I'm Garry Angus a postdoctoral fellow at the Astronomy, Cosmology and Gravity Centre (ACGC) at the University of Cape Town. The centre is composed of the Astronomy department and the Cosmology and Gravity group. I'm funded by the Claude Leon Foundation and a University Research Centre (URC) Fellowship (from UCT)

I currently investigate a controversial aspect of modern cosmology:

Modified Newtonian Dynamics (MOND)

MOND is an alternative theory of gravity proposed by Mordehai Milgrom in a series of seminal papers from 1983. It is defined by an acceleration scale, a_o=10^-10ms^-2 => such that gravitational fields stronger than a_o are unaltered from Newtonian dynamics, but when the field is weaker than a_o then the dynamics are modified, roughly according to g=(g_na_o)^1/2. Anyone's first port of call for info about MOND has to be the impeccably maintained MOND PAGES of Stacy McGaugh (or see his latest review with Benoit Famaey).

MOND, Sterile Neutrinos and Cosmology

Although MOND works arguably remarkably well for explaining the dynamics of individual spiral galaxies, it cannot account for the dynamics of clusters of galaxies, nor the many cosmological probes that depend on an expansion history that requires a large component of dark matter.

In Angus (2009) I proposed, through fitting the cosmic microwave background (CMB), that 11eV sterile neutrinos could solve all problems MOND has with clusters and cosmology. The fit to the CMB can be seen at the top of the page. Unfortunately, 11eV sterile neutrinos are too light to facilitate the formation of the large scale structure of the Universe see Angus & Diaferio (2011)

I recently wrote my own cosmological N-body code based on the QUMOND theory. The figure below is a z=0 output from a 512Mpc/h box with 128³ grid cells and particles using 11eV sterile neutrinos and QUMOND. Clicking on the image will download a movie which shows the structure formation process from z=200->0.

This paradigm cannot form the correct mass function of groups and clusters of galaxies. In fact, it severely over predicts the number of superclusters. I'm currently investigating more massive sterile neutrinos (between 30eV and 1keV) along with some other ideas which may be able to produce the correct cluster mass function.

If you are interested in reading more about this paradigm, please read my detailed blog entry on Scilogs.

Why Bother With Alternative Theories of Gravity & Cosmology

The wealth of cosmological data is currently interpreted with the Lambda-CDM model: Einstein's cosmological constant and cold dark matter.

However, there is no obvious physical mechanism that can generate the cosmological constant or more exotic dark energy fluids, nor is there any direct or indirect confirmation of the existence of the cold dark matter particle.

There are also many puzzling aspects of related physical fields: inflation is still not well understood, we do not comprehend the full details of neutrino oscillations and we do not have a theory of quantum gravity.

The Milky Way Satellites

On the right is a figure showing the 3D positions of the satellite galaxies relative to the Milky Way. One significant problem for the LCDM paradigm is why there are so few satellites (Klypin et al. 1999 and Moore et al. 1999 and references to these articles) and why the satellites have correlated orbits and positions - which makes them appear like a disk of satellites (see Kroupa et al. 2010 for a review). I recently tried to argue (see Angus, Diaferio & Kroupa 2011) that the small proper motions of the satellites prohibits them from being accreted along filaments. In related topics, in Angus (2008) I fitted the losVDs of the 8 classical Milky Way dwarf spheroidal galaxies (as measured by Walker et al. 2007) in MOND and showed that the 4 with the best measured photometry and kinematica data are in excellent agreement with the theory. The other 4 require updated photometry and more complete set of line of sight velocities, but currently do not fit in with MOND. In Angus & Diaferio (2009) we use an orbit integrator and a prescription for dynamical friction in MOND to show that it is not a problem for the Fornax dSph to have 6 globular clusters with relatively small angular separations from the dSph and for it to have no central bulge.

.