Active CO2 reservoir management for carbon storage: Analysis of operational strategies to relieve pressure buildup and improve injectivity

Abstract

For industrial-scale CO2 injection in saline formations, pressure buildup can limit storage capacity and security. Active CO2 Reservoir Management (ACRM) combines brine production with CO2 injection to relieve pressure buildup, increase injectivity, manipulate CO2 migration, and constrain brine leakage. By limiting pressure buildup, in magnitude, spatial extent, and duration, ACRM can reduce CO2 and brine leakage, minimize interactions with neighboring subsurface activities, allowing independent assessment and permitting, reduce the Area of Review and required duration of post-injection monitoring, and reduce cost and risk. ACRM provides benefits to reservoir management at the cost of extracting brine. The added cost must be offset by the added benefits to the storage operation and/or by creating new, valuable uses that can reduce the total added cost. Actual net cost is expected to be site specific, requiring detailed analysis that is beyond the scope of this paper, which focuses on the benefits to reservoir management. We investigate operational strategies for achieving an effective tradeoff between pressure relief/improved injectivity and delayed CO2 breakthrough at brine producers. For vertical wells, an injection-only strategy is compared to a pressure-management strategy with brine production from a double-ring 9-spot pattern. Brine production allows injection to be steadily ramped up while staying within the pressure-buildup target, while injection-only requires a gradual ramp-down. Injector/producer horizontal-well pairs were analyzed for a range of well spacings, storage-formation thickness and area, level and dipping formations, and for homogeneous and heterogeneous permeability. When the producer is downdip of the injector, the combined influence of buoyancy and heterogeneity can delay CO2 breakthrough. Both vertical and horizontal wells can achieve pressure relief and improved CO2 injectivity, while delaying CO2 breakthrough. Pressure buildup and CO2 breakthrough are sensitive to storage-formation permeability and insensitive to all other hydrologic parameters except caprock-seal permeability, which only affects pressure buildup for injection-only cases.

Full text

(This is a sample cover image for this issue. The actual cover is not yet available at this time.)
This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright

Author's personal copy
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
Contents
lists
available
at
SciVerse
ScienceDirect
International
Journal
of
Greenhouse
Gas
Control
j
our
na
l
ho
me
p
age:
www.elsevier.com/locate/ijggc
Active
CO2reservoir
management
for
carbon
storage:
Analysis
of
operational
strategies
to
relieve
pressure
buildup
and
improve
injectivity
Thomas
A.
Buschecka,∗,
Yunwei
Suna,
Mingjie
Chena,
Yue
Haoa,
Thomas
J.
Wolerya,
William
L.
Bourciera,
Benjamin
Courtb,
Michael
A.
Celiab,
S.
Julio
Friedmanna,
Roger
D.
Ainesa
aLawrence
Livermore
National
Laboratory,
P.O.
Box
808,
Livermore,
CA
94550,
USA
bDepartment
of
Civil
and
Environmental
Engineering,
Princeton
University,
Princeton,
NJ
08544,
USA
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
29
May
2011
Received
in
revised
form
9
November
2011
Accepted
10
November
2011
Keywords:
CO2capture
and
storage
CO2capture
Utilization
and
storage
Brine
production
Pressure
management
Injectivity
CO2migration
a
b
s
t
r
a
c
t
For
industrial-scale
CO2injection
in
saline
formations,
pressure
buildup
can
limit
storage
capacity
and
security.
Active
CO2Reservoir
Management
(ACRM)
combines
brine
production
with
CO2injection
to
relieve
pressure
buildup,
increase
injectivity,
manipulate
CO2migration,
and
constrain
brine
leakage.
By
limiting
pressure
buildup,
in
magnitude,
spatial
extent,
and
duration,
ACRM
can
reduce
CO2and
brine
leakage,
minimize
interactions
with
neighboring
subsurface
activities,
allowing
independent
assessment
and
permitting,
reduce
the
Area
of
Review
and
required
duration
of
post-injection
monitoring,
and
reduce
cost
and
risk.
ACRM
provides
benefits
to
reservoir
management
at
the
cost
of
extracting
brine.
The
added
cost
must
be
offset
by
the
added
benefits
to
the
storage
operation
and/or
by
creating
new,
valuable
uses
that
can
reduce
the
total
added
cost.
Actual
net
cost
is
expected
to
be
site
specific,
requiring
detailed
anal-
ysis
that
is
beyond
the
scope
of
this
paper,
which
focuses
on
the
benefits
to
reservoir
management.
We
investigate
operational
strategies
for
achieving
an
effective
tradeoff
between
pressure
relief/improved-
injectivity
and
delayed
CO2breakthrough
at
brine
producers.
For
vertical
wells,
an
injection-only
strategy
is
compared
to
a
pressure-management
strategy
with
brine
production
from
a
double-ring
9-spot
pattern.
Brine
production
allows
injection
to
be
steadily
ramped
up
while
staying
within
the
pressure-buildup
target,
while
injection-only
requires
a
gradual
ramp-down.
Injector/producer
horizontal-well
pairs
were
analyzed
for
a
range
of
well
spacings,
storage-formation
thickness
and
area,
level
and
dipping
formations,
and
for
homogeneous
and
heterogeneous
permeability.
When
the
producer
is
downdip
of
the
injector,
the
combined
influence
of
buoyancy
and
heterogeneity
can
delay
CO2breakthrough.
Both
vertical
and
hori-
zontal
wells
can
achieve
pressure
relief
and
improved
CO2injectivity,
while
delaying
CO2breakthrough.
Pressure
buildup
and
CO2breakthrough
are
sensitive
to
storage-formation
permeability
and
insensitive
to
all
other
hydrologic
parameters
except
caprock-seal
permeability,
which
only
affects
pressure
buildup
for
injection-only
cases.
©
2011
Elsevier
Ltd.
All
rights
reserved.
1.
Introduction
In
order
to
stabilize
atmospheric
CO2concentrations
for
climate
change
mitigation,
CO2capture
and
storage
(CCS)
implementation
must
be
increased
by
several
orders
of
magnitude
over
the
next
two
decades
(Fig.
3
of
IEA,
2009).
CCS
in
deep
geological
formations
is
regarded
as
a
promising
means
of
lowering
the
amount
of
CO2emit-
ted
to
the
atmosphere
and
thereby
mitigate
global
climate
change
(IEA,
2007).
In
order
for
widespread
deployment
of
industrial-
scale
CCS
to
be
achievable,
a
number
of
implementation
barriers
must
be
addressed.
Previously
identified
barriers,
such
as
CO2
∗Corresponding
author
at:
Lawrence
Livermore
National
Laboratory,
P.O.
Box
808,
L-223,
Livermore,
CA
94551,
USA.
Tel.:
+1
925
423
9390;
fax:
+1
925
423
0153.
E-mail
address:
buscheck1@llnl.gov
(T.A.
Buscheck).
capture
cost,
absence
of
CO2transport
network,
sequestration
safety,
legal
and
regulatory
barriers,
and
public
acceptance
have
been
recognized
for
a
number
of
years,
as
discussed
in
the
Spe-
cial
Report
on
CCS
(SRCCS)
(IPCC,
2005).
Implementation
barriers
receiving
more
recent
attention
include
water-use
demands
from
CCS
operations
and
pore-space
competition
with
emerging
activ-
ities,
such
as
shale-gas
production
(Court
et
al.,
2011a).
The
implementation
barrier
of
water-use
demands
for
CCS
may
be
particularly
acute
in
regions
where
water
resources
are
already
scarce.
A
comprehensive
review
is
presented
by
Court
et
al.
(2011a)
of
progress,
since
the
SRCCS,
on
several
of
these
large-scale
CCS
implementation
challenges:
water,
sequestration,
and
pore-space
competition;
legal
and
regulatory;
and
public
acceptance.
Active
CO2Reservoir
Management
(ACRM),
in
conjunction
with
CO2Capture,
Utilization,
and
Storage
(CCUS),
is
being
considered
as
a
means
of
addressing
some
of
these
implementation
barriers
1750-5836/$
–
see
front
matter
©
2011
Elsevier
Ltd.
All
rights
reserved.
doi:10.1016/j.ijggc.2011.11.007

Author's personal copy
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
231
(Buscheck
et
al.,
2011a,b,c;
Court
et
al.,
2011a,b;
Court,
2011).
In
this
paper,
we
first
discuss
the
challenges
of
saline-formation
CCS
in
Section
1.1.
We
then
suggest
potential
operational
and
licensing
benefits
of
ACRM
in
Section
1.2,
and
how
brine
produced
by
ACRM
may
enable
utilization
aspects
of
CCUS
in
Section
1.3.
In
Sections
2
and
3,
we
examine
how
ACRM
may
enhance
CO2storage,
through
an
analysis
of
operational
strategies
to
manipulate
CO2migration,
relieve
reservoir
pressure
buildup,
and
improve
injectivity,
while
delaying
the
breakthrough
of
CO2at
brine
producers.
1.1.
Background
The
most
promising
settings
for
widespread
deployment
of
industrial-scale
CCS
are
depleted
oil
reservoirs,
particularly
those
suited
to
CO2-based
Enhanced
Oil
Recovery
(CO2-EOR),
and
deep
saline
formations,
with
each
having
the
advantage
of
being
well
separated
from
the
atmosphere.
Industrial-scale
CO2storage
will
involve
large
volumes
of
injected
fluid
and
a
need
for
signifi-
cant
formation
storage
capacity
(Buscheck
et
al.,
2011c).
A
distinct
advantage
of
CO2-EOR
is
that
it
involves
fluid
production
(i.e.,
extraction),
which
increases
CO2storage
capacity
and
relieves
pres-
sure
buildup,
while
injection-only,
saline-formation
CCS
does
not
(Buscheck
et
al.,
2011c).
Yet,
because
of
limitations
in
the
volume
and
distribution
of
depleted
oil
reservoirs
and
the
large
volumes
and
widespread
availability
of
saline
formations,
CO2storage
in
saline
formations
is
likely
to
play
a
more
dominant
role
in
CCS
(IPCC,
2005).
The
absence
of
fluid
production
in
injection-only,
industrial-
scale,
saline-formation
CCS
may
result
in
a
large
pressure
buildup,
particularly
in
closed
or
semi-closed
formations,
persisting
both
during
and
long
after
injection
has
ceased.
Such
large
and
last-
ing
pressure
perturbations
will
require
careful
monitoring
and
may
require
restriction
of
injection
pressures
to
prevent
increasing
“fail-
ure”
risks
of
caprock
fracturing,
leakage
up
abandoned
wells,
and
induced
seismicity
(Morris
et
al.,
2011;
Rutqvist
et
al.,
2007;
Bachu,
2008).
If
not
sufficiently
controlled,
high
pressures
may
drive
CO2
and
brines
through
leakage
pathways
and
threaten
water
qual-
ity
in
shallower
water-supply
aquifers
(Bachu,
2008;
Carroll
et
al.,
2008).
Thus,
pressure
buildup
is
considered
to
be
a
limiting
factor
on
CO2storage
capacity
and
security,
and
storage-capacity
esti-
mates
based
on
effective
pore
volume
available
for
safe
trapping
of
CO2may
have
to
be
substantially
reduced
(Birkholzer
and
Zhou,
2009).
A
basin-scale
reservoir
model
showed
large
enough
pressure
interference
between
neighboring
CCS
operations
to
suggest
that
the
potential
area
to
be
characterized
in
a
CCS
permitting
process,
including
the
Area
of
Review
(AoR),
could
be
quite
large,
and
pre-
clude
the
possibility
of
permits
being
granted
on
a
single-site
basis
alone
(Birkholzer
and
Zhou,
2009).
1.2.
Active
CO2Reservoir
Management
(ACRM)
Active
CO2Reservoir
Management
(ACRM)
is
being
developed
to
enable
CO2Capture,
Utilization,
and
Storage
(CCUS)
in
saline
formations.
ACRM
combines
brine
production
with
CO2injection
(Buscheck
et
al.,
2011c;
Court
et
al.,
2011a,b)
with
the
primary
goal
of
enhancing
reservoir
performance,
thereby
enabling
more
secure
and
cost-effective
CO2storage.
The
specific
reservoir
per-
formance
objectives
of
ACRM
are
to
relieve
pressure
buildup,
increase
CO2injectivity,
increase
available
pore
space
and
storage
capacity,
manipulate
CO2migration,
and
constrain
brine
migra-
tion.
ACRM
is
being
considered
for
specific
CCS
sites
in
the
state
of
Wyoming
(Surdam
et
al.,
2009)
and
off
the
coast
of
Norway
(Bergmo
et
al.,
2011).
ACRM
practice
is
inherent
to
all
CO2-EOR
operations.
Because
CO2-EOR
utilizes
CO2in
a
beneficial
fashion,
it
is
a
CCUS
process.
ACRM
provides
an
opportunity
to
pursue
a
second
goal,
developing
potential
utilization
options
for
produced
brine
(Buscheck
et
al.,
2011a,b).
ACRM
has
the
potential
of
providing
reservoir
performance
advantages
for
CO2-storage-formation
siting,
site
characterization,
model
calibration,
uncertainty
reduction,
and
permitting.
Com-
pared
to
injection-only
CCS,
ACRM
enables
greater
control
of
subsurface
fluid
migration
and
pressure
perturbations.
Brine
pro-
duction
allows
for
“push–pull”
manipulation
of
the
CO2plume,
which
can
expose
less
of
the
caprock
seal
to
CO2and
more
of
the
storage
formation
to
CO2,
with
a
greater
fraction
of
the
storage
for-
mation
utilized
for
trapping
mechanisms
(Buscheck
et
al.,
2011c).
Another
form
of
CO2plume
manipulation
involves
reinjection
of
brine
on
top
of
the
CO2plume
to
accelerate
CO2dissolution
and
increase
solubility
trapping
(Keith
et
al.,
2004;
Hassanzadeh
et
al.,
2009).
When
CO2-storage
capacity
is
increased
and
brine
migration
reduced,
the
area
required
for
securing
mineral
rights
is
reduced
per
unit
of
stored
CO2.
If
the
net
extracted
volume
of
brine
is
equal
to
the
injected
CO2volume,
pressure
buildup
and
the
areal
extent
of
pressure
perturbations
are
minimized,
reducing
the
Area
of
Review
(AoR)
by
as
much
as
two
orders
of
magnitude
(Buscheck
et
al.,
2011c).
Definition
and
determination
of
the
AoR
is
presented
in
Court
(2011).
Their
work
demonstrated
the
potential
of
AoR
reduc-
tion
from
brine
production
using
a
simple
analytical
model
(Court,
2011).
Court
(2011)
also
provide
a
reservoir-scale
quantification
of
the
reduction
of
CO2and
brine
leakage
through
thousands
of
abandoned
wells,
resulting
from
pressure
relief
caused
by
brine
production.
ACRM
has
the
potential
of
reducing
other
risks
associ-
ated
with
pressure
buildup,
such
as
induced
seismicity.
ACRM
also
has
the
potential
of
reducing
the
volume
of
rock
over
which
brine
may
migrate
by
more
than
two
orders
of
magnitude
(Buscheck
et
al.,
2011c).
Brine
producers
can
function
as
actively
controlled
monitoring
wells,
providing
valuable
information
about
CO2-plume
migration
when
CO2breakthrough
occurs,
which
supports
history
match-
ing
and
model
calibration
and
also
reduces
uncertainty.
The
use
of
“smart-well”
technology,
with
down-hole
sensors
and
multiple
independently
controlled
production
zones
(Brouwer
et
al.,
2001;
Brouwer
and
Jansen,
2004;
Sudaryanto
and
Yortsos,
2001;
Alhuthali
et
al.,
2007),
could
extend
the
useful
lifetime
of
a
brine
producer
beyond
when
CO2is
first
detected.
Down-hole
zonal
control
could
also
be
applied
to
CO2injectors,
which
would
further
enhance
the
ability
to
manipulate
the
CO2plume.
After
CO2breakthrough
can
no
longer
be
mitigated
by
zonal
control,
it
could
be
possible
to
convert
a
brine
producer
to
a
CO2injector,
which
would
increase
overall
CO2injectivity
and
facilitate
additional
zonal
control
of
CO2-plume
migration.
Reducing
the
areal
and
vertical
extent
of
pressure
pertur-
bations
and
fluid
migration
would
lessen
the
possibility
of
imposing
operational
constraints
on
adjacent
subsurface
activi-
ties,
including
neighboring
CCS
operations.
Minimized
pressure
and
fluid-migration
interactions
between
neighboring
CCS
operations
facilitates
independent
planning,
assessment,
and
permitting
of
each
CCS
operation
within
a
basin.
It
would
also
reduce
pore-space
competition
with
other
subsurface
activities,
such
as
shale-gas,
deep
liquid-waste
injection,
and
geothermal
energy
production.
Thus,
ACRM
could
allow
CCS
sites
to
operate
in
closer
proximity
to
other
subsurface
activities
than
possible
for
injection-only
CCS
operations.
These
benefits
can
help
streamline
permitting.
In
this
study
we
assume
an
extraction
ratio
of
one,
which
is
a
volumetric
balance
between
injected
CO2and
the
net
extrac-
tion
(extraction
minus
reinjection)
of
brine.
It
is
important
to
note
that
this
is
just
one
end
member
of
ACRM,
which
can
also
involve
extraction
ratios
less
than
one.
Depending
on
storage-formation
depth,
an
extraction
ratio
of
one
requires
the
removal
of
between
1.25
and
1.5
m3of
brine
per
ton
of
injected
CO2.
For
a
1
GWe
coal
plant
this
would
require
the
net
removal
of
about
10–12
million
m3
(8100–9700
acre
feet)
of
brine
per
year
from
the
storage
formation.

Author's personal copy
232
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
Therefore,
a
major
challenge
for
ACRM
is
developing
cost-effective
solutions
to
reducing
the
volume
of
brine
in
the
storage
formation,
which
is
discussed
in
the
next
section.
1.3.
CO2Capture,
Utilization,
and
Storage
(CCUS)
ACRM
provides
benefits
to
reservoir
management
at
the
cost
of
extracting
brine.
This
added
cost
must
be
offset
by
the
added
ben-
efits
to
the
storage
operation
(e.g.,
fewer
injection
wells,
reduced
CO2compression
cost,
smaller
AoR,
and
reduced
duration
of
mon-
itoring)
and/or
by
creating
new,
valuable
uses
that
can
reduce
the
total
added
cost.
Utilization
options
of
choice
for
a
particular
CCUS
site
depend
on
the
chemical
composition
and
temperature
of
the
produced
brine,
as
well
as
the
proximity
to
the
potential
markets.
Useful
products
may
include
freshwater,
saline
cooling
water,
make-up
water
for
oil,
gas,
and
geothermal
energy
pro-
duction,
and
direct
recovery
of
geothermal
energy
(Harto
and
Veil,
2011;
Bourcier
et
al.,
2011;
Buscheck,
2010;
Buscheck
et
al.,
2011b).
Because
brine
disposal
is
a
major
challenge
for
ACRM,
a
key
objective
for
brine
utilization
is
to
provide
environmentally
safe,
cost-effective
solutions
for
brine
disposition.
Because
of
its
impor-
tance
to
the
viability
of
ACRM,
we
present
an
overview
of
potential
brine
utilization/disposition
options.
Brine-utilization
options
involve
a
full
range
of
treatment
possi-
bilities,
from
desalination
to
produce
freshwater,
to
softening
(e.g.,
ion
exchange
or
nanofiltration)
and/or
the
addition
of
corrosion
inhibitors
to
produce
saline
cooling
water
for
power
plants,
to
pos-
sibly
no
treatment
for
make-up
water
that
is
injected
for
pressure
support
in
oil,
gas,
and
geothermal
energy
production.
When
it
is
feasible
to
use
brine
as
make-up
water
there
is
no
brine
disposal
issue.
When
brine
is
used
as
a
feedstock
to
produce
freshwater
or
for
saline
cooling
water,
there
is
the
need
to
either
dispose
of
or
to
reinject
the
residual
brine,
either
into
the
CO2storage
formation
itself
or
into
a
separate
formation.
The
net
volume
of
produced
brine
can
be
reduced
by
partial
treatment
to
yield
freshwater
along
with
more
concentrated
brine
which
is
returned
to
the
reservoir
with
net
volume
reduction.
While
this
may
be
the
most
valuable
option,
on
a
per
unit
basis,
it
also
involves
the
most
expensive
forms
of
brine
treatment
because
it
requires
desalination,
such
as
Reverse
Osmosis
(RO).
Produced
brine
can
also
be
used
for
cooling
purposes,
such
as
in
saltwater
or
brackish-water
cooling
towers
(Maulbetsch
and
DiFilippo,
2010),
with
cooling
water
blowdown
(concentrated
residual
brine)
either
disposed
of
or
returned
to
the
reservoir.
Compared
to
desalina-
tion,
this
option
requires
less-costly
treatment,
such
as
softening
by
ion
exchange
or
nanofiltration
and/or
the
addition
of
corro-
sion
inhibitors
(Duke,
2007).
Evaporation
is
inherent
to
utilizing
water
for
cooling
purposes.
The
benefit
of
using
brine
for
cooling
is
to
supplant
the
need
to
consume
valuable
freshwater
resources
by
evaporation
during
the
cooling
process.
For
ACRM,
the
bene-
fit
of
evaporation
is
that
it
reduces
the
volume
of
residual
brine.
An
important
consideration
in
the
feasibility
of
utilizing
brine
for
cooling
purposes
is
the
cost
of
brine
transportation
between
the
CO2-storage
formation
and
cooling-water
user
(Harto
and
Veil,
2011).
Bourcier
et
al.
(2011)
conducted
a
preliminary
cost
estimate
for
RO
desalination
of
produced
brine
associated
with
CO2storage.
They
found
current
RO
technology
capable
of
treating
salinities
up
to
about
85,000
mg/L
TDS
(total
dissolved
solids),
which
is
a
value
not
exceeded
in
about
half
of
the
sampled
formation
brines
in
the
United
States
(Aines
et
al.,
2011;
based
on
data
from
Breit,
2002).
For
fresh
water
production,
Bourcier
et
al.
(2011)
estimated
RO
desalination
costs
ranging
from
$0.32
to
$0.80/m3of
permeate
(fresh
water).
Their
estimates
included
costs
of
all
surface
facilities,
transfer
pumps,
heat
exchangers
for
cooling,
and
piping,
but
did
not
include
the
cost
of
brine
production
and
reinjection
wells,
which
will
be
site
dependent.
For
a
net
removal
of
1.25
to
1.5
m3of
brine
per
ton
of
injected
CO2,
those
treatment
costs
translate
to
$0.40
to
$1.20/ton
of
CO2.
Offsetting
that
cost
would
be
the
market
value
of
produced
fresh
water.
As
discussed
in
Section
3,
brine
produc-
tion
has
the
potential
of
increasing
CO2-well
injectivity,
which
can
reduce
the
total
number
of
required
wells
and
CO2compression
cost.
The
capacity
of
currently
operating
RO
desalination
plants
is
large
compared
to
the
scale
of
net
brine
reduction
associated
with
ACRM
as
discussed
in
this
paper.
The
Perth,
Australia
seawater
RO
desalination
plant
has
been
operating
since
2006
with
a
capacity
of
52
million
m3per
year
(Sanz
and
Stover,
2007).
This
plant
has
an
overall
recovery
rate
of
42%
and
consumes
less
than
4.2
kWe-
h/m3of
permeate,
including
intake,
pretreatment,
both
RO
passes,
post-treatment,
potable
water
pumping,
and
all
electrical
losses.
Applied
to
ACRM
and
an
extraction
ratio
of
one,
a
plant
of
this
scale
could
consume
enough
brine
to
sequester
between
35
and
46
million
tons
of
CO2per
year;
which
is
the
amount
of
CO2emitted
from
coal
plants
generating
4.4–5.8
GWe.
It
is
worth
noting
that
the
desalination
capacity
of
the
Perth
RO
plant
is
an
order
of
magnitude
larger
than
what
would
be
required
for
the
ACRM
cases
analyzed
in
Section
3
of
this
paper.
There
is
an
expanding
industrial
experience
base
in
the
use
of
saline
cooling
water.
One
sector
(Maulbetsch
and
DiFilippo,
2010)
consists
of
otherwise
conventional
power
plants
that
use
estuarine
water
or
seawater
in
slightly
oversized
cooling
towers.
The
sec-
ond
sector
(Duke,
2007)
consists
of
power
plants
that
utilize
the
“zero
liquid
discharge”
(ZLD)
concept,
in
which
no
residual
liquid
is
returned
to
the
original
source.
ZLD
attempts
to
vaporize
a
large
fraction
of
the
water
in
the
cooling
process
by
making
a
large
num-
ber
of
cycles
so
as
to
minimize
the
amount
of
blowdown
for
final
disposal.
In
ZLD,
the
input
water
is
usually
pretreated
by
softening,
normally
by
ion
exchange.
This
reduces
the
scaling
potential
asso-
ciated
with
Ca
and
Mg.
Another
pretreatment
is
to
raise
the
pH,
which
acts
to
prevent
precipitation
of
silica.
These
steps
appear
to
not
only
effectively
control
scaling,
but
also
metal
corrosion
(Duke,
2007).
After
treatment
to
ensure
chemical
compatibility,
brine
can
be
utilized
directly
as
injected
make-up
water
for
pressure
support
of
oil
and
gas
production,
enhanced
geothermal
systems,
and
geother-
mal
power
recovered
from
hydrothermal
systems
(Harto
and
Veil,
2011).
Enhanced
geothermal
systems
(EGS)
are
typically
located
in
geologic
settings
lacking
formation
water
and
permeability
(MIT,
2006).
Hence,
EGS
may
require
water
to
stimulate
fracture
perme-
ability,
act
as
a
working
fluid,
and
to
make
up
for
injected
water
lost
to
the
formation.
Primary
issues
are
chemical
compatibility
with
the
receiving
formation
and
cost
of
brine
transportation
between
the
CO2-storage
formation
and
brine-receiving
formation
(Harto
and
Veil,
2011).
For
crude
oil
production,
the
rates
of
co-produced
water
can
be
large
compared
to
brine
production
for
ACRM.
In
the
state
of
Wyoming,
2.12
billion
barrels
(337
million
m3)
of
water
were
co-produced
in
2002,
along
with
54.7
million
barrels
of
oil
(Veil
et
al.,
2004).
Applied
to
ACRM,
brine
production
of
this
scale
would
sequester
between
225
and
270
million
tons
of
CO2per
year,
which
is
the
amount
of
CO2emitted
from
coal
plants
generating
about
28–34
GWe.
Geothermal
energy
can
be
directly
recovered
from
the
pro-
duced
brine
to
help
offset
the
increased
operational
cost,
when
combined
with
one
of
the
previous
volume
reduction
methods.
Geothermal
energy
production
can
be
limited
by
pressure
deple-
tion,
whereas
pressure
buildup
is
the
limiting
consideration
for
CO2storage
capacity
and
security.
These
two
complementary
sys-
tems
can
be
integrated
synergistically,
with
CO2injection
providing
pressure
support
to
maintain
the
productivity
of
geothermal
wells,
while
the
production
of
geothermal
brine
provides
pressure
relief
and
improved
injectivity
for
CO2-injection
wells
(Buscheck,
2010;

Author's personal copy
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
233
Buscheck
et
al.,
2011b).
An
integrated
geothermal–CCS
system,
actively
managed
to
yield
a
volumetric
balance
between
injected
and
produced
fluids,
mitigates
the
risks
of
reservoir
overpressure
(CCS
concern)
or
underpressure
(geothermal
concern),
including
induced
seismicity,
insufficient
well
productivity
or
injectivity,
subsidence,
and
fluid
leakage
either
to
or
from
overlying
forma-
tions.
1.4.
Objectives
of
this
study
From
a
reservoir-performance
perspective,
the
key
objective
for
ACRM
is
for
brine
production
to
relieve
pressure
buildup
driven
by
CO2injection.
Another
perhaps
less
intuitive
objective
is
to
reduce
total
operating
costs
of
CO2storage,
on
a
per
unit
of
stored-CO2
basis,
through
the
reduction
of
the
total
number
of
wells
and
the
cost
of
CO2compression.
Other
components
of
CO2storage
cost
are
infrastructure
costs,
such
as
those
related
to
obtaining
mineral
rights,
liability
insurance,
site
characterization,
and
monitoring.
As
discussed
in
Section
3,
ACRM
has
the
potential
of
reducing
many
of
these
costs.
These
and
other
costs,
such
as
those
associated
with
the
disposition
of
the
produced
brine,
are
likely
to
be
site
specific.
Because
of
the
breadth
and
complexity
of
determining
total
CO2
storage
costs,
it
is
beyond
the
scope
of
this
paper.
We
defer
eco-
nomic
analyses
of
the
net
cost
(and
benefit)
of
ACRM
to
future
studies.
Brine
production
can
eventually
cause
CO2breakthrough
at
brine
producers.
The
operational
challenge
for
ACRM
is
that
pres-
sure
relief
increases
with
decreasing
spacing
between
CO2injectors
and
brine
producers,
while
CO2-breakthrough
time
decreases.
Thus,
there
is
a
tradeoff
between
achieving
sufficient
pressure
relief
and
delaying
CO2breakthrough.
There
are
several
operational
strategies
that
can
better
achieve
this
trade-off.
One
strategy
is
to
successively
produce
brine
from
a
series
of
production
wells
that
are
incrementally
spaced
farther
from
the
injection
well
(Buscheck
et
al.,
2011a,c).
A
second
strategy,
which
could
be
used
in
combina-
tion
with
the
first
strategy,
involves
the
use
of
horizontal
injection
and
production
wells.
A
third
strategy,
which
could
be
combined
with
the
other
strategies,
is
the
use
of
“smart
wells”
(Brouwer
et
al.,
2001;
Brouwer
and
Jansen,
2004;
Sudaryanto
and
Yortsos,
2001;
Alhuthali
et
al.,
2007),
with
down-hole
sensors
and
multiple
inde-
pendently
controlled
production
and
injection
zones
to
extend
the
useful
lifetime
of
brine-production
wells
beyond
when
CO2is
first
detected.
For
this
study,
we
conduct
reservoir
analyses
to
investigate
the
first
strategy
of
producing
brine
from
successively
increasing
dis-
tances
from
the
injection
well,
applied
to
vertical
injectors
and
producers,
and
then
investigate
the
second
strategy,
for
horizontal
injector/producer-well
pairs.
We
defer
the
investigating
the
third
strategy
(smart
wells)
to
future
studies.
For
the
vertical-well
study
we
apply
the
conceptual
model
used
in
earlier
studies
(Buscheck
et
al.,
2011c;
Zhou
et
al.,
2008),
which
considered
homogeneous
formations.
For
the
horizontal-well
study,
we
modify
that
concep-
tual
model
to
include
a
wide
range
of
storage-formation
thickness,
dipping
and
level
formations,
various
caprock
thicknesses,
and
heterogeneous
permeability
distributions.
We
also
investigate
the
sensitivity
of
pressure
relief
and
CO2breakthrough
to
the
key
hydrologic
parameters.
2.
Methodology
In
this
study,
we
used
the
NUFT
(Nonisothermal
Unsaturated-
saturated
Flow
and
Transport)
code,
which
was
developed
at
Lawrence
Livermore
National
Laboratory
to
simulate
multi-phase
multi-component
heat
and
mass
flow
and
reactive
transport
in
unsaturated
and
saturated
porous
media
(Nitao,
1998;
Buscheck
Table
1
Summary
of
hydrologic
property
values
used
in
the
study.
Property
Storage
formation
Caprock
seal
Horizontal
and
vertical
permeability
(m2)
Homogeneous
case 10−13 10−18
Heterogeneous
case
110
−13 and
10−14 10−18
Heterogeneous
case
2
10−13 and
10−15 10−18
Pore
compressibility
(Pa−1)
4.5
×
10−10 4.5
×
10−10
Porosity
0.12
0.12
van
Genuchten
(1980)
m
0.46
0.46
van
Genuchten
˛
(Pa−1)
5.1
×10−55.1
×10−5
Residual
supercritical
CO2saturation
0.05 0.05
Residual
water
saturation 0.30
0.30
et
al.,
2003;
Johnson
et
al.,
2004a,b;
Carroll
et
al.,
2009;
Morris
et
al.,
2011).
The
pore
and
fluid
compressibility
are
4.5
×
10−10 and
3.5
×10−10 Pa−1,
respectively.
Water
density
is
determined
by
the
ASME
steam
tables
(ASME,
2006).
The
two-phase
flow
of
CO2and
water
was
simulated
with
the
density
of
supercritical-CO2deter-
mined
by
the
correlation
of
Span
and
Wagner
(1996)
and
viscosity
determined
by
the
correlation
of
Fenghour
et
al.
(1998).
Because
we
focused
on
the
response
in
the
storage
formation
and
adjoining
seal
units,
the
simulations
were
conducted
for
isothermal
conditions
at
a
fixed
temperature
of
45 ◦C.
Because
we
did
not
consider
the
rein-
jection
of
brine
in
our
study,
we
did
not
address
salinity-dependent
brine
density
and
viscosity.
The
influence
of
salinity-dependent
brine
density
and
viscosity
will
be
addressed
in
future
work
that
will
consider
reinjection
of
residual
brine.
Also,
the
influence
of
geomechanical
coupling
(Morris
et
al.,
2011)
and
geochemical
reac-
tions
resulting
from
CO2injection
were
not
considered.
To
simulate
CO2injection
and
brine
production,
we
applied
two
model
geometries:
(1)
3-D
models
of
a
vertical
CO2injec-
tor
surrounded
by
a
ring
(or
rings)
of
vertical
brine
producers
(Fig.
1a)
and
(2)
2-D
vertical-cross-sectional
models
of
horizontal
injector/producer
pairs.
The
numerical
grid
refinement
used
in
the
models
is
as
follows:
•3-D
models
of
vertical
wells:
200-m
×200-m
horizontally
by
10-
to
25-m
vertically
•2-D
models
of
horizontal
wells:
200-m
horizontally
by
20-m
ver-
tically
The
3-D
models
used
in
the
vertical-well
study
(Section
3.1)
represent
a
250-m-thick
storage
formation,
as
modeled
by
Zhou
et
al.
(2008)
and
Buscheck
et
al.
(2011c),
with
the
top
of
the
stor-
age
formation
located
1200
m
below
the
water
table
and
bounded
by
60-m-thick
seal
units.
The
outer
boundaries
have
a
no-flow
condition
to
represent
a
semi-closed
system
for
a
1256-km2stor-
age
formation.
The
lower
boundary
of
the
model,
1800
m
below
the
water
table,
has
a
no
flow
condition.
The
upper
1140
m
and
lower
290
m
of
the
model,
called
the
overburden
and
underburden,
have
the
same
hydrologic
properties
as
the
CO2storage
forma-
tion.
Hydrologic
properties
of
the
storage
formation
and
seal
units
(Table
1)
are
similar
to
previous
studies
(Zhou
et
al.,
2008;
Buscheck
et
al.,
2011c),
except
that
a
seal
permeability
of
10−18 m2is
used.
CO2injection
occurs
in
a
50-m
×
50-m
zone
in
the
lower
half
of
the
storage
formation,
at
a
rate
of
3.8
million
tons/year
for
an
injection
period
of
30
years,
unless
otherwise
noted.
Note
that
in
this
study
brine
production
always
occurs
at
a
specified
rate.
Brine
is
pro-
duced
in
the
lower
half
of
the
storage
formation
in
100-m
×
100-m
zones.
For
the
cases
with
brine
production,
we
maintained
a
vol-
umetric
balance
between
produced
brine
and
injected
CO2.
We
did
not
explicitly
represent
reinjection
of
brine
in
our
study.
The
vertical-well
study
only
considered
homogeneous
permeability
in
the
storage
formation.

Author's personal copy
234
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
Fig.
1.
(a)
A
plan
view
of
the
well
patterns
analyzed
in
the
vertical-well
study,
including
5-spot
patterns
with
5-
and
10-km
spacing
between
the
CO2injector
and
brine
producers
and
a
double-ring
9-spot
pattern
with
5-
and
10-km
well
spacing.
Note
that
the
5-spot
pattern
with
3-km
well
spacing
is
not
shown.
(b)
Pressure
buildup
histories
in
the
storage
formation
adjacent
to
the
top
of
the
CO2injector
for
5-spot
patterns
with
producers
at
5
and
10
km,
respectively,
and
of
a
double-ring
9-spot
pattern
(see
Fig.
1a),
with
brine
production
at
the
4
inner
producers
at
5
km
for
the
first
10
years,
ramping
from
full
to
zero
production
from
10
to
15
years.
At
10
years,
brine
production
begins
at
the
4
outer
producers
at
10
km,
ramping
from
zero
to
full
production
from
10
to
15
years,
and
continuing
until
the
end
of
injection.
CO2injection
rate
is
3.8
million
tons/year
for
30
years.
The
2-D
cross-sectional
models
used
in
the
horizontal-well
study
(Section
3.2)
included
one
representing
a
level
storage
for-
mation
(dip
angle
of
0◦)
and
the
other
representing
a
storage
formation
with
a
10%
slope
(dip
angle
of
5.7◦),
which
is
in
the
range
of
dip
angles
found
in
typical
sedimentary
formations.
The
models
represent
a
semi-closed
reservoir
system
that
is
40
km
in
the
lateral
direction
(orthogonal
to
the
well
axes)
and
4
km
in
the
longitudinal
direction
(parallel
to
the
well
axes),
with
no-
flow
boundaries
at
the
basement
of
the
storage
formation
and
at
the
lateral
and
longitudinal
boundaries.
This
is
representative
of
a
semi-closed
system
for
a
160-km2storage
formation.
For
level
formations,
calculations
were
also
made
for
a
1600-
and
16,000-
km2storage
formation
to
investigate
the
influence
of
basin
size.
Storage-formation
thicknesses
of
50,
100,
200
and
400
m
are
con-
sidered,
underlain
by
an
impermeable
basement
and
overlain
by
a
caprock
unit
with
thicknesses
of
50,
100,
200,
and
400
m.
A
con-
stant
pressure
boundary
is
maintained
at
the
top
of
the
caprock.
For
the
model
with
level
formations,
the
basement
is
1800
m
below
the
water
table.
For
the
model
of
the
dipping
formations,
the
base-
ment
is
1800
to
4800
m
below
the
water
table.
Hydrologic
property
values
(Table
1)
are
the
same
as
those
used
in
the
vertical-well
study.
The
horizontal-well
study
also
considered
cases
using
a
sim-
ple
conceptual
model
of
layered
heterogeneity
in
the
storage
formation,
with
40-m-thick
layers
of
alternating
high
and
low
permeability
and
permeability
contrasts
of
10
and
100.
The
homo-
geneous
and
layered-heterogeneous
conceptual
models
used
in
this
study
are
useful
and
appropriate
for
conducting
the
very
broad
range
of
sensitivity
analyses
addressed
in
this
study.
Real
sites
will
have
randomly
distributed
heterogeneity.
More
realistic
represen-
tation
of
heterogeneity,
with
randomly
distributed
permeability
pertaining
to
real
sites,
will
be
considered
in
future
studies
of
ACRM.
The
CO2injection
well
is
located
at
the
lowermost
20
m
of
the
storage
formation.
The
brine-production
well
is
also
located
at
the
lowermost
20
m
of
the
storage
formation,
either
5,
10,
15,
or
20
km
from
the
injection
well.
For
the
dipping
case,
the
brine-production
well
is
located
downdip
of
the
CO2injection
well.
Injection
periods
of
30
to
up
to
100
years
are
considered.
CO2injection
rates
of
0.475,
0.95,
1.9,
3.8,
and
7.6
million
tons/year
are
considered.
Because
a
2-
D
model
is
used,
the
injection
rate
is
distributed
over
a
longitudinal
distance
of
4
km.
We
conducted
grid-sensitivity
analyses
(Section
3.3)
to
demon-
strate
the
insensitivity
of
the
reservoir
analyses
to
the
numerical
grid
refinement
used
in
this
study.
We
also
conducted
a
parameter-
sensitivity
study
(Section
3.4)
that
showed
pressure
buildup
and
CO2breakthrough
time
to
be
most
sensitive
to
storage-formation
permeability
and
insensitive
to
all
other
hydrologic
parameters
except
caprock-seal
permeability,
which
only
affects
pressure
buildup
for
cases
with
no
brine
production.
3.
Results
The
following
two
sections
discuss
vertical-
and
horizontal-well
studies
of
operational
strategies
to
achieving
adequate
reser-
voir
pressure
relief,
while
delaying
CO2breakthrough
at
brine
producers.
We
also
investigate
how
brine
production
from
a
hor-
izontal
well
can
manipulate
CO2migration
from
a
horizontal
injector.
3.1.
Vertical-well
study
We
consider
CO2injection
from
a
single
vertical
well
and
ana-
lyze
the
influence
of
brine
production
from
patterns
of
vertical
wells.
We
first
examine
the
relationship
between
pressure
relief
of
pressure
buildup
at
the
CO2injector
and
the
distance
between
the
CO2injector
and
the
brine
producers.
We
then
compare
two
approaches
to
reservoir-pressure
management:
with
and
without
brine
production.
3.1.1.
Relationship
between
pressure
relief
and
spacing
between
the
CO2injector
and
brine
producers
We
analyzed
three
5-spot
patterns,
with
brine
producers
at
the
corners
of
the
patterns,
spaced
3,
5
and
10
km
from
the
CO2injec-
tor
at
the
center
of
the
pattern
(Fig.
1a).
The
case
with
no
brine
production
was
also
analyzed.
The
CO2injection
rate
is
3.8
mil-
lion
tons/year
for
30
years
and
a
volumetric
balance
is
maintained
between
injected
CO2and
produced
brine.
For
these
cases,
the
stor-
age
formation
has
an
area
of
1256
km2.
The
influence
of
brine
production
on
pressure
relief
in
the
storage
formation
adjacent
to
the
CO2injector
decreases
with
increasing
well
spacing
and
increases
with
time
(Table
2).
For
all
cases,
pressure
buildup
(P)
immediately
builds
up
to
a
peak
value

Author's personal copy
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
235
Table
2
Pressure-buildup
in
the
storage
formation
adjacent
to
the
top
of
the
CO2injector
for
an
injection
rate
of
3.8
million
tons/year
for
30
years
for
5-spot
patterns
with
well
spacings
of
3,
5,
and
10
km
between
the
CO2injector
and
brine
producers.
Time
(yr) Peak
pressure
buildup
(MPa)
No
brine
production Brine
production
at
the
indicated
well
spacing
3
km 5
km 10
km
0.01
2.95
2.95
2.95
2.95
1.0
2.72
1.51
1.90
2.28
5.0
2.40
0.66
1.07
1.58
10.0
2.49
0.41
0.74
1.24
15.0
2.68
0.29
0.56
1.05
20.0
2.87
0.23
0.46
0.92
30.0
3.23
0.14
0.35
0.74
31.0
2.29
−0.34
−0.29
−0.08
100.0
0.66
−0.09
−0.06
−0.06
of
2.95
MPa
(Table
2).
At
0.01
years
it
is
too
early
for
brine
produc-
tion
to
influence
P
for
any
of
the
well
spacings.
The
initial
decline
in
P
corresponds
to
the
fact
the
injection
well
is
perforated
in
the
lower
half
of
the
storage
formation
and
that
it
takes
time
for
the
influence
of
the
overlying
low-permeability
caprock
to
influence
pressures
adjacent
to
the
CO2injector.
After
the
pressure
perturba-
tion
fully
feels
the
influence
of
the
overlying
caprock,
P
adjacent
to
the
CO2injector
begins
to
increase
for
the
no-production
case,
which
continues
to
increase
for
the
duration
of
the
30-year
injec-
tion
period.
For
the
cases
with
brine
production,
the
influence
of
brine
production
on
relieving
P
occurs
prior
to
pressure
interfer-
ence
with
the
overlying
caprock.
For
this
discussion
we
define
pressure
relief
to
be
the
reduction
of
P
with
brine
production
divided
by
P
without
brine
produc-
tion.
Within
1
year,
the
CO2injector
experiences
pressure
relief
for
all
well
spacings.
Within
5
years,
pressure
relief
is
at
least
50%
for
well
spacings
of
3
and
5
km;
at
10
years,
pressure
relief
is
at
least
50%
for
all
well
spacings.
After
injection
ceases,
P
around
the
CO2injector
abruptly
drops
to
small
negative
values
(slightly
below
ambient
pressure)
for
all
well
spacings,
while
for
the
no-production
case,
P
persists
beyond
100
years.
With
brine
production,
buoy-
ancy
is
the
only
post-injection
driving
force
for
CO2and
brine
migration,
which
is
small
(compared
to
injection-driven
P)
for
CO2migration
and
negligible
for
brine
migration.
The
large
per-
sistent
post-injection
P
for
the
no-production
case
will
continue
to
drive
CO2and
brine
migration,
while
for
cases
with
brine
pro-
duction,
CO2migration
will
be
minor,
largely
occurring
updip,
fully
within
the
storage
formation,
while
outward
brine
migration
will
virtually
cease.
When
buoyancy
is
the
only
driving
force,
leakage
up
abandoned
wells
and
faults
is
less
of
a
concern.
The
large
reduction
of
post-injection
P
resulting
from
brine
production
could
have
a
positive
impact
on
post-injection
monitoring
requirements
and
on
the
cost
of
liability
insurance.
From
the
insights
gained
concerning
the
relationship
between
the
pressure
relief
and
well
spacing,
we
developed
a
well
pattern,
called
the
double-ring
9-spot
(Fig.
1a),
which
is
an
example
of
the
pressure-management
strategy
where
brine
is
produced
from
mul-
tiple
rings
of
brine
producers
spaced
incrementally
farther
from
the
CO2injector.
This
pattern
has
an
inner
ring
of
4
producers
5
km
from
the
CO2injector
and
an
outer
ring
of
4
producers
10
km
from
the
CO2injector
(Fig.
1a).
The
outer
ring
of
producers
is
rotated
by
45◦relative
to
the
inner
ring
in
order
to
“pull”
on
the
CO2plume
from
different
directions,
thereby
manipulating
the
plume
into
a
cylindrical
shape.
Brine
production
occurs
entirely
from
the
inner
4
producers
during
the
first
10
years;
during
the
next
5
years,
brine
production
is
gradually
shifted
to
the
outer
4
producers,
while
maintaining
the
same
total
brine
production
rate.
For
the
first
10
years,
the
P
history
for
the
double-ring
9-spot
follows
that
of
the
5-spot
with
5-km
spacing
(Fig.
1b).
As
brine
production
is
shifted
to
the
outer
ring
of
producers,
the
P
history
for
the
double-ring
9-spot
shifts
from
that
of
the
5-spot
with
5-km
spacing
to
that
of
the
5-spot
with
10-km
spacing.
Thereafter,
the
P
history
of
the
double-ring
9-spot
follows
that
of
the
5-spot
with
10-km
spacing.
3.1.2.
Comparing
pressure
management
approaches:
ACRM
versus
injection-only
To
illustrate
two
approaches
to
achieving
pressure
manage-
ment,
we
modified
the
double-ring
9-spot
example,
with
60
years
of
injection.
The
first
(ACRM)
approach
primarily
relies
on
brine
production
to
achieve
a
desired
(“target”)
value
of
Ppeak.
The
sec-
ond
(injection-only)
approach
relies
entirely
on
adjusting
the
CO2
injection
rate
to
achieve
a
target
value
of
Ppeak.
For
this
exam-
ple,
we
chose
a
Ppeak target
of
1.08
MPa,
because
it
is
close
to
the
value
of
P
at
5
years
(1.07
MPa)
for
the
case
with
brine
produc-
ers
at
5
km
(Table
2).
In
general,
a
target
value
of
Ppeak would
be
related
to
mitigating
risks,
such
as
those
related
to
the
potential
for
fracture
initiation
(Morris
et
al.,
2011)
or
fault
activation
(Rutqvist
et
al.,
2007).
The
no-production
case
with
a
constant
CO2injection
rate
of
3.8
million
tons/year
for
60
years
(Fig.
2b)
results
in
an
initial
Ppeak
of
2.95
MPa
and
an
ultimate
Ppeak of
4.06
MPa,
occurring
at
the
end
of
injection
(Fig.
2a).
For
the
double-ring
9-spot
case
with
a
constant
initial
CO2injection
rate
of
3.8
million
tons/year,
P
is
initially
2.95
MPa,
declining
to
1.07
MPa
at
5
years.
To
keep
P
just
at
the
target
value
(1.08
MPa),
CO2injection
rate
is
reduced
to
an
initial
value
of
1.2
million
tons/year,
ramped
up
to
4.0
mil-
lion
tons/year
at
5
years,
and
held
constant
until
15
years
(Fig.
2b).
The
yellow
area
in
Fig.
2a
represents
the
“overpressure”,
relative
to
the
P
target,
while
the
yellow
area
in
Fig.
2b
represents
the
required
reduction
in
CO2injection
rate
to
stay
within
the
P
target.
At
15
years,
brine
production
has
completely
shifted
to
the
outer
4
producers
at
10
km.
Brine
production
at
10
km
results
in
a
P
of
1.05
MPa
at
15
years
(Table
2),
which
is
just
below
the
target
value.
Because
pressure
relief
from
brine
production
at
10
km
increases
with
time,
it
is
possible
to
continuously
ramp
up
the
CO2injection
rate
from
4.0
to
8.6
million
tons/year
for
duration
of
the
injection
period
(Fig.
2b)
and
remain
close
to
the
P
target
(Fig.
2a).
To
keep
P
at
the
target
value
for
the
no-
production
case,
CO2injection
rate
is
reduced
to
an
initial
value
of
1.1
million
tons/year,
slowly
increased
to
1.5
million
tons/year
at
5
years,
then
gradually
reduced
to
0.7
million
tons/year
at
60
years.
A
way
to
quantify
the
pressure-relieving
benefit
of
brine
produc-
tion
is
the
injectivity
ratio,
determined
by
dividing
CO2injection
rate
for
the
ACRM
case
by
CO2injection
rate
for
the
no-production
case
for
the
same
value
of
P.
Thus,
injectivity
ratio
varies
con-
tinuously
with
time.
For
this
pressure-management
example,
the
injectivity
ratio
starts
at
a
value
of
1.1,
increasing
to
2.7,
5.4,
and
12.2
at
5,
30,
and
60
years,
respectively.
For
the
60-year
injection
period,
brine
production
enables
a
5.5-fold
increase
in
stored
CO2,
compared
to
the
no-production
case.
3.2.
Horizontal-well
study
We
consider
CO2injection
from
a
single
horizontal
well
and
ana-
lyze
the
influence
of
brine
production
from
a
single
horizontal
well
at
distances
of
5,
10,
15,
and
20
km
from
the
CO2injector.
We
inves-
tigate
the
influence
of
brine
production
on
CO2plume
migration,
pressure
buildup
(P)
in
the
storage
formation
adjacent
to
the
CO2
injector,
injectivity
of
the
CO2injector,
and
CO2breakthrough
at
the
brine
producer.

Author's personal copy
236
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
Fig.
2.
Pressure
buildup
history
in
the
storage
formation
adjacent
to
the
top
of
the
CO2injector
(a)
and
CO2injection
rate
(b)
are
plotted
for
two
“no-production”
cases
and
two
“double-ring
9-spot”
cases,
all
with
60
years
of
injection.
The
no-production
cases
include
“constant
injection”
with
a
CO2injection
rate
of
3.8
million
tons/year
and
“ramped
injection”
with
injection
rate
reduced
just
enough
to
keep
pressure
buildup
below
a
specified
value
(dotted
green
curve).
The
double-ring
9-spot
cases
only
differ
by
virtue
of
the
initial
CO2injection
rates,
as
depicted
by
the
yellow
area
in
(b),
with
“constant
initial
injection”
having
an
initial
rate
of
3.8
million
tons/year
and
“ramped
injection”
having
the
initial
rate
reduced
just
enough
to
keep
pressure
buildup
below
the
“target”
value.
The
yellow
area
in
(a)
shows
the
influence
that
the
initial
CO2
injection
rate
reduction
has
on
pressure
buildup.
Later,
the
CO2injection
rate
for
the
double-ring
9-spot
cases
is
gradually
increased
just
enough
to
keep
pressure
buildup
below
the
target
value.
3.2.1.
Manipulating
CO2plume
migration
The
use
of
brine
production
to
manipulate/steer
the
CO2plume
has
been
analyzed
for
CO2injection
from
a
single
vertical
well
surrounded
by
rings
of
vertical
production
wells
(Buscheck
et
al.,
2011c;
Court,
2011).
Figs.
3
and
4
from
Buscheck
et
al.
(2011c)
showed
significant
steering
potential;
however,
those
analyses
are
applicable
to
many
rings
of
CO2brine
producers,
which
is
unlikely
to
be
economically
practical.
Court
(2011)
found
a
single
ring
of
4
vertical
production
wells,
placed
outside
of
the
outer
extent
of
the
CO2plume
(in
order
to
avoid
CO2breakthrough),
has
negligi-
ble
steering
potential
on
the
CO2plume
and
they
concluded
that
more
complex
vertical-well
strategies
would
need
to
be
investi-
gated.
In
this
section,
we
investigate
the
effectiveness
of
CO2-plume
manipulation/steering
for
a
horizontal
injector/producer-well
pair.
We
examine
the
influence
of
brine
production
on
CO2plume
migration
in
dipping
storage
formations,
for
cases
with
homoge-
neous
and
heterogeneous
permeability
in
the
storage
formation.
We
consider
a
CO2-storage
formation
that
is
400
m
thick,
overlain
by
a
400-m-thick
caprock,
and
with
a
10%
slope
(5.7◦dip
angle).
The
brine
producer
is
located
10
km
downdip
from
the
CO2injector
and
the
injection
period
varies,
depending
on
when
CO2breakthrough
occurs.
Supercritical
CO2largely
flows
preferentially
through
the
high-permeability
layers
(Fig.
3),
while
diffusion
of
aqueous-phase
CO2distributes
CO2into
the
low-permeability
layers
(Fig.
4).
Homogeneous
permeability
in
the
storage
formation
allows
buoyancy
to
strongly
drive
CO2updip
for
the
case
with
no
brine
production
(Figs.
3a
and
4a).
The
addition
of
brine
production
10
km
downdip
of
the
CO2injector
largely
negates
the
influence
of
buoyancy
(Figs.
3b
and
4b),
pulling
the
CO2plume
down
to
the
brine
producer,
where
breakthrough
occurs
at
46
years.
The
influence
of
layered
heterogeneous
permeability
in
the
storage
for-
mation
impedes
the
buoyancy-driven
migration
of
the
CO2plume
(Figs.
3c
and
4c).
When
the
permeability
contrast
is
increased,
lay-
ered
heterogeneity
more
strongly
impedes
the
buoyancy-driven
migration
of
the
CO2plume
(Figs.
3e
and
4e),
so
much
so
that
the
CO2plume
is
almost
symmetrical
about
the
CO2injector.
For
the
ACRM
case,
layered
heterogeneous
permeability
in
the
storage
formation
causes
the
CO2plume
to
be
more
evenly
distributed
ver-
tically
in
the
storage
formation,
which
delays
the
arrival
of
the
CO2
plume
at
the
brine
producer,
increasing
breakthrough
time
to
50
years
(Figs.
3d
and
4d).
When
the
permeability
contrast
is
increased,
layered
heterogeneity
much
more
evenly
distributes
the
CO2plume
vertically
in
the
storage
formation,
which
further
delays
the
arrival
of
the
CO2plume
at
the
brine
producer,
increasing
breakthrough
time
to
75
years
(Figs.
3f
and
4f).
3.2.2.
Pressure
relief
and
injectivity
We
consider
the
relationship
between
Ppeak,
injectivity,
and
CO2breakthrough
time,
starting
with
a
400-m-thick
storage
forma-
tion
with
an
area
of
160
km2.
Ppeak increases
with
CO2injection
rate
and
well
spacing
between
the
producer/injector
pair
(Fig.
5a
and
Table
3).
The
pressure-relieving
effect
of
brine
production
is
seen
as
a
reduction
in
slope
of
Ppeak versus
CO2injection
rate
(Qinj)
curve.
Because
pressure
relief
increases
with
decreasing
well
spacing,
the
slope
is
reduced
with
decreasing
well
spacing.
Con-
versely,
with
increasing
well
spacing,
the
Ppeak versus
Qinj slope
increases;
for
large
enough
well
spacing,
it
approaches
that
of
the
no-production
case.
Heterogeneity
has
a
modest
influence
on
Ppeak for
the
no-
production
case,
while
for
cases
with
brine
production
the
influence
on
Ppeak is
much
stronger.
Compared
to
the
homogeneous
case,
the
heterogeneous
case
has
the
effective
permeability
in
the
horizontal
direction
reduced
by
45%
(5.5
×
10−14 m2versus
1.0
×
10−13 m2).
Accordingly,
the
influence
is
equivalent
to
nearly
doubling
the
well
spacing
for
a
level
formation
(Table
3).
As
discussed
earlier,
injectivity
ratio
is
a
useful
way
to
quan-
tify
the
pressure-relieving
benefit
of
brine
production.
For
the
horizontal-well
study,
we
define
the
injectivity
ratio
with
respect
to
the
Ppeak,
rather
than
for
the
P
at
a
specific
point
in
time.
Thus,
injectivity
ratio
is
the
CO2injection
rate
for
the
ACRM
case
divided
by
the
CO2injection
rate
for
the
no-production
case
for
the
same
value
of
Ppeak.
For
the
horizontal-well
study,
injectivity
ratio
pertains
to
the
injection
period
as
a
whole.
Because
the
ACRM
case
has
twice
the
number
of
wells
as
the
no-production
case
and
because
the
CO2injector
and
brine
producer
have
the
same
perfo-
rated
lengths
(4
km),
an
injectivity
ratio
greater
than
2
indicates
a
savings
in
well-drilling
costs.
Thus,
improved
injectivity
facilitated
by
ACRM
may
reduce
well-drilling
costs.
In
the
following
analyses
brine
production
in
horizontal
wells
was
always
found
to
result
in

Author's personal copy
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
237
Fig.
3.
Liquid
saturation
contours
show
CO2-plume
migration
driven
by
CO2injection
from
a
horizontal
well
for
no
brine
production
(a,
c,
and
e),
and
for
brine
produced
in
a
horizontal
well
10
km
downdip
from
the
injection
well
(b,
d,
and
f).
Storage-formation-permeability
cases
are
(1)
homogeneous
(a
and
b),
(2)
layered
heterogeneity
with
a
permeability
contrast
of
10
(c
and
d),
and
(3)
layered
heterogeneity
with
permeability
contrast
of
100
(e
and
f).
The
heterogeneous
cases
have
alternating
40-m-thick
layers
of
high
and
low
permeability.
The
horizontal
injectors
and
producers
are
located
in
the
lower
20
m
of
the
storage
formation.
The
formation
dip
angle
is
5.7◦and
the
vertical
scale
in
the
plot
is
exaggerated
by
a
factor
of
5.
an
injectivity
ratio
greater
than
2,
with
injectivity
ratio
often
being
much
greater
than
2.
Because
pressure
relief
decreases
with
increasing
well
spacing,
injectivity
ratio
also
decreases
with
well
spacing
(Fig.
5b).
Injectiv-
ity
ratio
is
seen
to
be
relatively
insensitive
to
CO2injection
rate,
as
evidenced
by
similar
injectivity
ratio
versus
well
spacing
curves
for
CO2injection
rates
of
3.8
and
7.6
million/year.
Because
Ppeak is
more
sensitive
to
heterogeneity
for
the
ACRM
cases
than
for
the
no-
production
cases,
injectivity
ratio
is
less
in
the
heterogeneous
cases.
It
is
worth
noting
that
if
the
average
horizontal
permeability
been
kept
fixed
between
the
homogeneous
and
heterogeneous
cases,
it
is
likely
that
the
injectivity
ratio
would
not
have
been
reduced
by
us
much
as
a
factor
of
2
in
the
heterogeneous
case.
Therefore,
what
is
actually
being
exhibited
is
that
injectivity
ratio
decreases
with

Author's personal copy
238
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
Fig.
4.
Aqueous-phase
CO2-concentration
contours
show
CO2-plume
migration
driven
by
CO2injection
from
a
horizontal
well
for
no
brine
production
(a,
c,
and
e),
and
for
brine
produced
in
a
horizontal
well
10
km
downdip
from
the
injection
well
(b,
d,
and
f).
Storage-formation-permeability
cases
are
(1)
homogeneous
(a
and
b),
(2)
layered
heterogeneity
with
a
permeability
contrast
of
10
(c
and
d),
and
(3)
layered
heterogeneity
with
permeability
contrast
of
100
(e
and
f).
The
heterogeneous
cases
have
alternating
40-m-thick
layers
of
high
and
low
permeability.
The
horizontal
injectors
and
producers
are
located
in
the
lower
20
m
of
the
storage
formation.
The
formation
dip
angle
is
5.7◦and
the
vertical
scale
in
the
plot
is
exaggerated
by
a
factor
of
5.
decreasing
permeability,
not
necessarily
by
virtue
of
the
existence
of
heterogeneity.
3.2.3.
CO2breakthrough
at
brine
producers
As
expected,
CO2breakthrough
time
increases
with
well
spac-
ing
(Fig.
6a).
Because
of
the
large
thickness
of
the
storage
formation
(400
m)
and
the
CO2injector
and
brine
producer
being
at
the
bottom
of
the
storage
formation,
the
slope
of
the
CO2breakthrough
time
versus
well
spacing
curve
is
less
than
1
for
smaller
well
spac-
ing,
increasing
to
a
slope
of
nearly
1
for
larger
well
spacing.
The
CO2
plume
rises
to,
and
flows
along,
the
top
of
the
storage
formation,
until
it
overlies
the
brine
producer,
where
it
is
pulled
downward
to
the
brine
producer.
For
smaller
well
spacing,
the
vertical
travel
dis-
tance
is
a
significant
portion
of
the
overall
travel
distance
between

Author's personal copy
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
239
Fig.
5.
(a)
Peak
pressure
buildup
in
the
storage
formation
adjacent
to
the
CO2injector
versus
CO2injection
rate
is
plotted
for
horizontal-well
cases
with
no
brine
production
and
with
brine
produced
from
producers
spaced
5,
10,
15,
and
20
km
from
the
injector.
The
storage
formation
is
level,
400
m
thick,
overlain
by
a
400-m
thick
caprock.
Heterogeneous
cases
have
alternating
40-m-thick
layers
of
high
and
low
permeability,
with
a
permeability
contrast
of
10.
(b)
Injectivity
ratio
is
plotted
versus
well
spacing
between
injector/producer
pairs
for
CO2-injection
rates
of
3.8
and
7.6
million
tons/year.
The
injection
period
is
30
years
and
storage-formation
area
is
160
km2.
Table
3
Pressure
buildup
in
the
storage
formation
adjacent
to
the
CO2injector
is
listed
for
cases
with
no
brine
production
and
with
brine
production
wells
at
the
indicated
spacing
for
a
400-m-thick
storage-formation
with
an
area
of
160
km2.
CO2injection
rate
of
3.8
million
tons/year.
For
the
cases
with
brine
production,
peak
pressure
buildup
occurs
at
the
beginning
of
the
injection
period.
Storage-formation
permeability
distribution
Formation
dip
angle
(degrees)
No
brine
production
pressure
buildup
(MPa)
at
indicated
time
Brine
production
at
the
indicated
well
spacing
peak
pressure
buildup
(MPa)
30
yr 50
yr
75
yr
100
yr
5
km
10
km
15
km
20
km
Homogeneous
0
16.8
25.1
34.5
43.2
1.0
2.1
2.7
3.9
5.7
16.6
24.8
34.7
44.0
0.9
1.9
2.5
3.5
Layered
heterogeneous
with
10:1
contrast
0
17.5
25.6
35.1
43.9
2.1
3.7
5.0
7.1
5.7
19.9
27.8
37.5
46.9
2.6
4.9
6.4
8.9
Layered
heterogeneous
with
100:1
contrast 0 18.8
27.0
36.6
45.5
3.4
4.6
6.0
9.7
5.7
21.8
29.8
39.2
47.9
4.2
6.8
8.9
11.9
the
CO2injector
and
brine
producer.
For
larger
well
spacing,
the
vertical
distance
is
a
smaller
portion
of
the
overall
distance.
To
first
order,
the
CO2travel
distance
is
5.8,
10.8,
15.8,
and
20.8
km
for
the
5-,
10-,
15-,
and
20-km
well
spacings,
respectively.
For
a
level
formation,
heterogeneity
causes
preferential
flow
of
CO2that
reduces
CO2breakthrough
time.
For
10-km
well
spacing
and
a
CO2injection
rate
of
3.8
million
tons/year,
CO2
breakthrough
occurs
at
45
and
30
years
for
the
homogeneous
and
heterogeneous
cases,
respectively,
while
for
an
injection
rate
of
7.6
million
tons/year,
CO2breakthrough
occurs
at
19
and
18
years.
Heterogeneity
can
have
the
opposite
influence
on
CO2break-
through
time,
depending
on
formation
dip
and
where
the
brine
producer
is
located,
relative
to
the
CO2injector.
Compared
to
level
placement
in
a
level
storage
formation,
placing
a
brine
pro-
ducer
downdip
of
the
CO2injector
can
increase
CO2breakthrough
time,
particularly
for
heterogeneous
storage-formation
permeabil-
ity
(Fig.
6b).
Therefore,
it
is
possible
to
take
advantage
of
the
Fig.
6.
(a)
CO2breakthrough
time
is
plotted
versus
well
spacing
between
horizontal
CO2injectors
and
brine
producers
for
the
set
of
ACRM
cases
plotted
in
Fig.
5b.
(b)
The
influence
of
formation
dip
on
CO2breakthrough
time
is
shown
for
a
CO2-injection
rate
of
3.8
million
tons/year.
Heterogeneous
cases
have
alternating
40-m-thick
layers
of
high
and
low
permeability,
with
a
permeability
contrast
of
10.
The
storage
formation
is
400
m
thick
with
an
area
of
160
km2and
is
overlain
by
a
400-m-thick
caprock.

Author's personal copy
240
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
Fig.
7.
(a)
Injectivity
ratio
and
(b)
CO2breakthrough
time
are
plotted
versus
well
spacing
between
horizontal
injector/producer
pairs
for
storage-formation
thicknesses
of
50,
100,
200
and
400
m.
The
storage
formation
has
homogeneous
permeability,
the
caprock
is
400
m
thick,
the
injection
period
is
30
years,
and
storage-formation
area
is
160
km2.
influence
of
buoyancy
flow
with
respect
to
CO2breakthrough.
The
beneficial
influence
of
buoyancy
on
delaying
CO2breakthrough
increases
with
dip
angle
and
with
permeability
contrast
(compare
Fig.
3d
and
f).
3.2.4.
Influence
of
storage-formation
thickness
and
area
The
previous
discussion
pertains
to
a
relatively
thick
(400
m)
storage
formation
and
a
relatively
small
storage-formation
area
(160
km2).
Therefore,
we
investigated
the
influence
of
storage-
formation
thickness
and
area
on
Ppeak and
injectivity,
including
storage-formation
thicknesses
of
50,
100,
200
and
400
m
and
storage-formation
areas
of
160,
1600,
and
16,000
km2.
Injectiv-
ity
ratio
decreases
with
decreasing
storage-formation
thickness
(Fig.
7a).
Reducing
storage-formation
thickness
from
400
to
200
m
has
a
relatively
small
effect
on
injectivity
ratio.
CO2breakthrough
time
is
reduced
nearly
linearly
with
storage-formation
thickness
for
larger
thicknesses
(Fig.
7b).
For
thinner
storage
formations,
the
decrease
in
CO2breakthrough
time
is
slightly
less
than
linear
because
the
CO2plume
occupies
a
greater
portion
of
the
storage
formation
for
thin
formations
than
for
thick
formations.
We
decided
to
increase
the
storage
formation
area
by
factors
of
10
and
100
to
see
when
the
influence
of
increased
area
has
a
diminishing
effect
on
pressure
buildup
and
injectivity
ratio.
In
other
words,
we
wanted
to
establish
when
the
formation
was
effec-
tively
infinite
in
areal
extent.
We
found
that
factors
of
10
and
100
yielded
the
same
Ppeak for
the
no-production
case
(Fig.
8a);
thus,
a
storage-formation
area
of
1600
km2is
effectively
infinite
in
areal
extent
for
this
problem.
We
also
found
that
increasing
the
storage-
formation
area
by
a
factor
of
either
10
or
100
reduces
Ppeak around
the
CO2injector
by
30–40%
(with
this
effect
increasing
with
CO2
injection
rate)
for
the
no-production
case
(Fig.
8a).
This
is
expected
because
there
is
considerably
greater
area
for
pressure
buildup
to
be
dissipated
through
the
caprock
and
greater
storage-formation
volume
over
which
fluid
compression
can
occur.
For
ACRM
cases,
Ppeak is
insensitive
to
storage-formation
area
for
well
spacings
of
5,
10,
and
15
km
and
slightly
sensitive
for
20-km
well
spac-
ing
(Fig.
8a).
Accordingly,
increasing
the
storage-formation
area
by
a
factor
of
10
has
the
effect
of
reducing
injectivity
ratio
by
about
30–40%
(with
this
effect
increasing
with
CO2injection
rate)
(Fig.
8b).
Injectivity
ratios
are
still
much
greater
than
2
for
well
spacings
of
10
km
or
less
and
are
greater
than
2
for
well
spacings
of
15
and
20
km.
Because
storage-formation
area
does
not
influence
CO2breakthrough
time,
it
was
not
necessary
to
include
plots
of
that
influence.
3.2.5.
Scalability
of
CO2storage
with
ACRM
As
discussed
earlier,
the
definition
of
injectivity
ratio
used
in
the
horizontal-well
study
is
based
on
Ppeak,
rather
than
P
as
a
function
of
time.
Because
of
when
Ppeak occurs
for
cases
with
and
without
brine
production,
this
definition
does
not
fully
quantify
the
beneficial
influence
of
brine
production
on
pressure
relief.
Ppeak
occurs
relatively
early
during
the
injection
period
for
cases
with
brine
production
(Fig.
9).
Without
brine
production,
P
increases
throughout
the
injection
period,
reaching
its
peak
at
the
end
of
injection
(Fig.
9);
P
increases
nearly
linearly
with
time
for
a
rel-
atively
small
storage-formation
area
(Fig.
9a),
while
the
rate
of
increase
of
P
decreases
with
time
for
a
storage
formation
of
very
large
(effectively
infinite)
areal
extent
(Fig.
9b).
After
Ppeak occurs,
the
influence
of
brine
production
on
pres-
sure
relief
continues
to
increase
with
time.
For
the
160-km2storage
formation,
this
reduces
P
to
zero
at
7,
10,
and
25
years
for
well
spacings
of
5,
10,
and
15
km,
respectively
(Fig.
9a).
For
the
1600-km2
case,
P
is
reduced
to
zero
at
8
and
28
years
for
well
spacings
of
5
and
10
km,
respectively
(Fig.
9b).
The
difference
between
the
small
and
large
storage
formation
indicates
the
influence
of
pressure
relief
being
stronger
in
a
more
areally
confined
storage
formation.
Because
Ppeak occurs
relatively
early
during
injection,
Ppeak is
insensitive
to
storage-formation
area.
The
strong
influence
of
brine
production
on
pressure
relief
indicates
that,
after
Ppeak is
attained,
it
should
be
possible
to
continuously
ramp
up
the
CO2injection
rate
while
remaining
just
below
a
target
value
of
P,
as
done
in
the
double-ring
9-spot
pressure-management
example
discussed
earlier
(Fig.
2).
This
demonstrates
the
inherent
scalability
of
ACRM
and
has
useful
implications
on
the
feasibility
of
large-scale
imple-
mentation
of
CCS.
Within
a
region,
as
more
CCS
emitters
become
equipped
to
capture
CO2,
it
would
be
possible
to
add
their
CO2
output
to
an
existing
CO2-storage
operation
that
utilized
ACRM.
We
also
considered
caprock
thicknesses
of
50,
100,
and
200
m,
in
addition
to
the
cases
discussed
above
that
had
a
400-m-thick
caprock.
For
this
range
of
caprock
thickness,
we
found
that
Ppeak,
injectivity
ratio,
and
CO2breakthrough
time
are
insensitive
to
caprock
thickness.
Thus,
it
is
unnecessary
to
include
any
plots
of
that
influence.
3.2.6.
Impact
of
multiple
horizontal
brine
production
wells
In
this
study
we
limited
ourselves
a
single
horizontal
brine
pro-
duction
well.
Consideration
of
Figs.
5
and
7a,
Figs.
8
and
9
indicate
the
advantage
of
successively
producing
brine
from
more
than
one
horizontal
well.
For
a
400-m-thick
formation
and
an
injection
rate

Author's personal copy
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
241
Fig.
8.
(a)
Peak
pressure
buildup
in
the
storage
formation
adjacent
to
the
CO2injector
versus
CO2injection
rate
is
plotted
for
horizontal-well
cases
with
no
brine
production
and
with
brine
produced
from
horizontal
wells
spaced
5,
10,
15,
and
20
km
from
the
injector
for
storage-formation
areas
of
160,
1600,
and
16,000
km2.
The
storage
formation
is
level,
200
m
thick,
with
homogeneous
permeability,
and
overlain
by
a
400-m
thick
caprock.
(b)
Injectivity
ratio
is
plotted
versus
well
spacing
between
injector/producer
pairs
for
CO2-injection
rates
of
3.8
and
7.6
million
tons/year.
The
injection
period
is
30
years.
Curves
for
storage-formation
areas
of
1600
and
16,000
km2are
the
same.
of
3.8
million
tons/year,
brine
production
from
a
well
5
km
from
the
CO2injector
could
increase
injectivity
by
a
factor
of
at
least
10
(Fig.
5b).
As
CO2approached
the
brine
producer,
at
some
time
greater
than
20
years
(Fig.
6b),
brine
production
could
be
gradually
shifted
to
a
well
20
km
from
the
CO2injector,
providing
the
same
degree
of
pressure
relief
achieved
at
early
time
from
the
producer
at
5
km
(Fig.
9),
while
delaying
CO2breakthrough
to
60
years
or
more
(Fig.
6b).
By
using
two
or
more
horizontal
brine
producers
it
could
be
feasible
to
sustain
an
injectivity
ratio
of
10
or
more,
and
a
tenfold
increase
or
greater
in
storage
capacity.
The
same
principal
could
be
also
applied
to
thinner
storage
formations.
3.3.
Grid-sensitivity
study
Grid-sensitivity
analyses
were
conducted
for
the
horizontal-
well
cases
for
a
400-m-thick
storage
formation
with
zero
dip.
Compared
to
the
base
mesh
used
in
this
study,
the
vertical
and
horizontal
grid
refinement
was
both
decreased
and
increased
by
a
factor
of
two,
with
grid
spacing
of
400-m
horizontally
by
40-m
vertically
for
the
coarse
mesh
and
100-m
horizontally
by
10-m
ver-
tically
for
the
fine
mesh,
versus
200-m
by
20-m
for
the
base
mesh.
Grid
refinement
has
a
negligible
influence
on
pressure
buildup
and
CO2breakthrough
time
(Table
4)
for
the
homogeneous
case
and
10:1
heterogeneous
case,
demonstrating
the
insensitivity
of
our
simulations
to
the
numerical
refinement
used
in
this
study.
For
the
100:1
heterogeneous
case,
which
was
only
considered
in
the
dipping
formations,
the
coarse
mesh
yields
a
greater
CO2break-
through
time,
while
the
base
and
fine
meshes
produced
similar
CO2breakthrough
times.
3.4.
Parameter-sensitivity
study
Sensitivity
analyses
of
key
hydrologic
parameters
were
con-
ducted
for
horizontal-well
cases
for
a
400-m-thick
storage
formation
with
zero
dip
(Table
3).
We
start
with
parameters
having
a
negligible
influence
on
pressure
buildup
(P)
and
CO2
Fig.
9.
Pressure
buildup
history
in
the
storage
formation
adjacent
to
the
CO2injector
is
plotted
for
horizontal-well
cases
with
no
brine
production
and
with
brine
produced
from
horizontal
wells
spaced
5,
10,
15,
and
20
km
from
the
CO2injector
for
a
storage-formation
area
of
(a)
160
km2and
(b)
1600
km2.
The
storage
formation
is
level
400
m
thick,
with
homogeneous
permeability,
overlain
by
a
400-m-thick
caprock.
The
injection
period
is
30
years.

Author's personal copy
242
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
Table
4
CO2breakthrough
time
and
pressure
buildup
in
the
storage
formation
adjacent
to
the
horizontal
CO2injector
are
listed
for
the
coarse,
standard,
and
fine
numerical
mesh
for
the
base
case
with
a
400-m-thick
storage
formation
with
an
area
of
160
km2and
400-m-thick
caprock.
Dependent
variable
Well
spacing
(km)
Homogeneous
10:1
Heterogeneous
100:1
Heterogeneous
Coarse
mesh
Base
mesh
Fine
mesh
Coarse
mesh
Base
mesh
Fine
mesh
Coarse
mesh
Base
mesh
Fine
mesh
CO2breakthrough
time
(yr)
10
45
45
46
32
30
30
52
42
38
No
production NA NA NA NA
NA
NA
NA
NA
NA
Peak
pressure
buildup
(MPa)
10
2.05
2.08
2.09
3.70
3.69
3.63
4.25
4.61
4.62
No
production
16.75
16.81
16.73
17.70
17.53
17.41
18.87
18.81
18.92
Table
5
CO2breakthrough
time
and
pressure
buildup
in
the
storage
formation
adjacent
to
the
horizontal
CO2injector
are
listed
for
a
range
of
storage-formation
van
Genuchten
˛.
The
storage
formation
is
400-m
thick,
with
an
area
of
160
km2,
overlain
by
a
400-m-thick
caprock.
The
base
case
is
shown
in
bold.
Dependent
variable
Well
spacing
(km)
van
Genuchten
˛
(Pa−1)
5.1
×
10−62.55
×
10−55.1
×
10−51.2
×
10−45.1
×
10−4
CO2breakthrough
time
(yr)
10
43
45
45
45
45
No
production
NA
NA
NA
NA
NA
Peak
pressure
buildup
(MPa)
10
2.13
2.09
2.08
2.08
2.07
No
production
16.97
16.84
16.81
16.79
16.78
Table
6
CO2breakthrough
time
and
pressure
buildup
in
the
storage
formation
adjacent
to
the
horizontal
CO2injector
are
listed
for
a
range
of
caprock-seal
van
Genuchten
˛.
The
storage
formation
is
400-m
thick,
with
an
area
of
160
km2,
overlain
by
a
400-m-thick
caprock.
The
base
case
is
shown
in
bold.
Dependent
variable
Well
spacing
(km)
van
Genuchten
˛
(Pa−1)
5.1
×
10−65.1
×
10−55.1
×
10−4
CO2breakthrough
time
(yr)
10
45
45
45
No
production
NA
NA
NA
Peak
pressure
buildup
(MPa) 10
2.08
2.08
2.08
No
production
16.81
16.81
16.81
Table
7
CO2breakthrough
time
and
pressure
buildup
in
the
storage
formation
adjacent
to
the
horizontal
CO2injector
are
listed
for
a
range
of
residual
supercritical
CO2saturation.
The
storage
formation
is
400-m
thick,
with
an
area
of
160
km2,
overlain
by
a
400-m-thick
caprock.
The
base-case
is
shown
in
bold.
Dependent
variable
Well
spacing
(km)
Residual
supercritical
CO2saturation
0.05 0.12 0.15
0.25
0.30
CO2breakthrough
time
(yr)
10
45
50
53
69
81
No
production NA
NA
NA
NA
NA
Peak
pressure
buildup
(MPa)
10
2.08
2.14
2.17
2.25
2.34
No
production
16.81
16.99
17.08
17.43
17.66
Table
8
CO2breakthrough
time
and
pressure
buildup
in
the
storage
formation
adjacent
to
the
horizontal
CO2injector
are
listed
for
a
range
of
storage-formation
porosity.
The
storage
formation
is
400-m
thick,
with
an
area
of
160
km2,
overlain
by
a
400-m-thick
caprock.
The
base-case
is
shown
in
bold.
Dependent
variable
Well
spacing
(km)
Storage-formation
porosity
0.06
0.12
0.24
CO2breakthrough
time
(yr)
10
45
45
45
No
production
NA
NA
NA
Peak
pressure
buildup
(MPa)
10
2.08
2.08
2.08
No
production
16.81
16.81
16.81
Table
9
CO2breakthrough
time
and
pressure
buildup
in
the
storage
formation
adjacent
to
the
horizontal
CO2injector
are
listed
for
a
range
of
storage-formation
permeability.
The
storage
formation
is
400-m
thick,
with
an
area
of
160
km2,
overlain
by
a
400-m-thick
caprock.
The
base
case
is
shown
in
bold.
Dependent
variable
Well
spacing
(km)
Storage-formation
permeability
(m2)
1.0
×
10−14 5.0
×
10−14 1.0
×
10−13 2.0
×
10−13 1.0
×
10−12
CO2breakthrough
time
(yr)
5
32
30
32
55
246
10 67
40
45
71
250
No
production NA NA
NA
NA
NA
Peak
pressure
buildup
(MPa)
5
10.20
2.09
1.03
0.49
0.11
10 20.40
4.03
2.08
1.02
0.19
No
production
31.61
18.11
16.82
16.21
15.73

Author's personal copy
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
243
Table
10
CO2breakthrough
time
and
pressure
buildup
in
the
storage
formation
adjacent
to
the
horizontal
CO2injector
are
listed
for
a
range
of
caprock-seal
permeability.
The
storage
formation
is
400-m
thick,
with
an
area
of
160
km2,
overlain
by
a
400-m-thick
caprock.
The
base
case
is
shown
in
bold.
Dependent
variable
Well
spacing
(km)
Caprock-seal
permeability
(m2)
1.0
×
10−20 1.0
×
10−19 1.0
×
10−18 1.0
×
10−17 1.0
×
10−16
CO2breakthrough
time
(yr)
10
45
45
45
46
50
No
production
NA
NA
NA
NA
NA
Peak
pressure
buildup
(MPa) 10 2.11 2.11 2.08
2.06
1.93
No
production 19.57
18.80
16.82
12.42
4.16
breakthrough
time.
Tables
5
and
6
show
the
negligible
influence
of
van
Genuchten
˛
(van
Genuchten,
1980),
which
is
inversely
pro-
portional
to
air-entry
pressure
and
a
direct
measure
of
capillarity.
Table
7
indicates
that
residual
supercritical
CO2saturation
in
the
storage
formation
has
a
negligible
influence
on
pressure
buildup
and
that
the
value
used
in
this
study
(0.05)
conservatively
predicts
CO2breakthrough
time.
Table
8
indicates
the
negligible
influence
of
storage-formation
porosity.
The
most
sensitive
parameters
are
storage-formation
permeability,
followed
by
caprock-seal
perme-
ability.
With
no
brine
production,
P
increases
with
decreasing
log10 of
storage-formation
permeability,
while
it
increases
linearly
with
decreasing
storage-formation
permeability
for
cases
with
brine
production
(Table
9).
The
influence
of
storage-permeability
permeability
on
CO2breakthrough
time
is
related
to
its
influence
on
buoyancy.
The
base-case
value
of
storage-formation
permeabil-
ity
used
in
study
is
close
to
the
value
resulting
in
a
minimum
CO2breakthrough
time
and
is
therefore
conservative
in
that
sense.
Caprock-seal
permeability
has
a
negligible
influence
on
P
and
CO2breakthrough
time
for
cases
with
brine
production
(Table
10).
With
the
exception
of
very
high
permeability
values,
the
influ-
ence
of
caprock-seal
permeability
on
P
is
roughly
half
that
of
storage-formation
permeability
for
cases
with
no
brine
produc-
tion
(Tables
9
and
10).
Without
brine
production,
brine
leakage
into
the
caprock
can
provide
significant
pressure
relief,
while
for
cases
with
brine
production,
the
lack
of
sensitivity
to
caprock-seal
permeability
is
indicative
of
minimal
brine
leakage
(Table
10).
4.
Summary
and
conclusions
For
injection-only,
industrial-scale,
saline-formation
geologic
CO2storage,
pressure
buildup
can
limit
CO2storage
capacity
and
security.
Moreover,
water
demand
and
parasitic
energy
costs
asso-
ciated
with
CO2capture
and
storage
operations
are
large.
Active
CO2Reservoir
Management
(ACRM),
which
combines
brine
pro-
duction
with
CO2injection,
has
the
potential
of
addressing
these
challenges.
We
demonstrate
how
ACRM
can
potentially
enhance
reservoir
performance
in
two
important
ways.
First,
it
can
provide
more
secure
CO2storage
by
enabling
pressure
relief,
spatial
and
temporal
control
of
brine
migration,
and
CO2plume
manipulation.
Second,
it
can
result
in
more
cost-effective
CO2storage
by
improv-
ing
CO2injectivity
and
storage
capacity.
Brine
production
may
also
enable
development
of
utilization
options,
including
freshwa-
ter
production,
saline
cooling
water
for
power
plants,
geothermal
power,
and
make-up
water
for
oil,
gas,
and
geothermal
energy
pro-
duction.
These
options
are
important
to
the
economic
feasibility
of
ACRM
because
they
can
reduce
the
volume
of
brine
requiring
reinjection.
The
key
reservoir-performance
objective
for
ACRM
is
to
relieve
pressure
buildup
driven
by
CO2injection.
For
economic
and
oper-
ational
reasons,
it
is
important
to
delay
CO2breakthrough
at
brine
producers.
We
investigated
two
operational
strategies
for
balanc-
ing
these
objectives:
(1)
vertical
wells
with
multiple
rings
of
brine
producers
and
(2)
horizontal
injector/producer-well
pairs.
For
vertical
wells,
an
injection-only
strategy
was
compared
to
a
pressure-management
strategy
with
brine
production
from
a
double-ring
9-spot
pattern.
Except
for
early
time,
pressure
man-
agement
can
be
entirely
achieved
with
brine
production.
Because
pressure
relief
increases
with
time,
the
CO2injection
rate
can
be
ramped
up,
while
staying
within
a
pressure-buildup
target,
while
for
the
no-production
case,
injection
rates
must
be
gradually
decreased
to
stay
within
target.
Brine
production
causes
pres-
sure
buildup
to
abruptly
drop
to
zero
after
injection
ceases,
while,
without
brine
production,
pressure
buildup
can
persist
long
after
injection
ceases.
For
horizontal
wells,
we
find
that
layered
heterogeneous
per-
meability
in
the
storage
formation
causes
preferential
flow
of
CO2,
which
can
reduce
CO2breakthrough
time
at
brine
producers.
Brine
produced
downdip
of
CO2injection
can
strongly
influence
CO2
migration.
Without
brine
production,
buoyancy
drives
CO2updip,
unless
impeded
by
layered
heterogeneity.
With
brine
production,
the
combination
of
buoyancy
and
layered
heterogeneity
can
cause
CO2to
be
more
evenly
distributed
vertically
in
the
storage
for-
mation,
which
delays
CO2breakthrough
at
the
brine
producer.
Pressure
buildup
and
CO2breakthrough
time
are
found
to
be
sensi-
tive
to
storage-formation
permeability
and
insensitive
to
all
other
hydrologic
parameters
that
we
investigated,
with
the
exception
of
caprock-seal
permeability,
which
only
affects
pressure
buildup
for
cases
with
no
brine
production.
Without
brine
production,
brine
leakage
into
the
caprock
can
provide
significant
pressure
relief;
with
brine
production,
the
lack
of
sensitivity
to
caprock-seal
per-
meability
is
indicative
of
minimal
brine
leakage.
Brine
production
from
a
horizontal
well
can
strongly
relieve
pressure
buildup
at
a
horizontal
CO2injector,
which
improves
CO2
injectivity.
Pressure
relief
and
injectivity
improve
with
decreas-
ing
well
spacing,
while
CO2breakthrough
time
is
reduced.
When
injectors
and
producers
are
at
the
same
depth
in
non-dipping
for-
mations,
layered
heterogeneity
decreases
CO2breakthrough
time,
while
when
brine
is
produced
downdip
of
CO2injection,
het-
erogeneity
can
delay
CO2breakthrough.
Injectivity
ratio,
which
quantifies
the
improvement
to
injectivity
caused
by
brine
pro-
duction,
is
insensitive
to
CO2injection
rate,
and
is
somewhat
dependent
on
storage-formation
thickness.
The
benefit
of
pressure
relief
is
stronger
for
areally
smaller
storage
formations
than
for
larger
formations.
However,
injectivity
ratios
were
generally
quite
large
(and
always
greater
than
2),
even
if
the
storage
formation
is
of
infinite
areal
extent.
Because
an
injectivity
ratio
greater
than
2
decreases
the
total
required
number
of
wells,
our
results
indicate
that
brine
production
may
reduce
well
infrastructure
costs.
The
following
are
key
findings
from
our
reservoir
study.
•Pressure
management
without
brine
production
may
require
a
combination
of
(1)
CO2-injection-rate
reduction
with
time,
(2)
large
storage-formation
area,
and
(3)
large
spacing
from
adjacent
CCS
operations
or
other
subsurface
activities.
If
the
storage
for-
mation
is
not
large
enough
or
neighboring
subsurface
operations
are
too
close,
this
can
constrain
CO2injection
rates
and
storage
capacity.
•Pressure
management
with
brine
production
may
allow
a
large
increase
in
CO2injectivity
and
storage
capacity.
CO2injec-
tion
rate
is
not
constrained
by
storage-formation
area
or
by

Author's personal copy
244
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
proximity
to
neighboring
CCS
operations
or
other
subsurface
activities.
Minimized
pressure
and
fluid-migration
interactions
between
neighboring
CCS
operations
can
help
facilitate
inde-
pendent
planning,
assessment,
and
permitting
of
each
CCS
operation
within
a
basin.
Pressure
relief
increases
with
time,
which
allows
the
CO2injection
rate
to
be
continuously
increased.
Thus,
additional
CO2emitters
could
be
continuously
brought
online.
•Post-injection
pressure
buildup
persists
long
after
injection
ceases
for
CO2storage
without
brine
production,
while
with
brine
production
it
abruptly
drops
to
zero.
With
brine
produc-
tion,
the
only
post-injection
driving
force
is
buoyancy,
which
is
only
strong
enough
to
drive
CO2updip,
fully
within
the
stor-
age
formation.
Buoyancy
is
of
much
less
concern
(than
pressure
buildup)
for
diffuse
leakage
and
for
leakage
up
abandoned
wells
and
permeable
faults.
With
brine
production,
the
driving
force
for
post-injection
brine
migration
can
be
eliminated.
The
differ-
ence
in
post-injection
pressure
buildup
with
and
without
brine
production
could
affect
post-injection
monitoring
requirements
and
the
cost
of
liability
insurance.
•Control
of
CO2-plume
migration
with
vertical
brine
producers
may
require
too
many
wells
to
be
practical,
while
for
horizontal
wells
it
appears
to
be
quite
promising,
as
it
can
counteract
the
influence
of
buoyancy.
•Managing
the
tradeoff
between
pressure
relief/injectivity
improvement
and
delayed
CO2breakthrough
appears
to
be
fea-
sible
for
both
vertical
and
horizontal
wells.
The
use
of
horizontal
wells
appears
to
be
more
promising
with
respect
to
improving
injectivity
and
reducing
the
total
number
of
wells.
The
following
are
implications
and
recommendations
derived
from
our
study.
•Consideration
of
brine
utilization/disposition
options
could
play
a
role
in
the
site
selection
process.
The
feasibility
of
various
options
for
a
particular
site
depends
on
the
chemical
composi-
tion
and
temperature
of
the
produced
brine,
and
proximity
to
potential
markets.
Of
particular
interest
is
whether
salinity
in
the
storage
formation
allows
for
cost-effective
treatment,
which
is
important
to
the
economic
feasibility
of
ACRM
at
sites
where
using
brine
as
make-up
water
is
not
an
option.
•Greater
selectivity
in
choosing
CCS
sites
may
be
enabled
by
ACRM.
If
sites
are
found
where
brine
production
and
disposal
is
economically
feasible,
this
may
facilitate
a
large
increase
in
injectivity
and
storage
capacity,
reducing
the
number
of
CCS
sites
required
for
a
region,
which
allows
greater
selectivity
in
choosing
sites.
Larger
storage
capacity
also
enables
greater
leveraging
of
infrastructure
costs,
such
as
those
associated
with
siting,
permit-
ting,
and
monitoring
activities.
We
recommend
consideration
of
the
concept
of
“cherry
picking”,
first
searching
for
“upper
ech-
elon”
sites
where
CCS
may
be
deployed
at
reduced
cost
and
risk.
In
summary,
ACRM
provides
benefits
to
reservoir
management
at
the
cost
of
extracting
brine.
This
added
cost
must
be
offset
by
the
added
benefits
to
the
storage
operation
and/or
by
creating
new,
valuable
uses
of
brine
that
can
reduce
the
added
cost.
The
results
from
this
study
should
motivate
future,
detailed
studies
of
approaches
and
costs
that
will
answer
the
question
of
the
applica-
bility
of
ACRM
to
specific
CO2sequestration
situations.
Acknowledgements
This
work
was
sponsored
by
USDOE
Fossil
Energy,
National
Energy
Technology
Laboratory,
managed
by
Andrea
McNemar,
and
by
the
Carbon
Mitigation
Initiative
at
Princeton
University
and
by
the
Environmental
Protection
Agency
under
Cooperative
Agreement
RD-83438501.
The
authors
acknowledge
the
review
of
Pat
Berge
at
Lawrence
Livermore
National
Laboratory
(LLNL).
The
authors
also
appreciate
the
comments
from
the
reviewers,
which
helped
this
paper
become
a
more
thorough
and
comprehensive
treatment
of
this
subject.
This
work
was
performed
under
the
aus-
pices
of
the
U.S.
Department
of
Energy
by
LLNL
under
contract
DE-AC52-07NA27344.
References
Aines,
R.D.,
Wolery,
T.J.,
Bourcier,
W.L.,
Wolfe,
T.,
Haussmann,
C.W.,
2011.
Fresh
water
generation
from
aquifer-pressured
carbon
storage:
feasibility
of
treating
saline
formation
waters.
Energy
Procedia
4,
2269–2276.
Alhuthali,
A.H.,
Oyerinde,
D.,
Datta-Gupta,
A.,
2007.
Optimal
waterflood
manage-
ment
using
rate
control.
SPE
Reservoir
Evaluation
&
Engineering
10
(October
(4)),
539–551,
doi:10.2118/102478-PA.
ASME,
2006.
ASME
Steam
Tables
Compact
Edition.
ASME,
Three
Park
Avenue,
New
York,
NY,
USA.
Bachu,
S.,
2008.
CO2storage
in
geological
media:
role,
means,
status,
and
barriers
to
deployment.
Progress
in
Energy
and
Combustion
Science
34,
254–273.
Bergmo,
P.E.S.,
Grimstad,
A.-A.,
Lindberg,
E.,
2011.
Simultaneous
CO2injection
and
water
production
to
optimize
aquifer
storage
capacity.
International
Journal
of
Greenhouse
Gas
Control
5,
555–564.
Birkholzer,
J.T.,
Zhou,
Q.,
2009.
Basin-scale
hydrogeologic
impacts
of
CO2storage:
capacity
and
regulatory
implications.
International
Journal
of
Greenhouse
Gas
Control
3,
745–756.
Brouwer,
D.R.,
Jansen,
J.D.,
2004.
Dynamic
optimization
of
water-flooding
with
smart
wells
using
optimal
control
theory.
SPEJ
– Society
of
Petroleum
Engineers
Journal
9
(4),
391–402,
doi:10.2118/78278-PA,
SPE:
78278-PA.
Brouwer,
D.R.,
Jansen,
J.D.,
van
der
Starre,
S.,
van
Kruijsdkijk,
C.P.J.W.,
Berentsen,
C.W.J.,
2001.
Recovery
increase
through
waterflooding
with
smart
well
tech-
nology.
In:
Paper
SPE
68979
Presented
at
the
SPE
European
Formation
Damage
Conference,
The
Haque,
21–22
May,
doi:10.2118/68979-MS.
Bourcier,
W.L.,
Wolery,
T.J.,
Wolfe,
T.,
Haussmann,
C.,
Buscheck,
T.A.,
Aines,
R.D.,
2011.
A
preliminary
cost
and
engineering
estimate
for
desalinating
produced
forma-
tion
water
associated
with
carbon
dioxide
capture
and
storage.
International
Journal
of
Greenhouse
Gas
Control
5,
1319–1328.
Breit,
G.N.,
2002.
Produced
Waters
Database:
U.S.
Geological
Survey
Provisional
Database,
Available
from:
http://energy.cr.usgs.gov/prov/prodwat/index.htm.
Buscheck,
T.A.,
Glascoe,
L.G.,
Lee,
K.H.,
Gansemer,
J.,
Sun,
Y.,
Mansoor,
K.,
2003.
Validation
of
the
multiscale
thermohydrologic
model
used
for
analysis
of
a
pro-
posed
repository
at
Yucca
Mountain.
Journal
of
Contaminant
Hydrology
62
(3),
421–440.
Buscheck,
T.A.,
2010.
Active
management
of
integrated
geothermal-CO2-storage
reservoirs
in
sedimentary
formations:
an
approach
to
improve
energy
recov-
ery
and
mitigate
risk.
In:
Proposal
in
Response
to
DE
FOA-0000336:
Energy
Production
with
Innovative
Methods
of
Geothermal
Heat
Recovery,
LLNL-PROP-
463758.
Buscheck,
T.A.,
Sun,
Y.,
Wolery,
T.J.,
Court,
B.,
Celia,
M.A.,
Friedmann,
S.J.,
Aines,
R.D.,
2011a.
Active
CO2reservoir
management
in
saline
formations:
utilizing
syner-
gistic
opportunities
to
increase
storage
capacity,
reduce
cost,
and
mitigate
risk.
In:
Presentation
at
the
SPE
International
Workshop
on
Carbon
Capture,
Utiliza-
tion,
and
Storage
(CCUS):
Environment,
Energy
Security,
and
Opportunities
for
the
Middle
East,
Abu
Dhabi,
UAE,
7–9
March.
Buscheck,
T.A.,
Sun,
Y.,
Hao,
Y.,
Chen,
M.,
Court,
B.,
Celia,
M.A.,
Wolery,
T.J.,
2011b.
Geothermal
energy
production
from
actively-managed
CO2storage
in
saline
formations.
In:
Proceedings
for
the
Geothermal
Resources
Council
35th
Annual
Meeting,
San
Diego,
CA,
USA,
23–26
October
2011.
Buscheck,
T.A.,
Sun,
Y.,
Hao,
Y.,
Wolery,
T.J.,
Bourcier,
W.L.,
Tompson,
A.F.B.,
Jones,
E.D.,
Friedmann,
S.J.,
Aines,
R.D.,
2011c.
Combining
brine
extraction,
desali-
nation,
and
residual-brine
reinjection
with
CO2storage
in
saline
formations:
implications
for
pressure
management,
capacity,
and
risk
mitigation.
Energy
Procedia
4,
4283–4290.
Carroll,
S.,
Hao,
Y.,
Aines,
R.D.,
2008.
Transport
and
detection
of
carbon
dioxide
in
dilute
aquifers.
In:
Proceedings
for
the
9th
International
Conference
on
Green-
house
Gas
Control
Technologies,
Washington,
DC,
USA,
16–20
November
2008.
Carroll,
S.A.,
Hao,
Y.,
Aines,
R.D.,
2009.
Geochemical
detection
of
carbon
dioxide
in
dilute
aquifers.
Geochemical
Transactions
10,
4.
Court,
B.,
2011.
Safety
and
Water
Challenges
in
CCS:
Modeling
Studies
to
Quantify
CO2and
Brine
Leakage
Risk
and
Evaluate
Promising
Synergies
for
Active
and
Integrated
Water
Management,
Ph.D.
Dissertation,
Princeton
Uni-
versity,
Princeton,
NJ,
available
at
http://ddspace.princeton.edu/jspui/handle/
88435/dsp01ms35t861f.
Court,
B.,
Elliot,
T.R.,
Dammel,
J.,
Buscheck,
T.A.,
Rohmer,
J.,
Celia,
M.A.,
2011a.
Promis-
ing
synergies
to
address
water,
sequestration,
legal,
and
public
acceptance
issues
associated
with
large-scale
implementation
of
CO2sequestration,
Special
Issue
on
Carbon
Capture
and
Storage
(CCS)
“Five
years
after
the
IPCC
Special
Report
on
CCS:
state
of
play”.
Mitigation
and
Adaptation
of
Strategies
for
Global
Change
Journal,
doi:10
1007/s11027-011-9314-x.

Author's personal copy
T.A.
Buscheck
et
al.
/
International
Journal
of
Greenhouse
Gas
Control
6
(2012)
230–245
245
Court,
B.,
Celia,
M.A.,
Nordbotten,
J.M.,
Elliot,
T.R.,
2011b.
Active
and
integrated
management
of
water
resources
throughout
CO2capture
and
sequestration
operations.
Energy
Procedia
4,
4221–4229.
Duke,
D.,
2007.
ZLD:
new
silica
based
inhibitor
chemistry
permits
cost
effective
water
conservation
for
HVAC
and
industrial
cooling
towers.
In:
IWC
Report
07-11,
Proceedings
for
the
69th
Annual
International
Water
Conference,
San
Antonio,
Texas,
October
26–29,
2008.
Fenghour,
A.,
Wakeham,
W.A.,
Vesovic,
V.,
1998.
The
viscosity
of
carbon
dioxide.
Journal
of
Physical
and
Chemical
Reference
Data
27
(1),
31–44.
Harto,
C.B.,
Veil,
J.A.,
2011.
Management
of
Water
Extracted
from
Carbon
Sequestra-
tion
Projects.
Argonne
National
Laboratory,
ANL/EVS/R-11/1.
Hassanzadeh,
H.,
Pooladi-Darvish,
M.,
Keith,
D.,
2009.
Accelerating
CO2dissolution
in
saline
aquifers
for
geological
storage:
mechanistic
and
sensitivity
studies.
Energy
and
Fuels
23,
3328–3336.
IPCC
(Intergovernmental
Panel
on
Climate
Change),
2005.
IPCC
Special
Report
on
Carbon
Dioxide
Capture
and
Storage.
Cambridge
University
Press,
New
York.
IEA
International
Energy
Agency,
2007.
Greenhouse
Gas
R&D
Programme.,
http://www.
ieagreen.
org.
uk/.
IEA
International
Energy
Agency,
2009.
How
the
energy
sector
can
deliver
on
a
cli-
mate
agreement
in
Copenhagen.
In:
Special
Early
Excerpt
of
the
World
Energy
Outlook
2009
for
the
Bangkok
UNFCCC
Meeting,
http://www.iea.org/weo/docs/
weo2009/climate
change
excerpt.pdf.
Johnson,
J.W.,
Nitao,
J.J.,
Knauss,
K.G.,
2004a.
Reactive
transport
modeling
of
CO2stor-
age
in
saline
aquifers
to
elucidate
fundamental
processes,
trapping
mechanisms
and
sequestration
partitioning.
In:
Baines,
S.J.,
Worden,
R.H.
(Eds.),
Geological
Storage
of
Carbon
Dioxide.
Special
Publications,
vol.
223.
Geological
Society,
London,
pp.
107–128.
Johnson,
J.W.,
Nitao,
J.J.,
Morris,
J.P.,
2004b.
Reactive
transport
modeling
of
cap
rock
integrity
during
natural
and
engineered
CO2storage.
In:
Benson,
S.
(Ed.),
CO2
Capture
Project
Summary,
vol.
2.
Elsevier.
Keith,
D.,
Hassanzadeh,
H.,
Pooladi-Darvish,
M.,
2004.
Reservoir
engineering
to
accelerate
dissolution
of
stored
CO2in
brines.
In:
Gale,
J.,
Wilson,
M.
(Eds.),
Proceedings
for
the
7th
International
Conference
on
Greenhouse
Gas
Control
Technologies.
Elsevier.
Maulbetsch,
J.S.,
DiFilippo,
M.N.,
2010.
Performance,
cost,
and
environmental
effects
of
saltwater
cooling
towers.
In:
California
Energy
Commission,
PIER
Energy-
Related
Environmental
Research
Program.
CEC-500-2008-043,
http://www.
energy.ca.gov/2008publications/CEC-500-2008-043/CEC-500-2008-043.pdf.
MIT,
2006.
The
Future
of
Geothermal
Energy:
Impact
of
Enhanced
Geother-
mal
Energy
Systems
(EGS)
in
the
United
States
in
the
21st
Century.,
http://www1.eere.energy.gov/geothermal/egs
technology.html.
Morris,
J.P.,
Detwiler,
R.L.,
Friedmann,
S.J.,
Vorobiev,
O.Y.,
Hao,
Y.,
2011.
The
large-
scale
geomechanical
and
hydrological
effects
of
multiple
CO2injection
sites
on
formation
stability.
International
Journal
of
Greenhouse
Gas
Control
5,
69–74.
Nitao,
J.J.,
1998.
Reference
Manual
for
the
NUFT
Flow
and
Transport
Code,
Version
3.0.
Lawrence
Livermore
National
Laboratory,
UCRL-MA-130651-REV-1.
Rutqvist,
J.,
Birkholzer,
J.,
Cappa,
F.,
Tsang,
C.-F.,
2007.
Estimating
maximum
sustain-
able
injection
pressure
during
geological
sequestration
of
CO2using
coupled
fluid
flow
and
geomechanical
fault-slip
analysis.
Energy
Conversion
and
Man-
agement
48,
1798–1807.
Sanz,
M.A.,
Stover,
R.L.,
2007.
Low
energy
consumption
in
the
Perth
seawater
desali-
nation
plant.
In:
Presented
at
IDA
World
Congress,
Maspalomas,
Gran
Canaria,
Spain,
21–26,
2007.
Span,
R.,
Wagner,
W.,
1996.
A
new
equation
of
state
for
carbon
dioxide
covering
the
fluid
region
from
the
triple-point
temperature
to
1100
K
at
pres-
sures
up
to
800
MPa.
Journal
of
Physical
and
Chemical
Reference
Data
25,
1509–1596.
Sudaryanto,
B.,
Yortsos,
Y.C.,
2001.
Optimization
of
displacements
in
porous
media
using
rate
control.
In:
Paper
SPE
71509
Presented
at
the
SPE
Annual
Technical
Conference
and
Exhibition,
New
Orleans,
30
September–3
October,
doi:10.2118/71509-MS.
Surdam,
R.C.,
Jiao,
Z.,
Stauffer,
P.,
Miller,
T.,
2009.
An
Integrated
Strategy
for
Carbon
Management
Combining
Geological
CO2Sequestration,
Displaced
Fluid
Produc-
tion
and
Water
Treatment.
Wyoming
State
Geological
Survey,
Challenges
in
Geologic
Resource
Development
No.
8.
van
Genuchten,
M.T.,
1980.
A
closed
form
equation
for
predicting
the
hydraulic
conductivity
of
unsaturated
soils.
Soil
Science
Society
of
America
Journal
44,
892–898.
Veil,
J.A.,
Puder,
M.G.,
Elcock,
D.,
Redwick
Jr.,
R.J.,
2004.
A
White
Paper
Describing
Produced
Water
from
Production
of
Crude
Oil,
Natural
Gas,
and
Coal
Bed
Methane.
US
DOE
National
Energy
Technology
Laboratory
Report,
http://www.evs.anl.gov/pub/doc/ProducedWatersWP0401.pdf.
Zhou,
Q.,
Birkholzer,
J.T.,
Tsang,
C.-F.,
Rutqvist,
J.A.,
2008.
A
method
for
quick
assess-
ment
of
CO2storage
capacity
in
closed
and
semi-closed
saline
formations.
International
Journal
of
Greenhouse
Gas
Control
2,
626–639.