Reducing Uncertainty in Horizontal Well Placement for Improved Field Development

Abstract Determining optimal locations for unconventional wells in tight oil reservoirs is a complex process that has significant economic impact on reservoir management. In this case study, a rapid-deployment well placement workflow is implemented where geological, geophysical, geomechanical and petrophysical parameters were integrated using seismic and well data from a basin in China. The overall goal of the integrated workflow was making the best use of available data to reduce uncertainties and improve field development. More specifically the workflow was focused on best placement of a horizontal well trajectory to maximise reservoir exposure and optimizial orientation for fracturing to increase production. Integrated technology enables improved drilling efficiency, which reduces nonproductive time, better reservoir definition and management, decreases uncertainties, increases production and improves the direct and intangible interpretation of seismic data so as to interpret meaningful geological boundaries. The primary challenge was to predict an optimized seismic well trajectory amid limited a priori information. Available image logs in addition to geological, petrophysical and geophysical information are interpreted independently and then integrated into a conceptual geological model. A model-based seismic inversion algorithm was implemented to predict acoustic impedance (AI). To predict porosity from AI, a linear relationship between impedance and neutron porosity was established and applied to impedance volume. A geomechanical model was developed for the field, which provided the reservoir pressure, in-situ stresses and rock properties to predict the optimal well trajectory. An optimum mud weight program was designed to drill the well with minimum non-productive time. The image lithofacies method was compared with core photographs and open-hole logs. Image log analysis shows that the reservoir comprises a series of amalgamated meandering channels and crevasse splays separated by layers of fine sediments that act as local seals. Vertical lithofacies analysis indicates the reservoir channels are connected laterally and vertically. Post- stack seismic and well impedance was used to derive a best fit wavelet that results in quantitative measures of synthetic to seismic goodness-of-fit and wavelet accuracy. In conjunction with geomechanical models, integrated interpretation of predicted AI and porosity was used to optimize the drilling location. The seismic amplitude time slices generated at the top of the target show a wide zone of amplitude anomalies corresponding to the meandering channels interpreted from the image log analysis. The horizontal well was successfully drilled and encountered more than 600 m of reservoir sands. Seismic inversion-driven porosities were more appropriately used as a steering component to derive trends for larger-scale volumetric estimations and to identify reservoir “sweet-spots.” Thus, seismic inversion can be a valuable tool for reservoir characterization for field development. The geomechanical model enabled the optimization of multistage transverse fracturing of the horizontal well to increase production.