In mankind's attempts to gain some understanding of this marvelous place in which we live, we have slowly come to accept some principles to help guide our search. One such principle is that the Universe, on a large enough scale, is homogeneous, meaning that one part looks pretty much like another. Recent studies by a group of Australian researchers have established that, on sizes greater than about 250 million light years (Mly), the Universe is indeed statistically homogeneous, thereby reinforcing this cosmological principle.
How can we ever hope to get a grasp on something like the Universe? The three main principles upon which modern cosmology is based are:
- 1. The universal physics principle – the laws of physics are the same everywhere and everywhen;
- 2. The Copernican principle – the Earth is not in a special location within the Universe; and,
- 3. The Cosmological principle – at any given time the Universe is homogeneous and isotropic (uniform in all orientations) at large distances.
None of these principles is obviously true. In particular, the universal physics principle is tested regularly by experiment, but, lacking evidence for changes, it is simplest to assume physical laws do not vary, then keep a sharp eye open for contradictions. (Actually, there is some evidence that one of the fundamental parameters describing physical laws is a bit different when you look one direction into the cosmos than another, but this is still a controversial result.)
The Earth is in a particular location within the Universe - right under the computer I am using to write this - but not a special location. What the Copernican principle says is that if the Solar System were in a reasonably quiet region of a different galaxy, there would be different constellations, but at large distances the galaxies of the Universe would still have a similar distribution. This need not be true. If the Universe had an edge, and Earth was located just next to it, half the sky would be dark. Near the edge would therefore be a special location, as we would have an atypical view of the Universe. So far, observationally the Universe looks the same in every direction, consistent with Earth not having a special location.
That leaves us with the Cosmological principle, that the Universe on large length scales is homogeneous and isotropic. What this really implies is that the effects of gravitation on the overall structure of the Universe are the same in large samples of the Universe wherever they are centered.
All-sky map of the cosmic microwave background from data from the Wilkinson Microwave Anisotropy Probe - the colors indicate differences of energy of about 0.01 percent (Photo: NASA)
One bit of evidence pointing to a homogeneous Universe is the isotropy of the cosmic microwave background (CMB). The CMB has very nearly the same intensity regardless of which direction you look from Earth - it is isotropic from our vantage point. Moreover, that it still appears isotropic after travelling through the Universe for 13.7 billion years suggests that the early Universe was highly homogeneous, and that it has remained rather homogeneous since then. The problem is that isotropy in the early Universe does not necessarily imply the Universe is homogeneous now.
The universe within a billion light years of Earth, showing local superclusters of galaxies - approximately 63 million galaxies are shown (Image: Richard Powell)
Indeed, it is obvious that the Universe isn't homogeneous on short length scales - if I take the Earth to a point between galaxies, the Universe isn't going to look the same. On a larger size scale, the Universe is filled with a filamentary structure, with filaments composed of dark matter and clusters of galaxies. The distance between these filaments is a few tens of Mly, making it clear that the Universe is not homogeneous when regions this size are considered. If the Universe is homogeneous in larger regions, those regions must be large enough to average over the filamentary structures.
Fortunately, our rapidly improving instruments can study a considerable part of the Universe in detail – roughly out to some 12 billion light years (Bly) by the BOSS study, although reasonably complete coverage together with distance information has only been achieved out to about six billion light years. This sort of data allows to form a map showing the distribution of mass over a considerable portion of the observable Universe.
The 3.9 meter (12.8 foot) Anglo-Australian telescope, the last major telescope made with an equatorial mount (Photo: Australian Astronomical Observatory)
The Australian study of the homogeneity of the Universe is based on images and redshifts of roughly 180,000 blue galaxies up to a distance of about 6 Bly. The data was obtained by the WiggleZ Dark Energy Survey, carried out at the Anglo-Australian telescope at the Siding Spring Observatory just beyond the boundaries of the Warrumbungle National Park about 350 km (220 miles) northwest of Sydney. The data does not cover the entire sky, rather being restricted to less than 1 percent thereof, a patch measuring thirty by thirty degrees.
On the left appears a fractal having a two-dimensional structure not unlike the filamentary structure of the Universe. On the right appears the same fractal averaged over regions having about one-tenth of the area of the original (Image: Brian Dodson)
One might well imagine that the filamentary structure of the cosmos might be best described by a fractal. The figure above shows a fractal structure with a dimension of 1.5 having filament-like structures. On the left, the mass distribution is clearly inhomogeneous. However, if you average over larger regions, the result is the almost constant mass distribution shown on the right. In fact, it is difficult to distinguish the mass distribution on the right from a totally homogeneous distribution. Had the structure been inhomogeneous on the scale of the blurring, the blurred image would show varying mass density in different parts of the blurred figure.
When the Australian researchers started blurring their data (a simple expression for a complex procedure taking them years to carry out), they found a highly homogeneous mass distribution in the Universe. Taking the "homogeneity scale" as the size of regions that, when averaged out, were within one percent of homogeneous, they found the homogeneity scale to be about 250 Mly, the exact number varying slightly with distance in accord with the most widely accepted cosmological models. Considering their data covered distances about 25 times larger than this, it does appear that, at least for the last 6 billion years, the mass distribution of the Universe is homogeneous on scales larger than about 250 Mly.
Further work is clearly required to fully pin down this result. In the future researchers will cover more of the sky at larger distances, and thereby reach a final resolution of the validity of the Cosmological principle. But this study is the first serious step toward that resolution.
Source: The WiggleZ Dark Energy Survey: the transition to large-scale cosmic homogeneity[PDF], Australian Astronomical Observatory