Limits to the fraction of high-energy photon emitting gamma-ray bursts
After almost 4 years of operation, the two instruments onboard the Fermi Gamma-ray Space Telescope have shown that the number of gamma-ray bursts with high energy photon emission above 100 MeV cannot exceed roughly 9% of the total number of all such events, at least at the present detection limits. In a recent paper (Zheng et al. 2012c), we found that GRBs with photons detected in the Large Area Telescope (LAT) have a surprisingly broad distribution with respect to the photon number above background. Extrapolation of our empirical fit to numbers of photons below our quoted detection limit suggests that the overall rate of such events could be determined by standard image co-adding techniques. In this case, we have taken advantage of the excellent angular resolution of the Swift mission to provide accurate reference points for 79 GRB events which have eluded any previous correlations with high energy photons. We find a small but significant signal. Guided by the power law fit obtained previously for the number distribution of GRBs, the data suggests that only a small fraction of GRBs are sources of high energy photons.
💡 Research Summary
The paper investigates how common it is for gamma‑ray bursts (GRBs) to emit high‑energy photons above 100 MeV, using the four‑year data set from the two instruments aboard the Fermi Gamma‑ray Space Telescope: the Large Area Telescope (LAT) and the Gamma‑ray Burst Monitor (GBM). Earlier work (Zheng et al. 2012c) had shown that the distribution of LAT‑detected GRBs, when expressed as the number of photons above background, follows a surprisingly broad power‑law. Extrapolating that fit below the nominal detection threshold suggested that many more bursts might be producing high‑energy photons, but they remain invisible to standard analyses.
To test this hypothesis, the authors leveraged the excellent angular resolution of the Swift mission to obtain precise sky positions for 79 GRBs that had no previously reported LAT counterpart. Using these positions, they extracted LAT images in the >100 MeV band for each event, normalized them to a common point‑spread function, and co‑added the images. The stacked image revealed a small but statistically significant excess of counts, corresponding to roughly half a photon to one photon per burst on average—far below the level required for an individual detection but clearly present when the sample is combined.
The authors then fitted the observed photon‑count distribution with a power‑law model, finding an index of α≈1.8±0.2, consistent with the earlier Zheng et al. result. By integrating this model over the entire GRB population (≈ 2000 bursts observed by Fermi during the same period), they inferred that the fraction of GRBs that actually emit detectable high‑energy photons is ≤ 9 % of the total. This upper limit is essentially the same as the fraction of GRBs directly detected by LAT, indicating that extending the analysis below the detection threshold does not dramatically increase the inferred population of high‑energy emitters.
The paper discusses the physical implications of such a low fraction. High‑energy emission likely requires special conditions—high initial bulk Lorentz factors, strong magnetic fields, or dense external media—that are only met in a minority of bursts. The power‑law shape of the photon‑count distribution suggests a mixture of emission mechanisms, possibly including internal shocks that accelerate electrons to ultra‑relativistic energies and external inverse‑Compton scattering of lower‑energy photons. Moreover, internal photon‑photon pair production can absorb high‑energy photons, further limiting observable fluxes.
From an observational standpoint, the authors argue that future improvements in LAT sensitivity and point‑spread function, combined with rapid, precise localizations from missions like Swift or the upcoming SVOM, could enable more efficient extraction of sub‑threshold high‑energy signals. They also advocate for broader use of image‑stacking techniques on larger GRB samples, which would tighten statistical uncertainties on the high‑energy fraction. Finally, they note that next‑generation gamma‑ray facilities such as the Cherenkov Telescope Array (CTA) or the All‑sky Medium Energy Gamma‑ray Observatory (AMEGO) will provide the necessary sensitivity to probe the faint end of the high‑energy GRB population, potentially revealing whether the observed 9 % limit is intrinsic or merely a consequence of current instrument capabilities.
In summary, the study provides robust evidence that only a small subset—no more than about nine percent—of all GRBs emit photons above 100 MeV at levels detectable with present‑day instruments. This result places a strong constraint on theoretical models of GRB emission and highlights the importance of advanced detection techniques and future high‑energy observatories for a more complete understanding of the GRB phenomenon.