Time domain trace interpolation for irregularly sampled seismic data by using a localized dip search approach

Abstract

Spatial aliasing of seismic data is caused by inadequate spatial sampling of higher frequency dipping events. In seismic data acquisition there is always a trade‐off between the absolute optimum field parameters and the cost of the acquisition. The acquisition design is usually such that primary events are sampled adequately, but in many cases broadband noise events with low apparent velocities cannot be sampled in a cost effective manner. As a result, these steeply dipping events are recorded as aliased data (Li, 1987, and Zhou, 2000). Spatial aliasing of coherent noise events causes problems for many of the various noise suppression algorithms because the algorithms they use cannot determine the true dip of the aliased events and therefore, these data are not handled properly. A trace interpolation which can predict the true dip of the noise correctly, can interpolate the aliased events properly. Moreover the increase of the spatial sampling of the data will increase the effectiveness of the multichannel noise suppression algorithms commonly used in data processing (Mostafa, 2008, and Porsani, 1999). This paper begins with a basic overview of spatial aliasing and then describes which aliasing conditions typically cause problems in common interpolation algorithms. It then presents a new anti‐aliasing trace interpolation method based on the use of 2D localized dip searching. Results from this new interpolation method will be shown for both synthetic and real data examples. The results will be compared with the results generated using standard F‐X trace interpolation (Spitz, S., 1991)