'CT Scan' of Distant Universe Reveals Cosmic Web in 3D
October 16, 2014
|Questions and Answers||High-resolution Images|
On the largest scales, matter in the Universe is arranged in a vast network of filamentary structures known as the 'cosmic web', its tangled strands spanning hundreds of millions of light years. Dark matter, which emits no light, forms the backbone of this web, which is also suffused with primordial hydrogen gas left over from the Big Bang. Galaxies like our own Milky Way are embedded inside this web, but fill only a tiny fraction of its volume.
Now a team of astronomers led by Khee-Gan Lee, a post-doc at the Max Planck Institute for Astronomy, has managed to create a three-dimensional map of a large region of the far-flung cosmic web nearly 11 billion light years away, when the Universe was just a quarter of its current age. Similar to a medical CT scan, which reconstructs a three-dimensional image of the human body from the X-rays passing through a patient, Lee and his colleagues reconstructed their map from the light of distant background galaxies passing through the cosmic web's hydrogen gas.
The use of the combined starlight of background galaxies for this purpose had been thought to be impossible with current telescopes – until Lee carried out calculations that suggested otherwise. Lee says: "I was surprised to find that existing large telescopes should already be able to collect sufficient light from these faint galaxies to map the foreground absorption, albeit at a lower resolution than would be feasible with future telescopes. Still, this would provide an unprecedented view of the cosmic web which has never been mapped at such vast distances."
Lee and his colleagues obtained observing time on one of the largest telescopes in the world: the 10m-diameter Keck I telescope at the W. M. Keck Observatory on Mauna Kea, Hawaii – but were plagued by a problem more terrestrial than cosmic. "We were pretty disappointed as the weather was terrible and we only managed to collect a few hours of good data. But judging by the data quality as it came off the telescope, it was already clear to me that the experiment was going to work," says Joseph Hennawi (MPIA), who was part of the observing team.
Although the astronomers only observed for 4 hours, the data they collected was completely unprecedented. Their absorption measurements using 24 faint background galaxies provided sufficient coverage of a small patch of the sky to be combined into a 3D map of the foreground cosmic web. A crucial element was the computer algorithm used to create the 3D map: due to the large amount of data, a naïve implementation of the map-making procedure would have required an inordinate amount of computing time. Fortunately, team members Casey Stark and Martin White (UC Berkeley and Lawrence Berkeley National Lab) devised a new fast algorithm that could create the map within minutes. "We realized we could simplify the computations by tailoring them to this particular problem, and thus use much less memory. A calculation that previously required a supercomputer now runs on a laptop", says Stark.
The resulting map of hydrogen absorption reveals a three-dimensional section of the universe 11 billions light years away – the first time the cosmic web has been mapped at such a vast distance. Since observing to such immense distances is also looking back in time, the map reveals the early stages of cosmic structure formation when the Universe was only a quarter of its current age, during an era when the galaxies were undergoing a major 'growth spurt'. The map provides a tantalizing glimpse of giant filamentary structures extending across millions of light years, and paves the way for more extensive studies that should reveal not only the structure of the cosmic web, but also details of its function – the ways that pristine gas is funneled along the web into galaxies, providing the raw material for the growth of galaxies through the formation of stars and planets.
The work described here will be published as K.G. Lee et al., "Lyα Forest Tomography from Background Galaxies: The first Megaparsec-Resolution Large-Scale Structure Map at z > 2" in the Astrophysical Journal Letters.
• ADS entry of the article
The team members are Khee-Gan Lee, Joseph F. Hennawi, and Anna-Christina Eilers (Max Planck Institute for Astronomy), Casey Stark and Martin White, (UC Berkeley and Lawrence Berkeley National Laboratory), J. Xavier Prochaska (University of California at Santa Cruz, Lick Observatory), David Schlegel (Lawrence Berkeley National Laboratory), and Andreu Arinyo-i-Prats (Universitat de Barcelona).
This research received financial support from the National Geographic Society/Waitt Grants Program.
What is new/important about the result?
This is the first time the cosmic web has been mapped at such a large distance of 11 billion light years away. Also, it is the first time that distant galaxies have been used as a 'backlight' to study foreground hydrogen 'Lyman-alpha forest' absorption, which has previously required bright quasars in the background.
The current measurements also serve as a successful feasibility study for a much more extensive survey. The elongated shape of the map is due merely to the fact that the observations were limited by bad weather – the astronomers simply did not have sufficient observing time to cover a larger patch of sky, and thus a larger volume of space. The technique itself is remarkably efficient, and it would not take long to obtain enough data to cover volumes hundreds of millions of light years on a side. This is the goal of the more extensive CLAMATO survey (COSMOS Lyman-Alpha Mapping And Tomography Observations), which is aimed at creating a tomographic map covering one square degree (=50%) of the Cosmic Evolution Survey (COSMOS) field, one of the most extensively studied regions in the sky.
From this much larger map – the next step on the agenda – the astronomers hope to glean insights not only about the structure of the adolescent cosmic web, but also about its function. During that period of cosmic history, galaxies underwent rapid growth; the cosmic web plays a crucial role in funneling gas, the raw material of star formation, onto the galaxies. Previous studies have been able to map the cosmic web in the immediate vicinity of a bright quasar, showing the inflow of pristine cosmic gas (cf. MPIA Science Release 2014-01) onto galaxies; a larger-scale map would enable astronomers to explore such processes in 3D, follow the flows over longer distances and to reconstruct the cosmic logistics of star formation on a large scale.
Furthermore, large tomographic maps would allow astronomers to study the roots of cosmic inequality: One of the most striking features of the present-day Universe is the dichotomy between 'rich' massive clusters of galaxies and the huge surrounding voids that are almost completely starved of matter. The 3D maps from the new technique will allow astronomers to identify proto-clusters and proto-voids in the distant Universe, at a time when the universe was a quarter of its current age, or less. Such maps would provide a unique historical record on how the galaxy clusters and voids grew from the tiny inhomogeneities nascent in the Big Bang.
How does "Lyman-alpha forest tomography" work?
The cosmic web is as elusive as it is important: dark matter is invisible, and the embedded hydrogen gas is so rarefied that it emits little light. For decades, astronomers have resorted to indirect detection with the help of quasars. Quasars are the nuclei of distant galaxies which, powered by the infall of matter onto a supermassive black hole, shine as the brightest objects in the Universe. As their light travels towards Earth, it encounters the rarefied cosmic gas filling the void between galaxies, and some of the light is absorbed. Crucially, this absorption occurs at very specific frequencies. When astronomers on Earth split a quasar's light, rainbow-like, into its different component colors, the absorption creates a characteristic pattern in the spectrum, narrow darker regions which astronomers call absorption lines. The pattern and shape of these lines allow astronomers to study the distribution of the absorbing gas.
Since our universe is expanding, the location of a gas cloud's absorption lines within the spectrum depends on the cloud's distance from Earth ("cosmological redshift"). This is because the quasar light stretches as it travels toward Earth, thus as the light passes through various gas clouds the absorption signature is imprinted on a different region of the quasar's spectrum. The quasar light thus bears the imprint of all the clouds it encountered; for each cloud, the position of the absorption line in the spectrum contains information about the cloud's distance from Earth. The most prominent of these absorption patterns is caused by the Lyman-alpha absorption line of hydrogen gas. Collectively, the pattern of Lyman-alpha lines, each associated with a different cloud, is known as the "Lyman-alpha forest".
But bright quasars are very rare, and hence sparse on the sky. In consequence, the sightlines where the cosmic web can be traced using quasar light are so few and far between that they do not provide nearly enough information to construct a three-dimensional map.
For this reason, astronomers considered using the combined starlight of distant galaxies as an alternative light source. Galaxies are nearly 100 times more numerous than quasars. If they could be used as backlights for absorption studies, this would enable a high-fidelity 3D tomographic mapping of cosmic structure – similar to computer tomography (CT) methods used in medical imaging. The snag was that galaxies are about 15 times fainter than quasars. The prevailing view was that this is simply not bright enough for the experiment to be feasible, and it appeared that the next generation of gargantuan telescopes, such as the European Extremely Large Telescope (E-ELT) with a 40m mirror diameter, would be required to construct 3D maps of the cosmic web. The results presented here show that, on the contrary, this kind of map is feasible with existing telescopes and instruments.
What is the location and size of the region that has been mapped?
The observation targeted part of the COSMOS field in the constellation Sextans, which because it has been targeted repeatedly with multiple instruments and telescopes, provides a very detailed view of the distant cosmos far beyond our own home galaxy. Light from the 3D region that was mapped takes 11 billion years to reach Earth; consequently, we see this region as it was 11 billion years ago, less than 3 billion years after the big bang (red-shift: z = 2.3). At the time of the snapshot, the region that was mapped measured roughly 2.5 million by 6 million by 100 million light years across (the latter value describes the radial extent: the most distant parts of this region are 100 million years more distant from us than the nearest parts). Since then, cosmic expansion has made all its length scales expand by a factor of 3.3.
Hasn't the cosmic web already been mapped?
Yes, but never before at such a large distance. Previous maps of the cosmic web had relied on using the positions of individual galaxies as tracers – map a sufficient number of such galaxy position, and you should be able to perceive the shape of the underlying filaments of the cosmic map. But the only existing maps covering large volumes have been confined to the local universe (mapping galaxies up to distances of 3 billion light years, z = 0.3), while smaller maps have covered distances of up to 7 billion light years (z = 1). At even larger distances, where the galaxies are fainter, maps of this kind – one data point per galaxy – are far too sparse and much too inefficient to produce a high-fidelity map. In contrast, the Lyman-alpha absorption technique described here – one line-of-sight per galaxy – proves much more efficient, since it continuously probes the distribution of the cosmic web over hundreds of millions of light years in front of each galaxy, allowing an efficient way of mapping the Universe even at this vast distance.
Which instruments and telescopes were used in these results?
The observations were carried out with the Low-Resolution Imaging Spectrograph (LRIS) on the 10m-diameter Keck I telescope at the Keck Observatory on Mauna Kea, Hawaii.