Max Planck Institute for Gravitational Physics (Albert Einstein Institute), Potsdam-Golm (Germany)
Local Project ID:
HPC Platform used:
SuperMUC of LRZ
Scientists from the Albert Einstein Institute in Potsdam/Germany used HPC system SuperMUC to study the dynamics of compact-object binaries, i.e. neutron stars and black holes, and improve our understanding of strong gravity.
One of the last predictions of general relativity that still awaits direct observational confirmation is the existence of gravitational waves. Those fluctuations of the geometry of space and time are expected to travel with the speed of light and are emitted by any accelerating mass. Only the most violent events in the universe, such as mergers of two black holes or neutron stars, produce gravitational waves strong enough to be measured with present detectors. These waves are extremely weak when arriving at Earth, and their detection is a formidable technological challenge for advanced detectors such as Virgo, and LIGO, which are expected to observe around 40 events per year.
To interpret the observational data, theoretical modelling of the sources is a necessity and requires numerical simulations of the equations of general relativity and relativistic hydrodynamics. Such computations can only be carried out on large-scale supercomputers, given that many scenarios need to be simulated, each of which typically occupies 500 CPU cores over 10 days.
The main goal of the research group at the Max Planck Institute for Gravitational Physics (Albert Einstein Institute) in Potsdam/Germany under leadership of Prof. Luciano Rezzolla is to predict the gravitational wave signal from the merger of two compact objects. Comparison with future observations will provide important insights into the fundamental forces of nature in regimes that are impossible to recreate in laboratory experiments. The waveforms from binary black-hole mergers will allow to test the correctness of general relativity in previously inaccessible regimes. The signal from binary neutron stars mergers will provide input for nuclear physics, because the signal depends strongly on the unknown properties of matter at the ultra high densities inside neutron stars, which cannot be observed in any other astrophysical scenario. Besides mergers, the team also improved the analytic models of close encounters, i.e. black-hole scattering, providing information on which approximations provide accurate estimates.
The results might be used to determine the distance to a source, by exploiting the fact that observed frequencies of distant sources are reduced due to the expansion of the universe. Independent distance measures are very valuable for cosmology, improving estimates of the current and past expansion rate of the universe and predictions of its future fate. Finally, the simulations also helped to shed some light on the mystery of so called short gamma ray bursts, intense and sudden bursts of gamma radiation that puzzled astronomers for (since) decades, and which are believed to be caused by neutron-star mergers. In particular, the researchers found that the strongly magnetized and rapidly rotating merger remnant could be used to explain the X-ray afterglows which often accompany those bursts.
Computations were executed on HPC system SuperMUC of LRZ Garching.
W. Kastaun, L. Rezzolla, D. Siegel, K. Takami
Max Planck Institute for Gravitational Physics (Albert Einstein Institute), Potsdam-Golm
Science Park Potsdam-Golm
Am Mühlenberg 1, D-14476 Potsdam-Golm/Germany