Learning about Our Galaxy from its Nearest Galactic Neighbours

9 June 2011

Is the Milky Way a typical galaxy? What can we learn about it from looking at its nearest neighbours, the Magellanic Clouds? Comparing a giant numerical simulation of millions of galaxies, with observations of galaxies from the largest survey of the sky carried out to date, an international team of astrophysicists including Oxford's Phil Marshall set out to answer these questions.

Cosmologists generally suppose that the Milky Way is a fairly ordinary galaxy in the overall galaxy population, which is one aspect of the Copernican Principle that the Universe does not revolve around us. The standard theory for how those galaxies formed - known as the Cold Dark Matter model - has been remarkably successful: it beautifully reproduces the number and large-scale spatial distribution of the bright galaxies that we observe with our telescopes (as this movie shows). However, today's most detailed computer simulations of galaxies do not reproduce an important feature in the Milky Way: they do not have satellite galaxies that are as bright as the Magellanic Clouds. There are two possible explanations for this: either there is a problem with the Cold Dark Matter theory, or else the Milky Way is somewhat more unusual than we assume.

To settle the issue, the team, led by Stanford University Assistant Professor Risa Wechsler, combined observational data from the Sloan Digital Sky Survey with a cutting-edge new computer simulation, called Bolshoi. They identified galaxies that looked similar to the Milky Way in both the Sloan data and the Bolshoi simulation, in terms of their brightnesses and their distances from other large systems. The team then searched for fainter galaxies near these Milky Way analogs, with brightnesses like the Magellanic Clouds. The data, visualised in this movie, showed that only a very small fraction - a few percent - of galaxies like the Milky Way have two satellites like the Magellanic Clouds. The Bolshoi simulation matched this result almost perfectly, the agreement providing strong confirmation of the Cold Dark Matter model.

But what about the Copernican Principle? In essence, this implies that the Milky Way should look like a galaxy chosen at random from the general galaxy population. But since galaxies, like people, are extremely diverse in their properties, it should not be surprising if our particular home galaxy is a bit unusual in at least one respect, just as a randomly selected person would probably have at least one rather odd feature. In this sense, the Magellanic Clouds may be like the Milky Way's big ears or crooked nose: a bit strange, to be sure, but not necessarily any weirder than what you might find in any other galaxy.

These distinguishing features allow us to learn more about our home galaxy. Using Bolshoi, the team identified a sample of simulated galaxies that had satellites matching the Milky Way's in terms of their locations and speeds. Because the Magellanic Clouds are unique, these model galaxies should be a lot like the Milky Way. For example, the total mass of dark matter contained in the Milky Way is not easily measured, but the models predict that our galaxy weighs about 1 thousand billion times the mass of the sun. They also make a prediction for how the dark matter is distributed within our galaxy, important for researchers trying to detect dark matter particles on Earth.

Even more difficult to constrain through direct observations is the formation process of the Milky Way. However, we can use the simulations to look back in time and infer how the Milky Way was assembled. In a paper led by Michael Busha (ITP/UZH,Zurich), the team found that the Magellanic Clouds are very likely to be in the process of falling, together, into the Milky Way - as you can see in this movie of one of the Bolshoi model Milky Ways. Our galaxy is growing, as all galaxies do, by eating its neighbors. All of this information comes from some very simple observations of the Magellanic Clouds, combined with a giant simulation of all the galaxies in the local Universe -- and when you have such a detailed model of the Universe, to understand the Milky Way all you have to do is look up!

Contact: Dr. Phil Marshall (Oxford)

Project website, with links to journal articles

Interview with Phil at the Oxford Science Blog

NSF press release

Co-authors: Prof. Risa Wechsler (KIPAC/Stanford), Dr. Michael Busha (Zurich), Lulu Liu (KIPAC/Stanford), Dr. Brian Gerke (KIPAC/Stanford), Peter Behroozi (KIPAC/Stanford), Prof. Anatoly Klypin (NMSU), Prof. Joel Primack (UCSC).

Simulation visualizations: Dr. Ralf Kaehler (KIPAC/SLAC)

The Bolshoi Simulation (Klypin et al 2011) was run on the Pleiades machine at NASA Ames.
SDSS data was taken from the DR7 NYU-VAGC (Blanton et al 2005, Abazajian et al 2009).
Image of the SDSS Telescope from Fermilab Visual Media Services.
Photographs of the Magellanic Clouds: Rob Geraghty, Akira Fujii/David Malin Images.