MBDCI

Stress Changes during Production

4-E Stress Changes During Production

Maurice Dusseault
 MBDCI

Production Engineering

Typical V processes in production

 Pressure depletion leads to shrinkage
 High p injection leads to expansion
4-E Stress Changes During Production

 Cooling leads to thermal shrinkage

 Heating leads to thermal expansion

 Solids production is a -V process

 Solids injection is a +V process

To simulate, predict and manage, we must

understand and quantify these

Volume changes drive stress changes!
 MBDCI

Example of Redistribution

v, section A-A
v

A must be
always constant
gain gain
4-E Stress Changes During Production

loss
B p
h

A zone A
of -p*
p
h,
B
*-p causes -V B-B
 MBDCI

p-Induced Stress Changes

V = -Cc (linear approximation)

Can we assume that p = ?
 This is the widely used Terzaghi assumption
 It is not valid in production geomechanics (see
4-E Stress Changes During Production

more detain in well test geomechanics section)

What happens in a reservoir?
 p leads to some V, therefore some
 But, V leads to redistribution of v

 Thus, v is not constant, hence p

Coupled modeling is required to simulate

 MBDCI

Non-Linear Stress Changes

From elasticity hmin in the reservoir
1 2
h = p
1
= Poissons ratio
4-E Stress Changes During Production

However, reservoirs
measured
behave differently
 Geological history theory

 Stress conditions

 Porosity, rock type (example only)



Good data now exist pressure - p

 MBDCI

Reservoir Stresses - Depletion

Depletion changes the stresses

This means that the following things
happen:
 The fracture gradient changes
4-E Stress Changes During Production

 The stiffness of the rock can change

 The drillability of the rock may change (i.e.

the strength can increase substantially)

Also, in drilling into and through a depleted
reservoir These effects can
be remarkably
 Blowout and LC risks increase costly if they
 Extra casing strings often used are unanticipated!
 MBDCI

Issues in Reservoir Stress Path

What do we mean by the stress path

How do we measure the stress path?
 Hydraulic fracture tests, including MiniFrac
 LOT, XLOT tests
4-E Stress Changes During Production

 Step-rate injection tests

Can we predict the stress path in advance?

 Theoreticalapproaches with mathematical models
 Empirical methods, measurements, then models

 Comparing to existing data bases (analogues)

 MBDCI

Stress Definitions
We need 3 y 3 a = 1
ppl stresses, Principal
max planes
3 directions, x Stresses
2
& one pore
pressure: 1
1 r = 3 r
seven
4-E Stress Changes During Production

parameters 2
3
z 1 > 2 > 3 Triaxial
Test a
We usually Stresses
assume v is a y
principal stress
x
v r
HMAX > hmin

HMAX r
ri
hmin z In Situ Borehole
Stresses Stresses
 MBDCI

Depletion Stress Path

Depletion has two major effects
 Total lateral stresses drop (hmin & HMAX)
 Effective stresses rise (v, hmin, HMAX )

This results in the following

4-E Stress Changes During Production

 PFdrops in the depleted zone

 Rock is stronger (confining stress increases)

 Rock compacts (sometimes, a lot e.g.: Chalk)

Consequences
 Slower drilling, incorrect bit choice in zone,
well control risks (blowout, LC), more casing
strings, LCM squeezes
 MBDCI

Stress Redistribution
stress Pressure depletion leads
h]o lateral stress to an increase in v, h
pressure
depletion
redistribution and a decrease in h
in stratum This is the Poisson
po
pf effect, related to /(1-)
4-E Stress Changes During Production

h h Increased means the

reservoir rock is stronger
depleted sand and tougher to drill
p
The h lost must be
h]f redistributed above and
stress below the reservoir
concentration

depth Blowout and LC!!

 MBDCI

Measuring the Stress Path

VALHALL: a North Sea Chalk field

55
Fracturing Pressure (MPa)

50 Slope:
45
Stress Path Parameter
4-E Stress Changes During Production

40

35 y = 10.9 + 0.90x
30 R2 = 0.95

25

20

15
15 20 25 30 35 40 45
Reservoir Pressure (MPa)

Stress
StressPath:
Path: Using
Usingfracture
fracturetests,
tests,measure
measurehh @
@various
variousPPres
res
 MBDCI

Hydraulic Fracture Test

pBD pBD - Break-down pressure
Pp - Fracture propagation pressure
pISIP - Instantaneous shut-in pressure
pCL - Fracture closure pressure
Bottomhole pressure

p
ISIP
4-E Stress Changes During Production

pCL
pP

shut-in pressure
Stop pumping &
Start pumping
at a low rate

p from p decay by flow

flow effects into the formation

Initial mud pressure

in the borehole (>po)
Time (constant pumping)

Classic
ClassicHydraulic
HydraulicFracture
FractureTest
TestTrace...
Trace...
 MBDCI

Step Rate Test for PF

pressure Step-rate test data

fracture
flow
PF
4-E Stress Changes During Production

matrix At each stage, inject, select

flow
p value consistently
Plot the data
Determine intersection

rate
4-E Stress Changes During Production Step-Rate Tests Raw Data
MBDCI
 MBDCI

4-E Stress Changes During Production Interpretation of Step-Rate Test

 MBDCI

Minifrac and Step-Rate Test

FIELD WELL MINIFRAC SRT DIFFERENCE

Field 1 A-15 396.8 402.45 -5.65
A-37 289.1 294 -4.9
B-19A 341.5 345 -3.5
C-12 310 313.5 -3.5
4-E Stress Changes During Production

C-11 324.6 323.8 0.8

P-34 415 413 2
Field 2 P-17 395 389 6
P-37 388 372 16
P-25 386.5 363.7 22.8

Simultaneous Minifrac and SRT tests on wells in the

North Sea show clearly that PF from SRT is not the
same as pCL from a careful hydraulic fracture test.
This shows the importance of quality control,
consistency, and care.
 MBDCI

Minifrac or SRT??
Two North Sea Fields - 2003
25
Minifrac minus SRT (bar)

Difference:
Difference:Thermal
ThermalEffects!
Effects!
20

15
4-E Stress Changes During Production

10

-5

-10
A-15 A-37 B-19A C-12 C-11 P-34 P-17 P-37 P-25
Well number
Minifrac: close to the wellbore only.
SRT, samples much larger volume, injects @ different T!
 MBDCI

Injection leads to T = !!

Waterflooding changes T

Drop in T causes rock Temperature
shrinkage (V = T)

This causes a loss of
4-E Stress Changes During Production

lateral stress in the Temperature of

injected fluid
reservoir zone
geothermal

Locally, a lower PF gradient - /km

Thus, the stress path is Original T

in the rock
complicated by T

Depth
 MBDCI

Repeated Measurements with T

PCL

Far-field h Measured PCL (= 3)

4-E Stress Changes During Production

PCL1
Successive cycles
of injection of low-
temperature fluid

Tres - Tinj
After
Tinj1 A larger temperature difference
some
time leads to a larger drop in PCL
 MBDCI

PF Evolution with Time - Depletion

Fracture Gradient (MPa/100m) North Sea Field 1
1.90
1.80 T E T/(1-
) = 100 bar
Early Value
1.70
1.60
4-E Stress Changes During Production

1.50 Later Value

1.40
1.30
1.20
1.10
1.00
18-Feb-82 14-Nov-84 11-Aug-87 07-May-90 31-Jan-93 28-Oct-95 24-Jul-98

Date of Test
With continued depletion, the reservoir fracture gradient
dropped from 1.65 to 1.44.
 MBDCI

Why Predict the Stress Path?

Assessment of the on-set of delayed sand
production in wells (time & risk analysis)

Design of hydraulic fractures and fracture
containment analysis
4-E Stress Changes During Production

Predicting permeability changes, especially in

fractured (mainly carbonate) reservoirs

Evaluation and prediction of subsidence

Drilling plans for asset redevelopment

Drilling through the reservoir to access deeper
targets (LC and blowout control)
 MBDCI

Example: Fractured Rock & Stress

Representative laboratory tests are not possible
for the k = () relationship

Well tests after stress changes to determine k
Rock vs Rock mass
4-E Stress Changes During Production

5 cm
--Intact rock
50 cm
--Single discontinuity

--Two discontinuities 20-100 m

--Several, etc.
--Rockmass

Stress path data can then predict changes in k!

 MBDCI

Exampe: Boundary Conditions

In the laboratory, for BCs we use either:
 No-lateral-strain (oedometric condition)
 Constant stress
4-E Stress Changes During Production

L/
V/

Constant stress No-lateral-strain on

on all boundaries, side boundaries,
3-D compaction 1-D compaction
 MBDCI

Laboratory Results

stic

Laboratory results from

r ela
high quality core are

in e a
useful, but insufficient

-l

ma hing
Non

al
teri
They allow you to

s

Cru
4-E Stress Changes During Production

estimate the behavior of

stic
a small specimen of rock

ela
You can estimate the

ear


Non-linear,

Lin
permanent strain res- non-elastic

ponse of the rock matrix

But, you cannot extra-
polate directly to the V
reservoir Permanent strain
 MBDCI

In the Field

Must incorporate the Reservoir 1 km 1 km 30 m
stress path of rock mass!

So that effects of T, p
Earth model 15153 km
are properly distributed
4-E Stress Changes During Production

BCs of the reservoir are Reservoir with

special
neither constant stress or boundary
zero displacement! elements on the
surface

This means that real field
measurements of stress The real reservoir

path are needed for model

calibration, and use.
Reservoir model: brick-shaped elements
 MBDCI

Data From Many Reservoirs

Data from 33 cases, 15 fields
Fracture gradient (MPa/100 m)

1.70

1.60
2
y = 0.5308x - 0.2447x + 1.0983
A>1
1.50 2
R = 0.7665
4-E Stress Changes During Production

1.40

1.30

1.20

1.10

1.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40

Pore pressure gradient (MPa/100 m)

The response of reservoirs is non-linear; each reservoir

responds differently, but some generalizations are possible
 MBDCI

Adriatic Basin

Adriatic Gas Field

60
Virgin fracture pressure (MPa)

50
4-E Stress Changes During Production

40

30
y = 1.5461x - 16.363
2
R = 0.9958
20

10

0
0 5 10 15 20 25 30 35

Virgin reservoir pressure (MPa)

 MBDCI

From Elastic Theory (h ~ 0)

Assumptions
elastic behavior constant - Poissons ratio
4-E Stress Changes During Production

 Linear

 Flat-lying and laterally extensive reservoir

 Uniform reservoir depletion, no arching effects

Poissons Ratio
 If ~ 0.25, h 0.67p
 If ~ 0.10 (fractured rock), h 0.90p

 If ~ 0.40 (highly ductile rock), h 0.33p

 MBDCI

Stress Path Behavior

Minimum in situ stress hmin

Initial stress path

No regional
depletion
4-E Stress Changes During Production

Reservoir-wide
depletion -path
Reservoir pressure
Some of these early time effects are
related to stress arching effects
 MBDCI

Reservoir Stress Path - I

An operational parameter (PF, drilling)

It cannot be measured in the lab because of:
 Sample damage
4-E Stress Changes During Production

 Scale effects (especially with natural fractures)

 Boundary conditions in lab are incorrect

Therefore it is difficult to predict in advance

However, it can be measured
 Hydraulic fracture tests, several cycles is best
 SRT but be careful about thermal effects

 Minifrac only measures hmin near the wellbore

 MBDCI

Reservoir Stress Path - II

The stress path following by a reservoir is not
linear, and strains can be largely permanent

The lateral total stress will decline, but it is
difficult to predict exactly because of
4-E Stress Changes During Production

 Non-linear stress-strain behavior

 Presence of fractures in the reservoir

 Other effects (e.g. massive cold water injection)

It is best to measure the stress path in real

situations. However, some cautions
 Noton massive cold or hot fluid injection wells
 Multicycle larger hydraulic fracture is best
 MBDCI

Delay Through Arching

Stress arching is taking place around wells
arching occurs
until the p zones
production interact at the
drawdown reservoir scale
zones
4-E Stress Changes During Production

compaction impeded
overburden stresses flow around the p, V zone
Example of multiple well and reservoir scale interaction

Local drawdown stress change effects are not the same as

those that occur upon reservoir-wide depletion.
 MBDCI

General Drawdown: no more Arching

z

full when zones interact,

subsidence arching is destroyed,
develops full
recompaction full compaction occurs
4-E Stress Changes During Production

triggered

stresses now flow uniformly without arching around zones

If large p values maintained by aggressive production &
injection, then uniform stresses will not develop over time
 MBDCI

Lateral Stressing Effects

lithotype
stress

UC sand
reducing
pressure
shale
4-E Stress Changes During Production

sandstone

salt increasing
pressure

limestone
assumed
depth initial h

The nature of the stress changes will also be affected by

stiffness differences, although these are not plotted here
 MBDCI

Thermal Effects

Stress Path is affected by cooling or heating

These lead to -V, +V (eg H2O flood, steam)

Massive can therefore take place
(Undrained p in low permeability shales)?
4-E Stress Changes During Production

The approach to analysis is:

 Calculatethe volume changes in a coupled model
 Apply to determine the stress changes

For thermal effects, this requires good data

Also, T data (thermal compressibility)
 MBDCI

Some T Consequences

Cool water injection to sustain pressure can

lead to breaching of cap rock
 Shale cools, -V, h until < p, fracturing...
4-E Stress Changes During Production

 Injection water is lost to adjacent zones

Steam or hot water injection

 Reservoir expands faster than shale
 Shear of well casings at the interface

However, shear dilation (high p, T) can be

beneficial, giving higher kw
 An aid to heavy oil extraction processes
 MBDCI

Shale Fracture by Cooling

To stress
Fracture breakthrough
po
thief zone, high k
4-E Stress Changes During Production

thin shale k~0

pinj v
fracture 3]init
plane sandstone,
high k

low k shale
net
Tinj frac
pres
3]final
T distribution near well

depth
 MBDCI

Cooling-Induced Stress Changes

To Water
displacement
To front
T
4-E Stress Changes During Production

T
front
T
HMAX

hmin

T
 MBDCI

Heating and Stress Fields

Near-field stresses are altered by T, p

Temperature increase
primary fracture causes stresses to
increase near fracture
4-E Stress Changes During Production

+T
secondary fracture

3 +3 heated zone
fracture tip

high pressure zone

p > 3(original)
 MBDCI

T and Shear Dilation

in weak rocks, such as

cool region many high porosity
hot region sandstones, dilation
expansion occurs during shearing
4-E Stress Changes During Production

compressional r
+V extensional triaxial test analogy

r
 MBDCI

Thermal Shearing

adjacent wells region of high shear

shearing no shearing

overburden

unheated HMAX > v

4-E Stress Changes During Production

reservoir +T

+ve +T T in the reservoir

max shear
-ve low shear stress because of symmetry
of displacements around the well
 MBDCI

The Inclusion Concept

Stress trajectories
overburden
deflection Slip along
4-E Stress Changes During Production

interface

Reservoir Zone
of -V

Zones of low
r, high
Giving rise to shear of casing, microseismicity
 MBDCI

Production Issues and

Very large changes in T, p, V

This leads to large stress changes
 Shearingof bedding planes, faulting
 Compaction, dilation, frac gradients, etc.
4-E Stress Changes During Production

 Changes in k (fractured reservoirs)

Coupled modeling is necessary

 Compressibilities must be used T, p,
 Non-linearities are important

Monitoring! (p, , microseismic,

deformation) and stress path clarification.