Is it possible to trace an impending earthquake's occurrence from seismo-ionospheric disturbance using principal component analysis? A study of Japan's Iwate–Miyagi Nairiku earthquake on 13 June 2008

Abstract

This research uses principal component analysis (PCA) to investigate global ionospheric integrated electron content map (GIM) anomalies corresponding to Japan's Iwate–Miyagi Nairiku earthquake of 13 June 2008 (UTC) ( M w =6.9). The PCA transform is applied to GIMs for 20:00–22:00 on 08, 11, and 12 June 2008 (UTC). To perform the transform, image processing is used to subdivide the GIMs into 100 (36° longitude and 18° latitude) smaller maps to form transform matrices of dimension 2×1. The transform allows principal eigenvalues to be assigned to ionospheric integrated electron content anomalies. Anomalies are represented by large principal eigenvalues (i.e., >0.5 in a normalized set). The possibility of geomagnetic storms and solar flare activity affecting the results is done through examining the Dst index for the corresponding days. The study shows that for the Iwate–Miyagi Nairiku earthquake, PCA possibly determined earthquake-related ionospheric disturbances for the whole region, including the epicenter.