Seismic multi-resolution velocity analysis in UDWT domain

Summary Different methods are used for seismic velocity analysis, including spectral analysis based on semblance, which is a reliable way to estimate the stacking velocity. The proposed method in this research is superior over other seismic velocity analysis methods because in the proposed method, the wave due to the absorption of the earth makes a reduction of the frequency content of the source in time, and results in poor temporal resolution. In this paper, the velocity analysis performed in the wavelet domain using undecimated discrete wavelet transform (UDWT). Random noise is often placed in high-frequency sub-bands and their effect leads to a reduction in coherency than other sub-bands. In this way, data analyses to high-pass and low-pass sub-bands and velocity analysis at any scale are performed using discrete wavelet transform and scaling filters. A Comparison of the obtained results showed that the spectral analysis of the velocity in the wavelet domain increases the accuracy of the velocity analysis in each band of frequencies.
Introduction One of the valuable information that can be obtained using seismic waves is the velocity of the earth layers, which can be useful for identifying the properties and petrophysical parameters of the layers. The velocity information is usually estimated when processing of seismic reflection data using the CMP velocity analysis is made. Velocity analysis is a powerful tool for detecting reflections and determines the stacking velocity of seismic data. Different methods are presented for analysis of seismic velocity. The UDWT method for signal decomposition has been introduced by Guo (1995).
Methodology and Approaches For a flat layer, the shape of the move out curve is defined by the hyperbolic relationship between the zero offset time and the velocity. Several methods of velocity analysis have been used in the past, but today, most velocities are selected interactively using combined displays on the processing workstations. Nevertheless, velocity analysis is still one of the most time-consuming parts of seismic processing. This is also the most critical step since velocity analysis is an initial interpretation of the data and it is important as the seismic interpreter is involved in the analysis and quality control steps. Velocity analysis is often performed several times during processing, which results in an iterative improvement in velocity estimation. The velocity spectrum display is calculated by determining how a given hyperbolic event matches actual events on the central CMP gather. These represent much more velocity trials than can be done using CVS or FVS analysis. The maximum coherence amplitude is expected when the hyperbola corresponds best fits to a given large amplitude seismic event. The coherence measurement most often used is called semblance, which is robust to noise, spatial aliasing, and lateral amplitude variations. There are various methods of displaying semblance. The average of too many gathers would increase computation time and could begin to filter geological variations. Wider peaks in the deeper part of the section indicate reduced resolution. The velocity spectrum is also good for identification of multiple reflections.
Results and Conclusions In this research, an improvement for the analysis of seismic velocity was made. In this paper, we have investigated the coherence technique based on discrete wavelet transform. We conclude that the cumulative spectrum in the UDWT domain gives a precise velocity spectrum.