How Gravitational Waves Were Found

On September 14, 2015, at 5:51 AM Eastern time, a signal lasting 0.2 seconds swept through two detectors in Washington State and Louisiana. It was a rising tone — a chirp — climbing from 35 Hz to 150 Hz before ending abruptly. The signal was so faint that it took weeks of analysis and months of cross-checks before a team of over a thousand scientists was confident enough to call it real. When they did, on February 11, 2016, they announced to the world that humanity had directly detected gravitational waves for the first time.

The announcement was celebrated globally as one of the greatest scientific achievements in history. It confirmed a prediction Einstein had made 100 years earlier while doubting it could ever be verified. It opened an entirely new window on the universe. It earned Kip Thorne, Rainer Weiss, and Barry Barish the 2017 Nobel Prize in Physics. And it was the culmination of one of the most audacious engineering projects in human history — an effort spanning 50 years, consuming billions of dollars, and requiring the construction of instruments capable of measuring displacements one thousand times smaller than a proton.

Why Einstein doubted his own prediction

Einstein predicted gravitational waves in 1916, one year after completing general relativity. His equations showed that accelerating masses would produce ripples in the fabric of spacetime propagating at the speed of light. He then calculated the amplitude of these waves from realistic astronomical sources and concluded they would be immeasurably small. A binary star system at typical astronomical distances would produce waves that stretch and compress space by a fraction so tiny — roughly 10⁻²¹ — that no instrument could conceivably detect them. Einstein concluded in 1936, with co-author Nathan Rosen, that gravitational waves didn't exist at all — a paper he tried to publish, then retracted after a referee pointed out an error. The waves were real; they were just very small.

In the 1950s and 60s, physicists debated whether gravitational waves were even physically meaningful — whether they carried real energy that could affect detectors, or were merely mathematical artifacts. The debate was not fully resolved theoretically until 1957, when Richard Feynman gave his famous "sticky bead" argument: a gravitational wave would move beads on a stick, generating friction and heat, proving the waves carry energy and are physically real. If they could heat beads, they could in principle move mirrors in a detector.

The laser interferometer — the idea that made detection possible

The key insight came in the early 1970s from Rainer Weiss at MIT. Rather than looking for resonant vibrations in a metal bar, Weiss proposed using laser interferometry — splitting a laser beam into two perpendicular arms, bouncing it off mirrors at the end of each arm, and recombining the beams. When the arms are exactly equal length, the beams cancel perfectly in destructive interference and no light reaches the detector. A gravitational wave would stretch one arm and compress the other by different amounts, breaking the perfect cancellation and allowing light to reach the detector. The size of the light pulse encodes the wave's amplitude.

The elegance of the design masked an almost incomprehensible engineering challenge. The strain of astrophysically realistic gravitational waves — 10⁻²¹ — means that for a 4-kilometer arm, the mirror displacement is 4 × 10⁻¹⁸ meters. That is roughly 1,000 times smaller than the diameter of a proton. Measuring it requires a laser that is stable enough not to fluctuate at this level, mirrors that are isolated from every conceivable source of vibration, and a vacuum system (the laser beams travel through tubes evacuated to one-trillionth of atmospheric pressure) that does not allow any residual air molecules to scatter the beams. Every noise source — thermal vibration of mirror atoms, quantum fluctuations of the laser photons, seismic noise from Earth, distant ocean waves, passing trucks — had to be identified, modeled, and reduced below the target sensitivity.

The morning of September 14, 2015

LIGO had just undergone a major upgrade — "Advanced LIGO" — that increased its sensitivity by a factor of 3–4 and extended its frequency range. The upgrade had just been completed. The detectors were in an engineering run — not yet officially collecting science data — being tested and characterized. At 5:51 AM Eastern, the Livingston detector in Louisiana registered a signal. Seven milliseconds later, the Hanford detector in Washington State registered the same signal — the 7-millisecond delay corresponding to the travel time of a gravitational wave between the two sites at the speed of light.

The signal was immediately flagged by automated detection algorithms. It was a chirp — rising in frequency from 35 to 150 Hz, lasting 0.2 seconds. Its amplitude corresponded to a strain of about 10⁻²¹. Analysis of the waveform — fitting it against banks of theoretical templates calculated from general relativity for merging compact objects — indicated a merger of two black holes, one of 29 and one of 36 solar masses, at a distance of about 1.3 billion light-years. The merger had released energy equivalent to 3 solar masses in 0.2 seconds — making it briefly more luminous than all the stars in the observable universe combined.

The team spent four months verifying the signal before announcing it. They tested whether it could be terrestrial noise — an earthquake, a distant storm, a passing vehicle. They injected fake signals and confirmed the detection pipeline would have flagged them correctly. They ran sophisticated noise analyses. They convened review committees. The signal was real. The probability that it was a noise fluctuation mimicking a gravitational wave signal was less than one in 3.5 million — the "five sigma" threshold for discovery in physics. On February 11, 2016, they told the world.