Principal Component Analysis of Ground Level Enhancement of Cosmic Ray Events

Abstract

We applied principal component analysis (PCA) to the study of five ground level enhancement (GLE) of cosmic ray (CR) events. The nature of the multivariate data involved makes PCA a useful tool for this study. A subroutine program written and implemented in an R software environment generated interesting principal components. Analysis of the results shows that the method can distinguish between neutron monitors (NMs) that observed Forbush decrease (FD) from those that observed GLE at the same time. The PCA equally assigned NMs with identical signal counts with the same correlation factor (r) and those with close r values equally have a close resemblance in their CR counts. The results further indicate that while NMs that have the same time of peak may not have the same r, most NMs that have the same r also had the same time of peak. Analyzing the second principal component yielded information on the differences between NMs having opposite but the same or close values of r. NMs that had the same r equally had the tendency of being in close latitude.