Transcript stress - University of Waterloo

Module B:
In Situ Stresses
Earth 437
Maurice B. Dusseault
University of Waterloo
Common Symbols in Earth Stresses









sv,shmin,sHMAX : Vertical, minor and major
horizontal stresses (usually sv  to surface)
Sv,Sh,SH: Same as above, different symbols
s1,s2,s3: Major, intermediate, minor stress
s1,s2,s3: Effective or matrix (solid) stress
E, n:
Young’s modulus, Poisson’s ratio
f:
Porosity (e.g. 0.25, or 25%)
r, g, po: Density, unit weight, pore pressure
k:
Permeability (kv, kh…)
These are the most common symbols used in
discussing stresses in the earth
Stresses in the Earth: Intro I

In situ stresses: a vital initial condition for
all geomechanics issues, not just drilling!
 To
carry out any quantitative analysis, it is
necessary to start from the initial stress state
 For example, deep reservoir depletion can lead to
a Δp of perhaps -75 MPa, so that Δs’v = +75 MPa.
 The stress change is what is important; it is
defined as Δs’ = s’ final - s’ initial
 This Δs’ value is used to compute subsidence,
rock behavior (shearing, collapse), and so on

In hard rocks (mining), [s]ij can be calculated
from direct strain measurements – [De]ij
Stresses in the Earth: Intro II

In sedimentary rocks (oil and gas
applications), it is far more difficult
 The
locations are deep, hard to get to
 And, the strains are small, hard to measure
 The rocks are porous, poor strain response
So… hydraulic fracture-based methods are
widely used - Minifrac™, LOT, XLOT
 +Core-based methods (DSCA, vP(q), …)
 +Geophysical logging based methods
 +Geological inference (burial and tectonic
histories of the basin give excellent clues)

Stress Definitions
s1 > s2 > s3
s3
Principal
s2 Stresses
s1
s1
s2
max planes
sa = s1
sr
sr = s3
s3
slip
planes
Triaxial
Test
Stresses
sa
y
We usually
assume sv is a
principal stress
x
sv
sHMAX > shmin
sr
sq
sHMAX
shmin
In Situ
Stresses
q
r
ri
z
Borehole
Stresses
Local, Reservoir and Regional Scales

~200 km
Regional Scale Stresses
 Basin
scale: 50 km to 1000 km
 Often called “far-field stresses”

Reservoir Scale Stresses
~4 km
A
reservoir, or part of a reservoir
 Scale from 500 m to several km
 Salt dome region: 5-20 km affected zone

Local Scale Stresses
~400 m
 Borehole
region: 1-5 m
 Drawdown zone (well scale) 100-1000 m

Small Scale Stresses (less than 10-20 cm)
Common Stress Regimes

The most common stress regimes are:
 Relaxed,
or non-tectonic (no faulting, flat-lying):
vertical stress, sv, is = s1 (major stress)
 Normal fault regime: sv is s1
 Thrust fault regime: sv is s3 (least stress)
 Strike-slip regime: sv is s2 (intermediate stress)
 Listric (growth, down-to-sea or GoM) fault regime:
sv changes from s1 to s3 at depth, then back to s1
Most sedimentary basins with O&G have
relatively simple stress regimes
 But, there are local complications, such as
multiple faults, salt domes, uplift, etc.

Faults and Plate Tectonics
The Big Picture!
Compression region
Regions of crustal extension
Where Are Tectonics Important?





Near active plates (eg: California, Sumatra,
Colombia), tectonics governs stresses
Near mountains, tectonic forces dominate
Away from plate margins and mountains
(eg: Williston Basin, Kalimantan, GoM),
other factors are important
In continental margin basins salt tectonics
(domes and tongues) can be very important
In non-tectonic intracontinental basins
(Michigan, Williston, Permian…), the shape
and burial/erosion history are more
important than tectonics
Basins: Major Examples in USA
Rockies Foreland Basins, compressive
stresses controlled by mountain thrust
Thrust basins
WILLISTON
BASIN
Powder
River B.
San Joachim,
a rift valley
Paradox B.
Southern CA
basin complex,
strike-slip and
normal faulting
MICHIGAN
BASIN
MIDCONTINENT
BASINS
Atlantic coastal
plain and
offshore basin
complexes,
passive margins
Non-Tectonic Regimes

On stable continental plates, far from
active plate boundaries. Some examples
 Mid-continent
basins: Williston, Michigan,
Permian age basins, East Texas, Songliao Basin
(richest Chinese basin), interior Russia…
On passive continental plate oceanic
margins such as GoM, Kalimantan, Nova
Scotia, NW Norway coast, Angola, etc.
 Basin geometry, history of sedimentation,
compaction, burial, erosion, diagenesis, salt
dissolution features…, salt tectonics affect
stress states locally & substantially

Stresses and Basin Shape
USA
Cross-section
New Orleans
shoreline
Houston
Regional
s3 directions
Gulf of Mexico
listric
faults
Edge of continental shelf
GoM example: regional stress directions
are dominated by the continental slope,
except locally near salt domes and a few
structures such as the Mississippi canyon
Normal Fault Regime
The San Joachim Valley
in California, the Rhine
Valley between France
and Germany, the Gulf
of Thailand are all
normal fault grabens
graben
horst
extension
sv = s1
sHMAX
= s2
Horst-graben structure
shmin
= s3
The normal fault regime is also
called the extensional regime. It
is characteristic of shallow rocks
in all non-tectonic sedimentary
basins without large erosion.
Some Classic Normal Fault Areas
Red Sea
Around UK, Ireland
Normal Fault Zones (Pull-Apart)
Mid-ocean rifts
 East African Rift
 Upper zones, GoM
 Gulf of Thailand
 Upper Cook Inlet
 On flanks of thrust
faults, etc.

Normal Faulting Regimes








High angle faults at surface (60°-70° dip)
This indicates that sv = s1 when faulting
occurred. (But, is the fault old or active?)
Also, sHMAX = s2 and shmin = s3
Characteristic of extensional strain
Also, typical of non-tectonic basins
Hydraulic fractures are vertical,  to shmin
However, high angle surface faults may
“flatten” at greater depth (as in the GoM)
Many continental margins, passive basins,
regions of crustal “pull-apart” …
Strike-Slip or Wrench Fault
Block diagram
shmin = s3
acute
angle
sv = s2
sHMAX
= s1
~vertical
fault plane
shmin
sHMAX
Associated
normal faults
Surface
view
Strike-Slip Stress Regime







Very high angle faults (>80° usually)
Indicates sv = s2 (sHMAX = s1, shmin = s3)
when the fault formed
Characteristic of plate margins
Common at depth in eroded basins
Common some distance from compression
Usually, normal faults are found nearby at
the surface, away from the main fault
trace, to accommodate strata movements
Hydraulic fractures vertical,  to shmin
Small Window Basins

Small “window” basins
between strike-slip
faults have complex
stress conditions
Small basin opened up
between parallel strike-slip
faults. Locally, stresses
can vary from normal to
thrust regimes, very
complex
Graben basin
San Joachim
Valley Basin
Southern CA
basin complex
Small
basin
15 km
Thrust Fault Regime and Structures
The shale bed in zone A has gone through
one hinge point, through two in zone B, and
through three hinge bends in zone C.
sv = s3
sHMAX = s1
hinge points
overthrust sheet
strong
lateral
thrust
B
A
high-p
shale
static basal sheet
compression
highly fractured zone
C
largely unfractured shale
Thrust and Reverse Faults
Less than 45° angle on fault plane
 If less than 20-25°, it is almost always
called a thrust fault rather than a reverse
fault

This angle is always less than
45°, usually less than 30°
Sometimes, thrust faults can take on
very complex, stacked structures
Thrust Faults and Mountains
NWT
Canadian Shield
Nunavut
Athabasca Basin
(Precambrian)
Alberta
Syncline
BC
Basin
edge
Tectonic
stress
Edmonton
ALTA
Canadian Shield
Breakouts
 to sHMAX
SASK
Williston Basin
MAN
USA
The Western Canadian Sedimentary Basin
Alberta is the “classic” compressional (thrust fault) regime
Compressional Basin Section
not to
scale
Thrust faults
SW
Rockies
Alberta Syncline
Massive heavy oil
deposits
Salt solution and
Edmonton
collapse features
NE
+
+
Cretaceous
sands, shales
Regional
Cretaceous
unconformity
Jurassic and
older carbonates,
sandstones, shales
+
+
+ +
+
+ + + +
+
+ +
+ +
+ + +
+
+
+ +
+
++
+ +
+
+
+ +
+
+
+ +
+
+ +
+
+
+ +
+
+
+ +
+
+
++
Precambrian rocks
Schematic cross-section
through Edmonton, Alberta
+
Thrust Faults






Low angle faults (dip of 0° to 30° usually)
Indicates sv = s3 (sHMAX = s1) when the
fault formed (or if it is still active)
Characteristic of compression regions,
associated with thrust mountain ranges
Same stress condition can often be found
at shallow depths in eroded basins
Usually, thrust fault “sheets” are bounded
by systems of normal and strike-slip faults
Hydraulic fractures will be horizontal (in
fact, usually they propagate gently upward)
Cross Section: Stress & Structure
forebasin
Mountains 

Golden Colorado 
Banff Alberta










 Distant plains
 Eastern Colorado
 Calgary Alberta
sHMAX
Near mountains:
•Very high sHMAX
•For great depth, sv = s3
•Thrusts, folds…
•Fractured strata
•Low to modest po
sHMAX
Distant from mountains:
•Moderate to high sHMAX
•For some depth, sv = s3
•Flat-lying, no faults
•Strata are relatively intact
•Low pressures
Sing07.024
Generally very high pore pressures are not found in thrust regimes and
their forebasins, as rocks are somewhat fractured, pressures dissipate
Real Thrust Faulting Structures
Rocks are permeable because of fractures
and folds, po is rarely overpressured
High stresses near the
tectonic compression front
Lower stresses far from the
tectonic compression front
Folding Thrust sheets Thrust fault planes
Folded belt in front of last thrust
Sing07.024
Undeformed sediments
Moderate deep overpressure may
remain in the deepest part of the
foreland basin (po ~ 1.3-1.4 gw·z)
Thrusting Aided by High po
Axis of geosyncline
Zone of abnormal fluid pressure
Normal faults
Undeformed
Eroded
Thrust fault
Undeformed strata
30
20
10
0
10
20
MILES
Sing07.025
This is a massive gravitational “landslide” (Wyoming), similar to listric faults.
This could not be possible without high local pore pressures in shales, which
allowed the fault block to virtually “float” along the fault plane
30
Listric Faulting and Stresses
grabens
steep
at top
“down-to-the-sea” faults
Listric faults on continental margins
lead to unusual stress regimes where
the major stress changes from vertical
to near-horizontal at depth
sea
stress
sv
sh
slip planes
zone where faults coalesce
(detachment or décollement zone)
depth
Stresses
change
with z!
Listric Faults






Characteristic of passive continental
margin basins that are “open-to-the-sea”
(GoM)
Look like normal faults at the surface
At depth, the faults flatten to become
thrust faults
Stress regimes change with depth!
Often associated with overpressured zones
These faults are like massive landslides
India and Tibet Examples
Active basins
Passive basins
Bay of Bengal Region and North






A continental margin basin exists offshore
south of Dacca and Calcutta, we will expect
a relaxed stress condition, GoM features
A strong thrust basin to the north, along
the Himalaya front, fractures, no oil
Strike-slip to the east (Sagaing zone)
Shan-Thai Plateau is partly a zone of
extension, some N-S faults are normal
Sichuan Basin, relatively undeformed, but
under strong compression
Etc.
Tectonic
Structure Map of
the Region
Tectonics Give Stress State Clues
This NASA image of Ganymede shows complex tectonism, giving
clues about the stresses and dynamics which caused the structures
Normal faulting
Shear zones
Extraterrestrial Geomechanics!
North Sea Stress Trajectories
From the World
Stress Map Project
 In the central part,
complex of grabens
and wrench faults
 Many “blocks”, each
with a stress pattern
 Farther north, the
Continental Shelf is
“open to the sea”
 Breakouts, LOT, HF
tests…

North America (World Stress Map)
(available online at www.world-stress-map.org)
Stress Map of Europe
Many solutions for
earthquake focal
mechanisms in
southern Europe give
the dense stress
coverage
 In the hard-rock areas
– strain relief methods
 In quiescent basins,
data from breakouts,
hydraulic fracturing,
LOT

Geological History!!
This basin opened, filled, was compressed
(thrusts and folds), uplifted and eroded
 Later, it subsided with new sediment fill
 The different lithologies compacted
differently, leading to normal faults

clays and silts
gravels
Normal faults
Relaxed stresses
3-10 km
Thrust condition
20 – 100 km
Folds and closed
structures
Conclusions on Tectonics and Faults

The tectonic condition and the nature and
orientation of faults give important clues:
 The
principal stress directions
 The relative magnitude of the stresses
 Whether stresses are intense or relaxed
To be confident of the stress conditions,
the faults must be shown to be “active”
 Geological history can be complex, giving
different stress fields at different depth
 The first task in a new area is to study the
stresses and tectonic features

Burial and Diagenetic History








What controls stresses during burial?
How do stresses change with diagenesis?
What happens during uplift and erosion?
Do all rocks behave the same?
What happens if pore pressures change?
When there is tectonic loading or
unloading, how are stress changes
partitioned in strata?
Hydrocarbon generation effect?
Etc, etc… (it gets complicated…)
Stresses at Depth
σv from density logs, σhmin, σHMAX from
various methods (geological estimation,
HF…)
 We often use the “K” coefficient.

sh min  p o sh min
K 

s v  po
sv
Ratio of least horizontal effective stress
to the vertical effective stress (in situ)
 <1 – vertical fracturing
 >1 – “horizontal” fracturing

Friction Angle Control of Stresses







If soft sediment is in a state of plastic
yield during sedimentation and burial:
K’]min = (1 - sin f)/(1 + sin f) (soft seds)
f is the Mohr-Coulomb friction angle
For loose sand: f = 30°, thus K’]min = 0.33
For shales, much lower friction angles
f = 10°: this gives K’ = 0.70
We observe that horizontal stresses in
soft shale are much higher than in sands
during burial, until sediments are indurated
Upper GoM, Gulf of Thailand…
Burial Stresses, Friction Control
These values are the limits,
not actual values in situ
E = stiffness
UC sand
0.5E
shale
0.75E
Ka = 0.70
salt is
viscoplastic
salt
sandstone
Ka = 0.33
Ka = 1.0
Ka = 0.33
E
Note: s = s - po
(Terzaghi’s law)
po
Frictional Control of Stresses





In fact, the strata we encounter are rarely
purely frictional materials
They also have cohesion
The frictional stress control “model” is only
intended to give the theoretical lower
bound of shmin for high porosity strata
If rocks are strongly cemented, it is
possible to have stresses lower than this
In exceptional cases, open fractures!
 Shallow,
above flanks of salt domes
 In mountainous areas
Stresses In and Around Salt

Salt is a very special material:
 Highly
soluble
 Low density (2.16 g/cm3 or 18 ppg equivalent)
 Viscoplastic, so all stresses are the same
Drilling long sections of salt is a challenge
 Drilling near salt structures such as diapirs
and sand tongues is challenging
 Therefore, a special Module on salt drilling
is included, and not treated here…
 See Module G for a full discussion…

Deep Salt Diapir Example
Gas Pull Down
Mid-Miocene regional pressure boundary
Top Balder
Top Chalk
Intra Hod/Salt
Sands and Shales, sh vs. Depth
Ko - effective stress ratio
stress, units of density (s/z)
1.0
1.5
2.0
mud
2.5
0.5
0
n ~ 0.45
sands &
sandstones
n ~ 0.25
1.0
clays &
shales
(n = Poisson’s ratio)
clay
mudstone
hydrostat = gwz
sv, vertical stress
sh
sand
sh
shale
This model
applies only to
upper 2000 m of
soft sediments and
no tectonics!
sh/sv sh/sv
shale
depth
Stress plot (density)
2.0 = 16.7 ppg
depth
Stress plot (ratio of sh / sv)
Porosity-Depth Relationships
0
0.25
0.50
0.75
1.0
porosity
sands &
sandstones
mud
clay & shale,
“normal” line
clay
mudstone
effect of
overpressures
on porosity
4-8 km
depth
shale
The specific details of
these relationships are
a function of basin age,
diagenesis, heat flow ...
Diagenesis and Rocks





Mechanical compaction, most important in
shales, drives the particles closer together
Pressure solution, important in sands but
not shales, lowers porosity substantially
Cementation, usually SiO2 or CaCO3, bonds
grains together, reducing porosity as well,
most important in sands (sandstones)
Very deep, clay minerals change, leading to
fracturing & stress changes (shales only)
Diagenesis rate and intensity is ƒ[T, s(ie:
depth), t, chemistry…]
Stress and Diagenesis
diagenesis
sv
burial
Sand burial in a non-tectonic
environment results in sh < sv
because of friction. Diagenesis
seems to reduce this stress
difference slowly over long
periods of time.
f is the friction angle for sand
sh
Stress & Diagenesis, no Tectonics





If there has been no tectonic activity, sh is
less than sv
In sands, the ratio Ko (defined as the ratio
of horizontal to vertical stress, sh/sv),
can be as low as 0.3, usually 0.4 – 0.6
Shales have a low angle of friction, usually
Ko is 0.6 – 0.8, even as high as 0.95 in muds
Thus, the fracture gradient is higher in
mud or shale in non-tectonic areas (GoM)
Deep burial and diagenesis tend to reduce
the stress differences
Diagenesis and Strength
max planes
shear stress - 
chemical cementation
diagenesis
effects on the
strength
sa
diagenetic
strength
increase
s3
s1
strength
increase
Principal stresses – see inset
slip
planes
sr
sr = s3
densification
(more interlock)
cohesion
sa = s1
Triaxial
Test
Stresses
original
sediment
normal
stress - sn
Diagenesis, Strength, Stiffness





Mechanical compaction in the early burial
stages increases the friction angle, f
Chemical effects increase cohesion, c,
often cementing particles together
The sediments also become stiffer (higher
Young’s modulus)
In general, rocks become stronger, ƒ(f)
However: deep, intense shale diagenesis
often generates shrinkage fractures; they
become weaker, shmin and sHMAX , k, even
though they are lower porosity
Erosion and Stresses
s’v
diagenesis
sv
sh
Elastic behavior governs
unloading because the rock is
stiff and strong; lateral
stresses increase naturally
s’h
Effect of Erosion
Once a sediment is buried and diagenetically indurated, it behaves elastically
 Direct erosion without tectonic loading
leads to the so-called “Poisson effect”:

n
Ds'h 
Ds' v
1 n

Thus, erosion naturally leads toward the
shallow condition K’o > 1.0 (except for salt,
which behaves as a viscous fluid)
Rocks, Stresses in an Eroded Basin
s, po
Erosion has created a
thrust
“skin” near the surface
stress
where sHMAX = s1, and
sv
state
sv = s3
sh sv = s3
 Deeper, a strike-slip
regime condition sHMAX
= s3, sv = s2 is found
sv = s2
 I.e. fracture gradient
increases with depth
strike-slip
stress
 Rocks are stronger,
state
stiffer
 High pore pressures
Z
(po > 1.3 sv) are rare in
Assuming that both horizontal stresses are equal eroded basins

Eroded Basin





The “Poisson effect” during unloading
generates a region at shallow depth where
horizontal stresses are larger than vertical
Also, the rocks are strong
Drilling underbalanced is becoming common
in such regions because of rock strength
Pore pressures in such regions are rarely in
the overpressure domain
For large overpressures, there is a special
module
Conclusions on Erosion





Generation of high Ko values requires
plastic deformation (incl. diagenesis)
This happens naturally in sedimentary
basins, even without tectonics, during the
burial phase, however, after induration…
The unloading phase is largely elastic
It is unlikely that glaciation has a large
effect on previously eroded basins
Any eroded basin will have a skin that is
“overstressed”, with sv = s3 to some depth
Changes in Stresses
Opening a hole (drilling) changes stresses
massively sr is now s3 at a local scale
 In drilling, heating or cooling the wall can
change the stresses, affecting stability
 High mud pressures can cause joints to slip
 Stresses changed by injection or production

 E.g.:
Depletion of the reservoir decreases the
fracture gradient (loss of horizontal stress)
Hot or cold water injection changes stresses
 Pressurization or depletion can result in
shearing of casing, faulting, minor seismicity

In Situ Stresses: Summary





Stresses are needed for casing programs,
borehole stability analysis, etc. etc.
Pore pressures are a vital aspect as well
Often stress directions and relative
magnitudes can be estimated from the
tectonic and burial history of the
sedimentary basin, or its structure
The most recent fault patterns often
reflect the stress regime
Features such as salt domes, etc.,
invariably indicate locally altered stresses
Lessons Learned

Stresses in the earth arise from:
 Gravity
effects (rock bulk density – sv)
 Tectonic effects (compression, salt tectonics…)
Recent faults indicate stress patterns
 Basin shape is a stress pattern indicator
 Geology and history allow us to estimate
relative stress magnitudes and orientations
 Stress is involved in many basin processes

 Basin
fabric, diagenesis, overpressure, oil
migration, gas and oil valving…
Recommendations
Offshore or inshore, it pays to have some
stress information for drilling, hydraulic
fracturing, reservoir modeling…
 The first step is to use geological history to
build a regional stress model
 The pore pressure conditions should be
inferred as well, (also offset well data…)
 Then, examine reservoir & local scale factors

 Faults,
salt features, reefs and drapes,
hydrothermalism, and other features
 These may “perturb” regional stresses
Extra Materials
Breakouts and Natural Stresses
Vertical
borehole
sHMAX
principal
stresses,
s1 > s3
shmin
breakouts
damage,
ravelling
high
sq
Breakouts are evidence of
stress anisotropy and are
caused by shear rupture of
the borehole wall
However, care must be
taken in assessing
breakouts, as other factors
can “interfere”
Use only vertical wells
(10) to get good stress
orientations
Some “Confusing” Effects
sHMAX
material
anisotropy
shmin
jointed
laminated
shale
slabbing,
ravelling
Material anisotropy means
the mechanical properties
are different in different
directions, as in a fissile
shale
bedding
planes
s3
ravelling
s1
s3
ravelling
sHMAX
Borehole Features: Wall Scan
0
90
180
270 360
axial
fractures
Stress
directions
sHMAX
“en-echelon”
axial fractures
if hole is
slightly inclined
shmin
breakouts
large
washout
Sinusoidal
fracture
traces
higher angle
of intersection
(joint plane)
low intersection
angle (bedding?)
Geometry of
joint plane
intersection
breakouts
no breakouts
axial fractures
Reconstructed breakout data from
Schlumberger borehole scanner logs
Axial fractures and breakouts are stress
direction indicators. If the stress difference
is large, breakouts are also larger (deeper
and wider).
s Directions: Breakouts, Fractures
This is a LWD log trace taken during a trip, so resolution is poor
Natural
fracture
plane
Borehole wall
tensile fractures
Small breakouts (90°
to tensile fractures)
shmin is at 40°Az in this example
Natural
fracture
plane
Modest breakouts,
no tensile fractures
Use of Breakouts, Axial Fractures







For s orientations, use only wells that are
vertical +/- 10° (rarely more inclined)
Establish quality control on your data
(length, symmetry across hole, quality…)
Grade your data (“A” “B”, “C” quality…)
Breakouts: sHMAX is at 90° to breakout axis
Borehole wall axial fractures: sHMAX is
parallel to the fractures axis
Combine with geology, sv calculations from
density log data, LOT data, HF data…
Build a stress map for your region & use it
More about Breakouts, Fractures





Don’t confuse breakouts with hole
enlargement (breakouts are symmetrical,
and the minor axis ~ hole gauge size)
Don’t confuse breakouts with sloughing in a
fissile shale when the hole dip is close to the
dip of the shale fissility
Joints and planar features trace sinusoidal
patterns on the borehole wall; induced axial
fractures do not
4-arm dipmeter data must show ~symmetry,
consistency, reasonable length, etc. (QC)
Full wall scans are easier to interpret
Log Data as Stress Indicators
4-arm dipmeter response
Some other log properties may
give stress orientation information
Formation, po
Conductivity response
Pad region
b
borehole
c
borehole,
a pw
d
sHMAX
Anisotropically invaded zone,
& stress microfissured region,
giving anisotropic permeability
shmin
Mud Weight Window Prognosis
1.1
1.3
1.5
1.7
1.9
2.1
prognosis
for shmin
prognosis
for po
Convert to
density units
by dividing
pressures and
stresses by
vertical depth
density, g/cm3
MW
=1.92
sv
XLOT shmin
value
overpressure
transition zone
2.3
Previous
casing
string
shoe
shoe location for
next casing string
area indicates
possible MW
depth
strong overpressure zone
This is a conservative approach: often, we can stretch the lower limit a bit, carefully
Mud Weight Prognoses
Data garnered from many sources, offset
wells, geology, XLOT, MWD…
 The interval defined is the secure window:

 MW
below pfrac (= shmin) at the shoe
 MW above pmax, usually at or near the bit

However, the window can be pushed a bit!
 Drilling
MW + ECD can be a bit above pfrac
 If shales are strong and no high k sands, drilling
may take place slightly below the pore pressure
Tricks such as high weight pills on trips
 Underbalanced drilling in strong shales

Pushing the Envelope in Drilling!
(2.0 = 16.7 ppg) 1.1
1.3
1.5
1.7
1.9
2.1
2.3
density, g/cm3
prognosis
for shmin
prognosis
for po
Convert to
density units
by dividing
pressures and
stresses by
vertical depth
MW
=1.92
sv
XLOT shmin
value
overpressure
transition zone
area indicates
possible MW
depth
Previous
casing
string
shoe
deeper shoe for
casing string!
strong overpressure zone
Using high weight trip pills and careful monitoring, the lower limit can be extended
Pore Pressure Generation
sv
Poisson
effect
Dsh = Dsv(n/1-n)
faulting
n/(1-n)
Pore pressure generation
reduces effective stresses, but
not in a path parallel to
hydrostatic because of
Poisson effect. This can lead
to rock shear if po is high
sh
Diagenesis, Burial, Erosion
25
sv - MPa
diagenesis
-2000 m-
20
This stress path
explains the
presence of 15
high horizontal
stresses near
the surface 10
Burial to 2000 m,
erosion to 500 m
erosion, n ~ 0.2
-500 m5
sh =17 MPa
sv = 7 MPa
Stress path
sh - MPa
0
0
5
10
15
20
A simple calculation of the probable effect of erosion of 1500 m
of rocks on the stresses. We assumed initial stress state (red
star), took a reasonable Poisson’s ratio for erosion (0.2), and
made the calculation. (Assume that po is always 8.33 ppg)
25
Fluid Flow in Shales






Shales are almost always the source rocks;
sands, limestones the reservoirs
Most shales are water-wet, except for the
rich kerogenous shales
Gas or oil as the non-wetting phase flows
only with gradients of > 500
Two-phase permeability of shales = zero
Thus, HC flow must be through fissures,
fractures, or other macrodiscontinuities
Shales are originally intact, so O&G expulsion
is likely through induced fractures
Stress and Petroleum Migration
How are microfissures generated?
 When s3 normal to fissures is < po

 po
is the pressure in the shale pores
 s3 is the lateral confining stress at the scale of
the microfissure

This can occur by various geological
and tectonic processes:
 Processes
which increase po
 Processes which decrease s3
 These are linked to shrinkage of shale, loading…
Reduction in s3
Tectonic unloading reduces s3 (shmin)
 Any shrinkage in the material reduces s3

-Clay compaction, thermal shrinkage (uplift)
-Loading of anisotropic shales
-Smectite to mixed-layer to illite changes
When s3 drops below po, fissures can open
stably, and remain open
 Fissures can be dominantly vertical (usual
case), or also horizontal
 Now, fluids can easily flow through the open
cracks, migrate to traps & accumulate

Increase in po








HC generation increases pore pressures
Retarded compaction (overpressure)
Smectite diagenesis releases H2O
Thermal pressuring increase po
Accumulation of a thick gas zone in a
reservoir with good vertical closure
Tectonic loading (?) may increase po
Other processes (e.g. gypsum dewatering)
These processes give the driving forces
for fluid flow and HC migration
Can Oil or Gas Flow in Shales?
clay
particle
Typical pore size in shale
0.1-0.5mm
gas or oil bubble
clay
particle
water
pw
po
r
oil or
gas
Dp = g/2r
g = interfacial tension
r = radius of curvature
Dp = po - pw
Darcy flow requires huge Dp
Conclusion: oil or gas cannot flow through intact shale!
Can Water Flow Easily in Shale?
Cations
Hydrated
cations
Bound water
Na+
H2O
molecules
Much of the water in shales is not
free to move easily:
Adsorbed on the clay fragments
Hydrated onto cations
Pore throats are “blocked” by
adsorbed water
Bound water
Free water, partly on ions
Conclusion: oil or gas cannot flow through intact shale!
Fluid Generation & Fracturing
Flow in shales must
be through fractures!
3-20 mm
sv
shale
T, p, s
increase
kerogen
microfissure
fluid
flow
sv
high T, p, s
oil and gas
generation of hydrocarbon fluids
semi-solid
organics,
po < sh < sv
po = sh < sv,
fractures
develop and
grow
fluids are
expelled through
the fracture
network,
po declines
Effects of Increasing po
This is called “valving” of gas
A
stresses along A-A
stress
gas cap
effect
oil, density
= 0.75-0.85
A
po
depth
sh
hydraulic fracture
when po > sh
gas cap,
low density
sn = sn - poil
o
fault slips when
 > (sn - po)·tanf,
i.e. when
 > sn·tanf
Fluid migration is a function of stress as well as pressure!


sn
Valving of Reservoirs & Stresses






Gas is generated and accumulated,
increasing the gas cap height
Because gas is light, it has a different
gradient of pressure (0.1 instead of 0.8 for
oil and ~1.04 to 1.10 for brine)
This causes po to increase at the crest
When po exceeds sh, hydraulic fracture!
This is like a valve that opens episodically
over geological time, releasing pressure
Responsible for shallow gas in some areas
Tectonic Stressing
lithotype
stiffness
UC sand
0.5E
shale
0.75E
stress
mud
salt is
viscoplastic
salt
sandstone
E
limestone
1.5E
stresses are
isotropic
loading
unloading
depth
assumed
initial sh
Tectonic Stressing





From its “virgin state”, unloading (rifting)
or loading (compression) can occur
This has a relatively larger effect in the
stiffer rocks (small strains = large Dsh)
Stress contrasts can be generated among
the beds (except salt)
This was proven by measurements in the
Appalachian Basin thrust zones in 1980s
This is easily simulated by numerical
models, but real data need measurements
Typical Stress Conditions
stress (or pressure)
stress (or pressure)
vertical stress. sv
vertical stress, sv
horizontal stress. sh
horizontal stress, sh
pore pressure, po
pore pressure, po
strongly
overpressured
region at depth
4 km
depth
4 km
depth
mild
overpressure
a. Gulf Coast of USA
b. Western Alberta, 100 km from Rockies
Relaxed continental margin
Tectonically stressed rocks
A Well Plan, North Sea
• classical mud weight window is too narrow; cannot avoid
instability
• low mud weight  breakouts
• high mud weight  destabilized fractured zones & losses
• breakout problems are controllable by good hole cleaning;
fracture zones are uncontrollable
Strategy:
• keep mud weight low
• manage breakouts with good hole cleaning before
increasing mud weight
• monitor cavings and mud losses for warning of
fractured zones
Stresses and Drilling
To increase hole stability, the
best orientation is that which
minimizes the principal stress
difference normal to the axis
Favored hole
orientation
sv
60-90° cone
sHMAX
shmin
sv
Drill within a 60°cone
(±30°) from the most
favored direction
sHMAX
shmin
sv >> sHMAX > shmin
sHMAX ~ sv
>> shmin
sv
sHMAX
shmin
sHMAX >> sv > shmin