Peng S., Zhang J., Engineering Geology for Underground Rocks

Подождите немного. Документ загружается.

3 In-situ stress and pore pressure

3.1 In-situ stress regimes

In-situ stress magnitudes and orientations play a very important role in

geological engineering, and they are the most basic parameter inputs in de-

sign of underground structures. One of the main functions of rock mechan-

ics has been to determine in-situ stress. In-situ stress is characterized by

the magnitudes and directions of three principal stresses. Generally, in-situ

principal stresses are consisted of three mutually orthogonal stresses, i.e.

vertical stress (

), minimum horizontal stress (

), and maximum hori-

zontal stress (

). In different geographic, geologic, and tectonic regions,

in-situ stress magnitudes and orientation are very different. The three in-

situ stresses correspond to three principal stresses, namely greatest stress

(

), intermediate stress (

), and least stress (

). According to the rela-

tionships between principal stresses and vertical, minimum and maximum

horizontal stresses, three in-situ stress regimes (Fig. 3.1) can be used to de-

scribe the stress states, i.e.:

1. Normal faulting stress regime. In this case, gravity or vertical stress

drives normal faulting and fault slip occurs when the minimum

stress reaches a sufficiently low value (Zoback et al. 2003). In this

stress state the vertical stress is the greatest principal stress (

), i.e.

2. Strike-slip faulting stress regime. In this case, the vertical stress is

the intermediate principal stress (

). When the difference be-

tween maximum and minimum stresses is sufficiently large, slip

faulting occurs. Therefore, this case also represents slip faulting re-

gion. In this stress state, one has

3. Reverse (or thrust) faulting stress regime. In this case, the vertical

stress is the least principal stress (

). When the stress field is

very compressive, both horizontal stresses exceed the vertical stress,

and folding and reverse faulting could occur when the maximum

horizontal principal stress is sufficiently large relative to the vertical

46 3 In-situ stress and pore pressure

stress. Some thrust fault regions belong to this category. In this

stress state, one has

Fig. 3.1. Illustration of different faulting stress regimes: a. Normal faulting; b.

Reverse faulting; and c. Strike-slip faulting. In the figure,

 2/

It can be seen that knowing the stress regime can better constrain in-situ

stress magnitudes. Assuming there are critically oriented faults constrain-

ing stress magnitudes, the Mohr-Coulomb criterion in faults can be ex-

pressed as follows:

qC d

(3.1)

where

is the maximum effective stress, and p

DVV



;

is the

minimum effective stress,

DVV



; C

is the cohesion of the fault;

)sin1/()sin1(

fff

 ;

is the angle of internal friction of the

fault, ranging from 30 to 45q; p is the pore pressure.

In significant depth, the effect of cohesion of a fault is relatively unim-

portant in terms of the stress magnitudes, thus the cohesion can be ne-

glected. Consequently the stress relationship in faults can be rewritten as

the following form, assuming

=1:



(3.2)

In different stress regimes, according to stress magnitudes and Eq. 3.2

the following relationships exist:

1) Normal faulting stress regime



(3.3)

2) Strike-slip faulting stress regime

3.2 Overburden stress 47



(3.4)

3) Reverse/thrust faulting stress regime



(3.5)

Equations 3.2 to 3.5 give a bound between in-situ stresses. These equa-

tions allow to define the limiting stress magnitudes for different tectonic

environments: in normal faulting areas

# 0.6

, in reverse faulting areas

# 2.3

, and in strike-slip faulting areas if

# (

)/2,

# 2.2

(Zoback and Zoback 1989).

It is noted that the world stress map presents globally measured data of

in-situ stresses, where both stress regimes and maximum stress directions

can be obtained in some areas (Reinecker et al. 2006).

3.2 Overburden stress

Overburden stress, also called vertical stress, is induced by the weight of

the overlying formations. If the overlying formations have an average den-

sity of

, then the overburden stress can be calculated by:

(3.6)

where g is the acceleration due to gravity.

If density varies with depth, the vertical stress can be calculated by inte-

gration of rock densities from the surface to the depth of interest, z, i.e.:

gdzz

)(

(3.7)

where

(z)is the rock density as a function of depth.

In offshore areas, water depth (z

) needs to be considered to calculate

the vertical stress, that is:



bwwV

gdzzgz )(

UUV

(3.8)

where

is the density of water.

48 3 In-situ stress and pore pressure

Density log can be used to calculate overburden stress. However, den-

sity log is usually not recorded at shallow depth. In this case some empiri-

cal methods can be used to estimate rock density at shallow depth. Most

clastic rocks have an average density of 2.3 g/cm

, as shown in Chap. 1.

Overburden stress gradient is normally in range of 18 to 25 kPa/m (about

1.0 psi/ft). Figure 3.2 gives an example to calculate overburden stress and

stress gradient using density log.

(a) (b) (c)

Fig. 3.2. Density log data versus the true vertical depth (TVD in meters) and

Drillworks

calculation of overburden gradient and overburden stress in offshore

Drillworks is a suite of software developed by Knowledge Systems, Inc. for pre-

dicting in-situ stress, pore pressure and wellbore stability.

3.3 Minimum horizontal stress 49

drilling: a. density log data in g/cm

; b. overburden gradient in specific gravity

(sg); and c. overburden stress in MPa.

Worldwide in-situ stress measurements show that the vertical stress and

depth have linear relationships, as shown in the following:

)0028.00266.0()26.19.1( rr

(3.9)

027.0

(3.10)

where

is the vertical stress in MPa and z is the depth in m.

3.3 Minimum horizontal stress

In normal faulting stress regime, the minimum horizontal stress is the

minimum principal stress and can be calculated by the following equation:

)(

pKp





(3.11)

where p is the pore pressure;

is the Biot’s constant; K

is the effective

stress ratio; and



(3.12)

where

is the Poisson’s ratio. The effective stress ratio, K

, ranges from

0.33 to 1.

According to the joint industry projects conducted at Knowledge Sys-

tems (2001), the K

of 0.8 is fairly accurate to predict the minimum hori-

zontal stress and fracture gradient in shaly formations in depth of below

3000 m in the Gulf of Mexico.

By considering horizontal strain and deformation effect, Hooke’s law

can be applied to derive the horizontal minimum stress and strain relation-

ships. The following equation can be obtained, and it can be used to calcu-

late the minimum horizontal stress with tectonic strain effects:

hHVh

)(

1 











(3.13)

where E is the elastic modulus of the rock;

and

are the tectonic strains

in maximum and minimum horizontal stress directions, respectively.

Using fracture data from the US Gulf Coast, Breckels and van Eekelen

(1982) proposed the relationships between minimum horizontal stress and

50 3 In-situ stress and pore pressure

depth. In these relationships the abnormal pore pressure effects were also

considered. They can be written as follows:

)500,3()0105.0(46.07.310264.0

)500,3()0105.0(46.00053.0

145.1

mDDpD

!



(3.14)

where D is the depth in meters; p is the pore pressure in MPa.

is the

minimum horizontal stress in MPa.

Breckel and van Eekelen pointed out that their minimum stress predic-

tion is fairly reliable in the depth range down to 3500 m, and it can be used

with a fair degree of confidence in other tectonically relaxed areas such as

the North Sea (Fjær et al. 1992).

Hydraulic fracturing is a common technique for stress measurements in

deep boreholes. Hydraulic fracturing stress determination aims to deter-

mine both minimum and maximum horizontal stresses. Hydraulic pressuri-

zation is conducted with specialized equipment for high pressure injection.

During hydraulic fracturing, the borehole is pressured by fluid injection,

and the pump injection flow and the well pressure are measured (Rutqvist

et al. 2000). A similar test, leak-off test (LOT), has been conducted in

drilling industry to evaluate cement job and estimate formation fracture

gradient. Hubbert and Willis (1957) concluded that hydraulic fractures in

the earth always propagate perpendicular to the orientation of the least

principal stress, because it is the least energy configuration. This provides

a fundamental principle for conventional hydraulic fracturing measure-

ments and LOT.

Figure 3.3 illustrates the pressure, volume/time plot for a typical hydrau-

lic fracturing or LOT measurement with the first and second pressurization

cycles.

In a hydraulic fracture test, an interval of borehole is isolated and sealed

with inflatable packers. In a LOT, only the hole below the casing and any

new formation drilled prior to the test are exposed (Edwards et al. 2002).

The fluid is then pumped into the hole with a constant flowrate. The pres-

sure increase in the hole is typically linear as long as there are no leaks in

the system, and the exposed formation is not highly permeable. As the

pressure increases at some point, the rate of pressure increase changes such

that the pressure-time curve departs from linearity. This departure from

linearity is called (Fig. 3.3) fracture initiation pressure (p

After departure from linearity, pressure is typically seen to increase at a

lower rate until a maximum pressure is reached, and this pressure is rock

breakdown pressure (p

). After this point the pressure falls rapidly, or re-

mains steady. Horizontal stress magnitudes can be estimated from the frac-

ture breakdown pressure (Haimson and Fairhurst 1970):

3.3 Minimum horizontal stress 51

3 Tpp

Hhb



(3.15)

where p is the pore pressure; p

is the fracture breakdown pressure; and T

is the tensile strength of the rock.

This relationship assumes that the rock behaves elastic and isotropic. It

indicates that fracture breakdown pressure is greater in magnitude than the

minimum horizontal stress.

After the rock is broken down (hydraulic fracture created), at some

point the pressure in the hole levels off and remains fairly constant (p

prop

)

at the same flowrate. During this stage, the fracture is propagating. When

the pumps are turned off, the pressure immediately drops to the instantane-

ous shut-in pressure (p

isip

). After the well is shut-in, the pressure begins to

drop as fluid leaks into the formation. As the pressure declines, the fracture

starts to close. If the fracture is vertical and largely in the far field, then the

stress acting to close the fracture is equal to the minimum horizontal stress

(

isip

= S

Pressure

Volume/time

Volume Volume

prop

isip

= S

Pressure

Volume/time

Volume Volume

prop

Fig. 3.3. Schematic pressure versus volume/time plot of a hydraulic fracture proc-

ess. The figure also shows the determination of minimum stress/closure pressure

) using the tangent method.

52 3 In-situ stress and pore pressure

In the LOT test the minimum horizontal stress (

or S

) is determined

from the shut-in pressure. The minimum horizontal stress is equal to the

closure pressure (p

), as shown in Fig. 3.3.

(3.16)

3.4 Maximum horizontal stress

3.4.1 Determinations from field tests

The maximum horizontal stress can be estimated from the LOT with frac-

ture reopening test, as shown in Fig. 3.3. In an ideal case, when no fluid

penetrates the fracture until fracture reopening, the maximum horizontal

stress (

) may be calculated using the following equation for a vertical

borehole (Bredehoeft et al. 1976):

3 pp

rhH



(3.17)

where p

is the reopening pressure and p

is the pore pressure in the frac-

ture.

This equation is derived from the Kirsch solution for a circular hole sub-

jected to an internal pressure in an isotropic, homogeneous, and linear elas-

tic medium. The assumption in the derivation is that the reopening occurs

when the fluid pressure applied on the borehole wall is high enough to

cancel out the minimum tangential stress at the borehole wall.

If the fracture is slightly conductive, the fluid pressure at the wellbore

wall may only partially penetrate, and the pressure inside the fracture at the

wellbore wall is equal to the well-pressure. In this case, the maximum

horizontal stress can be obtained by (Ito et al. 1997):

rhH

p23



(3.18)

In the extreme case with a more conductive fracture, the fluid pressure

may completely penetrate to the crack tip. This implies that the reopening

is dominated by total force formed by the fluid pressure on the fracture

surfaces, and reopening pressure would be equal to the minimum horizon-

tal stress (Rutqvist et al. 2000):

(3.19)

3.4 Maximum horizontal stress 53

3.4.2 Determinations from various equations

If there is no direct measurement available for determination of the value

of the maximum horizontal stress, the following relationship can be used to

estimate the maximum horizontal stress:

)(

hVhH





(3.20)

where m is a constant, normally m = 0 - 2. In the normal faulting stress re-

gime, m can generally be taken as 0.5.

While drilling, when the mud weight in a borehole is high enough, the

mud pressure will start to fracture the rock, and tensile failure will occur.

At this moment, the minimum tangential stress around wellbore is equal to

rock tensile strength. From the Kirsch equation (refer to Chap. 7) the fol-

lowing relationship can be obtained:

3 Tpp

mhH



(3.21)

where p is the pore pressure; p

is the mud pressure in the borehole.

Using elasticity theory and the Mohr-Coulomb failure criterion for slip-

page on the faults, the relationship between horizontal stress and pore pres-

sure during mobilization of normal and thrust faulting are as follows (Ad-

dis et al. 1996):

Normal faulting:





)(

sin1

(3.22)

Thrust faulting:

]})1([{

)1(

QQV





fhf

qpq

(3.23)

where is

is the internal friction angle of the fault and

sin1





When the stress concentration around a wellbore exceeds the rock com-

pressive strength, breakouts are induced around the wellbore rocks. Be-

cause breakout width is expected to remain stable as breakout growth oc-

curs, Barton et al. (1988) presented a method to determine the maximum

horizontal stress utilizing observations of breakout width when the rock

uniaxial compressive strength is known, that is:

54 3 In-situ stress and pore pressure

ppUCS

TVV

2cos21

)2cos21(





(3.24)

where UCS is the uniaxial compressive strength of the rock;

is the

breakout angle;

is the thermal stress arising from the difference (

between the mud temperature (T

) and the formation temperature (T

) , and



)(

fmt

TTE

(3.25)

where

is the thermal expansion coefficient of the rock; E is the Young’s

modulus of the rock.

Worldwide in-situ stress measurements show that the mean horizontal

stress components and vertical stress at a depth of  3000 m have the fol-

lowing relationship (Hoek and Brown 1980):

5.0

1500

3.0

100



(3.26)

where z is the burial depth in meters; k is defined as the ratio of the average

horizontal stress,

2)(

 , to the vertical stress,

. That is:



In-situ stress measurements show that the maximum and minimum hori-

zontal stress components have the following relationship:

1.4 to 3.3

3.5 Pore pressure prediction

Pore pressure is one of the most important parameters for geomechanical

and geological analyses. One cannot obtain a correct prediction for any ge-

omechanical model without a right pore pressure input for the porous for-

mations (Zhang et al. 2006, Madge et al. 2006). Pore pressure varies from

hydrostatic (pressure gradient of about 1.0 MPa/km, or 0.44 psi/ft), which

is normal pressure, to severely overpressured (48% to 95% of the overbur-

den stress). The abnormal pore pressure is caused by abnormal formation

compaction. When sediments compact normally, their porosity is reduced

at the same time as pore fluid is expelled. This causes hydrostatic pore