Speaker
Description
Gamma-ray experiments like BATSE, the Fermi GBM and LAT, and SWIFT have provided a wealth of data on gamma-ray bursts (GRBs), but prompt observations of multiwavelength counterparts of short bursts have been limited. To facilitate such observations, this work describes progress toward production of real-time GRB alerts that include detailed likelihood maps computed entirely aboard a gamma-ray telescope. Real-time, onboard map generation is a planned feature of the Advanced Particle-astrophysics Telescope (APT) and the Antarctic Demonstrator for APT (ADAPT), which will enhance these instruments' contributions to GRB science.
When a GRB occurs, observations of its gamma rays can alert instruments with narrower fields of view to look for prompt or afterglow emission in other wavelength bands. Particularly for short-duration GRBs, whose optical counterparts may fade within seconds to minutes, rapid production of an informative alert is critical to direct fast-slewing ground-based instruments toward the GRB. Examples of such instruments that could cooperate with a future ADAPT mission include 7DT, ATOM, LSGT, RAPTOR, and TURBO. Future orbital missions may include such secondary instruments on board or as part of a constellation of detectors, which would eliminate communication delays to the ground and permit even faster follow-up observation.
GRB localization by a gamma-ray instrument is subject to uncertainty due to measurement limitations and background radiation. The primary product of a GRB alert is therefore not a single source location but rather a likelihood map, which divides the sky into pixels and assigns a likelihood that the GRB lies within each pixel. Likelihood map production from a set of observed gamma rays is compute-intensive and so is typically deferred to ground-based computation, which increases the time to produce an alert, perhaps to minutes or even hours, and reduces the time available for follow-up observation of short-duration GRBs.
This work demonstrates the feasibility of real-time, onboard likelihood map generation for GRB alerts. We produce likelihood maps with resolutions of 2-4 degrees, built using a detailed instrument response matrix and fitting of the GRB's intensity, in one second or less using computing capabilities that could be flown aboard a future gamma-ray observatory. To generate a map rapidly, we both exploit parallelism and devise multiresolution methods that focus computation on the small part of the sky that is likely to contain the GRB. We evaluate our approach using simulated GRBs and instrument response matrices computed for the planned Compton Spectrometer and Imager (COSI) mission and made available as part of their series of Data Challenges.
| Collaboration(s) | the Advanced Particle-astrophysics Telescope consortium |
|---|
