Implications of the AMS-02 positron fraction in cosmic rays

The AMS-02 collaboration has just released its first result of the cosmic positron fraction $e^+/(e^-+e^+)$ with high precision up to $\sim 350$ GeV. The AMS-02 result shows the same trend with the previous PAMELA result, which requires extra electron/positron sources on top of the conventional cosmic ray background, either from astrophysical sources or from dark matter annihilation/decay. In this paper we try to figure out the nature of the extra sources by fitting to the AMS-02 $e^+/(e^-+e^+)$ data, as well as the electron and proton spectra by PAMELA and the $(e^-+e^+)$ spectrum by Fermi and HESS. We adopt the GALPROP package to calculate the propagation of the Galactic cosmic rays and the Markov Chain Monte Carlo sampler to do the fit. We find that the AMS-02 data have implied essential difference from the PAMELA data. There is {\rm tension} between the AMS-02 $e^+/(e^-+e^+)$ data and the Fermi/HESS $(e^-+e^+)$ spectrum, that the AMS-02 data requires less contribution from the extra sources than Fermi/HESS. Then we redo the fit without including the Fermi/HESS data. In this case both the pulsars and dark matter annihilation/decay can explain the AMS-02 data. The pulsar scenario has a soft inject spectrum with the power-law index $\sim 2$, while the dark matter scenario needs $\tau^+\tau^-$ final state with mass $\sim 600$ GeV and a boost factor $\sim 200$.