Scientists have successfully detected an exoplanet by using physics-based modeling to strip away the intense glare of its host star, according to research published in the journal Nature.
The team demonstrated this technique using infrared data from the James Webb Space Telescope (JWST), showing that the method can reveal worlds previously obscured by stellar light.
Astrophysicists typically struggle to image exoplanets because the light from a host star is often at least 100,000 times brighter than the planet orbiting it. While coronagraphs can block some direct starlight, residual glare often remains. Traditional image processing attempts to remove this by subtracting a reference star’s light, but small vibrations or optical distortions in the telescope can undermine these efforts.
To solve this, researchers at Northwestern University and several other U.S. institutes developed a technique called Physics-based Learning for Imaging Planets (FLIP).
Rather than relying on a reference star, they built a digital twin of the JWST’s Near-InfraRed Camera (NIRCam) to simulate how the telescope’s specific optics shape incoming light. By adjusting millions of parameters to match the real-world data, the model can accurately reproduce and remove the stellar glare.
Detecting 51 Eridani b in JWST Data
The researchers tested their technique on two known planetary systems: HIP 65426 and 51 Eridani. The planet 51 Eridani b, a gas giant roughly 11 times the mass of Jupiter, was initially discovered in 2015 using ground-based telescopes. While it had not appeared in previous JWST observations due to the host star’s overwhelming signal, the FLIP model successfully extracted the planet’s light.
The new technique outperformed standard empirical reference-star processing, known as KLIP, by doubling the signal-to-noise ratio in specific infrared filters. It even enabled the detection of 51 Eridani b at a 4.3-micrometer wavelength—a band corresponding to carbon dioxide absorption that is inaccessible from Earth.
Future Potential for Earth-Like Worlds
The research team anticipates that FLIP could be instrumental in identifying planets that resemble a young Saturn, which often remain hidden in stellar glare.
“JWST is the first observatory with the capabilities to directly image planets like this, but young Saturn analogues typically remain hidden in the glare of the star,” Rodrigo Ferrer-Chavez of Northwestern University said. “Our technique can drastically help uncover planets like these if they are indeed hiding.”
Beyond gas giants, the team suggests this physics-based approach could eventually assist in the discovery of "exo-Earths." These planets are estimated to be 10 billion times fainter than their host stars. By refining how telescopes account for optical instabilities, the researchers believe their method might simplify the hardware requirements for NASA’s proposed Habitable Worlds Observatory.
How FLIP Differs from Traditional Telescope Image Processing
How does FLIP differ from traditional telescope image processing?
Traditional methods subtract light by comparing the target star to a separate reference star. FLIP instead creates a high-fidelity digital twin of the telescope’s own optics to simulate and remove the specific glare patterns caused by the telescope’s unique mechanical instabilities.
Why is the 4.3-micrometer wavelength detection significant?
This wavelength is critical because it captures carbon dioxide absorption bands, which cannot be observed from the ground. Scientists plan to use this data to analyze the chemical composition of exoplanet atmospheres and better understand how those planets formed.
Could this technique work on all existing space telescope data?
The researchers said that more than 100 stars have already been observed by JWST, and they expect the discovery potential for these existing surveys to be high. By injecting fake planet signals into data, the team found that FLIP can be up to 15 times more sensitive than previous methods in detecting new planets.