Chasing Shadows!


Quantum Scribe
Chasing shadows 21 Apr 01

The idea that invisible galaxies haunt the Universe has got astronomers peering into the darkness. But what they find may dismay them, says Stuart Clark

GHOSTS gather in the shadows. Just outside the cosy circle of our own galaxy-which is lit by the fires of a hundred billion stars-is a host of wraith-like galaxies made of almost nothing but exotic invisible matter. Instead of shining like celestial beacons across the Universe, they are virtually indistinguishable from the blackness of space.

"Dark galaxies might outnumber normal galaxies by a hundred to one," says Neil Trentham of Cambridge University. Trentham, like most astronomers, believes dark galaxies must be out there somewhere. He is attempting the almost impossible task of trying to see these sinister clouds of darkness.

But other astronomers warn that the search is a waste of time. Like ghosts, dark galaxies may be a figment of the imagination. If so, our theories of how galaxies form are wrong, and we may have to change our ideas about what makes up most of the matter in the Universe, or rewrite the story of the Universe's first moments.

Galaxies are thought to have formed from the sea of gas left behind by the big bang. If some patches of gas were slightly denser than others, their gravity would have pulled in surrounding material.

But with the gravity of ordinary gas alone, this process would have taken far too long-galaxies would still be forming even today. So cosmologists have been forced to assume that there is also a lot of invisible matter in the Universe, outweighing the normal matter many times. This "dark matter" can also explain how galaxies spin so fast without breaking apart. A large spherical halo of dark matter surrounding each galaxy could provide enough gravity to balance the spin, gluing the galaxy together.

Physicists' attempts to merge the fundamental forces of nature have thrown up any number of candidates for the stuff of dark matter (New Scientist,16 January 1999, p 24), called weakly interacting massive particles, or WIMPS. These hypothetical exotic particles are supposed to have been formed along with normal matter in the furnace of the big bang. They are invisible because they don't feel all the forces that ordinary matter does, and light just ignores them. Astrophysicists usually assume that these dark matter particles are "cold"-that is, fairly heavy and slow moving, so they tend to clump together.

According to the conventional theory, this clumpiness allowed cold dark matter to give birth to galaxies. Way back in what astronomers call the dark ages, when the Universe was a mere billion or so years old, cold dark matter gathered itself under gravity into giant blobs called haloes. These then attracted normal matter to form stars, turning into bona fide galaxies.

The problem is that in its simplest form, this theory says there should be an awful lot of little galaxies-the failed relics of galaxy formation that never managed to grow into giant elliptical galaxies or majestic spirals like the Milky Way.

Astronomers already know of two species of galactic minnow. The small round galaxies called dwarf spheroidals and their untidier cousins, the dwarf irregulars, both weigh in at about ten million times the mass of the Sun, or only one ten thousandth that of the Milky Way. But there are far too few of them to agree with the theoretical models of cold dark matter, which predict ten or a hundred times as many.

To save the theory, astrophysicists assume that these small galaxies do exist-only they're invisible. Somehow, most of the smaller dark-matter haloes must have been unable to form stars. "The smallest dwarf galaxies could be the very rare, one-in-a-hundred cases which, for some reason, do form stars," says Trentham.

So what stops stars forming in the remainder? Here, dark-galaxy pundits split into two camps. Either something stops the gas entering the dark matter halo, or it falls in and then is somehow prevented from making stars.

Trentham believes that dark galaxies failed to attract any normal matter because they missed out on a feeding frenzy during the dark ages, when gas was plentiful. "To produce normal galaxies, dark matter haloes grow in the early Universe, pulling in more dark matter and gas," he says. And according to his simulations, still unpublished, the smaller haloes would be celestial late developers. "Imagine a small dark halo growing five to ten billion years after the big bang. The gas between the galaxies has been heated up by the light from stars, and it is now moving so fast that it cannot be pulled in."

Others are not convinced. Frazer Pearce of Durham University believes that even long after the dark ages, there was plenty of cold gas around to be captured. Astronomers see traces of dark gas clouds throughout intergalactic space, revealed because they absorb some of the light from the distant celestial beacons known as quasars. And these clouds are relatively cool.

"The gas is at 20,000 K," says Pearce. And that's too cool to escape even a modest gravitational pull. "Even when you have gas at tens of millions of kelvin, you can't keep it out of haloes forever."

Pearce thinks that these intergalactic clouds are already sitting inside their own very small haloes of dark matter. So what stops them from forming stars?

The tiniest galaxies might suffer a kind of boom-bust economy, he suggests. When the small dark haloes form, they attract some gas and stars form. The most massive of these burn quickly and explode as supernovae after just a few million years. That heats the remaining gas and flings it in all directions. A few supernovae could make the gas hot enough to eject it back into intergalactic space. "We are hoping that supernovae will blow these clouds to pieces and stop stars forming," says Pearce.

The hot, scattered gas will eventually radiate its excess energy, slow down and get pulled in again by the halo, beginning another boom-bust cycle. Each short period of intense star formation, lasting a few tens of millions of years, would be followed by a dormant period of up to a billion years.

But once again, there are dissenters. "The idea that it is really easy to blow the gas out of galaxies is problematic," says George Lake of the University of Washington. "People have tried to do simulations, overestimating the effect of supernovae, and they still can't get the stuff flung out. I think the whole idea is misguided."

Lake thinks dark galaxies are a mirage. "It is not that these dark galaxies lie below current detection limits, they are just not there. They do not exist at all."

The only way to settle the argument is to look for these spectral galaxies. Astronomers have a few ideas about how to catch them (see "Ghost hunting"), but what if, after the trawling is over, they return with empty nets?

Warm dark matter

Pearce believes that the conventional theory of galaxy formation could be repaired. Instead of cold dark matter, the Universe may be filled with another, slightly different strange substance.

The lighter a particle is, the faster it is likely to be moving. And fast-moving particles are much less likely to clump together. Very lightweight particles such as neutrinos would be a kind of "hot dark matter". These zippy particles would be so resistant to clumping that they would tend to smooth out structures even on the scale of large galaxies-so they can't make up most of the dark matter in the Universe, or we wouldn't be here.

Instead, Pearce thinks a dearth of little dark galaxies could be explained by "warm dark matter", of an intermediate mass and speed. Warm dark matter would happily form big clumps that make ordinary galaxies, but would move too fast to be captured by the weak gravity of small haloes.

The mass of a warm dark matter particle would need to be around 10-33 kilograms-a millionth the mass of a proton. This presents no problem for particle physicists, who can tweak their speculative theories to produce particles of virtually any mass required.

But Lake thinks this running repair is useless. He points out that the conventional theory of structure formation also predicts too many biggish objects, larger than galaxies but smaller than the most common kind of galaxy cluster, which contains hundreds of galaxies. It will take more than tinkering to repair this anomaly. Theorists may be able to use warm dark matter to wash out little galaxies, but it won't get rid of these heavier structures.

Lake believes that there is something fundamentally wrong with our theories about the early Universe. For structures to form at all in the Universe, there must be some initial variations in the density of gas. Cosmologists think that these variations were created by quantum fluctuations in the first fraction of a second after the big bang, and then magnified by a process called inflation (New Scientist, 16 December 2000, p 26). But how big were these fluctuations? The usual assumption is that, like a fractal pattern, the Universe was as lumpy on small scales as it was on large scales-a "scale-free fluctuation spectrum".

This is a catch-all solution that Lake thinks is used to mask our almost total ignorance of the early Universe. There is no strong evidence for it, and yet it has become entrenched in cosmological orthodoxy. "The strange thing is that people now treat it as though it were a unique prediction of inflation."

Lake believes this is where the problem lies, because the assumption that there are no special scales doesn't fit the shortage of small galaxies and small clusters. Galaxies and galaxy clusters mark distinct peaks in the fluctuation spectrum, rather than being part of a smooth continuum of structures, he says. "It seems like we have some notes or harmonics in the Universe." If he is right, then we can use this knowledge to work out just how the fluctuations formed. This could sound the death knell for dark matter. Powerful peaks in the fluctuation spectrum suggest bigger variations in gas density on the right scale to produce galaxies. And with a better head start, the gravity of ordinary gas would have been enough to make galaxies, obviating the need for dark matter.

That still leaves the problem of why some galaxies manage to spin so fast without falling apart, of course, but there may be an another explanation for that. Some researchers maintain that Newton's law of gravity might be slightly different on large scales. According to the theory of Modified Newtonian Dynamics, gravity pulls a little harder at large distances than conventional wisdom dictates, enabling it to hold fast-spinning galaxies together.

If Lake is right, astronomers will have to let the whole notion of dark matter slip away quietly into the night. The ghosts will have been banished for good.

Ghost hunting
How do you look for a black cloud in space? It's a riddle astronomers will have to answer if they want to find dark galaxies.

There may already be some circumstantial evidence for these shadowy objects. A few isolated galaxies look as though an invisible rival is pulling them to bits. UGC 10214, for example, has a conspicuous bridge of material extending into space towards-apparently-nothing.

There are also the small, faint objects called Blue Compact Galaxies, which are furiously forming stars. They cannot have been building stars so quickly for long, otherwise they would be packed with stars and therefore shining far more brightly. Trentham suggests that perhaps a dark matter halo has passed by each of these galaxies, causing gas clouds to collapse prematurely and form stars.

If nothing else, these shreds of evidence could help to narrow down the search for dark galaxies to a few promising sites. And astronomers will need all the help they can get.

If dark galaxies hold some gas or a few dead stars, conventional methods might just work. Brown dwarfs-small, failed stars-might collect within a dark galaxy, softly glowing with infrared light. The next generation of infrared satellites, such as NASA's Space Infrared Telescope Facility, will survey the Universe in the right wavelength range, and could spot them. A few white dwarfs, the cores of extinct stars, might also be around, but they would be almost impossible to spot with any existing or planned instruments.

However, Neil Trentham of Cambridge University thinks that there will be little or no ordinary matter in dark galaxies. If so, the search becomes fiendishly difficult.

Gravitational lensing might be the only way. A dark-matter halo would bend light slightly with its gravity. If it drifted between us and a more distant source, it would slightly distort the image of the source. Unfortunately, the technique used for spotting these distortions is still far too crude to see dark matter halos. But all may not be lost.

As the light rays detour through the dark matter halo, they take paths of different lengths. So rays of light emitted simultaneously no longer reach the Earth at the same instant. If the original source is variable, those changes will be staggered when viewed from Earth through the lens.

A rapidly varying source would be essential to detect these short time delays. Nial Tanvir of the University of Hertfordshire thinks that distant explosions called gamma ray bursts may do the job. When seen through a dark galaxy, rays from a burst would rise to a peak of brightness, dim a little and brighten again as the late-comers arrived.

The problem is that gamma ray bursts are few and far between, so we'll have to wait a very long time before one happens to explode right behind a dark galaxy. However, if in the future more sensitive gamma-ray satellites prove that there is a plethora of weaker, currently undetectable gamma ray bursts, then this method might be in business.

Stuart Clark is an astronomy writer and Director of Public Astronomy Education at the University of Hertfordshire

From New Scientist magazine, vol 170 issue 2287, 21/04/2001, page 38

Further reading:

: Completely dark galaxies, their existence, properties, and strategies for finding them, by Neil Trentham and others, at:]
The Bigger Bang, by James E. Lidsey, Cambridge University Press (2000)