“There is a single light of science, and to brighten it anywhere is to brighten it everywhere.” -Isaac Asimov
One of the most spectacular and successful ideas of the 20th Century was Einstein’s General Relativity, or the idea that matter and energy determines the curvature of spacetime, and the curvature of spacetime in turn determines how gravitation works.
From the orbits of planets to the bending of starlight, General Relativity governs all gravitational phenomena in the Universe, and accurately describes every observation we’ve ever made.
And the more concentrated energy is — in any form — in a given region of spacetime, the more severe the curvature that results. This means that if we can probe systems that are more extreme than we find in our own Solar System, we can potentially observe some aspects of General Relativity that are only theoretical predictions at this time.
While our Sun curves spacetime significantly enough to do things like bend starlight and cause a time-delay in communication signals, there are examples in the Universe that allow us to have a lot more fun with General Relativity.
In particular, white dwarfs — like Sirius B, illustrated above — condense a Sun’s worth of mass into a volume of only the Earth, creating spacetime that’s curved 100 times more severely than at our Sun’s photosphere. One of the most extreme things that happens in our Universe is that we find multiple white dwarfs orbiting one another, creating a binary white dwarf system. (Binaries are very common; although our Sun isn’t a binary system, our closest neighboring star is!)
According to General Relativity, these binary orbits are inherently unstable, and will eventually spiral in towards one another, coalescing, and eventually triggering a Type Ia supernova!
But objects that fall deeper into a gravitational potential well must somehow lose all that gravitational potential energy; where does it all go?
According to General Relativity, it gets radiated away, but not in the conventional form of radiation. Not in the form of known particles like photons and neutrinos; instead, this is a special type of radiation: gravitational radiation or gravitational waves.
Rippling through spacetime at the speed of light, these gravitational waves should be emitted each time a pair of masses orbits one another or each time a rotating object changes its shape. White dwarfs provide an increased curvature to spacetime compared to our Sun, but the most severely curved spacetime comes from neutron stars (which are a further factor of 1000 smaller than white dwarfs) and, in the most extreme case, black holes.
All that’s missing to complete the picture is a direct detection of this gravitational radiation. It’s challenging, because a gravitational wave, when it passes through an object, only distorts its shape in this weird way, as the animation below shows.
It appears to make a normal-shaped object first squished and fat, then normal again, then thin and stretched, then normal again, etc. Because all of spacetime is undergoing this, the object itself may have extraordinary difficulty noticing it, but there’s no problem with seeing it for an outside observer.
On the other hand, this is happening in three-dimensional space, not in an idealized 2-D case, and they’re coming from a source that is (probably) elliptical, rather than circular, in nature. So we have to take those things into account when we’re looking for gravitational waves in reality.
But the biggest problem with detecting them is that these gravitational waves are maddeningly weak. The amount of stretching/compressing that occurs in one direction relative to another is so small that, for even the Nobel-prizewinning binary pulsar, the number would be just 10-26, meaning that a 1-meter object would stretch/compress by 10-26 meters as a gravitational wave passed through.
So we fight this from two different angles: we look for sources that are even more powerful than the inspiraling neutron stars we talked about earlier, and we try to build detectors that are sensitive to the tiniest change in distance possible. The detector setup is known as an interferometer.
Shine a monochromatic laser through a beam splitter, and send the two light beams off at 90 degrees to one another. This ensures that if a gravitational wave passes through, one “arm” will contract while the other expands, and vice versa. If the shift is sufficiently large, the interference pattern that shows up on the screen will shift in a predictable fashion, as dictated by General Relativity!
In practice, we have a few stations around the world that use this concept with two additions:
- The entire apparatus is incredibly insulated from vibrations, motions, temperature variations, etc., here on Earth, and
- They bounce the light back-and-forth along mirrors many times to artificially increase the path length.
This is what the LIGO collaboration/experiment, or Laser Interferometer Gravitational-Wave Observatory, is all about.
The difficulty is that it’s only sensitive to signals stronger than about 10-20, which is a very big number in this field, and it’s only sensitive to very fast frequencies, or gravitational wave signals that repeat hundreds of times per second. The only astrophysical sources even capable in theory of being detected by LIGO are neutron star-neutron star mergers (not inspirals), neutron star-black hole mergers, black hole-black hole mergers, or a very fortuitous Type II (core-collapse) supernova.
Unfortunately, this does not include the most common black hole mergers in the Universe: the ones occurring at the galactic center. They’re so supermassive (and hence, their event horizons are so large) that they simply cannot be detected by the frequencies LIGO is sensitive to; the wavelength of the gravitational radiation will be too long for LIGO’s “arms”.
What makes this even more challenging is how close these (very rare) objects need to be to us in order to even have a shot at their gravitational waves. You see, the further away you are, the smaller the signal becomes. As always, we could get lucky.
In 1987, the light from a Type II supernova that went off in the Large Magellanic Cloud reached Earth for the first time. It wasn’t quite within our Milky Way, but it was close: just 168,000 light years (about 51 kpc) away. If LIGO was fully operational at the time, depending on the correct model for gravitational waves from Type II supernovae, it could have been detected.
You have to get your amplitude up, at the proper frequency, into the detectable range. And it’s difficult to know whether that’s going to happen or not. After all, there are uncertainties here, and difficulties in modeling how something as complex as a supernova occurs!
Although supernovae are copious throughout the Universe, finding one within the 10 kpc (about 30,000 light years) normally used in simulations is an extremely rare occurrence, having happened maybe 9 times in recorded history, and only 2 of which were type II.
So rather than wait for the Universe to come to us, we can do something about it, and expand our sensitivity, which is exactly what LIGO’s working on doing right now.
Expanding our search by upgrading to advanced LIGO should allow us to detect objects that are more than an order-of-magnitude farther away, while close by, we should be able to detect lower-amplitude objects.
One illustration of how the search expands from LIGO-in-its-current-form to the anticipated (2014) advanced LIGO is shown below, courtesy of David Shoemaker.
This may sound promising, but remember that our galaxy has had only 2 type II supernovae in the past 2000+ years, while the rate of binary mergers of collapsed objects is unknown but possibly far lower. Expanding our search range by a factor of between 10-to-maybe-50 (optimistically) could, conceivably, give us a few detections per year, but we could also see zero, and that wouldn’t be surprising either. (Right, Clara, Sean, Nick?)
The fact of the matter is that LIGO isn’t the ideal tool for the job of finding gravitational waves: LISA is, but we didn’t fund it. Famous astrophysicist Kip Thorne may be predicting that LIGO will find gravitational waves by 2017, and this may happen, but it’s unnecessarily optimistic. Advanced LIGO should come online in 2014, and assuming it reaches design sensitivity immediately (it took the first LIGO many years to get there), the event rates are on the high end of estimates, and the optimistic gravitational wave models are accurate, we’ll definitely see multiple sources by 2017. Also, Kip Thorne will probably win a Nobel Prize; he’ll be 77 in 2017.
But if even one of those assumptions is flawed, we have every reason to believe that gravitational waves still exist, and that Advanced LIGO won’t see them. It’s a spectacular chance for gravitational waves, but if we don’t see them, it doesn’t mean that relativity is wrong; it means we need to build a better tool for the job. I’m hoping this works as ideally planned, that it’s working perfectly in 2014, and that we enter a new era of gravitational-wave astronomy. But if this isn’t our entry, I’m not going to be surprised either, and neither should you.
The science is amazing enough without the sensationalism.