This answer to How precise are the observational measurements for the speed of gravity? says:

...in 2013 a Chinese group built a model using Earth's tides that helped them narrow it down.

... [T]he speeds of gravity are from 0.93 to 1.05 times the speed of light with a relative error of about 5%. This provides first set of strong evidences to show that the speed of gravity is the same as the speed of light.

This is so far the most accurate measurement I've seen. See the paper for more.

In the near future, LIGO may be able to provide more accurate measurements by comparing the distance among detectors and the delay of observation.

In case the links break, the papers are:

  • "Observational evidences for the speed of the gravity based on the Earth tide" TANG KeYun et al. Chinese Science Bulletin, February 2013 Vol.58 No.4-5: 474-477 doi: 10.1007/s11434-012-5603-3
  • "Bounding the speed of gravity with gravitational wave observations" Neil Cornish, Diego Blas, Germano Nardini 2017, https://arxiv.org/abs/1707.06101

update: As noted in @amateurAstro's comment and linked well-sourced answer the time between a gravitational wave detection and X-ray burst of GW170817 and GRB 170817A constrains the difference between the speed of gravity and the speed of light to be "...between $-3 \times 10^{-15}$ and $+7 \times 10^{-16}$ times the speed of light..."

So my updated question is:

Question: Have more recent LIGO/VIRGO gravitational wave measurements narrowed down the speed of gravity further?

  • 2
    $\begingroup$ See this related question and answers. $\endgroup$ Commented Aug 13, 2019 at 2:26
  • $\begingroup$ @amateurAstro I've updated the question to reflect your answer. It was a toss-up; I could have voted to mark this as duplicate as originally written, but this way allows for an updated answer, as well as any further analysis of that event in the subsequent years. Thanks for linking to it! It turns out I'd written about that event a few years ago as well; “Who saw” the binary neutron star merger first? What was the sequence of events? (GRB/GW170817) but completely forgot about it. $\endgroup$
    – uhoh
    Commented Aug 13, 2019 at 3:25
  • 2
    $\begingroup$ Unfortunately there hasn't been a second GW with electromagnetic counterpart. And judging from the citations in recent papers the analysis of GW170817 by Abbott, B. P., et al. 2017 is still be the most accurate one (and will likely remain so). $\endgroup$
    – SpaceCore
    Commented Aug 13, 2019 at 12:02
  • $\begingroup$ @SpaceBread then that is the answer to this question. If things change in a few more years (which they certainly might), an answer can be updated or a new answer added. $\endgroup$
    – uhoh
    Commented Aug 13, 2019 at 13:16

1 Answer 1


To get these kind of measurements one needs a gravitational wave detection with a electromagnetic counterpart. The big problem is, that the now quite common BH mergers are not expected to produce a detectable electromagnetic signature.
Therefore those can not be used to measure the speed of gravitational waves. At least not to any precision.
What you need is a binary neutron star merger. During the destruction and following accretion of the two NSs, these produce photons over the whole electromagnetic spectrum.

Until now, there have only been detected two such mergers, GW170817 and S190425z.
As mentioned by @amateurAstro, GW170817 gave us the precise constraint added in the question. Judging from the citations in recent papers (e.g. this) the analysis by Abbott et al. 2017 is also still up to date. For their constraint Abbott et al. assumed that the gamma-ray burst was emitted between 0 and 10 seconds after the gravitational wave. If future simulations would be able to further constrain that range, this will also give us a more precise limit on the speed of gravitational waves.

The other one is S190425z. The problem here is that the only 1/4 detectors were working so almost a quarter of the sky had to be searched. It was also quite far away. So unfortunately no counterpart was found.

So until we have a better theoretical understanding of NS merger or/and we observe another one, the constraint by Abbott et al. 2017 stands.

  • $\begingroup$ Thanks for the review and summary of the current situation! $\endgroup$
    – uhoh
    Commented Aug 13, 2019 at 15:39

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .