Available online at www.sciencedirect.

com

Energy Procedia 37 (2013) 3742 – 3746

GHGT-11

CO2 Leakage Potentials Prediction based on Geostatistical

Simulations
Weon Shik Hana*, Eungy Parkb, Kue-Young Kimc
a
University of Wisconsin-Milwaukee, Milwaukee, USA
b
Kyungpook National University, Daegu, Repulic of Korea
c
Korea Institute of Geoscience and Mineral Resources, Daejeon, Repulic of Korea

Abstract

We studied the uncertainties in the predictions of supercritical CO2 plume behavior and leakage potentials within
heterogeneous media. For this purpose, hypothetical heterogeneous field with different geostatistical features were
developed. The heterogeneous field represents a weakly non-stationary field, in which the geostatistical patterns of
the mid-depth geology are similar. Two conditional non-parametric geostatistical simulation models, the Sequential
Indicator Simulation, SISIM, and Generalized Coupled Markov Chain, GCMC, were applied for the stochastic
characterizations. Then, predictive CO2 plume migration simulations based on multiple stochastic realizations were
performed and summarized

©©2013
2013The
TheAuthors.
Authors.Published by Elsevier
Published Ltd.Ltd.
by Elsevier
Selection
Selectionand/or peer-review
and/or under
peer-review responsibility
under of GHGT
responsibility of GHGT
Keywords: CO2 sequestration; Leakage; Geostatiscs

1. Introduction

Prior to commercial-scale CO2 injection, various preceding assessments from laboratory- to pilot-scale
verifications and numerical simulations have been conducted by international scientific and engineering
communities in pursuit of (1) safe and secure storage [1,2,3], (2) short- and long-term monitoring [4,5,6],
and (3) risk assessment [7,8,9]. The majority of these studies reached similar conclusions, suggesting that
sound characterizations of the subsurface heterogeneity are the most important requirement for
constructing conceptual models and conducting associated analyses [10,11,12]. Nevertheless, the detailed
characterization of such complex subsurface heterogeneities at the site-wide scale is often constrained due
to issues related to the project resources. For this reason, investigators often rely on the limited numbers
of hard data collected within the boundaries of the sites (e.g. withdrawn borehole core or logging data). To
overcome such limitations at the site-wide scale, geostatistics and the associated simulation methods could
be applied. The uncertainties associated with subsurface characterizations play important roles in the
sound predictions, which suggest the necessity of the stochastic approaches. The present study emphasizes

1876-6102 © 2013 The Authors. Published by Elsevier Ltd.

Selection and/or peer-review under responsibility of GHGT
doi:10.1016/j.egypro.2013.06.269
 Weon Shik Han et al. / Energy Procedia 37 (2013) 3742 – 3746 3743

to understand the usefulness of the stochastic and non-parametric geostatistical characterization due to the
nature of abrupt transitions in targeted and capping formation. For this purpose, we chose two
representative non-parametric geostatistical simulation models, SISIM [13], and the recently developed
model of GCMC [14]. To compare the heterogeneity fields developed from these two geostatistical
models, we designed a hypothetical case describing the CO2 injection formation under the low-k caprock.

2. Non-Parametric Geostatistical Models

Fig. 1 show distributions of three rocks, in which the rock with medium-permeability (k) (RT1: 10-14
m ) occupies the upper half and the rock with high-k (RT2: 10-13 m2) occupies the lower half of the
2

domain. The low-k rocks (RT3: 10-16 m2) with elongated-lens shape are distributed around the interface
between the high- and medium-k rocks. The model domain is 4000 m×100 m size and is uniformly
discretized by 40 m×1 m grids. The hypothetical model represents a weakly non-stationary field where
the distributions of low-k lenses were concentrated around the mid-x and -y locations. In these
heterogeneity fields, artificial boreholes are assigned in which the x-positions are randomly selected. To
further mimic actual borehole drilling, the maximum depths of the boreholes are also chosen randomly.

Fig. 1. Hypothetical and deterministic k field.

With a hypothetical and deterministic k distribution and their stochastic realizations, two-dimensional
cross-sectional models were designed. The models represent the candidate formation for geologic CO2
sequestration with depths ranging from 1,000 m at the top to 1,100 m at the bottom. For initial conditions,
the model is fully saturated with brine and assigned a hydrostatic pressure and temperature ranging from
10.0 MPa to 11.0 MPa and from 35°C to 40°C, respectively. In all simulations, supercritical CO2 was
injected at a depth of 1,080 m during a 2-year period with a constant rate of 0.5 million metric tons per
year.

3. CO2 Injection Simulations on Deterministic and Stochastic Cases

Figs. 2a-d show the snap-shots of CO2 mass distributions at years 2, 50, 100, and 200 from the
deterministic CO2 injection simulation. Two different transports controlling the CO2 plume distribution
around and within the low-k lenses are observed. First, when an upward migrating CO2 plume
encountered a low-k lens, the CO2 plume accumulated and spread under the low-k lens, displacing the
brine. The CO2 plume kept expanding underneath the low-k lens until it reached the lateral margin of the
low-k lens where the CO2 plume migrated negative to the gravitational direction. The escaping CO2
plume at the margin of the low-k lens was impeded again when the plume encountered a subsequent low-
k lens. In contrast to the fast-flowing CO2 plume by detouring around the low-k lenses, certain amounts of
3744 Weon Shik Han et al. / Energy Procedia 37 (2013) 3742 – 3746

CO2 directly penetrated across the low-k lenses at a condition in which the cumulative pressure
underneath the low-k lenses was large enough to overcome the capillary entry pressure of the low-k
lenses.
a. b.

c. d.

Fig. 2. CO2 mass distribution [kg] in the deterministic domain: (a) 2 years, (b) 50 years, (c) 100 years, and (d) 200 years.

Figs. 3a and 3c show the k-distribution from randomly selected GCMC and SISIM realizations,
respectively. From two sequences of simulations, dissimilarity of the CO2 plume distributions is observed
at 100 years (Figs. 3b and 3d). The most peculiar difference is the migration pathways and the amount of
CO2 plume arrived at the upper boundary of the model domain. The CO2 plume reaches only a few
locations of the upper boundary in the GCMC. In contrast, the CO2 plume has arrived at large portions of
the upper boundary in the SISIM realization. Preliminarily judging from the single realizations at GCMC
and SISIM, the CO2 migration simulation based on a SISIM realization overestimated the leakage of the
CO2 plume to the upper formation (RT1).

a. b.

c. d.

Fig. 3. (a) Representative k field chosen from GCMG realizations, (b) CO2 mass distribution [kg] at 100 years, simulated in the
chosen GCMC k field, (c) Representative k field chosen from SISIM realizations, and (d) CO2 mass distribution [kg] at 100 years,
simulated in the chosen SISIM k field.

In Fig. 4, the mass fraction evolutions of supercritical CO2 and dissolved CO2 in brine from GCMC
and SISIM realizations are plotted. The solid-black lines represent the deterministic simulations and the
dashed-black lines show the arithmetic means calculated from GCMC and SISIM realizations. The pink-
and blue-shaded areas represent 99% and 80%-prediction intervals, respectively. In the GCMC
realizations, the deterministic CO2 mass fraction evolutions of the upper formation (RT1), lower
formation (RT2), low-k lenses (RT3), and dissolved CO2 in brine fall within the 80% prediction intervals.
In detail, the deterministic lines do not coincide with the simulated mean dash lines. Rather, the CO2 mass
fraction from multiple realizations showed a certain range of variation relative to the deterministic
prediction, which may be biased. The simulated mean of the CO2 mass fraction in RT1 was positively
biased compared to the deterministic line, while those in RT3 and the dissolved CO2 in brine were
slightly underestimated in their deterministic predictions. The overestimation of the CO2 mass fraction in
RT1 directly links to the underestimation of RT3 because the performance of low-k lenses is critical to
 Weon Shik Han et al. / Energy Procedia 37 (2013) 3742 – 3746 3745

determine the amount of CO2 mass leaving the lower half of the simulated domain. The dissolved CO2 in
brine is dependent on the detouring length of migration pathways, which determines the probability of the
CO2 plume contacting with the CO2-unsaturated brine. For this reason, the slight underestimation of the
dissolved CO2 in brine relative to the deterministic prediction implies that the predicted migration length
of the CO2 plume in the GCMC prediction is shorter than that of the deterministic prediction.
In contrast to the transient evolution of CO2 mass fraction through the GCMC realizations, the SISIM
realizations showed significant over- or underestimations subjected to the deterministic predictions.
Especially in the later time, most curves of the deterministic CO2 mass fraction deviate outside of the
99% prediction intervals, except the CO2 mass fraction in RT2. The asymptotic levels of CO2 mass
fractions in RT2 predicted from GCMC, SISIM, and the deterministic simulations were coincident after
100 years, suggesting that the amount of residually trapped CO2 in RT2 was not sensitive to the
configurations or distributions of the low-k lenses (RT3).

GCMC SISIM

Fig. 4. Mass fraction evolution of supercritical CO2 predicted from 50 GCMC and 50 SISIM realizations.

4. Conclusion

The presented study focused on understanding uncertainties of two non-parametric geostatistical

applications (GCMC and SISIM) in subsurface heterogeneity and their associated effects on CO2 plume
migration predictions. The hypothetical model representing a weakly non-stationary field was developed.
The SISIM underestimated the horizontal extent of the low-k lenses relative to the GCMC prediction
resulting in more CO2 leakage to the upper formation. For this reason, the GCMC-based simulations
represented the deterministic prediction better than those from the SISIM realizations. The simulation
suggested that the sizes of the low-k lenses are critical to determine the amount of CO2 leaking to RT1.
However, the amount of residually trapped CO2 remaining in RT2 was fairly consistent in all realizations
and was less sensitive to the variations of the low-k lens sizes or structures. Finally, CO2 mass-dissolved
in brine was governed by the length of the migration pathway. Throughout the comprehensive “Monte
Carlo”-typed analyses, the study suggests that geostatistical tools are useful in the prediction of CO2
plume behaviors under the heterogeneous subsurface conditions. However, proper geostatistical
techniques must be chosen carefully, depending on their geologic environments.

5. Acknowledgements
3746 Weon Shik Han et al. / Energy Procedia 37 (2013) 3742 – 3746

This work was supported by the Technology Development for CO2 Geologic Storage Demonstration
through Participating in the Canadian projects of the Korea Institute of Energy Technology Evaluation
and Planning (KETEP) grant funded by the Korea Government Ministry of Knowledge Economy.

References

[1] Audigane P, Gaus I, Czernichowski-Lauriol I, Pruess K, Xu T. Two-dimensional reactive transport modeling of CO2
injection in a saline aquifer at the Sleipner site, North Sea. Am J Sci 2007;307:974-1008.
[2] Han WS, McPherson BJ, Lichtner PC, Wang FP. Evaluation of trapping mechanisms in geologic CO2 sequestration: case
study of SACROC northern platform, a 35-year CO2 injection site. Am J Sci 2010;310:282-324.
[3] Juanes R, Spiteri EJ, Orr FM, Blunt MJ. Impact of relative permeability hysteresis on geologic CO2 storage. Water Resour
Res 2006;42, W12418, doi:10.1029/2005WR004806.
[4] Bateson L, Vellico M, Beaubien SE, Pearce JM, Annunziatellis A, Ciotoli G, Coren F, Lombardi S, Marsh S. The application
of remote-sensing techniques to monitor CO2-storage sites for surface leakage: method development and testing at Latera (Italy)
where naturally produced CO2 is leaking to the atmosphere. Int J Greenhouse Gas Contr 2008;2:388-400.
[5] Emberley S, Hutcheon I, Shevalier M, Durocher K, Gunter WD, Perkins EH. (2004) Geochemical monitoring of fluid-rock
interaction and CO2 storage at the Weyburn CO2-injection enhanced oil recovery site, Saskatchewan, Canada. Energy
2004;29:1393-1401.
[6] Lewicki JL, Hilley GE, Oldenburg CM. An improved strategy to detect CO2 leakage for verification of geologic carbon
sequestration. Geophys Res Lett 2005;32: L19403, doi:10.1029/2005GL024281.
[7] Kopp A, Binning PJ, Johannsen K, Helmig R, Class H. A contribution to risk analysis for leakage through abandoned wells
in geologic CO2 storage. Adv Water Resour 2010;33:867-879. doi:10.1016/j.advwatres.2010.05.001.
[8] Rohmer J, Bouc O. A response surface methodology to address uncertainties in cap rock failure assessment for CO2
geological storage in deep aquifers. Int J Greenhouse Gas Contr 2010;4:198-208.
[9] Siirila ER, Navarre-Sitchler AK, Maxwell RM, McCray JE. A quantitative methodology to assess the risks to human health
from CO2 leakage into groundwater. Adv Water Resour 2012;36:146-164.
[10] Ambrose WA, Lakshminarasimhan S, Holtz MH, Núñez-López S, Hovorka SD, Duncan I. Geologic factors controlling
CO2 storage capacity and permanence: case studies based on experience with heterogeneity in oil and gas reservoirs applied to CO2
storage. Environ Geol 2008;54:1619-1633.
[11] Lengler U, De L, M., Kühn M. The impact of heterogeneity on the distribution of CO2: Numerical simulation of CO2
storage at Ketzin. Int J Greenhouse Gas Contr 2010;4:1016-1025.
[12] Jahangiri HR, Zhang D. Effect of spatial heterogeneity on plume distribution and dilution during CO2 sequestration. Int J
Greenhouse Gas Contr 2011;5:281-293.
[13] Deutsch CV, Journel AG. GSLIB: Geostatistical Software Library and User's Guide. 2nd ed. Oxford University, New York,
1998.
[14] Park E. A multidimensional, generalized coupled Markov chain model for surface and subsurface characterization. Water
Resour Res 2010;45:W11509,doi:10.1029/2009WR008355.