Imagine standing in the Albert Hall, and just before the performance begins, as the orchestra is tuning-up, you hear a perfect, virtuosic Mozart melody coming through the noise. Ten years ago, to the day, the world of astronomy (along with the rest of the world) was shaken as the first gravitational waves ever to be detected were measured with the newly-upgraded LIGO detectors. While they were still tuning-up. The gravitaitonal wave took just a fraction of a second to pass through the Earth, but in its wake it created a new field of astronomy.
Monday 14 September 2015 is a day which has been cemented in the history of astronomy for the first detection of gravitational waves, the first direct detection of a black hole, and completing a century-long process which started with the modern formulation of our theory of gravity in 1915 by Albert Einstein in his General Theory of Relativity, and later prediction that gravity should have its own form of radiation similar to the electromagnetic radiation which is a much more normal part of everyday life. This didn’t, however, make any headlines. Indeed, the headlines that day were dominated by the arrest of a school student in Texas for bringing a clock to school. It was the end of a long summer: I’d finished my masters degree in June, and I was waiting to start my PhD on 1 October. I’d spent the last couple of months working at the University’s observatory, mostly writing bits and pieces of software. I was rather keen to get started on research, and that weekend I’d applied to join the LIGO Scientific Collaboration. On Monday morning my application was approved.
Detecting gravitational waves requires a lot of people, and a lot of coordination. It turns out that email is the way everything is coordinated. A lot of emails. The first thing which I did after joining the collaboration was to sign up to a couple of mailing lists my supervisor had suggested. I signed up to the “bursts” list, which was all about the kind of signal I was going to spend my PhD working on. There was quite a lot of chatter on that list, and I didn’t really understand any of it.
In September 2015 the two LIGO detectors in the USA had just been turned back on after a five year period where a lot of the technology in the detectors was replaced to massively improve their sensitivity. For the first time they were expected to be sensitive enough to actually detect gravitational waves, probably from the collision between two neutron stars, which were expected to be the most common kind of source at the time. There was real excitement, because something had been detected. It was a gamma-ray burst: these were believed to be produced by neutron star mergers. Had LIGO detected anything?
Buried in the noise of all this chatter there was another conversation going on, as it would turn out. One which I was completely oblivious to at the time.
I can’t remember when I first actually realised that something had happened right in front of me, but it was at least a few days later. Back in 2015 gravitational-wave people tended to be the butt of jokes between astronomers. They’d spent enormous amounts of money building massive detectors which didn’t detect anything. For the first few months of my PhD it was sometimes difficult to not let slip that maybe things had changed when someone made one of these jokes. You see, we didn’t immediately go to The Times to announce our success.
There was (and indeed, still is) a lot of caution in the modern gravitational-wave astronomy community about claiming first detections. The first “first detection” was made in 1969 by Joseph Weber, but Weber’s results were never widely recognised by the scientific community. Other groups, including the group at Glasgow, failed to find anything similar with their own detectors. Meanwhile, it became clear that the amount of energy which would be involved in creating gravitational waves like Weber was reporting would require physical phenomena which were implausible. All this isn’t to say that there isn’t a possibility that the first detection did happen in this era: the Glasgow group published an event detected on 5 September 1972. However, this first era of “detection” would pass with no convincing evidence that gravitational waves really existed.
Gravitational waves were the last measurable prediction of General Relativity to be directly observed (observational tests of the theory go all the way back to Arthur Eddington in 1919 who observed the predicted deflection of star light around the Sun during a total solar eclipse), but they were first observed indirectly. In 1974 a binary star system was observed for the first time which was composed of two neutron stars, one of which is a pulsar: a neutron star which produces highly-beamed radiation from two poles. Pulsars produce an incredible predictable signal: by measuring changes in this pulsar’s signal it’s possible to measure the movement of a pulsar in the binary. What Hulse and Taylor discovered was that the binary was shrinking by a few metres every year: the neutron stars were getting closer to each other, meaning that energy as being lost from the orbit. That energy was being emitted as gravitational waves. Hulse and Taylor went on to win the 1993 Nobel Prize for Physics. The stakes had risen. It was very clear that the first person to make a reliable direct detection of gravitational waves would be a dead-cert for the Nobel.
In 2014 the science media was ablaze with the news of another first detection. It even made it into the normal newspapers. This was very very bad news for LIGO. By this stage I’d spent a bit of time working in the Institute for Gravitational Research at Glasgow, and I knew people were going to be pretty upset that they’d missed their chance at glory. This detection, which was claimed by an experiment at the South Pole, called BICEP-2, was made by measuring the polarisation of the cosmic microwave background: what is effectively the heat left over from the Big Bang. In the end this detection also proved to be premature, and was explained-away as the result of a much closer-to-home phenomenon which hadn’t been correctly accounted for in the analysis.
After so many false starts and analyses which proved to be less-than-entirely-robust, it was vital that LIGO got things right. It would take months to complete the analyses, check every single part of the analysis, write the paper, and then debate the wording of the paper. I remember being on one telecon where the results of an important investigation were being discussed. The entire digital process which converted the output signal into the data which is used in the actual data analysis had been checked through. Constructing the analysable data is non-trivial, and lots of corrections need to be made to what comes out of the detector to make it usable. During this check the teams were able to confirm that nobody had been walking around near either detector - their footsteps would have shown up on the seismometers. That meant that the only way someone could have faked the signal was to find a way to get close to the detector, somehow plug something like an ipod into it, play a fake signal into the detector, without walking near it. The Mission Impossible scenario seemed more implausible than the Universe having given us our big break. (Plus they’d have needed a perfectly-timed accomplise at the other detector to also fake a signal at just the right time thousands of miles away).
That meant that we needed to keep everything a secret until the big reveal in February 2016. We didn’t do too badly at this, though by the new year I think people in our department may have started to wonder why a lot of senior people seemed to be more sleep-deprived than usual, and kept booking lecture theatres for secret meetings. That said, there were some controversial leaks which ultimately may have backfired on their originator. I was lucky enough to attend the announcement party in Glasgow when things did finally get announced.
We were also very lucky with what we observed. The signal, which has the name GW150914_095045 (celebrating the date and time it was observed on), or GW150914 to its friends, was very loud. With only very minimal processing you can in fact see the signal by eye in the data. That’s rare. When it was detected the detectors were in an “engineering run”, basically a time when all the technology in the detector is turned on, but it’s being tuned and tweaked to make sure everything’s running as expected. You can’t normally make observations during it, because someone might be making adjustments to one of the thousands of optics to improve the sensitivity. (During an “observing run” this sort of tweaking is much more controlled). However, it was still the weekend in the USA at 10:50 am in the UK as the wave passed through. They’d left the detector running. It wasn’t a science run, but it might as well have been.
If the detector had been turned off over the weekend, and we’d missed GW150914, the next signal which we’d observe, in October 2015, was much less convincing. It would have been agonising: was it real, was it not? (We actually took years to decide that what we would eventually accept was our second detection was in fact real). Our first clear detection would have come right at the end of the observing run instead, GW151226, the “boxing day event”. I’m not even convinced we’d still have been observing over the holidays had we not had our real detection.
It’s been a singular experience, being able to be part of a new field of science developing, literally from day 1. On 13 September 2015 there were gravitational-wave astrophysicists. People who worked on the theory behind the observations. But on 14 September we got a new observational field: gravitational-wave astronomy.
Since then I’m very lucky to have worked with a lot of excellent people to analyse hundreds of gravitational wave signals. Two years after the first detection we’d finally observe a pair of colliding neutron stars, GW170817. In another gift from the cosmos, we’d see it simultaneously with electromagnetic emission: what was hailed as the holy grail of astronomy. By 2020 I’d finished my PhD, and started a postdoctoral position, still at Glasgow, and slowly got caught up in the work for creating the mammoth “catalogue” analyses, which work out the properties of the black holes which are involved in each signal. In the last year or so I’ve helped to lead teams which have analysed over 200 of these signals.
Not in my wildest dreams, signing up to the collaboration ten years ago to the day, could I have guessed this is what I’d do.
~~~
If you’re interested in how the story played-out for some other people in the field, the University of Glasgow have made a short documentary about the big day. It’s an excellent 8 minute watch: