Pulsars as the Sources of High Energy Cosmic Ray Positrons

TL;DR

The paper investigates whether mature pulsars can explain the PAMELA positron excess by injecting high-energy electron-positron pairs and propagating them through the Galaxy with diffusion and energy losses. It analyzes both the cumulative Galactic pulsar population and nearby sources like Geminga and B0656+14, finding that plausible efficiencies and spectra can reproduce the observed positron fraction. It further proposes an observable dipole anisotropy in the high-energy electron-positron flux, detectable by Fermi, to distinguish pulsar origins from dark matter scenarios. The work emphasizes that pulsars offer a natural, non-exotic explanation consistent with gamma-ray and antiproton constraints, and motivates future population studies and anisotropy measurements.

Abstract

Recent results from the PAMELA satellite indicate the presence of a large flux of positrons (relative to electrons) in the cosmic ray spectrum between approximately 10 and 100 GeV. As annihilating dark matter particles in many models are predicted to contribute to the cosmic ray positron spectrum in this energy range, a great deal of interest has resulted from this observation. Here, we consider pulsars (rapidly spinning, magnetized neutron stars) as an alternative source of this signal. After calculating the contribution to the cosmic ray positron and electron spectra from pulsars, we find that the spectrum observed by PAMELA could plausibly originate from such sources. In particular, a significant contribution is expected from the sum of all mature pulsars throughout the Milky Way, as well as from the most nearby mature pulsars (such as Geminga and B0656+14). The signal from nearby pulsars is expected to generate a small but significant dipole anisotropy in the cosmic ray electron spectrum, potentially providing a method by which the Fermi gamma-ray space telescope would be capable of discriminating between the pulsar and dark matter origins of the observed high energy positrons.

Figures (5)

• Figure 1: The spectrum of cosmic ray positrons (left) and the positron fraction (right) resulting from the sum of all pulsars throughout the Milky Way. Also shown as a dashed line is the prediction for secondary positrons (and primary and secondary electrons in the right frames) as calculated in Ref. moskalenkostrong. In the right frames, the measurements of HEAT heat (light green and magenta) and measurements of PAMELA Adriani:2008zr (dark red) are also shown. We have used the injected spectrum reported in Eq. (\ref{['eq:inje']}). In the lower frames, the upper (lower) dotted line represents the case in which the injection rate within 500 parsecs of the Solar System is doubled (neglected), providing an estimate the variance resulting from the small number of nearby pulsars contributing to the spectrum.
• Figure 2: The spectrum of positrons (left) and ratio of positrons to electrons plus positrons (right) from the pulsar Geminga, with the dashed lines as in Fig. 1. In the right frames, the measurements of HEAT heat (light green and magenta) and measurements of PAMELA Adriani:2008zr (dark red) are also shown. Here we have used an injected spectrum such that $dN_e/dE_e \propto E^{-\alpha}\, \exp(-E_e/600\,{\rm GeV})$, with $\alpha=1.5$ and 2.2. The solid lines correspond to an energy in pairs given by $3.5\times 10^{47}$ erg, while the dotted lines require an output of $3\times 10^{48}$ erg.
• Figure 3: As in Fig. \ref{['geminga']}, but from the nearby pulsar B0656+14. The solid lines correspond to an energy in pairs given by $3\times 10^{47}$ erg, while the dotted lines require an output of $8\times 10^{47}$ erg.
• Figure 4: The positron spectrum and positron fraction from the sum of contributions from B0656+14, Geminga, and all pulsars farther than 500 parsecs from the Solar System.
• Figure 5: The dipole anisotropy in the electron+positron spectrum from a source 110,000 years old at a distance of 290 pc (B0656+14-like) and from a source 370,000 years old at a distance of 157 pc (Geminga-like). In each case, we have normalized the energy output to match the PAMELA data and have used a spectral shape of $dN_e/dE_e \propto E_e^{-1.5} \exp(-E_e/600\,\rm{GeV})$. Also shown as dashed lines is the sensitivity of the Fermi gamma-ray space telescope to such an anisotropy (after five years of observation). The Fermi sensitivity shown is for the spectrum integrated above a given energy.