Does light speed vary in different gravitational situations?

[narrator]

I understand that the speed of light is always constant for the observer, but I keep coming across references that suggest it varies in different situations.

What situations, and what frame of reference?

I saw an interesting documentary many years ago. They talked about light moving at different speeds depending on gravitational forces. To test this, they had a plane flying at high altitude. From the ground, they shot laser pulses at it and timed the reflected response, comparing it to a laser shot across land. The experiment worked and there was cheering in the streets ;)

 

[zhermes]

The speed of light is a constant for all observers, for all of everything, always.

It sounds like the experiment was observing http://en.wikipedia.org/wiki/Time_dilation"

 

[IsometricPion]

The speed of light is a constant for all observers, for all of everything, always.

He may have been referring to the speed of light in a http://en.wikipedia.org/wiki/Refractive_index" as demonstrating a varying speed of light. However, all such effects in GR are global measures of the speed of light, which is not required to be constant. Rather it is local measures (since locally GR reduces to SR) of the speed of light that must (and do) always yield c.

They talked about light moving at different speeds depending on gravitational forces. To test this, they had a plane flying at high altitude. From the ground, they shot laser pulses at it and timed the reflected response, comparing it to a laser shot across land. The experiment worked and there was cheering in the streets ;)

Without more information I can't say for sure but, the Shapiro delay could be measured that way if one had a very accurate clock and the changing index of refraction of air (due to air currents) didn't cause too much trouble (the effect would only be of order 1 nanosecond based on the formula on the wikipedia page).

 

[zhermes]

He may have been referring to the speed of light in a http://en.wikipedia.org/wiki/Refractive_index" .

Absolutely, my apologies---I should have been more precise---'speed of light in a vacuum always everywhere constant yadda yadda'

Also, one could interpret results such as the http://en.wikipedia.org/wiki/Shapiro_delay" as demonstrating a varying speed of light.

I disagree with you semantically/philosophical... but I think that's fairly irrelevant.

the Shapiro delay could be measured that way if one had a very accurate clock and the changing index of refraction of air (due to air currents) didn't cause too much trouble

The Shapiro effect is measurable from the earth---I think its actually been measured with interferometers on (effectively) just the surface of the Earth as-well (e.g. 100 meter scale).

 

[bcrowell]

The way I would put it is that there is no well-defined notion of velocity in GR except for local velocity. So in an experiment like Shapiro's, I would say that it's not even valid to assign a velocity to the ray.