False positive seismic "bright spots" and discriminination between gas and brine saturation using multiple avo analysis techniques in the Columbus Basin, offshore Trinidad

Exploratory drilling in the Tertiary Columbus Basin, offshore Trinidad has targeted high amplitude anomalies or “bright spots” in stacked sections with the expectation that these high amplitude reflectors will be gas-saturated sands. Many of these “bright’ reflection events are brine-saturated sands or tight, uncommercial gas-saturated silts. Thus, they are often characterized as “False Bright Spots”. This study investigates the use of Amplitude Variation with Offset (AVO) techniques to discriminate bona fide gas-saturated “bright spots” from false brine-saturated “bright spots”. Three AVO analysis techniques utilizing well log data and 3-D seismic data were applied to 4 different study areas within the Columbus Basin: Intercept (A) and Gradient (B), Elastic Impedance (EI) inversion and Lambda-Mu-Rho (LMR) inversion. A total of 20 prospective intervals were evaluated. The three AVO analysis techniques were all successful in determining the type of fluid saturation in the vast majority of the case studies. In most cases simple A-B crossplotting was sufficient to discriminate gas-saturation from brine-saturation and in two exceptions where it was not effective, the other methods, EI and LMR inversion, were successful in all but 1 of the 20 intervals examined. Application of all three techniques is recommended as a means of cross-validating results and increasing the reliability of results and interpretation. In addition, the more complex inversion methods demonstrated potential to be lithology indicators between sandy and shaly lithologies. The Rock Physics and AVO analyses presented in this study provide a reliable method to discriminate gas-saturated and brine-saturated sediments, including “False Bright Spots”, and contributes to the overall geophysical understanding and characterization of the Columbus Basin; an area where research and investigations of this nature are rare. Further, this is also a new case study of LMR inversion involving younger Tertiary sediments. As part of this study, a new “mudrock” line with coefficients relative to the Columbus Basin has been developed for estimation of seismic shear wave velocities from conventional sonic log data.