3-D Grain-Scale Simulation of Diagenesis and Rock Properties
In this presentation at the AAPG Annual Meeting Geocosm authors discuss concepts associated with Cyberstone, Geocosm's revolutionary and patented Digital Sedimentary Petrology model.
Fundamental geomechanical, petrophysical, and fluid-flow properties of clastic rocks reflect grain-scale compositions and textures. There is a revolution underway in the ability to predict these properties via “digital physics” simulations at the grain scale using high-resolution three-dimensional images of sample material as input constraints. Comparatively little work has been done, however, to develop “digital sedimentary petrology” models capable of predicting 3D grain-scale compositions and textures as a function of sediment depositional properties and diagenetic histories. Yet such models are intriguing because, when combined with digital physics simulations, they could provide a uniquely powerful basis for predicting rock properties away from well control, reconstructing the evolution in rock properties through geologic time, and forecasting rock response to engineering activities. Consequently, we are developing a 3D petrologic modeling system (“Cyberstone”) that incorporates the following components: (1) simulation of the physics of grain deposition where shapes, size distributions, compositions, and physical properties of grains reflect the characteristics of sediments in nature, (2) simulation of the processes responsible for compaction including grain rearrangement; elastic, plastic, and brittle deformation; and contact dissolution (“pressure solution”), and (3) simulation of the effects of geochemical reactions such as quartz overgrowth cementation. An important input for the simulator is the stress and thermal history of the simulated rock. In parallel with the development of the modeling system we are conducting a suite of laboratory experiments that yield fundamental insights into the controls on depositional and diagenetic processes while providing benchmarks for evaluating the model performance. We also rely on detailed analysis of geologic data for process understanding, constraining material properties and geochemical kinetics, and testing the efficacy of model assumptions.