--> ABSTRACT: Structural Curvature versus Amplitude Curvature, by Chopra, Satinder; Marfurt, Kurt J. ; #90142 (2012)

Datapages, Inc.Print this page

Structural Curvature versus Amplitude Curvature

Chopra, Satinder *1; Marfurt, Kurt J.2
(1) Arcis Corporation, Calgary, AB, Canada.
(2) University of Oklahima, Norman, OK.

Seismic curvature attributes can enhance subtle information that may be difficult to see using attributes such as dip-magnitude and dip-azimuth. Recently, curvature attributes and associated interpretation workflows have been developed by a broad range of interpreters such that curvature computations have found their way into most commercial interpretation software packages. Initially introduced as 2D computations on picked seismic surfaces, curvature computations from volumetric estimates of inline and crossline dip components followed soon after. Volumetric curvature is generated by taking derivatives of volumetric estimates of reflector dip and azimuth that best represents the best single dip for each single sample in the volume. We refer to these calculations as structural curvature, as the calculations are carried out on reflector depth or time.

We can also compute curvature attributes using seismic amplitude and refer to such a computation as amplitude curvature. Horizon-based amplitude curvature is in the hands of most interpreters. First, we generate a horizon slice through a seismic amplitude, RMS amplitude, or impedance volume. Next, we compute the inline and crossline derivatives of this map. Computing derivatives of these gradients (or second derivatives of amplitude) gives us amplitude curvature. Such maps can often delineate the edges of bright spots, channels, and other stratigraphic features at any desired direction.

Volumetric estimates of reflector amplitude gradients are computed in small windows along previous estimates of structural inline and crossline dip. To minimize the negative impact of noise on such computations, we favor computing derivatives of the coherent component of the data using a covariance matrix and principal component analysis such as commonly used in structure-oriented filtering. For data processed with an amplitude-preserving sequence, amplitude variations are diagnostic of geologic information such as changes in porosity, thickness and /or lithology. We have found that the computation of curvature on amplitude gradients furnishes higher level of lineament information that appears to be promising. The application of amplitude curvature to impedance images is particularly interesting where low-impedance, diagenetically-altered cracks can be nicely-highlighted.

As with structural curvature, one can generate rose diagrams of the lineaments that can be compared with similar roses obtained from image logs.

 

AAPG Search and Discovery Article #90142 © 2012 AAPG Annual Convention and Exhibition, April 22-25, 2012, Long Beach, California