Variability Signatures of a Burst Process in Flaring Gamma-ray Blazars

TL;DR

This paper introduces an autoregressive inverse gamma model to describe the heavy-tailed gamma-ray flux variability in blazars, linking burst processes to observed flux distributions and variability timescales.

Contribution

It proposes a novel autoregressive inverse gamma model that explains blazar gamma-ray variability as a burst-driven shot-noise process, extending previous heavy-tailed flux distribution analyses.

Findings

The model reproduces observed flux distributions of FSRQs and BL Lacs.

Fractional variability decreases with increasing time bin duration.

Model parameters correspond to physical burst properties.

Abstract

Blazars exhibit stochastic flux variability across the electromagnetic spectrum, often exhibiting heavy-tailed flux distributions, commonly modeled as lognormal. However, Tavecchio et al. (2020) and Adams et al. (2022) found that the high-energy gamma-ray flux distributions of several of the brightest flaring Fermi-LAT flat spectrum radio quasars (FSRQs) are well modeled by an even heavier-tailed distribution, which we show is the inverse gamma distribution. We propose an autoregressive inverse gamma variability model in which an inverse gamma flux distribution arises as a consequence of a shot-noise process. In this model, discrete bursts are individually unresolved and averaged over within time bins, as in the analysis of Fermi-LAT data. Stochastic variability on timescales longer than the time bin duration is modeled using first-order autoregressive structure. The flux distribution becomes approximately lognormal in the limiting case of many weak bursts. The fractional variability is predicted to decrease as the time bin duration increases. Using simulated light curves, we show that the proposed model is consistent with the typical gamma-ray variability properties of FSRQs and BL Lac objects. The model parameters can be physically interpreted as the average burst rate, the burst fluence, and the timescale of long-term stochastic fluctuations.