Chandra Catches a Supernova Impostor in the Act

Chandra detected X-rays from AT 2016blu in March 2026, at the peak of an outburst the star's 113-day cycle had predicted in advance. In the telescope's archive from 2001 and 2002, the same position shows nothing.

That detection, and that non-detection, carry most of the weight in a new paper led by Mojgan Aghakhanloo of the University of Virginia. AT 2016blu sits about 29 million light years away and holds roughly 33 times the mass of the Sun. It is a luminous blue variable, and for a decade it has been the leading "supernova impostor": a star that erupts like a supernova on the verge of exploding, does it 27 times, and does not explode. The paper's conclusion is that a compact companion is stripping mass off it, and that AT 2016blu is the first known luminous blue variable whose outbursts are powered this way.

Stars of 33 solar masses do end as supernovae. Many supernovae show eruptions in the years before they go. AT 2016blu has flared 27 times since it was flagged in 2012, on a quasiperiod of roughly 113 days. A massive star doing that on a schedule looked like a countdown.

Aghakhanloo's earlier papers, in 2023 and 2025, argued that a binary system was responsible and that no explosion was imminent. Those papers left many questions unanswered, and the new X-ray observations are aimed at them. The reason the team got those observations is that the periodicity turned a guess into an appointment. "We knew when the next outburst should be," Aghakhanloo said. Multiple telescopes and more than 30 amateur astronomers watched the March 2026 peak, and the team then triggered a Target of Opportunity request on Chandra. The telescope interrupted its own observing schedule to point at the star.

A Target of Opportunity request is a standing arrangement rather than a snapshot of luck: an observatory agrees in advance to break its planned schedule if a nominated event happens. Usually the trigger is something nobody could book, a gamma-ray burst or a supernova in a nearby galaxy. Here it was a calendar.

Xs.

The X-rays Chandra collects come from gas heated as it falls onto a compact object. Falling onto something compact means falling deep into a gravity well, and the infalling gas converts that energy into heat. The gas reaches temperatures where it radiates X-rays rather than visible light, which is why the emission shows up in Chandra and why it points at a companion instead of at the star's own surface.

The chain runs from the companion's gravity into the star's wind. A luminous blue variable is already throwing off material, and the compact object's pull drags part of that stream toward it, heating it as it comes. The star's own photosphere is far too cool for the same trick: it radiates in the visible and the ultraviolet, not at the energies Chandra collects. An X-ray signal at that position therefore tags the companion rather than the star.

Chandra measured the luminosity during the outburst, and from that number the researchers calculated the mass accretion rate needed to produce it. The calculation is bookkeeping. Accretion turns gravitational energy into light, and the amount of light released per unit of mass depends on how deep the well is. Roughly a tenth of the rest energy of the infalling gas comes out as radiation when the accretor is a neutron star, which has a surface; a black hole does better or worse depending on how fast it spins, but stays in the same range. That fraction is the conversion factor, and it is what makes the measurement quantitative. Take the luminosity Chandra recorded, divide by the energy released per unit mass, and what is left is the rate at which material has to arrive. What is measured is the luminosity. The accretion rate is inferred from it. That rate is consistent with a compact companion. In the authors' words, AT 2016blu is "the first known case of an LBV SN impostor whose outbursts are driven by intermittent accretion onto a compact object." A massive star plus a compact companion in orbit is a high-mass X-ray binary.

The two objects travel an eccentric orbit, so they are close only near one end of it, and the flare recurs each time they pass that point. The system's brightness tracks the shape of the orbit rather than a steady burn. A comet that grows a tail only near perihelion behaves the same way. The paper calls the periastron trigger presumed, which is the honest framing: the timing is measured, the mechanism producing the timing is inferred.

Those were not planned exposures of this star; the field happened to fall in a pointing aimed elsewhere. The inadvertent Chandra observations from 2001 and 2002 showed no X-rays at that position. At some point in those 24 years, accretion onto the companion switched on. A detection on its own would not have said when that happened. The blank field does.

The stacked March 2026 images also contain NGC 4559 X7, a separate ultraluminous X-ray source nearby.

Luminous blue variables are a short-lived phase of the most massive stars, and the phase is poorly understood because the stars are rare and brief. AT 2016blu, at 33 solar masses, sits in the mass range where that phase happens. If a companion is draining one of them on a 113-day cycle, that changes the star's mass and its rotation, and therefore changes what it becomes.

The paper describes the companion as a compact remnant, a black hole or a neutron star, and does not settle which. It also leaves open whether other impostors work this way. AT 2016blu is by definition the first case found, and a first case is often the easiest one to spot rather than the representative one.

One caveat on the result itself. The paper is on arXiv and forthcoming in The Astrophysical Journal, and the X-ray picture of AT 2016blu currently rests on a small number of observations at a single predicted outburst.