Nearly 50 years ago, the first radio signals from cosmic ray air showers were detected. After many successful studies, however, research ceased not even 10 years later. Only a decade ago, the field was revived with the application of powerful digital
signal processing techniques. Since then, the detection technique has matured, and we are now in a phase of transition from small-scale experiments accessing energies below 1018 eV to experiments with a reach for energies beyond 1019 eV. We have demonstrated that air shower radio signals carry information on both the energy and the mass of the primary particle, and current experiments are in the process of quantifying the precision with which this information can be accessed. All of this rests on a solid understanding of the radio emission processes which can be interpreted as a coherent superposition of geomagnetic emission, Askaryan charge-excess radiation, and Cherenkov-like coherence effects arising in the density gradient of the atmosphere. In this article, I highlight the state of the art of radio detection of cosmic rays and briefly discuss its perspectives for the next few years.
CoREAS is a Monte Carlo simulation code for the calculation of radio emission from extensive air showers. It is based on the endpoint formalism for radiation from moving charges implemented directly in CORSIKA. Consequently, the full complexity of th
e air-shower physics is taken into account without the need for approximations or assumptions on the emission mechanism. We present results of simulations for an unthinned shower performed with CoREAS for both MHz and GHz frequencies. At MHz frequencies, the simulations predict the well-known mixture of geomagnetic and charge excess radiation. At GHz frequencies, the emission is strongly influenced by Cherenkov effects arising from the varying refractive index in the atmosphere. In addition, a qualitative difference in the symmetry of the GHz radiation pattern is observed when compared to the ones at lower frequencies. We also discuss the strong increase in the ground area subtended by the radio emission when going from near-vertical to very inclined geometries, making very inclined air showers the most promising ones for cosmic ray radio detection.
We present the endpoint formalism for the calculation of electromagnetic radiation and illustrate its applications in astroparticle physics. We use the formalism to explain why the coherent radiation from the Askaryan effect is not in general Cherenk
ov radiation, as the emission directly results from the time-variation of the net charge in the particle shower. Secondly, we illustrate how the formalism has been applied in the air shower radio emission code REAS3 to unify the microscopic and macroscopic views of radio emission from extensive air showers. Indeed, the formalism is completely universal and particularly well-suited for implementation in Monte Carlo codes in the time- and frequency-domains. It easily reproduces well-known classical mechanisms such as synchrotron radiation, Vavilov-Cherenkov radiation and transition radiation in the adequate limits, but has the advantage that it continues to work in realistic, complex situations, where the classical mechanisms tend to no longer apply and adhering to them can result in misleading interpretations.
Over the previous decade, many approaches for the modelling of radio emission from cosmic ray air showers have been developed. However, there remained significant deviations between the models, reaching from important qualitative differences such as
unipolar versus bipolar pulses to large variations in the predicted absolute amplitudes of up to factors of 20. Only recently, it has been realized that in the many models predicting unipolar pulses, a radio emission contribution due to the time-variation of the number of charged particles or, equivalently, the acceleration of the particles at the beginning and the end of their trajectories, had not been taken into account. We discuss here the nature of the underlying problem and demonstrate that by including the missing contribution in REAS3, the discrepancies are reconciled. Furthermore, we show a direct comparison of REAS3 and MGMR simulations for a set of prototype showers. The results of these two completely independent and very different modelling approaches show a good level of agreement except for regions of parameter space where differences in the underlying air shower model become important. This is the first time that two radio emission models show such close concordance, illustrating that the modelling of radio emission from extensive air showers has indeed made a true breakthrough.
