A data-driven method for measuring corner-clipping probabilities in segmented particle detectors

Abstract

The accuracy of particle counting in highly segmented detectors is limited by the corner-clipping effect, in which a single ionizing particle generates signals in adjacent detection elements. This phenomenon introduces a direction-dependent overcounting bias that distorts reconstructed observables and is commonly corrected using Monte-Carlo simulations, thereby inheriting modeling uncertainties. We present a fully data-driven method to directly measure the single-particle corner-clipping probability, exploiting the nanosecond timing resolution of modern detectors to statistically distinguish genuine corner-clipping events from random coincidences, with non-neighboring detection elements serving as an intrinsic control sample. The technique is validated using detailed simulations of the Underground Muon Detector of the Pierre Auger Observatory, reproducing the true angular dependence of the corner-clipping probability with absolute deviations below 0.01. To parameterize the results, we introduce a compact analytical model incorporating detector geometry, minimum detectable path length, and orientation-independent contributions. The proposed methodology and parameterization enable the direct incorporation of data-driven corner-clipping corrections into reconstruction algorithms, mitigating the overcounting bias and ultimately yielding a more accurate determination of the muonic component of extensive air showers. These developments are broadly applicable to any segmented detector with sufficient timing resolution, making them relevant to a wide range of experiments in high-energy and astroparticle physics.