Detection of anomalous radioxenon concentrations: A distribution‐free approach

Abstract

The detection of anomalous atmospheric radioxenon concentrations plays a key role in detecting both underground nuclear explosions and radioactive emissions from nuclear power plants and medical isotope production facilities. For this purpose, the CTBTO's International Data Centre uses a procedure based on descriptive thresholds. In order to supplement this procedure with a statistical inference‐based method, we compared several non‐parametric change‐point control charts for detecting shifts above the natural radioxenon background. The results indicate that the proposed methods can provide valuable tools for the institutions responsible for the verification and classification of anomalous radioxenon concentrations.

Automatic deforestation detectors based on frequentist statistics and their extensions for other spatial objects

“…Notably, the application of Bayesian hierarchical models has enriched understanding of land cover changes (Pirzamanbein & Lindström, 2022), while data science's role in environmetrics, especially in analyzing large datasets including satellite imagery, has become increasingly pivotal (Rodrigues & Carfagna 2023). Furthermore, the employment of nonparametric methods for anomaly detection (Scagliarini et al, 2023) and quantile regression for clustering satellite time series data (Musau et al, 2022) share methodological similarities with our work. Additionally, investigations into spatial dependence (Shooter et al, 2021) and atmospheric motions modeling (Sahoo et al, 2023) from satellite data reflect the broader utility of statistical techniques and satellite imagery in environmental monitoring, aspects our research also engages with.…”