Log-normal flux distribution of bright Fermi blazars

TL;DR

This study analyzes the gamma-ray flux distribution of the brightest Fermi blazars, finding that most follow a log-normal distribution and revealing correlations with spectral index, which impacts flux estimation methods.

Contribution

It is the first comprehensive analysis showing that gamma-ray fluxes of bright blazars follow a log-normal distribution, with implications for understanding blazar variability and classification.

Findings

Fluxes of most bright blazars follow a log-normal distribution.

Standard deviation of flux distribution correlates with spectral index.

Unclassified blazars may be associated with FSRQs based on flux distribution features.

Abstract

We present the results of the $\gamma$-ray flux distribution study on the brightest blazars which are observed by the \emph{Fermi}-LAT. We selected 50 brightest blazars based on the maximum number of detection reported in the LAT third AGN catalog. We performed standard unbinned maximum likelihood analysis on the LAT data during the period between August 2008 and December 2016, in order to obtain the average monthly flux. After quality cuts, blazars for which at least 90\% of the total flux was survived were selected for the further study, and this includes 19 FSRQs and 19 BL Lacs. The Anderson-Darling and $\chi^2$ tests suggest that the integrated monthly flux follow a log-normal distribution for all sources, except for three FSRQs for which neither a normal nor a log-normal distribution was preferred. A double log-normal flux distribution tendency were observed in these sources, though it has to be confirmed with improved statistics. We also found that, the standard deviation of the log-normal flux distribution increases with the mean spectral index of the blazar, and can be fitted with a line of slope 0.24$\pm$0.04. We repeat our study on three additional brightest unclassified blazars to identify their flux distribution properties. Based on the features of their log-normal flux distribution, we infer these unclassified blazars may be closely associated with FSRQs. We also highlight that considering the log-normal behavior of the flux distribution of blazars, averaging their long term flux in linear scale can largely under estimate the nominal flux and this discrepancy can propagate down to the estimation of source parameters through spectral modeling.