On Optimally Selecting Candidate Detectors with High Predicted Radio Signals from Energetic Cosmic Ray-Induced Extensive Air Showers

Monte Carlo simulations of induced extensive air showers (EASs) by ultra-high-energy cosmic rays are widely used in comparison with measured events at experiments to estimate the main cosmic ray characteristics, such as mass, energy, and arrival direction. However, these simulations are computationally expensive, with running time scaling proportionally with the number of radio antennas included. The AugerPrime upgrade of the Pierre Auger Observatory will feature an array of 1660 radio antennas. As a result, simulating a single EAS using the full detector array will take weeks on a single CPU thread. To reduce the simulation time, detectors are commonly pre-selected based on their proximity to the shower core, using a selection ellipse based on the Cherenkov radiation footprint scaled by a fixed constant factor. While effective, this approach often includes many noisy antennas at high zenith angles, reducing computational efficiency. In this paper, we introduce an optimal method for selecting candidate detectors with high predicted signal-to-noise ratio for proton and iron primary cosmic rays, replacing the constant scaling factor with a function of the zenith angle. This approach significantly reduces simulation time—by more than 50% per CPU thread for the heaviest, most inclined showers—without compromising signal quality.