The inverse data space : examples and practical aspects

It has been proposed that in the case of strong multiple scattering, the data matrix P is replaced by its inverse P−1 before starting seismic processing. In the inverse data space, all surface-related multiples map onto the origin, whereas the inverse of the primaries map mostly at negative times. This interesting property opens up new ways to separate primaries from multiples and apply other preprocessing steps. In the previous chapter the concept was introduced and illustrated with two simple examples. In this chapter, the inversion of seismic data is demonstrated for a more complicated 2D example. Furthermore, a new strategy for obtaining the inverse of the data is developed, without involving explicit multi-dimensional inversion.