Another wave, and a smile

Last week the international gravitational wave collaboration* announced its third very secure detection. Our new acquaintance GW170104 — named for its arrival date — passed through our two LIGO detectors and whispered to them about the coalescence of a pair of black holes in a binary system that had taken place almost three billion years ago. And then it raced on, hardly affected at all by its encounter with us and our planet. In that respect it was much like its two predecessors, GW150914 and GW151226. But every detection is special, and this one is very special indeed.

One reason is simple: it brought a big smile to all of us in the collaboration, and allowed us a big sigh of relief — our first two detections were not just a lucky fluke, nor a super-mysterious instrumental malfunction. The dates of the detections tell the story: the first two in 2015, this one at the start of 2017. The big gap is there because the LIGO detectors had been shut down for most of 2016 in order to improve their sensitivity. Once we started up again, with significantly modified detectors, Nature provided us with helpful reassurance: the waves keep coming, and they look just the same as they did in the previous version of the detectors. (Goodness knows what we missed during 2016!)

The second important thing about GW170104 is that it has started our transition from mainly doing fundamental gravitational physics to primarily doing gravitational-wave astronomy. The first detection GW150914 was the science event of the decade because it showed how right Einstein was with general relativity: his amazing fundamental theory of gravity had predicted gravitational waves and black holes, and in just one event, lasting only about two tenths of a second, both predictions were spectacularly confirmed. But now, as we continue to observe more such events, we become more and more focussed on what they tell us about the Universe. We will of course use all new data to further test our confidence in Einstein’s general relativity. But the new detections are going to explore more and more of what Kip Thorne, one of the founders of LIGO, famously called the Dark Side of the Universe.

A hint of spin

But the really special aspect of the new detection is the hint it gives us about the way the black holes may have been spinning. Black holes spin, as do all astronomical bodies. And the normal expectation is that the spins in a binary system of black holes will be in a consistent direction, just like the spins of planets in the solar system. The Sun and the planets all formed together out of a rotating cloud of gas, so it is natural that they inherited their spins from that cloud, so that these spins are all roughly in the same sense, which is also the same as the sense of the orbital rotation of the planets around the Sun, and indeed of the spin of the Sun itself. In our previous detection, GW151226, we found evidence that the more massive of the two holes was spinning, and the evidence was that its spin was probably aligned with the sense of the orbital rotation. But in GW170104, there is just a hint that the black holes’ spins were either anti-aligned with each other, or they were anti-aligned with the sense of their orbital motion about one another before they merged.

if this was the case, then it would suggest that the two black holes had not been formed together from a single primeval cloud of gas. Instead, they were probably unrelated single black holes, which just by chance linked up into a binary system. This ought to sound bizarre to you, wildly improbable. After all, black holes, even the rather massive ones in this system, are only something like 100 km across, maybe twice the size of London. How could two such small objects ever encounter one another if they were wandering through the vast emptiness of space between the stars in their galaxy? And you would be right: that just wouldn’t happen. But it would be much more probable if it happened inside the big dense clusters of stars that we call globular clusters.

These globular clusters are roughly spherical micro-galaxies, and in them heavy things like black holes sink, over millions of years, down toward their centers, where there can be thousands of stars in a volume that would contain just one star in neighbourhood of our Sun. In these conditions, near encounters between black holes do sometimes happen. What’s more, if a third object, even a normal star, is nearby, then sometimes the third object can be a kind of catalyst, helping the black holes to form a binary system by stealing and speeding off with some of their energy. So the hint that the spins in GW170104 don’t match up is a hint that this system might have been formed in this haphazard way.

The magnetic side of gravity

But here’s the amazing part: how did we get this hint in the first place, about how things spin in a system that is totally in the Dark Side? What did we read in the signal we call GW170104 that tells us about the spins? The answer lies again in the fundamentals of Einstein’s theory. Einstein predicted that spin creates a different form of gravity, a magnetic form. Think of a spinning charge, or of the electric current winding round and round through the coil in an electromagnet: this creates magnetism, which we know is the partner of the electric force in the theory of electromagnetism. Magnets exert forces on other magnets and on moving charges. In a closely analogous way, a spinning mass creates a form of gravity that we call gravitomagnetism, through which spinning masses exert gravitational forces on other spinning masses and on moving masses. So our spinning black holes are gravitomagnets, which interact with one another and with their orbital motion, and this slightly changes the orbital periods of the last few orbits before the holes merge together. We always look in our data for these changes, and in this case we found a hint not only that the spins were significant, but that they might be anti-aligned.

But it is only a hint. The odds are only 4-1 in favor of the spins’ being anti-aligned rather than their being consistently aligned. It could be simply a distortion in the signal caused by a slightly improbable level of noise in our detectors. But since we will never get any more information about this particular system, we will never be able to say for sure whether this was how GW170104 in fact formed. So GW170104 is not only special, it is also frustrating!

What sort of physics are we doing?

This frustration has pushed the global gravitational wave collaboration to confront the question of whether we are still primarily doing fundamental physics or whether we have started the transition to astronomy. One of the differences between these two disciplines is their expectation of how much uncertainty they can tolerate in a measurement. The observation of gravitomagnetism is of course very fundamental, even though this prediction of Einstein has already been verified by an experiment in orbit around the earth [1] and by observations of the orbits of satellites themselves [2]. A fundamental physicist wants to pin down the laws of physics, and that needs a great deal of certainty. As a measurement of gravitomagnetism, our hint of spin doesn’t do as well as the previous experiments have done, and is certainly not up to our standard for deciding we have detected something in the first place. While the actual detection of GW170104 has very high significance (odds of better than 70,000 to 1 that no chance noise-generated event this strong would have happened in coincidence between the two detectors in a whole year’s worth of data), the information we can glean about spin from the slight spin-induced nuance in the orbital motion has only 4-1 odds of being real.  So a number of physicists in the collaboration did not want to see the spin result highlighted strongly in the announcement of our observation.

Astronomers, by contrast, have learned to live with uncertainty: you can never do a controlled experiment on a star, so you have to be content with whatever data it sends you. Some of the astronomers in our collaboration felt that the possible implications of this hint of spin for our understanding of the formation history of this system was interesting enough to deserve prominence in our announcement, even at this low significance level. The philosophy is to put the data out there so that other astronomers can judge whether they find it interesting or not. This clash of scientific cultures led to a rather lively discussion inside the collaboration on how to report the tantalising but rather uncertain measurement on the spins of the black holes, and it even delayed the announcement of the detection by some weeks. What you can read in the paper we wrote and the publicity we put out is, of course, our compromise. As you might guess from this post, I find myself on the side of the astronomers.

But it was already 5 months ago…

The detection we announced last week happened 5 months ago. Has anything else interesting been detected since then? Well, by the confidentiality rules of our collaboration, I can’t answer that question. But what I can say is that the increases in sensitivity that we looked forward to a year ago are taking more time than we hoped. These incredibly complicated detectors, with their astonishing sensitivity to such weak gravitational waves, are also immensely sensitive to all kinds of disturbances, from outside the detectors and especially from inside. The two LIGO detectors are indeed more sensitive now than they were before they shut down at the beginning of 2016, but they will need much more work before they reach the ultimate goal that we call Advanced LIGO. The Italian-French detector Virgo, near Pisa, has also had a number of challenges in improving its sensitivity. We are hoping it will start observations soon, finally giving us the three-detector network that is required in order to get accurate information about the location of events on the sky and their distances.

We can expect a further planned shutdown of LIGO, and probably of Virgo, late this year or early next year, and another long wait for data as the sensitivity is further improved. But this kind of wait should be worthwhile. A factor of 2 increase in sensitivity corresponds to a factor of 2 increase in the range of our detectors, in the maximum distance to which we can detect events. This implies an increase in the volume we can survey by a factor of 8, so it would bring us more events in 2 months than we presently get in a year.

Gold from gravitational waves

And higher sensitivity might also get us a new kind of event: mergers of neutron stars from binary orbits. They are less massive than black holes, so their signals are weaker, but they are no less interesting. Whereas black holes merge, well, blackly, neutron stars merge in a burst of color. They should produce a spectacular display of light, radio waves, X-rays, and gamma-rays, easily detectable by telescopes on Earth.

Besides all the physics and astronomy that we will learn from neutron star mergers, here is the really spooky part: when we finally observe such an event, we may be observing a re-run of our own cosmic history. Astronomers believe that most of the common heavy elements on Earth — gold, silver, platinum, mercury, uranium and many more — may have been created in just one such event, the merger of two neutron stars. They were ejected by the explosion triggered by the merger, polluting the nearby gas cloud that would eventually condense into our Sun and solar system. In fact the explosion might even have triggered that cloud to condense, setting off the long chain of events that led to our own existence. The rings on our fingers, and perhaps even our fingers themselves, may have come from a nearby gravitational wave event that happened long long before we were able to build detectors to observe it!

So there is much more to come, with patience.


* The collaboration consists of the LIGO Scientific Collaboration (LSC) and the Virgo Collaboration. Altogether there are over 1000 scientific members as authors to the papers.


1. Everitt; et al. (2011). “Gravity Probe B: Final Results of a Space Experiment to Test General Relativity”. Physical Review Letters. 106 (22): 221101.

2. Ciufolini, I.; Pavlis, E. C. (2004). “A confirmation of the general relativistic prediction of the Lense–Thirring effect”. Nature. 431, 958–960. Ciufolini, I., et al (2016): “A Test of General Relativity Using the LARES and LAGEOS Satellites and a GRACE Earth’s Gravity Model”.Eur. Phys. J. C. 76, 120.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s