Earth's magnetic field is roughly half a gauss. Jupiter's radio storms come from above its poles, where the field is orders of magnitude stronger than Earth's. For more than three decades, astronomers have wanted to repeat this observation beyond the Solar System: hear a planet's own radio storms.
The difficulty is not sensitivity but attribution. A radio telescope's beam is far wider than the angular separation between a planet and its star, so a received signal cannot easily be assigned to one or the other.
In September 2026, a team led by Kevin N. Ortiz Ceballos of the Harvard-Smithsonian Center for Astrophysics posted a paper on arXiv (arXiv:2609.16720) reporting the direct localization of auroral radio emission to the exoplanet Beta Pictoris b.
Why This Star
Beta Pictoris lies 19.63 ± 0.06 parsecs away (about 63–64 light-years), and is roughly 23 million years old. The host is an A6V-type star with a debris disk and at least three known giant planets.
The largest, Beta Pictoris b, has a mass of about 12 Jupiter masses, orbits at roughly 10 AU with a period of about 24 years, and reaches a maximum angular separation of about 0.55 arcseconds from the star. It is a young planetary-mass object with spectral type L2 ± 1, effective temperature around 1700 K, and a rotation period of just eight to nine hours.
One practical reason it was chosen: the host star is radio-quiet. Only when the star is quiet can a planet be heard.
Four Observations, Two Bands
The team made four observations with the MeerKAT array in South Africa between 2025 and 2026. L-band covers 0.856–1.712 GHz; S-band covers 1.75–3.5 GHz.
In all four observations, a radio source appeared in the direction of the Beta Pictoris system. The remaining task was attribution.
Three Checks That Pinned the Position
First, reference-frame tie-down. The researchers aligned the radio images to the Gaia celestial reference frame using nine compact radio sources corresponding to Gaia-identified quasars, plus one VLBI calibrator, fitting an affine correction (translation, rotation, scale, and shear).
After the correction, the radio source agreed with Beta Pictoris b's known position: the offset corresponds to a Mahalanobis radius of 1.1, with a probability of about 0.53 for a deviation of this magnitude given combined uncertainties. Conversely, it differed from the host star by 4.4 standard deviations and from planet c by 4.8 standard deviations, with systematic uncertainties included.
Second, stability. Repeatedly dropping single and paired reference sources from the fit — 55 leave-one-out and leave-two-in fits — the radio position remained consistent with Beta Pictoris b every time. On median, the radio source sits about 100 milliarcseconds from the planet and about 434 milliarcseconds from the star, while the reference-frame correction itself introduces a median shift of only ~4 milliarcseconds.
Third, calibration of the positional uncertainty itself. The researchers injected 160 artificial point sources with the same brightness and beam shape as real sources into the images and recovered them with the same fitting pipeline; the scatter of recovered positions was 1.1 times the reported fitting uncertainty, a factor folded into the final systematic error budget.
A further test addressed ionospheric refraction: measuring 40 bright compact sources across eight S-band sub-bands showed a common chromatic shift of 9.8 ± 2.6 milliarcseconds across the field, absorbed by the reference-frame correction; the median residual between sightlines, 32 milliarcseconds, was also conservatively included in the positional uncertainty.
| Check | Method | Result | |---|---|---| | Reference-frame tie | Nine quasars plus VLBI calibrator, affine correction | Matches planet, Mahalanobis radius 1.1 | | Distinguishing from star | Systematics included | 4.4 sigma | | Distinguishing from planet c | Systematics included | 4.8 sigma | | Fit stability | 55 leave-one/two-out fits | Consistent with planet every time | | Positional uncertainty | 160 artificial sources injected | Scatter 1.1× reported value | | Ionospheric shift | 40 sources across eight sub-bands | Common shift 9.8 ± 2.6 mas |
What the Signal Looks Like
The radiation appears in bursts: rapidly varying, about 40–70% circularly polarized, with weaker persistent emission between bursts. The bursts are broadband with a relatively flat spectrum spanning both observing bands.
These features point to the electron cyclotron maser instability. Energetic electrons moving along field lines convert energy into radio emission above the planet's polar regions — the same mechanism behind Earth's auroras and Jupiter's radio storms.
The paper includes a blunt exclusion: no known physical mechanism capable of producing radio emission on an early-type star can explain the observed radiation.
How Small Is a Hundred Milliarcseconds
An angular separation of 0.55 arcseconds is, on everyday scale, comparable to the gap between a one-dollar coin and a fifty-cent coin placed side by side, viewed from 100 km away. Resolving planet from star at this scale relies mainly on the arithmetic of tying the images to a reference frame afterward.
After correction, the radio source's median offset is about 100 milliarcseconds from the planet versus about 434 milliarcseconds from the star — a difference of more than fourfold. The frame correction itself moved things by only ~4 milliarcseconds, showing the result does not hinge on the correction magnitude.
The Number 1250 Gauss
The cyclotron maser frequency is directly tied to the local magnetic field strength. Since the radiation extends up to 3.5 GHz, the team infers a magnetic field of at least 1250 gauss at the emission source.
This is the first direct measurement of an exoplanet's magnetic field strength. All previous claims about exoplanet fields came from model extrapolation.
As an order-of-magnitude reference, Earth's surface field is about half a gauss. The 1250-gauss lower limit is consistent with theoretical predictions for field generation in young, massive giant planets, as the team notes in the paper.
The planet's eight-to-nine-hour rotation period is regarded as one of the main conditions driving this dynamo.
The inference chain has just two steps: the maser frequency is proportional to the local field strength, and the highest observed frequency is 3.5 GHz — so the lower limit on the source-region field is set.
Next: The Next Generation of Telescopes
The team did not stop at one planet. They have listed seven more giant-planet candidates across five nearby systems that could be tested the same way.
The threshold is written into the paper: studying these targets requires radio telescopes five to seven times more sensitive than existing facilities. That number places future progress in the hands of next-generation instruments.
Three Caveats
First, the paper is currently an arXiv preprint and has not been peer-reviewed.
Second, 1250 gauss is a lower limit, not a measurement. It comes from the highest detected frequency; the true field could be stronger.
Third, the word "first" needs precise bounds. Coherent radio emission had previously been detected from exoplanet systems, but the signal could not be distinguished from the star. This time, the increment is pinning the source on the planet.
Watch List
- Peer review: whether the three attribution checks hold up, especially the 4.4 and 4.8 sigma figures.
- Repeat observations: whether burst periodicity recurs with rotation phase across the 2025–2026 observations.
- Follow-up candidates: seven more giant planets in five nearby systems; the five-to-seven-fold sensitivity requirement defers this to next-generation facilities.
- Interior inference: how much the magnetic field strength constrains young giant planets' internal structure and dynamo depth once it enters models.
- Stellar-wind interaction: with a field reading, models of atmospheric escape and star–planet wind interaction finally have a directly comparable measured point.
References
1. Kevin N. Ortiz Ceballos et al., Discovery of radio emission from the exoplanet β Pictoris b, arXiv:2609.16720 2. Phys.org, Astronomers detect radio signals coming from an exoplanet for the first time, 2026-09-22 3. Science/Astronomy media coverage of the auroral radio detection, 2026-09 4. Science News Today compilation of Beta Pictoris b observation details, 2026-09 5. MeerKAT array observing information and European Southern Observatory background material