Using Geophysical Logs to

Estimate Porosity,
^Vater Resistivity, and
Intrinsic Permeability

United States
Geological
Survey
Water-Supply
Paper 2321
Using Geophysical Logs to
Estimate Porosity,
Water Resistivity, and
Intrinsic Permeability

By DONALD G. JORGENSEN

U.S. GEOLOGICAL SURVEY WATER-SUPPLY PAPER 2321

DEPARTMENT OF THE INTERIOR
DONALD PAUL MODEL, Secretary

U.S. GEOLOGICAL SURVEY

Dallas L. Peck, Director

Any use of trade, product, industry, or firm names in this publication is for
descriptive purposes only and does not imply endorsement by the U.S.
Government.

UNITED STATES GOVERNMENT PRINTING OFFICE: 1989

For sale by the

Books and Open-File Reports Section
U.S. Geological Survey
Federal Center
Box 25425
Denver, CO 80225

Library of Congress Cataloging-in-Publication Data

Jorgensen, Donald G.
Using geophysical logs to estimate porosity, water resistivity, and intrinsic
permeability.
(U.S. Geological Survey Water-Supply Paper; 2321)
Bibliography: p.
Supt. of Docs, no.: I 19.13:2321
1. Water, Underground Methodology. 2. Geophysical well logging.
3. Porosity Measurement. 4. Groundwater flow Measurement.
5. Rocks Permeability Measurement.
I. Title. II. Series.
GB1001.72.G44J67 1988 622'.187 87-600025
CONTENTS
Nomenclature IV
Conversion factors V
Abstract 1
Introduction 1
Estimating porosity and lithology 2
Estimating water resistivity 4
Qualitative methods 4
Spontaneous-potential method 4
Cross-plot method 7
Estimating permeability 11
Background and research 11
Cross-plot method 17
Summary 19
References cited 20
Appendixes:
A. Estimating dissolved-solids concentration 22
B. Estimating sodium-chloride concentration 22
C. Estimating density 22
D. Estimating viscosity 24

FIGURES

1. Dual-porosity, gamma-ray, and caliper logs 3

2. Idealized dual-porosity log calibrated to limestone matrix and gamma-ray
log 4
3. Electric log (spontaneous-potential and resisitivity logs) 5
4. Idealized electric log of shale and sandstone sequence containing fresh
and saline water 6
5. Flow chart of spontaneous-potential method of estimating resistivity of
formation water 7
6-14, Bl. Graphs showing:
6. Measured and estimated resistivity of water 8
7. Cross plot of resistivity and porosity measured on dolostone
cores 9
8. Porosity and lithology from formation-density log and
compensated-neutron log data 10
9. Cross plot of geophysical-log values of Ro and n 11
10. Viscosity, density, and temperature of freshwater 12
11. Water viscosity at various temperatures and percentages of
salinity 13
12. Water density at various temperatures and percentages of
salinity 14
13. Relation between porosity factor and intrinsic permeability 16
14. Permeability and porosity 17
Bl. Resistivity of water as a function of salinity and temperature 23

Contents III
 TABLES

1. Water-resistivity data 8
2. Porosity and permeability data 14
3. Permeability from formation factor and pore shape 17
4. Typical cementation factors, porosities, and permeabilities and calculated
permeabilities for various lithologies 18
5. Typical range of permeabilites of various consolidated and unconsolidated
rocks 18

NOMENCLATURE

A Coefficient relating dissolved-solids concentration to sodium-chloride

concentration.
API Units defined by the American Petroleum Institute.
BHT Bottom-hole temperature in a borehole, in units specified.
Cds Concentration of dissolved solids in units specified.
CNaCt Concentration of sodium chloride, in units specified.
C(I) Tortuosity function.
Dt Total depth of borehole, in feet.
DS Dissolved-solids concentration, in units specified.
F Formation factor, dimensionless.
G Specific gravity, dimensionless.
HR Hydraulic radius, in units of length specified.
K Constant at a specified absolute temperature relating to spontaneous
potential, in millivolts.
K Hydraulic conductivity, in feet per day.
L Length of sample, in feet.
Le Effective length of flow path, in feet.
Ro Resistivity of water-rock system, in ohm-meters.
Rm Resistivity of mud, in ohm-meters.
Rmf Resistivity of mud filtrate, in ohm-meters.
Rmfe Resistivity of mud-filtrate equivalent, in ohm-meters.
Rw Resistivity of water equivalent, in ohm-meters.
Rw: Resistivity of water at 77°F, in ohm-meters.
Rw. Resistivity of water at 75 °F, in ohm-meters.
SC Specific conductance, in microsiemens.
SF Geometric shape factor, dimensionless.
Specific surface area of solids per unit volume of solids, in feet"1 .
SP Specific surface, in feet"1 .
SP Spontaneous potential, in millivolts.
Ta Absolute temperature, in degrees Rankine.
Tc Temperature, in degrees Celsius.
Tf Temperature, in degrees Fahrenheit.
Tf Temperature of formation, in degrees Fahrenheit
Tma Mean annual temperature, in degrees Fahrenheit.
Tmf Temperature of mud filtrate, in degrees Fahrenheit.
G Kozeny coefficient relates to pore geometry, dimensionless.
P Porosity factor, dimensionless.
T Tortuosity factor, dimensionless.
a Empirical constant related to lithology, dimensionless.
Ad Change of density in grams per cubic centimeter.
d Density, in units specified of mass per unit volume.
db Bulk density, which is density of matrix and fluid, in units specified of
mass per unit volume.
IV Contents
dj Density of fluid, in units specified of mass per unit volume.
dma Density of matrix, in units specified of mass per unit volume.
g Acceleration due to gravity, in specified units of length per unit of
time squared.
k Intrinsic permeability, in millidarcies.
n Porosity, dimensionless.
nd Apparent porosity from a density log, dimensionless.
nn Apparent porosity from a neutron log, dimensionless.
m Cementation factor, dimensionless.
m* Factor related to the number of reductions of the pore-size openings,
dimensionless.
p Pressure, in Pascal.
r2 Coefficient of determination, dimensionless.
\L Dynamic viscosity, in units specified units of mass per length-time.

CONVERSION FACTORS

Factors to convert measurements in units other than the International System of Units (SI units)
are presented. SI units are modernized metric units. The unit of time is s (second). The unit of mass is
kg (kilogram). The unit of length is m (meter). The unit of force is N (newton) and is that force which
gives a mass of 1 kilogram an acceleration of 1 meter per second per second. The unit of pressure or stress
is Pa (pascal), which is 1 newton per square meter.
SI units may use the following prefixes:
tera T 1012
giga G 109
mega M 106
kilo k 103
milli m 10"3
micro \>. 10"6
nano n 10~9
pico p 10" 12
The following conversions may be useful to hydrologists, geologists, and geophysicists:
acceleration due to gravity (g) = 9.807 m/s2
centipoise (cp)=1.000x 10"3 Pa.s (pascal second)
degree Fahrenheit (°F) = (1.8 TQ + 32)
foot per day (ft/day) = 2.633 X 104 m/s
foot per second-second (ft/s2) = 3.048x 10' 1 m/s2
foot (ft) = 3.048X10-' m
gallon per day per square foot [(gal/d)/ft2] =4.716X 10'7 m/s
grams per cubic centimeter (g/cm3) = 1.000 X 103 kg/m3
inch (in.) = 2.540 X 10'2 m
kilogram per cubic centimeter (kg/cm3) = 9.807 XlO4 Pa
millidarcy (mD) = 9.87x lO' 15 m2
milligram per liter (mg/L)=1.000x 10'3 kg/m3
millimho = millisiemen
parts per million (ppm) = milligrams per liter/relative density
percent salinity of sodium chloride = 10,000 ppm NaCl
pounds per square foot (lb/ft2 or psf) = 4.788XlO 1 Pa
pounds per square foot (lb/in2 or psi) = 6.895x 103 Pa
square foot (ft2) = 9.290 XlO'2 m2

Contents
Using Geophysical Logs to
Estimate Porosity,
Water Resistivity, and
Intrinsic Permeability
By Donald G. Jorgensen

Abstract where
.p=
Geophysical logs can be used to estimate porosity,
formation-water resistivity, and intrinsic permeability for This equation for k was based on a selected data base
geohydrologic investigations, especially in areas where meas- of 10 sets, which included in-place measurements of
urements of these geohydrologic properties are not available. permeability, in-place measurements of porosity from two dif-
The dual density and neutron porosity logs plus the gamma- ferent types of porosity logs, and a measurement of bulk
ray log can be used to determine in-situ porosity and to resistivity of the rock and water from a resistivity log. The equa-
qualitatively define lithology. Either a spontaneous-potential log tion has a coefficient of determination (r2) of 0.90. The rela-
or a resistivity log can be used to estimate relative resistivity tion is applicable to permeable rocks in which surface
of the formation water. conductance along grains is not dominant, such as most car-
The spontaneous-potential and the cross-plot methods bonate rocks, fractured rocks, and coarse elastics. Calculated
were tested for their usefulness as estimators of resistivity of permeability values for different lithologies using typical values
the formation water. The spontaneous-potential method uses of porosity and cementation factors compare well with typical
measurements of spontaneous potential and mud-filtrate permeability values of the different lithologies. Some data were
resistivity to estimate the formation-water resistivity. The cross- available to support the equation, but considerably more data
plot method uses porosity values and observed resistivity of will be needed to better test the equation established for k.
saturated rock to estimate the formation-water resistivity.
Neither method was an accurate estimator. However, in areas
of no data the methods can be used with caution.
A review of the literature of the basic relations for forma-
INTRODUCTION
tion factor (F), porosity (n), and cementation factor (m) implies
that the empirical Archie equation The problem of estimating rock, fluid, and
hydraulic properties becomes more important as hydrolo-
gists are asked to solve problems related to ground-water
flow in rock material about which little is known. For
is applicable to carbonates, coarse-grained elastics, and uni- example, most studies of flow systems in deeply buried
formly fractured media. The relation off, n, and m to intrinsic
formations have little or no hydraulic data because those
permeability (k) was also investigated. Merging the well-known
Archie equation with the Kozeny equation establishes an equa-
data are generally obtained from water wells, which are
tion for estimating intrinsic permeability. The resulting equa- rare in deeply buried aquifers or aquifers containing saline
tion implies that intrinsic permeability is the function of a water. Therefore, hydraulic property values of deep
medium factor, (GISS2), and a porosity factor -P, [n m + 2/(1-n) 2]. aquifers commonly are estimated using sparse and in-
Where G is the Kozeny coefficient, 5S is the specific surface of direct data. Available analytical methods have been
grains per unit of volume of solids. Unfortunately, G and Ss developed by petroleum engineers or geologists and are
values generally are impossible to determine from wireline not widely known to hydrologists who study ground-
geophysical logs.
water flow systems.
A plot of the porosity factor as a function of intrinsic
The purpose of this paper is to describe selected
permeability defined the following empirical equation for k,
in millidarcies:
techniques of using borehole-geophysical logs for water-
resources investigations. The procedure for each tech-
/c=1.828x105 (-P1 ' 10), nique, the assumptions upon which it is based, and

Introduction
measured data for comparison and evaluation are geophysical logs with special reference to water is rele-
presented for each technique. Special emphasis is given vant. A method allowing a "quick" qualitative lithology
to techniques for estimating intrinsic permeability and interpretation is available that uses the same borehole-
water resistivity. Discussion of porosity and lithology is geophysical logs used to determine porosity and to
limited to those techniques directly related to the tech- estimate water resistivity and permeability.
niques used for determining permeability and water A suitable geophysical log for determining poros-
resistivity. ity and lithology combines a compensated-neutron poros-
This paper is directed to hydrologists investigating ity log and a compensated formation-density log,
water resources in aquifers for which hydraulic data are especially if traces of both are printed on the same log
sparse, such as most deeply buried rock sections, and chart as is shown in figure 1 (right side). The combina-
assumes a cursory knowledge of geophysical logging. tion log is termed a "dual-porosity log" in this paper.
Hydraulic property values are expressed in their Figure 1 also shows a gamma-ray trace and a caliper trace
customary units. Calculated estimates are compared with (both left side). The gamma-ray log supplements the dual-
measured values allowing evaluation of the results. In porosity log in lithologic interpretations. The dual-
addition, methods of estimating in-situ water-quality porosity log is especially powerful because not only are
properties, such as dissolved-solids concentration, porosity values recorded, but also the position of the den-
sodium-chloride concentration, density, and viscosity are sity trace with respect to the neutron trace generally in-
discussed in the appendixes. dicates lithology. For example, if the density trace is to
Water resistivity is a measure of the resistance of the right of the neutron trace on a dual-porosity log
a unit volume of water to electric flow and is related to calibrated to limestone, either shale or dolostone or
water chemistry and temperature. Water that has a low dolomite is implied. The gamma-ray trace defines the
concentration of dissolved solids has a high electrical shale sequence; thus, shale and dolomite can be easily dif-
resistance, whereas water that has a high concentration ferentiated (fig. 2). If the density trace is close to the
of dissolved solids has a low resistance. Quantitative tech- neutron trace, calcite, such as limestone, is indicated. If
niques are available for identifying water resistivity a the density trace is to the left of the neutron trace, silica,
characteristic related to water chemistry. The relation of such as sandstone and chert, or a gas effect is indicated
dissolved solids and common chemical constituents to (fig. 2). As an example, figure 1 shows the top of a sand-
water resistivity commonly is known or can be deter- stone that underlies a dolostone at 2,000 ft.
mined. (See appendixes B and C.) Quantitative techniques are available for identify-
Intrinsic permeability, usually called permeability, ing simple lithologies and for correcting for shaly condi-
is a measure of the relative ease with which a medium tions. Most of the techniques use neutron, density
(rocks) can transmit a liquid under a potential gradient; (gamma-gamma), dielectric, and sonic logs to define
it is a property of the medium alone. The cross-plot porosity for cross-plotting methods. These techniques are
method determines intrinsic permeability from borehole described in log interpretation texts and manuals; some
geophysical logs using the cementation factor. uses of these techniques are presented by MacCary (1978).
Methods of estimating water resistivity and Before the compensated-porosity logs were avail-
permeability in aquifers of a porous medium whose able, porosity was determined using uncompensated
freshwater contains less than about 700 mg/L dissolved- sonic, gamma-gamma, and neutron logs". Interpretations
solids (Huntley, 1986, p. 469) were described by Biella were difficult because of unknown lithologic differences
and others (1983), Jones and Buford (1951), Alger (1966), and variations in hole size. MacCary (1983) discussed how
Pfannkuch (1969), Worthington (1976), and Urish (1981). to use uncompensated and (or) uncalibrated porosity or
The methods usually require detailed information about other geophysical logs in investigating carbonate aquifers.
the water chemistry and about the nature of the porous The resistivity versus porosity cross-plotting method can
medium. For example, to estimate permeability of a be used with these logs.
coarse-grained porous medium, the average grain size and Porosity values determined from sonic logs nor-
the uniformity coefficient, as well as the surface conduct- mally are assumed to represent nonfracture porosity.
ance of the water molecules surrounding the grains must Therefore, fracture porosity may be estimated if the
be known. Those data are seldom known. porosity determined from sonic logs is subtracted from
total porosity determined from compensated dual-
porosity logs. However, porosity values from some sonic
ESTIMATING POROSITY AND LITHOLOGY logs sometimes exceed total porosity values. Thus, the
assumption that sonic-porosity values represent only non-
Because the techniques discussed rely on data about fractured porosity cannot be made without additional in-
porosity, lithology, and rock and water resistivity from formation about conditions in the hole or about shaliness.
wireline-geophysical logs, some discussion of wireline Likewise, because the fracture porosity is usually small,

Using Geophysical Logs to Estimate Porosity, Water Resistivity, and Permeability

 CALIPER
DIAMETER IN INCHES 13 POROSITY, IN PERCENT (LIMESTONE MATRIX)
COMPENSATED FORMATION-DENSITY POROSITY
30 20 10 0 -10
GAMMA RAY, IN API UNITS
150 COMPENSATED-NEUTRON POROSITY
20 10 0

Density porosity

Neutron porosity

Figure 1. Dual-porosity, gamma-ray, and caliper logs.

Estimating Porosity and Lithology 3

 Qualitative Methods
GAMMA-RAY LOG DUAL-POROSITY LOG
Depth
Gamma-ray units Porosity Two little-known but easy to apply methods for
1
qualitatively estimating relative water resistivity use the
1
Compensated resistivity log and the spontaneous-potential log. Figure
Shale neutron \ 4 shows that within sandstone D, the resistivities, as
Gamma ray^ ^-porosity c ompensated
density recorded on both the shallow and deep traces, are equal.
porosity Therefore, the formation-water resistivity can be assumed
Limestone to be equal to the mud filtrate as measured by the shallow
trace. Thus, the formation-water resistivity is equal to the
1
l
mud-filtrate resistivity. Mud-filtrate resistivity usually is
Sandstone 1 measured, and its value can be found on the log heading.
1
1 Spontaneous potential is a function of the log-
1 arithm of the ratio of the ionic activity of the formation
1
l water to the mud filtrate. Therefore, for the SP deflec-
Dolostone 1
1 tion to be zero, as shown for sandstone D in figure 4,
1
the ratio is 1 and the activites are equal. If resistivities
and activities are assumed to be proportional, the usual
Figure 2. Idealized dual-porosity log calibrated to limestone
matrix and gamma-ray log. assumption in interpreting SP logs, it follows that the
resistivity of the water equals the resistivity of the mud
filtrate, which is usually recorded in the log heading. This
slight errors in either or both porosity logs can make frac- unique condition is useful in quickly estimating the water
ture porosity impossible to detect. resistivity at one point and for qualitatively evaluating
the relative water resistivity in overlying or underlying
permeable rocks, assuming other conditions are equal.
ESTIMATING WATER RESISTIVITY

Resistivity and spontaneous-potential logs contain Spontaneous-Potential Method

information useful in estimating water resistivity.
Resistivity logs are the most widely used and most com- The two quantitative techniques typically used to
monly available type of geophysical log. A resistivity log estimate water resistivity are (1) the spontaneous-potential
records resistance of the electrical current flow in rock method, which is the most common, and (2) the cross-
with depth. There are many types of resistivity logs; most plot method. Although both methods are usually
variations refer to the specific method of measuring, such presented in most well-logging manuals and texts, their
as a lateral log, an induction log, or a conductivity log. accuracy is not. In the present paper, estimated values
A typical resistivity log is shown in figure 3. are compared with measured values to evaluate the use
The spontaneous-potential log records the spon- and accuracy of the method.
taneous potential (SP) of the fluid-filled borehole (fig. 3). The spontaneous-potential method can be used to
The SP measured is largely an electromotive potential be- estimate resistivity of sodium-chloride-type water. The
tween the mud filtrate, the water within the rock, or the method is widely known and described in nearly all texts
adjacent saturated rock materials. and well-logging manuals, such as "Application of
The combined SP log and resistivity log is common borehole geophysics to water-resources investigations"
and is here called an electric log. A hypothetical electric (Keys and MacCary, 1971), and is based on the equation:
log of a sandstone and shale sequence is shown on figure
4. Each deeper sandstone contains water of increased Rmf
SP = -K log (1)
salinity. Two resistivity traces are shown a deep and a R\v
shallow trace. Deep implies that the resistivity measured
represents material at some distance from the well bore. where SP is the spontaneous potential, in millivolts, at
In addition, deep resistivity measurements generally the in-situ (formation) temperature; K, in millivolts, is
represent undisturbed formation water and rock. The a constant porportional to its absolute temperature within
shallow trace measures the resistivity of the material ad- the formation; Rw is resistivity of water equivalent, in
jacent to the borehole. This shallow resistivity is most ohm-meters, at in-situ temperature; and Rmf'is the
likely to be affected by the invading drilling fluid. The resistivity of mud filtrate, in ohm-meters, at in-situ
following methods can be used to estimate water temperature. This method of calculation requires spon-
resistivity. taneous potential from an electric log and the mud-filtrate

4 Using Geophysical Logs to Estimate Porosity, Water Resistivity, and Permeability

 RESISTIVITY, IN OHM-METERS LATERIAL LOG
SPONTANEOUS
POTENTIAL 02 1.0 10 100 1000 2000
I 1
MEDIUM INDUCTION
I 0.2 1.0
20 , -- 10
T -- ---100
I"' 1000 2000
MILLIVOLTS DEEP INDUCTION
0.2 1.0 10 100 1000 2000

-1,900-

;.- _g

VX^I-8*6 '8' '°9

-2,000-

-2,100-

Figure 3. Electric log (spontaneous-potential and resistivity logs).

Spontaneous-Potential Method
 ELECTRIC LOG
Rmf = 0.75 Rm (2)
^Spontaneous potential - Depth
*
Resistivity ^
The in-situ (formation) temperature (Tf) is rarely known
Shale unless temperature was measured in the borehole after
Sandstone
drilling had been completed. However, Tf may be
with
A
1 "Deep" trace-^. estimated by assuming that the temperature between the
very | "Shallow" mean annual temperature near the surface and the
fresh water trace
temperature at the bottom of the borehole (BHT) in-
Shale creases linearly with depth; mathematically, Tf can be
- Sandstone 1
shown as
with
B
fresh
water (BHT-Tma) (Df)
Tf= Tma + (3a)
Shale Dt
Shale-
Sandstone 1
with where Df is formation depth, and Dt is the total hole
]
1
C
fresh depth at which BHT-was measured. A second and similar
water 1
method of estimating 7/uses information on the geother-
Shale
mal gradient of the area:
Sandstone
with
D Tf= Tma + (geothermal gradient) (Df), (3b)
saline
water

Shale where geothermal gradient is in degrees per unit depth.

The procedure for the spontaneous-potential
Sandstone
with very 1 method to estimate the water resistivity is:
saline E
_ water 4-J1 1. Determine Rmf and Tmf. Read values from the
log heading. If Rmf is not available, estimate from Rm
using equation 2.
Figure 4. Idealized electric log of shale and sandstone se- 2. Determine SP from the spontaneous-potential
quence containing fresh and saline water. Letters identify
sandstone units whose waters range in salinity from very fresh
trace on the electric log.
(A) to very saline (E). 3. Determine 7/from temperature log or estimate
using equations 3a or 3b.
4. Determine formation-water resistivity (Rw)
resistance measurement, generally recorded on the log using the algorithm shown in figure 5.
heading. The spontaneous-potential method was used to
The spontaneous-potential method is commonly estimate Rw for 11 rock sequences in boreholes for which
used because SP logs are readily available. (More elec- the formation-water resistivity had been measured.
tric logs are available than any other type of geophysical Results of comparing Rw estimated from SP method
log.) This method of estimating water resistivity is useful versus Rw measured from chemical analyses are listed in
in sand-shale sequences where good SP differences ex- table 1 and are shown in figure 6. A least-squares analysis
ist, but reportedly is not usable or works poorly in car- for a linear relation indicates a coefficient of determina-
bonate rocks (MacCary, 1980, p. 3). However, no terms tion (r2) of 0.66. A coefficient of determination of zero
in equation 1 refer to lithology; thus, the equation should indicates no correlation and a value of 1 indicates perfect
be applicable to any permeable rock type. correlation. The value of 0.66 indicates that a correla-
An algorithm, similar to the one presented by tion exists. Examination of figure 6 indicates that a scat-
Bateman and Konen (1977), for using SP to determine ter of about 1 order of magnitude might be expected.
Rw is shown in figure 5. Mud-filtrate resistivity values The coefficient of determination of 0.66 may be
(Rmf) at specific temperature (Tmf) and the in-situ typical for the method if logs from the petroleum industry
temperature of the permeable material (Tf) at which the are used. The value of the coefficient might be interpreted
SPis measured are required. The SP value, in millivolts, as an inaccurate estimator. However, water resistivity in
is the signed (4- or -) difference between the potential nature ranges from about 0.01 to more than 10 more
of the aquifer material and the potential at the reference than 3 orders of magnitude. Thus, if an accuracy of plus
shale line (vertical line along which most shale or clay or minus one-half order of magnitude is acceptable, the
sequences plot). If the value of Rmf is not known, an method may be used with caution to estimate water
Rmf can be estimated from the mud resistivity (Rm): resistivity in areas of no measured data.

Using Geophysical Logs to Estimate Porosity, Water Resistivity, and Permeability

 Rmf, Tmf, SP, Tf
EXPLANATION
K Spontaneous potential constant at a specific temperature
Rmf Resistivity of mud filtrate, in ohm-meters
Rrnfe Resistivity of mud filtrate equivalent, in ohm-meters

Rm/75 Resistivity of mud at 75°F, in ohm-meters

Rw Resistivity of water, in ohm-meters
Rwe Resistivity of water equivalent, in ohm-meters

Ru>75 Resistivity of water at 75°F, in ohm-meters

SP Spontaneous potential, in millivolts
K=60+0.133 Tf
Rwe=Rmfe WSP/K Tf Temperature of formation, in degrees Fahrenheit (°F)
Tmf Temperature of mud filtrate, in degrees Fahrenheit (°F)

NO

75 Rwe+5 ( 0^8 n\(069Ku,e-024A

Rw75 u;75 (( . 8 0) )
146-337 Rwe

1
Ru>=Ru> 75(82/(T/+7))

Figure 5. Flow chart of spontaneous-potential method of estimating resistivity of formation water.

The 11 rock sequences tested were mostly car- and a line is fitted to the points. Ideally, the points define
bonates. No evidence was found to indicate that the a straight line, and the intercept of the line projected to
method was better suited to sandstone than any other type the 100-percent porosity value represents the water
of rock if the shale line for the SP curve could be
established; however, only three sandstone sequences 100
were used in the test. Method of Determination
A15

The accuracy of the method depends on the ac-

A Cross plot
curacy of the spontaneous-potential measurement, which o Spontaneous potential
is difficult to measure accurately because spurious elec- I 10
tromotive forces are inadvertently included in the o
measurement. Equation 1 is most applicable if the for-
mation water is saline, sodium and chloride are the
predominant ions, and the mud is fresh and contains no
unusual additives.

Cross-Plot Method
8,9° °
The cross-plot method for determining water resis-
tivity is discussed in MacCary (1978). It is also referred
to as the "carbonate" method or the "Pickett" cross-
plot method and is not as widely used as the spontaneous- 0.01
0.01 0.1 1.0 10 100
potential method.
MEASURED WATER RESISTIVITY, IN OHM-METERS
As the name implies, the method requires a cross
plot of resistivity and porosity values of saturated Figure 6. Measured and estimated resistivity of water.
material. These values are plotted on log-log graph paper Numbers are those of boreholes listed in table 1.

Spontaneous-Potential Method
 Table 1. Water-resistivity data
[Leaders ( ) indicate data not available]

Resistivity of water, in ohm-meters Rock-section depth, Name and location of

Hole lithology, and borehole
No. red Estimated
formation
Spontaneous- Cross-plot
potential method method
0.12 0.38 0.80 Depth 2,420-2,444 ft, vuggy dolostone, DC & FA #1: SE1/4NW1/4NW1/4,
Jefferson City Dolomite. sec. 13, T. 12 S., R. 17. E.
Douglas County, Kansas.
.79 .86 .44 Depth 615-955 ft, limestone and shale, Do.
Pennsylvanian Lansing-Kansas City Groups.
.27 .36 1.9 Depth 2,240-2,335 ft, cherty Do.
dolostone, Ordovician Viola Limestone.
.19 .26 .80 Depth 2,398-2,508 ft, dolostone and Do.
sandstone, Ordovician and Cambrian
Simpson and Arbuckle Groups.
.14 Depth 2,934-3,985 ft, calcareous
sandstone and granite, Cambrian Lamotte
Sandstone and Precambrian rock.
.61 .48 .20 Depth 3,343-3,665 ft, sandy Gels #1: SW1/4SW1/4SW1/4,
dolostone and cherty dolostone, sec. 32, T. 13 S., R. 2 W.,
Simpson and Arbuckle Groups. Saline County, Kansas
.15 .09 .15 Depth 1,980-2,200 ft, limestone, Do.
Lansing and Kansas City Groups.
.10 .09 .22 Depth 2,616-2,804 ft, limestone, Do.
Mississippian Warsaw, Keokuk,
Burlington, and Fern Glen Limestones.
.10 .09 .35 Depth 2,970-3,070 ft, dolostones, Do.
Devonian and Silurian Hunton Group.
.67 .20 Depth 3,384-3,498, dolostone, slightly Do.
cherty, Arbuckle Group.
.17 .07 .38 Depth 1,760-2,179 ft, dolostone and Watson #1: SE1/4SW1/4SE1/4,
sandstone, Arbuckle Group and sec. 18, T. 18 S., R. 23 E.
Lamotte Sandstone. Miami County, Kansas.
.38 .07 .70 Depth 1,245-1735 ft, limestone and Do.
dolostone, Hunton and Arbuckle Groups.
2.09 3 Depth 900-1,500 ft, dolostone and sandy Ordinance #1: Center SW1/4,
dolostone, Ordovician Cotter, sec. 22, T. 31 S., R. 20 E.
Jefferson City, and Gasconade, Labette County, Kansas.
Dolomites, and Roubidoux Formation.
.16 .20 Depth 1,501-1,816 ft, dolostones with Do.
vuggy zones, Ordovician and Cambrian
Gasconade, Eminence, and Bonneterre
Dolomites.
5.31 56 Depth about 2,600 ft, dolostone, Madison #1: NE1/4SE1/4,
Mississippian Mission Canyon Limestone. sec. 15, T. 57 N., R. 65 W.,
Crook County, Wyoming.

resistivity (Rw). This hypothesis was tested using data The most accurate porosity values are obtained
from two dolostone cores (DC & FA #1 and Geis #1) from from a dual-porosity log, such as the log shown on figure
Douglas and Saline Counties, Kansas. The cores were 1. However, other porosity logs, such as sonic, neutron,
saturated with water of known resistivity. The results are density, or dielectric, could also be used. Porosity deter-
shown in figure 7 and are encouraging because the pro- mined from a resistivity log cannot be used.
jection of a line, which fitted most of the data from the The procedure for determining water resistivity
Geis #1 core, intercepted the 100-percent porosity line using a dual-porosity log (including a gamma-ray trace),
close to the measured Rw value. However, DC & FA #1 and an electric log (including a resistivity trace) is as
core results show a few data points falling below the follows:
straight line, probably because the Ro values for poros- 1. Identify the lithology on the dual-porosity
ity values less than 3 percent may be affected by surface log.
conductance along the grains. (Ro is the combined 2. Select a rock sequence of uniform lithology to
resistance of water and saturated rock.) analyze.
Porosity and resistivity values for the cross plot can 3. Record both neutron and density porosity values
be obtained from geophysical logs. Homogeneous lith- for the selected points. Be certain to include the maximum
ology, constant water resistivity, and 100-percent water and minimum porosities as well as values within the
saturation are assumed. Resistivity values from logging range. About 10 points typically are needed to define the
devices that "look deep" into a formation are preferable. line.
Suitable logs might be a deep-induction log (as shown in 4. Calculate porosity values using a graph similar
fig. 3), a long-lateral log, a deep-conductivity log, and so to that shown in figure 8. (See log-interpretation manuals
forth. (Conductivity is the inverse of resistivity, generally for more details and for the appropriate graphs for the
recorded in millimhos or micromhos per meter.) tool used.)

8 Using Geophysical Logs to Estimate Porosity, Water Resistivity, and Permeability

 100

EXPLANATION
DC&FA #1, Core, dolostone
Geis #1, Core, dolostone
Measured water resistivity
Cementation factor

1.0 10 100 1,000

RESISTIVITY OF ROCK WATER SYSTEM (Ro), IN OHM-METERS

Figure 7. Cross plot of resistivity and porosity measured on dolostone cores.

5. Record resistivity values from the deep resistivity obtained from the log will be larger than Ro values of
trace of the same points selected in step 3. the aquifer material if it had been 100-percent saturated.
6. Plot the resistivity and porosity values as shown The values would plot to the right of the line defining
in figure 9 using log-log graph paper. n versus Ro for a 100-percent water-saturated section. The
7. Fit a straight line to the data points. During the unusually large resistivity values may indicate an un-
straight-line fitting, give less weight to the lower poros- saturated zone, hydrocarbons, or gases.
ity values because their resistivities may be affected by Because the porosity and resistivity logarithms
conductance along the grain. The resistivity indicated by define a straight line, standard least-squares techniques
this intercept is the water-formation resistivity. Project can be applied to determine the standard estimate of
the straight line to 100-percent porosity. error. Accordingly, the standard estimate of error for Rw
When using the cross-plot method, points defining can also be defined. Note that the estimated Rw is for
a straight line to the degree of desired accuracy cannot formation conditions. The accuracy of the method was
always be selected. Logs with expanded depth scales are tested by estimating Rw for 15 rock sections for which
easier to use in selecting better porosity and resistivity Rw had been measured (fig. 6). A least-squares analysis
value sets because the same point on all logs can be gave a coefficient of determination of 0.88, a value that
located more accurately. An example of the method is may be typical of estimates based on available logs. Figure
shown in figure 9 using data from the logs shown in 6 shows variations or scatter of about 1 order of
figures 1 and 3. magnitude in a range of more than 3 orders of magnitude
If the rock section is not 100-percent saturated with that can be expected in nature. Thus, based on the results
water, the rock-water system resistivity (Ro) values shown in figure 6, the method did not accurately estimate

Spontaneous-Potential Method
 FRESHWATER, LIQUID-FILLED HOLES

-1-15

0 10 20 30 40
POROSITY FROM NEUTRON LOG (n n ), IN PERCENT (APPARENT LIMESTONE POROSITY)

Figure 8. Porosity and lithology from formation-density log and compensated-neutron log (from Schlumberger Well Surveying
Corp., 1979). Data are from figures 1 and 3. Arrows show how to solve the example problem.

/?w; however, the method can be used to estimate to the range of porosity measured within the section of
formation-water resistivity in areas of no data if an interest. The wider the range, the more accurately the line
estimated accuracy of plus or minus one-half order of can be defined. The recorded measurement can be read
magnitude is acceptable. The method is applicable ir- more accurately from the log (trace) if the scale is ex-
respective of whether the water is saline or fresh. panded. Also, the accuracy of the measurement is of con-
The accuracy of the method generally is proportional cern. The consistency of resistivity values as measured by

10 Using Geophysical Logs to Estimate Porosity, Water Resistivity, and Permeability

 100
Rw = 0.38
ohm-meters
EXAMPLE
Determine: Water resistivity (Rw), cementation factor (m)
and formation temperature (Tfj
Given: Geothermal gradient =0.0133° Fahrenheit
per foot
Average depth =2055 feet
10 Mean annual temperature =55° Fahrenheit
z
LLJ
o Solution: Rw=Ro at 100-percent porosity =0.38
cc m=1.37/l = 1.37
LLJ
P-
T/=55+(0.0133) (2055)=82° Fahrenheit

O Data Depth, Ro n
oc point in feet (fig.3) (fig-1)
O
Q_ 1.0
1 2,020 9.5 9.5
2 2,026 12 10
3 2,036 18 6
4 2,041 26 3.5
5 2,059 6 13
6 2,071 4.5 16
7 2,092 6 7

0.1
0.1 1.0 10.0 100 1000
RESISTIVITY OF THE ROCK-WATER SYSTEM (Ro), IN OHM-METERS

Figure 9. Cross plot of geophysical-log values of Ro and n. See table 1 for description of boreholes.

the induction tools, lateral-resistivity tools, and conductiv- properties of rocks, such as intrinsic permeability or
ity tools are generally unknown to the user. For example, hydraulic conductivity, have been measured. Of the few
tools may be adjusted in the field to produce a "sharp" measurements available, most are of intrinsic permeability
trace irrespective of the ohm-meter scales used on the logs. and most have been made on rocks in the search for oil
The Rw value may be used to estimate water and gas. Intrinsic permeability (k) is a measure of the rela-
chemistry if the relation has been established. For exam- tive ease with which a medium can transmit a liquid under
ple, the sodium-chloride content can be estimated for potential gradient, and it is a property of the medium
many saline waters if the in-situ temperature of the water alone. Hydraulic conductivity (^0 is a measure of the ease
is available or can be estimated. Turcan (1966) used of the flow of water at a specific viscosity through a rock.
resistivity logs to determine Rw, which he then correlated Hydrologists studying shallow ground water generally use
with chloride or dissolved-solids concentration. the term "hydraulic conductivity^ whereas petroleum
Dissolved-solids concentration can be estimated engineers and geologists use "intrinsic permeability!' The
from specific conductance (conductance at 75 °F), usually relation between the two properties is
expressed in microsiemens or micromhos per centimeter,
if the relationship between specific conductance and K = kdg (4)
dissolved solids is known. A method of estimating
dissolved-solids concentration in water from Rw is given
in appendix A. where K is hydraulic conductivity,
k is intrinsic permeability,
ESTIMATING PERMEABILITY d is the density of the fluid,
g is the acceleration due to gravity, and
Background and Research H is the dynamic viscosity of the fluid.

In the deep subsurface and (or) areas where the Both density and viscosity are functions of temperature,
media (rocks) contain saline water, few hydraulic salinity, dissolved gases, and, to a lesser degree, pressure.
Estimating Permeability 11
Water viscosity and density are functions of temperature TEMPERATURE(7) ),IN DEGREES FAHRENHEIT (°F)
and salinity. Density, viscosity, and temperature relations
32 50 68 96 104 122 140 158
are shown in figures 10, 11, and 12. The viscosity-to-
temperature relation for various sodium-chloride solu-
tions shown in figure 11 can be approximated by an equa-
tion derived by Weiss (1982), which is given in appendix
D. Methods of estimating density from dissolved-solids
concentration are given in appendix C.
Most intrinsic-permeability values are determined
from testing cores in the laboratory or from a drill-stem
test. Most drill-stem tests are conducted on petroleum
reservoirs, and the results, even if accurate, must be used
with care because they may represent hydrocarbon reser-
voirs rather than aquifers.
Intrinsic-permeability values determined from core
tests usually do not completely represent conditions in
the rock because any method of collecting cores disturbs
the rock. Also, recovery of unconsolidated material is dif- 0.976 Q
"0 10 20 30 40 50 60 70
ficult and rarely successful. Laboratory tests should be TEMPERATURE (Tc ), IN DEGREES CELSIUS (°C)
run under conditions duplicating those in the subsurface.
Determining k values of fractured rock from cores is ex- Figure 10. Viscosity, density, and temperature of fresh-
tremely difficult because fractured core pieces are nearly water (data from Weast, 1984).
impossible to arrange exactly as they were positioned in
the subsurface. Also, the scale problem must be con-
sidered. Does a small volume of core represent the large where L is sample length, and Le is effective electrical-
volume of rock and fluid being considered? The scale flow path length. Note that the tortuosity factor of equa-
problem arises when determining aquifer permeability tion 6 differs from the tortuosity defined by Winsauer
that may be considered homogeneous over a thickness and others (1952, p. 255). Equation 5 indicates that for-
of tens or hundreds of feet, but that is extremely variable mation factor depends on pore size and its reduction, and
over a short distance, such as 1 inch. For example, the tortuosity. Equation 5 may be useful, but a procedure
intrinsic permeability values for sample 10 listed in table to apply data from geophysical logs to equations 5 and
2 ranged from 4,890 mD to 0.01 mD. Obviously, an 6 has not yet been fully developed.
averaging technique, such as the geometric mean or a The relations among the resistivity recorded on a
thickness-weighted mean, is required to determine the ef- geophysical log, water resistivity, and rock-material
fective permeability of the core. Thus, even if an un- resistivity is not entirely straightforward. Archie (1942)
disturbed fractured core could be obtained, how many assumed that the rock material was nonconductive and
point samples would be necessary to define the effective derived empirical equations to define observed resistivities
intrinsic permeability of the aquifer? From the discus- in terms of reservoir properties. The most generalized
sion, obviously the k value determined from a core or form of the Archie equation (Archie, 1942, p. 56) is
a drill-stem test must be carefully evaluated before assum-
ing the value is representative of a large rock mass. F = n~ (7)
The concept of relating formation factor to intrin-
sic permeability is appealing and has been investigated where F is the formation factor (dimensionless),
by many. Bear (1972, p. 113-117) related formation n is porosity (dimensionless), and
factor (F) to a tortuosity factor by m is the cementation factor (dimensionless).
F= (5) The relation of formation factor to pressure and
temperature is not completely known. In reference to
where C(1) is some tortuosity function, n is porosity, and temperature effect, Somerton (1982, fig. 12, p. 188)
m* is a function of the number of reductions in pore size showed that the logarithm of the ratio of the formation
openings. The function C(T^ may be 1 or less because the factor at a specific temperature to formation factor at
tortuosity factor is 1 or less. The tortuosity factor is de- a specified reference temperature for the Berea Sandstone
fined as: varied nonlinearly with temperature change.
For increasing pressure, Helander and Campbell
T = (L/LJ2 (6) (1966, p. 1) reported that formation factor changes can
12 Using Geophysical Logs to Estimate Porosity, Water Resistivity, and Permeability
 2.1

2.0
1.14

1.9

1.8

1.7

1.6

1.5

1.4

1.3
1 00
0 100 200 300 400

1.2 (7)), IN DEGREES FAHRENHEIT (°F)

Pressure correction factor (fp) for water versus (7))

1.1
Presumed applicable to brines but not confirmed
experimentally
1.0 Viscosity at elevated pressure ju. pT = (jn)(/p)

8 °-
0.8

0.7
Viscosity at 1 atmosphere pressure below
212° F or at saturation pressure of
0.6 water higher than 212° F

0.5

0.4

0.3

0.2

_ (1 percent concentration of sodium chloride = 10,000 parts per million)

0.1

r\ r\ _______|______|_______|_______|_______|_______|_______j_______|_______j______|______|_______|_______|_______|_______|_______|_______|_______|
'40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400
TEMPERATURE, (T,), IN DEGREES FAHRENHEIT (°F)

Figure 11. Water viscosity at various temperatures and percentages (%) of salinity (modified from Matthews and Russell, 1967).

be attributed to the following: (1) The increase in length the smallest pores, and (3) the effect of the double layer
of the mean free path for current (increased tortuosity) is increased by the reduced area of the pores as porosity
results from increased constriction as pores close, (2) the decreases.
amount of constriction is largely due to the closing of The cementation factor (rri) is a function of

Estimating Permeability 13
 tortuosity and pore geometry. Tortuosity is the ratio of
1.25 the fluid path length to the sample length. Aquilera (1976)
(1 percent concentration of sodium
cc chloride = 10,000 parts per million) studied the effect of fractured rock on the formation and
UJ 1.20
UJ
cementation factor, and used a double-porosity model
to define m. The model implies that m will approach 1
£ 1.15
LU
for a rock mass in which all porosity is the result of frac-
O tures (that is, there is no interconnected primary porosi-
O 1.10
m ty). Because the length of the flow path in a fractured
o 1.05 medium is much shorter than the length of flow path in
DC
LU
a porous medium, the tortuosity of the fractured medium
C/5 1.00
is small; the cementation factor is also small and ap-
<
proaches 1. The relation of porosity (ri), formation factor
O .95 (F), and the tortuosity factor (1) is:
z
S .90 F= 1
(8)
T/2
.85
Archie further defined
.80
F = Ro/Rw (9)
.75
C\jCMCMC\JC\JCOCOCOOOCO'fr where Ro is resistivity observed (log resistivity), and Rw
TEMPERATURE, (Tf) IN DEGREES FAHRENHEIT (°F) is the formation-water resistivity. (In the present paper,
Ro is assumed to be the bulk water and rock resistivity
Figure 12. Water density at various temperatures and percent- unaffected by fluid invasion, sometimes called true
ages (%) of salinity (data from Arps, 1953). resistivity or Rt.)

Table 2. Porosity and permeability data

[k, intrinsic permeability in millidarcies (mD); n, porosity, dimensionless; m, cementation factor, dimensionless; Rw, resistivity of water,
in ohm-meters; F, formation factor, dimensionless; leaders ( ) indicate data not available]

Hole Rock-section depth, lithology Name and location of

k n m Ra F
No. remarks , and formation borehole (source)
6.45 0.105 2.01 0.144 Depth 2,420-2,443 ft, vuggy dolostone, DCL & FA #1: SE1/4NW1/4NW1/4, sec.
maximum k=17 mD, minimum k=0.1 mD, 13, T. 12 S., R. 17 E., Douglas
core, Jefferson City Dolomite. County, Kansas (Gogel, 1981).
4.9 .13 2.37 Depth 1,980-2,200 ft, oolitic limestone, Geis #1: SW1/4SW1/4SW1/4, sec. 32,
drill-stem test, Lansing and Kansas T. 13 S., R. 2 W., Saline County,
City Groups. Kansas (Gogel, 1981).
4.74 .074 1.86 1.01 Depth 3,482-3,493 ft, dolostone, Do.
maximum k=29 mD, minimum k=0.02 mD,
core, Roubidoux Formation.
4.54 .092 2.15 Depth 2,934-2,985 ft, calcareous sandstone DCL & FA #1: SE1/4NW1/4NW1/4,
and granite, drill-stem test No. 1, sec. 13, R. 17 E., Douglas
Lamotte Sandstone and Precambrian rock. County, Kansas (Gogel, 1981).
101 .181 1.78 Depth 2,616-2,804 ft, cherty limestone Geis #1: SW1/4SW1/4SW1/4, sec. 32,
and dolostone, drill-stem test, Warsaw, T. 13 S., R. 2 W., Saline County,
Keokuk, and Burlington Limestones. Kansas (Gogel, 1981).
300 .125 1.09 Depth 2,944-3,046 ft, porous dolostone, Do.
drill-stem test, Hunton Group.
7 1,147 .22 1.10 14.1 5.29 Depth 210-610 ft, limestone, aquifer GF 18: sec. 20, T. 7 S., R. 10 W.
test on well tapping the Floridian Gulf County, Florida (Kwader,
aquifer. 1982).
.10 1.26 5 Depth 2,500-2,760 ft, dolomite, breccia, Madison #1: NE1/4SE1/4, sec. 15,
some intensely fractured, some very T. 57 N., R. 65 W., Crook County,
dense, flow-dye test, Mission Canyon Wyoming (Blankennagel and others,
Limestone (maximum k=789 mD, minimum 1977).
k=0.4, from core test).
.136 1.49 Depth 2,760-3,030 ft, limestone and Do.
dolostone with some thin anhydrite and
interbedded shale, flow-dye test
(maximum k=320 mD; minumum k=0.1 mD).
360 .149 1.49 Depth 3,100-3,400 ft, dolostone, some Do.
breccia texture, some crystalline,
some vuggy, some dense; flow-dye test
(from core test, maximum k=4,890 mD,
minimum fe=0.1 mD).

14 Using Geophysical Logs to Estimate Porosity, Water Resistivity, and Permeability

 Combining equations 7 and 9 yields: Croft (1971) successfully correlated the ratio
Ro/Rw with intrinsic permeability. However, the correla-
Rw = Ro nm (10) tion requires information about water chemistry, which
is not available from geophysical logs. The method works
If Rw is constant, equation 10 will yield a straight line well locally where the change in Ro is primarily due to
with the slope of -m on a log-log plot of n versus Ro as a change in porosity. MacCary (1984) pointed out that
shown in figures 7 and 9. in thick hydrologic units, neither Rw nor Ro can be con-
Sethi (1979) comprehensively reviewed the work of sidered constant.
many researchers in defining formation-factor relations. Stephens and Lin (1978) derived an equation for
The paper includes data originally presented by Winsauer intrinsic permeability that includes a geometric shape
and others (1952) in their resistivity study of brine- factor (SF) of the pore and the hydraulic radius (HR),
saturated sands and pore geometry. Raiga-Clemenceau which is defined as the ratio between area of cross sec-
(1977) reviewed the derivation and accuracy of the com- tion of the pore and its wetted perimeter. Their equation is
mon forms of the modified Archie equation:
k = (16)
F= a (11)

where a is an empirically determined constant probably SF is 2 for circular pores and 3 for narrow cracks.
related to lithology. Some authors term "a" the cemen- Stephens and Lin reported good correlation between
tation factor and "m" the cementation exponent (Dewan, calculated values of intrinsic permeability and the meas-
1983, p. 19). The two terms are covariant because they ured permeability values presented by Brace (1977).
are not independent variables. Two variations of equa- If equation 7 is substituted into equation 16,
tion 11 are
k = (HR2/SF)-nV-5m-0-5) (17)
F = 0.62 AT2- 15 , and (12)
Equation 17 implies that intrinsic permeability is a func-
F = 1.0/7-2 . (13) tion of the medium, the first set of terms, and a term that
includes porosity and cementation factor.
Equation 12 is sometimes called the Humble equation and The Kozeny equation is a common starting point
is used for elastics. Equation 13 is sometimes called the for relating intrinsic permeability and resistivity of porous
carbonate equation. Raiga-Clemenceau (1977) noted that media. This equation as stated by Bear (1972, p. 166) and
both equations are empirical and concluded that intrin- Herdan (1960, p. 196) is
sic permeability might be used to better define F. Accord-
ingly, he chose to define Fby setting a equal to 1, which k = G (18)
is the Archie equation, and making m a function of k.
The equation yielded a formation factor with less error
than the formation factor estimated by the Humble equa- where Gis the Kozeny coefficient, and Sp is the specific
tion. In conclusion, the Archie equation (eq. 7) may be surface, which is defined as the total interstitial surface
as appropriate for elastics as it is for carbonates, especial- area of the pores per unit volume of the medium.
ly considering the empirical nature of the equations. The relation between specific surface area of the
The following discussion relates formation factor, solids per unit volume of solids (S5) to specific surface
cementation factors, and tortuosity factor to intrinsic (V is
permeability. Regression analysis is one technique for
defining intrinsic permeability in terms of formation
factor. Carothers (1968) derived two equations for intrin- (19)
sic permeability (in millidarcies):
k = 4.0X108/F3 -65 (14) Substituting equations 8 and 19 into equation 18 yields:
G *2
for limestone, and k = (20)
FS 2 (I-/*)2
k = 7.0X108/F4-5 (15)
Equation 20 would be especially useful if F and Rw were
for sandstone. Ogbe and Bassiouni (1978, p. 10) used a independently known, but generally they are not. If equa-
similar approach. tion 7 is substituted into equation 20:

Estimating Permeability 15
 k = nm + 2 10,000
(21)

Equation 21 indicates that permeability is a function of ^

a medium factor (first set of terms right of the equal sign) d 1,000
and a porosity factor (P) (the last set of terms). Specifical- z
ly, the porosity factor is ^
m+2 100
P= n (22)

The numerator of P indicates that as the cementation fac- 10

tor (m) decreases, the porosity factor and the intrinsic
permeability increase. Similarly, the denominator in-
dicates that as the porosity increases, the porosity factor
and permeability increase. All values for the porosity
,-5
factor can be evaluated from information from geophys- 1x10 1x10 1x10 1x10"' 1x10
ical logs. The factor Gcan be evaluated; however, efforts POROSITY FACTOR (P)
to evaluate Ss from geophysical logs are not straight-
forward. Nevertheless, equation 21 does imply that any Figure 13. Relation between porosity factor and intrinsic
permeability.
relation for permeability should include a medium factor
and a porosity factor.
A search was made to find data sets of intrinsic many different samples have been plotted; no single line
permeability, porosity, and cementation factor for un- fits the data. Variations should be expected because the
consolidated, consolidated, and fractured rock of various basic implied assumption that the flow of electricity in
lithologies. The literature reports numerous values for an aquifer is in all ways analogous to the flow of water
cementation factor, core permeability, and empirical con- is not completely true. Permeability is a function of
stant (a), but little is said about how the properties were porosity, surface area, and tortuosity. Electrical conduc-
measured or what equation was used to determine the tivity is a function of rock conductivity, ion mobility,
factors. Because several different procedures are com- temperature, pressure, surface area, charge on the sur-
monly used, finding reported data to compare is difficult. face area, and the conductivity of the double layer sur-
The approach used here was that of Raiga-Clemenceau rounding the grains. The straight line shown in figure 13
(1977); that is, equation 7 was considered an appropriate might be assumed to define an approximate relation ap-
form of the Archie equation to use. The cementation plicable to coarse-grained elastics (such as sandstone and
factor was calculated from equation 10 or it was deter- siltstone, most porous carbonates, and uniformly frac-
mined from the Ro versus porosity cross plot. Special tured rocks) because surface conductance along the pore
preference was given to in-situ permeability and poros- walls is not dominant. The relations in equation 23 are
ity measurements from dual-porosity logs. Only a few shown in figure 14, which relates porosity to permeabili-
sites were found that met the requirements (table 2). Near- ty for various cementation factors. The Humble sand-
ly all data in table 2 are for carbonate rocks. The data stone reference samples reported by Winsauer and others
from table 2 are plotted in figure 13, which shows poros- (1952) whose permeability exceeds 1,000 mD (not shown
ity factor versus intrinsic permeability. Because of the on figure 13) correlate well with estimates using equation
wide range in values, a log-log plot was chosen. 23. Equation 23 was tested against data collated by Brace
A least-squares line was used to fit data as shown (1977) as listed in table 3. The correlation between
on figure 13, and the regression in equation of the line is calculated and measured permeabilities is good except if
the hydraulic radius, in micrometers, was less than 1.
k = 1.828X105 (P1 - 10) (23) Again, this comparison shows the importance of in-
cluding a geometric factor, such as Ss, Sp, or SF, to
The correlation of determination (r2) is 0.90, which is a better estimate intrinsic permeability of a medium with
good correlation. With the limited data available, it can- large surface areas.
not be ascertained if the exponent 1.10 is significantly dif- To better test the applicability of equation 23 for
ferent from 1 if the factor 1.828X105 were reduced other regions and for other lithologies of rock sections
accordingly. Equation 23 does not include the medium listed in table 2, typical porosities and cementation factors
factor as defined in equation 21. Medium factors for (table 4) were input into equation 23 to calculate

16 Using Geophysical Logs to Estimate Porosity, Water Resistivity, and Permeability

 10,000 is good. Considering that the range of typical perme-
abilities in nature is about 12 orders of magnitude (from
10~4 to 108 mD as listed in table 5), the accuracy and
usefulness of equation 23 are rather surprising because
it allows use of data from borehole-geophysical logs to
estimate intrinsic permeability with an accuracy of less
than one-half an order of magnitude (fig. 3). The equa-
tion, based on available data, successfully estimates
across a range of about 10~3 to 103 .

Cross-Plot Method

Equation 23 implies that intrinsic permeability can

be estimated if the cementation factor and porosity are
known. The value of m can be found from a resistivity-
to-porosity cross plot, such as that shown in figure 9. The
inverse of the slope of the cross plot is the cementation
factor, m. The value of m can be scaled directly from
the resistivity-to-porosity cross plot as follows. The slope
of the cross plot (m) is

(1-n) 2
m = (horizontal distance)/(vertical distance).
m = cementation factor
porosity
The cross-plot method of estimating intrinsic
permeability is as follows:
1-7. The first seven steps are the same as used in
determining resistivity of water (Rw) using a resistivity-
to-porosity cross plot.
10 15 20 8. Determine m from cross plot.
POROSITY (n), IN PERCENT 9. Calculate the thickness-weighted mean porosi-
ty of the section being tested.
Figure 14. Permeability and porosity.
10. Calculate k using equation 23.
The cross-plot method assumes that lithology is
permeabilites of other lithologies. The comparison be- constant except for variations in porosity. The method
tween typical permeabilities and calculated permeabilities also assumes that Rw in the section is constant.

Table 3. Permeability from formation factor and pore shape

[fim, micrometer]

Hydraulic Intrinsic permeability (fe) 1 ,

Formation Cementation radius in millidarcies
Porosity factor factor f ITP\
Rock (n) (m) in ym Calculated Observed Reference
Western granite:
50 megapascals 0.01 3,100 1.746 0.1 l.lxlO" 3 6.3xlO- 5 Brace and others (1968).
0.4 gigapascals .01 16,300 2.106 .1 1.7xlO~4 4xlO~ 6 Do
Sherman Granite - .03 880 1.933 6 .05 0.10 Do
Berea Sandstone - .22 11.7 1.624 7 750 890 Wyllie and Ros< (1950).
.37 4.4 1.490
.29 6.2 1.474 12 3,400 3,900 Do
2.2 990 Do
Nichols Buff - .20 12.5 1.569 4.2 538 230 Do
sandstone.
Eocene sandstone .22 13 1.694 4.4 670 340 Wyllie and Spangler (1952).
Pennsylvanian------ .16 20 1.635 4.1 180 120 Do.
sandstone 1.
.21 13 1.644 5.5 590 520
sandstone 2.

Calculated fc=1.828x lO^P1 - 10), where P = (n m+2)/(l-n) 2 .

Estimating Permeability 17
 Table 4. Typical cementation factors, porosities, and permeabilities and calculated permeabilities for
various lithologies
[mD, millidarcies]

Commonly observed Average k , in mD

Intrinsic
Cementation permeability Calculated Commonly Calculated
factor Porosity of (k), range of k, observed
Lithology (m) (n) in mD in mD1

Sand, fine to medium 2 1.3- 1.4 4 15-30 4 1, 000-3, 500 220-5,100 2,300 2,600
Sandstone, slightly
10-30 6 2,500 2,200
Sandstone, well
cemented 3 1.5-2.0 5 13-18 5 1-200 30-380 100 200
Limestone, crystalline 2 1.5-1.8 5 8-18 h-300 6-380 150 190
5o 1 Q 5 1-300 150 110
5 7-25 1-770 390
Dolostone, crystalline 2 2.2-2.4 5 8-18 5 1-300 1-100 150 50
Rocks, dense, fractured 1.1-1.8 2-20 7 0. 001-1, 000 0.015-870 500 440

'Calculated using typical m, typical n, and Ar=l,828xl05 P1 - 10, where Pn(m+2>/(l-n)2 .

2Kwader (1985).
3Coates and Dumanoir (1973).
4Jorgensen (1980).
5Core Laboratories Inc. (1983).
6This study.
7Brace (1980).

Table 5. Typical range of permeabilities of various consolidated and unconsolidated rocks

INTRINSIC PERMEABILITY (fe)

ID' 12 10'9 10'7

Square meter io" 19 io" 16
i I
10' 11 10'8
Square foot 10' 18 10' 15
i i
Millidarcy 1Q-4 1Q-3 1Q-2 1Q-1 1 1Q 1 1Q 2 108
10 3 104 105 106 107
I I I I I I I I I

Very fine sand; silt;mixtures of "Clean" well-

Unweathered "Clean" sand; mixtures
Unconsolidated rocks sand, silt and clay; glacial till; sorted gravel
clay of clean sand and gravel
stratified clay or cobble

Shale deeper than 250 Sandstone; dolostone; lime- Rocks with

Extensively fractured
feet; evaporites; lime- stone shallower than 15,000 larger than
Consolidated rocks rock; weathered lime-
stone deeper than feet; shale shallower than pore-size
stone or dolostone
15,000 feet 250 feet openings

Permeability Very
Very slightly permeable Slightly permeable Permeable
characteristic permeable
I 5 I 4 I 3 I 2 I , I I I 2 I 3 I 4 I 5 I 6
Gallons per day per square foot 10 10 10 10 10 1 10 10 10 10 10 10
' .ft .«? -4 7 .*> ' 1 9 ' * 'A ' 5
Footperday 10* 10 5 10^ 10 3 10^ 10 1 1 10 10* 10"3 104 10°

Foot per second 10~ 11 10" 10 10"9 10"8 10"7 10~6 10*5 10"4 10"3 10"2 10" 1 1

HYDRAULIC CONDUCTIVITY (It), AT 1.005 CENTIPOISE

18 Using Geophysical Logs to Estimate Porosity, Water Resistivity, and Permeability

 The data may not always closely define the permeability, especially in areas where few hydrologic
resistivity-porosity curve. Resistivity and porosity logs with data are available.
expanded depth scales are easiest to use and generally yield Dual density and neutron porosity logs can be used
the most accurate data. If Rw is known from other sources in conjunction with gamma-ray logs to determine in-situ
and is plotted at 100-percent porosity, the line and its slope porosity and to qualitatively identify lithology. Either a
can be easily and accurately defined. Rw might be spontaneous-potential log or a resistivity log can be used
available from chemical analyses of a water sample. The to define relative water resistivity in a rock section.
value of m can sometimes be estimated if the type of The spontaneous-potential and cross-plot methods
porosity and lithology are known (Asquith, 1985). of estimating water resisitivity were tested. The
Porosity values from other types of porosity logs, spontaneous-potential method uses spontaneous-potential
such as density, neutron, sonic, and dielectric, can also log measurements and mud-filtrate resistivity to estimate
be used in the cross plot. Logs not in porosity units also water resistivity. The cross-plot method uses the relation
can be used in cross plots to determine m (Aquilera, 1976, between porosity and the observed resistivity of the
p. 767; Pickett, 1973). For example, a sonic log, which saturated rock to estimate water resistivity in the rock.
records transit time, can be used. A plot on log-log paper Estimates of water resistivity were compared with meas-
of the difference between the transit time from the log ured values. A coefficient of determination of 0.66 for
minus transit time of the matrix versus rock-water the spontaneous-potential method and 0.88 for the cross-
resistivity (Rd) will define a straight line with the slope plot method were determined. Plots of the estimated
of -m. Density logs recording bulk density can be used values relative to the measured values show variations of
similarly as a plot on log-log paper of grain density minus about 1 order of magnitude. The methods are not ac-
bulk density versus Ro will define a straight line with a curate estimators, but can be used with caution to
slope of -m. Porosity values from an epithermal-neutron estimate water resistivity if no measurements are
log versus Ro when plotted on log-log paper also will
available.
define a straight line with the slope of -m. Values, in
The relations among resistivity measured on a
either counts per second or API units, from thermal
geophysical log, formation-water resistivity, rock-matrix
neutron logs can be plotted on a linear scale versus values
resistivity, and degree of cementation have been in-
of Ro on a log scale. The slope of the line defined will
be -m/D, where D is a function of borehole size and scale vestigated by many, such as Archie (1942) and Winsauer
function. Thus, ideally, even uncalibrated logs can be and others (1952). Several empirical relations have been
used to determine m. proposed and are described in the literature. Raiga-
The resistivity-to-porosity cross-plot method gives Clemenceau (1977) investigated the relations among
average values within the logged section because the log- cementation factor, permeability, and formation factor
ging tools have a "radius of investigation" and the values and questioned the validity of many of the empirical for-
measured are a volumetric average for the material within mation factors that have been proposed since Archie's
their radius of investigation. The radius of investigation original work. Raiga-Clemenceau empirically defined the
for a "resistivity-logging tool" might be 10 ft or more. cementation factor as a function of permeability.
A dielectric logging tool measures to about 2 in. depth By merging the well-known Archie equation with
beyond the borehole wall. The radius of investigation for the Kozeny equation, an equation for the intrinsic
a "radioactive-logging tool" might be 6 in. to 1 ft. permeability is derived as shown:
Recorders for radioactive tools commonly average emis- ,m + 2
sions over a time period, which results in additional k = (21)
averaging. Also, the cross-plot method requires several
sets of readings at different depths to define the empirical
linear relationship; thus "averaging," which is needed The equation implies that permeability is a function of
to determine effective intrinsic permeability of the entire the medium, represented by the first set of terms, and
thickness of a formation, is inherent in the procedure. a porosity factor (P) represented by the last set of terms.
The accuracy of the procedure can be checked and Porosity values can be determined from logs, such as
improved if the procedure can be calibrated to local neutron, density, sonic, and dielectric. A cross plot of
aquifer conditions. For example, the results can be observed resistivity to porosity defines the cementation
checked against permeability values from an aquifer test factor. However, the Kozeney coefficient and surface area
or a drill-stem test. are not easily determined from borehole-geophysical
logs.
SUMMARY The following linear regression equation, estab-
lished on the basis of 10 carefully selected data sets,
Geophysical logs can be used in geohydrologic describes the relationship between the porosity factor and
studies to estimate porosity, water resistivity, and intrinsic intrinsic permeability plot.

Summary 19
 k = 1.828X105 (P1 - 10). (23) Coates, G.R., and Dumanoir, J.L., 1973, A new approach to
log-derived permeability: Society of Professional Well Log
The regression equation has a coefficient of determina- Analysts, 14th Annual Logging Symposium, Transactions,
tion of 0.90 and applies to rocks in which surface con- p. 1-28.
ductance along grains is not dominant, such as fractured Core Laboratories Inc., 1983, Variations of permeability with
rock, coarse-grained elastics, and most carbonates. Com- rock types, in Dewan, J.T., Modern open-hole log interpreta-
tions: Tulsa, Okla., Penn Well Publishing, figs. 7-8, 361 p.
monly observed porosities and cementation factors for
Croft, M.G., 1971, A method of calculating permeability from
different lithologies were used in the equation to calculate electric logs: U.S. Geological Survey Professional Paper
permeabilities. The calculated permeabilities compared 750-B, p. 265-269.
well with typical permeabilities for the different litholo- Desai, K.P., and Moore, E.J., 1969, Equivalent NaCl solutions
gies, indicating the general usefulness of the equation. from ionic concentrations: Log Analyst, v. 10, no. 3, p. 12.
However, additional data, which include in-situ perme- Dewan, J.T., 1983, Essentials of modern open-hole log inter-
ability measurements, are needed to better evaluate the pretation: Tulsa, Oklahoma, Penn Well Publishing, 361 p.
relations. Gogel, Tony, 1981, Preliminary data from Arbuckle test wells,
Miami, Douglas, Saline, and Labette Counties, Kansas:
U.S. Geological Survey Open-File Report 81-1112, 155 p.
REFERENCES CITED Hela»der, D.P., and Campbell, J.M., 1966, The effect of pore
configuration, pressure, and temperature on rock resistivi-
Alger, R.P., 1966, Interpretation of electric logs in fresh water ty: Society of Professional Well Log Analysts, 7th Annual
in unconsolidated formations: Society of Professional Well Logging Symposium, Transactions, 29 p.
Log Analysts, 7th Annual Logging Symposium, Tulsa, Herdan, George, 1960, Small particle statistics: London, But-
Oklahoma, May 1966, Transactions, p. CC1-225. terworth Scientific Publications, 418 p.
Aquilera, R., 1976, Analysis of naturally fractured reservoirs Huntley, David, 1986, Relation between permeability and elec-
from conventional well logs: Journal of Petroleum Tech- trical resistivity in granular aquifers: Ground Water, v. 24,
nology, p. 764-772. no. 4, p. 466-474.
Archie, G.E., 1942, The electrical resistivity log as an aid in Jones, P.H., and Buford, T.B., 1951, Electric logging applied
determining some reservoir characteristics: American In- to ground-water exploration: Geophysics, v. 16, no. 1,
stitute of Mining and Metallurgical Engineers Transactions, p. 115-139.
v. 146, p. 54-62. Jorgensen, D.G., 1980, Relationships between basic soil-
Arps, J.J., 1953, Technical note 195: Journal of Petroleum engineering equations and basic ground-water flow equa-
Technology, sec. 1, p. 18. tions: U.S. Geological Survey Water-Supply Paper 2064,
Asquith, G.B., 1985, Handbook of log evaluation techniques 40 p.
for carbonate reservoirs: American Association of Petro- Keys, S.W., and MacCary, L.M., 1971, Application of borehole
leum Geologists Methods in Exploration 5, 47 p. geophysics to water-resources investigations: U.S.
Bateman, R.M., and Konen, C.E., 1977, The log analyst and Geological Survey Techniques of Water-Resources In-
the programmable pocket calculator: Log Analyst vestigations, book 2, chap. El, 123 p.
(September-October), v. 18, no. 5, p. 3-10. Kwader, Thomas, 1982, Interpretation of borehole geophysical
Bear, Jacob, 1972, Dynamics of fluids in porous media: New logs in shallow carbonate environments and their applica-
York, American Elsevier, 764 p. tion to ground-water resources investigations: Tallahassee,
Biella, Giancarlo, Lozei, Alfredo, and Tabacco, Ignazio, 1983, Florida State University Ph.D. dissertation, 186 p.
Experimental study of some hydrogeophysical properties _____1985, Estimating aquifer permeability from formation
of unconsolidated porous media: Ground Water, v. 21, no. resistivity factor: Ground Water, v. 23, no. 6, p. 762-766.
6, p. 741-751. MacCary, L.M., 1978, Interpretation of well logs in a carbonate
Blannkenagel, R.K., Miller, W.R., Brown, D.L., and Gushing, aquifer: U.S. Geological Survey Water-Resources In-
E.M., 1977, Report on preliminary data for Madison vestigations Report 78-88, 30 p.
Limestone test well No. 1, NE, SE, sec. 15, T. 57 N., R. _____1980, Use of geophysical logs to estimate water-quality
65 W., Crook County, Wyoming: U.S. Geological Survey trends in carbonate aquifers: U.S. Geological Survey
Open-File Report 77-164, 97 p. Water-Resources Investigations Report 80-57, 23 p.
Brace, W.F., 1977, Permeability from resistivity and pore shape: _____1983, Geophysical logging in carbonate aquifers: Ground
Journal of Geophysical Research, v. 82, no. 23, p. Water, v. 21, p. 334-342.
3343-3349. ____1984, Relation of formation factor to depth of burial
____1980, Permeability of crystalline and argillaceous rocks: along the Texas Gulf Coast, in Surface and borehole
International Journal of Rock Mechanics, Mineral geophysical methods in ground-water investigations: U.S.
Sciences, and Geomechanics Abstracts, v. 17, p. 241-251, Environmental Protection Agency and National Water
Brace, W.F., Walsh, J.B., and Framgos, W.T., 1968, Well Association (Feb. 6-9, 1984), p. 722-741.
Permeability of granite under high pressure: Journal of Matthews, C.S., and Russell, D.C., 1967, Pressure build up and
Geophysical Research, v. 73, no. 6, p. 225-2236. flow tests in wells: Dallas, Texas, Society of Petroleum
Carothers, J.E., 1968, A statistical study of the formation factor Engineers of American Institute of Mining, Metallurgical,
relation: Log Analyst, September-October, p. 13-20. and Petroleum Engineers Monograph 1, 158 p.

20 Using Geophysical Logs to Estimate Porosity, Water Resistivity, and Permeability

Meyer, C.A., Mclntosh, R.B., Silvestri, G.J., and Spencer, Log Analysts, 20th Annual Logging Symposium, Transac-
R.C., 1968, 1967 steam tables (2d ed.): New York, tions, 11 p.
American Society of Mechanical Engineers, 328 p. Somerton, W.H., 1982, Porous rock-fluid systems at elevated
Miller, R.W., 1977, A Galerkin finite-element analysis of steady- temperatures and pressures: Geological Society of America
state flow and heat transport in the shallow hydrothermal Special Paper 189, p. 183-197.
system in the east Mesa area, Imperial Valley California: Stephens, D.R., and Lin, W., 1978, Relationships of permeabil-
U.S. Geological Survey Journal of Research, v. 5, no. 4, ity, porosity, and formation factor: University of Califor-
p. 497-508. nia Research Letter 80856, 9 p.
Ogbe, David, and Bassiouni, Zaki, 1978, Estimating of aquifer Turcan, A.N., Jr., 1966, Calculation of water quality from elec-
permeabilities from electric well logs: Log Analyst, trical logs Theory and practice: Louisiana Geological
September-October, p. 21-27. Survey Water Resources Pamphlet 19, 23 p.
Pfannkuch, H.O., 1969, On the correlation of electrical con- Urish, D.W., 1981, Electrical resistivity-hydraulic conductiv-
ductivity properties of porous systems with viscous flow ity relationships in glacial outwash aquifers: Water
tranport coefficients: Haifa, First International Symposium Resources Research, v. 17, no. 5, p. 1401-1408.
of the Fundamentals of Transport Phenomena in Porous Weast, R.C., ed., 1984, CRC Handbook of chemistry and
Media, International Association of Hydraulic Research, physics: Boca Raton, Florida, CRC Press, p. F-ll, F-38.
p. 42-54. Weiss, Emanuel, 1982, A computer program for calculating
Phillips, S.L., Igbene, A., Fair, J.A., and Ozebek, H., 1981, relative transmissivity input arrays to aid model calibra-
A technical databook for geothermal energy utilization: tion: U.S. Geological Survey Open-File Report 82-447,
Lawrence Berkley Laboratory, University of California 18 p.
LBL-12810, 46 p. Winsauer, W.O., Shearin, H.M., Masson, P.H., and Williams,
Pickett, G.R., 1973, Pattern recognition as a means of forma- M., 1952, Resistivity of brine-saturated sands in relation
tion evaluation: Society of Professional Well Log Analysts, to pore geometries: American Association of Petroleum
14th Annual Logging Symposium, Transactions, Geologists Bulletin, v. 36, no. 2, p. 253-277.
p. A1-A21. Worthington, P.F., 1976, Hydrogeophysical equivalence of
Raiga-Clemenceau, J., 1977, The cementation exponent in the water salinity, porosity, and matrix conduction in arena-
formation factor-porosity relation The effect of ceous aquifers: Ground Water, p. 224-232.
permeability: Society of Professional Well Log Analysts, Wyllie, M.R.J., and Rose, W.D., 1950, Application of the
18th Annual Logging Symposium, Transactions, 13 p. Kozeny equation to consolidated porous media: Nature,
Schlumberger Limited, 1972, Log interpretation. Volume no. 165, p. 972.
1 Principles: New York, 112 p. Wyllie, M.R.J., and Spangler, M.B., 1952, Application of elec-
Schlumberger Well Surveying Corporation, 1979, Log inter- trical resistivity measurements to the problem of fluid flow
pretation charts: Houston, Texas, 97 p. in porous media: American Association of Petroleum
Sethi, D.K., 1979, Some consideration about formation resistiv- Geologists Bulletin, v. 36, no. 2, p. 359-403.
ity factor-porosity relations: Society of Professional Well

References Cited 21
APPENDIX A Most of the curves on figure Bl can be approximated by
the equation:
Estimating Dissolved-Solids Concentration
NaCl,ppm
Dissolved-solids concentration can be estimated where CNaC1 ppm is sodium-chloride concentration, in
from either resisitivity or specific conductance measure- parts per million, and Rw15 is the resistivity, in ohm-
ments. Specific conductance is a measure of the conduct- meters, of the solution at 75 °F. 75 °F is the usual reference
ance of electrical current through a fluid and is usually temperature in geophysical logging.
expressed in microsiemens or micromhos. A specific- Values of resistivity, in ohm-meters, at any
conductance measurement is referenced to a specified temperature, Tx, in degrees Fahrenheit, can be converted
temperature, usually 25 °C (77 °F). to resistivity at 75 °F by the equation:
Specific conductance (SC), which is the resistivity
reciprocal at 77 °F, can be calculated from the equation: (TX +1)/(S2) (B2)
1X104 Sodium-chloride concentration, in milligrams per
SC = (Al)
Rw77 liter, can be converted to concentration in parts per
million:
where Rw17 is the water resistivity at 77 °F, in ohm- (B3)
'NaCl.ppm 'NaCl,mg/L
meters.
Resistivity of water at any temperature, Tx, in where G is specific gravity and is the solution density
degrees Fahrenheit, can be converted to RW-J-J by: divided by pure-water density.
Concentration of a sodium-chloride solution in
(A2) milligrams per liter (CNaC1 mg/L) can be converted to
estimated concentrations in part per million (CNaC1 )ppm)
The relation between dissolved-solids concentra- by the following equation (Robert Leonard, U.S.
tion, in parts per million, and specific conductance, in Geological Survey, written commun., 1984):
microsiemens is
NaCl,ppm -V- 6.7 X 10-7) CNaC1>mg/L (B4)
DS ^ (P) (SC) (A3)
Investigators, such as Turcan (1966) and Desai and
where P is a factor to be determined for each water, but Moore (1969), list multipliers that allow resistivity of
which typically is about 0.67 for many ground waters if equivalent sodium-chloride solutions to be estimated for
specific conductance is in microsiemens. solutions that contain ions other than Na and Cl.
Dissolved-solids concentration can also be esti-
mated from the sodium-chloride concentration. The em-
pirical relation between sodium-chloride concentration APPENDIX C
(CNaC1) and dissolved-solids concentration (DS) is:
Estimating Density
DS = (A) (CNaC1) (A4)
Density is a function of temperature, pressure, and
where DS and CNaC1 are both in the same units. The salinity (fig. 12). Miller (1977) reported a relation derived
coefficient A should be determined for each water type; by J.W. Mercer (U.S. Geological Survey, written com-
however, the value of A generally used is about 1.04 for mun., 1975) to estimate density (d) of freshwater, in
many natural waters. kg/m3 (kilograms per cubic meter) as a function of
temperature and pressure:

APPENDIX B dTp = 2.0743 X103 + 5.56172X10-9/?

-4.26895 X\Q-5(\/Ta)
Estimating Sodium-Chloride Concentration
-2.35494X10-3 Ta2
Many useful interpretations of geophysical-log data + 4.95912 Xl07(l/rfl2) ; (Cl)
are based on relations for sodium-chloride solutions.
Relations among water resistivity, temperature, and where p is pressure, in Pa (pascals); and Ta is absolute
dissolved-solids concentration are shown on figure Bl. temperature.

22 Using Geophysical Logs to Estimate Porosity, Water Resistivity, and Permeability

 SODIUM-CHLORIDE CONCENTRATION (CNaC1 ) IN PARTS PER MILLION (PPM)

50
<v ^SS&Bfc
EXAMPLE
Determine C.^ /->i
Given: Rw=0.1 ohm-meters
75
T= 125°F
Solution: CN <-,, =40,000 ppm

100

CO
UJ
UJ 125
cc
CD
UJ
Q
150

175

200

250

300

350
400 * in <p N; opo>q
CO TJ- IO <D q q q q qqqo
q q q qqd d d C» odd*-1 co *» 10 <b "
o d d> d do
RESISTIVITY OF SOLUTION, IN OHM-METERS

Figure B1. Resistivity of water as a function of salinity and temperature (modified from Schlumberger, 1972). Arrows show
how to solve the example problem.

Ta can be calculated by In borehole geophysical logging applications, many

useful interpretations are based on the properties of
Ta = Tc + 273.15 (C2) sodium-chloride solutions. Phillips and others (1981)
derived a regression equation of aqueous sodium-chloride
Equation Cl is based on a least-squares regression solutions:
of data by Meyer and others (1968) for a range of 0° to
350°C and l.OxlO5 to 1.75X107 Pa . d = A + Bx + Cx 2 + Dx3 (C5)
Curves shown by Schlumberger Limited (1972, fig. and
8-5, p. 47) indicate that density (d) is affected by the
sodium-chloride concentration: x =
where d is density, in grams per cubic centimeter;
Ad = 7.3 X 10-4 CNaappm (C3) Tc is temperature, in degrees Celsius;
where Ad is change of density in grams per cubic cen- p is pressure, in mega pascals (MPa);
timeter and CNaC1 ppm is sodium-chloride concentration
mol is molal concentration of NaCl;
Cl = -9.9595; C2 = 7.0845; C3 = 3.9093;
in parts per million. Thus, an estimate of density as a
a\ = 0.004539 (a\ = -0.00500 for vapor saturated
function of pressure, temperature, and salinity might be
made by the use of both equations Cl and C3: values);
a2 = -0.0001638; a3 = 0.00002551; A = -3.033405;
d * dTp + Ad (C4) B = 10.128163; C = -8.7507567; D = 2.663107;
and
where d and Ad are in the same units. e = the exponential constant.
Appendix C 23
Phillips and others (1981) reported that the experimen- Miller (1977) reported on the work of Meyer and
tal data were reproduced to a maximum deviation of ± 2 others (1968) to estimate viscosity. If viscosity is a func-
percent over the entire range of concentration, for tion of temperature alone, viscosity, in pascal-seconds,
temperatures of 0°-350°C, and for pressure as high as can be estimated by:
50 MPa.
An empirical relation derived by Robert Leonard /*S (10-7) (241.4X10*) (D2)
(U.S. Geological Survey, written commun., 1984) to esti-
mate specific gravity, which is solution density divided The exponent x is:
by density of pure water, is
x = (247.8)/(rc + 133.15)
G 3 i/(l- 6.7 XlO-7cNaC1)ppm (C6)
where Tc is temperature in degrees Celsius. The equation
is reported valid for liquid water along the saturation line
where G is specific gravity and CNaCljppm is sodium- from 0-300 degrees Celsius.
chloride concentration in parts per million. Phillips and others (1981) reported a correlation for
sodium-chloride solutions using more than 1,500 selected
experimental values. The model was originally developed
APPENDIX D
for colloidal suspensions and solutions of nonelectrolytes;
however, the equation has been used extensively for elec-
Estimating Viscosity trolytes, such as NaCl, after suitable modification:

An equation to estimate dynamic viscosity of water, =1 + a mol + b mol2 + c mo/3

(n), was derived by Emanuel Weiss (1982, p. 4). The equa- + d Tc(\ - el m°l) (D3)
tion approximates the curves of Matthews and Russell
(1967) shown in figure 11. The approximate equation is:
where /* is viscosity of sodium-chloride solution, in
centipoise;
/* = [(38.37r/- 5) - (14.6/7}-0- 25) + 1.48] fjLw is viscosity of water, in centipoise;
a = 0.0816; b = 0.0122;
(Dl) c = 0.000128; d = 0.000629;
i = -0.7; Tc is temperature, in degrees Celsius;
Where n is dynamic viscosity, in centipoise; mol = molal concentration, in gram-moles NaCl
Tj- is temperature, in degrees Fahrenheit; and per kilogram of water;
C-ds.mg/L is dissolved-solids concentration, in and e is the exponential constant.
milligrams per liter. Phillips and others (1981) reported that the equa-
Weiss (1982) reported that the largest error is at 55 °F, tion reproduces data to an average of better than ± 2 per-
at which temperature the value of the calculated /* from cent over the range 10 to 350 °C, 0.1 to 50 MPa, and 0-5
equation Dl is inaccurate by nearly 10 percent. mol.

24 Using Geophysical Logs to Estimate Porosity, Water Resistivity, and Permeability

* U.S. GOVERNMENT PRINTING OFFICE: 1989-673-047/86048