Epilogue

"We've done these calculations in a standard A—cold dark matter universe." The energetic young speaker at the front of the Phillips Auditorium at the Center for Astrophysics, Kathryn Johnson, a professor from Wesley an, was setting the stage for presenting her new results on galaxy cannibalism. There were 100 people in the room for the Thursday Astronomy Colloquium, Kathryn had a lot of new results to share, and she wasn't wasting any of her time or theirs by justifying the cosmology she had assumed.

Nobody blinked. Nobody asked a question. But my mind, always unreliable after 4 p.m. in a darkened room, started immediately to drift into speculation How could a "A" universe, two-thirds dark energy and one-third dark matter, be the "standard" picture in the autumn (^f 2003? Just 5 years earlier, cosmic acceleration had seemed unbelievable, and dark energy, in its guise as the cosmolog-ical constant, had been a notoriously bad idea, personally banished by Albert Einstein. What had changed?

Part of the answer is that the supernova results had gelled to become more solid. Further work by our high-z supernova search team had just come out in the September 1, 2003, issue of The Astro-physical Journal, and the Supernova Cosmology Project (SCP) had another paper in the works, too. Both teams still found that the universe was accelerating, 14 billion years after the Big Bang and 5 years after the first supernova results. We hadn't been fooled by bad luck or bad data—the new samples of type la supernovae at a distance of 5 billion light-years were, once again, 20% fainter than they would have been without cosmic acceleration. We could keep our goldfish, houses, and dogs.

And part of the answer was that some of the alternatives had been tracked down and shown not to be the cause of that slight dimming.

It probably wasn't pink pixie dust—this dust would have changed the colors of distant supernovae. We had measured those colors, at least for one supernova, and did not find any change. We have several more color measurements underway, but I am pretty sure we won't find a glaring problem with pixie dust.

And it probably was not the age of the stars. Both teams had used the Hubble Space Telescope (HST) to look at the galaxies in which distant supernovae were detected. On our team, Craig Hogan and Ben Williams had looked at the galaxies where supernovae were found. For the SCF, Richard Fllis and Mark Sullivan had inspected the sites of the explosions. When they sorted the host galaxies into elliptical galaxies (where there arc very few young stars) and spirals (where there are many recently formed stars), they still got the same result for cosmic acceleration.

And it wasn't something that just depended on how long the light from a distant supernova look to get to us. Adam Riess had flipped his transparencies in public and shown that for one object at the highest rcdshift, SN 1997ff, the SN la was not fainter than you'd expect in a coasting universe, but brighter. This was the smoking gun of chapter 11 that showed cosmology was really responsible for what wre see, not some error that growrs larger as you look farther back. We live in a mixed dark-matter and dark-energy universe. At first, for about 7 billion years, the universe would slow down due to dark matter, then shift over to acceleration as expansion diluted the matter, while dark energy began to drive a more rapid cosmic expansion. The transition from slowing down to speeding up should be roughly 7 billion years in the past, halfway back to the Big Bang; SN 1997ff demonstrated that we could find and measure supernovae that far back, at least with the Hubble Space Telescope.

SN 1997ff was only one object, but in Hawaii, John Tonry and his student Brian Barris were leading the way to higher redshifts from the ground, using the big camera on the Subaru telescope to find a handful of supernovae out beyond rcdshift 1. The supernovae found in Hawaii also pointed in the direction of a stop-and-go universe. Adam Riess and the higher-2 team were putting the new camera on the Hubble Space Telescope to good use, finding a dozen more objects that promised to confirm the earlier work.

Despite these improvements in the supernova data, I don't think that's the reason why our speaker spent no lime in selling out the pros and cons of a A cosmology. The real reason was a sudden convergence of many independent lines of research (in the very same values for the contents and age of the universe, weaving a web of evidence. Measurements of galaxy clustering from the Two Degree Field Galaxy Redshift Survey carried out in Australia were much more comprehensive than the Las Campanas Redshift Survey data shown in chapter 5, and the new data pointed to a universe with - 0.3- This was the same value we were getting from the supernova analysis.

The same was true with cosmic ages. Once you factored in the cosmic history prescribed by the supernova data of deceleration due to dark matter followed by acceleration due to dark energy, measurements of the Hubble constant from HST observations of cepheids (done just the way Hubble had done them in 1929 except in galaxies 25 times further away) gave an age for the universe of about 13-6 billion years. This was a very good match to the age of the oldest stars, as judged from the time it takes stars to fuse hydrogen into helium in their cores. The oldest globular clusters were recently gauged to be around 12.5 billion years old, allowing just enough time for them to form in the first 1 billion years after Lhe Big Bang. When completely independent paths lead to the same place, it makes you think something good is happening.

All of this was very satisfying, but the most powerful new set of information came from better measurements of the cosmic microwave background (CMB). Since 1998, pioneering experiments had measured and reported the subtle texture of the exceptionally smooth glow from the Big Bang. Some used sensitive detectors carried high into the atmosphere on balloons, and others measured the CMB fluctuations from high, dry sites in the Atacama Desert of Chile and stations in Antarctica. Those preliminary results, when combined with the supernova measurements, matched well with a universe that had £2iri of 0.3 and QA of 0.7, as discussed in chapter 11. Like Los Angeles, the universe was one-third substance and two-thirds energy.

But the best was yet to come. In 2001, the Wilkinson Microwave Anisotropy Probe (WMAP), a satellite to measure the CMB, was launched into a unique orbit at 4 times the distance to the

Moon. From this superb perch, it patiently mapped the whole sky for a year.

'ITic first data from WMAP were released in February 2003, and those results changed the tone of the discussion in observational cosmology from cautious and tentative to confident and quantitative. Just as the earlier measurements had indicated less precisely, the WMAP results confirmed that the universe has the large-scale geometry of flat space. This means that the 3-dimensional space we live in has none of the tricky curvature that Gauss imagined and Einstein showed how to compute. Just as expected, tf inflation is right, the large-scale geometry of the universe is the geometry you learned in high school: parallel lines don't meet, the angles inside a triangle add up to 180 degrees, and the surface area of a sphere is 47cR2. Although general relativity provides for many other possibilities, the universe we Jive in is the simplest of these. When you combine the WMAP measurements with other evidence, the best age for the universe is about 13-4 ±0.2 Gyr, and the best estimate for the present-day composition of the universe is flra = 0.27 and = 0.73.

These results are qualitatively the same as the earlier ones, so there was no big surprise when the WMAP results were presented (with vigorous tub-thumping from NASA) in February 2003, but rather an audible sigh of relief. Even though the CMB measurements don't detect cosmic acceleration directly, as the supernova measurements do, taken together, they point with good precision to a universe with both dark matter and dark energy. Things were fitting together—and the better you measured them, the better they fit. Quantitative agreement is the ring of truth. This is the reason why, by the autumn of 2003, our colloquium speaker didn't bother to make the case that a A-dominated universe was the right picture.

So, what's next? We arc confident that dark energy is real, but what is iV.^Supernovae led the way in revealing cosmic acceleration. Can they now be used to pin down the nature of dark energy? We want to End out whether this weird stuff is really the cosmological constant Einstein created and discarded or possibly a more general "quintessence" that changes over time. If we could trace the onset of acceleration more precisely, we could tell if we were seeing something that is constant or something that is changing. But to do this, we need to improve our technique. More of the same isn't good enough. We need better precision.

I lere's part of the plan: find 200 SN la in the next 5 years. Measure the light curve for each one in the same way and get a good spectrum of every one. If you do that, you should be able to tell whether dark energy has the same properties as the cosmological constant. Technically, we will find the "equation of state"—tracing the way the energy density of dark energy changes as the universe expands. This cannot fail to be interesting—either dark energy is the cosmological constant or it isn't. Either way, it is a deep mystery: there is still no explanation why the cosmological constant should be 10120 times smaller than the simplest theoretical estimates, and if dark energy turns out to be something else, that's also of tremendous interest. After all, whatever it is, it makes up two-thirds of the universe! We should take an interest.

To measure this property of dark energy, you need enough telescope time to find and follow 200 faint supernovae, enough computer power to sift through the data immediately before the supernova fades away, and a good acronym. We have all three. The ESSENCE (Equation of State: SupErNovae trace Cosmic Expansion, pronounced just like "SNs") program has been granted time on the 4-meter telescope at Cerro Tololo for 5-years worth of supernova hunting. We go to a small number of fields every 4 nights—that's frequent enough to get good light curves for the supernovae we discover, and the fields are big enough, rhank.s to the large CCO camera at the 4-meter, to make it very likely we will have some live supernovae in each month's series (if observations. Chris Stubhs, recently at the University of Washington, but now at Harvard, has built a powerful dedicated computer system that can lake each picture from an observing night, compare it to earlier images, and find ali the new objects in our data by the next morning. We have time at the Gemini Observatories, VLT, Keck, and on the Magellan telescopes to get the spectra that will show us whether these are type la supernovae, and tell us whether the distant objects are the same as those nearby.

This enterprise, and similar work being carried out at the Can-ada-Francc-I fawaii Telescope, should build up a precise set of data for the era about 5 billion years ago, when the universe was switching over from deceleration due to dark matter to acceleration due to dark energy. The pace of the acceleration will tell whether this results from a dark energy that behaves like the cosmological constant or a dark energy that behaves differently as the universe expands. In a few years, we should have a grip on the nature of dark energy.

Similarly, the push to redshifts beyond 1 will show whether this picture is complete The new "Advanced Camera for Surveys" on the I lubble Space Telescope is spectacularly good for searching out distant supernuvae, measuring their light curves, and even at getting their spectra. In the next 2 years, we should go from a sample of 1 to a sample of 10 to a well-distributed sample of a few dozen high-rcdshift super novae. These will show if we really do live in a stop-and-go universe with 7 billion years of cosmic deceleration from dark matter's drag followed by 7 billion years of cosmic acceleration powered by dark energy's push.

Technically speaking, the rate of change in an object's position is called velocity, the rate of change in velocity is called acceleration, and, to the unending amusement (if physics students everywhere, the change in acceleration is called "jerk." So the search for the switch from cosmic deceleration by dark matter to cosmic acceleration by dark energy is a search for cosmic jerk. When Adam Riess gave a talk in October 2003 about preliminary results on the higher-jr search that showed evidence for changes in cosmic acceleration, the headline writer at the New York Times couldn't resist running his picture under the banner: "A Cosmic Jerk that Reversed the Universe."

The Hubble Space Telescope is old enough now that we arc beginning to think about the endgame for this splendid machine and the transition to the next big thing—the James Webb Space Telescope (JWST), a large orbiting telescope designed to function in the infrared, where the high-redshift supernovae shine. Earth's atmosphere is very thin at the altitude of 380 miles, where I lubble orbits, but there is a miniscule drag that, little by little, is lowering the orbit and, in time, will make HST spiral inward and enter the thick part of Earth's atmosphere, where it will burn up. NASA is planning to send the space shuttle up scxin, once they are confident it will be safe to do this (perhaps in 2005), to install a new instrument on HST and boost the telescope to a higher orbit that should keep 11ST out of trouble for several more years.

NASA's original plan for the endgame was to use the space shuttle to go up to HST in 2012 (or whenever is the right time), put the telescope in the cargo bay, and bring it down to Earth to hang in the Air and Space Museum on the National Mall in Washington, D.C. After the Columbia tragedy in 2003, this no longer seemed like such a good idea. It's one thing to risk astronauts' lives to do something important that advances science—and science education is important—but it doesn't seem right tu lake that risk for a museum piece.

So now the discussion centers on ways to extend the life of HST and to bring the 12-ton satellite down in a controlled way when the time comes. We want to be sure it eventually ends up in the Indian Ocean, not in Indianapolis. NASA engineers are working hard to design a small rocket that could be attached to I (ST during a shuttle servicing mission in 2010. The shuttle could boost HST for operation until JWST is ready, and then that small rockct could be used to bring down HST in a controlled way, sometime after 2012.

As a scientist, I think it would be a good thing—if NASA must take the risk and bear the expense of making a shuttle trip to install that bring-it-down rocket in 2010—to use that final servicing mission to install an even better camera that could be used for the dark-energy problem. A high-powered committee, headed by John Bahcall, has recommended that NASA consider how to lisc that shuttle trip to get more science from HST, if possible. We will see how all of this turns out. We need to balance the desire to get on with JWST against continued operation of the existing space telescope. One possibility would be to simplify the operations of HST in its final years by concentrating on the dark-energy problem with a wide-field camera taken up on the final servicing mission.

Meanwhile, the Lawrence Berkeley Lab, home of the Supernova Cosmology Projcct, is investigating the possibilities for SNAP (Su-pcrNova Acceleration Probe), a satellite dedicated to the study of dark energy, It is a very ambitious program, with a proposed telescope almost as big as HST, a billion-pixel camera with hundreds of CCDs, and a spectrograph envisioned to work superbly from ultraviolet to infrared wavelengths. If SNAP is built and performs as specified, it would provide a large and uniform set of 2000 well-observed supernovae that should narrow down the nature of dark energy to a few percent by some time around 2014. Like all large satellite programs, the SNAP team will require persistence, fortitude, good luck, and a boatload of money to reach its goals. I wish them well.

And finally, we turn to the realm of theory and experiment on the very small scale. As a (slightly fraLidulent) member of the International Brotherhood of Theorists (through my experience at the Institute for Theoretical Physics), i hope that there will be a theoretical breakthrough to match the progress in observations. There are now 1,545 papers on the High Energy Physics preprint server that cite our original evidence for an accelerating universe. Most of them arc theory papers. 1 can't claim to have read them all. But, as I understand it, there's no great progress to report on answering the question, Why is the vacuum energy so much smaller than predicted from a simple calculation? Although it may be possible to make a string theory of all the forces that reside in 11 dimensions, and gravity may seep into those extra dimensions, we don't yet know why gravity would make the cosmic acceleration small but real. One possible path might come from the experimental world of panicle accelerators, where investigators hope to see signs of "supcrsymmetry," a theory of particle physics that goes beyond today's Standard Model. In supersymmetry, the lightest particle, dubbed the "neutralinofrt might be a g<x>d candidate for the dark matter, and a cancellation (if sorts might be able to make the dark energy small. Of course, this discussion would be more convincing if experiments at Fermilab or at CF.RN show that the neutralino exists in the real world as well as in the minds of theoretical physicists.

But it is certainly possible that a conceptual advance will show us why the cosmological constant is 0.7. The present quantitative disagreement is so large it would count as a great success if the prediction were 7 or 70 or 700 in those units. We shall see. Cosmology is on the minds of the string theorists, and those are very active minds.

The discovery (if cosmic acceleration has been a tremendous adventure in finding out how the world works. Although we bravely claim to be entering the era (if "precision cosmology," there are still great voids in our understanding. While we arc beginning to determine the amounts of dark energy and dark matter to a few pcrccnt, we still don't know what either of them is. But we do know what to do next, and we arc eager to get on with the hunt. This story is not over In fact, the fun has just begun.

0 0

Post a comment