See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/28128647

Estimation of soil petrophysical parameters from resistivity data: Application to

oil-contaminated site characterization

Article  in  Geofísica Internacional · July 2006

Source: OAI

CITATIONS READS

36 381

6 authors, including:

Omar Delgado-Rodríguez Aleksandr Mousatov

Instituto Potosino de Investigación Científica y Tecnológica Instituto Mexicano del Petroleo
77 PUBLICATIONS   575 CITATIONS    90 PUBLICATIONS   867 CITATIONS   

SEE PROFILE SEE PROFILE

David Flores
Instituto Mexicano del Petroleo
4 PUBLICATIONS   64 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Joint interpretation of geoelectrical and volatile organic compounds data: A case study in a hydrocarbons contaminated urban in Mexico City. View project

Characterization of agricultural soils contaminated with arsenic. View project

All content following this page was uploaded by Omar Delgado-Rodríguez on 28 September 2017.

The user has requested enhancement of the downloaded file.

 Geofísica Internacional (2006), Vol. 45, Num. 3, pp. 179-193

Estimation of soil petrophysical parameters from resistivity data:

Application to oil-contaminated site characterization

Vladimir Shevnin1, Omar Delgado Rodríguez1, Aleksandr Mousatov1, David Flores Hernández1, Héctor
Zegarra Martínez1 and Albert Ryjov2
1
Mexican Petroleum Institute, Mexico, D.F., Mexico
2
Moscow State Geological Prospecting Academy, Geophysical Faculty, Moscow, Russia

Received: May 23, 2005; accepted: August 1, 2006

RESUMEN
El método Sondeo Eléctrico Vertical (SEV), conocido desde 1912, ha cambiado sustancialmente durante los últimos 10
años, apareciendo una nueva tecnología llamada Imagen de Resistividad (IR) con interpretación 2D de los datos de resistividad.
Otra vía posible de desarrollo del método SEV es, partiendo de las relaciones existentes entre la resistividad eléctrica y los
parámetros petrofísicos (PP), estimar estos últimos a partir de datos de IR. Para la realización práctica de este concepto fue
desarrollada la teoría del problema directo e inverso que relaciona la resistividad eléctrica con los PP. Cada trabajo de campo
deberá incluir un levantamiento de SEV (IR), mediciones de resistividad eléctrica del agua subterránea con el objetivo de
determinar su salinidad y la recolección de algunas muestras representativas de suelo del sitio con mediciones hechas en
laboratorio de la resistividad eléctrica como función de la salinidad del agua de poro, creando el modelo petrofísico del suelo de
este sitio. Esta tecnología puede ser utilizada tanto para la caracterización de sitios limpios como contaminados por hidrocarburos.
Para el caso de sitios contaminados, los valores de los PP determinados en laboratorio, salinidad de agua y los datos de IR,
permiten establecer la frontera petrofísica entre suelo limpio y contaminado, y por consiguiente, configurar la pluma contaminante.
En este trabajo se incluyen, como ejemplos prácticos, los resultados de la aplicación de esta tecnología en algunos sitios
contaminados por hidrocarburos en México.

PALABRAS CLAVES: Parámetros petrofísicos de suelo, modelación petrofísica, contenido de arcilla, porosidad, capacidad
de intercambio catiónico, Sondeo Eléctrico Vertical, Imagen de Resistividad 2D, contaminación por hidrocarburos.

ABSTRACT
Vertical Electrical Sounding (VES) method, known from 1912, has changed greatly during the last 10 years, into a new
technology named Resistivity Imaging (RI) with 2D data interpretation. Another possible development for VES method is
estimating petrophysical parameters (PP) from RI data, using the relationship between electrical resistivity and PP. In order to
reach this purpose, the theory of the forward and inverse problem that relates the electrical resistivity with PP was developed.
Each field survey should include a VES (RI) survey, groundwater resistivity measurements in order to determine the groundwater
salinity, and collecting some representative soil samples in the study site for resistivity measurements as function of pore water
salinity in laboratory, creating a soil petrophysical model of the site. This technology can be used for the characterization of
uncontaminated and oil contaminated sites. For the case of contaminated site PP values determined in laboratory, groundwater
salinity and RI data help to define the petrophysical boundary between contaminated and uncontaminated soil, and consequently,
to obtain the contamination plume. In this work, the results of the application of this technology in some hydrocarbon contaminated
sites in Mexico are presented.

KEY WORDS:Petrophysical parameters of soil, petrophysical modeling, clay content, porosity, cation exchange capacity,
Vertical Electrical Sounding, 2D Resistivity Imaging, oil contamination.

INTRODUCTION of traditional 1D model (horizontal layering) to 2D (and 3D)

models for interpretation in heterogeneous media. Field
Vertical Electrical Sounding (VES) is a classical technology of VES method was transformed from
method of applied geophysics used for distant and non- performance of soundings made in separated and
destructive study of the upper part of geological medium. It independent points, with spacing of current electrodes
uses direct current (DC) injected in the ground surface to growing in logarithmic scale, to measuring system with
investigate the underground electrical resistivity. For the last multi-electrode array distributed along profiles (known as
10 - 15 years this method has changed greatly, from solution Resistivity Imaging - RI). In the case of RI, the step between

179
V. Shevnin et al.

sounding sites is equal or proportional to the inter-electrode Sauck (1998) in his contamination model proposed that
distance. For the 3D survey, the information is collected on low resistivity anomalies in mature contaminated sites
a grid of profiles distributed on the earth surface. There are resulted in an increase of salt content in pore water.
many publications on the theory of forward and inverse According to Sauck, low resistivity is created by intense
problem solution (including interpretation algorithms) for bacteria action on petroleum products and chemical
DC resistivity method above 2D and 3D heterogeneous interactions of contaminants and their products with soil
media (Loke and Barker, 1995, 1996). RI technology and grains in a lower part of vadose zone. In this model, bacteria,
2D data interpretation improved greatly the detail and through biodegradation, produce organic and inorganic acids
accuracy of inhomogeneous media study. increasing the dissolution of minerals and releasing the ions
that increase the levels of total dissolved solids (TDS) in
At the same time there is another important possibility pore water.
in VES method development. This is the relation between
soil resistivity and petrophysical parameters. Well known The model of Sauck was modified by Atekwana et al.
Archie formula (Archie, 1942) displayed rock’s resistivity (2001; 2003), and Abdel Aal et al. (2004). Atekwana et al.
relation with groundwater salinity, porosity and formation (2003) found that there is no tight correlation between soil
factor. Soil studies uncovered also the influence of humidity, and water resistivity in contaminated zones, so, another cause
volumetric clay content, grain size and packing, cation of low resistivity anomaly should be found. Abdel Aal et al.
exchange capacity and contaminants on resistivity. All this (2004) explained that the main cause of low resistivity
knowledge allowed developing a theory of soil resistivity anomaly is superficial conductivity increase in soil pores
and algorithms of forward problem solution (calculation of compared with electrolytic conductivity. Resulting process
resistivity using known PP) and inverse problem solution is reflected in apparent changes of PP estimated from soil
(estimation of some PP from soil resistivity, taking into resistivity.
account that all others PP are known or fixed).
Shevnin et al. (2005a) found that, in the case of clean
There are many papers on electrical properties of hard soil, PP values estimated from resistivity are close to real
rocks (mainly for oil well logging). Quantity of papers about values, but in oil contaminated zones PP becomes
loose soils electrical properties is noticeably less (Marion et anomalous. For example, in contaminated zone, clay content
al., 1992; Klein and Santamarina, 2003; Revil and Glover, estimated from resistivity (it is called quasi-clay) seems
1998). In papers of Ryjov (1987) and Ryjov and Sudoplatov increased noticeably. Electrical conductivity in clay resulted
(1990) was developed algorithm of resistivity calculation in superficial conductivity and this conductivity increases
for sand-clay soils, based on two-component model of soil in contaminated soil due to biodegradation products. That
(sand and clay). is why increasing quasi-clay is found.

To determine PP from resistivity one need to provide The basic principle of petrophysical interpretation
in the theory and in experiments the sufficient amount of consists in analysis of relation between soil resistivity and
information to develop some methodology of quantitative pore water salinity. To estimate clay content, porosity and
PP determination. Estimation of PP with VES method does cation exchange capacity from soil resistivity, finding or
not cancel traditional methods of PP determination in fixing all other PP values as much as possible close to reality
geological, chemical and agronomical laboratories, but can is needed. To find them representative soil samples are
add to these some data obtained distantly, rapidly and with collected in the site and their resistivity as function of water
high spatial density. Traditional laboratory methods are salinity ρ(C) is measured. This curve ρ(C) is interpreted to
accurate, but punctual and expensive. Rational integration find such PP as sand porosity, clay capillary radii, clay
of traditional laboratory methods and VES can provide PP content and clay CEC. These parameters can be used as a
estimations with higher density and lower cost in optimized petrophysical model in interpretation of soil resistivity and
time and with increasing quality. pore water salinity to estimate clay content, porosity and
CEC of soil.
VES method offers great advantages at oil
contamination studies. Oil contaminants cause series of VES METHOD WITH RI TECHNOLOGY
changes in physical, chemical and biological properties of
soil (Modin et al., 1997; Sauck, 1998; Atekwana, 2001), Apparent resistivity (ρa) pseudo-section (Figure 1A),
mainly during first several months after contamination. Just which is a result of RI technology, can be interpreted in terms
after contamination a high resistivity anomaly marks of 2D model (Loke and Barker, 1995, 1996), giving a true
contaminated zone, but after several months, as a result of or soil resistivity (ρ) cross-section (Figure 1B). 2D model
biodegradation of contaminants under the influence of consists of rectangular cells reaching infinitely in the
bacteria, this zone reveals a low resistivity anomaly. horizontal direction perpendicular to the profile. All cells

180
 Estimation of soil petrophysical parameters

have the same vertical thickness in one depth interval. That cases physical properties change more in vertical, than in
is why such level can be considered as a layer (not as a real lateral direction. In this situation 2D interpretation (with
geological layer, but some model layer). Number of layers Res2DInv) increases the inversion quality due to more
and their thicknesses are fixed for the whole site. RI correct accounting of near-surface and deep
technology in this paper was applied to study the soil inhomogeneities’ influence, and regularization in inversion
structure until 10 - 20 m deep. In this case geological sections process makes its results more stable.
are mainly horizontally-layered (consist of different
geological layers) or pseudo-layered ρ(z) structures (due to Traditional result of 1D interpretation is a horizontally
the influence of groundwater level and weathering). In both layered model with layers spreading infinitely in lateral

Profile M N

1
2
A
3
4
AB/2 (m)

5
6
7
8
9

Profile
1
2 B
3
4
Depth (m)

7
8

Fig. 1. Presentation of measured apparent resistivity values (A) and interpreted (with RES2DINV) soil resistivity values (B) for one VES
profile. The vertical axis of pseudo-section (A) is the half-distance between current electrodes, whereas the vertical axis of the cross-section
(B) is the depth.

181
V. Shevnin et al.

directions. Nevertheless during nearly one century 1D results PETROPHYSICAL INTERPRETATION OF SOIL
were used to study 2D and 3D geological structures on a RESISTIVITY
grid of VES. 2D interpretation gives more stable geological
results along each profile and it can be also used to study The important step in VES method improvement is
3D structures, especially in a situation with nearly horizontal petrophysical interpretation of soil resistivity. This
layering. With some restrictions an assumption can be made technology was developed by A. Ryjov (Ryjov, 1987; Ryjov
that 2D interpretation gives information about some ρ(z) and Sudoplatov, 1990) and then applied in Mexico at more
distribution (restricted volume instead of infinite horizontal than 12 sites with oil contamination (Ryjov and Shevnin,
prisms) below each VES point, all VES being distributed 2002; Shevnin et al., 2004; 2005a, b; 2006). The forward
on the site area along profiles. In case of cells with vertical petrophysical problem consists in the calculation of soil
thickness 1 m, soil volume for resistivity determination resistivity values on the base of PP of some soil model
amounts several cubic meters. These assumptions are (mixture of sand and clay). The inverse problem consists in
suitable for pseudo-layered models and allow creating a soil PP estimation on the base of soil resistivity and pore water
resistivity cube (Figure 2) (Shevnin et al., 2005a) for an salinity, taking into account petrophysical soil model of the
easy preparation of vertical sections for any profiles and site. For practical realization of petrophysical interpretation
horizontal maps (lateral ρ changes) for some fixed depths. in each site it is necessary to obtain VES data, groundwater
salinity and representative soil samples to measure their
Theoretically evident solution to use 3D inversion resistivity versus pore water salinity in laboratory and then
algorithms is not practical. Existent algorithms of 3D to interpret data creating soil model of the site.
interpretation give unstable solutions or need to apply very
detailed and complicated 3D field survey. Now more The basic model for loose soils is the sand and clay
practical solution is to make 2D interpretation for each profile mixture. Grain sizes of sand and clay differ in four orders.
in the study site and then to create 3D model in spite of all If clay content is less than sand porosity smaller clay particles
restrictions of this approach. are situated in greater sand pores. When clay content is more

Profile 6
Profile 5
Profile 4
Profile 7
Profile 3 Profile 8

Profile 2
Profile 1
Layer 1
Layer 2
Layer 3
..........
..........
..........
..........
..........
Layer N
Fig. 2. Soil resistivity cube as a result of interpretation of profiles grid by using the inversion software RES2DINV.

182
 Estimation of soil petrophysical parameters

than sand porosity, sand grains are dispersed in clay matrix. than 5 g/l) are situated in parallel to the lines for sand and
Sand and clay capillaries are completely (or partially) filled water with distribution along vertical axis according to soil
with water. Clay capillaries radii are comparable with electric porosity. The curves for clay and clay-sand mixture for
double layer (EDL) thickness. As a result, clay conductivity salinity less than 5 g/l are not in parallel to the lines for sand
is formed under the EDL influence. Sand capillaries radii and water, cross water curve and their vertical distribution
are much greater than EDL thickness, and EDL does not depends on clay content. Such non-linear behavior of the
influence on sand conductivity. curves for clay - rich soil resulted in clay EDL and CEC
influence.
An example of the forward petrophysical problem
solution, (that is theoretical resistivity calculation for a given An important peculiarity of sand and clay resistivity
soil model) is presented in Figure 3A. Soil resistivity curves should be commented: their difference depends on water
for different sand and clay proportions are here displayed: salinity. When water salinity is 0.01 g/l, sand to clay
the curve for clean sand (in the upper part of Figure 3A, 0% resistivity ratio is equal to 780, for salinity 0.1 it is equal to
of clay) and the curve for pure clay (in the lower part of 80, for salinity 1 it is equal to 10 and for salinity 10 it is
Figure 3A, 100% of clay). This model curves were calculated equal to 3 (Figure 3).
for pore radii in sand 10-4 m, in clay 10-8 m, sand porosity
0.25, clay porosity 0.55, CEC of clay 3 g/l and full water So, resolution of resistivity method to differentiate soil
saturation of soil pore space. Figure 3B shows theoretical lithologies depends on groundwater salinity and decreases
dependence of porosity from clay content for the model A. with salinity increasing.
Porosity curve begins at 25% (sand porosity), has minimum
when clay content is equal to sand porosity and all sand This also shows that clay content determination on soil
pores are filled with clay, and then increase until clay porosity resistivity cannot be performed without knowledge of pore
(55%). Left part of curve was formed under the influence of water salinity.
sand porosity, while the right part is influenced by clay
content and clay porosity. INTERPRETATION OF SOIL SAMPLES
RESISTIVITY MEASUREMENTS AS FUNCTION
Theoretical curve for sand resistivity vs water salinity OF WATER SALINITY
(Figure 3, curve 0) is practically straight line situated in
parallel and above water resistivity curve. These curves Representative soil samples collected in each site are
decline of straight lines at high salinity (near solubility limit). dried and homogenized in a laboratory. The soil is distributed
The clay and clay-sand mixture curves at high salinity (more into 4-5 resistivimeters (with volume about 200 ml each)

Resistivity, Ohm.m
1000
0 Clay content, %

Water
2
100 4

10 60
20 Porosity, %

10
20 A 50
B
30
40
50 40
70
100
30
1 T=20 C
o

20
CECClay=3 g/l
10
C(NaCl), g/l Clay content, %
0.1 0
0.01 0.1 1 10 100 0 10 20 30 40 50 60 70 80 90 100

Fig. 3. A. Theoretical resistivity curves versus pore water salinity for different clay content in sand-clay soils; B: Relation between soil
porosity and clay content for the model A.

183
V. Shevnin et al.

and filled until full saturation with NaCl solutions having position in Figure 5 depends on CEC. That is why a soil
salinity in interval 0.5 - 50 g/l with the step 2-3 times. resistivity model of the site is needed for petrophysical
Experimental resistivity curve ρ(C) (see example in Figure interpretation.
6) is interpreted by fitting with theoretical curves calculated
for petrophysical model (sand and clay capillary radii, Table 1
porosity, formation factor and CEC of clay). Some model
parameters can be adjusted during interpretation. Final best- Values of cation exchange capacity for different materials
fit model gives clay content in soil, soil porosity and adjusted (from Rowell, 1993)
sand and clay components parameters. This soil model is
then used for soil resistivity (obtained from VES) and
groundwater salinity petrophysical interpretation. Material CEC, meq/100 g

In Figure 4 there are graphs of errors in soil PP Organic matter 130-500

estimation as a function of fitting error for the case of soil Vermiculite 100-150
samples measured in laboratory. Fitting error is the RMS Montmorillonite 29-150
error between experimental and theoretical ρ(C) curves. Hydro mica 10-40
Errors were estimated with the help of statistical approach, Kaolin 3-15
developed by Goltsman (1971) and Tarantola (1994). Errors Fine sand 0.8
in clay content estimation are 6-17% when fitting error is 2- Sand 1 -4
5%. Experimental studies of calibrated soils in laboratory Loamy sand 2 - 12
gave us mean error in clay content estimation equal to 18% Sandy loam 7 - 16
(Shevnin et al., 2006). Clay 4 - 60

Sand and clay curves from Figure 3 are displayed in

Figure 5 as groups of curves. Each group is presented by CATION EXCHANGE CAPACITY (CEC)
series of curves depending on sand porosity, and clay CEC.
Clay and organic matter particles act as giant anions,
The sand curve depends closely on the sand porosity with their surface covered with a net of negative charges.
(Figure 5). Unlike the sand, clay porosity changes are not Cations are attracted to the surface and exchange takes place.
so high. Montmorillonite (bentonite) counts on 60% porosity, CEC is simply a measure of the quantity of sites on the soil
kaolin - 50% and hydromica - 55%. There is another factor grains surface that can retain positively charged ions
that influences on clay resistivity: CEC of clay. Clay line (cations) by electrostatic force. The electrostatically retained

70
Error in parameters
60 estimation, % C
CE
50 ad ius
ry r
Model with 10% of clay a p illa
40 yc t
Cla y c o nten
30 Cla ity
l a y poros
C
20
10 Sand porosity

0
0 2 4 6 Fitting error, % 10
Fig. 4. Errors in soil petrophysical parameters as a function of fitting error.

184
 Estimation of soil petrophysical parameters

cations are easily exchanging with other cations in soil pore CEC depends on clay content in soil and clay CEC of the
solution. The cations exchange sites are mainly located on site and it can be determined in this way from resistivity
clay grains surface. Normal range of soil CEC is in interval values.
from < 4 meq/100g (for sandy soil) to > 25 meq/100g (for
rich - clay soil). Increase of organic material in soil notably RELATION BETWEEN CONTAMINATED ZONE
increases CEC. AND LOW RESISTIVITY ANOMALY

AN IDEA OF CLAY CONTENT, POROSITY AND When a leakage of light non-aqueous phase liquids
CEC DETERMINATION ON VES RESISTIVITY (LNAPL) takes place, it penetrates into subsoil, changing
its physical - chemical properties, electrical resistivity
Knowing soil resistivity and pore water salinity, it is between them. The resistivity difference between
possible (with the help of Figure 3A) to estimate clay content, contaminated and clean soil depends on such factor as the
and then to estimate soil porosity from clay content by using age of contamination event. In case of recent leakage, the
Figure 3B. contaminated soil creates high resistivity anomalies, being
generally proportional to a contamination level. On the other
Let’s consider the situation when sand-clay soil hand, when the leakage is aged (from four months to a year
resistivity is 10 Ohm.m and water salinity is 0.01 g/l. In this after the contamination, or more) oil contamination creates
case, clay content estimated on Figure 3A is 30% and low resistivity anomalies.
porosity found on Figure 3B is 15%. If salinity is equal to 1
g/l for the same soil resistivity, clay content would be 20% The main factors that influence on clay sand mixture
and porosity found from clay content would be 16%. In resistivity are the following: clay content, water salinity and
uncontaminated sites clay CEC is the same for the whole conductivity, clay and sand capillaries structure. Mature oil
area. Clay CEC is determined in laboratory by measuring contamination slightly increases water solution conductivity,
and interpreting soil resistivity vs salinity curve. The soil changes pore structure by filling pores with biodegradation

10000
Resistivity, Ohm.m

1000 Sand porosity

10, 15, 20, 25, 30, 35, 40%

100 Water

10 0.21 - CEC of clay, g/l

0.58
1
2
3
5
7
9
15

C(NaCl), g/l
0.1
0.01 0.1 1 10 100
Fig. 5. Change of sand curve position depending of sand porosity and change of clay curve depending of clay cation exchange
capacity (CEC).

185
V. Shevnin et al.

products and changes superficial conductivity in pores the point with salinity 0.22 g/l, so water salinity is
(Abdel Aal et al., 2004). This effect on soil resistivity is determined. All possible soil resistivity values of this site
equivalent to the increase of clay content that can be found (at complete water saturation) are on this line - 0.22 g/l.
in the clay content estimation process and used as criteria to This soil sample is typical for clean soil of the site and shows
discriminate oil contaminated zones. clay content 43%. From geological data maximum clay
content is in the first layer and should not be more than 43%,
PETROPHYSICAL MODELING FOR MATURE OIL but there are some soils here with less clay content until
CONTAMINATED SITE pure sand. This information helps to conclude that all soils
resistivities of the site are in interval 14-110 Ohm.m, from
Petrophysical modeling consists in some analysis of soil with 43% of clay until pure sand (along line B).
information for each site (soil resistivity from VES, soil Boundary F of minimum soil resistivity of the area is equal
samples resistivity curve ρ(C) and its petrophysical model, to 14 Ohm.m. Statistical distribution of soil resistivity values
and water salinity), in order to compare all information and of the site has values in the areas A and E. This site was
estimate its consistency. All information obtained is contaminated by oil well many years ago. Mature oil
displayed on the template with coordinates: soil resistivity contamination gives low resistivity anomalies. From
– water salinity (Figure 6). The template itself is calculated petrophysical modeling it can be concluded that soil with
for a soil model obtained from soil curves–ρ(C) resistivity below the boundary value 14 Ohm.m corresponds
interpretation. This interpretation gives clay and sand to contaminated soil. A clay content exceeding 43% (for soil
parameters (porosity, capillary radii, clay CEC) and clay resistivity below the boundary F) is obtained in soil resistivity
content in soil. Template in Figure 6 was calculated for clay interpretation with PP estimation. For soil resistivity
CEC 0.12 g/l, sand porosity 0.25, clay porosity 0.55. Each interpretation between 3 and 5 Ohm.m clay CEC value
theoretical curve is marked with clay content and soil should be increased, because 100% clay with CEC=0.12 g/
porosity. There is also experimental curve for soil sample– l can not give resistivity below 5 Ohm.m. Low resistivity in
ρ(C) (Line D). Groundwater resistivity of this site is 27 the area was created by oil contamination, so clay content
Ohm.m. Horizontal line 27 Ohm.m crosses water curve in in soil more than 43% is indicator of contamination and does

200 Resistivity, Ohm.m

100 B
4-23 0-25 Clay content , %; porosity, %
70 2-24
10-20 Soil sample model:
50 Clay content 43%,
Porosity 24%,
30 20-16 A
27 Clay CEC 0.12 g/l
30-16
20
40-22 F
14
10 50-27 D
70-38

5 100-55
E

2
Water
0.22
1
0.1 0.2 0.5 1 C, g/l NaCl 10 20
Fig. 6. The example of petrophysical modeling.

186
 Estimation of soil petrophysical parameters

not show real increase of clay content. This effect is called petrophysical parameters (Figures 7 B-D) of the
quasi-clay. Such petrophysical modeling is performed for contamination plume traced similar zones as the soil
each site. resistivity anomaly does (Figure 7A).

Bold line in each map means boundary between clean

PRACTICAL EXAMPLE: MECATEPEC-3,
and contaminated soil. Boundary value was estimated
VERACRUZ
through petrophysical modeling by integration of
groundwater salinity, soil resistivity, and soil petrophysical
The results of 2D resistivity interpretation in terms of characteristics (clay content, porosity and CEC). Boundary
PP for the site Mecatepec-3, Veracruz are presented in Figure values (separating clean and contaminated soils) were
7. Four maps were obtained for the 3rd layer of 2D estimated for resistivity equal to 10 Ohm.m, for clay content
interpretation with depth from 2 to 3 m. It shows that - 80%, for porosity - 46%, for CEC - 0.5 g/l.

Y,m Y,m
1050

1040
Resistivity Rho,
Ohm.m
1050

1040
Clay
%
120 1030 100
1030
90
1020 74 1020 80
1010 45 1010 70
60
27
1000 1000 50
2
16 40
990 990
10 30
980 980 20
1 6
970 970 10
3.7 0
960 A 960 B
950 X,m 950 X,m
110 120 130 140 150 160 170 180 190 200 110 120 130 140 150 160 170 180 190 200
Y,m Y,m
1050
Porosity %
1050
CEC CEC,g/l
1040 1040

1030 50 1030 0.74

1020 45 1020
0.45
1010 40 1010
1000 35 1000 0.27
990 990
30
980 980 0.17
25
970 970
20 0.1
960 C 960 D
950 X,m 950 X,m
110 120 130 140 150 160 170 180 190 200 110 120 130 140 150 160 170 180 190 200

Fig. 7. Maps of soil resistivity (A) and petrophysical parameters: clay content (B), porosity (C) and CEC (D). Bold lines on the maps
marked contaminated zones estimated on VES. Points on A - VES positions.

187
V. Shevnin et al.

1000 0 0 0
Resistivity, Ohm.m
A B C D
0 - Clay content, %
500
2 2 2 2
4
200
4 4 4
10
100

20 6 6 6
50

8 8 8
20

Humidity, % Rho, Ohm.m Clay content, % Humidity, %

10 10 10 10
10 20 50 100 Z, m Z, m Z, m
10 20 50 100
10 20 50 100 500 0 2 4 6 8 10

Fig. 8. A. Theoretical dependence of soil resistivity versus soil humidity for different clay content in sand; B-D - Mean changes of soil
resistivity (B), clay content (C) and humidity (D) with depth, estimated from VES.

PRACTICAL EXAMPLE: KM124, TABASCO (Figure 9). As a result in maps visualization a higher
resolution can be reached with the possibility of some weaker
Site Km124 was contaminated as a result of a spill anomalies locating. To adjust resistivity range a statistical
from the pipeline. This site is in sandy soil with low clay analysis was applied, allowing to increase visualization
content (<10%). Groundwater level, at the moment of the resolution (Figure 9, Table 2) and scale range. Without
resistivity study was at a depth of 2.5 m, but it changes from special adjustment of resistivity range it is frequently not
1 to 4 m deep due to annual changes in precipitation, as well possible to locate oil contamination anomalies on sections
as the influence of a pond nearby and the high hydraulic and maps. In case of aged contamination resistivity ratio
conductivity of the soil. As a result, contaminated soil between clean and contaminated soil is from 2 to 5. To locate
smeared with oil was found in a depth interval from 1 to 3 contamination anomaly, maximum resolution of visual
m. The main problem with VES data from this site was presentation is needed and it is easier to obtain in case of
interpretation of soil resistivity and groundwater salinity in maps. Both sections and maps help us to have an idea of
terms of petrophysical parameters in conditions of vadose contamination distribution in space.
zone (incomplete water saturation) with noticeable changes
of humidity. To perform this operation, soil samples were
Algorithm of Ryjov was created for soil resistivity
collected in different parts of the site and clay content and
interpretation in terms of petrophysical parameters for all
CEC were found by means of laboratory measurements of
data with equal soil humidity. For data interpretation in the
resistivity vs salinity curves and its interpretation. Clay
vadose zone a separate interpretation for different layers
content was below 10%. Then theoretical dependencies of
needs to be used, each layer counting on its own humidity.
soil resistivity from clay content and humidity were
calculated (Figure 8A) by using Ryjov’s algorithm (Ryjov
and Sudoplatov, 1990). After 2D interpretation of the VES Thus, resistivity interpretation in terms of petrophysical
data, the mean soil resistivity distribution (Figure 8B) and parameters for each layer obtained in 2D VES interpretation
then clay content and humidity distribution with depth is performed taking into account soil humidity. Soil
(Figures 8C, 8D) were found. resistivity map for the second layer (0.7-1.6 m depth) is
displayed in Figure 10. Petrophysical maps for this layer
Which is the best form of VES data visualization, have a similar outline of the contaminated zone. Boundary
through sections or maps? Visualization in sections has less line for the second layer separating clean and contaminated
interpolation distance between measuring points. To create soil (Figure 10) is equal to 170 Ohm.m.
maps a distant interpolation between profiles is necessary.
However maps have lower resistivity range than sections Two months before resistivity survey at the Km124
(electrical properties change more with depth than in plan) site, measurements of soil gases were performed to detect

188
 Estimation of soil petrophysical parameters

Fig. 9. Histograms for each horizontal layer in data cube for the Km124 site (1-7) and histogram for all soil resistivity data (0).

189
V. Shevnin et al.

Table 2

List of initial and adjusted histogram ranges (Figure 9)

Sample collection Rho min, Rho max, Range Adjusted Adjusted Adjusted
Ohm.m Ohm.m max/min Rho min, Rho max, range
Ohm.m Ohm.m max/min

All data 10 1600 160 35 700 20

Layer 1 20 350 17 30 220 7.3
Layer 2 65 1000 15 120 700 5.8
Layer 3 100 1500 15 200 1300 6.5
Layer 4 60 1600 27 140 1100 7.9
Layer 5 55 1000 18 67 420 6.3
Layer 6 38 140 7 38 140 3.7
Layer 7 6 80 13 13 65 5
Mean value of range 16 6

Fig. 10. Soil resistivity map of the second layer in Km 124 site. Bold line is a boundary between clean and contaminated soil.

concentrations of volatile organic compounds (VOC). Soil Two superposed maps of clay content (estimated from
resistivity and VOC values have correlation coefficient -0.62 resistivity) and VOC concentrations are presented in Figure
and their dispersion is shown in Figure 11. When VOC 12. Anomalies in these maps are congruent despite of the
increasing (reflecting growth of contamination grade) fact that data for both maps were measured in different time
resistivity decreasing about 5 times. periods and with different grids. It is necessary to note that

190
 Estimation of soil petrophysical parameters

clay content anomaly actually resulted into an increase of

superficial conductivity in contaminated soil (Abdel Aal et
al., 2004). Borehole 2 was drilled at the point with the highest
VOC value and detected free product at a depth of 1.5 m
and groundwater level at 1.6 m.

CONCLUSIONS

Resistivity Imaging technology along with 2D

interpretation (with Res2DInv software) is a very useful
instrument for shallow 3D study with creating soil resistivity
cube and presenting results in sections or maps for different
layers.

To obtain maximum resolution of visualization,

sufficient for contaminated zones localization, layers maps
are more suitable.

Interpretation of resistivity values in terms of

petrophysical parameters means a step forward in geological
characterization of uncontaminated and oil contaminated
sites, studied by means of resistivity method.

ACKNOWLEDGMENTS

We are indebted to The Mexican Petroleum Institute,

where this study was performed. We thank anonymous
Fig. 11. Correlation between soil gases concentration and soil reviewers for their comments that significantly improved
resistivity for contaminated zone. the manuscript.

260
Y,m VOC,
Clay,% ppm
250

240
7 60
230

220
5 40

210 Borehole 2
Clay
VOC X,m
200 3 30
370 380 390 400 410 420 430 440 450
Fig. 12. Superposed maps of clay content and soil gases intensity for the site Km 124.

191
V. Shevnin et al.

BIBLIOGRAPHY VLADOV, 1997. Investigations of oil pollution with

electrical prospecting methods. 3rd Meeting
ARCHIE, G. E., 1942. The Electric Resistivity Logs as an environmental and engineering geophysics.
Aid in Determining some Reservoir Characteristics. Proceedings. Aarhus, Denmark, 8-11 September 1997.
SPE-AIME Transactions, 146, 54-62. p.267-270.

ABDEL, AAL, G. Z., E. A. ATEKWANA, L. D. SLATER, REVIL, A. and P. W. J. GLOVER, 1998. Nature of surface
and E. A. ATEKWANA, 2004. Effects of microbial electrical conductivity in natural sands, sandstones, and
processes on electrolytic and interfacial electrical clays. Geophys. Res. Lett., 25, 5, 691-694.
properties of unconsolidated sediments. Geophys. Res.
Lett., 31, 12, L12505 10.1029/2004GL020030 ROWELL, D. L., 1993. Soil science: methods and
applications. Longman Scientific and Technical. P. 133.
ATEKWANA, E., D. P. CASSIDY, C. MAGNUSON, A. L. ISBN 0-470-22141-0
ENDRES, D. D. WERKEMA, JR and W. A. SAUCK,
2001: Changes in geoelectrical properties RYJOV, A., 1987. The main IP peculiarities of rocks. In:
accompanying microbial degradation of LNAPL. In: “Application of IP method for mineral deposits’
Proceedings of the Symposium on the Application of research”. Moscow, 1987, 5-23. (In Russian).
Geophysics to Engineering and Environmental
Problems. 1 – 10. RYJOV, A. and V. SHEVNIN, 2002. Theoretical calculation
of rocks electrical resistivity and some examples of
ATEKWANA, E. A., E. A. ATEKWANA and R. S. ROWE, algorithm’s application. In: Proceedings of the
2003. Relationship Between Total Dissolved Solids and Symposium on the Application of Geophysics to
Bulk Conductivity at a Hydrocarbon-Contaminated Engineering and Environmental Problems.
Aquifer, SAGEEP Proceedings, pp. 228-237.
RYJOV, A. A. and A. D. SUDOPLATOV, 1990. The
BOBACHEV, A.A., 1994. IPI2Win software: http:// calculation of specific electrical conductivity for sandy
geophys.geol.msu.ru/rec_labe.htm . - clayed rocks and the usage of functional cross-plots
for the decision of hydrogeological problems. In:
BOBACHEV, A.A., 2003. X2IPI software: http:// “Scientific and technical achievements and advanced
geophys.geol.msu.ru/x2ipi/x2ipi.html . experience in the field of geology and mineral deposits
research. Moscow, pp. 27-41. (In Russian).
GOLTSMAN, F. M., 1971. Statistical models of
interpretation. Moscow., 328 pp. (In Russian). SAUCK, W. A., 1998. A conceptual model for the
geoelectrical response of LNAPL plumes in granular
KLEIN, K. A. and J. C. SANTAMARINA, 2003. Electrical sediments. In: Proceedings of the Symposium on the
conductivity in soils: Underlying phenomena. J. Application of Geophysics to Engineering and
Environ. Engin. Geophys., 8, 4, 263-273. Environmental Problems, 805-817.

LOKE, M. H. and R. D. BARKER, 1995. Least-squares SHEVNIN, V., O. DELGADO RODRÍGUEZ, A.

deconvolution of apparent resistivity pseudosections. MOUSATOV and A. RYJOV, 2004, Soil resistivity
Geophysics, 60, 1682-1690. measurements for clay content estimation and its
application for petroleum contamination study.
LOKE, M. H. and R. D. BARKER, 1996. Rapid least-squares SAGEEP-2004, Colorado Springs. p. 396-408.
inversion of apparent resistivity pseudosections using
a quasi-Newton method.Geophys. Prospect., 44, 131- SHEVNIN, V., O. DELGADO RODRÍGUEZ, A.
152. MOUSATOV, H. ZEGARRA MARTÍNEZ, J. OCHOA
VALDÉS and A. RYJOV, 2005a. Study of petroleum
MARION, D., A. NUR, H. YIN and D. HAN, 1992, contaminated sites in Mexico with resistivity and EM
Compressional velocity and porosity in sand-clay methods. SAGEEP Proceedings, Atlanta, Georgia,
mixtures. Geophysics, 57, 554-563. pp.167-176.

MODIN, I. N., V. A. SHEVNIN, A. A. BOBATCHEV, D. SHEVNIN, V., O. DELGADO-RODRÍGUEZ, L.

K. BOLSHAKOV, D. A. LEONOV and M. L. FERNÁNDEZ-LINARES, H. ZEGARRA

192
 Estimation of soil petrophysical parameters

MARTÍNEZ, A. MOUSATOV and A. RYJOV, 2005b.

Geoelectrical characterization of oil contaminated site
in Tabasco, Mexico. Geofís. Int., 44, 3, 251-263.

SHEVNIN, V., A. MOUSATOV, A. RYJOV and O.

DELGADO, 2006. Estimation of clay content in soil
based on resistivity modeling and laboratory
measurements. Geophysical Prospecting. In print.

TARANTOLA, A., 1994. Inverse problem theory. Method

for data fitting and model parameter estimation. Elsevier
Science, Amsterdam, 600 p.

__________

Vladimir Shevnin1, Omar Delgado Rodríguez1,

Aleksandr Mousatov1, David Flores Hernández1,
Héctor Zegarra Martínez1 and Albert Ryjov2
1
Mexican Petroleum Institute, Eje Central Lázaro Cárdenas
152, 07730, Mexico D.F. E-mail: vshevnin@imp.mx
2
Moscow State Geological Prospecting Academy,
Geophysical faculty, Volgina str., 9, 117485, Moscow, Russia.

193

View publication stats