Cherenkov Telescope Array sensitivity to the putative millisecond pulsar population responsible for the Galactic center excess

TL;DR

This paper evaluates CTA's ability to detect inverse-Compton gamma-ray signals from millisecond pulsars in the Galactic bulge, which could explain the Galactic center excess, considering systematic uncertainties and spectral properties.

Contribution

It provides a realistic forecast of CTA's sensitivity to MSP-induced IC emission, accounting for diffuse emission modeling uncertainties and spectral variations.

Findings

CTA can potentially detect MSP IC signals if injection spectra are hard enough.

Detection depends on the efficiency of $e^ extpm$ injection and spectral hardness.

CTA can distinguish MSP origin from dark matter if a signal is observed.

Abstract

The leading explanation of the $\textit{Fermi}$ Galactic center $\gamma$-ray excess is the extended emission from a unresolved population of millisecond pulsars (MSPs) in the Galactic bulge. Such a population would, along with the prompt $\gamma$ rays, also inject large quantities of electrons/positrons ($e^\pm$) into the interstellar medium. These $e^\pm$ could potentially inverse-Compton (IC) scatter ambient photons into $\gamma$ rays that fall within the sensitivity range of the upcoming Cherenkov Telescope Array (CTA). In this article, we examine the detection potential of CTA to this signature by making a realistic estimation of the systematic uncertainties on the Galactic diffuse emission model at TeV-scale $\gamma$-ray energies. We forecast that, in the event that $e^\pm$ injection spectra are harder than $E^{-2}$, CTA has the potential to robustly discover the IC signature of a putative Galactic bulge MSP population sufficient to explain the GCE for $e^\pm$ injection efficiencies in the range $\approx 2.9-74.1\%$, or higher, depending on the level of mismodeling of the Galactic diffuse emission components. On the other hand, for spectra softer than $E^{-2.5}$, a reliable CTA detection would require an unphysically large $e^\pm$ injection efficiency of $\gtrsim 158\%$. However, even this pessimistic conclusion may be avoided in the plausible event that MSP observational and/or modeling uncertainties can be reduced. We further find that, in the event that an IC signal were detected, CTA can successfully discriminate between an MSP and a dark matter origin for the radiating $e^\pm$.