Astronomers discover hidden alien planet in James Webb images

Finding these planets is one challenge. Actually seeing them is another. Most exoplanets orbit stars that shine so brightly that the planets become nearly impossible to distinguish from the surrounding glare. Even the most advanced space telescopes struggle to separate the faint light of a distant planet from the overwhelming brightness of its host star.

Now, researchers have developed a powerful new method for removing that starlight from telescope images, revealing planets that were previously hidden. The technique uses a sophisticated computer simulation of the James Webb Space Telescope (JWST) and has already exposed an exoplanet that was concealed in existing observations. In tests involving known planets, it achieved two to 15 times greater detection sensitivity than conventional approaches.

A Digital Twin of James Webb Reveals Hidden Planets

The breakthrough centers on a virtual replica of Webb’s Near Infrared Camera (NIRCam), an instrument designed to observe the universe in infrared light. By recreating how the camera operates, researchers can identify and correct tiny imperfections that interfere with its observations.

The virtual model, often called a “digital twin,” allows scientists to simulate the telescope’s behavior and determine how starlight spreads through its optical system. Once they understand the patterns produced by the star, they can digitally subtract much of that glare, making faint planets easier to detect.

“Directly imaging exoplanets is incredibly challenging because even the tiniest imperfections in a telescope can make it difficult to separate the faint light of a planet from the overwhelming glare of its star,” explains Caltech’s Katie Bouman, professor of computing and mathematical sciences, electrical engineering and astronomy, a Rosenberg Scholar, and a Heritage Medical Research Institute Investigator, who is a co-author of a Nature paper about the new method published online November 4.

“What’s exciting is that we’re showing how computation can help overcome limitations of the telescope itself,” Bouman continues. “By combining a physical model of the instrument with optimization techniques from machine learning, we have been able to remove starlight much more precisely, making it easier to detect faint exoplanets and even reveal one that was previously hidden in the JWST data.”

How Scientists Make Blinding Starlight Disappear

The researchers named their approach Physics-based Learning for Imaging Planets (FLIP). Rather than relying entirely on improvements to telescope hardware, the system uses advanced computation to correct problems in images that have already been collected.

To develop FLIP, the team created a detailed physical model of Webb’s infrared camera. They then used computational techniques to examine millions of possible imperfections, subtle changes, and other unknown factors that could influence the telescope’s observations.

These small variations matter because even an extremely slight distortion can alter how a star appears in an image. When astronomers attempt to remove the star’s light, any remaining glare can overwhelm the much dimmer signal from a nearby planet.

By accounting for these effects, FLIP produces a highly precise model of the star’s appearance. Researchers can then subtract that modeled starlight from the original observations, allowing previously obscured objects to become visible.

“This is the first time that we have demonstrated that these little imperfections or instabilities are not a limiting factor in the kinds of planets we can image,” says lead author Rodrigo Ferrer-Chavez, a graduate student at Northwestern University in the lab of astronomer Jason Wang, who is also a co-author of the paper. “As you go through this iterative process of trying to find what the telescope actually is doing during observations, you basically see the glare of the star disappearing in very bizarre ways. It gets dimmer and very deformed until at the end you get a pretty clean image and, if you’re lucky, there’s a blob there, which is a glowing planet.”

A Failed Experiment Led to a Major Breakthrough

The idea behind FLIP began taking shape when Ferrer-Chavez was an undergraduate student at the Autonomous University of Yucatán in Mexico. He had already participated in a Summer Undergraduate Research Fellowship project with Wang, who was then a 51 Pegasi b Postdoctoral Fellow (named after the first exoplanet discovered orbiting a Sun-like star) in the laboratory of Dimitri Mawet.

Mawet is the David Morrisroe Professor of Astronomy at Caltech and a senior research scientist at NASA’s Jet Propulsion Laboratory, which is managed by Caltech. He also serves as director of instrumentation for the Caltech Optical Observatories.

At the time, Ferrer-Chavez was looking for additional opportunities to pursue research. Meanwhile, Wang had been discussing a potential collaboration with Bouman that would combine her expertise in computational imaging with his knowledge of astronomical instruments. Their shared goal was to improve the ability of telescopes to directly image exoplanets.

The two researchers brought Ferrer-Chavez into the project and began mentoring him. However, the initial effort did not produce the results they had hoped for.

“Rodrigo’s senior thesis was an exploration of possible approaches, none of which ended up being what we did for this paper, but it helped rule out what not to do and also helped us understand the problem better,” Wang says. “I believe our recent success was only possible because we did that exercise first.”

Ferrer-Chavez originally attempted to train machine learning models to predict the patterns of starlight appearing in telescope data. That strategy proved unsuccessful, prompting the team to rethink how the technology could be used.

Instead of asking artificial intelligence to predict what the glare should look like, they turned to computational tools developed for machine learning to perform detailed physical calculations.

These methods allowed the researchers to adjust millions of variables, much like turning an enormous collection of control knobs. Each adjustment helped refine their simulation of the telescope until it more closely reproduced the actual observations. Performing this many calculations manually would be impractical.

The resulting approach combines the physical laws governing light and telescope optics with the computational power of modern optimization methods. Rather than simply guessing what should be removed from an image, FLIP uses a model of how the telescope produces that image in the first place.

James Webb’s Archives May Contain More Hidden Worlds

One of the most exciting possibilities is that astronomers may not need new telescope observations to benefit from the discovery.

Webb has already collected enormous amounts of astronomical data, including images of stars that could harbor planets too faint to detect using earlier analysis methods. FLIP offers a way to revisit these observations and search for objects that were overlooked.

“We now have a tool that lets us discover and study planets that we were not able to ‘see’ before, even if that planet was already in an image in the JWST archive,” Ferrer-Chavez explains. “We can try to find planets that may have been missed there, and also planets that perhaps we tried to image, but we could not actually detect at the time.”

This means some planets that were previously considered beyond the reach of direct imaging could potentially be identified without sending another telescope into space.

It also changes how astronomers can think about existing observations. An image that appears to contain nothing but a brilliant star may hold additional information about planets hidden beneath its glare.

Pushing Telescopes Beyond Their Physical Limits

Astronomers have spent decades developing technology to reduce the overwhelming brightness of stars. One important tool is a coronagraph, an instrument designed to block starlight so that faint objects nearby can be observed more clearly.

For example, NASA’s recently launched Nancy Grace Roman Space Telescope carries a Coronagraph Instrument developed by researchers at Caltech, JPL, and IPAC (a science and data center for astronomy at Caltech). The instrument is specifically designed to suppress starlight and improve the direct imaging of exoplanets.

But even exceptionally precise telescope hardware has unavoidable limitations. Wang explains that physical instruments alone cannot match the level of correction achieved by FLIP when processing the images.

“The James Webb Space Telescope, which we demonstrated on, is optically stable to the nanometer level, which just sounds pretty awesome, but that’s still limiting,” he says. “We’re excited because this technique is not limited specifically to JWST. Because of some very specific design, this technique could be used on any telescope.”

A nanometer is one billionth of a meter, meaning Webb’s optical system maintains extraordinary precision. Yet even disturbances on this incredibly small scale can interfere with the detection of planets next to bright stars.

FLIP offers a different way around that problem. Instead of requiring the telescope itself to eliminate every tiny imperfection, the technique mathematically accounts for those effects after the observations have been made.

The researchers now intend to refine FLIP and investigate how it could work with additional instruments, including the Roman Space Telescope and telescopes operating from Earth’s surface.

A Powerful New Tool in the Search for Another Earth

The implications may extend well beyond today’s observatories. The researchers believe their approach could help shape future space missions, particularly those designed to search for planets resembling Earth.

One major possibility is NASA’s proposed Habitable Worlds Observatory (HWO). The mission is intended to become the first telescope specifically designed to search for evidence of life on planets orbiting other stars.

A central objective would be to photograph Earth-like planets within the habitable zones of stars similar to our Sun. A habitable zone is the range of distances around a star where temperatures could allow liquid water to exist on a planet’s surface, assuming suitable atmospheric conditions.

But the technical challenge is enormous. An Earth-like planet orbiting a Sun-like star can be approximately 10 billion times fainter than the star itself, making the planet extraordinarily difficult to distinguish from the surrounding light.

“The goal of HWO is to find Earth-like planets and the habitable zones of Sun-like stars through a direct imaging method,” Ferrer-Chavez says. “The engineering challenge of trying to image something that is around 10 billion times fainter than something right next to it is a brutal — not just to clean the image but to build a telescope that can take that image.”

FLIP could make that challenge more manageable. If astronomers can digitally remove more unwanted starlight after capturing an image, future telescope designs may not need to overcome every source of optical instability through hardware alone.

The approach could therefore give engineers greater flexibility while allowing astronomers to extract more information from the light their instruments collect.

“I feel we just unlocked a new frontier,” Wang says. “We’ve now gone beyond the telescope stability limit for the first time with this new technique, and this new paradigm, I think, can be applied to every instrument that’s trying to image exoplanets.”

Bringing the Search for Alien Life Into Focus

For Bouman, the long-term ambition is much larger than simply finding more distant planets. Astronomers ultimately hope to obtain direct images of worlds resembling Earth and examine them for possible signs of life.

Reaching that goal will require major advances in telescope engineering, astronomical instruments, and image processing. But the new research demonstrates that some of the most important improvements may come from better ways of interpreting data rather than building increasingly precise hardware.

“There’s still a long way to go, but this work shows how much more is possible when we combine advances in telescope technology with computational imaging,” Bouman says. “Sometimes, the key to seeing something new isn’t just building a better instrument but finding a better way to make sense of the light it collects.”

The findings were published in Nature in a paper titled “Imaging exoplanets below the telescope stability limit.” Additional co-authors include William T. Freeman and Brandon Y. Feng of MIT, along with Laurent Pueyo and Marshall D. Perrin of the Space Telescope Science Institute in Baltimore.

The research received support from the Space Telescope Science Institute, the Sloan Foundation, an Amazon AI4Science Discovery Award, and the National Science Foundation.