Thursday, January 14, 2016

Have Gravitational Waves Been Detected?

Since Monday, when Lawrence Krauss tweeted and confirmed a rumor, there is a lot of traffic repeating and questioning the rumor that LIGO has detected gravitational waves.

It's worth emphasizing that there is little-to-no doubt that gravitational waves exist. That evidence is very strong, though indirect.

As I wrote about for Physics World a few months ago, after visiting LIGO's Hanford facility in Washington state, Einstein's equations of general relativity predict that gravitational waves should be emitted by any object (mass, or pure energy) in motion. But these waves aren't sine waves, like the canonical image of electromagnetic waves we all have, self-propagating through space.

That's because there is only one type of mass -- positive -- whereas there there are two types of electric charge -- positive and negative. So unlike for electric charge, there is no "dipole moment" for mass.

Instead, a "quadrupole monent" is the simpliest gravitational wave.

I'm not going to write here exactly what this means technically. But it does mean that gravitational waves are unlike other waves that we're used to. As they move through space, they squeeze it in one dimension, but expand it in the other.

The analogy I used was that of puckering up, and puckering down, for a kiss -- your top and bottom lips go up and down, while the side of your mouth comes in and out.

The indirect evidence for gravitationl waves was made by Joseph Taylor Jr and Joel Weisberg, utilizing a new type of pulsar discovered by Taylor Jr. and Russell Hulse in 1974. They won the Nobel Prize for the discovery of that pulsar. (See the "Pulsar Detectives" box on the second page of my article).

As you'd expect, a pair of stars orbiting around one another would get closer and closer, and the energy they lose is released as gravitational waves. This can be calculated from Einstein's equations -- it's not as easy calculation, by any means, but it is straightforward -- and that prediction is in spectacular agreement with what was observed by Hulse and Taylor's stellar pair. (See graph to the right.)

It's worth nothing that that, while Hulse and Taylor discovered this special pulsar as part of a pair, it was Taylor and Weisberg who showed that its decaying orbit was in agreement with Einstein's prediction. People often get this wrong. Perhaps Weisberg should have gotten the Nobel Prize too.

But while we know gravitational waves exist, detecting them directly, as LIGO is trying to do, would still be a big deal. Gravitational scientists hope that detection of gravitational waves might soon become routine, just as detecting stars via visible light and radio waves and x-rays is routine today.

If so, this would provide astronomers with a new "sense," a new way of not unlike how hearing is a different sense to seeing.

Those new senses brought about a truly new view of the universe -- from the placid, steady state found by Hubble et al, to the wild and racuous view of stars and galaxies we know today.

Direct detection of gravitational waves could inspire the the same type of shift.

If this detection is true -- and Lawrence Krauss backed off a bit on Tuesday -- it suggests that detecting gravitional waves directly could be easy with Advanced LIGO, their latest and last-planned upgrade, since they just started running their machine in full detection mode this past autumn.

This massive tomb is its own black hole.
A direct detection would defintely be a most important result here -- easily worthy of a Nobel Prize for trhe project leaders and also especially to Kip Thorne (now retired from Caltech, and an advisor on the movie Interstellar), who has been pushing for the LIGO machine since Day One and was instrumental in not just getting the project going way back in the 1980s, but who has fundamental to its design since then.

As a theorist, Thorne has had a powerful influence on the progress of general relativity since the 1970s. All real students of science, at least my generation, drag around his 5.6 lb textbook published with Charles Misner and John Wheeler in 1973. (I used to say its cover was black because no light would possibly escape its huge mass.)

Thorne should definitely win the Nobel Prize, especially if LIGO has really detected gravitational waves. I'm not sure who else would get it, on the experimental side. On the theoretical side, I think Stephen Hawking and Roger Penrose deserve it too, but it's unlikely the Nobel committee would award it to three theorists and at least one LIGO experimentalist.

No comments: