Prestack Depth Migration of Ultrashallow Seismic Data. A Critical Analysis

Abstract

PSDM is increasingly becoming rule rather than exception in near surface seismic imaging when the goal is to get detailed sub-surface images and accurate velocity field. PSDM produces the seismic image of an earth model in depth converting the seismic reflection data into a depth section by means of a velocity field. Although different implementations of velocity model building tools exist, they usually start with an estimation of an initial velocity model using moveouts in the time domain and proceed with an iterative model refinement using residual moveouts in the depth domain and seismic reflection tomography.In addition, if available, a priori velocity/depth information, is also considered to improve the accuracy of the velocity model.In shallow seismic reflection imaging, instead, refraction tomography has been used not only to improve the velocity model built with PSDM but also as the exclusive tool to build the velocity model. In our tests, building a velocity model by means of a priori information unrelated with the recorded data or by means of the exclusive use of refraction tomography, PSDM produces depth images that have an imprint of the velocity models and not a result coming only from seismic reflection data.