ArticlePDF Available

Abstract and Figures

Jerusalem was hit by earthquakes several times in its history, in the course of which none of the holy sites of the three main faiths of the western world escaped damage. Intensities of the last ML 6.2, July 11, 1927 Dead Sea earthquake, reached MSK VIII in the Old City of Jerusalem and the surrounding villages. As future strong earthquakes are inevitable, the need for the evaluation of earthquake-related hazards is obvious. Only general geotechnical properties of the section exposed in the mountainous area of Jerusalem are available; therefore, the hazard assessment was conducted from a geological perspective. The hazards identified in this study are: (1) amplification of seismic acceleration due to soft rock and soil conditions; (2) amplification due to mountainous topography; (3) dynamic instability of natural slopes; and (4) potential failure of slopes that have undergone engineering development and were weakened due to damaging, steepening, overloading, and wetting beyond their natural state. We formulated relative grades of vulnerability for each of the hazards and delineated the zones that require further specific investigation. For practical use we constructed a summary map that combines the different hazard categories. Looking at the summary map, the ground at the central N–S axis zone across Jerusalem is the least vulnerable. The bedrock there is mostly hard carbonate, the topography is mild, and thus only the alluvial cover, if thicker than 3m, should be considered sensitive. Yet although the natural hazard in this area is limited, the risk should not be underrated. Much of the city lies there, including buildings constructed before antiseismic codes were regulated, and traditional engineering practice should not be taken for granted as antiseismic proof either. Eastwards, the shear wave velocity (Vs) contrast between the hard and soft rocks as well as the notable topography in places, impose the potential for amplification. Slopes, either naturally or artificially cutting into the soft chalk, may expose the area to dynamic instability; thus, the ongoing extensive development of the city in this direction should certainly take into account all of this. West of the central axis, the potential of failure of both steep natural and urbanized slopes appears. Being a plausible direction for future urban expansion, these areas specifically call for careful environmental and engineering planning. For engineering purposes, however, a specific site investigation is still necessary. Nevertheless, the summary map established in this study sets up for Jerusalem, for the first time, a practical tool for environmental and municipal planning, emergency response planning, and civil protection. KeywordsJerusalem-Earthquake-related hazards-Ground amplification-Topographic amplification-Slope stability
Content may be subject to copyright.
Zones of required investigation for earthquake-related
hazards in Jerusalem
Amos Salamon Æ Oded Katz Æ Onn Crouvi
Received: 12 December 2007 / Accepted: 20 July 2009 / Published online: 4 September 2009
Springer Science+Business Media B.V. 2009
Abstract Jerusalem was hit by earthquakes several times in its history, in the course of
which none of the holy sites of the three main faiths of the western world escaped damage.
Intensities of the last M
6.2, July 11, 1927 Dead Sea earthquake, reached MSK VIII in the
Old City of Jerusalem and the surrounding villages. As future strong earthquakes are
inevitable, the need for the evaluation of earthquake-related hazards is obvious. Only
general geotechnical properties of the section exposed in the mountainous area of Jeru-
salem are available; therefore, the hazard assessment was conducted from a geological
perspective. The hazards identified in this study are: (1) amplification of seismic accel-
eration due to soft rock and soil conditions; (2) amplification due to mountainous topog-
raphy; (3) dynamic instability of natural slopes; and (4) potential failure of slopes that have
undergone engineering development and were weakened due to damaging, steepening,
overloading, and wetting beyond their natural state. We formulated relative grades of
vulnerability for each of the hazards and delineated the zones that require further specific
investigation. For practical use we constructed a summary map that combines the different
hazard categories. Looking at the summary map, the ground at the central N–S axis zone
across Jerusalem is the least vulnerable. The bedrock there is mostly hard carbonate, the
topography is mild, and thus only the alluvial cover, if thicker than 3 m, should be
considered sensitive. Yet although the natural hazard in this area is limited, the risk should
not be underrated. Much of the city lies there, including buildings constructed before
antiseismic codes were regulated, and traditional engineering practice should not be taken
for granted as antiseismic proof either. Eastwards, the shear wave velocity (Vs) contrast
between the hard and soft rocks as well as the notable topography in places, impose the
potential for amplification. Slopes, either naturally or artificially cutting into the soft chalk,
may expose the area to dynamic instability; thus, the ongoing extensive development of the
city in this direction should certainly take into account all of this. West of the central axis,
the potential of failure of both steep natural and urbanized slopes appears. Being a
Electronic supplementary material The online version of this article (doi:10.1007/s11069-009-9436-6)
contains supplementary material, which is available to authorized users.
A. Salamon (&) O. Katz O. Crouvi
Geological Survey of Israel, 30 Malkhe Israel St, Jerusalem 95501, Israel
Nat Hazards (2010) 53:375–406
DOI 10.1007/s11069-009-9436-6
plausible direction for future urban expansion, these areas specifically call for careful
environmental and engineering planning. For engineering purposes, however, a specific
site investigation is still necessary. Nevertheless, the summary map established in this
study sets up for Jerusalem, for the first time, a practical tool for environmental and
municipal planning, emergency response planning, and civil protection.
Keywords Jerusalem Earthquake-related hazards Ground amplification
Topographic amplification Slope stability
1 Introduction
Strong earthquakes rocked Jerusalem and its environs several times along its history, the last
of which occurred in 1927, along the Dead Sea Transform (DST) in the northern Dead Sea
about 30 km southeast of Jerusalem, and reached M
6.2 (Shapira et al. 1993; Avni 1999).
These earthquakes caused loss of life and damage on both sides of the DST and since this is
an active plate boundary, such strong earthquakes are expected to strike the region again.
According to the current antiseismic Israeli Building Code (IBC-413, 1995), there is a
90% probability for Jerusalem that no horizontal acceleration higher than 0.1–0.15 g will
occur in the next 50 years. The Code relates to rock conditions (shear wave velocity of
620 m/s) and flat topography, with some corrections for soil type, but ignores other factors
that may intensify the effect of the seismic waves, such as the topography and dynamic
(earthquake induced) slope failure. Only a few site-specific examinations have been
conducted so far in Jerusalem, mainly for producing response spectra for private projects.
The data is restricted, and is anyhow not sufficient for a quantitative microzone mapping.
Here, we aim to bridge the gap and characterize, define and map the potential of the
earthquake-related hazards in Jerusalem that are beyond those addressed by the present
Methods of hazard evaluation are based on a quantitative or a relative assessment (e.g.,
Ziony 1985; Monahan et al. 2000), focus on the potential risk (e.g., Erdik et al. 2003),
examine the expected effects of a given scenario on the areal infrastructure and lifelines
(e.g., Toppozada et al. 1994), and more. There are also detailed guidelines for evaluation
and mitigation of seismic hazards (e.g., California Geological Survey [CGS], SP-117).
Unfortunately, while the available methods strongly rely on dense measurements,
intensive data collection in the area of Jerusalem is not foreseen. To bridge the gap we
developed hazard analysis, which relies on the geology of the area, in light of the existing
methods. Based on a qualitative approach, we analyzed each of the relevant hazard factors
according to the local geology and constructed a series of thematic maps that point to the
vulnerable areas in the city. The findings are all summarized in a map that combines all the
hazard categories and the map defines the zones that require investigation of the earth-
quake-related hazards. Thus, the summary map becomes a guide for conducting site-
specific examinations as well as a tool that enables the city planners to evaluate and
mitigate these hazards.
2 Geological background
Jerusalem is located on the northern part of the Judea Mountains (Fig. 1), from around
420 m above sea level at the lowest seasonal stream channels to about 930 m at the
376 Nat Hazards (2010) 53:375–406
mountain peaks, on both sides of the regional watershed line that separates the Mediter-
ranean and the Dead Sea catchments.
Continental breakup and uplift are the two main geological processes presently active in
the region. The breakup separates the Arabian plate in the east from the Sinai subplate in
the west (see Garfunkel 1981; Garfunkel and Ben-Avraham 2001, for background and
references). It has accommodated a sinistral offset of about 105 km between these plates
since the Early Miocene (e.g., Quennell 1959) and is thus a continental transform. This is
the most active fault system closest to Jerusalem, with a relative velocity of about 0.5 cm/
year in long geologic terms (e.g., Freund et al. 1970) and about 0.4 cm/year for recent
times (Wdowinski et al. 2004). Coupled with the shear along the southern part of the DST
is a small component of transverse extension between Arabia and Sinai–the Jordan and the
Arava valleys (Fig. 1) (Garfunkel 1981; Salamon et al. 2003). The Dead Sea basin, located
in-between the two valleys, is considered a large pull-apart, rhomb-shaped graben.
Associated with the breakup is the uplift of the margins of that rift (Wdowinski and
Zilberman 1997), including the Judea and Samaria mountains west of it, where Jerusalem
is located. The last stage of this took place after the Pliocene and reached an amplitude of
about 200 m., that is, a rate of about 0.01 cm/year (Begin and Zilberman 1997). There is
insufficient evidence to determine that the uplift continues nowadays; however, since it
seems to be associated with the DST, we may assume so. The following background briefs
the geology of the study area.
Jerusalem is located on top of a broad, uplifted anticlinorium that trends north–south,
parallel to the DST, widely referred to as the Judea and Samaria mountains. The structure
is about 70 km wide and stretches from the Judea foothills in the west to the DST in the
east (Picard 1943). The most prominent secondary folds of that anticlinorium are the
Ramallah and Hebron monoclines that trend NNE in a left en-echelon arrangement. The
saddle between the two monoclines is where Jerusalem is situated (Fig. 1).
Two main fault systems are recognized in the study area: (1) a NNW-SSE system,
mainly in the southwest, in which some of the faults are long with displacements of up to
about 50 m; and (2) W-E faults, north and northwest of the city, with displacements that
may reach several tens of meters (Israeli 1977). Some of these faults were marked on
previous maps as displacing alluvium (Arkin 1976; Israeli 1977) and may therefore be
mistaken as active. However, they actually reflect the mapping methodology at the time
which did not consider the alluvium as a potential indicator for active faulting (Y. Arkin
and A. Israeli, personal communication, 2006). Bartov et al. (2002) mapped the suspected
active faults in Israel and did not single out any in the Jerusalem area as such.
The rocks in the Jerusalem area are intensively fractured. The fractures are mostly
vertical, some are closed and some are open with a soil and clayey fill or secondary calcite
crystal growth of karst origin (Israeli 1977).
Fig. 1 Location maps: a plate tectonic setting. Note the century-long M
C 4 seismicity (from Salamon
et al. 2003), including the M
6.2, July 1927 epicenter, in relation to Jerusalem (gray square) and to the Dead
Sea Transform (DST), Cypriot Arc (CA) and Suez Rift (SR) plate boundaries. Jerusalem is about 25 km west
of the DST, in northern Judea (Ju), south of Samaria (Sa). b Main structural elements. The study area (gray
square) is located between the Hebron and Ramallah anticlines. C Carmel fault, G Galilee, J Jericho, SG Sea
of Galilee. c Jerusalem area in and around the Old City, from the geological map of Arkin (1976). See map
limits in Fig. 3 and symbols of geological unit in Fig. 2. White circle outlines the Old City area. Localities: 1
el-Eizariya (el-Azarije), 2 Holy Sepulcher, 3 Mt. of Olives, 4 Mt. Scopus, 5 Temple Mount, Dome of the
Rock (Omar) and al-Aqsa mosques, 6 el-Eizariya slump (Fig. 7)
Nat Hazards (2010) 53:375–406 377
378 Nat Hazards (2010) 53:375–406
2.1 Stratigraphy and lithology
Knowledge of the near surface geology is required in order to evaluate the potential of
seismic amplification due to local site conditions and dynamic slope stability. Most of the
rock-section exposed in and around Jerusalem (e.g., Arkin et al. 1965; Roth 1969; Braun
1970; Israeli 1977; Lewy 1989, 1991; Gill 1996) is composed of hard limestone and
dolomite rocks and a few marly and clayey units, all part of the Judea Group (JG) of
Albian–Cenomanian–Turonian age. The JG is overlain by soft chalk, marl, and chert of the
Senonian–Maastrichtian–Paleocene Mount Scopus Group (MSG). In the subsurface, below
the JG, there is a thick unit of mainly loose sands of the Kurnub Group of Lower Creta-
ceous age (Mimran 1969, 1995). Figure 2 presents the litho- and chrono-stratigraphy of the
formations that are included in the JG and MSG. The geological map of the area was
compiled by Arkin (1976) on a 1:50,000 scale.
The uppermost cover consists of alluvium of Quaternary age, and appears in three modes:
(1) deposits of silt, clay, and local rock fragments and gravels up to several meters thick in
seasonal stream channels and low areas; (2) dark brown, compact clay with some stone
fragments, up to 10 m thick; and (3) ‘Rubble soil’’ (Israeli 1977) of anthropogenic origin,
Fig. 2 Stratigraphy and geotechnical units exposed in the Jerusalem area, from Arkin (1976) and Israeli
(1977), respectively
Nat Hazards (2010) 53:375–406 379
composed of silt and clay with artificial fragments of stones and shards originating from the
wreckage of destroyed structures and dumps that have accumulated in the course of Jeru-
salem’s long history. In places, the material is likely to reach a thickness of 10 m and is
common in the area of the Old City and the Temple Mount and its environs (Fig. 1c).
The common soils are Terra–Rossa (American classification: Xerochrepts, FAO:
Luvisols and Cambisols) that developed on the limestone and dolomite units and penetrates
karst cavities and open fractures to a depth of several meters, and Rendzina (American:
Haploxerolls, FAO: rendzinas) that developed over chalk and marl units. Calcrete, (lime
crusts, caliche, locally termed ‘Nari’), 1–3 m thick, which is a variegated hardened pet-
rocalcic horizon, composed of rock fragments and soil cemented by calcium carbonate, is
abundant on the soft calcareous rocks, mainly chalk or chalky limestone outcrops, and in
places underneath the Rendezina soil (Dan 1977).
The transition between the hard JG carbonates and the soft MSG chalks and marls
constitutes a N–S trending structural and morphological marker in the regional geology,
Fig. 3 Map of the geotechnical units in the study area, compiled and modified from Israeli (1977) and
Arkin (1976). See the correlation between the geotechnical units and the stratigraphy in Fig. 2. The square
shows the limits of map 1C
380 Nat Hazards (2010) 53:375–406
just east of the Old City (Fig. 3). Eastwards, the landscape is characterized by rounded hills
and, to a large extent, covered by calcrete.
2.2 Geotechnical mapping
Geotechnical mapping in Jerusalem was carried out by Israeli (1977) on a 1:12,500 scale.
Other geotechnical studies were limited to specific sites (e.g., Arkin 1984; Arkin et al. 1993,
1994; Gilat et al. 1992; Michaeli and Arkin 1994; Mimran 1995). Israeli (1977) divided the
exposed section into six units on the basis of the following parameters: uniaxial compressive
and tensile strength (where very low, low, medium, high, and very high grades represent the
values\28 MPa, 28–56 MPa, 56–112 MPa, 112–225 MPa, and[225 MPa, respectively);
change of strength (sensitivity) with water (moisture) content; Young modulus; density;
bearing capacity; permeability; bedding; karst and development of soil. The resulting
geotechnical units and the formations included in it are described in Table 1, presented in
the stratigraphic section in Fig. 2 and in the map of the study area (Fig. 3).
3 Earthquakes affecting Jerusalem
The earliest notice of a seismic event in Jerusalem was described in the Bible (Zechariah
14: 4–5) and it left its mark at that period (Amos 1:1, The Holy Scriptures, 1942, Jewish
Publ. Society, Philadelphia): The words of Amoswhich he saw concerning Israel in the
days of Uzziah king of Judah, two years before the earthquake.’ (e.g., Bentor 1989;
Ben-Menahem 1991). Yet a careful examination of this event by Ambraseys (2005)
yielded the conclusion that ‘We could find no direct or indirect evidence that Jerusalem
was damaged’’ !
Since in the course of history damaging earthquakes did hit Jerusalem and its vicinity,
there is a need to reconstruct the actual effects and damage that occurred there. This is not
a simple task, as Ambraseys (2005), after Ambraseys and Karcz (1992), has already shown
in regard to the 1546 earthquake:
Table 1 The geotechnical units in the study area
Unit Geotechnical properties Formations included
I Very high to medium strength and low
sensitivity to water
Kefira [Kk], ‘Amminadav [Kua], Weradim
[Kuw], Bi’na [Kub]
II Medium to low strength and low sensitivity
to water
Giv’at Ye’arim [Kugy], Soreq [Kus],
Kesalon [Kuke], Bet Me’ir [Kubm]
III Medium to very low strength, in parts with
high sensitivity to water
Kefar Sha’ul [Kuks] Lower Menuha [Kum
IV High strength and low sensitivity to water,
but highly fractured (chert)
Mishash [Kumi]
V Soft rocks of very low strength and high
sensitivity to water
Moza [Kumo] Upper Menuha [Kum
VI Loose material of very low strength and high
sensitivity to water
Alluvium, soils, Nari
Although not classified by Israeli (1977), the Nari was also included in this unit. In places it appears as a
compact and strong unit, but in general the Nari is heterogeneous, varies highly in composition and
thickness with open cracks filled with clay and soil. Uniaxial compressive strength, as determined by
Schmidt Hammer (Israeli 1977), shows that the Nari is the weakest lithological unit in the area
Nat Hazards (2010) 53:375–406 381
it is a typical example of how widely the effects of an earthquake can become known
and also magnified, not so much because they were very serious but chiefly because of
the desire of contemporary writers to draw theological and political morals from a
natural disaster, particularly when the earthquake happens to be in the Holy Land.
Likewise, Shalem (1949) arrived at the conclusion that Jerusalem was relatively less
damaged than other nearby cities, and that its unique status and long historic record gave
the false impression of its being exceptionally vulnerable.
We compiled preliminary lists of historic events that damaged Jerusalem and of historic
events that were only felt (Electronic Supplementary Material [ESM] 1) or thought to have
caused damage (ESM 2), based mainly on studies that directly evaluate original contem-
poraneous reports (e.g., Guidoboni et al. 1994; Guidoboni and Comastri 2005). In light of
Ambraseys (2005) conclusions, further detailed examination is required.
Overall, the most damaging earthquakes in Jerusalem were those of 363 A.D. and 746 A.D.,
for whichthe chroniclesreportthatmore than half ofthecitywas destroyed, whereas damage and
casualties from other events were limited. Interestingly enough, the Temple Mount, the largest
structure ever built in the Holy Land, which is an elevated 480 9 280 m platform on which the
Second Temple was built as well as the Dome of the Rock and the al-Aqsa mosques, has been hit
several times but the outer walls and the major underground halls were not too severely damaged
from the time that King Herod built it two millennia ago. The original al-Aqsa mosque, however,
did not survive past quakes and the present structure, which is from medieval times, has already
undergone several repairs of seismic damage (Avni and Seligman 2001).
The sparse seismicity recorded during the last 25 years in and around Jerusalem (GII
(Geophysical Institute of Israel) 2006) has not reached M
[ 3, and no M
[ 4 events
have occurred there during the last century (Fig. 1a). In contrast, numerous earthquake foci
were observed in the DST, 25 km east of Jerusalem. Avni (1999) studied the strongest of
the recorded events (in 1927, M
6.2) and noted that ‘The number of casualties from the
earthquake in Eretz–Yisrael (Palestine at the time) and Jordan together came to 285 dead
and around 940 injured. It is the first destructive earthquake in the region that was
instrumentally recorded, and it enabled us to cross-check historic documentation and
modern quantitative data.’ In Sect. 5, we examine to what degree the damage caused in
Jerusalem and the surrounding villages from that event can be correlated with the geology.
The ongoing seismicity along the DST near Jerusalem (GII (Geophysical Institute of
Israel) 2006) as well as morphotectonic evidence (Belitzky 1996), including the presence of
neotectonic faults (Bartov et al. 2002), all indicate present activity of the DST. Most of the
significant events in this area were concentrated in a narrow belt along the transform fault
(Salamon et al. 1996) with only a few M
C 4 off the belt. Accordingly, that seismogenic belt
releases most of the seismic moment, although the total sum seems to account for only a part
of the overall accumulated motion (e.g., Jackson and McKenzie 1988; Salamon et al. 1996).
Earthquake mechanisms of the DST close to Jerusalem are left-lateral with some extension
(Salamon et al. 2003, and references therein) and most of the seismicity and the seismic
moment within that part are generated within the Dead Sea basin. Thus, the DST is the major
seismogenic structure in the region and the closest source of strong earthquakes to Jerusalem.
4 Delineating the zones of required investigation for earthquake-related hazards
Assessment of the zones within the city of Jerusalem that are expected to be affected
during a strong earthquake by each of the relevant hazard factors is described below and
382 Nat Hazards (2010) 53:375–406
outlined in Fig. 4. Since site-specific examination of earthquake-related hazards should be
based on in situ and quantitative measurements, our analysis is limited only on to delin-
eating the zones that require site-specific investigation. The analysis was based on char-
acterization of the geological and geotechnical units, as they appear in Arkin (1976) and
Israeli (1977) maps, respectively.
The maps were prepared as GIS layers of data and cover the municipal area of Jeru-
salem, which is about 150 km
. The maps were scanned, rectified according to the new
Israeli grid, digitized, and compiled, yielding a polygonal vector layer of the geological
and geotechnical units (Fig. 3). For the topography we used the Digital Terrain Model
(DTM) of Hall (1993), with a 25 m spatial resolution. According to Hall et al. (1999), the
horizontal error for this DTM ranges between 10 and 50 m, whereas in the vertical
dimension, the error is less than 5 m for about 80% of all the data, and less than 10 m for
95% of the data.
The information provided through the evaluation of the geotechnical data was then
combined to form a summary map that defines the zones requiring further site-specific
investigation. Thus, it was able to microzone the potentially vulnerable areas expected to
be affected by the various earthquake-related hazards within the city of Jerusalem.
4.1 Ground amplification of seismic accelerations
In its most simple appearance, amplification occurs when there is a considerable decrease
in shear wave velocity (Vs) toward the surface layers (e.g., Dobry et al. 2000; Kawase
2003). In terms of general field definitions, amplification occurs when a soft rock or soil
overlies a hard rock, but practically, the specific response of a given site may vary widely
and is difficult to predict (Boore 2004). Zaslavsky et al. (2003) and Gvirtzman (2004) have
shown examples in Israel where this effect is emphasized in areas of a sharp velocity
contrast between the two upper units, including in places where the stratigraphy is similar
to that of Jerusalem. Specifically, Gvirtzman (2004, Fig. 23) shows that the resonance
amplification in the case of class C (MSG in Jerusalem area) above class A (JG in
Jerusalem), increases from 3 to 7, for a thickness range of the overlying unit (class C) that
decreases from 200 to 20 m, respectively. These factors are higher than the site coefficients
GeologyDTM Slopes
Potential of Amplification:
Lithology Topography
Potential of Slope Failure:
Natural Urbanized
Summarizing Map
Data Base
(GIS layers)
Fig. 4 Methodology of the earthquake hazard evaluation in Jerusalem
Nat Hazards (2010) 53:375–406 383
suggested by both the IBC-413 and NEHRP (1997). Nevertheless, the combined effect of
having the IBC-413 based on too low Sv bedrock (620 m/s, which results in overestimating
the PGA) and too low amplification coefficients (results in underestimating the expected
resonance effect), is yet to be examined.
Shear wave velocities of a given type of rock vary broadly due to the extent of
weathering, intensity of fracturing, depth below the surface, and more. Specific factors
common to Jerusalem are karst, natural pits, and ancient artificial subsurface cavities,
which sometimes reach several meters in depth and are usually filled with alluvium, soil or
artificial fill, but are open in places. Although these factors are typical of certain geological
formations (Israeli 1977), they change locally and cannot be predicted. Therefore, the exact
seismic velocities in the area of Jerusalem cannot be determined without in situ measuring
(Shtivelman 1996; and V. Shtivelman, personal communication, 2002).
So far, only a few surveys have been carried out in Jerusalem to determine seismic
velocities (Ronen 1995; Wachs and Levitte 1983; Ezersky 2001) and specific site-effects.
The data does not systematically cover the study area or the stratigraphic section, and
therefore is not sufficient for detailed mapping (microzoning) throughout the city. The
Fig. 5 Potential for ground amplification. Units should be more than 3 m (10 ft) thick in order to be
sensitive to amplification (NEHRP 1997)
384 Nat Hazards (2010) 53:375–406
Table 2 Correlation between site-class definitions and rock units in Jerusalem
Site class definitions
From NEHRP (1997) and Wills et al. (2000)
Rock units in and around Jerusalem
Class Vs30 m/s SPT Lithology Ramla–Lod Vs (m/s) from
Zaslavsky et al. (2001)
Geotechnical units
from Israeli (1977)
Lithology (Formations)
[1500 Hard rock
Dolomite and limestone,
JG 1100–2300
I–II Dolomite & limestone
(Kk, Kugy, Kuke, Kubm,
Kua, Kuw, Kub)
760–1500 Rock II–III Marly carbonate (Kuks*, Kus)
555–1000 Marl and chalk, MSG 400–1000 III–IV Chalk (Kum1, Kumi)
360–760 [50 Soft rock and
very dense soil
Clay (Yafo group) 300–620 V Marl (Kum2, Tlt**)
270–555 Marl, alluvium, loam, sandy
limestone 200–1170
V–VI Clayey marl (Kumo***, Al)
180–360 15–50 Stiff soil VI Alluv., soils, rubble soil (Al)
\180 \15 Soil
‘Soft’ soils
* Ezersky (2001), ** Tlt Fm., geological map of Arkin (1976), *** SPT in Kumo (A. Israeli, pers. comm., 2000): weathered limestone 38, [50; clay [25, 32
Nat Hazards (2010) 53:375–406 385
current IBC-413 allows for a maximal amplification of up to a factor of 2 due to the soil
type, but ignores the possible effect of Vs resonance in the shallowest layers and disregards
the damaging frequencies. The new version proposed to update that code, relies on NEHRP
(1997) with coefficients higher than those of the present IBC-413, but still there is a need to
verify that the new coefficients do satisfy the resonance effect. Given the above-mentioned
limitations and difficulties, and until sufficient data is collected, only a semiquantitative
evaluation of the amplification potential can be made.
First, we assigned Vs to the various geological units that crop out in Jerusalem
(Table 2), according to the following data and principles:
1. Velocities (Ronen 1995; Ezersky 2001) and Standard Penetration Test (SPT, A. Israeli,
personal communication, 2000) obtained from seismological and geotechnical
investigations made in Jerusalem.
2. Vs30–the average Vs in the top 30 m, in accordance with NEHRP site class definitions
(NEHRP 1997, para.; Wills et al. 2000). This is based on accumulated
experience and measurements in California.
3. SPT values that relate to Vs30 (NEHRP 1997).
4. The relative grading of the geologic formations in Jerusalem according to their
geotechnical characteristics defined by Israeli (1977) (Sect. 2.2, Table 1 and Fig. 2).
5. Vs in the Ramla and Lod municipal area, about 25 km west of Jerusalem. Although
some of these measurements relate to the subsurface only, this is the closest area to
Jerusalem in which microzoning based on specific site-response surveys was carried
out (Zaslavsky et al. 2001), and the geologic section there is more or less similar to
that of Jerusalem.
With the assigned Vs, it was possible to determine the velocity contrast between each
configuration of the two uppermost units that occur in the study area. The degree of
contrast that predicts the resonance effect enabled us to define the potential of amplification
Table 3 Relative Vs gradient between two upper rock units
Upper (shallowest) unit Lower unit Relative vs contrast
potential for
Rock unit** Site class (Vs30) Rock unit Site class (Vs30)
Kumo CD
Kubn AB
Judea Gr.***
Mt. Scopus Gr.
Rubble soil (Al) D
Judea Gr.*** AB
Mt. Scopus Gr.
(Kum, Kumi)
Judea Gr. (Kub) AB
No potential
Tlt C
Kumi BC
Judea group*** AB
Judea Gr.*** AB
The relative velocity contrast between the two upper rock units is defined here as the Potential for
Amplification. The uppermost unit is the one exposed on the surface and the lower unit is the unit below it.
Site Classes is according to Table 2
* See text for explanation
** If thickness is [10 ft (3 m) (NEHRP 1997, See text
*** All JG sub-units except Kumo (marl)
386 Nat Hazards (2010) 53:375–406
and rank it in the following way: a relatively small difference of one Vs30 site-class grade,
(e.g., between A’ to B’, or AB’ to BC’) was defined as ‘low’ potential, and a difference of
two grades (e.g., A’ to C’, or AB’–CD’) was defined as ‘moderate’. Larger contrasts (of
three grades) were not found (Table 3).
Due to the original deposition or erosion, the thickness of the surface units varies from
place to place and this value is not evident from the geological map. Therefore, we follow
NEHRP (1997, and recommend that a minimum thickness of 3 m (10 ft) is needed
for a unit to be considered sensitive to amplification, especially for all types of alluvial and
soil materials.
In the last stage we marked the potential of each of the outcropping geological for-
mations to amplify seismic accelerations on a map (Fig. 5). Maximum potential appears in
exposures of alluvial materials and marl (Moza Fm.). In NEHRP (1997) terms, this should
be paralleled mostly with class D, and it means that the highest potential in Jerusalem
should be regarded as a ‘medium’ level only. The artificial rubble soil, however, is a
unique unit (Katz and Crouvi 2007) and in our opinion should be suspected as class E or F
for engineering purposes. Nonvanishing (‘Low’) potential was attributed to chalks and
marls of the MSG, mainly due to its stratigraphic setting above the hard carbonates of the
4.2 Topographic amplification
Records of strong ground motion (Sepu
lveda et al. 2005) as well as of weak motion
(Stewart et al. 2001), mainly of shear waves, have shown that seismic accelerations at tops
of mountains and cliffs are stronger than those in the nearby valleys. The amplification
does not depend on the internal geologic structure or soil type of the relief [overview by
Geli et al. (1988), and references therein, and observations in Israel by Zaslavsky and
Shapira (2000a, b)], although the total could be a combined effect of all the factors. The
greater the ratio between the height of the mountain peak above its surroundings and its
width, the higher the amplification is.
So far there is no strong motion data for Jerusalem to approve the effect of surface
topography on site amplification, yet the topography in the area suggests that the contri-
bution of this effect to hazard should be considered. Zaslavsky Y. (personal communi-
cation, 2002, 2008) proposed that the amplification found in site-response surveys, which
were conducted on a steep hill (Holy Land project) and elsewhere in the city, is due to the
topographic effect. This was inferred from weak motion of remote moderate earthquakes,
as well as from background seismic noise. On the other hand, investigations of recordings
taken by the seismographic station of Jerusalem (JER) which is located on a moderate hill
(Giv’at Ram), showed no amplification at all (Y. Zaslavsky, personal communication,
Regarding Jerusalem, we presume the following: The range of Vs in the JG, which
forms the base of all the local mountains, is 1,100–2,300 m/s (Zaslavsky et al. 2001) and
the range of frequencies that may put man-made structures at risk is 0.5–10 Hz. Assuming
that amplification occurs when the width of the topographic structure is similar to the
length of the seismic shear wave, then hills 100–5,000 m wide would be most vulnerable.
Maximum amplification occurs when the seismic wavelength is similar to the width of the
topographic structure and when the relation between the height of the structure and half of
its width is higher than 0.2 (Geli et al. 1988). This means that the width and height of the
vulnerable mountains would be in the range of 100–5,000 m and 10–500 m or higher,
Nat Hazards (2010) 53:375–406 387
Amplification also occurs on cliffs and slopes steeper than 60 (Ashford and Sitar 1997;
Ashford et al. 1997), particularly on top of cliffs with heights (H) that equal one-fifth of the
seismic wave length (k). Maximum amplification occurs in frequency (f) of:
f ¼ Vs=5H: ð1Þ
Given that cliffs in and around Jerusalem are formed by JG rocks, the minimal height of
a cliff exposed to this effect (Vs and damaging frequencies as mentioned above) would be
around 20 m and higher.
Still, these parameters cannot be simply transformed into mapping criteria because of
the yet unexplained discrepancy between the models and the actual observations (Geli
et al. 1988). Also, the role of asymmetry and shape of the mountain and the relevant height
level of the mountain at which the topographic effect becomes a significant factor are thus
far unknown.
Despite the afore-mentioned, we qualitatively delineated the ridges that are prominent
relative to their surroundings, such as Mt. Scopus and Mt. Gillo, and the cliffs that are
Fig. 6 Potential for topographic amplification. Note the zones vulnerable to both topography and lithology
effect on the amplification. Localities: 1 Ben Hinnom Valley, 2 seismographic station of Jerusalem (JER), in
Giva’t Ram, 3 Holy Land project, 4 Mt. Gillo, 5 Mt. Scopus, 6 Nahal Qidron
388 Nat Hazards (2010) 53:375–406
20 m and higher, such as in the valleys that encircle the Old City from around the south
and east (Ben-Hinnom and Qidron valleys) (Fig. 6). The result is a preliminary map that
will certainly need further refinement. Yet in its present form, it stresses the need to
quantify this factor and incorporate it into any comprehensive seismic hazard evaluation.
4.3 Dynamic slope stability
Slope failure is one of the primary seismic hazards (e.g., Keefer 1984). For Jerusalem we
followed the recommended procedure of the California Geological Survey (CGS, SP-117,
1997), with necessary modifications. It is based mainly on screening the areas sensitive to
seismic slope failure by reviewing previous studies, analyzing air-photographs and field
work and on quantifying the potential for landslides during a strong earthquake by defining
sensitivity grades.
The screening is aimed at finding indicators for slope failure, while areas with no such
signs can be regarded as having negligible potential. The indicators are:
(1) Evidence of past or presently active landslides: It is assumed that slopes that are
statically unstable or slopes that failed in past earthquakes will most probably be
reactivated by a future strong earthquake. While typical failure through a rotational
(slump) mechanism or rockslide such as hummocky structure, scars, landslide bodies
and traces were pointed out in the field and on air-photos (Wachs and Levitte 1983,
1984; Wachs et al. 1990; Gilat et al. 1992; Arkin et al. 1993), direct evidence for
current active failure was not observed.
Overall, we looked at the morphology and slope gradient of sites that were suspected
or mentioned in previous studies as potentially unstable (ESM 3): local collapses in
the highly fractured dolomite and marl layers of the Soreq Formation, in most cases
in cut-slopes; slides of large dolomite blocks of the ‘Amminadav Formation above
the marl of the Moza Formation; soil creep above the calcrete cover; slides and
detachment of calcrete slabs over the chalk of the Menuha Formation; and slides of
rubble soil recently dumped near the eastern walls of the Old City.
(2) Geologic formations and soil materials sensitive to slope failure: This is based on the
geotechnical properties (Israeli 1977). Slopes composed of hard limestone and
dolomite of geotechnical unit (GT) I to III were defined as not sensitive (Table 4).
Included here are the Kefira, Giv’at Ye’arim, Kesalon, ‘Amminadav, Kefar Sha’ul,
Weradim, Bi’na and the lower member of Menuha formations.
Slopes composed of geotechnically weak rocks such as chalk and marl of GT IV–VI,
were defined as sensitive to failure. This includes the Moza, Upper Menuha, and
Mishash formations, as well as alluvium, soil materials, and Nari that were graded the
weakest. The ‘strong’ Soreq and Bet Me’ir formations that were originally graded as
GT II, were suspected as ‘sensitive to slope failure’ because of the presence of thin
marl layers and indeed, in the frame of this work and that of Katz (2004), evidence for
landslides were found in the Soreq Formation.
(3) Landscape indicators for sensitive slopes: Keefer (1984) characterized the gradients
of slopes that failed in earthquakes by the type of failure, and found that rocks and
soils failed at a moderate slope gradient of 15 and 5, respectively, whereas rock-fall
(i.e., free fall of rock blocks) occurred at slope gradients of 40 and over. These
characteristics were then projected onto the conditions in Jerusalem. Since GT VI
consists of soils and GT IV–V consist of weak rocks that tend to lose strength in the
Nat Hazards (2010) 53:375–406 389
presence of water (Israeli 1977), it is assumed that at a gradient higher than 5,GT
IV–VI might be sensitive to failure, possibly by a slide mechanism.
The most important finding in Jerusalem is the steepest place where the Menuha
Formation outcrops at a site near el-Eizariya (A. Gilat, personal communication). The
morphology there resembles a large landslide (slump), 350 m long and 150 m wide,
with a scar as high as about 10 m (Fig. 7a). The gradient of the adjacent intact slopes
reaches around 30, whereas in other places the gradients are a little over 20
(Fig. 7b). The slope of the landslide body, however, is less than 20, while at the front
of the landslide, over the toe, in the areas that did not slump, the slope increases to
20–30 and shallow rockslides appear at the base of the calcrete cap. From this, we
estimate empirically that rotational landslides in the Menuha Formation are likely to
occur at gradients higher than 30, whereas shallow rockslides at the base of the
calcrete are likely to develop at slope gradients of 20–30. The existing data,
however, does not enable us to determine whether the suspected site failed during a
static state or at the time of an earthquake.
(4) Hydrological conditions that may destabilize slopes: We found no specific natural
conditions or effects that might decrease dynamic slope stability, except at times in
winter, which is the wet season in Jerusalem (average precipitation is about 600 mm/
year in the western part and 400 mm/year in the eastern part).
Table 4 Failure style and sensitivity of rock units exposed in natural slopes in Jerusalem
Rock unit Style of potential failure Sensitivity to
slope failure
Formation Lithology Geotech. unit
Al Rubble soil
Soil, clay
Creep, rotational
Creep, rotational
Rock fall, block detachment
Kumi Chert, chalk, marl IV Sensitive
Kum2 Chalk, marly chalk V Shallow rock slides in slope
gradient of 20–30
Rotational slides in slope
gradient of [30
Kum1 Chalk, limy chalk III Not sensitive
Kub Limestone I Not sensitive
Kuw Dolomite I Not sensitive
Kuks Clayey limestone, chalk III Not sensitive
Kua Dolomite, limestone I Not sensitive
Kumo Marl, limestone V Slides in natural and cut
Kubm Dolomite, some marl II Not sensitive
Kuke Dolomite II Not sensitive
Kus Dolomite, marl II Old natural rotational slides,
shallow slides and rock
falls in cut slopes
Kugy Dolomite II Not sensitive
Kk Limestone, dolomitic limestone II Not sensitive
390 Nat Hazards (2010) 53:375–406
Fig. 7 The slump near el-Eizariya. a Southward view of the slump showing typical Scar, Body, and Toe
morphology, b Topographic map of the slump area (Scar, Body, and Toe are marked) showing calculated
slope gradients (0–20: no color, 20–30: light gray,30\ : gray). Direction of view of 7a is also shown
Nat Hazards (2010) 53:375–406 391
(5) Slope failure as a result of anthropogenic activity is known from exposures of the
middle part of the Menuha Formation. Static failure observed in a few places,
probably due to artificial steepening and overloading (U. Saltzman, personal
communication, 2002; D. Wind, personal communication, 2002), and dynamic
failure were reported in cut-slopes after the 1927 earthquake (Avni 1999).
Following Keefer (1984) and the findings from the screening process, we define four
sensitivity grades for dynamic failure of natural slopes in Jerusalem (Fig. 8). The slopes
were determined from the DTM by using the ArcGIS slope function. According to Jibson
et al. (1998), the gradient calculated for very steep slopes ([60) is usually lower than the
real slope, but such slopes are rare in the study area.
(1) No sensitivity: stable slopes. The upper limit was empirically determined at 5 in soft
rocks and 10 in hard rocks, and this sets the lower threshold for seismogenic
(2) Low sensitivity: this is the lowest range of slopes in which landslides may occur.
(3) Moderate sensitivity: here, shallow soil- or rockslides are possible during earth-
quakes. The lower boundary is estimated as 20 in soft rocks and 30 in hard rocks.
(4) High sensitivity: Potential for rotational slumps in soft rocks in slopes steeper than
30; and free fall of blocks of hard rocks in slopes larger than 40 (this last condition
hardly exits in the study area, except for the cliffs south and east of the Old City).
Using these grades, we produced the sensitivity map for dynamic failure of natural
slopes (Fig. 9). All in all, the sensitive areas are outcrops of formations with evidence of
slope failure (Soreq, Moza and Upper Menuha formations, and alluvium and soil mate-
rials); and formations built of weak rocks of GT IV–VI (Moza, Upper Menuha and Mis-
hash formations, and alluvium and soil materials). No evidence for slope failure in the Bet
Me’ir Formation was found and, therefore, we assumed that its failure potential is low.
It is important to note that ground or topographic amplification may further decrease the
dynamic stability of slopes, and therefore increase the hazard. In fact, this was already
Fig. 8 Sensitivity grades for dynamic natural slope failure
392 Nat Hazards (2010) 53:375–406
taken into account, for all weak GT units that we considered more sensitive to failure
(Fig. 8) are soft rock with low Sv (Tables 1, 2, 3, 4, 5), and the low Sv rocks also have the
potential for amplification (Tables 2, 4, 5). Future studies, however, will have to estimate
quantitatively how much the horizontal acceleration reduces the dynamic stability of the
Generally, most of the area along and around the N–S center axis of Jerusalem is of no-
to low- sensitivity. Medium sensitivity can be identified southeast of the city within the
Nahal Qidron basin, and west of the city within the Nahal Soreq and Nahal Refa’im basins
(Fig. 9). High sensitivity appears only in limited locations within these basins.
4.4 Vulnerability of urbanized slopes
Using failure criteria learnt from intact natural slopes in the analysis of urbanized slopes
may result in underestimating its failure potential. Construction and engineering
Fig. 9 Potential for dynamic failure of natural slopes. Localities: 1 Mt. Scopus, 2 Mt. of Olives, 3 Nahal
Qidron, 4 Nahal Refa’im, 5 Nahal Soreq
Nat Hazards (2010) 53:375–406 393
intervention may degrade the strength of the rocks, mainly of weak rocks, due to one or
more of the following effects: (1) the removal of the hard calcrete cap that covers the
moderate-steep slopes of the soft carbonate rocks; (2) an increase in slope gradient; (3)
overloading due to multistorey buildings; (4) an increase in the water content due to
leakage from water, sewage, and drainage systems as well as due to blockage of natural
permeable flow and drainage; and (5) local gardening irrigation. As a result, the
mechanical properties of the rocks may change locally from slope to slope, and a site-
specific study is necessary to evaluate the stability of such urbanized slopes.
With this understanding, we decided that in addition to traditionally mapping the slopes
vulnerable to failure in natural conditions, we will also evaluate slopes that have undergone
or may undergo engineering development (Fig. 10). The map intends to single out all the
areas in Jerusalem in which weak rocks are exposed, independent of the slope gradient.
The sensitive rocks in this group are thus the same weak rocks that were mentioned in
constructing the sensitivity map of natural slopes (Sect. 4.3).
Fig. 10 Potential for dynamic failure of urbanized slopes
394 Nat Hazards (2010) 53:375–406
4.5 The summary map–zones of required investigation in Jerusalem
In order to get a comprehensive perspective, especially where several earthquake-related
hazards coexist in the same place and may affect each other and increase the hazard, we
constructed a summary map (Fig. 11). The map combines all the information resulting
from the evaluation of the geological and geotechnical data (in a way, similar to Monahan
et al. 2000) as follows: The various earthquake-related hazards, as analyzed, defined and
delineated in this study, were grouped into four main categories according to the criteria
explained below and detailed in Table 5:
(1) Normal. Hard rock areas and low to moderate topography. No potential of seismic
hazards beyond those foreseen in the IBC-413 was identified in these regions. These
are the white zones in Fig. 11.
Fig. 11 The summary map: categories of earthquake-related hazards in and around Jerusalem, based on
combining the potentials of the various hazard factors (Table 5). Units should be more than 3 m (10 ft) thick
in order to be sensitive to amplification (NEHRP 1997). For simplification, municipal limits are omitted and
the small irregular polygons of natural slope failure cover the hazard categories below it
Nat Hazards (2010) 53:375–406 395
Table 5 Criteria for defining the hazard categories in the summary map
Geology Hazard categories
Group Formation,
geotechnical unit
Lithology Amplification Slope stability
Litho. Topo. Natural slopes* Urbanized
Al, VI Alluvium, soil, rubble soil,
calcrete, clay
Sensitive M [ 20
H [ 30
Mount Scopus Tlt, V Marly chalk Sensitive M [ 20
H [ 30
Kumi, IV Chert, chalk, marl Sensitive
, III/V Chalk Sensitive
Judea Kub, I Limestone Not sensitive M [ 30
H [ 40
Not sensitive
Kuw, I Dolomite Not sensitive
Kuks, III Clayey limestone, chalk Not sensitive
Kua, I Dolomite & some limestone Not sensitive
Kumo, V Marl, limestone Sensitive M [ 20
H [ 30
Kubm, II Dolomite, some marl Not sensitive M [ 30
H [ 40
Not sensitive
Kuke, II Dolomite Not sensitive
Kus, II Dolomite, marl Not sensitive M [ 20
H [ 30
Kugy, II Dolomite Not sensitive M [ 30
H [ 40
Not sensitive
Kk, II Limestone, dolomitic limestone Not sensitive
* M moderate grade, H high grade
If on top of a mountain or a cliff
396 Nat Hazards (2010) 53:375–406
(2) Amplification. Areas in which there is a potential for amplification of seismic
acceleration due to the lithology (Low and Intermediate grades in Fig. 5), or the
topography (from Fig. 6), or both (also from Fig. 6). These are the light gray zones in
Fig. 11.
(3) ‘Urbanized’ slope failure. Areas of weak rocks, independent of the slope gradient.
These are the dotted zones in Fig. 11. Such places (Fig. 10), if inappropriately
developed (e.g., slope steepening, weakening, and wetting), might become vulnerable
to dynamic slope failure. This category overlaps zones of amplification (gray color) in
eastern Jerusalem, and partly covers such zones in western Jerusalem.
(4) Failure of natural slopes. Areas where natural slopes are steeper than the critical
threshold determined for dynamic failure, and thus may fail in a strong earthquake.
Following the criteria presented in Figs. 8 and 9, the critical threshold was set for the
Moderate grade (dark gray zones in Fig. 11)at20 in weak (geotechnically weak)
rocks and 30 in strong (geotechnically strong) rocks. The High grade zone (black
color in Fig. 11) was set at 30 for weak rocks and at 40 for strong rocks (relevant to a
few cliffs only). Low potential is not presented for the sake of simplicity and because
for weak rocks it mostly coincides with category (3)–failure of urbanized slopes. In the
eastern part of the city this category (4) overlaps category (2) (amplification).
The areal distribution of the four categories was then delineated on the map of the study
area. It thus shows the potentially vulnerable areas expected to be affected by the various
earthquake hazards within the city limits, and allows noticing the full scope of the
earthquake-related hazards for each place in Jerusalem. The resolution of the map depends
on the accuracy and scale of the original geotechnical map, which is 1:12,500. Practically,
the summary map directs attention to the zones that require further site-specific investi-
gation and focuses on problematic localities, for example, where amplification in slopes
composed of soft rocks may further reduce their dynamic stability.
5 Relationship between seismic intensities of the 1927 earthquake and the geology
in Jerusalem
It is now possible to examine the summary map in light of the understanding gained
through the analysis of the damage caused by the 1927 earthquake. Historic records show
that earthquakes damaged some structures such as the al-Aqsa mosque and the church of
the Holy Sepulcher again and again (ESM 1). It is not yet clear if this was because of an
exceptional site effect, poor foundation, bad construction, or biased reporting that gave
preference to religious and public structures over ordinary buildings. The 1927 event,
however, affords a special opportunity in that it was extensively investigated and evaluated
on the MSK scale (Avni 1999), and it covers an area around and outside the city, never
reported before.
Looking at the macroseismic data, some localities such as Mount Scopus and the Mount
of Olives seem to have suffered higher intensities. We therefore examined the relationship
between the seismic intensity and the geology in the given sites for a possible systematic
pattern of damage distribution. The analysis was carried out as follows:
1. The localities damaged by the 1927 earthquake were identified on rectified geographic-
historic maps (Jerusalem map 1917, 1934), in order to determine their true coordinates.
2. These coordinates were projected on the modern geological map of Jerusalem, in order
to identify the surface geology of the damaged locality.
Nat Hazards (2010) 53:375–406 397
3. The 1927 epicenter was taken from Shapira et al. (1993) and Avni et al. (2002) who
located it in the northern Dead Sea, at N31.6, E35.4, with a range of error of about
10 km.
4. At this stage we calculated the expected intensity in the given localities according to
the typical ratio for Israel. This ratio, between the expected seismic intensity (I) on the
MSK scale (Grunthal 1993), and the local magnitude (M
), was empirically
determined by Feldman and Shapira (1994) on the basis of macroseismic data of
earthquakes that were instrumentally recorded and felt in Israel:
I ¼ 0:2 þ 1:6M
2:5log rðÞ0:003r ð2Þ
and r is defined as:
¼ R
þ 15 ð3Þ
where R is the distance from the epicenter.
In this empiric equation, I is the intensity of the average damage for an area with a
radius of 5 km and the standard deviation is r = 0.7.
5. Now it was possible to find the difference (dI) between the observed (Avni 1999) and
the calculated intensities (Table 6; Fig. 12). Relating to the gradient dI instead of the
absolute levels, enables reducing possible bias due to the type of intensity scale in use
and subjective determination of the observed intensities, as well as uncertainties in
magnitude–intensity relations which in this case is strictly a function of the epicenter
6. Last we compared the value of ‘dI’ with the local geology. For the sake of clarity, we
present the data in Table 6 in an ascending order of dI.
It appears that more than 80% of the dI values in the range of 1.5–2.0 (i.e., observed
intensity larger than expected) characterize sites of soft rock (chalks and marls of the
MSG), mountain ridges and cliffs, and localities where landslides and rock falls have
occurred. On the other hand, low dI values, in the range of ?0.5 to -1.0 (i.e., observed
intensity equal or smaller than expected), characterize mostly sites constructed on lime-
stone and dolomites of the JG, and this is, more or less, within the range of standard
deviation of the I calculated by Feldman and Shapira (1994). Low dI values also typify
modern constructions which seem to be more resistant than the old traditional ones. The
actual damage on the Temple Mount was higher than the calculated, possibly due to the old
and poor construction, or because it is founded partly on thick artificial fill.
Sites with notable topography (symbolized ‘
’ in Table 6) seem to correlate with higher
dI values, although less consistent than dI correlates with the lithology. Several sites are
characterized by both topography and soft MSG rocks, but the relative contribution of each
of the two factors for the total increase in dI cannot be resolved with the given information.
We may now extrapolate from the experience gained by the 1927 earthquake and
assume that higher-than-expected intensities may affect man-made structures located on
MSG soft rocks, topographic highs and cliffs, and traditional dwellings. Moreover, since
the 1927 event has not directly been used in developing the summary map, and since the
hazard maps are based on the geology, it is possible to test the summary map by retro-
spectively comparing it with the damage caused by the 1927 event. It appears then, that
intensities within zones of hazard categories (2) and (3) tend to increase higher than
expected. Although it was not possible to resolve the relative role of the soft MSG rocks or
the topography, or both, in amplifying the damage, the end result is clear.
398 Nat Hazards (2010) 53:375–406
Table 6 Observed and calculated seismic intensities of the 1927 earthquake in Jerusalem and its surroundings (modified from Avni 1999)
Locality NER(km) Observed I
Calculated I
dI Group; Formation; Lithology
1. Bet Hakerem
631885.7 218184.2 31 5.5 6.5 -1 JG; Kuw; dolomite
2. Moza
633520.7 214968.6 34 6 6.5 -0.5 JG; Kubm; dolomite
3. Ramat Rahel
627184.2 220762.0 26 6.5 6.5 0 MSG; Kum-Kumi; chert, chalk, marl
4. Beit Surik
636725.5 214256.7 37 6 6 0 JG; Kuke; dolomite
5. El-Bira
645985.2 220363.7 41 6 6 0 JG; Kus; dolomite, marl
6. Silwan 630858.0 222562.9 27 7 6.5 0.5 JG; Kub; limestone
7. Jerusalem
631988.8 221033.5 29 7 6.5 0.5 JG; Kub; limestone
8. Batir 626210.5 213319.9 31 7 6.5 0.5 JG; Kus; dolomite, marl
9. Deir el Sheikh 628513.4 206441.9 38 6.5 6 0.5 JG; Kus; dolomite, marl
10. Abu Ghosh 634853.9 210431.8 39 6.5 6 0.5 JG; Kus; dolomite, marl
11. Beit Jala
624886.7 217760.8 27 7.5 6.5 1 JG; Kush; chalky limestone
12. Temple Mount
631674.4 222393.3 28 7.5 6.5 1 Al-JG; Al-Kub; rubble soil, limestone
13. En Kerem 630598.2 215486.6 32 7.5 6.5 1 JG; Kua-Kumo-Kubm; dolomite, marl
14. Qiryat Anavim
635330.2 211565.2 38 7 6 1 Al-JG; Al-Kus; Alluv., dolomite, marl
15. Jerusalem-Jericho Rd 634296.0 238443.9 24 8 6.5 1.5 MSG; Kug; marly chalk
16. Jerusalem-Jericho Rd 633875.0 240450.2 24 8 6.5 1.5 MSG; Kum; chalk
17. Abu Dis
630031.8 225342.9 25 8 6.5 1.5 MSG; Kum; chert, chalk, marl
18. Bethlehem
623808.2 219303.1 25 8 6.5 1.5 MSG; Kum; chalk
19. Ein el Qilt 638058.9 236027.1 28 8 6.5 1.5 JG; Kub; limestone (rock-falls)
20. Ramallah
645749.3 218755.9 42 7.5 6 1.5 JG; Kus; dolomite, marl
21. Commissioner’s Pal
629100.5 222491.1 26 8.5 6.5 2 MSG; Kumi; chert, chalk, marl
22. Mt. of Olives
632165.1 223558.8 28 8.5 6.5 2 MSG; Kum-Kumi; chert, chalk, marl
23. Mt. Scopus
633259.1 223444.4 29 8.5 6.5 2 MSG; Kum-Kumi; chert, chalk, marl
Nat Hazards (2010) 53:375–406 399
Table 6 continued
Locality NER(km) Observed I
Calculated I
dI Group; Formation; Lithology
24. A-Ram
640244.1 222168.4 35 8 6 2 MSG; Kum; chalk
25. Mukhmash 642291.4 226339.9 35 8 6 2 MSG; Kum; chalk
N–E coordinates, New Israeli Grid (Fig. 12), R Distance from epicenter in km, Observed I: Actual damage, Calculated I: according to Feldman and Shapira (1994), rounded,
dI: The difference between the observed and the calculated intensities, Group, Formation, Lithology: near surface geology in that location
Modern construction, as opposed to traditional
Site located on a mountain or range or cliff
400 Nat Hazards (2010) 53:375–406
6 Discussion and conclusions
The evaluation performed here makes use of the already known geology of Jerusalem
rather than of geotechnical and site-specific examinations, which are preferred but
unfortunately not yet available in detail. For now, we delineated the zones of required
investigation for each of the hazard factors that were not yet foreseen by the present IBC-
413 but may yet affect the city, and hope that a quantitative calibration will be followed.
The findings of this survey constitute a professional basis for defining earthquake hazards
in Jerusalem. In addition to the conventional approach, we introduce the concept of ‘failure
of urbanized slopes’, since analyzing such slopes as if they are still in their natural state
may underrate their actual vulnerability.
Field observations show that the Jerusalem area is not particularly sensitive to a static
failure of natural slopes. The broad appearance of the mostly intact calcrete cap indicates
stability of the slopes in the past tens and hundreds of thousands of years (Dan 1977). In
fact, the calcrete cap may have even stabilized the slopes.
Fig. 12 The difference between the observed and the calculated intensities (dI) of the 1927 earthquakes
(white star) in and around Jerusalem. White circles denote the expected calculated intensities. Locality
numbers, as in Table 6. Thin white line denotes the municipal limits of Jerusalem
Nat Hazards (2010) 53:375–406 401
Evidence of natural failure in large volumes is ambiguous. The Soreq Formation may
have produced some large landslides and rock falls, but evidently they are ancient and
sometimes even covered by the calcrete. The morphology typical to the chalk in the
Menuha Formation, if not calcrete capped, is of hummocky slopes, 1–4 m high. Com-
monly, this form develops on slopes of weak rocks as a result of rotational landslides;
however, we did not find conclusive evidence for such processes. In places there is agri-
cultural cultivation and it is not clear if the agriculture exploits this naturally formed
terrace morphology or if the slopes were artificially built specifically for agricultural needs.
It is also possible that alternations between chalk and marly chalk layers and creeping soil
produced these steps and were later exploited by man.
The critical question is thus whether a significant slope failure is expected during the
next strong earthquake. Evidently, the 1927 earthquake, which occurred in the dry summer,
did not cause any large landslide in natural slopes (Wachs and Levitte 1983; Avni 1999).
The only reported slope failure from that earthquake relates to road cuts east of Jerusalem
(Avni 1999). Nevertheless, modern construction and road cuts on slopes of weak rocks are
more widespread today than in 1927, mostly in eastern Jerusalem (on MSG chalk expo-
sures). Katz (2004) concluded that significant failure of a natural slope may occur only in
the case of Mw [7 earthquake (recurrence time of 10
years; Begin 2005; Begin et al.
2005). Thus, it is reasonable to assume that failure in natural slopes will not be widespread
in the next strong earthquake, although failure of road cuts and constructed slopes, espe-
cially of GT IV–VI units, is to be expected.
Looking at the full spectrum of the earthquake-related hazards (Fig. 11), the central
N–S axis zone across Jerusalem is the least vulnerable area. The bedrock there is mostly
hard carbonate, the topography is mild, and thus only the alluvial cover, if thicker than
3 m, should be considered as vulnerable. But much of the city lies along this axis,
including buildings constructed prior to the implementation of the IBC-413, and to be on
the safer side, traditional engineering practice should not be taken for granted as antise-
ismic proof.
Eastwards, the hazard is characterized by the soft MSG chalks that overlie the hard JG
carbonates and the moderate topography. The Vs contrasting between the hard and soft
rocks as well as the topography in places imposes the potential of amplification. Slopes,
either natural or artificially cut into the soft chalk, may expose the region to dynamic
instability. The added effect of amplification may further increase the instability. Most
vulnerable are the few places in the southeast where the potential for amplification as well
as for failure of both steep natural and urbanized slopes, exists (black areas in Fig. 11).
Thus, the ongoing extensive development of the city in this direction, should take all this
into account.
West of the central axis, the picture may seem to be complex, although it simply derives
from the incision of deep valleys into the local strata. This process configured the high
mountains, the steep slopes including the exposures of soft marl units, and the alluvial
cover along the stream beds. All these forms constitute the potential for earthquake haz-
ards. The most severe hazard exists where the potential for failure of both steep natural and
urbanized slopes appear (black in Fig. 11). Being a potential direction for expansion of the
city, these areas specifically call for careful environmental and engineering planning.
Overall, the summary map is an essential layer that can be used for environmental and
municipal planning, land use determination, emergency response planning, and lifeline and
utility vulnerability studies, especially in densely populated areas with so many historic,
cultural, and religious monuments. The map can also assist in guiding site-specific geo-
technical examinations, necessary for engineering planning.
402 Nat Hazards (2010) 53:375–406
Incorporating the hazard maps in the city master plan will raise the need for regulating
planning and antiseismic building practice and increase public awareness. It will also
enable the municipality to enforce its responsibility in controlling and supervising its
growth, expansion, and sustainable development, in regard to the seismic hazards. The
building inventory and architectural style typical of Jerusalem are partly vulnerable, for
example the numerous ancient historic structures unparalleled in religious and cultural
importance, the common first level pillar building, and the mandatory stone cover on the
outside walls of all buildings. With proper modification, the results of this study can help
the city preserve its style despite the hazards posed by future earthquakes.
The summary map depicts the type of hazards in the different parts of the city; however,
this is not intended to limit or forbid construction in any of the areas. It is for the planner and
engineer to determine, according to the hazard factors, what can be built, where, and how.
Following the 1927 earthquake, Willis (1928) demonstrated how poor the local building
practice performed and the certain regulations he suggested in order to go far toward
reducing the damage from any future shock to a minimum’’, were indeed influential.
Retrospectively analyzing the same event, we suspect that higher-than-expected intensities
were not limited to traditional dwellings but also affected man-made structures located on
MSG soft rocks and topographic highs. Hopefully, the understanding gained in this study
will serve for routine practice rather than for outlining the pattern of damage in the next
Acknowledgments We received much help from colleagues at the Geological Survey of Israel: Yoav
Avni, Rivka Amit, Ya’aqov Arkin, Gidi Baer, Arie Gilat, Ariel Heimann, Dubi Levitte, Yoav Nachmias,
Amihai Sneh, Dani Wachs and Ezra Zilberman. Constructive review by Beni Begin, Gidi Baer and Rivka
Amit improved this work considerably. Channa Nezer-Cohen and Bat-Sheva Cohen are greatly appreciated
for preparing the figures. Bevie Katz helped with the editing. Careful reading and constructive suggestions
by Amotz Agnon from the Hebrew University of Jerusalem and two anonymous reviewers significantly
improved our manuscript and are much appreciated. Avi Shapira, Alex Beck and Yuli Zaslavski of the
Geophysical Institute of Israel were involved in the beginning stages of this project. Yuli was a partner to
our dilemmas and we greatly benefited from his professional experience. We specially thank Uri Shitrit, the
chief engineer of Jerusalem at the time. Thanks are due also to Ofer Manor, Ami Kaplan and Amiram
Rotem, all from the Jerusalem municipality. The steering committee, particularly the late engineer Dan
Wind as well as engineers Shemuel Mahala, Uzi Saltzman, Eduard Liebovitz, Shulamit Sa’ar, David David,
and the late Haim Marx, helped improving professional aspects. Amos Israeli, who prepared the geotech-
nical map of the city, contributed from his experience. The project was commissioned and funded by the
Jerusalem municipality as part of its new master plan.
Ambraseys NN (2005) Historical earthquakes in Jerusalem–a methodological discussion. J Seismol 9:329–340
Ambraseys NN, Karcz I (1992) The earthquake of 1546 in the Holy Land. Terra Nova 4:253–262
Arkin Y (1976) Jerusalem and vicinity, Geological map, 1:50,000, Isr Geol Surv
Arkin Y (1984) A sinkhole near the damascus gate, old city of Jerusalem. Int Symp On Land Subsidence,
3RD, Venice, Proc, 565–574
Arkin Y, Braun M, Starinsky A (1965) Type sections of cretaceous formations in the Jerusalem-Bet
Shemesh area, GSI Stratigraphic sections No 1, Part 1, Lithostratigraphy, Isr Geol Surv
Arkin Y, Michaeli L, Wolfson N (1993) Har Hazofim tunnel: geological and geotechnical site investigation,
Isr Geol Surv, GSI/18/93
Arkin Y, Michaeli L, Liberman A (1994) Har Hazofim tunnel, additional boreholes Z6 and Z7, Isr Geol
Surv, TR-GSI/22/94
Ashford A, Sitar N (1997) Analysis of topographic amplification of inclined shear waves in a steep coastal
bluff. Bull Seism Soc Am 87(3):692–700
Nat Hazards (2010) 53:375–406 403
Ashford A, Sitar N, Lysmer J, Deng N (1997) Topographic effects on the seismic response of steep slopes.
Bull Seism Soc Am 87(3):701–709
Avni G, Seligman J (2001) The temple mount 1917–2001: documentation, research and inspection of
antiquities. Israel Antiquities Authority 2001, 44pp
Avni R (1999) The 1927 Jericho earthquake, comprehensive macroseismic analysis based on contemporary
sources, PhD thesis, Ben GurionUniversity of the Negev, Beer-Sheva, Israel (in Hebrew with English abst)
Avni R, Bowman D, Shapira A, Nur A (2002) Erroneous interpretations of historical documents, related to
the epicenter of the Jericho earthquake in the Holy Land. J Seismol 6:469–476
Bartov Y, Sneh A, Fleischer L, Arad V, Rosensaft M (2002) Potentially active faults in Israel, Isr Geol Surv,
Stage B, Rep GSI/29/2002
Begin BZ (2005) Destructive earthquakes in the Jordan Valley and the Dead Sea–their recurrence interval
and the probability of their occurrence, Geol Surv Israel, Rep. GSI/12/2005
Begin BZ, Zilberman E (1997) Main stages and rate of the relief development in Israel, Isr Geol Surv, Rep
GSI/24/97 (in Hebrew, English abst)
Begin ZB, Steinberg DM, Ichinose GA, Marco S (2005) A 40,000 year unchanging seismic regime in the
Dead Sea rift. Geology 33(4):257–260
Belitzky S (1996) Tectonic geomorphology of the Lower Jordan Valley–An active continental Rift, PhD
Thesis, Hebrew University of Jerusalem, 98pp (in Hebrew, English abst)
Ben-Menahem A (1991) Four thousand years of seismicity along the Dead Sea Rift, J Geophys Res 96
Bentor YK (1989) Geological events in the Bible. Terra Nova 1(4):326–338
Boore DM (2004) Can site response be predicted? J Earthq Eng 8(1):1–41
Braun M (1970) Facies changes of the Judea Group in Judea and Samaria. In: Stratigraphic problems in the
Judea Group, Isr Geol Surv Ann Meet Jerusalem, pp 29–35
California Geological Survey (CGS) (1997) Special bulletin 117: guidelines for evaluation and mitigating
seismic hazards in California
Dan J (1977) The distribution and origin of Nari and other lime crusts in Israel. Isr J Earth Sci 26:68–83
Dobry R, Borcherdt R, Crouse C, Idriss IM, Joyner WB, Martin GR, Power MS, Rinne EE, Seed RB (2000)
New site coefficients and site classification system used in recent building seismic code provisions.
Earthquake Spectra 16(1):41–67
Erdik M, Aydınog
lu N, Barka A, Yu
, Siyahi B, Durukal E, Fahjan Y, Akman H, Birgo
ren G, Alpay-Biro
Y, Demircioglu B, Ozbey C, Sesetyan K (2003) Earthquake risk assessment for the Istanbul metropolitan
area, Final Report, Bogazici University, Kandilli Observatory and Earthquake Research Institute
Ezersky M (2001) Refraction survey at the Holyland site, Jerusalem. Prepared for the Holyland Park
Company Ltd, Geophy Inst Isr, Rep 205/162/01 (in Hebrew)
Feldman L, Shapira A (1994) Analysis of seismic intensities observed in Israel. Nat Hazards 9:287–301
Freund R, Garfunkel Z, Zak I, Goldberg M, Weisbrod T, Derin B (1970) The shear along the Dead Sea.
R Soc Lond Philos Trans, Ser A 267:107–130
Garfunkel Z (1981) Internal structure of the Dead Sea leaky transform (rift) in relation to plate kinematics.
Tectonophysics 80:81–108
Garfunkel Z, Ben-Avraham Z (2001) Basins along the Dead Sea transform In: Ziegler PA, Cavazza W,
Robertson, AHF, Crasquin-Soleau S (eds) Peri-Tethys Memoir 6: Peri-Tethyan Rift/Wrench Basins
and passive margins. Mem. du Musee Nat d’Hist naturell, Paris, v 186
Geli L, Bard PY, Jullien B (1988) The effect of topography on earthquake ground motion: a review and new
results. Bull Seism Soc Am 78(1):42–63
GII (Geophysical Institute of Israel) (2006) Search for information on earthquakes in Israel and adjacent
areas, Seismology Division, Geophy Inst Isr,
Gilat A, Wolfson N, Michaeli L. Arkin Y (1992) The eastern ring road, geological and geotechnical survey.
Isr Geol Surv, Rep GSI/22/92, 61pp (in Hebrew)
Gill D (1996) The geology of the city of David and its ancient subterranean waterworks. City of David
excavations, Final report IV. Hebrew University, Jerusalem, pp 1–28
Grunthal G (ed) (1993) European macroseismic scale 1992 (up-dated MSK-scale). European Seismological
Commission, Luxembourg
Guidoboni E, Comastri A (2005) Catalogue of earthquakes and tsunamis in the Mediterranean area from the
11th to the 15th century, INGV-SGA, Italy
Guidoboni E, Comastri A, Traina G (1994) Catalogue of ancient earthquakes in the Mediterranean area up to
the 10th century. ING-SGA, Bologna, Italy
Gvirtzman Z (2004) Ground motion amplification in the Israeli foothills: empiric relations between reso-
nance of ambient noise and geological structure. Isr Geol Surv, Rep GSI/17/2004, 42pp (in Hebrew,
English abst)
404 Nat Hazards (2010) 53:375–406
Hall JK (1993) The GSI digital terrain model (DTM) completed. Isr Geol Surv Current Res, Jerusalem 8:47–50
Hall JK, Weinberger R, Marco S, Steinitz G (1999) Test of the accuracy of the DTM of Israel. Isr Geol Surv,
Rep TR-GSI/1/99
IBC-413 (Israel Building Code 413) (1995) Provisions for earthquake resistance design of structures. The
Standards Institution of Israel
Israeli A (1977) Geotechnical map of Jerusalem and surroundings, Scale 1:12,000, Isr Geol Surv, Rep
MM12/77 (in Hebrew)
Jackson J, McKenzie D (1988) The relationship between plate motions and seismic moment tensors, and the
rates of active deformation in the Mediterranean and Middle East. Geophys J 93:45–73
Jerusalem Map (1917 Sheet XVII, Palestine exploration fund map
Jerusalem Map (1934) Sheet 8, 1:100,000 series, Survey of Palestine
Jibson RW, Harp EL, Michael JA (1998) A method for producing digital probabilistic seismic landslide
hazard maps: an example from the Los Angeles, California, area. U.S. Geol Surv Open-File Rep
Katz O, (2004) Evaluation of earthquake induced landslide hazard in the city of Jerusalem area, Isr Geol
Surv Rep GSI/12/2004, 34pp (in Hebrew)
Katz O, Crouvi O (2007) The geotechnical effects of long human habitation (2000 \ years); earthquake
induced landslide hazard in the city of Zefat, northern Israel. Engin Geol 95:57–78
Kawase H (2003) Site effects on strong ground motion. In: Lee WHK, Kanamuri H, Jenning PC, Kisslinger
C (eds) International handbook of earthquakes and engineering seismology, chap 61, Academic Press,
pp 1013–1030
Keefer DK (1984) Landslides caused by earthquakes. Bull Geol Soc Am 95:406–421
Lewy Z (1989) Correlation of lithostratigraphic units in the upper Judea group (Late Cenomanian–Late
Coniacian) in Israel. Isr J Earth Sci 38:37–43
Lewy Z (1991) Periodicity of Cretaceous epeirogenic pulses and the disappearance of the carbonate platform
facies in Late Cretaceous times (Israel). Isr J Earth Sci 40:51–58
Michaeli L, Arkin Y (1994) Gillo viaduct investigations; Geological and geotechnical survey, foundation pit
DO. Isr Geol Surv Rep TR-GSI/26/94, 19pp (in Hebrew)
Mimran Y (1969) The geology of Wadi El-Malich area, MSc thesis, Hebrew University of Jerusalem, Israel,
68pp (in Hebrew)
Mimran Y (1995) Stratigraphic and geotechnical evaluation of the chalk in the Mt. Scopus tunnel, Isr Geol
Surv Rep GSI/42/95
Monahan PA, Levson VM, McQuarrie EJ, Bean SM, Henderson P, Sy A (2000) Relative earthquake hazard
map of greater Victoria, showing areas susceptible to amplification of ground motion, liquefaction and
earthquake- induced slope instability. British Columbia Geol Surv Ministry of Energy and Mines,
Geoscience Map 2000–2001
NEHRP (1997) NEHRP recommended provisions for seismic regulations for new buildings and other
structures, Part 1: provisions (FEMA 302). Building Seismic Safety Council of the National Institute of
Building Sciences, chap 4, pp 33–41
Picard L (1943) Structure and evolution of Palestine, Bull Geol Dept, Hebrew University, Jerusalem IV
Quennell AM (1959) Tectonics of the Dead Sea Rift. Int. Geol. Congr. 20th, Mexico 1956, Assoc Servic
Geol Afric 385–405
Ronen A (1995) Seismic refraction survey using P and S waves Rout #4–North, Ramot Intersection. Inst
Petrol Res Geophysi, Rep No K09/304/95
Roth I (1969) The geology of Wadi El Qilt region. MSc thesis, Hebrew University of Jerusalem, Israel, 52pp
(in Hebrew)
Salamon A, Hofstetter A, Garfunkel Z, Ron H (1996) Seismicity of the eastern mediterranean region:
perspective of the sinai subplate. Tectonophysics 263:293–305
Salamon A, Hofstetter A, Garfunkel Z, Ron H (2003) Seismotectonics of the Sinai subplate–The eastern
mediterranean region. Geophys J Int 155:149–173
lveda SA, Murphy W, Jibson RW, Petley DN (2005) Seismically induced rock slope failures resulting
from topographic amplification of strong ground motions: The case of Pacoima Canyon, California.
Engin Geol 80 (3–4):336–348
Shalem N (1949) The earthquakes of Jerusalem. Yerushalayim 2 (1–2):22–54. In: Articles of N Shalem,
Jerusalem, Kiryat Sefer, 1973, pp 270–308
Shapira A, Avni R, Nur A (1993) A new estimate for the epicenter of the Jericho earthquake of 11 July 1927.
Isr J Earth Sci 42:93–96
Shtivelman V (1996) Estimation of shear wave velocity distribution in the shallow subsurface and its
implications, Inst Petrol Res Geophysi, Work No K57/173/95
Nat Hazards (2010) 53:375–406 405
Stewart JP, Chiou S-J, Bray JD, Graves RW, Somerville PG, Abrahamson NA (2001) Ground motion
evaluation procedures for performance based design. Rpt No PEER-2001/09, Pacific Earthquake
Engineering Research Center
Toppozada T, Borchardt G, Hallstrom CL, Youngs LG (1994) Planning scenario for a major earthquake on
the rodgers creek fault in the Northern San Francisco Bay Area. CDMG Spec Publ 112
Wachs D, Levitte D (1983) Earthquake risk and slope stability in Jerusalem, Isr Geol Surv, Rep EG/2/83
Wachs D, Levitte D (1984) Earthquake in Jerusalem and the mount of olives landslide, Israel–Land and
Nature 9(3):118–121
Wachs D, Zilberman E, Arad A (1990) Geo-engineering evaluation of the stability of the Armon Hanaziv
historical tunnel, Isr Geol Surv, Rep GSI/29/90, 17pp (in Hebrew)
Wdowinski S, Zilberman E (1997) Systematic analysis of the large scale topography and structure across the
Dead Sea Rift. Tectonics 16(3):409–424
Wdowinski S, Bock Y, Baer G, Prawirodirdjo L, Bechor N, Naaman S, Knafo R, Forrai Y, Melzer Y (2004)
GPS measurements of current crustal movements along the Dead Sea Fault. J Geophys Res
109:B05403. doi:10.1029/2003JB002640
Willis B (1928) Earthquakes in the holy land. Bull Seismol Soc Am 18:73–103
Wills CJ, Petersen M, Bryant WA, Reichle GJ, Saucedo GJ, Tan S, Taylor G. Treiman J (2000) A site-
condition map for California based on geology and shear wave velocity. Bull Seismol Soc Am 90
Zaslavsky Y, Shapira A (2000a) Experimental study of topographic amplification using the Israel seismic
network. J Earthq Eng 4(1):43–65
Zaslavsky Y, Shapira A (2000b) Observation of topographic site effects in Israel. Isr J Earth Sci 49(2):111–125
Zaslavsky Y, Gorstein M, Kalmanovich M, Giller V, Livshits I, Giller D, Dan I, Shapira A, Fleischer L,
Leonov J, Peled U (2001) Microzoning of the earthquake hazard in Israel, Project 1: Seismic mi-
crozoning of Lod and Ramla. Geophys Inst Isr, Job No 569/143/01
Zaslavsky Y, Shapira A, Leonov J (2003) Empirical evaluation of site effects by means of H/V spectral
ratios at the locations of strong motion accelerometers in Israel. J Earthq Eng 7(4):655–677
Ziony JI (ed) (1985) Evaluating earthquake hazards in the Los Angeles region–An earth science perspective:
US Geol Surv Profes Paper 1360, 505pp
406 Nat Hazards (2010) 53:375–406
... Typically, reported damage associated with an earthquake is assessed at numerous sites and the earthquake magnitude is determined from the distribution of intensities (Tinti et al., 1987;Gasperini et al., 1999;Serva et al., 2007). However, due to the fact that the intensity values are assigned at each location based on historically reported damage to man-made structures and other natural surficial deformation features, site effects, which might amplify selectively the local earthquake induced damage, are usually not considered in the analysis (Michetti et al., 2004;Hough and Bilham, 2008;Salamon et al., 2010). Thus, the evaluated earthquake magnitude and the overall seismic hazard can be over-estimated (Hough et al., 2000), unless based on locally calibrated magnitudeintensity relations (e.g. ...
... The damage reports have been used to assess the magnitude of these earthquakes (Ambraseys and Barazangi, 1989;Ambraseys, 1997;Avni et al., 2002;Marco et al., 2003), with no consideration of the expected local site effects due to its vulnerable geotechnical state. Similarly, many more cities in the area, e.g., Jerusalem and Nablus, were repeatedly damaged in historical earthquakes, and the damage reports have been used to study these earthquake magnitudes (Ambraseys and Karcz, 1992;Salamon et al., 2010). Hough and Avni (2009) noted that the intensities reported from large earthquakes in the Dead Sea Fault region (Near East) are higher than the ones reported from California in comparable-sized earthquakes, and suggested lower attenuation in the Dead Sea crust. ...
... As is the case in the Old City of Jerusalem, many modern cities of the Near East still have old residential quarters built in the historical location of the city and on archeological mound structures, e.g. Zefat (Katz and Crouvi, 2007); Jerusalem (Salamon et al., 2010). These quarters are subject to higher seismic hazard due to site amplification, as was also noted by Wachs and Levitte (1978). ...
Throughout historical times, human habitation in the Near East was limited by its arid climate to sites with strategic advantages. These sites underwent millennia of recurring destruction and rebuilding, resulting in the accumulation of artificial mounds, known as ‘Tells’, consisting of multiple archeological debris strata of ruined cities, which could pile up to a height of a few tens of meters. Intensity-based magnitude evaluation of Near East historical earthquakes often utilizes reported damage from cities built on earlier archeological mounds. Moreover, old residential quarters of many modern Near Eastern cities are founded on such mounds. In this study we measured the subsurface shear-wave velocity (Vs) profile on top of a typical mound structure as well as the Vs profile of the bedrock on which the mound was accumulated. The results suggest that the mound structure is likely to amplify seismic ground motions. These results imply that intensity-based magnitude evaluation of Near East historical earthquakes that are based on reported damage from cities built on archeological mounds, might be biased and could lead to over-estimations of past earthquake magnitudes. Furthermore, many old residential quarters of modern Near Eastern cities are commonly founded on such mounds and might be exposed to a higher seismic hazard than nearby modern neighborhoods, thus requiring special attention in seismic hazard mitigation programs and seismic design.
... Israel is situated along the Dead Sea Fault, a tectonically active plate boundary (e.g., Salamon et al. 2003;Garfunkel 2014) which has been the origin of intensive earthquakes causing extensive destruction for the past 2000 years and more. According to the historical records, almost every major city in Israel was damaged several times during the last two millennia by earthquakes, including Jerusalem that was damaged in the more recent past (e.g., Salamon et al. 2010;Avni 1999) and is still vulnerable to the effects of an earthquake (Katz 2004;Salamon et al. 2010). Although there has been no major earthquake in the region in the last 90 years, experts predict that strong tremors will occur in the near future (e.g., Begin 2005;Shapira et al. 2007;Hamiel et al. 2009). ...
... Israel is situated along the Dead Sea Fault, a tectonically active plate boundary (e.g., Salamon et al. 2003;Garfunkel 2014) which has been the origin of intensive earthquakes causing extensive destruction for the past 2000 years and more. According to the historical records, almost every major city in Israel was damaged several times during the last two millennia by earthquakes, including Jerusalem that was damaged in the more recent past (e.g., Salamon et al. 2010;Avni 1999) and is still vulnerable to the effects of an earthquake (Katz 2004;Salamon et al. 2010). Although there has been no major earthquake in the region in the last 90 years, experts predict that strong tremors will occur in the near future (e.g., Begin 2005;Shapira et al. 2007;Hamiel et al. 2009). ...
Full-text available
The use of casualty modeling in the field of disaster management is well established. Nevertheless, it is currently based almost exclusively on damage to the built environment and fails to consider additional factors that may influence the number of casualties in a given event, such as behavioral features of the exposed population. The present study has taken an innovative approach and integrated behavioral traits of residents in a high-risk area in northern Israel, near the Dead Sea Transform, into a well-known casualty estimation simulation. The expected behavioral characteristics of residents during an earthquake, in city sectors with different socioeconomic rankings, were assessed using a designated survey and were applied into the casualty estimation process. In order to test the sensitivity of the behavioral factor, twelve synthetic earthquake scenarios were designed. The results shed light on the relationship between specific behavioral strategies and casualty projections and suggest that loss estimation models that do not take behavioral factors into account may overestimate the projected number of casualties. Households with low socioeconomic status were found to be more vulnerable in terms of risk of injury and death compared with those ranked higher. The present study shows the importance of raising public awareness regarding proper behavior prior to and during the event which can help increase resilience of communities, mitigate risks and losses and ultimately save lives. Further implications of these results and possible ways of improving casualty modeling and community resilience are also discussed.
... An ultra-Orthodox neighborhood is defined as a neighborhood where over 50 % of the Jewish population aged 15 or older are ultra-Orthodox (Korach and Choshen, 2020, p. 25). The background map outlines soil with a tendency to amplify shaking (Salamon et al., 2010). This is based on soil and rock classification and needs verification. ...
... This finding aligns with previous studies on minority groups' low level of self-protection, preparedness, and hazard knowledge in comparison to the majority group (e.g., Lucini, 2014;Maldonado et al., 2016). Specifically, half of our survey respondents indicated that the buildings they live in do not meet the legal safety standards despite the fact that some of their neighborhoods are in areas with increased ground shaking hazard during an earthquake (Salamon et al., 2010) (Fig. 2). Alarmingly, none of our respondents recommended retrofitting, a crucial strategy for minimizing the harm caused by earthquakes (Bertero and Bozorgnia, 2004). ...
Full-text available
To work effectively, emergency management systems that deal with earthquake threats must consider the needs of religious minority groups. Studies regarding earthquake preparedness among marginalized social–cultural groups can highlight ways to improve it. Recently, some research has focused on the effect of religion on earthquake preparedness. However, very few studies have connected the two and examined earthquake preparedness among religious groups that are also a social–cultural minority in relation to the authorities. This study examines the effects of religious beliefs and customs on earthquake preparedness among the Jewish ultra-Orthodox community in Israel, a significant religious minority with unique social, cultural, and economic characteristics. Data were obtained using mixed methods including a survey, in-depth interviews, and focus groups. Results demonstrated that the majority of the community had a low level of hazard knowledge and a high level of disbelief that a devastating earthquake would occur in their area in the near future. This is despite a long-documented history of earthquakes that devastated the Levant and, in particular, dwelling locations for this community. Low exposure to media, insularity of educational institutions, and suspicion toward state authorities were shown to hinder preparedness, while strong social capital improved it. This research is unique for it studies a religious group that is also a cultural minority, which, therefore, requires special adaptations. Some of the recommended adaptations include receiving support from religious leaders, publishing preparation guidelines in proper settings, working with civilian organizations that are seen as legitimate by the religious communities, and adapting technologies and information to be religiously appropriate. To conclude, this research offers a perspective on the complex reality of hazard preparedness in a religiously diverse country. The conclusions are applicable to other countries and natural hazards.
... Recurrence time for an M w 6 and M w 7 earthquake is to the order of 10 2 and 10 3 years, respectively (Begin 2005;Hamiel et al. 2009). According to historical records, almost every major city in Israel has suffered earthquake damage several times over the last two millennia (Zohar et al. 2016;Salamon et al. 2010;Avni 1999). ...
Full-text available
Building damage probabilities are invaluable for assessing short-term losses from natural hazards. In many countries however, the individual building level data required for assessing reliable damage are usually unavailable. This paper shows how the post-processing of aggregate HAZUS earthquake damage assessments can yield building-level damage probabilities. On the basis of three plausible scenarios for Northern Israel, we generate and visualize a building-level combined damage probability index. We use the tools of exploratory spatial data analysis to purge any causal influences in the spatial pattern of these calculated damage probabilities. The costs and benefits of our approach are discussed.
... This is then expanded to include cascading effects in the form of reduced amenities. We adopt a spatial equilibrium approach to the local labor market as pioneered by Rosen (1979) and Roback (1982) and extended by Moretti (2011). At the outset we are interested in two markets: the labor market where local supply is upward sloping and demand is downward sloping and the land market where supply is upward facing but with different elasticities for residential (housing) and commercial (industrial) demand for land by workers and firms. ...
We present an economic definition of cascading effects of a disaster on the labor market over the medium to long term. Cascading effects are considered events that alter local amenities. In the context of the labor market, the standard conception of a cascade as a sequence of events that alter the capital stock, may not be very instructive as the immediate time horizon is not the relevant economic timeframe. We outline some of the theoretical implications arising from this definition and give them some intuition based on an agent based simulation model. The model is used to simulate two cascade-type scenarios following an earthquake in the city of Jerusalem. Results indicate that a strong cascading effect in the labor market depends on serious functional change in the physical environment i.e. land-use change. Flow-related changes in labor and population movement are less likely to create effects that cascade into other sub-markets. Implications of these findings point to the key role of labor mobility as workers seek solutions outside the area struck by disaster.
... Geotechnical engineering also has a very important role to play because there are quite a number of historic sites and monuments affected by geotechnical risks of different types [2]. As all architectural heritages are standing on ground, the seismic risk assessments for architectural heritage can be initiated from geological and geotechnical approaches to understand the composition and the material property distribution of the beneath and surrounding ground, local site effect, and seismic zonation [3][4][5][6]. For the structural and architectural engineering perspective, the seismic behaviour of historic monument and architectural heritage composed of stone block or masonry structure have been evaluated analytically, numerically [7][8][9][10][11] and experimentally [12][13][14][15]. ...
Full-text available
The Gyeongju Historic Areas, which include the millennium-old capital of the Silla Kingdom, are located in the region most frequently affected by seismic events in the Korean peninsula. Despite the numerous earthquakes documented, most of the stone architectural heritage has retained their original forms. This study systematically reviews and categorises studies dealing with the seismic risk assessment of the architectural heritage of the historic areas. It applies research methodologies, such as the evaluation of the engineering characteristics of subsoil in architectural heritage sites, site-specific analysis of the ground motions in response to earthquake scenarios, geographic information system (GIS)-based seismic microzonation according to the geotechnical engineering parameters, reliability assessment of dynamic centrifuge model testing for stone masonry structures and evaluation of seismic behaviour of architectural heritage. The M 5.8 earthquake that hit Gyeongju on September 12, 2016 is analysed from an engineering point of view and the resulting damage to the stone architectural heritage is reported. The study focuses on Cheomseongdae, an astronomical observatory in Gyeongju, whose structural engineering received considerable attention since its seismic resistance was reported after the last earthquake. Dynamic centrifuge model tests applying the Gyeongju Earthquake motions are performed to prove that it is not a coincidence that Cheomseongdae, a masonry structure composed of nearly 400 stone members, survived numerous seismic events for over 1300 years. The structural characteristics of Cheomseongdae, such as the well-compacted filler materials in its lower part, rough inside wall in contrast to the smooth exterior, intersecting stone beams and interlocking headstones are proven to contribute to its overall seismic performance, demonstrating outstanding seismic design technology.
... The location of the area is only 30 km away from the active Dead Sea fault. The city center houses many structures that do not ascribe to modern building codes thus elevating the risk for extensive damage in the case of a major earthquake [25]. We run the model 25 times with no shock and 25 times with a shock, located randomly in space (to avoid spatially biased results), occurring at day 60. ...
This paper presents two opposite perspectives on the labor market in the aftermath of a disaster. The first posits a production sector that is non-tradeable and a labor market with total mobility. This is modeled using agent based simulation. The second presents a production sector that is fully tradeable and a labor market that is perfectly immobile. This is modeled using traditional micro-economic modeling and numerical simulation. Outcomes from the two approaches are compared. In the no-disaster case, participation rates and wages under both approaches settle down to a low-level equilibrium albeit at different rates. In the case of a disaster, outcomes are very different. Under the agent based model labor market mobility results in solutions being found outside the area. In the micro-economic approach workers absorb the recovery process within the area readjusting their demand for labor. When population movement is introduced the system reorganizes at a new equilibrium. The results highlight first, the importance of labor mobility and flexibility and second, the divergent absorption costs in determining the long-term outcomes of a disaster.
... These areas were found highly sensitive to liquefaction due to the presence of massive Holocene and Pleistocene sands and artificial fill, very shallow groundwater and the expected high seismic accelerations (Salamon et al. 2006;Frydman et al. 2007). Based on historical evidences (Salamon et al. 2010), the recommendation was to exercise a tsunami along the northern coast of Israel, in accordance with the vulnerable inundation area delineated in the tsunami hazard map of the Haifa Bay (Fig. 3). For practice purposes, it was also recommended to extend the impact of the synthetic tsunami further south and include the Tel Aviv and the Ashdod-Ashqelon shores, in accordance with the maps of the areas vulnerable to flooding there (Salamon 2009). ...
Full-text available
National exercises are an excellent opportunity to practice earthquake preparedness. Such exercises can greatly benefit from productive communication between the civil protection agencies (CPs) and the earth sciences community (SC). The challenge of the scientists in this interaction is to properly formulate their message and convey their perspective in a manner understandable to the responsible emergency agencies. On October 2012, Israel held its first national earthquake emergency exercise (TP6) that examined the response of the country’s systems at large to an Mw ~ 7 earthquake. The exercise greatly benefited from brain storming meetings between the CPs and researchers from the Geological Survey of Israel (GSI) that were held prior to the drill. These helped in choosing the earthquake scenario and establish the concept of the exercise. Geological hazards and damage maps, including numerous discrete events, were prepared in advance with the HAZUS Multi-Hazard Loss Estimation software and were conveyed to the drilled authorities during the exercise. The exercise also benefitted from close collaboration between researchers of FEMA and the GSI. During the drill, the GSI and its relevant scientists practiced the preparation and transfer of the relevant material to the decision makers in “real time.” The drill provided the following lessons: (1) In real time, the damage maps should be delivered by earthquake researchers, thereby helping the CP agencies to grasp the information. (2) Damage maps should be prepared in advance and accessibly stored by the CP agencies for a range of probable scenarios. (3) Damage maps based on dot density that represent number of buildings damaged, number of casualties and weight of debris were found to be the most comprehensible when presenting the scope of the damage. The lessons learned from the collaboration between the CP and SC in TP6 provide an excellent example for optimal planning of national earthquake exercises, thereby helping in minimizing the anticipated impact of destructive earthquakes.
... Landslides in Israel have occurred in historical earthquakes and extreme rainstorms. Recently, landslide hazard was quantitatively assessed at a local scale (Katz and Crouvi, 2007;Salamon et al., 2009) and at national scale (Crouvi, 2001;Katz and Almog, 2006;Katz et al., 2008;Katz, 2012). At a local scale, anthropogenic soil-like material, located mostly in ancient cities (e.g., Zefat, Jerusalem), was found to be mechanically weak, and thus susceptible to slope failure (Katz and Crouvi, 2007). ...
Full-text available
... While Jerusalem is located 30 km southeast of the active Dead Sea Fault line, the last major earthquake in the city occurred in 1927. The city center lies in a relatively stable seismic area but many of its buildings were constructed prior to the institution of seismicmitigation building codes making it prone to damage (Salamon et al. 2010). The study area houses 22,243 inhabitants, covers 1.45 km 2 and is characterized by low-rise buildings punctuated by high rise structures. ...
This paper illustrates how synthetic big data can be generated from standard administrative small data. Small areal statistical units are decomposed into households and individuals using a GIS buildings data layer. Households and individuals are then profiled with socio-economic attributes and combined with an agent based simulation model in order to create dynamics. The resultant data is ‘big’ in terms of volume, variety and versatility. It allows for different layers of spatial information to be populated and embellished with synthetic attributes. The data decomposition process involves moving from a database describing only hundreds or thousands of spatial units to one containing records of millions of buildings and individuals over time. The method is illustrated in the context of a hypothetical earthquake in downtown Jerusalem. Agents interact with each other and their built environment. Buildings are characterized in terms of land-use, floor-space and value. Agents are characterized in terms of income and socio-demographic attributes and are allocated to buildings. Simple behavioral rules and a dynamic house pricing system inform residential location preferences and land use change, yielding a detailed account of urban spatial and temporal dynamics. These techniques allow for the bottom-up formulation of the behavior of an entire urban system. Outputs relate to land use change, change in capital stock and socio-economic vulnerability.
Data from 40 historical world-wide earthquakes were studied to determine the characteristics, geologic environments, and hazards of landslides caused by seismic events. This sample was supplemented with intensity data from several hundred US earthquakes to study relations between landslide distribution and seismic parameters. Correlations between magnitude (M) and landslide distribution show that the maximum area likely to be affected by landslides in a seismic event increases from approximately 0 at M = 4.0 to 500 000 km2 at M = 9.2. Each type of earthquake-induced landslide occurs in a particular suite of geologic environments. -from Author