Tegmark and his colleagues based their analysis on an early look at the second data release from the ongoing Sloan Digital Sky Survey (SDSS). In two comprehensive papers submitted to the Astrophysical Journal and Physical Review D, they conclude that their new findings are in excellent agreement with earlier results. Among other things, the new best value for the age of the universe is 13.5 ± 0.2 billion years. This compares to 13.7 ± 0.2 billion years as determined early this year with the help of data gathered by the Wilkinson Microwave Anisotropy Probe (WMAP), which is analyzing the cosmic microwave background radiation from the Big Bang.
Other new refinements, based on Sloan and WMAP combined, include the amount of unknown "dark energy" filling space (70 percent of all existing matter and energy, compared to a previous best value of 73 percent) and the amount of exotic or "non-baryonic" dark matter: 26 percent, compared to the WMAP announcement of 23 percent.
"Different telescopes, different galaxies, different astronomers, and a different analysis — and we still get the same numbers. That’s quite comforting," says Tegmark.
Other astronomers agree. "It appears to be a very thorough analysis," comments cosmologist Sarah Bridle (University of Cambridge, UK).
How do galaxy positions tell about such fundamental properties of the universe? In principle, it’s easy. Had the age, the cosmic expansion history, or the amounts of dark matter and dark energy in the universe been different, the clustering properties observed among galaxies would also be different. Conversely, the clustering properties, statistically described in a graph known as a power spectrum, can be tied to a particular set of cosmic parameters. The uncertainties become much smaller when the power spectrum of primordial density fluctuations in the young universe is also taken into account. These fluctuations show their presence as minute temperature variations in the cosmic background radiation observed by experiments such as WMAP.
In practice, however, things are less clearcut. For instance, it’s not entirely clear how well the distribution of visible galaxies matches the distribution of the more ubiquitous invisible dark matter. Moreover, the number of cosmological variables that go into a particular model of the universe turns out to be 13, so the observations can still be matched by a range of slightly different models. The numbers quoted above involve some judgment calls in this regard, even though the Sloan study has narrowed the need for such assumptions.
Finally, the statistical analysis, which had to be tested by running hundreds of computer simulations on different model universes, is extremely cumbersome. Says Tegmark: "It’s very boring to read about, and in fact it’s very boring to work on, too."
But the result is anything but dull. In some cases, the error bars on the cosmological parameters derived from earlier projects such as WMAP and the Australian 2dF galaxy survey could be cut in half by including Sloan’s galaxy-clustering data. According to Tegmark, the main reason is that the Sloan survey is all digital — from telescope to final galaxy catalog — leading to fewer calibration worries and higher confidence. Also, analysis methods have been improving; "Even with the same number of galaxies, we can do better," says Tegmark.
Another recent discovery in the Sloan data is the largest structure yet found in the universe. This is a wall of many thousands of galaxies, at a distance of about 1 billion light-years, measuring almost 1.4 billion light-years across, report J. Richard Gott III (Princeton University) and colleagues. The "Sloan Great Wall" is three times farther away and 80 percent longer than the famous Great Wall of galaxies uncovered in 1989 by Margaret Geller and John Huchra (Harvard-Smithsonian Center for Astrophysics). Although unexpected by cosmologists, the existence of such a gigantic structure is not inconsistent with current ideas about the origin and evolution of the universe. "In a universe like ours, there’s about a one-in-six chance of encountering something this large," says Tegmark. "It’s not very unlikely, just a bit."
In the near future, constraints on the major cosmological parameters will get tighter and tighter. "We now know that the universe mainly consists of mysterious dark energy and dark matter," says Tegmark, "but future observations will hopefully tell us about the nature of both." The completion of the Sloan survey (foreseen for 2006) will help improve the accuracy of the results. So will the ambitious Large Synoptic Survey Telescope (LSST) project, the planned (but as yet unfunded) successor to Sloan.
Then again, techniques other than galaxy surveys might take the lead. Says Bridle, "I believe that the next few years will be dominated by constraints from the 'cosmic shear' technique: using the apparent alignment of galaxy shapes in random patches of sky due to gravitational lensing."