Particle Acceleration at the Earth’s Bow Shock

abstract

Shocks can be viewed as sites where upstream bulk flow energy is converted into downstream thermal energy. Collisionless shocks have the further property that they are sites of particle acceleration, as some fraction of the flow energy is diverted to a small number of energetic particles. Shocks are thus important in many astrophysical and solar environments. The best studied example is the Earth’s bow shock, which has the benefit of high time resolution, in situ multi-point measurements. Models and mechanisms developed for the Earth’s bow shock are useful for understanding the behaviour of shocks in the corona and solar wind. The Earth’s bow shock is curved, and the main controlling factor of its structure is the angle between the magnetic field and shock normal, which is used to classify the shock as quasi-parallel or quasi-perpendicular. We review observations of accelerated electrons at the quasi-perpendicular shock and the standard model of acceleration, namely adiabatic reflection. We describe the weaknesses of this model and suggest a new mechanism, based on simulations, which predicts power law energy spectra. The quasi-parallel shock is a region of large amplitude turbulence, containing coherent structures and wavetrains. Cluster multi-point observations of these large amplitude pulsations are presented as a review of the present understanding of this type of shock. In terms of particle acceleration, quasi-parallel shocks are important since they are believed to be the best examples of sites for Fermi acceleration∈dexFermi acceleration. Cluster observations are reviewed indicating unambiguously that energetic particle diffusion, a fundamental assumption of the Fermi process, is operating at the quasi-parallel bow shock. Finally, a class of transient events observed at the bow shock, Hot Flow Anomalies, is described. The causative mechanism and implications for particle acceleration are discussed, and a possible role in other heliospheric environments is suggested.