Ripples in the fabric of spacetime could someday provide observational evidence for the goings-on in the earliest instants of the universe, revealing high-energy processes that currently remain opaque to even the largest particle colliders.

So-called gravitational waves are a prediction of Albert Einstein's theory of general relativity—moving objects perturb spacetime, generating waves like a boat moving across a lake. But the waves tend to be subtle, and only celestial heavyweights are expected to produce detectable effects. To date, only indirect evidence for gravitational waves has been found, although supremely sensitive detectors have been built to hunt for more direct proof in the form of waves emanating from nearby cataclysms such as a collision between two ultradense neutron stars.

A review paper in the May 21 issue of Science presents the prospects for detecting more primordial gravitational waves—those produced in the early universe that may still be detectable by the imprint they left billions of years ago or by the ripples that persist today.

Such primordial waves might offer the best means for testing cosmological models such as inflation, which holds that the newborn universe ballooned from a tiny pocket to something roughly 1026 times larger in just a sliver of a second. "It's hard to imagine a mechanism that's going to give us a direct window to a time closer to the instant of creation," says study co-author Lawrence Krauss, a theoretical physicist at Arizona State University. (Krauss, who writes a monthly column for Scientific American, is a member of the magazine's board of advisers.)

The first place to look for the mark of gravitational waves is in the cosmic microwave background, or CMB, remnant radiation from only 380,000 years after the big bang. The European Space Agency's Planck satellite, launched in 2009, is now following up its NASA predecessor, the Wilkinson Microwave Anisotropy Probe (WMAP), in measuring temperature fluctuations across the CMB. Those temperature fluctuations trace regions of greater and lesser density in the infant universe, providing important clues to how the universe and its component structures formed.

The CMB maps made by WMAP provided a boost to inflation, broadly affirming the inflationary model's predictions for what the early universe should look like, and more precise measurements could provide further confirmation. "The same events that we believe formed the hot spots in the cosmic microwave background could have produced gravitational waves, and we can estimate their magnitude," Krauss says. "It's at least possible that with the next generation of satellites we'll be able to observe their effects."

Gravitational waves passing through space would have left their imprint on the photons of the CMB in subtle polarization patterns. WMAP's measurements set upper bounds for the prevalence of those waves, and with Planck's greater sensitivity the newer spacecraft "might get lucky" and detect polarization from primordial gravitational waves, Krauss says.

Yale University cosmologist Richard Easther notes that CMB measurements are already yielding clues, albeit not huge ones, to the dawn of the universe.* "In fact, some inflationary scenarios are already ruled out because they would produce more gravitational waves than current measurements, mainly from the WMAP mission, would allow," he says. Planck and other experiments are now working to push those limits even lower. "So, if nature has been kind to us, we could have the first evidence for inflationary gravitational waves in the next few years," Easther says. If that evidence escapes Planck and its contemporaries, a more specialized polarization-measurement mission may be needed.

A CMB imprint from primordial gravitational waves would be indirect evidence, like the high-water mark left by a receding tide, but what of detecting the waves themselves? At present, the instruments most sensitive to locally produced gravitational waves are interferometers such as the dual Laser Interferometer Gravitational Wave Observatory (LIGO) installations. Each sends laser light down four-kilometer arms to look for interference effects that would be caused by passing gravitational waves stretching space in one direction and compressing it in the other. In future decades, space-based interferometers could use the same principles on much larger scales, bouncing laser light across the solar system, to directly detect the fainter primordial gravitational waves.

"Those kind of things could potentially—not in the near term but in the next generation—directly detect them," Krauss says. "We'd be able to look at the spectrum, and we'd be able to get some conclusive proof of what's going on." Easther says that the defining feature of a gravitational wave background produced by inflation would be that across all wavelengths, "from perhaps a few meters to the current size of the visible universe," the amplitude of the waves would be roughly the same. "To put this in perspective," he says, "if we could build a piano that produced gravitational waves instead of sound, the keyboard would need about a thousand keys to produce a big enough range of frequencies, and inflation manages to hit each note with almost exactly the same strength."

Although specific predictions of inflation might be vetted by detecting gravitational waves, a nondetection might not paint as clear a picture. "All inflationary scenarios produce gravitational waves, but the signal from some inflationary models is incredibly faint," Easther says. "Consequently, there is no consensus that failing to detect an inflationary gravitational wave background at some level can be taken as proof that inflation itself did not happen—although a huge number of specific inflationary scenarios would be ruled out."

Andrew Jaffe, a cosmologist at Imperial College London, portrays experiment and theory as participants in a sort of cat-and-mouse game. "We are rapidly getting to the point where the simplest versions of inflation may be ruled out, but it's very easy to build models that will evade the coming experimental bounds," he says. "In that case, we'll have to figure out much more clever ways to do our tests, or to attach the predictions of inflation to broader theoretical models of particle physics."

*CORRECTION (5/21/2010): This sentence originally referred to "CMB polarization measurements," but Easther was referring to measurements of the CMB's temperature fluctuations.