Skip to main content
Sandbox Physics

Gravitational-Wave Standard Sirens

How can a gravitational wave measure distance without a standard candle? The loudness of a merging binary’s gravitational waves follows from general relativity, so the waveform itself tells how far away the source is. Pair that distance with the redshift of its host galaxy and you get the expansion rate. Without a light flash, every galaxy at the right distance is a candidate, so one event says much less.

Drag the binary distance, orbit tilt and host galaxies. The wave amplitude gives DLD_L; a host redshift, or a whole galaxy catalog, turns it into H0H_0.

Interactive modelGravitational-Wave Standard Sirens
Observed signal-to-noise ρ\rho28.528.5
Distance from the wave DLD_L78−10+7.6 Mpc78^{+7.6}_{-10}\,\mathrm{Mpc}
Bright siren H0 (km s−1 Mpc−1)H_0\ (\mathrm{km\,s^{-1}\,Mpc^{-1}})69.7−7.2+1169.7^{+11}_{-7.2}
Dark siren H0 (km s−1 Mpc−1)H_0\ (\mathrm{km\,s^{-1}\,Mpc^{-1}})94.3−43+4594.3^{+45}_{-43}

Synthetic record, distance likelihood and Hubble constant

Synthetic: one injected event (true Hubble constant 70) analysed in this browser. No detector data.

Polarisation amplitudes h+, h×h_+,\ h_\times
  • Measured h+h_+
  • Measured h×h_\times
  • Injected

a+=20.1, a×=20.2a_+=20.1,\ a_\times=20.2 in noise units. The chirp shape is schematic; only the two amplitudes enter the fit.

Likelihood over distance and tilt (DL, cos⁡ι)(D_L,\ \cos\iota)
  • 68% and 95% regions
  • Injected source

Horizontal: DL/MpcD_L/\mathrm{Mpc}. Vertical: cos⁡ι\cos\iota, 1 is face-on. The ridge is the distance–tilt degeneracy.

Hubble constant p(H0∣x)p(H_0\mid x)
  • Bright: host redshift
  • Dark: whole catalog
  • H0inj=70H_0^{\mathrm{inj}}=70

Horizontal: H0/(km s−1 Mpc−1)H_0/(\mathrm{km\,s^{-1}\,Mpc^{-1}}). Violet ticks: the value each galaxy would imply. Curves include the selection correction.

Reading a distance from a waveform

BackgroundA supernova is a standard candle only after it is calibrated against nearer distances. A compact binary needs no ladder: general relativity fixes how loud its waves are for a given chirp mass, so the measured amplitude gives the luminosity distance DLD_L directly.

Why it mattersMove the binary farther away, tilt its orbit, and drag galaxies around to see how one distance turns into an expansion rate.

One-sentence intuition
The amplitude falls as 1/DL1/D_L, but a tilted orbit is fainter too. The ratio of the two polarisations measures the tilt, so the distance can be separated from it, best for orbits seen nearly edge-on.

Core mathematical model

Two polarisations

h+=ADL 1+cos⁡2ι2cos⁡Φ,h×=ADLcos⁡ι sin⁡Φh_+=\frac{\mathcal A}{D_L}\,\frac{1+\cos^2\iota}{2}\cos\Phi,\qquad h_\times=\frac{\mathcal A}{D_L}\cos\iota\,\sin\Phi

Leading-order amplitudes of a circular inspiral with inclination ι\iota. For a face-on orbit the two are equal, so a far face-on binary and a nearer tilted one look alike.

Distance and redshift

cz≈H0DL(z≪1),DL(z)=c(1+z)H0∫0zdz′E(z′)cz\approx H_0 D_L\quad(z\ll1),\qquad D_L(z)=\frac{c(1+z)}{H_0}\int_0^z\frac{dz'}{E(z')}

A host-galaxy redshift turns the distance into H0H_0. Nearby, a galaxy’s own peculiar velocity of a few hundred kilometres per second is a large part of its measured czcz.

Without a counterpart

p(H0∣x)∝p(H0)β(H0)∑i=1N∫dz p(x∣DL(z,H0)) p(z∣vˉi)p(H_0\mid x)\propto\frac{p(H_0)}{\beta(H_0)}\sum_{i=1}^{N}\int dz\,p(x\mid D_L(z,H_0))\,p(z\mid \bar v_i)

Every catalog galaxy ii with measured velocity vˉi\bar v_i is a possible host. One event gives several peaks; only many events stacked together isolate the true value.

Selection correction

β(H0)=∫0zmax⁡Pdet(DL(z,H0)) p0(z) dz\beta(H_0)=\int_0^{z_{\max}}P_{\mathrm{det}}\big(D_L(z,H_0)\big)\,p_0(z)\,dz

Only loud events are analysed. Dividing by the fraction of the catalog volume that would be detected for each H0H_0 removes the pull toward high values.

Common difficulties

The waveform alone does not give the redshift

Typical misconceptionGravitational waves measure both distance and redshift, so they give the Hubble constant by themselves.

Better mental modelThe waveform measures the redshifted chirp mass (1+z)M(1+z)\mathcal M, which mixes mass and redshift. A host galaxy, a catalog or an assumed mass distribution has to supply the redshift.

One synthetic event is not a measurement

Typical misconceptionIf the bright-siren curve here lands near 70, the Lab has measured the Hubble constant.

Better mental modelEvery number here comes from a synthetic record with injected values. The real GW170817 result, about 70 with an uncertainty of roughly 10 to 15 percent, came from detector data and an observed host galaxy.

Run the experiment

  1. 01

    Watch the distance band

    Drag the host galaxy along the cone from near to far.

    What to observe: The signal-to-noise ratio drops as one over distance and the cyan distance band widens.
  2. 02

    Tilt the orbit

    Drag the tilt handle on the orbit inset from face-on to edge-on, or compare the two degeneracy presets.

    What to observe: Face-on, the likelihood map stretches into a long ridge in distance. Edge-on, the cross polarisation vanishes, the source is fainter, but its distance is pinned down.
  3. 03

    Lose the flash

    Raise the number of catalog galaxies and compare the gold and violet curves.

    What to observe: The bright siren keeps one peak. The dark siren spreads over every galaxy inside the distance band.
  4. 04

    Drag one candidate

    Drag a violet galaxy into the cyan band.

    What to observe: A new peak appears in the dark-siren curve at the Hubble constant that galaxy would imply.

Related experiments and articles

What this page links to directly

See every article series on the content map →

Gravitational-wave standard sirens · Reading path

From a wave amplitude to the Hubble constant

Articles on this path that use this experiment, followed by the other experiments they suggest.

  1. ArticleHow can a gravitational wave tell us how far away it came from?
  2. ArticleWaveform amplitude, orbit tilt and luminosity distance
  3. ArticleFrom one distance and one redshift to the Hubble constant
  4. ArticleDark sirens, galaxy catalogs and the selection normalisation
  5. ArticleRecomputing a bright-siren and a dark-siren Hubble constant
  6. ArticleReading the GWTC-3 dark-siren cosmology paper
Something broken, confusing, or missing on this page?Tell us
What kind of feedback?

Sent with this page’s address, its Lab ID, the site language and your device type. Please leave out personal details. Feedback is kept for up to 12 months.Privacy policy