Astronomers' understanding of the history of galaxy formation has, until now (2006), depended on observations using visible light. It now seems that these studies are missing 50% or more of the star formation in distant galaxies. Telescopes operating in the sub millimetre and far-infrared have now begun to uncover a hidden secret life of galaxies. And what they are finding is both new, and oddly familiar.
A decade ago, 1996 was a big year for observational cosmology, with two key results. The most famous was the Hubble Deep Field (HDF) - the most sensitive optical image ever taken, providing images of galaxies all the way to the earliest stages of the universe. By comparing the emission of galaxies in the HDF with their distance from us, it was possible to see how the star formation rate of the universe changed with the age of the universe. Comparison with similar measures in the local universe allowed the star formation history of the universe to be mapped for the first time. It revealed a peak in star formation when the universe was about 2 billion years old. This suggested that the long sought epoch of galaxy formation had been found.
The second discovery of 1996 came from the COBE satellite. It was designed to study the Cosmic Microwave Background (CMB), the dim echo of the heat of the Big Bang left behind by the beginning of the universe. However, COBE also had a second mission: to explore the large region of unexplored spectrum at far-infrared and submillimetre wavelengths. It was wholly unclear what secret history of the universe might be hidden at these wavelengths. COBE was thus looked in the far-infrared and sub-millimetre as well as at microwave wavelengths. It found something rather surprising, showing that there was as much energy coming from a Cosmic Infrared Background (CIB) as from all the distant optically detected galaxies seen in the HDF.
Whatever was making this infrared background was as significant in the history of energy generation in the universe as all the stars seen in all the HDF galaxies. What was it, and what did it have to say about the history of the universe?
Interstellar dust particles are small particles (about 1/10,000th of a millimetre across) scattered into space by supernova explosions and by the winds from old stars. Dust is mostly made from graphite and silicate grains coated with a variety of organic molecules. Just like the dust from pollution in our own atmosphere, interstellar dust absorbs optical light in such a way that longer, redder wavelengths are less affected than shorter, bluer wavelengths. This is the effect that can make the sun seem red at dusk as it sets behind the smoky atmosphere of a city. The energy of the absorbed light goes to heat up the dust particles, and they begin to emit electromagnetic radiation of their own. However, interstellar dust is much colder than the stars that heat them. Stars, at about 6000 degrees, emit mostly in the optical part of the spectrum. This is why our eyes function at these wavelengths. In contrast, interstellar dust, even when heated by young powerful stars, only reaches about 50 degrees above absolute zero. Its emission then peaks at much longer wavelengths than the sun, in the far-infrared or sub-millimetre part of the spectrum. A galaxy containing a lot of dust particles could be quite luminous in the far-infrared and sub-millimetre, but its optical emission would be suppressed by the dust absorbing optical light from stars. A specific class of galaxies, the so-called Ultraluminous Infrared Galaxies (ULIRGs), are very powerful far-infrared emitters. As much as 99% of their energy output can be in the far-IR and sub-millimetre. ULIRGs are among the most luminous objects in our neighbourhood, and generate power at a rate greater than a million, million (1012) suns. Their main power source is thought to be a massive burst of star formation, triggered by a collision between two or more galaxies. ULIRGs are very rare today but may have been more common when the universe was younger. Observations and simulations of local ULIRGs suggest they might be merging spiral galaxies transforming into elliptical galaxies. An increased number of ULIRGs in the early universe might explain the CIB, but if ULIRG-like objects are responsible for the CIB, what does this mean for the history of galaxy formation?
The first clear result is that the old, optical picture of the history of star formation in the universe is wrong. There is much more room for star formation in the first billion years of the universe than was originally thought, and the total amount of star formation has to be significantly greater. The details of the star formation history are still, quite literally, shrouded in dust, but it appears that the famous peak in star formation at an age of 2 billion years is probably not correct, and that the history of star formation stretches back to earlier ages of the universe. While the bad news is that we now have a less clear idea of the overall history of galaxy formation, one of the great mysteries might also have been solved. This is the origin of elliptical galaxies. These systems are full of old stars, and do not contain the gas and dust that are typical of a spiral galaxy. There is no ongoing star formation in them, and they appear as if they have burnt out all the fuel needed for star formation in some spectacular episode in their distant past. However, until the discovery of the CIB, nobody had seen any signs of any galactic conflagrations in the early universe that could have led to the current state of ellipticals. The discovery of the CIB might have solved all that. If ULIRG-like galaxies are responsible for the infrared background, then they could also be the progenitors of elliptical galaxies. The starbursts that power ULIRGs are triggered by galaxy collisions which turn spirals into elliptcials, and the ensuing star formation consumes all the dust and gas in the resulting system. We can now see a new picture of a dynamic early phase in galaxy evolution, with colliding galaxies triggering vast amounts of star formation shrouded in dust.
But if numerous dust enshrouded galaxies are responsible for the CIB, then how are we to find them?
Uncovering the Secret History
The key to finding CIB galaxies are observations in the far-infrared and sub-millimetre where the CIB dominates. Unfortunately, water vapour renders the Earth's atmosphere opaque at these wavelengths. Telescopes like the UK's James Clarke Maxwell Telescope (JCMT), perched on high, dry mountain sites, are able to peer through a few windows in this obscuration. The JCMT's sub-millimetre camera SCUBA was among the first instruments to start to see individual CIB objects, but has so far only managed to account for about1/5th of the CIB emission. Its successor, SCUBA2, and other telescopes now being constructed, should allow much more extensive studies of the CIB galaxies.
While ground-based telescopes can work in the sub-millimetre, satellites are needed for the far-infrared where the CIB is strongest. ISO was the first satellite to try to find individual CIB sources where it is brightest. The next few years, though, will see major advances for the far-infrared. The Spitzer satellite is currently flying and ASTRO-F was launched early in 2006. These will provide sensitive all-sky surveys and detailed studies of objects at the peak of the infrared background. The CIB, though, makes contributions throughout the far-infared and sub-millimetre. Two new ESA satellites, Planck and Herschel, due to be launched late 2007 or early 2008, will fill in the gap between far-infared satellites and ground-based sub-mm observatories, and should allow a detailed understanding of the objects. The future of far-infrared and sub-millimetre astronomy, and our chances of peering through the dust that obscures the secret life of galaxies, thus looks very bright.
Dr David Clements works with Spitzer and ASTRO-F data, and on preparations for the Herschel and Planck satellites at Imperial College London. He is also one of a small band of scientists (from various disciplines) who relates science to the SF community.Originally commissioned by Concatenation this article appeared in Astronomy Now in a slightly different form.
[Up: Science Index | Top: Concatenation]
[Updated: 06.5.10 | Contact | Copyright | Privacy]