Gravity waves spread through space and time like ripples on a pond, warping the fabric of the universe as they pass. The largest waves emanate from the most cataclysmic events in the universe: stellar explosions, mergers of black holes, and the violent first moments of cosmological history. Or so the venerable theory of general relativity goes—although many predictions of Albert Einstein's theory of gravity have been proved, only indirect evidence for gravity waves has been found.

An experiment looking for gravity waves directly has, in the course of not finding them, placed new upper limits on how noisy the universe's gravity-wave background could be. Those limits, the researchers say, can refine or even rule out cosmological models that predict large backgrounds originating from processes in the early universe, and more implications should be forthcoming as the experiment, known as the Laser Interferometer Gravitational-Wave Observatory (LIGO), ramps up in sensitivity.

The LIGO team, working with the European Virgo collaboration, presents its findings from data collected between 2005 and 2007 in a recent issue of Nature. (Scientific American is part of the Nature Publishing Group.)

Vuk Mandic, an astrophysicist at the University of Minnesota and a member of the LIGO team, says the observatory is providing a taste of things to come in the field. "This is the first time that this type of experiment, directly searching for gravitational waves, has reached the sensitivity that is sufficient to start probing cosmology and early-universe models," Mandic says.

LIGO comprises two observatory sites, in Washington State and Louisiana, each of which hosts an L-shaped laser interferometer with four-kilometer-long arms. A laser beam originating at the elbow splits into two and travels down each of the arms, before being reflected back by mirrors at the ends.

A passing gravity wave would temporarily stretch one arm of the detector while squashing the other, and the two beams, having traversed different distances, would be out of phase when recombined.

The challenge facing LIGO researchers is immense—as California Institute of Technology physicist Marc Kamionkowski noted in a commentary accompanying the new paper, gravity wave detectors "require detection of minute changes—a mere fraction of the size of an atomic nucleus—in kilometer-scale separations between free-floating masses."

In order to rule out spurious signals, LIGO has a suite of passive and active noise suppressors to counteract vibrational effects from passing pedestrians or trucks. The mirrors at the end of each arm, for instance, are suspended on pendulums to insulate them from vibrations, and a forthcoming improvement to the observatory (called Advanced LIGO) will daisy-chain four pendulums together to multiply their suppression effects, Mandic says.

Noise from seismic activity, on the other hand, can be suppressed by searching for gravity waves at relatively high frequencies—the LIGO paper sets limits on gravity waves in the band around 100 hertz, or cycles per second, whereas seismic noise is generally below 25 hertz.

By improving on LIGO's past limit by an order of magnitude, Mandic and his colleagues have further constrained the amplitude of the gravity-wave background that emerged from the universe's first moments. That background, like the cosmic microwave background—which affords cosmologists their best estimate of the age of the universe—should be packed with information about the early universe. But gravity waves have a unique value: they carry information about the all-important first minute following the big bang—farther back in time than we can currently probe.