Discovery of Two New Interstellar-Meteor Candidates
In a new paper that I co-authored with my brilliant postdoc Richard Cloete, available here, we report the identification of two previously unrecognized interstellar meteor candidates in NASA’s CNEOS fireball database.
By exploiting an empirically calibrated uncertainty model from 2025 (reported here), we find two events that robustly exceed the escape velocity from the Solar System. CNEOS-22 (detected on 2022–07–28 in the eastern tropical Pacific) exceeds escape by 8.7 standard-deviations and CNEOS-25 (2025–02–12) exceeds escape by 5.5 standard deviations. For both events, none of a million Monte-Carlo realizations yield an orbit that is gravitationally bound to the Sun. The adopted error model would need to underestimate the true uncertainties by factors of 5–9 for either candidate’s unbound status to be marginal.
The discoveries of 1I/`Oumuamua, 2I/Borisov, and 3I/ATLAS demonstrated that interstellar objects larger than a football field, transit the inner Solar System. Although smaller bodies evade telescope detection, they can reveal themselves as fireballs when they enter Earth’s atmosphere at speeds exceeding the local escape velocity from the Solar System.
The CNEOS fireball database, maintained here by NASA’s Jet Propulsion Laboratory, is a global, space-based catalog providing event-level velocity vectors for bolide detections from U.S. Government sensors. These velocity measurements enable computation of heliocentric orbits and testing whether an impactor was gravitationally unbound to the Sun. However, CNEOS publishes no per-event uncertainties, and the accuracy of reported velocities has varied across sensor generations.
Previous interstellar meteor reports have focused on IM1 from 2014–01–08 and IM2 from 2017–03–09, based on nominal heliocentric speeds exceeding escape velocity. Independent analysis of IM1 by the U.S. Space Command confirmed its interstellar identification (as reported here), but some astronomers raised concerns about the velocity accuracy of pre-2018 CNEOS data — arguing that without a calibrated uncertainty model, it is impossible to distinguish genuinely unbound trajectories from measurement artifacts.
The situation improved substantially with a 2025 empirical calibration (reported here) which cross-matched CNEOS events with independent ground-truth networks. This analysis reveals two regimes: a high-discrepancy regime for pre-2018 events and a low-discrepancy regime for post-2018 events. This calibration provides, for the first time, an empirically grounded basis for formal statistical assessment of interstellar candidacy.
We analyzed the complete CNEOS fireball catalog of events with reported latitude, longitude, altitude, and three-component velocity vector. For each event, the Earth-fixed velocity was transformed to inertial geocentric coordinates, accounting for Earth rotation, precession, nutation, and polar motion. After the Earth’s gravitational influence was removed, the heliocentric velocity was inferred by subtracting the Earth’s heliocentric velocity at the event time. An event was classified as an interstellar candidate when its speed exceeded the escape speed from the Solar System, given the Earth-Sun separation at the event time.
Restricting attention to the post-2018 low-discrepancy era, two events emerge as statistically-robust interstellar candidates. Both have positive heliocentric specific energy at their nominal velocities and remain unbound to the Solar System in all of a million Monte-Carlo realizations. Neither has been previously identified as an interstellar candidate.
CNEOS-22: 2022–07–28
This fireball occurred at 01:36:07 UTC over the eastern tropical Pacific Ocean, approximately 600 kilometers west of Peru, at an altitude of 37.5 kilometers. Its impact energy was equivalent to 0.69 kiloton of TNT and its impact speed was 30 kilometers per second.
The heliocentric speed of this bolide is 46.98 kilometers per second, exceeding the solar escape speed of 41.79 kilometers per second. Under the Monte Carlo analysis, the deviation is statistically significant at a level of 8.7 standard deviations. None of the million draws yield an orbit bound to the Solar System.
The impact energy and speed imply a bolide with a mass of 6.4 metric tons and a radius of about 0.9 meters for a solid density of 2 grams per cubic centimeter. The air’s ram-pressure at the fireball’s peak-brightness altitude was 5.2 megapascals.
CNEOS-25: 2025–02–12
This fireball occurred at 04:33:39 UTC over the Barents Sea, between Novaya Zemlya and Franz Josef Land in the high Arctic, at 42-kilometer altitude. Its impact energy is equivalent to 0.13 kiloton of TNT and its impact speed is 22 kilometers per second.
The heliocentric speed is 45.63 kilometers per second, exceeding the escape speed of 42.4 kilometers per second with a difference that is 5.5 standard deviations. None of the million Monte-Carlo draws yield a bound orbit.
The impact energy and speed imply a bolide with a mass of 2.2 metric tons and a radius of about 0.6 meters for a solid density. The air’s ram-pressure at the fireball’s peak-brightness altitude was 1.5 megapascals.
***
The diameters of both objects, 1.8 meters for CNEOS-22 and 1.2 meters for CNEOS-25, are comparable to the height of an adult person or a 7-year-old kid respectively.
The heliocentric interstellar speeds of the two candidates, 21.5 kilometers per second for CNEOS-22 and 16.9 kilometers per second for CNEOS-22, sample the low end of the velocity distribution expected for objects from the Galactic solar neighborhood. For comparison, 1I/`Oumuamua entered the Solar System with 26 kilometers per second and 2I/Borisov with 32 kilometers per second, whereas 3I/ATLAS entered with 58 kilometers per second. The lower excess speeds of the fireball candidates are qualitatively consistent with the expectation that smaller, fainter impactors are preferentially detected at lower encounter speeds, as long as their atmospheric luminosity is sufficient to trigger space-based sensors.
CNEOS-22 occurred over the open Pacific at 3–4 kilometer depth, presenting deep-ocean recovery challenges similar to the IM1 expedition under my leadership with its results summarized here. The higher impact energy of CNEOS-22 suggests a more massive impactor with potentially more recoverable debris than CNEOS-25.
However, CNEOS-25’s Arctic location (Barents Sea) offers the advantage of a shallow continental shelf (200–400 meters) but the constraints of sea-ice logistics. Crucially, CNEOS-25 is recent (February 2025), and rapid mobilization could maximize recovery prospects before material redistribution by ice drift and currents.
***
We live in exciting times, with multiple opportunities to study interstellar packages that just arrived in our mailbox here on Earth. Future expeditions hold the promise of revealing the contents of these packages. Rest assured, I will be the first to board any interstellar expedition ship.
ABOUT THE AUTHOR
Avi Loeb is the head of the Galileo Project, founding director of Harvard University’s — Black Hole Initiative, director of the Institute for Theory and Computation at the Harvard-Smithsonian Center for Astrophysics, and the former chair of the astronomy department at Harvard University (2011–2020). He is a former member of the President’s Council of Advisors on Science and Technology and a former chair of the Board on Physics and Astronomy of the National Academies. He is the bestselling author of “Extraterrestrial: The First Sign of Intelligent Life Beyond Earth” and a co-author of the textbook “Life in the Cosmos”, both published in 2021. The paperback edition of his new book, titled “Interstellar”, was published in August 2024.
https://avi-loeb.medium.com/
https://www.youtube.com/@ProfessorAviLoeb
https://open.spotify.com/show/1zhndXkvSY2b8FdjspFpCd
https://x.com/ProfAviLoeb
