Reservoir Characterization through Integrated Stochastic Inversion

Abstract We present an efficient inversion method, One-Dimensional Stochastic Inversion (ODiSI), which integrates petrophysical data, seismic data and geological information to provide an estimate of reservoir properties and associated uncertainties. We show results from the application of this inversion method to fields in the Nile Delta, offshore West Africa, and to a tight-gas field in the Middle East. ODiSI provides 3D volumes of reservoir properties and their associated uncertainties, and is becoming an integral part of our reservoir characterisation process. The inversion algorithm works by matching large numbers of stochastically simulated 1D stratigraphic profiles (pseudo-wells) to actual seismic traces. The method first establishes rock property relationships and vertical statistics at the reservoir level from existing well data. This information is used to generate thousands of pseudo-wells with a range of consistent and realistic reservoir properties. Synthetic seismic traces, derived from the pseudowells using extended elastic impedance (EEI), are compared with traces of input colour-inverted angle stacks. The best matches are selected and averaged to provide an estimate of reservoir properties and the associated uncertainties. In early applications, ODiSI was used primarily to predict net-to-gross sandstone for relatively simple siliciclastic fields. The output net-to-gross volumes were used to create net-to-gross maps, define key geological features, assist with well planning, inform volumetric estimates, and to provide input for reservoir modelling. We have also extended the application of ODiSI to a tight sand reservoir in the Middle East, where good quality well ties were observed between inverted and measured Vshale logs. The porosity estimates showed spatially consistent patterns indicative of geological features. The output net-to-gross sandstone and porosity volumes were used in uncertainty analysis to assess the spatial extent of good quality sand bodies. ODiSI products are becoming a standard input to our seismic analysis and interpretation workflow. We continue to expand the capability of the tool to enable its application to other geological settings.