All-sky search for long-duration gravitational-wave transients in the second Advanced LIGO observing run

We present the results of a search for long-duration gravitational-wave transients in the data from the Advanced LIGO second observation run; we search for gravitational-wave transients of 2–500 s duration in the 24–2048 Hz frequency band with minimal assumptions about signal properties such as waveform morphologies, polarization, sky location or time of occurrence. Signal families covered by these search algorithms include fallback accretion onto neutron stars, broadband chirps from innermost stable circular orbit waves around rotating black holes, eccentric inspiral-merger-ringdown compact binary coalescence waveforms, and other models. The second observation run totals about 118.3 days of coincident data between November 2016 and August 2017. We find no significant events within the parameter space that we searched, apart from the already-reported binary neutron star merger GW170817. We thus report sensitivity limits on the root-sum-square strain amplitude $h_{\mathrm{rss}}$ at 50% efficiency. These sensitivity estimates are an improvement relative to the first observing run and also done with an enlarged set of gravitational-wave transient waveforms. Overall, the best search sensitivity is $h_{\mathrm{rss}}^{50\%}=2.7\times {10}^{-22}{\mathrm{Hz}}^{-1/2}$ for a millisecond magnetar model. For eccentric compact binary coalescence signals, the search sensitivity reaches $h_{\mathrm{rss}}^{50\%}=9.6\times {10}^{-22}{\mathrm{Hz}}^{-1/2}$.

Figure 1

Time-frequency representations of a few model signals used in the search, showing a mix of chirp-up (FA, ECBC) and chirp-down (Magnetar, ADI) astrophysical waveforms as well as a linearly decreasing ad hoc waveform (LINE). Descriptions of these waveforms and others are given in Sec. 4. The harmonics of ECBC are also visible. The full set of waveforms ($\sim 70$) chosen for this analysis fully covers the search frequency band of 24–2000 Hz. The waveforms are shifted in time to show how they cover the parameter space in this axis as well.

Figure 2

Upper limits on gravitational-wave strain vs frequency for sources detected with 50% efficiency and a false alarm rate of 1 event in 50 years. The lowest value among all four pipelines is represented on the plots. The left figure shows the ad hoc waveforms’ results while the “physical” waveforms are represented on the right. The average amplitude spectral density curves for both Hanford and Livingston are also shown.

Figure 3

Upper limits (marginalizing over the second observation run amplitude calibration errors) on eccentric compact binary coalescences as a function of the distance at a 90% confidence level considering the best results for each waveform. The inset shows the distance at 50% detection efficiency for the pipelines in this analysis for comparison. ECBC_A, ECBC_B, and ECBC_C are $1.4–1.4$ solar mass binaries with eccentricities of 0.2, 0.4, and 0.6 respectively, while ECBC_D, ECBC_E, and ECBC_F are $3.0–3.0$ solar mass binaries with eccentricities of 0.2, 0.4, and 0.6 respectively, where the masses are quoted in the detector frame.