Characterizing Sedimentary Cycle Using Synchrosqueezing Three-parameter Wavelet Transform
Y. Tian, J. Gao, N. Liu, Q. Wang and H. Chen
Event name: 81st EAGE Conference and Exhibition 2019
Session: Seismic Attributes - Spectral Methods
Publication date: 03 June 2019
Info: Extended abstract, PDF ( 1.74Mb )
Price: € 20
Sedimentary cycle division is significant in the sequence stratigraphy. Time-frequency analysis is a widely used method for the division of the sedimentary cycles. However, traditional time-frequency analysis methods can not characterize the sedimentary cycles accurately because of the huge difference between the time-frequency basic atoms and the waveforms of different seismic wavelets. In this paper, the precise recognition of the sedimentary cycles is realized by introducing the synchrosqueezing three-parameter wavelet transform (STPWT), in which the time-frequency basis atom (e.g. the three parameter wavelet) can match well with different seismic wavelets. Hence, the STPWT achieves good time-frequency localization characteristics and distinguishes the spectrum units accurately corresponding to different sedimentary cycle units in the time direction. In addition, the frequency direction resolution is improved by synchrosqueezed operation. The time-frequency analysis results of STPWT can clearly depict the time-frequency characteristics of the main frequency component corresponding to the positive cycle unit from high to low or the reverse cycle unit from low to high. Experiments on synthetic and field data demonstrates its validity and effectiveness.