Astrophysics – New Simulations Revolutionize Theory of Galaxy Formation

The cosmic processes we discovered through our ChaNGA simulations evaded detection for over a decade because they could not be captured in lower-resolution models. We’re looking forward to finding out what upcoming galaxy simulations have in store for us.

Maureen Teyssier, Rutgers University

To further our understanding of the nature and evolution of cosmic structures, black holes, and dark matter, scientists use high-resolution simulations to follow the formation of galaxies ranging from the mass of the Milky Way to those 100 times less massive. Using a new code called ChaNGA (Charm N-body GrAvity solver) on NASA’s Pleiades supercomputer, researchers from Rutgers University and the University of Washington are running the most realistic galaxy simulations ever produced, resolving structures down to the scale of star clusters only a few hundred light years across. The cutting-edge ChaNGA simulations use an accurate physical model that captures both gravity and gas hydrodynamics, and includes many processes that act on both large and small scales. Results from these simulations have already revolutionized scientists’ view of galaxy formation by showing that supernovae in high-density regions where stars are born can change the dark matter profile of galaxies. Insights from these simulations explain a number of long-standing challenges in galaxy formation theory, and are being used to help interpret data gathered by NASA’s Hubble Space Telescope.

Quick Facts

• The ChaNGA simulations resolve galaxy structures down to several hundred light years (about 100 parsecs), and explain several long-standing observational challenges to galaxy formation theory based on the standard Lambda Cold Dark Matter (CDM) cosmological model.
• Simulation results are being used to quantify the redistribution of matter in galaxies, to explore the growth of black holes, and to determine whether the ultraviolet light from stars in galaxies can "escape" to re-ionize the universe.
• Results are helping to interpret observations gathered by NASA missions such as the Hubble Space Telescope.
• The ChaNGA code scales efficiently to hundreds of thousands of cores to efficiently run large-scale galaxy simulations, which can require up to 1 million processor hours on NASA’s Pleiades supercomputer.

Key Image

Image Caption

The filamentary nature (“cosmic web”) of dark matter in a volume of the universe, from a simulation produced with the ChaNGA (Charm N-body GrAvity ) code. In this image, the universe is only 3.5 billion years old. Brighter points indicate denser regions. The volume is 81.5 million light years (25 megaparsecs) per side, and contains 2 billion dark matter particles. Dense knots indicate highly dense regions where galaxies form and cluster together. Maureen Teyssier, Rutgers University; Andrew Pontzen, University College London
Download High-Res Image (11 MB)

More Media

See additional videos related to this demo in our media gallery.

Contacts

More Information

• SC14 Demo Abstract
• Pleiades Supercomputer
• University of Washington Astronomy Department
• Rutgers Department of Physics and Astronomy
• ChaNGA (Charm N-body GrAvity solver)