Wavelet Transformation And Its Application In Analyzing Seismic Electromagnetic Signal

A large amount of observations and studies suggest that electromagnetic (EM)anomalies appear before moderate and large earthquakes, which can be detected byboth stations on the ground and satellite. Using these data, scientists have beenmaking efforts to reveal the relationship between EM anomalies and earthquakes intime and space, which may contribute to successful earthquake prediction. To attainthis purpose, raw EM data should be firstly processed and then analyzed to facilitatefurther identification of reliable EM anomalies related with earthquakes. It remains,however, an unsolved, or at least a debated issue, how to extract meaningful EMsignals from flood data. Thus it needs to develop new methods. This study attempts toapply the wavelet methods to analysis of the EM signals recorded during theaftershocks of the2013Lushan earthquake, and tries to seek the possible relationshipbetween variations of the EM fields and release of seismic energy.The contents of this thesis are summarized as follows:1introduction of the development of wavelet transform and its application ingeophysics. Wavelet transform put forward the progress of the signal processing,making up the weakness of the Fourier transform and short time Fourier transform Sofar, not only theory of the wavelet analysis continuously made breakthrough,theapplication of the method have also been very extensive. In geophysical area,thewavelet method were applied in many aspects mainly include: the seismic signalanalysis, the analysis of the gravity anomaly and the underground material singularitydetection, and the electromagnetic signal noise suppression research2Mathematical principles of continuous and discrete wavelet transforms. Thewavelet transform has the characteristic of time-frequency domain analysis. Itovercomes the limitations of the Fourier transform and the short-time Fouriertransform in analyzing non-stationary signals. The diversity of wavelet basis functionmakes the wavelet transform have good flexibility and adaptability. This work tried to introduce the wavelet transform to the analysis of EM signals to reveal theinformation contained. In this thesis, two kinds of functions of the wavelet transformare used. One is its filter function of the discrete wavelet transform which candecompose the signal into different scales of approximate components and detailcomponents in some multi-scaled space, and the approximate and detailedcomponents correspond to separate band of frequency. The other is the detection ofthe anomaly in the signal using the wavelet maxima method. The local regularity ofsignal can be expressed by the attenuation of the wavelet coefficient with the changeof the scale. At the same time the position of the singularity can be detected by theappearance of the modulus maxima at the fine scale.3Processing and analysis of EM data recorded during the aftershocks of the2013Lushan earthquake using wavelet transforms. EM Data processing isperformed in two aspects: Firstly, this work decomposes the original time series of themagnetic field into different scales and then calculates the energy variation in thewhole observation time. It is seen that pulses become intensive around theearthquakes. And then，this work processes the data using the traditional MT method,and controls the quality by the smoothing the apparent resistivity curves, yielding thevariation of the autopower spectrum of the eligible data. For identifying the EManomalies, this work calculates the wavelet maxima lines of the autopower spectrumcurves for six frequencies. The identified anomalies are further examined byLipschitz-exponent α. By comprehensive analysis of the anomaly distribution of thethree magnetic field components and the maxima line of the aftershocks’ energyvariation, this study suggests three possible EM anomalies associated withearthquakes, which appeared on15May,24May and6June of2013, respectivelyThis thesis analysis the electromagnetic signals recorded during the aftershockperiod of Lushan earthquake using the wavelet methods and try to find the innerconnection between the different methods for the purpose of mining the anomalies ofthe electromagnetic fields related with the earthquakes. And it presents another point of understanding view of the electromagnetic fields.