Please use this identifier to cite or link to this item:
Type: Artigo de periódico
Title: Inverse Common-Reflection-Surface
Author: Perroud, H
Tygel, M
Freitas, L
Abstract: The Common-Reflection-Surface (CRS) stack method is a powerful tool to produce high-quality stacked images of multicoverage seismic data. As a result of the CRS stack, not only a stacked section, but also a number of attributes defined at each point of that section, are produced. In this way, one can think of the CRS stack method as a transformation from data space to attribute space. Being a purely kinematic method, the CRS stack lacks amplitude information that can be useful for many purposes. Here we propose to fill this gap by means of a combined use of a zero-offset section (that could be a short-offset or amplitude-corrected stacked section) and common midpoint gather. We present an algorithm for an inverse CRS transformation, namely one that (approximately) transforms the CRS attributes back to data space. First synthetic tests provide satisfying results for the two simple cases of single dipping-plane and single circular reflectors with a homogeneous overburden, and provide estimates of the range of applicability, in both midpoint and offset directions. We further present an application for interpolating missing traces in a near-surface, high-resolution seismic experiment, conducted in the alluvial plain of the river Gave de Pau, near Assat, southern France, showing its ability to build coherent signals, where recording was not available. A somewhat unexpected good feature of the algorithm, is that it seems capable to reconstruct signals even in muted parts of the section.
Subject: Numerical approximations and analysis
Controlled source seismology
Country: EUA
Editor: Wiley-blackwell Publishing, Inc
Rights: aberto
Identifier DOI: 10.1111/j.1365-246X.2010.04762.x
Date Issue: 2010
Appears in Collections:Unicamp - Artigos e Outros Documentos

Files in This Item:
File Description SizeFormat 
WOS000284274700023.pdf839.8 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.