Halos gone MAD

Distribution of approximately 3 dark matter when the universe billions years old, a numerical simulation of galaxy formation taken from was. In the left pane shows the continuous distribution of dark matter particles, shows the typical wispy structure of the cosmic Web, with a network of metal sheets and filaments, while in the right pane that dark matter of halos, represents the most efficient mass of 300 billion times higher than in Star burst underlines cosmic Web sites for the formation of galaxies with a minimal dark matter Halo of the credit so.: VIRGO Consortium/Alexandre Amblard / ESA

One of the successes of the ΛCDM model of the universe is the way for models structures similar to that with scales and distributions that we see today in the universe. Or, at least that is what astronomers tell us. While computer simulations numeric universes in a box can recreate, is this mathematical approximations interpret a challenge in and of itself. To identify the components of simulated space, astronomers had to tools structure to search for develop. The results is almost 30 independent computer programs since 1974. Each promises that reveal the structure of the universe to find form regions in the dark matter of halos. To test these algorithms from a Conference was held in Madrid, Spain to see how well they stacked were made during the May 2010 with the title "Halons go MAD" in which 18 of this code to the test.

Numerical simulations of universes, such as the famous Millennium simulation start with no more than "particles". While these were certainly small cosmological scale, is such a particle dark matter BLOBs with million or billion solar masses. Because time runs forward, they may interact our best understanding of physics and the nature of such matter with another following rules the collapse. This leads to an evolving universe of astronomers need to use complicated codes would make finding the clusters of dark matter in galaxies.

One of the most important methods that use such programs is to small Overdensities are looking for and then grow until the density at a negligible factor is a spherical shell around it. Most are then cut the particles within the band that are not gravity bound to ensure, that the mechanism for which recognize take not only on a short, transient clustering, decay in time. Other techniques include search other phase spaces for particles with similar speeds all nearby (a sign that they have bound).

Two tests set to each of the algorithms as came to compare it, were. The first involved a series of deliberately created dark matter of halos with embedded sub-halos. Because the particle was deliberately placed, the output of the programmes should correctly the Center and size for the halos. The second test was a full simulation of the universe. In this respect the allocation would be unknown, but the sheer size would make it possible to watch different programs on the same record, as well as she interpreted a common source to be compared.

Both tests all finders carried out generally well. In the first test, it gave up some discrepancies based such as various programs which defines the location of the halos. Some define it as the peak density, while others defined it as a center of gravity. In search of sub-halos those who the phase space approach used seemed to smaller formations was can reliably detect but could not always detect the particles in the lump is actually committed. For the complete simulation, all algorithms agreed exceptionally well. Due to the nature of the simulation, not small scales were well represented, these structures, see the understanding as each was limited.

Prefer the combination of these tests not a particular algorithm or method above all others. It revealed that each generally good in terms of an other functions. The possibility for so many independent code, which means with protocol-independent methods, that the findings are extremely robust. Know that they share our understanding of the universe as developed can be fundamental comparisons make astronomers of the observable universe, to test the such models and theories.

The results of this test were collected in a paper, which is scheduled for publication in an upcoming issue of the monthly notices of the Royal Astronomical Society.

Tagged as: cosmology, simulations

View the original article here