Practical Benefits of Kirchhoff Least-squares Migration Deconvolution

Kirchhoff images can suffer from uneven illumination and contamination by migration artefacts, particularly in regions of more complex geology. One issue is that migration is not a true inverse operation – it is based on the adjoint of the forward modelling operator. In contrast, least-squares migration approximates the inverse of the forward modelling and, with a sufficiently accurate velocity model, the impact of the detrimental effects on the image can be reduced. Here, we describe the benefits of a non-iterative Kirchhoff least-squares method, usually referred to as migration deconvolution. We present a practical workflow to demonstrate that the method can be used to attenuate image artefacts, particularly the strong wave-front swings, to help balance image illumination, and increase clarity and robustness of AVO attributes. We illustrate our workflow on a real data example from offshore Gabon, in which we are careful to ensure that both the conventional and migration deconvolution have the same pre-processing sequence applied prior to imaging. This sequence includes elements (de-ghosting, spectral balancing, regularization and interpolation) designed to improve the quality of the conventional Kirchhoff image, as well as de-multiple and de-noise processing. We observe a distinct, yet realistic, uplift from the migration deconvolution over conventional migration.