Data Releases for Events

This page shows data for validated gravitational wave data surrounding confirmed discoveries, as well as data corresponding to important non-detections where detectable gravitational waves might plausibly have been expected, e.g., the data surrounding one or more gamma-ray bursts, supplying evidence for or against a scientific model.

Binary black hole (BBH) events in Advanced LIGO

Events from the O2 run

2016 Nov 30 16:00 UTC to mid-2017 (TBD)

Events from the O1 run

2015 Sep 12 0:00 UTC to 2016 Jan 19 16:00 UTC

Audio files

The binary black hole merger events the LIGO has detected are in the audio band. The they can be converted to sound (.wav) files, so that you can hear them. (Of course, GWs are not sound, and they travel at the speed of light, not sound. However, many of us listen to music on the radio; radio waves arent's sound, and they travel at the speed of light as well; but our radios have no problem converting that to sound, as well).

Some subtleties:

  • LIGO data has lots of noise at both low and high frequencies, so it's virtually impossible to "hear" the weak signals unless the data are first whitened and bandpassed to suppress the low- and high-frequency noise, leaving the quiet band between 40 Hz and 300 Hz untouched.
  • We also whiten the waveform template derived from GR that best matches the data.
  • The signals are at the low end of the audio band (below a few hundred Hz). They sound like faint thumps; it's hard to hear the "chirp". We can frequency-shift the data and the waveform template by +400 Hz, into the sweet spot of our hearing range; then we can hear the chirp better.
  • All the signal processing details are in our binary black hole event tutorials; for example, for GW150914.
You can download the resulting 4-second-long *.wav files, here:

Important Non-detections in Initial LIGO

Rapid Triggers from LIGO Data

The LIGO Scientific Collaboration (LSC) and the Virgo Collaboration currently started taking data in 2015, and we expect the sensitivity of the network to improve over time. Gravitational-wave transient candidates will be identified promptly upon acquisition of the data; we aim for distributing information with an initial latency of a few tens of minutes initially, possibly improving later.