Target-oriented Migration Deconvolution: Practical Application and Implementation

Conventional seismic migration usually suffered from the presence of migration artefacts, poor illumination and deblurring. This problem arise not only because of physical constraint during seismic acquisition but also due to the use of adjoint operator during seismic migration. Least square migration helps to mitigate some of this issue like poor illumination better focusing but it was computationally intensive and time consuming. Therefore, the use of migration deconvolution process is a better alternative in order to obtained similar results compared to least square migration. This method requires only single iteration to obtained the deblurring filter compared to 10 iteration that were needed for least square migration to achieve convergence. In this paper we proposed combining the migration deconvolution approach with target-oriented migration in order to achieve faster runtime making it more efficient to apply for commercial usage. The proposed workflow are applied on both synthetic and real dataset, and the results obtained prove to be promising for further future application.