Castagna, John P.2014-07-222014-07-22May 20142014-05http://hdl.handle.net/10657/692Stepwise linear regression of a database of 177 Class III hydrocarbon prospect outcomes and associated descriptions of Direct Hydrocarbon Indicator (DHI) observations indicate that the seismic characteristics can be used to predict well outcomes with a success rate better than 74% for out of sample tests. The most important seismic characteristics are presence of a phase change at the down dip edge of the anomaly, down dip conformance of the anomaly to structure (fit to closure), lack of unexplained anomalies in the same stratigraphic interval in the area, down-dip extent of the anomaly consistent with sealing capacity, and presence of prospect analogues. AVO analysis and results consistent with rock physics trends are also found to be significant factors in success/failure analysis. As seal capacity is an often neglected factor, its high ranking in the stepwise regression has significant practical implications. The mean-squared prediction error and residuals for all of the predictions are within acceptable limits. This shows that there is a relationship between the characteristics and quality of the interpreted DHI anomalies and the prospect outcome.application/pdfengThe author of this work is the copyright owner. UH Libraries and the Texas Digital Library have their permission to store and provide access to this work. Further transmission, reproduction, or presentation of this work is prohibited except with permission of the author(s).DhiProspectRisk analysisMultipleLinear regressionRegressionSaamSeismicAmplitudesAnalysisGeophysics2014-07-22Thesisborn digital