Petroleum Geology

Resrvoir Rocks or Rocks as Pore Systems

Abid Hussain
Geologist
Pakistan Petroleum Limited

Aim of this Section

How the pore network controls the storage and flow of fluid through that network
The geology that controls the pore network
The petrophysics we use to measure and describe the properties of the pore
network and the reservoir engineering of fluid movement
Drainage:
Migration of non-wetting phase
fluid & displacement of the
wetting phase
(e.g. filling of HC traps)

Imbibition:
Injection of wetting phase fluid
& displacement of the nonwetting phase
(e.g. waterflooding)

Porosity
The most important fact about reservoir rocks is that, by definition,
they are not completely solid, but rather are porous to one degree or
another. The degree to which they are porous is quantified by a
parameter known as the porosity.
The fraction of the cylinder that is occupied by pore space is known
as the porosity, and is usually denoted by :
= Vp /Vb
Effective porosity which measures only the pore space that is
interconnected and can potentially form a flow path for the
hydrocarbons.

ROCKS AS PORE SYSTEMS 4INTRO

What is the role of the geologist?

To understand the acquisition and use of wireline log data to calculate porosity
and saturation

Persuade your local petrophysicist that the geologist has an important

contribution to make

Decide quantitatively what is reservoir and what is not, based on log response
and (maybe) core and test data

Divide the reservoir into rock types whose properties can be calibrated by logs (or
seismic) according to relationships you have defined

Map those rock types between wells and predict them beyond

Recognise and understand small-scale variation in flow and storage properties

and know how to average or upscale it so that the important flow effects are
preserved

Pore Type Classifications

Many ways to cut and slice it:

Mode of formation (primary, secondary, dissolution)

Descriptive (inter-, intra-granular)

Size (macro-, micro-porosity)

Connection (catenary, cul-de-sac, isolated or vug)

Special terms for carbonates

In this class: nice pores and naughty pores

All classifications are aids to thinking and a means to an end. Lets not get too hung up

Key Objective: To Represent This By a Handful of Numbers

White is grain, blue is pore and the field of view is about 2 cm high. It is a sandstone of Permian age and aeolian in
origin. I guess porosity at 20-25% and permeability in the low hundreds of mD. Cute quartz overgrowths are visible
around the coarser grains. The dark clusters are remnants of rock fragments; almost all the rest of the grains are quartz.
Note that the pore system is patchy and heterogeneous but homogeneously heterogeneous. The areas dominated by
large pores would tend to contain a higher saturation of oil and most of the permeability.

AND THESE

THIS GRAIN IS
0.1 mm

0.2 mm

THESE TOO

Sandstone Pore Systems

Key Variants
1.

Grain size, sorting

2.

Compaction, cementation

3.

Clay content and distribution

4.

Cement type and content: overgrowth (quartz), pore lining, pore blocking

5.

Matrix content (usually compacted, sand-sized mud or lithic clasts)

6.

Grain types (e.g. lots of feldspars to dissolve)

Sandstone problems* usually involve clays (authigenic or detrital),

or extreme lowness of permeability

* Problems = requiring more care than usual in analysis: departing from the routine

Pore Type Variety

Sandstones tend to be straightforward

Grain size is almost always the primary control thus facies

Diagenesis is generally a secondary influence

Carbonates have a greater range of pore types

Grain type before size

Carbonate diagenesis starts early and is much

more capable of creating porosity

Grains v matrix: sorting

Carbonate mudstone can be reservoir

Effect of granular texture on porosity

After Beard & Weyl, 1973

Effect of granular texture on permeability

After Beard & Weyl, 1973

Influence of textural & mineralogical maturity on sandstone pore

space properties & reservoir quality

Nagtegaal, 1978

Effect of clay mineral distribution

Modified after Serra, 1984

Effect of authigenic clay mineral types & habit

Types of dispersed clay in sandstone reservoirs:
discrete particle kaolinite

Photo Courtesy of PetroTech Associates 2002, Neasham 1977

RMS021882A1

Effect of authigenic clay mineral types & habit

Types of dispersed clay in sandstone reservoirs:
pore lining chlorite

Photo Courtesy of PetroTech Associates 2002, Neasham 1977

RMS021882A1

Effect of authigenic clay mineral types & habit

Types of dispersed clay in sandstone reservoirs:
pore bridging illite

Photo Courtesy of PetroTech Associates 2002, Neasham 1977

RMS021882A1

Discrete particle kaolinite:

mainly reduces porosity

Pore lining chlorite:

may retard later diagenesis

Pore-bridging illite:
about as bad as it gets for permeability reduction!

RMS021882A1

From the Middle Jurassic Brent Group,

Cormorant Field (Block II), North Sea

Pc curves vs type of dispersed clay

Pore-bridging
(e.g. illite)

Pore-lining
(e.g. chlorite)

Discrete particle
(e.g. kaolinite)

Neasham, 1977
DGH861883

Carbonate Pore Systems (1)

Key Variants
1.

Mouldic / intragranular /
intergranular

2.

Sorting: mudstone
wackestone - packstone
- grainstone

3.

Diagenesis: secondary
porosity, karst,
crystallinity

Carbonate Pore Systems (2)

Diagram re-drawn from F. Jerry Lucia, 1995. Rock-Fabric/Petrophysical Classification of Carbonate Pore Space for
Reservoir Characterization, BAAPG, 79(9), 1275 - 1300; see also Lucia, 2002, Carbonate Reservoir
Characterization, Springer, 226 pp

Carbonates

Carbonate problems
usually involve
extremes of pore
connectivity

Carbonates

Oolites as an
example of
diversity

Effectiveness of Pore Types

Connected macroporosity nice pores

Has a higher saturation than the bulk rock of the

least-wetting phase

Contains most of the permeability

Relatively easy to deal with and measure

The Rest naughty pores

Microporosity

Vugs

Fractures

Nice porosity
- always effective

Naughty Pores

Microporosity

Largely ineffective for hydrocarbon storage and

fluid flow

Vugs

Enhances conductivity and capillarity

May hold hydrocarbon, but ineffective for flow

Fractures

So effective for flow that other pore systems are

likely to be bypassed

May be counted but not quantified by wireline

Key point: to be able to recognise when nonstandard circumstances arise

Summary of carbonate pore systems

Weber & van Geuns, 2004

Porositypermeability relationships, for the Tor, Ekofisk and Lower

Cretaceous Sola and Tuxen Formation carbonates in the Danish Central
Graben.

For the same porosity, the matrix

permeability of the Ekofisk will be
ca. one tenth that of the Tor
F
r

Tor = Maastrichtian (U. Cretaceous); Ekofisk = Danian (L. Tertiary)

Influence of burial & compaction on Chalk reservoir properties:

Chalk porosity preservation during burial history

Porosity-depth relationships, Dan Field, Denmark

Subsea Depth (m)

Daan Field Porosity vs Depth

Dan Field
Porosity-Depth
0
-100
-200
-300
-400
-500
-600
-700
-800
-900
-1000
-1100
-1200
-1300
-1400
-1500
-1600
-1700
-1800
-1900
-2000
-2100
-2200
-2300

FY Log
H eidim iao Log
Putaohua Log
FY C ore

10

20
Porosity

30

40

50

Dan Field
Top Chalk
Depth Map

ROCKS AS PORE SYSTEMS 4

4.83%

PORE TYPES

0.02 mD

Measuring pore size

Inject mercury into a cleaned

and dried core plug, record
pressure and saturation

Injection record (top right) can

be converted to pore-size
distribution (bottom right)

It measures THROATS!

19.3%
1702 mD

And the Point of All That?

To think about pore types in terms of their

effectiveness

Theres physics in it and theory out there if

you want it

There are geological techniques that bridge

the gap toward petrophysical rock
characterisation

ROCKS AS PORE SYSTEMS 4 PORE NETWORKS

Summarising: (1of 3)
1. In describing the characteristics of rocks that govern fluid storage and flow we are
mainly concerned with the pore space.
2. Grain texture and composition, and their modification by diagenesis, are of interest
only insofar as they determine the physics and chemistry of the pore walls.
3. The same physics and chemistry apply in conventional and unconventional reservoirs,
although the weighting of forces and the importance of particular characteristics
increase in unconventional reservoirs.

4. Real pores are irregular, tortuous and complicated spaces but they result from
geological processes and it is part of the geologists job to understand the rock from
the pore-scale upwards.
5. The tasks are to understand the system well enough to be able to give an accurate

description and prediction, and to recognise a non-standard rock when you see it,
adapting your analysis accordingly.
6. The key collaborator of the geoscientist in this task is the petrophysicist.

ROCKS AS PORE SYSTEMS 4 PORE NETWORKS

Summarising: (2 of 3)
7. Coarse-grained, well-sorted rocks have lots of big pores with wide, clean
throats, high permeability, low capillary pressure, low irreducible water

saturation, low residual oil saturation and thus a high displacement

efficiency.
8. They also permit easy and accurate log analysis and conform to standard
formation evaluation methods.

9. All these characteristics are reversed in fine-grained, poorly-sorted rocks: it

is the fundamentals of the pore network which limit hydrocarbon saturation
and recovery in poor reservoir and which also make evaluation and recovery
more difficult.

10. Conversely, the same pore network (capillarity) processes control initial trap
filling and the determination of reservoir/seal boundaries (both stratigraphic
& fault-related).

ROCKS AS PORE SYSTEMS 4 PORE NETWORKS

Summarising: (3 of 3)
11.The main reason why carbonate pore systems can be more difficult than
sandstone is solubility.

12.A sandstone is unlikely to be cemented at shallow burial and then leached

deeper but a carbonate can.
13.A clastic mudrock is probably non-reservoir forever, but a carbonate
mudrock may change (e.g through dissolution, dolomitisation, etc.).

14.The solubility of carbonates leads to a greater variety of pore types, with the
possibility of having very poorly connected pores (high porosity, low
permeability) or the opposite.
15.The physics of pore systems is deep, but its out there if that helps.

16.Pore systems can be classified according to a variety of schemes, but select

the system that addresses the question in hand.

Contents

Pore types: sandstone/carbonate/nice/naughty

Henri Philibert Gaspard Darcy

Core analysis

Permeability in lab and reservoir

Porosity-permeability plots

Capillarity

Wettability and rel perm

Henri Philibert Gaspard Darcy: Permeability

Q/A = (k/)(P/L)
Flow per unit area is proportional to mobility and pressure gradient. Mobility
is permeability divided by viscosity
Note that for constant viscosity, flow is linear with permeability. Plotting
permeability on a log scale may be convenient but is misleading
Q/A = flow per unit area
k = permeability of the rock
= viscosity of the flowing fluid

Pic 2

P/L = pressure gradient

HPGD, 1803-1858
Pic 3

and his
experiment

Units
Permeability has the dimension L2 e.g. m2.

1 m2 is a bit big

Oilfield units
Darcies and millidarcies

1 Darcy allows a fluid of 1cP viscosity to flow at 1 cm3s-1 through an

area of 1 cm2 under a pressure gradient of 1 atm/cm

1 D = 0.987 x 10-12 m2
1 mD = 0.987 x 10-15 m2

Flow is Linear With Permeability

1000

900

800

700

600

500

400

300

200

100

0.001
2770

2770

2870

2870

2970

2970

3070

3170

0.010

0.100

1.000

3070

3170

LINEAR SCALE:

LOG SCALE:

PROPERLY REPRESENTS
FLOW POTENTIAL

CLUTTER!

10.000

100.000

1000.000

ROCKS AS PORE SYSTEMS

Contents

Pore types: sandstone/carbonate/nice/naughty

Henri Philibert Gaspard Darcy

Core analysis

Permeability in lab and reservoir

Porosity-permeability plots

Capillarity

Wettability and rel perm

Routine Core Analysis

Porosity

A known volume of helium is expanded into the sample: Boyles Law does the
rest

Sample can be any shape

Accurate to 0.5 p.u.

Also measures grain density

Permeability

Measure flow rate of helium or nitrogen through a confined sample

Only on regular plugs

Good to 15% mid range; 50% low and high*

KEY POINT: Be aware of problems in sample preparation and handling

* I have not found an analysis of this. In my experience, the accuracy of the measurement is much
less of a problem than the preparation of the sample and the heterogeneity of the rock

Routine Core Analysis Process

Remove from sleeve Inspect Continuity log

Core Gamma CT scan Preserved samples

Take plug samples
Clean and dry Plugs
Porosity permeability
Grain Density
Overburden Phi/K

SCAL subset
Hotshot Phi/K

Bag and label trims

Fluid saturations

Hg-Injection

Petrography

Klinkenberg

Slab Core Resinate - Photograph

Final Report

A CONSTANT DIALOGUE BETWEEN CLIENT PETROPHYSICIST/OPS

GEOLOGIST AND LAB IS CRUCIAL

Routine Core Analysis

Core Labs kit, the CMS-300

Manufacturers spec:

18 samples at a time,
automated

0.05 D to 15 D

0.01-40%

280-9,800 psi

Cleaned and dried samples

ROCKS AS PORE SYSTEMS 4 METHODS

From Reservoir to Lab

1.

Sit around quietly for 100 million years

2.

Notice approaching vibration

3.

Flushed with mud filtrate

4.

Rattled around in core barrel

5.

3000 psi pore pressure, 3000 psi confining pressure to 14 psi and zip in a few hours

6.

100C to 10 C

7.

Dropped on rig floor

8.

Laid in core box, dropped, shuffled, put back in core box upside down

9.

Truck / boat / helicopter ride

10.

Sit around on a pallet for 3 months in baking sun / freezing cold / driving rain

11.

Plugged

12.

Boil in solvent

13.

Bake in oven

..Measure porosity and permeability

Delicate Clay Textures

Clay textures can be wrecked in the process of preparing samples: the effect of
different drying methods on illite texture is well documented

Dried at the critical-point: fluffiness

preserved

Air-dried: bad hair day

From
Kantorowicz,
1992

Grain Density
2.70

2.75

2.80

2.85

2.90

2.95

3.00

3000

The data are a by-product of the

porosity measurement

Most useful geologically in

3025

3050

carbonates: calcite v dolomite v

anhydrite

In clastics, can be a check for

3075

3100

anomalous mineralogy or a measure

3125

of carbonate cementation
3150

Used by petrophysicists to calibrate

density logs, thus in calculating

3175

porosity and acoustic impedance

3200

CALCITE

DOLOMITE

ANHYDRITE

How Much Data is Enough?

In principle, every bed should be sampled with sufficient frequency to give

a stable mean

There is no answer to this that is both practical and statistically rigorous in

vertical sections of sedimentary rocks

A 2-step empirical approach is necessary:

1.

has the core sampled the full range of log response of the
reservoir?

2.

What is the visible heterogeneity of the core compared to

the 1-ft or 0.3-m plugging?

Core plugs are likely to miss the extremes, which is important only if the
extremes are features big enough to influence well performance

A 25x25x1 m geomodel cell is 8 million times bigger than the 3 core plugs
it might contain

GO SEE THE CORE

Step 1: Representative geological & petrophysical core analysis

measurements of all key facies & flow units

RMS976006

Modified After Ebanks, Scheihing, Atkinson, 1992

Step 2: Full petrophysical & petrographic analyses

Sneider et al, 1984

Step 3: Facies/rock type/flow unit to log calibration

Sneider et al, 1984

Step 3: Rock type/flow unit to log calibration (cont.)

Sneider et al, 1984

Step 4: Rock type/flow unit to fluid flow calibration

Can. Well Log, Soc.

After Connally & Reed, 1983

Step 5: Net pay definition

STEP

PROCEDURE

1. Geologically
Characterize Reservoir

Core Description, Wireline Log Calibration, Lithofacies

Determination, Depositional Environment Analysis

2. Determine Reservoir
Properties

Core Analysis (Porosity, Permeability, Fluid Saturation),

Wireline Log Analysis (Porosity, Fluid Saturation)

3. Delineate Reservoir
& Nonreservoir
Rocks & Characterize
Pore Space Geometry

Porosity / Permeability Crossplots, Thin Section Petrography,

Pore Cast Electron Microscopy, Mercury Injection Capillary
Analysis; Apply Cut-off Criteria

4. Evaluate Pay and

Nonpay

Mercury Injection Capillary Analysis, Fluid Saturation

Analysis; Apply Cut-off Criteria

5. Confirm Pay Zones

Measure Well Performance Using Spinner, Temperature,

Flowmeter Data And Production Results; Observations
Noted During Drilling, Including Shows
After Gaynor & Sneider, 1992

Step 5: Net pay definition-summary

Contents

Pore types: sandstone/carbonate/nice/naughty

Henri Philibert Gaspard Darcy

Core analysis

Permeability in lab and reservoir

Porosity-permeability plots

Capillarity

Wettability and rel perm

Something to Bear in Mind.

Lab-measured permeability is lovely, but for the prediction of

reservoir behaviour we need to know the relative permeability to
hydrocarbon at reservoir conditions, and its behaviour as saturation
varies

ROCKS AS PORE SYSTEMS 4 METHODS

Correcting to In Situ Conditions

Labs measure the permeability to nitrogen at 400 psi. Few reservoirs

match those conditions

It is necessary to correct for liquid, overburden and two phases

1.

Liquid

2.

44

Calibration

Measure a subset at confining pressure

At least 2 phases

Klinkenberg correction

Because gas flow in pores is less affected by grain surfaces

Overburden

3.

44

44

Relative permeability

SCAL (Special Core Analysis)

SADLY, ALL THESE CORRECTIONS ARE DOWNWARDS

ROCKS AS PORE SYSTEMS 4 METHODS

Correcting to In Situ: Klinkenberg

Make measurements at
different gas pressures, plot k

against 1/P (reciprocal

pressure) and extrapolate to
1/P = 0 to find Klinkenberg
corrected perm

Done by the core analysis lab

as a standard procedure

average gas permeability, mD

40
35
30
25
20
15
10
5
0
0

0.2
1/P, 1/atm

0.4

0.6

ROCKS AS PORE SYSTEMS 4 METHODS

Correcting to In Situ: Overburden

Routine phi v. percent left at overburden
100

Charts show the percentage of the

80

property measured at standard

60

conditions (400 psi) which is

retained at overburden pressure

40

20

The reservoir is at 2900 m,

normally pressured, mechanically

0
0

10

20

30

robust.

40

Routine k v. percent retained at overburden

100

1 across a range of percentage

80

60

Perm correction looks wicked.

Multiply by 0.8 would cover a lot of

40

it, but is the scatter to do with

20

geology? More analysis required

0
0.00

Porosity correction is easy: subtract

0.01

0.10

1.00

10.00

100.00

1000.00

ROCKS AS PORE SYSTEMS 4 METHODS

Correcting to In Situ: Overburden

SPE 56813

Chart shows the cumulative

frequency of permeability retained at

pressure: 50% of the samples (red
arrow) retain 40% or less of
unstressed k (blue arrow)

GoM and California sands: typically

30% porosity, 2 D pemeability at
standard conditions (300 psi), so
they are mechanically weak

ROCKS AS PORE SYSTEMS 4 METHODS

Exceptions and Pitfalls

Acquisition, Transport and Preparation

You have choices in all these which will affect the quality of the data
Consult the experts: the drillers, your petrophysicist, your Ops group and the
service companies

Sampling and Measurement Processes

Rough plugs

Isolated vugs

Damage to rock during coring: weak rocks and delicate textures

Damage to rock texture during preparation

The Dataset

How much data is enough?

Bad sampling

Under-sampling the crumbly bits

Go see the core and check

ROCKS AS PORE SYSTEMS

Contents

Pore types: sandstone/carbonate/nice/naughty

A very small piece of pore network theory

Henri Philibert Gaspard Darcy

Core analysis

Permeability in lab and reservoir

Porosity-permeability plots

Capillarity

Wettability and rel perm

Gus E Archie

M. Gassman and Herr Zoeppritz

ROCKS AS PORE SYSTEMS 4 PHI v k

Porosity v Permeability

Lets say that the

colours represent
porosity and there are 4
different facies in this
volume

How then do you do the

permeability?

ROCKS AS PORE SYSTEMS 4 PHI v k

Porosity v Permeability

Fundamental to understanding reservoir rock types

Not much use unless integrated with grain size, facies, thin section

Replete with waffly, ineffective description but limited understanding

There is theory, but rocks are messy

Applications:

Define rock types

Pick layers

Selection and application of SCAL data

Whats a rock type?

A lithology that can be distinguished stratigraphically or geographically or by its wireline log

characteristics and which has flow properties that can be deduced reliably in uncored wells. Something
measurable from the logs (e.g. porosity, GR value) can be used as input to a rule for calculating
permeability and choosing cap pressure and rel perm curves. This definition assumes you want to
know about flow. If you just want storage (volumetrics), life is easier.

Sandstone 1: Grain-size Control

9/8a-B02 C ore Plu g Data

10000

MU l
ML u
l
TheML
standard
is to plot

1000
1000

porosity linear on the xFU u

Permeability, mD

Khg
(mD)

axis, permeability
FU l

100
100

FL u

logarithmic on the y
FL l

Colours mark different

grain-size classes

10
10

Black squares are the

coarsest facies, red is

11

0.1

finer, green is finer again

10
0

11

12

13

14 10
15

16

17

18

19

Porosity, %
Porosity (%)

20
20 21

22

23

Sandstones 2
10000
1000
100

Permeability, mD

10

Several datasets

are shown in order

to see behaviour

0.1

across 9 orders of
magnitude of

0.01

permeability, from

Tight Gas Sand, East Texas

0.001
0.0001

Uncemented sst, Germany, fg or greater

ten darcies to ten

Uncemented sst, Germany, vfg or less

nano-darcies

Brent Group, Ninian

0.00001
0

0.05

0.1

0.15

0.2

Porosity

0.25

0.3

0.35

A Model Sandstone

Bryant and Cade, 1993. Permeability prediction from

geologic models, BAAPG, 77, 1338-50
Cade, Evans and Bryant, 1994. Analysis of permeability
controls a new approach. Clay Minerals, 29, 491-501

The chart is from BP work dating from

the early 90s. It shows the calculated
permeability and porosity of a
digitised random bead pack

They took a random pack of spherical

1000 um (vc/c)

500 um (c/m)

250 um (m/f)

125 um (v/vf)

62.5 um (vf/silt)

Compaction + pore-fill

Pore-fill only

1,000,000

beads, all the same size. The pack

100,000

was solidified, sectioned and the

Knowing the bead centres and bead

diameter, porosity and permeability
could be calculated

Compaction was modelled by moving

the bead centres closer vertically:
cementation by growing the beads or
by nucleating blockages in pore space

There are some approximations along

the way, but the output of the model
matches nicely what rocks do, and
heres the important bit helps in

understanding the rocks

Permeability, mD

position of the bead centres recorded

10,000
1,000
100
10
1
0.00

0.05

0.10

0.15

0.20

Porosity

0.25

0.30

0.35

A Model Sandstone So What?

106

1,000,000

Probably not great for

predicting permeability in real
rocks

100,000

Permeability, mD

10,000

103

1,000

100

K DISAPPEARS STEEPLY
10

0.00

0.05

0.10

0.15

0.20

Porosity

0.25

0.30

0.35

But it sure helps in interpreting

porosity-permeability plots

1 ,0 0 0 ,0 0 0

1 0 0 ,0 0 0

ROCKS AS PORE SYSTEMS 4 PHI v k

1 0 ,0 0 0

1 ,0 0 0

10000

Sandstone
control
9/8a-B021:C great
ore Plugrain-size
g Data

1 00

1 0

0. 00

0. 05

0. 10

0. 15

0. 20

0. 25

MU l
1 ,0 0 0 ,0 0 0

ML u
1 0 0 ,0 0 0

1000
1000
1 0 ,0 0 0

Permeability, mD

Khg
(mD)

1 ,0 0 0

100
100

ML l
FU u
FU l
FL u

1 00

FL l
1 0

10
10

0. 00

0. 10

0. 15

0. 20

0. 25

0. 30

The colour-coding indicates

grain size: Black squares are
the coarsest facies, red is
finer, green is finer again

11

0.1

0. 05

10
0

11

12

13

14 10
15

16

17

18

19

Porosity, %
Porosity (%)

20
20 21

22

23

Gradient and relative

positions match the model

0. 35

0. 30

0. 35

1 ,0 0 0 ,0 0 0

1 0 0 ,0 0 0

Sandstones 2

1 0 ,0 0 0

1 ,0 0 0

1 00

1 0

0. 00

0. 05

0. 10

0. 15

0. 20

0. 25

0. 30

10000
1000
100

125 m, compaction
line: right where the
green triangles should
be

Blue triangles sit on a

finer grain-size trend

0.1
Brent samples show a
wide range of grain size
and a trend indicating
cementation and
compaction

0.01
0.001
0.0001

Permeabilities by Rock Type

Uncemented sst, Germany, fg or greater

1,000.000

Uncemented sst, Germany, vfg or less

100.000

Brent Group, Ninian

0.00001
0

0.05

0.1

0.15

Not nearly
as good a fit

0.2

Porosity

0.25

Per m eabi l i t y, D

Permeability, mD

10

10.000

Type 1
Type 2
Type 3

1.000

0.3

0.35

0.100
0.1

0.15

0.2
Porosity

0.25

0.3

0. 35

Travis Peak Fmn, E Texas a Tight Gas Sand

LIFE

MODEL

100

1,000,000

100,000

10
10,000

A microscope view.
White is pore, black
is grain

1,000

100

0.1
10

0.01
1
0.05

0.10

0.15

0.20

0.001

0.25

0.30

0.35

Permeabilities by Rock Type

1,000.000

Pore-filling cement

0.0001

100.000

Tight Gas Sand, East Texas

0.00001
0

0.05

0.1

0.15

0.2

No use

Per m eabi l i t y, D

Permeability, mD

0.00

10.000

Type 1
Type 2
Type 3

1.000

0.100
0.1

0.15

0.2
Porosity

Porosity

0.25

0.3

SPE17729, 39917

Modified after Lucia 1995, BAAPG. See also Carbonate Reservoir Characterisation by Jerry Lucia (Springer, London, 366 pp., 2007)
1 ,0 0 0 ,0 0 0

1 0 0 ,0 0 0

Carbonates 1: Lucias system

1 0 ,0 0 0

1 ,0 0 0

1 00

1 0

0. 00

0. 05

0. 10

Grain-dominated

1000.0

Mixed

Permeability, mD

100.0

10.0

Crystalline
1.0

Mud-dominated
0.1

These must be porous wackestones

and mudstones and they sit on a grainsize trend

0.0
0.00

0.05

0.10

0.15
Porosity

0.20

0.25

0.30

0.35

0. 15

0. 20

0. 25

0. 30

0. 35

1 ,0 0 0 ,0 0 0

1 0 0 ,0 0 0

Carbonates 2: Chalk

1 0 ,0 0 0

1 ,0 0 0

1 00

1 0

100.00

0. 00

0. 05

0. 10

0. 15

0. 20

0. 25

0. 30

Mid East Aptian Chalk

Permeability, mD

10.00

1.00

These are porous

wackestones and
mudstones and they sit
on grain-size trends: the
Danian Chalk is finer

Maastrichtian Chalk
0.10

Danian Chalk
0.01
0.00

0.05

0.10

0.15
0.20
Porosity

0.25

0.30

0.35

0. 35

1 ,0 0 0 ,0 0 0

1 0 0 ,0 0 0

Carbonates 3: Suspect Device Field

1 0 ,0 0 0

1 ,0 0 0

1 00

1 0

0. 00

1000.000

Permeability, mD

100.000

10.000

1.000

0.100

Coloured according to reservoir layer. Theres

a lot of rock here and the chart is a big mess,
but the layering is starting to sort things out

0.010

0.001
0%

5%

10%

15%

20%

Porosity, %

25%

30%

35%

40%

0. 05

0. 10

0. 15

0. 20

0. 25

0. 30

0. 35

1 ,0 0 0 ,0 0 0

Carbonates 3: Suspect Device Field

1 0 0 ,0 0 0

1 0 ,0 0 0

1 ,0 0 0

1 00

1 0

0. 00

Porosity
1000.000

Permeability, mD

100.000

10.000

1.000

0.100

0.010

0.001
0%

5%

10%

15%

20%

25%

30%

35%

Same original facies (oolites again), similar porosity,

buried in the same location 400 ft apart vertically:
1000x difference in perm

40%

0. 05

0. 10

0. 15

0. 20

0. 25

0. 30

0. 35

1 ,0 0 0 ,0 0 0

1 0 0 ,0 0 0

Suspect Device: Dolomites 1

1 0 ,0 0 0

1 ,0 0 0

1 00

1 0

0. 00

1000

100

This lot looks grainy and

are organised along a
grain-size trend

Permeability, mD

10

0.1

These are crystalline dolomites: the

steepest phi-k relationship youll see
other than fractured plugs

0.01

2 wells: Coloured according to reservoir layer

0.001
0%

5%

10%

15%

20%
25%
Porosity, %

30%

35%

40%

45%

0. 05

0. 10

0. 15

0. 20

0. 25

0. 30

0. 35

1 ,0 0 0 ,0 0 0

1 0 0 ,0 0 0

Suspect Device: Limestones

1 0 ,0 0 0

1 ,0 0 0

1 00

1 0

0. 00

1000

0. 10

0. 15

0. 20

No correlation between
por and perm: these are
oomouldic

100

Permeability, mD

0. 05

10

0.1

This layer is a bit of a mess, but

theres a trend to part of it
suggesting grainstones

0.01

2 wells: Coloured according to reservoir layer

0.001
0%

5%

10%

15%

20%

Porosity, %

25%

30%

35%

40%

0. 25

0. 30

0. 35

1 ,0 0 0 ,0 0 0

1 0 0 ,0 0 0

Suspect Device: Classification Without Understanding

1 0 ,0 0 0

1 ,0 0 0

1 00

1 0

Permeability, mD

0. 00

Porosity, %

0. 05

0. 10

0. 15

0. 20

0. 25

0. 30

0. 35

1 ,0 0 0 ,0 0 0

1 0 0 ,0 0 0

And the Point of All That?

1 0 ,0 0 0

1 ,0 0 0

1 00

1 0

0. 00

Most wells are not cored, so permeability has to be derived, usually from porosity

Understanding a bit about pore networks and poroperm relationships allows more

0. 05

0. 10

0. 15

effective selection of rock types, which should allow more effective description of the
reservoir (and better selection of cap curves, rel perm)

Understanding comes from looking down the microscope, visiting the core and talking
to your colleagues

This report is a compilation of 70 published datasets

http://pubs.usgs.gov/of/2003/ofr-03-420/ofr-03-420.html

0. 20

0. 25

0. 30

0. 35

ROCKS AS PORE SYSTEMS 4 PHI v k

Who Does What?

Petrophysicist

Coordinator of core analysis programme

Custodian of core analysis data

Geologist

Informed user of core analysis data

Coordinator of core description programme

Reservoir Engineer

Informed user of core analysis data

Service Company

Will have broad experience and advice, but not necessarily the detailed
knowledge of your reservoir

N.B. Workflow sheet

Contents

Pore types: sandstone/carbonate/nice/naughty

Henri Philibert Gaspard Darcy

Core analysis

Permeability in lab and reservoir

Porosity-permeability plots

Capillarity

Wettability and rel perm

Capillarity

Contents

Some basic science

The practical effects of the physics how it works in rocks

(it controls the static distribution of fluids in the reservoir)

Capillary entry pressure

Saturation v height and free-water levels

How we deal with it measurement and quantification

Bouyancy, entry pressure and saturation

IN A NUTSHELL
The greater the
buoyancy force,
the smaller the
pores that can be
entered by the oil,
overcoming
capillary
resistance

Basic concepts

Buoyancy versus capillary pressure

Fluid pressure gradients v. API gravity

Some typical fluid pressure gradients

Material
A: Sediment/rock
B: Salt sat. brine
C: Pure water
D: 60 API oil
E: Dry gas

Pressure Gradient
A: 1.00 psi/ft
B: 0.50 psi/ft
C: 0.433 psi/ft
D: 0.321 psi/ft
E: 0.100 psi/ft

ROCKS AS PORE SYSTEMS 4 CAPILLARITY

The Result is a Hydrocarbon Column

RFT data
Top of transition zone

Depth

Transition Zone

Irreducible water

Immobile Oil
Entry pressure

Deepest oil production

Practical OWC range
Lowest detectable oil
Free Water Level
0

Swirr

Water Saturation

Sor

Simplified air-mercury injection capillary pressure

laboratory measurements

After Core Labs

Simplified air-mercury injection capillary pressure

laboratory measurements: Step 1 (initial conditions)

After Core Labs

Simplified air-mercury injection capillary pressure

laboratory measurements: Step 2 (initial Pd)

After Core Labs

P1= Pd = Displacement Pressure

Simplified air-mercury injection capillary pressure

laboratory measurements: Step 3 (intermediate Pc)

After Core Labs

Simplified air-mercury injection capillary pressure

laboratory measurements: Step 4 (Pc at Swi)

After Core Labs

Typical air-mercury Pc curve

Range of Pc curves for different rock types

Drainage v. Imbibition
Drainage:
Migration of non-wetting phase
fluid & displacement of the
wetting phase
(e.g. filling of HC traps)

Imbibition:
Injection of wetting phase fluid
& displacement of the nonwetting phase
(e.g. waterflooding)

Pore Throats

Permeable rock may be treated as a

network of capillary tubes

Oil must overcome capillary forces to pass

through pore throats

Standard technique is mercury porosimetry

Cap pressure curves may change with

overburden stress

To get to Sw v h, convert air-mercury to

oil-brine (PcB = PcA(.cos)B/(.cos)A)
and pressure to height (h = Pc/. g)

ROCKS AS PORE SYSTEMS 4 CAPILLARITY

Apparent Stepped OWCs

Are the lower 4 horizons coarsening or fining up?

ROCKS AS PORE SYSTEMS 4 CAPILLARITY

In Practice

Very commonly, Sw v h is
picked from log data

FWL may not be known

Rocks are messy

But each rock type should

have a distinctive cap curve

The Leverett J-Function is a normalisation of capillary pressure for permeability and

porosity
Diagram from GTA Manual M30

ROCKS AS PORE SYSTEMS 4 CAPILLARITY

Curious Transition Zone Effect

PORE AND
THROAT
HIGH IN
THE OIL
COLUMN

IN THE
TRANSITION
ZONE

LARGE PORE

MEDIUM PORE
SMALL PORE

Residual oil saturation is less in the transition zone

Check out SPE87921: the saturation effect is small but the rel perm
difference is very significant: more oil can be recovered Diagram from SPE71609

ROCKS AS PORE SYSTEMS 4 CAPILLARITY

In Practice

Capillary forces determine the static state of the reservoir

Volumetric calculations need to take account of Sw v h and thus

capillary pressure, but small in good reservoir and negligible for
gas

Reservoir simulation models need cap pressure information

because it affects hydrocarbon movement

Cap pressure is relatively easy to relate to geologic controls

because it is strongly determined by grain size and sorting

Cap pressure by mercury-vacuum, air-brine or oil-brine is

standard SCAL technique

ROCKS AS PORE SYSTEMS

Contents

Pore types: sandstone/carbonate/nice/naughty

A very small piece of pore network theory

Henri Philibert Gaspard Darcy

Core analysis

Permeability in lab and reservoir

Porosity-permeability plots

Capillarity

Wettability and rel perm

Gus E Archie

M. Gassman and Herr Zoeppritz

ROCKS AS PORE SYSTEMS 4 WETTABILITY

Oil-wet v Water-wet

N.B. This is
a real pore
system

DIAGRAMS MODIFIED FROM GTA MANUAL M6

In the oil-wet case, the pore wall

has a greater electro-chemical
affinity for oil than for water

ROCKS AS PORE SYSTEMS 4 WETTABILITY

Example of a Non-Wetting Surface

(when properly polished, of course)

ROCKS AS PORE SYSTEMS 4 WETTABILITY

Wettability Tips
Tending to oil-wet:
Oil with a high concentration of polar compounds
Kaolinite, chlorite

Coarse-grained, high porosity (because lower Swirr)

Heavy oils in carbonates
Tending to water-wet:
Oil with low concentration of polar compounds
Quartz, feldspar, illite
Fine-grained, low porosity (because higher Swirr)
Water-wetness is predominant (all reservoirs start out water-wet), although in detail
many rocks are probably of mixed wettability, simply because they are so
heterogeneous at the microscopic scale. Wettability is a complex phenomenon of
hydrocarbon and surface chemistry and electrostatics
Modified from Worden et al, Petroleum Geoscience, 4, 130

ROCKS AS PORE SYSTEMS 4 2-PHASE FLOW

Relative Permeability
FULLY SATURATED WITH
OIL: NO PERMEABILITY
TO WATER

Kr = FRACTION OF
ABSOLUTE

PERMEABILITY

BOTH WATER AND OIL

FLOWING

A
0
0

SW
OIL NO LONGER
CONTINUOUS:

residual oil saturation

irreducible water
saturation

DIAGRAMS MODIFIED FROM GTA MANUAL M6

Relative permeability curves

Relative permeability v. Capillary Pressure

ROCKS AS PORE SYSTEMS 4 2-PHASE FLOW

Tight Gas Rel Perm

Tight gas
reservoir
No fluids flow
THIS IS CONTROVERSIAL, but
it allows me to make a point
about the importance of rel
perm curves

Regular
reservoir
Two fluids flow

From Shanley et al., BAAPG, 2004

And the Point of All That?

Capillary pressure and wettability are the rock properties that determine

fluid saturation

Wettability affects electrical properties, rel perm and reservoir

performance (accelerates water breakthrough; retards, but may not

reduce, oil recovery)

Fluid movement is powerfully determined by rel perm. Rel perm can be

difficult to measure, especially in poor reservoir. Tweaking the rel perm
curves is a favourite dodge of reservoir engineers

ROCKS AS PORE SYSTEMS 4 ROCK PROPERTIES

What I Was Trying to Say (1)

1. This section took us from the porosity-permeability territory that is most familiar to
the geoscientist through to the flow properties of pore systems occupied by more than
one phase of fluid.
2. Capillary pressure describes the way a pore system prefers to hold on to a particular
phase. When oil or gas are migrating into a reservoir the capillary pressure resists their
entry and has to be overcome by buoyancy. The balance between buoyancy and

capillary pressure is what determines the hydrocarbon saturation.

3. Poor reservoir (smaller pores) has a greater cap pressure so it is more difficult to get
hydrocarbon in. Any reservoir will show different hydrocarbon saturation between the
larger and smaller pores, if they could be isolated.
4. During a waterflood, the high capillary pressure rock preferentially attracts the water.
At the small scale the pace of the displacement is accelerated or retarded according to
capillary pressure behaviour.

ROCKS AS PORE SYSTEMS 4 ROCK PROPERTIES

What I Was Trying to Say (2)

5. The wettability of a rock is determined by the electro-chemistry of the hydrocarbon,

water and mineral surfaces.

6. All reservoirs are initially water-wet, which is to say that water lines the mineral wall
while the hydrocarbon sits in the middle, not touching the mineral surface at all.
7. Some reservoirs become oil-wet, lots probably are mixed.
8. Oil-wetness affects the flow behaviour. From an oil-wet reservoir production overall
will be more watery (the water rel-perm curve is higher than for oil-wet) but if you can
handle the produced water the ultimate oil recovery can be greater because the oil
tends to remain as a continuous film to low saturations. This film can be squeezed and
flowed (the oil rel perm curve goes to zero only at very high water saturation).

ROCKS AS PORE SYSTEMS 4 ROCK PROPERTIES

What I Was Trying to Say (3)

9. Relative permeability curves are what controls most directly the flow in the reservoir.
The fluids in the pore space share between them the absolute permeability of the dry
rock, which is what we measure in routine core analysis.
10. The relative permeability curves describe how this sharing occurs as saturation
changes.

11. Getting the relative permeability curves right is extremely important for modelling
reservoir performance and is measured in the lab as part of the bag of routines we call
Special Core Analysis (SCAL).
12. It can be expensive and time-consuming and is usually done on not nearly enough
samples.
13. There is geology in the choice and use of relative permeability curves, but these
secrets are usually held deep in the heart of reservoir engineering territory.