Technical Note

Landslides Tommaso Beni · Daniela Boldini · Giovanni Battista Crosta ·

DOI 10.1007/s10346-023-02122-7
Received: 24 February 2023 William Frodella · José Ignacio Gallego · Edoardo Lusini ·
Accepted: 21 July 2023 Claudio Margottini · Daniele Spizzichino
© Springer-Verlag GmbH Germany,
part of Springer Nature 2023
Rock instabilities at the archaeological site
of Dadan (Kingdom of Saudi Arabia)

Abstract This paper provides a summary of the investigations heritage protection for the detection of structural damage, moisture,
carried out for the preliminary assessment of potential geohaz- water seepage, plaster detachments, fractures, cavities, and ledge-
ards affecting the archaeological site of Dadan in the Kingdom of niche systems (Cabrelles et al. 2009; Teza et al. 2015; Frodella et al.
Saudi Arabia. The site is characterized by the presence of Cambro- 2017; Guerin et al. 2019). Several studies have proven that the com-
Ordovician sandstone cliffs (Siq formation), which were quarried bination of the above-mentioned RS-NDT methods integrated with
in ancient times (Dadan/Lihyan kingdom) for building materials. a traditional approach, encompassing field surveys and laboratory
Both the steep quarried portion (Upper Siq) and the gentler under- testing, can yield promising results in the field of rockfall hazard
lying slope (Middle Siq) contain tombs of significant archaeologi- assessment (Margottini 2015; Menegoni et al. 2019; Weidner and
cal value. Landsliding and erosion are the main geomorphological Walton 2021; Beni et al. 2022), discontinuity mapping (Salvini et al.
processes affecting the site, posing risks to the safety of visitors, 2017; Battulwar et al. 2021), geomorphological studies (Sturzenegger
archaeological workers, and the preservation of the site. The pri- and Stead 2009; Gracchi et al. 2021; Weidner et al. 2021), and slope
mary processes affecting the rock cliffs, the underlying slope talus, stability analysis and landslide monitoring (Spizzichino et al. 2016;
and the tombs were identified using geological and geomechanical Balek and Blahůt 2017; Pellicani et al. 2019).
surveys, in addition to various geomatic acquisitions. The mechani- The archaeological area of Dadan in AlUla is one of the most
cal properties of the rock formations and discontinuity sets were important sites in the Kingdom of Saudi Arabia (KSA). In this
identified through laboratory testing and in situ surveying, respec- framework, the protection and conservation of cultural heritage
tively. This information represents the first step in promoting fur- are becoming one of the main goals for the future, not only from an
ther actions for risk mitigation and site management. archaeological perspective but also for social implications (Vision
2022). With the dual objectives of protecting cultural heritage and
Keywords AlUla · Dadan · Kingdom of Saudi Arabia · Slope improving the management of the archaeological park from the
stability · Geohazards · Cultural heritage preservation perspective of a secure touristic fruition, the purpose of this study
was to provide a preliminary assessment of the geomorphological
Introduction processes and potential geohazards affecting both the slopes and
Preservation of archaeological and heritage sites, by safeguard- tombs. The methodological approach adopted in the investigation,
ing archaeologists, tourists, and the structural stability of herit- combining new emerging technologies with traditional field work,
age remains, is mandatory for the fruition and the development of can be successfully applied to sites characterized by similar geologi-
connected economic activities. The stability of rock-carved herit- cal and geomorphological contexts.
age is particularly complex because it involves both natural and
human-related features, and their interaction can be influenced by The Dadan archaeological site in the context of the AlUla region
environmental conditions and their variability on different scales Dadan is a paradigmatic site from various perspectives. It is one of
(Margottini et al. 2016). Rock-carved structures can be frail if they the few places across antiquity for which something as essential as
are located in weak rocks, such as sandstones, which are often char- its proper name and language are known (Hidalgo-Chacón Díez
acterized by high porosity, weak bonding, and fractures. Several 2014). From numerous rock inscriptions, hundreds of personal sto-
multidisciplinary non-invasive monitoring techniques and meth- ries were collected, which give rise to a similar number of questions
odologies have been adopted to assess and inventory instability about the society, history, and politics of the Dadan inhabitants.
phenomena affecting cultural heritage, rupestrian settlements, and The archaeological area is composed of a conglomeration of ele-
rock-carved monuments (Remondino and Campana 2014; Boldini ments that testify to the human presence in the AlUla region over
et al. 2018; Frodella et al. 2021; Pires et al. 2021; Sampietro-Vattuone several centuries. Although there is evidence of human presence
and Peña-Monné 2021; Flora 2022). Remote sensing (RS) meth- at least since the Middle Bronze Age (XIII century BCE), the main
ods, such as terrestrial (TLS) and airborne laser scanning (ALS), civilization period occurred during the Kingdom of Dadan/Lihyan.
unmanned aerial vehicle digital photogrammetry (UAV-DP), and Its capital, from its evolution throughout the last millennium BCE,
satellite imaging, represent the most commonly used techniques to occupied most of the current archaeological site. Dadan was an
obtain high-level spatial data for the studied area. Infrared thermog- important way station on the incense route from Yemen and South-
raphy (IRT) is a versatile non-destructive testing (NDT) technique ern Saudi Arabia to the Mediterranean. Following the Nabataean
widely used in the field of slope instability analysis and cultural occupation of the north-western Arabian Peninsula (Rohmer and

Landslides
 Technical Note
Charloux 2015), the Kingdom of Dadan/Lihyan disappeared as a inscription on one of the tombs associates it with a member of the
political structure around the I century BCE, but the settlement was Minaic community (from Ma’in, today’s Yemen) of Dadan.
maintained. It is still not known if this was permanently occupied,
but the present-day evidence tells us that what can be currently Geological setting
observed was maintained until the XI century ACE. Evidence of a The AlUla region lies in the western Arabian shield that has been
fortified monumental building and its associated settlement, cor- involved in regional tectonics since the Palaeozoic, particularly
responding to the initial five centuries of Islam in Wadi AlUla, was from the rifting of the Red Sea area, which started around 30 Myr
also found. ago (Martinez and Cochran 1988). In the area, a thick Cambrian-
Today, the remains of the city, with its numerous houses and Ordovician clastic succession (500–450 Myr old) is conformably
public spaces, as well as its necropolis, which unfolds in the form of superimposed over the eroded Precambrian basement rocks of the
a multitude of niches and hypogea excavated in sandstone cliffs, are Jibalah Group (Tuvia and Sneh 2002).
clearly recognizable. The tombs were mostly carved in the massive The outcropping clastic succession (Fig. 2), Early to Late Cam-
sandstone layer at the bottom of the slope (Fig. 1). This was con- brian in age, includes from bottom to top: the Siq sandstone (S), a red-
firmed by the quarried profile of the most cemented and reachable dish-brown massive and arkosic arenite, the Quweira sandstone (Q),
sandstone layers. yellow conglomerates, quartzites, and current-bedded sandstones;
Several archaeological missions have been carried out in Dadan only marginally the Saq sandstone outcrops in the study area (Hus-
since its rediscovery at the beginning of the twentieth century seini 1990; Whabi 2014). Both the Siq and Quweira sandstones are
by both Saudi and international universities, as well as research informally subdivided into Lower, Middle, and Upper Units. The
centers. Currently, under the auspices of the Royal Commission Dadan cliff is characterized by the superposition of the Lower
for AlUla (RCU) and its Archaeology and Conservation Research Quweira Formation over the Siq formation (see the geological
Institute, the Kingdoms Institute, an international and multidisci- sketch in Fig. 2a, b).
plinary team led by King Saud University and the CNRS (Rohmer The Middle Siq unit (Sm, up to 70–80 m thick) is limited by
et al. 2020) is engaging in a scientific project to unravel unknowns conformal contact with both the Lower and Upper Siq units (Su;
about Dadan. The so called “Lion Tombs” (al-Aswad Tombs, sixth Fig. 2c) and can be classified as feldspathic litharenite to subli-
century BCE), represent the most iconic feature of the site (Fig. 1d, tharenite. According to Whabi (2014), the unit suggests a progres-
e). They are square-shaped cavities approximately 2 m deep, cut at sive transition from braided fluvial deposition to an estuarine and
different heights, carved by the Lihyanites, and decorated with two successively a tidal-dominated one, with a progressive decrease in
lions each. The tombs are positioned to protect the buried, showing environmental energy and grain size. The Middle Siq unit forms
the importance of their owners and representing an outstanding the lower part of the slope at the base of the rocky cliff, which is
example of Mesopotamian influence (e.g., the Lion of Babylon). An morphologically evident because of the gentler slope angle. This

Fig. 1  Setting of the Dadan cliff: a geographical location; b view of the southern rock slope of Dadan, where famous Lion Tombs were carved
(details of the tombs are shown in c–e). In c, the quarried layer of the Upper Siq unit in which tombs were carved is visible; on the left is a
schematic sketch of the most relevant layers involved in this study: the silty-clayey beds ­SC1 and ­SC2 (see Fig. 2) and the massive sandstone
layers ­MS1, in which tombs were carved, and the upper M ­ S2

Landslides
 Fig. 2  Geological insights of the Dadan archaeological area: a the geologic map projected on the satellite image (from Google Earth©); the
blue lines represent the debris from the quarrying activity, as well as the debris cones and the talus accumulation; “Sm” stands for Middle Siq
sandstone, “Su” for Upper Siq sandstone, “Q” for Quweira sandstone, and “A” for alluvial deposits; b a simplified geological cross section; c the
conformal contact between the upper part of the Sm (represented by the ­SC1 layer defined in Fig. 1) and the lower part of Su (the base of the
­MS1 layer shown in Fig. 1); d cross bedding sandstone at the base of Su; e the massive strata of Su (­ MS1) in which tombs were carved

part of the Siq formation contains a series of small tombs that are and, in particular, of fracture propagation arrest points also known
completely or partially covered by the talus and quarried material. as ribs and locally by hackles. Ribs of growing size occur at the mac-
Indeed, the Su unit base M ­ S1 layer was heavily quarried over the roscale, suggesting extensional propagation through multiple steps
centuries, accommodating the Lion Tombs and approximately 87 or pulses that could be directly related to block wedging activities.
rock-carved tombs in the form of single subhorizontal burial pits These important structures demonstrate that fractures propagated
or burial chambers. The Quweira sandstone lies conformably over from the upper quarry limit, where the wedges were placed, toward
the Siq formation. the base of the slope.
This study mainly focuses on the central part of the 200-m cliff, The upper part of the cliff starting from above the quarry face is
which is characterized by a 20- to 25-m-thick massive sandstone layer­ vertical to subvertical with bedding subhorizontal to gently dipping
(MS1) lying directly over the transition from the Middle Siq to the to the N and with abundant vertical joints and a change towards
Upper Siq unit (characterized by thin silty to clayey beds, such as­ more yellowish tones. This change suggests a passage from the Siq
SC1). The M
­ S1 layer was quarried because of its position and the physi- to the Quweira sandstone formation.
cal and mechanical characteristics that make the rock material suit-
able for construction. Excavation was performed using different tech- Material and methods
niques that could be identified by the presence of natural and induced Most of the geological and geomechanical data described in the
fractures or linear pick marks. Extraction was performed both following sections were collected during two field campaigns con-
parallel and perpendicular to the slope face. Some evidence of the ducted in 2020 (Gallego et al. 2022) and 2021. The methodological
quarrying activities is the presence of holes for placing scaffolding- approach adopted in this study was developed in three steps (see
like structures or for moving along the slope or for anchoring ropes. Fig. 3): (1) data collection, including in situ data collection through
Quarrying was performed by taking advantage of the weak bedding direct measurements (JCS, JRC, discontinuity dip/dip direction),
or lamination planes and open or latent discontinuities. Chisels, remote sensing surveys (IRT, UAV-DP, TLS), and rock sample col-
points, hammers, picks, saws, and wedges were used to cut, excavate, lection; (2) data analysis, involving geotechnical characterization
and open artificial fractures or to remove large blocks by progres- and creation of 2D-3D surface data and maps; and (3) data inter-
sively lowering the upper rock face surface. The downward opening pretation, in order to provide reliable preliminary insights about
of artificial cracks affecting the entire quarry face is suggested by potential criticalities affecting the archaeological area of Dadan, in
the presence of clear plumose (e.g., feather-like) features (Fig. 2e) terms of degradation and instability processes.

Landslides
 Technical Note

Fig. 3  Flowchart of the proposed methodological approach

Data collection Gate. The joint roughness coefficient (JRC) and joint compressive
strength (JCS) were evaluated using a Barton profilometer and an
Geomechanical surveys were carried out, following ISRM (1978a) L-type Schmidt hammer, respectively. In this study, a decametric
recommendations, during the 2020 and 2021 campaigns at the scale of roughness was considered (ISRM 1978a).
foot of the Dadan cliff, where the Upper Siq sandstone outcrop
and most of the tombs were carved (e.g., sectors A, B, and C in Data analysis
Fig. 4). Three horizontal scanlines were acquired at the base of the
­MS1 quarried layer (sectors A, B, and C) (Fig. 4). Additional dis- Geotechnical characterization
continuity orientations were measured in sectors A, B, and C both During the 2020 and 2021 campaigns, seven rock samples were collected
inside and outside the tombs. Sector A (~ 55 m long) consisted of (named 3, 4, 5, SX, DX-A, DX-B, and DX-C; see Appendix 1, Fig. 14) at
five horizontal scanlines (from A1 to A5) that were variously ori- different locations, primarily near the Lion Tombs (Fig. 4). The calcium
ented following the slope face at the Lion Tombs location. Sector carbonate mass percentage was assessed using a Dietrich-Frühling cal-
C (~ 25 m) included two horizontal and perpendicular scanlines cimeter on the rock powder. In addition, a Pascal 240 mercury porosim-
(C1 and C2) unrolled next to the third tomb group from the South eter (produced by Thermo Fisher Scientific), operating up to a maximum

Fig. 4  Aerial view of the study area: the location of surveyed sectors (A, B, C) and the sample collection points are shown; additional disconti-
nuity orientations were measured randomly along the base of the entire cliff. Sector A: Lion Tombs

Landslides
Fig. 5  Results of the laboratory analysis: a longitudinal wave velocity (Vp) versus total porosity (n) is plotted; b, c uniaxial compressive
strength (UCS) and tensile strength (σt) values are plotted against the total porosity. Uniaxial compression was performed both on dry and
water saturated specimens

pressure of 200 MPa, was used to measure the distribution of pore size and the presence of voids, the available specimens were characterized by
and volume in representative samples (ASTM 2018). On a total of 61 speci- very different dimensions and diameter-to-height ratios (see Table 5 in
mens prepared (Fig. 15 in Appendix 1), 56 were used for mechanical tests Appendix 1). A few irregular or undersized samples were used to assess
(Table 5 in Appendix 1). The ultrasonic velocities of longitudinal P waves the physical properties. To compare the results of the mechanical tests
were determined for all specimens undergoing mechanical tests, in many of the undersized samples (d/h values higher than 0.5 and up to 1.006
cases both in dry and saturated conditions, by measuring the transit time according to Table 5), the following empirical correction, valid in the inter-
of a square wave (ISRM 1978b) generated by the Pundit pulser through val 3 > d/h > 1/2, was applied to account for the different diameter-to-height
Panametrics piezoelectric transducers (25 mm in diameter), with a typi- ratios (Obert et al. 1946; Ribacchi et al. 2018):
cal frequency of 1 MHz. Uniaxial compression and Brazilian tests were 𝜎f d
performed applying a loading rate of 0.5 MPa/s (ISRM 1978b). Triaxial = 0.78 + 0.22 ⋅ (1)
𝜎f 0 h
tests were performed using a Hoek cell (ISRM 1983) with a computer-con-
trolled axial load and a hydraulic hand pump for the confining pressure.
where d and h are the diameter and height of the specimen, respec-
22 and 7 specimens were subjected to uniaxial compression tests under
tively; σf is the measured uniaxial compressive strength (UCS) of
dry and water-saturated conditions, respectively, while 14 specimens were
the specimen and σf0 is the UCS of a specimen with d/h = 1. Before
subjected to the Brazilian test (ISRM 1978c) and 3 to the triaxial test (ISRM
the mechanical tests, the specimens were oven dried at 105 °C for
1983) (Fig. 6). Measurements of specimen deformations by strain gauges
the time required to reach a constant mass md determined on a pre-
failed owing to the relatively high glue stiffness compared to that of the
cision balance. The values of the dry density ρdry were estimated by
rock material. Because of the extremely irregular shape of the blocks
measuring the specimen volume using a precision caliper, together
with the solid matrix density ρs, determined by a water pycnometer
(ISRM 1977) on the finely ground materials passing through the No.
200 ASTM sieve. All the specimens were later saturated in deaired
water for about 24 h, and their increase in mass Δm, after wiping
water from their surface, was determined to obtain the saturated
mass ms = md + Δm.
The total porosity n was determined after measuring the mass of
the saturated specimens immersed in water mh. Knowing the three
masses (e.g., dry, saturated, and immersed) and the solid matrix den-
sity, the total porosity was assessed using the following equations:
md
𝜌b = ⋅𝜌
ms − mh w (2)

pb
n= ⋅ 100(%) (3)
ps

Fig. 6  Failure points and Mohr–Coulomb and Hoek–Brown strength where 𝜌w is the water density at 22 °C. This method was chosen
criteria. The red dotted circle highlights values that fall outside the because of the imperfections in the specimen geometry.
failure domain (relative to samples #3 and #4)

Landslides
 Technical Note
Table 1  Technical and optical features of the camera and lens mounted on the DJI Inspire 2 UAV
Camera lens Sensor (mm) Max aperture (mm) Image dimensions (rate | Resolution (dpi) Focal
pixels) length
(mm)

ZENMUSE X7 DL 24 mm f/2.8 23.5 × 15.7 f/2.8 3:2 | 6016 × 4008 72 24

LSASPH

Three‑dimensional data analysis values of each image (the available digital images covered only the
Three geomatic techniques were employed: a laser scanner from an lower slope, where the old quarrying and tombs are located).
aerial survey (ALS), a laser scanner from a ground survey (TLS),
and a drone survey (unmanned aerial vehicle-based digital photo- Two‑dimensional temperature maps
grammetry, UAV-DP). The ALS point clouds cover the entire area The IRT survey was performed in hand-held mode using a FLIR
with a uniform resolution (0.4 × 0.4 m). This is not sufficient for system SC620 thermal camera with a focal plane array (FPA) micro-
structural analysis, especially for subvertical slopes, but could sup- bolometer sensor (FLIR 2009a, b; Table 2). A short-term survey
port the georeferencing of more detailed products and modeling. was performed on December 2021, from 10:30 to 11:30 a.m. (air
The TLS point cloud has centimetric resolution, but is limited to temperature and humidity ranged from 20 to 24.2 C° and 37–22%,
the lower part of the slope, where the archaeological remains are respectively). The duration of the IRT analysis was limited due the
located. Therefore, a drone survey was conducted and several UAV restricted access to the site and the logistics of the field surveys;
acquisition campaigns were carried out in the last 3 years. UAV- nevertheless, the timing proved to be a good compromise for avoid-
DP acquisitions were achieved using a DJI Inspire 2 quadcopter ing shadowing effects on the cliff and for the analysis of the rock
(see Table 1 for the optical and technical features) without ground mass heating phase. The thermograms were acquired at an average
control points (GCPs). distance of 180 m from the west-facing cliff sector encompassing
Four flights were performed with an image overlap higher than the Lion Tombs, allowing for a pixel dimension (instantaneous field
80%. The structure from motion (SfM) process was carried out of view) of 11.7 cm. Data correction, thermal focusing, histogram
using Pix4Dmapper software (Pix4D 2022) on the four datasets equalization, and mosaicking were performed by means of FLIR
including more than 2500 images. A ground sample distance (GSD) Tools + software (FLIR 2009b). The ESRI ArcMap package was used
of approximately 2 cm was achieved due to the different acquisi- to create classified surface temperature maps using the quantile
tion heights. To overcome the absence of GCPs, a semi-automatic function (ESRI 2019).
workflow using the iterative closest point algorithm (ICP) and In the analysis of slope instability phenomena, a qualitative
manual transformations on the CloudCompare open-source soft- analysis of thermographic images based on the detection and inter-
ware (CloudCompare 2022) was applied, in order to scale and align pretation of thermal anomalies can reveal the presence of poten-
the obtained point clouds to the available and georeferenced TLS tial criticalities, such as (i) structural discontinuities (due to the
survey and digital terrain model (DTM) of the study area. This cooling/heating effect of air circulating within open fractures and
georeferencing process resulted in decimetric resolution (due to the to the different thermal transfer capacity of the infilling material
internal camera parameters; Table 1) and a metric precision (due to with respect to the exposed sound rock); (ii) moisture or seepage
the available 0.4 × 0.4 m DEM), which is acceptable given the extent zones (due to the surface cooling caused by water evaporation); (iii)
of the area. To cover as much rock slope as possible, the alignment ledge-niche protruding systems, scarps (due to the slope aspect and
process was repeated three times on each point cloud, which were topographic roughness, which creates shadowing effects during a
then aligned and merged together with the DTM-derived cloud, daily cycle of solar radiation); and (iv) loose coarse debris covering
obtaining a 150 million point single point cloud. The UAV-DP point sectors (due to the cooling effect of air circulating within the voids)
cloud was finally textured with real colors extracted from the RGB (Frodella et al. 2017).

Table 2  Technical features of the adopted thermal camera and IRT survey parameters
Camera Sensor type Spectral Thermal Sensor spatial IR resolution Field of Sensor— Instantaneous
type range (µm) sensitivity resolution (pix) view (°) target field of view (m)
(mK) (mrad) distance
(m)

FLIR Focal plane array 7.5–13 40 0.65 640 × 480 24 × 18 180 1.17
SC620 microbolometer

Landslides
Results whereas the values determined for samples #3 and #4 (red dotted
circle in Fig. 6) fall outside the failure domain.
Geotechnical properties

The samples #3 and #4 exhibited a finer grained matrix, DX-A con- Geostructural analysis
tained an alternation of finer and coarser grained sandstone layers,
and sample SX appeared less cemented with respect to the others. A three-dimensional model was obtained by merging the ALS and
The average values of the physical and mechanical properties of UAV-DP data (Fig. 7), whereas the TLS data were only considered
each sample are summarized in Table 3. for the georeferencing process.
All samples have pore sizes mostly in the range 10–100 µm, with From the stereoplot analysis, three main sets of discontinuities
the DX-A sample presenting two peaks respectively at about 0.4 were identified: bedding planes (BG) and two orthogonal joint sets
µm and 7 µm. The porosity varies significantly among the samples JN1 and JN2 (Fig. 8). Owing to their wide spacing and subhorizontal
(Table 3) from 15.9 to 24.2% for sample #3 and DX-C, respectively. orientation, the bedding planes were under-sampled (only a few
Samples #3 and #4 show similar values of ρdry and n, whereas all the measures of dip/dip direction value) along the achieved horizontal
others are similar to those of DX-C. Considering the ultrasonic data, scanlines (see Fig. 8). The two sets of joints are uniformly distrib-
a marked reduction in the longitudinal wave velocity was observed uted, and JN2 joints are more spaced. Joints in sector A are generally
with the increase in the total porosity (Fig. 5a). Samples #3 and #4 less open and contact surfaces more resistant than those of sector
are characterized by porosity values lower than 20% and Vp values C. The mean orientation of set JN1 is rotated eastward from sector
higher than 3.5 km/s, whereas for larger values of n, the estimated A to sector C by about 20–30°. The geomechanical characteristics
velocity is spread over the 2.25–3.25 km/s interval. As expected, satu- of each set of discontinuities are summarized in Table 4. Detailed
ration is associated with an increase in Vp. Figure 5b and c summa- survey data are also presented in Table 6 (Appendix 2).
rizes the strength values obtained through the uniaxial compression Coltop3D, a full featured software for the structural geologi-
and Brazilian tests, respectively. cal analyses of 3D point clouds (Jaboyedoff et al. 2004; Coltop3D
The UCS ranges from about 11 MPa to over 40 MPa and is halved 2022), was used for the geostructural characterization, in combi-
for saturated specimens. The tensile strength ranges between 1.4 and nation with field work. It is designed for the interactive analysis of
8.2 MPa, with values higher than 3.5 MPa for samples #3 and #4 (see geological structures from high-resolution point clouds and digi-
Table 3). The influence of porosity on the rock strength, especially tal elevation models. For each point cloud, the software automati-
on the tensile strength, is shown in Fig. 5c. The UCS and triaxial cally computes the point normals and produces a point cloud rep-
peak strength values for specimens of the same size (e.g., 38 mm resented by a hue saturation intensity color scale (HSI) through
diameter), are shown in Fig. 6 together with the Mohr–Coulomb which the dip direction and the dip of the normals (poles) are
and Hoek–Brown fitted criteria. represented, respectively, by the hue and saturation values. In the
Considering that the specimens were prepared from the SX, present case study, the “normals” are computed for a radius of 0.5
DX-B, and DX-C samples, the derived strength parameters are m, with a minimum number of points of 3. Structural features
representative of a rock with a porosity of more than 20%. Hence, (e.g., joints, faults, folds) are finally easily detected. After manually
higher values should be anticipated for materials with lower poros- selecting few polygons that are representative of the main discon-
ity, such as samples #3 and #4 (see Table 3). The Brazilian test results tinuity sets, Coltop3D automatically detects the parallel surfaces,
were plotted on the same graph but were not used to estimate the with a certain angular tolerance (− 5° in this case study).Two dis-
parameters of the two failure criteria. The lower values of the ten- tinct automatic geostructural characterizations were performed
sile strength are in agreement with the Hoek–Brown criterion,

Table 3  Average values of the physical and mechanical properties of each rock sample. Bracketed values were obtained under saturated con-
ditions. Empty cells indicate missing data
Sample # CaCO3 (%) ρdry (Mg/m3) ρs (Mg/m3) n (%) Vp (km/s) UCS (MPa) σt (MPa)
3 3.33 2.255 2.679 15.9 3.87 38.3 7.68
4 5.00 2.222 2.678 16.4 3.67 41.2 6.46
5 2.08 1.983 2.694 24.1 2.88 (3.36) 24.8 (20.53) 2.74
SX - 2.019 2.653 23.9 2.77 (2.89) 17.0 (4.59) -
DX-A - 2.036 2.658 21.3 2.75 (3.19) 22.3 2.51
DX-B - 1.994 2.653 23.3 3.06 (3.35) 26.1 (17.48) -
DX-C - 1.992 2.658 24.2 2.88 (3.24) 31.8 -

Landslides
 Technical Note

Fig. 7  Three-dimensional model of the archaeological area of Dadan: a point clouds of the whole rock mass created by merging UAV-DP data
together with the 0.4 × 0.4 m ALS one (less dense point cloud); the red dot indicates the Lion Tombs location sector; b relative mesh (only the
UAV-DP mesh had available RGB data)

by using Coltop3D, having as input the Lidar and the combination two main sets of subvertical joints dipping EW (090–095° and
of Lidar and photogrammetric point clouds. Figures 9 and 10 illus- 270–275°) and NW–SE (145–150° and 325–330°).
trate the comparison of stereonet and rosette plots obtained for
the full cliff with the ALS survey only and by combining the UAV Hydrological analysis
and ALS surveys, respectively. The automatic extraction provided
more than 1,000,000 planes in the first case and about 718,000 To investigate the hydrological surface network of the rock mass,
for the second; the obtained stereoplots confirm the presence of a GIS-based hydrological analysis was conducted using the ESRI

Fig. 8  Stereoplot analysis of discontinuities of the Dadan cliff collected in the field: a the pole projection on an equal area stereonet plot rela-
tive to the scanline C2 of the sector C is shown; three joint sets are recognized JN1, JN2, and the bedding (BG); b pole projection relative to
the scanline A2 of the sector A; the same three sets are present (JN1, JN2, and BG) but the clusters result slightly rotated with respect to a

Landslides
Fig. 9   Coltop3D© semi-automatic discontinuity extraction: a ALS point clouds for the entire cliff and b the stereonet and relative rosette
plots. In upper a, the colors correspond to the aspect of the full cliff face (dip angle and dip direction), expressed in terms of hue value and
saturation (Metzger et al. 2009), while the lower one is only reporting the selected discontinuities, with colors also corresponding to dip angle
and dip direction

ArcMap Hydrology Tools package (ESRI Inc 2019). The available and to highlight the presence of open cracks, rills, and moisture
DTM allowed to establish the flow direction, flow distance, flow sectors (Fig. 11).
accumulation, and stream order functions to highlight the slope-
scale drainage system, infer the surficial water run off trajecto- Weathering‑erosional phenomena
ries, and categorize the stream order. In addition, the watershed
function allowed for the identification of five basins across the Peculiar lithological and geotechnical features of the sandstone
entire area (Fig. 11). The latter was used to visualize how the water (Wangler et al. 2011) were confirmed by laboratory analysis car-
runoff concentration areas could affect the rock-carved cultural ried out on the Upper Siq unit rock samples belonging to the M­ S1
heritage. The erosion channels present along the southern sector layer (low UCS and σt values; Figs. 5 and 6) and by the structural
of the cliff, shaped by ephemeral streams, could induce insta- and morphologic setting of the slope (e.g., dominant disconti-
bility processes and create a preferential path for rock fall and nuity sets and fracture-hydrographic networks). The tests con-
debris flow. Hydrographic characterization was combined with firmed the reduction of UCS under saturated conditions (Fig. 5b).
the IRT surveys to confirm the drainage network of the slope In particular, the dissolution of the sandstone carbonate calcium

Table 4  Mean characteristics of discontinuity sets of Dadan area from scanline surveys in the sectors A and C
Sector Set Spacing (m) Dip (°) Dip direction (°) Length (m) Opening (mm) JRC JCS (MPa)
A BG - 0 220 - - - -
JN1 1.7 90 186 5.2 1.8 11.4 27.9
JN2 2.8 87 269 5.4 1.4 10.3 31.8
C BG - 10 335 - - - -
JN1 1.6 88 218 4.0 2.6 11.0 21.1
JN2 2.8 78 285 6.8 3.4 10.5 19.9

Landslides
 Technical Note

Fig. 10   Coltop3D© semi-automatic discontinuity extraction, a combining the UAV-DP and ALS point clouds for the entire cliff and b the rela-
tive rosette and stereonet plots. In upper a are the two used data bases; in middle a, the colors correspond to the aspect of the full cliff face
(dip angle and dip direction), expressed in terms of hue value and saturation (Metzger et al. 2009), while the lower one is only reporting the
selected discontinuities, with colors also corresponding to dip angle and dip direction

percentage was reported for the SX rock sample (collected in the cold thermal anomalies (down to 12.1–13.2 °C on an average rock
rill area on the left of the Lion Tombs, see Fig. 4), which is char- cliff surface temperature of 17.2 °C) corresponding to persistent
acterized by lower cementing with respect to the others. This fact fractures developing along the entire height of the slope. Warm
was confirmed by the IRT surveys (Fig. 11c), which highlighted thermal anomalies (average 23.4 °C) were instead located in

Landslides
Fig. 11  Evidence of weathering and erosional phenomena acting on the Dadan cliff: a map of the ephemeral drainage network watersheds
(the dashed black rectangle shows the Lion Tombs area); b modeled drainage network map of the Lion Tombs area (dashed red rectangle;
the black arrows mark the three main identified ephemeral creeks); c mosaicked surface temperature map of the Dadan cliff (rainbow color
bar; 1205 × 772 IR resolution) showing rills and moisture sectors (dashed white lines and arrows; d panorama picture showing rill erosion
(dashed blue line) in the slope debris talus at the foot of the cliff; e field evidence of the undermining of the thick sandstones due the ero-
sion of the underlying siltstones in correspondence to the slope break that marks the lithological passage from the lower Sm to the upper Su
units; f detail of the rill slope break

correspondence to the slope debris talus and the eastern sun- the ­MS1 layer (Fig. 11c). Both the pattern and location of the ther-
exposed cliff surfaces (generally connected to JN1 discontinui- mal anomalies suggested their connection to moist sectors, rep-
ties). Cold thermal anomalies (down to 13 °C) were detected in resenting large areas more susceptible to weathering with respect
correspondence with rills cutting the debris talus at the foot of to the rest of the slope. Furthermore, field observations show how

Landslides
 Technical Note
these rills (still bearing moisture, as marked by bushes or small Unstable block assessment
trees; Fig. 11d) tend to erode the finer silty-clayey layers (e.g., S­ C1
in Fig. 1c), which underlie the thick M ­ S1 layer, thereby undermin- The geomechanical field surveys coupled with the analysis of both
ing their stability (Fig. 11e, f). A hydrological GIS-based analysis high-resolution point clouds and IRT data were aimed at identify-
of the hydrographic network not only confirmed that these moist ing potentially unstable blocks. These are located in and above the
sectors and rills represent the ephemeral drainage pattern of the quarried thick sandstone layer of the site (­ MS1 and M
­ S2 in Fig. 1c),
slope (Fig. 11b) but also allowed the identification of five main in correspondence with the Lion Tombs sector (Fig. 12). To quantify
basins in the whole area, each of which conveys the surface run- the potentially unstable block volumes and evaluate the connected
off in three steep and narrow creeks up to the 7th stream order hazard scenarios, the geometrical features (maximum length, maxi-
(Fig. 11a). The creek of watershed no. 3 in Fig. 11a–f, as shown by mum width, outcropping surface, volume, and height of the block
surface temperature maps (Fig. 11c) and field surveys (Fig. 11d–f), above the cliff base) were extracted.
cuts the Dadan cliff, just a few meters east of the Lion Tombs The surface temperature map of the M­ S1 layer (Fig. 12b) allowed
area. Rill erosion also affects the lower Middle Siq formation, to locate cold thermal anomalies (12.2–15.7 °C classes) located in
represented by the gentle foot slope of the cliff. correspondence with open cracks (e.g., block nos. 3–5, 8, 14, and

Fig. 12  Unstable block assessment in the Lion Tombs sector by means of field survey integrated with UAV-DP and IRT data: a unstable
blocks mapped on the high-resolution point cloud (red areas mark the block delimitation); the inferred geometric features of each block are
reported in Table 7 of Appendix 3; in the inset, a frequency distribution plot for the block volumes identified in the Lion Tombs cliff sector of
Dadan (Fig. 12a) is shown. All the data were extracted from the UAV-DP obtained point cloud and were confirmed by the in situ survey and
the comparison with respect to the terrestrial/aerial photos; b detail of the blocks on the classified surface temperature map (“quantile” clas-
sification; 428 × 228 IR resolution)

Landslides
21–28), while warm anomalies (19–26.7 °C class) highlighted niches affecting the archaeological site of Dadan. The weathering processes
and protruding sectors caused by differential erosion of the overly- of sandstone material can accelerate the instability mechanisms
ing fine sand-siltstone layer S­ C2, shown in Fig. 1c (e.g., block nos. active on the Dadan slope (e.g., rock falls, sliding, toppling), which
6–7, 9, 15, 17). Other blocks (e.g., nos. 10–13, 16, 18–20) seem delim- could in turn damage the rock-carved cultural heritage, causing
ited by a combination of the above-mentioned elements. The pat- serious threats for the future fruition of the archaeological site and
tern of these thermal anomalies well integrated field inspections for the visitors’ safety (Tunusluoglu and Zorlu 2009).
(Fig. 8), as well as the visual mapping of the UAV-DP data (Fig. 12a).
All these structural and morphological elements (mainly connected Integration of multidisciplinary methods
to BG and JN1 discontinuities; Figs. 8 and 9) allowed to delimit
thirty potentially unstable blocks, with volumes ranging from 0.2 The preliminary geotechnical characterization of the rock at the
­ 3 (see inset in Fig. 12a and Table 7 in Appendix 3).
to 116 m Dadan site (i.e., longitudinal wave velocity, total porosity, uniaxial
compressive strength, and tensile strength; Table 3 and Appendix 1
Discussion and 2) offer valuable insights for the stability analysis of the cliffs
Several rock-carved cultural heritage sites with rock-carved struc- and of the tombs and to evaluate the potential risks associated with
tures are situated in areas characterized by the presence of “weak rock slope instabilities. Endangered slope sectors and most critical
rocks,” which exhibit unfavorable geotechnical properties (Canuti tombs and cavities or features could be identified to design stabi-
et al. 2009; Margottini and Spizzichino 2014, 2022; Polimeni et al. lization work and to outline areas exposed to instability and their
2019). These weak rocks, while facilitating easy and rapid quarry- possible evolution.
ing, are susceptible to weathering, deterioration, and slope instabil- The integration of geological, UAV-DP, laser scanning, and tra-
ity processes. Consequently, such rock-carved sites across the globe ditional scanline methods has shown great potential for assessing
are frequently subjected to geohazards, posing significant risks to rock mass characteristics and the involved instability mechanisms.
their preservation (Margottini and Spizzichino 2022). Among these These methods complement each other and provide a comprehen-
sites, the infiltration and runoff of surface water are primary factors sive understanding of the geological and structural characteristics
accelerating weathering processes, leading to a reduction in the of the site including those associated to the quarrying activities
unconfined compressive strength (UCS) of the rocks (Weiss et al. that have occurred over time.
2004). Furthermore, water runoff contributes to the partial dissolu- UAV-DP, with its high-resolution aerial imaging and 3D recon-
tion of calcium carbonate in sandstone, worsening the geotechni- struction capabilities, offers a powerful tool for visual inspections
cal parameters of rocks. Lastly, the presence of water can lead to and structural assessments. Detailed 3D models enable precise
moisture expansion (hygric dilatation) caused by the swelling and measurements of the block dimensions (see Appendix 3), distances,
shrinking of clayey minerals (in turn originating from the chemical and angles, facilitating the evaluation of various discontinuity
weathering of the sandstone silicate minerals of the weak rocks; parameters. These quantitative data greatly contribute to a robust
Ruedrich et al. 2011). Thus, the presence of water in sandstones understanding of the stability of rock slopes and the identifica-
can be the main cause of several weathering processes, including tion of potential failure mechanisms and aid in making informed
increased porosity, exfoliation, detachment, arenitization, crust for- decisions regarding preservation and conservation. By integrating
mation, and decohesion (Marszałek et al. 2014). Surface rills and field observations from scanlines with high-resolution imagery and
discontinuity network areas are particularly prone to concentrated detailed 3D models obtained through UAV-DP, a more accurate rep-
erosion caused by heavy rainfall infiltration and seepage. resentation of unstable block sizes can be achieved supporting the
The Dadan cliff, throughout its existence since the I millennium simulation of rockfalls.
BCE, has experienced weathering, natural instabilities, and anthro-
pogenic activities, continuously shaping the morphological features Main instability processes affecting Dadan’s cliff
and altering the stability conditions of the rock slope housing the
carved tombs, primarily within the M ­ S1 layer. In this framework, According to the multidisciplinary investigation of the area
a multidisciplinary approach that combines archaeological and described in the previous paragraphs, the following processes were
geological expertise, traditional field surveys, laboratory analysis, recognized as potentially relevant in the evolution of the rocky cliff:
remote sensing techniques, and non-destructive methods becomes
indispensable for effective cultural heritage surveying and monitor-
1. Potential planar and wedge failures, toppling, and free fall
ing. This comprehensive approach allows for data collection from
(from small to large dimensions) distributed along the whole
hazardous or inaccessible areas while ensuring the safety of both
rocky cliff, both in the Upper Siq sandstone (Fig. 13a) and in
personnel and the analyzed monument (only fallen rock blocks were
the topmost Quweira sandstone
collected for laboratory analysis). Over the past decade, this kind
2. Free falls reaching the lower part of the cliff and in a few cases
of approach has proven its efficacy in assessing potential geohaz-
beyond the talus limit (Fig. 13b)
ards in similar geological contexts characterized by weak, weath-
3. Secondary falls of loose blocks and debris from the morpho-
ered, and fractured rocks along subvertical slopes, as evidenced by
logical terraces above the quarried layer (Fig. 13c)
a number of successful studies (Gigli et al. 2014; Margottini et al.
4. Rill erosion in the Middle Siq formation, represented by the
2016; Themistocleous et al. 2016; Margottini and Spizzichino 2017;
gentle slope partially covered by collapsed blocks and debris
Boldini et al. 2018; Frodella et al. 2021). The adopted methodological
from the quarry, during heavy rainfall (Fig. 13d)
approach (Fig. 3) allowed to obtain preliminary results and data to
assess weathering-erosional phenomena as well as rock instabilities

Landslides
 Technical Note

Fig. 13  Potential instabilities affecting the rock slope and impact on cultural heritage: a potential planar and wedge failure, block toppling,
flexural toppling, and free fall affecting the slope; b example of past morphological processes that reached the talus limit; c secondary falls of
loose blocks and debris from the morphological terraces above the quarry; d rill erosion in the Middle Siq formation; e unstable block prone
to free fall above one of the Lion Tombs, on the touristic path; f blocks falling from the upper part of the cliff and potentially affecting the visi-
tors; g rill erosion damaging the tombs excavated in the Middle Siq formation; h example of a very endangered tomb affected by roof and
façade collapse

5. Weathering-erosional phenomena influenced by layering, run- ological park. Damage to the tombs located in the Middle Siq
off, and seeping water and changes in moisture content formation can also derive from these phenomena (Fig. 13f).
3. Rill erosion may damage the tombs excavated in the Middle
These processes can affect the exposed elements at the Dadan Siq formation (Fig. 13g).
site. In particular, it can be noticed that:
Furthermore, well-developed tension cracks located along the
1. The collapse of small/medium sized blocks from the quarry tomb façade, or even within the excavations, characterized by open-
walls may damage the tombs carved at the base of the cliff, ings up to a few tens of centimeters were observed (Fig. 13h).
either by direct impact or by directly cutting away from the In situ observations suggest that tombs were excavated both
tomb facades (Fig. 13e). before and after the quarrying of the Upper Siq layers. In fact, hol-
2. Blocks falling from the upper part of the cliff (Su unit) or lows and pits within the Middle Siq unit were found to be covered
removed by secondary falls may endanger visitors or operators by both talus and quarry debris (Fig. 13g). In other cases, as for
at the toe of the slope or along the white road inside the archae- the Lion Tombs, inscriptions and carvings suggest an age of the

Landslides
tombs successive to quarrying (Fig. 13e). These differences in the traditional survey and laboratory analyses on collected samples,
time and sequence of excavation can support further analyses and considerable geological, geomorphological, geotechnical, hydro-
the assessment of the rock fall frequency and of the progressive logical, and geomechanical information was collected. This expe-
rock degradation by wetting–drying cycles. Unfortunately, it is not rience demonstrates how essential a multidisciplinary approach is
easily determined if the tomb façades were quarried or affected to develop reliable data-driven field geohazard investigations to
by an instability within the quarry face. Moreover, beyond stress support future mitigation and monitoring plans. The results of the
concentration, water infiltration, and runoff from the upper cliff performed activities are the following:
during rare intense rainfall, the wind action and the sharp thermal
excursions may be listed among the predisposing factors that may • The geotechnical characterization of the collected rock samples
trigger rock falls and further damage. highlighted the range values of the physical and mechanical
properties of the Upper Siq Unit showing the relevant decrease
from dry to saturated conditions
Safe touristic fruition • Direct geostructural measurements at the foot of the slope iden-
tified the main discontinuity sets affecting the whole rock slope
The fruition of a natural or cultural heritage site requires a guaran- and revealed the geological features and failure mechanisms
tee of the safety conditions for the heritage and the structures, as controlling block detachments and tomb instabilities
well as for the personnel and visitors. With KSA being a developing • The contactless detection of discontinuities using point clouds and
country from a touristic point of view, the goal of this study was specific software confirmed that these techniques support in situ sur-
to assess and quantify the extent and evolution of the potential veys to extend data to slope sectors that are otherwise inaccessible
weathering-erosional phenomena and instability processes act- • The unstable block assessment based on the obtained data and
ing on the Dadan cliff. This should represent a starting point for expert judgment allowed the definition of involved volumes to
further risk mitigation strategies and monitoring plans. Due to develop rockfall modeling and the design of mitigation and
the high relief topography, the runout of collapsing blocks can be stabilization works
extremely dangerous for visitors when located close to the cliff • The infrared thermography technique can integrate field and
itself. During rare but intense rainfall events, flash floods, rock photogrammetric analysis for the detection of critical sectors
falls, and debris flows could represent serious hazards for the of the cliff with respect to weathering, erosion, and instability
Dadan archaeological remains and related structures (e.g., the • A hydrological GIS-based analysis can provide the identification
entrance building, the pathway at the base of the cliff and the road of most critical streams that can concentrate runoff, therefore
at the talus toe), but also for the personnel and tourists. Consider- enhancing weathering and erosion up to instability in the rock cliff
ing climate change, despite the latitude, the increasing occurrence
of intense rainfall events must be considered as a factor of growing These collective findings offer valuable insights for informed
importance for the erosion and weathering processes of the cliff. decision-making and preservation efforts related to rock-carved
This risk requires the urgent development of a proper conserva- cultural heritage sites. The multidisciplinary approach provided a
tion plan focusing on both the mitigation of potential hazards and preliminary understanding of the intensity and extension of insta-
reduction of exposure and vulnerability of visitors. The mitiga- bility processes affecting the site with the aim of implementing a
tion actions should include all measures suitable to stabilize the mandatory mitigation and monitoring plan for a safe tourist frui-
potential collapsing blocks (using the geotechnical and geome- tion of the archaeological site of Dadan.
chanical data obtained from this preliminary study) and limit the
runout of falling boulders. Particular attention should be paid to Acknowledgements
the tombs located in the intermediate clayey Siq formation, where The reported activities were carried out thanks to the collabora-
rill erosion during heavy rainfall may damage or even destroy tion between the UNESCO Chair on Prevention and Sustainable
the archaeological evidence. Beyond these structural measures, a Management of Geo-Hydrological Hazards of the University of
proper management of visitors, cars, and buses passing through Florence, the Italian Institute for Environmental Protection and
the site is required to minimize the runout of falling blocks and Research (ISPRA), the Sapienza University of Rome, and the Uni-
the detachment of loose blocks at the top of the quarried area or versity of Milano Bicocca, coordinated by the Royal Commission
even from the tombs. Non-structural measures should include a for AlUla (RCU) and the French Agency for AlUla Development
careful planning of touristic paths outside of most endangered (AFALULA), in the framework of the International Programme
areas. Finally, in consideration of the steep morphology, a proper on Landslides IPL-259 project “Landslide Risk assessment in
monitoring network combining remote survey (e.g., ground-based AlUla Archaeological sites–Kingdom of Saudi Arabia.” ALS data
radar interferometry) and local gauges measuring deformation were provided by the Royal Commission for AlUla. We deeply
and tilting of potential unstable blocks should be implemented thank Factum Foundation for making available the TLS survey
and complemented by a weather station. and Drone photos. LAGIRN laboratory of the University of Bolo-
gna, in the person of Dr. Fausto Peddis and Dr. Barbara Ravaglia,
Conclusions is highly acknowledged for the activity of laboratory testing.
Two field and remote surveying campaigns were conducted in the Jhaed Kamuzaman Abul provided also his help in this stage of
archaeological site of Dadan. By combining close-range remote the research.
sensing, non-destructive testing techniques together with in situ

Landslides
 Technical Note
Author contribution Daniele Spizzichino; writing review and editing: Tommaso Beni,
Conceptualization and methodology: Daniela Boldini, Giovanni Daniela Boldini, Giovanni Battista Crosta, William Frodella,
Battista Crosta, Claudio Margottini, and Daniele Spizzichino; José Ignacio Gallego, Edoardo Lusini, Claudio Margottini, and
formal analysis and investigation: Tommaso Beni, Daniela Daniele Spizzichino; Supervision: José Ignacio Gallego, Claudio
Boldini, Giovanni Battista Crosta, William Frodella, José Ignacio Margottini, and Daniele Spizzichino.
Gallego, Edoardo Lusini, Claudio Margottini, and Daniele
Spizzichino; writing original draft preparation: Tommaso Beni,
Daniela Boldini, Giovanni Battista Crosta, William Frodella, Appendix 1
José Ignacio Gallego, Edoardo Lusini, Claudio Margottini, and

Fig. 14  The seven rock samples collected in the archaeological site of Dadan during 2020 and 2021 campaigns

Fig. 15  Core and disk specimens prepared from the Upper Siq unit (Su) samples for the mechanical tests: a from 2020 campaign and b 2021
relatively. See Table 5 for sample dimensions

Landslides
Table 5  Geometrical characteristics of the specimens used for the mechanical tests. In the Test column, the following acronyms were used:
VP P wave velocities, VP_SAT P wave velocities in saturated conditions, UCS uniaxial compressive test, UCS_SAT uniaxial compressive test in
saturated conditions, BRA indirect tensile test, TRIAX triaxial compressive test. In the Note column, the following acronyms were used: DL dou-
ble lithology, EL eroded level, CS chopped specimen
Sample Specimen h (mm) d (mm) d/h Test Note
3 A 50.40 24.89 0.494 VP, UCS
3 B 45.91 24.80 0.540 VP, UCS
3 C 45.86 25.02 0.546 VP, UCS
3 D 24.35 49.87 2.048 VP, BRA
3 E 25.35 49.83 1.97 VP, BRA
4 A 36.95 25.01 0.677 VP, UCS
4 B 25.50 49.81 1.953 VP, BRA
4 C 24.59 49.85 2.027 VP, BRA
5 A 49.39 24.90 0.504 VP, UCS
5 B 49.63 24.78 0.499 VP, UCS
5 C 46.57 24.73 0.531 VP, UCS
5 D 47.28 24.77 0.524 VP, UCS
5 E 47.74 24.80 0.5195 VP_SAT, UCS_SAT
5 F 49.54 24.85 0.502 VP_SAT, UCS_SAT
5 G 49.31 24.74 0.502 VP_SAT, UCS_SAT
5 H 49.88 24.70 0.495 VP
5 I 50.26 24.78 0.493 VP_SAT, UCS_SAT
5 L 51.02 24.86 0.487 VP_SAT, UCS_SAT
5 M 48.73 24.85 0.510 VP
5 N 50.22 24.90 0.496 VP
5 O 24.91 49.87 2.002 VP, BRA
5 P 25.32 49.87 1.970 VP, BRA
5 Q 24.99 49.81 1.993 VP, BRA
Sx 1 75.33 37.90 0.503 VP, VP_SAT, UCS
Sx 2 70.02 37.90 0.541 VP, VP_SAT, UCS
Sx 3 68.16 37.90 0.556 VP, VP_SAT, UCS
Sx 4 63.09 37.90 0.601 VP, VP_SAT, UCS
Sx 5 49.01 24.60 0.502 VP, VP_SAT, UCS
Sx 6 24.63 24.60 0.999 VP, VP_SAT, UCS
Sx 7 26.20 24.60 0.939 VP, VP_SAT, UCS
Sx 8 24.54 24.60 1.002 VP, VP_SAT, UCS
Sx 9 38.57 37.80 0.980 VP, VP_SAT, UCS
Sx 10 48.85 24.60 0.504 VP, VP_SAT, UCS_SAT
Dx-A 1 48.69 24.60 0.505 VP, VP_SAT, UCS DL, 1/3 fine grain
Dx-A 2 47.29 24.60 0.520 VP, VP_SAT, UCS DL, 2/5 fine grain

Landslides
 Technical Note
Table 5  (continued)
Sample Specimen h (mm) d (mm) d/h Test Note

Dx-A 3 48.97 24.60 0.502 VP, VP_SAT, UCS DL, 1/4 fine grain
Dx-A 4 23.89 24.60 1.030 VP, VP_SAT, UCS DL, 2/5 fine grain
Dx-A 5 25.11 24.60 0.980 VP, VP_SAT, UCS
Dx-A 6 24.58 24.60 1.000 VP, VP_SAT, UCS
Dx-A 7 24.75 49.79 2.012 VP, VP_SAT, BRA DL, high % of fine
grain
Dx-A 8 24.60 49.79 2.024 VP, VP_SAT, BRA Coarse grain, EL
Dx-A 9 26.17 49.79 1.903 VP, VP_SAT, BRA DL, < 50% of fine grain
Dx-A 10 25.33 49.79 1.966 VP, VP_SAT, BRA DL, high % of fine
grain
Dx-A 11 25.44 49.79 1.957 VP, VP_SAT, BRA Coarse grain
Dx-A 12 24.12 49.79 2.064 VP, VP_SAT, BRA Coarse grain
Dx-A 13 25.72 49.79 1.936 VP, VP_SAT, BRA Coarse grain
Dx-B 1 75.83 37.90 0.500 VP, VP_SAT, TRIAX CS
Dx-B 2 76.00 37.90 0.499 VP, VP_SAT, TRIAX
Dx-B 3 75.41 37.90 0.503 VP, VP_SAT, TRIAX
Dx-B 4 47.91 24.60 0.513 VP, VP_SAT, UCS Slightly CS
Dx-B 5 48.18 24.60 0.511 VP, VP_SAT, UCS_SAT CS
Dx-C 1 37.78 38.00 1.006 VP, VP_SAT, UCS
Dx-C 2 38.33 38.00 0.991 VP, VP_SAT, UCS
Dx-C 3 38.83 24.60 0.634 VP, VP_SAT, UCS
Dx-C 4 38.47 24.60 0.640 VP, VP_SAT, UCS
Dx-C 5 38.67 24.60 0.636 VP, VP_SAT, UCS

Landslides
Appendix 2

Table 6  Data obtained from the geomechanical scanlines

Line N Progressive (m) Dip (°) Dip direction (°) Opening (mm) Length (m) JRC JCS (MPa)
A1 1 0.00 80 170 2 6

A1 2 1.70 84 191 3 6

A1 3 4.70 68 2 3 5

A1 4 5.05 89 19 5
A1 5 7.79 80 188 3 3.5

A1 6 11.30 88 67 2 1

A1 7 11.60 64 190 2 6

A1 8 12.50 90 1 2 2.5

A1 9 13.05 86 188 1 3 9
A1 10 13.35 66 190 1 2

A1 11 14.60 60 185 5 9 15
A1 12 16.50 85 10 1 2 13
A2 1 2.10 80 190 1 1.7 19
A2 2 5.32 83 189 2.5 4.5 11
A2 3 6.71 90 347 0.5 1.5

A2 4 9.00 90 266 3 6 15 33.6

A2 5 11.20 86 191 2.5 11

A2 6 12.70 85 280 6 7 9
A2 7 12.90 76 53 14 11
A2 8 13.30 79 4 2 7
A3 1 0.22 85 230 2 29.1

A3 2 0.50 87 272 2
A3 3 1.88 82 175 0 2

A3 4 3.10 75 30 0 1.7 11
A3 5 3.60 89 310 0 1.7 9
A3 6 4.00 90 277 0 2 7 28.8

A3 7 4.00 80 0 1 4 7
A3 8 5.10 88 276 2 3.5 17
A3 9 5.30 90 252 3.4 9
A3 10 6.65 87 270 1 2.4 11
A3 11 8.60 81 61 4 3.4 11 31.5

A3 12 8.90 71 351 2 6 7
A3 13 9.20 78 346 0 2.1

A3 14 10.40 85 286 2 6 9 33.0

Landslides
 Technical Note
Table 6  (continued)
Line N Progressive (m) Dip (°) Dip direction (°) Opening (mm) Length (m) JRC JCS (MPa)
A3 15 10.70 90 289 0 1 7
A3 16 10.85 90 289 0 4 17

A3 17 11.10 85 89 0 0.5

A3 18 11.90 0 220 9
A3 19 12.10 72 148 0.5 3.5 11
A4 1 0.00 85 280 2 20 7 35.8

A4 2 0.06 77 20 1 3.5

A4 3 1.72 89 2 2 9 9
A4 4 1.84 - - - -

A4 5 1.96 - - - -

A4 6 2.06 - - - -

A4 7 2.23 - - - -

A4 8 2.39 84 12 4 20 17
A4 9 5.75 - - 1 6

A4 10 6.00 82 187 1 4 9 27.9

A4 11 6.40 85 2 0.5 20

A4 12 6.50 90 15 0 2

A4 13 7.53 70 20 3 5

A5 1 0.00 88 260 1 8 7 28.3

A5 2 0.50 88 260 0.5 5

A5 3 1.73 66 260 0 2

A5 4 1.92 81 268 0 3 7
A5 5 2.60 83 246 3 7.5 11 31.2

A5 6 3.70 82 119 11 28.5

A5 7 4.70 68 222 1 1.8

A5 8 4.70 89 83 2 6

A5 9 5.50 66 230 2 7 25.1

A5 10 5.60 86 273 1 15

C1 1 0.40 78 215 0 6

C1 2 0.62 88 215 0 0.5

C1 3 0.75 58 305 0 2

C1 4 1.35 84 20 3 6

C1 5 2.94 90 12 2
C1 6 3.30 88 270 2
C1 7 5.00 86 298 2 3.5

C1 8 7.10 72 225 15

C1 9 12.30 85 210 1 2

Landslides
Table 6  (continued)
Line N Progressive (m) Dip (°) Dip direction (°) Opening (mm) Length (m) JRC JCS (MPa)
C1 10 12.70 84 198 5 3

C1 11 13.00 84 274 15

C1 12 14.00 80 235 1 6

C1 13 14.50 90 297 15

C1 14 15.50 88 188 5 2

C1 15 16.80 73 267 10 5

C1 16 18.80 87 231 3 1

C1 17 19.50 87 232 3 7

C1 18 19.80 87 230 10 7

C1 19 20.00 70 350 5 4

C2 1 0.85 78 37 1 0.4

C2 2 1.00 88 54 1 1

C2 3 1.20 78 60 2 1

C2 4 1.50 84 299 5 10

C2 5 1.90 90 220 1.5

C2 6 3.10 87 280 2 2

C2 7 3.60 54 275 3 2

C2 8 4.80 10 335

Landslides
 Technical Note
Appendix 3

Table 7  Geometrical features of the unstable blocks identified in the Lion Tombs cliff sector of Dadan (Fig. 13a). All the data were extracted
from the UAV-DP obtained point cloud and were confirmed by the in situ survey and the comparison with respect to the taken terrestrial/
aerial photos
Block ID Block base height from the ground (m) Max length (m) Max width (m) Max thickness (m) Surface ­(m2) Volume ­(m3)
01 15 1.8 1.6 0.8 3.5 2.6

02 18 1.6 1.2 0.7 2.7 3.7

03 6 7.1 6.9 1 47.1 30

04 4 5 2.5 0.4 11.5 4.7

05 9 3.9 0.7 0.7 2.1 1

06 17 1.9 1.8 0.8 4.2 4.5

07 16 2.2 1.9 1.2 5.3 5.5

08 8 7.1 5.3 1.5 29.7 46.1

09 5 1.4 0.8 0.2 0.9 0.4

10 4 1 0.8 0.4 0.9 0.2

11 5 6.2 1.8 0.4 7.7 2.3

12 2 4.2 0.6 0.3 2.2 0.5

13 4 7 1.8 2.5 21 21.6

14 3 2.5 0.9 0.4 3.7 1.3

15 16 8.8 3.9 2.5 14.5 116

16 7 6.6 5.1 2.1 32.6 59.9

17 16 2.4 1.4 0.5 4.2 2.7

18 9 5.9 1.4 0.8 8.3 6.4

19 8 6.8 0.8 0.7 4 6

20 9 5.3 1.5 0.8 4.4 8.5

21 16 1.7 0.5 0.9 1.6 0.7

22 16 1.6 0.7 0.8 2.2 2.6

23 19 3.9 1.9 2 9.5 13

24 19 2.6 1.3 1.4 5.7 5.6

25 16 1.3 0.8 0.3 1.3 0.2

26 16 2.3 1.2 0.2 3.2 1

27 17 1.8 0.9 0.8 2 1.2

28 17 0.8 0.6 0.3 0.5 0.2

29 17 5.5 4.4 3.6 30 83

30 17 8.2 4.3 2.5 33 65

Landslides
Funding Gracchi T, Rossi G, Tacconi Stefanelli C et al (2021) Tracking the evolu-
The field work was funded by the French Agency for AlUla Devel- tion of riverbed morphology on the basis of UAV photogrammetry.
Remote Sens 13:829. https://​doi.​org/​10.​3390/​rs130​40829
opment (AFALULA), on behalf of the Royal Commission for AlUla Guerin A, Jaboyedoff M, Collins BD et al (2019) Detection of rock bridges
(RCU). by infrared thermal imaging and modeling. Sci Rep 9:13138. https://​doi.​
org/​10.​1038/​s41598-​019-​49336-1
Declarations Hidalgo-Chacón Díez M del C (2014) Place names in the Dadanitic
inscriptions of al-ʿUḏayb. Adumātu 15–30
Competing interests The authors declare no competing interests. Husseini MI (1990) The Cambro-Ordovician Arabian and adjoining plates:
a glacio-eustatic model. J Pet Geol 13:267–288. https://​doi.​org/​10.​
1111/j.​1747-​5457.​1990.​tb008​47.x
ISRM (1978a) International society for rock mechanics commission on
References standardization of laboratory and field tests. Int J Rock Mech Min Sci
Geomech Abstr 15:319–368. https://​doi.​org/​10.​1016/​0148-​9062(78)​
ASTM (2018) Standard test method for determination of pore volume 91472-9
and pore volume distribution of soil and rock by mercury intrusion ISRM (1978b) Suggested methods for determining sound velocity. Int J
porosimetry Rock Mech Min Sci Geomech Abstr 15:53–58. https://​doi.​org/​10.​1016/​
Balek J, Blahůt J (2017) A critical evaluation of the use of an inexpensive 0148-​9062(78)​91678-9
camera mounted on a recreational unmanned aerial vehicle as a tool ISRM (1983) Suggested methods for determining the strength of rock
for landslide research. Landslides 14:1217–1224. https://​doi.​org/​10.​ materials in triaxial compression: revised version. Int J Rock Mech
1007/​s10346-​016-​0782-7 Min Sci Geomech Abstr 20:285–290. https://​doi.​org/​10.​1016/​0148-​
Battulwar R, Zare-Naghadehi M, Emami E, Sattarvand J (2021) A state- 9062(83)​90598-3
of-the-art review of automated extraction of rock mass discontinuity ISRM (1978c) Suggested methods for determining tensile strength of
characteristics using three-dimensional surface models. J Rock Mech rock materials. Int J Rock Mech Min Sci Geomech Abstr 15:99–103.
Geotech Eng 13:920–936. https://​doi.​org/​10.​1016/j.​jrmge.​2021.​01.​008 https://​doi.​org/​10.​1016/​0148-​9062(78)​90003-7
Beni T, Gigli G, Lombardi L et al (2022) Route stability index (RSI): an index ISRM (1977) Suggested methods for determining water content, poros-
for the assessment of rockfall-related hazards in rock slopes equipped ity, density, absorption and related properties and swelling. Int J Rock
for sport climbing. Geoheritage 14:80. https://​d oi.​o rg/​1 0.​1 007/​ Mech Min Sci Geomech Abstr 16:141–156
s12371-​022-​00715-7 Jaboyedoff M, Baillifard F, Couture R, Locat J, Locat P, Rouiller JD (2004)
Boldini D, Guido GL, Margottini C, Spizzichino D (2018) Stability analy- New insight of geomorphology and landslide prone area detection
sis of a large-volume block in the historical rock-cut city of Vardzia using DEM. In: 9th International Symposium on Landslides, W.A. Lacerda,
(Georgia). Rock Mech Rock Eng 51:341–349. https://​doi.​org/​10.​1007/​ M. Ehrlich, A.B. Fontoura and A. Sayo (eds.), Balkema 199–205
s00603-​017-​1299-7 Margottini C (2015) Engineering geology in shaping and preserving
Cabrelles M, Galcerá S, Navarro S et al (2009) Integration of 3D laser the historic urban landscapes and cultural heritage: achievements
scanning, photogrammetry and thermography to record architectural in UNESCO world heritage sites. In: Lollino G, Giordan D, Marunteanu
monuments. In: Proceedings of the 22nd International CIPA Sympo- C et al (eds) Engineering geology for society and territory -, vol 8.
sium. Kyoto, Japan 11–15 Springer International Publishing, Cham, pp 1–28
Canuti P, Margottini C, Fanti R, Bromhead E (2009) Cultural heritage and Margottini C, Gigli G, Ruther H, Spizzichino D (2016) Advances in geo-
landslides: research for risk prevention and conservation. In: Sassa K, technical investigations and monitoring in rupestrian settlements
Canuti P, (eds.) Landslides: disaster risk reduction. Springer Verlag, inscribed in the UNESCO’s world heritage list. Procedia Earth Planet
ISBN: 978–3–540–69966–8 Sci 16:35–51. https://​doi.​org/​10.​1016/j.​proeps.​2016.​10.​005
CloudCompare (2022) CloudCompare. https://​www.​danie​lgm.​net/​cc/. Margottini C, Spizzichino D (2014) How geology shapes human settle-
Accessed 27 Apr 2022 ments. In Bandarin F., van Oers R. (eds.) Reconnecting the city. The
Coltop3D (2022) https://​www.​terra​num.​ch/​produ​its/​colto​p3d/. Accessed historic urban landscape approach and the future of urban heritage.
15 Oct 2022 Chichester: Wiley Blackwell
ESRI Inc (2019) ArcMap 10.6 datasheet. https://​www.​esri.​com/​conte​nt/​dam/​ Margottini C, Spizzichino D (2017) Historical accesses to UNESCO cul-
esris​ites/​en-​us/​media/​pdf/​produ​ct/​deskt​op/​ArcGIS-​10.6-​Deskt​op-​Arc- tural heritages: engineering geology for the sustainable conserva-
Map-​Funct​ional​ity-​Matrix.​pdf tion of Petra Siq. Innov Infrastruct Solut 2:25. https://​doi.​org/​10.​1007/​
FLIR (2009a) FLIR ThermaCAM SC620 technical specifications. www.​flir.​ s41062-​017-​0074-7
com/​cs/​emea/​en/​view/?​id=​41965 Margottini C, Spizzichino D (2022) Weak rocks in the Mediterranean
FLIR (2009b) FLIR tools + datasheet. https://​www.​infra​redca​meraw​areho​use.​ region and surroundings: threats and mitigation strategies for
com/​conte​nt/​FLIR%​20Dat​ashee​ts/​FLIR%​20Too​lsPlus%​20Dat​asheet.​pdf selected rock-cut heritage sites. Eng Geol 297:106511. https://​doi.​
Flora A (2022) Taking care of heritage, a challenge for geotechnical org/​10.​1016/j.​enggeo.​2021.​106511
engineers. In: Proc. Geotechnical engineering for the preservation of Marszałek M, Alexandrowicz Z, Rzepa G (2014) Composition of weath-
monuments and historic sites III. Naples, Italy 19–54 ering crusts on sandstones from natural outcrops and architectonic
Frodella W, Gigli G, Morelli S et al (2017) Landslide mapping and char- elements in an urban environment. Environ Sci Pollut Res 21:14023–
acterization through infrared thermography (IRT): suggestions for a 14036. https://​doi.​org/​10.​1007/​s11356-​014-​3312-y
methodological approach from some case studies. Remote Sensing Martinez F, Cochran JR (1988) Structure and tectonics of the northern
9(12):1281 Red Sea: catching a continental margin between rifting and drifting.
Frodella W, Elashvili M, Spizzichino D et al (2021) Applying close range Tectonophysics 150:1–31. https://​d oi.​o rg/​1 0.​1 016/​0 040-​1 951(88)​
non-destructive techniques for the detection of conservation prob- 90293-4
lems in rock-carved cultural heritage sites. Remote Sens 13:1040. Menegoni N, Giordan D, Perotti C, Tannant DD (2019) Detection and
https://​doi.​org/​10.​3390/​rs130​51040 geometric characterization of rock mass discontinuities using a 3D
Gallego JI, Margottini C, Spizzichino D et al (2022) Geomorphological high-resolution digital outcrop model generated from RPAS imagery
processes and rock slope instabilities affecting the AlUla archaeo- – Ormea rock slope, Italy. Eng Geol 252:145–163. https://​doi.​org/​10.​
logical region. In: Geotechnical engineering for the preservation of 1016/j.​enggeo.​2019.​02.​028
monuments and historic sites III, 1st edn. CRC Press, London 456–466 Metzger R, Jaboyedoff M, Oppikofer T (2009) Coltop3D: a new software
Gigli G, Frodella W, Garfagnoli F et al (2014) 3-D geomechanical rock mass for structural analysis with high resolution 3D point clouds and DEM.
characterization for the evaluation of rockslide susceptibility scenarios. In: 2009 CSPG CSEG CWLS Convention. Calgary, Alberta, Canada
Landslides 11:131–140. https://​doi.​org/​10.​1007/​s10346-​013-​0424-2

Landslides
 Technical Note
Obert L, Windes SL, Duvall WI (1946) Standardized tests for determining Ioannides M, Fink E, Moropoulou A, et al. (eds) Digital heritage. Pro-
the physical properties of mines rocks. US Bureau of Mines Report of gress in cultural heritage: documentation, preservation, and protec-
Investigations, No:3891 tion. Springer International Publishing, Cham 91–98
Pellicani R, Argentiero I, Manzari P et al (2019) UAV and airborne LiDAR Tunusluoglu MC, Zorlu K (2009) Rockfall hazard assessment in a cultural
data for interpreting kinematic evolution of landslide movements: and natural heritage (Ortahisar Castle, Cappadocia, Turkey). Environ
the case study of the Montescaglioso landslide (Southern Italy). Geo- Geol 56:963–972. https://​doi.​org/​10.​1007/​s00254-​008-​1198-z
sciences 9:248. https://​doi.​org/​10.​3390/​geosc​ience​s9060​248 Tuvia W, Sneh A (2002) Sedimentology and paleogeography of the Late
Pires F, Bilotta E, Flora A, Lourenço PB (2021) Assessment of excavated Precambrian-Early Cambrian arkosic and conglomeratic facies in the
tunnels stability in the Maya archeological area of Copán. Honduras northern margins of the Arabo-Nubian Shield. Geol Surv Isr Bull 87.
Int Journal of Architectural Heritage 17(3):431–449. https://​doi.​org/​ Vision (2022) Vision 2030. In: Vis. 2030. https://​www.​visio​n2030.​gov.​sa
10.​1080/​15583​058.​2021.​19317​30 Wangler TP, Stratulat A, Duffus P et al (2011) Flaw propagation and
Pix4D (2022) Pix4D. https://​w ww.​p ix4d.​c om/​p rodu​c t/​p ix4d​m apper-​ buckling in clay-bearing sandstones. Environ Earth Sci 63:1565–1572.
photo​gramm​etry-​softw​are. Accessed 10 Jun 2022 https://​doi.​org/​10.​1007/​s12665-​010-​0732-y
Polimeni B, Bixio R, Galeazzi C, Germani C, Parise M, Saj S, Sammarco M Weidner L, Walton G (2021) Monitoring the effects of slope hazard mitigation
(2019) Creating a map of the underground heritage in the Mediter- and weather on rockfall along a Colorado highway using terrestrial laser
ranean area: a visual representation for a comprehensive research. In: scanning. Remote Sensing 13:4584. https://​doi.​org/​10.​3390/​rs132​24584
Amoruso, G., Salerno, R. (Eds.) Cultural landscape in practice. Conser- Weidner L, Walton G, Krajnovich A (2021) Classifying rock slope materials
vation vs. Emergencies. Lecture Notes in Civil Engineering. Springer in photogrammetric point clouds using robust color and geometric
Verlag features. ISPRS J Photogramm Remote Sens 176:15–29. https://​doi.​
Remondino F, Campana S (eds) (2014) 3D recording and modelling in org/​10.​1016/j.​isprs​jprs.​2021.​04.​001
archaeology and cultural heritage: theory and best practices. Archae- Weiss T, Siegesmund S, Kirchner D, Sippel J (2004) Insolation weathering
opress, Oxford, England and hygric dilatation: two competitive factors in stone degradation.
Ribacchi R, Rotonda T, Graziani A et al (2018) Meccanica delle Rocce. Environ Geol 46. https://​doi.​org/​10.​1007/​s00254-​004-​1041-0
Teoria e Applicazioni nell’Ingegneria. Edizioni Efesto, Italia Whabi AM (2014) Sedimentological and stratigraphic studies of the
Rohmer J, Alsuhaibani A, Lesguer F et al (2020) The Dadan Archaeologi- Cambro-Ordovician succession in northwest Saudi Arabia. Master the-
cal Project. Report on the first field season (2020), edited by J. Rohmer, sis, King Fahd University of Petroleum & Minerals, Dhahran – 31261,
A. Alsuhaibani and F. Lesguer. Centre national de la recherche sci- Saudi Arabia
entifique (CNRS), Agence française pour le développement d’AlUla
(Afalula), Royal Commission for AlUla (RCU)
Rohmer J, Charloux G (2015) From Liḥyān to the Nabataeans: dating Tommaso Beni · William Frodella
the end of the Iron Age in north-west Arabia. Proc Semin Arab Stud Department of Earth Sciences, University of Florence, Via G. La Pira 4,
45:297–319
50121 Florence, Italy
Ruedrich J, Bartelsen T, Dohrmann R, Siegesmund S (2011) Moisture
expansion as a deterioration factor for sandstone used in build- Tommaso Beni
ings. Environ Earth Sci 63:1545–1564. https://​d oi.​o rg/​1 0.​1 007/​ Division of Florence, National Institute for Nuclear Physics INFN, Via
s12665-​010-​0767-0 Bruno Rossi 1, 50019 Sesto Fiorentino, Italy
Salvini R, Mastrorocco G, Seddaiu M et al (2017) The use of an unmanned
aerial vehicle for fracture mapping within a marble quarry (Carrara, Daniela Boldini (*)
Italy): photogrammetry and discrete fracture network modelling. Department of Chemical Engineering Materials Environment,
Geomat Nat Hazards Risk 8:34–52. https://​d oi.​o rg/​1 0.​1 080/​1 9475​ Sapienza University of Rome, Via Eudossiana 18, 00184 Rome, Italy
705.​2016.​11990​53 Email: daniela.boldini@uniroma1.it
Sampietro-Vattuone MM, Peña-Monné JL (2021) Application of 2D/3D
models and alteration mapping for detecting deterioration processes Giovanni Battista Crosta
in rock art heritage (Cerro Colorado, Argentina): a methodological Department of Earth and Environmental Sciences, University of Milan
proposal. J Cult Herit 51:157–165. https://​doi.​org/​10.​1016/j.​culher.​ Bicocca, Piazza Della Scienza 4, 20126 Milan, Italy
2021.​08.​006
Spizzichino D, Margottini C, Chiessi V, Boldini D (2016) Assessment of the William Frodella · Claudio Margottini · Daniele Spizzichino
stability conditions of a large-volume sandstone block in the north- UNESCO Chair On the Prevention and Sustainable Management
ern sector of the Siq of Petra. In: Landslides and engineered slopes. of Geo-Hydrological Hazards, University of Florence, Largo Fermi 2,
Experience, theory and practice: Proceedings of the 12th International 50125 Florence, Italy
Symposium on Landslides (Napoli, Italy, 12–19 June 2016). CRC Press
1851–1858 José Ignacio Gallego
Sturzenegger M, Stead D (2009) Close-range terrestrial digital photo- RCU–Royal Commission for AlUla,, AlUla, Kingdom of Saudi Arabia
grammetry and terrestrial laser scanning for discontinuity characteri-
zation on rock cuts. Eng Geol 106:163–182. https://​doi.​org/​10.​1016/j.​ Edoardo Lusini
enggeo.​2009.​03.​004 Department of Civil, Chemical Environmental and Materials
Teza G, Marcato G, Pasuto A, Galgaro A (2015) Integration of laser scan- Engineering, University of Bologna, Via Terracini 28, 40131 Bologna,
ning and thermal imaging in monitoring optimization and assess- Italy
ment of rockfall hazard: a case history in the Carnic Alps (North-
eastern Italy). Nat Hazards 76:1535–1549. https://​doi.​org/​10.​1007/​ Daniele Spizzichino
s11069-​014-​1545-1 Depart. for Geological Survey of Italy, ISPRA, Via Vitaliano Brancati 48,
Themistocleous K, Cuca B, Agapiou A et al (2016) The protection of 00144 Rome, Italy
cultural heritage sites from geo-hazards: the PROTHEGO project. In:

Landslides