Two observatories received the same brief pattern
On 14 September 2015, a short rising vibration passed through LIGO’s two observatories in the United States. Livingston, Louisiana, received it first. Hanford, Washington, followed about seven thousandths of a second later. During the loudest fraction of a second, the oscillations grew faster and stronger before fading. The pattern, named GW150914, was the first directly detected gravitational-wave signal. Its changing pace and final decay pointed to two black holes spiralling together and forming a single remnant.
The instruments had recorded no sound travelling through air. A gravitational wave changes the separations between freely moving objects in a characteristic pattern. LIGO translates that motion into changes in laser light. Scientists can later turn the measured oscillations into an audible chirp, but the original measurement is a tiny fractional change in length, called strain. The peak strain was about one part in a thousand billion billion.
Each observatory has two perpendicular arms, four kilometres long. A beam splitter sends laser light down both arms, and mirrors return it to the junction. A passing wave can lengthen one direction while shortening the other. The returning waves of light then combine differently at the output. Suspended mirrors, repeated light trips inside the arms and a carefully controlled laser let an enormous instrument respond to an extraordinarily small displacement.
The rising pitch told a physical story
An orbiting compact pair radiates gravitational waves and loses orbital energy. As the orbit shrinks, the bodies speed up. Their faster motion drives a faster-changing wave, so the crests crowd together as the signal grows. This rising pitch is the chirp. Its rate of increase carries information about a particular combination of the two masses. Astronomers call that combination the chirp mass.
In GW150914, roughly eight cycles led up to the amplitude peak over about two tenths of a second, with the frequency rising from around 35 to 150 cycles per second. The signal continued through the merger and the fading vibration of the remnant. The rapidly increasing frequency required objects close together; the mass scale inferred from that increase required objects much heavier than ordinary neutron stars. Black holes could supply both the mass and the necessary compactness.
To follow the final plunge, the collaboration compared the measurements with waveforms calculated from general relativity, including numerical solutions for the merger. The reported source masses were approximately 36 and 29 solar masses, with substantial uncertainties. The final black hole was about 62 solar masses. Roughly three solar masses’ worth of energy had escaped as gravitational radiation. These numbers came from fitting a family of physical waveforms to noisy detector data.
A second detector made the question sharper
A disturbance can shake a mirror, alter a control system or briefly spoil a detector’s output. One sharp trace on a screen would therefore leave a difficult question: did spacetime move, or did the apparatus misbehave? The separation between Hanford and Livingston offered a powerful check. A passing wave should appear in both places with a delay shorter than the light travel time between them. The wave should also produce related shapes after the different detector orientations are taken into account.
The teams examined the instruments and their surroundings. Seismometers, microphones and many control channels helped test whether a local or shared disturbance could explain the signal. Independent searches found GW150914: some looked for a coherent burst with few assumptions about its source, and others compared the data with predicted compact-binary waveforms. The two approaches addressed the same event with different expectations.
The search teams also shifted one detector’s data far enough in time to break any true astrophysical coincidence. Repeating this exercise produced many artificial pairings of noise. Those pairings showed how often noise could imitate a candidate as strong as GW150914 under the search procedure. The matched-filter search estimated a false-alarm rate below one event per 203,000 years. That rate summarizes the noise background estimated from the repeated artificial pairings.
The event date and the prize date belong to different parts of the story
The LIGO Scientific Collaboration and Virgo Collaboration announced the result on 11 February 2016. Only the two LIGO detectors were observing this event; Virgo’s detector was undergoing an upgrade. Virgo scientists nevertheless shared in the analysis through the joint collaboration. The discovery was the work of a large experimental and theoretical community, including the people who built the observatories, controlled their noise, calculated the waveforms and examined the data.
The 2017 Nobel Prize recognized Rainer Weiss, Barry Barish and Kip Thorne for decisive contributions to the detector and the observation. The signal had opened an observational route that complements light. A dark binary could tell astronomers about its masses and motion through the changing shape of a wave. The essential step was a physical measurement repeated at two sites, supported by a quantitative account of the noise surrounding it.