Uncertainty Quantification of CO2 Leakage through a Fault with Multiphase and Geo-mechanic Effects
Abstract
The potential for CO2 leakage through a permeable fault is a key concern for geologic CO2 sequestration (GCS) in saline formations, which typically involves CO2 being injected under supercritical conditions. If CO2 migrates vertically upward through a fault from the storage reservoir to an overlying fresh-water aquifer, phase change can occur because temperature and pressure decrease with decreasing depth. The decrease in CO2 density during phase transition causes an additional reduction in temperature (Joule-Thomson cooling effect), which reduces CO2 viscosity, resulting in an increase in the CO2 leakage rate. In this paper, we present a computational model for simulating the behavior of a leaky fault connecting a saline CO2 storage reservoir and an overlying fresh-water aquifer. We address phase transition, considering the nonlinear CO2 enthalpy and viscosity functions, the impact of geo-mechanics on rock hydraulic properties. The model is initialized to represent the hydrostatic and geothermal temperature gradients in the model domain, which extends from 230 to 2000 m below the ground surface. Time-varying boundary conditions of pressure and CO2 saturation are specified at the lower boundary to a hypothetical fault to represent conditions in an operating CO2 storage reservoir during both the injection and post-injection periods. The model results indicate that the CO2 leakage rate initially increases when CO2 migration is driven by both buoyancy and overpressure during the period of injection-driven pressure buildup in the reservoir. For the post-injection period when overpressure in the reservoir dissipates, CO2 leakage is only driven by buoyancy and the leakage rate decreases. The deterministic model of this faulted reservoir system is used within an uncertainty quantification (UQ) sampling framework in order to rigorously quantify the sensitivity of the brine and CO2 leakage response to the uncertain model parameters, including the input constitutive properties, geo-mechanics properties and boundary conditions. The results show that fault permeability is the most sensitive factor affecting CO2 leakage rate. The sensitivity analysis is used to screen input parameters for model reduction, using the PSUADE code. The reduced-order models of CO2 and brine leakage are developed for the emulation of a large number of sample points, from which probability distributions can be derived. The reduced-order models can be incorporated in risk assessment of groundwater contamination resulting from CO2 and brine leakage.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2012
- Bibcode:
- 2012AGUFM.H13L..06L
- Keywords:
-
- 1832 HYDROLOGY / Groundwater transport;
- 1875 HYDROLOGY / Vadose zone