The historic discovery of gravitational waves announced this week involved the work of more than a thousand scientists working tirelessly in several different institutions, across many different countries and timezones.
Why is an entire village, albeit a diverse and disparate one, required to verify experimentally the last of Einstein’s major predictions in his theory of general relativity? And how does such a village function and coordinate in such a way that maximises scientific output?
A tour of the village
LIGO Scientific Collaboration consists of two individual experiments, located at two sites in the United States, separated by 3,000 kilometres. At each site, a single, very high-power laser beam is split in two, and travels down two perpendicular four kilometre-long vacuum tunnels.
At the ends of these tunnels the laser hits large, 40-kilogram mirrors suspended by an intricate series of pendula to reduce shaking from external forces.
The laser light returns along the same tunnel, and recombines. Gravitational waves cause the actual length of each arm to change. The way the laser light recombines is used to determine this change.
In order to make a detection, the LIGO instruments needed to measure a change in arm length equal to 1,000th the diameter of a proton. Performing such a measurement is a remarkable technological feat that involves development across multiple scientific streams.
These fields include, but are not limited to; quantum physics and quantum metrology; high-powered optics; mechanical systems including thermal and vibrational control systems; general relativity and gravitation; theoretical astrophysics and traditional astronomy; large-scale computing … the list goes on.
The multidisciplinary nature of this experiment is reflected in the structure of the LIGO Scientific Collaboration (LSC), which looks more like a corporate entity than a traditional scientific collaboration.
Among other things, there are many, many science working groups which fall within the scope of three main themes: instrument science, detector characterisation and data analysis.
Alongside the science working groups sit groups such as Education and Public Outreach, Diversity, and the Presentation and Publications Committee.
Each working group is a dynamic, scientific collaboration all unto themselves. Each has a chairperson, or multiple co-chairs, who report to the theme leaders who, in turn, report to the LSC spokesperson, executive committee and council.
Who works in the village?
So exactly how many scientists does it take to detect a gravitational wave? This particular effort took 1,006 scientists working tirelessly in 16 countries in 83 different institutions, located in 14 different timezones!
Research for the discovery was done all over North America, Brazil, throughout Europe, Russia, India, China and South-East Asia and Australia.
We, along with about 50 colleagues, work on this experiment in Australia. A majority of the leadership group work in institutes in the US, at places such as CalTech and MIT.
The result of this unfortunate circumstance is that full, collaboration-wide teleconferences typically take place between 2am and 4am in Australian time. Over the past two months, building to the announcement, this has affected our lives many times!
In general, science working groups hold weekly teleconferences. Many of us are part of working groups that only exist on two continents, making it possible to schedule meetings that also allow for a relatively normal existence.
Many of us also work in groups that have numerous members on three or more continents; very early, or very late teleconferences are not uncommon, but remind us of the scale of the collaboration and the international effect of our work.
What do they do?
As mentioned before, this is a precision measurement! Every aspect of the experiment is incredibly finely-tuned. For example, multiple groups and individuals around Australia work on the technology and design of the mirrors.
Monash University researcher Yuri Levin, while a PhD student of Kip Thorne’s at CalTech, developed the theoretical framework for computing thermal noise (which is now widely used within the collaboration). From this work it became clear that LIGO mirrors require exceptionally high-quality reflective coatings.
The coating noise Levin anticipated is now considered to be among the most serious sources of noise in the LIGO experiment.
Scientists at Adelaide University developed, installed and commissioned wavefront sensors for the LIGO mirrors that measure the mirrors’ change in shape due to the temperature of the high-powered laser, and corrects these distortions.
Researchers at CSIRO developed mirror coatings and polishing techniques for the initial phase of the LIGO experiment that lasted from 2002 to 2010. A team at the Australian National University developed tip-tilt mirror suspension systems that can be used to steer the laser light with remarkable accuracy.
A group at the University of Western Australia have built a mini-LIGO experiment that is used, among other things, to study an instability the high-powered laser can induce on the mirrors, causing them to wobble uncontrollably.
Each element and each component of the incredibly complex LIGO system undergoes incredible levels of development and scrutiny.
This is perhaps best exemplified in the data analysis sphere. In Australia, we have strong groups at Monash University, the universities of Melbourne and Western Australia, the Australian National University and Charles Sturt University.
We all work on developing and running computer software that can pick a tiny signal out of noisy data streams. Somewhat infamously, LIGO puts itself through a process called blind injections.
Blind injections are performed by a very small group of people. The team inject a fake signal into the data stream by artificially shaking the mirrors of the detector in such a way that makes it looks like a gravitational wave has passed through.
The unsuspecting data analysts play their usual games of analysing this data and, lo and behold, inevitably find the signal.
An early result?
The most famous of these blind injections occurred in September 2010. Very soon after the signal was automatically injected into the detectors, it was picked up with the initial data analysis algorithms.
The purported signal looked like it to came from the constellation Canis Major, and the event was subsequently called the “Big Dog”. The collaboration then went through a six-month process of vetting, checking, and re-checking the analysis, and even wrote up a full paper to be submitted to the journal.
An independent Detection Committee reviewed all of the results, and a collaboration-wide vote was held on whether to submit the paper for peer review – the result was an anonymous “yes”.
And then the envelope was opened: the signal was fake.
That exercise, while incredibly painful to many, shows just how seriously the LIGO Scientific Collaboration takes its science. That this latest detection was not a blind injection has been known by the entire collaboration for a long time – the experiment was only beginning to collect data, and the blind injection software had not yet been set up properly.
Less than one hour after the LIGO experiment wobbled from the gravitational wave on that fateful day on September 14, 2015, one of us (Lasky) and fellow Monash academic and LIGO researcher Eric Thrane, who sits on the fake injection committee, were sitting at our laptops at home when we both received an email titled “Very interesting event on ER8” (ER8 stands for Engineering Run 8, which was the name of the pre-science phase of the experiment).
A quick Skype conversation quickly ensued:
Thrane: Have you seen the email?
Lasky: Yes. Is it a false injection?
Lasky: Did we just detect a gravitational wave?
Thrane: I think we did.
And the rest, as they say, is history.
This is truly the dawn of a new age of discovery. The gravitational-wave universe has many untold stories to tell, and scientists across Australia are striving to tell the tale along with the rest of the world.