3 ms·

For 130 million light years the difference between gravitational wave signal and gamma burst is 2s. That constrains the speed difference to 1.6e-18 m/s. This is

For 130 million light years the difference between gravitational wave signal and gamma burst is 2s. That constrains the speed difference to 1.6e-18 m/s. This is fascinating number especially given that speed of light is 3e8.

alok-g 9y ago
Newbie question: While this is a very small fractional difference, could it be theoretically significant? What is the explanation of this difference? My best guess is that this difference must also be there in when the waves _started_.
- SideburnsOfDoom 9y ago
  I'm not a physicist, but I did do the 101 course a long time ago. I read "That constrains the speed difference to 1.6e-18 m/s" as "we measured as accurately as we could, and if they do differ, they must differ by this amount or less, which could just be the tolerances of our experiment" In other words, there's always a confidence interval, the trick is to measure it and minimise it: The speeds could be identical numbers, but what was measured was an either no or a very small difference. Larger differences have been ruled out. Identical numbers are _suggested_.
  - alok-g 9y ago
    What is observed is the delay between the two signals. Fractional speed difference is calculated from it. As I understand, there is little chance that the 2 second delta isn't real.
    - SideburnsOfDoom 9y ago
      OK, but there are other possible explanations besides fundamental physics: either the order of events at the source, or the effects of the interstellar medium along the way.
- fpoling 9y ago
  As neutron stars are small, like 20 kilometers, while the light travels 600 000 km within 2 seconds and one can roughly assume that speed of processes that generates the gamma burst during the stars collisions matches the speed of light the delay may need some explanations. It could be just scattering of gamma rays by intergalactic medium that does not affect the gravitational waves, but what ever it is, I suppose it matches models or this delay will be in the news.
- pjungwir 9y ago
  It reminds me of a point from non-Euclidean geometry. In Euclidean geometry a triangle has 180 degrees, but in hyperbolic or elliptic geometry it has more or less (I forget which way). So you could start measuring triangles to find out whether our universe is Euclidean or not. But since every measurement has an error interval, e.g. 180 +/- 0.000001 degrees, you may some day prove the universe is non-Euclidean, but you could never prove it is Euclidean! It may be "skewed", but just less skewed than your instruments can measure.
- CamperBob2 9y ago
  Simplest explanation I can think of: there's no reason to model this event as if all of its energy is emitted from a point source. As the NYT article put it, the size of the explosion is comparable to the orbit of Neptune. Meanwhile, two seconds at the speed of light is less than the orbit of Earth's moon. So, if the gravitational effects originate from the center of mass of the explosion, and the gamma rays originate from some kind of Big Bang-like recombination phenomenon happening a few hundred million km away from the center of the expanding shell, that would easily account for the difference.
ncallaway 9y ago
> For 130 million light years the difference between gravitational wave signal and gamma burst is 2s. That constrains the speed difference to 1.6e-18 m/s. This assumes that the production of gravitational waves and the gamma rays occurred at the same time.

alok-g 9y ago

Newbie question: While this is a very small fractional difference, could it be theoretically significant? What is the explanation of this difference? My best guess is that this difference must also be there in when the waves _started_.

SideburnsOfDoom 9y ago
I'm not a physicist, but I did do the 101 course a long time ago. I read "That constrains the speed difference to 1.6e-18 m/s" as "we measured as accurately as we could, and if they do differ, they must differ by this amount or less, which could just be the tolerances of our experiment" In other words, there's always a confidence interval, the trick is to measure it and minimise it: The speeds could be identical numbers, but what was measured was an either no or a very small difference. Larger differences have been ruled out. Identical numbers are _suggested_.
- alok-g 9y ago
  What is observed is the delay between the two signals. Fractional speed difference is calculated from it. As I understand, there is little chance that the 2 second delta isn't real.
  - SideburnsOfDoom 9y ago
    OK, but there are other possible explanations besides fundamental physics: either the order of events at the source, or the effects of the interstellar medium along the way.
fpoling 9y ago
As neutron stars are small, like 20 kilometers, while the light travels 600 000 km within 2 seconds and one can roughly assume that speed of processes that generates the gamma burst during the stars collisions matches the speed of light the delay may need some explanations. It could be just scattering of gamma rays by intergalactic medium that does not affect the gravitational waves, but what ever it is, I suppose it matches models or this delay will be in the news.
pjungwir 9y ago
It reminds me of a point from non-Euclidean geometry. In Euclidean geometry a triangle has 180 degrees, but in hyperbolic or elliptic geometry it has more or less (I forget which way). So you could start measuring triangles to find out whether our universe is Euclidean or not. But since every measurement has an error interval, e.g. 180 +/- 0.000001 degrees, you may some day prove the universe is non-Euclidean, but you could never prove it is Euclidean! It may be "skewed", but just less skewed than your instruments can measure.
CamperBob2 9y ago
Simplest explanation I can think of: there's no reason to model this event as if all of its energy is emitted from a point source. As the NYT article put it, the size of the explosion is comparable to the orbit of Neptune. Meanwhile, two seconds at the speed of light is less than the orbit of Earth's moon. So, if the gravitational effects originate from the center of mass of the explosion, and the gamma rays originate from some kind of Big Bang-like recombination phenomenon happening a few hundred million km away from the center of the expanding shell, that would easily account for the difference.

ncallaway 9y ago

> For 130 million light years the difference between gravitational wave signal and gamma burst is 2s. That constrains the speed difference to 1.6e-18 m/s. This assumes that the production of gravitational waves and the gamma rays occurred at the same time.