You are familiar with waves created by a piston moving through water:
You can perceive electromagnetic waves with your eyes. One way to create them is to wiggle a charged electric particle: the frequency of its oscillation will be the frequency of the light wave it emits.
Java applet from the excellent NTNUJAVA Virtual Physics Laboratory site. Thanks!
It turns out that two massive objects moving rapidly around each other will create gravitational waves, ripples in the very fabric of space itself.
Movie of black hole binary inspiral courtesy of the SXS project.
Gravitational waves have some very interesting properties. For example, when they pass through a region, they "squeeze" the space in that region in alternating directions: first it expands horizontally while contracting vertically, then contracts horizontally while expanding vertically.
If we put a painting on that wall through which the wave passes, we might see it distort sort of like this:
There's just one problem: even the most massive, rapidly moving objects in the sky -- binary black holes orbiting around each other every few minutes or hours -- produce gravitational waves with small amplitudes. Really small amplitudes. Really, really, REALLY SMALL AMPLITUDES.
How small? Physicists can describe the amplitude of a gravitational wave with a fraction: how much longer (or shorter) does an object become when the wave passes through it? For example,
Example: Joe waves his magic wand and creates a gravitational wave with an amplitude of A = 10 percent = 0.1. The wave moves towards his little brother Billy, who is H = 1.2 meters tall under ordinary circumstances. How tall will Billy be when the wave is passing through him?
A gravitational wave of that amplitude would be easy to detect.
Unfortunately, the real gravitational waves generated by astrophysical objects aren't quite that strong. For example, the amplitude of the wave created by a pair of neutron stars if they were to spiral into each other at a distance of about 20 Mpc (that's roughly the distance of the nearest cluster of galaxies) is about
-20 A = 10 Q: Joe holds a meter stick in front of him and measures it very carefully. If a gravitational wave with this amplitude were to pass through the meter stick, how much longer would it grow? Just how easy would it be to detect this sort of change in length?
The answer is: really, really hard. For reference, here are some sizes of objects you might try to use in the measurement procedure.
Golly. Is there ANY way to detect a gravitational wave?
(Note: much of the following material is shamelessly taken from Alan Weinstein's 2006 lectures on gravitational waves and LIGO, which can be found at the Undergraduate Resources section of the the main LIGO web site.
The Michelson interferometer is a device which splits a single beam of light into two beams. It sends each beam down a long tunnel and back using mirrors, the two beams travelling in perpendicular directions. Then the beams are brought back together so that they can interfere with each other.
LIGO -- the Laser Interferometry Gravitational wave Observatory -- has two of these interferometers. One is in Hanford, Washington, and other in Livingston, Lousiana.
At each site, there are tunnels about 4 km long running in perpendicular directions.
A gravitational wave with an amplitude of -20 A = 10 passes through the Earth near the Hanford LIGO site. It happens to run vertically down into the Earth's surface, so that its distortion first makes one leg longer and then other other leg longer. Q: How large is the difference in distance travelled by light going along the North leg, compared to the light going along the East leg? Just how easy would it be to detect this sort of change in length?
This is a REALLY tough engineering job.
There were some problems ...
... but things are going very well now. LIGO has been gathering data for several years. It's not the only gravitational wave observatory, either: scientists in other countries have built their own.
After LIGO performed a series of five science runs, scientists decided to improve the sensitivity of the interferometer by making a number of improvements to the equipment.
The schedule shows that work should be finished at the end of this year:
Alas, the answer so far is ... NO.
How far away COULD we detect merging binary systems?
Hmmm. That's quite a distance. The problem is that merging binary systems are not very common. They are so rare, in fact, that the rate at which they are predicted to merge within the range of LIGO is, well, something like one every year. Or maybe one every ten years. So, the fact that we haven't detected one YET really doesn't tell us very much.