Prompt emission from GRB 150915A in the GeV energy range detected at ground by the New-Tupi detector: A review

TL;DR

This paper reports on the detection of prompt GeV emission from GRB 150915A using the ground-based New-Tupi detector, analyzing its significance and likelihood of being a background fluctuation.

Contribution

It presents the first analysis of GeV-range prompt emission from GRB 150915A detected by a ground-based scintillator array, using the scaler technique.

Findings

Detected a counting rate excess with 4.4 sigma significance

Estimated a lower limit fluence of 1.3 x 10^-6 erg/cm^2 in GeV range

Event likely a background fluctuation with 5 x 10^-6 probability

Abstract

Since 2014, a new detector (New-Tupi) consisting of four plastic scintillators ($150 \times 75 \times 5 cm^3$) placed in pairs and located in Niteroi, Rio de Janeiro, Brazil, has been used for the search of transient solar events and photomuons from gamma-ray bursts (GRBs). On September 15, 2015, at 21:18:24 UT, the Swift Burst Alert Telescope (BAT) triggered and located GRB 150915A (trigger 655721). The GRB light curve shows a weak complex structure of long duration $T_{90}=164.7 \pm 49.7 $ s, and a fluence in the 15-150 keV band of $8.0 \pm 1.8 \times 10^{-7}erg/cm^2$. GRB 150915A was fortuitously located in the field of view of the New-Tupi detector, and a search for prompt emission in the GeV energy range is presented here. The analysis was made using the "scaler" or "single-particle" technique. The New-Tupi detector registered a counting rate excess peak of duration $T_{90}=(6.1\pm 0.6)$ s with a signal significance $(4.4\pm 0.5)\sigma$, (and not $6.9\sigma$ as reported in the previous version). The signal is within the T90 duration of the Swift BAT GRB, with an estimated "excess" fluence of $F_S(E>0.1 GeV)=1.3 \pm 0.3 \times 10^{-6} erg/cm^2$. This value can be considered the lower limit of the gamma ray fluence in the GeV energy region. However, the Poisson probability of the event to be a background fluctuation is $5.0 \times 10^{-6}$ and it appears in the counting rate of the New-Tupi detector with an annual rate $\sim 76$. In addition, the signal has a significance of only $2\sigma$ in the time profiles with a bin above 2 seconds. Thus we conclude that the event has a high probability to be background fluctuation.