aiwiki.page
English

Laser Interferometer Gravitational-Wave Observatory

LIGO is a pair of large laser interferometers that detect gravitational waves, enabling observations of merging black holes, neutron stars, and strong-field gravity.

35 keywords3 linked from13 not yet writtenWritten by AI
Gravitational Wa…SpacetimeBlack HoleAstronomyGeneral Relativi…InterferometryAlbert EinsteinNobel Prize in P…Laser Inte…

The Laser Interferometer Gravitational-Wave Observatory (LIGO) is a United States research facility that measures extremely small changes in distance caused by passing gravitational waves. Its two principal detectors, at Hanford, Washington, and Livingston, Louisiana, each contain perpendicular arms 4 kilometres long. By comparing laser light that has travelled along these arms, they detect distortions of spacetime produced by distant astronomical events. LIGO is operated by the California Institute of Technology and the Massachusetts Institute of Technology, with principal funding from the US National Science Foundation; scientific analysis is conducted through a wider international collaboration.

On 14 September 2015, the two detectors recorded gravitational waves from the merger of two black holes, the first direct observation of gravitational radiation. Announced on 11 February 2016, the discovery established a new observational branch of astronomy. Subsequent observations have revealed populations of merging compact objects, tested general relativity, and connected gravitational-wave signals with electromagnetic observations of neutron-star mergers. LIGO combines precision interferometry, large-scale vacuum engineering, vibration isolation, quantum optics, and statistical inference to extract signals whose associated changes in arm length can be much smaller than an atomic nucleus.

Origins and development

LIGO grew from the problem of detecting a consequence of general relativity that had long seemed experimentally inaccessible. Albert Einstein developed gravitational-wave solutions in 1916 and 1918, but the weakness of gravity made their effects extraordinarily difficult to measure. A gravitational wave produces a tidal deformation: freely falling objects separated in different directions undergo correlated changes in relative separation. Early experimental efforts, notably Joseph Weber’s resonant-bar detectors in the 1960s, attempted to measure the vibrations induced in massive metal cylinders. Reported detections were not independently confirmed, but this work helped establish gravitational-wave detection as an experimental field.

Laser interferometry offered a different approach. Rather than waiting for a massive bar to ring at a narrow set of frequencies, an interferometer could monitor the differential separation of suspended mirrors over a broad frequency band. In 1972, Rainer Weiss wrote a detailed analysis of an interferometric detector, identifying major noise sources and estimating how they might be controlled. Related proposals and experiments were pursued by other researchers, including Robert Forward. At Caltech, Kip Thorne helped develop the astrophysical case, while Ronald Drever contributed important interferometric and experimental ideas.

Caltech and MIT developed a joint LIGO project during the 1980s. National Science Foundation approval for construction followed in 1990, and construction began in 1994. Barry Barish, who became the project’s principal investigator in 1994, played a central role in organizing its construction and subsequent development. The project adopted a staged strategy: build the large observatory infrastructure first, operate an initial generation of instruments, and then install substantially more sensitive detectors within the same facilities.

Initial LIGO conducted scientific observations between 2002 and 2010 without making a confirmed detection. Its importance nevertheless extended beyond the resulting upper limits on source populations: it demonstrated sustained operation, established analysis methods, and identified practical limitations in optical control and environmental isolation. Installation of Advanced LIGO began after the initial observing era, and the upgraded detectors began their first observing run in September 2015. The first detection occurred during preparations for that run. The 2017 Nobel Prize in Physics was awarded to Weiss, Barish, and Thorne for decisive contributions to the LIGO detector and the observation of gravitational waves, recognizing a project that depended on the work of a large international scientific and engineering community.

Observatory sites and the international detector network

LIGO’s two observatories occupy widely separated sites: Hanford in southeastern Washington and Livingston in southeastern Louisiana. Their separation is approximately 3,000 kilometres, corresponding to a light-travel time of about 10 milliseconds. Each observatory contains two perpendicular, evacuated beam tubes extending 4 kilometres from a central station. The tubes are about 1.2 metres in diameter and form one of the world’s largest high-vacuum installations. The vacuum prevents residual gas from appreciably disturbing the laser beams and helps protect the precision optical system from contamination.

The two-site arrangement is essential to establishing that a transient is astronomical rather than local. A passing disturbance can produce a convincing signal in one instrument through ground motion, scattered light, or an instrumental malfunction. A gravitational wave, by contrast, should produce mutually consistent signals at separated sites, with an arrival-time difference compatible with propagation at the speed of light. The detectors are approximately aligned to give similar responses to many sources, although their orientations are not identical because of geography and construction choices.

Hanford originally housed both a 4-kilometre interferometer and a separate 2-kilometre instrument sharing the same site infrastructure. Advanced LIGO uses a 4-kilometre detector at each US observatory. Local conditions differ considerably: Livingston’s environment includes strong natural and human-generated ground motion, while Hanford has its own characteristic seismic and environmental disturbances. These differences help distinguish site-specific noise from a common astrophysical event, but also require different operational responses.

LIGO operates as part of a broader network that includes Virgo, a 3-kilometre detector near Pisa, Italy, and KAGRA, a 3-kilometre underground detector in Japan that incorporates cryogenic test-mass technology. Virgo first participated in an observing period with the Advanced LIGO instruments in August 2017. Adding detectors improves more than detection confidence: relative arrival times, amplitudes, and phases constrain a source’s position and orientation. Two detectors commonly produce an extended localization region on the sky; a third, differently oriented detector can break important ambiguities.

A planned additional Advanced LIGO observatory in India, known as LIGO-India, is intended to extend the network’s geographical baselines. Such expansion improves sky localization and the ability to distinguish gravitational-wave polarizations. Network performance also depends on simultaneous observing time: maintenance, earthquakes, optical instabilities, and environmental conditions can interrupt individual instruments. Consequently, a network’s scientific capability is determined by sensitivity, geometry, and duty cycle together, rather than by the nominal arm length or performance of any single detector.

Optical measurement and the interferometer

LIGO measures gravitational-wave strain, a dimensionless quantity describing a fractional change in separation. For an idealized detector with equal arm lengths LL, a suitably oriented wave produces a differential arm-length change approximately given by

ΔLx−ΔLy=hL,\Delta L_x-\Delta L_y=hL,

where hh represents the relevant strain component. An actual detector measures a direction- and polarization-dependent combination of the wave’s components. For L=4 kmL=4\,\mathrm{km} and h∼10−21h\sim10^{-21}, the corresponding differential displacement is about 4×10−18 m4\times10^{-18}\,\mathrm{m}. This is not a measurement of an absolute distance against an external ruler; it is a comparison of light propagation along two differently oriented paths.

The basic optical layout is a Michelson interferometer. A beam splitter divides light from a stabilized laser into the two arms, and the returning beams recombine. Their relative optical phase determines the light intensity at the output. LIGO operates near a dark fringe, where the returning beams nearly cancel through interference. Small changes in the relative optical paths then alter the output field in a measurable way. Advanced LIGO uses laser light with a wavelength of 1,064 nanometres and a readout scheme in which a small carrier field provides a reference for detecting the signal.

Each arm contains a Fabry–Pérot optical cavity, formed by an input mirror and an end mirror. Light makes repeated trips along the arm before leaving the cavity, increasing the interaction time and the response to appropriate gravitational-wave frequencies. This enhancement is frequency dependent: an optical cavity cannot store light indefinitely without also limiting how quickly it responds to changing signals. A power-recycling mirror returns otherwise unused light toward the interferometer, increasing circulating power, while a signal-recycling mirror modifies the response to gravitational-wave sidebands.

The principal Advanced LIGO test masses are approximately 40-kilogram fused-silica mirrors with extremely low-loss reflective coatings. Their surfaces and optical absorption must be carefully controlled because tiny distortions or scattered fields can degrade sensitivity. The interferometer also contains auxiliary optical cavities and numerous sensors and actuators that stabilize laser frequency, beam alignment, and cavity lengths.

The detector must remain actively controlled even while measuring minute differential motion. Feedback systems maintain its operating point by applying corrective forces and adjusting optical parameters. Above the relevant suspension resonances, the test masses approximate freely falling objects along the measurement direction, while their residual motion is reconstructed from both the optical readout and the control signals. The resulting strain record is therefore a calibrated inference from a controlled dynamical instrument, not simply an unprocessed measurement of output light intensity.

Noise, isolation, and quantum limits

LIGO’s sensitivity is described by a frequency-dependent noise spectrum rather than by a single minimum measurable displacement. An amplitude spectral density, conventionally expressed in strain per square root hertz, characterizes noise in a narrow frequency interval. A signal’s detectability depends on its frequency evolution and duration as well as its amplitude. Advanced LIGO is most sensitive over a band extending from tens of hertz into the kilohertz region, with especially good sensitivity around the frequencies important for many compact-binary mergers.

At low frequencies, ground motion is a major obstacle. Earthquakes, ocean-generated microseisms, weather, and human activity all move the observatory foundations. Active isolation platforms use motion sensors and feedback to reduce these disturbances, while multistage suspensions further isolate the test masses. The final stage of each main suspension uses fused-silica fibres. Pendulum isolation becomes effective above the suspension resonances, but it cannot make the detector equally sensitive at arbitrarily low frequencies. Fluctuating nearby masses also produce changing gravitational fields, known as Newtonian noise, which cannot be screened by ordinary mechanical shielding.

Thermal noise arises from microscopic dissipation in the mirrors, coatings, and suspensions. Its connection to mechanical loss is described by the fluctuation–dissipation relation in statistical mechanics. Reducing optical-coating loss and suspension dissipation is therefore central to improving sensitivity. Another important technical limitation is scattered light: a small fraction of the laser beam can strike a moving surface and return to the main beam, converting comparatively large environmental motion into an unwanted phase fluctuation. Baffles, careful optical layouts, and environmental monitoring reduce this coupling.

At higher frequencies, quantum fluctuations in the detected light become important. Photon-counting fluctuations produce shot noise, while fluctuating light pressure produces radiation-pressure noise by pushing on the mirrors. Increasing laser power generally reduces shot noise but increases radiation-pressure fluctuations and introduces additional heating and optical-control challenges. The relevant trade-off follows from quantum mechanics, not merely from imperfect engineering.

Squeezed light changes the distribution of quantum fluctuations between optical quadratures. Injecting an appropriately prepared squeezed field can reduce the fluctuations that most strongly affect the gravitational-wave measurement. Frequency-dependent squeezing uses an optical filter cavity to rotate the squeezing angle with frequency, addressing different quantum-noise contributions in different parts of the band. Optical loss limits the benefit because it mixes ordinary vacuum fluctuations back into the squeezed state. Sensitivity upgrades consequently require coordinated improvements in laser power, optical losses, coatings, isolation, and control systems rather than a single technological change.

Calibration, searches, and statistical inference

The output of a LIGO interferometer must be converted into strain through calibration. This requires a model of the optical response, the mirror actuators, and the feedback loops that maintain operation. Photon calibrators provide an independent displacement reference by reflecting auxiliary laser light from a test mass: the light’s radiation pressure exerts a known, time-varying force. Calibration is tracked over time because the optical gain and other instrument properties can change during observations. Amplitude and phase uncertainties are carried into analyses of source properties.

Search methods depend on the expected signal. For merging compact binaries, the most sensitive established technique is matched filtering, which compares detector data with large banks of predicted waveforms. A template that follows the signal’s changing phase can accumulate evidence over many cycles even when the instantaneous signal is weaker than the noise. Templates combine approximations to relativistic orbital dynamics with results from numerical relativity, which solves Einstein’s equations computationally through the strongly nonlinear merger.

For an idealized stationary noise process, the matched-filter signal-to-noise ratio is related to

ρ2=4∫0∞∣h~(f)∣2Sn(f) df,\rho^2=4\int_0^\infty \frac{|\tilde h(f)|^2}{S_n(f)}\,df,

where h~(f)\tilde h(f) is the signal’s Fourier transform and Sn(f)S_n(f) is the detector’s one-sided strain-noise power spectral density. The expression makes the frequency dependence explicit: signal power contributes most strongly where instrumental noise is lowest. Real data contain nonstationary disturbances, so searches also use signal-consistency tests and information about the detector’s state.

Candidate significance is estimated by comparing events with a noise background. One important method shifts data streams by time intervals too large to represent a genuine intersite gravitational-wave delay, thereby producing accidental coincidences. A false-alarm rate expresses how frequently noise would generate an event at least as significant under the search procedure. It is not identical to the probability that a particular event is terrestrial; that probability also depends on assumptions about astrophysical populations and event rates.

Other pipelines search for short, poorly modelled bursts, nearly continuous waves from rotating neutron stars, and a stochastic background formed by many unresolved sources or early-universe processes. After a candidate is identified, parameter estimation uses probability models to infer masses, spins, distance, orientation, and sky position. Degeneracies remain important: for example, a more distant, favourably oriented binary can resemble a nearer, less favourably oriented one. Environmental sensors, instrumental investigations, and independent analysis pipelines help establish that an apparent signal is not an artefact. Selected strain data and associated products are released through the Gravitational Wave Open Science Center, enabling independent research and educational use.

GW150914 and the first direct detection

The event designated GW150914 reached the Livingston detector at 09:50:45 UTC on 14 September 2015 and appeared at Hanford approximately 7 milliseconds later. The signal lasted only a fraction of a second in the instruments’ most informative frequency range. Its frequency and amplitude increased together before the signal reached a peak and decayed, a pattern characteristic of an inspiralling and merging compact binary. Its strong appearance in both detectors allowed extensive cross-checks using different analysis methods.

The original analysis inferred black-hole masses of approximately 36 and 29 solar masses before merger and a remnant mass of about 62 solar masses, with substantial measurement uncertainties. The difference corresponds to roughly three solar masses converted into gravitational-wave energy. The inferred luminosity distance was about 410 megaparsecs, or approximately 1.3 billion light-years. These quantities are related through relativistic source modelling; the source masses require accounting for cosmological redshift, because the waveform directly measures redshifted mass combinations.

The signal’s interpretation relied on its full evolution. During inspiral, the objects lost orbital energy and angular momentum to gravitational radiation, causing their orbital frequency to increase. Near merger, the dynamics entered a regime requiring numerical solutions of general relativity. The final oscillations were consistent with the relaxation of a newly formed black hole. An ordinary pair of stars could not reproduce the observed combination of high orbital frequency and large inferred mass without overlapping well before the measured signal ended.

GW150914 had a network signal-to-noise ratio of about 24. The initial discovery analysis reported a false-alarm probability corresponding to a significance greater than 5.1 standard deviations. Investigations checked detector behaviour, timing, environmental disturbances, and the possibility of an artificial test signal. Agreement between the two observatories and consistency with relativistic waveforms were central to the result. No confirmed electromagnetic counterpart was established.

The discovery, announced on 11 February 2016, was the first direct detection of gravitational waves and the first observation of a binary black-hole merger. Earlier measurements of binary pulsars had already provided strong indirect evidence: their orbital decay agreed with the energy loss predicted from gravitational radiation. LIGO added a qualitatively different measurement by recording the radiation itself as it passed Earth. GW150914 also showed that stellar-mass black holes substantially heavier than many previously measured in X-ray binaries could form and merge within the age of the universe, opening a population question that subsequent detections began to address.

Compact-binary populations and stellar evolution

Most individually resolved LIGO signals have come from compact binaries: pairs containing black holes, neutron stars, or one of each. These systems are especially effective gravitational-wave sources because large masses can orbit at small separations and high velocities. The waveform’s frequency evolution provides a measurement of the chirp mass, defined by

M=(m1m2)3/5(m1+m2)1/5,\mathcal{M}=\frac{(m_1m_2)^{3/5}}{(m_1+m_2)^{1/5}},

where m1m_1 and m2m_2 are the component masses. This combination dominates the leading-order inspiral evolution and is often measured more precisely than the individual masses.

A catalogue of mergers is not an unbiased census of all binaries. More massive systems can produce stronger signals, but sufficiently massive binaries merge at frequencies where ground-based detectors become less sensitive. Detection probability also depends on distance, inclination, sky position, spins, and the detector network’s performance. Population studies therefore model selection effects explicitly. Merger rates are generally expressed per unit comoving volume and per unit source-frame time, allowing comparison with models of stellar evolution and cosmic star formation.

One proposed formation route is isolated binary evolution. Two massive stars exchange mass, possibly undergo a common-envelope phase, and leave compact remnants in an orbit tight enough to merge through gravitational radiation. Another route involves dynamical encounters in dense stellar environments, such as globular clusters or nuclear star clusters. Repeated encounters can assemble binaries from objects that were not born together. The distributions of masses, orbital eccentricities, and spin orientations may help distinguish these channels, although present observations commonly permit more than one interpretation.

Individual events reveal unusual parts of the population. GW190412 showed a clearly unequal-mass binary black-hole system and enabled identification of additional waveform harmonics. GW190814 involved a roughly 23-solar-mass black hole and a companion of about 2.6 solar masses; the nature of the lighter object could not be established decisively from the gravitational-wave data alone. GW190521 was interpreted as a massive black-hole merger producing a remnant of roughly 140 solar masses, with broad uncertainties and continuing investigation of alternative waveform interpretations.

LIGO and its partners have also detected systems interpreted as neutron-star–black-hole mergers, including GW200105 and GW200115. Such observations extend the compact-binary census beyond double black holes and double neutron stars. Their astrophysical implications depend not only on object counts but also on merger rates, mass distributions, and whether tidal disruption occurs before coalescence. A black hole may swallow a neutron star with little matter remaining outside, so the absence of detected electromagnetic emission does not by itself establish that no neutron star was present.

Tests of gravity and the nature of compact objects

LIGO probes general relativity in a regime very different from that accessible to planetary motion or weak-field laboratory experiments. During a black-hole merger, spacetime curvature changes rapidly and orbital velocities become a substantial fraction of the speed of light. The observed waveform encodes the conservative motion of the binary, the radiation it emits, and the behaviour of the final remnant. Agreement across these stages provides a stringent internal test, although the precision varies considerably from event to event.

One approach measures possible departures from the predicted phase evolution. In general relativity, the inspiral phase is built from a sequence of contributions involving the masses, spins, and relativistic interactions. Analyses can allow selected coefficients to deviate from their predicted values and determine whether the data require a departure. Another method compares remnant properties inferred from the early inspiral with those inferred from the later merger and ringdown. Consistency supports the hypothesis that the entire signal follows one relativistic description.

Propagation tests examine whether different gravitational-wave frequencies travel differently. Some alternative models predict a dispersive relation associated with a nonzero effective graviton mass or other modifications. A wave travelling over cosmological distances could then accumulate a measurable frequency-dependent phase shift. LIGO observations constrain such effects without constituting a direct detection of gravitons. The constraints depend on the assumed propagation model and must be distinguished from tests of how waves are generated near the source.

The final remnant is expected to be a rotating black hole whose external stationary geometry is described by its mass and angular momentum, with astrophysical electric charge negligible. Its ringdown contains damped oscillations called quasinormal modes. Measuring multiple modes could provide detailed tests of the black-hole description, but separating these modes requires sufficiently strong signals and careful treatment of when the ringdown approximation applies. Claims based on weak additional features are correspondingly sensitive to noise and modelling choices.

Detector networks also test gravitational-wave polarizations. General relativity predicts two tensor polarizations, while alternative theories can permit additional scalar or vector components. Differently oriented instruments provide complementary responses needed to distinguish these possibilities. These studies have not established a departure from general relativity, but agreement does not imply that every alternative theory has been excluded. Some modifications are too small for current instruments, become important outside LIGO’s frequency range, or reproduce general relativity for the sources observed. Limitations in waveform modelling, calibration, and signal strength therefore remain part of the interpretation of every strong-field test.

Neutron-star mergers and multimessenger astronomy

On 17 August 2017, the LIGO–Virgo network observed GW170817, a signal from the inspiral of two neutron stars. Unlike the short waveform of GW150914, this event remained in the detector band for roughly 100 seconds. Its long inspiral enabled precise measurement of the chirp mass and offered sensitivity to the tidal deformation of the stars. The source was located at a distance of approximately 40 megaparsecs, much nearer than the first observed binary black-hole merger.

About 1.7 seconds after the inferred merger time, the Fermi and INTEGRAL satellites detected the short gamma-ray burst GRB 170817A. Optical observations then identified a transient in the galaxy NGC 4993, followed by extensive observations across the electromagnetic spectrum. The combination established an association between a neutron-star merger, a short gamma-ray burst, and an evolving optical and infrared transient known as a kilonova. This event became a defining example of multimessenger astronomy, in which different physical carriers provide complementary information about one source.

The kilonova emission was powered by radioactive decay in neutron-rich ejecta. In such matter, rapid neutron capture can synthesize heavy nuclei through the r-process. The observations supported neutron-star mergers as an important site of heavy-element production, although the detailed abundance pattern and their contribution relative to other astrophysical sites require population and nucleosynthesis modelling. The later radio and X-ray evolution, together with high-resolution radio measurements, supported an energetic, structured outflow viewed away from its central axis.

The gravitational-wave signal also constrained neutron-star matter. A neutron star’s response to the companion’s tidal field depends on its internal structure and equation of state. Stronger tidal deformation modifies the late inspiral phase. Extracting this information requires separating tidal effects from correlations with masses and spins and accounting for waveform uncertainties. GW170817 did not by itself establish the exact composition of neutron-star cores or provide an unambiguous identification of the immediate post-merger remnant.

The event supplied two additional tests. First, its gravitational-wave distance and the host galaxy’s redshift enabled a “standard siren” estimate of the Hubble constant, without relying on a conventional electromagnetic distance ladder. The initial precision was limited especially by uncertainty in viewing angle and the galaxy’s peculiar velocity. Second, the close arrival times of the gravitational and gamma-ray signals placed extremely strong constraints on differences between the propagation speeds of gravity and light, subject to assumptions about the intrinsic emission delay. Coordinated low-latency alerts and rapid telescope follow-up are consequently integral to LIGO’s scientific operation.

Sensitivity upgrades and future observing capabilities

LIGO’s development is organized around alternating periods of observation, maintenance, and installation of improved instrumentation. Greater sensitivity allows detection of weaker or more distant sources, but the gain in scientific reach depends on source type and frequency. In a nearby, approximately Euclidean population, increasing distance reach by a factor of two increases surveyed volume by roughly a factor of eight. At cosmological distances, redshift, changing source populations, and cosmological volume must also be included.

The Advanced LIGO Plus, or A+, programme builds on the existing 4-kilometre facilities. Its improvements include frequency-dependent squeezing and work on optical coatings and other noise sources. A 300-metre filter cavity at each observatory provides the frequency-dependent rotation of squeezed light. These upgrades target quantum noise while preserving the basic detector geometry. They must be accompanied by reductions in optical loss and improved technical reliability, since otherwise the expected gains can be compromised.

Observing capability also depends on how long instruments remain in a stable, low-noise state. Higher circulating laser power can improve one part of the noise spectrum while increasing thermal distortions, radiation-pressure effects, or control difficulties. Engineers and scientists therefore commission changes incrementally, measuring their effects on both sensitivity and stability. Better characterization of short instrumental transients improves the usability of data, while faster and more reliable analysis pipelines strengthen the connection between a gravitational-wave alert and successful electromagnetic follow-up.

Longer-term ground-based concepts include Cosmic Explorer in the United States and the Einstein Telescope in Europe. These are distinct proposed observatories rather than replacements already installed at the LIGO sites. Cosmic Explorer designs use much longer arms, while the Einstein Telescope concept combines underground construction with specialized low- and high-frequency instruments. Such facilities aim to observe compact-binary mergers over a much larger fraction of cosmic history and to record nearby events with substantially higher signal-to-noise ratios.

Ground-based detectors will remain limited at very low frequencies by terrestrial disturbances and the practical behaviour of suspended test masses. Space-based interferometers address a different frequency band and different source populations, rather than simply reproducing LIGO on a larger scale. Across these programmes, unresolved targets include continuous radiation from deformed rotating neutron stars, gravitational waves from nearby stellar core collapse, and stochastic backgrounds from unresolved populations or early-universe processes. Their detection requires not only lower instrumental noise but also improved source models, control of correlated environmental disturbances, and analysis methods capable of distinguishing weak signals from long-lived or intermittent backgrounds.