Update on Results of SECARB Test of Monitoring Large Volume Injection at Cranfield

Presented to United States Energy AssociationApril 8, 2010Washington DC

Susan HovorkaGulf Coast Carbon CenterBureau of Economic GeologyJackson School of GeoscienceThe University of Texas at Austin

Funded through Southern States Energy Board by DOE National Energy Technology Lab

SECARB Cranfield Research: Theoretical Approaches Through Commercialization

Theo

ryan

d la

bFi

eld

expe

rimen

tsH

ypot

hesi

s te

sted

Commercial Deployment by Southern Co.

CO2 retained in-zone-document no leakage to air-no damage to water

CO2 saturation correctly predicted by flow

modeling

Pressure (flow plus deformation)

correctly predicted by model

Above-zone acoustic monitoring (CASSM) & pressure monitoring

Contingency planParsimonious public

assurance monitoringTow

ard

com

mer

cia-

lizat

ion

Surface monitoring: instrument verificationGroundwater programCO2 variation over time

Subsurface perturbation predicted

Sensitivity of tools; saturated-vadose

modeling of flux and tracers

CO2 saturation measured through time – acoustic impedance + resistivity

Tomography and change through time

Microseismic test, pressure mapping

3- D time lapse surface/ VSP seismic

Acoustic response to pressure change over

time

Dissolution and saturation measured via tracer breakthrough and chromatography

Lab-based core response to EM and acoustic under various saturations, tracer

behavior

Advanced simulation of reservoir pressure field

Cranfield “Early” Field Test Collaboration

Gulf Coast Carbon Center Industrial Associates

Denbury Onshore LLC

Schlumberger Carbon Services

LBNLLLBL USGSORNLNETL QEA

U MississippiMiss State

UTPGEUT DoG

University TennesseeBP

PrincetonStanford

University Edinburgh

Gulf Coast Carbon Center StaffSusan HovorkaRamon TrevinoTip MeckelChangbing YangJiemin LiuKatherine RomanakRebecca SmythSigrid CliftMasoumeh KordiStuart ColemanYihua CaiHamid Lashgari

BEG staffTongwei ZhangJeff PaineBob ReedyRobert Reed Kitty Millikan

Sandia Technologies

Completed Phase IIEOR/brine Storage tests

Black WarriorAlabama Geological Survey

VirginiaVirginia Tech

CranfieldPhase IIIearly

Plant DanielsSouthern ARI

Plant BarrySouthern ARI

Jackson Dome Natural CO2 source

B’

Natural CO2 Available Now in large Volumes Shipped via Sonant Pipeline to Test Lower Part of the Gulf

Coast Wedge

Onshore salt basins

Cranfield test site

Relatively young sandstones with shale sealsHeterogeneous, high porosity sediments

Salt tectonics and growth faultsHeavy industry

Characteristics of the Gulf Coast wedge

Galloway et al

Phas

e II

Site

de

velo

pmen

t

Cha

ract

eriz

atio

nCranfield Progress

2007

Phas

e II

Site

sel

ectio

n

2008 2009

Firstcored well,brine samples

baseline seismic S

oil g

as b

asel

ine

2006

Phase IIPhase III

2010 2011 2012

Pha

se II

I NE

PA

Phase III injection

Phase II injection

Projected 1.5 million tons phase III

May 1 million tones injectedP II + III

End

SE

CA

RB

Ear

ly

December 20Last well for 1 million tones/year rate

Drill Phase III3 DAS wells

Monitoring

Assuring Permanent Storage

• Know the geology of the reservoir – Characteristics that will accept and retain CO2

• Predict the area and magnitude of pressure increase at planned injection rate– Required for any injection in US by Safe

Drinking Water Act, 1974• Predict the distribution of CO2• Make measurements that document that

these predictions are correct.

DAS

A B

3D Denbury - interpretation Tip Meckel BEG

Characterization of the ReservoirTu

scal

oosa

Fm

Tuscaloosa D-E reservoir

Oil-water contactBased on log annotation and recent side-walls

Tuscaloosa confining system

Phase II

Baseline Cross Well tomogram

Z-Seis & Tom Daley Jonathan Franklin in review at LBNL

F1F2F3West East

112 m

Upward fining fluvial sandstone and conglomerates of the lower Tuscaloosa Fm

Jiemin Lu BEG

Confining zone Reservoir

Injector

Producer (monitoring point)Observation Well

HiVITP II Obs well

GMT

Phase II

Psite

Go to the field to test

Pipeline head&Separation facility

5km

Simplified CO2 injection rate

BEG Observation well

4000

4500

5000

5500

6000

2/22/2008 11/18/2010 8/14/2013 5/10/2016 2/4/2019

Date

Pres

sure

(psi

)

Measurement

Calculated

Model –history match pressure at real-time monitoring well

Injection rates

Modeled pressureMeasured pressure

Results of 1 year modelcontinuous pressure data

7/2008 12/2009

Rock and fluid properties in simulator

Obs well EGL7

JP Nicot Jong Won Choi BEG

Using pressure to show no leakage

surface

Remaining open annulus between rock and casing=Potential leakage path for CO2 or displaced brine?

Add CO2 for EOR

Surface casingCemented in

Non-cementedlong string

Cement to isolate

Production/injection zone

AZMITime

Pre

ssur

e

Injection zone

AMZI

Seal = No fluid communication

Injector

Producer (monitoring point)Observation Well

HiVIT

DASGMT

Phase II

Psite

Look in Detail at Flow

Detailed Study Area (DAS)

Pipeline head&Separation facility

5km

DAS MonitoringInjectorCFU 31F1

Obs CFU 31 F2

Obs CFU 31 F3

Above-zonemonitoringF1 F2 F3

Injection Zone

Above Zone Monitoring

10,500 feet BSL

Closely spaced well array to examine flow in complex reservoir

68m

112 m

Phase III Research Observation well construction for both wells

Fiberglass non-conductive casing

Tuscaloosa DE

ERT – 20 electrodes

Casing-conveyed pressure sensor

100’

U-tube sampler1/4 “SS

2 7/8” tubing

BHP+ T

Seismic sources/receivers

Distributed temperatureand heater loop

BEG LBNL LLNL USGS ORNL Sandia Technologies

Cross well array in two wellsHigh injection volumesFar-field monitoring microseismic, P&T, chemistry, surface seismic

200’

Start injection at DAS Dec 1, 2009175 kg/min step up to 350kg/min

Start injection at DAS Dec 1, 2009175 kg/min step up to 350kg/min

Bot

tom

hol

e pr

essu

re

Elapsed time

Dec 1

400

bar

340

psi

Injector BHP Observation well BHP

It’s all about pressure

Measuring distribution of CO2 in the reservoir

• Well-based methods– Wireline logs in time lapse -RST– Temperature

• Cross well methods– Time- lapse ERT– Time – lapse acoustic (seismic)

Wireline Formation Evaluation - ELAN – RST CFU 31 – F#3

GRWashouts

ResistivityOHPorosity Sigma

RSTPorosity

Perm

CO2 Volume

CO2Saturation

RST12/12/09

RST12/15/09

RST12/31/09

Bob Butch

What happened at the wells?

Obs

erva

tion

wel

l F2

packer

Inje

ctio

n W

ell F

1

Obs

erva

tion

wel

l F3

Dec

1 C

O2

flow

s in

to fo

rmat

ion packer

Dec 1pressure changes right away, but no CO2

Day 9

Obs

erva

tion

wel

l F2

packer

Inje

ctio

n W

ell F

1

Obs

erva

tion

wel

l F3

Dec

1 C

O2

flow

s in

to fo

rmat

ion packer

Dec 9CO2 detected in top of well interval

Day 13

Obs

erva

tion

wel

l F2

packer

Inje

ctio

n W

ell F

1

Obs

erva

tion

wel

l F3

Dec

1 C

O2

flow

s in

to fo

rmat

ion packer

Dec 13 still minor amounts ofCO2 detected in top of well interval and maybe some thin zones

Dec 13CO2 detected in top of well interval and maybe some thinzones

Day 31

wel

l F2

packer

Inje

ctio

n W

ell F

1

wel

l F3

Dec

1 C

O2

flow

s in

to fo

rmat

ion packer

Dec 30 large amounts ofCO2 detected in well interval and some thick zones in lower part of formation

Dec 31 large amounts ofCO2 detected in well interval upper part of formation

Injector

Direction of CO2 plume

x

x

x

x

x

x

Cross Well ERT tells us how flow occurred

Charles Carrigan, LLNL

50ft

Obs

erva

tion

wel

l F3

elec

trode

s

Obs

erva

tion

wel

l F2

elec

trode

s

Conductive plume= workover fluids?

Resistive plume = CO2 in reservoir

SecondResistive plume out of section migration

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

11/29/09 0:00 12/4/09 0:00 12/9/09 0:00 12/14/09 0:00 12/19/09 0:00 12/24/09 0:00 12/29/09 0:00 1/3/10 0:00

SAMPLING TIME

CO

NC

ENTR

ATI

ON

CH4

CO2

High frequency fluid sampling via U-tube

yields data on flow processes

BEG, LBNL, USGS, ORNL, UTDoG, data compiled by Changbing Yang BEG

Breakthrough of of CO2

Additional flow paths – more methane extracted

Double injection rate

Small diameter sampler with N2 drive brings fluids quickly tosurface with tracers intactCO2 dissolution into brine liberates dissolved CH4

Originallybrine methane saturated

Injector

Producer (monitoring point)Observation Well

HiVIT

P Site

GMT

Phase II

Is it possible to find leakage at

surface ? P-Site tests

Pipeline head&Separation facility

5km

DAS

1‐BG

1‐00

4‐02

1‐011‐02

1‐03

1‐041‐05

2‐01

2‐02

3‐02

3‐01

4‐01

Trans 1

Trans 2Trans 3

Trans 4

1950’s pit

road

Assessment of near surface techniques“P Site”

AW

PitPadPlantsP&A well

Preliminary Soil Gas data

CO2 (vol %)

O2 (vol %)

CH4 (vol %)

Pressure (inches H2O)

Katherine Romanak and Changbing Yang, BEG

Interim Conclusions of Study at Cranfield

• Phase III 1 million ton/year rate achieved Dec 20, 2009, 2 Million tones monitored since July 2008

• Rate to be maintained >15 months• Monitored with standard and novel approaches

– History match pressure response– No leakage into Above-Zone Monitoring Interval– Fluid flow measured/monitored with multiple tools in complex

flow field– First US use of Electrical Resistance Tomography (ERT) for

sequestration– Quantification of dissolution

• Export to commercial EOR/sequestration projects

Goals of monitoring at a long term, full scale commercial project

• Confirm that the predictions of storage security based on site characterization are valid

• Confidence to continue injection is gained from monitoring observations that are reasonably close to model predictions

• Confirm that no unacceptable consequences (risks or liabilities) result from injection.

• Monitoring during injection should be designed to prove-up sequestration so that monitoring frequency could be diminished through the life of the project and eventually stopped, allowing the project to be closed.

www.gulfcoastcarbon.org