analysis of co2 sequestration in the - ceri-mines.org sequestration in the southwestern u.s. ......

Analysis of CO2 Sequestration in the Southwestern U.S.

Oil Shale Symposium

October 18, 2006

Brian J. McPhersonEnergy and Geosciences Institute

andDepartment of Civil and Environmental Engineering

University of Utah

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1000

2000

3000

4000

5000

6000

7000

Glo

bal E

mis

sion

of C

arbo

n as

CO

2

Global Carbon

Six warmest years of global record all after 1990.

Premise: Enhanced Greenhouse EffectG

loba

l Tem

pera

ture

Ano

mal

y (o

C)

(milli

on m

etric

tons

)

2000195019001850Years

Specific Research Interest: CO2 Sequestration

From the November 1st, 2002 issue of Science: Hoffertet al., “Advanced Technology Paths to Global Climate Stability: Energy for a Greenhouse Planet”

Goal: Mitigate Climate Change

National Portfolio

Partnerships

MRCSP

MGSC

SECARB

SWRP

WESTCARB

Big Sky

PCOR

Oil bearing

Gas bearing

Saline aquifer

Coal seam

Terrestrial

Field Test Type

Southwest Portfolio

Partnerships

MRCSP

MGSC

SECARB

SWRP

WESTCARB

Big Sky

PCOR

Oil bearing

Gas bearing

Saline aquifer

Coal seam

Terrestrial

Field Test Type

The Southwest Carbon Sequestration Partnership

In all partner states:• major universities• geologic survey • other state agencies• science lead: University of Utah

as well as• Western Governors Association• five major utilities• seven energy companies• three federal agencies• the Navajo Nation• many other critical partners

Southwest CO2

Sources

Figure complimentsof Rick Allis, UGS

- electrical power plants- cement & other plants- urban centers- non-point sources

Total regional point source emissions ~108 t/yr.

Figure complimentsof Genevieve Young, CGS

Southwest Major

CO2 Sinks

Southwest Regional Partnership on Carbon Sequestration

Phase I Task: Link Sources to Sinks

Phase I Primary Tasks:• Characterize region’s sources and sinks

• Identify best options by tying sources to sinks

• Outcome: In Southwest,“first opportunities” lie along existing CO2 pipelines

?

?Phase II Concept:“String of Pearls”

Pilot demonstrations will test short-term strategy:

sequester along pipelines

Characterization Demos Deployment Regional Pilot Full-Scale

Phase I Phase II Beyond Phase II


1,23

4

January, 2007150,000 tons/year

Project Portfolio

Four of over 100 geologic options were selected as the most promising opportunities for evaluation:

1) combined enhanced oil recovery with sequestration and

2) Deep brine reservoir sequestration testing, Paradox Basin, Utah January, 2007


Four of over 100 geologic options were selected as the most promising opportunities for evaluation:

(3) combined enhanced coalbedmethane production and sequestration testing, San Juan Basin, NM

1,23

4

September, 200775,000 tons/year

Project Portfolio


(4) combined enhanced oil recovery and sequestration testing, Permian Basin, TX

Renewed focus:Monitoring of CO2operations at SACROC

1,23

4

March, 2008300,000 tons/year

Project Portfolio


One site that was on the short-list but did not make it to the final portfolio of tests: Uinta Basin, Utah

We did study the site carefully because of its combined oil-shale and sequestration potential.

Project Portfolio

Research Focus Fundamental research questions include:(1)Is it possible for significant overpressure be

induced by CO2 injection?

Yes, injection rates of ~millions tons/year make it a possibility.

Bonanza - 4.3 Mton/y

Huntington - 6.2 Mton/y

Hunter - 10.6 Mton/y

Intermountain - 16 Mton/y

West Valley - 383,000 ton/y

Gadsby - 220,000 ton/y

Carbon - 1.3 Mton/y

Research Focus Characterization of the Uinta basin revealed a fundamental issue: the Duchesne Fault Zone may be directly related to observed subsurface overpressures.

Fundamental research questions include:(1) Will significant overpressure be induced by injection?Yes, injection rates of ~millions tons/year make it a possibility.

(2) Can the overpressure be controlled?Yes, near the wellbore. One uncertainty is that sustained injection may ultimately propagate overpressure far distances.

Research Focus Characterization of the Uinta basin revealed a fundamental issue: the Duchesne Fault Zone may be directly related to observed subsurface overpressures.

Fundamental research questions include:(1) Will significant overpressure be induced by injection?Yes, injection rates of ~millions tons/year make it a possibility.

(2) Can the overpressure be controlled?Yes, near the wellbore. One uncertainty is that sustained injection may ultimately propagate overpressure far distances.

(3) If overpressure is propagated far distances, will significant strain (e.g., fracture re-activation) be possible at those distances?Our study suggests this may be possible.

Study Area: Uinta Basin, UT

DuchesneRiverFormation

Uinta Formation Oligocene

LateEocene

Age

Limestone

Shale

Sandstone + ShaleDolomite

MahoganyOil Shale

Gre

en R

iver

For

mat

ion

FlagstaffPaleocene

Eocene

Dolomite / Mudstone / Shale

Sandstone

-4

-3

-2

-1

0

1

2

3N S

SunnysideAltamont

UDR

NH

MM

FS

TCM

UM

UDR = Uinta and Durchesne River FormationsUM = Upper Marker of Green River FormationMM = Middle Marker of Green River FormationTCM = Carbonate Marker of Green River Fm.FS = Flagstaff Member, Green River FormationNH = North Horn Formation (Cretaceous)

Ele

vatio

n (k

m)

Study Area: Uinta Basin, UT

Altamont RedWash

East-West Axis : 190 km

North-South Axis: 135 km

2000

4000

6000Head (m

)

Sunnyside

FaciesTransition

Observed Overpressures(1) Will significant

overpressure be induced by injection?

(2) Can the overpressure be controlled?

(3) If overpressure is propagated far distances, will significant strain (or fault activation) be possible at those distances?

Observed Overpressure:

• Steady-state code originally developed by Ge and Garven (1992)

• Code revised by McPherson and Garven (1999) to be temporal (non-steady-state)

• Additional revisions by McPherson and Garven (1999) include displacement boundary conditions (original code only had stress BCs)

Poroelasticrock strain

(plane strain)

Processes

Testing the Conceptual Model: Numerical Modeling

FLOW

Pressure/Temperature

DEFORMATION

Sequential Iterative (SIA)hydrodynamicadvectiondiffusiondispersion{

heat transferadvectionconductionsingle phase{

THRUST2D

“Hydromechanical” data required to parameterize model, for example:

• permeability, porosity• compressibility, density• thermal conductivity,

heat capacity• rock strength• poroelastic parameters:

– Young’s modulus, E – Poisson’s ratio, ν– Shear modulus, G = f

(E,ν)

Processes

hydrodynamicadvectiondiffusiondispersion{

heat transferadvectionconductionsingle phase{

THRUST2D

Testing the Conceptual Model: Numerical Modeling

T1ˆ ˆˆ 2 2 2

1 2 1 2xx xx kkG G P G Tν νσ ε ε α αν ν

+= + + +

− −

T1ˆ ˆˆ 2 2 2

1 2 1 2yy yy kkG G P G Tν νσ ε ε α αν ν

+= + + +

− −

T1ˆ ˆˆ 2 2 2

1 2 1 2zz zz kkG G P G Tν νσ ε ε α αν ν

+= + + +

− −ˆ ˆ ˆ2 , 2 , 2 .xy xy yz yz xz xzG G Gσ ε σ ε σ ε= = =

Poroelastic rock strain(plane strain)

Experimental Testing:Benchscale Rock Mechanical Data

• HYDROLOGICAL– Porosity– Permeability

• MECHANICAL– Tensile failure strength– Compressive failure strength

Elastic parameters

• PHYSICAL– Rock Framework– Mineralogy– Cementation

Model Parameters Evaluated

DUCHESNEMYTON

40° N

110°

W

N

#1-1

5

#21-26

#32-34#31

#27-30

#40-41

#38-39

#35-37

#42, 47#48

~5 km

Tensile Strength• Method: • Brazil Tests

• “Typical” Values: – 4-14 MPa

• Average: 10.6 MPa (UB only)

• NORTH: 11.9 MPa• in DFZ: 10.3 MPa• SOUTH: 8.8 MPa

– TRENDS?

Average Tensile Strength vs. Location

0

4

8

12

16

Samples

Tensi

le F

ailu

re S

trength

(M

Pa)

NORTH

in FZ

SOUTH

Compressive Strength• Method: Uniaxial compression

• “Typical” Values: – 60-270 MPa

• Average: 132 MPa– Amoco: 124 MPa– UB: 149 MPa

• NORTH: 136 MPa• in DFZ: 160 MPa• SOUTH: 120 MPa

Compressive Srength vs. Location

0

50

100

150

200

250

Samples

Com

pre

ssiv

e Fa

ilure

Str

ength

(M

Pa)

NORTHAmoco (NORTH)in FZSOUTH

Compressive Strength vs. Depth

0

50

100

150

200

250

2000 4000 6000 8000 10000 12000

Depth (ft bgs)

Com

pre

ssiv

e Fa

ilure

S

tren

gth

(M

Pa)

horizontalvertical

Elastic Parameters All

Samples

UB

Samples

North of

DFZ

In DFZ South of

DFZ

Amoco

Samples

N 43 14 3 9 2 27

Minimum 1.7 x104 2.8 x104 3.2 x104 2.8 x104 3.6 x104 1.7 x104

Maximum 6.9 x104 6.1 x104 5.4 x104 6.1 x104 3.6 x104 6.9 x104

Mean 4.5 x104 4.7 x104 4.5 x104 4.9 x104 4.1 x104 4.5 x104

Median 4.7 x104 4.9 x104 5.0 x104 5.1 x104 4.1 x104 4.5 x104

You

ng’s M

odulu

s

(GPa

)

Standard

Deviation 1.4 x104 1.0 x104 1.2 x104 1.1 x104 8.1 x103 1.5 x104

N 17 14 3 9 2 3

Minimum 0.10 0.14 0.14 0.14 0.21 0.10

Maximum 0.26 0.26 0.24 0.26 0.26 0.17

Mean 0.19 0.20 0.18 0.20 0.23 0.13

Median 0.19 0.20 0.15 0.19 0.23 0.12

Poisso

n’s R

atio

(n

o

units)

Standard

Deviation 0.05 0.04 0.05 0.04 0.04 0.04

N 17 14 3 9 2 3

Minimum 4.0 x104 4.0 x104 4.0 x104 4.2 x104 7.3 x104 5.1 x104

Maximum 1.3 x105 1.3 x105 9.5 x104 1.3 x105 7.9 x104 8.9 x104

Mean 7.7 x104 7.9 x104 6.8 x104 8.4 x104 7.6 x104 6.5 x104

Median 7.3 x104 7.3 x104 6.8 x104 7.3 x104 7.6 x104 5.5 x104

Bulk

Modu

lus

(GPa

)

Standard

Deviation 2.3 x104 2.4 x104 2.7 x104 2.5 x104 4.6 x103 2.1 x104

Mean ValuesYoung’sModulus45,000 GPa

Poisson’sRatio0.19

BulkModulus77,000 GPa

• 40 CO2 injection wells penetrating the Flagstaff Formation• no production wells• dissolved phase shown here (follows separate phase, for most part)• overpressures develop in lower permeability marginal lacustrine facies• overpressures terminate at higher permeability open lacustrine facies• CO2 migrates more freely in higher permeability open lacustrine facies

3-D Modeling Analysis: CO2 Migration and Pressures

indicates wells

Migration Distance and Rate are Controlled by: Absolute & Relative Permeability, Capillarity

-16 -15 -14 -13 -120

10

20

30

40

50

60

70

80

90

100

Log Absolute Permeability (m2)

Mig

ratio

n D

ista

nce

(km

) in

1000

yrs

123 m/yr

21 m/yr

2 m/yr25 cm/yr

Results specific to thisUinta BasinSimulation (e.g., itsstructural geometry, topographically driven flow, etc.)

Coupled Stress-Strain Modeling Results:Neotectonic Forces but No Overpressure

εyy x 10-9

-6 -4 -2 0 2 4 6y

x

2

1

0

-1

-2

-3

Assigned Stress Boundary Condition = Stress Rate Required to Induce Observed Fault Offset

Elev

atio

n (k

m)

135 km

DFZ

Cross-Section Location

Coupled Stress-Strain Modeling Results:Neotectonic Forces but No Overpressure

εyy x 10-9

-6 -4 -2 0 2 4 6y

x

2

1

0

-1

-2

-3


Elev

atio

n (k

m)

135 km

DFZ

εyy x 10-9

-6 -4 -2 0 2 4 6y

x

Elev

atio

n (k

m)

135 km

2

1

0

-1

-2

-3

DFZ

Observedoverpressure(assignedcondition)


Coupled Stress-Strain Modeling Results:Neotectonic Forces with Overpressure

Summary and Major Conclusions

(1) Will significant overpressure be induced by CO2 injection?Yes, injection rates of ~millions tons/year make it a possibility. It should be controllable through reservoir engineering.

(2) If overpressure is propagated far distances, will significant strain (or fault activation) be possible at those distances?

• Observations of the Uinta system suggest that natural overpressures propagated far distances, implying that injection-induced overpressures may propagate as well

• Calibrated numerical models of the Uinta system suggest that high fluid pressures can contribute to rock strain

• Results of this study are guiding reservoir engineering of CO2 injection pilot tests that will begin in early 2007

0 10 20 30 40

Porosity (%)

10-19

10-18

10-17

10-16

10-15

10-14

10-13

10-12

Per

mea

bilit

y (m

2 )

Samples with insignificant claySamples with clay content > 5%Data of Pitman et al., 1982

Trend for single fracture

Initial Trend

analysis of co2 sequestration in the - ceri-mines.org sequestration in the southwestern u.s. ......

Documents