ULF geomagnetic changes possibly associated with the 2008 Iwate–Miyagi Nairiku earthquake

• Published in: Journal of Asian earth sciences 2011-06, Vol.41 (4), p.442-449
• Main Authors: Hirano, Takuya, Hattori, Katsumi
• Format: Article
• Language: English
• Subjects: Normalized spectrum density ratio; Spectrum density ratio analysis; The 2008 Iwate–Miyagi Nairiku earthquake; ULF geomagnetic anomaly; Wavelet transform
• ISSN: 1367-9120; 1878-5786
• DOI: 10.1016/j.jseaes.2010.04.038

There are many reports on earthquake-related electromagnetic phenomena. Anomalous ULF geomagnetic field changes associated with earthquake is one of the most convincing and promising phenomena due to deeper skin depth. Since ULF signals associated with large earthquakes are weak, effective signal discrimination methods should be required. Several methods for the signal discrimination have been developed so far: which are spectrum density ratio analysis, geomagnetic transfer function analysis, fractal analysis, principal component analysis, direction finding analysis, and so on. In this study, we investigate ULF geomagnetic changes possibly associated with the 2008 Iwate–Miyagi Nairiku earthquake based on spectral density ratio analysis. Geomagnetic data observed at Esashi, where the epicentral distance is about 47 km and Kakioka, the distance is about 317 km, and as a reference station have been analyzed. Wavelet transform have been performed for the spectral density analysis instead of the conventional FFT method. Before the earthquake, the variation of spectral density ratio, S z / S x and S z / S y , at the nearest station of Esashi exhibits an apparent increase from the trend. On the contrary, there are no corresponding significant changes at a remote station of Kakioka. After investigating the singularity of the increase using normalized spectrum density ratio, the enhancement is the most significant in intensity and duration for the all analyzed period. The level of peak is beyond the 3 σ and its duration is 3 days. The lead time is about 3–4 weeks before the earthquake. At the periods from 3 to 105 s, similar anomalous changes occurred. These facts suggest the anomalous change is a possible candidate of earthquake-related ULF magnetic change.