Umm al Qanatir

Transliterated Name	Language	Name
Umm el-Qanatir	Arabic	ام القناطر
Umm el-Kanatir	Arabic	ام القناطر
Ein Keshatot	Hebrew
Rehavam's Arches
Kantur
Qamtra

Introduction

Fairly extensive archeoseismic evidence is present at Umm el-Qanatir, a site that appears to have been largely abandoned in the middle of the 8th century CE. An old landslide scar is also present at the site. Wechsler et al (2008) used a 2D slope stability model with measured geotechnical properties of three soil units to infer a minimum PGA of 0.36 g to initiate the landslide. They also estimated a minimum moment Magnitude of 7.0 for the earthquake which is presumed to have caused the slide.

Description and Exploration

from Zvi Uri Ma'oz in Stern et. al. (1993 v.2)

Umm el-Qanatir is built on a limestone ledge under the basalt escarpment of the Golan, near a northern tributary ofNahal Samakh (map reference 2195.2506). The Arabic name of the site means "Mother of the Arches," after a nymphaeum with two arches built over the spring to the south of the synagogue. The site was discovered by L. Oliphant in 1884 and surveyed by G. Schumacher in 1885. A limited excavation was conducted in 1905 by H. Kohl and C. Watzinger, who published a very accurate plan of the remains; however, their reconstruction of the building was faulty. In 1932, the site was surveyed by E. L. Sukenik and in 1980 by Z. Ma'oz.

The synagogue (13.8 by 18.8 m) is preserved to a height of several courses of ashlars, particularly in the southern wall. The structure is oriented north-south, with its facade in the south. Outside the entrance is a small portico with a stone floor that is reached by three steps. The porch has two columns, topped by corbel capitals. The main entrance (1.63 m wide) was slightly to the east of the building's central axis, probably in order to leave room for the Ark of the Law inside. The doorposts were diagonally molded. The lintel was not found, but nearby was a stone from a relieving arch, decorated with bands of circular, zigzag, and bud patterns. At the top of the arch was a decoration resembling a three-branched menorah (?). The facade had decorated windows with lintels in the form of Syrian pediments, as well as pilasters with capitals decorated with bands of meander, egg-and-dart, and den til motifs. Found in the debris of the synagogue were two similar stones, each of which bore a three-dimensional depiction of the forepart of a lion (of which only the legs have survived). Kohl and Watzinger suggested that these stones had stood on either side of a window, but it seems more probable that they adorned the base of the Ark of the Law. The unmutilated head of one of these lions was later found at the site. Its face is schematic and naively rendered. The building was crowned with a gable topped by an eagle with outspread wings. A side entrance to the hall, 1.55 m wide, was found in the western wall. There were three rows of columns in the prayer hall: two rows of five columns each, parallel to the long walls, and an additional row at right angles, parallel to the northern wall. The columns supported architraves of the 'En Nashut type, one of them decorated with egg-and-dart motifs and a vine branch. Worthy of mention is a capital consisting of a pair of colonnettes with a relief featuring an eagle with outspread wings and rosettes; it was probably part of the stone-built aedicula of the ark.

The synagogue building probably collapsed in an earthquake; most of its walls and architectural elements were found in the debris among the remains. Kohl and Watzinger, as well as Ma'oz, identified an ancient repair: the gaps between the columns in the western row were sealed to form a wall incorporating various architectural elements. The modification was probably necessary because of the apparent earthquake damage to the building. The synagogue at Umm el-Qanatir is unique among the Golan synagogues for its architectural style and decoration, being so far unparalleled by anything revealed by the Golan survey. Kohl and Watzinger believed it had been built in the fifth century CE, but M. Avi-Yonah, F. Hiittenmeister, and G. Reeg dated it to the third century CE. A comparison of its artistic style with that of other synagogues in the Golan indicates, in this writer's opinion, a date in the second half of the fifth century CE, however.

Exploration and Excavations

from Dray et al. (2017:209-211)

The site was first surveyed by Sir Laurence Oliphant¹ in 1884, when he visited the area together with a local sheikh and the German engineer Gottlieb Schumacher (Oliphant 1887: 267). Oliphant described the spring arches and the ruins of the large structure, which he identified as a synagogue. Schumacher also recorded his impressions (Schumacher 1888: 260–265), although he did not make any such identification.

In 1905, the synagogue was first excavated, albeit only partially, by the Galilean synagogue scholars Heinrich Kohl and Carl Watzinger (1916: 134–135). The plan of the synagogue produced by the excavation, only four days long, served as the basis for research for the subsequent century. Kohl and Watzinger proposed reconstructing the structure with its southern doorway in the centre, although their precise sketch of the building shows a different location. In addition, the northern directional arrow is off by about 60°. Scholars only noticed these errors after 1967 (Urman 1995: 547; Maʿoz 1993: 452–453; Dray, Gonen and Ben David 2007: 62–63). Kohl and Watzinger dated the synagogue to the fifth century CE on the basis of the basket capital uncovered in the debris in front of the southern entrance.

In 1928 Eliezer Lipa Sukenik visited the area (Sukenik 1935: 172–174) and noted that he accepted Kohl and Watzinger’s late dating based on the location of the main entrance in the east, unlike other Galilean synagogues.²

According to a 1967–68 survey, architectural elements from the synagogue were incorporated into houses of the modern-day Bedouin village, including an eagle capital (Epstein and Gutman 1972: 283, site 148). Dan Urman, who surveyed the site in 1968, noted the mistake in the northern directional arrow in Kohl and Watzinger’s plan (Urman 1995: 547, n. 589) and described several architectural elements unmentioned in the previous survey.

In his entry on synagogues in the Encyclopedia of Archaeological Excavations in the Holy Land, Michael Avi-Yonah (1978: 1137–1138) ascribed an earlier date to the synagogue than Kohl and Watzinger, but did not specify the date. Avi-Yonah’s earlier dating stemmed, among other reasons, from the fact that no apse was discovered in the structure. Hüttenmeister and Reeg (1977: 468) followed suit, dating the structure to the third century CE. Foerster (1972: vol. 2, 5, n. 54), however, ascribed the synagogue to the Byzantine period.

Maʿoz (1995: 114) was the first to mention that the main doorway at Umm el-Qanatir was built east of the centre of the southern wall, as at Ein Neshut, proposing that this had been done to reserve the wall’s centre for the Torah shrine³ (Maʿoz 1995: 83–85). Based on artistic style, he dated the Umm el-Qanatir synagogue to the second half of the fifth century CE (1995: 129).

The ceramic survey at the site led Ben David (2005: 172) to posit that the settlement was established in the second or early third century CE and continued to stand throughout the Byzantine period.

In the summer of 2003, excavation and study of the building began, serving as a test case for the feasibility of its full reconstruction.⁴ There have since been six excavation seasons at the site, with the last three focusing on reconstruction.⁵

Footnotes

1 A British author, traveler, diplomat and mystic, who sought to establish a Jewish state in Palestine.

2 Surprisingly, Sukenik, who visited the site in person, relied on Kohl and Watzinger’s erroneous measurements.

3 Maʿoz suggested that this was also the case in the synagogues of Zumimra and Kh. Zavitan.

4 The excavation and reconstruction project is directed and implemented by Y. Dray (Restoration of Ancient Technology Center, Binyamina). Archaeologists: I. Gonen and C. Ben David. Scientific sponsorship: the Institute of Archaeology of Bar-Ilan University and the Kinneret Institute for Galilean Archaeology. Funding: Golan Regional Council and the Israel Government Tourist Corporation. Excavation team: T. Kaʿatabi, N. Aharoni, C. Tal, S. Chamami, M. Golan and H. Kinsbursky. Pottery reading: B. Johnson and M. Hartal. Conservation architect: G. Solar. Mapping: Mabat 3D Technologies. Drawings and surveying: D. Porotsky and R. Mishayev. Photography: Y. Dray and I. Gonen.

5 In November 2010 the site was selected as one of the heritage sites to be fostered by the Landmarks Project, a special programme funded by the Prime Minister’s Office. The site was also given a new Hebrew name ––ʿEin Qeshatot, meaning ‘Arches Spring’.

Maps, Aerial Views, Plans, and Photos

Maps and Aerial Views

Umm al Qanatir in Google Earth

Umm al Qanatir

click on image to explore this site on a new tab in Google Earth
Umm al Qanatir on govmap.gov.il

Umm al Qanatir

click on image to explore this site on a new tab in govmap.gov.il
Aerial View of Umm al-Qanatir

Umm el-Qanatir synagogue from the air (2013)

click on image to open a high res magnifiable image in a new tab

Avram Gracier - Wikipedia - CC BY-SA 3.0

synagogue from Wikipedia

Plans

Normal Size

Fig. 4 Plan of the synagogue

Fig. 4

Schematic plan; last phase of the building shown in red

Dray et al. (2017)

at Umm al-Qanatir from Dray et al. (2017)
Fig. 1 Images of the three

Fig. 1

Computer imaging of 3D mapping: three upper layers of the collapsed structure of the synagogue

Dray et al. (2017)

upper layers of the collapsed structure of the synagogue at Umm al-Qanatir from Dray et al. (2017)

Magnified

Fig. 4 Plan of the synagogue

Fig. 4

Schematic plan; last phase of the building shown in red

Dray et al. (2017)

at Umm al-Qanatir from Dray et al. (2017)
Fig. 1 Images of the three

Fig. 1

Computer imaging of 3D mapping: three upper layers of the collapsed structure of the synagogue

Dray et al. (2017)

upper layers of the collapsed structure of the synagogue at Umm al-Qanatir from Dray et al. (2017)

Landslide Photos

Fig. 2a Overhead Aerial photo of

Fig. 2a

Aerial photo of study area.

Wechsler et al (2008)

landslide from Wechsler et al (2008)
Fig. 2a Ground-based Photo of

Fig. 2b

A northward look at the Umm-El-Qanatir site.

Wechsler et al (2008)

landslide looking north from Wechsler et al (2008)

Archaeological Photos

Fig. 2 Collapsed elements

Fig. 2

Collapsed elements on the synagogue floor and the column bases in situ (view to the north)

Dray et al. (2017)

on the synagogue floor and the column bases in situ from Dray et al. (2017)
Fig. 3 Collapsed eastern

Fig. 3

The collapsed eastern wall (view to the south)

Dray et al. (2017)

wall from Dray et al. (2017)
Fig. 9 Secondary construction

Fig. 9

Secondary construction of rooms in the northern part of the synagogue (view to the south)

Dray et al. (2017)

of rooms in the northern part of the synagogue from Dray et al. (2017)
Fig. 5 Secondary construction

Fig. 5

The synagogue suffered two damaging events. The firs led to its abandonment, as evidenced by a makeshift house (marked with dashed line), which was built on the floor of the synagogue using many of its finely carved stones. The second event caused the destruction of this house and triggered the abandonment of the whole site, as evidenced by absence of post mid-8th century artifacts

Wechsler et al (2008)

of rooms in the northern part of the synagogue from Wechsler et al (2008)
Fig. 4a Horizontal shift of

Fig. 4a

Horizontal shift of a large masonry basalt blocks at the synagogue wall

Wechsler et al (2008)

large masonry basalt blocks of the synagogue wall from Wechsler et al (2008)
Fig. 4b Horizontal shift of

Fig. 4b

Horizontal shift of a large masonry basalt blocks at the synagogue wall

Wechsler et al (2008)

large masonry basalt blocks of the synagogue wall from Wechsler et al (2008)
Fig. 4c Slipped arch voussoir

Fig. 4c

Keystones of an arch that slid out.

Wechsler et al (2008)

(possibly a keystone) from Wechsler et al (2008)
Fig. 4d Aligned fallen columns

Fig. 4d

Aligned fallen columns

Wechsler et al (2008)

from Wechsler et al (2008)
Fig. 4e Westward collapse of

Fig. 4e

Westward collapse of the wall of the synagogue

Wechsler et al (2008)

a wall of the synagogue from Wechsler et al (2008)
Fig. 4f Sheared water pool

Fig. 4f

A water pool, part of the spring complex, displaced left- laterally 95 cm. This water pool is displaced by the landslide and is used for the slope stability analysis

Wechsler et al (2008)

of the Spring complex from Wechsler et al (2008)
Plate XIXa Arch built over

Plate XIXa

Umm el Qanatir: Arch built over spring

Sukenik (1935)

spring from Sukenik (1935)
Plate XIXb Ruins of Synagogue

Plate XIXb

Umm el Qanatir: View of ruins of Synagogue

Sukenik (1935)

from Sukenik (1935)

1st Earthquake - mid 8th century CE ?

Figures/Photos

Fig. 4f Sheared water pool

Fig. 4f

A water pool, part of the spring complex, displaced left- laterally 95 cm. This water pool is displaced by the landslide and is used for the slope stability analysis

Wechsler et al (2008)

of the Spring complex from Wechsler et al (2008)

Discussion

Wechsler et al (2008) report that artifacts from Umm el-Qanatir all date between the Byzantine and Early Arab periods and that no Abbasid period or later artifacts have been found in the ruins. This, according to the authors, suggests abandonment of the site in the mid 8th century CE. Wechsler et al (2008) suggest that abandonment was precipitated by one of the Sabbatical Year Quakes which damaged the town's spring and may have compromised it's main industry of fine linen production. Pottery shards unearthed from an abandoned and damaged pool associated with the spring complex all date up to the mid 8th century. The damaged pool, which was apparently sheared during the landslide (see Fig. 4f), shows no signs of repair. An 8th century CE terminus post quem for damage appears to be well established but, due to a lack of post Umayyad remains, a terminus ante quem is not well established. One of the Sabbatical Year Quakes and/or later earthquakes could be responsible for the damage observed at the site.

2nd Earthquake (?) - after the mid 8th century CE

Figures/Photos

Fig. 9 Secondary construction

Fig. 9

Secondary construction of rooms in the northern part of the synagogue (view to the south)

Dray et al. (2017)

of rooms in the northern part of the synagogue from Dray et al. (2017)
Fig. 5 Secondary construction

Fig. 5

The synagogue suffered two damaging events. The firs led to its abandonment, as evidenced by a makeshift house (marked with dashed line), which was built on the floor of the synagogue using many of its finely carved stones. The second event caused the destruction of this house and triggered the abandonment of the whole site, as evidenced by absence of post mid-8th century artifacts

Wechsler et al (2008)

of rooms in the northern part of the synagogue from Wechsler et al (2008)
Fig. 4 Plan of the synagogue

Fig. 4

Schematic plan; last phase of the building shown in red

Dray et al. (2017)

at Umm al-Qanatir from Dray et al. (2017)
Fig. 1 Images of the three

Fig. 1

Computer imaging of 3D mapping: three upper layers of the collapsed structure of the synagogue

Dray et al. (2017)

upper layers of the collapsed structure of the synagogue at Umm al-Qanatir from Dray et al. (2017)

Discussion

Wechsler et al (2008) report that a makeshift house (see Fig. 9) found inside the seismically damaged synagogue was presumed to represent post-abandonment occupation of the site. Agricultural tools such as spades, hoes, and sickles were found below a collapse layer which suggest a second unexpected destruction event. Dray et al. (2017:221) reports that the fill of room L1240 revealed an assemblage of complete pottery vessels, including an intact juglet and a cooking-pot lid on top of a jug inside a cooking bowl, within which another intact juglet was found. Since these vessels are described as being in the fill of the secondary structure (aka makeshift house), they would have presumably been buried during the first earthquake when the site was possibly abandoned. Those vessels dated to the first half of the eighth century CE. Since none of the agricultural implements found in the collapsed house were dated, the 2nd earthquake can only be specified as striking after the mid 8th century CE.

References

Dray et al. (2017)

from Dray et al. (2017:219-220)

Figures/Photos

Fig. 9 Secondary construction

Fig. 9

Secondary construction of rooms in the northern part of the synagogue (view to the south)

Dray et al. (2017)

of rooms in the northern part of the synagogue from Dray et al. (2017)
Fig. 5 Secondary construction

Fig. 5

The synagogue suffered two damaging events. The firs led to its abandonment, as evidenced by a makeshift house (marked with dashed line), which was built on the floor of the synagogue using many of its finely carved stones. The second event caused the destruction of this house and triggered the abandonment of the whole site, as evidenced by absence of post mid-8th century artifacts

Wechsler et al (2008)

of rooms in the northern part of the synagogue from Wechsler et al (2008)
Fig. 4 Plan of the synagogue

Fig. 4

Schematic plan; last phase of the building shown in red

Dray et al. (2017)

at Umm al-Qanatir from Dray et al. (2017)
Fig. 1 Images of the three

Fig. 1

Computer imaging of 3D mapping: three upper layers of the collapsed structure of the synagogue

Dray et al. (2017)

upper layers of the collapsed structure of the synagogue at Umm al-Qanatir from Dray et al. (2017)

Discussion

The Last Phase of the Building

The above-mentioned remains of secondary construction found inside the synagogue abutted the entire length of its northern wall (W1036), utilizing architectural elements from the upper part of the building, such as bases, capitals, engaged columns and cornices, and creating four rooms (fig. 9). The rest of the interior was found devoid of fragments of benches or roof tiles. Immediately to the south of these rooms, along W1071, three capitals and bases were found in a row. Along the western wall (W1062), immediately to the north of the western doorway, three adjacent stretchers were uncovered. Along the benches in the east (L1080), at the point of ascent to the eastern doorway from the earlier phase, two doorpost stones and a threshold stone were found next to one another.

The four rooms (fig. 4) extended over an area of 68 m² (all rectangular, L1240: 2.40 × 5.50 m; L1241: 2.50 × 6.60 m; L1242: 2.00 × 5.80 m; and L1243: 2.50 × 5.20 m). One interior wall (W1065) is fenestrated, found in a ‘domino collapse’ arrangement. Significantly, when W1063 was dismantled, two architectural elements in secondary use were found to have been used in its construction. The faces of these elements, which bore reliefs, had been set toward the interior and were thus not visible. One was the base of an engaged column decorated with a double meander and an animal; the other was the capital of an engaged column bearing a seven-branched menorah, shofars, a bird holding a ring in its beak and a double meander.

Under the collapse, in the fill of the rooms, which consisted mainly of light-coloured clayish soil, a variety of items were found, including basalt rollers, mortars and a quern (jarusha), as well as an assemblage of iron agricultural tools such as pick-axes, double-axes, sickles, a knife, a pruning hook, a mattock and the components of a complete disk-thresher. The fill of room L1240 revealed an assemblage of complete pottery vessels, including an intact juglet and a cooking-pot lid on top of a jug inside a cooking bowl, within which another intact juglet was found. These vessels are dated to the first half of the eighth century CE (fig. 10).

Outside the synagogue’s western doorway an improvised staircase (L1118) was found, consisting of architectural elements in secondary use, such as a rounded cornice stone. Alongside the steps to the north, three column shafts were laid side by side, while alongside the steps on the south, two beams were set one on top of the other. The upper beam had a broken end, and the lower one bears a geometric and vegetal relief at the narrow side.

The eastern wall of the building from this phase was built without doorways. Four additional courses were laid above the original ‘header course’ –– which was set obliquely, low in the south and rising toward the north, 0.27–0.53 m above the benches. The stones in these courses were of various sizes in secondary use. They were carelessly laid, with their carved face even set toward the interior. The wall from this phase was preserved to a height of 2.00 m above the floor.

Meagre remains found outside the synagogue made use of architectural elements from the upper parts of the building, including cornices and capitals: about 5 m east of the synagogue, the foundations of two rooms can be seen above the collapsed eastern wall, while next to the southeastern corner of the synagogue and atop its southern portico, remains were found of additional walls, too poorly preserved for their nature to be discerned.

1st Earthquake - mid 8th century CE ?

Effect	Location	Image(s)	Description
Displaced masonry blocks	entrance pier and southern wall of the synagogue Fig. 4 Schematic plan; last phase of the building shown in red Dray et al. (2017)	Fig. 4a Horizontal shift of a large masonry basalt blocks at the synagogue wall Wechsler et al (2008) Fig. 4b Horizontal shift of a large masonry basalt blocks at the synagogue wall Wechsler et al (2008)	Other typical earthquake-triggered damage is the horizontal shift of the entrance pier and of large basalt blocks (~1m³) in the southern wall of the synagogue (Fig. 4a, b) - Wechsler et al (2008)
Collapsed Walls	western wall (W1062) of the synagogue Fig. 4 Schematic plan; last phase of the building shown in red Dray et al. (2017)	Fig. 4e Westward collapse of the wall of the synagogue Wechsler et al (2008)	The western wall and most of the columns of the synagogue, collapsed westward (Fig. 4e). This direction is similar to the directional of fall of the columns in three basilicas at Sussita, an archaeological site about 10 km to the southwest of Umm-El-Qanatir, which was affected by one of the mid-8th century earthquakes (Segal et al. 2003) - Wechsler et al (2008) The way the [eastern and western] walls fell shows that they were pushed from the top, that is, at the moment of collapse, the roof, attached to the tops of the eastern and western walls, pushed these walls outward. - Dray et al. (2017:228) The synagogue was determined to have been two storeys high on the basis of the column parts uncovered, which had two different diameters, the thickness of the walls and the articulated collapse of the eastern and western walls. These walls were found with their courses having fallen in rows and their ashlars facing downward in the fill. Fourteen courses of the outer face of the western wall (W1062) were uncovered, six of them in situ, while the rest of the wall, consisting of eight collapsed courses, was found to extend over 4 m to the west of the wall. That was the case for the northern part of the western wall. The southern part extended even farther, but the debris was disturbed by the later phase of the structure - Dray et al. (2017:221, 223)
Collapsed Walls	eastern wall (W1024) of the synagogue Fig. 4 Schematic plan; last phase of the building shown in red Dray et al. (2017)	Fig. 3 The collapsed eastern wall (view to the south) Dray et al. (2017)	Excavation around the synagogue revealed the articulated collapse of its walls (Fig. 3) - Dray et al. (2017:213-214) The synagogue was determined to have been two storeys high on the basis of the column parts uncovered, which had two different diameters, the thickness of the walls and the articulated collapse of the eastern and western walls. These walls were found with their courses having fallen in rows and their ashlars facing downward in the fill. - Dray et al. (2017:221) Sixteen courses of the original phase of the eastern wall (W1024) were found, five of them in situ and the other eleven collapsed courses extending up to 4.50 m east of the wall. The height of the wall may be estimated based on how far its collapse extended, subtracting the spaces between the stones that resulted from inertia. - Dray et al. (2017:223) The way the [easetern and western] walls fell shows that they were pushed from the top, that is, at the moment of collapse, the roof, attached to the tops of the eastern and western walls, pushed these walls outward. - Dray et al. (2017:228)
Fallen columns	synagogue interior Fig. 4 Schematic plan; last phase of the building shown in red Dray et al. (2017)	Fig. 4d Aligned fallen columns Wechsler et al (2008) Fig. 2 Collapsed elements on the synagogue floor and the column bases in situ (view to the north) Dray et al. (2017)	The western wall and most of the columns of the synagogue, collapsed westward (Fig. 4e). This direction is similar to the directional of fall of the columns in three basilicas at Sussita, an archaeological site about 10 km to the southwest of Umm-El-Qanatir, which was affected by one of the mid-8th century earthquakes (Segal et al. 2003) - Wechsler et al (2008) JW: The synagogues and columns also collapsed in a downhill direction
Sheared Water Pool	Pool in Spring Complex Fig. 2a Aerial photo of study area. Wechsler et al (2008)	Fig. 4f A water pool, part of the spring complex, displaced left- laterally 95 cm. This water pool is displaced by the landslide and is used for the slope stability analysis Wechsler et al (2008)	Typical earthquake damage to the spring area is apparent in the form of a tilted and broken pavement and a displacement of the water pool by a total of ~1 m to left-laterally (Fig. 4f). - Wechsler et al (2008) we argue that the cause of damage in the spring area is most probably an earthquake-induced landslide. This conclusion is based on the typical landslide morphology of the spring area—the slope is quite steep (up to 40") and its crescent shape is compatible with a slide scarp (Fig. 2a, b); on the weak geological material the site is founded on, which enables slope failure; and on the existing water at the slope surface (the spring) that also promotes slope failure. We did not find any fault scarps in the area of the Umm-El-Qanatir site, the underlying strata are not faulted, and no active faults have been mapped in the site in previous surveys (Bartov et al. 2002). This further supports the landslide origin for the displacement of the pool. - Wechsler et al (2008)
Landslide	downslope from the synagogue	Fig. 2a Aerial photo of study area. Wechsler et al (2008) Fig. 2b A northward look at the Umm-El-Qanatir site. Wechsler et al (2008)	we argue that the cause of damage in the spring area is most probably an earthquake-induced landslide. This conclusion is based on the typical landslide morphology of the spring area—the slope is quite steep (up to 40") and its crescent shape is compatible with a slide scarp (Fig. 2a, b); on the weak geological material the site is founded on, which enables slope failure; and on the existing water at the slope surface (the spring) that also promotes slope failure. We did not find any fault scarps in the area of the Umm-El-Qanatir site, the underlying strata are not faulted, and no active faults have been mapped in the site in previous surveys (Bartov et al. 2002). This further supports the landslide origin for the displacement of the pool. - Wechsler et al (2008)
Re-used building elements	secondary structure (aka makeshift house) inside the synagogue along with other areas in the synagogue and outside the synagogue Fig. 4 Schematic plan; last phase of the building shown in red Dray et al. (2017) Fig. 9 Secondary construction of rooms in the northern part of the synagogue (view to the south) Dray et al. (2017) Fig. 5 The synagogue suffered two damaging events. The firs led to its abandonment, as evidenced by a makeshift house (marked with dashed line), which was built on the floor of the synagogue using many of its finely carved stones. The second event caused the destruction of this house and triggered the abandonment of the whole site, as evidenced by absence of post mid-8th century artifacts Wechsler et al (2008)		The above-mentioned remains of secondary construction found inside the synagogue abutted the entire length of its northern wall (W1036), utilizing architectural elements from the upper part of the building, such as bases, capitals, engaged columns and cornices - Dray et al. (2017:219) when W1063 was dismantled, two architectural elements in secondary use were found to have been used in its construction. The faces of these elements, which bore reliefs, had been set toward the interior and were thus not visible. One was the base of an engaged column decorated with a double meander and an animal; the other was the capital of an engaged column bearing a seven-branched menorah, shofars, a bird holding a ring in its beak and a double meander. - Dray et al. (2017:219,221) Outside the synagogue’s western doorway an improvised staircase (L1118) was found, consisting of architectural elements in secondary use, such as a rounded cornice stone. Alongside the steps to the north, three column shafts were laid side by side, while alongside the steps on the south, two beams were set one on top of the other. The upper beam had a broken end, and the lower one bears a geometric and vegetal relief at the narrow side. The eastern wall of the building from this phase was built without doorways. Four additional courses were laid above the original ‘header course’ –– which was set obliquely, low in the south and rising toward the north, 0.27–0.53 m above the benches. The stones in these courses were of various sizes in secondary use. They were carelessly laid, with their carved face even set toward the interior. The wall from this phase was preserved to a height of 2.00 m above the floor. Meagre remains found outside the synagogue made use of architectural elements from the upper parts of the building, including cornices and capitals: about 5 m east of the synagogue, the foundations of two rooms can be seen above the collapsed eastern wall, while next to the southeastern corner of the synagogue and atop its southern portico, remains were found of additional walls, too poorly preserved for their nature to be discerned. - Dray et al. (2017:221)

2nd Earthquake (?) - after the mid 8th century CE

Effect	Location	Image(s)	Description
Collapsed Walls	makeshift house inside synagogue Fig. 4 Schematic plan; last phase of the building shown in red Dray et al. (2017) Fig. 5 The synagogue suffered two damaging events. The firs led to its abandonment, as evidenced by a makeshift house (marked with dashed line), which was built on the floor of the synagogue using many of its finely carved stones. The second event caused the destruction of this house and triggered the abandonment of the whole site, as evidenced by absence of post mid-8th century artifacts Wechsler et al (2008) Fig. 9 Secondary construction of rooms in the northern part of the synagogue (view to the south) Dray et al. (2017)		A small 4 x 6 m makeshift house was found inside the synagogue (Fig. 5). It was built on the synagogue’s tiled floor with many of the synagogue’s recycled fancy stones, after the synagogue was damaged. Agricultural tools such as spades, hoes, and sickles, which were found below the rubble, indicate that this secondary building belonged to a farmer who took advantage of the ruined structure. It is highly unlikely that someone would build such a house while the synagogue was active. The farmer’s house itself had collapsed and buried these useful artifacts, indicating a sudden, unexpected (second) disaster. - Wechsler et al (2008) The rooms [of the makeshift structure], which were one storey high, were apparently roofed with wooden planks and mud. When the roof collapsed, it caused the two stone rollers to fall into the rooms and the mud preserved the iron agricultural tools. It also sealed and ended use of the rooms, which contained pottery vessels typical of the first half of the eighth century CE. This dating conforms to the event following which the building was abandoned–– the earthquake of 749 CE. - Dray et al. (2017:228-229)
Collapsed Wall	Wall W1065 in the secondary structure Fig. 4 Schematic plan; last phase of the building shown in red Dray et al. (2017) Fig. 9 Secondary construction of rooms in the northern part of the synagogue (view to the south) Dray et al. (2017) Fig. 5 The synagogue suffered two damaging events. The firs led to its abandonment, as evidenced by a makeshift house (marked with dashed line), which was built on the floor of the synagogue using many of its finely carved stones. The second event caused the destruction of this house and triggered the abandonment of the whole site, as evidenced by absence of post mid-8th century artifacts Wechsler et al (2008)		One interior wall (W1065) is fenestrated, found in a ‘domino collapse’ arrangement. - Dray et al. (2017:219)

1st Earthquake - mid 8th century CE ?

Modified by JW from Fig. 4 of Dray et al. (2017)

Deformation Map

lightly modified by JW from Fig. 4 of Dray et al. (2017)

1st Earthquake - mid 8th century CE ?

Modified by JW from Google Earth capture

Landslide Map

Modified by JW from Google Earth capture

1st Earthquake - mid 8th century CE ?

2 D Model

2D Slope Stability Model of Wechsler et al (2008)

Fig. 2c - Cross section showing geological units and sample numbers. Water table is assumed. This section is used to model slope stability ...

Wechsler et al (2008)

Soil Properties

Cohesion (C) and the Angle of Internal friction (Φ) was determined for all three soil units (KA-2, KA-4, and KA-5) via a direct shear test performed under an undrained consolidated setting. KA-4, a red clay paleosoil which is believed to have been the layer responsible for the slope failure during the landslide yielded values of C = 4.0 kN/m and Φ = 31º. The other soils had similar friction angles and roughly five times higher higher cohesion (C).

Static Analysis

A Factor of Safety (FS) for the slope was estimated first to determine if the slope was stable under aseismic conditions. Two methods were used - the Fellenius method and the modified Bishop method. Both methods divide the soil model into vertical slices and determine Factor of Safety as a ratio from a sum of moment balances performed on each vertical slice. This accounts for changes in topography and lateral changes in unit thicknesses and elevation of the water table. The generalized equation is shown below.

FS = Σ M_r / Σ M_d

where

M_r = Resisting Moment
M_d = Driving Moment

Resisting Moment is the ability of the soil mass to avoid moving. The driving moment is the forces such as gravity and water saturated layers pushing the soil mass down slope. FS less than 1 indicates that the slope is unstable. FS greater than 1 indicates that the slope is stable. Wechsler et al (2008) evaluated FS for a variety of conditions using both the Fellenius and modified Bishop method and came up with a range of FS values from 2.05 to 3.23 - indicating that the slope is aseismically stable. The lowest value of FS represents a condition where the water table comes right up to the surface. The highest value of FS corresponds to a condition where the water table was at the same elevation as the impermeable soil unit KA-4. Wechsler et al (2008) noted that the existence of the year-round flowing spring (and several others in the vicinity) indicates a high water table, and the weather patterns (rainy winter with ~900 mm average annual rainfall) suggest that at times it can be as high as the topography. Thus, lower values of FS may represent realistic conditions for the slope. The commercial software Slope/W™ was used to perform the Static Analysis. Modeling indicated that slope stability was largely determined by the weakest layer (lowest Cohesion) - paleosoil KA-4.

Dynamic Analysis

The Newmark Displacement method was used to evaluate slope stability during an earthquake. First a critical acceleration was estimated. Critical acceleration is the minimum earthquake induced acceleration that initiates slope movement. Critical acceleration was computed as follows:

a_c = (FS-1)sinβ

where

a_c = critical acceleration measured in g's (1 g = 9.81 m/s²)
FS = Factor of Safety (computed previously for a range of conditions)
β = Thrust angle (degrees)

Thrust Angle

Newmark (1965)'s slope stability model

Fig. 15 a

O = Circle center
c.g. = Center of Gravity
β = Thrust angle

Newmark (1965)

showing thrust angle (β)

For a planar slide, thrust angle (β) is the direction the COG of the sliding block moves when displacement initially occurs. Ideally this is the dip of the bedding plane. In regional studies this is typically approximated by the slope angle (Katz and Crouvi, 2007 citing Miles and Keefer, 2001). For rotational movement on a circular surface, Newmark (1965) showed that the thrust angle is the angle between the vertical and a line segment connecting the center of gravity of the landslide mass and the center of the slip circle ( Jibson, 1996:310).

Wechsler et al (2008) used a value of β = 20º. A lack of recorded strong motion records from the area prompted Wechsler et al (2008) to estimate the Newmark displacement D_N using an empirically derived formula from data from Southern California. The empirical relationship comes from Jibson et al (2000: p. 8 eqn. 3) which estimates the critical displacement D_N at which the slope begins to fail.

lnD_N = 1.521*ln(I_α) - 1.993ln(a_c) - 1.546

where

D_N = Newmark Displacement (cm.)
I_α = Arias Intensity (m/s)
a_c = Critical acceleration (g's)

Arias Intensity (I_α) was related to Magnitude and Fault Distance using an equation derived for the Dead Sea by Katz and Crouvi (2007):

log₁₀I_α = 1.2M_W - 2.2log₁₀R - 4.9

where

I_α = Arias Intensity
M_W = Moment Magnitude
R = Fault Distance (km.)

By assuming a D_N of 5 or 10 cm., they were able to solve for I_α and then M_W as a function of R. Based on their plots of M_W as a function of R, they chose an R representative of the distance to the nearest active fault capable of producing a strong earthquake. The fault they selected was the The Jordan Gorge Fault ~15 km. away from Umm al-Qanatir. This led to an estimated minimum moment magnitude of ~7.0. They concluded that the landslide at Umm al-Qanatir was caused by an earthquake with a moment magnitude of at least 7.0. These calculations are reproduced in the calculator below:

Calculator

Variable	Input	Units	Notes
FS		unitless	Wechsler et al (2008) constrained FS from 2.05 to 3.23
β		degrees	Wechsler et al (2008) used a value of 20
D_N		cm.	Wechsler et al (2008) recommends a value of 5 or 10
R		km.	Distance to nearest earthquake producing fault
Variable	Output (No Site Effect)	Units	Notes
a_c		g	minimum acceleration to induce slide
I		unitless	Conversion from a_c to I using Wald et al (1999)
M		unitless	Attenuation relationship of Hough and Avni (2009) used to calculate Magnitude from I and R
I_α		m/s	Calculated from eqn. 2 of Wechsler et al (2008)
I_α		m/s	Calculated from eqn. 3.17 of (Kramer, 1996:87)
M_w		unitless	calculated from eqn. 3 of Wechsler et al (2008) which comes from Katz and Crouvi (2007)
Variable	Input	Units	Notes
Seismic Amplification		unitless	Site Effect due to Topographic or Ridge Effect (set to 1 to assume no site effect)
Variable	Output (Site Effect)	Units	Notes
I		unitless	Intensity with Topographic Effect removed
M		unitless	Magnitude with Topographic Effect removed - using Hough and Avni (2009)
M_W		unitless	Moment Magnitude with Topographic Effect removed M_W from eqn. 3 of Wechsler et al (2008) which comes from Katz and Crouvi (2007)

caveat

Site Effect Removal Methodology

Figure 13a from Massa et al (2010)

Figure 13. (a) Residuals between predicted acceleration response spectra (5% damping) for reference station (NRN2) and for station NRN7, for periods up to 1 s. The results are reported for the west—east component and are presented in terms of logarithmic differences. PGHA indicates peak ground horizontal component (in this case, the east—west component).

Massa et al (2010)

Output with site effect removed assumes that PGA is higher than it would be if there was no site effect. In this situation, Intensity (I) with site effect removed is calculated pre-amplification (i.e. it will be lower). This is because an Intensity estimate with the site effect removed is helpful in producing an Intensity Map that will do a better job of "triangulating" the epicentral area.

Site Effect is based on Equation 2 and Figure 13 a of

Massa et al (2010). In their study, they estimated a frequency dependent additional PGA (St in Eqn. 2) which is added by a topographic site effect. The additional topographic site effect PGA varied from ~0.1 g to 0.5 g for dominant frequencies of approximately 1 - 5 Hz.. Higher PGA's were shown to be present for higher frequencies which are more likely to occur when the earthquake producing fault is closer to the site. They also noted that a greater topographic effect was observed when the seismic energy arrived orthogonal (perpendicular in their words) to the ridge. Both of these considerations suggest that a topographic ridge effect should be considered when other evidence suggests that a nearby fault broke during the earthquake. The additional Site Effect PGA is linearly scaled from 0 - 0.5 g for site effects where amplitude increases from 1x to 10x. It's not the greatest transform to remove site effect from the Intensity estimate but may be useful for crude estimates.

1st Earthquake - mid 8th century CE ?

Earthquake Archeological Effects chart

Earthquake Archeological Effects (EAE)

Rodríguez-Pascua et al (2013: 221-224)

of Rodríguez-Pascua et al (2013: 221-224)

Effect	Location	Image(s)	Description	Intensity
Displaced masonry blocks	entrance pier and southern wall of the synagogue Fig. 4 Schematic plan; last phase of the building shown in red Dray et al. (2017)	Fig. 4a Horizontal shift of a large masonry basalt blocks at the synagogue wall Wechsler et al (2008) Fig. 4b Horizontal shift of a large masonry basalt blocks at the synagogue wall Wechsler et al (2008)	Other typical earthquake-triggered damage is the horizontal shift of the entrance pier and of large basalt blocks (~1m³) in the southern wall of the synagogue (Fig. 4a, b) - Wechsler et al (2008)	VIII +
Collapsed Walls	western wall (W1062) of the synagogue Fig. 4 Schematic plan; last phase of the building shown in red Dray et al. (2017)	Fig. 4e Westward collapse of the wall of the synagogue Wechsler et al (2008)	The western wall and most of the columns of the synagogue, collapsed westward (Fig. 4e). This direction is similar to the directional of fall of the columns in three basilicas at Sussita, an archaeological site about 10 km to the southwest of Umm-El-Qanatir, which was affected by one of the mid-8th century earthquakes (Segal et al. 2003) - Wechsler et al (2008) The way the [easetern and western] walls fell shows that they were pushed from the top, that is, at the moment of collapse, the roof, attached to the tops of the eastern and western walls, pushed these walls outward. - Dray et al. (2017:228) The synagogue was determined to have been two storeys high on the basis of the column parts uncovered, which had two different diameters, the thickness of the walls and the articulated collapse of the eastern and western walls. These walls were found with their courses having fallen in rows and their ashlars facing downward in the fill. Fourteen courses of the outer face of the western wall (W1062) were uncovered, six of them in situ, while the rest of the wall, consisting of eight collapsed courses, was found to extend over 4 m to the west of the wall. That was the case for the northern part of the western wall. The southern part extended even farther, but the debris was disturbed by the later phase of the structure - Dray et al. (2017:221, 223)	VIII +
Collapsed Walls	eastern wall (W1024) of the synagogue Fig. 4 Schematic plan; last phase of the building shown in red Dray et al. (2017)	Fig. 3 The collapsed eastern wall (view to the south) Dray et al. (2017)	Excavation around the synagogue revealed the articulated collapse of its walls (Fig. 3) - Dray et al. (2017:213-214) The synagogue was determined to have been two storeys high on the basis of the column parts uncovered, which had two different diameters, the thickness of the walls and the articulated collapse of the eastern and western walls. These walls were found with their courses having fallen in rows and their ashlars facing downward in the fill. - Dray et al. (2017:221) Sixteen courses of the original phase of the eastern wall (W1024) were found, five of them in situ and the other eleven collapsed courses extending up to 4.50 m east of the wall. The height of the wall may be estimated based on how far its collapse extended, subtracting the spaces between the stones that resulted from inertia. - Dray et al. (2017:223) The way the [easetern and western] walls fell shows that they were pushed from the top, that is, at the moment of collapse, the roof, attached to the tops of the eastern and western walls, pushed these walls outward. - Dray et al. (2017:228)	VIII +
Fallen and oriented columns	synagogue interior Fig. 4 Schematic plan; last phase of the building shown in red Dray et al. (2017)	Fig. 4d Aligned fallen columns Wechsler et al (2008) Fig. 2 Collapsed elements on the synagogue floor and the column bases in situ (view to the north) Dray et al. (2017)	The western wall and most of the columns of the synagogue, collapsed westward (Fig. 4e). This direction is similar to the directional of fall of the columns in three basilicas at Sussita, an archaeological site about 10 km to the southwest of Umm-El-Qanatir, which was affected by one of the mid-8th century earthquakes (Segal et al. 2003) - Wechsler et al (2008) JW: The synagogues and columns also collapsed in a downhill direction	V +
Sheared Water Pool	Pool in Spring Complex Fig. 2a Aerial photo of study area. Wechsler et al (2008)	Fig. 4f A water pool, part of the spring complex, displaced left- laterally 95 cm. This water pool is displaced by the landslide and is used for the slope stability analysis Wechsler et al (2008)	Typical earthquake damage to the spring area is apparent in the form of a tilted and broken pavement and a displacement of the water pool by a total of ~1 m to left-laterally (Fig. 4f). - Wechsler et al (2008) we argue that the cause of damage in the spring area is most probably an earthquake-induced landslide. This conclusion is based on the typical landslide morphology of the spring area—the slope is quite steep (up to 40") and its crescent shape is compatible with a slide scarp (Fig. 2a, b); on the weak geological material the site is founded on, which enables slope failure; and on the existing water at the slope surface (the spring) that also promotes slope failure. We did not find any fault scarps in the area of the Umm-El-Qanatir site, the underlying strata are not faulted, and no active faults have been mapped in the site in previous surveys (Bartov et al. 2002). This further supports the landslide origin for the displacement of the pool. - Wechsler et al (2008)	IV +
Landslide	downslope from the synagogue	Fig. 2a Aerial photo of study area. Wechsler et al (2008) Fig. 2b A northward look at the Umm-El-Qanatir site. Wechsler et al (2008)	we argue that the cause of damage in the spring area is most probably an earthquake-induced landslide. This conclusion is based on the typical landslide morphology of the spring area—the slope is quite steep (up to 40") and its crescent shape is compatible with a slide scarp (Fig. 2a, b); on the weak geological material the site is founded on, which enables slope failure; and on the existing water at the slope surface (the spring) that also promotes slope failure. We did not find any fault scarps in the area of the Umm-El-Qanatir site, the underlying strata are not faulted, and no active faults have been mapped in the site in previous surveys (Bartov et al. 2002). This further supports the landslide origin for the displacement of the pool. - Wechsler et al (2008)	IV +

The archeoseismic evidence requires a minimum Intensity of VIII (8) when using the Earthquake Archeological Effects chart of Rodríguez-Pascua et al (2013: 221-224).

2nd Earthquake (?) - after the mid 8th century CE

Earthquake Archeological Effects chart

Earthquake Archeological Effects (EAE)

Rodríguez-Pascua et al (2013: 221-224)

of Rodríguez-Pascua et al (2013: 221-224)

Effect	Location	Image(s)	Description	Intensity
Collapsed Walls	makeshift house inside synagogue Fig. 4 Schematic plan; last phase of the building shown in red Dray et al. (2017) Fig. 5 The synagogue suffered two damaging events. The firs led to its abandonment, as evidenced by a makeshift house (marked with dashed line), which was built on the floor of the synagogue using many of its finely carved stones. The second event caused the destruction of this house and triggered the abandonment of the whole site, as evidenced by absence of post mid-8th century artifacts Wechsler et al (2008) Fig. 9 Secondary construction of rooms in the northern part of the synagogue (view to the south) Dray et al. (2017)		A small 4 x 6 m makeshift house was found inside the synagogue (Fig. 5). It was built on the synagogue’s tiled floor with many of the synagogue’s recycled fancy stones, after the synagogue was damaged. Agricultural tools such as spades, hoes, and sickles, which were found below the rubble, indicate that this secondary building belonged to a farmer who took advantage of the ruined structure. It is highly unlikely that someone would build such a house while the synagogue was active. The farmer’s house itself had collapsed and buried these useful artifacts, indicating a sudden, unexpected (second) disaster. - Wechsler et al (2008) The rooms [of the makeshift structure], which were one storey high, were apparently roofed with wooden planks and mud. When the roof collapsed, it caused the two stone rollers to fall into the rooms and the mud preserved the iron agricultural tools. It also sealed and ended use of the rooms, which contained pottery vessels typical of the first half of the eighth century CE. This dating conforms to the event following which the building was abandoned–– the earthquake of 749 CE. - Dray et al. (2017:228-229)	VIII +
Collapsed Wall	Wall W1065 in the secondary structure Fig. 4 Schematic plan; last phase of the building shown in red Dray et al. (2017) Fig. 9 Secondary construction of rooms in the northern part of the synagogue (view to the south) Dray et al. (2017) Fig. 5 The synagogue suffered two damaging events. The firs led to its abandonment, as evidenced by a makeshift house (marked with dashed line), which was built on the floor of the synagogue using many of its finely carved stones. The second event caused the destruction of this house and triggered the abandonment of the whole site, as evidenced by absence of post mid-8th century artifacts Wechsler et al (2008)		One interior wall (W1065) is fenestrated, found in a ‘domino collapse’ arrangement. - Dray et al. (2017:219)	VIII +

The archeoseismic evidence requires a minimum Intensity of VIII (8) when using the Earthquake Archeological Effects chart of Rodríguez-Pascua et al (2013: 221-224).

Site Effect

Wechsler et al (2008) noted that the Umm al-Qanatir site could be subject to an effect known as the topographic effect or ridge effect. This is an effect where a resonance condition can develop when the wavelength of the seismic wave is equal to the length of the ridge or landform resulting in seismic amplification. This is a complex frequency dependent and angle of incidence dependent phenomenon that cannot be expected for all earthquakes and is thus fairly unpredictable for a specific historical earthquake. However since, as noted by Wechsler et al (2008), Umm al-Qanatir sits atop a river gorge, it may be subject to such effects. Modeling this effect was deemed beyond the scope of the paper but the authors noted that by making conservative choices about the Newmark Displacement and the Factor of Safety, we assume that the topographic effect is indirectly partially compensated for.

References

Articles and Books

Dray, Y., et al. (2017). "The Synagogue of Umm el-Qanatir Preliminary Report." Israel Exploration Journal 67(2): 209-231.

Seismic Design for Nuclear Power Plants (1970:438-483)

Sukenik, E. L. (1935). The Ancient Synagogue of El-Hammeh, Hammath-by-Gadara, an Account of the Excavations Conducted on Behalf of the Hebrew University, Jerusalem, Rubin Mass.

Wechsler, N., et al. (2009). "Estimating location and size of historical earthquake by combining archaeology and geology in Umm-El-Qanatir, Dead Sea Transform." Natural Hazards 50(1): 27-43.

Websites

Umm al-Qanatir at Manar Al-Athar (Oxford University)

Bibliography from Stern et. al. (1993 v.2)

L. Oliphant, PEQ !7 (1885), 89-91

G. Schumacher, The Jaulan, London 1888,260- 265

Kohl-Watzinger, Synagogen, 59-70, 125-134

Goodenough, Jewish Symbols!, !99, 206-208

E. L. Sukenik, The Ancient Synagogue ofel-Hammeh, 85-87

id., JPOS 15 (1935), 172-174

M. Avi-Yonah, QDAP 14(1950), 57

Hiittenmeister-Reeg, AntikenSynagogen, 465--468

Z. Ilan, ES/6(1987-1988), 110; Z. U. Ma'oz, BA 5! (1988), 116-128.

Notes

The Original definition of the Arias Intensity can be found in Seismic Design for Nuclear Power Plants (1970:438-483).

Wikipedia page for Umm el-Qanatir

from Wikipedia