An automatic time imaging using Common Scatter Point gathers

Abstract

Prestack time migration (PreSTM) still represents the majority of seismic imaging activities in the industry. The reason for this is the speed and robustness of time imaging and its ability to focus seismic events for most geological settings. However, well defined migration velocities are necessary. When we speak about migration velocities and migration operators we implicitly assume time migration velocities and time migration operators. Usually, the migration velocities are not known and, therefore, approximated by stacking velocities. The stacking velocity is attached to the stationary point of the migration operator, however the migration velocity is defined at the apex of the migration operator. Due to this discrepancy, the stacking velocity may deviate from the migration velocity, e.g., in the case of dipping reflectors, and the final time image may be unfocused. In order to correct for the velocity difference and to enhance the time imaging, the velocity model needs to be refined. There are several techniques to provide an updated velocity model. However, all the techniques are based on an iterative approach requiring a strong manual interaction. We propose a new workflow for the time imaging which includes an automatic update of time migration velocities. The workflow is based on an automatic Common‐Reflection‐Surface (CRS) stack of Common Scatter Point (CSP) gathers and provides an improved time migration velocity model and, hence, a highly focused time image.