Astronomers have developed a new technique called FLIP that combines a digital replica of the James Webb Space Telescope with machine learning to uncover exoplanets hidden in stellar glare, successfully detecting the previously hidden giant planet 51 Eridani b in JWST data.
The method, described in a paper published Nov. 4 in the journal Nature, addresses a fundamental challenge in exoplanet imaging: distinguishing a faint planet from the overwhelming light of its star. Planets are typically a million times fainter than their stars, study first author Rodrigo Ferrer-Chavez, an astronomer at Northwestern University, said in a statement. To image exoplanets, he said, we need to remove that glare.
Telescopes like JWST use coronagraphs to block some starlight, but even the best coronagraphs cannot eliminate all the glare, and the telescope itself vibrates subtly during observations. These tiny movements create mismatches between the image and what the instrument actually sees, enough to obscure even large planets.
To overcome this, Ferrer-Chavez and his team built a digital twin of JWST that recreates how light travels through the telescope's detectors without any movement-related imperfections. They fine-tuned the replica using computational models based on real physics, alongside machine learning techniques similar to those used to train AI. However, the authors stressed that this is not an AI model. Because we already know the physics, we don't need AI to understand what's happening, study co-author Jason Wang, also an astrophysicist at Northwestern, said in a statement.
The researchers named their approach Physics-based Learning for Imaging Planets, or FLIP. They applied it to data on three stars, including 51 Eridani, about 97 light-years from Earth, which was already suspected of hosting a giant exoplanet. That planet, 51 Eridani b, had been detected by ground-based telescopes but was drowned out in previous JWST observations. With FLIP, it popped out of the dataset.
The scientists hope the technique will help identify more hard-to-spot exoplanets, particularly ones that resemble planets in our own solar system. In the nearly 35-year history of exoplanet hunting, the vast majority of found planets have been radically different from those in our solar system, partly because larger and more reflective worlds are easier to see with current telescopes. It's not that [solar-system-like planets] don't exist, Ferrer-Chavez said; it's that we can't see them.
How it's landing
Posts on X from people the desk follows.
Analysis
Why This Matters
- The new method could uncover a population of exoplanets that are otherwise invisible to current telescopes, potentially including worlds more like those in our solar system.
- By removing the limitations of telescope stability, FLIP enhances the science return of JWST without requiring hardware changes.
- The approach demonstrates how combining physics-based models with machine learning can solve observational problems, offering a template for other instruments.
Background
The James Webb Space Telescope, launched in 2021, is the most powerful space telescope ever built, designed to observe the universe in infrared light. One of its key science goals is direct imaging of exoplanets, taking actual pictures of planets orbiting other stars. Coronagraphs block most starlight, but residual glare and tiny vibrations from the telescope's moving parts still obscure faint planets. Previous direct imaging efforts have mostly found young, massive, self-luminous planets that are easier to see, while older and smaller planets like those in our solar system remain hidden.
Key Perspectives
Rodrigo Ferrer-Chavez (Northwestern University, study first author): says planets are a million times fainter than their stars, so to image exoplanets astronomers need to remove that glare. He adds it's not that solar-system-like planets don't exist, it's that we can't see them.
Jason Wang (Northwestern University, study co-author): says because we already know the physics, we don't need AI to understand what's happening.
What to Watch
- Whether FLIP is applied to more JWST datasets and uncovers additional hidden exoplanets, especially smaller or older ones.
- Potential adaptation of the digital twin approach to other telescopes, such as ground-based observatories or future space missions like the Nancy Grace Roman Space Telescope.
- Follow-up studies to characterize the atmosphere and orbit of 51 Eridani b using FLIP-enhanced JWST data.
Checked
12 claims in this story checked against 1 source: 12 supported.
Show each claim
- SupportedA new method removes glare from a James Webb image to reveal a hidden exoplanet.Latest from Live Science: “Using a combination of modeling and machine learning techniques, they picked out a previously hidden giant exoplanet in data from JWST.”
- SupportedAstronomers used a digital twin of JWST and machine learning to spot a giant planet previously drowned out by starlight.Latest from Live Science: “To address this, Ferrer-Chavez and his team developed a "digital twin" of JWST that virtually recreates how light travels through the telescope's detectors, minus any movement-based imperfections. To fine-tune this digital replica, the researchers used computational models based in real-world physics alongside machine learning techniques”
- SupportedThe researchers named the technique FLIP and used it to detect 51 Eridani b in JWST data.Latest from Live Science: “The researchers named their approach Physics-based Learning for Imaging Planets, or FLIP. Using FLIP, the team analyzed data from three stars, including 51 Eridani”
- SupportedThe paper describing the method was published Nov. 4 in Nature.Latest from Live Science: “The results were published Nov. 4 in the journal Nature .”
- SupportedFerrer-Chavez said planets are typically a million times fainter than their stars and that imaging them requires removing the glare.Latest from Live Science: “"Planets are a million times fainter than their stars," study first author Rodrigo Ferrer-Chavez , an astronomer at Northwestern University, said in a statement . "So, to image exoplanets, we need to remove that glare."”
- SupportedJWST coronagraphs block some starlight but cannot remove all the glare, and slight telescope movements can obscure large exoplanets.Latest from Live Science: “they can't block all the glare, and the telescopes themselves move. Even the most stable telescopes vibrate almost imperceptibly as they collect light, so the images a telescope produces contain ever-so-slight mismatches compared with what the instrument actually sees. These slight shifts can be enough to obscure even very large exoplanets .”
- SupportedWang said, “Because we already know the physics, we don't need AI to understand what's happening.”Latest from Live Science: “"Because we already know the physics, we don't need AI to understand what's happening," study co-author Jason Wang , also an astrophysicist at Northwestern, said in a statement.”
- SupportedThe researchers tested FLIP on data from three stars; 51 Eridani is about 97 light-years away and was already suspected of hosting a giant exoplanet.Latest from Live Science: “Using FLIP, the team analyzed data from three stars, including 51 Eridani, which is located around 97 light-years from Earth and was already suspected to have a giant exoplanet orbiting it.”
- Supported51 Eridani b had been detected by ground-based telescopes but was drowned out in earlier JWST observations; FLIP made it stand out in the dataset.Latest from Live Science: “Although the planet, called 51 Eridani b , had been detected by ground-based telescopes, it was drowned out by light in previous JWST observations. With FLIP, however, 51 Eridani b popped out from the dataset.”
- SupportedThe scientists hope FLIP will help identify hard-to-spot exoplanets, especially ones like planets in our own solar system.Latest from Live Science: “The scientists hope this approach will help astronomers pick out even more hard-to-spot exoplanets, especially ones that look like the planets in our own backyard.”
- SupportedIn nearly 35 years of exoplanet hunting, most discovered planets have differed radically from those in our solar system, partly because larger, more reflective planets are easier to spot with current telescopes.Latest from Live Science: “In the nearly 35-year history of exoplanet hunting, the vast majority of planets we've found have been radically different from those in our own solar system , partially because larger and more reflective worlds are easier to spot with current telescopes.”
- SupportedFerrer-Chavez said, “It's not that [solar-system-like planets] don't exist,” but “It's that we can't see them.”Latest from Live Science: “"It's not that [solar-system-like planets] don't exist," Ferrer-Chavez said. "It's that we can't see them."”
