Yes, we've found another one

After the excitement of February when we (the LIGO Scientific Collaboration & Virgo Collaboration, or LVC for short) announced the first direct detection of a gravitational wave signal a lot of people having understandably been asking "Well, did you see any more?". The analysis performed for the announcement of the first detection (of the source called GW150914) used just over a month of data from the start of a longer observing run (that we called O1), which ran from the 12^th September 2015 until the 19^th January 2016. So we did have more data "in the can". And, as it happened that additional data did indeed provide us with another highly significant detection^*. This new signal was observed in what was the early hours of Boxing Day in the UK, although it was still Christmas Day in the US when it hit the two LIGO detectors - so we can call it a late Christmas present. As we work in Coordinated Universal Time (UTC), which follows Greenwich Mean Time (GMT), the signal has been given the title GW151226 (i.e. it arrived on 26^th December 2015), but internally has generally been called "The Boxing Day Event".

So, is this signal different from the first one? Well, like GW150914 it appears to be the result of two black holes inspiralling in to each other and merging, although the two merging black holes are smaller at roughly 14 and 8 times the mass of the Sun. A nice illustration of where these black holes sit, in terms of mass and radius, compared to other known black holes is shown here. However, unlike GW150914 we can pretty definitively say that at one of the merging black holes is spinning. Another thing to note is that if you look at Figure 1 from our detection paper for GW151226 you can't really see the signal in the data time series (whereas GW150914 stuck out like a sore thumb!) and it pretty much looks like noise.  As the system was less massive (but at a similar distance to) GW150914 the amplitude of the signal was intrinsically smaller. The saviour to this was that it also lasts longer in the detector's sensitive frequency bands (see Figure 1 in this paper that discusses all the detection's together) which means that you can integrate (basically sum together) over the longer signal and still "see" it in the noise.

Given that we'd already announced the detection you may be wondering what's important about this new one. The main thing is that we are now starting to reveal a population of objects rather than a single one. From looking at the population you can start to understand the distribution of source properties and investigate how they form. Admittedly with just two (and a bit) sources you really can't say much - it would be hard to work out the distribution of everyone's height by measuring just two people, but you at least get a rough idea of the likely range. It also allows us to be sure that the first signal wasn't a fluke, and suggests that we'll see many more of these objects in our upcoming observing runs (the next one, O2, should start this autumn with slightly better sensitivity than O1, and hopefully include the Virgo detector).

We often say that these gravitational waves are opening a new astronomical window on the Universe. And they really are! Imagine that the sky had always been covered in cloud, so you'd never been able to see the Moon, planets or stars (although in this scenario assume that you had a pretty good theory that the diffuse light coming through the clouds was being emitted by very distant objects called "stars".). Then, imagine that one night there's a slight chink in the clouds and through that you see a black sky with a single shining point of light in it. Wow! Your theory about "stars" was right! As the nights go on the clouds clear even more and reveal even more stars and other astronomical objects and the wonders of the Universe (and the exciting physics they reveal(!)) open up to you. It's a slightly tortured analogy, but you can kind of see that we're just seeing the first few points of light as the clouds are just starting to clear.

Some further information/reading:

• The science summary of these results can be found here.
• The GW151226 detection paper is here and a paper detailing all the binary black hole events detected in O1 is here.
• The data, and tutorials on using it (with ipython/jupyter notebooks), for GW151226 can be found here.
• A (probably non-exhaustive) list of blog posts by other LVC members about this detection (in no particular order):
• The Wave that Stole Christmas by Daniel Williams
• Merry Christmas, LIGO: Another Gravitational Wave! by Amber Stuver
• The Cosmic Classroom on Boxing Day by Shane Larson
• The Boxing Day Event by Christopher Berry
• Gravitational waves found again: here’s how they could whisper the universe's secret by Graham Woan
• Black holes rule! by Mark Hannam

^*The eagle-eyed of you may have know that amongst all the papers produced about our detection announcement there was also mention of a candidate gravitational event that was within the originally analysed dataset. We've estimated that this candidate, dubbed LVT151012 (for LIGO-Virgo Trigger), has a roughly 90% chance of being a real astrophysical signal, but we like to be far more certain than that to claim it as a definite signal.

The wait is over

Did you feel anything odd at around 09.50am GMT on 14th September 2015 (I'll let you do the time zone conversion)? Did you notice a disturbance in the force? Did you feel a tingle down your spine? Did you have butterflies in your stomach? Or, did you just feel a little bit wibbly?

No!? Well, at around that time a gravitational wave slammed into you at the speed of light, tried to rip your component atoms apart and then pull them together again¹, and then passed out the other side of you. But you didn't even notice, did you! It's not particularly surprising you didn't feel anything as the disturbance the wave produced was spectacular mainly in its minuscule effect - the waves would have attempted¹ to 'wibble' you from head-to-toe by only about 0.000000000000000000001 m (see Fig. 1 for an illustration of the effects of such a wave!).

Fig. 1. The effect of a passing gravitational wave on an alligator (as found around the LIGO Livingston observatory), a tumbleweed (as found in plentiful supply around the LIGO Hanford observatory) and myself (as found in the School of Physics & Astronomy at the University of Glasgow). Note that there is no discernible effect on any of these, except maybe as slight noticeable increase in my happiness at the prospect of the last 13 years of my working life having not been futile!

However, we (humanity in general, but large teams of scientists - many within the LIGO Scientific Collaboration [LSC] and Virgo Collaboration, including me - more specifically) have managed to build instruments that did indeed feel something on this date and time - a signal which we've given the catchy name GW150914 (or "The Event" as it was known for a while within the collaboration). These instruments, in this case the two US-based LIGO observatories (now entering their advanced phase), one in Hanford, Washington and the other in Livingston, Louisiana², both felt the waves' passing and saw a very consistent signal (see Fig. 2) - other than it looking like exactly what we'd expect from a gravitational wave source, our confidence that this was a real signal came from empirically estimating how often such a consistent and strong signal would have be seen by chance (i.e., from random [generally non-Gaussian] noise fluctuations in the detectors), which we work out as being less than once per 200,000 years. So, we're pretty sure (greater than 5.1σ in annoyingly frequentist statistical terminology), i.e. certain, that the signal was real. And, what's more, we've been able to use the pattern of wibbles the instruments felt to work out that this gravitational wave was emitted by two black holes, both tens of times more massive than the Sun, whacking into each other at about half the speed of light, to form the largest small(!) (~solar mass) black hole we know of. The amount of energy this event emitted was a whopping 5×10⁴⁷ Joules (quite possibly the most luminous event we've ever observed), equivalent to three times the mass of the Sun being converted directly to energy (remember E=mc²). Or, if you're into some "fun" energy conversions this is apparently equivalent to "the number of kilocalories in 2×10²⁸ cubic kilometres of butter (that's the volume of 14 billion Suns of pure butter!)" or "3 quadrillion times the energy required to destroy the planet Earth"!

Fig. 2. The GW150914 gravitational wave signal observed in the two LIGO detectors (this is figure 1 from the discovery paper, Abbott et al., Phys. Rev. Lett., 116, 061102 (2016)) (Credit: The LIGO Scientific Collaboration and the Virgo Collaboration)

This is the first ever detection³ we've had and it is quite a big deal, both scientifically (there's a whole load of awesome astrophysics that has been done using the signal, and it opens up a whole new area of astronomy) and personally. Some people have been in the gravitational wave detection game for almost five decades, but we've all had to wait patiently without seeing any definite sign of them in our detectors⁴ (this has occasionally been to the amusement of colleagues in other areas of physics and astronomy). As a member of the LSC myself since starting studying for my PhD at the University of Glasgow in the autumn of 2002 (when what is now known as the "initial" LIGO detectors had just started taking data) I've only been waiting 13 years, but that's still my entire working life. For some additional context here's what I wrote in my thesis acknowledgements section back in 2005:

I was attracted to the field of gravitational wave research due to the promise that we would be entering exciting times with several large scale projects bringing in unprecedented amounts of new data. Given this the discovery of gravitational waves would be just round the corner, opening up gravitational wave astronomy for real. Little did I know that this has been exactly what’s been said for around 30 years! Despite this I do actually believe that I’ve entered the field at a prime time and finding gravitational waves is just around, if not the first corner, then the next one.

So, ten years later, we have now turned that "next corner" and gravitational wave astronomy is finally with us! Many more detections should now be forthcoming in our future observation runs, hopefully including other exciting sources as well as more merging pairs of black holes.

Finally, the other rather cool, and timely, thing is that the signal arrived a century after Einstein published his General Theory of Relativity from which the prediction of gravitational waves arises. Einstein's prediction of gravitational waves and also Schwarzschild's solution to Einstein's equations from which predictions of black holes would arise, were both published a century ago in 1916.

The paper describing the detection and analysis of GW150914 has been peer reviewed and is now published (Abbott et al, Phys. Rev. Lett., 116, 061102, 2016) and further papers detailing the detectors, analyses and science results can be found here. Also, summaries (at a less detailed level) of the main science we've obtained from the signal can be found here. The data containing the signal and some example codes showing how to view, and hear(!), it are available here, so you should go ahead and take a look yourself.

More information, reactions and opinions about this amazing discovery (and the time-line of how the detection happened⁵) can be found in blog posts by several fellow collaboration members linked below and within a special edition of the LIGO Magazine:

• Amber Stuver (Post-doc, LIGO Livingston Observatory)
• Andrew Williamson (PhD student, Cardiff University)
• Bangalore Sathyaprakash (Professor, Cardiff University)
• Brynley Pearlstone (PhD student, University of Glasgow)
• Christopher Berry (Post-doc, University of Birmingham)
• Chris North (Lecturer, Cardiff University)
• Daniel Hoak (Post-doc, Virgo)
• Daniel Williams (PhD student, University of Glasgow)
• Mark Hannam (Professor, Cardiff University)
• Rebecca Douglas (PhD student, University of Glasgow)
• Roy Williams (Research Scientist, Caltech)
• Sean Leavey (PhD student, University of Glasgow)
• Shane Larson (Research Associate Professor, Northwestern University)

Also, be sure follow @ligo and @ego_virgo on twitter along with the hastags #GravitationalWaves, #EinsteinWasRight, #BinaryBlackHole and #AdvancedLIGO.

P.S. If you want to know what I was doing when GW150914 passed by it probably involved nappies (diapers for those in the US), feeding a child/cleaning bottles, or doing laundry, as I was on the final day of paternity leave following the birth of my second child (here's my first child one simulating a gravitational wave chirp). I didn't see the growing emails about the signal until early that afternoon when I thought I should try clearing out my inbox before returning to work the following day. It definitely made going back to work that bit more exciting. However, all the work to get the analyses of this event checked has slightly eaten into my main job, which is to search for gravitational waves from pulsars.

¹ Don't worry, gravity is very weak. The forces keeping your various constituent bits and bobs together are far more than enough to overcome any piddling gravitational wave that passes through you. Any displacement (stretching or squeezing) is only noticeable (if you have an exquisitely sensitive gravitational wave detector at least) between freely falling objects in the same local frame, i.e. if there are effectively no other external forces acting on the objects.
² Note that another detector called Virgo is also due to start taking data later this year, but wasn't operational at the time, and that a smaller (and unfortunately less sensitive, but still important) detector called GEO600 was operational, but not observing when the event passed by.
³ Prior to this direct detection (some quibbles over the "directness" of detection can be found here) many gravitational waves have obviously continuously been impinging on the Earth and passing through us, but this is the first time we've had the technology to catch one in the act.
⁴ There are very good reasons why they've not been seen until now (basically boiling down to us not having been able to build sensitive enough detectors), but that hasn't made us any less impatient.
⁵ The signal was first "detected" about three minutes after it arrived by online analysis software that looked for generic transient (short-duration) coherent signals, i.e. blips in the data that appeared at the same time in both detectors and looked similar. The first reasonably detailed estimates of the source parameters (i.e. that is was two black holes merging) were with us within about a day. After a couple of days we release our estimate of the location of the source in the sky and released it to selected astronomy groups to point their telescopes at. Following that full and proper detailed studies of the signal, and very careful checks on the performance of the detectors, have taken many months of painstaking work and placed great strain⁶ on the collaboration. But, given the general inertia that you get within a large collaboration (which we've experienced releasing results that didn't contain any signals) we've actually turned around a finished detection paper (after 14 draft iterations), and twelve(!) companion papers in remarkably short time.
⁶ This is a hilarious gravitational wave pun.

Disclaimer: everything on this blog is my own personal opinion and any mistakes are my own.