Previous Talks

Wednesday, March 9th, 2011

Permeability evolution in fracture networks in response to mechanical and chemical processes

Joseph P. Morris, Schlumberger Doll Research


Whether by design or as an undesired side-effect, many subsurface activities trigger large changes in permeability due to combinations of mechanical and chemical effects involving fractures. For example, large-scale carbon capture and sequestration (CCS) projects will induce pressure and stress gradients that could activate existing fractures and faults, or drive new fractures through the caprock. In addition, over longer periods of time, geochemical processes will lead to evolution in the porosity and permeability of the storage domain, possibly enhancing or limiting containment of the CO2. Another example is that of gas shale where effective production requires creation of connected fracture networks that communicate well with the trapped gas. In this context, the challenges include creation of the fractures themselves, propping the fractures to maintain permeability and avoiding deleterious chemical effects that might limit effective production. Both applications involve processes which couple over multiple scales and require development of appropriate simulation capabilities. I will present results of investigations using two classes of computational tool: Distinct Element Methods (DEM) and Boundary Element Methods (BEM).

The DEM is naturally suited to simulating extensive fracture networks because it can explicitly accommodate the blocky nature of natural rock masses, however, such approaches presume constitutive models for predicting the response of the individual fractures. In contrast, it is also of interest to consider detailed coupling between various processes within the individual fractures. In practice, it is too computationally expensive to include the fine structures typical of a natural fracture (such as individual asperities) using finite elements or distinct elements. As an alternative I will present a Boundary Element Method (BEM)-based code where each fracture is approximated by potentially millions of asperities that interact according to boundary element interaction terms. As the confining stress on the fracture changes, the Fast Multipole Method is used to obtain a rapid prediction of the evolution in aperture within the fracture. I will present demonstrations of these different methods with relevance to CO2 sequestration and gas shale production. This will include simulations of induced microseismicity, fracture network creation and combined chemical-mechanical evolution which highlight the advantages of explicitly including the response of individual fractures within the formation.

Speaker Biography

Joseph P. Morris is a Principal Research Scientist at Schlumberger Doll-Research, Cambridge, USA where he provides expertise in the area of computational geomechanics and computational physics. Prior to joining Schlumberger in 2010, he worked as a researcher at Lawrence Livermore National Laboratory for over 10 years, contributing to both defense and energy related projects. Dr. Morris has led development of combined finite element-discrete element codes for predicting failure of jointed rock masses. In addition, he has developed boundary element based techniques for predicting deformation in fracture networks while including aspertity-level details. He has a B.A. with Honours in Science and Ph.D. in Mathematics both from Monash University, Australia.