MacResearch

This project was about studying an astrophysical event that commonly occurs in galactic nuclei i.e. the capture of a small star by a supermassive black hole. The goal of the project was to use computer simulations to obtain information about the gravitational waves that are radiated away during such events. Gravitational waves are "ripples" in space-time that travel at the speed of light. These were theoretically predicted by Einstein's Relativity, but have never been directly observed. Currently, there is an extensive search being performed for these waves by the newly constructed NSF LIGO laboratory and various other such observatories in Europe and Asia. NASA also has a mission planned in the near future, the LISA mission, that will also be attempting to detect these waves. To learn more about these waves and the recent attempts to observe them, please visit the LISA mission website.

We performed a large set of simulations that correspond to different orbits typically taken by small stars, as they are captured by a massive black hole. In this manner, we obtained various patterns of emitted gravitational radiation and also their intensities. The results will not only help in establishing the type of events these various gravitational wave observatories would be able to witness, it would also help in understanding certain aspects of the astrophysics behind these complex systems. In time, results from these simulations will be published in a research journal and OpenMacGrid and its contributors will be formally acknowledged.

Each available Mac on the OpenMacGrid ran its own copy of our "Teukolsky Code" and had its own orbit to work on. Orbits are characterized by three parameters: semi-latus rectum, eccentricity and inclination angle. We distributed orbits based on a large set of appropriately chosen values for these parameters and also included some redundancy. In total, over 600 runs were completed in just over a week! Such a large number of runs would have taken several months to finish on the modest number of workstations that our research group has. The OpenMacGrid ran at its full current capacity of about 200 GHz (approximately 70 Macs) for several days, but there were a few days during the week when it maintained a rather low performance of about 40 GHz due to the unavailability of a large number of Macs from one laboratory. As agents from a variety of sources from within the community join, there would little chance of that happening again.

Below, we include some sample results from our simulations. The orbit under consideration here, is a "zoom-whirl" orbit around a spinning black hole. The basic idea is that for certain elliptic orbits the star spends a large amount of time close to the black hole, so that it traces a quasi-circular path close to the hole before traveling back further out. For high-eccentricity orbits, the star appears to “zoom in” towards the black hole, spend a number of circular revolutions (“whirls”) and then “zoom out”. These orbits are very interesting, especially for the case of rapidly spinning black holes, because of the large number of “whirls” they perform close to the hole. When the star is close to the hole, it radiates very strongly, which makes it very relevant for observational purposes. In the plots below, we show a sample "zoom-whirl" orbit and the resulting gravitational wave pattern ("waveform"). I won't get into the units used on these plots, but it should be sufficient to know that in the waveform plot, the y-axis is a measure of the amplitude of the wave, while the x-axis is a measure of time.

In the waveform plot, ignore the noise in the first "burst" of radiation. It is due to a well understood, unphysical artifact in our code and it should be ignored. The repetitive "bursts" you see in the waveform are from the "whirls" in the orbit, because that is when the star is really close to the black hole and interacting strongly with it.

Looks like my students have plenty of data to play with for a while!

Gaurav Khanna, Gravity Group at UMass Dartmouth, MA.