Gravitational waves are generated when massive objects undergo huge non-spherical accelerations. Pairs of compact objects, black holes and/or neutron stars, that become gravitationally bound to each other are among the strongest sources of gravitational waves that we know of today. The gravitational waves emitted by these binary compact objects carry energy away from the system. This loss of energy is balanced by a tightening of the orbit, which results in an increase in the speed of the objects (and hence their orbital acceleration) and the strength of the gravitational waves emitted. Over time, the compact objects gradually spiral inward and eventually crash into each other in a violent cosmic event.
This paper describes the first joint search for gravitational waves from binary mergers with data from the Laser Interferometer Gravitational-wave Observatory (LIGO) and Virgo detectors. LIGO and Virgo aim to detect gravitational-wave signals between 40 and 1000 Hz. It is significant to have at least three different sites (two LIGO and one Virgo) observing data because this allows a direction of origin to be determined for any gravitational-wave signal that is coincidently detected at each location.
Each LIGO/Virgo detector is an interferometer with two orthogonal arms. Laser beams travel down the individual arms to suspended mirrors and bounce back. If gravitational waves have 'stretched' or 'compressed' space-time along the directions of the arms, then it would be evident in the interference pattern formed when the the two laser beams meet back at the center of the interferometer. LIGO operates a 4-km detector in Livingston, Louisiana and two colocated detectors (2-km and 4-km) in Hanford, Washington. The 3-km Virgo detector is in Cascina, Italy.
This search looks for gravitational-wave signals that have waveforms similar to theoretical predictions for compact binary mergers. It focuses on mergers of objects with individual masses between 1 and 34 solar masses and no more than a total of 35 solar masses. Single detector data are matched to theoretical waveforms, however this procedure alone leads to a high false alarm rate (labeling noise as signal). To reduce the false alarm rate, we require time and mass coincidence to identify a signal - meaning that events must occur at the same time and with the same mass pair in multiple detectors. The methodology adopted in this search also represents a substantial advance over previous searches for binary mergers because it takes into account the fact that the sensitivity varies for each detector.
Although no gravitational waves were identified, an upper limit was set on the rate of gravitational waves from binary mergers. The upper limit establishes how loud the gravitational waves could be and still be consistent with the results of this search.