M51, also known as the Whirlpool Galaxy, is one of the brightest spiral galaxies in the night sky. The finely detailed spiral structure is thought to be the result of an interactions with its close galactic neighbor, NGC 5195.
Credit: NASA/ESA
A FEW YEARS AGO Avishai Dekel gave up chess in favour of mud wrestling. Dekel is a cosmologist and he isn’t known to frequent strip clubs.
But there are two types of cosmologists: those who study fundamentals, like the initial conditions and content of the early universe, and those who immerse themselves in the messier problem of the evolution of galaxies, replete with gas and stars that heat and cool, form jets, make black holes, and sometimes explode.
Martin Rees of the University of Cambridge in England calls the two classes of cosmologists chess players and mud wrestlers. Cosmology is “a fundamental science just as particle physics is,” says Rees. “The first million years [of the universe] is described by a few parameters ... but the cosmic environment of galaxies and clusters is now messy and complex.”
Now that the chess players have established those basic parameters – such as the relative amounts of invisible dark matter, the even-more mysterious dark energy, and ordinary matter – more cosmologists are turning their minds to mud. Recent surveys of the shapes, colours, and masses of galaxies have put a new focus on the nitty-gritty of galaxy formation.
“Now that we know the cosmological parameters, it’s really time to understand how galaxies form,” says Dekel, of Israel’s Hebrew University of Jerusalem. To do that, “we have to trace the gas”, not dark matter, because it’s the gas that forms stars. “That’s where the action is.”
The physics governing large-scale gas interactions, or ‘gastrophysics’, is much more complicated than that of dark matter. Gas molecules respond to a host of forces, while dark matter is simple to model because it responds predominantly to just one force: gravity. Nonetheless, says Dekel, he is a recent convert to gastrophysics.
Through the 1980s and 1990s, Dekel spent most of his time trying to estimate the density of matter in the universe by mapping the velocities at which galaxies and matter move through the vast, invisible reaches of dark matter. Although no one knows what dark matter is made of, it appears to constitute 85% of the mass of the universe.
And simply because there’s so much of it, the stuff provides the gravitational scaffolding that pulls together ordinary gas – electrons, protons, atoms, etc. – to make stars and galaxies.
The behaviour of dark matter has thus been considered a reliable map for the path of galaxy formation. Every galaxy is nestled within a halo of cold dark matter, composed of exotic particles that move much slower than the speed of light (this relatively slow pace is why this dark matter is dubbed ‘cold’).
“The universe started with lots of dark matter everywhere, so gravity caused bits to fall together. Bigger clumps attracted even more dark matter,” says Joss Bland-Hawthorn of the Sydney Institute for Astronomy at the University of Sydney. “The fraction of the mass in the form of gas is crucial [for galaxies to form]. Some halos may have little gas, others have a great deal more. We think the maximum gas fraction is [about] 17% of the dark halo mass.”
The halos start out small but continually merge to grow bigger, dictating that all structure in the universe should evolve in the same way – from little to big. The growing clumps of dark matter form the backbone of a cosmic web, with clusters and superclusters of galaxies falling into place along the densest filaments, like paint onto a dark canvas.
There is some debate about whether galaxies grow largely by merging with one another, or if fresh gas is primarily drawn to galaxies along the filaments of this vast web of dark matter (see “Big babies in a cosmic web”).
Nevertheless, on the largest scales in the universe, dark matter accounts amazingly well for galactic structure, such as where and how galaxies concentrate, says Piero Madau of the University of California at Santa Cruz, in the USA.
But in 2003, Dekel and others became intrigued by a discovery they believe dark matter alone cannot explain. Astronomers have known since the 1920s that the modern-day universe consists mainly of two galaxy types: disc-shaped spirals such as the Milky Way; and football-shaped ellipticals.
Ellipticals have a reddish tinge – an indication that they are old and finished forming stars long ago – while spirals have a bluish tinge, a sign of recent star formation.
Several years ago, researchers found that in the universe today, these two populations divide sharply by weight. An analysis of the Sloan Digital Sky Survey, which has recorded more than a million nearby galaxies of the northern sky, revealed that the ‘red and dead’ ellipticals nearly always tip the scales at masses greater than the Milky Way, while the starforming spirals fall below that weight.
SOMEHOW, STAR BIRTH was systematically and dramatically quenched in the big guys, but proceeded unimpeded in the spiral small fry. The puzzle deepened in 2005 when Sandy Faber also at the University of California at Santa Cruz and her colleagues announced that they found the same galactic dichotomy when the universe was just seven billion years old – half its current age.
Faber’s team used a spectrometer she designed for the Keck Observatory atop Hawaii’s Mauna Kea to measure the mass of distant galaxies, as part of a survey into what composed the universe at seven billion years of age. She reviewed the results of the survey, known as Deep-2, in a January 2008 meeting of the American Astronomical Society in Austin, Texas.
At first glance, the dichotomy would seem to conflict with cold dark matter theory. A preponderance of ‘red and dead’ massive galaxies early in the universe might indicate that halos can start out as giants and then break apart into smaller bodies – the opposite trend of what dark matter would produce.
Dekel and his colleagues, including Yuval Birnboim, now at the Harvard-Smithsonian Centre for Astrophysics in Boston have an explanation that would fit with cold dark matter theory – but it requires combining gastrophysics with dark matter.
Gas pulled inside a dark-matter halo would normally fall into the centre, where it would cool and grow dense enough to make stars. But as the universe ages, dark-matter halos merge and grow more massive, some becoming greater than about a trillion times the mass of our Sun.
When a halo reaches this critical value, the stage is set for a galactic divide, say Birnboim and Dekel. Their calculations show that the infalling gas rams into the cold, stationary gas already at the halo’s centre. The collision creates a long-lasting shock that heats the cold gas, causing it to exert a pressure.
That pressure pushes on infalling gas, hurling the material back to the halo’s outskirts, where it remains like some exile in galactic Siberia, unable to coalesce and make more stars.
As long as the material in the central part of the halo maintains its outward pressure, the supply of fresh gas is choked off, and the galaxy can no longer make stars. Over time, the massive galaxy growing inside the halo’s centre, once a hotbed of star birth, becomes red and dead.
Halos that remain less massive – and which therefore beget smaller galaxies –can’t forge such long-lasting shocks. Gas continues to stream unimpeded into the central region, enabling the birth of new generations of stars.
Simulations from several other groups – including those led by Dusan Keres, now at the Harvard-Smithsonian Centre for Astrophysics; Darren Croton, now at the Swinburne University in Melbourne; Richard Bower of Durham University in England; and Andrea Cattaneo, now at the University of Potsdam in Germany – have come up with similar findings.
“The idea is that big, central galaxies are quenched before [the universe is seven billion years old] because they are in massive halos ... while smaller galaxies are quenched later, if at all, when their parent halos reach the critical mass,” says Dekel.
One remaining puzzle, notes Dekel, is how gas within the centre of a massive halo can maintain, for up to 10 billion years of cosmic history, the outward pressure that keeps new gas at bay in the outer halo. He calculates that the pressure might last for only one-tenth that time. Some other source must keep star birth from turning back on.
Again delving into gastrophysics, he and other researchers point to the unusual role that black holes may play in staving off star birth in massive galaxies. Researchers now believe that every massive galaxy houses a central, heavyweight black hole, and that these gravitational monsters wield influence far beyond their immediate surroundings.
Packing the equivalent of billions of Suns into a volume no bigger than our Solar System, black holes don’t just pull matter in. Energy from the gas and stars spiralling into the hole also creates jets of matter stretching out a million light-years from the centre. In this way, a black hole can act to keep star forming gas far from the galaxy and this will regulate or even switch off star formation.
Moreover, researchers have found that black holes at galactic centres grow in lock step with the mass of stars in that galaxy’s hub: the holes always seem to be just one 500th the mass of those stars. That prescription means that the most massive galaxies house the heaviest black holes – also the most likely ones to have jets strong enough to interrupt star formation.
“What’s truly amazing is how tight the correlation seems to be” between the mass of a central supermassive black hole and a surrounding galaxy, says Tim Heckman of Johns Hopkins University in Baltimore, Maryland.
“I don’t think prior to 10 years ago you would have found one astronomer in 1,000 who thought black holes had some fundamental part in the formation of galaxies. We still don’t know whether a black hole dictates the formation of a galaxy, or the other way around.”
Dekel and Birnboim, along with Jerry Ostriker of Princeton University in New Jersey, recently began entertaining the idea that black holes might not be needed to explain the galactic divide after all. According to their calculations, the heat produced by gas falling into the centres of massive dark-matter halos might be enough to quench the supply of cold, star-forming gas.
But not everyone is convinced there is much of a mystery about the two types of galaxies. It’s simple, says Joss Bland-Hawthorn: “Elliptical galaxies form in very dense regions of the universe where a great deal of dark matter has collected. Spirals form in less dense regions.”
He argues that the ellipticals – which he describes as “train wrecks” where many smaller galaxies have collided – are old enough to have used up most of their gas – even the ones surveyed half-way through the life of the universe. “They formed a long time ago, and most of the gas got used up at that time to form stars … the remaining gas was heated and pushed out by large-scale galactic winds or swept out as galaxies moved in clusters.
“People tend to exaggerate the difficulty of forming these galaxies,” he added, arguing that large ellipticals formed by mergers were even present when the universe was just three to four billion years old.
A NEW STUDY GOES further back in time than ever before to probe the difference between galaxy types. Using distant quasars as searchlights, a team led by Art Wolfe of the University of California at San Diego says its search may have reached back to the era when massive galaxies were still forming stars – before the death knell was sounded for these heavyweights.
During their five-year study, Wolfe and his colleagues, including J. Xavier Prochaska of the University of California at Santa Cruz, used spectrometers at the Keck Observatory to study star formation in 143 dense gas clouds, each pierced by radiation from a different quasar. Astronomers generally agree that these clouds, known as damped Lyman-alpha systems, are the likely predecessors of modern-day galaxies.
They reveal what those galaxies were like when the universe was only about two billion years old. To assess the star-formation rate in the clouds, the team homed in on the abundance of carbon atoms stripped of a single electron.
Newborn stars readily excite these carbon ions: the higher their abundance, the higher the star formation rate. The team used spectra of another ion, silicon stripped of one electron, to indicate the masses of the dark-matter halos in which the dense clouds reside.
To the surprise of the researchers, the study confirmed that star birth was highest in those clouds that lie within the heaviest dark-matter halos. Those clouds are the likely progenitors of the most massive galaxies today, the team says in a 2008 Astrophysical Journal study.
That scenario contrasts with the current universe, “where [massive galaxies] exhibit little, if any, star formation,” says Wolfe. “But that’s just what the Dekel-Birnboim model predicts. That far back [in time], the high-mass galaxies are still forming stars at a high rate.”
Moreover, observations of distant galaxies by several researchers, including Chuck Steidel of the California Institute of Technology, also show that stars once formed at a feverish rate in massive galaxies. “We go back far enough to see the star-forming phase of the high-mass systems,” says Wolfe. It’s only later that star birth shuts down in the high-mass systems, a victim of overheated gas and possible interference by monster black holes, he notes.
Dekel, in the meantime, says he hasn’t entirely abandoned his interest in investigating the fundamental properties of the universe. It’s just that the evolution of galaxies provides such a messy, and thus intriguing, canvas for testing his ideas. “I see myself as a chess player who has waded into the mud,” he notes. “And that’s where all the fun is.”
BIG BABIES IN A COSMIC WEB
Dark matter theory dictates that galaxies start out small, when ordinary matter – in the form of gas and stars – gathers into clumps along filaments of dark matter. Galaxies eventually eat up other galaxies, and this process – known as hierarchical merging – has been credited with creating massive galaxies. It is supposed to be slow and gradual.
But a new study from Liverpool John Moores University in England may throw a cosmic spanner in the idea that galaxies get big through mergers. In Nature in April 2009, the British researchers detailed the discovery of galaxies that are nine billion years old – and therefore created only four billion years or so after the Big Bang. Their size? Five times bigger than they should be.
“By analysing the light from these remote galaxies, we have effectively weighed them. They are nine billion years old, but as the Big Bang took place 13.7 billion years ago, they are relatively young and we would have expected them to be much smaller,” said study author Chris Collins. If these galaxies were not formed by gravitational pull, which requires more time, then “the findings therefore suggest a different picture in galaxy formation,” he added.
As the rate at which gas can be sucked into galaxies to create stars has a limit – too fast and the heat and force creates a shockwave that prevents fresh gas from entering – the find has presented something of a mystery. How did these galactic youths swell so furiously?
It is not the only study that seems to show galaxies growing faster than was thought possible without mergers, either – a German team has observed other early galaxies with surprising rates of star formation. “The large galaxies, as they appear in this early stage, indeed created stars at a very rapid rate, but this does not appear to be at all a result of galactic mergers,” says Avishai Dekel of Hebrew University of Jerusalem.
Dekel’s own study, published in Nature in January 2009, might provide a solution to the problem. Instead of the conventional idea that large galaxies form mostly by mergers, his team’s new theoretical model proposes that gas is ‘pumped’ to galaxies through a network of filaments of dark matter that feed galactic halos. This ‘cosmic web’ would allow the gaseous fuel for star formation to be fed right into galaxies, without causing them to burn themselves out at an early stage.
Others disagree with the idea that the galaxies discovered by Collins’ group are bigger than they should be. Joss Bland-Hawthorn, an astrophysicist at the University of Sydney, argues that large elliptical galaxies were present earlier than many think and the merger of smaller galaxies is an adequate explanation for them without the need for more exotic theories. “In any case, when you’re looking at stars that are so far away, it’s difficult to see clearly what’s going on,” he says. — John Pickrell
Follow Cosmos on Twitter! twitter.com/cosmosmagazine
Ron Cowen is a science writer in Washington DC, whose articles have appeared in magazines including Science News and National Geographic. Additional reporting by John Pickrell.
