
Neutron stars are some of the most extreme objects in the universe. They can be detected as either pulsars or magnetars, both of which are rotating stars that expel powerful bursts of radiation, often at radio wavelengths. Although astronomers don’t fully understand the physics that govern those stars, basic categorical traits define them. The categories, however, may require updating.
Pulsars are cosmic lighthouses, rotating at a constant speed such that astronomers see a pulse of radio emission at an exact interval. That interval is so consistent over such a long time that it rivals the long-term accuracy of an atomic clock. Pulsars rotate so fast that most of them have a period of less than 12 s; millisecond pulsars rotate even faster. The speed of rotation is part of the process that creates the energy to emit beams of light.
Under current theories, magnetars have slightly slower rotation rates. They generate energy in their twisting magnetic fields. Magnetar emissions, however, are short-lived. The stars may emit energy for only a few months at a time or just once and are never detected again.
Those categories are convenient, and they have gone unchallenged—up until now. But the universe is full of surprises, and current models don’t explain GPM J1839-10, an ultralong-period radio transient recently discovered by Natasha Hurley-Walker (seen in figure 1), of the International Centre for Radio Astronomy Research in Australia, while she was using the Murchison Widefield Array (seen in figure 2). To find the star, she disregarded conventional wisdom on how to discover pulsars and instead processed the telescope data in a way that didn’t assume a short period.
Rather than running algorithms designed to detect pulsar-like short-period spikes in radio emission, she made images of every four seconds of data and used filters to detect pulse-like emission on time scales of seconds to minutes. Hurley-Walker and her team found what they were looking for: GPM J1839-10 has a period of 22 min. Already it exists outside the current understanding of what is possible for a pulsar. Something rotating that slowly shouldn’t be able to emit bright radio waves.

And yet it doesn’t seem to be a magnetar either. Collaborator Scott Hyman (an emeritus professor at Sweet Briar College in Virginia) searched through the archives of radio telescopes across the Southern Hemisphere and was able to find signals from the star stretching back to the dawn of digital radio astronomy. That’s three decades of metronomic radio emission—very un-magnetar-like behavior.
GPM J1839-10 isn’t the only ultralong-period radio transient. A handful of other objects are sporadically being discovered with incongruous properties (seen in figure 3). As more discoveries are made, more theories are being considered to explain them, but so far no one theory has included all the observed properties.

Hurley-Walker is currently conducting a survey on the Murchison Widefield Array with the goal of finding more ultralong-period radio transients. When more than a handful of examples as been accrued, patterns may begin to emerge, and perhaps a new class of astronomical objects will be defined. (N. Hurley-Walker et al., Nature 619, 487, 2023.)