A good trick to decrease the noise level in the migration result is to use the water-bottom reflection only for the source. This can be done most of the time by applying a simple mute to the data. Figure shows the up- and down-going wavefields if the water-bottom reflection is used as a source. Figure displays a comparison of the migration of multiples with all the primaries as the source (Figure a) and the water-bottom as the source (Figure b). The noise level has decreased without loosing structural information. This result proves that for these data, most of the surface-related multiples are water-column reverberations.
The need for separating multiples might be quite dissuasive when field data are imaged because it might involve heavy computations and/or an earth model that might not be known in advance. Let us consider for a moment that we do not want to do the multiple attenuation but still want to do some imaging with multiples. This goal requires that the recorded data with primaries and multiples are used for both up- and down-going wavefields (Figure ). Now I compare in Figure the migration results when only multiples (Figure a) and multiples plus primaries (Figure b) are extrapolated in the up-going wavefield. The migration of primaries and multiples yields a noisy image but never-the-less structurally interpretable.
In the next section I migrate multiples for two 2-D lines. One survey comes from the Gulf of Mexico and the other from the North Sea. They demonstrate potential strengths and weaknesses of the proposed method for multiple migration.