Fortress at Arad Quake

~250 BCE

by Jefferson Williams

Introduction     Textual Evidence     Archeoseismic Evidence     Tsunamogenic Evidence     Paleoseismic Evidence     Notes     Paleoclimate - Droughts     Footnotes     References


Archeoseismic Evidence from the Fortress at Tel Arad and elsewhere along with paleoseismic evidence from the Dead Sea suggests that a mid 3rd century BCE earthquake struck the area. This may or may not be related to tsunamogenic evidence from Elat.

Textual Evidence

Archeoseismic Evidence

Archeoseismic evidence is summarized below

Location Status
Fortress at Arad probable
En Erga possible to probable - I = IX - X epicenter to the ENE

Archeoseismic Evidence is examined on a case by case basis below
Fortress at Arad
Collapsed Roof of Water Reservoir at Fortress at Arad
Fig.3 Collapsed section of bedrock roof of water reservoir, looking east.
from Herzog, Z. (2002)

Herzog, Z. (2002) interpreted damage observed at the Fortress at Tel Arad during the Hellenistic Period and attributed the damage to a strong earthquake during the middle of the 3rd century BCE. Apparent seismic damage was observed at the southern and eastern wings of the fortress and in two cisterns where roof collapse was observed. Dating this damage is based on Hellenistic pottery shards found inside a debris filled depression that was presumed to have been caused by the earthquake and late Hellenistic structures built atop this debris filled depression and elsewhere. Hellenistic Structures were dated based on toothed chisel marks. Archeoseismic evidence for a mid 3rd Century BCE earthquake can best be described as possible. Relevant sections from Herzog's report are reproduced in the Notes section of this catalog entry.
'En Erga
Plan View of Seismic Damage at Ein Erga
Fig.3 Ein Erga. Plan of the fort with marked types of deformations
from Khorzhenkov and Erickson-Gini (2003)

Khorzhenkov and Erickson-Gini (2003) report seismic damage at 'En Erga in the 3rd century BCE with an local Intensity of IX-X. They also estimated that the direction of the epicenter from En' Erga was ENE. Archeoseismic Evidence for the Fortress at Arad Quake at En Erga is labelled as possible to probable (prossible). Some excerpts from their article follow:
These Nabataean forts [En' Erga and Ein Rahel] were constructed on the early branch of the Incense Road (the Darb es-Sultan) between Petra and Gaza during the Hellenistic period between the 3rd and 1st cent. B.C.. The fort at Ein Rahel was reoccupied by the Nabataeans in the early 1st cent. A.D. until the early 2nd cent. A.D. and it connected Petra with the Nabataean station at Ein Hazeva located further north in the central Arava.


Recent examination of the finds from the excavation of the fort [at En' Erga] indicates that it was constructed by the Nabataeans in the 3rd cent. B.C. However, this fort was abandoned before it was ever occupied, apparently as the result of an earthquake that damaged the structure, causing a large section of the southern wall to collapse. No evidence of occupation was found anywhere in the structure or surrounding area. A single Hellenistic incurved bowl dated to the 3rd cent. B.C. was found buried below the dirt floor of one of the rooms, probably placed there as a foundation deposit. This practice was found in Nabataean structures in later periods, including the second occupational phase at Ein Rahel and at Mampsis and Petra.


Today only the lower courses of the walls of this fort are visible. Upon examination the building appears to be unfinished as the result of a sudden and complete abandonment due to an earthquake. It is theorized that the fort was abandoned before it was completed and that a new fort was constructed a kilometer to the southwest, next to the spring of Ein Rahel and off of the main route (Y. Israel, 1998, personal communication).


Ein Erga

Collapse Features

In the ruins of the Ein Erga fort the walls facing the seismic wave collapsed systematically toward the seismically induced compression strain, whereas walls aligned parallel to the seismic wave lost support and collapsed in a random manner. A correlation between the orientation of construction elements and the direction of collapse was examined.


The wall oriented in the direction of NS180° in the Ein Erga fort reveals a clear picture of the collapse: the lower part of the wall is intact (as can be easily seen from its western side), whereas the upper its part collapsed southward (Fig. 6 a. b). This wall reveals collapse oriented towards E90°, whereas walls oriented in a perpendicular direction collapsed on both sides of the original wall‹s position. Fragments of the destroyed wall were thrown off up to 3.2 m from the structure. This would indicate that the direction of seismic wave propagation was roughly perpendicular to the NS oriented walls. The cone of collapse is asymmetric in form. This may indicate that the propagation of the seismic wave was in an E-W direction, but under some angle to the wall from ENE.

The Displacements of Rock Fragments and Building Elements

The shift of rock fragments and building elements may be used in a similar manner as wall inclination or block collapse. Some construction elements or rock fragments are shifted toward an epicenter due to inertia. Such examples were observed in the Ein Erga fort: three travertine blocks of the travertine plate underlying nearly the entire fort were thrown eastward (see Fig. 3) and rotated clockwise. One of these was displaced horizontally, 1.7 m (Fig. 7) eastward. A large block of travertine bedrock lies on the lower rows of the former wall and measures 1.40 m in length, 0.90m in width and 0.42m high. It weighs approximately 2 tons. These features indicate that the seismic energy radiated from the ENE.


The seismic intensity of the first earthquake [3rd century BCE] was about I = IX–X,
Archeoseismic evidence for the Fortress at Arad Quake is labeled as possible to probable.

Tsunamogenic Evidence

Goodman Tchernov et al (2016) identified a paleo tsunami deposit close to Elat from two submarine cores taken at North Beach and Tur Yam locations. They described the dating as follows:
The radiocarbon age from the North Beach places the maximum age at 100–400 BC (2 sigma error), while the Tur Yam radiocarbon age brackets the horizon as a minimum age of 100–500 BC (2 sigma error). Highest probability of these two radiocarbon ages place the event at about 2300 yBP, or around 200–300 BC (Data Repository).
The tsunamite deposit in the Tur Yam core was inferred from "an anomalous bed (~60cm) of more concentrated mixed shell and broken coral fragments of varying condition from pristine to heavily worn and eroded." The inferred tsunamite deposit in the North Beach core was encountered "at a depth of 160 cm down-core [where] the grain size increases to greater than coarse sand (>250 micron) and foraminifer abundances decrease significantly to either low or barren (0–150 individuals per cm3)." The anomalous bed in the North Beach Core was ~32 cm. thick.

Elat Cores
Fig 4 - Description and summary of analysis from Tur Yam and North Beach cores.
‘g’ = granule (2-4mm), ‘p’ = pebble (4-64mm), ‘c’ = cobble (64-256mm). Granulometry particle size distribution completed using Ocean Data View version 4.3.10. Correlation between the anomalous horizons of both cores presented. Detail of foraminifer counts and radiocarbon ages available in data repository. Examples of color ranges and corrasion of foraminifera in anomalous horizon as presented in Amphistegina lobifera (d’Orbigny 1826). Goodman Tchernov et al (2016)

Paleoseismic Evidence

Paleoseismic Evidence for the Fortress at Arad Quake is summarized below:

Location Status
Bet Zayda unlikely
Nahal Darga possible
En Feshka possible - several candidates various deformation types ~ 2 cm. thick
En Gedi none reported
Nahal Ze 'elim probable - 8 cm. thick intraclast breccia
Qatar Trench Jordan probable

Each site will now be discussed separately.
Bet Zayda
Wechsler at al. (2014) records event CH4-E6 with a wide modeled age range from 392 BCE – 91 CE in paleoseismic trenches at Bet Zayda just north of the Sea of Galilee (aka Lake Kinneret).
Bet Zeyda Earthquakes
Figure 9. Probability density functions for all paleoseismic events, based on the OxCal modeling. Historically known earthquakes are marked by gray lines. The age extent of each channel is marked by rectangles. There is an age uncertainty as to the age of the oldest units in channel 4 (units 490-499) marked by a dashed rectangle. Channel 1 refers to the channel complex studied by Marco et al (2005).

Dead Sea

Nahal Darga
In the coarser grained lithology present at Nahal Darga, Enzel et. al. (2000) report a 20 cm. thick seismite in Deformed Unit 8 in Stratigraphic Unit 10 which is dated to 250 BC +/- 200 (2000-2400 BP) (see Table 2).

En Feshka
Kagan et al (2011) in Table 3 report several seismites from En Feshka at depths of 425-447 cm. which might fit this earthquake.

En Gedi
Migowski et. al. (2004) do not report a seismite in the middle of the 3rd century BCE in the 1997 GFZ/GSI core DSEn (Table 2).

Nahal Ze 'elim (ZA2)
Kagan, et al (2011) in Table 3 lists a 8 cm. thick seismite at a depth of 552 cm. which might fit this earthquake.


Qatar, Jordan
Klinger et. al. (2015) hypothesized that Event E7 in a paleoseismic trench in the southern Arava near Qatar, Jordan may have been caused by an earthquake in ~150 BC or an earlier event (Date Range 338 BC – 213 BC). The Fortress at Arad Quake may be a better fit for this event than the Southern Dead Fish and Soldiers Quake of ~150 BC
Qatar Trench
Figure 6. Age model computed for the trench stratigraphy using OxCal v4.2 (Bronk-Ramsey et al. 2010) and IntCal13 calibration curve (Reimer et al. 2013). Light grey indicates raw calibration and dark grey indicates modeled ages including stratigraphic information. Phases indicate subsets of samples where no stratigraphic order is imposed. Klinger et al (2015)


Archeoseismic Evidence from the Fortress at Arad
Relevant sections from Herzog's (2002) report on the fortress at Arad are reproduced below:
pp. 12-13

Apparent damage to the southern and eastern wings of the fortress occurred during the Hellenistic period (3rd century BCE). The massive foundations, intended to guarantee the stability of a large tower erected at the centre of the site, completely destroyed remains of earlier periods. Additional severe damage resulted from the collapse of the rock roof of two of the water cisterns (Fig. 3). This event took place during the Hellenistic period, apparently the result of a strong earthquake. The collapse caused the complete destruction of all occupational remains and created a deep depression in the northwestern sector of the fortress. The depression was partly filled in with debris and partly built over by later Hellenistic-period structures. Moreover, the levelling of the depression with debris from the close surroundings eliminated most of the upper Iron Age remains (Strata VII and VI) in this area (Fig. 4). Consequently, the Hellenistic structures were erected at elevations similar to those of the Iron Age strata elsewhere. This chaotic process is responsible for the lack of architectural remains of the Iron Age strata in this area.
p 74

Based on evidence provided by the only intact cistern, the subterranean reservoir consisted of elliptical cisterns. There appear to have been three cisterns. The rock ceilings of two of these collapsed during the Hellenistic period. The considerable thickness of the rock layer that remained above the reservoir (approximately 2 m.) indicates that the collapse was not a result of the pressure of settlement layers, but the consequence of a powerful earthquake.
p 76

The water system of Arad is a unique example of a water storage system combined with a postern for emergency use. An earthquake apparently caused the collapse of the Arad water system as well as other systems in the south. From the excavations at Masada and Qumran, we know that earthquakes occurred during the 2nd and 1st centuries BCE (Karcz and Kafri 1978). During that same period, the water system at Tel Beersheba was also destroyed. Such a date is supported by the late Hellenistic sherds found amid the debris in depressions created as a result of the collapse (Fig. 4, above). The same episode probably also caused the collapse of the well in the lower city.

Paleoclimate - Droughts