[ Home ]

Abstract Summary: AVO crossplotting is a widely employed AVO analysis technique that has gained acceptance in the geophysical community over the past decade. Crossplotting AVO attributes such as the AVO intercept and gradient has proven useful in hydrocarbon diagnostics in unconsolidated clastic basins worldwide.  Improved understanding and interpretation of the crossplot results can be obtained through 3D visualization of the AVO crossplot.  Modeled and observed 3D seismic crossplots, through visualization techniques, demonstrate the improved recognition of background trends and associated AVO anomalies, which in turn improves interpretation and analysis.

Introduction : It has been demonstrated in the literature and through professional presentations that AVO analysis can be a diagnostic tool in hydrocarbon detection and delineation.  AVO analysis, using first two-dimensional data, and now three-dimensional data has proven its value in both exploration and production projects.  The present momentum toward 3D AVO analysis continues to yield positive results.

Over the past decade, AVO analysis has shifted from single (combined) attribute analysis to a multi-attribute approach, and AVO (attribute) crossplotting is one such technique that is widely practiced.  In addition to computing combined AVO attribute sections using the intercept, gradient, near angle stacks and far angle stacks, or their equivalents, both data sets can be plotted against one another through crossplotting techniques.  Under ideal situations, common lithologies and fluid types generally cluster together in AVO crossplot space permitting a straightforward interpretation. Together with AVO modeling and the AVO attribute section, the crossplot is used to evaluate the potential for discriminating fluid-type and or lithology of the objectives.  AVO crossplotting, which typically uses the AVO intercept (A) and gradient (B), assists standard AVO analysis techniques by identifying background trends and anomalous responses (off trend aggregations) that may or may not be associated with hydrocarbons (Castagna and Swan, among others, have presented work regarding these concepts).   Using the A and B data volumes and small target specific windows where Vp/Vs are nearly invariant, a background trend can often be determined which defines the wet-sand/shale interfaces and other similar lithologies.  AVO (A-B) pairs lying off the trend are considered anomalous. Through interactive testing, with a priori information through modeling, and with geological integration of the basin geology, the crossplot can be used to assess hydrocarbon-bearing strata or key lithologies.  In light of the aforementioned, crossplotting has evolved to be a crucial component in AVO analysis.

However, even with current computer speeds and inexpensive disk drive costs, interactive crossplotting of seismic data is computationally intensive.  A 23 square kilometer area with 25 meter square bins and 6 s records at a 4ms sample rate results in nearly 3.6x10 8 A-B pairs, and for a 100 sq. km area using a 1 s window, 4.0x10 7 A-B pairs.  Because of the large quantities of crossplot pairs, much of the seismic crossplotting performed to date on projects has taken the form of 2D profiles or small subsets of seismic volumes, resulting in a composite-summary crossplot of the A-B pairs.  These composite-summary crossplots have worked fairly well for the classic AVO anomalies, but can be confusing for the not-so-obvious objectives. The dominant reasons for the shortcomings of the composite-summary crossplot are: the immense amount of points associated with the background responses which clutters the displays (even for reduced subsets of the entire volume); and second, the three dimensional nature of the subsurface being analyzed. These two detractors to the composite-summary crossplot are linked, and through standard seismic visualization techniques can be overcome.

Biographical Sketch: Mark Sparlin works with Hampson-Russell in the capacity of Senior Research Geoscientist where he is currently involved with consulting services in AVO analysis and neural-network /geostatistical seismic attribute evaluation in applications to 3D seismic/borehole reservoir characterization projects. Mark has twenty years of experience working in the oil industry for major oil companies as an exploration geophysicist and Schlumberger Technology Corporation as Manager of Geophysical product development. He has also been employed with The University of Texas Research Campus involved in geophysical/geological evaluations in international reservoir characterization studies.

Mark received a B.S. in geology from the University of Akron with honors and an M.S. in Geophysics from Purdue University.  Email:  msparlin@hrs-us.com