Conclusions

The cross-correlation of the receiver wavefield by the source wavefield imaging condition has the advantage of being robust and honoring the kinematics of Claerbout's imaging principle. However, it has the disadvantage of losing resolution and creating image artifacts in the presence of multipathing.

A better imaging condition is obtained by computing the reflection strength as the zero lag value of the deconvolution of the receiver wavefield by the source wavefield. This approach was implemented in the time domain using least squares inverse filters and in the Fourier domain as the scale sum of the frequency components of the wavefields division.

Using the zero lag of the deconvolution imaging condition, we attenuate image artifacts due to wavefield multipathing through velocity anomalies. Also, we showed that it better handles amplitudes during imaging. The preceding statements are corroborated using two different synthetic data sets: one based on the convolutional model and the other based on a wave equation modeling program.