Kumiko Hattori

Reduction of ULF external source origin signals (periods t<1,000 s) using of interstation method and wavelet transform

Summary form only given. Ground-based observation of ULF geomagnetic field changes is one of the most promising methods to monitor seismic and volcanic activity due to the skin depth. The observed ULF magnetic and electric potential fields are considered to be the superposition of several possible signals. Intense external source fields in periods of 1-1000 sec include those associated with solar-terrestrial interactions, noise from DC motors, and local sources such as movement of magnetized objects at a site. Therefore, signal separation and effective methods for discrimination of ULF geomagnetic signals are significant. We present an effective time series analysis for reducing the field variations originating from external sources. Our approach, which uses reference geomagnetic data and ULF data collected at different sites, is called the interstation transfer function (ISTF) method. We have adopted a wavelet transform using the Morlet wavelet for the ISTF approach. We have applied the proposed ISTF method to actual data. Results indicate that the proposed ISTF method is effective in reducing signals originating from external source field variations in time periods less than 1000 sec and to enhance the other signals. Both model computation of SES detection and application to the 2002 Boso slip show the capacity of the proposed method.