Event D at Nahal Ze'elim site ZA-1 a 4 cm thick Type IV seismite with an estimated intensity of 9. ZA-1 was the first site in Nahal Ze'elim to be systematically investigated for Holocene paleoseismicity and established the basic framework for correlating earthquake-generated deformation structures with the historical earthquake record of the Dead Sea region. The site occupies a relatively landward position within the Ze'elim Terrace compared with the more seaward exposures examined in later studies such as Kagan et al. (2011). As a result, ZA-1 contains a greater proportion of shoreline and nearshore facies, whereas the later studies accessed more seaward sections characterized by thicker and more continuous lacustrine deposition. These more offshore settings preserved a larger number of earthquake-induced deformation horizons, resulting in a more complete Holocene paleoseismic record than was available at ZA-1.
Event A occurs within the northern gully at site ZA-1, ~395 cm above the base of the studied section, and is hosted within the Ze'elim Formation, a sequence of laminated aragonite and detrital sediments deposited in the paleo-Dead Sea following the retreat of Lake Lisan. The deformed horizon occurs within the upper part of the ZA-1 section below a major erosional unconformity. Like Events A, B, and C, it takes the form of a mixed layer produced by deformation of water-saturated lake-floor sediments during earthquake shaking. Although the mixed layer is only a few centimeters thick, Ken-Tor et al. (2001a) regarded it as evidence of a significant regional earthquake because it occurs in a stratigraphic position consistent with one of the best-documented seismic events of Late Antiquity.
Unlike Events A and B, Event D contains no datable organic debris and therefore could not be constrained by direct radiocarbon determinations. The nearest sample below the mixed layer yielded a calibrated age of 340-540 CE, while the nearest sample above yielded a calibrated age of 1030-1210 CE. Because the younger sample occurs above a major erosional unconformity and a hiatus in sedimentation, Ken-Tor et al. (2001a) rejected the very low sedimentation rate implied by the two dates and instead applied the higher sedimentation rates (4-9 mm yr-1) observed lower in the section. Using this approach, they estimated an age range of approximately 358-580 CE ( 2σ) for Event D and noted that a similar result is obtained by projecting upward from Event B, which served as chronological anchor horizon ( 31 BCE Josephus earthquake).
On the basis of this chronology, Ken-Tor et al. (2001a) correlated Event D with the 363 CE Cyril earthquake, although at the time it was not yet widely recognized that the historical evidence records two separate earthquakes in 363 CE rather than a single event. Likewise, Williams (2004), working with the same historical framework, also correlated Event D with the 363 CE earthquake. Subsequently, Revital Bookman (nee Ken-Tor) co-authored Leroy et al. (2010), in which the Event D horizon was still assigned to the 363 CE Cyril earthquake. Potential supporting evidence for this assignment is provided by the discovery of tombstones commemorating three 363 CE earthquake victims at Safi, approximately 35 km. south-southeast of ZA-1 and reports of extensive damage in Jerusalem ~50 km. to the north-northwest.
In contrast, Agnon et al. (2006: Figure 8) reassigned the Event D horizon to the 419 CE Monaxius and Plinta earthquake. Kagan et al. (2011) identified a seismite at Nahal Zeelim site ZA-2 that they also correlated with the 419 CE Monaxius and Plinta earthquake rather than one of the 363 CE Cyril earthquakes. The disagreement between the 363 CE and 419 CE correlations remains one of the more important chronological problems in the late Holocene Dead Sea paleoseismic record. One possibility is that one or both earthquakes were not expressed uniformly at all sites. Another possibility is that one of the two events has been misidentified because of uncertainties in age control. This issue is particularly relevant at En Gedi, where Migowski et al. (2004) identified a seismite that they correlated with the 419 CE earthquake but did not identify a corresponding horizon for the 363 CE earthquake. Their chronology relied heavily on varve counting, and the relevant part of the sequence lies several centuries from the nearest historically constrained anchor earthquake. As a result, small cumulative counting errors could potentially affect the final correlation. At present, the original assignment of Event D to one of the 363 CE Cyril earthquakes proposed by Ken-Tor et al. (2001a) and supported by Williams (2004) remains a plausible interpretation, although the alternative correlation with the 419 CE Monaxius and Plinta earthquake cannot be ruled out.
- Fig. 1c Oblique aerial photo of
SW Dead Sea showing Masada and Zeelim Plain from Agnon et al. (2006)
An oblique air photo looking southward at the southwestern part of the Dead Sea showing faults (solid lines) and locations of studied late Pleistocene sections M and PZ. MFZ—Masada Fault Zone. The photo was taken in 1940 when the lake level was 395 m below sea level. The current level is 420 m below sea level; additional area has emerged, including the Ze' elim fan outcrop.
click on image to open in a new tab in Google Earth
Agnon et al. (2006)- Nahal Ze'elim outcrop areas
in Google Earth
ZA-I is a guesstimate
click on image to explore this site on a new tab in Google Earth- Figure 2 Annotated Lithosection
of ZA-1 with interpreted ages noted from Ken-Tor et al. (2001a)
The lithology and chronology of a composite section exposed in Ze'elim Plain. The section is described from two outcrops exposed in different gullies 300 m apart. The correlation between the outcrops is based on the sedimentary sequence, laminae counting, and 14C dates. Ages presented in 14C years B.P. Deformed units (mixed layers and liquefied sands) are marked by capital letters.
slightly modified by JW - red dates as interpreted by JW added
click on image to open in a new tab
Ken-Tor et al. (2001a)- Figure 8 Age Model for ZA-1
from Agnon et al. (2006)
A modified age model for the Ze'elim section studied in outcrop (Ken-Tor et al., 2001a) and drill core (Migowski et al., 2004). A–H denote events discussed in the text. The present model was constrained by two rules:
(1) each event horizon (top of each intraclast breccia) matches a historical earthquake of notice.
(2) each continuous deposition segment shows a uniform deposition rate.
Two outcomes support the model. Two of the breccia layers match pairs of earthquakes (64–31 BCE; 1202–1210 CE) such that the earlier event horizon is within the breccia layer and the later event matches the top. With these assignments for the event horizons, the model gives a uniform sedimentation rate of 0.5 cm/yr during the three periods separated by hiatuses.
click on image to open in a new tab
Agnon et al. (2006)- Figure 3 Age Model for ZA-1
from Ken-Tor et al. (2001a)
Table 1
- Chronology of the deformed units (seismites) in the Ze'elim section. Solid dots represent 14C ages in years B.P. Error bars represent the ranges in the calibrated ages (2σ) of all samples in each stratigraphic horizon. Vertical thin lines represent historical earthquakes in the Dead Sea area, which were correlated to the deformed units in the Ze'elim section. Horizontal dashed arrows are drawn from the deformed units (listed in capital letters) to the correlative earthquakes.
- Sedimentation rates calculated for the lower part of the composite section. The longest calibrated range was used for calculating the minimum sedimentation rate, and the shortest range for calculating the maximum sedimentation rate. Two clear unconformities are evident. The upper one is dated to 1290-1420 A.D., and the lower one to 1030-1210 A.D. The lower unconformity is marked by a sharp decrease in the sedimentation rate. Vertical dashed lines represent earthquakes that lie within the sedimentological hiatuses. Sedimentation rate of the upper part of the section was not calculated because datable samples were insufficient.
click on image to open in a new tab
Ken-Tor et al. (2001a)- Table 1 Radiocarbon Table
from Ken-Tor et al. (2001b)
AMS results of 14C dating. Calibrated dates according to Stuiver et al. (1998). The samples are listed by stratigraphic height, from top to bottom. Samples collected from seismite units are shown in bold.
Click on image to open a magnifiable image in a new tab
Ken-Tor et al. (2001b)- Table 1 Radiocarbon Table
from Ken-Tor et al. (2001a)
AMS Results of Radiocarbon Dating
click on image to open in a new tab
Ken-Tor et al. (2001a)- Table 2 Refined Radiocarbon Table
from Ken-Tor et al. (2001a)
The 14C Chronology of the Deformed Layers (Seismites)
click on image to open in a new tab
Ken-Tor et al. (2001a)- Figure 4e Calibrated Radiocarbon Ages
from Ken-Tor et al. (2001b)
Chronology of the seismite record
click on image to open in a new tab
Ken-Tor et al. (2001b)- Fig. 2 Sediment Core comparisons
and Age-Depth Models for Ein Gedi, En Feshka, and Nahal Zeelim from Migowski et al. (2004)
Lithology of the sediment cores and the established age-depth models of the different profiles. The Ze'elim coring profile is paralleled by the Ze'elim gully wall. The Ein Gedi chronology is based on 20 radiocarbon dates and on the varve-counted section (black line) in the upper part.
click on image to open in a new tab
Migowski et al. (2004)- Correlated Trench Logs used
to Produce Composite ZA-1 Litholog from Revital Bookman (née Ken-Tor)
The final lithosection of Bookman (née Ken-Tor) is a composite litholog derived from multiple sites located up to approximately 300 m apart. ZA-1 refers to the location from which a large portion of the composite lithosection was constructed. More landward gullies were used to capture the most recent earthquakes, including Events G (1834 CE) and H (1927 CE). Bookman referred to site ZA-1 as Site 2.
click on image to open in a new tab
Revital Bookman (née Ken-Tor)Figure 3
Photograph of the location of [pollen] samples in the wall of gully A in and above the seismite of AD 363 in the Ze’elim site.
click on image to open in a new tab
Leroy et al. (2010)
