Incense Road Quake

110-114 AD

by Jefferson Williams

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


At the beginning of the second century AD, the Nabataeans controlled a profitable and likely very secretive trade route. Frankincense and myrrh, harvested in South Arabia, and products from India and the Horn of Africa were transported by camel up the Arabian Peninsula to port cities such as Gaza where they were shipped and sold all over the Mediterranean. The transport route the camel caravans followed is known as the Incense Road . The way stations along that route were oases of prosperity in the middle of the harsh desert. Partly in order to control this trade , Rome annexed Nabataea in 106 AD. Within a decade, a powerful earthquake apparently struck their newly annexed province. There is no clear mention of this earthquake in any known literary sources. The Nabataeans produced almost no surviving literature of their own. Rather, the evidence for this earthquake was discovered by a young Archaeologist named Ken Russell by examining excavation reports for the way station towns and cities that were along the Incense Road.

Textual Evidence

Chronicon by Eusebius

Eusebius wrote Chronicon in two parts in the early 4th century AD. Although the original Greek text has been lost, later Chroniclers preserved significant parts of the manuscript. Jerome translated all of the second part (Book 2) into Latin. In this translation (Eusebius Chronicon Book Two, page 282, 227th Olympiad), we read that in the 227th Olympiad
Nicopolis and Caesarea were ruined in an earthquake.
Eusebius dates this to the first year of the 227th Olympiad which corresponds to July 1, 130 AD – June 3- 131 AD. [1]. The problem with this date is that there is little paleoseismic or archeoseismic evidence to corroborate it. As a result, two schools of thought have developed in explained what appears to be a mistake by Eusebius.Russell (1985) has suggested that this terse entry from Eusebius, a native of Caesarea who should have been aware of earthquakes that struck the area, may describe the Incense Road Earthquake (110-114 AD) and that Eusebius, using unknown sources and writing 200 + years after the event, merely got his date wrong. Although Russell (1985) does not propose a reason why Eusebius’ sources may have gotten the date wrong, one possibility is that his sources may have reported an earthquake that occurred during Hadrian's rule when in fact the earthquake occurred during the rule of Trajan ; Hadrian's predecessor. If one changes Eusebius' date for the earthquake from Hadrian's 13th – 14th year (130/131 AD) to Trajan's 13th – 14th year (111/112 AD) , one arrives at a date which is within the 4 year time span (110 – 114 AD) when the archeoseismic evidence constrains the timing of the Incense Road Earthquake. Ambraseys (2009) suggests that Eusebius was not referring to Nicopolis and Caesarea of the Palestinian coast but rather to two like named cities in the northeastern Anatolian province of Pontus and that his 130/131 AD date was approximately correct.

Babylonian Talmud

(Salamon et al, 2010) noted that Shalem (1956), using Judaic sources, suggested that the coast between Caesarea and Yavne was hit by a tsunami around 115 AD. Salamon et. al. (2011) offered a quote from Karcz et al (1987) regarding those Judaic sources as follows : “Talmudic references are not specific neither in time nor location, but Yavne may have been affected”. Karcz (personal communication, 2014) indicated that the Judaic sources come from the Babylonian Talmud and are located in

• Megilla 3a
• Baba Metzia 59 B
• Hullin 59 B
Megilla 3a
In Megilla 3a, the following passage describing an earthquake can be found
R. Jeremiah — or some say R. Hiyya b. Abba — also said: The Targum of the Pentateuch [2] was composed by Onkelos the proselyte under the guidance of R. Eleazar and R. Joshua. The Targum of the Prophets was composed by Jonathan b. Uzziel under the guidance of Haggai, Zechariah and Malachi, and the land of Israel [thereupon] quaked over an area of four hundred parasangs (1463 km.) by four hundred parasangs [3], and a Bath Kol came forth and exclaimed, Who is this that has revealed My secrets to mankind? Jonathan b. Uzziel thereupon arose and said, It is I who have revealed Thy secrets to mankind. It is fully known to Thee that I have not done this for my own honour or for the honour of my father's house, but for Thy honour l have done it, that dissension may not increase in Israel. He further sought to reveal [by] a targum [the inner meaning] of the Hagiographa, but a Bath Kol went forth and said, Enough! What was the reason? — Because the date of the Messiah is foretold in it.

But did Onkelos the proselyte compose the targum to the Pentateuch? Has not R. Ika said in the name of R. Hananel who had it from Rab: What is meant by the text, And they read in the book, in the law of God, with an interpretation. and they gave the sense, and caused them to understand the reading? ‘And they read in the book, in the law of God’: this indicates the [Hebrew] text; ‘with an interpretation’: this indicates the targum, ‘and they gave the sense’: this indicates the verse stops; ‘and caused them to understand the reading’: this indicates the accentuation, or, according to another version, the masoretic notes? — These had been forgotten, and were now established again.

How was it that the land did not quake because of the [translation of the] Pentateuch, while it did quake because of that of the prophets?
Jonathan ben Uzziel was a Rabbinic sage who survived the Roman siege of Jerusalem and relocated to the town of Yavne where Yohanan ben Zakkai founded a school of halakha (Jewish religious law) in 70 AD. Jonathan wrote Targum of the Prophets in the second century AD while in Yavne. It is unclear if the passage above contains a memory of an earthquake but if it does, it’s date is not specific; occurring sometime early in the second century AD.
Baba Metzia 59 B
Baba Metzia 59 B may contain a description of a seismic sea wave or tsunami which if related to the earthquake of Megilla 3a, may provide some dating information.
A Tanna taught: Great was the calamity that befell that day, for everything at which R. Eliezer cast his eyes was burned up. R. Gamaliel too was travelling in a ship, when a huge wave arose to drown him. 'It appears to me,' he reflected, 'that this is on account of none other but R. Eliezer b. Hyrcanus.' Thereupon he arose and exclaimed, 'Sovereign of the Universe! Thou knowest full well that I have not acted for my honour, nor for the honour of my paternal house, but for Thine, so that strife may not multiply in Israel! 'At that the raging sea subsided.

Ima Shalom was R. Eliezer's wife, and sister to R. Gamaliel. From the time of this incident onwards she did not permit him to fall upon his face. Now a certain day happened to be New Moon, but she mistook a full month for a defective one. Others say, a poor man came and stood at the door, and she took out some bread to him.[On her return] she found him fallen on his face. 'Arise,' she cried out to him, 'thou hast slain my brother.' In the meanwhile an announcement was made from the house of Rabban Gamaliel that he had died.
Tana refers to the Tannaim ; Rabbinic sages working from approximately 10-220 AD whose views are recorded in the Mishnah. Included among the Tannaim were Rabbi Gamaleil or more specifically Gamaliel II who apparently died right before the Kitos War of 115-117 AD (Moed Kattan 27a; Yerushalmi Moed Kattan 82a - from Wikipedia) and Rabbi Eliezer ben Hyrcanus who was Gamaliel II’s brother in law and also was one of the Tannaim. Gamaliel II’s and Rabbi Eliezer ben Hyrcanus both resided in Yavne at the time of this account which appears to have occurred soon before Gamaliel II’s death. Thus, if the passage above does refer to an actual seismic sea wave, it’s date is sometime before 115 AD.
Hullin 59 B
In Hullin 59 B, one can read the following
The Emperor once said to R. Joshua b. Hananiah, 'Your God is likened to a lion, for it is written: The lion hath roared, who will not fear? The Lord God hath spoken, who can but prophesy? But what is the greatness of this? A horseman can kill the lion'! He replied: 'He has not been likened to the ordinary lion, but to the lion of Be-Ilai'i!' 'I desire', said the Emperor, 'that you show it to me'. He replied: 'You cannot behold it'. 'Indeed', said the Emperor, 'I will see it'. He [R. Joshua b. Hananiah] prayed and the lion set out from its place. When it was four hundred parasangs distant it roared once, and all pregnant women miscarried and the walls of Rome fell. When it was three hundred parasangs distant it roared again and all the molars and incisors of man fell out; even the Emperor himself fell from his throne to the ground. 'I beseech you', he implored, 'pray that it return to its place'. He prayed and it returned to its place.

Another time the Emperor said to R. Joshua b. Hananiah, 'I wish to see your God'. He replied: 'You cannot see him'. 'Indeed', said the Emperor.
Apparently the "lion hath roared" in the passage above alludes to the prophetic book of Amos which starts with reference to an earthquake perhaps likening it to a lion's roar and two chapters later continues with several lines of poetry again mentioning the lions roar. Nonetheless, it is unclear if this seismically enigmatic passage has any relationship to an actual earthquake.

Taken together, these three passages suggest that there may have been an earthquake and possible tsunami experienced in Yavne in the years before 115 AD. A seismic exegesis for some of these Talmudic references is apparently discussed in Krauss, S. (1914).

Silence of the Sources

It should be noted that the seismicity of the Arava and Negev is severely under reported during this time; presumably due to the low population density and nomadic lifestyles of many of its inhabitants. While the Nabataeans residents left inscriptions on buildings, there is very little extant written Nabataean literature. This has to be a consideration when confronting the lack of corroborating historical information about the Incense Road Quake and many other earthquakes with epicenters in the South Dead Sea or the Arava. It should also be noted that there is no equivalent source in the second century AD which provides such a wealth of information regarding events in Judea such as Josephus (37 AD - ~100 AD) provides for the first century AD. Russell (1985) further notes that Cassius Dio (155 – 235 AD), an important source for information in the Roman empire during this time, failed to record two significant earthquakes (in ~106 AD and ~122 AD) in Anatolia during the reigns of Trajan (98 – 117 AD) and Hadrian (117 -138 AD) and that the only earthquake he did record during their reigns was the Trajan Quake in Antioch in 115 AD possibly because Trajan was nearly killed by it.

Archeoseismic Evidence


According to Russell (1985), there is abundant well dated archeoseismic evidence that there was an earthquake, likely along the Arava fault, between 110 and 114 AD. Russell (1985) contends that the evidence is too wide widespread to support alternative explanations based on war or civil disturbance. Ambraseys (2009) agrees with Russell (1985) that an earthquake or earthquakes is the most likely explanation for this evidence although he cautioned that, based on the evidence available to him at the time, the earthquake assessment was not yet definitive. The archaeological evidence suggests an early-second century destruction at Petra, Masada, Avdat and several other sites along the Petra - Gaza road (Russell 1985, 40-41). There are claims of further evidence from cities not along the Petra – Gaza Road some of which, upon close examination, seem dubious. However, since Russell (1985) first published his article, paleoseismic and additional archeoseismic evidence has emerged supporting his thesis that an earthquake struck the Arava between 110 and 114 AD. The dates proposed by Russell (1985) are constrained by two pieces of evidence. A coin from 110/111 AD found under earthquake debris in Masada provides the earliest possible date. An inscription dedicated to Roman emperor Trajan associated with rebuilding in Petra provides the latest possible date (114 AD). A list of locations with archeoseismic evidence for the Incense Road earthquake is provided below accompanied by our assessment.

Location Status
Jerash unlikely
Heshbon indeterminate
Caesarea needs investigation
Masada probable
Khirbet Tannur possible
Alia probable
Petra probable
Avdat/Oboda possible
Mampsis indeterminate
Moje Awad indeterminate
Ein Rahel probable - I = VIII–IX epicenter ESE (~125 degrees) several km. away
'En Ziq needs investigation
'En Yotvata possible
Other sites in the Negev needs investigation

Each site will now be discussed separately.


Inscription On North Gate of Jerash
Inscription On North Gate of Jerash - photo by Jefferson Williams

Drawing of Inscription On North Gate of Jerash
Drawing of Inscription On North Gate of Jerash - photo by Jefferson Williams

According to Kraeling (1938, p.47), a new north gate was constructed in Jerash in 115 AD. The dedicatory inscription from the gate was discovered in 6 fragments (Kraeling, 1938, p. 424) where Trajan is referred to as the "savior and founder" of the city. While Russell (1985) speculated that the civic dedication may reflect imperial aid Trajan supplied to aid reconstruction after a disastrous earthquake, Kraeling (1938, p.47) attributes the dedication to the improvement of the roads out of Jerash; in particular the Road to Pella which enabled direct connections to the coastal cities of Caesarea and Ptolemais (aka Acre). If the Incense Road Earthquake was caused by a fault break on the Arava fault, seismic damage at Jerash would have been light so we prefer Kraeling’s (1938, p.47) explanation.

Tell Hesban (Roman Esbus) aka Heshbon aka Hesban

Mitchel (1980, Ch. 4 - Stratum 13 and 14) noticed a massive collapse of bedrock in underground structures at a location known as Tell Hesban. He attributed the collapse to an earthquake. In fill atop the bedrock collapse, he found a coin from Aretas IV (9 BC – 40 AD) along with a number of pottery shards dated to a wide spread of ages over hundreds of years but with the bulk of shards from Early Roman I-IV (63 BC – 135 AD). He surmised that the fill was deposited soon after the bedrock collapse because he saw no evidence for extended exposure before filling (silt, water-laid deposits, etc.) and the fill was relatively homogeneous, unstratified, and loose soil that “gave the appearance of rapid deposition in one operation”. Mitchel (1980) assigned a date of 130 AD to the destruction layer caused by an earthquake citing Chronicon by Eusebius as a historical reference ( see Jerome's Latin translation of Eusebius Chronicon Book Two, page 228, 227th Olympiad). Russell (1985) favored a date of 110-114 AD for this destruction layer.

In our opinion, this destruction layer is not well dated. It rests on an untested assumption that the fill above the bedrock collapse layer was deposited soon after the collapsed bedrock. None of Mitchel’s (1980) date information comes from beneath the collapsed layer. If one dispenses with the argument that the fill was deposited immediately after an earthquake and examines the variety of dated objects found in the fill, the objects suggest that the fill was deposited over some period of time and the most probable earthquake candidate for the bedrock collapse layer might be the 31 BC Josephus Quake. In our opinion, this archeoseismic evidence is indeterminate.


Russell (1985) claims that early 2nd century destruction evidence is archaeologically attested at Caesarea. Karcz and Kafri (1978) report the following: "Caesarea 2nd century AD displaced offshore moles and port installations tilted walls".


Yadin, Y. (1965, 30) noted that the Great Public Bath-House (Caldarium ) was filled, as a result of earthquakes, with massive stone debris concluding that the finds on the floor of the bathhouse represented the last stage in the stay of the Roman garrison left on Masada after the end of the first Jewish War in 73 AD [4]. This was based on the presence of “vouchers” written in Latin and coins which were found mostly in the ash waste of the furnace. One coin from the time of Trajan was found in the Caldarium which was struck in Tiberius in 99/100 AD (Yadin, Y., 1965, 36 and 118). The latest coin discovered from this occupation phase was found in one of the northern rooms of Building VII and dates to 110/111 AD (Yadin, Y., 1965, 119). According to Yadin (1965, 119), this meant that the Roman garrison stayed on Masada until at least 110/111 AD. According the Russell (1985), this coin supplies the earliest possible date for the Incense Road Earthquake.

Khirbet Tannur

Khirbet Tannur
Displaced Building Stones at Khirbet Tannur - photo by Jefferson Williams

Glueck (1965) excavated a Nabataean Temple at Khirbet Tannur in Jordan. He identified three separate building phases which he subdivided into Periods I, II, and III. Period III ended when “a violent earthquake undoubtedly destroyed entire temple” (Glueck, 1965, 122). The date of the earthquake that ended Period III is unknown. What can be surmised is when Period II ended and Period III began. Glueck (1965, 138) dated the rebuilding phase that started Period III to the first quarter of the second century AD based on the architecture and sculpture thinking the rebuild probably, but not definitively, began before the Roman annexation of Nabataea in 106/107 AD. The reason for the rebuild started in Period III is unknown. It could have been due to age of the Temple or previous earthquake destruction.

Thus, archeoseismic evidence for earthquake destruction at Khirbet Tannur during the Incense Road Earthquake of 110-114 AD is possible but indeterminate.

Alia (Aqaba, Jordan)

Thomas et. al. (2007) also saw probable archeoseismic evidence for this earthquake in Alia. Thomas et. al. (2007) examined a Nabataean Stone and mudbrick structure in Area J-East inside the modern city of Aqaba, Jordan. Area J-East was part of the ancient port city then known as Aela or Alia. Thomas et. al. (2007) noted that early excavations indicated that the site was first occupied in the first century AD and dated it to pre Roman annexation Nabatea based on pottery that was found there. That occupation ended with a violent structural collapse in the early part of the second century AD. The occupation deposits were covered by a thick layer of mudbrick collapse which Thomas et. al. (2007) labeled as Event EQ VIII. The thickness of the collapsed layers mostly varied from ~0.5 - 1.0 m (see Fig. 3 on p. 64) - sometimes exceeding 1 meter and the collapse dented the surfaces below indicating a violent fall of the structures. The cause of the collapse was thought to be related to either Rome’s annexation of Nabataea in 106 AD or the Incense Road Earthquake of 110 – 114 AD. Thomas et. al. (2007) prefered the Incense Road Earthquake as an explanation for the collapse. However, they noted that because there is ongoing debate about the degree of Nabataean resistance to the Roman annexation, it is not possible to rule out human agency for the collapse. Dolinka (2003) - Nabatean Alia from a Ceramic Perspective identified tumbled over mudbricks from the Area B.1/3 domestic complex in Alia which he interpreted as earthquake damage from the Incense Road Quake thus ushering in abandonment Phase II on the site (see discussion on p. 23 and Fig. 14 on p. 31). The earthquake evidence in Alia is probable but not definitive.

Incense Road

Russell (1985) relates that extensive damage was revealed during the excavation of cardo maximus in Petra. Russell (1985) further relates that archeologists believe that this damage could only have been caused by an earthquake although there are arguments that this destruction resulted from sacking by Safaitic and Thamudic hordes in the mid first century AD. A monumental commemorative arch dedicated to Trajan by the city late in 114 AD was recovered (Kirkbride 1960: 120) but the sections of the inscription that would have documented the reason for this dedication were not recovered. Nevertheless, Russell (1985) proposed that this inscription was a result of rebuilding efforts after the Incense Road Earthquake. As such, this inscription, according to Russell (1985) provides the latest possible date (terminus ante quem) for the Incense Road Earthquake. Russell also noted that during the 1976 excavations at Petra, a brass coin (sestertius) commemorating Trajan's alimenta italiae endowment was uncovered on a floor-slab next to several crushed unguentaria in a storage room of a collapsed house of the early 2nd century. Russell (1985) relates that Sestertii of this type were minted between 103 and 117 ( Robertson 1971: 57-59, and pl. 13, nos. 344, 350, 354). Unfortunately, the consulship was illegible in the obverse inscription which would have allowed for a more precise dating. Coins of the last Nabataean king, Rabbel II (71- 106), have been noted in association with this destruction evidence at Petra (Kirkbride 1960: 118- 19; Parr 1960: 129).

Kolb (2002) reported on the excavation of a large residential structure in ez-Zantur in Petra. They dated the earliest phase of the structure to the 20's CE and report that it was remodeled "in the early decades of the 2nd century CE" after apparent earthquake damage. A terminus post quem of 103-106 CE for the remodel was provided by a coin struck under King Rabbel II and found in some rough plaster (rendering coat) in Room 212 of site EZ III (Kolb, 1998:263).

Archeoseismic evidence for an earthquake around the time of the Incense Road Quake can be labeled as probable.
Avdat aka Oboda
Potter's workshop at Avdat
Potter's workshop at Avdat - photo by Jefferson Williams

Russell (1985) noted the following :
"At Avdat, an imperial coin struck at Alexandria and tentatively identified as Trajanic was apparently found in association with the collapse of the potter's workshop (Negev 1974: 24)."
Korzhenkov, A. and T. Erickson-Gini (2003) cast doubt on Russell's assertion of archeoseismic damage at Mamphis and Avdat stating that recent research indicates a continuation of occupation throughout the 1st and 2nd cent. A.D. citing an MA Thesis (A Nabataean Roman Settlement in the Central Negev Highlands in the Light of the Ceramic and Architectural Evidence Found in Archaeological Excavations During 1993–1994, Unpublished M.A. Dissertation, Tel Aviv University (1999)) and subsequent paper ( Erickson-Gini, New Excavations in the Late Roman Quarter in Avdat, Proceedings of the Twenty-Seventh Archaeological Congress in Israel, Bar Ilan University April 2–3, 2001). Continuous occupation could indicate that seismic damage was limited rather than absent.

Ambraseys (2009) supplied this observation and analysis
"Negev argues instead that these destructions were caused by invading Safaitic and Thamudic hordes in the mid first century ( Negev 1976), basing his thesis on the period of pottery debris found in a workshop at Oboda. This solution might seem preferable, since it is best not to assume an earthquake unless there is written evidence for it. However, apart from the complexity of the multiple dates of the pottery discovered by Negev (and the fact that later potters often imitated earlier styles), the appearance of a second-century coin among the pottery (Russell 1981, 8) seems to refute his thesis. Of course, this coin does not prove that Oboda was destroyed by an earthquake; it merely shows that Negev has made a mistake. What may suggest an earthquake is the sheer severity and extent of the destruction. Russell believes that neither a Roman annexation of the territory nor sacking by Safaitic or Thamudic hordes could, in any case, have done so much damage. It is also quite possible that these towns were damaged by a series of distinct earthquakes, since Petra and Oboda are some 80 km apart. Certainly, though, given that the history of this area in the early second century is relatively well documented, no invasions being recorded, seismic activity seems the most likely, but not definite, cause of damage in the absence of any other solution."
Archeoseismic evidence at Avdat could thus be labeled as possible.
Russell (1985) cites Negev (1971, p 166) for evidence of early second century earthquake destruction at Mamphis. Negev (1971) reports extensive building activity in Mamphis in the early second century AD obliterating much of the earlier and smaller infrastructure. However, neither a destruction layer nor an earthquake is mentioned in the article. Korzhenkov, A. and T. Erickson-Gini (2003) cast doubt on Russell's assertion of archeoseismic damage at Mamphis and Avdat stating that recent research indicates a continuation of occupation throughout the 1st and 2nd cent. A.D. citing an MA Thesis (A Nabataean Roman Settlement in the Central Negev Highlands in the Light of the Ceramic and Architectural Evidence Found in Archaeological Excavations During 1993–1994, Unpublished M.A. Dissertation, Tel Aviv University (1999)) and subsequent paper ( Erickson-Gini, New Excavations in the Late Roman Quarter in Avdat, Proceedings of the Twenty-Seventh Archaeological Congress in Israel, Bar Ilan University April 2–3, 2001). Continuous occupation could indicate that seismic damage was limited rather than absent.

This archeoseismic evidence is best labeled as indeterminate.
Moje Awad
Cohen (1982) relates that three building and occupation phases have been identified at Moje Awad. In the middle phase plentiful pottery dating until the end of the first century AD was present along with Nabatean coins from the reigns of Aretas II (103 BC – 96 BC), Aretas IV (9 BC – 40 AD), and Rabbel II (70 – 106 AD). In the final phase, ceramic vessels dating to the 2nd to 3rd century AD were present along with coins from Roman Emperors Trajan (98 – 117 AD), Commodus (161 – 192 AD), and Caracalla (188 – 217 AD). Although a destruction layer was not mentioned between the middle and final phase, both the Incense Road Earthquake of 110 – 114 AD and the Roman annexation of Nabataea in 106 AD neatly fits at the interface between the end of the middle phase and beginning of the final phase. This building phase could represent a response to earthquake destruction as hypothesized at Khirbet Tannur. However, since it cannot be ascertained whether the building phase is a result of Roman annexation of Nabatea in 106 AD or earthquake destruction between 110 and 114 AD, this earthquake evidence can be best described as indeterminate.

Erickson-Gini and Israel (2013) opined that:

The Early Roman phase of occupation in the site ended with extensive damage caused by an earthquake that took place shortly before the Roman annexation of the region in 106 CE (Korjenkov and Erickson-Gini 2003). The building in Area C and the kiln works were destroyed, and the cave dwellings were apparently abandoned as well. Reconstruction was required in parts of the fort. At this time, deposition from its floors was removed and thrown outside of the fort and a new bath as well as heating were constructed in its interior. Along its eastern exterior and lower slope, rooms were added. Thus, the great majority of the finds from inside the fort and its ancillary rooms date to the latest phase of its occupation in the Late Roman, post-annexation phase, the latest coins of which date to the reign of Elagabalus (219–222 CE).
This archeoseismic evidence is labeled as possible.

Nabataean Fort Ein Rahel

Plan View of Seismic Damage at Ein Rahel
Fig.5 Ein Rahel. Plan of the fort with marked types of deformations. Following Shamir (1999)
from Khorzhenkov and Erickson-Gini (2003)

Korzhenkov, A. and T. Erickson-Gini (2003) report that the fort at Ein Rahel was first constructed and occupied in the 3rd century BCE followed by a hiatus in occupation from ~100 BCE until the early first century CE after which it was re-occupied. The fort was then, according to the authors, re-abandoned in the early second century CE. The abandonment came after the fort's destruction due to what the authors beleive was a seismic event. The archeoseismic evidence for the early second century CE earthquake was described as follows:
In the surrounding casemate rooms the latest occupational phase (dating to the early 2nd cent. A.D.) was sealed by the collapse of the upper floor of the fort. Sections excavated in these rooms revealed clear collapse of the ceiling of the lower floor and the upper floor debris sealed by the upper floor roof. The ceiling and roof of the structure were made from woven organic matting and mud and were supported by wooden beams.

A rich ceramic assemblage was discovered in the fort as well as extensive organic finds and included reed-matting and wooden beams, almond shells, nuts and olive and dates stones. Several wooden lice combs and other wooden objects were found in excellent condition, as well as many shreds of textiles and leather. Two camel bones were found bearing inscriptions in black ink in the Nabataean script.
Shamir (1999) examined the textiles, basketry and cordage and reported that
Preserved by the arid climate, the perishables from `En Rahel include about 300 textile and basketry fragments, cordage, spindle whorls and needles. The Early Roman date of the material, provided by its archaeological context, differs slightly from its radiocarbon dating [1] (Carmi and Segal 1995:55).

[1] In the fall of 1991, a brown goat-hair textile fragment from L13, Basket 129 was submitted to I. Carmi and D. Segal at the 14C laboratory of the Weizmann Institute of Science, in order to verify the archaeological conclusions. Their results suggest the fabric was manufactured in the Roman period:

Sample d14C d13C yrs BP* Calendaric Age**
RT-1596 -209.2 ± 3.9 -15.95 1885 ± 40 82-204 CE
* Conventional radiocarbon years before 1950.
** Calculated using CALIB 3 (Stuiver and Reimer 1993).

Note by Jefferson Williams : The radiocarbon date reported in Carmi and Segal 1995:55 is a bit different (and earlier)

Sample d14C d13C yrs BP* Calendaric Age**
RT-1596 -209.2 ± 3.9 -15.95 1885 ± 40 66-145 CE (87%), 165-186 CE (13%)
* Conventional radiocarbon years before 1950.
** Calculated using CALIB 3 (Stuiver and Reimer 1993).

As for seismic effects, the authors mention the inclination of walls, the collapse of walls and lintels and the rotation of building elements. Based on their analysis of structural damage, they hypothesize that the epicenter of the causative earthquake was located several kilometers ESE (~125 degrees) of Ein Rahel and estimated local seismic intensity of VIII–IX.

Tali Erickson-Gini (Personal communication, 2021) relates that the date of seismic destruction at Ein Rahel was sometime during the late first century CE or early second century CE. This archeoseismic classified as probable.

Hellenistic-Nabataean Fort 'En Ziq

Erickson-Gini (2012) report that at the site of ’En Ziq in the Nahal Zin basin near Sde Boker, the same early 2nd century earthquake which destroyed the Nabatean Fort at Ein Rahel also destroyed the Nabatean Hellenistic fortress at 'En Ziq. Unfortunately, no details for this assessment at 'En Ziq were provided. Due to the absence of details in this reference, this archeoseismic evidence is classified as needs investigation.

Horbat Hazaza

Erickson-Gini (personal communication, 2021) relates that there is archeoseismic evidence at Horbat Hazaza for destruction and rebui9lding after an earthquake in the late 1st or early 2nd century CE. Erickson-Gini excavated this site after initial work by Rudolph Cohen. This archeoseismic evidence is classified as possible.

'En Yotvata

Erickson-Gini (personal communication, 2021) relates that there is archeoseismic evidence in a Nabatean Building at 'En Yotvata for seismic damage at 'En Yotvata. This archeoseismic evidence is labeled as possible.

Other sites in the Negev

Erickson-Gini and Israel (2013) noted that:
Evidence of an early second-century CE earthquake is found at other sites along the Incense Road at Nahal Neqarot, Sha'ar Ramon, and particularly at the head of the Mahmal Pass where an Early Roman Nabataean structure collapsed (Korjenkov and Erickson-Gini 2003; Erickson-Gini 2011). There is ample evidence of the immediate reconstruction of buildings at Moyat ‘Awad, Sha'ar Ramon, and Horvat Dafit. However, this does not seem to be the case with the Mahmal and Neqarot sites.
Archeoseismic Evidence at these sites is labeled as possible.

Tsunamogenic Evidence


Although Goodman-Tchernov and Austin (2015) and earlier researchers identified a tsunamite in cores taken immediately offshore of the harbor in Caesarea which they associated with the Trajan quake of ~115 AD, as noted in the Tsunamogenic Evidence section of the Catalog entry for the ~115 AD Trajan Quake, this association is unlikely. Salamon et al (2011) noted that the presence of a tsunami far south of the supposed epicenter of the Trajan Quake does not fit the usual pattern of tsunamis on the Israeli coast where most other tsunamis which hit the Israeli coast were generated by ruptures more or less opposite to the coast (e.g. from the Cypriot and Hellenic Arcs). While Salamon et. al. (2011) suggested a storm surge as a possibility, we are proposing that an offshore shelf collapse caused by the Incense Road Earthquake of ~110 - ~114 AD as the mostly likely cause of the tsunamite of offshore Caesarea.

Caesarea Tsunamites
Figure 2. Core stratigraphy. Terrestrial-nearshore (T-NS) and W areas are summarized stratigraphies from those areas and do not represent individual cores or true horizon dimensions (msl—mean sea level). Extended descriptions are available in Figure DR2 (see footnote 1).

**Ages summarize the range of all ages for that horizon (C—radiocarbon, O and OSL—optically stimulated luminescence, P—pottery), reported with 2σ error (Tables DR1 and DR2).

N/A—not available. Dates are from this study and previous reports (Raban, 2008; Reinhardt and Raban, 1999, 2008; Reinhardt et al., 1994).

Sea-level curve model (blue zone represents minimum and/or maximum values) from Israel demonstrates that sea-level change would not have altered coastal zones of horizons during past 4 ka (Sivan et al., 2001).

Goodman-Tchernov et al (2009)

Some simple calculations are performed below to see if the Incense Road Earthquake could have produced the tsunamogenic deposits in Caesarea due to a localized offshore shelf failure.

1. Calculate ahmax from seismite thickness (based on Williams (2004) – valid for Nahal Ze ‘elim)

One can calculate ahmax from h


h = thickness of seismite (cm.)
ahmax = Peak Horizontal Ground Acceleration - calculated from CCS

h ahmax
(cm) (g)
1.0 0.24
3.0 0.28
5.0 0.31
9.0 0.36

Kagan et. al. (2011) identified a 5 cm. thick seismite in Nahal Ze ‘elim which was likely caused by the Incense Road Earthquake of 110-114 AD resulting in:

ahmax = 0.31 g.

Calculate ahmax
2. Magnitude from fault distance and ahmax

Determine fault distance from location to nearest or most likely earthquake producing fault

Fault distance = 13 km.
Enter ahmax = 0.31 g
Magnitude = 7.5

3. Fault break length based on magnitude (Wells and Coppersmith, 1994)

Magnitude = 7.50
Fault Break = 124 km.

4. Compute ahmax base on Magnitude and distance to location

1. Magnitude = 7.5
2. Locate the nearest fault break to Caesarea.
3. Calculate the Distance (R) from fault break to Caesarea (150 km.).
4. Use a variety of attenuation relationships to estimate peak horizontal ground acceleration (ahmax) at Caesarea.

The result is a peak horizontal ground acceleration (ahmax) of 0.04 – 0.14 g in Caesarea.

Can this cause an offshore slope failure that would produce a localized tsunami in Caesarea ? ck Wechsler slope stability article

Paleoseismic Evidence

A list of locations with paleoseismic evidence for the Incense Road earthquake is provided below accompanied by our assessment.

Location Status
En Feshka no evidence
En Gedi good evidence - ~ 5 cm. thick - variable thickness - flowed during quake
Nahal Ze 'elim good evidence - 5 cm. thick
Taybe Trench good evidence
Qatar, Jordan possible

Each site will now be discussed separately.

Dead Sea

There appears to be a clear spatial pattern at play for an early second century earthquake with a 5 cm. thick seismite in the south (Nahal Ze ‘elim) and no seismites observed in the north (En Feshka). This, combined with paleoseismic evidence from the Arava, indicates that this early second century earthquake was caused by a fault break on the Arava Fault; likely associated with the Incense Road Earthquake of ~110 - ~114 AD.
En Feshka
Kagan et. al. (2011) did not see any evidence for a seismite created around this time.

En Gedi (DSEn)
Migowski et. al. (2004) assigned two seismites at depths of 264 and 265 cm. (2.64 and 2.65 m) at En Gedi to earthquakes in 112 and 115 AD. The 112 AD date refers to the 110-114 AD Incense Road Quake and the 115 AD date refers to the Trajan Quake which was too far away to have created a Dead Sea seismite. During field work in January 2014 in the nearby En Gedi Trench, Williams saw evidence for a sizeable earthquake around 112 +/- 8 AD which was probably created by the Incense Road Quake. Williams also observed two detachment planes in the Incense Road Quake seismite (use magnifying glass to see at high resolution) which might explain why Migowski, while doing microscope work on the En Gedi Core, identified two separate seismites from the same deformation event.

Nahal Ze ‘elim (Site ZA-2)
Kagan et. al. (2011) dated a 5 cm. thick seismite at a depth of 445 cm. to 86-164 AD (1 σ) and assigned a date of 115 AD. The 115 AD date refers to the Trajan Quake which was too far away to have created a Dead Sea seismite so a correction has been made to associate this seismite with the 110-114 AD Incense Road Quake.



Taybe Trench, Jordan
LeFevre et al. (2018) identified a seismic event (E4) in the Taybeh trench in the Arava which they modeled between 14 BC and 205 AD and associated with the Incense Road Earthquake which struck between 110 AD and 114 AD.

Taybeh Trench Earthquakes
Figure S5: Computed age model from OxCal v4.26 for the seismic events recorded in the trench

Qatar, Jordan
Klinger et. al. (2015) identified a seismic event (E6) in a trench near Qatar, Jordan in the Arava which they modeled between 9 BC and 492 AD. The large spread in age caused them to consider two possible earthquakes as the cause; the Incense Road Earthquake between 110 AD and 114 AD and the southern Cyril Earthquake of 363 AD. They preferred the Cyril Earthquake of 363 AD based on weighing other evidence [5] not related to their paleoseismic study and noted that further investigation was required.

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 modelled ages including stratigraphic information. Phases indicate subsets of samples where no stratigraphic order is imposed. Klinger et al (2015)


Russell (1985) reports that Elias of Nisbis writing around 1019 AD in his book Chronographia has an earthquake report similar to the one reported by Eusebius. In a French translation of Chronographia (top of p. 57) we can read (converted to English)
An 438 - In that year there was an earthquake: Nicopolis and Caesarea were overthrown.
A.G. 438 comes from the Seleucid calendar and is often abbreviated as A.G.(Anno Graecorum). Russell(1985) relates that A.G. 438 corresponds to 126/7 AD. Chronographia in original Syriac can be found here.

Paleoclimate - Droughts


[1] See Finegan (1998) Sections 185 – 187 for a discussion of the Olympiad calendar system.

[2] The Pentateuch is a Greek translation of the Torah - the first five books of the Hebrew Bible (aka Old Testament).

[3] A parasang is a Persian mile. There are differing accounts of the exact distance of a parasang. Karcz (2004) states that this is 4000 yards which is in approximate agreement with other estimates. Using the reckoning of Karcz (2004), 400 parasangs equals to 1463 km. Although 400 parasangs by 400 parasangs does seem like an impossible distance, this is not necessarily a reason to reject this account. The ancient sources frequently exaggerate when it come to numbers and this is particularly the case in religious/spiritual literature. In addition 400 parasangs may be a symbolic number and/or a euphemism for a wide area. 400 parasangs was also used in a Talmudic account to describe the extent of the Pig on the Wall Quake.

[4] The stationing of a Roman Garrison after the conquest of Masada is corroborated by Josephus in his Book The Jewish War where he says in Book VII Chapter 10 Paragraph 1
WHEN Masada was thus taken, the general left a garrison in the fortress to keep it, and he himself went away to Cesarea; for there were now no enemies left in the country, but it was all overthrown by so long a war.
The next occupation of Masada after the earthquakes was a Byzantine settlement in the 5th century AD.

[5] Archeoseismic Evidence, Historical Reports, and Dead Sea Seismite Evidence.

Missing references


Negev, A. (1974). The Nabatean Potter's Workshop at Oboda, Habelt.


Kirkbride, D. (1960). "A Short Account of the Excavations at Petra in 1955-56." Annual of the Department of Antiquities of Jordan: 4-5. Kirkbride 1960: 118-19;
Parr, P. J. (1960). "Excavations at Petra, 1958–59." Palestine exploration quarterly 92(2): 124-135.Parr 1960: 128-29;


Toombs. L. E. 1978 The Stratigraphy of Caesarea Maritima. Pp. 223-32 in Archaeology in the Levant: Essays for Kathleen Kenyon, ed. R. Moorey and P. Parr. Warminster. England: Aris and Phillips.


Shalem, N. (1956). "Seismic tidal waves (tsunamis) in the Eastern Mediterranean (in Hebrew)." Society for Exploration of Eretz Israel Exploration Journal(20): 159-170.