Abstract

Coherent noises exhibit the similar behavior as the signal in the time series. The problem is seen particularly for transient signals in geophysical time series. In this present study a hybrid approach based on Wavelet theory and Principal Component Analysis(PCA) theory is proposed for denoising the coherent noise in geophysical time series. The application is shown for a GPS time series to enhance slow slip earthquake events of Cascadia region in north west of North America. Exploitation of wavelet based denoising techniques such as soft or hard thresholding and then PCA has a pronounced effect on denoising of time series.