International Journal of Rock Mechanics & Mining Sciences 67 (2014) 78–87

Contents lists available at ScienceDirect

International Journal of
Rock Mechanics & Mining Sciences
journal homepage: www.elsevier.com/locate/ijrmms

Ground penetrating radar and structural geological mapping

investigation of karst and tectonic features in flyschoid rocks
as geological hazard for exploitation
Marjana Zajc a,n, Željko Pogačnik b, Andrej Gosar a
a
University of Ljubljana, Faculty of Natural Sciences and Engineering, Department of Mining and Geotechnology, Aškerčeva 12, Ljubljana, Slovenia
b
Salonit Anhovo d.d., Anhovo 1, Deskle, Slovenia

art ic l e i nf o a b s t r a c t

Article history: We evaluated the use of ground penetrating radar (GPR) in detecting karst cavities and discontinuities
Received 22 February 2013 that could form potential landslide surfaces in flyschoid rocks of the Rodež open pit mine in Anhovo
Received in revised form (W Slovenia). We recorded 21 GPR profiles in three consecutive benches with the unshielded 50 MHz
20 December 2013
rough terrain antenna (RTA) system, and correlated them with the results of detailed structural and
Accepted 8 January 2014
lithological mapping of the area. We located several karst cavities and confirmed the presence of
Available online 24 February 2014
discontinuities with the interpretation of GPR profiles alone. However, their correlation with geological
Keywords: and structural data gave a more precise insight into the structural setting of the studied area. The different
Ground penetrating radar discontinuity families specified in the Rodež open pit mine are mainly the result of the strike-slip tectonics
Rough terrain antenna system
and gradual anisotropic lithology. The complexity of the intersections of structural discontinuities and the
Karst formations
mechanical properties of rocks contribute to the formation of sliding surfaces and the developing of karst
Flyschoid rocks
Rodež open pit mine features. Cavities and phreatic channels develop in the vadose zone in calcarenites and result from complex
structural deformation and karstification factors at the open joints or larger structural fractures.
& 2014 Elsevier Ltd. All rights reserved.

1. Introduction open pit is developed in 14 benches and spans 300 m in elevation.

Altering sedimentological characteristics and complex geological
structures (e.g. joints and faults) of rock massifs make orientating
and positioning of slopes a major challenge in designing an open pit
mine. The initial movements in the slope are often associated with
stress relaxation, i.e. linear elastic deformation caused by unloading
mining works. The first clear evidence of such instability in an open
pit is the activation of sliding surfaces. The development of a
complete geotechnical model that consists of four interlinked
components (geology and structure of the ore body, open pit rock
mass characteristics and hydrogeology) can prevent such hazardous
incidents from happening. In this study we focused our research on
the structural and geological properties of the Rodež open pit mine,
developed in flyschoid rocks (Fig. 1).
The Rodež quarry is the main source of raw material for the
production of cement in the company Salonit Anhovo, which is
the largest cement producer in Slovenia. The quarry is located in
the Soča Valley (W Slovenia), 10 km north of Nova Gorica, on the
west side of Mt. Jelenk at the margin of Banjščice plateau. The
n
Corresponding author. Tel.: þ 386 1 4704 503.
E-mail address: marjana.zajc@ntf.uni-lj.si (M. Zajc).
The Palaeogene flyschoid sedimentary rocks in the studied area
belong to the Lower and Middle Palaeocene [1]. Detailed geologi-
cal mapping revealed a complex structural heterogeneity of
the beds.
Structural discontinuities are related to the formation of several
karst features (e.g. small cavities) in the carbonate units of
the Rodež megabed. The possible occurrence of hidden under-
ground cavities is very hazardous for exploitation, and can lead to
dangerous and costly accidents like collapses under quarry
machines (Fig. 2). In most cases, the underground karst structures
develop in the vadose zone and result from complex structural
deformation and karstification factors [2] at the open joints
or larger structural fractures. To reveal the spatial occurrence
of cavities, it is necessary to obtain a detailed insight into
the distribution of structural discontinuities for each part of the
mining area. The same discontinuities can also develop into sliding
plane structures and, depending on their orientation, can cause
slope failure during mining (Fig. 3).
For this reason, we decided to study the structural and
karst phenomena in the area with the low frequency ground
penetrating radar (GPR) method, which has been proven effective
in many different case studies (e.g. [3,4]). Using geophysical
investigations, we wanted to prove the existence of underground

http://dx.doi.org/10.1016/j.ijrmms.2014.01.011
1365-1609 & 2014 Elsevier Ltd. All rights reserved.
 M. Zajc et al. / International Journal of Rock Mechanics & Mining Sciences 67 (2014) 78–87 79

Fig. 1. (A) Geological position of studied area (blue area) [8] with corresponding lithostratigraphical interpretation. (B) Red polygon indicates area shown in Fig. 5.
(For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 2. Work accident during loading of raw material at E255 bench. (A) Roof collapse of phreatic channel along discontinuity (a) and joint structures, some with the same
dip as bedding plane (b); (B) karst channel with visible flow direction (yellow arrow); (C) sinking of loader wheel into karst cavity. (For interpretation of the references to
color in this figure legend, the reader is referred to the web version of this article.)

geomorphological structures (cavities, phreatic channels, etc.) in potential landslide surfaces or other geomechanical instabilities
anisotropic lithological horizons, which are related to gradual related to the structural discontinuities within gradual anisotropic
bedding. On the basis of the conducted research, we evaluated lithology. We correlated the GPR results with the results of struc-
the suitability of the GPR method for tracking karst structures and tural and lithological mapping.
80 M. Zajc et al. / International Journal of Rock Mechanics & Mining Sciences 67 (2014) 78–87
Fig. 3. Samples of differently orientated structural discontinuities at bench or ramp face with potential single or double plane sliding.
2. Study area
The sedimentological environment of the study area is part of
the Friuli Palaeogene basin [5], which also includes the Julian or
Slovenian basin [6,7]. From the structural point of view, the study
area is located on the north-eastern part of the Friuli Palaeogene
basin and the Trnovo nappe. The Palaeogene flyschoid sedimen-
tary rocks in the area belong to the Lower and Middle Palaeocene,
as determined with nanoplankton analysis [8]. During the detailed
geological mapping of the surrounding area and narrower zone of
Anhovo [9], a complex structural heterogeneity of the beds with
deep turbidite fans was revealed. Within the turbidite fans, the
Outer fan, with sediments transported in the axial direction of
the basin from the northwest [1,10], can be distinguished from the
upper part of the Second fan, which consists of calcareous mega
and proximal turbidites (alteration of siltstones, claystones, marl-
stones with carbonate breccia and grainstones; denoted as aSCLy
in Fig. 4) that filled up the sedimentary basin from the southeast
and partly east [1].
As seen in Fig. 4, the lowest part of our study area consists of
calcareous breccia (CB), followed upwards by coarse grained
carbonate sandstones (cGCS), middle and fine grained carbonate
sandstones (mfGCS) and marlstones (M). The highest part of the
study area consists of alternating siltstones and claystones (aSCly).
The rock mass structure of the Rodež quarry can be classified as
blocky (undisturbed rock mass consisting of angular to cubical
blocks formed by intersections of differently orientated disconti-
nuity sets) to very blocky (partially disturbed rock mass with
angular blocks formed by differently orientated joint sets). Pre-
vious and current mining works show that, in addition to
sedimentological and hydrological properties, the orientation of
structure discontinuities is also a controlling factor of potential
slope instability and therefore a potential geological hazard
(Fig. 3). The development of karst features along these disconti-
nuities can also lead to the formation of collapse structures. An
example of such an accident is shown in Fig. 2, where a roof of a
phreatic channel collapsed under the weight of a quarry machine,
creating a sink hole at the E255 bench. The phreatic channel
developed along a subvertical discontinuity and was at least 3 m
deep.
3. Methodology
3.1. GPR method
Due to its efficiency and non-destructive nature, the use of the
GPR method has been increasing rapidly in geological studies, e.g.
in the analysis of faults and fractures, in the detection of karstic
structures and cavities, in the mapping of sediments and in the
watertable depth determination [11]. It has also been applied in
turbidite outcrop imaging [12].
GPR is a geophysical method based on the principles of electro-
magnetic waves emitted from the transmitting antenna into the
subsurface and reflected from the boundaries between rock layers
and other objects with different dielectrical properties. The time
elapsed between the transmission of the signal and its return back
to the receiving antenna after reflecting in the subsurface (two-
way travel time) is measured and later converted to depth [11,13].
The depth range depends mostly on the frequency of the antennas
used, while a number of other factors, e.g. signal attenuation,
presence of water and clay fractions in the ground as well as
system dynamic range, are important as well [13,14].
The choice of antenna frequency depends on the desired
penetration depth and data resolution. Lower frequency antennas
allow for greater depth ranges, albeit at a reduced resolution,
which makes them most suitable for structural–geological surveys
where reaching greater depths while still achieving a satisfactory
resolution is key. Historical development, basic principles and
applications of the GPR method have been described in various
publications [11,13–18].
Its design and operating principle have proven the GPR method
to be a suitable tool for detecting cavities and other geomorpho-
logical structures beneath the ground surface in many case studies
like [19–22]. As stated in [23], the GPR techniques appear to be
the most popular geophysical tool for identifying and locating
subsurface karst features such as cavities, channels, conduits
and solutionally enlarged fractures. The efficiency of its appli-
cation generally increases with the use of low frequencies (below
100 MHz), which contribute to a weaker signal attenuation [23].
3.1.1. Data acquisition
The aim of the study was to evaluate the suitability of the low
frequency GPR method for tracking karst structures and potential
sliding surfaces or other geomechanical instabilities related to the
structural discontinuities within gradual anisotropic lithology.
A total of 21GPR profiles were measured on three consecutive
benches of the Rodež quarry (Fig. 5). The benches are labelled
according to their elevation (280, 300 and 320 m). On each bench,
the longitudinal profiles were oriented parallel to the quarry walls
to minimize the influence of scattered signals. The shorter trans-
verse profiles on the benches E280 and E300 were oriented more
or less perpendicular to quarry walls and parallel to each other.
We recorded 12 profiles on the E280 bench, 8 profiles on the E300
bench and 1 profile on the E320 bench (Fig. 5). Since the profiles
were recorded in straight lines and the surfaces of the quarry
benches were horizontal, this resulted in a grid-like configuration
of GPR profiles. The starting and ending points of profiles were
defined with differential GPS measurements.
In this study, we used the Mala ProEx GPR recording unit with
an unshielded 50 MHz rough terrain antenna (RTA). Compared to
normal rigid antennas, the flexible “snake-like” design of the RTA
system, recently developed by Mala, enables easier manoeuvring
 M. Zajc et al. / International Journal of Rock Mechanics & Mining Sciences 67 (2014) 78–87 81

Fig. 4. Lithological and structural map of study area with geological profile GEOL. Yellow rectangle indicates studied area shown in Fig. 5. (For interpretation of the references
to color in this figure legend, the reader is referred to the web version of this article.)

also through rugged terrain without affecting ground contact and
a considerable decrease in data acquisition time, providing good
results also in difficult environment. The length of 50 MHz RTA is
9.25 m and the distance between the transmitting and the receiv-
ing antenna is 4 m [24]. During the recording time, a trigger
mechanism with the time interval of 0.2 s was used which requires
steady movement along the profile lines.
Given the RTA system is unshielded, the signals are trans-
mitted in all directions and not only into the ground as is the case
with shielded antennas. Consequently, the receiving antenna also
detects incoming signals from different directions, including
those reflected from objects above ground, e.g. trees, power lines,
quarry walls, etc. This results in unwanted noise signals in the
form of “air reflections”, which can be identified during the
processing steps by velocity analysis. The hyperbolas fitted to
these reflections show the signals were travelling at the velocity
of 30 cm per nanosecond, which corresponds to the dielectric
constant ε ¼ 1. The only material that fulfils this criterion is air,
meaning the source of the incoming signals has to be above
ground.
82 M. Zajc et al. / International Journal of Rock Mechanics & Mining Sciences 67 (2014) 78–87
Fig. 5. Studied area of Rodež quarry with benches and positions of 21 GPR profiles,
geological profile (GEOL, Fig. 4) and 4 integrated geological—GPR (GGPR) profiles
(Fig. 9).
3.1.2. Data processing
The RadExplorer 1.4 software from DECO Geophysical [25] was
applied for the GPR data processing. We defined the geometry of
profiles by entering the coordinates of the start and end points of
profile lines. Since the data were acquired with the time trigger
and since maintaining a steady pace while measuring is not always
possible, Spatial Interpolation was applied first. This process
recalculates the data with the interpolation of traces in the
horizontal direction [25]. This enabled the program to calculate
the distances between the traces, which is necessary for the pro-
per functioning of some processing routines used later in the
process (e.g. hyperbola fitting) [25]. The processing steps that
followed were DC removal, time-zero adjustment, background
removal, amplitude correction and bandpass filtering.
Based on the diffractions that represent cavities and can be
seen in the GPR profiles in Fig. 8, velocity determination was
conducted through hyperbola fitting. Most diffractions gave the
velocity of 12.2 cm per nanosecond which corresponds to the
material dielectric constant ε ¼6.
3.2. Structural geological mapping
In order to explore the possibility of slope failure during mining
(Fig. 3), we investigated major structures (e.g. inter-ramp faults)
and other structural elements (e.g. joints, bench scale faults),
as well as anisotropic changes in sedimentological and mineralo-
gical features seen on the outcrops.
During geological mapping, the benches E320, E300, E280 and
E255 of the Rodež quarry were studied in detail. For the purpose of
determining the nature of structural discontinuities, the following
properties were surveyed: dip and strike, aperture and type of
filling material (clays or clayey material and fault gouge can have
an important impact on strength defects), and orientation of
bedding planes (to establish the structural impact of joint sets
orientated perpendicular to bedding).
The results of previous field studies [9] as well as the chemical
and mineralogical composition of lithological units [10] (Fig. 6)
were also included in the research.
4. Results
Based on the structural and lithological mapping (profile GEOL
in Fig. 4), the surveyed area includes several families of structural
discontinuities, including the structure in the coarse grained
carbonate sandstones (cGCS) (GGPR2 in Fig. 9) which has also
been mentioned by [9] in his structural study of the Rodež quarry.
A total of 70 discontinuities present within the CB and cGCS
horizons were measured. On the basis of the results, we were able
to define differently orientated discontinuity families that cut
through the entire carbonate lithology: a system with a 21 to
501 dip towards SSW with a mostly reverse sense (J1), a system
with a 481 dip towards NNE (Jb), a system with a 751 dip towards
NNE and a left-lateral system with a subvertical dip between 74
and 801 towards WNW (J2). This subvertical J2 system directs the
percolating water towards the less steep J1 system, contributing to
the development of karst structures along the latter. Among the
discontinuities, a clayey fault zone with a 451 dip and a 3101
azimuth (J3), which cuts off the phreatic channels developed on
the subvertical J2 system, was defined as well.
The rock competence enables the formation of open type
discontinuities (J1 and J2 in Fig. 7A). The internal friction angle
for CB, cGCS and mfGCS varies from 44 to 411 and shows a
geomechanical variety as a result of rheological characteristics of
turbidite flows. The J2 system serves as a collector of percolating
water, enabling the formation of karst features at the contact with
less permeable rocks. The ground water flow and consequently the
genesis of morphological structures are mainly controlled by
mineralogical properties (the relationship between the CaCO3,
kaolinite, chlorite and muscovite; Fig. 6) and by mechanical
properties, e.g. fracturing of lithostratigraphic units within the
carbonate complex [26,27]. Within the structural mesh formed by
differently orientated joint sets (J1 and J2 with Jb discontinuities),
rock disintegration at the intersection of two or more joints
resulted in the development of phreatic channels (Fig. 7B) and
cavities (Fig. 7C) that can be seen on the E280 bench face. Along
these discontinuity families, another structure was also mapped.
The clayey fault zone with a 451 dip and 3101 strike cuts off the
phreatic channels developed along the subvertical J2 system,
which can be seen on the E280 bench face (Fig. 7D).
With the use of GPR, we wanted to determine the presence of
subsurface karst features and the discontinuities along which they
can develop (as seen on the E280 bench face, Fig. 7). The profiles
that were used for the structural analysis of the quarry are marked
on the geological-GPR (GGPR) profiles in Fig. 9 (numbers 2–10, 14
and 20). The interpretations of some of these GPR profiles are
shown in Fig. 8. Since the deeper sections of profiles gave no
significant results, all profiles show only the first 350 ns of the
two-way travel time, which corresponds to about 20 m in depth,
the approximate height of the benches. Due to the fact that the
GPR profile 1 was recorded close to the edge of E280 (Fig. 5), the
scattered signals from the bench edge made it unsuitable for
interpretation. We found that the best position of longitudinal
profiles, which produced the least scatter from the parallel bench
edges and quarry walls, was in the middle of the benches. There-
fore, only the longitudinal profiles 3 (bench E280) and 14 (bench
E300) are shown (Fig. 5). Among all of the transverse profiles
recorded, the profiles 6, 7, 10 and 20 are shown. The transverse
profiles intersect the longitudinal profiles in key areas (Fig. 8).
In the GPR profile 3, we were able to determine two cavities at
the approximate depth of 5 m. Perpendicular to this profile and
close to the first cavity (between the 60 and 70 m mark), the
profile 7 was recorded. In this profile, we can see the same cavity
and the discontinuity along which it developed. The second cavity
in the profile 3 (between the 90 and 100 m mark), which was
visible in the lower E255 bench face, can also be seen in the profile
 M. Zajc et al. / International Journal of Rock Mechanics & Mining Sciences 67 (2014) 78–87 83

Fig. 6. Schematic cross section of cyclothemes for research area and results (normalised data) of chemical (weight amount of SiO2, Al2O3 and Fe2O3, defined by XRF method)
and mineralogical (sum weight amount of kaolinite—k, chlorite—k and muscovite—m; defined by XRD and TG methods for weight amount of CaCO3) analysis of all
lithological units in research area [after 10].

Fig. 7. Example of karst cavity and open type discontinuity on E280 bench face, (A) different discontinuity families (J1, J2 and Jb) that lead to formation of sliding surfaces
((a)–(c); pale blue area). (B) Phreatic channels (with traces of transported cave sediment—yellow dashed lines). (C) Roof of the cavity developed at Jb. (D) clayey fault zone J3
(red arrows) cutting off phreatic channel (yellow arrow). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this
article.)

10. Again, we can see the same discontinuity in both profiles.
Profile 6 (on the 50 m mark) shows only the discontinuity with no
signs of cavities. There are also no cavities visible at the same
location in the profile 3 (between the 50 and 60 m mark). Since
the vertical resolution of the 50 MHz RTA system used is about
1 m in carbonates [3], it is impossible to determine whether or not
there are more parallel discontinuities from GPR results alone.
Therefore, the results of structural and lithological mapping were
also used in the construction of four GGPR (geological-GPR) cross
sections (Fig. 9).
In the higher E300 bench, the longitudinal profile 14 shows
at least three cavities, located more or less at the same depth
(5–6 m). The third cavity (between the 70 and 80 m mark) is also
seen in the E280 bench face (Fig. 7B) and in the transverse profile
20. In this profile, we can see the same discontinuity dip as in the
transverse profiles recorded on the lower E280 bench.
84 M. Zajc et al. / International Journal of Rock Mechanics & Mining Sciences 67 (2014) 78–87

Fig. 8. Interpretation of two longitudinal GPR profiles (3 and 14) and four transverse GPR profiles (6, 7, 10 and 20), showing locations of karst cavities (arrows), air reflections
(square brackets) and discontinuities (yellow line). Some diffraction hyperbolas used for velocity determination are also shown (green line). (For interpretation of the
references to color in this figure legend, the reader is referred to the web version of this article.)

We correlated the GPR results with the results of structural and
lithological mapping by constructing four GGPR cross sections
(Fig. 9). The profile GGPR1 in Fig. 9 shows the correlation between
these cavities and the system of discontinuities. The cavities on the
E280 bench and the cavities on the E300 bench are connected with
the same discontinuity system that dips towards the southwest,
parallel to the lithological unit dip and perpendicular to the layers
of the exploitation edge on the benches. The detail of the profile
GGPR2 (Fig. 9) shows the position of discontinuities obtained from
the GPR results (blue lines) which correlate well with the mapped
 M. Zajc et al. / International Journal of Rock Mechanics & Mining Sciences 67 (2014) 78–87 85

Fig. 9. Geological profiles based on GPR and other data (GGPR profiles) with enlarged details. Locations of profiles are shown on left and in Fig. 5. (For interpretation of the
references to color in this figure legend, the reader is referred to the web version of this article.)

structural discontinuities (red lines). Note that the GPR profiles are
perpendicular to the GGPR2 profile and the blue lines dip in
apparent angles. The detail of the profile GGPR3 (Fig. 9) shows that
the karst cavity seen in the GPR 3 and 10 profiles (Fig. 8) is also
located along the discontinuity system. The GGPR4 profile runs
higher and parallel to the profile GGPR2. Based on the GPR profile
20 (Fig. 8), the cavity seen in the detail of GGPR4 is very close
to the surface of the E300 bench. Nearby, a cavity with a collap-
sed roof can be seen on the surface (image to the right of the
GGPR4 detail in Fig. 9). Based on its location and the GPR results, it
formed along the same discontinuity as the cavity seen in the GPR
profile 20.
5. Discussion
A similar contact karst phenomenon can also be seen in
the nearby Natisone Valley and Bernadia mountains of the Friuli
region (NE Italy) [28]. The karst structures in the flysch and
flyschoid rock formations develop mostly at the contact of carbo-
nate breccia layers (e.g. Smoganica cave [29]) or calcarenites with
marls in the footwall [30]. In our study, we thus assumed that the
karst features identified by the GPR method are located in the
phreatic zone between two aquifers. The upper aquifer consists of
coarse grained carbonate sandstone (cGCS), while the underlying
aquifer is made of marly calcareous breccia (MCB). The share of
non-carbonate minerals in the cGCS unit amounts to 15 wt%
(Fig. 6). The dip of this lithological unit ranges between 20 and
351. This coincides with the results of [28], where the dip of
lithological units with karst features also does not exceed 401.
The results of geological mapping of structural discontinuities
correlate well with the results of the GPR method. From the
stereographic projection of discontinuity poles (Fig. 10), we can
conclude that most karst features developed along the J1 discon-
tinuity family as a result of water flow directed through the J2
system. The diagrams in Fig. 10 show a good correlation between
86 M. Zajc et al. / International Journal of Rock Mechanics & Mining Sciences 67 (2014) 78–87
Fig. 10. Left: Intersection of discontinuities (J1 and JGPR—reconstructed from GPR profiles) with slope (SL); Right: Diagram of poles for main structures (J1, J2, Jb and JGPR).
the mapped J1 family and the discontinuities derived from the GPR
results. Based on the dispersion of the Jb and J2 poles, the
relationship between these two families can be defined as pseudo
orthogonal. This confirms that the orientation and density of such
structures influences the triggering process of a karst system
evolution, even in the case of carbonate lithology within a
complex turbidite body [31]. Considering the Hocking's criteria
[32], the J1 and J2 discontinuity families can lead to single plane
sliding, while the pseudo orthogonal J1 and Jb families present
potential areas of collapse due to the development of karst
features at their intersections (Figs. 2 and 3).
The correlation between the discontinuity data derived from
the GPR results and from the mapping of the J1 system represents
a good verification of the GPR method. Moreover, we were able to
verify the GPR results at the locations where two of the cavities
detected with GPR were visible on the bench faces (the cavity
at the intersection of profiles 3 and 10, and the cavity at the
intersection of profiles 14 and 20, Fig. 8) and during excavation of
two of the benches.
6. Conclusions
The different discontinuity families specified in the Rodež
quarry are mainly the result of the regional strike-slip tectonics
[8] and stratigraphy of lithological units [1,10]. The complexity of
the intersections of these structural discontinuities and the
mechanical properties of rocks contribute to the formation of
sliding surfaces and the developing of karst features.
The low frequency GPR proved to be a successful method for
detecting karst cavities and discontinuities in the flyschoid rocks
of the Rodež quarry. Since we used an unshielded antenna, all the
GPR profiles contain a certain amount of air reflections from the
quarry walls and scattered signals from the bench edges. This
makes the interpretation of the profiles measured close to the
bench edges or quarry walls more difficult. The best results were
acquired from the longitudinal profiles recorded in the middle of
the benches. In future surveys, profile positioning should hence be
taken into consideration. On the other hand, it was proved that the
RTA antennas used have several advantages compared to classical
rigid antennas when working in rough terrain.
By interpreting the GPR profiles, we not only located several
karst cavities but also confirmed the presence of discontinuities.
However, due to the limited vertical resolution of the 50 MHz
frequency antenna, it is impossible to conclude whether the
reflections on the recorded GPR profiles are the result of one or
more parallel discontinuities. Therefore, the GPR results were
correlated with the results of structural and lithological mapping
in the GGPR profiles. We found that the discontinuities defined by
GPR coincide with the location and dip of the mapped disconti-
nuities, and that the karst cavities seen in the GPR profiles are
located along these discontinuities.
On the basis of this study, we can conclude that the low
frequency GPR is a very efficient technique for detecting karst
cavities and discontinuities in flyschoid rocks. These features can
pose serious hazards in the quarry. Since they are most dangerous
when they occur in shallow depths, higher frequency antennas
could also be used to detect them and could provide an even more
precise image of the subsurface. A detailed GPR survey conducted
prior to exploitation could therefore diminish the possibility of
accidents that can happen as a result of sinking surfaces and land
sliding.
Acknowledgments
This study was conducted with the support of the research
programme P1-0011 and the Ph.D. grant 1000-10-310074 financed
by the Slovenian Research Agency.
References
[1] Skaberne D. Evolution of the karstic carbonate platform—excursions guidebook.
Trieste: Universita degli studi di Trieste, Instituto di geologia e paleontologia;
1987; 37.
[2] Panek T, et al. Gravitationally induced cavities and other discontinuities
detected by 2D electrical resistivity tomography: case studies from the Polish
Flysch Carpathians. Geomorphology 2010;123:165–80.
[3] Gosar A. Analysis of the capabilities of low frequency ground penetrating radar
for cavities detection in rough terrain conditions: the case of Divača cave,
Slovenia. Acta Carsol 2012;4:77–88.
[4] Guttierrez F, et al. Integrating geomorphological mapping, trenching, InSAR
and GPR for the identification and characterization of sinkholes: a review and
application in the mantled evaporite karst of the Ebro Valley (NE Spain).
Geomorphology 2011;134:144–56.
[5] Placer L, Vrabec M, Celarc B. The bases for understanding of the NW Dinarides
and Istria Peninsula tectonics. Geologija 2010;53:55–86.
[6] Alberti A, Lenaz D, Princivalle F, Tunis G. Mineralogical characterization of
flysch sequences of the SE Alps and Outer Dinarides (NE Italy, Slovenia,
Croatia). Geol Carpath 1999;50:9–10.
[7] Lenaz D, Princivalle F. The crystal chemistry of detrital cromian spinel from the
Southeastern Alps and outer Dinarides: the discrimination of supplies from
areas of similar tectonic setting? Can Mineral 2005;43:1305–14.
[8] Buser S. Basic geologic map of Yugoslavia. 1:100.000, sheet Tolmin and Videm,
Beograd: Federal Geological Survey; 1987.
[9] Kokošin J. Interpretation of discontinuities in the Rodež quarry. 2nd seminar
paper. Ljubljana: University of Ljubljana, Faculty of Natural Sciences and
Engineering; 2009; 21.
 M. Zajc et al. / International Journal of Rock Mechanics & Mining Sciences 67 (2014) 78–87
[10] Pogačnik Ž, Pavšič J, Meden A. The geological record as an indicator of the
mudstones thermal characteristics in the temperature range of decarbonatisa-
tion. Mater Technol 2009;43:157–63.
[11] Blindow N, Eisenburger D, Illich B, Petzold H, Richer T. Ground penetrating
radar. In: Knödel K, Lange G, Voigt HJ, editors. Environmental geology,
handbook of field methods and case studies. Berlin, Heidelberg, New York:
Springer; 2007. p. 283–335.
[12] Young RA, Slatt RM, Staggs JG. Application of ground penetrating radar
imaging to deepwater (turbidite) outcrops. Mar Petrol Geol 2003;20:809–21.
[13] Jol HM. Ground penetrating radar: theory and applications. Amsterdam,
Oxford: Elsevier Science; 2009.
[14] Davies JD, Annan AP. Ground-penetrating radar for high-resolution mapping
of soil and rock stratigraphy. Geophys Prospect 1989;37:531–51.
[15] Neal A. Ground-penetrating radar and its use in sedimentology: principles,
problems and progress. Earth Sci Rev 2004;66:261–330.
[16] Milsom J. Ground penetrating radar. Field geophysics. 3rd ed.. West Sussex:
John Wiley and Sons Ltd.; 2003; 167–78.
[17] Annan AP. GPR-history, trends and future developments. Subsurf Sens Technol
Appl 2002;3:253–70.
[18] Daniels DJ. Surface penetrating radar. London: Institute of Electrical and
Electronic Engineers; 1996.
[19] Pavlič MU, Praznik B. Detecting karstic zones during highway construction
using ground-penetrating radar. Acta Geotech Slov 2011;8:17–27.
[20] Puelo-Anchuela O, Pocovi-Juan A, Soriano MA, Casas-Sainz AM. Characteriza-
tion of karst hazards from the perspective of the doline triangle using GPR—
examples from Central Ebro Basin (Spain). Eng Geol 2009;108:225–36.
[21] Al-Fares W, Bakalowicz M, Guerin R, Dukhan M. Analysis of the karst aquifer
structure of the Lamalou area (Herault, France) with ground penetrating radar.
J Appl Geophys 2002;51:97–106.
87
[22] Spillmann T, Maurer H, Willenberg H, Evans KF, Heincke B, Green AG.
Characterization of an unstable rock mass based on borehole logs and diverse
borehole radar data. J Appl Geophys 2007;61:16–38.
[23] Chalikakis K, Plagnes V, Guerin R, Valois R, Bosch FP. Contribution of
geophysical methods to karst-system exploration: an overview. Hydrogeol J
2011;19:1169–80.
[24] Mala. ProEx—professional explorer control unit. Operating manual, Mala;
2009. p. 60.
[25] DECO Geophysical Co. Ltd. RadExplorer—the software for GPR data processing
and interpretation. User manual, Moscow 2005; p. 92.
[26] Cooke ML, Simo JA, Underwood CA, Rijken P. Mechanical stratigraphic controls
on fracture patterns within carbonates and implications for groundwater flow.
Sediment Geol 2006;184:225–39.
[27] Eaton TT, Anderson MP, Bradbury KR. Fracture control of ground water flow
and water chemistry in a rock aquitard. Ground Water 2007;45:601–15.
[28] Mochiutti A. Contact caves in flysch formations – Friuly region – Northeast
Italy. Acta Carsol 2001;30:157–64.
[29] Knez M, Slabe T, Šebela S. Smoganica—a cave developed in upper cretaceous
breccia. Acta Carsol 2005;34:425–38.
[30] Mochiutti A, Maddalena P. Chemical, geomechanical and geomorphological
aspects of Karst in sandstone and marl of flysch formations in north east Italy.
Acta Carsol 2005;34:349–68.
[31] Di Naccio D, et al. Role of mechanical stratigraphy on fracture development in
carbonate reservoirs: insight from outcropping shallow water carbonates in
the Umbria-Marche Apennines, Italy. J Volcanol Geoth Res 2005;148:98–115.
[32] Yoon WS, Jeong UJ, Kim JH. Kinematic analysis for sliding failure of multifaced
rock slopes. Eng Geol 2002;67:51–61.