I just read this article:
http://www.cnn.com/2003/TECH/space/01/08/gravity.speed.ap/index.html
So…um… how exactly does this work?
Which part don’t you get?
It has been assumed for a long time that gravitational effects travel at the speed of light. The alternative was that it was an instantaneous effect. Think what would happen if you could magically make our sun disappear. Would the earth immediately stop its circular orbit and hightail into deep space on a straight line tangential to its old orbit? Or, if gravity must actually travel, would it take 8 minutes or so before the sun’s disappearance was felt? Most had assumed it was the second possibility but it seems this experiment has confirmed it (to some extent at least).
In the Washington Post article I was more befuddled by the statement that gravity travels at 1.06 the speed of light (with a 20% margin for error).
What’s up with that?
Measurement error is a consequence of any experiment. Gravity propogation is notoriously difficult to measure at all, and the fact that a way was found that is within our technology is pretty cool.
So 20% of 1.06 is 0.212. This means that gravity travels between 1.06-0.212 and 1.06+0.212 of the speed of light, but we’re not entirely sure what its actual value is.
While this doesn’t prove that gravity travels at the speed of light, it gives extremely large amounts of evidence that gravity doesn’t travel instantaneously, and since the speed of light falls in the range of uncertainty, its reasonable to assume that it does travel at the speed of light.
Future experiments will narrow the gap, naturally. Maybe it doesn’t travel at the speed of light, but we now have more convincing evidence. It is almost certainly, as I said, not instantaneous.
Well, I certainly understand margin for error.
Let me be more clear:
If we believe those error bars there’s a 50%+ chance that gravity propagates faster than light.
That’s what I was getting at. How does that reconcile?
Whack-a-mole -
You explained it better than the article, frankly. Reading the article, I didn’t see the difference between the experiment they talk about and the one they did decades ago, where they looked up during a solar eclipse and found that the stars right near the blacked-out sun appeared to be not exactly where they should be.
I get what your post is saying, though. Now, please come back and explain how very precisely looking at Jupiter’s pull on the light coming from a quasar shows that the gravity took some positive length of time to (what? emanate?) from Jupiter’s center of gravity.
Oh - that’s another thing - from what point does the gravity start? Some of the gravity came from Jupiter’s side nearest the light, some from the side furthest, most from somewheres in the middle.
I’m reasonably certain that in cases like this gravity can be considered as a point source. That is, the gravity Jupiter exerts on something travelling nearby may be thought as emanating from the planet’s center. If you do the math and add up the greater gravity of the near side cancelled by the gravity of the far side it all cancels out to act as if it all comes from one point.
As to how the experiment specifically worked I have no clue. Watching light bending around gravity wells has been done for awhile now. How it can show the propogation of gravity I have no idea. I will throw the following out as food for thought. I could be light years off base here though so take it with a Jupiter sized hunk of salt.
Idea 1: That they chose a Quasar to view seems telling. IIRC Quasars are some of the most distant things in the Universe we can view. The light coming from them has been travelling for so long it started before Jupiter existed. Perhaps by measuring the redshift researchers can speculate on when the light from the Quasar was under the effect of Jupiter’s gravity.
Problems: The article specifically mentioned the bending of the light on not redshift leading me to believe this is not the case.
The effect of Jupiter’s gravity is likely so tiny far out that I doubt it can be measured. Not until something is relatively near the planet would I think the effect of Jupiter’s gravity to be sufficient to have a noticeable effect on anything.
I dunno…that they chose a Qusar to view seems to have something to do with the answer. Maybe Bad Astronomer will be along to enlighten us.
Sorry about that, Jonathan, didn’t mean any sort of slight. 
You’re right, there’s a chance that gravity is faster than light. You can’t eliminate that possibility based on the measurement made. That is, you can’t reject the possibility out of hand, unless you are assuming or have other evidence that gravity is exactly at light speed. Add up all the relevant sources of error in the measurements, and we have that 20% margin of error.
All that we know, based on the measurement they made, is the range that the propogation of gravity could be in. Strictly speaking.
Based on this, the big thing that we ‘know’ is that its not instantaneous. There is, however, a very good chance that gravity does propogate at the speed of light.
Do we know for sure? No. But there will always be uncertainty. There’s uncertainty on the speed of light itself, even though its been measured to goofy levels of precision.
I’m not sure where the concern is.
Seems like they used the same trick that Einstein used to show the bending of light via gravity in the first place – see the change in apparent position relative to expected position of a distant light source.
I’m guessing they used highly accurate orbital information about the Earth and Jupiter to form a precise idea of where Jupiter was, and see that as Jupiter approached the Earth/quasar line that the bending effect of gravity on the light lagged behind Jupiter’s approach. That is, that if the change in the position of the gravity well is always exactly centered on Jupiter as Jupiter moved, the bending would be some amount at a certain time, but that if the well took time to propogate then it would not be on-center with Jupiter but would ‘lag behind’ a bit.
Serious WAG. 
Don’t count on FTL gravity. Einstein’s theories predict light speed gravity and so far his theories have held up very well. The uncertainty is in the measurement being conducted to confirm the theory. Of course, it should still be checked. Yay research!
Gravity is the behavior of space in response to mass. The force of gravity essentially extends to infinity (although very very weakly). If mass changes (like two black holes combining into one), then the information about that change takes time to travel across the universe. If gravity was instantaneous, then we would be able to detect events (like merging black holes) before we see them occurring. But that does not seem to be the case. (But imagine being able to detect something billions of light years away instantaneously!!)
The LIGO project (info available on the 'net if you’re curious) is ongoing to try & detect gravity waves from events like black hole/neutron star collisions, etc.
Being so weak, gravity differences are notoriously hard to detect (readily swamped out by the fat ol’ Earth’s gravity in this neighborhood).
William_Ashbless - Yeah, that makes a lot of sense. The quasar appeared to ‘move’ at a certain time - a very small amount of time after* Jupiter was close enough to exert the gravity to cause that much displacement.
But Jupiter moves in front of stars all the time - I think Whack-a-Mole is right that the quasarity of the light was somehow important.
:smack:
There are doubters. See this article.
It’s an interesting result. But like all new observations, it needs some confirmations before it will be accepted.
I’m guessing something simple: They wanted a source that wasn’t moving across the sky (just away from us). Local stars, and even local galaxies probably have motions relative to Earth that are less than ideal because the line between here and there would change. A quasar, being so far away, has the advantage of moving almost exclusively away from us (everything, far enough away, does), and any sideways motion it does have causes virtually zero deviation in the line to it at a Jupiter’s distance.
Yeah, a quasar has the lowest possible paralax, so it makes a good choice when you have to define an absolute, non-moving point to make hyper-sensative measurements of the motion of Jupiter and the effects of its gravity.
PS- I’d be very wary of any article that refers to a quasar as a “star” no less than three times. Also, I fail to see what kind of “planetary alignment” happens only once in a decade which is relative to Jupiter occulting a quasar. Seeing as there are more than 12,000 known quasars, Jupiter ought to occult one fairly often.
How would this affect string theory? It looks like this experiment simply agrees with GR. Why would that affect other cosmolgies?
Space is big. Really big. Given how small Jupiter appears at its distance from the Earth, there could be a million known quasars with Jupiter still failing to intersect the Earth/quasar line.
Probably…if you were standing next to Jupiter I bet you could position yourself anytime to get a Qusar occluded. I think the issue for them is to have Jupiter occlude a Quasar and have the Earth along the same line of sight (hence the planetary alignment bit…Earth-Jupiter-Quasar alignment). That is probably what happens only once every ten years or so.
Perhaps. Still seems that it ought to happen considerably more often than that. I’ll defer to the opinion of a resident astronomer, though.