Plans for the LIGO–TAMA joint search for gravitational wave bursts

P. J. Sutton, Masaki Ando, P. R. Brady, L. Cadonati, A. Di Credico, S. Fairhurst, L. S. Finn, Nobuyuki Kanda, E. Katsavounidis, S. Klimenko, A. Lazzarini, Szabolcs Márka, J. W. C. McNabb, Saikat Ray Majumder, P. R. Saulson, Hideyuki Tagoshi, Hirotaka Takahashi, Ryutaro Takahashi, Daisuke Tatsumi, Y. Tsunesada, S. E. WhitcombClassical 2004 · DOI

TL;DR

This paper outlines the strategy for a joint search for gravitational wave bursts (GWBs) between the LIGO and TAMA collaborations, using data from early 2003. The core challenge is combining data from observatories with different characteristics, leading to a conservative search approach focused on high-frequency signals to minimize false alarms.

What Problem Does This Paper Address?

Detecting gravitational waves (GWs) is a monumental task, and confirming a signal requires high confidence to distinguish it from instrumental noise. While individual gravitational wave detectors like LIGO and TAMA are powerful, combining data from multiple observatories offers significant advantages: reduced false alarm rates, improved sky localization, and better signal characterization. However, these observatories are not identical. They have different noise profiles across the frequency spectrum and their antenna patterns (how sensitive they are to GWs from different directions in the sky) are misaligned. This paper addresses the challenge of designing a joint search that leverages the strengths of both LIGO and TAMA while accounting for these crucial differences, particularly for unmodeled GWBs from events like supernovae or black hole mergers, where theoretical templates are scarce.

Methods

The paper details an "untriggered" search for unmodeled GWBs, meaning they aren't looking for specific waveform shapes. Instead, each collaboration independently identifies candidate GWBs based on their own algorithms. These candidates are then compared for coincidence across all four interferometers (three LIGO detectors and one TAMA detector).

Key aspects of their method include:

  • Independent Candidate Generation: Each observatory uses its own burst detection algorithms. LIGO events are characterized by frequency range, while TAMA uses a fixed frequency band (700-2000 Hz).
  • Coincidence Requirement: A candidate GWB must be detected simultaneously (within a specified time window) by all four interferometers to be considered. This is the primary method for reducing false alarms.
  • Conservative Frequency Range: To maximize the chance of coincident detection by both LIGO and TAMA, the search focuses on the frequency range where their sensitivities overlap significantly, approximately 700-2000 Hz. This is a compromise, as it might miss low-frequency signals detectable by LIGO alone.
  • False Alarm Estimation: Accidental coincidences due to background noise are estimated by artificially shifting time stamps of the data.
  • Efficiency Estimation: Simulated signals are injected into the data to estimate the network's efficiency in detecting real GWBs.

Datasets

The paper uses data collected during the LIGO Science Run 2 (S2) and the TAMA Data Taking Run 8 (DT8), which occurred between February 14 and April 14, 2003. This dataset represents a specific period of operation for these gravitational wave observatories.

Key Insights and Ideas

  • The Power of Collaboration: Combining data from different observatories is crucial for increasing confidence in GW detection and reducing false alarms, even with the added complexity.
  • The Trade-off Between Sensitivity and False Alarms: Requiring coincidence across multiple, differently characterized detectors inherently reduces the network's overall sensitivity compared to individual detectors. However, this loss in sensitivity is traded for a significant reduction in the false alarm rate, which is paramount for making a credible detection.
  • Frequency Band Selection is Critical: The choice of the search frequency band (700-2000 Hz) is a direct consequence of the differing noise spectra of LIGO and TAMA. Focusing on this overlap region ensures that both types of detectors have comparable sensitivity, increasing the likelihood of a coincident detection.
  • Antenna Pattern Misalignment Complicates Sky Localization: The fact that LIGO and TAMA are most sensitive to different parts of the sky means that the combined network's sensitivity to any given point in the sky is limited by the least sensitive detector in that direction. This complicates sky localization efforts.

Key Results

The paper doesn't present final detection results but rather the plans for the search. The key "result" is the established strategy:

  • A conservative coincidence search requiring detection by all four interferometers.
  • A focus on the 700-2000 Hz frequency band to maximize overlap in sensitivity between LIGO and TAMA.
  • This approach is expected to yield a lower sensitivity of the network compared to individual interferometers but is the most conservative method for minimizing false alarms.

Limitations and Future Work

  • Reduced Sensitivity: The paper explicitly acknowledges that the requirement for four-fold coincidence, especially with differing detector characteristics, leads to a reduced network sensitivity compared to individual detectors.
  • Low-Frequency GWBs: The chosen frequency band means that low-frequency GWBs detectable by LIGO alone might be missed by this joint search. However, the authors note that LIGO conducts a separate low-frequency search, mitigating this risk.
  • Coherence Tests for LIGO-TAMA: While coherence tests are effective for LIGO-only searches, extending them to cross-correlate LIGO and TAMA events is noted as an area requiring further study due to alignment differences.
  • Future Work: The paper positions this joint search as a prototype for more comprehensive collaborative analyses in the future, suggesting that lessons learned here will inform subsequent, potentially more sensitive, searches. The study of extending coherence tests between LIGO and TAMA is also highlighted as future work.

Summary updated

Abstract

We describe the plans for a joint search for unmodelled gravitational wave bursts being carried out by the LIGO and TAMA Collaborations using data collected during February-April 2003. We take a conservative approach to detection, requiring candidate gravitational wave bursts to be seen in coincidence by all four interferometers. We focus on some of the complications of performing this coincidence analysis, in particular the effects of the different alignments and noise spectra of the interferometers.
BibTeX
@inproceedings{sutton2004plans,
  author = {P. J. Sutton and Masaki Ando and P. R. Brady and L. Cadonati and A. Di Credico and S. Fairhurst and L. S. Finn and Nobuyuki Kanda and E. Katsavounidis and S. Klimenko and A. Lazzarini and Szabolcs Márka and J. W. C. McNabb and Saikat Ray Majumder and P. R. Saulson and Hideyuki Tagoshi and Hirotaka Takahashi and Ryutaro Takahashi and Daisuke Tatsumi and Y. Tsunesada and S. E. Whitcomb},
  title = {{Plans for the LIGO–TAMA joint search for gravitational wave bursts}},
  booktitle = {Classical},
  year = {2004},
  doi = {10.1088/0264-9381/21/20/023},
}
Figure