Project T16 finds 10,091 new exoplanet candidates in the TESS archive
A team of astronomers analysed the light curves of more than 54 million stars from the first year of the TESS mission and, using the transit method, picked out 10,091 new candidates, to be published in the Astrophysical Journal Supplement.
The team of astronomers behind project T16 announced on 13 May 2026 that, by reanalysing light curves from the first year of the Transiting Exoplanet Survey Satellite (TESS), they had found 10,091 previously unrecorded candidate exoplanets, bodies that after further checks could be confirmed as planets outside our solar system. The preprint with the results is available on arXiv under the reference 2604.18579, with the final paper due in the Astrophysical Journal Supplement.
The TESS mission, which NASA runs in partnership with the MIT Kavli Institute for Astrophysics and Space Research, has been active since 2018 and is designed to find planets in orbit around bright, relatively nearby stars. The T16 team decided to go back to data from the first year of the mission, which TESS collected between July 2018 and July 2019, because in the meantime signal-processing techniques have been developed that allow transits to be detected around stars up to 16 times fainter than TESS routinely handled in real time.
The methodology rests on what is known as the transit method. When a planet passes in front of its host star from the observer’s line of sight, it blocks a small share of the starlight and leaves a periodic dip in the light curve. The T16 team analysed light curves from more than 54 million stars and identified 11,554 signals resembling transits. After removing 1,052 previously confirmed planets and 411 signals with fewer than three transits recorded, 10,091 new candidates with at least three transit events remained.
What this number does NOT mean
The “candidate” label matters methodologically. The fact that a signal has been detected does not mean a planet has been confirmed. A candidate can turn out to be an instrumental error, a binary star in the system or a pulsation of the host star. According to TESS statistics to date, about 60 percent of candidates pass all subsequent checks and are confirmed as real planets, while the rest fall out as false signals. In other words, of the current 10,091 candidates, statistically a few thousand are expected to be confirmed after further observations from ground-based telescopes and radial-velocity measurements.
In practical terms, the T16 team note that most of the candidates orbit stars cooler and smaller than the Sun, known as red dwarfs. That means the planets sit in very close orbits, with transits that are frequent and easy to detect, but they are correspondingly very hot, which in principle takes them away from the habitable zone. According to the NASA Exoplanet Archive, of the roughly 6,000 confirmed exoplanets to date, about half were recorded thanks to TESS, which makes this systematic reanalysis of a historical sample one of the largest one-off additions to the catalogue.
The next step will be to confirm the candidates through observing programmes with instruments such as HARPS-N on La Palma and NEID on Kitt Peak in Arizona. Only once the host star’s radial velocity has been measured is it possible to estimate the planet’s mass and, if confirmed, classify it as a rocky or gaseous world. Most candidates on the T16 list lie at distances between 100 and 1,500 light-years from Earth, putting them within reach of spectroscopic analysis by the James Webb telescope in the coming observing cycles.
Methodologically, T16 belongs to a family of projects that lean on rereading the archive rather than new observations. Similar approaches have been applied before to the Kepler archive: the best-known programme of this kind was the K2 extension from 2014 to 2018, which picked out more than 500 additional confirmed planets from data already collected. The Kepler archive experience suggests that most of the additional candidates are turned into confirmed planets over the following five to seven years, which means that the definitive tally for T16 may only be available between 2031 and 2033, if funding for the next observing programmes holds up.
The long-term significance of project T16 is also that, even after more than seven years of operation, the TESS archive still contains a large number of undocumented signals. That means that, even when the mission is formally over, it will remain a source of scientific discovery for many years to come. According to the MIT Kavli Institute, the total volume of TESS data collected by May 2026 stands at more than two petabytes, around 500 times the size of the Hubble archive for comparison.
