Did dark matter form the Gould Belt?

The Sun lies near the middle of a ring of stars thousands of light-years across. What formed this structure?

On interstellar scales, the solar neighborhood is an extraordinary place. Within a kiloparsec of the Sun, you can find the Rho Ophiuchi cloud complex, the Orion Nebula, the Taurus Molecular Cloud, and countless other nebulae and star-forming regions. Elegant and diverse, they’re the subjects of many exquisite images from Hubble, Spitzer, and other space telescopes. But what if I told you that many — most, even — of these molecular clouds are related, and indeed might have formed at the same time?

The Rho Ophiuchi cloud complex, as seen by the Wide-Field Infrared Explorer. Image credit: NASA/JPL-Caltech/WISE Team

Almost a century and a half ago, an astronomer named Benjamin Gould noticed that many nearby stars and star-forming regions lay in a strip across the sky distinct from the galactic plane. Later dubbed the Gould Belt, this region is about a kiloparsec in diameter and contains an unusually high number of OB stars and molecular clouds. This isn’t some statistical coincidence, though — it’s real. More than a century’s worth of data indicates that many of these stars formed at the same time, and that seemingly unrelated objects, like the Scorpius-Centaurus Association and the Orion-Monoceros complex, had a common past.

At the moment, nobody knows for sure what caused this large-scale star formation. An interaction with another galaxy can likely be ruled out, as we would see effects in a larger region of space. However, it’s possible that something similar happened — just on a smaller scale. A leading theory for the formation of the Gould Belt holds that about 30 million years ago, a high-velocity cloud impacted the galactic disk at a speed of about 100 km/s. As a shock wave formed at the interface, instabilities in the post-shock region led to the formation of the progenitors of current molecular clouds, which collapsed into smaller complexes and eventually stars.

The Gould Belt (in red) compared to the galactic plane (in blue). Image credit: Nick Wright/Thomas Dame

Recent simulations have suggested an alternate, but related situation, where a giant molecular cloud was hit by a massive clump of dark matter. Could this exotic explanation be correct? If so, it might have important consequences for the formation of the Milky Way and a major problem in galaxy formation: Why does the Milky Way have so few satellite galaxies?

A high-speed collision

The halo of the Milky Way is peppered with high-velocity clouds (HVCs). Weighing up to several million solar masses, these cool clumps of gas travel at speeds of around 100 km/s. While their origin is still unknown, their interactions with the galactic disk are an important topic of study. When an HVC passes through the disk, the impact can dissipate some of the cloud’s hydrogen, driving the evolution of the halo and the interstellar medium.

A Green Bank Telescope image of Smith’s Cloud, a massive HVC traveling at about 73 km/s. Image credit: Bill Saxton, NRAO/AUI/NSF, under the Creative Commons Attribution 3.0 Unported license.

More dramatic effects than long-term gas stripping are possible. Comeron & Torra (1994) were some of the first to suggest that the shock wave from such a collision could lead to the formation of a ring of “bound complexes”, the precursors of giant molecular clouds. They simulated the impact of several different HVCs (ranging from 10,000 to 1,000,000 solar masses) with the galactic disk, and tried to figure out whether the gas escaping from the shocked region would be stable to density or pressure perturbations.

Their results depended on two important factors: the rotation of the disk and the galactic magnetic field. In particular, axisymmetric perturbations were stabilized by magnetic field lines, whereas shear (radial) perturbations could be subject to instabilities, with growth rates dependent on the Oort constants. Over about 100 million years, cloud complexes would arise from these perturbations, with galactic rotation limiting their sizes to about 100 parsecs.

Fig. 1, Comeron & Torra (1994). Within 1–3 million years, all parts of the cloud should have entered and perhaps exited the shocked region.

The authors found that an HVC about 600 parsecs across — similar to their intermediate model — and around 450,000 solar masses could form a region similar in size and morphology to the Gould Belt. Interestingly enough, this mass appeared to match estimates for the total mass of cloud complexes in Orion and Monoceros. Three other regions of interest in the Belt weigh about half as much combined; when assuming that preexisting molecular clouds contributed to the formation of these star-forming regions, the model did an excellent job of reproducing many of the Belt’s properties. Assuming the impact began 40 million years ago, it appears to be a reasonable hypothesis for the formation of the Belt.

Could dark matter play a role?

The high-velocity cloud theory is an attractive one, and does a good job of explaining the masses of the largest cloud complexes in the Gould Belt. However, Comeron & Torra’s simulations weren’t the only ones to consider a collision; more recently, Bekki (2009) looked at a slightly more exotic alternative: A ten million solar mass dark matter clump (DMC) moving at about 90 km/s, impacting a giant molecular cloud about one-tenth as massive. The collision should form a density wave, transforming the oblate molecular cloud into a ring-like structure. Star formation would begin after about 7 million years, when the center of the cloud would be dense enough, and the ring would expand and evolve over tens of millions of years.

Fig. 1, Bekki (2009). 45 million years after a collision between a molecular cloud (magenta) and a clump of dark matter (blue), new stars (yellow) have formed in a ring several hundred parsecs across

This hypothesis can be tested in two important ways. First, the DMC should have resulted in an overdensity in the dark matter distribution within the Belt. If we could successfully detect such an overdensity, this would provide support for the theory. Additionally, the simulations also predicted the expected velocities of cloud complexes and stellar associations formed from the collision. Accurate kinematic measurements of OB associations are possible; if they match what the dark matter model predicts, it could provide strong supporting evidence.

Bekki’s simulations produced an interesting result: Given that there should be anywhere from 20–100 DMCs of ten million solar masses or more in the Milky Way’s halo, and given the known distribution of giant molecular clouds, it turns out that there should have been other collisions just like this over the course of the galaxy’s lifetime. In fact, we should see the same behavior in other galaxies. If — and this is a big if — we could definitively study the occurrence of belts like these in the Milky Way, perhaps we could compare it to the assumed DMC distribution, or, to go back to the high-velocity cloud theory, the impact rates of HVCs with the galactic plane.

Fig. 2, Bekki (2009). The modeled distribution of stars forms an inclined belt, shown in red, that matches the actual locations of a number of stellar associations (white circles).

The missing galaxies

I mentioned earlier that answering the question of the formation of the Gould Belt could give us insights into the formation of the Milky Way. Both the HVC and DMC hypotheses are clearly connected to galactic evolution; high-velocity clouds, for example, influence the composition and properties of the interstellar medium. One interesting line of investigation from Bekki’s model has to do with something called the missing satellites problem. Essentially, simulations predict the Milky Way should have more than ten times as many dwarf galaxies orbiting it as it actually seems to have.

The Fornax Dwarf Spheroidal Galaxy, one of the Milky Way’s satellites. There aren’t as many satellite galaxies as simulations tell us there should be. Image credit: ESO/Digitized Sky Survey 2, under the Creative Commons Attribution 4.0 International license.

A popular solution to the missing satellites problem involves the formation of “subhalos” of anywhere from one million to 100 million solar masses — small dark matter halos that are analogs of the Milky Way’s. You might notice that this mass range is centered on the assumed mass of Bekki’s DMC. Could studying DMCs like this shed light on subhalo populations in the galactic neighborhood? It might. This is another interesting reason to continue the search for other Gould Belts in the galaxy.

I should point out that there are other possible explanations for the existence of the Gould Belt. Some involve an expanding set of OB stars, or a series of supernovae, or other small-scale events. These are also plausible to varying degrees, and are also promising avenues of research. Regardless of which explanation is right, though, these hypotheses tell us a lot about both the Milky Way as a whole and our place in it. It’s truly extraordinary that many features in the solar neighborhood might have a common origin — and one in the not-too-distant past! The Gould Belt might have formed closer to the rise of humans than the extinction of the dinosaurs. It truly is a newcomer on the galactic stage.