Australasian Science Magazine

These frames simulate the evolution of large-scale structures in the universe, including galaxy clusters and cosmic filaments. The frames show the evolution of structures from 140 million light years ago (left) to the present day (right). Simulations performed at the National Center for Supercomputer Applications by Andrey Kravtsov (University of Chicago) and Anatoly Klypin (New Mexico State University). Visualisations by Andrey Kravtsov

By Jasmina Lazendic-Galloway

Cosmic filaments are the largest structures in the universe, and are the most likely places where the universe’s missing matter resides.

We live on a planet that is 13,000 km in diameter, orbits around a star that is 1.4 million km in diameter, and resides in a galaxy that measures 100,000 light years in diameter but only 1000 light years in height. The closest galaxy to us is the Canis Major dwarf galaxy, which was discovered in 2003 and is located about 25,000 light years from our solar system and 42,000 light years from the centre of our galaxy.

In the 1930s Harlow Shapley discovered that individual galaxies are grouped into clusters. In the following 30 years, thousands more galaxy clusters were discovered by George Abell and Fritz Zwicky. Abell suggested that clusters of galaxies are “fundamental condensations of matter in space” that can be used to investigate the formation and evolution of the universe. Clusters can contain as many as 10,000 galaxies, or as little as 30 in our “local group”.

Clusters of galaxies are part of the visible universe of stars, gas and dust. However, this visible (or baryonic) part of the universe makes up only 5% of the universe’s total mass–energy budget. The rest resides in dark matter (23%) and dark energy (72%) – intangible forms of matter and energy that have not yet been directly observed.

The discovery of clusters of galaxies in the 1930s prompted Zwicky to introduced the idea of dark matter to account for the missing baryonic mass needed to explain the orbital motions of the galaxies in clusters.

However, by comparing the baryonic mass budget from optical observations of the distant (i.e. younger) universe with that of today’s so-called local universe, it was found that some of this visible baryonic matter is missing in the local universe.

What has happened to it? How did we lose it? It turns out that all of the baryonic matter should still be there, but it has become too diffuse to be detected easily.

In between the clusters of galaxies there are vast and seemingly empty voids. They are very low in density, possibly as low as one hydrogen atom per cubic metre. These voids are delineated by filaments stretching between the clusters of galaxies. These filaments are thought to be 10–100 times denser than the voids.

The galaxies themselves contain only about 10% of the baryonic mass. The rest is expected to reside in the filaments stretching between them. In comparison, 99.86% of the total mass of our solar system resides in the Sun.

Detection of the CfA2 Great Wall in 1989 has shed light on just how large these structures can be. The CfA2 Great Wall is a filament of galaxies approximately 500 million light-years long, 300 million light-years wide and 15 million light-years thick. By comparison, our local group extends 10 million light-years in diameter, and is part of the Virgo Supercluster (containing more than 100 galaxy groups and clusters) that extends about 110 million light-years in diameter.

How Do We Find Cosmic Filaments?
Cosmic filaments are made up of intergalactic medium and individual galaxies. These galaxies flow along their filaments like cosmic highways towards clusters of galaxies as a result of gravitational attraction.

The hottest part of the intergalactic medium in filaments is called warm–hot interstellar medium because it is expected that the medium, consisting mostly of ionised hydrogen and helium, is heated to warm-to-hot temperatures of 105–107 °K. At these energies the gas emits mostly in X-ray wavelengths. In addition, emissions from highly ionised atomic species are also expected, mostly from ionised oxygen.

However, observational evidence for the existence of diffuse gas in cosmic filaments has come mostly from observations at ultraviolet wavelengths emitted by distant, bright objects like quasars. These galaxies from the early universe beam through the filaments like a flashlight beams through fog.

Such observations, however, are only capturing the smaller, cooler component of cosmic filaments, and may suffer confusion from other sources like galactic superwinds.

Cosmic filaments have been detected at optical wavelengths as strings of galaxies. While they can provide hints of their distribution, such observations are not able to probe the hot, diffuse gas that is expected inside the filaments.

Optical observations are sensitive to temperatures between 104–105 °K. Since the intergalactic medium was at similar temperatures in the earlier epochs of the universe, optical observations of the early universe could detect the matter that accounts for the full baryonic mass budget.

However, during the process of large-scale structure formation, as the matter condensed into different forms, gravitational shocks would have heated this medium to X-ray energies. Therefore, the large baryonic component is no longer observable in optical wavelengths in the local universe.

Furthermore, the X-ray emissions expected from cosmic filaments are not easy to detect since they are very diffuse. As a direct comparison supernova remnants, which have very similar physical properties to cosmic filaments but are much easier to detect in X-rays, are denser than the filaments on average by a factor of 10,000.

Astronomers have been trying to overcome this problem by employing different techniques. One of the first systematic searches for X-ray emissions from cosmic filaments in 1995 examined the space between the 50 galaxy clusters but did not detect any emissions, even when the observations from all the regions were averaged together. Most of the other X-ray searches have been targeted closer to the galaxy clusters, where hot, diffuse gas from the filaments might be denser and thus easier to detect.

Since the X-ray emissions of galaxy clusters are hotter (10^8 °K) than the filaments themselves, the various searches have been finding an “excess” of the lower energy X-ray photons (10^5–10^7 °K), that match the signatures of cosmic filaments. There are few claims for the detection of filaments in individual galaxy clusters, but it is still not clear if these are unusually denser regions or if they are representative of common cosmic filaments.

For example, the European Space Agency’s XMM-Newton X-ray Space Observatory detected a cosmic filament between two galaxy clusters, Abell 222 and Abell 223, located about 2.3 billion light-years away. The filament can be clearly seen as a bridge connecting the clusters, and its detection was credited to the favourable alignment of the filament, which concentrated X-ray emissions to a relatively small part of the sky. The filament was found to have a temperature of 10^7 °K and a density of 100 particles per cubic metre (an over-density of 100 in respect to the voids). Thus, the X-ray observations have traced the densest and hottest regions of that filament.

Why Are Cosmic Filaments Important?
Most of what we know about cosmic filaments comes from numerical simulations, and there are many things we still need to understand. Are filaments formed first and then fragment into clusters, or is the other way around? How will filaments evolve? Will they flow into the clusters? And most important of all, can we observationally prove that most of the missing baryonic matter in the local universe does reside in cosmic filaments?

Recently our team detected X-ray emissions from filaments towards 41 galaxy cluster pairs using archival observations by the ROSAT X-ray Observatory, which was operational between 1990 and 1999. While a couple of state-of-the-art X-ray satellites have been launched since then, the ROSAT data are still relevant for X-ray astronomers. ROSAT was designed for the study of faint extended astronomical sources, and it has surveyed the whole sky, making X-ray maps available for anyone to use.

Using these archival sky maps, we searched for X-ray emissions between galaxy cluster pairs using known optical positions of their filaments obtained from the Anglo-Australian Observatory’s 2dF Galaxy Redshift Survey.

We did not find X-ray emissions from individual filaments, which was not too surprising – ROSAT survey maps have been obtained in too brief a time interval to be sensitive to such faint and diffuse structures as cosmic filaments. Instead, we averaged the signal from all the filaments, assuming that they have more or less the same properties.

Indeed we detected a signal, implying that the average density of all the filaments we examined is slightly more than 100 particles per cubic metre, which agrees with theoretical predictions.

Our work is not the first detection of filaments in X-rays, but it is the first detection that includes a large number of sources, and as such has confirmed that a significant fraction of cosmic filaments are indeed made up of hot, diffuse gas and that the missing baryonic matter is very likely hidden there.

Although we now have detected a few cosmic filaments, most of their physical properties are still derived from numerical simulations, and models tend to disagree about the exact proportion of the baryonic mass budget is contained in them. We need to obtain high-quality data and measure the physical properties of these structures in detail to understand their origin and evolution.

Numerical simulations show that filaments of dark matter are formed along these visible filaments, and thus understanding them better will be significant for understanding the evolution of the whole universe.

While there is still much to be done with the current space- and ground-based telescopes, upcoming missions will offer even more sensitive tools for studying these elusive fibres of the universe.