Abstract
Gravitational wave (GW) bursts (short duration signals) are expected to be associated with highly energetic astrophysical processes. With such high energies present, it is likely these astrophysical events will have signatures in the EM spectrum as well as in gravitational radiation. We have initiated a program, "Locating and Observing Optical Counterparts to Unmodeled Pulses in Gravitational Waves" (LOOC UP) to promptly search for counterparts to GW burst candidates. The proposed method analyzes near real-time data from the LIGO-Virgo network, and then uses a telescope network to seek optical-transient counterparts to candidate GW signals. We carried out a pilot study using S5/VSR1 data from the LIGO-Virgo network to develop methods and software tools for such a search. We will present the method, with an emphasis on the potential for such a search to be carried out during the next science run of LIGO and Virgo, expected to begin in 2009.
Introduction
Gravitational wave detectors like LIGO and Virgo are "all-sky" instruments—they listen to the whole universe at once. However, they are notoriously bad at pinpointing where a signal is coming from. A typical GW detection might only be localized to a "patch" of sky covering tens of square degrees, which is far too large for most high-resolution telescopes to search efficiently.
Furthermore, GW signals are often buried in noise. A low-threshold signal (SNR 5-10) might be an astrophysical event or just a glitch in the instrument. Finding a simultaneous "optical counterpart"—a flash of light from the same event—would provide the "smoking gun" evidence needed to confirm a discovery, identify a host galaxy, and determine the event's distance (redshift).
The LOOC UP Pipeline
The authors propose and test a pipeline called LOOC UP (Locating and Observing Optical Counterparts to Unmodeled Pulses in Gravitational Waves). The workflow consists of four main steps:
- Real-Time Analysis: GW data is Fourier-transformed to find "bursts" (excess power in the time-frequency plane).
- Triple Coincidence: To filter out noise, the system looks for signals appearing simultaneously (within light-travel time) across three different detector sites (e.g., Hanford, Livingston, and Virgo).
- Position Estimation: Using "time-of-flight" (the slight delay in when the signal hits each detector), the system calculates a rough sky position.
- Galaxy Catalog Filtering: This is the "secret sauce." Because the error region is so large, the team uses a catalog of known galaxies within the detectors' range (initially ~20 Mpc). They only trigger telescope observations for the specific galaxies that fall within the GW error ellipse, drastically reducing the amount of sky that needs to be imaged.
The paper discusses two computational approaches: Incoherent (fast, uses simple trigger properties like peak time) and Coherent (slower but more accurate, combining raw data streams to resolve polarizations and break sky-position degeneracies).
Localization and Galaxy Targeting
The "natural" direction for multi-messenger astronomy is using GW-triggered EM searches because GW detectors see the whole sky, while telescopes must be pointed. Without a galaxy catalog, a standard telescope would need to take dozens of images to cover a GW error region. By targeting likely host galaxies, the search becomes feasible for narrow-field instruments.
To catch the "afterglow" of events like neutron star mergers (which peak and fade over hours or days), the pipeline must move fast. The pilot achieved a latency of ~45 minutes, with the goal of reaching under one minute. Furthermore, GW triggers might help find "orphan" afterglows—events where the high-energy gamma rays are beamed away from Earth, but the gravitational waves and slower optical light are visible to us.
Data and Pilot Study Results
The study utilized several key data sources:
- LIGO S5 / Virgo VSR1: Data from the 2007 science runs were used for the pilot study.
- Galaxy Catalog: A modified version of the LSC Compact Binary Coalescence group's catalog, containing 2,766 objects within 20 Mpc, including Milky Way globular clusters.
- Observational Data: During the pilot, the team used the 2.4m Hiltner and 1.3m McGraw-Hill telescopes (MDM Observatory) and the 1m Swope telescope (Chile).
The team successfully ran a pilot study in 2007, demonstrating that a real-time GW-to-telescope pipeline is logistically possible. Over three observing runs, the team identified and imaged 95 targets based on GW triggers. The paper estimates that by extending the search to 40 Mpc (the expected range for the 2009 S6 run), they would find roughly one target galaxy per square degree, making the search even more effective.
Limitations and Future Work
With only three detectors, the time-of-flight method often results in two distinct "mirror" regions on the sky. Moving toward "coherent" analysis or adding more detectors (like the planned Advanced LIGO/Virgo) is necessary to resolve this. Additionally, if a signal has complex polarization, the "peak time" might appear different to different detectors, leading to a 5–10 degree error in positioning.
While the triggers are fast, the authors note that they are still working on a real-time image processing pipeline to automatically identify transients among the thousands of stars in their images. Future iterations aim for fully robotic telescope integration to minimize human-induced delays.
Summary updated