Speaker
Mr
Soichiro Morisaki
(The University of Tokyo)
Description
Accurate, precise and prompt sky localization of gravitational-wave sources is essential to the success of multi messenger astronomy. One of the most accurate sky localizations we can obtain is from full signal parameter estimation obtained with the LIGO-Virgo LALInference software [1], done after a quick initial sky localization with the Bayestar software [2]. While more accurate, the improved analysis takes on the order of a day to complete for a binary neutron star event. In that time-scale, an optical counterpart signal has a high chance of being missed. In order to speed up that process, we designed a new automation scheme and improved significantly the speed of LALInference parameter estimation.
For the automation, we integrated LALInference parameter estimation with an infrastructure for automatic follow ups on gravitational-wave alerts, GWCelery [3]. Currently, it starts within less than 10 minutes after the detection of gravitational waves, and includes optimised choices for parameter estimation settings.
For the acceleration, we formulated a technique to restrict prior ranges of source parameters before parameter estimation utilizing the preliminary information provided by the detection pipelines. We found that the parameter estimation is sped up by an order of magnitude with this technique, delivering an improved sky localization in potentially less than an hour.
In this talk, I will explain the detail of the automatic infrastructure and our technique to speed up the parameter estimation, and I will also summarize the current status of the development of the infrastructure for the accurate and precise sky localization.
[1]: J. Veitch et al., "Parameter estimation for compact binaries with ground-based gravitational-wave observations using the LALInference software library", Phys. Rev. D 91, no. 4, 042003 (2015). [2]: L. P. Singer and L. R. Price, "Rapid Bayesian position reconstruction for gravitational-wave transients," Phys. Rev. D 93, no. 2, 024013 (2016). [3]: https://git.ligo.org/emfollow/gwcelery .
Primary author
Mr
Soichiro Morisaki
(The University of Tokyo)
Co-author
Dr
Vivien Raymond
(Cardiff University)
