Revisiting the AMS-02 antiproton excess: The role of correlated errors

Cosmic-ray antiprotons are a remarkable diagnostic tool for the study of astroparticle physics processes in our Galaxy. While the bulk of measured antiprotons is consistent with a secondary origin, several studies have found evidence for a subdominant primary component in the AMS-02 data. In this proceedings article, we revisit the excess considering systematic errors that could affect the signal. Of particular importance are unknown correlations in the AMS-02 systematic errors, the dominant of which are associated with the cross sections for cosmic-ray absorption in the detector. We compute their correlations in a careful reevaluation of nuclear scattering data, utilizing the Glauber-Gribov theory to introduce a welcomed redundancy that we explore in a global fit. The inclusion of correlated errors has a dramatic effect on the significance of the signal. In particular, the analysis becomes more sensitive to the diffusion model at low rigidities. For a minimal extension beyond single-power-law diffusion, the global significance drops below 1$sigma$ severely questioning the robustness of the finding.

Constraining dark matter annihilation with cosmic ray antiprotons using neural networks

The interpretation of data from indirect detection experiments searching for dark matter annihilations requires computationally expensive simulations of cosmic-ray propagation. In this work we present a new method based on Recurrent Neural Networks that significantly accelerates simulations of secondary and dark matter Galactic cosmic ray antiprotons while achieving excellent accuracy. This approach allows for an efficient profiling or marginalisation over the nuisance parameters of a cosmic ray propagation model in order to perform parameter scans for a wide range of dark matter models. We identify importance sampling as particularly suitable for ensuring that the network is only evaluated in well-trained parameter regions. We present resulting constraints using the most recent AMS-02 antiproton data on several models of Weakly Interacting Massive Particles. The fully trained networks are released as DarkRayNet together with this work and achieve a speed-up of the runtime by at least two orders of magnitude compared to conventional approaches.

Dark matter or correlated errors: Systematics of the AMS-02 antiproton excess

Several studies have pointed out an excess in the AMS-02 antiproton spectrum at rigidities of 10-20 GV. Its spectral properties were found to be consistent with a dark-matter particle of mass 50-100 GeV which annihilates hadronically at roughly the thermal rate. In this work, we reinvestigate the antiproton excess including all relevant sources of systematic errors. Most importantly, we perform a realistic estimate of the correlations in the AMS-02 systematic error which could potentially fake a dark-matter signal. The dominant systematics in the relevant rigidity range originate from uncertainties in the cross sections for absorption of cosmic rays within the detector material. For the first time, we calculate their correlations within the full Glauber-Gribov theory of inelastic scattering. The AMS-02 correlations enter our spectral search for dark matter in the form of covariance matrices which we make publicly available for the cosmic-ray community. We find that the global significance of the antiproton excess is reduced to below 1 $sigma$ once all systematics, including the derived AMS-02 error correlations, are taken into account. No significant preference for a dark-matter signal in the AMS-02 antiproton data is found in the mass range 10-10000 GeV.