## Final Activity Report Summary - RSSGWA (Realistic Supercomputer Simulations for Gravitational Wave Asteroseismology)

Many astrophysical phenomena can be simulated on computers using methods known as computational hydrodynamics. If typical velocities in the system are small and gravity is weak, it is sufficient to use the simple Newtonian approximation of the laws of motion and gravity.

However, in many environments, these approximations do not hold. To describe large amounts of matter compressed on small scales, one must resort to Einstein's theory of general relativity, a generalisation of Newton's theory of gravity. Such a situation is encountered near black holes (the prospective driving engines of astrophysical jets), as well as in core collapse supernovae, in collapsars (one possible source of gamma-ray bursts), or in neutron stars.

Some of these scenarios are emitting gravitational radiation. While light or sound waves propagate through spacetime, gravitational waves are ripples of spacetime itself. Such spacetime distortions have been predicted by Einstein in his general theory of relativity over 80 years ago, and are planned to be measured by laser interferometers or resonant bar detectors. Such experiments (like the British-German project GEO600, LIGO in the U.S., the French-Italian project VIRGO, the future E.U.-U.S. mission LISA, or IGEC in Japan) have recently started very sensitive measurements, and the first successful direct detection of gravitational waves can be envisaged within the next few years.

In order to accomplish a successful detection of gravitational waves, very efficient electronic filters have to be employed to extract a possible signal from the data measured by a detector. It is therefore of great importance to predict as precise as possible the signals from theoretical models of various astrophysical sources of gravitational radiation. This is typically achieved by either computing detailed so-called waveform templates which provide the qualitative shape of a complicated emitted radiation field, or by simply specifying some parameters like frequency and amplitude if the radiation is monochromatic.

In the projects related to this fellowship I have performed the up-to-date most realistic supercomputer-based numerical simulations of both rotational stellar core collapse to a hot proto-neutron star and pulsating cold neutron stars in general relativity. I have computed waveform templates of the gravitational radiation for a wide variety of models, for the first time including different microphysical zero or non-zero temperature equations of state (which specify the properties of matter at such high densities), neutrino effects, magnetic fields, and relativistic gravity.

For core collapse events I have found a very generic waveform type, which is in stark contrast to previous, less sophisticated simulations. In the case of pulsating neutron stars (whose oscillations may be excited by regular stellar core collapse, a contraction due to a phase transition to quark matter in the centre, or a migration from an unstable rotation state) I have extended the theory of asteroseismology, which describes how the frequency of those oscillations depends on other physical properties of the system, to rotating configurations.

I have also paved the way to simulate the formation of very compact objects like black holes with the mathematical formulation that is the basis of our computer code. For this I have discovered a reformulation of the equations that removes prohibitive obstacles like numerical instabilities and mathematical degeneracies. In the course of that aspect I have worked together with experts in mathematics and numerical theory.

This work has been done in close collaboration with more than 20 scientists from research groups in 7 different E.U. countries and the U.S. I have made extensive use of supercomputing facilities in various countries. I have actively disseminated these results in international meetings to the respective audience, I have created a publicly accessible website for gravitational radiation templates (which includes many details also in public outreach format), and I have given several lectures in summer schools to graduate students and young postdoctoral researchers.

However, in many environments, these approximations do not hold. To describe large amounts of matter compressed on small scales, one must resort to Einstein's theory of general relativity, a generalisation of Newton's theory of gravity. Such a situation is encountered near black holes (the prospective driving engines of astrophysical jets), as well as in core collapse supernovae, in collapsars (one possible source of gamma-ray bursts), or in neutron stars.

Some of these scenarios are emitting gravitational radiation. While light or sound waves propagate through spacetime, gravitational waves are ripples of spacetime itself. Such spacetime distortions have been predicted by Einstein in his general theory of relativity over 80 years ago, and are planned to be measured by laser interferometers or resonant bar detectors. Such experiments (like the British-German project GEO600, LIGO in the U.S., the French-Italian project VIRGO, the future E.U.-U.S. mission LISA, or IGEC in Japan) have recently started very sensitive measurements, and the first successful direct detection of gravitational waves can be envisaged within the next few years.

In order to accomplish a successful detection of gravitational waves, very efficient electronic filters have to be employed to extract a possible signal from the data measured by a detector. It is therefore of great importance to predict as precise as possible the signals from theoretical models of various astrophysical sources of gravitational radiation. This is typically achieved by either computing detailed so-called waveform templates which provide the qualitative shape of a complicated emitted radiation field, or by simply specifying some parameters like frequency and amplitude if the radiation is monochromatic.

In the projects related to this fellowship I have performed the up-to-date most realistic supercomputer-based numerical simulations of both rotational stellar core collapse to a hot proto-neutron star and pulsating cold neutron stars in general relativity. I have computed waveform templates of the gravitational radiation for a wide variety of models, for the first time including different microphysical zero or non-zero temperature equations of state (which specify the properties of matter at such high densities), neutrino effects, magnetic fields, and relativistic gravity.

For core collapse events I have found a very generic waveform type, which is in stark contrast to previous, less sophisticated simulations. In the case of pulsating neutron stars (whose oscillations may be excited by regular stellar core collapse, a contraction due to a phase transition to quark matter in the centre, or a migration from an unstable rotation state) I have extended the theory of asteroseismology, which describes how the frequency of those oscillations depends on other physical properties of the system, to rotating configurations.

I have also paved the way to simulate the formation of very compact objects like black holes with the mathematical formulation that is the basis of our computer code. For this I have discovered a reformulation of the equations that removes prohibitive obstacles like numerical instabilities and mathematical degeneracies. In the course of that aspect I have worked together with experts in mathematics and numerical theory.

This work has been done in close collaboration with more than 20 scientists from research groups in 7 different E.U. countries and the U.S. I have made extensive use of supercomputing facilities in various countries. I have actively disseminated these results in international meetings to the respective audience, I have created a publicly accessible website for gravitational radiation templates (which includes many details also in public outreach format), and I have given several lectures in summer schools to graduate students and young postdoctoral researchers.