Scientists uncover new insights about pulsar glitches

Multi-wavelength image of the Crab nebula, which hosts a glitching pulsar at its heart. Credit: NASA, ESA, J. Hester and A. Loll (Arizona State University)

Pulsars are dead stars that spin remarkably steadily – they are some of the most regularly ticking clocks in the Universe! However, every few years some pulsars ‘glitch’, and speed up a tiny amount almost instantaneously. Understanding what causes these glitches may unveil what’s really happening inside these super-dense dead stars. 

Detailed theoretical and computer models are hard to connect to real observations, so instead PhD student Julian Carlin and Chief Investigator Andrew Melatos, from the ARC Centre of Excellence for Gravitational Wave Discovery (OzGrav), built a ‘meta-model’ in a paper recently published in the Monthly Notices of the Royal Astronomical Society.
 
The meta-model relies on the idea that ‘stress’ builds up inside the pulsar until it reaches a threshold, and then some of this stress is released as a glitch. The interesting thing about this meta-model is that the stress increases by taking a ‘random walk’ upwards: like an intoxicated person returning home from the pub who might take two steps forward, one step back, then three steps forward. The randomness in how the stress builds is supported by some theoretical models, as well as a recent study of a glitch-in-action led by OzGrav researchers Greg Ashton, Paul Lasky, and others.
 
Meta-models make predictions about what we should see in the long term from glitching pulsars.
 
‘This meta-model predicts that there should always be a correlation between big glitches and the time until the next glitch: if a lot of stress is released, it takes longer on average for the pulsar to build up enough stress for another glitch,’ explains Carlin.
 
Using this prediction, Carlin and Melatos tried to falsify the meta-model, asking the question: ‘Are there long-term observations that can’t be explained?’. The answer depends on the pulsar. Some are well-explained by the meta-model, while others don’t quite match the predictions.
 
‘We need to see more glitches before this question can be answered for certain, but this work shows a way to answer it for many theoretical models, all at the same time,’ says Carlin.