Stars are grouped together in galaxies. Galaxies form clusters, filaments, and sheets. On a large scale, the universe appears highly organized. But just how was it able to form these patterns that alternate between empty voids and regions of highly concentrated matter?
In other words, how was the primordial soup able to give rise to such complexity? According to François Bouchet, from IAP, “ever since Georges Lemaître's work in the 1920s, the Belgian astronomer who first formulated the Big Bang theory, cosmologists have believed that the essential mechanism by which structures form is gravity. This is thought to have acted on the initial conditions of the universe, which contained tiny spatial fluctuations in density.” So a few lumps in the primordial soup, together with the force of gravity, were probably enough to give rise to the complexity we observe today.
© Projet Horizon (2005-2008)
The evolution of part of the universe–a chunk 150 million light years across–has been modeled by the Horizon project team. Long filaments of matter surrounded by almost totally empty voids can be seen.
Clues from Fossil Radiation
In the 1980s, cosmologists began hypothesizing that in the universe's very first moments, space-time must have been perturbed by irregular fluctuations of quantum origin. “These ripples then expanded to cosmological distances,” Bouchet continues. These first instants have bequeathed an invaluable legacy to astrophysicists, encoded in properties of the cosmic microwave background radiation (CMB) first discovered in 1964. In a way, CMB represents the first light emitted by the universe. NASA's Cosmic Background Explorer satellite (COBE), operational during the 1980s, revealed the existence of tiny irregularities in the fossil radiation, which are the imprint of these primordial fluctuations. Their amplitude and distribution in the sky were subsequently mapped more accurately by NASA's Wilkinson Microwave Anisotropy Probe satellite (WMAP). Once examined in the light of theory, these data enabled scientists to get a better understanding of the composition of the universe
. This was the missing ingredient needed to develop models that could accurately simulate the formation of the structures of the universe.
Such models are at the heart of the French project Horizon1
which has been ongoing since 2001. It is the most ambitious cosmological numerical simulation project in the world, and brings together around 20 researchers, several of whom are from CNRS. The aim is to simulate the evolution of a significant chunk of the universe over a period of 13.5 billion years. “Just as climatologists model climate by numerically solving the physical equations that govern its evolution from an initial configuration, we have reconstructed the history of the universe,” explains project-leader Romain Teyssier, from AIM,2
a leading French astrophysics laboratory. Scientists fed into their computer the initial content of the universe and the spatial distribution of the primordial fluctuations as revealed by the CMB, and started solving the equations of gravitation from 380,000 years after the Big Bang until today. “We placed ourselves in an expanding universe as described by general relativity, and which integrates a priori the effects of dark energy. And we computed the evolution of two 'fluids,' one representing dark matter, the other ordinary matter,” Teyssier explains. Modeling the Birth of the Universe
In an initial model carried out in 2006-2007, the Horizon researchers simulated the evolution of a chunk of the universe 150 million light years across. This enabled them to observe the formation of galaxies that were astonishingly close to reality. In summer 2007, they repeated the experiment using the CEA's Bull civil supercomputer. This time, they simulated a sphere with a radius of six billion light years with no less than 70 billion particles. Amazingly, the simulation reconstructed a universe as it would be seen by an Earthbound observer looking at half of the visible universe.
Once processed, the images obtained were staggering, and almost identical to real photographs. As their 'model' universe became older, the astrophysicists observed how the primordial fluctuations became amplified under the effect of gravity, with visible matter falling into increasingly concentrated regions of dark matter (dark matter “halos”), forming stars and galaxies. The galaxies organized themselves into a kind of web-like foam, leaving huge spaces completely devoid of matter. “Our results are in extremely close agreement with very large-scale surveys of the galaxies,” Teyssier points out. These include the VIMOS VLT Deep Survey (VVDS), which uses the Visible Multi-Object Spectrograph (VIMOS) on ESO's Very Large Telescope (VLT) in Chile, to quantify the evolution of 90% of the history of the universe.
“This agreement validates the scenario of a universe in which the formation of structures is dominated by the gravitational attraction exerted by the dark matter halos that emerged from the primordial fluctuations,” continues Teyssier. In other words, these famous primordial lumps would really be at the origin of our universe's organization. Seeing Beyond the Limits
So do scientists now have a definitive answer as to how these structures formed? Not yet. As Guilaine Lagache, from IAS3
in Orsay explains, “these simulations account for the evolution of dark matter halos very well. But we still have to use various empirical recipes for the physics of ordinary matter, hiding the fact that we can't grasp the physics of the phenomena as a whole.”
For instance, astrophysicists still have to make much theoretical headway before being able to model the formation of some very “real objects,” like the giant galaxies that appear some three billion years after the Big Bang, too early to fit with the theory. As for Teyssier, he's already dreaming of a future generation of computer models which would enable us to recreate the entire visible universe, or even beyond. “These types of models could let us see much farther than any real sky observation instrument we design,” he concludes.
Echoes of the Big Bang
Information about the history of the universe comes to us via electromagnetic radiation in various forms such as visible light, infrared, radio waves and high-energy gamma rays.1 Scientists are now also hoping to be able to detect gravitational waves, distortions in space-time that move at the speed of light. Predicted by Einstein's theory of general relativity, they are believed to have been produced at the birth of the universe or during cataclysmic events such as the explosion of stars at the end of their life. “Unlike light, these waves interact very weakly with matter and travel through space without being significantly altered, whatever the medium they pass through,” explains Éric Gourgoulhon, from LUTH2 in Meudon. But this property also makes them very hard to detect. In the next few months, this will be the mission of the European detector VIRGO, currently being built near Pisa, Italy. “VIRGO should enable us to access many compact objects from which light never escapes, such as the cores of supernovae, or systems of black holes that orbit each other,” Gourgoulhon says. In ten years or so, the space interferometer LISA may detect very low frequency gravitational waves that would be undetectable on Earth. This would provide us with information emitted just after the Big Bang, enabling us to go back beyond the fossil radiation emitted 380,000 years after the birth of the universe.
1. High energy photons emitted during very violent events, such as the explosion of massive stars (supernovae).
2. Laboratoire de l'Univers et de ses théories (CNRS / Observatoire de Paris / Université Paris-VII).
Contact: Éric Gourgoulhon,
The Composition of the Universe
The universe is thought to be made up of 0.4% visible ordinary matter, 26.6% “hidden” dark matter (of which 3.6% is non-visible ordinary matter), and 73% dark energy. Dark energy is a mysterious force which has existed since the first instants of the universe, and which has recently been accelerating its expansion.
2. Astrophysique interactions multi-échelles (CNRS / CEA / Université Paris-VII).
3. Institut d'astrophysique spatiale (CNRS / Université Paris-XI).