Removal of Outliers in Geomagnetic Field Time Series Using the Hampel Filter

The paper presents a methodology for removing outliers in geomagnetic field time series using the Hampel filter. Quality metrics for binary data classification based on the confusion matrix demonstrated the effectiveness of the method and are comparable to those for similar algorithms used for outlier removal in 1-second magnetograms of the international INTERMAGNET network. Outliers identified using the developed methodology for the period 2019-2022 at the Ak-Suu base station exhibit a seasonal distribution pattern that correlates well with thunderstorm activity. The method enhances the quality of preliminary data processing for the geomagnetic monitoring network of the Research Station of the Russian Academy of Sciences, specifically by automating the procedure of magnetograms outliers filtering.