Propagation source wavelet phase extraction using multi-taper method coherence estimation

• Main Authors: Diako Hariri Naghadeh, Christopher Keith Morley
• Format: Journal
• Published: 2018
• Subjects: Engineering; Environmental Science; Agricultural and Biological Sciences
• Online Access: https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85010669535&origin=inward; http://cmuir.cmu.ac.th/jspui/handle/6653943832/46777
• Institution: Chiang Mai University

© 2016 Sinopec Geophysical Research Institute. It is possible to use statistical methods to extract the propagation source wavelet phase from seismic data without getting information from a well log. Using kurtosis as a high-order statistics can preserve the phase of the signal but it is highly sensitive to outliers. A new method is introduced here called the multi-taper method coherence estimation. Two steps are required: first, a cosine function that includes the dominant frequency and maximum amplitude of signal is chosen. Secondly, the maximum coherence in the frequency band of the signal, which shows the best phase matching between the time series is determined. To validate this new method real data sets were chosen and the extracted wavelet phases for noise free and noisy data sets were compared with data extracted from a well log. Extracted wavelets using Kurtosis were also generated for comparison, and demonstrate the improved results using the new method.