A robust seismic time-frequency analysis method based on general linear chirplet transform

What is it about?

Time-frequency analysis (TFA) has been widely used in seismic processing and interpretation. We apply a robust seismic TFA method based on the general linear chirplet transform (GLCT). The algorithm follows such a simple logic and produces a broadband time-frequency spectrum free of cross-term interference, resulting in good performance characterizing the instantaneous spectral variations. Both the synthetic and field data demonstrate that GLCT can be a reliable tool for seismic facies interpretation improvement and direct hydrocarbon indication, particularly in noisy condition.

Featured Image

Why is it important?

To meet the requirement of subtle reservoir characterization, time-frequency analysis (TFA), an important tool used in spectral decomposition for seismic interpretation, has become a common practice for local frequency representation and attribute extraction. In the past few decades, it has taken effect in several aspects, such as stratigraphic visualization, reservoir thickness estimation, and direct hydrocarbon detection. Since the quality of time-frequency representation (TFR), refleted by such properties as energy concentration and the variance of derived parameter estimates, is dependent on how well-adapted the TFR is to the signal, the chirplet transform (CT) offers more flexibility for parsimonious representation than traditional TFA methods with unmodulated kernels. The GLCT method is an extended form of CT, which is a unifying framework encompassing the short time Fourier transform (STFT) and the continuous wavelet transform (CWT) using the chirplet atom as the kernel function instead of the sinusoidal wave or wavelets. GLCT gives a good trade-off between the time-frequency resolution, robustness and computational effeciency, which may serve as a promising tool for seismic data interpretation and reservoir characterization in practice.