Possible Quasi-Period Oscillation Signals in the Unique Event of GRB 250702DBE/EP250702a?

TL;DR

This study analyzes the ultra-long GRB GRB 250702DBE/EP250702a using Fermi-GBM lightcurves in the $8$ keV–$1$ MeV band to search for quasi-periodic oscillations (QPOs) across three sub-bursts. It employs FFT and WWZ analyses, models the red-noise-dominated PSD with extensive simulations ($N=20{,}000$) to assess significance, and cross-checks with autoregressive (AR) spectral analysis. The authors identify plausible QPOs at $f=0.046~\mathrm{Hz}$ ($P\approx 21.7~\mathrm{s}$) in GRB 250702B and $f=0.024~\mathrm{Hz}$ ($P\approx 41.7~\mathrm{s}$) in GRB 250702E, but after accounting for the red noise and three-segment trials, the global false-alarm probabilities are modest (roughly $\sim$23% for B and $\sim$5% for E, with Monte Carlo results similar). Spectral modeling across $8$ keV–$40$ MeV favors a Band function for the broad-band spectrum, while the QPO interpretations lean toward periodic jet phenomena (e.g., a helical jet or precession) or micro-TDE activity around an intermediate-mass black hole in a dense environment, suggesting a potentially unique GRB class; nevertheless, deeper multiwavelength follow-up is required for robust conclusions.

Abstract

GRB 250702DBE was time-consequently triggered by GBM onboard the Fermi satellite. It is uncertain which celestial catalog is suitable for this special ultra-long event to belong to. In this paper, we comprehensively investigate the lightcurves obtained by Fermi-GBM detectors. In the energy band of 8-1000 keV, no Quasi-Period Oscillation (QPO) signals are found in the lightcurve of the first burst 250702D, a possible QPO signal of 0.046 Hz corresponding to a period of 21.7 s is found in the lightcurve of the second burst 250702B, and a possible QPO signal of 0.024 Hz corresponding to a period of 41.7 s is found in the lightcurve of last burst 250702E. The significance level of the possible QPO signals is comprehensively examined. In addition, we examine the spectral properties of the sources. In general, a broken power law is suitable for modeling the spectral data from 8 keV to 40 MeV. We qualitatively suggest some kinds of celestial object with the periodic characteristic that might be the progenitors of this unique event.

Figures (6)

• Figure 1: Analysis of GRB 250702D lightcuvre. Top panel: the 256 ms binned lightcurve of GRB 250702D (nb) in the energy band of 8-1000 keV. Middle panel: the PSD of FFT. The dash-dotted line shows the best-fit noise model, with green (95%), yellow (99%), and red ($3\sigma$) dashed lines showing the confidence intervals. Bottom-left panel: WWZ power. Bottom-right panel: WWZ time-integrated PSD and FFT PSD.
• Figure 2: Analysis of GRB 250702B lightcuvre. Top panel: the 256 ms binned lightcurve of GRB 250702B (nb) in the energy band of 8-1000 keV. Middle panel: the PSD of FFT. The dash-dotted line shows the best-fit noise model, with green (95%), yellow (99%), and red ($3\sigma$) dashed lines showing the confidence intervals. Bottom-left panel: WWZ power. Bottom-right panel: WWZ time-integrated PSD and FFT PSD.
• Figure 3: Analysis of GRB 250702E lightcuvre. Top panel: the 256 ms binned lightcurve of GRB 250702E (n8) in the energy band of 8-1000 keV. Middle panel: the PSD of FFT. The dash-dotted line shows the best-fit noise model, with green (95%), yellow (99%), and red ($3\sigma$) dashed lines showing the confidence intervals. Bottom-left panel: WWZ power. Bottom-right panel: WWZ time-integrated PSD and FFT PSD. Besides the QPO signal at 0.024 Hz, another possible QPO signal at 0.096 Hz that is exceeding 3$\sigma$ confidence level is noted here. However, this QPO signal is only detected in n8 detector, while it is not shown in the lightcurves from other detectors. Thus, we are not sure for this identification as a possible QPO signal.
• Figure 4: Top panels: PSDs obtained from AR(2) spectral analysis to GRB 250702B lightcurves with 256ms (left) and 16ms (right) time bins. Bottom panels: PSDs obtained from AR(2) spectral analysis to GRB 250702E lightcurves with 256ms (left) and 16ms (right) time bins.
• Figure 5: Top panel: PSD comparison between FFT (blue solid line) and AR(2) model (purple dash-dotted line) for GRB 250702B. Dashed lines show AR(2) confidence levels: green (95%), yellow (99%), red (3$\sigma$). Bottom panel: Same as top panel but for GRB 250702E. To clearly show the possible QPO signals compared to the AR(2) model, the low frequency ($<10^{-2}$ Hz) region is not displayed in this figure.