Introduction

Focusing-effect AVO (FEAVO) consists of focusing seismic wavefield amplitudes through velocity lenses too small to generate proper triplications. The amplitude effects are large enough to thwart proper AVO analysis in the affected area Kjartansson (1979). Migration with a velocity model containing the lenses that cause FEAVO eliminates the effect Bevc (1994). However, the traveltime effects are too small to allow classical velocity analysis approaches such as Dix inversion or traveltime tomography to succeed. New approaches are needed to deal with such small velocity anomalies.

In previous work Vlad and Biondi (2002); Vlad (2002) Wave-Equation Migration Velocity Analysis (WEMVA; introduced by Biondi and Sava (1999)) was shown to resolve FEAVO-causing velocity anomalies, by optimizing image quality in the angle domain after prestack depth migration. The first step of the proof consisted in showing that FEAVO anomalies are recognizable in a field dataset migrated and transformed to the angle domain. WEMVA iterates by migrating with the current velocity model, extracting an image perturbation, and converting it into a velocity model update by inverting a linearized downward continuation operator. Vlad (2002) then showed that the inversion operator did not distort FEAVO anomalies into becoming unrecognizable, despite the linearizing approximations.

While the previous studies only demonstrated the possible suitability of WEMVA as a tool for the given problem, this paper shows the results of successfully running WEMVA on a FEAVO-affected synthetic dataset. In particular, it shows that the velocity lenses are found, and that the image migrated with the updated velocity model no longer exhibits FEAVO effects. This paper also explores the limitations of the type of WEMVA used and different ways of extracting the image perturbation to be fed into the inversion.