Estimate of the Fermi Large Area Telescope sensitivity to gamma-ray polarization

TL;DR

This paper evaluates the Fermi-LAT's potential to detect gamma-ray polarization, showing it could measure polarization degrees of around 37-50% for bright sources after a decade of data collection.

Contribution

It provides the first detailed sensitivity estimate of Fermi-LAT for gamma-ray polarization measurements, including statistical and systematic uncertainties.

Findings

Fermi-LAT can detect polarization degrees of 37-50% in bright gamma-ray sources.

Systematic uncertainties are likely smaller than statistical errors in polarization measurements.

A 5σ sensitivity limit of approximately 37% polarization degree was established.

Abstract

Although not designed primarily as a polarimeter, the \textit{Fermi}-Large Area Telescope (LAT) has the potential to detect high degrees of linear polarization from some of the brightest gamma-ray sources. To achieve the needed accuracy in the reconstruction of the event geometry, low-energy ($\leq200$ MeV) events converting in the silicon detector layers of the LAT tracker have to be used. We present preliminary results of the ongoing effort within the LAT collaboration to measure gamma-ray polarization. We discuss the statistical and systematic uncertainties affecting such a measurement. We show that a $5\sigma$ minimum detectable polarization (MDP) of $\approx30-50\%$ could be within reach for the brightest gamma-ray sources as the Vela and Crab pulsars and the blazar 3C 454.3, after 10 years of observation. To estimate the systematic uncertainties, we stack bright AGN, and use this stack as a test source. LAT sensitivity to polarization is estimated comparing the data to a simulation of the expected unpolarized emission of the stack. We measure a 5$\sigma$ sensitivity limit corresponding to a polarization degree of $\approx37\%$. This is in agreement with a purely statistical estimate, suggesting that the systematic errors are likely to be small compared to the statistical ones.