This is the second time that I’ve written a blog post under an embargo; the first time was a few months into my PhD, and it was about something which had happened just about the time I started. As a result, I’d not really had much to do with it, and I’d spent most of the time that other people were analysing data and writing papers finding my way into the collaboration. That wasn’t quite so true this time, as I found myself invovled with the public outreach effort for LIGO’s next big announcement. It turns out that it’s hard to condense a groundbreaking discovery, which took over 3000 people to make, into one A4 sheet of paper. I also learned all sorts of new things I never expected to during my PhD, like the niceties of colour theory. However, on with the story.

The depths of the cold war

A Vela Satellite under construction

There are lots of times and places where the story of GW170817 could start. One of those times is at 14:19 on 2nd July 1967; in the depths of the Cold War a pair of satellites, Vela 3 and Vela 4, detected something odd. They’d been launched to watch for the characteristic signatures of nuclear explosions, which produce highly-energetic electromagnetic radiation — gamma rays, to ensure that the USSR was complying with the Partial Test Ban Treaty. What they detected that day was a flash of gamma rays, but it didn’t have the characteristic appearance of a bomb test, so it filed away.

Over the next few years Vela would detect more of these gamma ray bursts, and by analysing the difference in detection time between different satellites it was possible to work out the general location of these phenomena in the sky, showing that these were events of extra trerrestrial origin, and weren’t being produced by lightning strikes on the Earth. In 1973 the data was declassified, and published in Astrophysical Journal Letters. The quest to explain what could produce such energetic radiation was just about to begin.

Early theories suggested that these bursts were the result of some phenomenon happening nearby — in our Galaxy. In the early Ninties a new space-based observatory, the Compton Gamma-ray Observatory introduced a new instrument to study these bursts. BATSE, the Burst and Transient Source Explorer, continued to observe Gamma Ray Bursts, and in 1992 a paper published in Nature showed that the bursts were observed to be coming from all parts of the sky; this effectively ruled out the possibility that they were produced by phenomenta within our Galaxy, and so must be happening much further away. As you move away from the source of radiation (electromagnetic radiation anyway) it becomes dimmer by the square of that distance (so if you move twice as far away from a light-bulb, for example, it will look one quarter as bright as it did before). Since these bursts were happening much further away than we’d thought, they must also be a lot brighter than we’d thought too.

Explosions in the sky

As decades’ worth of data became available it started to become clear that there were two different types of these gamma-ray bursts (which I’m going to start calling GRBs from now on). Short GRBs (sGRBs) last less than a second; around 30% of all GRBs are this type, while the rest are normally classified as long gamma-ray bursts, and they can last for as long as 30 seconds.

Short gamma-ray bursts are especially interesting, because producing the enormous amount of energy involved in these events in such a short period of time must involve something truely cataclysmic happening. The sort of cataclysmic events that only happen when the strangest objects in the universe collide.

Stars end their lives when they’re no longer able to produce enough energy to support themselves against their own weight; as stars “burn” gas (converting light elements like hydrogen into heavy elements like helium) they produce enough upward pressure to support themselves. Once they run out of light elements they can use as fuel they collapse on themselves. This is part of a much more complicated (and ultimately mysterious) process known as a “supernova”, which is an odd hybrid of an explosion and an implosion. When all of the material from the star collapses in on itself it forms a super-dense remnant. The pressure from that collapse causes the material it’s made of to turn into a mixture which is dominated by neutrons — one of the particles which make up the nucleus of atoms — at incredible densities.

One of these neutron stars weighs a little more than our sun, and has a diameter comparable to most cities. They are truely some of the oddest objects we know of in the universe, and because they’re so small, we’ve almost no means of telling anything about them since they were first discovered1.

A grand cosmic waltz

A long time ago, in a galaxy far, far, away, there was a violent explosion. That galaxy was 130 million light years away, and that explosion happened 130 million years ago. That explosion happened at the end of a cosmic waltz, after two neutron stars had spent thousands, or perhaps millions of years orbiting around each other, slowly spiralling in towards one another2. This is another place that I could legitimately start this story. This is where (and when) it all happened.

As the two neutron stars got close to each other their powerful gravitational fields started distorting each other, tidally deforming their super-dense, incredibly hard material. The hardest material known in the universe. About a minute or so before the merging of these two stellar corpses was complete the gravitational waves they were producing were so intense that they would be strong enough to be picked up 130 million light years away.

An email

The 17 August was like any other Thursday in the IGR3; the data analysis group had got together for our weekly meeting, and we’d all headed out for lunch afterwards. This is the final place that I could have started this story.

I’d recently (a month earlier) returned from the USA where I’d been working at the LIGO observatory in Louisiana. (LIGO is an enormous gravitational wave observatory, which works by measuring the distortion of space time over a distance of 4km in two directions). I’d been spending my time on developing some software to help monitor the data output from the detector, and to help monitor some of the control systems. I also spent some time at Georgia Tech talking about what neutron star merger signals would look like, amongst other things4. I’d started to get used to being back in Scotland by now, with the rain, the wind, and the proper-sized pints.

We were right at the end of the second Advanced LIGO observing run, and a few days before we’d had news of the first triple-coincidence detection of a binary black hole (BBH) signal (which was to be called GW170814). During an observing run a piece of software runs on a computer to listen for things which might be signals in the data coming from the detectors; the things it reports get called “triggers”, and most of them turn out to be the result of some source of noise near or in the detector. Over lunch I got an email telling me that a trigger had been identified, which looked like a binary neutron star (BNS) collision. I didn’t pay it much attention though, until I got back to the office and saw that that one email was accompanied by several other ones, and it started to become apparent that this was something a little bit different.


  1. The first observational evidence for neutron stars came from radio detections of objects called “pulsars”. The story of their detection is one of the greats of astronomical history, involving a PhD student, “little green men”, and ultimately, a nobel prize. That’s another story, however.

  2. While all of the orbits which are important to day-to-day life, like the orbit of the Earth around the sun, are pretty stable, so that there’s no immediate prospect of the Earth spiralling into the sun, the Theory of General Relativity tells us that no orbit is completely stable, and they all decay over time, by releasing gravitational waves.

  3. That is, the “Institute for Gravitational Research”; the research group at the University of Glasgow which I work in.

  4. Both of which deserve blog posts that I’ve yet to finish writing…