Characterizing an unconventional reservoir with conventional seismic data: a case study using seismic inversion for the Vaca Muerta Formation, Neuquen Basin, Argentina

Abstract

Reservoir characterization for unconventional shale plays ideally requires multi-component, wide-azimuth, long-offset surface seismic data. These data are generally not available, espe- cially in exploration or pre-development stages. Furthermore, it is common to have only a few wells over a large area, along with non-existent or scarce microseismic, engineering and production data. This thesis presents a methodology and workflow to deal with these cir- cumstances of limited data availability. By using a narrow-azimuth, regional P-wave seismic volume and integrating it with wireline logs, cuttings and PLT data, the variability in the geomechanical properties of the Vaca Muerta Formation in Argentina’s Neuquen Basin, and their relationships with lithology, stress state and total organic content, were analyzed. Post-stack and pre-stack inversions were performed on the seismic volume. The un- certainties inherent from limited well control in the estimation of elastic properties were investigated using blind well testing. Sensitivity and error analysis was conducted on post- stack vs pre-stack derived P-impedance, the choice of the inversion algorithm (model-based vs sparse-spike) and the definition of the low frequency model (simple kriging model vs com- plex model derived from multi-attribute stepwise regression) were examined. Also, the use of isotropic AVA equations to approximate the anisotropic (VTI) behaviour of the reservoir was evaluated, using estimates of Thomsen parameters and simple AVA modelling. The integration of the inversion results with the petrophysical analysis and the mechanical stratigraphy work by Bishop (2015), suggests that the rock composition has the largest influence on the geomechanical behaviour of the reservoir. Overpressure is also a major driving factor in that it controls changes in elastic properties. Bishop’s cluster analysis was used to identify good quality rock classes. The probabilistic interpretation of these rock classes from seismic inversion provides a quantitative measure of uncertainty and guides the selection of potential drilling targets.