Open this page in a new tab

En Gev Landslides and Trenches (aka Ha'on escarpment)

Morphotectonic map of the studied area Figure 2

Morphotectonic map of the studied area (coordinates are in Israeli TM grid). The map is based mainly on areal photos (dated 1945), interpretation, and complementary field work. Mapped are all lineaments, landslides, fluvial systems (with sites of channel displacement marked by open black circle) and alluvial fans located within the study area. Also shown are the sites of the paleoseismic trenches and the highest shore line of the Sea of Galilee, 170 m below sea level, achieved at 25 ka (Hazan et al., 2005).

JW: Ein Gev landslides are shaded purple

Katz et al (2009)


Introduction
Introduction

Katz et al (2009) found evidence for five to seven MW > 6 earthquakes from paleoseismic trenching and an analysis of paleo landslides around the Sea of Galilee. They dated five of these events to 45, 40, 35, 10, and 5 ka BP using the optically stimulated luminescence (OSL) method. They detected what may be a landslide evidence for an event younger than 5 ka BP which they suggested might have been caused by one of the mid 8th century CE earthquakes.

Stratigraphy

Miocene lacustrine and fluvial sediments compose the steep slopes of the study area (Michelson, 1979; Givon, 1984; Mor and Sneh, 1996). These sediments are a part of the Miocene Susita Fm. (mostly limestone, marl, and sandy-dolomite), the overlying Miocene Ein-Gev Fm. (mostly sandstone and some sandy limestone, limestone, and marl; Michelson, 1979), and the overlying Miocene Hordos Fm. (mostly conglomerates and sandstones, and some shales, marls, and limestone beds). This entire Miocene sequence is deposited on an Eocene basement (mostly chalk and limestone, exposed in places at the lower slopes) and is capped by the Plio-Pleistocene plateau basalt (Cover Basalt Fm.; Michelson, 1979). The Cover Basalt covers most of the southern Golan Heights with lows interbedded by layers of fossil clay-rich paleosols of various thicknesses.

Maps, Aerial Views, Plots, Tables, Sections, Trench Logs. Seismic Lines, and Photos
Maps, Aerial Views, Plots, Tables, Sections, Trench Logs. Seismic Lines, and Photos

Maps

Normal Size

Location Map


Figure 2

Morphotectonic map of the studied area (coordinates are in Israeli TM grid). The map is based mainly on areal photos (dated 1945), interpretation, and complementary field work. Mapped are all lineaments, landslides, fluvial systems (with sites of channel displacement marked by open black circle) and alluvial fans located within the study area. Also shown are the sites of the paleoseismic trenches and the highest shore line of the Sea of Galilee, 170 m below sea level, achieved at 25 ka (Hazan et al., 2005).

JW: Ein Gev landslides are shaded purple

Katz et al (2009)


Landslide Hazard Map


Figure 13

Map of landslide hazard around the Sea of Galilee (coordinates are in Israeli TM grid). This qualitative hazard map is a result of integration of the mechanical properties of the exposed rocks, the dip of the geological structure, and the slope gradient, where I to X mark the lowest to greatest hazard, respectively; Negl.—negligible hazard (Katz and Almog, 2006). Tiberias city premises and the studied area are marked by black lines; the sites of the studied landslide (EGLS) and the Berniki Beach landslide (BBLS) are also marked.

Katz et al (2009)


Magnified

Location Map


Figure 2

Morphotectonic map of the studied area (coordinates are in Israeli TM grid). The map is based mainly on areal photos (dated 1945), interpretation, and complementary field work. Mapped are all lineaments, landslides, fluvial systems (with sites of channel displacement marked by open black circle) and alluvial fans located within the study area. Also shown are the sites of the paleoseismic trenches and the highest shore line of the Sea of Galilee, 170 m below sea level, achieved at 25 ka (Hazan et al., 2005).

JW: Ein Gev landslides are shaded purple

Katz et al (2009)


Landslide Hazard Map


Figure 13

Map of landslide hazard around the Sea of Galilee (coordinates are in Israeli TM grid). This qualitative hazard map is a result of integration of the mechanical properties of the exposed rocks, the dip of the geological structure, and the slope gradient, where I to X mark the lowest to greatest hazard, respectively; Negl.—negligible hazard (Katz and Almog, 2006). Tiberias city premises and the studied area are marked by black lines; the sites of the studied landslide (EGLS) and the Berniki Beach landslide (BBLS) are also marked.

Katz et al (2009)


Aerial Views

  • En Gev in Google Earth
    En Gev in Google Earth

    click on image to explore this site on a new tab in Google Earth
  • Annotated En Gev in Google Earth
    Annotated En Gev - North and South Lanslides are outlined in purple

    click on image to explore this site on a new tab in Google Earth
  • En Gev on govmap.gov.il
    En Gev on govmap.gov.il

    click on image to explore this site on a new tab in govmap.gov.il

Dating Plots and Tables

Plots

Studied Trenches


Fig. 5

Age of earthquake events found in the studied trenches (for location, see Fig. 2). Age determined using OSL (solid diamonds; Table 1) or according to field considerations (solid diamonds with question marks). Also shown are: ages of landslide events (squares) and the age of the slopes away from the fault trace at the different trench sites (open diamonds). Error bars are from Table 1.

Katz et al (2009)


Studied Area


Fig. 12

Age and magnitude of earthquakes identified in the studied area. The moment magnitude is calculated using the observed (normal) slip on the fault surfaces and the regression presented by Wells and Coppersmith (1994). Error bars are 95% confidence interval of Wells and Coppersmith (1994). Age determined using OSL (solid diamonds) or ac cording to field considerations (open diamonds with question marks). Error bars for age determination are according to Table 1 or according to field consideration. One earthquake, Figure 12 with age of ~47 ka, has unconstrained magnitude (marked by triangles in the estimated possible magnitude range). Also shown are ages of landslides identified in the studied area.

Katz et al (2009)


OSL Table


Table 1

Summary of luminescence dating results

Katz et al (2009)


Landslide Section


Figure 9

Geometrical comparison of the northern landslide and the slope to its south. (a) Geological cross section of the slope south of the northern landslide
  • Me —Ein-Gev Fm.
  • Mh—Hordos Fm.
  • Pβc —Cover Basalt Fm.
  • Al— alluvium
The [blue] dotted line is a reconstruction of the sliding plane according to:
  1. point 1 is the slide base where the sliding plane daylights, points 2 and 3 are the landslide scar base and top, respectively.
  2. Geological cross section of the northern landslide (the bedrock is not exposed on the slope below the Pβc and therefore the mapped unit is LS: landslide material and colluvium).
  3. Geometrical comparison of the northern landslide (black line) and the slope to its south (gray line), showing the area of material depletion (landslide source) and area of material excess (landslide toe). Overlap is based on the contact of Mh and Pβc and the local slope profile at this contact.
Katz et al (2009)


Trench Logs and Photos

Location Map


Figure 2

Morphotectonic map of the studied area (coordinates are in Israeli TM grid). The map is based mainly on areal photos (dated 1945), interpretation, and complementary field work. Mapped are all lineaments, landslides, fluvial systems (with sites of channel displacement marked by open black circle) and alluvial fans located within the study area. Also shown are the sites of the paleoseismic trenches and the highest shore line of the Sea of Galilee, 170 m below sea level, achieved at 25 ka (Hazan et al., 2005).

JW: Ein Gev landslides are shaded purple

Katz et al (2009)


TEG-0 North Wall


Fig. 4

Log of the northern wall of trench TEG-0 (for location, see Fig. 2)

Katz et al (2009)


TEG-III South Wall


Fig. 6

Log of the southern wall of trench TEG-III (for location, see Fig. 2)

Katz et al (2009)


TEG-I South Wall


Fig. 8

Log of the southern wall of trench TEG-I (for location, see Fig. 2)

Katz et al (2009)


South Wall of Northern Landslide Trench


Fig. 11

Log (a) and a picture (b) of the southern wall of the northern landslide trench (for location, see Fig. 2). Hatched areas marked by A1–A3 are the colluvium proiles described in the text. Some basalt blocks are marked by dark gray numbers 1–3 on the log and on the picture.

Katz et al (2009)


Older Versions of Trench Logs (with different dates)

  • from Amit. R., Katz, O., Yagoda-Biran G., Hatzor, Y.H., 2009. Paleoseismology of the eastern Sea of Galilee . Dead Sea Workshop Field Guide. 49-53.
  • This may explain discrepancies between dates reported in theses of Braun and Kagan and the article by Katz et al (2009)
TEG-A (aka TEG-0)


Fig. 2a

Paleoseismic trenches: Two events: ~ 11ka and ~4ka

Amit et al. (2009)


TEG-B (aka TEG-III)


Fig. 2b

Paleoseismic trenches: two events: ~40ka and ~4 ka

Amit et al. (2009)


TEG-C (aka TEG-I)


Fig. 2c

Paleoseismic trenches: Two events: ~40ka and ~10ka

Amit et al. (2009)


Seismic Lines

Location Map

Figure 10a

High resolution reflection profile along the northern landslide (coordinates are in Israeli TM grid).
  • The trace of the reflection profile (in green)
  • the northern landslide outline (dashed line)
  • mapped faults (blue lines; appear also in Fig. 2)
  • projection of the faults (F) and the listric discontinuities (L) on the surface, interpreted from the reflection section
  • T marks the site of the trench (shown in Fig. 11)
Katz et al (2009)


Interpreted Seismic Line


Figure 10b

Interpreted reflection profile (time domain; two way travel).
  • projection of the faults (F) and the listric discontinuities (L) on the surface, interpreted from the reflection section
  • T marks the site of the trench (shown in Fig. 11)
Katz et al (2009)


Uninterpreted Seismic Line


Figure 10c

Uninterpreted reflection profile (time domain; two way travel).

Katz et al (2009)


Photos

Normal Size

  • Fig. 3 - Surface rupture and
    Figure 3

    Surface rupture and the scarp of the youngest fault studied

    1. Surface rupture of a colluvium unit. This site was later used for paleoseismic trenching (trench TEG-0 in Fig. 2).
    2. Surface rupture of bedrock (Eocene Maresha Fm.), about 100 m north of (a).
    3. Close-up of one of the scarps that appear in (b).


    Katz et al (2009)
    scarp of the youngest fault studied from Katz et al (2009)
  • Fig. 7 - Highly fractured
    Figure 7

    Highly fractured bedrock of Ein-Gev Fm. (marked by dashed ellipse) bounding the fault exposed in trench TEG-III (Unit 6b in Fig. 6).

    Katz et al (2009)
    bedrock of Ein-Gev Fm. exposed in Trench TEG-III from Katz et al (2009)

Magnified

  • Fig. 3 - Surface      
    Figure 3

    Surface rupture and the scarp of the youngest fault studied

    1. Surface rupture of a colluvium unit. This site was later used for paleoseismic trenching (trench TEG-0 in Fig. 2).
    2. Surface rupture of bedrock (Eocene Maresha Fm.), about 100 m north of (a).
    3. Close-up of one of the scarps that appear in (b).


    Katz et al (2009)
    rupture and scarp of the youngest fault studied from Katz et al (2009)
  • Fig. 7 - Highly fractured
    Figure 7

    Highly fractured bedrock of Ein-Gev Fm. (marked by dashed ellipse) bounding the fault exposed in trench TEG-III (Unit 6b in Fig. 6).

    Katz et al (2009)
    bedrock of Ein-Gev Fm. exposed in Trench TEG-III from Katz et al (2009)

Chronology
ca. 3000 BCE Earthquake

The northern Ein Gev landslide exhibited a multi-phase sliding history where the youngest slide was dated to ca. 5 ka BP which is equivalent to ca. 3000 BCE. According to Kagan (2011), the error bar on this event is ± 300 years meaning that it struck between 2700 and 3300 BCE.

mid 8th century CE (?) Earthquake

Figures

Figures

Normal Size

  • Fig. 3 - Surface rupture and
    Figure 3

    Surface rupture and the scarp of the youngest fault studied

    1. Surface rupture of a colluvium unit. This site was later used for paleoseismic trenching (trench TEG-0 in Fig. 2).
    2. Surface rupture of bedrock (Eocene Maresha Fm.), about 100 m north of (a).
    3. Close-up of one of the scarps that appear in (b).


    Katz et al (2009)
    scarp of the youngest fault studied from Katz et al (2009)

Magnified

  • Fig. 3 - Surface rupture and
    Figure 3

    Surface rupture and the scarp of the youngest fault studied

    1. Surface rupture of a colluvium unit. This site was later used for paleoseismic trenching (trench TEG-0 in Fig. 2).
    2. Surface rupture of bedrock (Eocene Maresha Fm.), about 100 m north of (a).
    3. Close-up of one of the scarps that appear in (b).


    Katz et al (2009)
    scarp of the youngest fault studied from Katz et al (2009)

Discussion

The northern Ein Gev landslide exhibited a multi-phase sliding history where the youngest slide was dated to before ca. 5 ka BP. Because the youngest slide had a fresh face with no colluviation (see Fig. 3), Katz et al (2009:289) suggested the possibility that an additional event might have occurred after 5 ka BP which might be related to one of the mid 8th century CE earthquakes.

Master Seismic Events Table
Master Seismic Events Table

Landslide analysis
Landslide analysis

The geometry and geology of the landslides

The study area hosts two large landslides (Fig. 2). The length of the northern and southern landslides (scar to toe) is 1500 m and 1000 m respectively, ranging in width from tens to hundreds of meters. The landslides span the entire slope height of about 500 m, and are the only places along the study area where large rock-blocks (volume >1 m3) of the Cover Basalt Fm. (exposed in situ at the upper part of the slope) are widely distributed at its foot. The scar and toe morphology suggests that sliding took place in a slump mechanism (Varnes, 1978). The landslides are spatially correlated with the studied fault segments, either covering the faults or being cut and displaced (normally) by them (Fig. 2). These field relations point to correlated events of earthquakes and landslides, thus to the working assumption that the landslides might be earthquake-induced.

Fig. 2
Figure 2

Morphotectonic map of the studied area (coordinates are in Israeli TM grid). The map is based mainly on areal photos (dated 1945), interpretation, and complementary field work. Mapped are all lineaments, landslides, fluvial systems (with sites of channel displacement marked by open black circle) and alluvial fans located within the study area. Also shown are the sites of the paleoseismic trenches and the highest shore line of the Sea of Galilee, 170 m below sea level, achieved at 25 ka (Hazan et al., 2005).

JW: Ein Gev landslides are shaded purple

Katz et al (2009)


The deep structure of the northern landslide

We studied the deep structure of the northern landslide and the sliding surface depth using geometrical analysis following Masson et al. (2002), as well as a high-resolution seismic reflection survey.

Assuming that the intact slope south of the land slide is similar in its topography and geology to the pre-failure slope of the studied landslide, a geometrical comparison by overlapping the geological cross sections of the intact and the landslide slopes, with hinge points along the contact of the Ein Gev with the overlaying Cover Basalt formations, reveals areas with respective material depletion (Fig. 9, intact slope proile above landslide profile) and material excess (Fig. 9, intact slope profile below landslide profile). These are the source and the deposition parts of the landslide, respectively. This structure of source and deposition martial is in agreement with the slump type assumed for this landslide according to the surface geometry. The vertical difference between the two profiles, about 30 m, is a reasonable approximation of the landslide thickness. This thickness is in accordance with other field-observed ratios of landslide thickness t, vs. landslide length l, of 0.05–0.15 (Guzzetti et al., 2009; where l is defined as √A and A is the landslide surface area). For the landslide studied, √A is ~500 m and t is ~30 m, thus t/l is ~0.06.

Fig. 9
Figure 9

Geometrical comparison of the northern landslide and the slope to its south. (a) Geological cross section of the slope south of the northern landslide
  • Me —Ein-Gev Fm.
  • Mh—Hordos Fm.
  • Pβc —Cover Basalt Fm.
  • Al— alluvium
The [blue] dotted line is a reconstruction of the sliding plane according to:
  1. point 1 is the slide base where the sliding plane daylights, points 2 and 3 are the landslide scar base and top, respectively.
  2. Geological cross section of the northern landslide (the bedrock is not exposed on the slope below the Pβc and therefore the mapped unit is LS: landslide material and colluvium).
  3. Geometrical comparison of the northern landslide (black line) and the slope to its south (gray line), showing the area of material depletion (landslide source) and area of material excess (landslide toe). Overlap is based on the contact of Mh and Pβc and the local slope profile at this contact.
Katz et al (2009)


The high resolution seismic reflection survey profile reveals several faults and discontinuities along the studied landslide (Fig. 10). All discontinuities can be detected very close to the slope surface and thus they are apparently displacing the mass of the landslide, and therefore are assumed to be active post sliding. Discontinuities mapped by the seismic reflection survey can be divided into two types:
  1. Faults, steeply dipping westwards and cutting the entire section (infferred depth of about 200 m). Most of these detected faults have an expression at the surface and were mapped north and south of the landslide (Fig. 10). Only one of the faults interpreted from the reflection proile ruptures the surface, across the landslide toe (Fig. 2).

  2. Listric faults. These are sub-vertical close to the landslide surface, and turn shallow at a depth of few tens of meters where they terminate, apparently along a common surface (Fig. 10).
This structure is interpreted as typical to slumps (Skempton and Hutchin son, 1969) where secondary sliding surfaces develop above the major sliding surface as part of post-failure slope stabilization processes. This interpretation is in accordance with the geometrical analysis (described above) suggesting a depth of a few tens of meters for the sliding plane.

Fig. 2
Figure 2

Morphotectonic map of the studied area (coordinates are in Israeli TM grid). The map is based mainly on areal photos (dated 1945), interpretation, and complementary field work. Mapped are all lineaments, landslides, fluvial systems (with sites of channel displacement marked by open black circle) and alluvial fans located within the study area. Also shown are the sites of the paleoseismic trenches and the highest shore line of the Sea of Galilee, 170 m below sea level, achieved at 25 ka (Hazan et al., 2005).

JW: Ein Gev landslides are shaded purple

Katz et al (2009)
Fig. 10 Seismic Lines

Location Map

Figure 10a

High resolution reflection profile along the northern landslide (coordinates are in Israeli TM grid).
  • The trace of the reflection profile (in green)
  • the northern landslide outline (dashed line)
  • mapped faults (blue lines; appear also in Fig. 2)
  • projection of the faults (F) and the listric discontinuities (L) on the surface, interpreted from the reflection section
  • T marks the site of the trench (shown in Fig. 11)
Katz et al (2009)


Interpreted Seismic Line


Figure 10b

Interpreted reflection profile (time domain; two way travel).
  • projection of the faults (F) and the listric discontinuities (L) on the surface, interpreted from the reflection section
  • T marks the site of the trench (shown in Fig. 11)
Katz et al (2009)


Uninterpreted Seismic Line


Figure 10c

Uninterpreted reflection profile (time domain; two way travel).

Katz et al (2009)


The shallow structure of the northern landslide

We opened a trench across the northern landslide where one of the faults leaves a morphological expression on the landslide surface (Fig. 2). The trench, 2 m deep and 13 m long, exposed a sequence of colluvial sediments (Fig. 11). Due to its relatively shallow depth, it did not reveal a discrete sliding plane marking the base of the landslide.

Fig. 2
Figure 2

Morphotectonic map of the studied area (coordinates are in Israeli TM grid). The map is based mainly on areal photos (dated 1945), interpretation, and complementary field work. Mapped are all lineaments, landslides, fluvial systems (with sites of channel displacement marked by open black circle) and alluvial fans located within the study area. Also shown are the sites of the paleoseismic trenches and the highest shore line of the Sea of Galilee, 170 m below sea level, achieved at 25 ka (Hazan et al., 2005).

JW: Ein Gev landslides are shaded purple

Katz et al (2009)


Fig. 11
Fig. 11

Log (a) and a picture (b) of the southern wall of the northern landslide trench (for location, see Fig. 2). Hatched areas marked by A1–A3 are the colluvium proiles described in the text. Some basalt blocks are marked by dark gray numbers 1–3 on the log and on the picture.

Katz et al (2009)


The colluvium at this site is composed of three colluvial units, in each of which a soil was developed. The lower unit A1 (Fig. 11) is composed of clayey material with 10% gravel (0.1 m in average). A well developed calcic soil was formed in the colluvium (A1) with a high amount of calcic nodules and gravel coated by secondary calcium carbonate. This unit is dated 65±5 ka and 69±5 ka (Table 1; Fig. 5). Unit A2 is a mixture of clay with 40% gravel. The gravel is bi-modal with 80% granules (2–3 mm) and ~20% large (>0.3 m) boulders. A moderately developed calcic soil was developed in this unit, with scattered calcic nodules and a small amount of coated gravel. This unit is dated 6.5±0.5 ka (Table 1; Fig. 5). The soils in Units A1 and A2 are calcic but differ in their degree of development, supporting different time periods of slope instability. Unit A3 is silty–clay with 20% gravel (0.03 m size, some >0.1 m) and weak AC soil, non-calcic. This colluvial unit, which covers the recent landslide surface, was dated at 5.2±0.4 ka (Table 1; Fig. 5). The relatively young age of this unit (~ ca. 5 ka) might explain the lack of diagnostic calcic horizon.

Fig. 5
Fig. 5

Age of earthquake events found in the studied trenches (for location, see Fig. 2). Age determined using OSL (solid diamonds; Table 1) or according to field considerations (solid diamonds with question marks). Also shown are: ages of landslide events (squares) and the age of the slopes away from the fault trace at the different trench sites (open diamonds). Error bars are from Table 1.

Katz et al (2009)


The ages of the colluvial units support different time periods of slope instability and rule out the possibility of a single sliding event. The lower unit (A1) is the oldest and the two upper units (A2 and A3) were deposited later, after a long period during which no colluviation occurred along the slope. This suggests multiple sliding events separated by long periods of quiescence.

The colluvium at this site shows local deviation from slope parallel deposition, probably as a result of changing local topography and relief above the trace of the fault (Fig. 11). The proximity of the landslide deposit, fault trace, and disturbed colluvium suggests both seismic triggering for the landslide, and deposition of the three colluvium units on an existing fault scarp.

Slope Stability Analysis

Slope stability analysis was performed on the northern Ein Gev landslide using a Pseudo-Static Back Analysis and the method of Slices (Morgenstern and Price, 1965 method) in two dimensions using the software SLOPE/W from GEO-SLOPE International Ltd. Because the northern Ein Gev landslide exhibited a multi-phase sliding history, the sandstone of the Ein Gev formation (the unit that failed) was mechanically tested in two different states to extract an Initial Peak Shear Strength and a Residual Strength that would exist after the earliest failure. Test results are listed below:

Sample Mechanical State Cohesion
(kPa)		 Friction Angle Factor of Safety
from Static Analysis		 Critical Acceleration
Pristine Rock Peak Shear Strength 376 43° 4.5 0.95 g
Deformed Rock Residual Strength 0 38° 2.8 0.37 g
In order to assess the possibility that a severe rain storm could have caused the observed landslides, additional static analysis was performed at full water saturation with the water table at ground level. The slopes were found to be stable under these conditions suggesting that the landslides were seismically induced. Paleoseismic trenching also associated a seismic event at 5 ka BP with a landslide failure. Critical acceleration to induce sliding was estimated to 0.95 g when the sandstone of the Ein Gev formation was at Peak Shear Strength and 0.37 g at Residual Strength. This translates into local intensities of 9.2 and 7.7 respectively when using Wald et al (1999) for the conversion. Thus minimum PGA is estimated at 0.37 g and minimum Intensity is estimated at 7.7 for the presumably seismically induced landsliding events observed in the northern Ein Gev landslide.

Paleoseismic Analysis
Paleoseismic Analysis

Location Map


Figure 2

Morphotectonic map of the studied area (coordinates are in Israeli TM grid). The map is based mainly on areal photos (dated 1945), interpretation, and complementary field work. Mapped are all lineaments, landslides, fluvial systems (with sites of channel displacement marked by open black circle) and alluvial fans located within the study area. Also shown are the sites of the paleoseismic trenches and the highest shore line of the Sea of Galilee, 170 m below sea level, achieved at 25 ka (Hazan et al., 2005).

JW: Ein Gev landslides are shaded purple

Katz et al (2009)


Introduction

We opened three trenches across three of the mapped N–S-oriented normal faults (Fig. 2). These faults are part of the mapped fault zone and are separated by ca. 100 m.

TEG-0

Fig. 4 - Trench Log


Fig. 4

Log of the northern wall of trench TEG-0 (for location, see Fig. 2)

Katz et al (2009)


A normal fault displaces the slope surface at the southern end of the study area, where chalky limestone (slightly sandy) of the Eocene Maresha Fm. crops out. In places the fault scarp is completely covered by colluvial material, and elsewhere along its trace it shows a recent scarp exposing the chalky limestone of the Maresha Fm. (Fig. 3). Paleoseismological analysis reveals two faulting events, and an additional young event exposing the above-described scarp. However, the dating results are highly scattered so it was not possible to determine confidently the age of the tectonic events, except for the younger event revealed in the trench. This is related to the fact that the samples for the OSL dating are very heterogeneous, with large proportion of aliquots representing the parent material and poorly bleached grains. In the following section we describe the sequence of events revealed in trench TEG-0.

Phase 1 - A faulting event that displaced the slope surface vertically by ca. 0.5 m. This event was followed by deposition of colluvial wedge (Fig. 4; Unit 3) that did not yield reliable age determination (Fig. 5). During the build up of this colluvial wedge, a weak AC soil profile was developed. The soil is composed of an organic A horizon and a C horizon composed of brown silty-clay matrix, weak soil structure (angular) with no calcic nodules or any other secondary deposition.

Fig. 5
Fig. 5

Age of earthquake events found in the studied trenches (for location, see Fig. 2). Age determined using OSL (solid diamonds; Table 1) or according to field considerations (solid diamonds with question marks). Also shown are: ages of landslide events (squares) and the age of the slopes away from the fault trace at the different trench sites (open diamonds). Error bars are from Table 1.

Katz et al (2009)


Phase 2 - A second event dated to about 5±0.3 ka (Table 1; Fig. 5) displaced the slope surface vertically by about 0.8 m (Fig. 4). This event caused the deposition of a short colluvial wedge (Unit 4) that buried part of the previous slope (Unit 3). The most indicative feature characterizing the two colluvial wedges is two organic horizons located at the top of each wedge. The soil developed along colluvial wedge 4 is more developed than the soil that was buried by colluvial wedge 3, suggesting a longer time of soil formation.

To summarize, the slope at the site of Trench TEG-0 was faulted at least twice (Fig. 5), with the second event occurring during the upper Holocene (ca. 5 ka). The weakly developed soil proile in colluvial Unit 3 might indicate that the first event was close in time to the second event (ca. 5 ka), probably occurred during the Holocene. The fresh rocky free face exposed further north along the fault suggests that an additional, third, faulting event, might have occurred after the last dated event (i.e., LT 5 ka).














TEG-III

Fig. 6 - Trench Log


Fig. 6

Log of the southern wall of trench TEG-III (for location, see Fig. 2)

Katz et al (2009)


The normal fault at the site of trench TEG-III displaces the Hordos and Ein Gev formations. This fault ruptures the surface about 200 m east of the fault studied at the site of trench TEG-O (Fig. 2). The expression of the fault at the surface is mild, with a ca. 1 m fault scarp. The faulted slope material, away from the fault-related colluvial wedges, has a minimal age of 82±7 ka (Table 1 and Fig. 5). In the following section we describe the sequence of events revealed in this trench.

Fig. 5
Fig. 5

Age of earthquake events found in the studied trenches (for location, see Fig. 2). Age determined using OSL (solid diamonds; Table 1) or according to field considerations (solid diamonds with question marks). Also shown are: ages of landslide events (squares) and the age of the slopes away from the fault trace at the different trench sites (open diamonds). Error bars are from Table 1.

Katz et al (2009)


Phase 1: A faulting event that displaced the slope surface vertically but with unknown displacement magnitude. This event was followed by deposition of gravelly alluvial material on top of the faulted block by a local alluvial fan draining the eastern steep slopes (Unit 1; Fig. 6). The age of the alluvial material, mostly reworked gravel from the Hordos Fm., is be tween 47.4±4 and 53±8 ka (Fig. 5). No soil was found on top of the gravelly material, suggesting a short time between the deposition of the alluvium and the burial of the down faulted block by the colluvial unit (Unit 2; Fig. 6).

Phase 2: A second faulting event displaced the slope by about 0.5 m, followed by the deposition of colluvial Unit 2 on top of Unit 1 (Fig. 6). This colluvial wedge (Unit 2) is composed of a large amount of intensely jointed and shattered rock blocks (ca. 50% of the rock blocks), most of which were derived from the fault zone and shattered in situ (e.g., 6b Fig. 6). This shattering (Fig. 7) is indicative of fault zones (Sagy et al., 2001; Katz et al., 2003). The paleosol developed in the colluvium (Unit 2) is calcic soil stage I, the silty–clay matrix is slightly cemented by calcium carbonate, with no clear calcic nodules or clay cutans and a low amount of organic material. The degree of soil development indicates a short time period of surface exposure. This event occurred at ca. 36±3 ka, a short time (several thousand of years) after the deposition of the alluvium of phase 1 (Figs. 5, 6).

Fig. 7
Figure 7

Highly fractured bedrock of Ein-Gev Fm. (marked by dashed ellipse) bounding the fault exposed in trench TEG-III (Unit 6b in Fig. 6).

Katz et al (2009)


Phase 3: A third faulting event displaced the surface by ca. 0.15 m. The colluvial wedge (Unit 3; Fig. 6) formed after this faulting event is composed mainly of silty–clay ine material and 20–30% gravel. Calcic soil developed in this unit reaches stage IV, with a Bk horizon. The upper 0.2–0.3 m has a high amount of root holes, fragments of snail shells, and borrows. Patinated gravel and rock fragments are scattered in the colluvial material derived from the damaged, faulted slope. In contrast to colluvial Unit 2, the degree of soil development in colluvial Unit 3 supports a long time period of slope stability. The colluvium age is 40±3, indistinguishable from phase 2 (Fig. 5).

Phase 4: A faulting event that displaced the slope by about 0.4 m (Fig. 6) occurred at ca.10±0.8 ka (Fig. 5), after a long period of quiescence (ca. 35 ka). The event was followed by the deposition of colluvial Unit 4 (Fig. 6). Most of the colluvium is composed of ine material (silty–clay) with a small amount of gravel and rock fragments. Most of the rock fragments are patinated, indicating sources from the pre-faulted slope surface. The soil of Unit 4 is calcic, with a clear organic A horizon and a B horizon slightly cemented by calcium carbonate. The lower part of the soil is less cemented by calcium carbonate, followed by deposition of gypsum mycelia.

To summarize, this slope has a long history of tectonic activity, only part of which is exposed in the trench. At least four events were identified (Fig. 5): three events during the upper Pleistocene and one during the Early Holocene after a long period of quiescence.

TEG-I

Fig. 8 - Trench Log


Fig. 8

Log of the southern wall of trench TEG-I (for location, see Fig. 2)

Katz et al (2009)


The trench at site TEG-I exposes a fault located 100 m east of the fault studied in trench TEG-III (Fig. 2). The trenched area is comprised mostly of colluvial material (Unit 3; Fig. 8) with no steep fault scarp. The colluvium partly derives from the sand stone of the Ein-Gev Fm. (Unit 5; Fig. 8) and partly consists of limestone and chert lithoclasts exposed at the upper part of the slope. The colluvium consists mostly of rock, supported by large angular rock blocks and gravel ranging in size from 0.5 m to ca. 0.15 m. The matrix comprises non-sorted coarse silt, sand, grit, and granules. Calcic soils were developed in the col luvial units and reached stages between II and III. At the ca. 1 m wide fault zone, colluvial Units 3 and 1 are intensely cracked and jointed (joints are a few mm to a few cm wide). In the following section we describe the sequence of events revealed in this trench.

Phase 1: Displacement of colluvial Unit 3 by about 1.60 m, which occurred ca. 35±3 ka. The faulted unit is composed of sand (coarse and fine) with a small amount of silt and gravel, 3–4 cm in size. The gravel consists of sandstone, limestone, and chert. Calcic soil developed in this unit and reached stage II-III, with calcium carbonate enveloping the gravel and disseminated in the sandy matrix. The age of this unit at the downthrown block is 251±14 ka (Fig. 8); this should be considered a minimum age. At the fault zone and at the up-thrown block close to the surface of Unit 3, the ages are 102±8 ka and 107±8 ka, respectively. The colluvial Unit 2, deposited on top of Unit 3 on the down thrown block (Fig. 8), consists of massive clay–silty colluvium mixed with small gravel (3–5 cm). A calcic soil developed in this unit and reached stage II-III. Most of the calcium carbonate is deposited at the up per 1 m of the unit, its amount decreasing downward. Orthic and disorthic calcium carbonate nodules are scattered in the fine matrix, most of them ca. 0.5 cm. Some of the gravel is coated with calcic crust. The disorthic nodules and the coated clasts scattered in the colluvium are a result of soil erosion and deposition on the down-faulted block. The orthic calcic nodules were formed during the quiescence in between the tectonic events.

Phase 2: A second tectonic event occurred at 9.2±1.9 ka, displacing the slope surface by ca. 0.4 m (Fig. 8). As a result, colluvial Unit 1 was deposited on the downthrown block and covered the entire slope. This colluvial unit (Unit 1) consists of non-sorted gravel ranging between 0.03 m and 0.2 m. In addition, 0.005-0.01 m granules are scattered in the matrix. The matrix consists of silt and fine sand, with a high amount of organic material. A weak clayey-gypsic soil developed continuously along the slope and welded into the calcic soil at Unit 2.

We identified two events in trench TEG-I, one event at ca. 35 ka (Fig. 5) and the other at ca. 9 ka. The well developed soil in Unit 2 and the weak soil profile in Unit 1 suggest long quiescence between events.

Fig. 5
Fig. 5

Age of earthquake events found in the studied trenches (for location, see Fig. 2). Age determined using OSL (solid diamonds; Table 1) or according to field considerations (solid diamonds with question marks). Also shown are: ages of landslide events (squares) and the age of the slopes away from the fault trace at the different trench sites (open diamonds). Error bars are from Table 1.

Katz et al (2009)


Paleoseismic interpretation of the faults and landslides

We observed evidence for at least five and as many as seven surface-rupturing earthquakes, with magnitudes of at least 6 (e.g., McCalpin, 1996). The oldest earthquake occurred in the upper Pleistocene, ca. at 45 ka (Figs. 5, 12; trench TEG-III). The second and third events occurred at 35–40 ka; the event at 35 ka was detected in two trench sites (Fig. 5; trench TEG III and TEG-I). A fourth event, in the early Holocene, ca. 10 ka, was detected in two trench sites (TEG-I and TEG-III; Fig. 5). A younger event was detected only in one site (trench TEG-0, Fig 5). Because of dating uncertainties (see Table 1) we could not determine a reliable age for this event; however, according to the weakly developed soil profile (Fig. 4), it is possible that the earthquake occurred not long before the dated one, meaning during the Holocene. This event might coincide with the event at ca. 10 ka found in trenches TEG-I and TEG-III). A fifth event, around 5 ka, was detected only in one site (trench TEG-0, Fig. 5). This event was detected at the westernmost and youngest fault studied (Fig. 2). The scarp of the youngest fault in this study exposes a fresh rock face with no colluviation, which suggests an additional event might have occurred on this fault post 5 ka. This might be related to the devastating historical earthquake that occurred at 749 CE (Ambraseys et al., 1994; Guidoboni et al., 1994; Karcz; 2004) in the northern Jordan valley. It is reported to have severely damaged the city of Tiberias, 10 km across the SOG (Marco et al., 2003), the city of Susita (Yagoda-Biran and Hatzor, 2010), the village of Umm El Kanatir (Wechsler et al., 2009), and a few other places, all near the SOG (Marco et al., 2003).

We note that aerial photos taken in 1945 reveal additional, probably younger, fault segments further to the west of the study area towards the current shores of the SOG (Fig. 2). These could not be studied due to current intense cultivation of this area. Using the observed normal slip on the fault surfaces and the regression presented by Wells and Coppersmith (1994), the calculated magnitudes for the paleo earthquakes in the studied sites range between MW 6.1 and MW 6.8, (Fig. 12). Co-seismic slip was probably to some extent larger than the observed estimates, because erosional processes likely reduced scarp heights over time. The calculated magnitudes are thus likely minimal estimates. According to paleoseismic observations, the recurrence interval for MW >6 earth quakes at this site is 7–10 ka, based on observations of 5–7 such earthquakes in the last 50 ka. However, integrating all paleoseismic observations along the northern Jordan Valley with historical and recent re cords yields shorter return periods of about 500 years and about 1500 years for earthquakes of MW >6 and MW >6.5, respectively (Zilberman et al., 2000, 2004; Begin, 2005; Hamiel et al., 2009). This discrepancy is apparently due to incompleteness of the paleoseismic record, owing to the fact that the current study is focused on the marginal faults zone which is only a part of the pull-apart structure of the SOG.

Morphological, structural, and mechanical data suggest that the landslides exposed in the studied area are earthquake-induced. The detailed analysis performed on the northern landslide reveals that at least one sliding event coincides temporally with earthquake event. Dates determined using OSL suggest that the younger phase of landsliding at 5.2±0.5 ka (Fig. 11) is temporally associated with the youngest event (5.0±0.3 ka, Fig. 12). The older landsliding event was not recognized in the trenches that exposed younger sediments. We thus conclude that it might have been triggered by a strong event prior to 65 ka.

The association of slope failure and fault activity discussed above is not unique to the study area. Across the SOG, on its western shore at Berniki Beach area (Fig. 1), a similar association of landslide and faults has been documented. In this area, a 1000-m-long landslide is crossed by a fault at its toe (Yagoda-Biran et al., 2010). All lineaments mapped in this area, which is a few hundred meters west of the inferred location of the SOG West Marginal Fault (Hurwitz et al., 2002), have a normal component revealed at the surface and partly also by a seismic survey (Medvedev, 2008). Ac cording to the ield relations at this site and the analysis made in the study area, we propose that the Berniki Beach landslide, situated within the sandstone of Hordos Fm., is also earthquake induced. Indeed Yagoda-Biran et al. (2010) concluded that seismic acceleration was the likely triggering mechanism, as calculations reveal that the slope would have remained statically stable even if the entire slope had been water saturated. The special geomorphologic–geologic relationship where weak sedimentary rocks compose steep slopes associated with faults is found on both the western and the eastern sides of southern SOG. These slopes are strong enough to be stable under static conditions, but strong earthquakes with epicentral distance of no more than a few km can induce dynamic instability and slope failure (Yagoda-Biran et al., 2010).

Calculator
Convert PGA to Intensity

Variable Input Units Notes
g Peak Horizontal Ground Acceleration
Variable Output - Site Effect not considered Units Notes
unitless Conversion from PGA to Intensity using Wald et al (1999)
  

Notes and Further Reading
References

Articles and Books

Amit. R., Katz, O., Yagoda-Biran G., Hatzor, Y.H., 2009. Paleoseismology of the eastern Sea of Galilee . Dead Sea Workshop Field Guide. 49-53.

Katz, O., et al. (2009). "Quaternary earthquakes and landslides in the Sea of Galilee area, the Dead Sea Transform: Paleoseismic analysis and implication to the current hazard ." Israel Journal of Earth Sciences 58: 275-294.

Morgenstern, N.R. and Price, V.E. (1965) The Analysis of the Stability of General Slip Surfaces . Géotechnique, 15, 79-93.

Yagoda-Biran, G., et al. (2010). "Constraining regional paleo peak ground acceleration from back analysis of prehistoric landslides: Example from Sea of Galilee, Dead Sea transform ." Tectonophysics 490(1–2): 81-92.

Slope Stability Modeling with Geo Studio by GEO-SLOPE international Ltd. 2004-2021

Websites

Dr. Gony Yagoda-Biran Researcher, Head of the Geological Hazards Division at Geological Survey of Israel

Dr. Oded Katz Senior researcher, Stratigraphy and Subsurface Research Division at Geological Survey of Israel

Wikipedia Pages

Slope Stability analysis



Rayleigh wave



Love wave



Paleoseismology of the Eastern Sea of Galilee - Amit et al. (2009)

  • from Amit. R., Katz, O., Yagoda-Biran G., Hatzor, Y.H., 2009. Paleoseismology of the eastern Sea of Galilee . Dead Sea Workshop Field Guide. 49-53.
  • Note that trench names changed from Amit et al. (2009) to Katz et al (2009)
  • TEG-A became TEG-0
  • TEG-B became TEG-III
  • TEG-C became TEG-I
Figure 1

Morphotectonic map of the Eastern Sea of Galilee

Amit et al. (2009)


Field guide

The study site is located between the western slopes of the Golan Heights and the eastern coast of the Sea of Galilee (Fig1) in a seismically active area as part of the Dead Sea Rift (DSR). In this area landslides and faulted and tilted blocks shape the steep, 600 m high slopes of the southern Golan Heights. The Miocene–to-Quaternary sequence which is exposed on these slopes is composed of sedimentary rocks such as marls, sand, fluvial sediments, lake sediments and basalts of Pliocene-Pleistocene age. This geological setting and lithology enhance development of compound landslides of several kinds (e.g. earth flow, slump, creep) with slide units comprising, in places, complete rock sequences. The faulting, the westward dipping strata and the steep slopes play significant roles in the sliding process, resulting in a terraced morphology of these steep slopes.

Three trenches were opened crossing three of the mapped N-S oriented normal faults composing 800 m wide fault zone (Fig 1). These faults are ~ 200 m apart from each other, crossing a series of tilted blocks .

TEG –A (Fig 2a)

A normal fault displacing the slope of one of the tilted blocks. This tilted block is composed mainly of intercalation of chalky limestone (slightly sandy) of the Maresha Formation (Middle Eocene). The fault scarp is partly covered by colluvial material towards the top and part of the scarp has a free face exposing the chalky limestone of the Maresha Formation. The paleoseismological analysis revealed that the fault is a multiple type with two identifiable faulting events. The age of the faulted slope is between 47±8 ka and 3.5±1.4. The slope which was stable during the upper Pleistocene was faulted twice during the Holocene with one event during the early Holocene (~ 11 ka) displacing ~ 1m (colluvial wedge 3) and another event during the late Holocene (~ 3.5 ka) displacing ~80 cm (colluvial wedge 4). The fresh rocky free face, which is exposed more to the north along the fault trace, might suggest that an additional faulting event occurred later than the last dated event (LT 3.5±1.4 ka).


Fig. 2a

Paleoseismic trenches: Two events: ~ 11ka and ~4ka

Amit et al. (2009)


TEG –B (Fig 2b)

A normal eastern fault crossing a tilted block is displacing the Hordos and En Gev formations. The faulted slope has a minimal age of 78.6±13 ka. Three events were identified; two that occurred close in time (during the late Pleistocene) and one that occurred during the Holocene after a long quiescence of about 35 ky. The last event, which occurred ~ 5.7±2.2 ka, displaces the slope by ~ 40 cm and was also detected in trench TEG – A, a parallel fault line located more to the west (Fig 1).


Fig. 2b

Paleoseismic trenches: two events: ~40ka and ~4 ka

Amit et al. (2009)


TEG –C (Fig 2c)

A fault displacing the toe of the two landslides in the study site was trenched (Fig 1). The trench was opened along the southern margin of the southern landslide (Fig 1). The trenched area is composed mostly of colluvial material (unit 3) partly derived from the white-yellowish sandstone of the En-Gev formation exposed at the upper part of the slope (unit 5) and partly composed of limestone and chert lithoclasts. The colluvium is mostly rock supported with large angular rock blocks, some of which reach 50 cm, and gravel ~ 15 cm. The matrix is composed of unsorted coarse silt, sand, grit and granules. Calcic soil developed in this unit. The calcium carbonate envelopes the gravel and is disseminated in the sandy matrix. The soil reached stage II-III. At the fault zone, which is ~ 1 m wide, the colluvial unit 3 and 1 are intensely cracked and jointed (between 1–2 mm and 5-10 cm wide joints) (Fig 2).

Phase 1: Displacement of unit 3 by about 1.60 m which occurred ~ 37±3 ka (at TEG - C ~ 36 ka). The faulted unit is composed of sand (coarse and fine) with small amounts of silt and gravel of 3-4 cm size. The gravel is composed of sandstone, limestone and chert. Calcic soil developed in this unit and reached stage II-III with calcium carbonate enveloping the gravel and disseminated in the sandy matrix. The age of this unit in the downthrown block is 255±19 ka (Fig 2c; TEG-45). At the fault zone (Fig 2c; TEG-43) and in the upthrown block close to the surface (Fig 2c; TEG 46) the age of unit 3 is 97±18 and 99±12 ka, respectively. The colluvial unit 2 which is deposited on top of unit 3 on the downthrown block is composed of massive clay-silty colluvium mixed with small gravel (3-5 cm). A calcic soil developed in this unit and reached stage II-III. Most of the calcium carbonate is deposited in the upper 1 m of the unit, with decreasing amounts downward. Orthic and disorthic calcium carbonate nodules are scattered in the fine matrix, most of them ~ 2 mm. Some of the gravel is coated with calcic crust. The disorthic nodules and the coated clasts scattered in the colluvium are the result of soil erosion and deposition on the downfaulted block. The orthic calcic nodules were formed during the quiescence in between the tectonic events.

Phase 1: The second tectonic event occurred ~ 9 ka ago (Fig 2c), displacing the slope by ~40 cm. As a result, colluvial unit 1 was deposited on the downthrown block and covered the whole slope. This colluvial unit (unit 1) is composed of unsorted gravel which ranges between 20 cm, the largest, and 3 cm, the smallest. In addition, granules of the size of 0.5 – 1 cm are also scattered in the matrix. The matrix is composed of silt and fine sand and a high amount of organic material. A weak clayey-gypsic soil developed continuously along the slope and integrated into the calcic soil at unit 2.

Two events were identified in this trench. One occurred at ~ 37 ka (Fig.2c ;TEG-42: 37±3) and another occurred ~ 9 ka ago (TEG- 44; 9.2±1.9). The well-developed soil in unit 2 and the weak soil profile in unit 1 point to a long quiescence between events.

Summary

The oldest event occurred ~ 37 ka ago and was detected in two of the trenches, TEG-B and TEG-C. Another event occurred ~ 11 ka ago and was detected in all three trenches studied. The youngest event was detected in TEG-A and occurred ~3 ka ago.

A magnitude of Mw 6.6 was estimated. The calculated magnitude ranges between Mw 6.3 and Mw 6.8 and was estimated as Mw 6.6. The recurrence time of large earthquakes on this segment is ~ 9 ky.


Fig. 2c

Paleoseismic trenches: Two events: ~40ka and ~10ka

Amit et al. (2009)