Turbulence is a common chaotic phenomenon that everyone has certainly experienced: you witness the development of turbulence when you stir a coffee too vigorously or watch the smoke of a cigarette or feel the aeroplane going up and down. What makes turbulence often spectacular is the presence of several vortices which are indeed seen in many atmospheric phenomena from the magnificent tornadoes and hurricanes to the tiny dust devils. Such vortices have a certain life-time that is set by the amount of energy that sustains them and that works against viscosity, which tries to re-establish the calm and quietness of the flow.
However, there exists also another type of turbulence, called “quantum turbulence“, that appears in some very strange fluids known as superfluids. These are fluids like those we are familiar with (water, air, etc…), with the remarkable difference that they have no viscosity at all. They have all other properties of fluids, but if you set them in motion they will flow without resistance. We don’t usually see them around us because superfluidity is a property that appears when the temperature is close to absolute zero (-273 °C). Despite the total absence of viscosity such superfluids can exhibit turbulence because of quantum effects and they too develop vortices here and there. The difference with the normal fluids is that superfluid vortices have a fixed size, look identical and never grow bigger and bigger as it might happen instead with normal fluids (e.g., a tornado).
In a paper that has appeared today on arXiv, an Australian and an American astrophysicist link such quantum vortices to the existence of irregularities in the spin evolution of a certain type of neutron stars known as pulsars (as if the quantum turbulence phenomena were not mind-blowing by themselves already…). Young pulsars emit pulsed radio waves and they rotate around their own axis every second or so. Thanks to their radio pulses one can time the pulsar rotational evolution with a precision that in some cases rivals that of atomic clocks. This is true for the so-called radio-millisecond pulsars, which are old and spin several hundred times per second. However, young radio pulsars are not rotating with such a wonderful stability and they show a still unexplained rotational wandering called “timing noise”.
Timing noise has been a mysterious component of all the thousands young radio pulsars discovered so far, and many different theories have appeared in the literature that try to explain what its origin is. The new calculations of quantum turbulence seem now to be able to explain many characteristics of the timing noise in radio pulsars. The authors show that when considering a large number of known radio pulsars (366 objects), several trends observed in the pulse times of arrival can be explained by the superfluid turbulence. Such quantum turbulence is a new possibility and probably even a likely one, as there is strong evidence for the existence of superfluidity in the interior of most neutron stars. As the authors conclude in their paper, more observations are needed to set the final word on the problem, but these quantum vortices might very well populate the sky and live in the cold interior of neutron stars.
October 15, 2013 at 5:13 am
It is not at all surprising that superfluids experience turbulence: generically, all fluid flows are turbulent unless viscosity is able to suppress it (this is what the Reynolds number measures, for example). Although superfluid flows are not viscosity-free (because of the inevitable presence of a normal fluid component) the viscosity on a pulsar scale is expected to be low. So, sure, turbulence, no problem. Unfortunately measuring red noise is extremely difficult, and it’s very easy to produce a 1/f^2 spectrum: any kind of white noise in the torque would do, including for example random switching between modes in the magnetosphere – which makes a certain sense in light of the paper out of Jodrell Bank claiming correlations between profile variations and timing noise. There’s more promise in the idea of looking at the scaling of timing noise across the population of pulsars, but unfortunately it’s so hard to come up with a good measure of timing noise that there isn’t much in the way of data sets to test theories against.
October 15, 2013 at 9:40 am
I agree this is not the final word on the problem and I also strongly feel that changes in pulse profiles have a big part in the timing noise we see. (I myself suggested this idea several years ago to some theorists but never published anything on the topic). But I also like very much the turbulence idea, after all as you say, turbulence is basically (almost) unavoidable in the superfluid interior of neutron stars. So we should be looking for hints in the observations and this paper provides some interesting calculations to compare the observations with. That the observations are tremendously complicated to analyse and the fact that timing noise can be generated by a number of different phenomena is true but I see a lot of room for more sophisticated analysis. What do you think about this last point ?
October 15, 2013 at 10:12 am
I agree that there’s room for clever new analysis techniques to describe red noise in pulsars – I did a poster on it, way back when. My techniques weren’t all that clever, but there are some new ideas that are kind of promising. I think timing noise is such a messy phenomenon, though, that developing techniques to measure it will involve a lot of work for a meager increase in understanding. Fortunately many of the techniques for measuring, or at least diminishing the effect of, red noise are going to need to be developed anyway for the pulsar timing array projects, which do have a promising scientific payoff at the end.
October 15, 2013 at 10:32 am
Is there a way I can look at your poster ? I’m always looking for new ideas/techniques on how to characterize and account for the presence of timing noise in pulsar timing.