LIGO

by Wm. Robert Johnston
last updated 3 October 2001

LIGO, or Laser Interferometry Gravitational Wave Observatory, is part of a $360 million NSF project to detect gravitational waves (GWs). It includes two facilities. The LIGO-Livingston Observatory is about 7 kilometers north of Livingston, Louisiana (between Baton Rouge and Hammond). The LIGO-Hanford Observatory is at the DOE Hanford Reservation in Washington state.

LIGO will begin attempting to detect GWs next year. The LIGO observatory consists of two tubes, each 4 kilometers long, connected in the shape of an "L". A laser beam generated at the corner of the "L" is split, sent down each arm, and trapped in each arm between two mirrors. The two separate beams are then compared. A GW will produce an unequal change in length of the two arms, which should (in principle) be revealed by that comparison.

In practice, the challenge is that the expected change in length is very small--over the 4-kilometer arm length, the length change is perhaps one-one thousandth of the diameter of the nucleus of an atom. The facility is at the frontiers of technology to try to isolate such a small signal. The beam tubes are pumped down to one of the best vacuums on Earth; the mirrors have a complex suspension system to eliminate stray vibrations. The twin LIGO sites (Louisiana and Washington) will allow cross checking of signals. Thus, when both sites show the same jiggle you can be more confident that it was a GW--as opposed to the vibration from a semi-truck hitting the bumps on I-12, 11 kilometers from LIGO-Livingston, or from a stray air molecule bumping one of the mirrors. They also use some complex computer programs to separate a real GW signal from all this stray noise.

This is a sketch of the configuration of each LIGO site (you can also see the diagram at http://www.ligo-la.caltech.edu/Posters/poster13.html).

1--laser
2--mirrors/optics to shape laser beam
3--beam splitter mirror
4--recycling mirror (one for each arm)
5--end station mirror (one for each arm)
6--light detector

Path of laser light: 1 to 2 to 3 to 4 (both) to 5 (both) to 4 (both) to 3 to 6

Parts 1-4 and 6 are all in the corner station building in a main room called the LVEA (for Laser Vacuum Equipment Area). The two paths from 4 to 5 are the 4-kilometer arms. The laser beam is actually trapped between mirrors 4 and 5 in each arm an average of 50 back-and-forth trips before getting back to 3 to 6. Note that these are the main mirrors: if you count all the lesser mirrors and lenses, they number in the dozens.

The LVEA is a large room (like a warehouse). In it the tunnel and associated assemblies are laid out on the ground in a "+" shape. The laser is at the end of the "east" branch. The laser used is infrared, meaning the laser light is invisible to the human eye--but direct or indirect exposure can still carry enough energy to damage the eye. The laser travels in the vacuum tube to the center of the "+" shape, the location of the beam-splitter mirror. This mirror splits the beam into two beams, one down the X-arm ("west" branch of the "+" shape) and one down the Y-arm ("south" branch of the "+" shape). Between the beam-splitter mirror and each arms' end station mirror, the laser beam is trapped, so to speak. A fraction of the light from each returning beam is directed from the center of the "+" shape to the "north" branch, where the recombined laser beams mutually interfere. This is what provides information on changes in the relative lengths of the arms, allowing (someday soon!) detection of gravitational waves (GWs). (The four branches form right angles, but not exactly N-S-E-W.)

This picture shows part of the interior of the LVEA. The pink curtains are "clean areas" protecting chambers while the interiors are accessed to adjust optics and other components. The beam splitter mirror is inside the assembly between the two pink curtained areas. The X-arm extends to the left and the Y-arm toward the foreground. The laser room is located on the far side of the beam splitter mirror.

The vacuum tubes of the two arms each extend 4 km to an end station building containing a chamber for the end mirror. The arms' tubes are 1.2 meters in diameter on the outside. They are made of steel using specialized manufacturing techniques to minimize the release of residual gases. A concrete shell covers the arms for protection; this is what is visible in the first picture above.

The LVEA contains much larger chambers for equipment such as described above. The vacuum tubes in the LVEA are actually a series of large roughly spherical chambers perhaps 3 meters across. Each one can be sealed off independently, allowing maintenance of a component in a single chamber without repressurized and subsequently depressurizing the whole 8 kilometers of tunnel. This saves time: removing the air from the whole tunnel takes a few days. LIGO's vacuum assembly, with a total volume equal to a cube about 20 meters on a side, will operate at one-one trillionth of normal atmospheric pressure. This is among the largest vacuum chambers in the world. The two LIGO facilities are probably the largest ultrahigh vacuum facilities in the world. The quality of the vacuum is a few hundred times better than space at the altitude of the International Space Station. Incidentally, the largest vacuum chamber in the world is NASA's Space Power Facility in Ohio, with a volume about three times greater than either LIGO site.

The data analysis software is essential to the success of LIGO. Any length changes from passing gravitational waves will have to be separated from a myriad of vibration sources and other sources of noise. Different programs are being developed to look for different GW signals. One general method is to look for a match between LIGO data and the theoretical predictions from different expected sources of GWs. Another is to look for statistical correlations between LIGO data and other astronomical observations, such as gamma-ray bursts or supernovae. Besides cross-checking data from the two sites, preparations have been made to cross-check with other GW observatories around the world.

You can browse additional LIGO pictures on the LIGO-Livingston web site at http://www.ligo-la.caltech.edu/photo.htm.

LIGO's future schedule over the coming months is:

A "mock data challenge" will happen about September 2001. This is essentially an exercise of the software alone. It will test the various data analysis software packages by running through fictitious data and seeing what comes out. (Note added 2 Oct: This did take place in early September at MIT in Boston; review of results is continuing.)
Fall 2001 is also a target for having LIGO-Livingston readjusted and closed up again. LIGO-Hanford is already at this point now (July 2001). When LIGO-Livingston is ready, another engineering run will be conducted. This is when the laser and everything else is operated to test for problems.
Between November 2001 and spring 2002 will be the first opportunity for the two LIGO sites to work in conjunction and do some low quality scientific listening (but not sensitive enough that detectable events are expected).
Between spring 2002 and 2003 is when planned operational sensitivity should be attained.
Starting in 2003 various proposed improvements are to be implemented at both sites. This upgrade is called LIGO II.

LIGO is not the only effort to detect gravitational waves:

Six other interferometer-type detectors are nearly complete or in the works. Several smaller prototypes have been operated in the past.
Another type of detector has been in use since the 1960s: the resonant bar detector. This method monitors a suspended metal bar for vibrations at its natural frequencies--including vibrations caused by gravitational waves. Work with resonant bars was pioneered by Weber (with no confirmed detections); currently six bar detectors--supercooled to reduce thermal vibrations--are in operation.
Work is proceding on three planned resonant sphere detectors. These use the same principle as the bar detectors, except with a spherical shape to allow detection of gravitational waves from all directions.
A couple of space-based interferometers are under development. Each would use three or four spacecraft separated by millions of kilometers, functioning as an laser interferometer with very long arms. Similarly, radio signals from several space probes to the outer solar system have been analyzed for possible gravitational wave signals.

Nearly all the operating (or soon-to-be-operating) detectors will compare results to improve the ability to confirm any detection of gravitational waves. For a list of GW detectors worldwide, click here.

To continue on to a discussion of black holes, click here.