By having a highly accurate value for the distance to the Large Magellanic Cloud galaxy, astronomers can get a better measure of cosmic 'dark energy'. Using binary stars, they have now achieved a value accurate to 2.2%. See Letter p.76
Distances to celestial bodies are crucial in astronomy. They allow astronomers to understand the structure of the Universe; for example, to see the organization of the Solar System and to recognize that galaxies lie beyond the Milky Way. The derived physical sizes of bodies scale with the distances adopted for them, whereas their energetics scale with the square of the distances. A current hot enterprise is to use distance measurements to the farthest supernovae to map out the expansion history of the Universe and to uncover the nature of the Universe's mysterious dark energy. Distances are deduced by means of a 'distance ladder': knowledge of the distances to nearby bodies is used to determine the distances of bodies farther out, and so on to yet more remote objects. On page 76 of this issue, Pietrzyński et al.1 claim to provide a much-needed, highly accurate measure of the distance to the Large Magellanic Cloud galaxy — the bottleneck in the ascent of the distance ladder.
Historically, the lowest 'rung' of the distance ladder, the size of Earth, was used to calibrate the timings of the transit of Venus across the Sun, and so to climb to the second rung, the Earth–Sun distance. The method of parallax — watching stars wobble back and forth as Earth orbits the Sun — was used to climb to the third rung, the distance to nearby stars. For cosmology and extragalactic astronomy, almost all ways of climbing the distance ladder pass through the rung of the Large Magellanic Cloud (LMC; Fig. 1), which is the evocatively named nearest galaxy to the Milky Way. So if the LMC distance is off by 10%, then the distances to all galaxies (as quantified with the Hubble constant, the expansion rate of the Universe) will be in error by the same 10%. When combined with measures of far-off supernovae, a highly accurate distance to the LMC will also substantially improve measurements of the properties of dark energy, and thereby potentially reveal its nature.
The distance to the LMC has long been a contentious question. Published values over the decades have been scattered, with values from the decade before 2001 alone having a scatter of 36%, despite the fact that the reported error bars are much smaller than this scatter2. So it is clear that the many methods used to determine the LMC distance had large and unidentified systematic errors. In 2001, the influential Hubble Space Telescope Key Project presented a middle value of the previous scatter with a reasonable error bar (50.1 ± 2.3 kiloparsecs, or 163,400 ± 7,500 light years)3. But its publication created a bandwagon effect, and for the next six years the astronomy community published distances that slavishly followed this value4. The 2011 Nobel laureate Adam Riess and the SH0ES project have since adopted a method pioneered by a group centred at Villanova University in Pennsylvania5,6,7,8 (involving four eclipsing binary stars) to obtain an LMC distance of 49.8 kiloparsecs, with an accuracy to 3% (ref. 9).
Now Pietrzyński et al., as part of the international Araucaria Project, derive a value that is accurate to 2.2%. They achieved this by measuring the distances to eight eclipsing binaries in the LMC. (Eclipsing binaries are two-star systems whose component stars regularly pass directly in front of each other as viewed from Earth.) They measured accurate sizes for both stars in each binary (from the durations of the eclipses and the measured velocities) and their surface temperatures (from spectra of the stars), thereby deriving the total intrinsic luminosity of each binary system. By comparing the total luminosity and the observed brightness, and using the inverse square law of light, they then determined the distances to the binaries. This method inspires confidence because it uses only undergraduate astronomy and physics (but very carefully done), and completely jumps over all the lower rungs of the distance ladder. The authors' new distance to the LMC, which is based on the average of the distances to the eight binaries, is 49.97±1.13 kiloparsecs.
A comparison of the new LMC distance with previously published values reveals three disturbing issues. First, the Araucaria Project had previously reported one of its eclipsing binaries to have a distance of 50.1±1.4 kiloparsecs10, but it now quotes 49.3±0.5 kiloparsecs for that binary, with no indication of how its analysis has changed. Second, and much more disturbingly, the four eclipsing binaries reported by the Villanova group5,6,7,8 have an average distance of 47.1±1.4 kiloparsecs, which is substantially different. The main difference between the groups' techniques is the use of cool stars by the Araucaria Project, together with an empirical surface-brightness-versus-colour relation to determine the stars' surface temperature, as opposed to the use of hot stars by the Villanova group and theoretical models to infer the temperature. Third, the new distance is remarkably close to the value obtained by the Key Project, with a difference of only one-tenth of the error bar quoted by Pietrzyński et al., so we should recall the history of the bandwagon effect.
What can we expect in the future? Studies of roughly a dozen more eclipsing binaries are in the pipeline, with results expected soon. Also, extensions of the technique will be made to the eclipsing binaries in the galaxy M33, in the constellation Triangulum. But the entire scene will change when the Gaia spacecraft is launched later this year, with its awesome capability for measuring accurate distances by means of the parallax, even out to the LMC. Gaia will provide definitive calibration of various 'standard candles' used to climb rungs above the LMC on the distance ladder, thereby eliminating the bottleneck at the LMC. Indeed, Gaia will obtain parallax data for many stars in the LMC with the resultant average accurate to 1%, and will map out its structure. So the time spent by the LMC's eclipsing binaries at the forefront of astrophysics will be limited to only the next few years.
About this article
CLUSTERING OF LOCAL GROUP DISTANCES: PUBLICATION BIAS OR CORRELATED MEASUREMENTS? I. THE LARGE MAGELLANIC CLOUD
The Astronomical Journal (2014)