Abstract
Radon exhalation is a natural process by which atoms of the radioactive gas radon diffuse in the soil and then exhale to an indoor and/or outdoor environment. High radon concentration levels, possibly from high radon exhalation rate levels, can generate an impact on public health and environmental safety, particularly in agricultural areas where prolonged exposure may affect nearby populations. While studies have examined radon exhalation, few have focused on modeling its behavior in agricultural settings or identifying key environmental and soil parameters that influence its variation. This study addresses this gap by applying Artificial Neural Network (ANN) models and Monte Carlo methods. Three distinct approaches were developed based on radon exhalation measurements from four Peruvian agricultural regions, incorporating meteorological and soil physicochemical data. First, the ANN model determined environmental factors affecting radon exhalation, achieving [Formula: see text] values of 0.7949 (training) and 0.7656 (validation). Second, simulations analyzed radon diffusion under varying wind conditions, assessing dispersion risks. Third, gamma radiation measurements quantified radon progeny contributions ([Formula: see text] efficiency) for soil moisture detection. This integrated methodology advances understanding of agricultural radon dynamics, supporting improved radiological safety protocols and soil monitoring techniques.