-
f A prestack interpolation primer Part 1: Tutorial
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 11th Simposio Bolivariano - Exploracion Petrolera en las Cuencas Subandinas, Jul 2012, cp-330-00111
Abstract
Prestack interpolation of land data is a topic of great interest in modern seismic processing. Prestack interpolation is used for diverse applications ranging from infilling of sparse acquisition in areas exhibiting complex topography or otherwise inaccessible terrain to data regularization for azimuthal processing required in fracture detection analysis to minimizing sampling-induced prestack migration artifacts in AVO analysis to harmonizing the sampling across different surveys in merge processing. Yet for all its popularity, prestack interpolation seem to be poorly understood by the geophysical community at large and misconceptions abound concerning the question of what prestack interpolation can and cannot do. The purpose of this tutorial presentation will be to be to provide the geophysical interpreter with a clear understanding of the algorithmic assumptions and relative pros and cons for two powerful interpolation techniques so that he/she can more reliably appraise their performance in a production environment. The first of these is the popular 5D interpolation by Fourier reconstruction approach (often called 5DMWNI) and the second is an approach based on local time-space dip-scans and slant stacks. We will focus on simplifying the basic theory and we will illustrate the strengths and weaknesses using very simple synthetic data testing. In particular we will distinguish between two very important and different interpolation tasks: regular upsampling and gap filling of random holes and we will carefully study the algorithms’ response to spatially aliased input data.