Science1 publisher2 min readPublished
The brightest transient in the VLA sky survey has been emitting radio waves since 2005
Jessie Miller's Caltech team followed VT J1906+0849 across radio, infrared, optical and X-ray data and favors sustained accretion onto a compact object, even though a Swift observation found no X-ray counterpart.
The Scientist · Science desk

What happened
- VT J1906+0849 is the brightest transient known in the Very Large Array Sky Survey, found in a search for radio transients at low galactic latitudes, and VLASS first detected it on Oct. 27, 2017.
- Archival MAGPIS data show the source was already bright in 2005, at about 35 millijansky at 4.86 gigahertz.
- Its brightness appears to have peaked above 200 millijansky around 2014, faded, and rose again in late 2025; the team puts the total brightening at a factor of at least six over roughly 21 years.
- A Swift observation found no X-ray counterpart, and the team estimates the source lies between 15 and 32 kiloparsecs from Earth.
- Very Long Baseline Array imaging in 2010, 2022 and 2025 shows an emitting region that has not expanded, even though spectral lines indicate gas moving at thousands of kilometers per second.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- capability The rebrightening in late 2025 makes this a live target, so the accretion model can be tested against new simultaneous observations instead of reconstructed from archives.
- constraint A luminosity uncertain by a factor of about 4.5 limits how firmly any class can be excluded on brightness alone; the exclusions of novae and cataclysmic variables are only as strong as the nearer end of the distance bracket.
- decision Sorting a genuinely new class from a familiar accretor in an unusual state is now an X-ray problem, and it needs a pointing deep enough to publish a limit.
- precedent The brightest object this survey found is unclassified, so further searching of low galactic latitudes is where the next unclassifiable Milky Way transients would come from.
Most radio transients rise fast and fade slowly [13]. This one has been detectable for more than two decades, and its properties match nothing in the catalog of known galactic radio sources [1]. The team wrote that those properties "favor sustained accretion onto a compact object rather than an impulsive explosion or passively evolving nebula" [19].
That leaves the size to explain. The spectral lines imply fast motion and the images show no growth, so the paper needs a way to hide the expansion, and proposes a jet from a black hole or a neutron star running into a dense disk wind, with the wind blocking part of the radio-emitting plasma so the visible source stays compact [20]. Broad near-infrared emission lines are the possible evidence for that wind [21].
The brightness temperature sits between 100 million and 10 billion kelvin [16]. Thermal processes, star formation among them, do not reach those values, which leaves a nonthermal origin near a compact object, most naturally synchrotron radiation from relativistic electrons spiraling through strong magnetic fields [16][17]. A young neutron star could supply the energy through spin-down and power a synchrotron-emitting pulsar wind nebula. Pulsar wind nebulae evolve over centuries or longer, and this source has changed substantially in a few years [18].
The eliminations of the faint classes run through luminosity, and luminosity runs through distance. The bracket the team quotes is 15 to 32 kiloparsecs, or 49,000 to 104,000 light-years [11]. The far end is 2.1 times the near end, and flux converts to luminosity with the square of distance, so the same measurement implies a luminosity range spanning a factor of about 4.5 [23]. Flare stars, magnetically active binaries, novae and cataclysmic variables are far too faint across that range, according to the paper [15].
The X-ray silence rests on one Swift pointing, and the account of the paper reports the non-detection with no exposure time and no flux limit [12]. The work is a preprint on arXiv [2].
Fast radio bursts and other extragalactic transients have been found in large numbers in recent years, while the Milky Way's own transient population remains much less explored [22]. One object cannot say how many similar sources the Galaxy holds, and it cannot supply a rate; a rate takes the searched area and the sensitivity behind it. And the two-decade history exists only because MAGPIS imaged the field in 2005, twelve years before VLASS flagged the source [8][25].
What to watch
- Whether an X-ray pointing during the current bright phase yields a detection or a quotable upper limit on the accreting engine.
- Whether new VLBA imaging taken during the late-2025 rebrightening finally shows the emitting region expanding.
- Whether peer review holds the 15-to-32 kiloparsec distance bracket, on which the luminosity-based class eliminations depend.