Space science: Cosmic rays beyond the knees

Journal name:
Date published:
Published online

The development of a radio technique for detecting cosmic rays casts fresh light on the origins of some of these accelerated particles, and suggests that they might have travelled much farther than was previously thought. See Letter p.70

A technique for measuring the composition of cosmic rays in the energy range 1017 to 1017 electronvolts — a range thought to mark the transition point from Galactic to extragalactic cosmic rays — is reported by Buitink et al.1 on page 70 of this issue. The findings have implications for our understanding of the sources of these mysterious rays.

Energetic protons and atomic nuclei arriving at Earth are classified as cosmic rays. The spread in energy of these particles is impressive, covering ten orders of magnitude (Fig. 1). The most energetic cosmic rays have energies more than 10 million times those achievable for protons at the Large Hadron Collider, the world's most powerful particle accelerator at CERN in Geneva, Switzerland. This begs the question of what cosmic sources can accelerate particles to such high energies.

Figure 1: The energy spectrum of cosmic rays.
The energy spectrum of cosmic rays.

When the number of cosmic rays per square metre that arrive at Earth's surface per day is plotted on a logarithmic scale against the energies of the cosmic rays, also on a logarithmic scale, the overall graph (yellow line) between 1014 and 1019 electronvolts contains four distinct linear regions, connected by changes in slope called the proton knee, the iron knee and the ankle. The slope changes correspond to transitions between the composition or sources of cosmic rays that dominate at different energies. Buitink et al.1 report on observations of cosmic rays between the iron knee and the ankle, providing clues about their origins.

Clues about the origin of cosmic rays come from both their composition and their energy spectra. Unfortunately, these particles arrive at Earth rather infrequently — only one particle with an energy greater than 1017 eV arrives each day in every 10,000 square metres. This scarcity makes it difficult to build detectors able to capture sufficient arrival events to draw statistically meaningful conclusions. At present, detector size and observation times are limited by the use of optical detectors2, which have areas of up to a few square kilometres and can operate only on clear, moonless nights.

The region between about 1015 eV and 1018 eV (the second energy value is called the ankle; Fig. 1) in the energy distribution of cosmic rays demarks an area of change. At energies higher than 1015 eV, the composition of cosmic rays changes from lightweight particles, such as protons and helium nuclei, to increasingly heavy ones, such as the nuclei of carbon and heavier elements. These changes are thought to be associated with the maximum energies up to which Galactic sources — possibly supernova remnants (SNRs) — can accelerate cosmic rays3. The maximum energy for Galactic protons (the proton knee) is about 1015 eV, whereas that for iron nuclei (the iron knee) is about 1017 eV. The ankle is thought to be dominated by particles from extragalactic sources4, 5, for example active galactic nuclei, γ-ray bursts or neutron stars. But, within this picture, the origin of cosmic rays between the iron knee and the ankle is unclear.

Buitink et al. investigated the cosmic-ray composition in this energy region using a new radio-based instrument. This can detect more cosmic rays in a given period than can optical detectors of equivalent effective area, because it can operate both day and night. The authors' detection method relies on descriptions6, 7 of the radio signals generated by air showers — the cascades of energetic charged particles and electromagnetic radiation produced when cosmic rays enter the atmosphere. The radio signal is produced by the interaction of the charged particles with Earth's magnetic field, and by the development of a charge imbalance within the shower through a phenomenon called the Askarayan effect8; the contributions of these two effects to the signal are of comparable magnitude.

This knowledge allowed Buitink and colleagues to probe the air-shower profile produced by cosmic rays in the energy range of interest. Air showers involve a chain of processes that rapidly increase the number of energetic particles (known as secondary particles) within them. The rate of transfer of a cosmic ray's energy to secondary particles depends on its nuclear composition. The authors could discriminate between different species of cosmic ray — that is, the type of particle that was originally accelerated — from the depth in the atmosphere at which the maximum number of air-shower particles occurred.

From a sample of 118 such profile measurements, the researchers concluded that the fraction of cosmic rays consisting of protons and helium nuclei in the energy band 1017 to 1017 eV is between 38% and 98%, with a 99% confidence level; the best-fit value for this light-mass fraction is about 80%. These results are consistent with previous results attained using optical techniques2, 9, but this is the first time that such a composition measurement has been made using a radio instrument.

The success of Buitink and co-workers' method provides fresh clues about the origin of cosmic rays between the iron knee and the ankle. It could be that cosmic rays in this energy range are associated with an extragalactic component, which would challenge the idea that the ankle represents the onset of this component in the energy distribution. Alternatively, these cosmic rays might have a Galactic origin. This would indicate the existence of a second population of Galactic sources capable of accelerating particles to considerably higher energies than those achievable by SNRs. Either way, the authors' findings support the idea of light-mass cosmic rays in the knee-to-ankle region that must now be explained.

The powerful method used to make this measurement might also seed a new era of cosmic-ray science, in which previously unknown spectral features and changes in composition at energies within the transition zone are probed with unprecedented detail. Perhaps in this way we will uncover further surprises and evidence of extragalactic sources.


  1. Buitink, S. et al. Nature 531, 7073 (2016).
  2. Berezhnev, S. F. et al. Proc. 32nd ICRC Beijing 1, 209212 (11–18 August 2011).
  3. Bell, A. R., Schure, K. M., Reville, B. & Giacinti, G. Mon. Not. R. Astron. Soc. 431, 415429 (2013).
  4. Hillas, A. M. Annu. Rev. Astron. Astrophys. 22, 425444 (1984).
  5. Waxman, E. Phys. Rev. Lett. 75, 386389 (1995).
  6. Huege, T., Ludwig, M. & James, C. W. AIP Conf. Proc. 1535, 128 (2013).
  7. Alvarez-Muñiz, J., Carvalho, W. R. Jr & Zas, E. Astropart. Phys. 35, 325341 (2012).
  8. Askarayan, G. Sov. Phys. JETP 14, 441443 (1962).
  9. Abu-Zayyad, T. et al. Astrophys. J. 557, 686 (2001).

Download references

Author information


  1. Andrew M. Taylor is in the High Energy Astrophysics Group, School of Cosmic Physics, Dublin Institute for Advanced Studies, Dublin 2, Ireland.

Corresponding author

Correspondence to:

Author details

Additional data