The strategy of WEMVA differs from that of all the previous attempts: instead of trying to fit the data, it tries to fit the image. The tactics are different as well: instead of integral (ray-based, Kirchhoff) operators, it uses differential (wave-equation) ones, with all their well-known advantages. Unlike most of the previous approaches, this one is capable of using a starting guess from classic velocity analysis. Therefore, it simply fine tunes the velocity model for small velocity anomalies. It requires no picking and tries to match the entire image using information contained both in the amplitudes and traveltimes.
This new approach, unlike the old ones, takes into account all the characteristics of the FEAVO anomalies, including the variation with depth. This goal is achieved by measuring anomalies on surfaces in the depth-midpoint-angle volume, instead of simple ``V''s in the midpoint-offset plane. An adaptation of the signal-noise separation technique described in Harlan (1986) will assure that the image perturbation contains only information related to the FEAVO anomalies.
My approach uses the strategy and framework of WEMVA as described in the corresponding section, but differs by the change of objective function in the inversion. The usual WEMVA criterion describing the quality of the image is flatness in angle gathers that is directly related to traveltime anomalies. The traveltime changes associated with the FEAVO effect are very small and do not produce whole-event curvature, but only wiggliness in angle gathers. Biondi and Sava (1999) show on a synthetic, and this paper will show on a real dataset, that FEAVO anomalies keep their ``V'' shapes through prestack migration and conversion from offset to angle gathers. Therefore, the fitting goal of the inversion must be related to the distribution of amplitudes in the midpoint-angle space. The desired image will not exhibit the characteristic ``V'' patterns in the midpoint-offset plane.
My method might be able to discriminate between amplitude anomalies caused by absorption and those caused by velocity because both kinds of anomalies have two different ``signatures'' in the image space. In the case of velocity, the high amplitudes are found close to the low amplitudes: the energy is not lost but it is only focused locally. In the case of absorption the FEAVO effects are not ``bipolar.''