Application of the empirical mode decomposition and wavelet transform to seismic reflection frequency attenuation analysis

Frequency attenuation analysis is a useful tool for direct hydrocarbon indication. Frequency attenuation gradient is more sensitive to the type of reservoir identification than other frequency properties such as center frequency, root-mean-square frequency. One of the derived properties for direct hydrocarbon detection is time-frequency spectral decomposition. Based on seismic attenuation theory in a fluid-filled porous medium, a new method combining the Empirical Mode Decomposition (EMD) and the continuous-wavelet transform named EMDWave is proposed as a high-precision frequency attenuation analysis and an improved time-frequency analysis methods. Compared to the Hilbert-Huang Transform (HHT) method, it reflects more details. The common frequency section calculated by the EMDWave method can improve the reservoir characteristics. First, the EMD method is used as multiband filtering in the temporal domain. The EMD method can decompose the original seismic signals into a finite number of Intrinsic Mode Functions (IMFs). All these IMFs can be expressed as gradual single-frequency signals that enhance the physical meaning of instantaneous frequencies and instantaneous amplitudes. After the correlation analysis of the original seismic signal and its corresponding IMFs, the IMF component which reflects more oil and gas information is selected for further hydrocarbon detection. The selected IMF with relatively narrow band can make the wavelet transform avoid the frequency loss incurred by a large scale distribution for broadband non-stationary signals. Second, the wavelet transform is applied to the selected IMF. The time-frequency spectrum obtained has a single-peaked spectrum with narrow side-lobes and it is good for computing the absorption coefficients. Then absorption coefficients are computed by curve fitting based on the least square method. The proposed method EMDWave effectively improves the precision of the conventional methods of Energy Absorption Analysis (EAA). Applications of the EMDWave method for hydrocarbon detection over a gas field located in western Sichuan Depression, China, show the effectiveness of gas bearing detection. It can improve the traditional attenuation analysis for better reservoir characterization.