GIVEN THE DAUNTING TASK of explaining the universe without the luxury of being able to visit much of it, scientists have done pretty darned well. If the history of the cosmos were a movie, they’ve used physics to rewind the film approximately 14 billion years, all the way back to a fraction of a moment after the Big Bang began. Yet, the crucial events of the unimaginably brief, earliest instant after creation are still a big mystery, the subject of only educated guesses. But those first events were critical to the development of the universe as we know it today. Cosmologists now hope that a new generation of telescopes will finally reveal clues to exactly how the Big Bang got started, and several research groups are engaged in the race.

Among these new research efforts, one innovative project stands out. With a modest budget and plenty of elbow grease, Adrian Lee and a small group of cosmologists at the University of California, Berkeley, plan to build PolarBear, the first of a new kind of radio telescope with supersensitive detectors. The sky might have been hiding a snapshot of creation all along, and with luck, PolarBear could be the first to see it.

Cosmology is in a revolutionary period, Lee says, and the textbooks for the field have to be rewritten almost every year. “All we want to do is set up an experiment where we can really contribute to this revolution,” he says.

Before the Big Bang, matter and light did not exist. In the most widely favored hypothesis of the universe’s birth, everything came out of almost nothing virtually in an instant — within the first trillionth of a trillionth of a second — in a process called inflation. Inflation is an appealing theory, experts say, because it fits well in the Big Bang’s timeline, and seems to solve a number of cosmological riddles. Inflation would explain, for instance, the smoothness of the early universe, and with it, the mostly uniform distribution of galaxies that astronomers see in the sky today. At the same time, inflation would have given rise to a universe lumpy enough that those galaxies — and the stars and planets they are made of — could form in the first place, under the pull of gravity.

But confirming the theory has been a challenge. The physical conditions of the early universe are, of course, impossible to recreate in the laboratory. And no one has ever observed any direct evidence that inflation did take place. But that could be about to change.

According to Albert Einstein’s theory of gravity, the Big Bang would have created ripples in the very geometry of space. These ripples are called gravitational waves. The waves unleashed in the first second of that unfathomably violent birth, physicists say, were so intense that they would still be shaking the cosmos now, 14 billion years later. But today they would be mere ghosts of their original strength. Physicists are building special detectors that, within a few years, could be sensitive enough to detect them directly, something that has never been done before. It would be like listening to the sounds of the Big Bang, hearing remote whispers that were once a cataclysmic roar. But cosmologists also predict that indirect traces of such whispers will show up in the images captured by radio telescopes such as PolarBear. If the experts are right, it would provide incontrovertible proof of inflation — and help them figure out which of a host of competing theories about the Big Bang is correct.

Through the study of gravitational waves, says Kip Thorne, a theoretical physicist at Caltech in Pasadena, California, cosmology will enter a golden age in which observation, not just theoretical guesses, will drive our understanding of the first instants of the universe’s life. By various research approaches, Thorne predicts, there will be “something like 35 years of exciting studies of the first second of the universe that will come from gravitational waves.” That golden age, Thorne says, will begin with the new radio telescopes.

ACCORDING TO EINSTEIN’S THEORY, the presence of matter curves the geometry of space. When any massive body accelerates, it will create gravitational ripples in the distorted pocket of space surrounding it. These waves propagate at the speed of light, and they are unlike anything we experience in everyday life.

Gravitational waves alter distances and distort shapes, similar to the seismic waves that constitute earthquakes. A gravitational wave that travels through the solar system and passes through Earth from pole to pole would turn the planet’s equator and all its other longitudinal circles into ellipses — squeezing them in one direction and stretching them in the other, then vice versa, swinging back and forth. However, gravitational waves propagate not through matter but through the fabric of space itself, even if it is empty. Therefore, counterintuitive though it might seem, unlike earthquakes, gravitational waves do not, technically, move anything.

Imagine, for example, two hypothetical spaceships in orbit at opposite points over the equator. The spacecraft would not feel any acceleration at the passage of a gravitational wave. Yet they would become alternately closer and farther from each other, not because they are moving back and forth, but because the space between them — and the Earth with it — has stretched and shrunk.

In practice, only the most dramatic cosmic phenomena, such as the collision of two black holes, release waves strong enough for scientists to measure. Physicists are already attempting to do just that. In Washington and Louisiana, two laboratories collectively called LIGO — for Laser Interferometry Gravitational Wave Observatory — are searching for relatively short-wavelength gravitational ripples from black hole collisions. Each lab shoots laser beams back and forth down two 3-kilometer-long vacuum tubes. LIGO will look for gravitational waves by measuring changes in the length of the laser’s paths, and thus changes in the distance between the two ends of each tube. With a sensitivity that will reach the scale of the width of an atomic nucleus, LIGO aims to detect collisions happening within 100 million light-years (the distance light travels in 100 million years) from Earth. Such events should happen on average about once every few years, experts say. Similar gravitational-wave hunting labs are under construction in Europe and Japan.

The same principle will be the basis of a trio of space probes that will orbit the sun in formation. This project is the Laser Interferometer Space Antenna (LISA), a joint mission between NASA and the European Space Agency scheduled to start in 2012. LISA’s probes will beam lasers at each other and reflect them back, measuring changes on the order of one-tenth the size of an atom in the distances from probe to probe. LISA is designed to detect waves at much longer wavelengths than LIGO. In addition to picking up signals of black hole collisions, LISA could detect gravitational waves coming from the most remote and dramatic cosmic phenomenon of them all — the creation of the universe.

Any measurement of those waves would be a breakthrough. Different models of the Big Bang predict gravitational ripples of varying wavelengths. “There’s a large range of different wavelengths,” says Caltech’s Thorne, “where we have a shot at probing the first second of the universe’s life.” Collecting actual data on gravitational waves would validate some models while ruling others out, or require new models altogether. LISA, Thorne says, could for instance yield clues about how the fundamental forces of nature were born, within the first trillionth of a second.

However, the crucial first trillionth of a trillionth of a second of the universe’s life could be beyond even LISA’s reach, experts say. Within that first instant, the theory of inflation goes, some sort of repulsive force caused the cosmic expansion, pushing space apart faster than the speed of light to create matter, energy and gravity out of nothing. Gravitational waves generated from this cataclysm would have been stretched out so dramatically by inflation that they would make all of the sky vibrate at once. And they would take billions of years to cross our solar system. But without an interstellar spaceship and a detector the size of a galaxy, scientists don’t yet know how to build a device that would measure the waves directly.

BUT IF THE WHOLE COSMOS is shaking, why not use the universe itself as a measuring instrument? That seemingly preposterous proposal came in 1997 from three cosmologists, Marc Kamionkowski, now at Caltech, Arthur Kosowsky, now at Rutgers University in Piscataway, New Jersey, and Albert Stebbins of Fermi National Accelerator Laboratory, near Chicago. Rather than detecting gravitational waves directly, the trio says, scientists could prove the existence of inflation’s ripples by showing that the ripples changed the very way the early cosmos looked.

For roughly 400,000 years after the Big Bang, the universe was a rarefied plasma, a mixture of protons, neutrons and electrons about as hot as the surface of the sun. Then the plasma started to cool down just enough to form the first atoms — mostly of hydrogen, the simplest element. Much of that hydrogen gas later would coalesce to form the early galaxies and stars.

Plasma is essentially opaque to light. But at the time of its transition to hydrogen, some of the plasma’s glow was able to escape. That light has been traveling ever since, and as the universe cooled and expanded, its wavelength grew longer. It’s similar to what happens to the sound waves inside a trombone when its player slides the handle to get a lower pitch; the waves flatten out because the space within the instrument has expanded. By today’s time, the light’s waves had descended into the invisible, infrared realm, and down to microwaves, the kind of high-frequency radio waves used for TV broadcasting. (The only difference between light and radio waves is in the wavelength; both are electromagnetic radiation.) About 1 percent of the “snow” picked up by a TV today — peaking around channel 69 — comes from this so-called cosmic microwave background. It’s the afterglow of the Big Bang.

To those who first discovered this afterglow in the mid-1960s, the cosmic microwave background seemed to come with equal intensity from all directions of space. But in 1990, an orbiting NASA telescope proved otherwise, finding fluctuations in intensity on the order of 0.001 percent, and revealing patterns in the sky. Cosmologists reckoned that those patterns were the early signs of the formation of the galaxies; Stephen Hawking called that finding the discovery of the century.

In 2000, a new NASA probe, the Wilkinson Microwave Anisotropy Probe (WMAP), took even more precise measurements, giving the most accurate picture yet of the universe at age 400,000 years. By Kamionkowski and his colleagues’ calculations, the passage of gravitational waves should show up in that “baby picture” of the universe. The waves released by inflation left an unmistakable signature, the cosmologists predicted: They would have deflected particles of light, or photons, and caused them to line up in certain ways — an effect known as polarization. But although both WMAP and a 1999 Antarctica experiment demonstrated that the microwave background is polarized, neither team of scientists discovered the signatures of inflation.

Far from being deterred, however, several research groups, including Adrian Lee’s team at Berkeley, are now honing their tools, hoping to find the missing polarization patterns. To succeed, they reckon they will need at least a 100-fold improvement in the sensitivity of current instruments. Lee and his PolarBear colleagues believe they can achieve that goal thanks to a new-generation detector that Lee calls “a real leapfrog step over what’s been done before.”

With a main mirror three meters (10 feet) wide, PolarBear will be twice as large as WMAP. But the critical difference will be in the detector at the new telescope’s core. While WMAP’s main purpose was to measure the small variations in the intensity of the background radiation between different points in the sky, PolarBear’s will be completely dedicated to measuring polarization.

WMAP and other traditional microwave detectors are typically made of a dozen or so sensors. Each sensor is essentially a small antenna that records a single pixel of data from the sky. PolarBear will host one of the first microwave detectors built by arranging thousands of these sensors in one array.

“The technology we’re using is actually very similar to the technology that a company like Intel would use to build computer chips,” says Roger O’Brient, a graduate student working with Lee. The technology is commonly used to make CCDs, the light-sensitive chips that power everything from consumer digital cameras to the Hubble telescope. Having a detector with thousands of pixels means taking more data at once. In the past few decades, Lee says, CCDs have brought a revolution in optical astronomy, and he anticipates that the same will soon happen in radio astronomy.

PolarBear’s polarization detector will also be the first to use superconductors to pick up and amplify signals. Superconductors are materials that, when cooled below a threshold temperature, cease to offer any resistance to the passage of electric currents. The detector will be housed in a liquid helium tank, close to the threshold temperature of half a degree above absolute zero (minus 273 Celsius, or minus 460 Fahrenheit). The passage of polarized waves through the array would generate a signal by releasing just enough heat to turn the superconductor off, O’Brient explains.

Like all radio telescopes, PolarBear will need a clear, quiet sky and a dry climate, since atmospheric humidity tends to absorb microwave radiation. Other than in space or at the South Pole, the best location is an arid mountaintop. Lee and his team chose Barcroft station, a desolate research camp at an altitude of 3,810 meters (12,500 feet) in the White Mountains range, between California and Nevada. The University of California originally established Barcroft in 1952 to study the long-term effects of high altitude on humans and animals.

The Berkeley physicists plan to start building the scope next year. With their small budget, they will have to haul their equipment up there by themselves, without the help of workers or engineers. “People in our field have this reputation of being cowboys,” says Huan Tran, a postdoctoral researcher at UC Berkeley who is designing PolarBear’s optics, “because in order to get the job done with limited funding, you have to do everything.” As a graduate student working with Lyman Page, a WMAP researcher at Princeton University in New Jersey, Tran helped build a telescope in the Atacama mountains in Chile. “It’s a very unique thing,” he says. “You pulled up to the mountaintop on a truck, you tightened every bolt to make this thing work.”

But one problem that could hamper observations is the so-called cosmic foreground. Polarized microwaves emanating from our galaxy — from interstellar dust or cosmic rays — could be strong enough to drown the signal in noise. “That could be worrisome,” says Lee, “but people don’t know yet.”

EVENTUALLY, A SPACE TELESCOPE mission would stand the best chance of success at finding gravitational waves, being free from the interferences from human activities — even a cell phone can ruin the day of a radio astronomer — and from the filter of the atmosphere. Moreover, a space telescope will be able to survey the whole sky, something impossible from anywhere on Earth. Sometime after 2007, NASA plans to launch CMBPol, a radio telescope dedicated to measuring polarization. For now, Lee says, even if PolarBear is unsuccessful, the experience that researchers gain from the project will advance the state of the art, and benefit the rest of the scientific community. “One thing that PolarBear will do,” he says, “is investigate techniques and technology that can be used on a satellite such as CMBPol.”

Meanwhile, a number of other projects are in the pipeline to measure the polarization of the cosmic background using traditional detectors. Stanford University’s Sarah Church and her colleagues are working on QUEST, a new detector for a South Pole radio telescope which should start taking data next March, before the end of the Antarctic summer. A team based at Caltech and the University of Chicago is working on a similar project. And two additional space experiments are in preparation: a radio telescope built at the University of Bologna, Italy, for the International Space Station, and a new probe called Planck, which the European Space Agency plans to launch in 2007.

But for the time being, technology will give PolarBear an edge over its competitors. “Everybody in the field will say this: The superconducting technology is going to be the future,” says Keith Thompson, a researcher in Stanford’s QUEST team.

Though virtually everybody in the field is excited to see what happens, some doubt that signatures of inflation will show up at all. The energy scale of inflation could be a problem: That is, if it was too fast, or too slow, no signatures would be visible at all, says Joel Primack, a cosmologist at the University of California, Santa Cruz. “The window where they could exist is very small,” Primack says. Still, he adds, “Lee’s is probably the group that’s likely to make the big discovery, if there is one.”

Lyman Page, the WMAP researcher at Princeton, thinks that, even if the signatures of inflation exist, distinguishing them from foreground noise might be impossible without sending a probe such as NASA’s planned CMBPol into space. “It’s just that measurements are very difficult, and the whole community has a lot to learn about how to make them.” Still, Page wants them to press on. “I am an ardent supporter of all these efforts,” he says,. “just because you have to do them. You have to try.”