Anomaly recognition in neutron monitor observations using wavelet decompositions and statistical decision theory rules
O.V. Mandrikova, B.S. Mandrikova
Institute of Cosmophysical Research and Radio Wave Propagation of the Far-Eastern Branch of Russian Academy of Science, Paratunka, Russia
Keywords: cosmic rays, neutron monitor observation, Forbush effect, space weather, wavelet decomposition, statistical decision theory rules
Abstract
During disturbances in near-Earth space, disruptions in ground-based and satellite technological systems, including catastrophic failures, occur. Therefore, the problem of developing methods for real-time analysis and monitoring of the natural environment with acceptable accuracy is particularly pressing. This article explores a new automated method for detecting anomalies in neutron monitor (NM) observations. This method is based on the synthesis of wavelet decompositions with the rules of statistical decision theory; it enables the detection of anomalies in NM observations and the assessment of their intensity. Discrete wavelet decompositions with adaptive threshold functions are used to detect anomalies. The parameters of the threshold functions are automatically estimated (as data entered the processing system) using the rules of statistical decision theory. The application of these rules yielded a solution with an error not exceeding an a priori specified value. The intensity of anomalies in NM observations is then calculated by summing the amplitudes of wavelet coefficients exceeding the estimated thresholds. The paper studies periods of strong and extreme geomagnetic storms in 2024-2025. Correlations between Dst index and the intensity of anomalies in NM observations are analyzed. The results showed a significant increase in the correlation between the Dst index and the intensity of anomalies in neutron monitor observations during geomagnetic storms. The obtained correlations reached their maxima with a delay of several hours, demonstrating the importance of neutron monitor observations and their consideration when solving space weather problems. The results of the work can be used in space weather forecasting for the early detection of sporadic variations in the cosmic ray flux.
|
