Leakage through Existing Wells: Models,
Data Analysis, and Lab Experiments
Michael A. Celia and George Scherer
Department of Civil and Environmental Engineering
Princeton University
Peter Jaffe
Satish Myneni
Catherine Peters
Jean Prevost
Princeton
Princeton University
University
Rob Bruant
Andrew Altevogt
Mileva Radonjic
Andre Duguid
Sarah Gasda
Li Li
Outline
‹ Overview
of activities in 2003
– Continue focus on leakage and wells.
– Refocus effort toward detailed studies of well cements.
‹ Brief
presentation of selected results
– Analysis of wells and tools for leakage estimation.
– Unsaturated zone and responses to elevated CO2.
‹ Cement
studies (presented by G. Scherer)
– New laboratory for cement experiments.
– Design of experiments and preliminary results.
– Focused activities for CY 2004 including local-scale
modeling.
Princeton
Princeton University
University
Mature Sedimentary Basins: Analysis of Wells
Viking Formation: 195,000 Wells
Number of clusters
Number of wells [% total]
Area [% total]
Mean intrinsic density
[wells/km2]
Princeton
Princeton University
University
Alberta Basin: 350,000 Wells
Very High-Density
(Heavy-oil)
HighDensity
MediumDensity
Low-Density
(Background)
32
5.2
0.1
17.1
268
28.0
2.7
3.75
963
28.6
10
1.13
-38.2
87.2
0.15
Number of Wells Impacted by ‘typical’ Injection
High-density clusters
Medium-density clusters
Low-density Background
Princeton
Princeton University
University
Mean
Median
Range
241.5
62.6
17.8
216
61
11
45-721
8-144
0-130
(Numbers from Gasda et al., 2003)
CO2 Injection and Leakage Pathways
Shallow-zone Effects
• Unsaturated Soils
• Groundwater Resources
• Numerical Simulations
• Analytical Solutions
Well-leakage dynamics
Flow Dynamics
• Spatial Locations
• Well Properties
• Cement Degradation
Princeton
Princeton University
University
Overview of Selected Results from 2003
‹
Simplified Solutions for CO2 Transport and Leakage Estimation (3
Papers):
– Analytical solutions to accommodate many potentially leaky wells and
multiple geological layers.
– Framework for linear equations (eg, hazardous waste injection), showing
importance of caprock thickness and storage in layers above injection.
– Energy minimization arguments yield simple equations for evolution of
CO2 plume and pressure field; solution applies to all practical ranges of
densities and viscosities (P,T,S) for CO2 injection in aquifers.
– Bounding calculations for leakage of brine and CO2 through leaky well.
– Comparisons to Eclipse show excellent match of results.
– Ongoing extensions: Intersecting CO2 plumes, variable density along the
well, redistribution of CO2 after cessation of injection.
‹
Unsaturated Zone Studies (2 Papers)
– Modeling studies for CO2 transport: dominant transport mechanisms.
– Modeling studies of geochemical response to elevated CO2 concentrations:
evolving redox profiles.
– Laboratory column study for geochemical tests.
Princeton
Princeton University
University
New Analytical Solutions
Princeton
Princeton University
University
New Analytical Solution: Two-Phase Flow
Normalized Depth: z/B
Analytical solution (sharp interface) versus Effective Saturation from Eclipse
Full Numerical Solution (from Eclipse)
Analytical Solution
⎤
b( r , t )
1 ⎡ λc λwV
=
− λw ⎥
⎢
2
B
λc − λw ⎣ πBr
⎦
Solution includes unfavorable mobility
r2/t
ratio and gravity
override
Princeton
Princeton University
University
Leakage Solutions
Princeton
Princeton University
University
Effect of CO2 Leakage on the Redox Profile in the Vadose Zone
10
8
Z(m) 6
4
2
0
10
8
Z(m) 6
4
2
0
10
8
Z(m) 6
4
2
0
10
8
Z(m) 6
4
2
0
0.02
O2
0.01 (mole/L)
0
10
10
10
20
30
20
30
20
30
40
20
15
10
5
0
NO3
(mole/L)
x10-2
1.0
0.8
0.6
0.4
0.2
0
Mn(IV)
(mole/L)
x10-4
40
40
6
4
2
0
10
Princeton
Princeton University
University
20
Y(m)
30
40
Fe(III)
(mole/L)
x10-4
Ongoing Efforts for 2004
‹
Leakage Modeling for specific field site:
– Apply numerical and analytical models to estimate leakage,
coupled with well statistics for spatial locations.
– Determine threshold statistics for well parameters
associated with target leakage amounts.
– Incorporate transient well parameters based on cement
degradation estimates.
‹
Unsaturated Zone Studies:
– Continue with laboratory column study.
– Test models against laboratory data.
‹
Well-cement experiments and analysis will be the major
effort for 2004.
Princeton
Princeton University
University
General Overview of 2003 – Topics Covered
Analysis of Well Statistics in Viking Formation, Alberta Basin:
• Gasda, S.E., S. Bachu, and M.A. Celia, "The Potential for CO2 Leakage from
Storage Sites in Geological Media: Analysis of Well Distribution in Mature
Sedimentary Basins", submitted to Environmental Geology, December 2003.
Simplified Solutions for CO2 Transport and Leakage Estimates:
• Nordbotten, J.M., M.A. Celia, and S. Bachu, "Analytical Solutions for Leakage
Rates through Abandoned Wells", revised version submitted to Water
Resources Research, December 2003.
• Nordbotten, J., M.A. Celia, and S. Bachu, "Injection and Storage of CO2 in
Deep Saline Aquifers: Analytical Solution for CO2 Plume Evolution during
Injection", submitted to Transport in Porous Media, October 2003.
• Nordbotten, J., M.A. Celia, S. Bachu, and H.K. Dahle, "Analytical Solution for
CO2 Leakage between Two Aquifers through an Abandoned Well", submitted
to Environmental Science and Technology, December 2003.
Large-scale Compositional Simulations:
• Prevost, J.H., R. Fuller, A.S. Altevogt, R. Bruant, and G. Scherer, “Numerical
Modeling of Carbon Dioxide Transport in Deep Saline Aquifers: Equation
Development”, under review, SPEJ, 2003.
Princeton
Princeton University
University
General Overview of 2003 – Topics Covered
Shallow-zone Transport and Geochemisttry:
• Altevogt, A. and M.A. Celia, "Modeling Carbon Dioxide Transport in
Unsaturated Soils", revised version submitted to Water Resources Research,
November 2003.
• Altevogt, A.S. and P.R. Jaffe, "Modeling the Effects of Gas-phase CO2
Intrusion on the Biogeochemistry of Variably Saturated Soils", to appear, Proc.
CMWR 2004 Conference, June 2004.
• Wang, S. and P.R. Jaffe, “Dissolution of Trace Metals in Potable Aquifers due
to CO2 Release from Deep Formations”, to appear, Energy Conversion and
Management, 2004.
• Giammar, D.E., R.G. Bruant, and C.A. Peters, "Forsterite Dissolution and
Magnesite Precipitation at Conditions relevant for Deep Saline Aquifer Storage
and Sequestration of Carbon Dioxide, Submitted to Chemical Geology, 2003.
Upscaling Studies:
• Gasda, S.E. and M.A. Celia, "Upscaling Relative Permeabilities in a Structured
Porous Medium", to appear, Proc. CMWR 2004 Conference, June 2004.
• Li, L., C.A. Peters, and M.A. Celia, "Examination of Upscaling Considerations
for Geochemical Reaction Rates using Network Model Simulations of Calcite
Dissolution", to appear, Proc. 11th Int. Sym. Rock-Water Interactions, June
2004.
Princeton
Princeton University
University