Tabby's Star (KIC 8462852)
Partially explainedSummary
A perfectly ordinary star, about 1,500 light-years away, dims by up to a fifth of its brightness, at no interval anyone can predict, for anything from a day to several months. Nothing in the catalogue of stellar behaviour does that. For two years the leading explanations included an alien megastructure, and serious astronomers said so in print.
The answer arrived from a direction nobody was watching: not from a bigger telescope, but from asking what colour the missing light was.
What is documented
The star. KIC 8462852 is an F3V main-sequence star. Spectroscopy shows it to be entirely unremarkable, with no radial velocity variation and none of the signatures of the known classes of variable star.
The dips. Kepler monitored it from May 2009 to May 2013. The light curve shows a series of dips in brightness that are deep, irregular in shape, and aperiodic. The largest reach roughly twenty to twenty-two per cent. They last anywhere from five to eighty days.
The scale is the point. A large planet crossing the face of a star of this size blocks on the order of one per cent of its light. Twenty-two per cent is not a planet. It is not a planet even if the planet is enormous, and no solid body of plausible size can produce dips of that depth in that pattern.
Found by amateurs. The dips were not spotted by a machine. They were flagged by volunteers working through Kepler light curves on the Planet Hunters citizen science project, and published by Tabetha Boyajian and colleagues in 2016.
The long fade. Beyond the dips there is a slow, secular dimming. Analysis of Kepler's full-frame images showed the star faded by about three per cent across the mission. Archival photographic plates suggested it had faded over a century: Schaefer, working from the Harvard plate collection, reported around sixteen per cent, and Castelaz and Barker reported a slightly smaller figure from the Maria Mitchell plates. Hippke and colleagues disputed the statistical significance of the plate results and found no significant dimming in the Sonneberg archive. That disagreement has never been fully resolved and the registry notes it rather than choosing a side.
The problem that made it strange. Dust was the obvious candidate from the beginning. Dust close to a star is warmed by it and glows in the infrared. Follow-up observations found no infrared excess, which set strict limits on how much warm dust could be in the system. The obvious answer appeared to be ruled out by the evidence, and that is what left the field open.
Leading explanations
Dust, established by the colour of the missing light. This is the resolution, and the reasoning is clean enough to state in a sentence.
An opaque object blocks every wavelength equally. A star behind a solid body dims by the same fraction in ultraviolet, in visible light, and in the infrared, because a solid body does not care what colour the light is. Fine dust behaves differently: grains comparable in size to the wavelength of light scatter short wavelengths more strongly than long ones. Dust reddens what passes through it. Solid objects do not.
Between 2016 and 2018, several teams measured the star in more than one waveband at once. Meng and colleagues combined Spitzer in the infrared, Swift in the ultraviolet, and ground-based visible light, and found the dimming stronger in the ultraviolet than in the infrared. Boyajian and colleagues, observing four dimming events in 2017 with the Las Cumbres Observatory network, found the same: different colours were being blocked by different amounts. Schaefer and colleagues, with nearly twenty thousand images across four filters, found the amplitudes in the redder bands systematically smaller than in the blue, and stated the conclusion in the strongest available terms: occultation by any star, planet, solid body, or optically thick cloud is ruled out.
Whatever is passing in front of KIC 8462852 is translucent. It is made of particles small enough to be selective about colour, down to around a hundred nanometres. That is dust.
Where the dust came from, which is not settled. The mechanism is established; the history is not. Proposals include swarms of exocomets, the debris of a shattered planet or moon, dust-shrouded planetesimals on eccentric orbits, and the aftermath of the star consuming a gas giant. The dust is real. Its origin story is still being argued.
The megastructure. In 2015 and 2016 the possibility of an artificial structure, a partial Dyson swarm, was discussed seriously, including by astronomers who did not believe it. It was never a leading hypothesis and it was never presented as one by the people doing the work. It is now dead, and the manner of its death is worth stating: it was not dismissed as absurd. It was tested, and it failed a measurement.
What the popular version gets wrong
"Astronomers thought it was aliens." They did not. They listed it among the possibilities because the observations were genuinely unexplained and honest scientists say so. Wright and Sigurdsson's survey of hypotheses put circumstellar material, intervening material and stellar processes ahead of it throughout.
"The megastructure theory was laughed off." It was killed by data. An artificial structure is opaque, and an opaque structure cannot dim a star more in ultraviolet than in infrared. That is a falsifiable prediction, it was tested, and it failed. This is the strongest possible reason to reject an idea, and it is much more interesting than ridicule.
"It has been solved." Half of it. What is blocking the light is settled. Where that material came from, and why it is distributed in clumps that produce dips of eighty days, is not.
"It dimmed by 22 per cent, so something enormous passed in front of it." Something diffuse passed in front of it. A cloud that blocks a fifth of a star's light while remaining invisible in the infrared has to be tenuous, spread out, and made of very small grains. The size of the dip is a measure of coverage, not of solidity.
Current status
Partially explained. The registry keeps this record as the clearest example in the archive of an anomaly resolved by a better question rather than a better instrument.
Everyone was asking how much light was missing. The answer came from asking what colour it was.
Sources
- Boyajian, T. S. et al. (2016). "Planet Hunters IX. KIC 8462852: where's the flux?" Monthly Notices of the Royal Astronomical Society.
- Meng, H. Y. A. et al. (2017). Spitzer, Swift and AstroLAB IRIS multi-wavelength observations; wavelength-dependent dimming.
- Boyajian, T. S. et al. (2018). Las Cumbres Observatory observations of the 2017 dipping events.
- Schaefer, B. E. et al. (2018). "The KIC 8462852 Light Curve From 2015.75 to 2018.18 Shows a Variable Secular Decline." BVRI photometry; occultation by solid bodies ruled out.
- Montet, B. T. and Simon, J. D. (2016). Secular dimming in the Kepler full-frame images.
- Schaefer, B. E. (2016), Castelaz and Barker (2018), and Hippke et al. (2017) on the disputed century-scale fading in archival plates.
- Wright, J. T. and Sigurdsson, S. (2016). Survey and classification of proposed explanations.
- NASA JPL, Spitzer and Swift mission material.
Last reviewed: July 2026. Records are provisional. Where the evidence changes, the entry changes. Found an error? Tell us.