Title: What can cosmic-ray knees reveal about source populations?

Abstract: Cosmic ray (CR) knees (spectral steepenings) encode information on CR accelerator populations. We seek population features that imprint onto knee observables in a manner that is robust enough to be discernible even in the presence of significant systematics in CR data. In particular, we explore how diversity among population members could imprint on the knee phenomenology, under the assumption that a knee is due to a fixed-rigidity cutoff in the source spectra. We use a simple theoretical model for a population of CR accelerators. Each population member accelerates CR to a power-law spectrum, up to a cutoff rigidity. We allow for variance among members, in cutoff rigidity and power-law slope. We find that: (a) the slope step of the spectrum is $\sim 0.5$, decreasing weakly with increasing spread in either property; (b) composition always breaks first; (c) the difference between the break energies in composition and flux increases with increasing diversity; (d) composition and flux break together only if population diversity is minimal. These trends are robust under our assumptions; deviations from them would indicate more complex physics than encoded in our simple model. Comparing these trends with observed CR knees, we conclude that: (i) the primary knee at $\sim 4\times10^{15}$ eV is consistent with a constant-rigidity cutoff according to KASCADE-Grande data processed with post-LHC hadronic models, but not according to other datasets; (ii) the second knee at $\sim 5 \times 10^{17}$ eV requires more complexity than our model; (iii) the spectral feature identified by Auger at $\sim 10^{19}$ eV is consistent with a constant-rigidity source cutoff only if there is a substantial spread in both cutoff rigidity and slope. Interestingly, a significant spread in slope would also result in spectral curvature before the break, which would in turn be contributing to the ankle feature.