Locating exploration and appraisal wells using predictive rock physics, seismic inversion and advanced body tracking: an example from North Africa
G. Pickering, E. Knight, J. Bletcher, R. Barker, M. Kemper, 2004. "Locating exploration and appraisal wells using predictive rock physics, seismic inversion and advanced body tracking: an example from North Africa", 3D Seismic Technology: Application to the Exploration of Sedimentary Basins, Richard J. Davies, Joseph A. Cartwright, Simon A. Stewart, Mark Lappin, John R. Underhill
Download citation file:
A case study is described that illustrates a complete reservoir property prediction workflow, from petrophysical analysis, through rock physics, impedance inversion, and on to interpretation and final drilling locations. This study is from onshore Algeria, where the hydrocarbons are found in several clastic reservoirs of varying ages and properties. The controls on the presence of both reservoir and hydrocarbon location are not straightforward, therefore reducing the risk of drilling locations has significant value. This study concentrated on the Triassic TAG-I formation, which forms the major reservoir in the study area. The prediction method used was an impedance based deterministic approach, using relationships based on rock physics, although the interpretation method includes fuzzy set classifications to take into account the knon-uniqueness inherent in any seismic attribute. The petrophysical work ensured that the analysis of each well reconstructed ‘virgin-zone’ conditions. Rock physics models were then used to predict shear wave velocities. Acoustic impedance (AI), shear impedance (SI), and elastic impedance (EI) profiles were derived. Shear impedance was the best pure lithology indicator, with acoustic impedance showing a good relationship to porosity and elastic impedance most sensitive to fluid content. As the near angle seismic data were too noisy, gradient/intercept analysis was impossible, so shear reflectivity and consequently SI could not be derived. Although AI showed some lithological discrimination between sands and shales, it was not sufficient to be used as a single discriminator, so interpreted horizons were used to separate reservoir and non-reservoir intervals. The far-angle data were inverted to EI and this was analysed using a fuzzy logic approach. This method produces a classification volume and 3D body tracking was then used to find the best drilling targets. Generally, the analysis correctly predicted the results of the wells, both discoveries and non-discoveries. However, some of the discoveries were not predicted, which appears to be where the classification was not correctly calibrated. Work is now underway to improve the accuracy of the prediction process.