The propagation of seismic waves through visco-acoustic media is affected by frequency dependent absorption which is often described by the quality factor Q where a low Q means more loss of signal strength and bandwidth. Complex variations in attenuation, if not accounted for, can severely compromise both the amplitude and phase of the migrated data. This in turn affects the ability to accurately predict reservoir properties. In this paper we propose a new tomographic approach using adaptive centroid frequency shift (CFS) information from surface seismic data to estimate Q.
By picking the events in the migrated section and ray-tracing back to the unmigrated data domain, the centroid frequency of the unmigrated data can be measured for the picked events in the depth migrated CIG gathers. After applying the correction generated on the fly for the given source wavelet, these adaptively corrected CFS will then be back-projected along ray path to reconstruct the attenuation distribution through our tomographic inversion.
A synthetic test and a real data example will be presented to demonstrate how our approach can accurately estimate a Q model and can be included in the Q compensation process to fully account for the frequency dependent attenuation effects observed on seismic data.
A key element of the CFS method is deciding what analytical function should be used to fit the amplitude spectrum of wavelets before and after passing through visco-acoustic media. However, we found the accuracy of Q tomographic inversion to be sensitive to the accuracy of the fitting and that the fixed wavelet fitting function cannot describe the real source wavelet accurately. An adaptive correction is applied to the observed centroid frequency to account for any deviation from the explicit relationship through tabulating the absorption effect for different accumulated dissipation time. These adaptively corrected centroid frequency shifts improve the stability and the accuracy of the inversion.
The propagation of seismic waves through viscoacoustic media is affected by frequency dependent absorption which results in loss of signal strength and bandwidth. Complex variations in attenuation, if not accounted for, can severely compromise both the amplitude and phase of the migrated data. This in turn affects the ability to accurately predict reservoir properties (Best et al., 1994). Thus, there is a need to compensate for the frequency dependent absorption during the processing of the data.
