Monte Carlo Aggregating Method for Radon Precursor-Based Earthquake Parameter Prediction in Java, Indonesia

This study introduces a novel method for predicting earthquake parameters using radon as a precursor to address uncertainties and limitations in the dataset. The dataset comprises radon observation data from Yogyakarta, Indonesia, and earthquake records collected from a radon monitoring site and the USGS earthquake database from December 11, 2022, to August 8, 2023. The proposed method was trained on 80 of the dataset, which was utilized to generate a probability distribution for the Monte Carlo process to handle the constraints of limited precursor data. The results from the Monte Carlo simulations were then used to develop a model for predicting earthquake parameters. Experimental results demonstrate that the proposed method performs well within a monitoring station's radius of 300 and 400 km. At 300 km, the method outperforms in predicting magnitude, distance, and time, with RMSE values of 0.48, 60.60 km, and 57.85 hours, respectively. At 400 km, it achieves excellent performance with RMSE values of 0.61, 76.29 km, and 46.69 hours. This study shows that the proposed method outperforms benchmark methods in predicting earthquake parameters using radon gas as an earthquake precursor. ©2025 The authors.