On the Spectral Peak Energy of Swift Gamma-Ray Bursts

Owing to the narrow energy band of the Swift Burst Alert Telescope (BAT), several urgent issues remain unsolved. We systematically study the properties of a refined sample of 283 Swift/BAT gamma-ray bursts (GRBs) with well-measured spectral peak energy (E_p) at a high confidence level greater than 3σ. We find that the duration (T₉₀) distribution of Swift bursts still exhibits an evident bimodality with a more reliable boundary of T₉₀ ≃ 1.06 s instead of 2 s as found for previously contaminated samples, including bursts without well-peaked spectra, which is very close to the ∼1.27 and ∼0.8 s values suggested in the literature for the Fermi/Gamma-ray Burst Monitor and Swift/BAT catalogs, respectively. The Swift/BAT short and long bursts have comparable mean E_p values of ${87}_{-49}^{+112}$ and ${85}_{-46}^{+101}$ keV, similar to what was found for both types of BATSE bursts, which indicates that the traditional short-hard/long-soft scheme may not be tenable for certain detector energy windows. We also statistically investigate the consistency of distinct methods for E_p estimates and find that a Bayesian approach and BAND function (Band et al.) can always provide consistent evaluations. In contrast, the frequently used cutoff power-law model matches two other methods for lower E_p and overestimates the E_p by more than 70%, as E_p > 100 keV. Peak energies of X-ray flashes, X-ray-rich bursts, and classical GRBs could be an evolutionary consequence of moving from thermal-dominated to nonthermal-dominated radiation mechanisms. Finally, we find that the E_p and the observed fluence (S_γ) in the observer frame are correlated as ${E}_{p}\simeq {[{S}_{\gamma }/({10}^{-5}\mathrm{erg}{\mathrm{cm}}^{-2})]}^{0.28}\times {117.5}_{-32.4}^{+44.7}$ keV, which might be a useful indicator of GRB peak energies.