ArticlePDF Available

An integrated geophysical and geological interpretation of the area around the Unicorn Cave (Southern Harz Mountains, Germany)


Abstract and Figures

The Unicorn Cave in the southern Harz Mountains in Germany is a cave developed in dolomitic host rocks of the Permian Zechstein period. The cave is located on a plateau, with a shallow overburden of 10-15 m. The cave is essentially a north-south trending sequence of shallow passages, interrupted by large rooms. The cave terminates in the south in a breakdown room, providing natural access through two daylight openings. In the north, shallow crawl ways drowned in sediment, terminate the cave. We have mapped the known and accessible cave with LIDAR scanning, providing a detailed model of the cave’s interior, and unravelling the structural control on the cave passages. With geophysical surveys (gravity, electric resistivity imaging, ground-penetrating radar, shear wave reflection seismic) above and in the cave, we have recovered samples of the substantial sediment infill of the cave, mapping material properties such as density, electrical resistivity, electrical permittivity, and shear modulus. Based on the indirect geophysical signals obtained above the cave, we have mapped the terrain south and north of the known cave with the same geophysical techniques. We identified the southern extension of the Unicorn Cave by geophysical means, which we then proved by drilling into the voids. The boreholes pass an air-filled passage with large sediment infill.
Content may be subject to copyright.
*Addresses of the authors:
1Landesamt für Bergbau, Energie und Geologie – Geologischer Dienst von Niedersachsen, Stilleweg 2, 30655 Hannover, Germany
2Gesellschaft Unicornu fossile e. V., Im Strange 12, 37520 Osterode am Harz, Germany (
3FU Berlin, Institut für Geologische Wissenschaften, Fachrichtung Geophysik, Malteserstraße 74–100, Haus D, 12249 Berlin,
Germany (
4Leibniz-Institut für Angewandte Geophysik (LIAG), Stilleweg 2, 30655 Hannover, Germany ( / / /
5UNESCO-Geopark Harz, Braunschweiger Land, Ostfalen, Niedernhof 6, 38154 Königslutter, Germany (h.zellmer@geopark-hblo.
An integrated geophysical and geological interpretation of the area around
the Unicorn Cave (Southern Harz Mountains, Germany)
Heinz-Gerd Röhling1,2, Ralf Nielbock2, Georg Kaufmann3, 2, David Colin Tanner4, Jan Igel4,
Ulrich Polom4, Henning Zellmer5, 2 & Detlef Vogel4*
Röhling, H.-G., Nielbock, R., Kaufmann, G., Tanner, D.C., Igel, J., Polom, U., Zellmer, H. & Vogel, D. (2019): An integra-
ted geophysical and geological interpretation of the area around the Unicorn Cave (Southern Harz Mountains, Germany).
– Z. Dt. Ges. Geowiss.
Abstract: The Unicorn Cave, in the southern Harz Mountains in Germany, is a natural cave that developed in dolomitic host
rocks of the Permian Zechstein period. The cave is located on a plateau, with a shallow overburden of 10–15 m. The cave is
essentially a north-south trending sequence of shallow passages, interrupted by large rooms. The cave terminates to the south
in a collapsed room, providing natural access through two daylight openings. To the north, shallow tunnels that are drowned
in sediment, terminate the cave.
We have mapped the known and accessible cave with a LIDAR scan, which provides a detailed model of the cave’s interior,
and allows the unravelling of the structural control on the development of the cave. With geophysical surveys (gravity, elec-
tric resistivity imaging, ground-penetrating radar, shear-wave reflection seismics) above and in the cave, we investigated the
substantial sediment infill of the cave, mapping material properties, such as density, electrical resistivity, electrical permit-
tivity, and shear modulus.
Based on the indirect geophysical signals obtained above the cave, we have mapped the terrain south and north of the known
cave with the same geophysical techniques. We have identified the hitherto-unknown southern extension of the Unicorn
Cave using geophysical methods, which we then proved by drilling. The boreholes pass through an air-filled passage with
large sediment infill.
Kurzfassung: Die im Südharz gelegene Einhornhöhle ist ein natürlicher Hohlraum im Dolomit-Gestein des Zechsteins des
Oberen Perms. Die auf einem Plateau gelegene Höhle besitzt eine Überdeckung von etwa 10–15 m. Das generell Nord–Süd
gerichtete Höhlensystem besteht aus einer Aneinanderreihung von flachen Passagen, die von großen Hallen und Domen
unterbrochen werden. Die Höhle endet in einem Höhlenraum mit zwei Deckeneinstürzen, den einzigen heute noch vorhan-
denen natürlichen Eingängen zur Höhle. Im Norden endet die Höhle an mit Sediment verfüllten Gängen.
Der bekannte und zugängliche Teil der Höhle wurde mit LIDAR vermessen, um ein detailliertes Modell des Hohlraums
zu erhalten und die strukturelle Ursache der Entwicklung der Höhle aufzudecken. Mittels geophysikalischer Erkundungsme-
thoden (Gravimetrie, Geoelektrik, Georadar, Eigenpotenzial, Scherwellenseismik), die sowohl auf der Erdoberfläche als
auch innerhalb der Höhle durchgeführt wurden, wurde die mächtige Sedimentverfüllung der Höhle untersucht und Materi-
aleigenschaften wie Dichte, spezifischer elektrischer Widerstand, die elektrische Permittivität und der Schermodul abgebildet.
Von der Erdoberfläche aus wurde das Gelände südlich und nördlich der bekannten Höhle ebenfalls geophysikalisch un-
tersucht. Dabei wurde eine südliche, bisher unbekannte Fortsetzung der Höhle lokalisiert, die anschließend durch Boh-
rungen aufgeschlossen wurde. Wie die schon bekannten Teile der Einhornhöhle ist der neue Hohlraum lufterfüllt, die Basis
bildet eine mächtige Sedimentfüllung.
Keywords: Unicorn Cave, geophysical survey, geological survey, borehole, karst
Schlüsselwörter: Einhornhöhle, geophysikalische Untersuchungen, Geologie, Bohrung, Karst
Z. Dt. Ges. Geowiss. (German J. Geol.) Open Access Article
Published online October 2019
DOI: 10.1127/zdgg/2019/0195
© 2019 The authors
E. Schweizerbartsche Verlagsbuchhandlung, Stuttgart, Germany,
2 Heinz-Gerd Röhling et al.
1. Introduction
In the southern Harz Mountains, caves are mainly found in
the Upper Permian Zechstein sediments (Table 1), which
crop out to the west and south of the Palaeozoic Harz rocks.
The Zechstein rocks consist of soluble rocks such as lime-
stone, dolomite, anhydrite, gypsum or salt. These rocks can
be dissolved physically by water and for limestone and dolo-
mite additionally by water enriched with carbon dioxide.
Fig. 1 shows the distribution of Zechstein rocks along the
western and southwestern margin of the Harz Mountains
built up of Palaeozoic rocks.
Often, the caves along the southern Harz are located in
the shallow subsurface, thus a variety of geophysical meth-
ods can be used to identify cave voids: Gravity maps the
density difference between host rock and air- or sediment-
filled cave; electrical resistivity imaging maps the electrical
resistivity contrasts between air-filled cave (very high resis-
tivities), host rock (high resistivities) and sediment (low re-
sistivities), which is mainly controlled by fluid circulation;
ground-penetrating radar identifies structural contrasts by
changes in both the electrical resistivity and the electric per-
mittivity; seismics also targets structural differences (e.g.
cave roofs) by identifying changes in the mechanical proper-
ties (bulk- und shear modulus, density) by mapped seismic
velocities. Additionally, geodetic methods such as LIDAR
scans provide an accurate map of the cave interior, from
which structural information such as the faults controlling
the cave evolution can be derived.
1.1 The Unicorn Cave
The Unicorn Cave has been known as a natural and cultural
monument for a long time and it is the only natural cave of
the southern Harz that is accessible to tourists (Nielbock
1990, 2002a, 2008, 2010; Röhling & Nielbock 2005). The
cave, first mentioned in a historic document in 1541 (de-
scription of local border lineament in the area), is the largest
show cave in the western Harz Mountains, located about
1.5 km northwest of the village of Scharzfeld, which is an
urban district of the city of Herzberg am Harz.
The cave developed in dolomitic rocks of the Zechstein
on the Brandköpfe Plateau (Fig. 2). The known cave is lo-
cated under the central part of the ridge. Looking from out-
side at the cave area, old entrances and cave portals cannot
be seen due to debris on the slopes. Recent research has indi-
cated that the cave is much larger than the part that can be
walked in today. It is really only the top portion of a large old
cave, which was gradually filled by sediments over hundreds
of thousands of years, as shown by drillings and geophysical
surveys. How the cave continues underground is largely un-
known. In addition, the exact dimensions of the continuation
of the cave to the north are not yet known. At the foot of the
Rottsteinklippen (not visible on Fig. 2), to the north, the old
entrance portal to the cave awaits discovery.
Clearly recognisable is the bare rock cover in the cave,
and the floor dropping from the Blue Grotto (“Blaue Grotte”)
towards the north. Depending on the distribution, density
and dip direction of fractures in the rock, high halls (with
strong ceiling corrosion) and low passageways form the var-
ious parts of the cave; the orientation of the main axis of the
cave corresponds to the main strike of the fracture system.
The Unicorn Cave is famous for the amount of fossil
bone fragments that have been brought to light. This has at-
tracted bone collectors, who dug for centuries for these
bones, especially for the coveted “unicorn” (Fig. 3 left).
Gott fried Wilhelm Leibniz, who visited the cave in 1686, re-
ported the trade with unicorn artefacts was taking place
(Leibniz 2014), which were used at that time for making
medicine. Based on Otto von Guericke’s discovery of some
ancient animal bones on the Zeunickenberg (Nielbock 2010),
a mountain located north of the Harz near Quedlinburg,
Leibniz misinterpreted the bones found in the Unicorn Cave
as relicts of the legendary “unicorn” and published a drawing
in his book “Protogaea” (Fig. 3 left; Leibniz 2014). But by
the 17th century, the fossil bones had also been correctly iden-
tified as fossil remains of large mammals, especially the cave
bear (Ursus spelaeus; see Rosendahl et al. 2005; Nielbock
2008, 2010; Röhling & Nielbock 2013; Nielbock et al. 2013),
and not the legendary unicorn. Nevertheless, the mythical
animal became eponymous for the cave. In 1872, Rudolf
Virchow carried out excavations and found the bones of ani-
mals like lions or cave bears (Nielbock 2002b). Further in-
vestigations (Nielbock 1987, 1989), showed that the Unicorn
Cave was not only inhabited by cave bears, but also by other
animals such as cave lions, hyenas or wolves. The cave envi-
ronment is very favourable for conservation of the entire
bone structure of the dead animals. So far, almost 70 verte-
Table 1: Stratigraphic subdivision of the Zechstein outcrop around
the Harz Mountains (modified after Paul in press). The Unicorn
Cave is located on the western flank of the Eichsfeld-Altmark
Swell, which was active during Upper Permian i.e. Zechstein times.
The table shows also the thickness variations of the Zechstein sub-
units on the Eichsfeld-Altmark Swell in comparison with the Sub-
hercynian Basin.
Formation Subformation Basin (m) Swell (m)
Fulda Fm. 40
Friesland Fm. 0–7
Ohre Fm 0–5
Aller Fm. 20
Leine Fm. Leine-Salz 100–150 0
Leine-Sulfat 30–50 30–50
Leine-Karbonat 0 1–40
Leine-Tonstein 5 10
Stassfurt Fm. Staßfurt-Salz 0–500 0
Staßfurt-Sulfat 1–5 0–70
Staßfurt-Karbonat 1–3 –100
Werra Fm. Werra-Salz 0–8 0
Werra-Sulfat 40–50 0–350
Werra-Karbonat 1–10 0–120
Kupferschiefer 0.3 0–2
An integrated geophysical and geological interpretation of the area around the Unicorn Cave
brate species have been identified (Nielbock 2010; Nielbock
et al. 2013; Röhling & Nielbock 2013).
The speleological research around the penultimate turn
of the century had primarily one goal, to find the “diluvial
man” – i.e. the people of the glacial age. At the end of the 19th
century, the physician and anatomist Rudolf Virchow exca-
vated the Unicorn Cave – he wanted to prove the existence of
the “primeval man”. However, this evidence was not discov-
ered until 1985 with the discovery of lithic tools from the
Palaeolithic (Fig. 3 right; Nielbock 1989, 2002b, 2004; Veil
1989). A large backfilled cave portal, used by Neanderthals
over many decades, with a “stone workshop” was discovered
during an excavation campaign in 1985. This site has not yet
been excavated. The other archaeological finds prove, espe-
cially for the Blue Grotto (Fig. 2) and thus for the Unicorn
Cave, continuous human use from the time of Neanderthal
man to modern day. Based on archaeological finds of stone
tools from the time of early man to a Coca Cola can from the
20th century, the Unicorn Cave is one of the most important
prehistoric cultural monuments in Central Europe – it is not
only a geotope and biotope, but also a significant under-
ground archaeotope (Nielbock 2004).
The Unicorn Cave has been open to visitors since 1885,
starting with guided tours from the southern end via the natu-
ral entrance of the Blue Grotto, and from 1905 onwards, af-
ter construction of a new access tunnel, from the northern
side. Since then, 270 m of the total cave length of more than
650 m has been open as a show cave. In order to present the
new scientific findings to a broader public, the Unicorn Cave
has been run as natural show cave by the new operator as-
sociation “Gesellschaft Unicornu fossile e. V.” since 2002
(Nielbock et al. 2004; Röhling & Nielbock 2005). The acqui-
sition of EU funding, through the Lower Saxony Ministry of
Economic Affairs as part of the Geopark project, made it
possible to set up a state-of-the-art “Haus Einhorn” infra-
structure with an UNESCO Global Geopark (certificated in
2017) and cave information centre, cafe and a small cave
Nowadays, the Unicorn Cave is a major geological at-
traction in the southern Harz Mountains, visited by nearly
30,000 visitors in 2018. Since 2006, the “bio-, geo- and ar-
chaeosite” Unicorn Cave (Röhling & Nielbock 2013; Niel-
bock et al. 2013) has been awarded the “National Geosite”
certificate (Nielbock et al. 2006). Today, the cave is an extra-
curricular place of learning and one of the information cen-
tres of the National and UNESCO Global Geopark Harz .
Braunschweiger Land . Ostfalen (Brauckmann et al. 2009,
2013). Apart from the general fascination with the cave,
there is also a great scientific interest. The Unicorn Cave to-
day is still subject of intensive geological, geophysical, pal-
aeontological, biological, archaeological and cave research
(e.g. Reinboth 1996; Nielbock 2002b; Vladi 2004a, b;
Kempe 2005; Meischner 2006a, b; Rackow 2004, 2014;
Meischner 2011; Hillgruber et al. 2014; Kaufmann et al.
2011, 2014, 2015; Röhling & Nielbock 2013; Nielbock et al.
2013; Kaufmann & Romanov 2017).
Fig. 1: Distribution of soluble Zechstein rocks (carbonates, dolomites, anhydrites, gypsum, grey-striped areas) along the western and south-
western part of the Harz Mountains. Also shown are cities (white circles), localities (white triangles), and caves (blue triangles).
4 Heinz-Gerd Röhling et al.
Fig. 2: The Unicorn Cave in the dolomitic rocks of the Brandköpfe Plateau. Top : Vertical profile with names of the cave parts. Middle:
Projection of the cave profile into the dolomitic rocks; image width: approx. 350 m (from Nielbock 2010; Tanner et al. 2012). Bottom: Blue
Grotto (“Blaue Grotte”), the natural entrance to the Unicorn Cave.
1.2 Motivation
As already mentioned above, at the end of the 19th century
intensive scientific investigations began in the Unicorn Cave
(Nielbock 2002a, b). Despite the many subsequent excava-
tions, however, until a few years ago, little was known about
the diversity and the actual age of fossil wildlife, the inspec-
tion of Neanderthals as well as the actual size of the unicorn
cave. New knowledge about the glacial inhabitants and the
dimensions of the cave began in 1985 with an interdiscipli-
nary excavation and research campaign of the Geological
Institute of the TU Clausthal, guided by the Gesellschaft
Unicornu fossile e. V. (GUF e. V.). In recent years, the GUF
e. V. initiated several investigations in and above the cave.
One of the aims of these activities was, among other geo-
logical, palaeontological and archaeological investigations,
using geophysical methods (e.g. gravity, electrical resistivity
tomography, ground-penetrating radar, shear wave reflection
seismics, self-potential) to find new cave chambers beyond
the previously known and accessible cave system.
An integrated geophysical and geological interpretation of the area around the Unicorn Cave
Fig. 3: Left: Drawing of the reconstruction of a vertebrate skeleton after a written submission by Otto von Guericke, interpreted by Gott-
fried Wilhelm Leibniz as “Unicornu fossile”, the excavated unicorn. In Protogaea, 1749, it appeared posthumously; original: Leibniz Li-
brary Hannover. Right: Scraper from the Middle Palaeolithic (length: ca. 5 cm), found in the Jacob Friesen Tunnel (“Jacob-Friesen-
Gang”), Unicorn Cave.
Fig. 4: Left: View of the ceiling of the Schiller Hall, with deep corrosion holes. Image width approx. 10 m. Right: Condensation corrosion
in the passage Hubertusgang.
6 Heinz-Gerd Röhling et al.
2. Geological situation
The Unicorn Cave was formed by karstification of the Upper
Permian Zechstein sediments. The Zechstein period in Eu-
rope was a time of large evaporitic deposits (Richter-Bern-
burg 1955a, b; Paul et al. 2018). During the Zechstein period,
several sequences of carbonates, sulphates, rock salts and
potassium and magnesium salts were deposited because of
seawater ingression into the Central European Basin from
the North German Basin. Each ingression was followed by a
subsequent evaporation phase. The evaporation left seven
Zechstein formations (z1 to z7, i.e. Werra, Staßfurt, Leine,
Aller, Ohre, Friesland und Fulda formations; Table 1; Rich-
ter-Bernburg 1955a, b; Paul et al. 2018).
In the southern Harz region, which was during the Per-
mian Zechstein period located close to the equator and part
of the southern margin of the marine North German Basin,
mainly shallow-water carbonates and sulphates were depo-
sited, especially during the formation of the Werra (z1),
Staßfurt (z2) and the Leine Formation (z3). However, the
region of the Unicorn Cave was part of the Eichsfeld-Alt-
mark Swell (EAS), active both during the Upper Permian
Zechstein and Triassic times (Paul 1993; Röhling 2013). The
EAS contains several smaller depressions and topographic
highs. In its area, the Zechstein Sea transgressed over an an-
cient land surface, which was not a peneplain, but rather had
a strong relief (Paul 1993; Paul & Vladi 2001).
On top of the EAS, the basal unit of the lowermost Zech-
stein (Werra) Formation, the Kupferschiefer (“Copper
Shale”) is missing (Jordan 1979; Paul 1982, 2006; Paul et al.
2018). The Copper Shale was deposited only in valleys of
the pre-Zechstein relief (Paul 1982, 2006), where the shales
reach thicknesses up to 2 m and show diverse facies and
thickness developments (Paul et al. 2018). The latter also ap-
plies to the subsequently deposited carbonates and sulphates,
which show in general higher thicknesses on the swell,
whereas the salts are only very thin. In the area of the south-
ern margin of the present Harz Mountains, where the Zech-
stein layers are now close to the surface, the salts were
leached and one usually only finds their residual sediments
(Herrmann 1956; Paul 1993; Paul & Vladi 2001; Paul et al.
2018). This is also evident in the area of the Unicorn Cave,
e.g. the Brandköpfe Plateau, where carbonates are deposited
directly on Palaeozoic greywackes.
The small-scale multiple alternations of slightly soluble
sulphate/carbonate and impermeable claystone horizons re-
sulted in complicated karst systems in the Zechstein sedi-
ments around the Harz Mountains. In this context, the com-
plex karst system of the Pöhlde Basin with the Rhume
Spring, one of the strongest karst springs in Central Europe,
is hydrogeologically very significant for the region. There
are numerous caves known in both the carbonates/dolomites
and anhydrites: the most well-known being the Barbarossa
Cave on the Kyffhäuser hills and the Unicorn Cave near
The Unicorn Cave was formed by natural solution pro-
cesses in dolomitic rocks of the Werra Carbonate (Werra
Formation), which developed thicknesses up to 80 m on top
of the EAS. These dolomites directly overlay folded
greywackes of the Upper Devonian/Lower Carboniferous,
which can be observed along several up to 20 m high cliffs,
such as the Kaiserklippen to the east and the Rottsteinklip-
pen to the north of the known cave.
The beginning of the formation of the Unicorn Cave
dates back more than 5 million years, at the end of the Ter-
tiary. In humid and warm climates, rainwater enriched with
carbonic acid penetrated into the fissures of the dolomite
rock through the forest floor. The dolomite was dissolved
and a huge cavity system with characteristic ceiling corro-
sion holes was created (Fig. 4). Large halls are connected by
shallower passages. In the southwest of the cave, in the so-
called Blue Grotto, there are two ceiling openings. These are
the only natural entrances to the cave that still exist today. In
a side corridor of the White Hall in the northeastern area of
the cave, a tunnel was drilled through the rock massif in
1905; it has been used as the main access to the cave since
During the Pleistocene, a thick sequence (10–35 m) of
layers of clay, dolomitic sand and river gravels gradually ac-
cumulated within the entire cave. The Unicorn Cave, as we
know it today, is the top of a large cave that has been mostly
filled with natural sediments. Boreholes in and above the
cave reveal, that the sediments below the present-day cave
floor are more than 30 m thick. In this large sediment de-
posit, approximately 250,000 m3 of sediment, the remains of
all glacial phases have been preserved. This valuable geo-
archive is just underneath the feet of the show-cave visitors
(Nielbock 2004). The age of the cave sediments is not known
yet. The lateral extension, natural entrances and outlets of
the fluvial system are also not well known.
Excavations in the upper layers of the cavern fill exhibit
sequences of Thanatocenoses (including among others cave
lion, cave bear or wolves), that reach back to the last inter-
glacial, the Eemian (Paul & Vladi 2001). Artefacts that were
found in the cave in the area of a buried cave entrance also
point to this age (Nielbock 2002b; Paul & Vladi 2001). The
artefacts are dated as Middle Palaeolithic Levellois (Niel-
bock 2002b).
3. Data analysis
In this section, we discuss the results obtained from the
measuring campaigns carried out during the last decades. In
each section, we briefly describe the methods involved and
their relation to identify the karst voids of the Unicorn Cave
and its surroundings, and then we emphasise the key results
of each method.
3.1 Geophysical results
In karst rocks, both subsurface features, such as fissures and
bedding surfaces enlarged by dissolution and possibly en-
larged to cave voids, and surface features, such as solution and
collapse sinkholes, can often be identified by differences in
An integrated geophysical and geological interpretation of the area around the Unicorn Cave
material properties between the karst structure (e.g. sinkholes
and caves) and its surrounding soluble rock (e.g. Butler 1984;
Al-fares et al. 2002; El-Qady et al. 2005; Dobecki & Upchurch
2006; Nyquist et al. 2007; Mochales et al. 2008; Kaufmann et
al. 2012, 2015). Geophysical surveys are useful tools to iden-
tify and map these material differences, and as a combination
of different geophysical methods they are able to identify dif-
ferent material parameters.
Cave voids, for example, can be either air- or water-filled
or filled with sediments, thus the density ρ [kg/m3] of the
infill (0/1000/1500–2000 kg/m3 for air, water and sediments,
respectively) and the surrounding soluble rock (2200/
2600/2800/2900 kg/m3 for gypsum, limestone, dolomite, an-
hydrite, respectively) differ. These density differences can be
mapped with a gravity survey (GRAV).
Fissures and bedding surfaces enlarged by dissolution
provide preferential pathways through the soluble rock. As
the electrical resistivity ρ [Ωm] of the subsurface is mainly
controlled by these circulating fluids, differences in electri-
cal resistivity arise between the rock matrix (high resistivi-
ties, 2000–6000 Ωm) and the fluid infill (10–100 Ωm). In
addition, sediments in the cave voids can be either dry or
saturated with water, resulting in either high or low electrical
resistivities. These differences in electrical resistivity can be
mapped with electrical resistivity imaging (ERT).
If water flows through karst rocks through preferential
flow paths (fissures and/or bedding surfaces), excess electri-
cal charge will be carried in the water, which in turn will re-
sult in a difference in electrical potential U [V]. This poten-
tial differences can be mapped with the self-potential
method (SP), indicating fluid flow.
Structures present in the soluble rock, either through lay-
ering during deposition, enlarged fractures or air- and water
filled cave voids, can be identified through changes of mate-
rial properties along these structural interfaces. In the case of
changes in the electrical conductivity σ = 1/ρ [S/m] and/or
the electrical permittivity ε [As/(Vm)], these structural
changes can be mapped with the ground-penetrating radar
(GPR). This method uses an electromagnetic wave pulse of
defined frequency, and records the reflections of this wave
pulse from boundaries at which either σ or ε changes, or
If the changes across a structural boundary are a result of
changes in the elastic material properties bulk modulus k
[Pa], shear modulus μ [Pa], or density ρ [kg/m3], then the
seismic method (SEIS) can be used to map the structural dif-
ferences. The elastic properties control the speed of elastic
waves, their kinetic energy during propagation, and their
propagation behaviour. Body waves (P-waves and S-waves)
and surface waves (Rayleigh waves and Love waves) are
used by active and passive methods, depending on the target,
depth and the resolution required. For cave detection, the
wavelength λ [m] controls the detection capabilities of the
methods, which is a result of the frequency content and the
propagation speed of the seismic signal. Since only struc-
tures down to a quarter of the wavelength can be detected,
low wave speed and high frequencies are desired for cave
detection. On the other hand, low wave speed and high fre-
quency content both restrict the penetration depth due to sig-
nal absorption and damping. P-waves are always faster (fac-
tor 1.7–>10) than S-waves and propagate a higher frequency
content (factor 2–4), therefore they commonly are used for
deep and expanded targets e.g. in the hydrocarbon explora-
tion industry.
For shallow applications, S-waves and surface waves,
which propagate with nearly the same speed, are advanta-
geous due to the smaller wavelengths. S-waves are not af-
fected by pore fluids, so their propagation behaviour in case
of a water- or air-filled cave is the same; they propagate
around the cave or reflect back the whole energy. Since the
propagation of surface waves is restricted to strong boundary
contrasts (e.g. the free surface), their depth penetration capa-
bility is typically restricted to one wavelength beside a
boundary contrast. Therefore, body S-waves enable the best
compromise of depth penetration and structure detection ca-
pabilities, which is used e.g. by the shear wave reflection
We have applied the above geophysical methods to iden-
tify karst structures in and around the Unicorn Cave, as we
will detail below.
3.1.1 Gravity
Gravity surveys were carried out using two Lacoste-Rom-
berg type D gravimeters. While the coordinates were ob-
tained from hand-held GPS units, elevation differences were
determined through levelling to a precision better than 3 cm.
From repeated readings of gravity at the base stations, we
estimate the uncertainty in the Bouguer gravity anomaly bet-
ter than 0.03 mGal. The raw survey data were processed with
the GRAViMAG software (developed at the FU Berlin), ac-
counting for (i) repeated measurements for instrument drift
correction, (ii) tidal corrections with the Eterna software
(Wenzel 1996), (iii) embedding relative gravity readings into
the official survey network via an absolute gravity point at
the nearby village of Scharzfeld, (iv) the derivation of the
Bouguer anomaly from latitude, free-air and Bouguer cor-
rections, including a local digital elevation model for terrain
Fig. 5 shows a map of all survey stations. The survey area
mainly covers the known parts of the Unicorn Cave (see out-
line), the part south of the collapse terminating the Blue
Grotto, and the northern end of the dolomite outcrop along
the Rottsteinklippen cliff.
The Bouguer anomaly (Fig. 6) is referenced to a typical
density of compact dolomite (ρref = 2800 kg/m3), thus
Bouguer values below zero map a mass deficit with respect
to the Werra dolomite. Focussing on the cave area first, we
can clearly identify the cave void with negative Bouguer
anomaly values between -0.2 and -0.5 mGal. This broad
minimum reflects the air-filled cave voids, but also (and to a
larger degree; see e.g. Kaufmann et al. 2011, 2012) the sedi-
ment infill of the passages (as we will discuss in the section
structural model). In some parts of the observed Bouguer
anomaly over the Unicorn Cave, a small-scale signal below
-0.7 mGal can be traced to known voids in the cave: Two
8 Heinz-Gerd Röhling et al.
Fig. 5: Map of gravity survey stations. Shown are the topography (grey shading), the roads (grey lines), the map of the known part of the
Unicorn Cave (black filled contour), locations of boreholes (white diamonds), the locations of outcrops (white triangles), and the gravity
survey stations (red dots).
Fig. 6: (a) Map of Bouguer gravity anomaly. Shown are the topography (grey shading), the Bouguer gravity (colour-coded), the map of the
known part of the Unicorn Cave (black contour), locations of boreholes (black diamonds), the locations of outcrops (white triangles), and
the gravity survey stations (red dots). (b) Section of Fig. 6a – the narrower cave area.
An integrated geophysical and geological interpretation of the area around the Unicorn Cave
mapped chimneys in the Leibniz Hall (G1), the collapse area
of the Blue Grotto (G2), a shaft in the southernmost Van-
Alten-Kapelle room (G3), and caves beyond the southern
limit of the mapped cave (G4, G5).
The latter two peaks and the broad negative Bouguer
anomaly south of the mapped cave (yellow-red colours) have
been interpreted as a southern extension of the cave (Kauf-
mann et al. 2011, 2012) and resulted finally in a drilling cam-
paign to verify and access to these cave voids.
In the northern part of the Unicorn Cave, close to the cliff
line, the Bouguer anomaly is characterised by broad negative
values, but without the typical negative peaks present in the
south. Only directly around the cliff line can a pronounced
minimum be found, which might reflect an old entrance to the
cave during its time as an active river cave. In the far west, at
the western end of the cliff line, the Bouguer gravity signal
becomes small and even zero, indicating non-karstified crys-
talline dolomite rock. Inspection of the cliff line reveals a
compact, crystalline dolomite without visible karst features.
As we find a similar Bouguer gravity signal with values
close to zero along the northern end of the known cave pas-
sages, we speculate that here the dolomite is very compact
and less karstified, probably explaining why the cave ends in
small tunnels here.
3.1.2 Electrical resistivity imaging (ERT)
Electrical resistivity measurements were carried out with a
Campus Tigre multi-electrode instrument and an array of
steel electrodes in different geoelectrical setup configura-
tions. For later surveys, we replaced the Campus Tigre with
a Geolog GeoTom MK8E1000. For both instruments, most
ERT profiles were measured using a Wenner and Dipole-
Dipole setup, the measurements based on the Geotom addi-
tionally with Schlumberger setups. The apparent resistivity
sections were inverted with the Res2DInv software (Loke &
Barker 1995, 1996), using the robust inversion method,
which we identified as most appropriate for the large resis-
tivity contrasts present in karst rocks over small distances.
Prior to every inversion, bad data points were identified with
the software and removed before the inversion. With coordi-
nates of the electrodes logged with a hand-held GPS, we ex-
Fig. 7: Map of ERT surveys. Shown are the topography (grey shading), the roads (grey lines), the map of the known part of the Unicorn
Cave (black contour), locations of boreholes (white diamonds), the locations of outcrops (white triangles), and the ERT profiles (blue lines,
start of profile marked with circle and number).
10 Heinz-Gerd Röhling et al.
tracted topographical data from the digital elevation model
and incorporated them into the inversion. As the groundwa-
ter table on the plateau of the Brandköpfe is around 50–
100 m below the surface near the Unicorn Cave, we thus
mapped mainly the unsaturated zone of the dolomite rock.
Fig. 7 shows the locations of all ERT profiles. We cov-
ered the entire area around the Unicorn Cave with ERT pro-
files, carried out a dense set of profiles south of the mapped
cave, and tried to identify the northern extensions with sev-
eral profiles perpendicular to the proposed strike direction of
the northern continuation of the cave. Additionally, three
ERT profiles were conducted in the cave (not shown in
Fig. 7).
From the wealth of ERT profiles collected, we discuss
typical profiles for the three parts above the cave, south of
the cave and towards the northern cliff line. We limit our-
selves to the inversion of the Wenner setup, except for pro-
file 26, for which the inversion of the Schlumberger setup
was used. For a more detailed picture, the reader is referred
to Kaufmann et al. (2011, 2012, 2015).
We start by discussing two ERT profiles that cross the
mapped cave. Fig. 8 shows the two profiles 12 and 9. Profile
12 (64 electrodes, 5 m spacing) traverses the entire southern
part of the mapped cave. The resistivities range from
100 Ωm, typical for wet soils, to more than 6000 Ωm, char-
acteristic for air-filled voids. The cave parts, Blue Grotto
(E01), Leibniz Hall (E02), and in parts Schiller Hall (E03)
are shown as highly resistive areas. The thick infill in the
southern part is also visible as high-resistivity anomaly
(E04). The low resistivity part (E05), just south of the
mapped cave, possibly maps a small shaft at the end of the
Van-Alten-Kapelle room.
Profile 9 (32 electrodes, 3 m spacing), crosses the small
side-passage Jacob Friesen Tunnel, which leads to an old en-
trance of the Unicorn Cave, nowadays collapsed. The air-
filled cave passage is clearly visible as highly resistive area
(E06) in the profile.
We continue with three ERT profiles (21, 24, 26), located
south of the mapped cave (Fig. 9). In profile 21 (32 elec-
trodes, 3 m spacing), the most prominent feature is a low-
resistivity anomaly (E08, 100–400 Ωm) in the central part of
the profile. This anomaly, which extends over the entire ver-
tical distance mapped, can be correlated to a large karstified
fissure that strikes south–north, and which is accessible
along a small outcrop to the south (marked as white triangle
in Fig. 7). It seems that this karstified fracture, which can be
seen in various ERT profiles, is filled with soil and thus cap-
tures and stores surface water, producing the low resistivity
signal. The cave void (E07), which has been identified in the
Bouguer anomaly (and subsequently was accessed by drill-
ing) was not reached with the ERT profile due to the steep
terrain in the east, limiting the setup. The resistivities in pro-
file 24 (32 electrodes, 4 m spacing), just north of the small
outcrop mentioned above, are also dominated by the low val-
ues above the karstified fracture (E09). Around this enlarged
fracture, resistivities are much higher (1000–2000 Ωm),
closer to the dolomitic rock. Profile 26 (32 electrodes, 4 m
spacing) links the two profiles just discussed above in an
oblique direction. In this ERT profile, we have identified the
air-filled continuation of the Unicorn Cave as a highly resis-
tive anomaly (E10), and again the karstified fracture (E11)
with its low resistivities.
Next, we move to the north of the Unicorn Cave (Fig. 10),
and discuss three ERT profiles (8, 13, 28). Here, profile 8
(64 electrodes, 5 m spacing) is the northernmost profile, just
south of the Rottsteinklippen cliff. The resistivity section is
dominated by a highly resistive area (E12), with resistivity
values above 4000 Ωm. The anomaly roughly corresponds to
Fig. 8: Selected ERT profiles above the cave. Shown are the topography (black line), the electrical resistivity (colour-coded), the RMS
value and labelled anomalies (explained in the text).
An integrated geophysical and geological interpretation of the area around the Unicorn Cave
the location of the negative Bouguer anomaly discussed be-
fore, a possible hint to an open void, not yet explored. Mov-
ing a bit further south, profile 13 (64 electrodes, 5 m spacing)
is characterised by a broad area of high resistivities (around
1000 Ωm and above), indicative of the soluble dolomitic
host rock. A high-resistivity anomaly (E13) might be an indi-
cation of the cave passage in the north. Going further south,
just above the northern end of the mapped cave passages,
profile 28 (50 electrodes, 5 m spacing) shows three isolated
high resistivity anomalies. Here, anomalies E14 and E15 can
be correlated with the narrow Hubertusgang (E15) and a
shallow passage in the cave leading to the entrance cliff
(E14). For anomaly E16, no mapped cave feature can be
found, which hints at possible continuations.
In this last part of the ERT profiles, we examine the cave
sediments in the known part of the Unicorn Cave. As the
sediment infill consists of sequences of dolomitic blocks,
fluvial gravel and even glacial infill, its composition and
structure is heterogeneous and thus can be mapped with geo-
physical methods. We choose ERT to identify wet and dry
sections as well as air-filled voids. The main ERT profile 2
(64 electrodes, 4 m spacing) in Fig. 11 starts in the Hubertus-
gang in the north and traverses the entire cave reaching the
Van Alten Kapelle in the south. The ERT section essentially
reveals two different resistivities: values around 100–
400 Ωm, representing wet softer sediments, such as cave
clays and gravels, and higher resistivities around 100–
2000 Ωm, which we identified as breakdown areas with
larger dolomitic blocks. With this identification in mind, the
profile reveals softer sediments beneath the White Room
(E17), then blockier collapse areas beneath the two large
rooms Schiller Hall (E18) and Leibniz Hall (E19), and wet-
ter, soft sediments in the area of the Blue Grotto (E20), with
its daylight opening through the large roof collapse.
3.1.3 Self-potential (SP)
SP measurements were conducted with non-polarisable cop-
per-copper-sulphate electrodes and a multimeter along the
positions of the three ERT profiles 20, 21 and 22. The result-
ing map of potential differences is shown in Fig. 12. Here, a
clear negative signal (SP1) with values going down to
-10 mV can be found just above the southern continuation of
the Unicorn Cave, mapped with gravity and subsequently
explored with drilling. The negative signal indicates flow in
the vadose zone towards the open voids of the southern con-
tinuation. Just west of this large negative SP anomaly, an-
other feature (SP2) with similar negative values can be seen,
which correlates with the proposed continuation of the open
fracture. The large negative signal (SP3) in the northernmost
Fig. 9: Selected ERT profiles south of the cave. Shown are the topography (black line), the electrical resistivity (colour-coded), the RMS
value and markers (used in the text).
12 Heinz-Gerd Röhling et al.
Fig. 11: ERT profile from within the cave. Shown the electrical resistivity (colour-coded), the RMS value and markers (used in the text).
Fig. 10: Selected ERT profiles north of the cave. Shown are the topography (black line), the electrical resistivity (colour-coded), the RMS
value and markers (used in the text).
part of the surveyed area is a result of a flow coming from the
topographical hill north.
Thus, the SP measurements support both the Bouguer
anomaly and the high resistive area on profile 26 and con-
firm the presence of a karst void (which has been accessed
by later drilling).
3.1.4 Ground-penetrating radar (GPR)
Structural information was obtained from ground-penetrat-
ing radar measurements. For profiles 1–13, we used a GSSI
SIR4000-console and a 100 MHz monostatic antenna. The
continuous GPR readings were processed with the ReflexW
software package (Sandmeier 2016). Each profile was pro-
cessed with the same stack of filters:
(i) Stack traces of 10 traces (0.3 m),
(ii) dewow (with a 20 ns window) and background removal,
An integrated geophysical and geological interpretation of the area around the Unicorn Cave
(iii) manual gain adjustment,
(iv) running average (over 5 traces),
(v) migration,
(vi) time-depth conversion and topographic correction.
For profiles L1 and L2, we used a GSSI SIR3000 system and
a multiple low frequency antenna with dipole lengths corre-
sponding to approx. 40 MHz centre frequency for high pe-
netration. Processing of the data comprised:
(i) time-zero correction,
(ii) amplitude balancing (header gain removal, compensa-
tion of geometric divergence and material attenuation),
(iii) dewow (with a 50 ns window) and bandpass filtering,
(iv) running average of 8 traces (0.8 m),
(v) migration,
(vi) time-depth conversion and topographic correction.
By fitting hyperbolas in the resulting dataset, we deduced a
typical signal velocity of 0.08–0.09 m/ns, which is used for
migration and to transform the two-way travel-time data to a
corresponding depth domain.
In Fig. 13, a map of all GPR profiles is shown. With the
100 MHz antenna, we surveyed the southern extension of the
Unicorn Cave (profiles 8 to 13) and the Rottsteinklippen cliff
(profiles 1 and 2), and then connected both working areas
with the long profile 4. The two profiles L1 and L2 were
surveyed with the 40 MHz antenna.
In Fig. 14, two representative profiles (01 and 04) are
shown. Profile 01 reveals structural information from the
Fig. 12: SP anomaly. Shown is the topography (grey shaded), the SP gravity (colour-coded), the map of the known part of the Unicorn Cave
(black contour), locations of boreholes (white diamonds), and the locations of outcrops (white triangles).
western part of the Rottsteinklippen cliff. The radargram is
fairly rich in continuous reflectors, revealing the internal
structure of the cliff. Three areas deserve special inspection:
Within the part of the dense dolomite without visible karst
features (R01), a strong reflection with almost non-decaying
amplitude is present (an indication of reverberations, for
which we do not have an explanation). In area R02, the dip-
ping reflectors possibly mark the transition from dense, crys-
talline dolomite to karstified dolomite. In area R03, which
has also been identified by a pronounced minimum in the
Bouguer gravity signal and a high resistivity signal in the
ERT profile, we find strong reflectors down to 12 m depth,
and a vertical structure (at 180 m), characterised by the inter-
ruption in reflections, with a similar structure at 200 m.
Profile 04, which is 650 m long, traverses the entire area
from north to south. In the northern part, we observe some
cross-layering at the top of the dolomite, which probably
stems from the cliff line. About 120 m into the profile (R07),
the signal at the depths of 3–6 m is characterised by various
small-scale reflections, with interrupted reflectors, possibly
karstified fractures or even the top part of a chamber. Around
300 m into the profile (R06), the area around the northern
end of the mapped Unicorn Cave is crossed. Here, a short
reflector at about 6–8 m depth is present. On the southern
side of the GPR profile, above both large rooms, the Leibniz
Hall (R05) and the Blue Grotto (R04), reflections from dis-
solutionally enlarged fractures (such as the mapped chim-
14 Heinz-Gerd Röhling et al.
neys), can be found in the radargram. The rooms themselves,
however, were not reached with the signal from the 100 MHz
antenna. With the estimated signal velocity of 0.08 m/ns, we
were able to reveal structures down to a depth around 12 m
below the surface.
Fig. 15 shows the processed low frequency GPR data
with penetration depth of more than 20 m. The first 100 m
are measured on identical tracks whereas the track splits into
two (see Fig. 13). The reflections within the upper 10 m are
probably caused by internal structures of the dolomite such
as, e.g., bedding or fractures. The known air-filled cave
causes a distinct reflection at 12–18 m depth, which fits very
well to the depth of the roof, as shown by the LIDAR scan
inside the cave (see 3.2). However, the reflection becomes
very weak at 75 m on both profiles. This can be explained by
the fact that the cave is completely filled with sediments be-
tween 70 and 100 m, which significantly reduces the reflec-
tivity of the electromagnetic waves at the roof of the cave.
On profile L2 (western track; Fig. 15a), a second strong re-
flector can be tracked from 105–140 m, where only a weak
and small reflector at around 110 m is visible on profile L1
(eastern track; Fig. 15c). The strong reflector from 105–
140 m on profile L2 was interpreted as a second air-filled
cavity. In contrast, there is no indication of a massive air-
filled cavity below profile L1 and the weak reflection might
be an off-plane reflection from the side or caused by a sedi-
ment-filled or smaller air-filled cavity. The existence of the
predicted 40 m long unknown cavity at 11–14 m depth has
been confirmed by boreholes. The exploration of the detailed
cave geometry would require a 3D GPR survey. This was not
possible due to the large size of the low-frequency antennas
(4 m dipoles) and the forest site location, so that we were
restricted to use the existing forest roads.
3.1.5 Shear wave reflection seismics
Two 2D profiles (Fig. 16) were carried out with a horizon-
tally-polarised (SH) configuration: One of 250 m length
above the cave using a landstreamer with 120 horizontal
geophones (10 Hz resonance frequency) at 1 m intervals
Fig. 13: Map of GPR profiles. Shown is the topography (grey shading), the roads (grey lines), the map of the known part of the Unicorn
Cave (black contour), locations of boreholes (white diamonds), the locations of outcrops (white triangles), and the GPR profiles (green
lines, start of profile marked with circle and number for the 100 MHz data and red lines for the 40 MHz data). Note that the two 40 MHz
profiles are shifted by ±10 m for visibility reasons.
An integrated geophysical and geological interpretation of the area around the Unicorn Cave
along a forest road track that crosses the cave location di-
agonally, near the middle of the profile. The intention of the
first profile was to detect the cave location, either by an am-
plitude anomaly caused by the air-filled cave space or by
diffraction responses from the borders of the cave’s roof.
The second profile was carried out within the cave, on the
cave floor, using a setup of 96 geophones (10 Hz) at 1 m
intervals attached to tripod bases, simply firmly placed on
the cave floor. This technique was chosen since conven-
tional planting of spike geophones could hardly carried out
due to the stiff soil mixed with rock and landstreamer opera-
tion was too difficult within the cave. The intention of this
144 m length profile was to detect the thickness of the sedi-
ments on the cave floor, down to the underlying bedrock. As
seismic source for both profiles, an electrodynamic Elvis
vibrator shaker (version 5) unit was used to generate hori-
zontally polarised waves in the direction of the geophone’s
shaking sensitivity. The source signal was set to 20–120 Hz,
10 s duration, for profile 1, and 30–160 Hz, 10 s duration,
for profile 2. Two signals with alternated polarity were gen-
erated at each source location to enable an S-wavefield sep-
aration by subtractive stacking of both recordings. To enable
a sufficient Common Midpoint (CMP) coverage for statisti-
cal analysis the source location interval was set to 2 m for
both profiles, resulting in a theoretical mean CMP-fold of
30 and 24 for profiles 1 and 2, respectively. Data recording
was carried out by a Geometrics Geode recording system
that consisted of five units of 24 channels each plus one unit
for auxiliary channel recording for profile 1, and 4 units
plus auxiliary unit for profile 2. All records were stored se-
parately in uncorrelated mode without filters applied to en-
able specialised deconvolution methods during data pro-
cessing later on.
Due to the environmental conditions, both profiles did
not meet perfect acquisition standards. The track of profile 1
followed a forest road that consists of a mostly soft weather-
ing surface, supporting the propagation of undesired Love
surface waves. Furthermore, the profile track could not be
arranged perpendicular to the cave long axis and slightly me-
andered, resulting in a scattering of the midpoints. This was
also the case of profile 2, which was restricted to the natural
track of the cave.
Reflection seismic processing was carried out using
VISTA Vers. 10 (Schlumberger) software using a processing
flow described, e.g., in Krawczyk et al. (2013) and Polom et
al. (2013, 2016), where Polom et al. (2013) focus on the
elimination of disturbing Love waves by FK-filtering. The
velocity field for final depth conversion was derived from
seismic data by interactive velocity analysis due to the lack
of other velocity information, e.g. from well logging.
Fig. 17 shows the results (depth-converted Finite Differ-
ence [FD] migration section) of profile 1, acquired at the
surface above the cave. In the profile centre, the strong am-
plitude responses of the top rock surface and at the cave roof
are imaged, probably showing a stack of events due to mul-
tiple wave paths between these two interfaces. The cave
Fig. 14: GPR profiles 1 and 4: Reflection amplitudes are normalised to largest value, topography used for time shift.
16 Heinz-Gerd Röhling et al.
area including the sediment infill below shows significant
lower amplitudes that result from wave pathways around
the cave space. To the NE, a buried sinkhole structure is vis-
ible, indicating a collapse of the cave roof rock. In the SW,
the top rock and cave-roof reflector events continue with
significant weaker amplitudes. Structure resolution is nearly
2 m for the cave roof and less than 1 m for the sinkhole
Fig. 18 shows the result (depth-converted FD migration
section) of profile 2, acquired along the cave floor in the sub-
surface. Despite the situation of full-space wave-propaga-
tion, the image shows laminated layers of weak amplitudes
at nearly 1 m resolution at the top, which represent the sedi-
ment infill of the cave. In general, the sediment infill thick-
ness varies from 10 to 20 m over the 0–100 m of the profile,
in the SE a thickness of >20 m is visible. Below strong am-
plitudes of >2 m resolution indicate massive bedrock and
probably collapsed roof rocks at the top. At position of bore-
hole Leibniz Hall, the first interpretation of top bedrock by
seismic signature differs from the borehole lithology and re-
quired an adaption to the borehole results. The included cave
in the lower sediment column of the borehole probably also
indicates collapsed roof rock at this location. The general dip
of the bedrock is to the SW.
3.2 Geodetical results
We surveyed the internal structure of the Unicorn Cave using
a LIDAR device (Light Detection And Ranging) of the type
Riegl VZ-1000 in 2012. This device uses near-infrared laser
pulses to measure distances and angles to approximately
122,000 point measurements per second (Table 2). Due to
the somewhat limited vertical measurement window of 100°,
010 20 30 40 50 60 70 80 90 100 110 120 130 140 15
levation [m]
distance [m]
010 20 30 40 50 60 70 80 90 100 110 120 130 140 15
air-lled cave air-lled cave
sediment-lled cave
distance [m]
010 20 30 40 50 60 70 80 90 100 110 120 130 140 150
elevation [m]
elevation [m]
ground surfac
Fig. 15: GPR profiles L2 (a) and L1 (c) measured from NE to SW with 40 MHz antennas and interpretation of profile L2 (b). The first
100 m of the profiles measured above the known part of the cave are identical and the last 60 metres of the profile are measured on differ-
ent tracks (see Fig. 13). The crosses in (b) at 20–70 m distance mark the roof of the cave derived from the LIDAR scan (see chapter 3.2).
An integrated geophysical and geological interpretation of the area around the Unicorn Cave
it was necessary to tilt the device out of the vertical in order
to detect the complete structure of the cave.
We chose a point grid of 4 cm at 10 m distance (this was
the typical distance between scanner and cave wall). This
meant that about 3 million measuring points were measured
per scan position. For the entire cave, we scanned from 34
scan positions, creating a total point cloud of about 100 mil-
lion points.
3.2.1 Post-processing and analysis of the scans
The 34 single point clouds were georeferenced using the
Riegl software “Riscan-Pro”. For further work, the amount
of data was reduced to about 279,000 points, so that the
points have a spacing of about 20 cm. We used the program
“MeshLab” to create a triangulated surface of the entire cave
from the point cloud.
Further analysis steps were carried out with the software
“Move” (Midland Valley Exploration Ltd, 2016). The soft-
ware offers various possibilities to analyse a surface. In this
work, two methods were used: the dip of individual triangles
and the curvature of the triangulated surface (Fig. 19). For
Fig. 16: Map of shear wave reflection seismic profiles. Shown is the topography (grey shading), the roads (grey lines), the map of the
known part of the Unicorn Cave (black contour), and the seismic profiles (profile 1: red solid line, profile 2: red dashed line; start of profiles
marked with circle and number) and the locations of outcrops (white triangles).
Table 2: Specification of the LIDAR device used (Riegl VZ-10 00).
Method Time-of-flight
range ca. 1400 m
laser classification 1, according to IEC 60825-1:2014
laser frequency near-infrared, 1550 nm
beam divergence 0.3 mrad (0.019°)
measuring accuracy @
100 m
<4 mm
measuring field
(vertical x horizontal)
100° × 360°
measurements/sec 122,000
camera Nikon D700, 4.5 mm hemispheric
both methods, the information was stored as attributes to the
individual triangulation points of the scanned surface by the
software. Subsequently, the cave model could be coloured by
these attributes.
18 Heinz-Gerd Röhling et al.
3.2.2 Dip analysis of the Unicorn Cave
The analysis of the dip of the structures of the Unicorn Cave
shows the probably sedimentary layering in the walls (Fig.
20). Throughout the cave system, this stratification is sub-
horizontal. However, the stratification trace lines do not
completely pass through the cave (Fig. 20a), but are often
interrupted, which is probably primarily due to the dolomite
sedimentary milieu and how it was deposited.
3.2.3 Curvature analysis and evaluation of the fracture
Fig. 21 shows the entire cave coloured by curvature (as de-
fined in Fig. 19). This type of analysis is good to identify
fracture systems. The long tunnel between the Blue Grotto
and the Schiller Hall is characterised by a NE–SW striking
fracture system, whereas the Schiller Hall contains W–E
striking fractures.
It was also possible, as the laser beam could penetrate
deeply (in some cases over 1 m) into large open fractures, to
identify and analyse the fracture surfaces themselves. Fig. 22
shows an evaluation of such fractures in the roof of the cave.
This analysis confirms that the majority of the fractures
strike NNE–SSW to NE–SW. Only in the Schiller Hall and
halfway between the Blue Grotto and Schiller Hall W–E and
WNW–ESE striking fractures occur.
3.3 Borehole results
Between 1984 and 1987, nearly 100 boreholes were carried
out in and around the Unicorn Cave, from which samples
were taken. Also in the cave, several dynamic samples from
the sediment infill were taken (e.g. Nielbock 1987). The
probes reveal both the composition and the thickness of the
sediment infill of the cave, ranging from 15 m in the White
Room in the north to more than 30 m in the Leibniz Hall in
the south. As an example, in Fig. 23 the probe Core2 is
shown, taken from the White Room and revealing almost
14 m of sediment infill, ending in the greywacke.
In 1987, a borehole was drilled from the surface above
the Leibniz Hall, which passed through the open cave void,
then penetrated about 35 m of variable cave sediments, end-
ing in the dolomitic host rock, close to the underlying
greywacke. A simplified lithology of this core is shown in
Fig. 23, with 14.5 m dolomite overburden above the cave.
The cave sediments below the Leibniz Hall are mainly com-
posed of dolomitic blocks, likely the result of roof and wall
collapse in the cave. Interbedded in between, several layers
Fig. 17: Depth-converted FD migration section of profile 1 at the surface above the cave and perpendicular projection of borehole Leibniz
Hall in the profile plane. The cave area is indicated by a strong reflection amplitude response from the cave roof, where the S-waves could
not propagate into the air filled area. Weak amplitudes below result from residual wave energy propagating around the cave and in the cave
sediment infill. In the NE, a buried sinkhole structure shows probably a collapsed cave roof.
An integrated geophysical and geological interpretation of the area around the Unicorn Cave
of muddy clay deposits with river gravels were found as
evidence of a fluvial phase in the evolution of the cave. In
the upper part of the sediment infill, glacial deposits from
the Pleistocene period mark infill during glacial times, pos-
sibly via the old side entrance in the passage Jacob Friesen
In 2014, after the prediction of the southern extension of
the Unicorn Cave by gravimetric, geoelectric and ground-
penetrating radar surveys, three boreholes were drilled just
south of the collapse area towards the south of the Blue
Grotto (Fig. 23). All three boreholes passed around 12–14 m
of dolomitic host rock, and then reached an air-filled cavity
(Fig. 24). While two neighbouring boreholes traversed an
air-filled room about 1–1.5 m in height, the easternmost
borehole just crossed the edge of this passage, with 50 cm of
air-filled passage. All three boreholes then continued for
about 5–8 m into the cave sediment.
Fig. 18: Depth-converted FD migration section of profile 2 along the cave floor and boreholes White Room (NE) and Leibniz Hall (SW).
A stack of fine-structured reflection responses at the top indicates the sediment infill in the cave, partly with coarse-structured events below
in the NE that probably indicate a pile of boulders. The bedrock is indicated by strong amplitude responses that dip 30° to SW. Interpreta-
tion of top bedrock by the significant change of seismic amplitude signature (dotted yellow line) misfits the lithology of borehole Leibniz
Hall and required an borehole adapted interpretation (dashed yellow line).
Fig. 19: (a) Two adjacent triangles (ABC and ACD) with their
poles (P1 and P2) in three-dimensional space. α is the dip of triangle
ABC. (b) Cross-section along the dashed line in (a). φ is the angle
and s the arc length between the poles. The curvature of the two
triangles (κ) is defined as φ/s.
20 Heinz-Gerd Röhling et al.
Fig. 20: Parts of the 3D triangulated surface of the Unicorn Cave. (a) Schiller Hall (view from the south). (b) Blue Grotto (view from
SSW); triangles are coloured by dip (see Fig. 1 for explanation). Clearly visible are thin (<50 cm) horizontal traces, which dip flatly (<40°),
between steeply (>70°) dipping areas. We interpret this pattern is due to differential erosion of the dolomite layers during the formation of
the cave, so that the pattern represents the sedimentary layering. Clearly visible in (a) are the thin “finger-like” lobes on the roof of the
Schiller Hall, which were created when the laser beam penetrated into open fractures.
Fig. 21: Curvature analysis of the cave system in (a) map view and (b) vertical projection, viewed from the SE, with topographic surface
for orientation. The highest curvature is created at the edge of fractures in the dolomites and is depicted with brighter colours.
An integrated geophysical and geological interpretation of the area around the Unicorn Cave
layers, three-dimensional structural objects (e.g. voids, infill,
…) can be present.
We start assembling the structural model with a pre-de-
fined two-dimensional topography, then define the different
lithological units, Werra dolomite and greywacke. The reso-
lution of both topography and subsurface structure is 2 m. A
structural model of this setup is shown in Fig. 25 (top): View
is towards the northwest, topography is shown as colour-
coded contours, the Unicorn Cave in red colours (air-filled
voids) and blue colours (sediment infill). Note that towards
Fig. 22: Rose diagrams to show the strike direction of certain large fractures in the roof of the cave (orange lines). Approximately 20 values
per diagram.
4. Structural model
In this section, we derive a three-dimensional structural
model of the Unicorn Cave and its surroundings. We use the
PREDICTOR package (e.g. Kaufmann et al. 2012, 2015),
which defines a three-dimensional setup of the subsurface,
based on a two-dimensional topographical model and differ-
ent lithological layers in the subsurface. Each layer in the
subsurface can have different material properties (e.g. den-
sity, magnetisation, resistivity, …), and embedded in these
22 Heinz-Gerd Röhling et al.
Fig. 23: Borehole lithologies. Top left: Deep borehole through Leibnizhalle. Top right: Probe in White Room. Bottom: New boreholes south of the mapped cave.
An integrated geophysical and geological interpretation of the area around the Unicorn Cave
the south, the mapped Unicorn Cave is extended (represent-
ing the new parts) to provide a basis for predicting the grav-
ity signal. The topography shows the small hill just west of
the cave, and the small valley with the entrance cliff (see
entrance roof of the show cave). Also shown are selected
boreholes, e.g. the larger borehole from the surface (above
the Leibniz Hall) through the cave into the sediments, reach-
ing the dolomite again and stopping close to the top of the
greywacke. The three boreholes for the investigation of the
new southern parts are also shown, and a small probe low-
ered from within the cave into the sediments of the White
In Fig. 25 (bottom), the topography is only shown as con-
tours, and the sediment infill of the cave (blue colours) as
well as the greywacke (dark colours) are shown instead. The
sediment infill has been extrapolated from the borehole logs,
with a thinner sediment thickness in the northern part of the
Unicorn Cave (around 15–20 m), and thicker sediment (up to
40 m) in the southern sections. Note that we continued with
this thick sediment infill in the newly discovered southern
extension. The greywacke dips from north towards the south,
as suggested by Paul & Vladi (2001), which is one reason for
the asymmetrical infill of the Unicorn Cave (Kaufmann et al.
2015). The three ERT profiles shown are indicative of the
supporting information obtained from this geophysical
method to derive the structural model.
In Fig. 26, we present the Bouguer gravity derived from
the structural model described above. In the left part of the
figure, the observed Bouguer gravity over the Unicorn Cave
and its southern extension is shown (as discussed earlier). In
the right part, the predicted Bouguer gravity is shown. The
prediction is based on the geophysical prediction part of the
PREDICTOR package, in this case based on the different
densities of the air-filled cave, the cave sediments, the Werra
dolomite and the underlying greywacke.
We observe a similar broad negative Bouguer signal for
the predicted anomaly, comparable in size and amplitude
(-0.5 mGal) to the observations. The two chimneys in the
Leibniz Hall are present in the prediction, though less pro-
nounced. The large negative signal (B) resulting from the
mass deficit of the rooms Leibniz Hall and Blue Grotto, and
of course their sediment infill, which also has a lower density
when compared to the dolomitic host rock, are clearly visible
in the predicted signal. The small shaft in the Van-Alten-
Kapelle (C) is not predicted, which is a consequence of the
coarse 2 m-resolution of the model. However, the Bouguer
gravity lows (D, E) over the southern extension of the Uni-
corn Cave are well detected by the assembled structural
5. Discussion
We used a combination of different geophysical and geodeti-
cal methods to map the area in and around the Unicorn Cave
in the southern Harz Mountains. Our main aim is to identify
Fig. 24: Drilling rig exploring the previously unknown cave area. Bottom right: Camera control during the borehole logging. Bottom left:
First shot of the new cave room. Drilling equipment provided by Rump & Salzmann, Osterode and Knauf-Gips, Iphofen; borehole camera
provided by Ingenieurbüro Reinhard Völker, Uftrungen; access permission granted by Dieter Sauerbrey. Photos: Nielbock, Völker 2014.
24 Heinz-Gerd Röhling et al.
Fig. 25: Structural model of the Unicorn Cave. Top: Topography (colour-coded), cave (red), hut (blue), boreholes (vertical cylinders). Bot-
tom: Topography (contours), cave (red) and cave sediments (blue), greywacke (dark colour), and ERT profiles 12, 27 and 28 (colour-
An integrated geophysical and geological interpretation of the area around the Unicorn Cave
the cave in the signal of the different geophysical methods,
and with this knowledge to estimate, if and where we can
find continuations of the known cave in northern and south-
ern directions.
5.1 Known cave and southern extensions
The gravity survey revealed a clear correlation between cave
voids and negative Bouguer anomaly. We were able to map
both the air-filled passages of the Unicorn Cave and the large
sediment infill of the cave. By calibrating the Bouguer grav-
ity signal to the known cave parts, we have been able to iden-
tify a southern continuation of the Unicorn Cave. Through
inverse modelling of the Bouguer gravity signal, this con-
tinuation could be located and additionally its sediment infill
has been estimated. Thus the gravity method, with its poten-
tial to map an indirect geophysical signal in rougher terrain,
has been shown to be an excellent method to identify shal-
low cave voids and their infill.
Air-filled voids of the known cave rooms were identified
with electrical resistivity imaging as high-resistivity areas.
With this strong correlation above the known cave parts,
ERT measurements beyond the surveyed cave were used to
identify yet unknown voids in the south and a possible con-
tinuation in the north. The ERT profiles thus provide a good
opportunity to identify voids in the host rock (in parts even
the sediment-filled voids), along with their approximate ex-
tensions and limitations.
Both ground-penetrating radar and shear wave reflection
seismics have been able to identify the cave roof of the known
air-filled parts of the Unicorn Cave through clear reflections
of the electro-magnetic and elastic wave patterns, respec-
tively. However, beyond the detection of the cave roof by a
strong amplitude response, shear wave reflection seismics
could not clearly delineate the shape of the cave below, be-
cause the wave propagation was hampered by irregular ray
geometries and scattering. This led to apparent amplitude re-
sponse images located in the cave area itself, even though no
shear wave energy could in fact pass the air-filled space of
the cave.
Especially the low frequency GPR surveys were helpful
in identifying the roof part of the air-filled cavities south of
the mapped Unicorn Cave. Here the combination of gravity,
electrical resistivity imaging, and ground-penetrating radar
results were successfully used to identify the location of
boreholes to prove the existence of a southern continuation
with air-filled passages.
The additional self-potential measurements, south of the
mapped cave, identified clear minima in the electrical poten-
tial both above the southern continuation and the large open
fracture identified in the ERT profiles. Thus, self-potential
mapping helps to unravel flow paths in the shallow vadose
With the LIDAR scan of the Unicorn Cave, we have
clearly shown the strong structural control of the cave pas-
sages, mainly aligned in NE–SW direction. A similar strike
direction has been identified for the open fracture close to
the surface from ERT measurements. At present, LIDAR is
the best choice for precise exploration of cave systems (e.g.
Canavese et al. 2009; Nothegger & Dorninger 2009; Plan &
Xavier 2010; Roncat et al. 2011; Buchroithner & Gaisecker
Fig. 26: Observed (left) and predicted (right) Bouguer anomaly. Labelled anomalies are discussed in text.
26 Heinz-Gerd Röhling et al.
2009). We show here that the laser scan, after post-process-
ing and analysis of the resulting surface, is suitable to study
the geological structure of the cave, for instance for both
sedimentary layering and fracture orientation. We believe
that these structural tools are an important complement to
LIDAR cave scanning, especially when emphasis is placed
on additional geological information. The Riegl Scanner
VL-1000 has a minimum range of about 3.5 m, which could
lead to problems in even narrower cave systems. There are
other LIDAR devices that use phase change of the returning
beam and can work in a narrower range.
5.2 Northern extensions
In the northern part of the Brandköpfe Plateau, the geophys-
ical signals are not clear. Bouguer gravity only reveals broad
negative anomalies; there are sharp peaks as in the south.
The Bouguer values close to zero are even slightly above
both along the western edge of the Rottstein Cliff and the
northern end of the mapped Unicorn Cave indicating a
change in the structure of the dolomite, from more fractured
and karstified (in vicinity of the Unicorn Cave) to denser and
probably no significant karstification in the northwest. The
central and eastern part of the Rottstein Cliff again are char-
acterised by negative Bouguer anomalies, a possible hint for
the northern continuation of the cave.
The ERT profiles in the north support these hypotheses,
with clearly bounded high-resistivity anomalies identified in
three profiles. Here, a denser mapping of the gravity signal is
required, along with more GPR and/or seismic profiles. A
probable location for another borehole can be the high-resis-
tivity anomaly E12 in ERT profile 8 (Fig. 10), with its pos-
sible continuation as anomaly E13 in profile 13.
5.3 Cave sediments
The cave sediments have already been sampled with numer-
ous hand-drilled cores (e.g. Nielbock 1987; Paul & Vladi
2001). From these samples, we know that the sediments are
a rich mixture of dolomitic breakdown blocks, river gravels,
ice-age tills, and residual lutite from the dissolution of the
dolomite. This complex setting is mirrored by geophysical
measurements taken in the cave. The electrical resistivity
imaging reveals a blockier structure in the southern part of
the cave (Leibniz Hall), identified by higher resistivities and
finer infill in the north, characterised by lower resistivities.
The shear wave seismics is able to provide the lower bound-
ary between cave sediments and dolomitic host rock, which
dips towards the south, where the cave sediments reach their
largest thickness. Correlation with the borehole lithologies
Leibniz Hall and White Room shows that the top bedrock
interpretation requires a modification beyond a pure seismic
amplitude response interpretation. It shows that the strong
lateral heterogeneities of the sediment infill probably mixed
with collapses from the cave roof and enclosed small cavities
may affect the seismics imaging locally.
In the southernmost part of the cave (Blue Grotto), the
infill changes again to low-resistivity material, probably be-
cause of the fine material washed in through the daylight
6. Conclusions
The Unicorn Cave in the southern Harz Mountains with its
long research tradition both in geological and archaeological
sciences provides a unique location for detailed geophysical
and geodetical investigations. Both its accessibility and the
shallow overburden can be exploited by several geophysical
methods to provide a comprehensive picture of the cave and
its surroundings from indirect measurements.
We have successfully used gravity and electrical resistiv-
ity imaging to map the known cave and its infill, and local-
ised the cave roof with ground-penetrating radar and seismic
measurements. From the signal gathered above and south of
the known cave, we have been able to propose the location
and depths of the continuation of the cave, which has been
confirmed by drilling. From additional inverse modelling,
we were able to estimate the thickness of the sediment infill
in locations, where borehole information is not given.
The LIDAR scan provides a very high-resolution picture
of the cave interior. The analysis of the survey data with tools
for structural geology enabled us to map the faults directly,
along which the cave passages are aligned.
The successful application of the different geophysical
and geodetical methods offers unique insight into a complex
karst object and provides a useful tool beyond the possibili-
ties of direct inspection.
7. Acknowledgements
GK acknowledges funding under DFG-grant KA1723/6-2.
The drilling equipment was provided by Rump & Salzmann,
Osterode, and Knauf-Gips, Iphofen, the borehole camera by
Reinhard Völker, Uftrungen. The Gesellschaft Unicornu fos-
sile e.V team (Ralf Nielbock, Heinz-Gerd Röhling, Henning
Zellmer) thanks everyone for technical support. Further-
more, we thank the reviewers, Christina Flechsig, Leipzig,
and Florian Bleibinhaus, Jena, for their constructive reviews.
8. References
Al-fares, W., Bakalowicza, M., Guérin, R. & Dukhand, M. (2002):
Analysis of the karst aquifer structure of the Lamalou area (Hé-
rault, France) with ground penetrating radar. – J. Appl. Geo-
phys., 51 (2/4): 97–106.
Brauckmann, C., George, K., Gröning, E., Kellner-Depner, C.,
Knolle, F., Leuschner, J., Linke, C., Müller, R., Nielbock, R.,
Radday, H., Röhling, H.-G., Vladi, F., Weber, K.-F., Wilde, V. &
Zellmer, H. (2009): Geopark Harz . Braunschweiger Land .
Ostfalen – Die klassischen Quadratmeilen der Geologie: 48 p.;
Königslutter (Geopark . Braunschweiger Land . Ostfalen GbR).
An integrated geophysical and geological interpretation of the area around the Unicorn Cave
Brauckmann, C., George, K., Gröning, E., Kellner-Depner, C.,
Knolle, F., Leuschner, J., Linke, C., Müller, R., Nielbock, R.,
Radday, H., Röhling, H.-G., Vladi, F., Weber, K.-F., Wilde, V. &
Zellmer, H. (2013): Geopark Harz . Braunschweiger Land .
Ostfalen – The classical square miles of geology: 48 p.;
Königslutter (Geopark . Braunschweiger Land . Ostfalen GbR).
Buchroithner, M.F. & Gaisecker, T. (2009): Terrestrial laser scan-
ning for the visualization of a complex dome in an extreme Al-
pine cave system. – Photogramm., Fernerkund., Geoinfo., 4:
329–339; DOI: 10.1127/1432-8364/2009/0025
Butler, D. (1984): Microgravimetric and gravity-gradient tech-
niques for detection of subsurface cavities. – Geophysics, 49
(7): 1084–1096.
Canavese, E.P., Tedeschi, R. & Forti, P. (2009): The caves of Naica
– Laser scanning in extreme underground environments. – The
American Surveyor, 6 (2): 3–10.
Dobecki, T.L. & Upchurch, S. (2006): Geophysical applications to
detect sinkholes and ground subsidence. – Leading Edge, 25
(3): 336–341.
El-Qady, G., Hafez, M., Abdalla, M. & Ushijima, K. (2005): Imaging
subsurface cavities using geoelectric tomography and ground-
penetrating radar. – J. Caves Karst Stud., 67 (3): 174–181.
Herrmann, A. (1956): Der Zechstein am südwestlichen Harzrand
(seine Stratigraphie, Fazies, Paläogeographie und Tektonik). –
Geol. Jb., 72: 1–72.
Hillgruber, F., Lehmann, J., Nielbock, R. & Terberger, T. (2014):
Die Einhornhöhle im Lichte alter und neuer Forschungen. –
Ber. Denkmalpfl. Nieders., 34 (4): 153–155.
Jordan, H. (1979): Der Zechstein zwischen Osterode und Duder-
stadt (südliches Harzvorland). – Z. Dt. Geol. Ges., 130: 145–
Kaufmann, G. (2014): Geophysical mapping of solution and col-
lapse sinkholes. – J. Appl. Geophys., 111: 271–288.
Kaufmann, G. & Romanov, D. (2017): The Jettencave, Southern
Harz Mountains, Germany: Geophysical observations and a
structural model of a shallow cave in gypsum/anhydrite-bear-
ing rocks. – Geomorphology, 298: 20–30.
Kaufmann, G., Romanov, D. & Nielbock, R. (2011): Cave detec-
tion using multiple geophysical methods: Unicorn cave, Harz
Mountains, Germany. – Geophysics, 76 (3): B71–B77.
Kaufmann, G., Jahn, G., Galindo-Guerreros, J. & Nielbock, R.
(2012): Geophysical exploration of cave sites: The case of the
Unicorn Cave, Scharzfeld/Harz, Germany. – Braunschw.
Naturk. Schr., 11: 69–80.
Kaufmann, G., Nielbock, R. & Romanov, D. (2015): The Unicorn
Cave, Southern Harz Mountains, Germany: From known pas-
sages to unknown extensions with the help of geophysical sur-
veys. – J. Appl. Geophys, 123: 123–140.
Kempe, S. (2005): Karstgebiete und Höhlen in Deutschland. –
Geogr. Rundsch., 57 (6): 44–52.
Krawczyk, C., Polom, U. & Beilecke, T. (2013): Shear wave reflec-
tion seismics as a valuable tool for near-surface urban applica-
tions. – Leading Edge, 32 (3): 256–263.
Leibniz, G.W. (2014): Protogaea sive de prima facie telluris et anti-
quissimae historiae vestigiis in ipsis naturae monumentis dis-
sertation. – Scheid, C.L. (ed.) mit der Übers. von Wolf von
Engelhardt. Mit einer Einführung von F.-W. Wellmer unter
Mitwirkung von Mike Reich u. a.; Nachdruck der Ausg. Göt-
tingen 1749 und Stuttgart 1949: 203 p., Hildesheim (Olms-
Loke, M.H. & Barker, R.D. (1995): Rapid least-squares inversion
of apparent resistivity pseudosections using a quasi-Newton
method. – Geophys. Prospect., 44 (1): 131–152.
Loke, M.H. & Barker, R.D. (1996): Practical techniques for 3D re-
sistivity surveys and data inversion. – Geophys. Prospect., 44:
Meischner, D. (2006a): Das Blaue vom Himmel herunter – Warum
ist die Blaue Grotte in der Einhornhöhle im Harz blau? – Mitt.
Verb. Dt. Höhlen- u. Karstforscher, 52 (1): 6–7.
Meischner, D. (2006b): Klima-Extreme in Winter und Sommer in
der Einhornhöhle bei Scharzfeld im Harz. – Mitt. Verb. Dt.
Höhlen- und Karstforscher, 52 (1): 8–13.
Meischner, D. (2011): Eingeregelte Fossilien im Lehm der Ein-
hornhöhle bei Scharzfeld im Harz. – Mitt. Verb. Dt. Höhlen-
und Karstforscher, 57 (2): 45–48.
Mochales, T., Casas, A.M., Pueyo, E.L., Pueyo, O., Roman, M.T.,
Pocovi, A., Soriano, M.A. & Anson, D. (2008): Detection of
underground cavities by combining gravity, magnetic and
ground penetrating radar surveys: A case study from the
Zaragosa area, NE Spain. – Environ. Geol., 53: 1067–1077.
Nielbock, R. (1987): Holozäne und jungpleistozäne Wirbeltier-
faunen der Einhornhöhle/Harz. – Diss. TU Clausthal: 194 p.
Nielbock, R. (1989): Die Tierknochenfunde der Ausgrabungen
1987/88 in der Einhornhöhle bei Scharzfeld. – Archäol. Kor-, 19: 217–230.
Nielbock, R. (1990): Die Einhornhöhle – Ein quartärwissenschaftli-
ches Kleinod im Südharz. – Mitt. Verb. Dt. Höhlen- und Karst-
forscher, 36: 24–27.
Nielbock, R. (2002a): Die Einhornhöhle – Forschungsstand und
Perspektiven. – Abh. Karst- u. Höhlenkd., 34: 5–11.
Nielbock, R. (2002b): Die Suche nach dem diluvialen Menschen –
oder: Die Erforschungsgeschichte der Einhornhöhle. – Die
Kunde, N. F., 53: 57–65.
Nielbock, R. (2004): Archäotop Einhornhöhle. – Mitt. Verb. Dt.
Höhlen- und Karstforscher, 50 (2): 42–43.
Nielbock, R. (2008): Die Harzer Dolomiten – Natur- und Land-
schaftsinterpretation unter und übertage. – In: Röhling, H.-G. &
Zellmer, H. (ed.): GeoTop 2008. Landschaften lesen lernen. –
Schriftenr. Dt. Ges. Geowiss., 56: 146–151.
Nielbock, R. (2010): Die Einhornhöhle. Die Welt der Einhörner,
Höhlenbären und Neandertaler: 48 p.; München (Pfeil).
Nielbock, R. (2014): Die Einhornhöhle im Südharz – Facetten-
reiche naturnahe Nutzung eines Geotops. – ZELT Forum, 7:
Nielbock, R., Röber, S., Röhling, H.-G., Thomae, M. & Zellmer, H.
(2004): Der Geopark Harz . Braunschweiger Land . Ostfalen –
Geologie-Erlebnis für Jedermann. – In: Friedel, C.H. &
Röhling, H.-G. (ed.): GeoLeipzig 2004 – Geowissenschaften
sichern Zukunft. – Schriftenr. Dt. Geol. Ges., 35: 9–42.
Nielbock, R., Röhling, H.-G. & Vladi, F. (2006): Wege in den
Untergrund – Die Zechstein-Karstlandschaft am Südharz. –
In: Feldmann, L. & Look, E.-R. (ed.): Faszination Geologie –
Die bedeutendsten Geotope Deutschlands: 14–18; Stuttgart
Nielbock, R., Röhling, H.-G. & Unicornu fossile (2013): Die Ein-
hornhöhle: Geotourismus, Umweltbildung und Forschung in
einem bedeutsamen Geotop im GeoPark Harz . Braunschweiger
Land . Ostfalen. – Schriftenr. Dt. Ges. Geowiss., 85: 635–636.
Nothegger, C. & Dorninger, P. (2009): 3D filtering of high-resolu-
tion terrestrial laser scanner point clouds for cultural heritage
documentation. – Photogramm., Fernerkund., Geoinfo., 2009
(1): 53–63; DOI: 10.1127/0935-1221/2009/0006
Nyquist, J.E., Peake, J.S. & Roth, M.J.S. (2007): Comparison of an
optimized resistivity array with dipole-dipole soundings in
karst terrain. – Geophysics, 72 (4): 139–144.
28 Heinz-Gerd Röhling et al.
Paul, J. (1982): Zur Rand- und Schwellenfazies des Kupferschie-
fers. – Z. Dt. Geol. Ges., 133: 571–605.
Paul, J. (1993): Anatomie und Entwicklung eines permo-triassi-
schen Hochgebietes: Die Eichsfeld-Altmark-Schwelle. – Geol.
Jb., A 131: 197–218.
Paul, J. (2006): Der Kupferschiefer: Lithologie, Stratigraphie, Fa-
zies und Metallogenese eines Schwarzschiefers. – Z. Dt. Ges.
Geowiss., 157 (1): 57–76.
Paul, J. (in press): 7.4. Zechstein am Harzrand und auf der Huns-
rück-Oberharz-Schwelle. – In: Deutsche Stratigraphische
Kommission (ed.; Koord. & Red.: J. Paul & H. Heggemann):
Stratigraphie von Deutschland, XII. Zechstein. – Schriftenr. Dt.
Ges. Geowiss., 89.
Paul, J. & Vladi, F. (2001): Zur Geologie der Einhornhöhle bei
Scharzfeld am südwestlichen Harzrand. – Ber. Naturhist. Ges.
Hannover, 143: 109–131.
Paul, J., Heggemann, H., Dittrich, D., Hug-Diegel, N., Huckriede,
H., Nitsch, E. & AG Zechstein der SKPT/DSK (2018): Erläu-
terungen zur Stratigraphischen Tabelle von Deutschland 2016:
die Zechstein-Gruppe / Comments to the Stratigraphic Chart of
Germany 2016: The Zechstein Group. – Z. Dt. Ges. Geowiss.,
169 (2): 139–145.
Plan, L. & Xaver, A. (2010): Geomorphologische Untersuchung
und genetische Interpretation der Dachstein-Mammuthöhle
(Österreich). – Die Höhle, 61: 18–38.
Polom, U., Bagge, M., Wadas, S., Winsemann, J., Brandes, C., Bi-
not, F. & Krawczyk, C.M. (2013): Surveying near-surface de-
pocentres by means of shear wave seismic. – First Break, 31:
Polom, U., Mueller, C., Nicol, A., Villamor, P., Langridge, R.M. &
Begg, J. (2016): Finding the concealed section of the Whaka-
tane Fault in the Whakatane Township with a shear wave land-
streamer system: A seismic surveying report. – GNS Science
Open File Report 2016/41: 41 p.
Rackow, W. (2004): Die Fledermausfauna der Einhornhöhle und
des Landkreises Osterode am Harz. – Mitt. Verb. Dt. Höhlen- u.
Karstforscher, 50 (2): 59–60.
Rackow, W. (2014): Positiver Trend des Fledermaus-Winterbe-
stands in der Einhornhöhle bei Scharzfeld, Landkreis Osterode
am Harz. – Mitt. Verb. Dt. Höhlen- u. Karstforscher, 60 (1):
Reinboth, F. (1996): Zur Geschichte der Höhlenforschung im Harz.
– Karst u. Höhle, 1994/95: 63–80.
Richter-Bernburg, G. (1955a): Über salinare Sedimentation. – Z.
Dt. Geol. Ges., 105: 593–646.
Richter-Bernburg, G. (1955b): Stratigraphische Gliederung des
deutschen Zechsteins. – Z. Dt. Geol. Ges., 105: 843–854.
Röhling, H.-G. (2013): Der Buntsandstein im Norddeutschen
Becken – Regionale Besonderheiten. – In: Deutsche Stratigra-
phische Kommission (ed.; Koordination u. Redaktion: J. Lep-
per & H.-G. Röhling für die Subkommission Perm-Trias):
Stratigraphie von Deutschland, XI. Buntsandstein. – Schriftenr.
Dt. Ges. Geowiss., 69: 269–384.
Röhling, H.-G. & Nielbock, R. (2005): Einhornhöhle und Rhume-
quelle – Herausragende Geotope im südniedersächsischen
Zechstein-Karst. – Eichsfelder Heimatzeitschrift, 49 (5): 161–
Röhling, H.-G. & Nielbock, R. (2013): Bio- und Geotop Einhorn-
höhle. Rezente und fossile Tierwelten. – Biologie in unserer
Zeit, 43 (3): 184–190.
Roncat, A., Dublyansky, Y., Spötl, C. & Dorninger, P. (2011): Full-
3D surveying of caves: A case study of Märchenhöhle (Aus-
tria). – IAMG 2011 Conf. Paper, Salzburg, September 5–9: 11
p.; DOI: 10.5242/iamg.2011.0074
Rosendahl, W., Döppes, D., Joger, U., Laskowski, R., López Cor-
rea, M., Nielbock, R. & Wrede, V. (2005): New radiometric dat-
ings of different cave bear sites in Germany – Results and inter-
pretations. – Bull. Soc. Hist. Nat. Toulouse, 141 (1): 39–46.
Sandmeier, K. (2016): REFLEXW Version 8.0 User Guide.
Tanner, D., Dietrich, P., Nielbock, R., Röhling, H.-G., Krawczyk,
C. & Vogel, D. (2012): Auswertung der Kluft- und Störungssys-
teme der Einhornhöhle, Niedersachsen, mit Hilfe eines ter-
restrischen LIDAR-Scans. – In: Haneke, J., Lang, R. & Röh-
ling, H.-G. (ed.): GeoTop 2012 – Landschaften und ihr Geo-
potential. – Schriftenr. Dt. Ges. Geowiss., 79: 59–65.
Veil, S. (1989): Die archäologisch-geowissenschaftlichen Aus-
grabungen 1987/88 in der Einhornhöhle bei Scharzfeld. –
Archäol., 19: 203–216.
Vladi, F. (2004a): Zur Geologie der Einhornhöhle. – Mitt. Verb. Dt.
Höhlen- und Karstforscher, 50 (2): 44–51.
Vladi, F. (2004b): Ein geologischer Gang durch die Einhornhöhle.
– Unser Harz, 2 (4): 28–33.
Wenzel, F. (1996): The Nanogal software: Earth tide data process-
ing package ETERNA 3.30. – Bull. Inf. Marees Terr., 124:
Manuscript received: 01.03.2019
Revisions required: 17.07.2019
Revised version received: 25.07.2019
Accepted for publication: 26.07.2019
Gedruckt mit freundlicher Unterstützung von
... Unfortunately, the method is most deficient in terms of the true subsurface stratum's horizontal compositions, actual determination of the contaminants and the nature of the underground water aquifers. Because of the robustness, and less prone to external interferences like other geophysical methods, such as the potential fields and seismic methods, Electrical Resistivity Tomography techniques, (ERT) are frequently applied to subsurface stratum differentiation, e.g., (Malehmir et al., 2016, Mary et al., 2016, Skenderija, 2018, Cracknell et al., 2019, Kayode et al., 2019a, Kayode et al., 2019b, Röhling et al., 2019. ...
... In another dimensions, Yasir et al. (2018) (2019), on the other hands, combined several geophysical methods with the ERT, to delineate lateral, and vertical extent of an anthropogenic deposits, so as to discriminate landfill wastes from ashes and lime. There are many more recent applications of the ERT methods in varieties of subsurface characterizations, e.g., (Butchibabu et al., 2019, Cardarelli et al., 2018, Giocoli et al., 2019, Gupta et al., 2019, Röhling et al., 2019, Zhang and Chen, 2019, Arifin et al., 2018, Arifin et al., 2019. Nevertheless, geophysical methods is not the simple, and seemingly all round magical solutions to a complicated subsurface stratum dilemma. ...
Full-text available
Environmental hazards, industrial, and municipal wastes geochemical and geophysical assessments were carried out at an industrial waste disposal (IWD) site at Bukit Kepong, Kuala Lumpur, Malaysia. RES2-D geophysical method was applied, capable of identification and quantification of the industrial wastes; buried hazardous materials (BHM) and their effects on the subsurface stratum, from the moderately saturated zones, to fully saturated zones housing the aquifer units underneath the water table. Six RES2-D survey profiles were respectively acquired along E-W, and N-S directions. The perpendicular arrangement of the RES2-D survey lines, was tenaciously designed to make possible, the industrial waste materials (IWM)and municipal solid waste (MSW) quantification, with sufficient length of survey lines set at 200 m, and electrode spacing of 5 m, to cover as much details segments of the IWM and MSW as possible. The six RES2-D inversion results, helped in the subsurface stratum classification into three layers, namely; soft layers, which encompasses the waste materials, with varied resistivity values i.e., 0–100 Ω-m, at 10-15 m depths. The consolidated layers produced varied resistivity values i.e., 101–400 Ω-m, at 15-20 m depths. The bedrock has the highest resistivity values i.e., 401 – 2000 Ω-m, at depths >20 m. The estimated volume of the waste materials was 312,000 m 3, using 3-D Oasis Montaj modeling via rectangular prism model generated from the inverted RES2-D. Results from the geochemical analysis helped in the validation of the site as a potential contaminated zone with severe health effects.
Full-text available
While there is substantial evidence for art and symbolic behaviour in early Homo sapiens across Africa and Eurasia, similar evidence connected to Neanderthals is sparse and often contested in scientific debates. Each new discovery is thus crucial for our understanding of Neanderthals’ cognitive capacity. Here we report on the discovery of an at least 51,000-year-old engraved giant deer phalanx found at the former cave entrance of Einhornhöhle, northern Germany. The find comes from an apparent Middle Palaeolithic context that is linked to Neanderthals. The engraved bone demonstrates that conceptual imagination, as a prerequisite to compose individual lines into a coherent design, was present in Neanderthals. Therefore, Neanderthal’s awareness of symbolic meaning is very likely. Our findings show that Neanderthals were capable of creating symbolic expressions before H. sapiens arrived in Central Europe. A 3D video of the engraved giant deer bone is available online. It is free to view at:
Full-text available
In the past few years, construction extended extraordinarily to the southeast of Cairo, Egypt, where limestone caves occur. The existence of caves and sinkholes represents a hazard for such new urban areas. Therefore, it is important to know the size, position, and depth of natural voids and cavities before building or reconstruction. Recently, cavity imaging using geophysical surveys has become common. In this paper, both geoelectric-resistivity tomography using a dipole-dipole array and ground-penetrating radar (GPR) have been applied to the east of Kattamya at Al-Amal Town, Cairo, to image shallow subsurface cavities. The state is planning to construct a new housing development there. The resistivity survey was conducted along three profiles over an exposed cave with unknown extensions. The radar survey was conducted over an area of 1040 m2, and both sets of data were processed and interpreted integrally to image the cave as well as the shallow subsurface structure of the site. As a result, the cave at a depth of about 2 m and a width of about 4 m was detected using the geophysical data, which correlates with the known cave system. Moreover an extension of the detected cave has been inferred. The survey revealed that the area is also affected by vertical and nearly vertical linear fractures. Additionally, zones of marl and fractured limestone and some karstic features were mapped.
Full-text available
Einleitung Die Einhornhöhle am Südwestlichen Harzrand war bereits seit Jahrhunderten Anlaufpunkt wissensbegieriger Menschen. Nachdem die Höhle zunächst vor allem von Einhorn-Ausgräbern, sie wurde aber auch schon früh von Geowissenschaftlern und Forschern aufgesucht. Hier sind besonders u. a. die Universalgelehrten Leibniz (1685), Goethe (1784) und später Virchow (1872) zu nennen. Bereits Ende des 19. Jahrhunderts begannen intensive wissenschaftliche Untersuchungen in der Einhornhöhle (Nielbock 2002). Trotz der vielen dann folgenden Ausgrabungen, war jedoch bis vor wenigen Jahren wenig über die Vielfalt und das tatsächliche Alter der fossilen Tierwelt, die Begehung durch Neandertaler als auch über die tatsächliche Größe der Einhornhöhle bekannt. Völlig neue Erkenntnisse über die eiszeitlichen Bewohner und die Ausmaße der Höhle dieser Höhle brachte jedoch erst eine 1985 zunächst durch das Geologische Institut der TU Clausthal begonnene interdisziplinäre Grabungs- und Forschungskampagne, die sich unter Anleitung der Gesellschaft Unicornu fossile e. V. (GUF e.V. (s. u.) bis heute fortsetzt. Die heute begehbare „Höhle“ ist nur ein Teil des lufterfüllten Dachbodens eines wesentlich großen Höhlensystems, das in seiner Gesamtheit seiner Entdeckung und Erforschung harrt. Erst geophysikalische Ansätze zeigen bereits die Möglichkeiten der modernen Technik in der Erkundung auf. Kaufmann et al. (2010) nutzten geoelektrische und gravimetrische Methoden an der Oberfläche über der Höhle, unter anderem, um weitere versteckte Höhlenteile zu finden. In den Jahren 2011 und 2012 wurden durch das Leibniz-Institut für Angewandte Geophysik (LIAG, Hannover) drei geophysikalische Untersuchungen durchgeführt. Zunächst wurde ein Scherwellen-Seismik-Profil an der Oberfläche über der Höhle gemessen; es befindet sich noch in der Auswertung. Begleitend zu dieser Messung wurde ein genaues LIDAR-Scan des Inneren der Einhornhöhle durchgeführt. Die Auswertung dieser 3D-Darstellung der Höhle ist das Thema der vorliegenden Arbeit. Noch im selben Jahr wurde ein zweites, 200 m langes Scherwellen-Seismik-Profil durch das Innere der Höhle gemessen. Dieses Profil, das zur Zeit ebenfalls noch in der Bearbeitung ist, soll dazu dienen, ein genaueres Abbild der Höhlenbasis zu erstellen.
The development of subsurface voids and cavities in soluble rocks is controlled by the hydrological and chemical processes in the host rock. Water (enriched with carbon dioxide) percolates through fractures and bedding partings of the host rock and removes material from the rock surface. As this enlargement is a highly heterogeneous process, only some fractures and bedding partings become significantly enlarged, evolving towards larger voids and caves. The size of the enlarged voids, often reaching the metre scale, can result in mechanically unstable structures, which, when close to the surface, are prone to collapse and thus are a hazard to infrastructure. We explored two caves in the anhydrite host rock of the Permian Zechstein sequences in northern Germany using geophysical measurements: the Kalkberghöhle close to Bad Segeberg (Hamburg region) and the Jettenhöhle close to Osterode (Harz region). Based on the results of gravity and electrical measurements, we were able to identify the cave voids and to characterize the local geological setting. Using these indirect geophysical observations, we deduced a structural model for both cave sites by numerical modelling. Our structural models were successfully calibrated against the Bouguer gravity data.
In soluble rocks (limestone, dolomite, anhydrite, gypsum, …), fissures and bedding partings can be enlarged with time by both physical and chemical dissolution of the host rock. With time, larger cavities evolve, and a network of cave passages can evolve. If the enlarged cave voids are not too deep under the surface, geophysical measurements can be used to detect, identify and trace these karst structures, e.g.: (i) gravity revealing air- and sediment-filled cave voids through negative Bouguer anomalies, (ii) electrical resistivity imaging (ERI) mapping different infillings of cavities either as high resistivities from air-filled voids or dry soft sediments, or low resistivities from saturated sediments, and (iii) groundwater flow through electrical potential differences (SP) arising from dislocated ionic charges from the walls of the underground flow paths. We have used gravity, ERI, and SP methods both in and above the Unicorn Cave located in the southern Harz Mountains in Germany. The Unicorn Cave is a show cave developed in the Werra dolomite formation of the Permian Zechstein sequence, characterised by large trunk passages interrupted by larger rooms. The overburden of the cave is only around 15 m, and passages are filled with sediments reaching infill thicknesses up to 40 m. We present results from our geophysical surveys above the known cave and its northern and southern extension, and from the cave interior. We identify the cave geometry and its infill from gravity and ERI measurements, predict previously unknown parts of the cave, and subsequently confirm the existence of these new passages through drilling. From the wealth of geophysical data acquired we derive a three-dimensional structural model of the Unicorn Cave and its surrounding, especially the cave infill.
Subsurface investigations in urban areas have to take sealed surfaces and densely built and populated areas into account. The shear-wave reflection seismic system developed at the Leibniz Institute for Applied Geophysics (LIAG) provides a solution for this acquisition technology problem using horizontally polarized shear waves (SH-waves) and noninvasive source and receiver designs.
Karst rocks such as limestone, dolomite, anhydrite, gypsum, or salt can be dissolved physically by water or chemically by water enriched with carbon dioxide. The dissolution is driven by water flowing through the karst aquifer and either occurs along fractures and bedding partings in telogenetic rocks, or within the primary interconnected pore space in eogenetic rocks. The enlargement of either fractures or pores by dissolution creates a large secondary porosity typical of soluble rocks, which is often very heterogenously distributed and results in preferential flow paths in the sub-surface, with cavities as large-scale end members of the sub-surface voids. Once the sub-surface voids enlarged by dissolution grow to a certain size, the overburden rock can become instable and voids and caves can collapse. Depending on the type of overburden, the collapse initiated at depth may propagate towards the surface and finally results at the surface as collapse sinkholes and tiangkengs on the very large scale. We present results from geophysical surveys over existing karst structures based on gravimetric, electrical, and geomagnetical methods. We have chosen two types of sinkholes, solution and collapse sinkholes, to capture and compare the geophysical signals resulting from these karst structures. We compare and discuss our geophysical survey results with simplified theoretical models describing the evolution of the karst structure, and we derive three-dimensional structural models of the current situation for the different locations with our numerical tool PREDICTOR.