 October 2026 gives e = 6.123, a perihelion of 1.352 AU on 29 October 2025, and an inclination of 175.2 degrees. Its closest approach to Earth was about 1.8 AU, and it posed no hazard.
3I was active from discovery, resolving a compact coma in deep ground-based stacks. Hubble imaged a teardrop-shaped dust envelope on 21 July 2025. As of 20 August 2025, NASA reported that the nucleus diameter was no larger than 5.6 km and could be as small as 440 m. JWST NIRSpec observations at 3.32 AU inbound (Cordiner et al. 2025) revealed a coma dominated by CO2, with H2O, CO, OCS, water ice and dust also present. The measured CO2/H2O mixing ratio of 7.6 ± 0.3 is among the highest ever observed in a comet. The authors suggest an intrinsically CO2-rich nucleus, either from heavier radiation processing of its ices or from formation near the CO2 ice line of its parent disc.
Because perihelion occurred at low solar elongation as seen from Earth, ground-based coverage was interrupted from September 2025 until the comet reappeared in early December. An unusually broad fleet of spacecraft filled the gap, including Psyche, Lucy, MAVEN, the Mars Reconnaissance Orbiter's HiRISE camera, the Perseverance rover, Parker Solar Probe and Europa Clipper.
Comparing the three trajectories
The three orbits span a wide range. Eccentricity rises from 1.20 for 'Oumuamua to 3.36 for Borisov and 6.12 for 3I/ATLAS, and the inbound excess speed from about 26 to about 32 to about 58 km/s. 'Oumuamua passed closest to the Sun (0.256 AU); Borisov never came closer than about 2 AU. The current JPL solutions for all three objects include fitted non-gravitational acceleration terms, a reminder that even the two plainly cometary objects show measurable departures from purely gravitational motion.
Expected detection rates in the Rubin era
Three interstellar objects were found between October 2017 and July 2025. The Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST) is expected to raise the discovery rate substantially. Simulating 'Oumuamua-like objects drawn from galactic kinematics, Hoover, Seligman and Payne estimate that LSST will detect of order 15 interstellar objects over its roughly 10-year campaign, assuming no cometary activity. They also find that detectable trajectories cluster toward the solar apex and anti-apex. The authors describe these numbers as lower limits, to be updated once the number density and size distribution of interstellar objects are better constrained.
Intercept missions
ESA's Comet Interceptor, adopted in June 2022 and led by ESA in collaboration with JAXA, is designed to wait in space for a target that has not yet been found. Its three spacecraft (a primary and two probes) will be delivered to the Sun-Earth L2 point after a launch planned for late 2028 or early 2029, then sent toward a long-period comet, or potentially an interstellar object, once a suitable target is found. Reaching an interstellar object is the harder case. Hoover et al. estimate only about 0.0007 to 0.001 LSST-detected interstellar objects would be reachable with Comet Interceptor-class propulsion, compared with about 1 to 3 for the more capable proposed Bridge concept. An interstellar flyby would therefore depend on an unusually favourable discovery.
Outlook
The three known interstellar objects are strikingly different: an inert, elongated body with an unexplained acceleration, a CO-rich comet, and a fast, CO2-dominated comet. A larger LSST sample will allow population-level questions about size, composition and origin to be asked for the first time.