Title: Global Sampling for Integrating Physics-Specific Subsystems and Quantifying Uncertainties of CO₂ Geological Sequestration

Abstract

The risk of CO₂ leakage from a deep storage reservoir into a shallow aquifer through a fault is assessed and studied using physics-specific computer models. The hypothetical CO₂ geological sequestration system is composed of three subsystems: a deep storage reservoir, a fault in caprock, and a shallow aquifer, which are modeled respectively by considering sub-domain-specific physics. Supercritical CO₂ is injected into the reservoir subsystem with uncertain permeabilities of reservoir, caprock, and aquifer, uncertain fault location, and injection rate (as a decision variable). The simulated pressure and CO₂/brine saturation are connected to the fault-leakage model as a boundary condition. CO₂ and brine fluxes from the fault-leakage model at the fault outlet are then imposed in the aquifer model as a source term. Moreover, uncertainties are propagated from the deep reservoir model, to the fault-leakage model, and eventually to the geochemical model in the shallow aquifer, thus contributing to risk profiles. To quantify the uncertainties and assess leakage-relevant risk, we propose a global sampling-based method to allocate sub-dimensions of uncertain parameters to sub-models. The risk profiles are defined and related to CO₂ plume development for pH value and total dissolved solids (TDS) below the EPA's Maximum Contaminant Levels (MCL) for drinking watermore » quality. A global sensitivity analysis is conducted to select the most sensitive parameters to the risk profiles. The resulting uncertainty of pH- and TDS-defined aquifer volume, which is impacted by CO₂ and brine leakage, mainly results from the uncertainty of fault permeability. Subsequently, high-resolution, reduced-order models of risk profiles are developed as functions of all the decision variables and uncertain parameters in all three subsystems.« less

Authors:

Sun, Y. ^[1]; Tong, C. ^[1]; Trainor-Guitten, W. J. ^[1]; Lu, C. ^[1]; Mansoor, K. ^[1]; Carroll, S. A. ^[1]

---

+ Show Author Affiliations

1. Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)

Publication Date:

Thu Dec 20 00:00:00 EST 2012

Research Org.:

Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)

Sponsoring Org.:

USDOE

OSTI Identifier:

1237556

Report Number(s):

LLNL-JRNL-569456
Journal ID: ISSN 1750-5836

Grant/Contract Number:  

AC52-07NA27344

Resource Type:

Accepted Manuscript

Journal Name:

International Journal of Greenhouse Gas Control

Additional Journal Information:

Journal Volume: 12; Journal ID: ISSN 1750-5836

Publisher:

Elsevier

Country of Publication:

United States

Language:

English

Subject:

58 GEOSCIENCES; 25 ENERGY STORAGE; uncertainty quantification; CO2 sequestrian; risk assessment; reactive transport; global sampling