Anatomy of landslides along the Dead Sea Transform Fault System in NW

Jordan

H.G. Dill, K. Hahne, F. Shaqour

PII: S0169-555X(11)00646-5
DOI: doi: 10.1016/j.geomorph.2011.12.031
Reference: GEOMOR 3859

To appear in: Geomorphology

Received date: 13 December 2010

Revised date: 4 October 2011
Accepted date: 15 December 2011

Please cite this article as: Dill, H.G., Hahne, K., Shaqour, F., Anatomy of landslides
along the Dead Sea Transform Fault System in NW Jordan, Geomorphology (2011), doi:
10.1016/j.geomorph.2011.12.031

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Anatomy of landslides along the Dead Sea Transform Fault System in NW

Jordan

Dill, H.G. a,*, Hahne, K. a and Shaqour F. b

T
IP
a
Federal Institute for Geosciences and Natural Resources P.O. Box 510163
D-30631 Hannover, Germany

R
b
Jordan University, Faculty of Sciences, Department of Geology, Amman, Jordan

SC
* Corresponding author. Tel: +49-(0)511 643 2361; Fax: +49-(0)511 643 2304
E-mail address: [email protected] (H.G. Dill)

Abstract. NU
MA
In the mountainous region north of Amman, Jordan, Cenomanian calcareous rocks
are being monitored constantly for their mass wasting processes which occasionally
cause severe damage to the Amman-Irbid Highway. Satellite remote sensing data
D

(Landsat TM, ASTER, and SRTM) and ground measurements are applied to
TE

investigate the anatomy of landslides along the Dead Sea Transform Fault System
(DSTFS), a prominent strike-slip fault. The joints and faults pertinent to the DSTFS
P

match the architectural elements identified in landslides of different size. This

CE

similarity attests to a close genetic relation between the tectonic setting of one of the
most prominent fault zones on the earth and modern geomorphologic processes. Six
AC

indicators stand out in particular:

1) The fractures developing in N–S and splay faults represent the N–S lateral
movement of the DSTFS. They governed the position of the landslides. 2) Cracks
and faults aligned in NE–SW to NNW–SSW were caused by compressional strength.
They were subsequently reactivated during extensional processes and used in some
cases as slip planes during mass wasting. 3) Minor landslides with NE–SW straight
scarps were derived from compressional features which were turned into slip planes
during the incipient stages of mass wasting. They occur mainly along the slopes in
small wadis or where a wide wadi narrows upstream. 4) Major landslides with curved
instead of straight scarps and rotational slides are representative of a more advanced
level of mass wasting. These areas have to be marked in the maps and during land
management projects as high-risk area mainly and may be encountered in large
 ACCEPTED MANUSCRIPT
2

wadis with steep slopes or longitudinal slopes undercut by road construction works.
5) The spatial relation between minor faults and slope angle is crucial as to the
vulnerability of the areas in terms of mass wasting. 6) Springs lined up along faults
cause serious problems to engineering geology in that they step up the behavior of

T
marly interbeds to accelerate sliding during mass wasting

IP
The most vulnerable areas prone to slope instabilities are those with compressional

R
tectonics followed by extensional movements, with fault bound springs and smectite-

SC
bearing marly layers interbedded with pure massive limestones. The semi-arid to arid
climate with periodic rainfalls combined with subsurface water circulation along the

NU
joints and faults can trigger mass wasting.
MA
Keywords: Dead Sea Transform Fault System, Landslide, Carbonate rocks,
Cretaceous, Jordan
D

1. Introduction
TE

Landslides are natural hazards that are very widespread across the globe and
different climatic zones. They can be harmful to humans, cause property damage
P

and remove soil to thereby degrade the productivity of agricultural land. (Alexander,
CE

1986; Ahmad et al., 1993; Easterbrook, 1999; Guzzetti et al., 1999; Paudel et al.,
2003; Sidle et al., 2005; Galadini, 2006; Chen and Wang, 2007; Guthrie and Evans,
2007; Hilker et al., 2009). When slow-moving or abrupt mass wasting interrupts the
AC

road network of a region and block the lines of transport these geohazards,
furthermore, impede traffic and transport. Over the last few decades, Jordan
witnessed extensive construction activity, including inter-city highways passing over
and intersecting sedimentary rocks mainly of calcareous lithologies. One of these
highways runs across Jordan from the Gulf of Aqaba in the south to the Syrian
border in the north, crossing through the municipalities of Amman (Fig. 1). While the
southern branch of this road (Desert Highway) is hardly susceptible to geohazards,
because it runs through a flat topography, the northern branch between Amman and
Irbid cuts through a mountainous area parallel to the Dead Sea Transform Fault
System (DSTFS) (Figs. 2 and 3a). It is hence vulnerable to mass wasting (Fig. 3b,c).
Immediately north of Amman where the route crosses Cretaceous limestones mass
wasting processes of different dimensions have already damaged the road and
 ACCEPTED MANUSCRIPT
3

adjacent small housing areas (Fig. 4; Sawariah and Barjous, 1993; Al-Homoud et al.,
1999; Malkawi et al., 2000). The aim of our study is to map and describe the
geomorphologic and lithological features of mass wasting and classify their types. It
is possible to assess the influence of structural geology on the rock mass fabric and

T
mass wasting processes by investigating the regional structural pattern with a view to

IP
active landslides (Coltorti and Ollier, 2000; Galadini, 2006). The generations of
landslides is discussed in this morphoclimatic zone transitional from semiarid to arid

R
and treated in view of the stress field and the hydrology of one of the most prominent

SC
fault zones on Earth. The Arabian Plate is bounded to the west versus the African
Plate by the well-known strike-slip Dead Sea Transform Fault System (DSTFS)

NU
whose morphological expression can be recognized from the Gulf of Aqaba all along
the Wadi Araba through the Jordan Valley (Quennell, 1958, 1959; Sneh, 1996;
MA
Sharland et al., 2001; Smit et al., 2008) (Fig. 1). The local scarp pattern of landslides
is discussed in relation to the DSTFS which runs parallel to the Amman-Irbid
Highway. The discussion closes with structural and geomorphologic suggestions to
D

improve the area along the eastern side of the DSTFS.

TE

2. Study area
P

The study area lies in hilly grounds between 600 and 750 m above mean sea level.
CE

The wavy topography gradually turns into a more rugged topography approaching
the Dead Sea in the Jordan Valley.
AC

The stratigraphy and lithology of Jordan are summarized in Table 1. The distribution
of sediments and volcanic rocks is mapped in Fig. 1 (Bender, 1968, 1974; Toloczyki
et al. ,1999). The northwestern part of Jordan is covered by Late Cretaceous
calcareous sediments which submerge towards the NE under the plateau basalts of
Cenozoic age (Abed, 2000; Dill et al., 2009, 2010). Stratigraphic series as old as
Triassic are exposed in the core zones of some anticlines whose axes strike NE–SW.
North of the town of Irbid, Paleogene sediments and Cenozoic volcanic rocks cover
vast areas. Neogene and Quaternary sediments occur along the Dead Sea – Jordan
Valley graben system hiding the DSTFS and also in some NW-SE-trending grabens
branching towards the W (Fig. 1).
 ACCEPTED MANUSCRIPT
4

The landslide-prone area is located about 20 km north of Amman City as shown in

Fig. 1. At different sites in this hilly terrain underlain by gently dipping, thickly-bedded
limestones, landslides severely damaged the Amman-Irbid Highway (Fig. 4a,b). The
calcareous bedrock is stratigraphically attributed to the Cenomanian Hummar

T
Formation (A4) resting on marly and clayey deposits of the Fuhais Formation (A3)

IP
(Fig. 1, Table 1). These argillaceous deposits, measuring about 30 m in thickness,
have low permeability and, hence, act as an aquiclude. The A4, about 60 m thick, is

R
considered as a minor aquifer horizon within the lower Ajlun Group. This unit is

SC
overlain by the Shuayb Formation, a combination of two units (A5 and A6) and
composed of marl/limestone alternation (see Table 2 for chemical composition).

NU
About two thirds of the formation, ~200 m thick and made up of marls, behaves as an
aquitard. The area prone to minor and major landslides is intersected by a fault
MA
whose strike is around NNE to NE (Fig. 2b). This structure terminates an NNE-
running anticline cored by sediments of the Naur Formation towards the N and
sediments of the Kurnub Sandstone Series in the S (Fig. 2). The landslides
D

categorized in the text as major and minor landslides do not differ in their major
TE

features but in the size. Here, the term major landslide is used to delineate a mass of
rock exceeding 500 m in width and length and at least 10 m thick slid off the
P

underlying marly sediments.

CE

Jordan is situated in the eastern Mediterranean region which is characterized by

temperate semi-arid to arid climates. It has warm and dry summers, with
AC

temperatures up to 38°C, and cool winter months during which most of the annual
precipitation occurs with occasional snowfall in winter. Annual rainfall rarely exceeds
500 mm even in the highland areas and most of the country receives less than 200
mm (WAJ, 1989). The average evaporation constitutes approximately 90% of the
total rainfall (WAJ, 1989).

3. Methodology
The state of surface of the slides has been examined for changes in friability and
competence of rocks together with the spatial data. This is to perform subsurface
predictions based on the surface expression during ground-based as well as remote
sensing surveys of the landslides, using aerial photographs or satellite images of
 ACCEPTED MANUSCRIPT
5

high resolution. Field investigations were conducted for a tripartite subdivision of

each landslide using the surface-subsurface interrelation within the landscapes
affected by the mass wasting processes.

T
Mapping was based upon topographic map sheets of the area available on a scale

IP
1:50,000 along the Amman–Irbid Highway and in the area along the eastern side of
the DSTFS. They were used in combination with the geological map of the area to

R
locate and evaluate the spatial data of the slopes and the geological formations

SC
within the area as well as ease for detection of the small-scale structural elements
including faults (Muneizel and Khalil, 1993; Sawariah and Barjous, 1993; Basem and

NU
Muneizel ,1998; Shawabkeh, 2001; Diabat and Abdelghafoor, 2004). Aerial
photographs and satellite images are essential for such studies because they provide
MA
the necessary information on the morphology of the landscape and thereby facilitate
the assessment of the regional stress pattern (Mantovani et al., 1996). Enhanced
remote sensing data and geocoded satellite images included Landsat TM acquired
D

on October 13, 2003, ASTER acquired on January 17, 2007 an SRTM digital
TE

elevation model and Google Earth–Quickbird images.

P

Linear and planar structural elements within and outside the landslides including
CE

strike, dip, width and spacing were measured along with the positioning of each
waypoint by a Garmin 60 CSx GPS with an average accuracy of ±5 m.
The collected data and the Rockware software provided rose diagrams to display and
AC

evaluate the orientation of structural elements as density plots in the stereo net. The
spatial data were mapped using Surfer Software. Although the study area is
lithologically rather homogeneous, some samples were taken on the surface of the
sliding rock mass and within some of those scarps in which the lithology changed
from hard limestones to weaker marls.

In order to access the landslide-prone areas along the Amman–Irbid Highway linked
up with the DSTFS, a study area of the Western Highlands, delimited by the Jordan
Valley Escarpment, was selected for a small-scale tectonic analysis to assess major
and minor faults, thrust faults as well as anticlinal and synclinal axes. The latter are
visible in satellite images and geological maps which were updated in the course of
our field work
 ACCEPTED MANUSCRIPT
6

4. Results
4.1. Morphological subdivision of the landslides
Different types of landslides, which all affect the Amman–Irbid Highway, were

T
identified and studied for their gravitational mass wasting. After a reconnaissance in

IP
the area, two landslides were chosen for their importance with a view to the major
lines of communication north of Amman and for their usefulness as reference types.

R
They were the only landslides that occurred during the period of investigation, but not

SC
the only features of mass wasting in the area. Mass wasting is common and
prevalent also in the region of the Wadi As Sir (Fig. 3c). The landslides developed on

NU
the slopes of wadis facing westward and attaining critical slope gradients of approx.
1/10.
MA
The larger landslides (type 1) typically have curved main scarps/slip planes (1st
order) and widely-spaced 2nd order fractures confined to the north-eastern tail end of
D

the scarps (horse-tailing) (Fig. 5a). The 1st order features measure by the meter in
TE

width; their length is larger than the less prominent 2nd order faults by more than 10
m. The general strike of the 2nd order faults fluctuates around N–S. They fade out in
P

the northern calcareous wall rocks which results in a gradual boundary between the
CE

solid rock and the calcareous debris in motion.

The smaller landslide (type II) is intersected by several 1st order straight scarps/ slip
AC

planes, and a set of closely-spaced 2nd order fractures that are more or less
homogeneously distributed across the entire sliding body. Contrary to type I, the type
II landslide features a northern boundary that is delineated by a set of minor
discontinuous WNW-striking scarps. They create a sharp boundary of the landslide
against its northern calcareous wall rocks (Fig. 5b).

Both landslides are delimited downslope by a gentle bulge which marks the toe of the
landslide towards the W. There the dual carriage way of the modern Amman–Irbid
Highway runs largely parallel to its stratigraphic boundaries (Figs. 4b and 5). In
general, the major scarps in both landslides strike at an azimuth of approximately 45°
and run at angles of approximately 45° to the contour lines. The old road from
Amman to Irbid runs west of the present-day highway through the thalweg of the
 ACCEPTED MANUSCRIPT
7

wadi at the foot slope of the hills. It is entirely unaffected by the present landslides
(Fig. 4a,b). Several small hamlets are scattered across the slopes, some of which in
parts are damaged by the still active mass wasting and, hence, had to be abandoned
by their owners for safety reasons (Fig. 4c).

T
IP
4.2. Lithology and state of surface of the landslides

R
There are three types of calcareous rocks involved in the mass wasting.

SC
Lithotype I: Gradual disintegration of massive limestone eventually resulted in some

NU
kind of matrix-supported breccia with subangular limestone fragments floating in
brown loam. A gradual transition from the fresh calcareous rock, giving way through
MA
calcareous corestones into brown loam, occurs mostly in places of structural
weakness (Fig. 7a,c). For the purpose of this study, the terminology on rock strength
has been adapted to that proposed by Migon (1997) for the description of weathering
D

phenomena. Such disintegration processes causing a decrease in rock strength

TE

leave behind a cobble-strewn surface with friable brown loam. In vertical sections,
these cobble-strewn surfaces grade into funnel-shaped zones of rock disintegration
P

ending in a fracture zone (“sealed fractures”). Moving downward a gradual

CE

coarsening of the fragments can be discerned. The size of clasts can attain boulder
size. The dip of the bedding has been rotated clockwise creating a dip of bedding
against the slope angle (Fig. 7b). Locally, tiny undulous calcite bands cross these
AC

fractures unaffected by this mass wasting. These calcite bands denote zones of
abandonment of motion in these hilly areas.

Lithotype II: Coherent limestone benches dipping parallel to the slope create some
kind of a calcareous duricrust armoring the surface of the slides (Fig. 7e, #1). The
carbonate layers whose dip runs conformably with the slope angle are bent up
toward the toe-slope (Fig. 7e, #2).

Lithotype III: On top of the hill, the carbonate carapace shows a set of minor karren
carved into the limestone knops and weathering pits filled with residual loam in
between. Area III differs from area II only by its topographic position but not by quality
 ACCEPTED MANUSCRIPT
8

of rock strength or lithology (Fig. 7d). These areas cover four zones different as to
their block motion and spatial arrangement, which were discussed in Section 5.

4.3. Chemical composition and mineralogy of the landslides

T
Between 96 and 90 vol. % of the calcareous rocks are made up of calcite. Dolomite

IP
does not play any part among the carbonate minerals of the carbonate rocks affected
by mass wasting. The insoluble mineralogical constituents consist of quartz, illite and

R
illite-smectite mixed layers. The calcareous rocks under study are pure calcitic

SC
limestones sensu Fookes (1978). The chemical composition of the calcareous rocks
undergoing mass wasting is shown in Table 2. Only a few trace elements are above

NU
detection limit of XRF and listed thus in the table. The marly sediments abundant in
smectite-group phyllosilicates and intercalated into the calcitic limestones constitute
MA
the detachment horizon of the landslides.

4.4. Morphology, dip and strike of architectural elements pertinent to the landslides
The spatial data were collected during ground follow-up measurements within the
D

landslide-prone areas. The measurements include spacing of fractures, width of

TE

cracks and fractures, their shape as well as dip and strike directions (Fig. 8). These
features of the major architectural elements were used to characterize the different
P

parts of the landslide; they are crucial to unravel the history of mass wasting in this
CE

area. We drew on structural features, which have been described in different

textbooks and comprehensive papers. These features were collected for the various
AC

subsections of landslides and plotted as rose diagrams and stereo nets (Fig. 8). A
synoptical view encompassing all of these elements is given in Fig. 9.

A set of selected rose diagrams and stereo nets shown here reveals that the fracture
pattern is governed by two major trends. Group-I fractures tend to run N–S with a
slight deviation from that direction towards the W (Fig. 8). The other array of fractures
runs almost perpendicular to this direction fluctuating around NE (Fig. 8). A more
complex picture arises when the geometrical properties of the 2D-data array are
illustrated in a stereographic or equiangular projection in the stereo net (Figs. 8 and
9). The scatter diagram of poles has been contoured according to density per unit
area of the projection. The diagrammatic representation of the projected data,
clustering in the center of the stereo net, mirror dip and strike of the bedding planes
(poles-to-bedding figure). These planar elements gently dip towards the E in the
 ACCEPTED MANUSCRIPT
9

western proportion of the slide, and dip at a more inclined angle towards the WNW in
the eastern section. The various fractures form girdles close to the margin attesting
to very steeply dipping scarps and fractures within the landslide body.

T
At more than 40 waypoints, each covering between 20 and 30 measurements, within

IP
the major landslide, spatial data were collected together with their length, width and
striae indicating the orientation of the displaced blocks. During investigation of the

R
minor landslide, data were not collected at the same level of intensity as in the major

SC
landslide and hence a statistical treatment was not conducted. The pitch values of
the striae linear features on the fracture planes reflect the motion of individual

NU
subsections of the landslide which are shown to illustrate its various entities (Fig. 9).
Four subsections can be delineated within the landslide based on the morphology,
MA
dip and strike of the structural elements (Fig. 9). Delimited to the W and N by two
wadis, the entire slide is blanketed at some sites at its SE corner and in the center by
a scree of calcareous rubble (Fig. 7c).
D
TE

The morphology and size of the scarps, made visible by the constant motion of the
landslide, differs significantly with respect to surface expression and size. The main
P

scarps or 1st order fractures show a curved shape, concave toward the W, with a
CE

central block surrounded by both boundary fractures. The scarps are curved
concavely towards the direction of displacement which may also be deduced by the
zone of rubble or friction zones between the central block and each boundary fault
AC

(“r” in Fig. 10a). This block is wrapped down untilted (Fig. 10a). At its northern ends
some of these scarps grade into several smaller 2nd order scarps (Fig. 10b). In
contrast to the main scarps, the central blocks were subjected to rotational sliding in
this late-stage fracture pattern (Fig. 10b). Some cracks were described as saw teeth
fractures in the current paper owing to their peculiar course and outward appearance;
they never have any intermediate blocks, rarely exceed 1 m in width, and intersect
the landslide down to a depth of as much as 5 m in some places (Fig. 10c). One set
strikes N–S; the elements of the other set are aligned in NE–SW direction (Table 3).

4.5. Structural geology of the sedimentary series surrounding the landslides

The pressure zones are marked by a system of valleys and ridges running almost
NE–SW (Figs. 11 and 12). Gray linear structures cut across these structural
 ACCEPTED MANUSCRIPT
10

elements and are located on both sides of the master fault of the DSTFS which is
highlighted by the Dead Sea and the Jordan Valley running along it. A basin-and-
range topography characterized by extensional tectonics can be ruled out for most of
the study area in view of the cross sections B–B’ and C–C’ presented in Fig. 13. A

T
horst-and-graben tectonic may be deduced from cross section D–D’ (Fig. 13). In

IP
addition, the geomorphological setting close to the Jordan Valley Escarpment with
tributaries deeply incised in the western block of the DSTFS provides excellent

R
conditions for these structural elements detected by remote sensing methods and to

SC
be verified in the field (Fig. 3c). Zooming-out part of the study area does not provide
a picture that varies from its small-scale counterpart (Figs. 11 and 12). The fluvial

NU
drainage-pattern and network of natural springs is a mirror image of the structural
pattern (Figs. 12 and 13a). The interpretation of the stress pattern and the
MA
displacements along the faults is made visible in overlays and discussed in the next
section.
D

5. Discussion
TE

5.1. Regional structural analysis of the eastern block of the DSTFS

P

The influence of structural geology on rock mass fabric and mass wasting processes
CE

can be assessed by going from the small-scale to large-scale mass movements and
from the regional structural pattern to the landslide (Coltorti and Ollier, 2000;
Galadini, 2006). This approach has not been employed in areas along the western
AC

edge of the Arabian Plate, although it is considered to be crucial for understanding

mass wasting processes (Sharland et al., 2001). Outside this area in Jordan different
studies were performed to address the interrelationship between landslides and
tectonic processes. Space borne differential synthetic aperture radar interferometry
(DInSAR) has been used for mapping ground deformation phenomena and its
relation to mass wasting (Colessanti, 2003). Le Roux et al. (2008) have further
integrated morphogenetic methods into the study of tectonic and mass wasting
processes. The research by Sanchez et al. (2010) rests on a cosmoclimatic
approach. An almost pure structural approach underlies Gupta’s (2005) observation
that the direction of failure planes of landslides is concentrated either towards the
minimum stress axis σ3 or along the intermediate stress axis σ2, but never along the
maximum stress axis σ1. The latter investigation, although conducted in a modern
 ACCEPTED MANUSCRIPT
11

fold belt area, comes closest to our studies along this extensional zone of the Earth
crust. Considering the inventory of methods applied to tackle these issues between
landslides and tectonics, the present investigation resembles the approach taken by
Wetzel et al. (2000) and Roessner et al. (2005), with their extensive use of remote

T
sensing and GIS techniques.

IP
The major faults of the working area II represent a system of conjugate faults striking

R
NW and NE, intersecting the DSTFS with angles between 35° and 40°. Directions of

SC
maximum principal stress can be derived from satellite imagery, regarding the x-
shaped stress indicators, which are found on both sides of the Jordan Rift Valley.

NU
The acute angle between conjugate faults is bisected by the maximum principal
stress direction (Pmax). Fig. 11 shows a large-scale selection of such fault systems,
MA
while Fig. 12 provides the regional geology and Fig. 13a the morphology of the
corresponding area in a satellite image, so as to draw a coherent picture of structural
geology, lithology, hydrology and geomorphology.
D
TE

The area selected for the analysis of regional structural geology is located at a
junction of two major faults, showing dextral movements in broad zones (Fig. 13a).
P

These zones intersect at an angle of approximately 45°. The dextral transform is

CE

indicated by s-shaped minor faults that define tilted blocks dipping SE and s-shaped
shear lenses. Slickensides found on fault planes during ground follow-up work
confirm the right stepping character of the faults (Fig. 13b). More tectonical features
AC

that support this idea are given in Fig. 13c–e. This regional structural pattern
corresponds to the larger tectonic pattern illustrated in the maps of Fig. 1. The right
lateral transform fault of the working area becomes also obvious in the geological
map, where thrusts of opponent directions can be recognized. As the working area is
situated at a fault junction, there are hundreds of minor faults at different scales.
Major valleys and faults are mostly represented by steep vertical offsets. Minor faults
often represent shallow, layer-parallel thrusts causing layer thickening (Figs. 14 and
15). Besides being tilting and faulting, most layers are also folded. Folds occur in
various scales from meters to kilometers. Close to faults interfering multi-order folds
are common. The same is true for overturned isoclinic folds which are often cut off
(Figs. 3c and 14, cross section C–C`). Particular attention has to be paid to the listric
shape of some of the minor faults illustrated in Fig. 14. They have an effect on the
 ACCEPTED MANUSCRIPT
12

slope instability and the way in which mass wasting took place in this area.
Numerous springs, which can be correlated with these minor and major/master faults
(see different thicknesses of red lines in Fig. 14), added to the problems of
engineering geology. They provided water to the marly lubricants and kick-started the

T
mass wasting during periods of hefty rainfalls (Figs. 12 and 14). If such listric faults

IP
dip along with slope angle, the slope stability will become as low as discussed for the
two case histories along the Amman–Irbid highway. In tightly folded areas these

R
listric faults dip away from the slope and minimize the risk of slope failures (Fig. 14).

SC
In regions with horst-and-graben tectonic that follows this compressional tectonic
style, these listric faults become the detachment zones for subsequent mass wasting

NU
creating a “high-risk zone” together with fault-related springs (Fig. 14).
MA
5.2. Local structural analysis of the landslide-prone areas
A composite view of, on the one hand, the satellite image and, on the other hand, the
two reference types of landslides reveals that the structural regional pattern accords
D

with the system of fractures in the landslide prone areas along the Amman–Irbid
TE

Highway. The landslides under investigation correspond to the satellite image as to

orientation of fractures but scarps are not to scale in Fig. 13c. The major direction of
P

displacement during the mass wasting does not run parallel to the dip direction of the
CE

slope. It is an asymmetric landslide whose major direction of mass transport runs

obliquely to the dip of slope and at a right angle to the pressure ridges detected
further W during analysis of the regional stress pattern (Fig. 13a). Remote sensing
AC

data are most suitable in evaluating landslide prone areas and decipher the structural
setting around. Aerial photographs of Landsat TM false color composite in a scale
1:25,000 might have been found very suitable for these investigations. The digital
elevation model (DEM) derived from SRTM data and Quickbird oblique-angle images
are an excellent supplement.

5.2.1. Rubble-strewn areas

Areas coded “b” in Fig. 9 denote rubble-strewn parts of the landslide which now
obscure/cover-up abandoned fractures underneath. These fractures were active
during pre-gravitational movements when this part of the crust was ongoing
compression in NW–SE direction (Figs. 7 and 9, Table 3).
 ACCEPTED MANUSCRIPT
13

5.2.2. Saw teeth fractures

The saw teeth cracks are interpreted as primarily compressional features known from
a great variety of sedimentary rocks on a macro- and micro scale (Fig. 10c, Table 1).
They are indicative of the initial stages of fracturing under the existing stress pattern

T
governed by the strike-slip movement of the DSTFS. They swiftly get backfilled with

IP
brown weathering loam when becoming opened up during subsequent extensional
processes and abandoned during mass wasting. Minor fractures horse-tailing from

R
the major scarps and aligned in N–S direction are a mirror image of the lateral

SC
movements along the DSTFS (Figs. 1 and 3, Table 1). The lateral movement of the
DSTFS may be deduced also by striae from the 2nd order fractures running almost

NU
N–S in the major landslide (Figs. 5b and 11). It is the morphological expression to
accommodation strains near fault terminations where splay faults were branching.
MA
5.2.3. Straight boundary fractures
WNW–ESE fractures that terminate the minor slides create an almost continuous
D

boundary fault towards the N (Fig. 5b) and are found in an embryonic stage in form of
TE

discontinuous NW–SE faults in the major slides (Fig. 5a). They make use of a
structural disposition (rubble-strewn-areas) and reactivate incipient structures (Table
P

3). The morphological expression of this fractures near the northern terminus of the
CE

slides is vague and the boundary is much more subjected to

morphological/exogenous constrains such as the direction and dip of wadi slopes
than any structural disposition created by an ancient stress pattern that has been
AC

abandoned for long. As a result it is nowhere well preserved.

5.2.4. Major scarps

The curved scarps fade out to the NE and become more pronounced towards the
SE, where the impact on the tarmac road is at its maximum. This climax zone of
slope failure with major damage inflicted on the highway coincides with the maximum
of block tilting and the dip of pressure ridges in Fig. 13. They are the most visible
gravitational response to the modern tectonic stress pattern so as to say
neotectonics. Toe-material pushed out at the base of the landslide at the SW beyond
the slip plane. The regional stress pattern with its NE–SW trend may be traced into
the landslides irrespective of the morphology of the scarps, which may be straight or
arcuate. The straight scarps are the ones which show the most immediate
 ACCEPTED MANUSCRIPT
14

morphological response to the modern displacements in minor and embryonic slides

(Fig. 5). By contrast, arcuate scarps, albeit following the same stress pattern, only
occur at an advanced stage, so that the geometric correspondence between
structural pattern and gravitational movement might sometimes be overlooked.

T
The geodynamic setting and stress pattern on the eastern block of the DSTFS was

IP
the most controlling effect for the asymmetric landslides to develop in the Upper
Cretaceous calcareous rocks. Therefore, landslide-prone areas along the highway

R
may be singled out by a simple structural analysis of the regional structural regime,

SC
combining remote sensing and ground follow-up measurements.

NU
5.3. Morphology of the landslides and shaping the landscape
A morphological analysis of scarps and fractures and an assessment of the way of
MA
motion along them conduced to the subdivision of the landslide into four entities.
They may eventually reduce the number of unwarranted assumptions in a risk
assessment of a landscape vulnerable to landslides. Slide segment I is characterized
D

by rotational movement with the main slip direction oriented to WNW (Fig. 9). This
TE

accords with the movements deduced from the structural processes recorded from
the cross section D–D´ (Figs. 12, 13a and 14). The major scarps strike NE and NNE.
P

Segment I exerts a major component together with segment II to the gravitational

CE

force that causes damage to the traffic lane. In slide segment II, block sliding towards
W and NW parallel to the slope is operative. The northern slide segment III is
dominated by block sliding again with a strong vertical and minor lateral component
AC

(Fig. 10a,b, Table 1). The mass moving downhill gave rise to the toe of the landslide
which is the result of the combined action of these three segments (Fig. 9). Fractures
developed at an intermediate to late stage of mass wasting. They were driven by
gravitational forces only. This is also true for the straight and arcuate scarps
irrespective of whether they are accompanied by simple block sliding. The
morphology, namely the question whether they are concave or straight, is not
independent of the processes shaping the landscape. Arcuate slip planes or 1st
order scarps reflect an advanced stage of mass wasting, following straight planes.
The shape of the scarps in the area is dictated by the presence or absence of an
abutment of the sliding mass. Moderate risk-areas exist mainly in small wadis or in
those parts where the width of a wadi narrows down. Minor landslides with straight
scarps are met where wadi 2 in Fig. 9 fades out towards the N (Fig. 5). Only a creep
 ACCEPTED MANUSCRIPT
15

may be observed towards wadi1 in slide segment III. High-risk areas occur in large
wadis, e.g. wadi 2, with naturally steep slopes or slopes undercut by man-made
processes and no natural “gabion” or ”dam” exists to counterbalance the sliding rock
mass.

T
IP
5.4. Exogenous and endogenous effects triggering or accelerating mass wasting
The chemical composition of the carbonate rocks points to a pure non-dolomitized

R
limestone. Neither diagenetic alteration (dolomite) nor solution processes that result

SC
in the formation of karst cavities can be held responsible for the vulnerability of these
limestones to mass wasting. This is evident given the chemical composition of the

NU
limestones under consideration which have neither a significant amount of smectite
nor that of sulfate-bearing evaporites. The SO3 content is very low and does not allow
MA
gypsum precipitation as it is the Al2O3 content that rules out the presence of
phyllosilicates. Only minor rillenkarren exist. However, no deep subrosion has
weakened the bedrock. Rock mechanical aspects and hydrology control the
D

gravitationally driven mass wasting.

TE

There are exogenous and endogenous processes that can trigger or accelerate mass
P

wasting in an area with a negative structural disposition, for instance along the
CE

eastern side of the DSTFS. The climatic conditions with dry summers and abundant
rainfall during wintertime may kick-start and/or accelerate downhill creeping or sliding
of rock masses. Structural analysis has revealed that subsurface water ways give
AC

way to natural springs along strike slip faults oriented N–S and NNE–SSW which are
directly connected with the major strike slip movements along the DSTFS (Figs. 12 to
14). Even if the displacement along the DSTFS and its parasite faults in the Jordan
Valley is very small relative to the dimension of the active mass wasting, micro
seismicity is another cause to reckon with in predicting geohazards like this.
Numerous small faults intersect the Jurassic and Cretaceous sediments near the
escarpment of the Jordan Valley, demonstrating the ongoing displacement (Fig. 15).
These normal faults are an ideal pathway for circulating waters, as evidenced by the
dark green lines of vegetation along these fractures in an otherwise yellow-brown dry
landscape. These modern fault displacements occur side-by-side with zones prone to
soil creep, which in some areas also developed further by herds of grazing sheep
(Fig. 9a). There is a gradual transition from folds of various dimensions into
 ACCEPTED MANUSCRIPT
16

paleolandslides/slumps as illustrated by the cross sections in Fig. 14 which runs

perpendicular to the swarm of faults drawn in Fig. 13. Folding, faulting and mass
wasting are coherent processes operative since the DSTFS at the western edge of
the Arabian Plate became an active structural system (Fig. 1). Modern structural

T
distortions of the Mesozoic sediments along the eastern block of the DSTFS exert

IP
the greatest impact on creeping and sliding soil and rock masses.

R
6. Conclusions and recommendations

SC
The Dead Sea Transform Fault System (DSTFS) represents a left-lateral strike-slip

NU
fault. Its displacement reflected in the ductile and brittle deformation of Mesozoic
sediments along the eastern block is responsible for slowly and rapidly moving soil
MA
and rock masses shaping the landscape along the eastern escarpment and affecting
the lines of traffic and housing areas.
D

The regional structural pattern elaborated for the eastern block of the DSTFS
TE

matches the pattern of fractures and cracks identified in the area prone to landslides.
This structural pattern furnishes clear evidence that the geomorphological setting and
P

its exogenous processes such as landslides did not work their way through northern
CE

Jordan independent of the tectonic processes which are still ongoing.

1. The fractures developing in N–S direction and splay faults represent the N–S
lateral movement of the DSTFS. They govern the position of the landslides.
AC

2. Cracks and faults aligned in NE–SW to NNW–SSW directions are primarily

compressional features. They were reactivated during extensional processes
and used in some cases as slip planes during mass wasting
3. Minor landslides with NE–SW straight scarps derived from compressional
features which were turned into slip planes during the incipient stages of mass
wasting. They mostly occur along the slopes in small wadis or where a wide
wadi narrows upstream.
4. Major landslides with curved scarps, i.e., rotational slides, mirror a more
advanced level of mass wasting. These areas have to be labeled mostly as
high-risk areas and can be encountered in large wadis with steep slopes or
slopes undercut by man-made processes inappropriate under such
circumstances.
 ACCEPTED MANUSCRIPT
17

5. The spatial relation between minor faults and slope angle in an area is chiefly
determinative of its vulnerabilityto mass wasting.
6. Springs lined up along faults or located near them cause serious problems to
engineering geology given that they improve the quality of marly lubricants.

T
The risk to traffic lanes is highest where they run at an acute angle to the regional

IP
fault planes. The most vulnerable areas, prone to slope instabilities are those
undergoing compressional tectonic being followed by extensional movements, with

R
fault bound springs and smectite-bearing marly layers interbedded with pure massive

SC
limestones (Figs. 2b, 13 and 14). A joint analysis of geomorphology, hydrology and
structural geology ahead of infrastructural planning is strongly recommended. It is

NU
ecologically and economically much preferable to monitoring slopes in the wake of
construction work already undertaken. Tunneling, though a costly option, has to be
MA
revisited as an alternative to constructing a tarmac road alongside a slope affected
by myriads of slope failures, which will likely be in constant need of monitoring for
creeping or sliding processes.
D
TE

Acknowledgements
We are indebted to F. Korte for XRF analysis and D. Weck for XRD analysis at the
P

Federal Institute for Geosciences and Natural Resources in Hannover, Germany.

CE

We express our gratitude to Janina Dill (Merton College Oxford, Great Britain) for
checking the manuscript for linguistic flaws. We thank the anonymous reviewers for
their help to improve an early draft of our paper and Takashi Oguchi for his
AC

comments.

References
Abed, A.M., 2000. Geology of Jordan, Its Environment and Water. Jordanian
Geologist Association, Amman, 571 pp. (in Arabic)

Ahmad, R., Earle, A.H., Hugues, P., Maharaj, R., Robinson, E., 1993. Landslide
damage to the board river water supply pipeline, Bromley Hill, Jamaica: Case study
of a landslide caused by hurricane Gilbert (1988). Bulletin of Engineering Geology
and the Environment 47, 59-70.

Alexander, D., 1986. Landslide damage to buildings. Environmental Geology 8, 147-

151
 ACCEPTED MANUSCRIPT
18

Al-Homoud, A, Prior, G., Awad, A., 1999. Modeling the effect of rainfall on instabilities
of slopes along highways. Environmental Geology 37, 317-325.

Basem, K.M., Muneizel, S., 1998. The geology of Al Salt area map sheet 3154 III-
National Resources Authority. Geological Bulletin 41, 1-67.

T
IP
Bender, F., 1968. Geologie von Jordanien. Gebrueder Borntraeger, Stuttgart, 230 pp.

R
SC
Bender, F., 1974. Geology of Jordan. Gebrueder Borntraeger, Stuttgart, 196 pp.

Chen, Z., Wang, J., 2007. Landslide hazard mapping using logistic regression model

NU
in Mackenzie Valley, Canada. Natural Hazards 42, 75-89.
MA
Colesanti, C. 2003. Monitoring landslides and tectonic motions with the permanent
scatterers technique. Engineering Geology 68, 3-14.
D

Coltorti, M., Ollier, C.D. 2000 Geomorphic and tectonic evolution of the Ecuadorian
TE

Andes. Geomorphology 32, Pages 1-19 .

P

Diabat, A., Abdelghafoor, M., 2004. Geological Map Sheet 3153 I – Amman, Scale
CE

1:50.000. National Resources Authority, Amman .

AC

Dill, H.G., Kus, J., Abed., A.M., Sachsenhofer, R.F., Abul Khair, H., 2009. Diagenetic
and epigenetic alteration of Cretaceous to Paleogene organic-rich sedimentary
successions typical of the western margin of the Arabian Plate, northwestern Jordan:
GeoArabia 14, 101-140.

Dill, H.G., Bechtel, A., Kus, J., Gratzer, R., Abu Hamad, A.M.B., 2010. Deposition
and alteration of carbonaceous series within a Neotethyan rift at the western
boundary of the Arabian Plate: The Late Permian Um Irna Formation, NW Jordan, a
petroleum system. International Journal of Coal Geology 81, 1-24.

Easterbrook, D. J., 1999. Surface Processes and Landforms. Prentice-Hall.

Publishing Co., 546 pp.
 ACCEPTED MANUSCRIPT
19

Fookes, P.G., 1978. The geology of carbonate soils and rocks and their engineering
characterization and description. In: Jewell, R.J., Korshid, M.S., (Eds.), Engineering
for Calcareous Sediments. Balkema, Rotterdam, pp. 787-806.

Galadini, F., 2006. Quaternary tectonics and large-scale gravitational deformations

T
with evidence of rock-slide displacements in the Central Apennines (Central Italy).

IP
Geomorphology 82, 201–22.

R
Gupta, V., 2005. The relationship between tectonic stresses, joint patterns and

SC
landslides in the higher Indian Himalaya. Journal of Nepal Geological Society. 31,
51-58.

NU
Guthrie, R.H., Evans, S.G., 2007. Work, persistence, and formative events: The
MA
geomorphic impact of landslides. Geomorphology 88, 266-275.

Guzzetti, F., Carrara, A., Cardinali, M., Reichenbach, P., 1999. Landslide hazard
evaluation: a review of current techniques and their application in a multi-scale study,
D

Central Italy. Geomorphology 31, 181-216 .

TE

Hilker, N., Badoux ,A., Hegg, C., 2009. The Swiss flood and landslide damage
P

database 1972–2007. Natural Hazards Earth System Science 9, 913-925.

CE

Le Roux, O., Schwartz, S., Gamond, J.-F., Jongman,s D., Tricart, P., Sebrier, M.,
AC

2008. Interaction between tectonic and erosion processes on the morphogenesis of

an Alpine valley: geological and geophysical investigations in the lower Romanche
valley (Belledonne Massif, western Alps). International Journal of Earth Sciences 99,
427-441.

Malkawi, A., Saleh, B, Al-Sheriadeh, M., Hamza, M., 2000. Mapping of landslide
hazard zones in Jordan using remote sensing and GIS. Journal of Urban Planning
and Development ASCE 126, 1-17.

Mantovani, F., Soeters,. R., Van Westen C.J., 1996. Remote sensing techniques for
landslide studies and hazard zonation in Europe. Geomorphology 15, 213-225.
 ACCEPTED MANUSCRIPT
20

Masri, M., 1963. Report on the Geology of the Amman-Zerqa Area. Central Water
Authority, Amman, Jordan (unpublished report), 74 pp.

Migon, P., 1997. Palaeoenvironmental significance of grus weathering profiles: a

T
review with special reference to northern and central Europe. Proceedings

IP
Geological Association 108, 57–70.

R
Muneizel S., Khalil B., 1993. Geological Map Sheet 3154 III – As Salt, Scale

SC
1:50.000. National Resources Authority, Amman.

Paudel, P.P., Omura, H., Kubota, T., Morita, K., 2003. Landslide damage and

NU
disaster management system in Nepal. Disaster Prevention and Management 12,
413-419.
MA
Powell, J.H., 1989. Stratigraphy and Sedimentation of the Phanerozoic Rocks in
Central and South Jordan. 1 : 50,000 Geological Mapping Series, National
D

Resources Authority, Geological Bulletin 11, 1-130.

TE

Quennell, A.M., 1958. The structural and geomorphic evolution of the Dead Sea rift.
P

Quarterly Journal of the Geological Society of London 114, 1–24.

CE

Quennell, A.M., 1959. Tectonics of the Dead Sea rift. In 20th International Geological
AC

Congress, Association of African Geological Surveys, pp. 385–405.

Roessner, S., Wetzel H.-U., Kaufmann H., Sarnagoev A., 2005. Potential of satellite
remote sensing and GIS for landslide hazard assessment in southern Kyrgyzstan
(Central Asia). Natural Hazards 35, 395–416.

Sanchez, G., Rolland, Y., Corsini, M., Braucher, R., Bourlès, D., Arnold, M.,
Aumaître, G., 2010. Relationships between tectonics, slope instability and climate
Change: Cosmic ray exposure dating of active faults, landslides and glacial surfaces
in the SW Alps. Geomorphology 117, 1-13.

Sawariah, A., Barjous, M., 1993. Geological Map Suwaylih Scale 1:50,000, Sheet no.
3154-II. Natural Resources Authority, Amman, Jordan
 ACCEPTED MANUSCRIPT
21

Sharland, P.R., Archer, R., Casey, D.M., Davies, R.B., Hall, S.H., Heward A.P.,
Horbury, A.D., Simmons, M.D., 2001. Arabian Plate Sequence stratigraphy.
GeoArabia, Special Publication 2, 1-371.

Shawabkeh, K.F., 2001. Geological map sheet 3153 IV – Karama, Scale 1:50.000. -

T
National Resources Authority, Amman.

R IP
Sidle, R.C., Kamai, T.; Trandafir, A.C., 2005. Evaluating landslide damage during the

SC
2004 Chuetsu Earthquake, Niigata, Japan . EOS, Transactions American
Geophysical Union 86, 133-136.

NU
Smit, J. , Brun, J.-P., Fort, X., Cloetingh, S., Ben-Avraham, Z., 2008. Salt tectonics in
pull-apart basins with application to the Dead Sea Basin. Tectonophysics 449, 1-16.
MA

Sneh, A., 1996. The Dead Sea Rift: lateral displacement and downfaulting phases.
Tectonophysics 263, 277-292.
D
TE

Toloczyki, M, Trurnit, P., Voges, A., 1999. International Geological Map of Europe
and the Mediterranean Regions 1:1 500 000, Map Sheet F 7 Amman. Federal
P

Institute for Geosciences and Raw Materials, Hannover.

CE

WAJ (Water Authority of Jordan), 1989. North Jordan Project—Amman-Zarqa Basin.

AC

Internal report Ministry of Water and Irrigation, Water Authority of Jordan , Amman,
Jordan.

Wetzel, H.-U., Roessner, S., Sarnagoev, A., 2000. Remote sensing and GIS based
geological mapping for assessment of landslide hazard in southern Kyrgyzstan
(Central Asia). In: Brebbia, C.A., Pascolo, P. (Eds.) Management Information
Systems 2000: GIS and Remote Sensing. WIT Press, Southampton, pp. 355–366,

Captions
 ACCEPTED MANUSCRIPT
22

Fig. 1. Map showing the geology of northern Jordan (modified from Bender 1968,
1974; Toloczyki et al., 1999), the major routes of transport in the Kingdom of Jordan
and the topographic position of the study area along the Amman-Irbid Highway. LT:
Lake Tiberias, DS: Dead Sea. Roman numerals denote the study areas of mass

T
wasting (I) and structural geology (II)

R IP
SC
NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
23

Fig. 2. Geological map of the landslide prone area along the Amman–Irbid Highway.
The inset map shows the location of the study areas and localities along the highway
from Aqaba to Irbid referred to in the text

T
R IP
SC
NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
24

Fig. 3. Geographic and geological setting of the highway from the Gulf of Aqaba to
Irbid (for location see Fig. 1). a) Southern branch between Aqaba and Amman mainly
through the western marginal zone of the Arabian Desert strewn with cobbles and
pebbles coated by desert varnish (dv). The tarmac road south of Amman passes

T
over outcrops of flat-lying Paleozoic and Mesozoic sedimentary rocks surrounding

IP
the crystalline basement of the Arabian Shield. b) Slightly tilted sandstones (sdst)
and claystones (cst) with embedded lignite (lig) seams of the Kurnub Sandstone

R
exposed in a road-cut of a river valley along the northern branch of the Amman–Irbid

SC
Highway (Northern Highway; see Fig. 1). These sediments are rather stable and not
affected by mass wasting. c) Recumbent fault of the Turonian Shuayb Formation

NU
(Powell, 1989) made up of marls and limestones exposed along the steeply dipping
hillslopes of the Wadi Es Sir (core of the syncline is composed of the upper Turonian-
MA
2 sediments), west of Amman.
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
25

Fig. 4. Morphology of landslides and damage of properties and roads. a) Mass

wasting pushing into the tarmac road, which narrows down the traffic lane of the
Amman–Irbid Highway, north of Amman. The lower set of gabions along this
landslide-prone stretch of the tarmac road has been tilted towards the road and

T
squeezed out of the gabion wall. b) Slump of the landslide with the dip of slope

IP
marked by arrowheads. The line where the dip of slope reverses is marked by a
stippled line and in the field by transverse cracks. The road affected by the landslide

R
is shown in the background running from top right to down left. c) House destroyed

SC
by the block tilting and sliding in the course of mass wasting. Two scarps are labeled
“sc”.

NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
26

Fig. 5. Two reference types of landslides along the Amman–Irbid Highway. a)

Landslide intersected by curved scarps (slip planes, 1st order) and widely spaced
2nd order fractures confined to the NE tail end of the scraps (horse-tailing), typical of
large wadis with slope gradients of 1:10. The regional tectonic stress pattern is still

T
preserved but reworked by rotational and block sliding at an advanced level. b)

IP
Landslide intersected by straight scarps (slip planes, 1st order) and closely spaced
2nd order fractures, typical of narrow wadis with slope gradients of 1:10. The regional

R
tectonic (compressional) stress pattern is still well-preserved.

SC
NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
27

Fig. 6. Two landslides along the Amman–Irbid highway illustrated on a relief map with
25-m contours (courtesy: Google Earth). They represent different levels of mass
wasting in the same tectonic setting but different geomorphologic terrains.

T
R IP
SC
NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
28

Fig. 7. Topstratum and state of surface of the landslide area. a) Gradual

disintegration of massive limestones eventually resulting in some kind of matrix-
supported breccia with subangular limestone fragments floating in brown loam. The
stippled line denotes the position of a slide-induced normal fault in the near-surface

T
part of the mass wasting body. An undulous calcite band is crossing the fracture

IP
unaffected and indicates that the fissure is no longer active. b) Funnel-shape of
structural weakness along a fracture zones whose surface expression is shown in

R
(c). Vertically downward, a gradual coarsening of the fragments may be recognized.

SC
The size of clasts may attain boulder size. The dip of the bedding has been rotated
and led to a dip of bedding against the slope angle. c) A cobble-strewn surface with

NU
friable brown loam, marking more intensive weathering and disintegration due to
some infiltration of surface water. The result at depth is displayed in (b). d) Carbonate
MA
carapace on top of the hill with minor karren (arrowhead) carved into the limestone
knops and weathering pits filled with residual loam in between. e) Coherent limestone
benches dipping parallel to the slope (1) constitute some kind of a carapace armoring
D

the surface. Beyond the slope-parallel wadi the limestone layers (2) are bent up
TE

toward the toe-slope.

P
CE
AC
 ACCEPTED MANUSCRIPT
29

Fig. 8. Architectural elements of the landslide and interpretation. a) Selected rose

diagrams based on the strike of fractures in the south-western parts of the landslide
shown in Fig. 5b. b) Selected stereo nets (Wulff net) with pole projection of the
bedding planes near the center (“ss”) and a girdle reflecting the dip direction and dip

T
angle of fractures (“f”) from the same part in the SW of the landslide.

R IP
SC
NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
30

Fig. 9. Synoptic view of all spatial and morphological data for 40 waypoints across
the major landslide. Lines: fractures and scarps (line thickness corresponds to the
fracture width, and line length corresponds to fracture length). “b”: boulder-strewn
areas. Gray solid arrows: dip of bed surface. Black open arrows: direction of motion.

T
Roman numerals I to IV are codes to show areas within the landslide. calcite (open)

IP
= open fracture whose fracture planes are lined with calcite.

R
SC
NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
31

Fig 10. Morphology of scarps. a) Main scarp with the central block between the
boundary fractures (dotted lines) wrapped down untilted (arrows). Viewing towards N.
The 1st order fracture starts horse-tailing or fanning out into several smaller curved
2nd order scarps or splay faults (see b). The zone of rubble (friction zone) is coded

T
with two “r”. b) Horse-tail scarps (2nd order) (stippled lines) with the central blocks

IP
undergoing rotational sliding. The arrows mark dip of surface of the tilted blocks. c)
Saw teeth fractures intersecting the landslide down to a depth of as much as 5 m in

R
some places. These fractures mark the initial stages of fracturing under the existing

SC
stress pattern and swiftly get backfilled with brown weathering loam when becoming
abandoned during mass wasting. Their surface expression is attested to by saw teeth

NU
boundary fractures and subsidences. MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
32

Fig. 11. Structural and geological setting. a) Regional overview with examples of x-
shaped maximum-principal-stress-indicators (Pmax) and locations of the working
areas (yellow rectangles). Conjugate major faults intersect the sinistral DSTFS at
angles between 35°–40°. Landsat TM false color composite, bands 7,4,1 (RGB),

T
UTM, WGS 84. Working area I marks the landslide-prone area along the Amman–

IP
Irbid Highway. Working area II bridges the gap within the DSTFS. The latter area was
used for studying the structural geology by satellite images and by ground-follow up

R
investigations. Insets illustrate the different types of landslides along the Amman-Irbid

SC
Highway with the regional stress pattern projected onto the landslides under
consideration. b) Major landslides with curved major scarps (combination of rotational

NU
and simple block sliding) and minor landslides with straight major scarps (mainly
simple block sliding). Both landslides correspond to the satellite image as to
MA
orientation of fractures but scarps are not to scale. For size and position of both
landslides see Fig. 5a.
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
33

Fig. 12. Updated geological map with locations of the cross sections B–B’, C–C’ and
D–D´. Modified after Diabat and Abdelghafoor (2004), Shawabkeh (2001), Sawariah
and Barjous (1993) and Muneizel and Khalil (1993). Springs related to these faults
and affecting slope instabilities are shown by red dots with their location names given

T
in rectangles.

R IP
SC
NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
34

Fig. 13. Structural features on different scales. a) Examples of major faults within two
broad dextral shear zones (thick grey arrows) intersecting at an angle of 45°.
Dextral movement is indicated by s-shaped tilted blocks (shades at steeper scarp
slopes; minor faults confining blocks are not shown) and s-shaped shear lenses.

T
White lines denoting locations of cross sections B–B’, C–C’ and D–D´ shown in Fig.

IP
14. Springs related to these faults are shown in red circles. b) Slickensides on fault
planes proving the dextral sense of movement. c) Tilted blocks dipping SE in a view

R
to the E (dashed grey rectangle in a). The dominant bank of Wadi As Sir Limestone

SC
Formation (A7) highlights the torn and tilted character of the blocks. Vertical
exaggeration is 2×. The digital elevation model (DEM) is derived from SRTM data,

NU
and the source of the high resolution Quickbird image is Google Earth. d) Field image
of slickenside on a fault plane proving the dextral sense of movement. e) Minor
MA
normal fault with a vertical displacement of 1.1 m. Blocks dipping NW; thin grey arrow
pointing at location.
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
35

Fig. 14. Endogenous fault movements and folding, triggering mass wasting along the
eastern block of the DSTFS. Cross sections B–B’, C–C’, and D–D´ illustrate layer
thickening caused by isoclinic folding and shallow thrusting defined by minor faults.
Locations are shown in Figs. 12 and 13a. Minor faults (thin red lines) show a listric

T
shape marking high-risk areas where they dip parallel to the slope and springs are

IP
located in the immediate vicinity, e.g. in section D–D´ at Ain Baqqouria. The header
for each cross section denotes the intensity of compressional and extensional

R
tectonics in the region and their bearing on mass wasting processes by generating

SC
listric faults, which translate into landslides illustrated in Fig. 5

NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
36

Fig. 15. Normal faults with small jumps in Jurassic sandstones along the Jordan
Valley Escarpment. Potential pathways for water are highlighted by the lines of
vegetations marking the fractures on the slope. Arrows denote zones of soil creep.

T
R IP
SC
NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
37

T
IP
R
SC
NU
MA
D
PTE
CE
AC
 ACCEPTED MANUSCRIPT
38

Table 1. Stratigraphical units in the study areas in NW Jordan (Source: Masri, 1963;
Powell, 1989).

T
R IP
SC
NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
39

Table 2. Chemical composition of the calcareous rocks undergoing mass wasting.

SiO2 to SO3 are given in wt. % and Ba through Zr in ppm.

T
min mean max
SiO2 0.80 3.95 7.94

IP
TiO2 0.01 0.05 0.10
Al2O3 0.25 0.88 1.98

R
Fe2O3 0.15 0.43 0.90
MnO 0.00 0.01 0.01

SC
MgO 0.64 4.45 11.27
CaO 35.28 47.13 54.32
Na2O 0.01 0.01 0.02
K2O 0.01 0.02 0.05

NU
P2O5 0.01 0.02 0.03
SO3 0.11 0.15 0.18
Ba 7 26.0 37
Cr 5 8.0 14
MA
Cu 15 38.0 58
Sr 127 239.3 453
V 6 18.0 30
Zn 6 9.3 16
Zr 6 16.3 29
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
40

Table 3. Architectural elements of the landslides and their origin.

T
Fracture Orientation Movement Class/size Relative age Force
type

IP
Rubble- NW–SE compression patchy incipient regional
strewn (abandoned distribution stage stress

R
areas structures) on the
surface

SC
measuring
several
square

NU
meters
across
Saw teeth NE–SW/ compression 1st order initial stage regional
MA
fractures NNE–WSW  extension and 2nd stress
order
fractures
Saw teeth approx. N–S lateral 2nd order initial stage regional
D

fractures movement fractures stress

transitional
TE

into straight
fractures
Straight WNW to NW vertical 2nd order intermediate gravitational
P

boundary movement fractures to late stage force

fractures
CE

Straight NE–SW vertical 1st order (early) late gravitational

scarps movement fractures stage force
block (and
AC

rotational
sliding)
Arcuate NE–SW vertical 1st order latest stage gravitational
scarps movement fractures force
block and
rotational
sliding
 ACCEPTED MANUSCRIPT
41

Highlights

We modeled the origin of mass wasting in Cretaceous carbonate rocks along a

prominent fault zone

Remote sensing data (Landsat TM, ASTER, SRTM) were used to correlate mass

T
wasting and tectonics

IP
Applied geomorphology, hydrogeology, hydrology, tectonics and engineering geology
are tools to monitor mass wasting

R
The anatomy of landslides is correlated with structural geology on different levels

SC
Zones of vulnerability are defined based on structural geology

NU
MA
D
P TE
CE
AC
 ACCEPTED MANUSCRIPT
42

AGE GROUP FORMATION FORMATION LITHOLOGY

(Masri, 1963) (Powell, 1989)
PALEOG.

Eocene Um Rijam Chert

T
Muwaqqar
Paleocene

IP
Muwaqqar Chalk
Mastrichtian Belqa

R
Phosphorite/
Alhisa with oysters
Campanian Amman

SC
Amman silicified
limestone Chert
L. CRETACEOUS

Santonian
G u d ra n Chalk

NU
Coniacian Ghudran

Phosphorite
Wadi Sir
Micrite
and sandy
MA

Sandy
phosphorite
Wadi Sir
Turonian
Ajlun Shuayb
D

Micrite and
Hummar marl
TE

Shuieb
Fuheis
Cenomanian
Hummar
P

Fuheis Marl and

Naur micrite
CE

Naur
E. CRET.

Fluvial and
AC

Albian Kurnub Baqa Subeihi marine

sandstones

Table 1
 ACCEPTED MANUSCRIPT
43

min mean max

SiO2 0.80 3.95 7.94
TiO2 0.01 0.05 0.10

T
Al2O3 0.25 0.88 1.98

IP
Fe2O3 0.15 0.43 0.90
MnO 0.00 0.01 0.01
MgO 0.64 4.45 11.27

R
CaO 35.28 47.13 54.32

SC
Na2O 0.01 0.01 0.02
K2O 0.01 0.02 0.05
P2O5 0.01 0.02 0.03
SO3 0.11 0.15 0.18

NU
Ba 7 26.0 37
Cr 5 8.0 14
Cu 15 38.0 58
MA
Sr 127 239.3 453
V 6 18.0 30
Zn 6 9.3 16
Zr 6 16.3 29
D
TE

Table 2
P
CE
AC
 ACCEPTED MANUSCRIPT
44

Fracture Orientation Movement Class/size Relative age Force

type
Rubble- NW–SE compression patchy incipient regional
strewn (abandoned distribution stage stress
areas structures) on the

T
surface

IP
measuring
several

R
square
meters

SC
across
Saw teeth NE–SW/ compression 1st order initial stage regional
fractures NNE–WSW  extension and 2nd stress

NU
order
fractures
Saw teeth approx. N–S lateral 2nd order initial stage regional
fractures movement fractures stress
MA
transitional
into straight
fractures
Straight WNW to NW vertical 2nd order intermediate gravitational
D

boundary movement fractures to late stage force

fractures
TE

Straight NE–SW vertical 1st order (early) late gravitational

scarps movement fractures stage force
block (and
P

rotational
CE

sliding)
Arcuate NE–SW vertical 1st order latest stage gravitational
scarps movement fractures force
AC

block and
rotational
sliding

Table 3