The great promise of amplitude-versus-offset (AVO) analysis lies in the dependence of the offset-dependent-reflectivity of reflected compressional waves on the elastic properties of the subsurface. As different lithologies may exhibit distinct Poisson's ratios, and gas bearing strata usually exhibit anomalously low Poisson's ratios, AVO has been recognized as a potential seismic lithology tool and direct hydrocarbon indicator. Unfortunately, experience has shown that the theoretical potential for AVO analysis is, all too often, not realized in practice. Despite some notable successes, it cannot be claimed that the economic impact of AVO analysis is on a par with other breakthroughs in geophysical technology such as bright spot analysis or 3D seismic surveys. This lack of penetration of AVO into the mainstream as an interpretation technology is not due to any inherent lack of validity of the concept. Rather, it is due to a lack of credibility of the concept in the minds of potential practitioners. The purpose of this presentation is to discuss the various reasons for this mixed acceptance by explorationists with the hope that better understanding of these factors will lead to better utilization of the method and more astutely targeted research and development.
Factors resulting in poor or incorrect utilization of AVO analysis can be divided into four categories: (1) lack of robustness; (2) interpretational complexity; (3) poor analysis of uncertainty, and (4) inappropriate utilization. I will comment on each of these:
Lack of Robustness
I hesitate to dwell too much on this aspect, because it can easily deter the faint hearted. However, one cannot overemphasize the need to understand the quality of the seismic data. This requires that the interpreter look at the CDP gathers, the acquisition geometry, and the processing sequence in detail. This should challenge interpreters of the "work station generation" and warm the hearts of surviving old timers who remember when interpreters needed to know geophysics! Keep in mind that AVO is bereft to a large extent of reflection seismology's most powerful weapon; COP stacking. Multiples and coherent noise wreak havoc on the extracted AVO and must be attenuated properly or appropriately factored in the uncertainty of the result. An incomplete list of bandwidth, and fold, unbalanced channels, coherent noise, complex geology, geometric spreading, focusing, surface roughness, dip, Fresnel zone effects, fault shadows, near surface effects of various kinds including source and receiver coupling and directivity, array effects, overburden attenuation, dispersion, anisotropy, and scattering, lateral velocity variations, improper deconvolution, FK filtering, improper mutes, non-amplitude preserving migration, inadequate petrophysical signal, inappropriate indicators, partial saturation, inversion non- uniqueness, incorrect NMO, etc. Overall, in most cases, AVO results must be considered inaccurate and imprecise. Under these circumstances, one must attach more significance to the presence of anomalies than their absence and rely on the skill of the interpreter to discriminate against false anomalies.
This ties into lack of robustness, but deserves special treatment, as even the best possible data can readily be misinterpreted. Reliance on rules of thumb or processed sections where "red" indicates hydrocarbons cannot replace a proper physical understanding of offset-dependent reflectivity. The most damaging myth associated with AVO analysis is the statement "Gas-sand amplitude increases with offset". This has lead to the false notion that an AVO anomaly must consist of an amplitude increase with offset and has led to inappropriate use of "product" indicators. In fact, gas-sand amplitude may increase, remain constant, or decrease with increasing offset depending on the circumstances. Even classical bright spots may exhibit decreasing AVO. The interpretational need to reduce CDP gathers to a manageable amount of data, has led to the frequent use of indicators. Unfortunately, this is followed by a tendency to ignore the CDP data, which is never desirable and which may be fatal if an incorrect indicator is applied. The need to rapidly model CDP gathers has led to frequent inappropriate use of primaries only ray-trace synthetics. If you are unsure of the computational scheme used by the modeling program at your disposal, I suggest you attempt modeling a thin coal bed. If amplitude increases rapidly with offset, you may need a new modeling package.
Analysis of Uncertainty
It is remarkable how little attention this item receives in practice. Ultimately, if AVO results are to be factored into risk assessment, it is necessary to provide answers with error bars. Granted, there are some key components to these uncertainties that we do not know how to properly quantify. However, there is no good reason not to quantify what we can.
AVO can be applied in a reconnaissance mode to generate prospects, or at a prospect level, to refine risk assessments and improve reservoir characterization. I contend, and many disagree with me, that the ability to illuminate prospects which cannot be seen otherwise, has generally greater economic advantage than the ability to better characterize existing prospects. Certainly, the common emphasis on the latter, when the method lacks sufficient credibility (0 significantly alter risk assessment, frequently results in the conclusion that AVO analysis provides no value added.
In conclusion, AVO analysis is a complex, inaccurate, and imprecise technology, which is frequently misapplied. Unfortunately, this has led many practicing explorationists to conclude that the method is of little practical importance, and can safely be ignored along with divining rods. To the contrary , AVO is based on very solid physical principles, but for reasons discussed here, it is not appropriate for use by those who are unwilling to invest the effort required to understand the technology. Thus, AVO analysis provides an excellent opportunity for those who can see through the complexity of the problem to beat the competition.
About the Author(s)
Dr. John P. Castagna began college in 1972 at age fifteen and received undergraduate and graduate degrees in geology and geochemistry from Brooklyn College of The City University of New York. He then studied exploration geophysics under Milo M. Backus at the University of Texas at Austin before joining ARCO's geophysical research group in 1980. There he conducted research in sonic logging and borehole geophysics. A few thousand miles later, he completed his doctorate in 1983, and became technical coordinator of ARCO's sonic logging research and technical services team. In 1986, he spent a year as a log analyst in a reservoir engineering services group, before moving on to lead ARCO's rock physics research team. He then moved into research management and served as Director of Geoseismic Interpretation Research and later manager of Seismic Analysis Research. In 1990 he began a developmental assignment as a seismic interpreter working offshore Gulf of Mexico exploration and development. Success in this endeavor led to an extension of the assignment to four years, followed by participation in a couple of bid rounds in China and Russia. These efforts were recently rewarded by a return to the serene world of research, where John currently has a dual appointment as Senior Principle Research Geophysicist for ARCO and Visiting Research Scientist at HARC's Geotechnology Research Institute.
John's background and experience in geology, geochemistry, exploration and development, rock physics, and log analysis in addition to exploration seismology has colored his research perspective. He is intrigued by the commonly unexploited potential for integration of these disciplines and strongly believes that research of the exploration process itself has received inadequate attention from geophysicists. He is a prolific writer, having authored about fifty external papers and numerous internal reports. His paper published in GEOPHYSICS in 1985 with co-authors Mike Batzle and Ray Eastwood defined the ARCO "mudrock" trend which is extensively used in AVO analysis. In 1993 he co-edited the SEG publication "Offset-Dependent Reflectivity - Theory and Practice of AVO Analysis" with Milo M. Backus. His efforts to popularize crossplotting of seismic parameters has helped extend the utility of AVO analysis to high impedance reservoirs. His paper on this subject at the 1993 Annual SEG Meeting received honorable mention for best presentation. He has also presented many invited papers at regional, national, and international meetings.
next board chairman. He was a member of the SPWLA publications committee from 1990-1992 and is an active member of SEG, AAPG, and more intermittently SPWLA and AGU. In 1993 he served on the State of Texas Higher Education Coordinating Board. He is a 1995 Petrobras instructor.