Abstract:
In this project, we simulated gravitational waves, disturbances in spacetime which are generated when masses are accelerated. Specifically, we focused on the gravitational waves from the mergers of binary systems of black holes and/or neutron stars. Since 2015, the observatories LIGO and Virgo have made observations of such gravitational waves. We used an outline of a computational simulation of gravitational waves by Buskirk and Babiuc-Hamilton (2019) and the underlying theoretical model they used by Huerta et al. (2017) as the basis to create a script that simulated the gravitational wave signals of binary mergers in the free-to-use programming language Python. We used a formula for the signalto- noise ratio (SNR) of gravitational waves by Barrett et al. (2018) and sensitivity curves from LIGO and Virgo to calculate SNRs for the gravitational waveforms from our first script. We then used these SNRs to calculate the horizon distances for the gravitational wave events and the probabilities for gravitational waves from a merger of given properties at a given distance to be detectable. We combined these probabilities with rates of the formation of mergers of specific properties in the Universe from the stellar population synthesis program BPASS (Eldridge, Stanway, & Tang, 2019; Eldridge et al., 2017) to create predictions of the rates at which gravitational wave events would be observable by the LIGO/Virgo detectors. Our gravitational waveform and SNR calculations passed several verification tests, including approximately matching real SNRs for the systems of observed gravitational wave transients, and the horizon distances matching the theoretical prediction that they should be proportional to the chirp mass raised to the (5/6)th power for low masses. Our final predicted rates for observable binary black hole mergers were equal to those observed in LIGO/Virgo’s third observing run to within experimental uncertainty. However, our predicted rates for black hole–neutron star mergers and binary neutron star mergers exceeded those observed by about a factor of two. We think that this discrepancy is caused not by the code created during this project but rather by the way the underlying BPASS data we used for the rates of formation mergers takes into account a phenomenon called a neutron star kick, in which neutron stars receive a peculiar velocity if the supernova that forms them is asymmetrical. The magnitude of such kicks would disproportionately affect the rates of formation of neutron star mergers, so we think the kick model of Bray and Eldridge (2018) that the BPASS data used is overestimating the kicks.