gravity and orbital dynamics

[toothdust]

I am really not sure what the consensus is on the speed of gravity. From what I have seen and read, it appears to me that it is instantaneous over a distance, otherwise things would not work the way they do. From my understanding of the warping of space-time, it would make sense if it were instantaneous because space-time is already warped. Is it better to think of gravity as just there, as opposed to something being given off by a mass in all directions like light?

My question is: Do orbital dynamics support an infinite "speed" for gravity to propagate. I am imagining the Earth and Moon system here. For the moon to keep its stable orbit around a moving body, would the speed not have to be instant for it to work? Lets imagine the Earth as sitting still, or not orbiting a star. In this case, a moon orbiting stably would make sense. Now add the Earth moving in a circular orbit around the Sun at a velocity. Why wouldn't the moon's orbit slowly "fall behind" over time and eventually break off and lose its orbit from Earth?

[Hornblower]

I am really not sure what the consensus is on the speed of gravity. From what I have seen and read, it appears to me that it is instantaneous over a distance, otherwise things would not work the way they do. From my understanding of the warping of space-time, it would make sense if it were instantaneous because space-time is already warped. Is it better to think of gravity as just there, as opposed to something being given off by a mass in all directions like light?

My question is: Do orbital dynamics support an infinite "speed" for gravity to propagate. I am imagining the Earth and Moon system here. For the moon to keep its stable orbit around a moving body, would the speed not have to be instant for it to work? Lets imagine the Earth as sitting still, or not orbiting a star. In this case, a moon orbiting stably would make sense. Now add the Earth moving in a circular orbit around the Sun at a velocity. Why wouldn't the moon's orbit slowly "fall behind" over time and eventually break off and lose its orbit from Earth?

If I am not mistaken, the consensus is:

1. As massive objects move, changes in the gravitational field propagate at the speed of light.

2. This non-instantaneous action is incompatible with the simple Newtonian law of gravitation. Planetary orbits would be unstable.

3. The general theory of relativity (GR) clears up this problem, but the motions are somewhat different from what Newton's formula predicts. For example, Mercury's perihelion would advance even in the absence of perturbations by the other planets.

My remarks are based on what I have read in previous discussions of this topic. If anyone out there sees errors or omissions, please speak up.

Finally, the matter of what does or does not "make sense". The cosmos is what it is and does what it does, and it does not care whether or not we think it makes sense. The challenge is for us to accept what we can observe and to develop the analytical tools we need to interpret it. If that means math that is vastly more complicated than what was adequate for the relatively crude observations in Newton's time, so be it.

[Tensor]

Publius has a good link here.

[Tim Thompson]

I am really not sure what the consensus is on the speed of gravity.

The consensus is that the speed of gravity is the speed of light. Only a few folks on the fringe think otherwise, and they all have stinky arguments in support of their own claims, and so we rightly ignore them.

My question is: Do orbital dynamics support an infinite "speed" for gravity to propagate.

Van Flandern says "yes", everybody else says "no", so that's a pretty good example of "consensus". Of course, just because something is a consensus does not make it right. But it does make it more likely to be right, especially in a complex technical discussion.

For the technical discussion see ... van Flandern, 1998, Marsh & Nissim-Sabat, 1999, van Flandern, 1999, Carlip, 2000, Will, 2003 & etc. Not all of these papers are available online, but if there is an "arXiv preprint" link, follow it and you can find a PDF pre-publication copy thereby.

[toothdust]

Using the analogy of a ball on a rubber sheet to demonstrate the curvature of space-time around a mass, its not as though the mass is sending out a curvature, the curvature is just there.

Lets say we have a mass on this rubber sheet that suddenly gains an amount of mass. Could be large or small, doesn't matter because gaining or losing any mass will affect the curvature, correct??

So would this object, a star that has just swallowed another star, a significant amount of mass, propagate a gravity wave out at the speed of light? Using the analogy, the sheet would depress uniformly, at the same time, out to the edge of its influence, correct?

There seems to be a disparity here. Is the answer only in mathematics, or has it been demonstrated or observed?

[mugaliens]

Using the analogy of a ball on a rubber sheet to demonstrate the curvature of space-time around a mass, its not as though the mass is sending out a curvature, the curvature is just there.

Using your analogy, if you were to begin, at T=-x, rolling a ball very rapidly from the edge to the center, where it arrived and ground to a halt at T=0 the curvature wouldn't "just be there" when the ball arrived. It would propagate out, and would reach half the way to the edge at T+y, and the edge at T+2y.

[toothdust]

So it would be more like dropping the ball on the sheet, and a wave spreading like a rock in water?

Does anyone know if there is any research exploring gravity waves? Do we have evidence of this?

[antoniseb]

Does anyone know if there is any research exploring gravity waves? Do we have evidence of this?

The most solid evidence for gravity waves comes from the observation of pulsars in close orbit around each other. The evidence is not detected waves, but rather loss of orbital energy from the system matching that predicted by general relativity.

Aside from this there doesn't seem to be any direct evidence concerning the speed of propagation of gravity.

[alainprice]

Is gravity probe B sensitive to the speed of propagation or only the actual 'deformation' of space itself?

[Tensor]

Does anyone know if there is any research exploring gravity waves?

Try here and herefor Earth based detectors. And here for a space based dectector.

Do we have evidence of this?

General Relativity (GR)starts with gravity and gravitational radiation moving at c. This paper shows some tests of gravity. If you drop down to section 17-3, it mention the finite propogation of gravitational interaction. If you scroll down further, to page 8, you will find figure 17.1. This show the curved line, which is the prediction of GR and the circles, which are the actual observations. GR explains the phase shifts with lost gravitational radiation(gravity waves). This is due to, partly, to the finite speed of gravity. This is an indirect detection of gravity waves, something like the detection of planets Neried mentioned in your other thread.

[Tim Thompson]

So it would be more like dropping the ball on the sheet, and a wave spreading like a rock in water?

Yes. In the plastic sheet analogy, if the mass is static, the "gravity well" that it bends in the sheet will also be static. However, if the ball moves or changes mass, the "gravity well" will change. The change or changes will propagate through the sheet at the speed of sound for the sheet, which is dependent on the physical properties of the sheet. So you can think of the speed of light as analogous to the speed of sound; the propagation of light waves through space is analogous to the propagation of elastic waves through the sheet.

Does anyone know if there is any research exploring gravity waves? Do we have evidence of this?

First, a purely pedantic point: A gravity wave is a pressure wave in an atmosphere, whereas a gravitational wave is a wave of gravity (or gravitation) propagating through space. Since I have worked in both fields of astrophysics & atmospheric physics, I am constantly trying to fix the usage of these words, but I suspect it is a forlorn hope.

There are now several custom built observatories looking for gravitational waves (i.e., LIGO or GEO600) in an effort to directly detect them, but so far the results are negative (i.e., Sintes, et al., 2008; Abbott, et al., 2008). This is not yet worrisome, since we are able now to operate only at the threshold of detection anyway. If gravitational waves remain undetected by direct means, even after we have gone well beyond the theoretical threshold for detection, then we will have something to worry about.

Meanwhile, there are examples of indirect detection of gravitational waves. Very small but very massive bodies in close binary orbit should behave in a manner that is contrary to the predictions of Newtonian gravity, but consistent with the predictions of general relativity. The reason for the difference is the loss of energy in the form of gravitational waves emitted by the orbiting bodies. This is something that can be predicted with reasonable precision from general relativistic theory, and compared with observations, assuming we can find very small but very massive things to observe. As it turns out we can do this by observing small but massive neutron stars, seen as pulsars, in close binary orbit. The binary pulsar PSR 1913+16 and the double pulsar binary PSR J0737-3039 are the only two examples I know of where such observations can be made. In both cases, observation and theory are mutually consistent, which means that observation supports the inference from theory that these systems emit gravitational radiation (see, i.e.., Hulse & Taylor, 1975; Taylor & Weisberg, 1982; Lyne, et al., 2004; Lorimer, 2005; Weisberg & Taylor, 2005; Stairs, 2005; Kramer, et al, 2006; Breton, et al., 2008).

[Tensor]

The most solid evidence for gravity waves comes from the observation of pulsars in close orbit around each other. The evidence is not detected waves, but rather loss of orbital energy from the system matching that predicted by general relativity.

Aside from this there doesn't seem to be any direct evidence concerning the speed of propagation of gravity.

I'd say any test that GR passes is direct evidence. As GR assumes c is the limit for propogation of any kind of information, including gravitational interaction. Remember, one of the two reasons Einstein wanted to generalize SR, was because he thought he could get a new theory of gravity out of it. After all, Newtonian gravitation didn't fit with SR as it assumed action at a distance at a speed greater than c.

[RegisteredUsername]

Thought experiment

Assuming a massive body moving forward at the speed of light, and you were to move alongside or slightly in front of it, you won't feel any gravity coming from the massive body, due to the fact that it is moving too 'fast' for the curvature in space to 'catch up'.

Am I correct in this?

[Hornblower]

Thought experiment

Assuming a massive body moving forward at the speed of light, and you were to move alongside or slightly in front of it, you won't feel any gravity coming from the massive body, due to the fact that it is moving too 'fast' for the curvature in space to 'catch up'.

Am I correct in this?

No. As I understand GR, no massive body will ever move relative to you at the speed of light. It will be a bit slower at best.

If you are moving along with it, everything about it and its gravitational field will appear normal, while the universe around you will appear warped beyond easy recognition.

[alainprice]

Yeah, you screwed yourself with that example. If you are moving along with it, the speed to consider is now 0.

[RegisteredUsername]

Assuming that the speed is possible, and we deal only with the curvature of space.

Yeah, you screwed yourself with that example. If you are moving along with it, the speed to consider is now 0.

Shouldn't the curvature be relative to the space instead of me?

Sorry, I am a teenager by best ;D

[undidly]

The most solid evidence for gravity waves comes from the observation of pulsars in close orbit around each other. The evidence is not detected waves, but rather loss of orbital energy from the system matching that predicted by general relativity.

Aside from this there doesn't seem to be any direct evidence concerning the speed of propagation of gravity.

An orbiting pair will lose energy with or without GR and gravitational waves.
Because gravity goes any distance ( weaker by inverse square ) all masses in the universe will be pulled more then less towards the pair at a rate which matches the orbit rate.If the orbit loss is by gravitational waves at C is compared with loss by faster than C then the only difference will be the PHASE of the impulses from the pair.
When (if ever) we can detect these pulses we will not know if they are slow (C) waves or are faster until we compare their phase with the light from the pair and watch for any change as the pair moves away (expansion of universe) or moves nearer.

The energy loss is there by Einstein or by Newton.

[Hornblower]

An orbiting pair will lose energy with or without GR and gravitational waves.
Because gravity goes any distance ( weaker by inverse square ) all masses in the universe will be pulled more then less towards the pair at a rate which matches the orbit rate.If the orbit loss is by gravitational waves at C is compared with loss by faster than C then the only difference will be the PHASE of the impulses from the pair.
When (if ever) we can detect these pulses we will not know if they are slow (C) waves or are faster until we compare their phase with the light from the pair and watch for any change as the pair moves away (expansion of universe) or moves nearer.

The energy loss is there by Einstein or by Newton.

Can you show us, mathematically, why you think Newton's theory predicts an energy loss?

[Hornblower]

Assuming that the speed is possible, and we deal only with the curvature of space.



Shouldn't the curvature be relative to the space instead of me?

Sorry, I am a teenager by best ;D

To assume the possibility of a massive object moving at the speed of light, we would need to rewrite the whole theory. Heaven only knows what it will look like by the time we get it into good agreement with observations. My guess is that Ockham's Razor will give it a close shave.

[RegisteredUsername]

To assume the possibility of a massive object moving at the speed of light, we would need to rewrite the whole theory.

Understood, thanks ;D

Side note though, are gravitational waves affected by the medium they're currently travelling in?

[Nereid]

An orbiting pair will lose energy with or without GR and gravitational waves.
Because gravity goes any distance ( weaker by inverse square ) all masses in the universe will be pulled more then less towards the pair at a rate which matches the orbit rate.If the orbit loss is by gravitational waves at C is compared with loss by faster than C then the only difference will be the PHASE of the impulses from the pair.
When (if ever) we can detect these pulses we will not know if they are slow (C) waves or are faster until we compare their phase with the light from the pair and watch for any change as the pair moves away (expansion of universe) or moves nearer.

The energy loss is there by Einstein or by Newton.

That may be so, or maybe not (I'm looking forward to your answer to Hornblower's question!).

You left out one critical part: the observed orbital decay of the binary pulsars matches predictions from GR to within the observational uncertainties.

How well do the observations match quantitative predictions from Newton? Once we have the math (hopefully part of your answer to H's question), we can - objectively - plug in the numbers, turn the handle, and see for ourselves exactly how well it matches.

[undidly]

That may be so, or maybe not (I'm looking forward to your answer to Hornblower's question!).

You left out one critical part: the observed orbital decay of the binary pulsars matches predictions from GR to within the observational uncertainties.

How well do the observations match quantitative predictions from Newton? Once we have the math (hopefully part of your answer to H's question), we can - objectively - plug in the numbers, turn the handle, and see for ourselves exactly how well it matches.

"observational uncertainties.". Yes ,that is what concerns me.How big are they.

Plug in the numbers?.Would love to.What are the numbers?.I can't find the numbers myself without million dollar equipment.
Turn the handle (switch on the computer).Love to.Need the numbers and also local conditions to take into account.
If our sun was a binary it would cause tides on earth to have a complicated tidal high pattern.Would not this tidal effect draw energy from the binary
system to appear as heat in Earth oceans and sap the binary energy?.
This is NEWTON stuff.
Don't know these numbers either.I could make up some numbers.

[Tensor]

"observational uncertainties.". Yes ,that is what concerns me.How big are they.

In the paper link to below, In table 1, the original measured parameters are given. The period uncertainty is on the order of 5 x 10^-12.

From the paper below:

"In most cases(particularly in the latter data), the measurement uncertainties are smaller than the line widths.

Plug in the numbers?.Would love to.What are the numbers?.

Here they are.

I can't find the numbers myself without million dollar equipment.

Well, I found them with my 450 dollar laptop and 35 dollar a month cable connection. That's a far cry from a million dollars worth of equipment. Or perhaps you were just using that as an excuse?

So, now you have the numbers. When can we expect your analysis?

[Hornblower]

"observational uncertainties.". Yes ,that is what concerns me.How big are they.

Plug in the numbers?.Would love to.What are the numbers?.I can't find the numbers myself without million dollar equipment.
Turn the handle (switch on the computer).Love to.Need the numbers and also local conditions to take into account.
If our sun was a binary it would cause tides on earth to have a complicated tidal high pattern.Would not this tidal effect draw energy from the binary
system to appear as heat in Earth oceans and sap the binary energy?.
This is NEWTON stuff.
Don't know these numbers either.I could make up some numbers.

Of course the orbit will change over time in the case of tidally induced orbit/spin interaction. My intent was to discuss the idealized case of an isolated binary with no such interaction, which I thought would be clear from the context of this thread. My apology if it was not clear.

1. Is it your opinion that Newton's theory predicts orbital decay for such an idealized system with no orbit/spin interaction?

2. If your answer is yes, can you show us some mathematical justification for that opinion?

[undidly]

In the paper link to below, In table 1, the original measured parameters are given. The period uncertainty is on the order of 5 x 10^-12.

From the paper below:

"In most cases(particularly in the latter data), the measurement uncertainties are smaller than the line widths.



Here they are.
So, now you have the numbers. When can we expect your analysis?

Thanks for the numbers.Saved for study.
Didn't see a mention of orbital decay rate without losses due to gravitational waves.
Even if both members of the binary pair were not spinning relative to each other (both always see same face of other) they would still loose energy by their G influence on other mass.(causing movement and tides)
At least one is spinning but in the same direction as the orbit or not?.
Do pulsars experience tides caused by a nearby mass (its partner),are there losses,are any tidal movements frictionless?.
The Earth is loosing spin energy to the moon.Earth spin is slowing down,moon is moving to higher orbit.

[undidly]

Of course the orbit will change over time in the case of tidally induced orbit/spin interaction. My intent was to discuss the idealized case of an isolated binary with no such interaction, which I thought would be clear from the context of this thread. My apology if it was not clear.

1. Is it your opinion that Newton's theory predicts orbital decay for such an idealized system with no orbit/spin interaction?

2. If your answer is yes, can you show us some mathematical justification for that opinion?

In an idealized system?.NO.

[Nereid]

Thanks for the numbers.Saved for study.
Didn't see a mention of orbital decay rate without losses due to gravitational waves.
Even if both members of the binary pair were not spinning relative to each other (both always see same face of other) they would still loose energy by their G influence on other mass.(causing movement and tides)
At least one is spinning but in the same direction as the orbit or not?.
Do pulsars experience tides caused by a nearby mass (its partner),are there losses,are any tidal movements frictionless?.
The Earth is loosing spin energy to the moon.Earth spin is slowing down,moon is moving to higher orbit.

Just out of curiosity, from the info you now have, what is the distance between the two objects (that have been reported to be in a death spiral, due to loss of energy from gravitational waves)? What are their masses?

In the case of the pulsar system, assuming tides, would you expect the orbits to be expanding or contracting (you should have the spins of the objects in your files now)? A simple analogy with the Earth-Moon system might be a good place to start.

Have you read anything about the stiffness of neutron stars, especially compared with that of rocky planets and moons (like the Earth and Moon)?

[John Mendenhall]

If I am not mistaken, the consensus is:

1. As massive objects move, changes in the gravitational field propagate at the speed of light.

2. This non-instantaneous action is incompatible with the simple Newtonian law of gravitation. Planetary orbits would be unstable.

3. The general theory of relativity (GR) clears up this problem, but the motions are somewhat different from what Newton's formula predicts. For example, Mercury's perihelion would advance even in the absence of perturbations by the other planets.

My remarks are based on what I have read in previous discussions of this topic. If anyone out there sees errors or omissions, please speak up.

Finally, the matter of what does or does not "make sense". The cosmos is what it is and does what it does, and it does not care whether or not we think it makes sense. The challenge is for us to accept what we can observe and to develop the analytical tools we need to interpret it. If that means math that is vastly more complicated than what was adequate for the relatively crude observations in Newton's time, so be it.

My apologies to Hornblower for quoting his entire post, but read it again, folks. He is exactly on the mainstream. Once again, I refer you to the article on 'Speed of Gravity' in Wiki. And thanks to HB especially for the last paragraph.

Regards, John M.

[undidly]

Just out of curiosity, from the info you now have, what is the distance between the two objects (that have been reported to be in a death spiral, due to loss of energy from gravitational waves)? What are their masses?

In the case of the pulsar system, assuming tides, would you expect the orbits to be expanding or contracting (you should have the spins of the objects in your files now)? A simple analogy with the Earth-Moon system might be a good place to start.

Have you read anything about the stiffness of neutron stars, especially compared with that of rocky planets and moons (like the Earth and Moon)?

^What are their masses?^
Don't know.Will read it again but it makes no difference.Bigger mass = more radiation.

^A simple analogy with the Earth-Moon system might be a good place to start.^
Did that,#25.

^Have you read anything about the stiffness of neutron stars.^
No but I will.It must be very theoretical even to the point of being fantasy.