Warwick's RAVEN AI Pipeline Validates 118 New Exoplanets in TESS Data, Adding Nearly 1,000 Fresh Candidates

Overview

Astronomers at the University of Warwick have used a new AI pipeline called RAVEN to comb through more than 2.2 million stars observed during the first four years of NASA’s TESS mission, statistically validating 118 new transiting planets and flagging nearly 1,000 high-quality candidates that had not previously been catalogued, according to a Warwick press release. The work is published in the Monthly Notices of the Royal Astronomical Society and centres on planets that complete an orbit in less than 16 days.

What We Know

The pipeline, named RAVEN — for “RAnking and Validation of ExoplaNets,” as Phys.org reports — was designed to handle every step of the planet-finding workflow, from detecting transit signals to filtering out false positives and producing a statistical validation. “In addition, RAVEN is designed to handle the whole process in one go, from detecting the signal, to vetting it with machine learning and statistically validating it,” Andreas Hadjigeorghiou, who led the pipeline’s development, said in the Warwick release.

The central challenge in transit surveys is that not every dip in starlight is a planet. “The challenge lies in identifying if the dimming is indeed caused by a planet in orbit around the star or by something else, like eclipsing binary stars, which is what RAVEN tries to answer,” Hadjigeorghiou said in the same release. NASA’s Transiting Exoplanet Survey Satellite, or TESS, as Space.com explains, “spots exoplanets by recording the tiny dips in starlight they cause when they pass in front of the face of the parent star.”

The headline result is summarised by lead author Marina Lafarga Magro, a postdoctoral researcher at Warwick: “Using our newly developed RAVEN pipeline, we were able to validate 118 new planets, and over 2,000 high-quality planet candidates, nearly 1,000 of them entirely new,” she said in the Warwick release. The peer-reviewed results appear in the team’s MNRAS paper, “Automatic search for transiting planets in TESS–SPOC FFIs with raven: over 100 newly validated planets and over 2000 vetted candidates,” published in volume 548, issue 3 of the journal.

Methodology

RAVEN was applied to data covering more than 2.2 million stars from TESS’s first four years of operation, the Warwick release notes. The team focused on planets with orbital periods of less than 16 days, as the Warwick release explains.

Accompanying papers turned the validated catalogue into a population study. The team reports that roughly 9 to 10 percent of Sun-like stars host close-in planets, with uncertainties about ten times smaller than previous Kepler-based measurements, according to the EurekAlert summary of the work. A separate analysis examined the so-called “Neptunian desert,” a region of parameter space where Neptune-sized planets on short orbits are unusually rare; the team estimates these planets are present around about 0.08 percent of Sun-like stars, as detailed in the Warwick release.

“For the first time, we can put a precise number on just how empty this ‘desert’ is,” Kaiming Cui, who led the desert study at Warwick, said in the Warwick release.

Why It Matters

The project is positioned as a way to convert TESS’s vast haul of light curves into a statistically clean sample. “RAVEN allows us to analyse enormous datasets consistently and objectively. Because the pipeline is well-tested and carefully validated, this is not just a list of potential planets — it is also reliable enough use as a sample to map the prevalence of distinct types of planets around Sun-like stars,” senior author David Armstrong, an associate professor at Warwick, said in the same release.

Among the validated and candidate worlds are ultra-short-period planets that complete an orbit in less than a day, along with examples drawn from the Neptunian desert, ScienceDaily reports. The MNRAS authors — M Lafarga, D J Armstrong, K Cui, A Hadjigeorghiou, V Kunovac, L Doyle, E M Bryant, R F Díaz, L A Nieto and A Osborn — describe the run as a step toward improving the candidate sample TESS leaves behind for follow-up spectroscopic confirmation, according to the journal listing.

What We Don’t Know

The paper validates planets statistically rather than confirming them with radial-velocity mass measurements, so individual masses, atmospheres and detailed orbital architectures remain to be characterised on a target-by-target basis. The roughly 1,000 newly surfaced candidates that have not previously been catalogued as TESS Objects of Interest will require additional follow-up before they can be elevated to validated status, the Warwick release notes. The team’s occurrence-rate measurements also remain limited to the short-period regime the pipeline was tuned for; longer-period planets and smaller, Earth-like worlds at habitable-zone distances are outside the current scope.