ANALYSIS OF EARTHQUAKE DATA USING WAVELET TECHNIQUES

This dissertation presents the analysis of earthquake data collected from four locations in the USA over the same duration. Data was collected from United States Geological Survey website. Primary location is prone to regular seismic activity, the other two locations are nearer to this location and the final location is relatively far. Analyses have been carried out using fast Fourier transform (FFT) and wavelet transform. The advantages of wavelet transform are ascertained. Then data is analyzed using Continuous Wavelet Transform (CWT) which noticeably shows the variation of intensity with respect to time and thus providing the trends and patterns of earthquake data in real time. This can be used to study earthquake behaviour, cycle etc. Further advanced features available in wavelet transforms such as independent thresholding, different levels of decomposition and db methods were used to eliminate the high frequency noise terms. The results illustrates that the time and frequency analysis carried out by wavelet transform provides a better analysis of a signal which is not possible through traditional methods like FFT. Wavelets give good time resolution at high frequencies and at low frequencies gives good frequency resolution. The report indicates the advantages of using advanced features in wavelet transform in removing noise from the earthquake data.SCHOOL OF ENGINEERIN