Improved High-Resolution Time-Frequency Decomposition for Detailed Reservoir Characterization

Time-frequency localization has a wide range of application in geophysics. In seismic stratigraphic interpretation, high-resolution spectrogram have uses both for reservoirs that lie below and above the seismic resolution limits. In case of the reservoir where the thickness is greater than tuning thickness, high-resolution spectral decomposition provides an indicator of the area of interest, whereas, in case of thin-bed reservoirs, advance spectral decomposition helps to remove the tuning effects and improved frequency-dependent AVO analysis. Commonly methods help to enhance the resolution, visualize structural and stratigraphic features, estimate thin bed thickness, spectral balancing as well as direct hydrocarbon indication (Burnett et al., 2003; Chen et al., 2008; Xiaoyang et al., 2014). In the paper, we present a new modification to the Modified Stockwell transform (MST)(Li et al., 2016), and incorporate both the benefits of MST and Gabor decomposition into a single algorithm. By doing so the robustness of MST is improved for small window size. We named this algorithm MST2. The MST2, is perfectly reconstructable which leads to high time-frequency signal processing for better reservoir characterization. We compared the MST2 algorithm with MST, Gabor decomposition, and Continuous Wavelet Transform (CWT) for its time-frequency localization and characterization. In the example shown in this paper, we also demonstrate the use of MST2 usage for improved structural visualization.