Quick Search:    advanced search

GSW Home   GeoRef Home   My GSW Alerts   Contact GSW   About GSW   Journals List   Help

This Article Figures Only Full Text Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Zhang, L. Articles by Bartel, D. C. Search for Related Content GeoRef GeoRef Citation

Classification of salt-contaminated velocities with self-organizing map neural network

ChevronTexaco Exploration and Production Technology Company, Houston, Texas, U.S.

David C. Bartel

ChevronTexaco USA Production Company, New Orleans, Louisiana, U.S.

Corresponding author: linz@chevrontexaco.com

The first 20% of the full text of this article appears below.

In this article we will discuss a self-organizing map neural network for distinguishing salt-contaminated velocities from sediment-only velocities. Our objective is to demonstrate the capability of the self-organizing mapping neural network in automatic classification of salt-contaminated velocities as a replacement for interpreter's picking. The classified salt-contaminated velocities are then used to build a velocity model for subsalt imaging. We will illustrate our work with a Gulf of Mexico data set.

Subsalt imaging is in demand for hydrocarbon exploration in the Gulf of Mexico where subsurface salt covers a large portion of the deepwater region. Because of a large velocity contrast between salt and the surrounding sediments, reflections from subsalt strata are generally weak. Imaging of the subsalt strata requires accurate information about salt geometries including top and bottom salt surfaces. The imaging is in general divided into several steps: top of salt picking, salt velocity flooding, bottom of salt imaging, and finally, subsalt imaging. This work is concerned with top salt picking.

To pick a top salt surface, a sediment-only velocity model must be built to do the initial depth migration. The initial sediment-only velocity model excludes all velocity curves that are contaminated by a salt body. In the ideal case in which seismic rays pass through a thick salt body, the velocity curves can be easily identified because of the high salt velocity. However, when seismic rays pass through a thin salt body such as the edge of a salt body, the corresponding velocities may be easily interpreted to be sediment-only ones. Determination of the area around a salt boundary (the halo) is to a large degree an interpretive step. Another practical problem with the current velocity classification approach is that it is time consuming. Manual classification of salt-contaminated velocities can take a significant amount of time, especially for . . . [Full Text of this Article]