In this chapter we summarize measurements of the spectrum and composition of cosmic rays with energies above 100 TeV and the implications for sources. We noted in Chapter 12 that the knee of the cosmic ray spectrum may coincide with the upper limit of shock acceleration by supernova remnants (SNR). We also noted that, whether the knee reflects the maximum energy of a class of accelerators or a rigidity-dependent change in propagation, the composition should change systematically from light to heavy in the knee region.

Particles with energies greater than 3×1018 eV are generally assumed to be from sources outside of the MilkyWay because they show no sign of the anisotropy that would be expected if they came from sources in the Galactic plane. The hardening of the spectrum at the ankle around this energy is often interpreted as the transition from Galactic to extragalactic cosmic rays [638]. If the knee reflects the upper limit of acceleration by SNR with a maximum energy for protons of ≈ 1015 eV, then the major nuclear groups would follow the Peters cycle in rigidity culminating with Emax(Fe) ≈ 3 × 1016 eV. Interpretation of the ankle as the transition to extragalactic cosmic rays would then require a second kind of Galactic source capable of accelerating protons to ≈ 1017 eV and iron to ≈ 3 × 1018 eV. Hillas [639] suggests this possibility and identifies the contribution of unknown origin as Population B. On the other hand, the population of extragalactic cosmic rays may extend down to lower energy, A 1017 eV, in which case an alternate explanation for the ankle is needed. Berezinsky et al. [308] proposed an extragalactic spectrum dominated by protons to explain the ankle as a pileup effect just below 3 × 1018 eV, the energy above which losses due to e+e− pair production by protons in the CMB become more important than adiabatic expansion (see Figure 10.2).

These conflicting possibilities require a detailed understanding of the spectrum and composition as a function of energy for their resolution.