The Dark Side of the Milky Way

Dark matter is not just a puzzle. It is a solution

 By Leo Blitz | Tuesday, September 20, 2011

Although astronomers only slowly came to realize dark matter’s importance in the universe, for me personally it happened in an instant. In my first project as a postdoc at the University of California, Berkeley, in 1978, I measured the rotational velocities of star-forming giant molecular clouds in the outer part of the disk of our Milky Way galaxy. I worked out what was then the most accurate method to determine those velocities, and I sat down to plot out the results (by hand on graph paper) in the astronomy department lounge. Two other experts on the Milky Way, Frank Shu and Ivan King, happened by. They watched as I filled in the velocities of the outermost clouds, and the pattern we saw made it clear at once that the Milky Way was rife with dark matter, especially in its outermost parts. We sat and scratched our heads, imagining what the nature of the dark matter could be, and all the ideas we came up with turned out in short order to be wrong.
This study was one of many in the 1970s and 1980s that forced astronomers to conclude that dark matter—a mysterious substance that neither emits nor absorbs light and reveals itself solely by its gravitational influence—not only exists but is the dominant material constituent of the universe. Measurements with the WMAP spacecraft confirm that dark matter accounts for five times as much mass as ordinary matter (protons, neutrons, electrons, and so on). What the stuff is remains as elusive as ever. It is a measure of our ignorance that the most conservative hypothesis proposes that dark matter consists of an exotic particle not yet detected in particle accelerators, predicted by theories of matter that have not yet been verified. The most radical hypothesis is that Newton’s law of gravity and Einstein’s general theory of relativity are wrong or, at the very least, require unpleasant modifications.
Whatever its nature, dark matter is already providing keys to unlock some persistent puzzles about how the Milky Way came to have certain of its features. For example, astronomers have known for more than 50 years that the outer parts of the galaxy are warped like a vinyl phonograph record left on a heater. They could not make a viable model for the warp—until they considered the effects of dark matter. Similarly, computer simulations of galactic formation based on the assumed properties of dark matter predicted that our galaxy should be surrounded by hundreds or even thousands of small satellite galaxies. Yet observers saw only about two dozen. The discrepancy led people to question whether dark matter had the properties they thought it did. But in recent years several groups of astronomers have discovered troves of dwarf satellites, narrowing the disparity. These newly located satellites are not only helping to resolve a long-standing mystery of galactic structure; they may also be teaching us something about the total cosmic inventory of matter.
Factoring in the Warp
A first step to understanding what dark matter tells us about the Milky Way is to get a general picture of how the galaxy is organized. Ordinary matter—the stars and gas—resides in four major structures: a thin disk (which includes the pinwheel-like spiral pattern and the location of the sun), a dense nucleus (which also harbors a supermassive black hole), an elongated bulge known as the bar, and a spheroidal “halo” of old stars and clusters that envelops the rest of the galaxy. Dark matter has a very different arrangement. Although we cannot see it, we infer where it is from the rotation velocities of stars and gas; its gravitational effects on visible material suggest it is approximately spherically distributed and extends far beyond the stellar halo, with a density that is highest at the center and falls off approximately as the square of the distance from the center. Such a distribution would be the natural result of what astronomers call hierarchical merging: the proposition that in the early universe, smaller galaxies accreted to build larger ones, including the Milky Way.
For years astronomers could get no further than this basic picture of dark matter as a giant, undifferentiated ball of unidentified material. In the past several years, however, we have managed to glean more details, and dark matter has proved rather more interesting than we had suspected. Various lines of evidence suggest that this material is not smoothly distributed but has some large-scale lumpiness to it.
Such unevenness would explain the existence and size of the galactic warp. When astronomers say the galaxy is warped, we are referring to a specific distortion in the outskirts of the disk. At distances beyond about 50,000 light-years from the center, the disk consists almost entirely of atomic hydrogen gas, with only a few stars. Mapped by radio telescopes, the gas does not lie in the plane of the galaxy; the farther out you go, the more it deviates. By a distance of about 75,000 light-years, the disk has bent about 7,500 light-years out of the plane.
Evidently, as the gas within the disk revolves around the galactic center, it also oscillates up and down, in and out of the plane. These oscillations occur over hundreds of millions of years, and we catch them at one moment in their cycle. In essence, the gas disk acts as a kind of giant gong vibrating in slow motion. Like a gong, it can vibrate at multiple frequencies, each corresponding to a certain shape of the surface. In 2005 my colleagues and I showed that the observed warp is the sum of three such frequencies. (The lowest is 64 octaves below middle C.) The overall effect is asymmetric: gas on one side of the galaxy is much farther from the plane than gas on the other side.
The radio astronomers who first noticed the warping in the 1950s thought it might result from gravitational forces exerted by the Magellanic Clouds, the most massive galaxies in orbit around the Milky Way. Because these satellite galaxies are orbiting out of the plane of the Milky Way, their gravity tends to distort the disk. Detailed calculations, however, showed that these forces are too weak to explain the effect because the Magellanic Clouds are puny in comparison to the Milky Way. For decades the reason for the pronounced warp remained an unsolved problem.
Dark Hammer
The recognition that the Milky Way contains dark matter, together with new estimates of the mass of the Magellanic Clouds (which showed them to be more massive than thought), raised a new possibility. If the gas disk acts as a giant gong, the orbit of the Magellanic Clouds through the dark matter halo can act as a hammer ringing the gong, sounding its natural notes or resonant frequencies, albeit not directly. The clouds create a wake in the dark matter, just as a boat forms a wake as it plows through the water. In this way, the clouds create some unevenness in the distribution of dark matter. That, in turn, acts as the hammer to cause a ringing in the low-mass, outer parts of the disk. The upshot is that even though the Magellanic Clouds are puny, dark matter greatly amplifies their effects.
Martin D. Weinberg of the University of Massachusetts Amherst put forward this general idea in 1998. He and I later applied it to observations of the Milky Way and found we could reproduce the three vibration patterns of the gas disk. If the theory is correct, the warp is an active feature of the Milky Way with a shape that continually changes as the Magellanic Clouds move through their orbits. The shape of the galaxy is not fixed but ever shifting. [Editors’ note: A video of this process is available at www.ScientificAmerican.com/oct2011/blitz.]
The warp is not the only asymmetry in the shape of the Milky Way. One of the most striking is the lopsided thickness of the outer gas disk, also discovered using radio telescopes. If one drew a line from the sun to the center of the Milky Way and extended it outward, one would find that the thickness of the gas layer on one side of this line is, on average, about twice that on the other side. This large asymmetry is dynamically unstable and, left to its own devices, would tend to right itself; its persistence requires some sort of mechanism to maintain it. For 30 years astronomers knew about the problem but swept it under the rug. They revisited it only very recently when a much improved new survey of the Milky Way’s atomic hydrogen, coupled with a better understanding of the noncircular motions of the gas, made the asymmetry impossible to ignore any longer.
The two leading explanations both involve dark matter. Either the Milky Way is spherical but not concentric with its dark matter halo, or as Kanak Saha of the Max Planck Institute for Extraterrestrial Physics in Garching and several collaborators recently argued, the dark matter halo is itself somewhat asymmetric. Both call into question astronomers’ old view that the Milky Way and the halo formed together from the condensation of a single gargantuan cloud of material; if it had, the ordinary matter and the dark matter should be centered on the same point. Therefore, the asymmetry is further evidence the galaxy formed from the merging of smaller units or grew by continued merging or accretion of intergalactic gas—processes that need not be symmetric. The center of the galaxy could be offset from the center of the dark matter because gas, stars and dark matter behave differently.
A way to cross-check this idea is to study the long, thin streams of stars that stretch through the outer reaches of the galaxy. These formations are the elongated remains of former satellite galaxies. The most common kind of galaxy to be found in orbit about the Milky Way system is known as a dwarf spheroidal because of its roundish shape and small mass of stars—typically only about one ten-thousandth that of the Milky Way. Over time its orbit decays, and the satellite becomes subject to the tidal forces of the Milky Way. These forces are the same as those produced by the moon on Earth, stretching out the mass of water on Earth as our planet rotates, producing the twice-daily ocean tides. The dwarf galaxy gets stretched out and can be reduced to a thin ribbon [see “The Ghosts of Galaxies Past,” by Rodrigo Ibata and Brad Gibson; Scientific American, April 2007].
Because the stars in these streams orbit the galaxy at large distances, where the gravitational effects of dark matter are large, the shapes of the streams probe the shape of the halo. If the halo is not perfectly spherical but somewhat flattened, it will exert a torque on the orbits of stars in the stream and cause a marked deviation from a great circle. As it happens, the streams are observed to be very thin, and their orbits around the galaxy are nearly great circles. Computer simulations by Ibata and his colleagues therefore suggest that the dark matter distribution is close to spherical, although it might nonetheless be as lopsided as suggested by Saha and his colleagues.
Galaxies Gone Missing
If the destruction of dwarf galaxies raises questions, so does their formation. In our current models, galaxies begin as agglomerations of dark matter, which then accrete gas and stars to form their visible parts. The process yields not only large galaxies such as ours but also numerous dwarfs. The models get the properties of these dwarfs about right but predict far more of them than observers see. Does the fault lie in the models or with the observations?
Part of the answer comes from new analyses of the Sloan Digital Sky Survey, a systematic scan of about a quarter of the sky. The survey has turned up about a dozen new, extremely dim galaxies in orbit around the Milky Way. Their discovery is astonishing. The sky has been so completely surveyed for so long that it is difficult to imagine how galaxies on our cosmic doorstep could have lain undiscovered all this time. These galaxies, known as ultrafaint dwarfs, in some cases contain only a few hundred stars. They are so feeble and diffuse that they do not show up on ordinary images of the sky; it requires special data-handling techniques to identify them.
Had the Sloan survey covered the entire sky when the ultrafaint galaxies were discovered, it might have discovered another 35 or so more. Still, that would not account for all the “missing” dwarfs. So astronomers have sought other possibilities. Perhaps more such galaxies are out there, too far away for existing telescopes to detect. The Sloan survey can find ultrafaint dwarfs out to a distance of about 150,000 light-years. Erik Tollerud and his collaborators at the University of California, Irvine, predict that as many as 500 undiscovered galaxies orbit the Milky Way at distances up to around one million light-years from the center. Astronomers should be able to find them with a new optical telescope called the Large Synoptic Survey Telescope, which will have eight times the collecting area of the Sloan telescope. Construction began on the observatory this past March.
Another hypothesis is that the Milky Way is orbited by galaxies even dimmer than the faintest ultrafaint dwarfs—so dim, perhaps, that they contain no stars at all. They are almost pure dark matter. Whether such galaxies could ever be seen depends on whether they contain gas in addition to the dark matter. Such gas might be sufficiently diffuse that it cools only very slowly, too slowly to have formed stars. Radio telescopes surveying large patches of the sky might nonetheless detect the gas.
If these galaxies lack gas, however, they would reveal their presence only indirectly, by their gravitational effects on ordinary matter. If one of these dark galaxies hurtled through the disk of the Milky Way or some other galaxy, it might leave a splash like that of a pebble thrown into a quiet lake—­ob­serv­able as perturbations to the distribution or velocities of stars and gas. Unfortunately, this splash would be very small, and astronomers would have to convince themselves that it could not be made in any other way—a daunting task. All spiral galaxies show disturbances through­­out their atomic hydrogen disks akin to waves in a rough sea.
If the dark galaxy is massive enough, a method devised by Sukanya Chakrabarti, now at Florida Atlantic University, and several collaborators, including myself, may provide the tools to discern its passage. We recently showed that the largest disturbances in the outskirts of galaxies are often tidal imprints left by passing galaxies, which can be differentiated from other perturbations. By analyzing the disturbances, we can infer both the mass and current location of the intrusive galaxy. This technique can discern galaxies as small as one-thousandth the mass of the primary galaxy. Applying this method to the Milky Way, our team inferred that an undiscovered possibly dark galaxy lurks in the plane of the Milky Way, about 300,000 light-years from the galactic center. Plans are now under way to hunt for this galaxy in near-infrared light using data collected by the Spitzer Space Telescope.
Too Little Light
Quite apart from the challenge of finding them, ultrafaint and dark galaxies in the Milky Way’s vicinity pose a deeper problem for astronomers in regard to the relative amounts of material they contain. Astronomers commonly measure the amount of matter in a galaxy in terms of its mass-to-light ratio: the mass of material divided by the total amount of light it gives off. Typically we give the ratio in solar units; the sun, by definition, has a mass-to-light ratio of 1. In our galaxy, the average star is somewhat less massive and much dimmer than the sun, so the overall mass-to-light ratio of luminous matter is closer to 3. Including dark matter, the total mass-to-light ratio of the Milky Way jumps to about 30.
Josh Simon, now at the Carnegie Institution of Washington, and Marla Geha, now at Yale University, measured the velocities of the stars in eight ultrafaint dwarfs to obtain the mass and luminosity of these galaxies. The mass-to-light ratios in some cases exceed 1,000—by far the highest of any structure in the known universe. In the universe as a whole, the ratio of dark to ordinary matter is almost exactly 5. Why is the mass-to-light ratio of the Milky Way system so much higher and the ultrafaint galaxies even more so?
The answer could lie in the numerator or denominator of the ratio: galaxies with mass-to-light ratios higher than the universal average either have more mass than expected or produce less light. Astronomers think that the denominator is to blame. A huge amount of ordinary matter does not radiate brightly enough to see, either because it has never been able to settle into galaxies and coalesce into stars or because it did settle into galaxies but was then expelled back out into intergalactic space, where it resides in an ionized form that is undetectable by present-day telescopes [see “The Lost Galaxies,” by James E. Geach; Scientific American, May]. Lower-mass galaxies, having weaker gravity, lose more of their gas, so their light output is disproportionately reduced. What a curious irony that the problems raised by one kind of unseen matter (dark matter) should give rise to yet another set (ordinary but undetected matter).
The puzzle of dark matter, which lay dormant for so many years, is now one of the most vibrant research areas in both physics and astronomy. Physicists are hoping to identify and detect the particle that composes dark matter, and astronomers are looking for more clues about how the stuff behaves. But puzzle or no, the existence of dark matter has provided the answer to a large range of astronomical phenomena.
Sciam