Correlation of near-surface stratigraphy and physical...
Transcript of Correlation of near-surface stratigraphy and physical...
www.elsevier.com/locate/jappgeo
Journal of Applied Geophysics 53 (2003) 121–136
Correlation of near-surface stratigraphy and physical properties of
clayey sediments from Chalco Basin, Mexico, using Ground
Penetrating Radar
Dora Carreon-Freyre*, Mariano Cerca, Martın Hernandez-Marın
Centro de Geociencias, Universidad Nacional Autonoma de Mexico, Campus Juriquilla, Queretaro, Apartado Postal 1-742,
Queretaro 76001, Mexico
Received 21 September 2001; accepted 19 May 2003
Abstract
Detailed measurements of water content, liquid and plastic limits, electric conductivity, grain-size distribution, specific
gravity, and compressibility were performed on the upper 7 m of the lacustrine sequence from the Chalco Basin, Valley of
Mexico. Eight stratigraphic units consisting of alternating layers of clay, silt, sand, and gravel of volcanic origin are described
for this sequence. The analysis of contrasts in the physical properties permitted to identify potential reflectors of radar waves: (i)
change in the electrical conductivity at 0.4 m depth; (ii) increment in the clay and water content at 0.8 m depth; (iii) bimodal
behavior of the water content at 1.3 m depth; (iv) increment in the sand content and decrease in water content at 2.6 m depth;
and (v) the presence of a pyroclastic unit at 3.7 m depth. Radargrams with frequencies of 900 and 300 MHz were collected on a
grid of profiles covering the study area. Correlation of radargrams with the reference section permitted the spatial interpolation
of variations in the physical properties and the near-surface stratigraphy. Contrary to the expected in these clayey sediments,
electric contrast enhanced by variations in water content and grain size permitted the recording of the near-surface sedimentary
structures. Distinctive radar signatures were identified between reflectors. Furthermore, lateral discontinuities of the reflectors
and their vertical displacements permitted the identification of deformational features within the sequence.
D 2003 Elsevier B.V. All rights reserved.
Keywords: Clayey sediments; Near-surface stratigraphy; Physical properties; Radargram; Radar signature; Fractures
1. Introduction
Sedimentary sequences filling the Valley of Mex-
ico, and other basins in Central Mexico, consist
0926-9851/03/$ - see front matter D 2003 Elsevier B.V. All rights reserve
doi:10.1016/S0926-9851(03)00042-9
* Corresponding author. Tel.: +52-442-238-11-04x112; fax:
+52-442-238-11-05.
E-mail addresses: [email protected]
(D. Carreon-Freyre), [email protected] (M. Cerca),
[email protected] (M. Hernandez-Marın).
generally of various types of fluvial and lacustrine
deposits with particle sizes varying from gravel, sand,
and silt to clays, often interbedded with pyroclastic
rocks and lava flows. These materials present a high
variability in their properties such as strength, grain-
size distribution, mineralogical composition, and wa-
ter content. Major heterogeneities can be present also
at near-surface sequences of soils or single layers. The
mineralogical composition of the shallower layers,
generally bearing clay-size particles, can vary lateral-
d.
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136122
ly to different kinds of crystallized and amorphous
clay minerals (Warren and Rudolph, 1997). Under-
standing the mechanical behavior and precise charac-
terization of the near-surface stratigraphy of these
sequences may be critical to geological engineering
and geotechnical studies for the development of urban
infrastructure.
It has been demonstrated that electrical and geo-
technical properties of clayey sediments are closely
related (Saarenketo, 1998; Fam and Santamarina,
1997; Ohtsubo et al., 1983). Since the mechanical
behavior of fine-grained sediments also depends on
water content, Ground Penetrating Radar (GPR) pro-
vides a useful nondestructive method for geotechnical
characterization of these sequences. In the last 25
years, GPR applications for the stratigraphic prospec-
ting of sediments and for engineering geology and
civil engineering studies have been widely presented
in literature (Kruse et al., 2000; Saarenketo and
Fig. 1. Location of the studied area. The Mexico Valley is located in the ce
lacustrine basins. The Chalco Basin is located in the southern part of the M
Chichinautzin ranges. Location of the studied sequence and previous stud
Scuillon, 2000; Van Overmeeren, 1998; Doolittle
and Collins, 1995; Lorenzo, 1994; Goodman, 1994;
Davis and Annan, 1989; Ulriksen, 1982; Morey,
1974). Nevertheless, a detailed correlation of the
reflected patterns of radar wave with physical and
mechanical properties of clayey sediments has not
been presented yet. The aim of this paper is to
correlate the spatial variation of water content,
grain-size distribution, and compressibility within a
sedimentary sequence of lacustrine-volcanic origin
with the recorded reflectors in GPR sections.
2. GPR prospecting of sedimentary sequences
The principles of the GPR method have been
explained extensively in literature (Davis and Annan,
1989; Ulriksen, 1982; Morey, 1974). GPR transmits
high-frequency electromagnetic pulses in which
ntral part of the Trans-Mexican Volcanic Belt and comprises several
exico Valley. It is bounded by the Santa Catarina, Las Cruces, and
ies are indicated.
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136 123
ranges the dielectric permittivity of materials predom-
inates over electrical conductivity (Ulriksen, 1982).
For simplicity, in this text we use the term ‘‘permittiv-
ity’’ to refer to relative permittivity (ratio between
permittivity of the prospected media and free space).
This parameter indicates the polarization of the ele-
ments forming geologic materials, caused by the
application of an electrical field. At radar frequencies
polarization affects mainly the bipolar molecules of
water (Topp et al., 1980) in saturated material such as
clayey materials. According to Fukue et al. (1999), the
physical characteristics of sediments such as micro-
structure, grain size, water content, and mechanical
characteristics in clay-bearing materials are also relat-
ed to other electric properties such as resistivity.
Furthermore, Saarenketo (1998) showed that varia-
tions of dielectric properties in clay and silty particles
are related with their Cation Exchange Capacity
Fig. 2. Correlation of the near-surface stratigraphy (7 m depth) in the Cha
relative positions from southwest to northeast: (a) Core B (Caballero and O
and Site 2 excavation presented in this work; and (d) Site 1 excavation (Za
size data presented were previously reported by Zawadski (1996).
(CEC). This author concluded that the lower the
CEC the more orderly is the arrangement of water
molecules around the soil particles.
In particular, Warren and Rudolph (1997) reported a
direct relation between the clay-mineral phase (allo-
phane, smectite) with its CEC, porosity, and water
content for the basin of Mexico. Thus, radar-wave
behavior is influenced by: water content, physico-
chemical characteristics of pore water, solid–liquid–
air proportion, structure, and voids ratio of the solid
phase. The contrasts in these properties can be
recorded as reflectors in radar sections or radargrams.
In addition, these parameters are directly related with
different geological and mechanical aspects of the
sedimentary materials (Fam and Santamarina, 1997).
Several authors have shown that variations in the water
content and water retention capacity of soil particles
are related with its geological depositional conditions
lco lacustrine plain. The stratigraphic columns are presented in their
rtega-Guerrero, 1998); (b) BH4 Core (Zawadski, 1996); (c) CH core
wadski, 1996). Note variations in depth of pyroclastics layer. Grain-
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136124
(see Chandler, 2000, for clay sediments; Van Over-
meeren et al., 1997; Freeland et al., 1998, for sandy
deposits). These studies validate GPR surveying as a
useful tool for physical and geotechnical prospecting
of clayey sequences.
Fig. 3. Diagram of the grid of GPR profiles collected in the studied
zone: a, b, c, and d collected in N 20j E direction; e, f, g and h
collected in N 70j W direction. Most of the profiles are 30 m in
length except for profile g where the length is 40 m. CH core and
Site 2 excavation are described in this study.
3. Near-surface stratigraphy of the Chalco site
The Valley of Mexico is a volcanic-bounded basin
located in the central part of Mexico within the Trans-
Mexican Volcanic Belt (Fig. 1). Volcanic activity has
been intense in this area from Pliocene to recent times.
A lacustrine system characterized the Valley of Mex-
ico until pre-Hispanic times. Later, the urbanization of
the valley has forced the withdrawal of groundwater
level leaving only lake remnants and large extensions
of plains. The near-surface clayey deposits of the
Mexico Basin are mainly composed by saturated
smectitic clay and amorphous material derived from
weathering of volcanic ash (Warren and Rudolph,
1997). The complex mechanical behavior of the
‘‘Clays of Mexico City’’, especially brittle failure
associated with high water contents on highly com-
pressible materials, has been extensively studied and
different mineralogical and grain-size distributions
have been reported (Peralta-y-Fabi, 1989; Mesri et
al., 1975; Marsal and Mazari, 1969; Lo, 1962).
The Chalco Basin is located at the southernmost end
of the Valley of Mexico (Fig. 1). It is a horizontal plain
(formerly a lake) filled with lacustrine sedimentary
sequences overlying a regional granular aquifer.
According to Urrutia-Fucugauchi et al. (1994), drain-
age was restricted about 780,000 years ago by the
formation of the Chichinautzin volcanic field to the
south. Important variations in salinity or water level
related to changes in weather have occurred through the
time in the Quaternary volcanosedimentary sequence
of the Chalco Basin (Urrutia-Fucugauchi et al., 1994;
Lozano et al., 1993). We have selected a study site in
the north margin of the Chalco Basin near the Santa
Catarina village. Access to the area is provided by a
north–south paved road and it is important to mention
that an electricity line runs adjacent to it (Fig. 3).
Urbanization in the Chalco area began in the early
1980s and has not covered the total extent of the plain;
in this part, the land is still used for agriculture.
Regional extraction of groundwater is made by a line
of 14 wells named Ramal Mixquic of the Santa Cata-
rina well field.
Detailed stratigraphic sections of the upper part of
the sequence are available for the marginal and central
parts of the Chalco lake. The upper 7 m record a time
span between ca. 17,450–14,500 years to the present
according to 14C dating made in a core from the
central part of the basin (Caballero and Ortega-Guer-
rero, 1998). Zawadski (1996) described the upper 6 m
of the sequence from the margin of the lake that
consists of 11 layers of alternating clay, organic silt,
sands and gravel of volcanic origin, with thicknesses
ranging from 5 to 60 cm. Caballero and Ortega-
Guerrero (1998) assigned four stratigraphic units
(brown silt with lapilli, brown silt, gray diatomite,
and black-brown clay from top to bottom) and seven
tephra layers for the first 7 m from the central part of
the lake sequence. The physical correlation of these
sequences is based on the reconnaissance of the
Tlapacoya 2 unit (Ortega-Guerrero and Newton,
1998) in the studied CH core (see Fig. 2).
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136 125
The study area is affected by a north–northeast
trending regional fracture of several tens of meters
long, with uplifted borders, and a surface opening of
more than 1.0 m that closes with increasing depth
(Figs. 3 and 4a). The fracture is filled with clayey
Fig. 4. (a) Panoramic view of the Chalco Plain to the southwest, the regi
indicated; (b) excavation in Site 2 permitted to detail the stratigraphy of
observed.
sediments but opens seasonally. This fracture is the
most evident structural feature observed in the sur-
face; however, it has been shown that numerous
subsurface fractures and subsidence affect the lacus-
trine sequences (Ortega et al., 1993).
onal fracture with uplifted borders and the location of CH core are
the upper 2 m, and small fractures within the clay sequence were
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136126
4. Characteristics of the GPR used
The GPR Zond 12c, manufactured in Latvia by
Radar Systems, was used for the soundings. A porta-
ble computer connected to the control unit permits the
direct recording of raw data collected by the radiat-
ing–receiving antennae. With this equipment, a single
operator moves the antennae along the surveyed
surfaces in monostatic mode (the antenna operates
both as radiating and receiver). Table 1 summarizes
the measurement parameters of the survey.
The alternate application of two transmission fre-
quencies in the same profile, using shielded 900- and
unshielded 300-MHz antennae, permitted the recog-
nition of the particular types of structural features
recorded by each frequency. The analysis of a nor-
malized frequency spectrum indicates that the effec-
tive frequencies through the studied media were ca.
500 and 100-MHz for the 900- and 300-MHz anten-
nae, respectively.
Since the available antennae of the GPR were
monostatic, the analysis of subsurface velocities pro-
vided by the common midpoint (CMP) method could
not be performed. However, the detailed near-surface
stratigraphy observations permitted a satisfactory
evaluation of investigation depths (see Figs. 3 and
4b). By matching the position in depth of a specific
Table 1
GPR measurement parameters and characteristics of the radar wave
in the Chalco site
Center frequency
of antennae (MHz)
300 900
Impulse power 400 400
Length of antennae (cm) 102 52
Resolution (cm) 100 20
Recording time
window (ns)
240 100
Battery power supply (V) 12 12
Characteristics of the geological media
Estimated frequency
through the media (MHz)
100 500
Permittivity 40 40 (55)
Velocity (cm/ns) 4.7 4.7 (4.0)
Maximum depth of
penetration (m)
f 4.5–5 f 1.4–1.9
Recorded feature stratigraphy and
deformation
change in physical
property and
stratigraphy
Thickness (cm) 10–20 40–100
reflector (a stratigraphic boundary associated with
contrast in physical properties in most of the cases)
with the reflectors on the radargrams, we obtained an
accurate evaluation of the bulk velocity of propaga-
tion of electromagnetic waves through the media. This
methodology allowed the time/depth conversion using
a bulk velocity of propagation for the sequence of 4.7
cm/ns. Vertical variations in the bulk velocity of
propagation could be taken into account only for the
900-MHz sections, in which we obtained a lower
velocity (4.0 cm/ns) by adjusting the depths of well-
known reflectors such as sand lenses.
Eight profiles were obtained in two perpendicular
directions (N 20j E and N 70j W) in order to identify
the spatial variations of the recorded reflectors (Fig. 3).
North–south profiles a, b, c, and d were obtained
parallel to the regional fracture observed in the area,
whereas the east–west profiles e, f, g, and h were
perpendicular to this structure. The length of each line
profile was 30 m, except profile g that had a total
length of 40 m.
5. Methodology of laboratory analysis
Assuming that physical and mechanical variations
can produce sufficient electric contrast to be recorded
in a radargram, then the depth of specific reflectors
can be obtained from the detailed analysis of a
reference stratigraphic section. This method has been
used recently with success for correlation of radar data
with different properties of soils (Oleschko et al.,
2003). Representative samples of approximately 20
cm each from the Chalco core (CH) were analyzed at
the Geomechanical Laboratory of the Center of Geo-
sciences (National University of Mexico, UNAM).
For the first 2 m, the data were complemented with
the analysis of disturbed samples collected in the Site
2 excavation (Fig. 4b). The sequence below the water
table that occurs at about 0.8 m was considered to be
in saturated conditions.
The main physical characteristics that were mea-
sured in the sequence in order to identify physical
discontinuities that can be related to electrical con-
trasts were:
(a) Gravimetric (Wgrav%) water content is the weight
ratio between pore water and solid fraction
Fig. 5. Composite stratigraphic column of the Chalco soils using CH
core and Site 2 excavation. Ten stratigraphic units (A–J) were
described in the upper 7 m; the geometry and filling of the regional
fracture are also shown. The position of physical contrasts found in
the sequence (i –v) is described in the caption of Fig. 6. These
contrasts were identified as potential reflectors of electromagnetic
waves.
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136 127
averaged from three measurements obtained by
the ovendrying method (American Society for
Testing and Materials, ASTM D2216-92, 1998a),
whereas volumetric water content (Wvol%) was
computed by the ratio between Wgrav and the
porosity (Santamarina et al., 2001); Wgrav% is a
parameter widely used for geotechnical character-
ization of soils (see Das, 1997), a value of 300%
frequently obtained for the clays of Mexico means
that for each gram of solid there are 3 g of water;
(b) Atterberg limits (ASTM D4318-95) were deter-
mined using the Casagrande coup for the liquid
limit (LL) and the Terzaghi method for the plastic
limit (PL);
(c) Electric conductivity (EC) was measured using a
Cole Parmer TDS Conductivity meter in a
suspension of soil–water (1:2.5) (saturated paste,
USDA, 1996);
(d) grain size was determined by hydrometer and sieve
analysis (ASTM D422-63, 1998b) of samples
representative of each 20 cm; clay and silt contents
were added and presented as fine-grained sedi-
ments in Fig. 6;
(e) specific gravity of particles was determined by the
pycnometer method (ASTM D854-92, 1998c),
and;
(f) compressibility index was computed as a variation
of the void ratio in a log pressure cycle of the
compressibility curve (Lambe and Whitman,
1969), determined by the one-dimensional con-
solidation test (ASTM 2435-96, 1998d).
6. Results
6.1. Description of the near-surface sequence of
Chalco
The description of the stratigraphic units defined in
the CH Core was based on a correlation with available
column descriptions. The BH4 core and Site 1 exca-
vation of Zawadski (1996) are located within the study
area, whereas Core B of Caballero and Ortega-Guer-
rero (1998) is located in the center of the lake (Figs. 1
and 2). Notably, greater thickness of the sequence
above the Tlapacoya 2 unit is observed in Core B.
Within the study area, smaller variations in thickness
can be related to local depositional changes and/or
deformation features. Eight stratigraphic units (Fig. 5)
are described for the first 7 m depth of the studied CH
core: (unit A) a layer of organic soil with abundant
vegetal roots of approximately 0.5 m with an upper
horizon of high-salinity soils; (unit B) a layer of pale-
brown silt with pyroclastic rock fragments that suggest
intense volcanic activity during the deposition of this
layer, with thickness ranging from 1 m in the center of
the lake (Core B) to 0.1 m in the margins; (unit C) a
brown clay of high plasticity; (unit D) underlying there
is a brown-reddish clay layer with carbonate concre-
tions reaching 2 m depth, vertical fractures were
observed in the lower part of this unit in Site 2; (unit
E) a brown clayey silt separated from the overlying
clay layer by a thin lens of sand (see Fig. 3b); (unit F) a
sandy layer of variable thickness (10–30 cm); (unit G)
black silty clay of approximately 1 m thick; (unit H) 25
cm of pyroclastic gravel that correlates with Tlapacoya
2 unit of Ortega-Guerrero and Newton (1998) reported
at f 5 m depth in Core B (Caballero and Ortega-
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136128
Guerrero, 1998); and (units I and J) comprises a
sequence of brownish to greenish gray clays charac-
teristic of redox lacustrine environment and a grey-
brown clay layer with pyroclastic rocks observed at 6
m depth. The results obtained in the characterization of
the upper 7.0 m are presented in Fig. 6.
The analysis of the physical properties permitted to
identify five potential reflectors of radar waves (Fig.
6): (i) a decrease from 15 to 3 mS/cm in the electrical
conductivity at 0.4 m depth, this contrast is slightly
above (f 10 cm) that can be added to the grain-size
contrast in the upper boundary of the pyroclastic silt
unit B; (ii) an increment in the clay content associated
with an abrupt increase in gravimetric water content
(from 100% to 300%) that is observed at the upper
boundary of unit C (0.8 m depth); (iii) a bimodal
water content in the brown-reddish clays associated
with the lower boundary of the unit C at 1.3 m depth
Fig. 6. Measured physical and mechanical properties of the upper 7 m of
water content, (c) Atterberg limits (L: liquid, P: plastic), (d) electrical c
compressibility (oedometer test). Contrasts in the sequence are be observed
change in grain size (from sand to clay) and increase of water content (from
increase of fractures below high-plasticity clays at 1.35 m depth; (iv) a var
depth related to, and 3.75 m depth (v) related to a lens of pyroclastic grav
methodology. Note that units C, D, and E are high compressible soils wi
and decrease in the clay content; it is important to
mention that fracturing of the clay layers increases
below this level; (iv) an abrupt increment in the sand
content (volcanic ash layer) and a decrease in water
content from 100% to 50% at 2.6 m depth; and (v) a
lens of pyroclastic gravel at 3.7 m depth.
Because of its heterogeneous composition and high
water content a ductile–brittle mechanical behavior
was expected in the Chalco clayey sequence. As
mentioned above, units C and E present water con-
tents higher than their liquid limit, which suggest that
these materials can behave as fluids when their
structure is disturbed. In particular, a strong contrast
in the physical properties is observed in unit C,
elevated liquid and plastic limits, high compressibility
index (Cc f 3), and decrease of specific gravity. We
emphasize that the water table occurs at about 0.8 m
coincident with the increment of the water content and
the Chalco sequence: (a) gravimetric water content, (b) volumetric
onductivity, (e) grain-size distribution, (f) specific gravity, and (g)
at: (i) a decrease of the electrical conductivity at 0.4 m depth; (ii) a
100 to 300) at 0.8 m depth; (iii) a decrease of water content and the
iation of grain size (sand unit) and a decrease in water content 2.6 m
el. See text for ASTM and USDA test numbers and explanation of
th natural water content bigger than their liquid limit.
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136 129
electric conductivity slightly increases below unit C.
Finally, unit F represents a major contrast in grain
size, with greater sand content, low water content, and
greater specific gravity.
6.2. Spatial correlation of physical and mechanical
properties with GPR profiles
We present the first GPR records of stratigraphic
and structural features for the lacustrine sequence of
Chalco, Mexico. GPR surveying has permitted the
Fig. 7. Correlation of the stratigraphic column and reflectors recorded in 9
surface stratigraphy was based on projection (showed with an arrow) of th
and continuous reflectors along the radargram were associated to changes
stratigraphic units is presented in Fig. 5. The stratigraphy of the first
electromagnetic waves of 4.0 cm/ns (permittivity 55 indicated in italics) ins
the whole profile. Calculated depths of investigation for both velocities ar
identification of subtle stratigraphic heterogeneities
related to differences in water content and grain size
within the studied sediments, and allowed interpola-
tion of sedimentary characteristics such as the conti-
nuity of sand or clay units. Radar sections collected
along profile a, with 900- and 300-MHz antennae (Fig.
7a and b), were used to correlate radar signatures with
the reflectors identified previously. The projected
location of the core on profile a is portrayed in radar-
grams in Fig. 7. The above-defined potential reflectors
were then identified in the radargrams based on the
00- and 300-MHz radargrams of profile a. Correlation of the near-
e CH core located at a distance of 0.5 m from this profile. Coherent
in physical properties or stratigraphic boundaries. Nomenclature of
1.5 m is better adjusted assuming a velocity of propagation of
tead of the calculated bulk velocity of 4.7 cm/ns (permittivity 40) for
e presented in the right side of the 900-MHz profile.
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136130
changes in well-defined radar signatures and the pres-
ence of coherent radar wave reflectors. The reconnais-
sance of a reflector assumed to be the upper boundary
of the sand layer at ca. 100 ns permitted the estimation
of a value of 40 for the bulk permittivity of the
sequence. However, for the 900-MHz antenna, best
adjusting of the depth of reflectors was achieved with a
permittivity of 55 (velocity of propagation f 4.0 cm/
ns). The velocity of propagation in the media is
assessed using the simplified formula v = c/e1/2, wherec is the velocity of propagation of waves in the air and
e is the real part of the permittivity. We present
between parenthesis and italics the calculated depths
and velocity obtained with both values of permittivity
for the 900-MHz radar sections through the figures.
The five reflectors were observed in the 300-MHz
antenna; however, best definition of reflectors i, ii, and
iii was obtained using the 900-MHz antenna.
6.3. 900-MHz profiles
The records of radargrams in two perpendicular
profiles with the 900 MHz are portrayed in Fig. 8.
Fig. 8. Two perpendicular 900-MHz radargrams corresponding to profiles
profile a in Fig. 7, profile b shows coherent reflectors between saline s
reflectors were recorded in profile e for the first 15 m of horizontal distance
diffraction hyperbola (noise with a branch slope of 30 cm/ns) probably g
Lateral continuity of coherent reflectors at constant
depths along the total distance is observed in both
profiles. Small disturbances in reflectors are observed
only within the profile e perpendicular to the regional
fracture in particular, before 10 m and between 15 and
20 m in distance. It is important to note that a
diffraction hyperbola probably related to the electric-
ity line was recorded at the eastern part of all the
profiles perpendicular to the fracture.
The shaded area in the upper part of the radargram
corresponds to the direct wave (from 0 to 10 ns); the
high attenuation of energy is controlled by the high
electrical conductivity caused by a superficial high-
salinity layer of soils. The position of reflector i along
the profiles is defined by the decrease of electrical
conductivity that occurs within the underlying organic
soil layer. The response of radar waves can be added to
the reflector located at the highly irregular upper
boundary of unit B. Reflector ii is located at the in-
crease in water content coincident with the contact
between unit B and C. Notably, a distinctive and homo-
geneous radar signature is observed between reflectors
ii and iii corresponding to the relatively homogeneous
b and e, in north–south and west–east direction, respectively. As in
oils, pyroclastics, and high-plasticity and fractured clay. Similarly,
; however, from 15 to 30 m, the useful information was masked by a
enerated by the electricity line (see Fig. 2).
Fig. 9. North–south radargrams collected with the 300-MHz antenna along profiles b, c, and d. The CH core is projected from distances of 5, 9,
and 17 m to profiles b, c, and d, respectively. Calculated depth of investigation using a velocity of 4.7 cm/ns is presented in each profile. The
main reflectors are indicated, question marks indicate not well-recorded positions.
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136 131
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136132
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136 133
clayey soil mass with high gravimetric water content
(f 300%) and without structural features.
6.4. 300-MHz profiles
Three profiles with the 300 MHz, parallel to the
regional fracture, are compared in Fig. 9 in order to
show the record of the identified reflectors and their
in-depth variation and structure on both sides of the
fracture. The depth of the recorded reflectors was
fitted with the depth of defined contrasts in the
projected position of the CH core. Best definition of
structure patterns was obtained with this antenna;
thus, disturbances in the radar signature can be related
with deformational features. Clearly defined radar
signatures observed between reflectors i– ii and ii–
iii correspond roughly to pyroclastic silt unit B and to
the high-plasticity clay unit C, respectively. The
irregular patterns of reflection show sinuous forms
that suggest differential deformation of the soils in this
part of the sequence. A decrease in thickness of the
clayey sequence above unit F towards the east is
evident by the gradually shallower position of reflec-
tors iv and v.
Two examples of profiles perpendicular to the
regional fracture are shown in Fig. 10a and b. In this
direction, lateral continuity of reflectors is highly
disturbed and heterogeneous radar signatures recorded
suggest that deformation effects are more intense in
this direction. A satisfactory position of the reflectors
was obtained by correlating the depth of reflectors
identified in the north–south profiles. Differences in
depth of reflectors become evident below reflector ii,
and distortion of layers is best observed in the radar
signature obtained between reflectors iv and v that
correspond to the units F and G. These radar sections
are not corrected topographically because small uplift
(f 20 cm) of the horizontal surface is only observed
in a 40-cm-wide band on both sides of the regional
fracture (see Fig. 4a). The flower-like geometry of the
Fig. 10. West–east radargrams collected with the 300-MHz antenna along
identified in north–south radargrams (Fig. 9) are indicated under its corresp
line and question marks indicate a not well-recorded reflector. Some usefu
slope of 30 cm/ns. The depth of this reflector decreases gradually in prof
caused by the electricity line. (c) Interpretative section of profile g; vari
differential deformation of high-plastic layers with high water content (ir
regional fracture is indicated with a dotted line in a schematic way. Note stra
located in both sides of the regional fracture.
regional fracture and the inferred effect of uplifting on
reflectors are presented in the interpretative section
(Fig. 10c).
7. Discussion
The Chalco near-surface lacustrine sequence con-
sists of alternating layers of plastic clay, volcanic
ashes, pyroclastic sand and gravel. The abundance
of pumiceous concretions within clayey units indi-
cates that volcanic activity was coeval with deposition
of the sequence. The upper unit of the sequence is
characterized by a high-salinity clayey layer formed in
an arid climate that prevailed from the beginning of
the Holocene (Urrutia-Fucugauchi et al., 1994).
The use of GPR has been particularly useful for the
characterization of the Chalco sequence because: (a)
the presence of volcanic layers within the clayey
sequence increases the reflection coefficient by en-
hancing physical and chemical contrasts (i.e. the high
water content of unit C); (b) the presence of deforma-
tion structures with a particular radar signature; and (c)
the detailed stratigraphic studies available that have
served as a basis for the correlation of the physical
properties and stratigraphic units.
The GPR method was expected to have a low
detection capacity in these high water retention ca-
pacity and low electrical resistivity clayey materials.
However, water content variations and grain-size
contrasts improved the capacity of detection consid-
erably. For example, assuming a bulk permittivity of
30 and 40 for saturated sand and clay layers, respec-
tively, the coefficient of reflection in this contact is
approximately 0.01, which lies at the limit of detec-
tion capacity of a standard radar (Annan, 1992; Davis
and Annan, 1989). In more conductive clay, permit-
tivity can increase considerably because of the salt
content of pore water. An increment of f 30% in the
permittivity of the upper part of the sequence (from 40
the: (a) profile e and (b) profile g. The i, ii, iii, iv, and v reflectors
onding crossing profile. The fracture zone is indicated with a dotted
l information was masked by a diffraction hyperbola with a branch
iles approaching to the electrical pole, suggesting that this noise is
ations in depth of the reflectors were correlated and interpreted as
regular patterns between reflectors ii and v). The geometry of the
tigraphical and structural changes of the soil mass in profiles c and d
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136134
to 55) is sufficient to double the reflection coefficient
(to 0.02) and a high dielectric contrast between the
clay and sand layers is recorded.
Ground Penetrating Radar has proven to be an
effective instrument for prospecting near-surface
stratigraphic variations in saturated lacustrine sedi-
ments due to the close relation between the behavior
of electromagnetic waves in the medium and physical-
mechanical characteristics of soils (water content,
salinity, grain size, compressibility, and plasticity).
The analysis of radar profiles permitted the spatial
correlation of near-surface stratigraphic characteristics
as well as the identification of discontinuities of the
recorded reflectors associated to subsoil deformation.
The identification of coherent reflectors along the
profiles related to the electric contrast was also useful
for the record of deformational features observed in
the clayey layers. In this way, the single vertical
fractures or fractured zones could be identified in
several radargrams.
The different estimated values of the velocity of
propagation of electromagnetic waves in the pro-
spected media and the achieved detection depths are
summarized in Table 1. Notably, the highest resolution
in stratigraphy was obtained for the first 1.5 m allowing
a better calibration of the reflectors obtained with the
900-MHz shielded antenna. Using the 900-MHz an-
tenna, the recorded reflectors correspond to vertical
variations in physical properties coincident with strati-
graphic boundaries. Thus, reflectors ii and iii are clearly
associated with the change in water content at the upper
and lower boundaries of unit C. One exception is
reflector i that is better linked to a vertical change in
electric conductivity because the corresponding upper
boundary of unit B is highly irregular. As expected,
best adjusting of the depth to reflectors was obtained
using a lower velocity of propagation because shallow
layers have higher values of electrical conductivity,
water content, and plasticity.
A discontinuous radar signature observed between
reflectors ii and iii in profiles with the 300-MHz
antenna suggests deformation of unit C. Differential
deformation of this high-plasticity compressible clays
layers can occur by self-weight consolidation, or
because they can behave as a fluid under certain
externally forced conditions (e.g. produced by seismic
activity or depressurization of subsoil). Lateral heter-
ogeneities of the clayey sequences can also have an
important influence on its mechanical behavior, par-
ticularly in the differential consolidation.
This deformation can also be related to the origin
and propagation of the regional fracture observed in
the study area. This fracture opens and closes season-
ally likely due to shrinking and swelling of the clayey
sequence. During dry weather conditions, the fracture
opens as a consequence of the water content loss and
is partially filled by detritus material. During the rainy
season, the clayey sequence expands closing the
fracture and, as new material has been added, the
deformation is accommodated upwards.
Detailed GPR field studies relating detection ca-
pacity to variations of velocity of propagation in
geological materials are scarce albeit theoretical stud-
ies have been developed about propagation of elec-
tromagnetic waves. A quantitative analysis should
involve complex calculations considering among oth-
er parameters, the efficiency of the used radar, the
transmitted frequency, electrical contrasts, and the
attenuation of the signal. The radargrams presented
in this paper show that some aspects of the velocity of
propagation variations within a stratified geological
sequence can be qualitatively assessed. Interpolation
of near-surface physical properties of soil using GPR
is at an initial stage in the case of clayey Chalco.
Nevertheless, encouraging results were obtained by
the correlation of recorded reflectors in profiles with
changes in specific physical properties and relevant
applications can be envisaged both in geological and
geotechnical studies.
8. Conclusions
A comparative analysis of GPR profiles supported
on detailed punctual analysis of mechanical and
physical properties permitted the extrapolation of
lateral and vertical variations of the near-surface
stratigraphy of a lacustrine sequence in the Chalco
Basin. Contrary to the expected in these saturated
sequences, electric contrast enhanced by abrupt
changes in water content and grain size permitted
satisfactory recording of the sedimentary structures.
Distinctive radar signatures were obtained for each
layer present between coherent reflectors. The best
definition in stratigraphy and structure was obtained
with the 300-MHz antenna, whereas grain size and
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136 135
physical variations were best recorded using the 900-
MHz antenna. An irregular radar signature with sin-
uous form recorded the presence and geometry of a
highly plastic layer. Furthermore, lateral discontinu-
ities of the reflectors and their vertical displacements
permitted the identification of fractures zones and
differential deformation within the sequence.
Acknowledgements
The authors thank Klavdia Oleschko for valuable
advice. We are grateful to Gabor Korvin and Barbara
Martiny for discussions and helpful comments on an
early version of the manuscript. Two anonymous
reviewers provided constructive criticisms that enor-
mously helped to improve this work. We thank
Ricardo Carrizosa for his help on the analysis of the
Chalco cores at the Geomechanics Laboratory of the
Center of Geosciences, National University of Mexico
(UNAM). Financial support for this research provided
by the Program of Support to Technologic Researches
and Innovations UNAM, PAPIIT 1N111398 is grate-
fully acknowledged.
References
American Society for Testing and Materials, 1998a. D2216-92
standard test method for laboratory determination of water
(moisture) content of soil and rock. ASTM standards. 4 pp.
American Society for Testing and Materials, 1998b. D422-63
(1990)e standard test method for particle-size analysis of soils.
ASTM standards. 7 pp.
American Society for Testing and Materials, 1998c. D854-92 stand-
ard test method for specific gravity of soil solids by water pyc-
nometer. ASTM standards. 7 pp.
American Society for Testing and Materials, 1998d. ASTM D2435-
96 standard test method for one-dimensional consolidation
properties of soils. ASTM standards. 5 pp.
American Society for Testing and Materials. D4318-95a standard
test method for liquid limit, plastic limit, and plasticity index of
soils. ASTM standards. 14 pp.
Annan, A.P., 1992. Ground Penetrating Radar Workshop Notes.
Sensors and Software, Ontario, Canada. 128 pp.
Caballero, M., Ortega-Guerrero, B., 1998. Lake levels since about
40,000 years ago at Lake Chalco, near Mexico City. Quaternary
Research 50, 69–79.
Chandler, R.J., 2000. Clay sediments in depositional basins: the
geotechnical cycle. Third Glossop lecture. Quarterly Journal
of Engineering Geology and Hydrogeology 33 (Part 1), 7–39.
Das, B.M., 1997. Advanced Soil Mechanics. Taylor & Francis,
USA. 125 pp.
Davis, J.L., Annan, A.P., 1989. Ground Penetrating Radar for high
resolution mapping of soil and rock stratigraphy. Geophysical
Prospecting 37 (5), 531–551.
Doolittle, J.A., Collins, M.E., 1995. Use of soil information to
determine application of Ground Penetrating Radar. Journal of
Applied Geophysics 33 (1–3), 101–108.
Fam, M.A., Santamarina, J.C., 1997. A study of consolidation using
mechanical and electromagnetic waves. Geotechnique 47 (2),
203–219.
Freeland, R.S., Yoder, R.E., Ammons, J.T., 1998. Mapping shallow
underground features that influence site-specific agricultural
production. Journal of Applied Geophysics 40, 19–27.
Fukue, M., Minato, T., Horibe, H., Taya, N., 1999. The micro-
structures of clay given by resistivity measurements. Engineer-
ing Geology 54, 43–53.
Goodman, D., 1994. Ground-penetrating radar simulation in engi-
neering and archaeology. Geophysics 59 (2), 224–232.
Kruse, S.E., Schneider, J.C., Campagna, D.C., Inman, J.A., Hickey,
T.D., 2000. Ground penetrating radar imaging of cap rock,
caliche and carbonate strata. Journal of Applied Geophysics
43 (2–4), 239–249.
Lambe, W.T., Whitman, R.V., 1969. Soil Mechanics. John Wiley
and Sons, Inc., New York, NY. 582 pp.
Lo, K.Y., 1962. Shear strength properties of a sample of volcanic
material of the Valley of Mexico. Geotechnique 12 (4), 303–310.
Lorenzo, C.E., 1994. Prospeccion geofısica de alta resolucion me-
diante Georadar. Aplicacion a obras civiles. PhD thesis, Univer-
sidad Compultense de Madrid. 203 pp.
Lozano, G.S., Ortega, G.B., Caballero, M.M., Urrutia, F.J., 1993.
Late Pleistocene and Holocene paleoenvironments of Chalco
Lake, Central Mexico. Quaternary Research 40, 332–342.
Marsal, R.J., Mazari, M., 1969. The subsoil of Mexico City.
School of Engineering, National University of Mexico (UNAM),
Mexico City. 377 pp.
Mesri, G., Rokhsar, A., Bohor, B.F., 1975. Composition and com-
pressibility of typical samples of Mexico City clay. Geotech-
nique 25 (3), 527–554.
Morey, R.M., 1974. Continuous subsurface profiling by impulse
radar. Proceedings of Conference on Subsurface Exploration
for Underground Excavation and Heavy Construction. Am.
Soc. of Civil Engineers, Henniker, NH, pp. 213–232.
Ohtsubo, M., Takayama, M., Egashira, K., 1983. Relationships of
consistency limits and activity to some physical and chemical
properties of Ariake marine clays. Soils and Foundations 23 (1),
38–46.
Oleschko, K., Korvin, G., Figueroa Sandoval, B., Vuelvas, M.A.,
Balankin, A., Flores, L., Carreon-Freyre, D., 2003. Radar fractal
scattering from soil. Physical Review E 67, 1–13 (041403).
Ortega, M.A., Cherry, J.A., Rudolph, D.L., 1993. Large-scale aqui-
tard consolidation near Mexico City. Ground Water 31 (5),
707–718.
Ortega-Guerrero, B., Newton, A.J., 1998. Geochemical character-
ization of Late Pleistocene and Holocene tephra layers from
the Basin of Mexico, Central Mexico. Quaternary Research 50,
90–106.
D. Carreon-Freyre et al. / Journal of Applied Geophysics 53 (2003) 121–136136
Peralta-y-Fabi, R., 1989. Sobre el origen de algunas propiedades de
la formacion arcillosa superior del Valle de Mexico. Simposium
sobre Topicos Geologicos de la Cuenca del Valle de Mexico,
Sociedad Mexicana de Mecanica de Suelos. SMMS, Mexico
City, pp. 43–53.
Saarenketo, T., 1998. Electrical properties of water in clay and silty
soils. Journal of Applied Geophysics 40, 73–88.
Saarenketo, T., Scuillon, T., 2000. Road evaluation with Ground
Penetrating Radar. Journal of Applied Geophysics 43, 119–138.
Santamarina, J.C., Klein, K.A., Fam, M.A., 2001. Soils and Waves.
Particulate Materials Behavior, Characterization and Process
Monitoring. Wiley, New York, NY. 485 pp.
Topp, G.C., Davis, J.L., Annan, A.P., 1980. Electromagnetic deter-
mination of soil water content: measurements in coaxial trans-
mission lines. Water Resources Research 16 (3), 574–582.
Ulriksen, P., 1982. Application of impulse Radar to Civil Engineer-
ing. Doctoral thesis, Lund University of Technology, Dept. of
Eng. Geol., Sweden. Coden: Lutvdg/(TVTTG-1001)/ 1–175.
Urrutia-Fucugauchi, J., Lozano, G.S., Ortega, G.B., Miranda,
C.M., Hansen, R., Bohnel, H., Negendank, J.F.W., 1994. Pa-
laeomagnetic and palaeoenvironmental studies in the Southern
basin of Mexico: I. Volcanosedimentary sequence and basin
structure of Chalco Lake. Geofısica Internacional 33 (3),
421–430.
USDA, 1996. 8A: saturated paste, mixed. United States Department
of Agriculture. Soil Survey Laboratory Methods Manual. Soil
Survey Investigation Report No. 42. National Soil Survey Cen-
ter, USA. 693 pp.
Van Overmeeren, R.A., 1998. Radar facies of unconsolidated sedi-
ments in the Netherlands: a radar stratigraphy interpretation
method for Hydrogeology. Journal of Applied Geophysics 40,
1–18.
Van Overmeeren, R.A., Sariowan, S.V., Geherls, J., 1997. Ground
Penetrating Radar for determining volumetric soil water content:
results of comparative measurements at two test sites. Journal of
Hydrogeology 197, 316–338.
Warren, C.J., Rudolph, D.L., 1997. Clay minerals in basin of Mex-
ico lacustrine sediments and their influential on ion mobility in
groundwater. Journal of Contaminant Hydrology 27 (3–4),
177–198.
Zawadski, A., 1996. Investigations regarding the origin and hydro-
logic activity of a large sediment-filled fissure in fractured la-
custrine clay near Mexico City. MSc thesis, University of
Waterloo, Ontario, Canada. 211 pp.