An ancient view of acceleration

The Universe is expanding ever faster — the effect of 'dark energy', most astronomers believe. Surveys of how galaxies were distributed in the past could provide precise clues to what is driving this acceleration.

Gravity holds our expanding Universe together, or so astronomers have long assumed. The mutual attraction of galaxies should counteract the expansion of space that started with the Big Bang, causing the galaxies to slow down. But a decade ago, observations of distant supernovae gave evidence that this simple picture is wrong. That evidence has since been strengthened by a series of other cosmological probes, and the Universe's expansion appears actually to be faster now than it was billions of years ago. Understanding this accelerated expansion is the most pressing problem facing cosmologists today. On page 541 of this issue, Guzzo et al.1 give them a new handle on the acceleration: a comparison of the distribution of galaxies in the Universe now and in the past.

Ideas of what is causing the cosmic acceleration fall into two competing categories. The first is that the Universe is permeated by strange stuff known as dark energy that causes gravitational repulsion. The second is that the equations of general relativity — the theory of gravity formulated by Albert Einstein — are flawed and need to be modified. As best we can understand, the only observational effects of the acceleration are on the history of the Universe's expansion and the rate at which the clustering of matter has increased over time. Thus, we have extraordinarily few observational clues to distinguish between competing models. Astronomers are hungry for additional tests.

Like other techniques used so far, Guzzo and colleagues' approach to understanding the physical nature of the acceleration makes use of a fundamental aspect of astronomical observations: because light's speed is finite, the photons we receive now from a far-off galaxy were emitted at some time in the distant past. That makes it possible to observe the Universe when it was much younger than it is today. By observing galaxies at different distances, we can see how the Universe's properties have evolved as it has expanded.

One of the manifestations of the Universe's expansion is that features such as absorption lines in the spectrum of light emitted by a galaxy are systematically shifted to longer wavelengths — the phenomenon called redshift — by an amount approximately proportional to its distance. Measuring the distance of a galaxy from us is therefore straightforward if one has measured its spectrum.

This situation is somewhat complicated by the fact that, in addition to motion caused by the overall expansion of the Universe, galaxies attract each other gravitationally. In particular, a region of space with more than the average concentration of matter will attract galaxies to it, giving rise to motions that contribute to their redshifts. These motions cause a subtle, but measurable distortion in maps of the galaxy distribution. Determining the amount of this distortion both in the nearby (that is, present-day) and in the distant (early) Universe allows us to learn how the clustering of matter has changed with time (Fig. 1). That yields another clue to the nature of the cosmic acceleration2.

Figure 1: Clustering over cosmic time.


Guzzo and colleagues' survey1 of galaxy clustering represents a 'pencil beam' covering a region 2° across on the sky, stretching away from us in distance and time (here broken into three contiguous chunks). Each of the 9,126 dots represents a single galaxy. Distances are given in megaparsecs (Mpc; 1 Mpc is about 3.26 million light years) multiplied by h−1. This parameter represents the uncertainty in the Hubble constant, which is a measure of the rate of expansion of the Universe. Distance and time are also measured in terms of redshift z, with z = 1 corresponding to a lookback time of roughly 8 billion years. Guzzo et al. derive their constraints on cosmological models by examining subtle differences between the clumping of the galaxy distribution in the recent (small z) and ancient (high z) Universe. (Figure from ref. 9.)

Guzzo et al.1 measured redshifts of a large sample of distant galaxies using the 8-metre-aperture Melipal telescope in the Chilean Andes (Fig. 2). With such a large telescope, they were able to obtain redshifts of almost 6,000 extraordinarily faint galaxies, so distant that we see them as they were when the Universe was only about half its present age, about 7 billion years ago.

Figure 2: Very large telescopes.


The four telescopes of the European Southern Observatory's Very Large Telescope array adorn the 2,635-metre summit of Cerro Paranal in the Chilean Andes; the Pacific Ocean, at a distance of 12 kilometres, can be seen in the background. The telescopes are named after objects in the sky in the local Mapudungun language: from the left at the back Antu (the Sun), Kueyen (the Moon) and Melipal (the Southern Cross); in the foreground is Yepun (the 'evening star', interpreted as meaning Venus). Guzzo et al.1 used the 8-metre-aperture Melipal telescope for their survey of galaxies so distant that we see them as they were at half the Universe's present age.

The authors measured the distortion in the clustering of these galaxies, and compared it with values from surveys carried out in the nearby Universe. The error bars on the measurements are large, and all that can be shown is broad consistency with currently favoured cosmological models, in which dark energy is responsible for the accelerated expansion. But the authors also show that the next generation of surveys, which will cover 100 times the volume, will have the potential to distinguish between competing models, and provide the explanation for the accelerated expansion of the Universe: dark energy, a change in our understanding of how gravity works, or something even more exotic.

Astronomers are currently gearing up for the next such large redshift surveys. The DEEP2 survey, carried out using one of the two Keck telescopes situated at the summit of Mauna Kea on Hawaii, has observed 40,000 galaxies and has probed roughly the same cosmic epoch3. The third phase of the Sloan Digital Sky Survey4, based in New Mexico, is planned to include 1.5 million galaxies over the next six years at a somewhat more recent epoch than Guzzo and colleagues' survey. Other large surveys are planned or proposed using the Hobby–Eberly Telescope in Texas5, the Subaru Telescope located next to Keck on Mauna Kea6, and various space-based missions7,8. All are motivated, at least in part, by the desire to study cosmic acceleration. The technique described by Guzzo et al.1 shows that these surveys will be even more powerful than was hoped in constraining the nature of that puzzling phenomenon.


  1. 1

    Guzzo, L. et al. Nature 451, 541–544 (2008).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Linder, E. V. preprint at (2008).

  3. 3

    Coil, A. L. et al. preprint at (2007).

  4. 4

    Schlegel, D. J. et al. Am. Astron. Soc. 211th Meet. Abstr. 132.29 available at (2007).

  5. 5

    Hill, G. J. et al. Proc. SPIE 6269, 79–89 (2006).

    Google Scholar 

  6. 6

    Glazebrook, K., Eisenstein, D., Dey, A. & Nichol, R. preprint at (2005).

  7. 7

  8. 8

    Albrecht, A. et al. preprint at (2006).

  9. 9

    Marinoni, C. et al. Astron. Astrophys. (in the press).

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Strauss, M. An ancient view of acceleration. Nature 451, 531–532 (2008).

Download citation


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing