Mapping a universe of black hole collisions
Monday 10th December 2012
Anti-aligned spins of two black holes
Researchers in the gravitational physics group are about to perform the world's most complete study to date of colliding black holes, making use of a European supercomputer allocation of over 37 million CPU hours -- the equivalent of 24-hour continuous access to 4000 computers all of next year. Their goal is to solve Einstein's equations of general relativity to calculate how systems of two black holes will spiral together and merge, and, of most immediate interest, the pattern of gravitational waves that are produced in the process.
The urgency is because an international network of gravitational-wave detectors of unprecedented sensitivity is due to come online late next year, giving us our best chance yet of finally observing the gravitational waves that Einstein first predicted almost 100 years ago. Encoded within the gravitational waves will be information about what produced them -- whether it was colliding black holes or neutron stars, or something even more exotic -- and in order to get that information out, we will need to be able to match up the signals against the signals predicted by Einstein's theory.
Numerical relativists within the Cardiff gravitational physics group, led by Professor Sathyaprakash, in collaboration with other groups in Spain, Germany, India and the US, have just been awarded the 37-million CPU-hour allocation as part of the European PRACE high-performance computing initiative. Over the next year they plan to simulate a wide range of black-hole binary configurations, and from those simulations work out a model for any binary, which will in turn make it easier to observe gravitational waves from colliding black holes with the advanced LIGO and Virgo detectors. Their model will not take into account *all* binary configurations -- sometimes the black holes are spinning on their axis in such a way that they create wild, complicated orbits. But that's for the next year...