ASTRONOMERS can watch dramas unfold in exotic galaxies billions of light years away, so it鈥檚 easy to forget that they still haven鈥檛 fathomed some of the things going on far closer to home. Take a host of giant gas clouds that are ripping their way through our celestial neighbourhood, some moving at more than 350 000 kilometres per hour. Astronomers have known for more than 30 years that the clouds are out there, but still can鈥檛 agree where they鈥檙e coming from.
The trouble is it鈥檚 easy to tell how fast the clouds are moving, but impossible to decide where exactly they lie. Are they outside our Galaxy, raining down from outer space? Or are they just a few tens of light years away, powered by the births and deaths of stars in the disc of the Milky Way? According to Leo Blitz, a radio astronomer at the University of California, Berkeley, the difference could mean a long life for our Galaxy, or imminent death.
The so-called high-velocity clouds made their debut in the 1960s amid lively debate about how our Galaxy formed. Astronomers agree that all galaxies were born in an enormous sea of hydrogen and helium that filled the early Universe. Some parts of this sea were slightly denser than others, and at these points isolated clumps of the gas collapsed under gravity, creating galaxies. As the galaxies grew, more and more of the primordial gas would have fallen onto these islands of matter, fuelling the formation of the first stars.
Advertisement
The question Dutch astronomer Jan Oort asked himself in 1963 was-why should that process ever stop? Oort believed that the remnants of the cloud from which the local group of galaxies formed would still be falling towards the galaxies now.
If so, Oort reasoned, we should be able to see clouds of hydrogen gas swooping down towards the Milky Way under its gravitational pull. Not that they鈥檇 be easy to see-neutral hydrogen gas does not emit visible light. But it does broadcast radio waves with a characteristic wavelength of 21 centimetres. Oort and his colleague Edith M眉ller used a 25-metre radio telescope at Dwingeloo in the Netherlands to see if they could catch the speeding clouds from outer space in action.
Sure enough, M眉ller and Oort found clouds moving towards the disc of the Galaxy at around 100 kilometres per second. 鈥淚t seemed pretty stunning confirmation of Oort鈥檚 idea,鈥� says Blitz. 鈥淚t set up a whole cottage industry of people looking for these clouds and mapping them.鈥�
鈥淭hat was a very strong indication,鈥� agrees Joel Bregman, an astronomer at the University of Michigan in Ann Arbor. 鈥淏ut things became much more complicated later on.鈥� For example, several clouds came to light that were not falling towards us. Far from plunging towards the Milky Way, they were fleeing outwards into space, something Oort鈥檚 idea could not explain.
That鈥檚 one reason why for more than 20 years, Bregman has backed an alternative explanation for the origin of the clouds. He believes that they started their lives in the gas-filled disc of our own Galaxy. Sedentary gas and dust that floats between stars in the galactic disc could be heated to over a million degrees centigrade by hot young stars and massive stars exploding at the ends of their lives. If so, the energy would send the gas flying out of the disc in a 鈥済alactic fountain鈥� towards the spherical outer halo of the Galaxy, where the gas would cool, then rain back down.
But the clouds鈥� origin is still in dispute because they give very little away. In particular, it is difficult to establish their distances from Earth. Astronomers have a wide repertoire of tricks to find out how far away astronomical objects are. For some variable stars, for instance, measuring the time they take to brighten and fade tells you their intrinsic brightness. So knowing how bright they appear from Earth, you can calculate how far away they lie.
Tug of war
But the only direct way to find out the distance to a gas cloud is to see where it lies in relation to stars at known distances. If the cloud absorbs light from a distant star, taking a 鈥渂ite鈥� out of its spectrum, it has to lie between the star and us. But because most of the clouds swarm outside the star-crammed disc of our Galaxy, it鈥檚 difficult to find any suitable bright stars to use as benchmarks. So far, astronomers have measured the distances to only a handful of high-velocity clouds, which have turned out to be less than 15 000 light years away-safely within the Milky Way. But that鈥檚 not to say that all the clouds are just as close.
High-velocity clouds also conceal their size. A map of the hydrogen emission from each cloud reveals how big a patch of sky it covers. But without knowing how far away the cloud is, it鈥檚 impossible to translate this into a real size. Finding out whether the clouds are mammoth or minuscule on astronomical scales would be an important clue to whether they are a product of our Galaxy. In fact, a few are known to be nearby. Everyone agrees that one group of high-velocity clouds exists thanks to the eternal tug of war between our Galaxy and its entourage of two small satellite galaxies, the Magellanic Clouds. The Milky Way鈥檚 gravity stretches the Magellanic Clouds as they orbit, and pulls material towards us in a filament called the Magellanic stream. It became clear in the early 1970s that some of the high-velocity clouds are part of such a stream. But there is no consensus about the remaining clouds.
Blitz became interested in the mystery about 20 years ago, pretty much by accident. The molecular clouds that he was studying in the plane of the Galaxy were too close to the horizon to observe, so he looked at the high-velocity clouds, and became intrigued. Since then, he and his colleagues have amassed a wealth of data that they say points to an extragalactic origin for the vast majority of the clouds. First, they contain only traces of elements heavier than hydrogen and helium. Stars process hydrogen and helium into heavier elements, then spray them into space as they explode at the ends of their lives. So if the high-velocity clouds contained much debris from supernovae, they should contain a high proportion of heavy elements. But observations of the way the clouds weakly absorb the light of distant galaxies show that the gas contains levels of magnesium and sulphur that are just a tiny fraction of the amounts in the Sun.
More evidence comes from one of the clouds, which spans a giant 25掳 of the sky-about 50 times the diameter of the full Moon. The cloud lies in the direction of the outer regions of the Galactic disc, and is moving at up to 200 kilometres per second with respect to the Milky Way. If it were colliding with the gas already sitting in the Galaxy鈥檚 disc, the energy release would be catastrophic. 鈥淚t would have created explosions equivalent to a thousand exploding stars,鈥� says Blitz.
With no such fireworks in sight, Blitz concludes that this cloud must lie beyond the dense Galactic disc at least 120 000 light years away. If this is right, the cloud must be huge at least 60 thousand light years across, containing 90 million times the mass of the Sun. 鈥淭hat鈥檚 unlike anything we see in the Milky Way,鈥� Blitz says.
Another line of evidence comes from farther afield. If high-velocity clouds are remnants of the embryonic cloud that formed the local group of galaxies, we鈥檇 expect to see others raining down on neighbouring galaxies. A giant high-velocity cloud in the direction of one nearby galaxy, called M33, came to light in the 1970s. And one turned up following a sky survey by radio telescopes in the Netherlands in 1992. A cloud 14掳 across sits 鈥渞ight smack towards the Andromeda galaxy鈥�, Blitz says. If the cloud is indeed close to the Andromeda galaxy, its diameter would be around 520 000 light years, and it would contain about 900 million solar masses of hydrogen.
Blitz鈥檚 colleague David Spergel of Princeton University in New Jersey has used a computer simulation to add weight to this idea. His program modelled the local group of galaxies as they formed roughly 14 billion years ago. It then simulated how the primeval gas left over from the formation of the galaxies would fall onto them, leading up to the present day. The simulation suggested that our Galaxy has already swallowed around a quarter of the mass in the embryonic cloud. The virtual Andromeda guzzled even more, around 50 per cent.
Spergel鈥檚 simulations predict that the rest of the gas should remain outside the galaxies, tending to gather along a filament connecting Andromeda and the Milky Way. That鈥檚 exactly where we see the main groupings of clouds today, according to Blitz, if you ignore the few clouds that are probably nearby. The simulations also predict cloud velocities in line with those that radio telescopes observe (The Astrophysical Journal, in press). 鈥淲e were knocked off our feet,鈥� says Blitz. 鈥淲hat this means is that these high-velocity clouds around the Milky Way and Andromeda are the material out of which the galaxies formed.鈥�
Bregman, however, sees things differently. He doesn鈥檛 think the high-velocity clouds form main groupings in any particular direction. 鈥淭he clouds are all over the sky,鈥� he says. And he takes issue with Blitz for ignoring the nearby clouds: 鈥淵ou really have to ask: was that a fair thing to do?鈥� Bregman also disputes Blitz鈥檚 arguments that the galactic fountain would produce clouds that appear to contain lots of heavy elements. 鈥淭hat鈥檚 a very weak argument,鈥� he says. He points out that many of the heavier elements would freeze out onto dust grains. Because only the gaseous component absorbs light, these elements would effectively be invisible.
Blitz counters that some key heavy elements are unlikely to freeze out onto solid grains, and contends that the galactic fountain could not accelerate the high-velocity clouds up to such enormous speeds. But the extragalactic theory is not entirely sewn up. Blitz鈥檚 team has yet to explain why the high-velocity clouds have any heavy elements at all, if they represent the light-element haze that filled the Universe before galaxies appeared. Blitz admits this is a problem, but says it鈥檚 not unique to his theory. Other gas that astronomers know to be intergalactic also contains a smidgen of heavier elements, though nobody knows why. Some suspect that a generation of stars formed in the Universe before galaxies existed to house them. Such itinerant stars would have manufactured some heavy elements and sprayed them into the pristine material that went on to form galaxies. If so, that could also explain why extragalactic clouds contain heavy elements.
Star turn
There鈥檚 another challenge, though, that is harder to answer. If the clouds have been swooping around for most of the age of the Universe, they must be neatly held together by their own gravity or they would have broken up and dispersed into space. But if they鈥檙e bound by gravity, why doesn鈥檛 gravity take over and collapse the clouds into stars? Maybe a tiny population of stars formed, Blitz says, their lights too dim to see. Or perhaps the clouds we see are among the few that have remained on the brink of making stars, being not quite dense enough. 鈥淭hese questions must be answered,鈥� Blitz concedes.
Blitz and Bregman agree that the answer to the problem lies in future observations that could pin down the distances of the high-velocity clouds once and for all. Bregman says if the clouds are relatively close, a new instrument due to be installed on the Hubble Space Telescope in 2002 should confirm this. The sensitive Cosmic Object Spectrograph will reveal whether high-velocity clouds are absorbing the light of stars up to 60 000 light years away.
If stars in our Galaxy are absorbing the light of high-velocity clouds in the direction of the Andromeda galaxy, Blitz and his colleagues admit they will have lost their case. They have pinned their hopes on finding signs of high-velocity clouds swarming around many other galaxies, something that observations to date may have missed.
Whatever the answer, there will be implications for the life expectancy of our galaxy and others. New generations of stars can only form if the gas in galaxies is plentiful enough to form clumps under gravity and collapse into hot, dense stellar cores. According to Spergel鈥檚 simulations, roughly 1.2 Sun鈥檚 worth of neutral hydrogen falls onto the Milky Way from extragalactic clouds each year. This would be just enough to maintain the current rate of star formation and keep stars burning until the Milky Way has doubled its present age. In other words for another 10 billion years.
But if no gas is falling on the Milky Way, Blitz says, the gas in the disc will become too wispy to form stars in just 300 million years-only 2 per cent of the Galaxy鈥檚 current age. 鈥淭his could be the last gasp for our Galaxy,鈥� he says. The clouds that break galactic speed limits may well decide when our lights will finally go out.
