# Galactic cosmic rays after the AMS-02 observations

## TL;DR

The AMS-02 data provide new insights into cosmic ray origin and propagation, revealing a break in diffusion energy dependence and element-specific injection spectra, with implications for cosmic ray models.

## Contribution

This paper introduces a detailed analysis of AMS-02 cosmic ray data, highlighting a diffusion coefficient break and element-dependent injection spectra, and presents a new parametrization of relevant nuclear cross sections.

## Key findings

- A break in the diffusion coefficient at ~300 GV.
- Different injection spectra for protons, helium, and heavier nuclei.
- Improved fit to B/C ratio with source grammage consideration.

## Abstract

The unprecedented quality of the data collected by the AMS-02 experiment onboard the International Space Station allowed us to address subtle questions concerning the origin and propagation of cosmic rays. Here we discuss the implications of these data for the injection spectrum of elements with different masses and for the diffusion coefficient probed by cosmic rays through their journey from the sources to the Earth. We find that the best fit to the spectra of primary and secondary nuclei requires (1) a break in the energy dependence of the diffusion coefficient at energies $\sim 300$ GV; (2) an injection spectrum that is the same for all nuclei heavier than helium, and different injections for both protons and helium. Moreover, if to force the injection spectrum of helium to be the same as for heavier nuclei, the fit to oxygen substantially worsens. Accounting for a small, $X_{s}\sim 0.4~\rm g~cm^{-2}$, grammage accumulated inside the sources leads to a somewhat better fit to the B/C ratio but makes the difference between He and other elements even more evident. The statistic and systematic error bars claimed by the AMS collaboration exceed the error that is expected from calculations once the uncertainties in the cross sections of production of secondary nuclei are taken into account. In order to make this point more quantitative, we present a novel parametrization of a large set of cross sections, relevant for cosmic ray physics, and we introduce the uncertainty in the branching ratios in a way that its effect can be easily grasped.