Pore Pressure and Methods of Analysis Explained

I have worked exclusively in pore pressure for many years now, culminating in this
course I wrote and taught internationally. The course requires two days of instruction.
Each day I will post sequentially course sections. Ultimately, at the end of the course,
you will have a greater understanding of what pore pressure is and how it is analyzed.

Purpose

This course was generated to give upstream exploration and development

professionals a familiarization with the most commonly used methods of creating a
pore pressure and fracture gradient prediction for locations to be explored, and
discoveries to be appraised and developed.

The course was designed to increase the understanding by engineers and geoscientists
of the causes of pore pressure, utilization of seismic, log and drilling information to
analyze pore pressure and fracture gradients, and how the models are adjusted and
calibrated to generate a pore pressure gradient and fracture gradient predrill prediction.

An understanding of the commonly used pore pressure and fracture gradient methods
of analysis will give the end user engineer and geoscientist a greater ability to judge
and apply the pore pressure and fracture gradient prediction when planning the well.
Scope

There are two commonly used methods for pore pressure analysis in the oil and gas
industry, equivalent depth method and ratio method. This course discusses the most
commonly used method, the equivalent depth method.

Several different pore pressure and fracture gradient equivalent depth method models
are used in the oil and gas industry. This course discusses the Eaton and Bowers pore
pressure compaction models, and the Matthews and Kelly fracture gradient model.

A generalized description of clastic deposition geology is described, then a description

of pore pressure generation in impermeable shale.

Impermeable rock pore pressure modelling principles are described. The relationship
between impermeable and permeable pore pressure is described.

An offset well is used to demonstrate how the data is analyzed to develop the
parameters used to predict pore pressure and fracture gradient at a prospect location.

Introduction

Pore pressure is assessed by analysis of “impermeable” rock formations subject to the

effects of compaction.
“Impermeable” is a time-dependent term. As burial progresses, so does compaction.
With increased compaction the porosity is reduced. As the porosity is reduced, the
water in the pore spaces being reduced must move. The flow rates of this movement
are relatively very low, referenced to geologic time. Formations with flow rates
referenced to clock time are considered “permeable”. Formations with flow rates so
low it is impractical to reference to clock time are considered “impermeable”.

The pore pressure of “impermeable” rock has never been measured. Therefore, there
are no standard measurements of the pore pressure of “impermeable” rock in specific
conditions to compare and use as a reference. Thus, the study of pore pressure is
nebulous and the results are inexact. This characteristic creates a significant range of
uncertainty.

Pore pressure assessment is accomplished by comparing available data to

observations, using mathematics and logic. This course describes the math, and
demonstrates examples of how logic is used.
Pore Pressure and Methods of Analysis
Explained, Section 1 Depositional Sequence,
Compaction Factors, and Porosity Decline
Part 1. Pore Pressure and Clastic Deposition

Pore pressure is the pressure of the fluid within the porosity of the rock.
Porosity is divided into two categories, effective porosity and ineffective
porosity. Effective porosity is pore spaces connected with each other that
allows fluid to flow from one pore space to another. Ineffective porosity is
isolated pore spaces which fluid cannot flow from one pore space to another.

There are no verifiable methods of assessing pore pressure of ineffective

porosity. Verifiable methods of assessing pore pressure of rocks with effective
porosity require the rock to be compactible by the overburden stress. These
methods are known as “compaction models”. If the rock has been compacted
to the degree that the rock matrix is capable of supporting the total overburden
stress, the compaction models are ineffective. For a compaction model to be
effective, the rock must be able to be further compacted to reduce porosity
with increased burial and overburden stress.

Compactible rocks are most commonly from clastic deposition. Clastic

deposition results from consolidated rock being eroded and transported to the
location of deposition. Erosion is simply the breaking of larger rocks into
smaller rocks. Rock particle sizes range from boulders to clay particles.
Transport is by wind or water. Wind is responsible for a very minor portion of
clastic deposits, and water is the dominant transport mechanism.
Figure 1 Sedimentary Depositional Environments

Part 2. Depositional Sequences, Compaction Factors, and Porosity Decline

with Depth

A depositional sequence is a continuous deposit of rock clastics from

the same source being eroded, and the resulting clastics being transported to
the location of deposition. The particle size range of clastics being deposited at
a specific location is dependent on the energy of the water at that location. A
specific energy level will move, thus the particle sizes deposited at a specific
location will change within a depositional sequence. As an example, the energy
of a marine depositional system is highest at the shoreline, thus the sand is
deposited, and lowest away from the shoreline where the clay particles
settle. For an example of the energy of the water changing, causing different
particle sizes to be deposited, transgressions and regressions cause the energy
of the water to change at a specific location.

Figure 2. Coastal Depositional Sequence

Because a depositional sequence requires a supply of clastics from the

same source, the geochemistry of the clay will not significantly change during
deposition. The clay mineral will have a certain rock strength to resist fracture.
After the initial reorganizing of the orientation of clay particles in the
shallowest depths, compaction requires the clay particles to fracture to
facilitate a reduction of porosity. The rock strength of the clay particles is a
dominant factor in the compactability of clay sediments, given a specific
change in overburden stress during a specific duration of time. The
compactability factor of a clay formation member will determine the
parameters of the pore pressure model used to assess the pore pressure of that
clay member.

Part 3. Particle Size of Clastic Deposits and Clay Particles

Compaction analysis requires the use of clay formation members. Clay

particle size is less than 0.002 mm, or less than 0.00008 inch (1/12,700 inch).
Figure 3. Largest Clay Particle Size

The formation fluid must migrate out of a pore space being decreased.
The thickness of the clay particle determines the width of the path the
formation fluid must transit. The thickness of the clay particle is nominally
about one tenth of the maximum lateral dimension. The thickness of the clay
particle determines the area of the path of the fluid that migrates out of a pore
space being reduced by compaction.

Figure 4. Clay Particle Thickness

Part 4. Compaction and Porosity Decline with Increased Burial

Clay particles are deposited with random orientation creating a

porosity of as much as 70% at the depositional surface. The initial compaction
with burial is accomplished by reorientation of the clay particles. As the clay
particle reorientation progresses, a threshold is reached where further
compaction is accomplished by fracturing the clay particles. Mostly the edges
of the clay particles are deformed, allowing the particles to become closer and
reducing the volume of the pore space between them.

Initially, normal clay porosity is greater than sand porosity. As burial

increases the clay is compacted more thoroughly than the sand, and the
porosity of clay is less than that of sand until the sand begins to develop
cementation in its pore spaces. The continued cementation of sand reduces the
porosity of sand at a greater rate with burial than compaction of clay.

Figure 5 Porosity Decline of Clay and Sand

Pore Pressure and Methods of Analysis
Explained; Section 2 Fluid Flow and Pore
Pressure Gradient in Clay

Part 1. Clay Diagenesis

Clay diagenesis is a complex subject. A detailed discussion of clay
diagenesis is beyond the scope of this course. The basic transition phases will
be identified: compaction, chemical reconstitution, and lithification. Because
pore pressure analysis models assess compaction, the course concentrates on
compaction.

Compaction results in the clay “dewatering”. The porosity decreases

with increased burial, thus there is less water in the bulk volume, causing the
density to increase.

The most common chemical reconstitution of clay is the transition of

smectite to illite, which generates water as a byproduct, adding to the water
that was squeezed out as a result of compaction.

Crystallization is the ordering of clay crystal lattices that occurs when

the clay particles are compacted sufficiently close for adjacent crystal lattices
to form a relatively weak chemical bond. The hardness of a shale is related to
the aggregate strength of the chemical bonds.
Part 2. Clay Compaction and Fluid Flow Path
In the shallowest section, clays have the fluid properties such as viscosity and
surface tension of the unbound water between clay particles to help support
the overburden. As burial begins, the clay formation member begins to
compact and the unbound water begins to be squeezed out from between clay
particles. Pore pressure will not begin to increase above hydrostatic gradient
until the unbound water feels a restriction to its generally upward movement
created by the clay particles becoming closer.
Figure 6 Compaction Sequence of Clay

The path of the water through clay is very indirect. Because clay particles
deeper have been subjected to greater compaction that resulted in decreased
porosity, and shallower less compaction, the general direction of the water is
upward.
Figure 7 Irregular Fluid Flow Path in Clay

The reduction of porosity of clays is the result of compaction due to increasing

overburden as additional deposition and burial progress. Gravity is the
dominant compaction dewatering mechanism in shallow sediments.

Compaction increases with depth reducing porosity, therefore, the path of least
resistance of water being squeezed out of the space between clay particles is
toward the less compacted sediment, or a generally upward direction. As the
clay particles become closer with increased burial, the water from deeper
sediments is increasingly restricted in its generally upward movement.

As compaction progresses, the percentage of the total rock becomes less water-
filled porosity, and more solid rock matrix. As the proportion of water-filled
porosity decreases, log responses will establish a trend in accordance with the
rate of decreasing porosity with depth. The trend established is a compaction
trend. A compaction trend should not be confused with a Normal Compaction
Trend. A Normal Compaction Trend is defined by normal pressure, which is
hydrostatic. Because the pressure is not established in the figure below, it
cannot be assumed the pressure is hydrostatic. Therefore, the only known is
the decreasing porosity with depth, and the pressure is unknown, thus it is a
simple compaction trend.

Figure 8 Decreasing Porosity with Depth and Associated Log Responses

Part 3. Inviscid Fluid Flow in Clay

Inviscid fluid flow is flow without the effect of viscosity.
There are no frictionless surfaces. The velocity of any fluid at the surface-fluid
interface is always zero. The flow velocity will increase with distance from the
surface-fluid interface. Ultimately the flow velocity will become constant when
sufficient distance from the surface-fluid interface allows inviscid flow (flow
without the effect of viscosity). The pore pressure gradient will increase only
when inviscid flow is not possible between the source of the formation water
and its destination.

Figure 9 Fluid Velocity Relationship with Fluid-Surface Interface

The horizontal axis is distance from the fluid-surface interface. The
vertical axis is flow velocity. The blue trace rising from the graph origin and
reaching an asymptote with the horizontal dashed green line is the flow
velocity of a fluid particle as it passes upward through the horizontal axis as
distance from the fluid-surface interface increase. The horizontal dashed green
line is the inviscid flow velocity. The vertical dashed blue line is the distance
from the fluid-surface interface where inviscid flow begins.

When inviscid flow is achieved, the pressure gradient is constant

throughout the inviscid flow region. Between the inviscid flow region and the
fluid-particle interface where the flow velocity is zero is the boundary layer,
labelled δ in Figure 10.

Figure 10 Fluid Velocity Boundary Layer and Inviscid Flow Region

When inviscid flow is possible between the clay particles, the only contribution
to the force felt at point B will be the in-situ force at point A and the addition of
the force due to gravity on the fluid between point A and point B. Therefore,
there is no change to the pressure gradient between point A and point B when
inviscid flow is possible.

When clay particles are close enough to not allow inviscid flow, a proportional
amount of the frictional force of the boundary region is added to the in-situ
force at point A and the force due to gravity on the fluid between point A and
point B. Therefore, the pressure gradient at point B is greater than at point A
due to the addition of the frictional force.
Figure 11 Inviscid Flow Region and Frictional Force on the Fluid Only
Within the Boundary Layer

To review some principles:

 The force of gravity on the overlying sediment causes clay deposits to
compact. The compacting clay forces clay particles closer to each other, forcing
the water to move out of the decreasing pore volume of the pore space below,
and into the pore space above.

The force of gravity on the water being displaced by the water coming
from the deeper, more compacted clay, hydrodynamic frictional losses along
the path the water travels, and the viscosity of the water determine whether
the pore pressure gradient remains constant along the path of flow, or
changes.

In the shallowest sediments the porosity of shale is so high, the space

between particles causes frictional resistance to flow and effects of viscosity to
be negligible.

“Normal Pressure” is hydrostatic pressure. When the hydraulic head of

the water above is hydrostatic, and when the total resistance to flow due to the
porosity/aspect ratio/viscosity/flow rate relationship of the shale and in-situ
water is negligible, the pore pressure remains hydrostatic throughout the flow
path.
When the hydraulic head of the water above is greater than
hydrostatic, and when the total resistance to flow due to the porosity/aspect
ratio/viscosity/flow rate relationship of the shale and in-situ water is
negligible, the pore pressure gradient is unchanged throughout the flow path.

To review the clay particle size and scale of flow regime:

Figure 3 repeat Clay Particle Size

Figure 4 repeat Clay Particle Thickness

Figure 12 Fluid Flow Path Dimensions in Clay

When the width of the path of the water between two clay particles is
greater than 2δ, there is inviscid flow between the clay particles, and there is
no difference of pressure gradient along the path of the inviscid flow. Thus, the
pore pressure gradient does not change.
Figure 13 Fluid Flow Path Between Clay Particles Allowing Inviscid Flow

Part 4. Non-inviscid Fluid Flow in Clay

When the width of the path of the water between two clay particles is
less than 2δ, the clay particles are too close to allow inviscid flow. The
frictional force from one fluid-particle interface meets the frictional force from
the opposite fluid-particle interface, and there is not an inviscid flow region
between the two boundary layers. Thus, the frictional force is added to the
force of the mass between points A and B, and the pore pressure gradient at
the source of the flow is greater than the pore pressure gradient at any point
along the flow path.
 The pore pressure gradient relationship between the source of the flow
and the destination is determined by the flow rate and ability to achieve
inviscid flow. There are two conditions that cause the fluid migrating from
compacting sediments to not achieve inviscid flow, the particles becoming
closer due to compaction, and increased flow rate.

Figure 14 Fluid Flow Path Between Clay Particles Not Allowing Inviscid
Flow
The blue traces represent the fluid velocity from each fluid-surface
interface. Assuming a constant flow rate, he clay particles have become
sufficiently close to not allow inviscid flow.
Figure 15 Increased Flow Rate Not Allowing Inviscid Flow
The blue traces represent flow velocity between the clay particles with
increasing flow rate, thus increased flow velocity with each increased flow rate.
Ultimately, a flow velocity is achieved that does not allow inviscid flow, and the
pore pressure gradient at the source of the flow is greater than at the
destination
Pore Pressure and Methods of Analysis
Explained; Section 3 Pore Pressure Modeling
Principles

Part 1. Compaction, Pore Pressure, and Porosity Decline Rate

Pore pressure is the pressure of the fluid within the pore space. Effective stress
is the grain-to-grain stress of the rock particles surrounding the pore space.
The overburden stress is the combined force of gravity acting on the total mass
above a specific depth of investigation.

“Normal” pressure is hydrostatic pressure.

All compaction models are impermeable rock models. The relationship

between the pore pressure of an impermeable rock and the pore pressure of a
permeable rock is assessed with the theory of centroid. Centroid calculations
will be addressed later in this section.

Compaction as a result of increased burial causes the porosity to decrease. As

the porosity decreases, the fluid contained in the pore space being decreased
migrates out of the pore space. The pore space will decrease at a rate with
increasing overburden stress with depth, given the fluid migrating out of the
decreasing pore space is not restricted. When the fluid migrating is restricted,
the pressure of the fluid increases and introduces an additional force resisting
compaction. With the addition of this additional force, the porosity decline
rate deviates from the porosity decline trend established at the shallower depth
where the fluid migrating was not restricted.
Pore pressure compaction models identify changes in porosity trends of
impermeable rock, and associate that porosity trend change with a change of
pore pressure gradient.

Part 2. The Equivalent Depth Method

A pore pressure model that uses the Terzaghi relationship between pore
pressure, effective stress and overburden stress is an equivalent depth method
model.

The normal compaction trend is the trend of any physical property (resistivity,
acoustic impedance, density, etc.) through the normally pressured strata.
Below the normally pressured strata the pore pressure gradient increases, and
the porosity decline rate will be different than in the normally pressured strata
above. Log data and interval velocity trends will also change in accordance
with the change of the porosity decline trend.

The equivalent depth method is defined by a relationship between pore

pressure, effective stress, and the overburden stress described by the Terzaghi
relationship. The Terzaghi relationship assumes the overburden is supported
by the sum of the pore pressure and the grain-to-grain contact, termed the
effective stress, of the rock matrix. The accepted relationship is expressed by
the Terzaghi Principle, which states the downward force of the overburden is
equal to the sum of the upward forces of pore pressure and the effective stress.
Figure 16 Terzaghi Principle

An equivalent depth method model requires the use of a normal compaction

trend. The equivalent depth method assumes that there is a depth section over
which the pore pressure is hydrostatic, and the sediments are normally
compacted because of the systematic increase in vertical stress with depth.
When the log of a measured physical property (resistivity, acoustic travel time,
density) is plotted as a function of depth, normal compaction trends are fitted
to the data over the normally pressured interval, and extended through a
minimum distance of at least the base of the depositional sequence. Where the
data set deviates from the normal compaction trend below the hydrostatic
pressure region, the difference between the data set and the normal
compaction trend at a specific depth is used to calculate a pore pressure
gradient not equal to hydrostatic pressure.

Because the value of the measured physical property is related to effective

stress, which in turn is related to the overburden stress, an approximation of
the pore pressure at any depth where the measured value is not on the normal
compaction trend line can be calculated.

Often, because the physical properties are similar, the same normal
compaction trend is adequate for the subsequent depositional sequence. More
often, the normal compaction trend must terminate at the base of the
depositional sequence where the hydrostatic pressure interval was used to
define the normal compaction trend, and a new normal compaction trend
must be established to model the deeper formations.

Part 3. Pore Pressure Model Principles

Most common pore pressure models use either resistivity, sonic travel
time, or interval velocity data. This course references models using these data
sets.

As depth increases, porosity declines due to compaction. As porosity declines,

the percentage of water-filled porosity of the rock declines. Log response will
be in accordance with the effect of decreasing water-filled porosity.

Where the pore pressure gradient is constant with increased depth, the
porosity decline trend will be unchanged.
 Where the pore pressure gradient increases with depth, compaction is
resisted by the increased pore pressure gradient, and the porosity decline rate
deviates from the porosity decline trend.

The log response trend will change as the trend of water-filled porosity
changes. The change in the trend of water-filled porosity with depth is
associated with a change of pore pressure gradient.

Pore pressure calculations are based solely on true vertical depth.

A hypothetical normal compaction trend assumed to be normally

pressured is plotted with the resistivity, sonic, or interval velocity data.

Impermeable intervals are selected for analysis, and the resistivity or

sonic travel time, value are used with the normal compaction trend value to
calculate a pore pressure for each shale interval selected.

Permeable intervals are excluded from the model.

Part 4. Resistivity Models

Resistivity responds to the geochemistry of the rock.

The resistivity of a formation is indicative of the geochemistry. The water-filled

porosity in the formation is part of the geochemical constituents the resistivity
tool is measuring. Formation water is approximately 400 times more
conductive than the typical rock matrix. A change in water-filled porosity is
detected by the resistivity tool, and the trend of porosity decline due to
compaction can be determined by applying a trend to the resistivity log at
similar shale intervals.

Figure 17. The Normal Compaction Trend of a Resistivity Model and Pore
Pressure Calculation

Part 5. Sonic Models

 Acoustic travel time responds to the physical property of the rock. The
same physical properties that dictate the sonic travel time recorded by a
logging tool also dictate the interval velocity. In a sonic log model, the clays
can be separated from sandier formations, where this is very difficult with
seismic interval velocity data. As a result, the sonic model can be more precise
because the clays can be separated from the sands, while interval velocity
averages the sands and clays within the practical depth of resolution of seismic
surveys. Aside from this difference, the model principles are the same for
interval velocity as for sonic travel time.

The acoustic travel time of a formation is indicative of the physical

property. The water-filled porosity in the formation contributes to the acoustic
travel time the sonic tool is measuring. The acoustic travel time of water is
approximately 3 to 4 times the acoustic travel time of the typical rock matrix. A
change in water-filled porosity is detected by the sonic tool, and the trend of
porosity decline due to compaction can be determined by applying a trend to
the sonic log at similar shale intervals.
Figure 18. The Normal Compaction Trend of a Sonic Travel Time Model
and Pore Pressure Calculation

Part 6. Centroid Relationship Between Permeable and Impermeable Rock

The centroid theory assumes a dipping permeable formation is bound
by an impermeable formation, and at some point along the boundary the
pressure of the permeable formation is equal to the pressure of the
impermeable formation. This depth is called the “centroid”.

The permeable rock has effective porosity (hydraulically connected

pore spaces). The pressure at any depth is equal to the hydraulic head of the
fluid column within the permeable rock. The impermeable rock does not have
effective porosity (refer to the introduction of this course and the discussion of
permeability with respect to clock time and geologic time). The pressure at any
depth is determined by the compaction of the surrounding geology. Below the
centroid, the impermeable rock pressure is greater than the permeable rock
pressure. Above the centroid, the impermeable rock pressure is less than the
permeable rock pressure.

Figure 19. Hydraulic Pressure Calculations Within a Permeable Sand

Bounded by Impermeable Clay

A well is drilled at balance and penetrated a sand body at the centroid depth of
10,000 ft. Pore pressure gradient of the sand and shale at the centroid is 11.6
ppg, which is 0.6 psi/ft. The pressure at the centroid is (0.0519*11.6*10,000)
psi, which is 6020 psi.
At 1,000 feet above the centroid, the shale pressure decreased by 0.6 psi/ft, or
602 psi. The shale pressure 1,000 feet above the centroid is 6020-602, which
is 5418 psi at 9,000 ft. The pressure gradient for 5418 psi at 9,000 feet is 11.6
ppg.

The sand contained water and the measured pressure at the centroid depth
was 6020 psi. The formation water has a pressure gradient of 0.44 psi/ft. At
1,000 feet above the centroid, the sand pressure decreased by 0.44 psi/ft, or
440 psi. The sand pressure above the centroid is 6020-440, which is 5580 psi.
The pressure gradient for 5580 psi at 9,000 feet is 11.94 ppg.

At 1,000 feet below the centroid, the shale pressure increased by 0.6 psi/ft, or
602 psi. The shale pressure 1,000 feet below the centroid is 6020+602, which
is 6622 psi at 11,000 feet. The pressure gradient for 6622 psi at 11,000 feet is
11.6 ppg.

At 1,000 feet below the centroid, the sand pressure increased by 0.44 psi/ft, or
440 psi. The sand pressure 1,000 feet below the centroid is 6020+440, which
is 6460 psi. The pressure gradient for 6460 psi at 11,000 feet is 11.32 ppg.
Figure 20. Centroid Example Calculation Solution Spreadsheet
The psi per foot is linear with depth. The equivalent pressure gradient
calculations are not linear with depth. Graphical solutions using a linear trend
to project a centroid value over a significant depth interval will not be accurate.
The centroid value for pressure gradient must be calculated.

Part 7. Model Calculations Reference Height

All pressure gradient calculations are equivalent to the density of the
height of a fluid from the reference height of the model to the depth of the
calculation. Given the height of the fluid column, the pressure generated will
be determined by the density of the fluid.

Pressure gradient units are usually the density of the hypothetical fluid (ppg,
sg, etc.), but sometimes are expressed in change in pressure given a change in
depth (psi/ft, pa/m, etc.). To convert from sg and g/cc to ppg, 1.0 sg is equal to
8.3454 ppg. The most common reference heights used are the surface
elevation, mean sea level, and the elevation at the drill floor (RKB).

Figure 21. Model Reference Heights of MSL and RKB

Figure 22. MSL and RKB Model Calculations: Water Depth of 700 feet
and RKB Height of 100 feet
Figure 23. MSL and RKB Model Calculations: Onshore
Pore Pressure and Methods of Analysis
Explained; Section 4 The Pore Pressure and
Fracture Gradient Modeling Process; The
Overburden Gradient

I begin to describe the mechanics of generating a PPFG model in this section. I

am segmenting section 4 into the individual phases of constructing a PPFG
model. This post includes Parts 1 and 2, culminating with the generation of the
calculated overburden gradient.

Section 4. The Pore Pressure and Fracture Gradient Modeling Process

Part 1. Compaction Model Characteristics

Resistivity and sonic travel time models detect changes in the trend of
porosity of impermeable formation members to calculate changes to the pore
pressure gradient. Where available, the data through the normally pressured
section of the well is used to establish the normal compaction trend. When the
normally pressured section of the well is not logged, the normal compaction
trend is established by assessment of a calculated pore pressure with the data
collected and observations made while the well was drilled.
The d-Exponent model is not a compaction model because it does not detect
changes in the porosity trend, whereas compaction models do. The d-
Exponent model uses the relationship between the hardness of the rock and
the difference between annulus pressure and pore pressure to calculate a
parameter used to calculate a pore pressure gradient.

There are seven basic steps to constructing a PPFG model:

Calculate OBG

Select shale intervals

Define normal compaction trend

Calculate pore pressure gradient

Calculate effective stress ratio

Calculate fracture gradient

Calculate permeable formation hydraulic heads

Part 2. Calculating the Overburden Gradient
Because equivalent depth models base all calculations on a
relationship with the overburden gradient, the Overburden Gradient is the
core element of the pore pressure model. If the overburden gradient is not
representative of the actual vertical stress, all calculations of pore pressure and
fracture gradient will be erroneous. Thus, a representative overburden
gradient is critical to the applicability and confidence of the pore pressure
model.

The calculated overburden stress at any depth is a function of the combined

effect of the force of gravity on the total mass above and tectonic stresses. In
the Oil and Gas Industry, the overburden stress has never been measured. The
stress has been measured in other industries such as mining. They utilize
strain gauges to measure the actual stress in the formation. Placing a strain
gauge in an open borehole has not been practical, thus the overburden stress
has never been measured in a well.

In pore pressure modelling, the overburden stress at any depth is assumed to

consist of the combined effect of the force of gravity on the total mass above
that depth without including the effect of tectonic stresses.

The calculated overburden stress at any depth is the sum of the force applied
by the total mass above that particular depth. As depth increases, the density
of the formation is not constant. The overburden stress is calculated by
dividing the strata into a continuous series of discrete elements from the
surface or mud line to a specified depth, then adding incrementally the force
due to gravity of each element, beginning at the model reference height. At any
specified depth, the calculated overburden stress will be the force due to
gravity applied by the total mass above that depth.

The calculated overburden gradient is equal to the equivalent density of a fluid

of a column height equal to the distance from the specified depth to the
reference height (such as a drill floor or mean sea level) necessary to generate a
force equal to the calculated overburden stress.

The densities used to calculate the overburden stress are from density logs,
density transforms using sonic travel time and interval velocity, and an
assumed shallow density profile. Very seldom is there density log
measurements for the shallowest portion of the subsurface. The shallow
density profile begins at the surface for onshore locations, and at the mud line
for offshore locations. Determining the shallow density for a surface location is
relatively easy. Go to the location, collect a sample of the surface strata, and
measure the density. For offshore locations the mud line density is almost
always not a data element collected when surface coring is conducted, and is
not known.

Most often, a mud line density and shallow density profile for the offshore
location must be assumed.

Part 2a. Shallow Density Profile Assumption

The most commonly applied method for assuming shallow density is the Miller
method. The Miller method assumes a non-linear increase of density from the
surface or mud line to a depth where normal compaction begins and a more or
less linear increase of density with depth as compaction increases.

The default Miller options are near sediment source, and distant from
sediment source. The near sediment source assumes larger average grain size
sandier deposition, and distant from sediment source assumes a smaller
average grain size clay fraction being deposited. The sediment source may be
the mouth of a river, or the termination of a submarine canyon, or any other
location where the sediment being deposited is transported from immediately
prior to deposition.

The Miller shallow density algorithm assumes the density of the clay at the
mud line for distant from sediment source is 1.435 g/cc, and for near the
sediment source is 1.516 g/cc.

The equation for the Miller shallow density profile is proprietary and cannot be
shown in this article.
Figure 24. Example Miller Shallow Density Profiles and Calculated
Overburden Gradients

The choice of shallow density options affects the shallowest portion of the
overburden gradient. The effect diminishes with increased depth. Also, this
example is at a water depth of approximately 6,700 feet. In shallower water,
the difference in calculated overburden gradients would be greater.

At the two-dimensional boundary of the water and seafloor the density of clay
is relatively constant over a large area. Immediately below this two-
dimensional surface a location will have a characteristic clay density called the
“mud line density”. The mud line density of clay varies with rate of deposition.
The more rapid the deposition, the less time the clay is compacted for a
specific depth. High rates of deposition will have lower densities in the
shallowest sections below the mud line than lower rates of deposition.

Local rate of deposition is affected by shallow geologic structures.

Time, in addition to the overburden stress, is a factor influencing the degree of
compaction. At the apex of a structure, the depositional sequence being
deposited today will be thinner than away from the crest of the structure. For a
given depth, the deposits at the crest of the structure will have been subjected
to compaction for a longer period of time than away from the crest of the
structure. Therefore, mud line density of clay is greater at the apex of the
structure than away from the crest of the structure.

Figure 25. Variation of Mud Line Density Related to Shallow Geologic

Structure
 The mud line density of the Miller shallow density profile is adjusted
to conform to the rate of deposition inferred by the shallow geologic structure.
Mud line density usually varies between 1.62 and 1.87 g/cc, depending on the
depositional environment of the location and the position on the local geologic
structure. However, where deposition is most rapid and consists of only clay
fraction particle size, mud line density can be as low as 1.44 g/cc. The pore
pressure analyst must determine the depositional environment of the location
and the most likely mud line density.

The choice of mud line density has a more significant effect on the
calculated overburden gradient in shallower water than deeper water. In the
examples below, the mud line density was varied from 1.51 to 1.87 g/cc. With a
water depth of 6,743 feet, the difference between overburden gradient
calculations at 2,500 feet below the mud line was 0.139 ppg. When the water
depth was changed to 4,132 feet, the difference between overburden gradient
calculations was 0.202 ppg.

In deep water exploration, the surface casing shoe is normally set at

approximately 2,500 feet below the mud line. A difference of overburden
gradient of 0.139 is not significant in the practical drilling environment.
However, when the difference becomes 0.2 ppg, it becomes significant when
analyzing problems encountered during drilling or running casing.

Most commonly, a mud line density measurement is not available. The

analyst must use his best judgment, considering the complete geologic system,
when choosing a shallow density option.
Figure 26. Water Depth 6,743 feet: Mud Line Density Variations Effect on
Overburden Gradient Calculation
Figure 27. Water Depth 4,132 feet: Mud Line Density Variations Effect on
Overburden Gradient Calculation

Part 2b. Deriving Density Below the Shallow Density Profile

Below the shallow un-compacted zone the densities are taken from
offset well density logs, or calculated using sonic log data or interval velocity
with a Gardner transform. If a partial density log is available from an offset
well, the Gardner transform must be calibrated to match the measured density.
 The formula for the Gardner Δt density conversion is:

RHOB = c (106/∆t)e
where:
∆t = sonic transit time, µsec/ft, µsec/m
c = empirical constant (default 0.23 when ∆t is expressed in μs/ft)
e = empirical constant (default 0.25 when ∆t is expressed in μs/ft)

The Gardner calculated density must agree with accepted industry

shallow density and deeper measured density. If there are no measured density
data to calibrate with, the uncertainty of the overburden calculation becomes
significant.
Figure 28. Gardner Density Transform Calibration

The composite density data set used to calculate the overburden

consists of the Miller shallow density estimate, the Gardner calculated density
from sonic data, and the measured wireline density.
Figure 29. Overburden Gradient Calculated Using Composite Density
Data Set

Part 2c. Effect of Water Depth on Calculated Overburden Gradient

The overburden gradient is calculated using the total mass above the
depth of the calculation. For different water depths, the percentage of density
attributed to water varies at a specific total depth will be different for each
different water depth. Water has a density of 1.0 to 1.05 g/cc, and rock
densities normally range from 2.2 to 2.5 g/cc below the depth of shallow
density. Given the same rock density below the mud line, the total overburden
stress at 30,000 feet and a water depth of 3,000 feet will be greater than the
total overburden stress at 30,000 feet true vertical depth and a water depth of
9,000 feet.

It is not appropriate to generate a general overburden gradient and

apply it to an area of varying water depths. The resulting pore pressure models
will not be representative of the environment.

The graph in Figure 30 shows the different overburden gradient

calculations below the mud line using the same density profile with different
water depths.
Figure 30. Effect of Various Water Depths on the Calculated Overburden
Gradient
Pore Pressure and Methods of Analysis
Explained; Section 4. Part 3. Selection of
Impermeable Strata and Elimination of Permeable
Formation Members

Part 3. Selection of Impermeable Strata and Elimination of Permeable

Formation Members
Of all the models I’ve reviewed enforcing QA and exercising QC, if
there were inadequacies and/or weaknesses in the models, the most common
was due to not sufficient time and effort applied to selection of impermeable
shale points for input into the model. Selecting shale points is tedious and
laborious. However, a rigorous examination of the logs and a thorough
evaluation and selection of shale points has a dramatic effect on the precision
and credibility of the pore pressure model.

After the overburden gradient is calculated, the modelling process can begin.
Compaction pore pressure models utilize impermeable formation members,
thus it is necessary to eliminate the permeable formation members from the
data being used to calculate pore pressure. To accomplish this, only clays are
selected for analysis, and sands and permeable limestones are eliminated from
the pore pressure calculations.

All pore pressure models are impermeable formation models. There is not a
verifiable model that determines the pressure of a permeable formation
because the pressure of the fluid in a permeable formation is a function of the
hydraulic head, and a change in fluid pressure has very little effect on
resistivity, sonic travel time, interval velocity, or density. Thus, the information
and data collected by oil exploration is not sufficient to directly predict the
pressure of a permeable formation. Prediction of pressure in a permeable
formation requires direct pressure measurement at an offset location, and/or
assumptions of continuity and hydraulic head calculations using the centroid
concept.

Geophysical models (resistivity, sonic, interval velocity) are compaction

models that detect changes in the porosity trend and associate that porosity
trend change with a change of pore pressure.

Shale intervals are selected for analysis in the resistivity and sonic models. The
seismic velocity will combine sands and shale. Sometimes the analyst may be
able to discriminate between thick shale formation members and sandy
members with seismic surveys. However, most commonly the resolution of
seismic is not sufficient to allow this with a significant degree of certainty.

For a single depositional sequence there will be a recurring shale deposit that
can be identified on the logs and used for analysis. On the surface of the area of
deposition, there are various energy levels determining the grain size
distribution of sediments deposited. As burial progresses, a specific grain size
distribution will migrate around regions within the area of deposition. A
specific grain size distribution will move away from a specific location, then
after an amount of burial return to the location. After multiple episodes of a
similar grain size distribution returning to the same location, in a vertical
sequence penetrated by a well bore there will be a recurring clay that can be
chosen for use in the pore pressure model.

It is beyond the point of diminishing returns to expend the time and effort to
attempt to identify the exact same grain size distribution in clay throughout a
depositional sequence. This is to be taken into consideration by the analyst,
and their selection of shale points included in their model must take this
characteristic of sedimentation into consideration. The more similar the shale
points selected, the more precise the model will be.

For a specific normal compaction trend, the shale intervals selected must be
similar in geochemistry for the resistivity model, and physical characteristics
for the sonic and interval velocity models.

Most commonly, only the Gamma Ray log is considered for selection
of clay deposits to include in the pore pressure model. All available logs should
be used to select shale intervals. Potassium-rich sources result in high Gamma
Ray for shales, and lower Gamma Ray for silica and carbonate sands. However,
there are occurrences of feldspar sands with a high Gamma Ray value. All
available logs should be used to verify shale picks.

Increased silt in shale will cause a slight reduction of Gamma Ray, often
undetectable, considering the characteristics of data collection of Gamma Ray
tools. When the silt content becomes significant, micro permeability develops
and introduces water into the model causing the resistivity to decrease
unrelated to pore pressure, as well as the Δt to increase unrelated to pore
pressure. Use of resistivity and sonic tolls help identify these zones so they can
be eliminated from the model.

Increased carbonate in shale will cause a reduction of Gamma Ray, and like
increased silt content, it is not precisely detected by the Gamma Ray tool. This
increase in carbonate will cause an increase in resistivity unrelated to pore
pressure, and cause the model to calculate lower pore pressure when the actual
pore pressure gradient did not change, or an increase in pore pressure gradient
was not calculated when the pore pressure gradient increased.

Cemented and lithified zones cause the resistivity to increase and the Δt to
decrease unrelated to pore pressure. Cementation and lithification will have
the same effect the above change in carbonate content had on the calculated
pore pressure gradient being less than actual.

For the purpose of simplicity and brevity necessary for the allowable duration
of this class, we will demonstrate clay selection characteristics using the
Gamma Ray log. It is the responsibility of the analyst to understand how to use
multiple logs to determine lithology, and to use that knowledge.
Figure 31. Example Shale Picks by Maximum Gamma Ray Peaks and Pore
Pressure Calculations
Shale points are selected by establishing a shale baseline in the
Gamma Ray track. For every Gamma Ray value greater than the value of the
Shale Baseline, the corresponding data points in the resistivity data set are
selected and included in the model for calculation of pore pressure.

For this example, only the peaks of the Gamma Ray log are chosen.
The black dots in the resistivity track are the resistivity data points
corresponding to the Gamma Ray values greater than the Shale Baseline. A
Normal Compaction Trend Line (NCTL) is established associated with the
selected resistivity data points in accordance with the definition of normal
compaction trend. The shallowest section is assumed to be hydrostatically
pressured, and the NCTL was drawn to approximate that portion of the data
set.

At approximately 5,900 feet the pore pressure calculation indicates a

slight rise above hydrostatic pressure, then returning to hydrostatic at 6,300
feet where it begins to rise again, returning to hydrostatic at 6,700 feet. Given
the geology is a continuous clastic depositional sequence with no effective
pressure seals, this is highly unlikely. The determination of what depth the
pore pressure rises above hydrostatic cannot be determined with any
significant degree of confidence. Also, the line drawn to connect the resulting
calculations appear to be very angular, and an increased degree of precision is
desired. To accomplish this, we increase the number of resistivity data points
selected by moving the Shale Baseline to the left (lower value of Gamma Ray),
selecting a wider range of clay characteristic.
Figure 32. Example Number of Shale Picks Increased by Moving Shale
Baseline to Increase Number of Resistivity Data PointsF
Shifting the Shale Baseline to the left increased the number of
corresponding resistivity data points selected for input into the pore pressure
model. In the above example, broad strokes of the Shale Baseline are used to
select shale points. Using broad strokes produced a relatively irregular line
when the pore pressure gradient calculations for each resistivity data point
selected are connected with a line. Again, there are two depths where the pore
pressure calculation indicates the pressure rises above hydrostatic, at 6,500
feet and 6,700 feet.
 To achieve a more representative resistivity data set of shale picks,
shale points can be individually selected using small segments of the Shale
Baseline. In significantly thick depositional sequences with relatively high
rates of deposition, this is not frequently necessary. However, in older rocks of
thinner depositional sequences, often this is the only method of acquiring a
representative resistivity data set of shale points.

Figure 33. Selecting Shale Points individually Using Small Segments of the
Shale Baseline
Individually selecting the shale points had a significant effect on the
smoothing of the calculated pore pressure gradient line. However, there are
still two possible depths where the pore pressure possibly rises above
hydrostatic, at approximately 6,300 and 6,600 feet.

A universally accepted scientific method of analyzing data within a

representative data set is the use of a moving average. Applying a moving
average to the resistivity data points selected generates a representative value
with depth of the shale selected for input into the pore pressure model.

Figure 34. Applying a Moving Average to Resistivity Data Points Selected

Applying a moving average allows the determination that the pore
pressure rises above hydrostatic at approximately 6,350 feet, and increases its
rate of increase at approximately 6,750 feet. Such precision of the model
significantly increases the confidence and validity of the pore pressure model.

Pore Pressure and Methods of Analysis

Explained; Section 4. Part 4. The Normal
Compaction Trend

Part 4. The Normal Compaction Trend

“Normal pressure” is hydrostatic pressure.

The normal compaction trend is the trend of a property of clastic

sediments (resistivity, sonic travel time, density, etc.) undergoing compaction
due to increased burial through the hydrostatically pressured region of the
depositional sequence. To properly create a normal compaction trend, it must
be created to equal the measurements of a physical property through the
normally pressured interval, then extended beyond the depth where the
measured physical property begins to deviate from the normal compaction
trend. Within a single depositional sequence, this deviation from the normal
compaction trend is assumed to represent a change in the porosity decline
rate, and is associated with a change of pore pressure gradient.

All compaction models are based on the principle that, if the pressure
remains hydrostatic, there will be a corresponding porosity decline trend with
increased depth of burial. Because the overburden stress is shared between the
pore pressure of the fluid in the pore spaces of the rock and the grain-to-grain
effective stress where the rock grains are in contact with each other, a change
in pore pressure gradient will cause a corresponding change in the ability of
the overburden stress to compact the sediments. Thus, the porosity decline
rate with depth will deviate from the porosity decline rate that would exist if
the pore pressure remained hydrostatic.

A common characteristic of the most commonly used equivalent depth

method models is they are based on a logarithmic relationship with depth.
Therefore, the normal compaction trend is not linear. On a lin-log graph, the
normal compaction trend may appear as a straight line, as with the Eaton
relationship. With the Bowers relationship, the normal compaction trend does
not appear as a straight line on a lin-log graph. A mistake I have occasionally
seen is graphically plotting a normal compaction trend as a straight line on a
linear graph. This does not allow a valid graphical solution, and the linear
graph does not allow an accurate perspective of the relationship between the
porosity decline trend and the normal compaction trend. Often it is useful to
visually examine a graph and mentally approximate the change of pore
pressure gradient. Plotting a normal compaction trend on a linear graph does
not allow this.
Figure 35. Plotting the Normal Compaction Trend on a Linear Graph

On the linear graph, the graphical representation causes the

appearance that the normal compaction trend was established in accordance
with the definition of normal compaction trend, being plotted through the data
in the normally pressured interval to a depth of approximately 17,500 feet,
then extended beyond that depth. When the same normal compaction trend is
viewed on a lin=log graph, it becomes obvious the normal compaction trend
plotted on the linear graph violates the definition of normal compaction trend,
and its values are not equal to the values of the data set in the normally
pressured interval.
 Logarithmic relationships should be displayed on a logarithmic graph.

The Normal Compaction Trend establishes the relationship between

compaction due to increased burial and the pore pressure gradient. For active
deposition, the shallowest clay sediments are suspended, and the effective
stress is zero. As compaction progresses from the surface or mud line with
depth, particles become closer and the porosity decreases. However, the pore
pressure gradient will remain hydrostatic until the porosity is reduced
sufficiently for the clay particles to become close enough to obstruct the
generally upward movement of the water being squeezed out of the decreasing
porosity below.

As the porosity decreases, the resistivity will steadily increase and the
sonic travel time will steadily decrease. The trend of the resistivity and sonic
values in this hydrostatic pressured zone defines the normal compaction trend.
Resistivity and sonic values deviating from the normal compaction trend
indicate a change in pore pressure gradient. When resistivity or sonic data are
not available through the hydrostatic pressured zone, the normal compaction
trend must be established to match existing information and data, assuming
that trend extended to the surface or mud line is representative of normal
pressure.

The most common models used to analyze pore pressure are the Eaton
resistivity and sonic models and the Bowers sonic model. The Eaton and
Bowers models are logarithmic mathematic relationships.
With the Eaton model, the normal compaction trend establishes the normal
pressure trend, and the exponent determines the change in pore pressure
associated with the difference between the normal compaction trend and the
measured resistivity or sonic value.

The Bowers model is a mathematical curve fitting model. The parameters are
adjusted for the best fit between measured sonic values and drilling
information.

The parameters established for the Eaton and Bowers sonic models should be
similar to the Eaton and Bowers interval velocity models at the same location.

After the normal compaction trend is established, the pore pressure

can be calculated based on the data from either logs or seismic surveys.

Part 4a. The Eaton Models

The Eaton models were invented using data from the depositional
basin of the northern Gulf of Mexico. The default exponents were derived from
wells drilled in this basin. The analyst must assess the adequacy of the Eaton
exponent for use in the basin within which they are working.

The default exponent for the sonic model is 3.0.

The default exponent for the resistivity model is 1.2.

Eaton pore pressure models associate a change in the trend of water-filled
porosity as depth increases with a change in pore pressure gradient. For a
given change in resistivity or sonic travel time, increasing the exponent will
increase the change of calculated pore pressure gradient.

The Eaton resistivity and sonic equations are basically the same.
However, the default exponents for each model is significantly different. Using
the sonic equation as an example:

PP = OBG – {(OBG – PP)[(∆tnct/∆tsh)^y ]}

The Eaton sonic equation is PP=[f(OBG, PP , Δtnct , Δtsh )]y, where OBG,
PP,Δtnct , Δtsh are independent variables with respect to the Eaton exponent
y, and y is an independent variable with respect to depth.
An increase in pore pressure with depth is

ΔP=dP/dD = [f(dOBG/dD, dΔtmeasured/dD, dΔtNCT/dD)]^y.

The exponent controls the change in calculated pressure for a given change of
the independent variables.

dΔP/dy= y{[f(dOBG/dD, dΔtmeasured/dD, dΔtNCT/dD)] ^(y-1)}

Conductivity unit of measurement is microsiemens/cm. (1,000
μs/cm=1mmho.) Water at 25,000 ppm will have a conductivity of
approximately 40 mmho. Rock matrix conductivity is about 100μS/cm, which
is 0.1 mmho. The ratio of conductivity of formation water to rock matrix is
approximately 400. The sonic travel time for water is approximately 200 μs/ft.
The sonic travel time for a rock matrix is 50 to 60 μs/ft. The ratio of Δtwater to
Δtmatrix is 3.3 to 4. Therefore, because the ratio of conductivity of formation
water to rock matrix is significantly greater than the ratio of sonic travel time
of water to rock matrix, the default exponent for a resistivity model is
significantly less than the default exponent for the sonic model.

The Eaton equation to calculate pore pressure gradient from resistivity is:
PP = OBG – {(OBG – PP) [(Rsh/Rnct)^x ]}
Where:
PP = Pore pressure gradient

OBG = Overburden gradient

PP = Normal pore pressure gradient

Rsh = Observed shale Resistivity

Rnct = Normal compaction trend shale Resistivity

x = Empirical exponent (default 1.2)

Comparing the Eaton sonic equation to the Eaton resistivity equation, the
parameter (Rsh/ Rnct) and (∆tnct/∆tsh) are similar, but not identically
representative of each other. The denominator for the resistivity equation is
the normal compaction trend value in the resistivity equation, while the
denominator for the sonic travel time equation is the log data set value. This is
due to the relationship of resistivity being the reciprocal of conductivity, and
sonic travel time being the reciprocal of acoustic velocity. If the resistivity
formula used conductivity instead of resistivity, or the sonic travel time
formula used acoustic velocity instead of sonic travel time, the relationship
between the normal compaction trend and the data set values would not be
reversed.

Figure 36. Eaton Resistivity Model

Figure 37. Eaton Sonic Model

Part 4b. The Bowers Models

The normal compaction trend in the Bowers model must adhere to the
definition of normal compaction trend, as is true with any equivalent depth
method compaction model.

The Bowers sonic formula to calculate the pore pressure gradient is

PP = OBG – {[1/(10^6 Δtsh) –1/(10^6 ∆tmudline )] /a}^(1/b)
Where
 a = empirical constant

b = empirical exponent

Δtsh= sonic travel time of the shale μsec/ft

∆tmudline = sonic travel time at the mud line μsec/ft

OBG = overburden gradient

PP = normal pore pressure gradient, density of the ground water

The default values for the parameters are: a = 14.2, b = 0.724, ∆tmudline =
200 μsec/ft, PP = 0.45 psi/ft. The normal compaction trend parameters are
adjusted to fit the model to the observations related to pore pressure.

(Please forgive that the image is missing. I am having severe problems getting
LinkedIn to upload images. I will continue my attempts to deliver a complete post.)
Figure 38. Bowers Sonic Model