The Value of Having Detailed Geologic Models in Unconventional Plays
As unconventional resource plays become more prevalent in the industry, an increasing amount of work is being done to characterize the geological details of these reservoirs properly. There is often a dis-connect, however, between the detailed characterization being performed at the core and log scale and the Geomodels being built for reservoir simulation, volumetric analysis and hydraulic fracture simulation. This disconnect is a result of the interpreter's desire to create a model that accounts for all of the available geologic detail and the simulator's desire for models that are small enough to run efficiently. The work presented here outlines a workflow-based approach for creating models that balance this tradeoff. Using data from an onshore US unconventional resource play, a workflow was generated to characterize the value of building detailed geologic models statistically in order to predict rock characteristics (e.g. porosity) away from well control. Different models of varying geologic detail were constructed. The results indicate that incorporating geologic detail into Geomodels does increase the ability of the model to predict porosity away from well control accurately (> 20% increase), but there is a point at which additional detail does not increase the predictive capability of the model. These results presume that the porosity depends on depositional environment or facies. 2D variogram analysis assists in quantifying the value of carefully modeling geologic detail in Geomodels. Within the study area, models with zonation schemes that lump differing geologic environments result in variogram maps containing too much information. However, properly splitting different geologic environments into their own model zones yields 2D variograms that contain adequate information and are easier to incorporate into models. The statistical and variogram analysis utilized in this study illustrates the impact of geologically reasonable models upon unconventional resource plays. These models take more time to build than models without built-in geologic concepts, but the resulting property distributions are more meaningful. Utilizing this workflow within a given reservoir of interest can indicate the correct number of zones for maintaining geologic heterogeneity, while helping to satisfy the demands of the simulator. Ultimately, accurate rock predictions will yield better STOOIP calculations, DFN models and well planning/geo-steering plans.
AAPG Datapages/Search and Discovery Article #90189 © 2014 AAPG Annual Convention and Exhibition, Houston, Texas, USA, April 6–9, 2014