Big Data Analyses of Neutron Monitor Measurements: An example with the Correlation between Forbush Decreases, Geomagnetic Storms and other Space-Weather Variables

Abstract The ground-level network of neutron monitors (NMs) remains the most important instrument for investigating cosmic ray (CR) time-intensity variations over timescales of up to seven decades. The very high counting rate and long-term reliability of NMs are competitive advantages they hold over detectors in space. It has long been established that a Forbush decrease (FD) is one of the most spectacular and significant short-term changes in CR flux. Many publications have demonstrated that FDs can be used as a crucial space-weather variable. This is because they frequently show a significant correlation with indicators of dangerous space-weather phenomena, such as the intense geomagnetic storm index (Dst). However, quick and successful predictions of these phenomena remain an open research interest. FD is generally considered a good space variable. Nevertheless, the manual/case event method of calculating FD events in preparation for subsequent quantitative and qualitative correlation analyses is slow and ineffective for predictive purposes. The same can be said of FD-based superposed epoch analysis (SEA), which is commonly employed in the field. To demonstrate the efficient predictive features of the fully automated method (FAM), we analyzed large volumes of hourly CR data, Dst and other variables for three solar cycles (1996-2024). Implementation of Big Data analysis in space-weather is timely since the automated data acquisition techniques applied in NM measurements allows each station to accumulate large volume of CR data. Our results show that the FAM replicates the SEA/case event studies. We conclude that since FAM is much faster than the manual techniques, it may be useful in the investigation and prediction of space-weather conditions.