# What did LIGO Actually See? (Gravitational waves discovery)

I am trying to find an original video/image of what LIGO actually saw, but all I can find is artist renditions of gravitational waves.

• As discussed in my answer below, LIGO is more like a microphone than a camera; so it makes more sense to talk about what we heard rather than what we saw. You can listen to the signal here: youtube.com/watch?v=TWqhUANNFXw Feb 17, 2016 at 14:29
• Wouldn't a better metaphor be a seismometer? Feb 17, 2016 at 22:42
• @user151841 Not really. Seismometers have three output data streams: acceleration in x, y , and z. Also, I think microphones are more intuitively familiar to the non-science public than seismometers. The LIGO detectors are also especially well suited to being compared to a microphone because the sensitive band of the detectors is completely within the range of human hearing. Feb 17, 2016 at 23:43
• If we want to get pedantic, technically LIGO's measurement is an actual video with an actual camera. All they do is take continual video feeds of the recombined laser's interference pattern. A lot of mathematical processing is necessary to produce the plots in the answers below. So really that video is what they actually "saw". Jun 5, 2017 at 13:07
• surely someone has "remixed" the audio in to human-hearable audio? where is that guys? it would be fantastic to listen to it, to get a sense of the attack/decay/length etc. surely this exists? all you'd have to do is modulate it up so many octaves right? Jun 5, 2017 at 13:56

## 7 Answers

The actual image isn't much. I was able to find it from Science, and this is all it is:

It's just a ripple, seen at slightly different times from two different observatories. The shift fits perfectly by shifting it by the speed of light difference in their locations. Thus is the proof of gravity waves.

It should be noted that the reason there are two instruments is to provide a cross check against other vibration sources. Each observatory works by detecting vibrations on a 4 km scale, down to a very small order of magnitude (1/10,000 the width of a proton). When the two are compared, then one can assume the signal must have come from a non-local source, which only Gravity Waves fit that definition.

• The original source is journals.aps.org/prl/pdf/10.1103/… Feb 17, 2016 at 12:03
• "The actual image isn't much", "this is all it is." Your tone understates how awesome it actually is IMO ;). Of course, I'm a little bit biased. Feb 17, 2016 at 19:31
• How do the two observation locations coordinate their times relative to a shared or common clock? Are they referring to the same atomic clock and making adjustments for "latency", the time it takes to get the time? Feb 18, 2016 at 21:28
• @TRomano We use GPS which is accurate to 10s of nanoseconds. You can read more about the aLIGO timing system here: authors.library.caltech.edu/20471/1/… Feb 19, 2016 at 14:03
• @ChrisMueller: I suspected it was GPS, but didn't have time to look it up at that moment. Thanks! Feb 19, 2016 at 14:07

First of all, I think your question belies a misunderstanding of the nature of the LIGO observatories. The nature of the detectors is that they act like a microphone, as opposed to a camera. What that means is that they are sensitive to gravitational waves coming in from most directions, but don't have an ability to distinguish where the waves came from. By using multiple detectors (which is also necessary for a confident detection) the difference in time between the detectors can be used to give some idea of the location of the source. It also means that the output of the detectors is a single stream of data.

This image from the paper in Physical Review Letters (not behind a paywall) is a better summary of what LIGO heard than the current accepted answer. I'll explain the panes from top to bottom.

1. The top panes show the 'raw' signals measured in the two detectors with the H1 data overlayed on the L1 data on the right.
2. The second row of panes shows a number of different simulations of what general relativity (Einstein's theory) predicts for the gravitational waves. These simulations are how LIGO is able to claim that they know that the wave was caused by two merging black holes.
3. The third row of panes is the 'raw' data minus the simulations.
4. The bottom panes are simply another way of plotting the 'raw' data called a time-frequency plot. Time is on the x axis and frequency is on the y axis. To a person from the field this signal is the most recognizable characteristic of a merger, known as a chirp. As time moves forward, the frequency shifts higher. You can actually listen to the 'raw' chirp here.

• It's not behind a paywall because the paper is open content—it is licensed under CC BY 3.0. Feb 17, 2016 at 19:25
• @bwDraco Good point. Feb 17, 2016 at 19:28
• Can you explain why the H1 observation in the upper right plot is marked as "inverted"? I haven't seen anywhere else before remark that H1 is inverted, but I can clearly see that is the case. What's the reason for this? Jun 5, 2017 at 13:16
• @zephyr: The two detectors are oriented differently (Hanford NW/SW, Livingston WSW/SSE), that might be the reason; I’m just guessing, though. Jun 6, 2017 at 7:43

LIGO didn't "see" anything. It monitors the relative lengths of the paths taken by two laser beams in vacuum pipes about 4km long (although the laser path consists of about 75 trips up and down the arms) and at right angles to each other.

A gravitational wave, travelling at the speed of light, changes the ratio of these lengths (one gets shorter, one gets larger, then they swap) by about plus or minus 1 part in $10^{21}$ (a billion trillion) about 30-200 times per second as it passes through the instrument.

The whole event lasted about 0.3 seconds and the trace (which has been all over the news) simply records the fraction by which the length of the arms changes as a function of time.

The event was (nearly) simultaneously recorded by two almost identical setups in different parts of the USA. The detection of the same signal in both detectors rules out a local cause of the disturbance, and the small time delay between the detections allows a rough location of the gravitational wave source on the sky.

• To me, not only is it an amazing accomplishment that we could detect such a small signal, but we could actually predict ahead of time what the signal would look like. I'm flabbergasted that by using models scientist can be fairly certain that the wave was produced by two 30 solar mass black holes colliding (the first publicly released discovery). Einstein rules!! Aug 29, 2017 at 3:53

According to the GW150914 tutorial, this is what Advanced LIGO L1 and H1 detectors originally saw:

You can download the raw data from this tutorial.

The other answers show already processed (whitened, filtered, shifted by 7 ms, inverted) waveforms.

• You are correct that this is what the raw data streams out of the detectors look like (note that I was careful in my answer to keep 'raw' in quotes). The sensitive band of the detectors ranges from 10 Hz to 100 kHz, but the raw data stream is dominated by the incredibly large (for LIGO) noise below 10 Hz. You can see this by comparing the units on your plot to those in the plots I posted. Part of the technologies LIGO employs to reach its unprecedented goal involve advanced signal processing. Feb 18, 2016 at 12:42
• You can see the actual noise curves of the detectors around the time of the detection here: dcc.ligo.org/public/0119/G1500623/001/… Feb 18, 2016 at 12:55

The actual mechanism of measurement that LIGO used is laser interferometry, so one reasonable interpretation of what LIGO "saw" would be the interference pattern caused by the gravity waves, which would "look" something like this:

Unfortunately I could not find an image of the actual laser interference that LIGO mentioned; it's probably too small for photography anyway.

All the other graphs people are linking do are just graphs of the interference pattern data. Showing a graph of the LIGO data as an answer to this question is like showing an image histogram as an answer to the question, "What does the Hubble space telescope see?"

• This is actually the interference pattern of two overlapping laser beams with different curvatures, and is what one might expect to see in a cheap interferometer (see e.g. Newton's rings). However, LIGO has incredibly well made mirrors so the interference at the output of the detector does not have any rings, and, in fact, is completely black on the scale of this image. Feb 18, 2016 at 12:46

I don't know whether it's interesting for you, but here is the link of the paper that was published about those observations:

http://journals.aps.org/prl/pdf/10.1103/PhysRevLett.116.061102

Once the answer above is quite straightforward! What the paper says (in short) is that LIGO has observed a transient gravitational-wave signal, and these observations match with the predictions of waveform derived by General Relativity for the system involving two black holes.

• Welcome to Astronomy! However, link only answers are usually dis encouraged. If you have something new to add, please sum it up in a few paragraphs. Feb 16, 2016 at 20:39
• LIGO update: Rumors are out .. sciencenews.org/article/… .. that LIGO may have observed two colliding neutron stars. This would be significant since it could be the first time that gravitational waves and electromagnetic waves are seen from the same source. Aug 29, 2017 at 4:03

It sees waveforms as shown in the drawns above.

Some examples you can get and explore by yourself from here: