Long wavelength estimation

Migration Velocity Analysis (MVA) techniques aim at determining the background velocity model for subsequent Full Waveform Inversion. The GPX team has defined different strategies for successful wave equation based MVA applications. In particular, research activities are conducted along the following lines:

-In which domain should MVA be formulated? Extra dimension could be classical surface offset, angle domain, sub-surface offset, time-shift, ... For a given Common Image Gather description, what is the optimal associated objective function (semblance, differential semblance, ...)?

-MVA techniques are classically restricted to reflected data. The GPX group has developed new strategies to take into account reflected and transmitted waves. Multiples have a broader illumination. This should offer the possibility to introduce more physics and to invert for anisotropy and in the future for attenuation and density models.

-It is also important to understand the current limitations for automatic applications on real data. For example, the GPX team has proposed a scheme to compensate for uneven illumination for a more robust velocity estimation. As for classical tomography and for full waveform inversion, it is also crucial to introduce suitable a priori information.

Contact: Chauris, Noble, Singh