Numerical Modelling of Geological Heterogeneity - Implications for CO2 Geological Storage
Abstract
CO2 geological storage is a proposed mitigation strategy currently being considered to reduce atmospheric greenhouse gas emissions. One factor often limiting the implementation of CO2 geological storage is the uncertainty associated with geological heterogeneities within storage reservoirs and how these heterogeneities will impact CO2 partitioning into the various storage mechanisms. Numerical models are a useful tool for integrating field and laboratory data to generate predictions on the extent of CO2 storage at larger spatial and temporal scales than experimental work is capable of undertaking alone. Numerical models use governing equations to simulate physical and chemical processes, such as the flow and transport of CO2 within the subsurface. Governing equations require the specification of a number of input parameters inherent to the porous medium. In nature, parameters such as porosity and permeability vary within and between different rock types, according to variations in factors such as grain size, sorting, cementation and structure. These variations lead to geological heterogeneity at a number of scales. However, geological heterogeneity is often oversimplified in numerical models, either due to a lack of geological data or to increase computational efficiency. Grid spacing is often coarse, leading to faster simulation times but a decrease in numerical accuracy. Further work is required to investigate how simplifying geological heterogeneity within numerical models affects short and long term CO2 storage predictions. To quantify the impact of geological heterogeneity, TOUGH2, a multiphase flow and transport code, is used to construct a series of simulations with increasing degrees of geological complexity. Comparisons are made between numerous scenarios, including discrete versus gradual progression into areas of heterogeneous rock types, continuous versus discontinuous layering, internal structures and anisotropy. Input parameters associated with different rock types are varied within reasonable ranges and grid spacing is refined to determine the sensitivity of the models to grid size. Variations between simulations are used to determine the differences in the partitioning of CO2 between its various storage mechanisms, and whether the differences are reflective of heterogeneities in the real system or attributed to numerical error. Initial results indicate that variations in certain parameters are more significant than others in terms of the movement and partitioning of CO2 into its various storage mechanisms. Variations in horizontal to vertical permeability contrasts, and residual liquid and gas saturation have significant impacts on the flow path of CO2 through the system, and therefore the amount of CO2 that becomes trapped residually within the pore spaces or dissolves into the formation brine. Gradual changes in heterogeneity do not seem to alter the results significantly in comparison to discrete changes, indicating that modelling heterogeneities as discrete bodies is an adequate assumption. Results imply that certain geological heterogeneities and associated parameters require more accurate representation then others when considering how CO2 will be stored within the subsurface. Although finer grid sizes increases the numerical accuracy of simulations, acute grid refinement may not be required for all purposes.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2012
- Bibcode:
- 2012AGUFM.H23A1327H
- Keywords:
-
- 1829 HYDROLOGY / Groundwater hydrology;
- 1832 HYDROLOGY / Groundwater transport;
- 1847 HYDROLOGY / Modeling;
- 1859 HYDROLOGY / Rocks: physical properties