<title>Migration deconvolution</title>

Abstract

Reflectivity images of the earth are calculated by migrating discrete grids of seismic traces. For 3D data, these traces, typically, are spatially undersampled and so give rise to migration noise sometimes known as the `acquisition footprint'. In this paper, we show how to partly deconvolve the acquisition footprint noise from migration sections. This is accomplished by spectrally dividing the migration section by the migration Green's function, subject to the assumption of a laterally homogeneous velocity. Results with synthetic and field data show that migration deconvolution can be noticeably improve the spatial resolution of migration images, decrease the strength of migration artifacts and improve the quality of the migration image.