Constraining Geostatistical Reservoir Descriptions with 3-D Seismic Data to Reduce Uncertainty

The geostatistical external drift method is used to integrate three-dimensional (3-D) seismic data into a reservoir description and is illustrated with an application on a west Texas Permian basin interbedded carbonate-clastic reservoir. Seismic reflection amplitude, inverted to acoustic impedance, supplements sparse well control to estimate interwell porosity. The North Cowden unit is a mature field and serves as a laboratory for many reservoir characterization experiments. The extensive wireline and core database available for the area covered by a high-resolution 3-D seismic survey was resampled to mimic scenarios similar to three stages in a reservoir's life: (1) a late exploration/appraisal phase; (2) a development phase; and (3) a mature production phase, typical of many Permian basin fields. Spatial interpolation by kriging porosity with and without seismic data are compared. Stochastic (Monte Carlo) simulations are used to evaluate interpolation uncertainty (standard error). Interpolation uncertainty is greatly reduced when seismic data are integrated into the reservoir description.