Timing of earthquake ruptures at the Al Harif Roman Aqueduct (Dead Sea fault, Syria) from archeoseismology and paleoseismology

Full text

243
The Geological Society of America
Special Paper 471
2010
Timing of earthquake ruptures at the Al Harif Roman aqueduct
(Dead Sea fault, Syria) from archaeoseismology and paleoseismology
Mohamed Reda Sbeinati
Department of Geology, Atomic Energy Commission, Qasr El Khair, Damascus, Syria, and Laboratory of Global Geodynamics,
Institut de Physique du Globe, UMR 7516, 5 rue René Descartes, 67084 Strasbourg, France
Mustapha Meghraoui*
Laboratory of Global Geodynamics, Institut de Physique du Globe, UMR 7516, 5 rue René Descartes, 67084 Strasbourg, France
Ghada Suleyman
Directorate General of Antiquities and Museums, Department of Archeology and Archeoseismology, Damascus, Syria
Francisco Gomez
Department of Geological Sciences, University of Missouri, Columbia, Missouri 65211, USA
Pieter Grootes
Marie-Josée Nadeau
Leibniz-Labor für Altersbestimmung und Isotopenforschung, Christian-Albrechts Universität, Max-Eyth Str. 11-13,
D-24118 Kiel, Germany
Haithem Al Najjar
Department of Geology, Atomic Energy Commission, Qasr El Khair, Damascus, Syria
Riad Al-Ghazzi
†
Higher Institute for Applied Sciences and Technology, PO Box 31983, Damascus, Syria
ABSTRACT
We studied the faulted Al Harif Roman aqueduct, located on the north–trending,
~90-km-long Missyaf segment of the Dead Sea fault, using four archaeological exca-
vations, three paleoseismic trenches, and the analysis of six tufa cores. Damage to the
aqueduct wall exhibits successive left-lateral fault offsets that amount to 13.6 ± 0.2 m
since the aqueduct construction, which is dated younger than 65 B.C. Radiocarbon
dating of sedimentary units in trenches, building cement of the aqueduct wall, and
tufa cores constrain the late Holocene aqueduct history. The building stone types,
*Corresponding author: m.meghraoui@unistra.fr.
†
Current address: Syrian Virtual University, Ministry of Higher Education Building, Damascus, Syria.
Sbeinati, M.R., Meghraoui, M., Suleyman, G., Gomez, F., Grootes, P., Nadeau, M.-J., Al Najjar, H., and Al-Ghazzi, R., 2010, Timing of earthquake ruptures at
the Al Harif Roman aqueduct (Dead Sea fault, Syria) from archaeoseismology and paleoseismology, in Sintubin, M., Stewart, I.S., Niemi, T.M., and Altunel, E.,
eds., Ancient Earthquakes: Geological Society of America Special Paper 471, p. 243–267, doi: 10.1130/2010.2471(20). For permission to copy, contact editing@
geosociety.org. © 2010 The Geological Society of America. All rights reserved.

244 Sbeinati et al.
INTRODUCTION
Large strike-slip faults are continental tectonic structures
and sources of seismic strain release during recurrent large earth-
quakes. The ~850-km-long Dead Sea fault constitutes a north-
south–trending plate boundary that accommodates most of the
left-lateral active deformation between the African (Sinai sub-
plate) and Arabia plates (Garfunkel et al., 1981; Barazangi et al.,
1993; Figs. 1A and 1B). The total left-lateral offset along the fault
reaches 105 km, of which ~45 km come from post-Miocene slip
as a result of seafl oor spreading in the Red Sea (Quennell, 1984).
However, the northern Dead Sea fault shows less than 25 km left-
lateral post-Miocene offset (Trifonov et al., 1991), the ~20 km
missing slip being possibly absorbed by shortening along the Pal-
myrides fold belt (Chaimov et al., 1990). Kinematic models of
the northern Dead Sea fault imply a transpressional fault system
that suggests an oblique relative plate motion and relative rota-
tion at ~31.1°N, 26.7°E at 0.40° ± 0.028 m.y.
–1
(Westaway, 2004;
Gomez et al., 2006). The northern section of the Dead Sea fault
(i.e., in Lebanon and Syria) can be considered among the main
seismogenic zones in the region, since it has a long (since 1365
B.C.), rich, and well-documented history of large destructive
earthquakes that severely damaged many ancient cities (Fig. 1A;
Ambraseys and Melville, 1988a; Guidoboni et al., 1994; Sbeinati
et al., 2005). In contrast, the instrumental seismicity during the
last century along the plate boundary is of low level and does not
refl ect the hazardous nature of the fault (Salamon et al., 2003).
The long-term faulting behavior needs to be investigated, and a
better constraint on the rate of active faulting is required for seis-
mic hazard assessment.
The Dead Sea fault has been the source of numerous large
earthquakes with surface faulting in historical time (Ambraseys
and Jackson, 1998). Although no recent surface ruptures have been
observed, combined analyses of historical seismology, paleo-
seismology, and archaeoseismology contributes to a better under-
standing of the relationship between large historical earthquakes
(Mw >7) and fault segments. The most recent large earthquake
reached Mw 7.2 and took place on 22 November 1995 offshore in
the Gulf of Aqaba at the southern end of the Dead Sea fault. His-
torical earthquake-faulting–related studies include, from north to
south (Fig. 1A), the 1408 earthquake and Jisr-Al-Shuggur fault
(Ambraseys and Melville, 1988a), the 1157 and 1170 earthquakes
and Apame and Missyaf fault segments, respectively (Meghraoui
et al., 2003; Sbeinati et al., 2005), the 1202 earthquake and the
Yammouneh fault (Ambraseys and Melville, 1988b; Ellenblum
et al., 1998; Daeron et al., 2005, 2007), the 1759 earthquake
sequence and the Serghaya-Rachaya fault branches (Gomez
et al., 2003; Nemer et al., 2008), the 1837 earthquake and the
Roum fault branch of the Lebanese restraining band (Nemer and
Meghraoui, 2006), the A.D. 749 earthquake and the Jordan Val-
ley fault (Marco et al., 2003; Ferry et al., 2007), and the 1068
earthquake and south Araba Valley fault (Zilberman et al., 2005).
The study of historical seismic events of the Dead Sea fault
and related area of maximum damage is associated with inves-
tigations into the possible extent of surface ruptures and related
major geometrical barriers. Earthquake parameters that include
individual or cumulative left-lateral offsets and rate of slip can be
obtained from paleoearthquake studies along the Dead Sea fault.
Numerous fault slip rates have been inferred from offset geologi-
cal units and geomorphologic features along the Dead Sea fault,
and the more recent investigations including stream offsets and
paleoseismic studies yield 4–7 mm/yr measured at time scales of
~10–100 k.y. (Garfunkel et al., 1981; Ginat et al., 1998; Klinger
et al., 2000; Niemi et al., 2001; Daeron et al., 2004; Gomez et al.,
2007; Ferry et al., 2007; Karabacak et al., 2010) and younger
than 10 ka (Marco et al., 2005; Meghraoui et al., 2003; Gomez
et al., 2003; Akyuz et al., 2006). Although an accurate measure-
ment of the present-day active deformation across the Dead Sea
fault requires a dense geodetic network combined with consis-
tent block models, the 3–6 mm/yr global positioning system
(GPS) velocities appear to be comparable to the geologic rate of
slip (McClusky et al., 2003; Wdowinski et al., 2004; Reilinger
et al., 2006; Gomez et al., 2007; Le Beon et al., 2008; Alchalbi
et al., 2009).
The Dead Sea fault crosses regions with abundant archaeo-
logical sites that evidence records of direct (fault offsets) or indi-
rect (damage to building) coseismic features. Previous studies of
archaeological sites from fi eld investigations or textual documents
related cement dating, and tufa deposits of the aqueduct indicate two reconstruction-
repair episodes in A.D. 340 ± 20 and A.D. 720 ± 20. The combined analysis of trench
results; successive building and repair of aqueduct wall; and tufa onsets, growths,
and interruptions suggests the occurrence of four faulting events in the last ~3500 yr,
with a cluster of three events in A.D. 160–510, A.D. 625–690, and A.D. 1010–1210,
the latter being correlated with the 29 June 1170 large earthquake. Our study pro-
vides the timing of late Holocene earthquakes and infers a lower and upper bound
of 4.9–6.3 mm/yr slip rate along the Missyaf segment of the Dead Sea fault in Syria.
The inferred successive faulting events, fault segment length, and related amount of
coseismic slip yield M
w
= 7.3–7.5 for individual earthquakes. The identifi cation of the
temporal cluster of large seismic events suggests periods of seismic quiescence reach-
ing 1700 yr along the Missyaf fault segment.

Timing of earthquake ruptures at the Al Harif Roman aqueduct (Dead Sea fault, Syria) 245
have revealed the occurrence of “earthquake storms” probably
associated with the Dead Sea fault (Nur and Cline, 2000). Indi-
rect earthquake features are, however, very often problematic,
and, unless dedicated to the specifi c study of known historical
earthquake damage (Stiros and Jones, 1996; Marco, 2008), most
of archaeological reports can hardly provide usable earthquake
parameters (Ambraseys, 2006). Recent studies that combine
archaeoseismic excavations and paleoseismic trenching provide
some constraints of the left-lateral strike-slip movements and
related past earthquake events. The Jordan fault segment and
related past earthquake ruptures offset the Vadum Jacob Crusader
Castle and Holocene deposits visible in trenches at Beyt Zayda
near the Sea of Galilee, yielding 3–4 mm/yr slip rate (Ellenblum
et al., 1998; Marco et al., 2005). North of our study area, archaeo-
logical sites are widespread in the Amik Basin, where the fault
crosses the ~5000 B.C. Tell Sicantarla and reveals 42.4 ± 1.5 m
cumulative left-lateral movement, thus yielding 6.0 ± 0.2 mm/yr
slip rate (Altunel et al., 2009).
Previous archaeo-paleoseismic work on the faulted Al Harif
Roman aqueduct revealed 13.6 m left-lateral offset and 6.9 ±
0.1 mm/yr slip rate that result from at least three earthquakes
(Meghraoui et al., 2003). The early aqueduct study and trench
Figure 1. (on this and following page). (A) Seismic-
ity (historical before 1900 and instrumental until
2004) along the Dead Sea fault (data from merged
ISC [International Seismological Centre], EMSC
[Euro-Mediterranean Seismological Centre], and
the APAME [Archeo-PAleoseismology in the
Middle East] Project catalogues). Focal mechanism
solutions are from Harvard centroid moment tensor
(CMT) catalogue.

246 Sbeinati et al.
A shed light on the relationships between the Roman building
and repeated faulting events but left open questions on (1) the
earthquake events scenarios and related reconstruction and repair
of the aqueduct after each coseismic slip; (2) the estimated long-
term averaged slip rate versus a temporal cluster of seismic events
over the past 2000 yr and its comparison to the present-day geo-
detic rate; (3) the constraint of the ~800-yr-long temporal and
spatial seismic gap on the Missyaf segment and the recurrence
interval of large earthquakes along the northern Dead Sea fault.
In addition to the paleoseismic trenching and archaeoseismic
excavations, tufa accumulation since the aqueduct construction
may constitute a real archive of the aqueduct history that records
the successive earthquake damages. Our study infers that the
aqueduct remains, the related tufa deposits, and faulted Holocene
sedimentary units contain comparable records of the most recent
surface rupturing events along the Missyaf fault segment.
In this paper, we present the study of the faulted Al Harif
Aqueduct site using archaeological excavations and paleoseismic
trenching across the fault zone coupled with total station surveys
and the coring of tufa accumulation on the aqueduct walls. We
fi rst describe the geomorphologic features and clear late Qua-
ternary active tectonics of the fault zone that belong to the Mis-
syaf segment of the Dead Sea fault. Archaeological excavations
of the aqueduct walls and bridge combined with three trenches
dug across the nearby fault zone and related radiocarbon dating
illustrate the timing of successive faulting episodes. The dated
Figure 1. (continued). (B) Fault zone (black line;
Meghraoui et al., 2003; Gomez et al., 2003; Elias,
2006; Nemer et al., 2008) and global positioning
system (GPS) velocities (Eurasia fi xed; Reilinger
et al., 2006; Gomez et al., 2007; Le Beon et al.,
2008; Alchalbi et al., 2009) emphasizing the left-
lateral movements between the Sinai block and
Arabia plate. Thick line is strike-slip fault; thin
line is thrust fault. DSF—Dead Sea Fault, EAF—
East Anatolian Fault, KF—Karasu Fault.

Timing of earthquake ruptures at the Al Harif Roman aqueduct (Dead Sea fault, Syria) 247
onset, major discontinuities, and interruptions of tufa cores are
correlated to the faulting events. The analysis and interpretations
of earthquake damage with probable rebuilding phase of aque-
duct wall and coseismic ruptures constrain the timing of succes-
sive faulting events. The Holocene faulting activity and related
seismic cycle of the Missyaf fault segment reveal the long-term
seismic strain release and hence determine the potential for a
future large earthquake along the Dead Sea fault.
ACTIVE FAULTING AND SEISMOTECTONIC
SETTING OF THE MISSYAF SEGMENT
The Missyaf fault segment is a section of the northern Dead
Sea fault located in the western coastal part of Syria (Figs. 1B,
2, and 3). The ~200-km-long northern Dead Sea fault is made of
(1) a 90 ± 10-km-long linear fault zone, i.e., the Missyaf segment,
limited by the Al Boqueaa and the Lebanese restraining bend to
the south and the Ghab pull-apart basin to the north, and (2) the
~10-km-wide Ghab pull-apart basin and related fault accompa-
nied in its northern termination by a complex system of fault
branches when reaching the Amik Basin and Karasu Valley in
Turkey (Figs. 2 and 3).
The Missyaf fault segment has a nearly north-south linear
trend that limits the coastal ranges to the west from the Mesozoic-
Cenozoic plateau to the east (Fig. 3). Near Al Boqueaa Basin, the
fault affects the Neogene basaltic formation of the Sheen Moun-
tains. Further north, the fault crosses the Neogene volcanic and
sedimentary formations and shows ~10–50-m-wide gouge zone,
Figure 2. Major historical earthquakes (white dots)
and areas of maximum damage (shaded) for the
29 June 1170 earthquake (Io = IX using EMS98
intensity defi nition of Grünthal, 1998) as recorded
along the northern Dead Sea fault (local intensities
are from Guidoboni et al., 2004b; Sbeinati et al.,
2005). The shaded area of maximum damage (Io
= IX) for the A.D. 1170 large earthquake is along
the Missyaf fault segment and Ghab Basin (see
also Figs. 1B and 3 for legend) and overlaps with
the maximum intensity VIII (MSK, black dashed
line) as drawn by Ambraseys (2009). DSF—Dead
Sea Fault, EAF—East Anatolian Fault, KF—
Karasu Fault.

248 Sbeinati et al.
breccias, and rupture planes that affect the Mesozoic limestone
of the coastal ranges. From Missyaf city to the Ghab Basin area,
the fault bounds the Mesozoic limestone mountain range to the
west from the Quaternary basins and Mesozoic-Cenozoic Aleppo
plateau to the east (Dubertret, 1955).
The left-lateral fault exhibits a clear geomorphologic
expression along strike and neotectonic features consistent with
the structural characteristics of the Al Boqueaa and Ghab pull-
apart basins (Fig. 3). The left-lateral movements are indicated
by the en-echelon right stepping fault strands, faulted alluvial
fans, defl ected large and small streams that fl ow from the west-
ern mountain range, and shutter ridges made of either volcanic
or limestone units (Figs. 3 and 4). The left-lateral slip is also
expressed in outcrops where fault breccias in limestones also
display en-echelon structures. Estimated from main channel
defl ections, observed in aerial photographs or satellite images, or
measured using total station, systematic left-lateral offsets visible
at different scales range from as low as 9 ± 1 m along strike to a
few hundred meters (Figs. 3 and 4).
The instrumental seismicity along the Missyaf fault segment
is scarce in comparison with that of the Lebanese restraining
bend to the south or the Karasu Valley and junction with the East
Anatolian fault to the north (Figs. 1A and 1B). Although the fault
zone corresponds to the Africa-Arabia plate boundary, the instru-
mental seismicity is low level, and magnitudes (M
l
) are less than
4.5. Focal mechanisms (CMT Harvard) of the few events with
Mw > 4.5 in the northern Dead Sea fault show strike-slip faulting
with predominant north-south–trending left-lateral fault plane
mixed with normal faulting solutions near the pull-apart basins.
The historical seismicity along the northern Dead Sea fault
reports the occurrence of large and destructive earthquakes in
1365 B.C., A.D. 115, 526, 859, 1063, 1139, 1156, 1170, and
1408 (Ambraseys and Melville, 1988a; Guidoboni et al., 2004a,
2004b; Sbeinati et al., 2005). Only a few historical contempora-
neous manuscripts, however, account for accurate damage distri-
bution and sometimes for coseismic surface breaks with enough
details that allow the correlation with fault segments (Ambra-
seys and Melville, 1988b). Most contemporaneous manuscripts
and inscriptions from Byzantine, Crusader, and Arabic sources
provide accurate damage descriptions of castles, churches and
mosques, villages, and cities, often accompanied with an esti-
mate of casualties. Based on our work on the catalogue of
historical earthquakes of Syria and paleo-archaeoseismic inves-
tigations (Mouty and Sbeinati, 1988; Meghraoui et al., 2003;
Figure 3. The 90-km-long Missyaf fault segment
and the Al Harif Roman aqueduct site. The back-
ground topography (SRTM 30 arc posting digital
elevation model; Farr and Kobrick, 2000) clearly
delineates the fault segment (arrowheads) in be-
tween the Ghab and Al Boqueaa pull-apart basins.
The Roman aqueduct at Al Harif (see also Fig. 4)
was designed to bring freshwater from western
ranges to Apamea and Shaizar. LRB—Lebanese
restraining bend.

Timing of earthquake ruptures at the Al Harif Roman aqueduct (Dead Sea fault, Syria) 249
Sbeinati et al., 2005), we have shown that the damage distri-
bution associated with the 29 June 1170 earthquake suggests
a correlation with the Missyaf fault segment. Using numerous
historical documents that report the 1170 earthquake damage,
Guidoboni et al. (2004b) provided a comparable damage dis-
tribution and suggested an epicentral location on the Missyaf
fault. However, overestimated damages at Aleppo (from misin-
terpretations of the Arabic chronicler Ibn Al Athir [1160–1233])
and poorly constrained seismotectonics inferences brought the
authors to the erroneous conclusion that the 90 ± 10-km-long
Missyaf fault segment alone could not have generated the Mw
1170 > 7 for the seismic event (Guidoboni et al., 2004b). Zones
of maximum damage should be identifi ed primarily from con-
temporary eyewitness accounts in manuscripts, corroborated by
present-day fi eld investigations on the active fault and damage of
ancient buildings. Furthermore, the geometrical structures (i.e.,
the Al Boqueaa pull-apart basin and Lebanese restraining bend to
the south and the Ghab pull-apart basin to the north) that limit the
fault segment are major obstacles to a coseismic rupture propa-
gation and, hence, constrain the earthquake fault dimension.
ARCHAEOSEISMOLOGY AND PALEOSEISMOLOGY
Site Description
The Al-Harif aqueduct is located ~4 km north of the city
of Missyaf, immediately west of a limestone shutter ridge and
related ~200 m left-lateral stream defl ection (Figs. 3 and 4).
According to the remaining aqueduct walls and related mills
in the region, the aqueduct was built during the Roman time
(younger than 65 B.C. in the Middle East) to drain freshwater
collected from springs of the western mountain range to the east-
ern semiarid plains. The remaining ruins of the aqueduct sug-
gest an ~40-km-long construction that may have included several
bridges over streams and landscape gorges.
The aqueduct building description and related age have not
been reported so far in any archive, manuscript, or in the litera-
ture. There is, however, an interesting anecdotal story from the
local tradition that it was built by a local prince to supply potable
water to Apamea and/or Sheizar cities, located northeast of the
aqueduct (Fig. 3). Apamea during that time was the most famous
and strategic city during the Hellenistic and Roman period,
whereas Sheizar is known to have been an important political
and military fortress during the Middle Ages (Ibn Al Athir, 1982).
In their description of the Dead Sea fault in Syria, Trifonov
et al. (1991) mentioned the existence of a faulted aqueduct near
the city of Missyaf, but neither the precise location nor the accu-
rate amount of offset walls was given. However, this early tec-
tonic observation was helpful and allowed us to discover the site
and consider a detailed study (Meghraoui et al., 2003), which is
extended here using combined methods in archaeoseismology,
paleoseismology, and tufa investigations. In addition, a micro-
topographic survey of measurements accompanied all fi eld stud-
ies (Fig. 5).
Previous investigations on the aqueduct (Meghraoui et al.,
2003) established: (1) an evaluation of its age based on an account
of the large size blocks, the dating of sedimentary units below the
aqueduct wall foundation, and dating of early tufa deposits on
Figure 4. (A) Satellite view from Google
Earth showing offset Al Harif aqueduct
(black arrow) along the Dead Sea fault
(white arrows); and (B) local geomor-
phologic framework of the aqueduct site
as interpreted from part A, indicating a
shutter ridge (Mesozoic limestone east
of the fault) and ~200 m of left-lateral
offset. Blue arrow is for stream fl ow.
See Figure 5 for the detailed aqueduct
map and location of excavations and
trenches.

250 Sbeinati et al.
the aqueduct wall, and (2) the identifi cation of the seismic fault-
ing origin of damage in nearby trench A. The building style, with
typical bridge arch and large stone size disposition (Opus cae-
mentum), suggested a Roman age, which was confi rmed by the
radiocarbon dating of sedimentary layers below the walls and the
early tufa deposits on the walls. The faulted aqueduct revealed
13.6 ± 0.20 m of total left-lateral offset and called for detailed
investigations on the characteristics and history of successive
fault movements.
The aqueduct design, with an open canal on top of the
4-m-high wall, allowed freshwater and carbonate-saturated water
to overfl ow and induce signifi cant tufa accumulation from 0.30 m
to 0.83 m in section (Figs. 6A, 6B, and 7). The carbonate-rich
and cool water collected from the nearby western range is associ-
ated with a semiarid and karstic area of the Mesozoic limestone
(Fig. 4) that favors rapid carbonate precipitation and tufa accumu-
lation. The tufa deposits show successive growths of lamination
carbonate with high porosity, banded texture, and rich organic
encrustations (Ford and Pedley, 1996). Field observations show
that tufa accumulation developed on both eastern and western
sections (from the fault line), but only on the north-facing wall,
likely due to a slight tilt of the damaged aqueduct wall, probably
after the two fi rst earthquakes (Fig. 7).
The following paragraphs present the fi eld investigations,
which consisted of: (1) four archaeoseismic excavations near the
aqueduct walls and remains, (2) four paleoseismic trenches across
the fault zone and the alluvial sediments, and (3) four cores (two
cores were previously studied in Meghraoui et al., 2003) of tufa
deposits collected from different sections of the aqueduct. More
than 200 samples of organic matter, charcoal fragments, and tufa
core pieces were taken for radiocarbon analysis in order to char-
acterize the timing of successive faulting and related damage of
the aqueduct construction. All radiocarbon dating were calibrated
(2σ range, 95.4% probability density) using Oxcal v4.0 (Bronk
Ramsey, 2001) and INTCAL04 calibration curve of Reimer et al.
(2004).
Figure 5. Microtopographic survey (0.05 m
contour lines) of the Al-Harif aqueduct and
related fl at alluvial terrace. The aqueduct
(thin blue crosses) shows a total of 13.6 ±
0.20 m left-lateral slip along the fault zone
(Meghraoui et al., 2003). Roman numbers
indicate archaeoseismic excavations (in red-
dish and orange, labeled I to IV) and letters
indicate paleoseismic trenches (in gray and
black, labeled A, B, C, and E). The dragged
wall fragment is located between excavation
IV and trench E and is marked by a dense
cluster of survey points.

Timing of earthquake ruptures at the Al Harif Roman aqueduct (Dead Sea fault, Syria) 251
Archaeoseismic Excavations
The remaining aqueduct construction forms an ~50-m-long,
~5-m-high, and 0.60-m-thick wall that includes an ~15-m-high
arch bridge in its eastern section (Figs. 5 and 6A). The outer part
is coated by a thick layer of tufa deposits, probably due to a long
period of freshwater fl ow. The construction material that may vary
with the successive building and repair ages is made of: (1) large-
size limestone blocks (Opus quadratum, 1.0 m × 0.5 m × 0.5 m;
see also http://www.romanaqueducst.info/aquasite/), similar to
the typical Roman archaeological constructions and visible at the
lower bridge (pier section) and wall sections, (2) medium-size
limestone blocks (Opus incertum; 0.50 m × 0.30 m × 0.30 m),
which form the foundation or the upper half wall section and
show visible small portions of cement, and (3) small sizes of
mixed stones of irregular shape with signifi cant portions of mor-
tar (cement), mostly visible in the apparently rebuilt part of the
wall. Figures 5 and 6A also show a detached small piece of the
aqueduct wall made of small-size stones and related cement
~3.5 m away from the eastern wall. Therefore, four areas (noted
I to IV in Fig. 5) were excavated near the aqueduct using proper
archaeological methods.
The large excavation I was dug on the fault zone near the
dragged wall fragment, in the area between the eastern and west-
ern aqueduct walls (Figs. 5, 8A, and 8B). The purpose of exca-
vation is here to study the relationships between the fault zone
and aqueduct. The excavation that has ~4.5 × 4.5 m surface and
~0.6 m depth exposed missed parts of the aqueduct. A buried
and fallen wall piece rotated and dragged parallel to the fault
and a remaining wall piece in an oblique position between two
shear zones were discovered. The buried wall fragments are not
Figure 6. (A) Schematic sketch of the aqueduct and loca-
tions of the selected cores BR-3, BR-5, and BR-6; BR-4
core sample consists of tufa accumulations at the location
of the missing (broken) piece of the aqueduct wall near
the fault. Mosaic of the archaeological excavation I is de-
tailed in Figure 8B (see also location in Fig. 5). (B) Core
section BR-4 showing the limit between the stone wall
and tufa deposits.

252 Sbeinati et al.
comparable to the Opus caementum (quadratum) of the original
construction and suggest a rebuilding phase. The excavation fl oor
displays oriented gravels and pebbles that mark the shear zones
and related fault branches also visible in the inner trench sec-
tion E (Figs. 8B and 8C).
We collected four samples in the fallen wall sections labeled
A, B, and C of excavation I (Fig. 8B): Two cement samples
(AQ-CS-1 and AQ-CS-4) found in between building stones are
made of typical medieval rubble mortars (mainly mud, gypsum,
and lime); the two other samples (AQ-CS-3-2 and AQ-CS-3-3)
are tufa deposits preserved on building stones. All four samples
contained enough organic matter to allow radiocarbon dating
(Table 1). Two dates of cement yield A.D. 532–641 (section A,
AQ-CS-4) for the large fallen wall in excavation I and A.D. 650–
780 (section C, AQ-CS-1) for the wall fragment piece in between
the walls (Fig. 8B). In addition, two tufa deposits on wall stones
provide consistent ages A.D. 560–690 (section B, AQ-CS-3-2)
and A.D. 639–883 (section C, AQ-CS-3-3) with cement ages.
The two different cement dates of the fallen wall and dragged
Figure 7. Schematic sections of the aqueduct western wall and related tufa deposits (B, C, D, and E indicate earlier core sections of tufa
deposits (Meghraoui et al., 2003). Tufa samples AQ-Tr-B13 and AQ-Tr-D5 (Table 1) are from cores B and D, respectively. The right
and left vertical sections show the relative tufa thickness of the originally built part (with Opus caementum and quadratum stones) and
the rebuilt part, respectively. The plan view indicates the variation of tufa deposition and shows the core distribution and related thick-
ness along the western wall of the aqueduct.
Figure 8. (on this and following page). (A) View from the western
aqueduct wall, the dragged wall piece, buried wall, and eastern wall
(string grid is 1 m × 1 m). Log of trench-excavation E is in Figure 7C.

Figure 8. (continued). (B) Mosaic of
excavation I exhibits the main fallen
wall (A and B) and dragged wall piece
(C), scattered wall pieces and the fault
zone; note also location of cement
sample CS-1-4 (see text for explana-
tion). (C) Trench E (excavation I, north
wall) exposes faulted sedimentary units
below the archaeological remains and
wall fragment C visible in bottom of
Figure 8B; fz—fault zone; sedimentary
units are similar to those of trenches A,
B, and C (see also Fig. 10); and dat-
ing characteristics are in Table 1. a—
present-day soil and alluvial terrace
(plough zone), d—reddish alluvial fi ne
gravel, e—dark-brown silty clay (with
rich organic matter), f—gravels and
pebbles in silty-clay matrix, g—massive
gey clay with scattered gravels.

TABLE 1. SAMPLE LIST AND RADIOCARBON DATING (ACCELEROMETER MASS SPECTROMETRY) AT THE AL HARIF AQUEDUCT SITE
Sample ID Sample name Fraction Trench-excav. unit &
core level
Analyzed
carbon (mg)
Carbon
(content %)
14
C date
(yr B.P.)
14
C age
Calibrated B.C./A.D. (95.4%)
Al Harif—Trenches
KIA 14261 BAL-TA-N23 Charcoal, alkali residue b & c (A) 5.87 71.9% 1015 ± 35 A.D. 960–1060
KIA 14263 BAL-TA-N27 Charcoal, alkali residue f (A) 4.09 67.4% 2335 ± 30 B.C. 520–350
KIA 14262 BAL-TA-N25 Charcoal, alkali residue e (A) 0.42 24.9% 2090 ± 50 B.C. 350–30 A.D.
AA 43995 EH-I-S7 Charcoal, alkali residue e (A) – – 2195 ± 40 B.C. 390–160
AA 43993 EH-I-TA-S33 Charcoal, alkali residue d (A) 4.33 – 1287 ± 36 A.D. 650–810
KIA 14264 BA-TA-N31 Charcoal, alkali residue a (A) 1.95 63.3% 875 ± 35 A.D. 1030–1250
KIA 14268 BAL-TN-61 Charcoal, alkali residue a (A) 1.02 1.0% 4555 ± 40 B.C. 3490–3090
KIA 14265 BAL-TA-N47 Charcoal, alkali residue g (A) 1.71 3.6% 7410 ± 45 B.C. 6400–6100
KIA 23856 AQ-TA-3 Charcoal, alkali residue e (II) 1.83 22.6% 2295 ± 30 B.C. 410–210
KIA 23855 AQ-TA-4 Charcoal, alkali residue e (II) 0.24 4.7% 2050 ± 70 B.C. 350–130 A.D.
KIA 23861 AQ-TB-1 Charcoal, alkali residue e (III) 1.77 22.4% 2250 ± 30 B.C. 400–200
KIA 23862 AQ-TB-2 Charcoal, alkali residue e (III) 1.64 25.9% 2200 ± 40 B.C. 390–160
KIA 23863 AQ-TB-3 Charcoal, alkali residue e (III) 4.23 49.8% 2235 ± 30 B.C. 390–200
KIA 23857 AQ-TB-4 Charcoal, alkali residue e (III) 0.38 4.0% 2460 ± 60 B.C. 770–400
KIA 23858 AQ-TC-S1 Charcoal, alkali residue
e (I–E)
0.17 2.1% 1930 ± 110 B.C. 200–400 A.D.
KIA 23859 AQ-TC-S2 Charcoal, alkali residue
f (I–E)
0.13 1.6% 2450 ± 140 B.C. 900–200
KIA 23860 AQ-TC-S3 Charcoal, alkali residue
f (I–E)
0.28 9.5% 2280 ± 70 B.C. 550–100
KIA 23903 EH II-7S Charcoal, acid residue a (C) 0.49 128.9% 290 ± 40 A.D. 1480
–
1800
KIA 23880 EH II-2N Charcoal, alkali residue a (C) 4.16 50.7% 280 ± 25 A.D. 1510
–
1670
KIA 23917 EH II-16S Charcoal, alkali residue b1 (C) 5.62 69.0% 1465 ± 30 A.D. 540–650
KIA 23911 EH II-11S Charcoal, acid residue f (C) 2.11 51.7% 2135 ± 30 B.C. 360–50
KIA 23915 EH II-10S Charcoal, alkali residue f (C) 1.54 57.5% 2150 ± 30 B.C. 360–60
KIA 23910 EH II-12S Charcoal, alkali residue f (C) 4.66 62.3% 2160 ± 30 B.C. 360–90
KIA 23920 EH II-18S Charcoal, acid residue f (C) 0.97 9.3% 2525 ± 40 B.C. 800
–
510
KIA 23909 EH II-5S Seed, alkali residue f (C) 0.03 4.5% 3420 ±570 B.C. 3400–300
KIA 23895 EH III-8S Charcoal, acid residue e (B) 2.40 38.4% 2110 ± 35 B.C. 350–40
KIA 23896 EH III-7S Charcoal, acid residue e (B) 1.19 45.2% 2215 ± 35 B.C. 410–90
KIA 23897 EH III-6S Charcoal, acid residue Fault zone (B) 0.24 6.3% 2390 ± 80 B.C. 800–200
KIA 23900 EH III-3S Charcoal, acid residue Fault zone (B) 1.05 64.0% 4375 ± 40 B.C. 3100
–
2980
KIA 23893 EH III-10S Charcoal, alkali residue d? (B) 0.12 5.5% 2680 ± 170 B.C. 1300–350
Al Harif—Cores & cement
KIA 16627 AQ-Tr-D5 Tufa, acid residue D, 0–5 cm 1.13 80.1% 1863 ± 29 A.D. 80
–
240
KIA 16628 AQ-Tr-B13 Tufa, acid residue B, 0–5 cm 2.39 2.7% 2030 ± 25 B.C. 110–60 A.D.
KIA 16628 AQ-Tr-B13 Tufa, humic acids B, 0–5 cm 3.72 41.3% 1880 ± 25 A.D. 70–230
KIA 22189 AQ-CS-1 Cement, alkali residue I – Wall piece 0.50 0.4% 1314 ± 37 A.D. 650–780
KIA 22191 AQ-CS-3-2 Tufa, alkali residue I – Fallen wall 0.99 2.9% 1400 ± 35 A.D. 560–690
(Continued
)

Timing of earthquake ruptures at the Al Harif Roman aqueduct (Dead Sea fault, Syria) 255
wall fragment can be correlated to the new tufa deposits that tes-
tify for two rebuilding phases. The dated buried fallen wall in
section B (CS-3-2) obtained from a thin (~5 cm) tufa accumu-
lation correlates with the similarly fallen wall in section A and
related cement date of CS-4 (Fig. 8B; and Table 1). In section C,
the tufa deposits and related dated sample CS-3-3 correlates with
the cement age of CS-1. The type and size of stones (opus incer-
tum) and thin tufa accumulation in sections A and B suggest an
early rebuilding phase postdating the fi rst damaging event that
may have occurred between the fi rst and sixth centuries A.D.
The different building layout of section C made of small sizes
of mixed stones of irregular shape, and dating of cement sample
CS-1 (A.D. 650–780) and tufa accumulation CS-3-3 (A.D. 639–
883) indicate a repair and rebuilding period postdating a second
damaging event at the end of the Byzantine time and beginning
of the Islamic period (seventh to eighth century A.D.). The dam-
aged and dragged most recent wall section C along the fault indi-
cates the occurrence of a third event, after which the aqueduct
was defi nitely abandoned.
Three small excavations II, III, and IV (1.5 m to 3.0 m
long, 1.0 m wide and 1.50 m deep) were dug in the base layer
of the western aqueduct wall in order to expose its foundation
and related sedimentary units underneath that predate the early
building phase (Figs. 5 and 9). Excavations II and III were dug
under the wall section with maximum (> 0.80 m) and minimum
(~0.30 m) thickness of tufa deposition, respectively (Figs. 9A and
9B). Excavation IV, already described in Meghraoui et al. (2003),
exposed the faulted foundation of the missing section of the west-
ern wall edge. The wall foundation reaches 1 m depth and shows
regular patterns of medium-size cut limestone blocks (0.50 m ×
0.30 m × 0.30 m) built over a dark brown clayey layer (unit e).
Charcoal samples collected in excavations I (trench E),
II, and III from unit e yield
14
C dates with ages spanning from
approximately the third century B.C. to third century A.D. (see
samples AQ-TA, TB, and TC in Table 1; Figs. 8C, 9A, and 9B).
Although in these excavations, the age range of unit e seems quite
large (probably due to detrital charcoal mixing), the younger
age, i.e., 350 B.C. to A.D. 130 (sample AQ-TA-4), is consistent
with other radiocarbon ages of unit e and related stratigraphic
succession in trenches (see Paleoseismic Trenches herein). In
excavation II, the large stone shape (Opus quadratum) with small
amount of cementing material and pottery fragments found on
the same level near the building base can be correlated with the
early Roman era (Fig. 9A). Large stones and tufa thickness led
us to consider this section of the aqueduct wall to be in original
condition, i.e., probably undamaged by large earthquakes.
Excavation III (1.65 m long, 1.0 m wide, and 1.2 m deep;
Fig. 9B) is similar to excavations II and IV, but the 1-m-deep
wall foundation and upper section show irregular shapes of
mixed medium- and small-size cut limestone blocks (0.10 × 0.20
× 0.15 m). Excavation III was realized at the location of the thin-
nest tufa deposits (< 0.30 m). The size of stones, cement texture,
and irregular shape of building wall suggest that this building
section was rebuilt (Fig. 9B). The
14
C dating of unit e below the
TABLE 1. SAMPLE LIST AND RADIOCARBON DATING (ACCELEROMETER MASS SPECTROMETRY) AT THE AL HARIF AQUEDUCT SITE
Sample ID Sample name Fraction Trench-excav. unit &
core level
Analyzed
carbon (mg)
Carbon
(content %)
14
C date
(yr B.P.)
14
C age
Calibrated B.C./A.D. (95.4%)
KIA 22191 AQ-CS-3-3 Tufa, alkali residue I – Fallen wall 0.57 1.9% 1283 ± 44 A.D. 639–883
KIA 22192 AQ-CS-4 Cement, alkali residue I – Fallen wall 5.33 0.3% 1497 ± 24 A.D. 532–641
KIA 26056 BR-3-1/SYR Al Harif Tufa, alkali residue 3-1, 0
–
0.5 cm 0.53 17.3% 1570 ± 35 A.D. 410
–
600
KIA 26056 BR-3-4/SYR Al Harif Tufa, alkali residue 3-4, 32.5–33 cm 2.28 2.3% 1180 ± 20 A.D. 770–940
KIA 26057 BR-4-1/SYR Al Harif Tufa, alkali residue 4-1, 10.5–11 cm 0.71 6.7% 1465 ± 35 A.D. 530–660
KIA 26057 BR-4-3/SYR Al Harif Tufa, alkali residue 4-3, 23.5–24.5 cm 0.14 22.6% 1310 ± 110 A.D. 540–980
KIA 26058 BR-5-2/SYR Al Harif Tufa, alkali residue 5-2, 4.5–5.0 cm 0.51 1.7% 1995 ± 45 B.C. 110–130 A.D.
KIA 26058 BR-5-7/SYR Al Harif Tufa, humic acids 5-7, 32
–
33.5 cm 3.44 40.5% 1090 ± 25 A.D. 890–1020
KIA 26059 BR-6-1/SYR Al Harif Tufa, alkali residue 6-1, 0–0.5 cm 0.15 1.1% 2020 ± 110 B.C. 400
–
250 A.D.
KIA 26059 BR-6-8/SYR Al Harif Tufa, alkali residue 6-8, 38–39 cm 0.58 1.2% 1020 ± 35 A.D. 900–1160
Notes: All samples have been calibrated using the Oxcal program v. 3.5 (Bronk-Ramsey, 2001) and calibration curve INTCAL04 (Reimer et al., 2004), and adopted age ranges are
equivalent to calibrated 2σ ranges (95.4%), in A.D. and B.C. Trench and excavation units from trenches A, B, and C and excavations I, I–E, II, and III appear in parentheses. Location
of cores B and D is in Figure 7.

256 Sbeinati et al.
wall yielded a comparable age range to that obtained in excava-
tions II (see AQ-TA, TB, and TC in Table 1).
Trench section E (4.30 m long, 0.70 m wide, and 1.30 m
deep; Figs. 5, 8A, 8B, and 8C) was dug within excavation I in
order to see in section the fault zone that affects the archaeologi-
cal fl oor units. The trench wall exposes similar sedimentary units
to those visible in excavations II, III, and IV that are affected by
two main fault branches of the shear zone visible in the fl oor
layer of excavation I. The
14
C dating of samples AQ-TC-S1, S2,
and S3 of units f and e indicates 900 B.C. to A.D. 400 maximum
and minimum age range, respectively (Fig. 8C; Table 1), which
is comparable to the age range obtained in excavations II and III
for unit e (Figs. 8B and 9A; Table 1). However, as here again the
large age range can be due to charcoal mixing, the dating of unit
e is obtained by comparison to the dated stratigraphic succession
of units in trenches (see section Paleoseismic Trenches).
Paleoseismic Trenches
Two trenches, B and C (Figs. 5 and 10, trenches B and C),
were dug across the Dead Sea fault north of the aqueduct in addi-
tion to the previously studied trench, A (Fig. 10A; Meghraoui
et al., 2003). The two trenches exposed an ~1.5-m-wide fault
zone that affects a succession of 2–3-m-thick fi ne and coarse allu-
vial sedimentary layers similar to the alluvial deposits of trench
A. Alluvial units visible in all trenches exhibit here similar tex-
tures, structures, and color, and correspond to the same layers that
belong to the same alluvial terrace. Although the three trenches A,
B, and C may not expose a completed stratigraphic section, the
comparisons among sedimentary units, faulting events, archaeo-
seismic observations, and tufa accumulation limit the possibility
of a missing earthquake event that affected the aqueduct.
In trench B (south wall), the fault zone shows three main
fault branches that affect sedimentary units g to d and form
a negative fl ower structure. The central and western main
branches are truncated by unit a, which forms a stratifi ed
0.3–0.4-m-thick deposit of coarse gravels in a sandy matrix. The
eastern fault branch is buried below unit d, made of well-sorted
reddish fi ne gravels. Unit e, a 0.2–0.5-m-thick dark-brown silt-
clay, thickens toward east. Units f and g are made of scattered
clasts in a massive clay matrix of dark-brown and light-brown
color, respectively. Although intense warping and faulting are
marked by contrasting color and texture of unit e, faulted sedi-
mentary layers of this trench do not allow the identifi cation of
all faulting events. However, buried fault branches indicate a
faulting event postdated by unit d (event Y), while the other
fault branches show at least another faulting event (event Z)
overlain by unit a. While clearly visible in other trench walls,
event Y is here likely concealed by the complex fault branches
truncated by unit a.
Trench C (Fig. 10C) exposes a stratigraphic succession
affected by at least fi ve main fault branches (labeled I to V in
Fig. 10C). From trench bottom, fault branch I, which affects unit
g, is overlain by unit f. A similar observation can be made for
fault branch II, which also affects all units below unit d. Further-
more, the trench wall exposes an ~0.60-m-thick well-stratifi ed,
coarse and fi ne gravel layer above unit e and across the fault
zone. Unit d thins signifi cantly west of fault branch III and is
overlapped by relatively thick coarse gravel units, which dis-
play a mix of fi ne and coarse gravels between fault branches III
and IV, and unit d shows a succession of well-stratifi ed alluvial
units west of fault branch IV (Fig. 10C). Taking into account
its alluvial origin made of well-stratifi ed fi ne and coarse grav-
els, west of fault branch IV, unit d is subdivided into d1, d2, d3,
Figure 9. Excavations II (A) and III
(B) that expose the aqueduct wall foun-
dation (see also Fig. 5) and related
sedimentary unit e underneath. The dif-
ference in the size of stones between
excavation II (A) and excavation III
(B) implies a rebuilding phase of the
latter wall.
AB

Figure 10. Trench logs A, B, and C north
of the aqueduct site (see location in
Fig. 5). All trenches display the Dead Sea
fault zone as a negative fl ower structure
affecting all alluvial units below unit a.
Calibrated
14
C dates are in Table 1. Fault
branches in trench C are labeled I to V
(see text for explanation). The sedimen-
tary units are very comparable and show
three to four faulting events denoted W
to Z (see text for explanation). Trench
log A is in meters.

258 Sbeinati et al.
and d4. Faulting movements at this site allows truncation of unit
d1 (equivalent to d east of fault branch III) and sedimentation of
units d2 to d4 (in a likely small pull-apart basin). Unit d3 consists
of an ~0.20-m-thick dark-brown silt-sand overlain by unit d4,
which is made of light-brown fi ne silt-sand. Below the plough
zone a2, the well-stratifi ed unit a1 shows fl at-laying pebbles and
gravels and intercalated fi ne gravels covering previous units and
the fault zone.
Fault branches I to V in trench C indicate a negative fl ower
structure that intersects a sedimentary sequence and reveal at
least four faulting events (Fig. 10C): (1) Event W, identifi ed on
fault branch I, is older than 800–510 B.C. (EH II-18S) in the low-
ermost layers of unit f and is younger than unit g, which was
dated with sample EH II-5S (3400–300 B.C.). (2) Next to fault
branch II, buried below unit d, the vertical offsets between unit e
and units d and d1 across fault branch III, and the absence of unit
e between fault branches III and IV, determine the faulting event
X between unit e and unit d. Since unit d overlies an erosional
surface of unit e, faulting event X may have formed a depression
(i.e., a small pull-apart basin) that allowed the deposition of d1 to
d4 next to a thick unit d east of fault branch III. The faulting event
X is here predated by 360–90 BC (EH II-12S), 360–50 BC (EH
II-11S), and 360–60 BC (EH II-10S) of the uppermost layers of
unit f (event X is postdated by sample EH I-TA-S33 of trench A).
(3) Faulting event Y can be identifi ed at the westernmost fault
branch V between unit d2 and unit d3. The dating of sample EH
II-16S in d3 postdates event X to younger than A.D. 540–650,
which we consider as a reliable age, taking into account its high
carbon content (event Y is predated by sample EH I-TA-S33 of
trench A). (4) Faulting event Z corresponds to the main fault
branches III and IV, which are overlain by the stratifi ed unit a2
below the plough zone. Fault rupture IV affects unit d4 and indi-
cates that the faulting event Z is older than radiocarbon age A.D.
1480–1800 (EH II-7S) and A.D. 1510–1670 (EH II-2N) of unit
a2 and younger than unit d4.
Summary of Faulting Events from Archaeoseismology
and Paleoseismology
The analysis of faulting events from the aqueduct (damage
and reconstruction) and from trenches A, B, and C can be pre-
sented as following:
1. Event W is older than unit f (i.e., 800–510 B.C.) and
younger than unit g (i.e., 3400–300 B.C.) of trench C.
The bracket of event W is here diffi cult to assess since the
detrital charcoal sample in unit f was not taken from the
base of unit f. According to
14
C dates, the faulting event
can be estimated as younger than 3400 B.C. and older
than 510 B.C. However, taking into account the rate of
sedimentation in unit f, we may estimate a minimum age
of 962 B.C. for event W.
2. Event X, the fi rst faulting event that affected the aque-
duct, is bracketed between the fi rst and sixth centuries
A.D. In trenches, a large bracket of this event is between
350 B.C. and A.D. 30 and A.D. 650–810 (as obtained
from dated units of trench A).
3. Event Y, characterized from paleoseismology, appears
to be older than A.D. 650–810 (unit d, trench A) and
younger than A.D. 540–650 (unit d3 in trench C). The
results of archaeoseismic investigations indicate that ages
of CS-1 (A.D. 650–780) and tufa accumulation CS-3-3
(A.D. 639–883) postdate event Y.
4. Event Z is the last faulting event that affected the aque-
duct, after which it was defi nitely abandoned. In trenches
A and C, event Z is older than A.D. 1480–1800, A.D.
1510–1670, and A.D. 1030–1260 and younger than A.D.
960–1060.
TUFA OF THE AL-HARIF AQUEDUCT
The tufa thickness accumulated on the northern face of
aqueduct wall suggests a continuous water fl ow during a rela-
tively long period of time and may include the record of large
earthquakes that affected the aqueduct. Hence, the relationships
between tufa accumulation and earthquake events are established
through the simultaneous major tufa interruptions and restarts
observed in different cores. Except during major changes in the
water-fl ow conditions, the permanent water fl ow coming from
the nearby spring was responsible for the tufa accumulation that,
in principle, is not interrupted on the western wall section (with
regard to the fault). On the eastern wall section (and bridge) and
broken pieces of western wall, however, the tufa accumulation
was likely episodic due to the earthquake damage and related
faulting events; new tufa accumulation appears in subsequent
building-repair. Previous radiocarbon dating of early tufa depos-
its (A.D. 70–230 and A.D. 80–240; Table 1) postdated the initial
construction of the aqueduct and revealed a Roman age consis-
tent with the dates obtained from the archaeological and paleo-
seismic investigations (Meghraoui et al., 2003).
Six tufa cores (named Tr-B13, Tr-D5, and BR-3, BR-4,
BR-5, and BR-6) reaching the stone construction were collected
from the aqueduct wall in order to date major catastrophic events
and infer the relationship with large earthquakes (Fig. 11). Tr-B13
and Tr-D5 were previously collected and analyzed mainly to date
the early tufa deposits, which provide the maximum age of the
aqueduct construction (Meghraoui et al., 2003). A subsequent
selection of core locations on both eastern and western sections
of the aqueduct wall was performed to study the completed tufa
accumulation and successive growth. Figures 6 and 7 show the
drilled wall location with the early cores Tr-B13 and Tr-D5 and
three cores (BR-4, BR-5, and BR-6) on the western wall and
one core (BR-3) on the eastern wall next to the bridge. Cores
BR-5 and BR-6 correspond to the thickest tufa section. BR-4 is
on the eastern edge of the west aqueduct wall, a section prob-
ably exposed after earthquake damage that induced the collapse
of a 2.5-m-long wall section next to the fault zone. Each core is
described to illustrate fabric (structure, texture, and color) and
lamination changes, which provide evidence of tufa precipitation

Timing of earthquake ruptures at the Al Harif Roman aqueduct (Dead Sea fault, Syria) 259
and successive growths (Fig. 11). Although marked by a high
porosity, the cores were carefully drilled in order to preserve
their structure and length continuity. An analysis in progress of
cores using computer tomography (CT) and climatic-stratigraphy
correlation details the physico-chemical and biochemical pro-
cesses of tufa growth (Grootes et al., 2006). The cores show a
variety of porous, dense, and biogenic tufa with growth laminae
and stromatolitic markers of different colors. The end of tufa
growth (i.e., very porous tufa in Fig. 11) and onset of biogenic
tufa (indicating only a seasonal growth) can be interpreted as epi-
sodes of decreased accumulation, or a signifi cant decrease in the
chemical precipitation due a major change in the environmental
conditions (Fig. 11). Discontinuities of tufa deposits marked by
the interruption of core growths and initiation of biogenic tufa
are interpreted as major changes in environment with a possible
correlation with large earthquakes. The early tufa deposits on the
aqueduct wall provide A.D. 70–230 and A.D. 80–240 (samples
Tr-B13 and Tr-D5 in Table 1) ages, which postdate the aque-
duct building and early function (Meghraoui et al., 2003). The
tufa accumulation in BR-3 (core in eastern wall near the bridge,
Fig. 6) started sometimes before A.D. 410–600 (sample Br 3-1,
Table 1) and may have resulted from a repair of the aqueduct
with water overfl owing the eastern wall (and bridge) after a
major damaging event. Similarly, the location of a growth inter-
ruption (very porous tufa, Fig. 11) in BR-5 at ~6 cm after Br-5-2
(110 B.C.–A.D. 130) and onset of biogenic tufa in BR-6 after
Br-6-1 (400 B.C.–A.D. 250) coincide with the occurrence of the
fi rst damaging event X. In parallel, the beginning of BR-4 and
tufa accumulation at the damaged eastern edge of the western
wall (Fig. 6) and sample Br-4-1, dated A.D. 530–660 (Fig. 11;
Table 1), postdates the occurrence of a major damaging event.
Both Br-3-1 and Br-4-1 postdate here the record of a major dam-
aging event that affected the aqueduct. However, while BR-4
may have accumulated only after a major damage, BR-3 deposits
could only have accumulated after the repair of the aqueduct. It
implies that the fi rst major damaging event on the aqueduct took
place between A.D. 70–230 and A.D. 410–600.
The interruption of tufa growth in BR-3 a few centimeters
before sample Br-3-4, dated A.D. 770–940, probably resulted
from a second damaging event. This observation coincides with
the restart of BR-4 after a major interruption 3–4 cm after Br-4-3,
dated at A.D. 540–980 (Fig. 11;Table 1). Furthermore, the sharp
change (second interruption) from dense tufa to biogenic tufa
in BR-5 and BR-6 may also have been contemporaneous with
the damaging event. The age of this second damaging event can
be bracketed between Br-4-3 (A.D. 540–980) and Br-3-4 (A.D.
770–940). Unless simply broken, the defi nite interruption of
BR-3 (~10 cm after sample Br-3-4) marks the end of water over-
fl ow on the eastern aqueduct wall (and bridge) after the second
damaging event.
The growth of dense tufa in BR-4 and biogenic tufa in
BR-5 and BR-6 in the fi nal sections of cores indicates a con-
tinuous water fl ow on the western aqueduct wall after the second
damaging event. The almost simultaneous arrest of tufa growth
~2 cm after Br-5-7 (A.D. 890–1020), ~1 cm after Br-6-8 (A.D.
900–1160), and ~7 cm after Br-4-3 (A.D. 540–980) suggests the
occurrence of a major damaging event. Indeed, the arrest of tufa
accumulation (in core samples Br-3-4, Br-5-7, and Br-6-8) prob-
ably occurred after A.D. 900–1160 (Br-6-8, Table 1) and indi-
cates the fi nal stoppage of water fl ow over the aqueduct.
Figure 11. Synthetic description of cores with lithologic content and sample number for radiocarbon dating (see Table 1 and Fig. 6 for
core locations); I stands for major interruption. The very porous tufa indicates major interruptions in tufa growth (e.g., a major inter-
ruption of core growth in BR-3 is visible at ~22 cm (Br-3-4 sample; see text for explanation). The correlation between major interrup-
tions of tufa growth and faulting events in trenches and archaeoseismic building constrains the timing of repeated earthquakes along
the Missyaf segment of the Dead Sea fault.

260 Sbeinati et al.
TIMING OF EARTHQUAKE FAULTING AND
CORRELATION AMONG ARCHAEOSEISMIC
EXCAVATIONS, PALEOSEISMIC TRENCHES,
AND CORES
The analysis of fi eld data in archaeoseismology, paleoseis-
mology, and tufa coring provides some constraints on the suc-
cessive past earthquakes along the Dead Sea fault at the Al Harif
Roman aqueduct site (Figs. 12 and 13). The damage and repair of
the aqueduct are here related to the total 13.6 m of left-lateral fault
offset since construction of the aqueduct (Fig. 5). In addition, the
tufa successive growth and interruptions visible in cores provide
a direct relation between the water fl ow and the aqueduct function
east and west of the fault zone. The correlation and timing coin-
cidence between the faulting events visible in trenches, aqueduct
construction damage and repair (see also Summary of Faulting
Events from Archaeoseismology and Paleoseismology section),
combined with tufa growth and interruptions, provide a better
constraint on the timing of the successive large earthquakes:
Event W, observed in trench C, occurred before 800–510
B.C. (unit f) and after 3400–300 B.C. (unit g). This faulting event
can be determined only in trench C and hence cannot be cor-
related with damaging events in the aqueduct archaeoseismic
excavations and tufa cores. However, we suggest two possible
ages for this event: (1) according to the textual inscriptions found
in different archaeological sites in Syria, a damaging earthquake
sequence around 1365 B.C. affected Ugharit near Latakia in
Syria, and Tyre further south in Lebanon and east of the Dead Sea
fault (Sbeinati et al., 2005) may be correlated to event W; or (2)
the rate of sedimentation in unit f of trench C implies a minimum
age of 962 B.C. for event W.
Event X, identifi ed in trenches A and C between 350 B.C.–
A.D. 30 and A.D. 532–641, postdates the construction of the
aqueduct (younger than 65 B.C., i.e., the onset of Roman time in
the Middle East and older than A.D. 70–230 of early tufa depos-
its). Event X also predates the onset of BR-3 tufa growth (see
Br-3-1 dated A.D. 410–600). Similarly, the tufa growth interrup-
tion in BR-5 (after Br-5-2 dated 110 B.C.–A.D. 130) and onset
of tufa in BR-6 (after Br-6-1 dated 400 B.C.–A.D. 250) coincide
with the occurrence of the fi rst damaging event X. The fi rst earth-
quake faulting that damaged the aqueduct took place between
A.D. 70–230 and A.D. 410–600.
Event Y is younger than A.D. 650–810 (unit d in trench A)
and older than A.D. 540–650 (unit d3 in trench C). This event
postdates the fi rst rebuilding phase of the aqueduct recognized
from the fallen wall in excavation I and related cement sample
AQ-CS-4 (A.D. 532–641) and tufa sample AQ-CS-3-2 (A.D.
560–690). Event Y predates the dragged wall fragment and
related cement sample AQ-CS-1 (A.D. 650–780) and tufa sam-
ple AQ-CS-3-3 (A.D. 639–883; Table 1). Core samples of tufa
deposits provide a bracket of the second damaging earthquake
faulting between Br-4-3 (A.D. 540–980) and Br-3-4 (A.D. 770–
940). The second interruption in both BR-5 and BR-6 may also
have been contemporaneous with the damaging event. Taking
into account only the archaeoseismic results, we can conclude
that event Y likely occurred between A.D. 560–690 and A.D.
650–780; however, the consistency between all dates of paleo-
seismic, archaeoseismic, and tufa analysis suggest an earthquake
event close to A.D. 650. Cement samples CS-1 and tufa sample
CS-3-3 also indicate a rebuilding period after event Y, at the end
of the Byzantine time and beginning of the Islamic period (fi fth
to sixth century A.D.).
Event Z, observed in trenches A, B, and C, is identifi ed as
younger than A.D. 960–1060, and older than A.D. 1030–1260.
The defi nite interruption of tufa growth in all cores and mainly
BR-5 and BR-6 indicates the fi nal stoppage of water fl ow over
the bridge section. The interruption postdates sample Br-6-8
(A.D. 900–1160) and can be correlated with the 29 June 1170
large earthquake that affected the Missyaf region (Mouty and
Sbeinati, 1988; Sbeinati et al., 2005).
The Missyaf segment of the Dead Sea fault experienced four
large earthquakes: event W in 3400–510 B.C., event X in A.D.
70–600, event Y in A.D. 560–780 (probably close to A.D. 650),
and event Z in A.D. 960–1260 (probably in A.D. 1170). Using the
Oxcal program (Bronk Ramsey, 2001), an attempt of sequential
ordering of dates and events, presented in Figure 12, provides a
time probability density function for events W (2300–500 B.C.),
X (A.D. 160–510), Y (A.D. 625–690), and Z (A.D. 1010–1210).
The timing of events obtained from the correlation and sequential
distribution clearly indicate a temporal clustering of three large
seismic events X, Y, and Z (Fig. 12) after event W, which may
indicate a relatively long period of quiescence. Although our data
and observations cannot precisely constrain event W, it may be
correlated with the 1365 B.C. large earthquake that affected sev-
eral sites between Lattakia and Tyre, as reported in the historical
seismicity catalogue of Syria (Sbeinati et al., 2005). The Missyaf
fault behavior is comparable to the temporal cluster of large seis-
mic events that have occurred on other comparable major strike-
slip faults (e.g., San Andreas fault—Weldon et al., 2004; Jordan
Valley fault segment of the Dead Sea fault—Ferry et al., 2007).
DISCUSSION AND CONCLUSION
We conducted four archaeoseismic excavations, three pale-
oseismic trenches, and obtained the radiocarbon dating of six
cores at the Al Harif Aqueduct site along the Missyaf segment
of the Dead Sea fault. The combined study allows us to obtain
a better constraint on the timing of past earthquakes, with four
large seismic events during the last ~3400 yr. The occurrence of
three seismic events X, Y, and Z (A.D. 70–600, ca. A.D. 650, and
A.D. 1170, respectively) since the construction of the aqueduct
is attested by faulting events in trenches, the damage and repair
of the aqueduct wall, and the tufa growth and interruptions since
Roman time (Fig. 13). These results point out a temporal cluster-
ing of three large earthquakes between A.D. 70 and A.D. 1170
along the Missyaf fault segment (Fig. 14).
The 90 ± 10-km-long and linear Missyaf segment experi-
enced the A.D. 1170 earthquake recorded in trenches, aqueduct

Timing of earthquake ruptures at the Al Harif Roman aqueduct (Dead Sea fault, Syria) 261
construction, and tufa deposits. In this tectonic framework, the
large (10-km-wide) Ghab pull-apart basin to the north and the
Al Bouqueaa pull-apart and onset of the restraining bend to the
south (Fig. 3) may constitute endpoints for earthquake rupture
propagation, as observed for other large continental strike-slip
faults (Klinger et al., 2003; Wesnousky, 2006). The size of the
Ghab Basin and the sharp bend of the Lebanese fault system
may act as structural control of fault-rupture initiation and propa-
gation. Furthermore, the damage distribution of the A.D. 1170
earthquake, well located on the Missyaf segment, is limited to
the north by the A.D. 1156 large earthquake and to the south by
the A.D. 1063 and A.D. 1202 earthquakes (Fig. 2; Sbeinati et al.,
2005). The 20-km-thick seismogenic layer (Brew et al., 2001)
correlates with the ~90 km fault length estimated from fi eld map-
ping (Fig. 3). Fault dimensions are consistent with the ~4.3 m
maximum characteristic slip inferred from the warping of the
aqueduct wall east of the fault (and west of the bridge). Here,
we assume that successive faulting episodes maintained the early
~4.3 m warping of an already ruptured strong building. Taking
an average 2.0 m coseismic slip along the fault, the obtained
seismic moment is Mo = 1.05 × 10
20
N m (Mw 7.3; Wells and
Coppersmith, 1994), which is comparable, for instance, with the
seismic moment of the 1999 Izmit large earthquake (Mw 7.4) of
the North Anatolian fault.
Figure 12. (A) Calibrated dating of sam-
ples (with calibration curve INTCAL04
from Reimer et al. [2004] with 2σ age
range and 95.4% probability) and se-
quential distribution from Oxcal pro-
gram (see also Table 1; Bronk Ramsey,
2001). The Bayesian distribution com-
putes the time range of large earth-
quakes (events W, X, Y, and Z) at the
Al Harif aqueduct according to fault-
ing events, construction and repair of
walls, and starts and interruptions of the
tufa deposits (see text for explanation).
Number in brackets (in %) indicates
how much the sample is in sequence;
the number in % indicates an agreement
index of overlap with prior distribution.

262 Sbeinati et al.
The Faulted Aqueduct: Earthquake Damage and
Successive Offsets
The consistency among the timing of faulted sedimentary
units in trenches, the age of building and repair of the aqueduct
wall, and the dating of tufa interruptions and restart episodes
determines the completeness of a sequence of earthquake events.
The dating of three episodes of fault slip X, Y, and Z is consis-
tent with the two phases of aqueduct wall repair, and the two
interruptions of the longest tufa deposits BR-5 and BR-6, and
interruptions and restart in BR-3 and BR-4. Our observations
indicate that the aqueduct was repaired after the large seismic
events X and Y but abandoned after the most recent faulting event
Z. Building repair after a damaging earthquake is very often nec-
essary because it is a vital remedial measure of water supply in
order to avoid a decline of the local economy (Ambraseys, 2006).
The repair has the benefi t of leaving critical indicators of previ-
ous damage and, in some cases, of the fault slip characteristics.
For instance, the eastern wall of the Al Harif aqueduct shows a
clear warping that confi rms the left-lateral movement near the
fault zone. As observed for coseismic surface ruptures crossing
buildings, fences, and walls during large strike-slip earthquakes
(Yeats et al., 1997), warped walls that may record a coseismic
slip are often observed along strike-slip faulting. Warping that
amounts to 4.3 m can be interpreted as the individual coseismic
slip during event X. The warping can be due to the opposite lat-
eral movements across the fault constrained by the bridge cohe-
sion to the east and wall solidity to the west. While the western
aqueduct wall section was built straight on the fl at alluvial terrace
and ends abruptly against the fault, only the section between the
bridge and the fault zone (which is partly built on loose sediments
and bridge ballast) presents some warping and dragging (possibly
separated from the alluvial substratum; Fig. 14). The warped sec-
tion near the bridge displays one generation of cracks fi lled with
tufa that attests to the early bridge damage and possible correla-
tion with event X (Meghraoui et al., 2003). Similar warped walls
Figure 13. Correlation of results among paleoseismic trenching, archaeoseismic excavations, and tufa analysis. In paleo-
seismic trenching, the youngest age for event X is not constrained, but it is, however, limited by event Y. In archaeo-
seismic excavations, the period of fi rst damage overlaps with that of the second damage due to poor age control. In tufa
analysis, the onset and restart of Br-3 and Br-4 mark the damage episodes to the aqueduct; the growth of Br-5 and Br-6
shows interruptions (I) indicating the occurrence of major events. Except for the 29 June 1170 event, previous events have
been unknown in the historical seismicity catalogue. The synthesis of large earthquake events results from the timing cor-
relation among the faulting events, building repair, and tufa interruptions (also summarized in Fig. 12 and text). Although
visible in trenches (faulting event X), archaeoseismic excavations (fi rst damage), and fi rst interruption of tufa growth (in
Br-5 and Br-6 cores), the A.D. 160–510 age of event X has a large bracket. In contrast, event Y is relatively well bracketed
between A.D. 625 and 690, with the overlapped dating from trench results, the second damage of the aqueduct, and the
interruption and restart of Br-3 and onset of Br-4. The occurrence of the A.D. 1170 earthquake correlates well with event
Z from the trenches, the age of third damage to the aqueduct, and the age of interruption of Br-4, Br-5, and Br-6.
Date (AD)
Archaeological
excavations
Paleoseismic
trenching
Tufa
analysis
Historical
seismicity
Civilizations
Synthesis
of EQ events
(see Fig. 12)

Timing of earthquake ruptures at the Al Harif Roman aqueduct (Dead Sea fault, Syria) 263
and fences were observed after the 17 August 1999 earthquake
and along the North Anatolia fault in Turkey (Barka et al., 2002).
Subsequent faulting movements Y and Z would have affected an
already broken aqueduct wall (even if rebuilt) with less strength
at the fault zone than for the initial building conditions (Fig. 14).
Furthermore, the 4.3 m can be considered as a characteristic slip
at the aqueduct site; such characteristic behavior with repeated
same amounts of coseismic slip has already been observed and
inferred from paleoseismic trenches along major strike-slip faults
(Klinger et al., 2003; Rockwell et al., 2009). If the warped aque-
duct wall is random and not representative of a coseismic slip, the
alternative solution is quite similar if we consider a 4.5 m average
individual slip from the cumulative 13.6 m left-lateral offset and
the X, Y, and Z large seismic events at the aqueduct site.
Earthquake Records in Cores
Another key issue is the relationship between the aqueduct
damage and the start and interruption of tufa accumulation with
past earthquakes (Figs. 11 and 13). Indeed, the water fl ow may
be interrupted anytime due to, for instance, the actions of man
(warfare) or the onset of a drought period and climatic fl uctua-
tions that may infl uence the water fl ow. These possibilities seem
here unlikely because the only two interruptions in cores BR-5
and BR-6 coincide with earthquake events X and Y, and no other
additional interruptions were here recorded. This is also attested
by the two interruptions in cores BR-3 and BR-4 that correlate
with earthquake events X and Y. The difference between the tufa
accumulation in BR-4, BR-5, and BR-6 located on the wall sec-
tion west of the fault, and BR-3 located on the wall section next
to the bridge, east of the fault, provides a consistent aqueduct
damage history (Fig. 13). The onset of BR-3 after event X is the
sign of an extensive damage that tilted the bridge and allowed
overfl ow with tufa accumulation on the aqueduct northern side.
The subsequent interruption (repair) and restart of BR-3 that
coincides with event Y illustrate the successive aqueduct damage.
Located on the broken western wall section (Fig. 6), the onset of
BR-4 after event X and restart after event Y are consistent with
BR-3 tufa growth and accumulation. As illustrated in Figure 13,
the coincidence among faulting events X, Y, and Z from paleo-
seismic trenches, the three building damage and repair episodes
from archaeoseismic investigations, and tufa growth and inter-
ruption constrains the earthquake-induced damage and faulting
episodes across the aqueduct.
Figure 14. Schematic reconstruction (with fi nal stage from Fig. 5) of the A.D. 160–510, A.D. 625–690, and A.D. 1170
large earthquakes and related faulting of the Al Harif aqueduct. Except for the A.D. 1170 earthquake (see historical cata-
logue of Sbeinati et al., 2005), the dating of earthquake events are from Figure 12. The white small section is the rebuilt
wall after event X (see buried wall A and B in Fig. 8B); the subsequent gray piece corresponds to the rebuilt wall after
event Y (see wall section C in Fig. 8B), which was damaged and dragged after event Z. The earlier aqueduct deformation
(warping of the eastern wall near the fault rupture) may have recorded ~4.3 m of coseismic left-lateral slip that remained
relatively well preserved during the subsequent fault movements.

264 Sbeinati et al.
Missyaf Segment Seismic Gap and Fault-Rupture Behavior
The Al Harif aqueduct, located at the mid-distance of the
Missyaf fault segment, documents the size and rate of fault slip
associated with large earthquakes. The numerous stream defl ec-
tions observed along the fault segment imply cumulative left-
lateral coseismic offsets consistent with the total aqueduct wall
displacement. Stream defl ections and wall offset result from the
succession of large earthquakes and illustrate the long-term and
short-term fault behavior, respectively. The previous 6.9 mm/yr
slip rate obtained from the temporal cluster of large earthquakes
X, Y, and Z (Meghraoui et al., 2003) clearly overestimates the
long-term fault behavior because it was limited to the past
2000 yr time window. The occurrence of earthquake events W,
X, Y, and Z in the past 3500 yr or so and related inferred 4.3 m
characteristic left-lateral fault slip lead to a slip rate of 4.9–6.3
mm/yr (Fig. 15). Although the inferred age of event W from
trench C is not well constrained, the correlation with the 1365
B.C. seismic sequence and related extension of damage from
Latakia (in Syria) to Tyre (in Lebanon) reported in the histori-
cal catalogue (Sbeinati et al., 2005) suggests 1700–1300 yr of
seismic quiescence in the sequence. The fault activity is here
comparable to the Wallace-type behavior that describes the suc-
cession of temporal clusters of large earthquakes separated by
periods of seismic quiescence (Shimazaki and Nakata, 1980).
The mean recurrence time of large seismic events on the Missyaf
fault segment can be estimated between 550 and 850 yr during
the temporal cluster. This recurrence time increases to ~1077 if
we take into account the maximum estimated age of event W
obtained from the whole earthquake sequence of Figure 12. A
comparable ~1100 yr mean return period is obtained from an
~6000 yr paleo-earthquake record on the juxtaposed southern
Yammouneh fault segment (Daeron et al., 2007). With a limited
number of radiocarbon ages and a possible overlap of the 1408
and 1872 earthquake ruptures, Akyuz et al. (2006) suggested
a minimum 464–549 yr recurrence interval of surface faulting
in the past 1000 yr on the northern end of the Dead Sea fault.
However, the inferred large range in estimates of 500–1100 yr
for recurrence intervals of earthquake faulting confi rms the vari-
ability of earthquake occurrence and slip rates determined by the
Figure 15. Estimated fault-slip behavior and related slip rates (obtained from regression lines) from two scenarios
of possible earthquake occurrence taking into account timing for paleo-earthquakes as in Figure 12 (with average
X [A.D. 160–510] 375 ± 175, average Y [A.D. 625–690] 640 ± 32, Z [A.D. 1170], and two different time frames
for W [historical event of 1365 B.C. and 962 B.C.). In both cases, the two regression lines indicate a minimum and
maximum slip-rate estimate. In parallel, we assume an average 4.3 m characteristic individual slip consistent with the
cumulative 13.6 m measured on the aqueduct (Fig. 5). If we assume a minimum age A.D. 962 for W (according to
the dating in unit f, related rate of sedimentation, and the interface between unit f and unit g in trench C), the slip rate
ranges between 6.1 mm/yr and 6.3 mm/yr (dark regression line with 80% correlation coeffi cient), implying that a large
seismic event is overdue. If we consider the historical catalogue and the 1365 B.C. earthquake sequence along the
Dead Sea fault for W (gray regression line with 78% correlation coeffi cient), the slip rate reduces to 4.9–5.5 mm/yr.
The question mark indicates that for both scenarios, a large earthquake is overdue along the Missyaf fault segment
(according to the seismic gap and the 4.0 m slip defi cit). The temporal cluster of three large earthquakes in less than
1000 yr suggests a Wallace model of fault behavior with periods of seismic quiescence reaching ~1700 yr.

Timing of earthquake ruptures at the Al Harif Roman aqueduct (Dead Sea fault, Syria) 265
relatively long periods of quiescence along the Dead Sea fault
(Ferry et al., 2007).
The instrumental seismicity in Figure 1A shows a seismic
gap in Syria that also corresponds to more than 800 yr quiescence
since the A.D. 1170 earthquake along the Missyaf segment. The
seismic strain distribution is time predictable if we assume a
constant characteristic slip at the aqueduct location. Taking into
account a minimum 962 B.C. age of event W (Fig. 15), the 6.1–
6.3 mm/yr slip rate along the fault (Fig. 15) is comparable to other
slip rates along the northern Dead Sea fault obtained from geol-
ogy or GPS (McClusky et al., 2003; Altunel et al., 2009; Karaba-
cak et al., 2010). If the 1365 B.C. large earthquake involved the
Missyaf fault segment, it implies a 4.9–5.5 mm/yr slip rate, in
agreement with other long-term slip rates of the southern Dead
Sea fault (Klinger et al., 2000; Niemi et al., 2001; Daeron et al.,
2004; Marco et al., 2005; Reilinger et al., 2006; Ferry et al.,
2007; Gomez et al., 2007; Le Beon et al., 2008). The estimated
5.5 mm/yr slip rate and seismic quiescence since A.D. 1170
advocate ~4.0 m slip defi cit and indicate that more seismic stress
accumulation (which may correspond to one to two centuries to
reach a 4.3 m characteristic slip) is needed for a rupture initiation.
Our study shows that the integration of results from archaeoseis-
mology, paleoseismology, tufa deposits, and historical seismicity
is helpful to constrain the timing and characteristics of past earth-
quakes. However, the Dead Sea fault and related Wallace-type
behavior require further paleoearthquake investigations that span
several temporal clusters of seismic events.
ACKNOWLEDGMENTS
This research was funded by the European Commission–funded
APAME Project (contract ICA3-CT-2002-10024) and by the
UMR 7516 of Centre National de la Recherche Scientifi que in
Strasbourg. This research benefi ted from fi eld support from the
Syrian Atomic Energy Commission (SAEC), the Directorate
General of Antiquities and Museums (DGAM), and the Higher
Institute of Applied Sciences and Technology (HIAST) in
Damascus. We are grateful to Ibrahim Osman (director general
of the Atomic Energy Commission of Syria), Muawia Baraz-
angi (Cornell University), Abdal Razzaq Moaz and Tammam
Fakoush (DGAM), and Mikhail Mouty and Khaled Al-Maleh
(Damascus University) for their constant support during the
5 yr study of the Al Harif archaeo-paleoseismology site. We
are indebted to Tony Nemer, Ihsan Layous, Ryad Darawcheh,
Youssef Radwan, and Abdul Nasser Darkal for fi eld assistance
and to Matthieu Ferry and Ersen Aksoy for critically reading an
earlier version of this manuscript. We thank the two anonymous
reviewers who signifi cantly helped to improve our manuscript.
Some fi gures were prepared using the public domain GMT soft-
ware (Wessel and Smith, 1998). This article is a contribution to
the United Nations Educational, Scientifi c and Cultural Orga-
nization–funded International Geoscience Programme IGCP
567, “Earthquake Archaeology: Archaeoseismology along the
Alpine-Himalayan Seismic Zone.”
REFERENCES CITED
Akyuz, S., Altunel, E., Karabacak, V., and Yalciner, C., 2006, Historical earth-
quake activity of the northern part of the Dead Sea fault zone, south-
ern Turkey: Tectonophysics, v. 426, p. 281–293, doi: 10.1016/j.tecto
.2006.08.005.
Alchalbi, A., Daoud, M., Gomez, F., McClusky, S., Reilinger, R., Abu Romeyeh,
M., Alsouod, A., Yassminh, R., Ballani, B., Darawcheh, R., Sbeinati, R.,
Radwan, Y., Al Masri, R., Bayerly, M., Al Ghazzi, R., and Barazangi, M.,
2009, Crustal deformation in northwestern Arabia from GPS measure-
ments in Syria: Slow slip rate along the northern Dead Sea fault: Geo-
physical Journal International, doi: 10.1111/j.1365-246X.2009.04431.x.
Altunel, E., Meghraoui, M., Karabacak, V., Akyüz, S.H., Ferry, M., Yalçıner,
C., and Munschy, M., 2009, Archeological sites (Tell and Road) offset
by the Dead Sea fault in the Amik Basin, southern Turkey: Geophysical
Journal International, v. 179, p. 1313–1329, doi: 10.1111/j.1365-246X
.2009.04388.x.
Ambraseys, N.N., 2006, Earthquakes and archeology: Journal of Archaeologi-
cal Science, v. 33, p. 1008–1016, doi: 10.1016/j.jas.2005.11.006.
Ambraseys, N.N., 2009, Earthquakes in the Mediterranean and Middle East: A
Multidisciplinary Study of Seismicity up to 1900: Cambridge, UK, Cam-
bridge University Press, 947 p.
Ambraseys, N.N., and Jackson, J.A., 1998, Faulting associated with historical
and recent earthquakes in the Eastern Mediterranean region: Geophysi-
cal Journal International, v. 133, p. 390–406, doi: 10.1046/j.1365-246X
.1998.00508.x.
Ambraseys, N., and Melville, C.P., 1988a, Historical evidence of fault-
ing in Eastern Anatolia and northern Syria: Annali di Geofi sica, v. 28,
p. 337–343.
Ambraseys, N., and Melville, C.P., 1988b, An analysis of the Eastern Medi-
terranean earthquake of 20 May 1202, in Lee, W.H.K., Meyers, H., and
Shimazaki, K., eds., Historical Seismograms and Earthquakes of the
World: San Diego, Academic Press, p. 181–200.
Barazangi, M., Seber, D., Chaimov, T., Best, J., Litak, R., Al-Saad, D., and
Sawaf, T., 1993, Tectonic evolution of the northern Arabian plate in west-
ern Syria, in Boschi, E., et al., eds., Recent Evolution and Seismicity of
the Mediterranean Region: Dordercht, the Netherlands, Kluwer Aca-
demic, p. 117–140.
Barka, A., Akyüz, H.S., Altunel, E., Sunal, G., Çakir, Z., Dikbas, A., Yerli, B.,
Armijo, R., Meyer, B., de Chabalier, J.B., Rockwell, T., Dolan, J.R., Har-
tleb, R., Dawson, T., Christofferson, S., Tucker, A., Fumal, T., Langridge,
R., Stenner, H., Lettis, W., Bachhuber, J., and Page, W., 2002, The sur-
face rupture and slip distribution of the 17 August 1999 Izmit earthquake
(M 7.4), North Anatolian fault: Bulletin of the Seismological Society of
America, v. 92, p. 43–60, doi: 10.1785/0120000841.
Brew, G., Lupa, J., Barazangi, M., Sawaf, T., Al-Imam, A., and Zaza, T., 2001,
Structure and tectonic development of the Ghab Basin and the Dead Sea
fault system, Syria: Journal of the Geological Society of London, v. 158,
p. 665–674, doi: 10.1144/jgs.158.4.665.
Bronk Ramsey, C., 2001, Development of the radiocarbon program OxCal:
Radiocarbon, v. 43, p. 355–363.
Chaimov, T.A., Barazangi, M., Al-Saad, D., Sawaf, T., and Gebran, A., 1990,
Crustal shortening in the Palmyride fold belt, Syria, and implications for
movement along the Dead Sea fault system: Tectonics, v. 9, p. 1369–1386,
doi: 10.1029/TC009i006p01369.
Daeron, M., Benedetti, L., Tapponnier, P., Sursock, A., and Finkel, R., 2004,
Constraints on the post–25-ka slip rate of the Yammouneh fault (Leba-
non) using in situ cosmogenic
36
Cl dating of offset limestone-clast fans:
Earth and Planetary Science Letters, v. 227, p. 105–119, doi: 10.1016/
j.epsl.2004.07.014.
Daeron, M., Klinger, Y., Tapponnier, P., Elias, A., Jacques, E., and Sursock, A.,
2005, Sources of the large A.D. 1202 and 1759 Near East earthquakes:
Geology, v. 33, p. 529–532, doi: 10.1130/G21352.1.
Daeron, M., Klinger, Y., Tapponnier, P., Elias, A., Jacques, E., and Sursock, A.,
2007, 12,000-year long record of 10 to 13 paleo-earthquakes on the Yam-
mouneh fault (Levant fault system, Lebanon): Bulletin of the Seismologi-
cal Society of America, v. 97, p. 749–771, doi: 10.1785/0120060106.
Dubertret, L., 1955, Carte Géologique de Syrie avec Notice Explicative:
Damascus, République Syrienne, Ministère des Travaux Publics, scale
1:200,000.

266 Sbeinati et al.
Elias, A., 2006, Le Chevauchement de Tripoli-Saïda: Croissance du Mont
Liban et Risque Sismique [Ph.D. thesis]: Paris, Institut de Physique du
Globe de Paris, 230 p.
Ellenblum, R., Marco, S., Agnon, A., Rockwell, T., and Boas, A., 1998, Cru-
sader castle torn apart by earthquake at dawn, 20 May 1202: Geology,
v. 26, p. 303–306, doi: 10.1130/0091-7613(1998)026<0303:CCTABE
>2.3.CO;2.
Farr, T., and Kobrick, M., 2000, Shuttle Radar Topography Mission produces a
wealth of data: Eos (Transactions, American Geophysical Union), v. 81,
p. 583–585.
Ferry, M., Meghraoui, M., Abou Karaki, N., Al-Taj, M., Amoush, H., Al-Dha-
isat, S., and Barjous, M., 2007, A 48-kyr-long slip rate history for the
Jordan Valley segment of the Dead Sea fault: Earth and Planetary Science
Letters, v. 260, p. 394–406, doi: 10.1016/j.epsl.2007.05.049.
Ford, T.D., and Pedley, H.M., 1996, A review of tufa and travertine deposits
of the world: Earth-Science Reviews, v. 41, p. 117–175, doi: 10.1016/
S0012-8252(96)00030-X.
Garfunkel, Z., Zak, I., and Freund, R., 1981, Active faulting in the Dead Sea
rift: Tectonophysics, v. 80, p. 1–26, doi: 10.1016/0040-1951(81)90139-6.
Ginat, H., Enzel, Y., and Avni, Y., 1998, Translocated Plio-Pleistocene drainage
systems along the Arava fault of the Dead Sea transform: Tectonophysics,
v. 284, p. 151–160, doi: 10.1016/S0040-1951(97)00165-0.
Gomez, F., Meghraoui, M., Darkal, A., Hijazi, F., Mouty, M., Sulaiman, Y.,
Sbeinati, R., Darawcheh, R., Al-Ghazzi, R., and Barazangi, M., 2003,
Holocene faulting and earthquake recurrence along the Serghaya
branch of the Dead Sea fault system in Syria and Lebanon: Geophysi-
cal Journal International, v. 153, p. 658–674, doi: 10.1046/j.1365-246X
.2003.01933.x.
Gomez, F., Khawlie, M., Tabet, C., Darkal, A.N., Khair, K., and Barazangi, M.,
2006, Late Cenozoic uplift along the northern Dead Sea transform in Leb-
anon and Syria: Earth and Planetary Science Letters, v. 241, p. 913–931,
doi: 10.1016/j.epsl.2005.10.029.
Gomez, F., Karam, G., Khawlie, M., McClusky, S., Vernant, P., Reilinger,
R., Jaafar, R., Tabet, C., Khair, K., and Barazangi, M.,, 2007a, Global
positioning system measurements of strain accumulation and slip trans-
fer through the restraining bend along the Dead Sea fault system in
Lebanon: Geophysical Journal International, v. 168, p. 1021–1028, doi:
10.1111/j.1365-246X.2006.03328.x.
Gomez, F., Nemer, T., Tabet, C., Khawlie, M., Meghraoui, M., and Barazangi,
M., 2007b, Strain partitioning of active transpression within the Leba-
nese restraining bend of the Dead Sea fault (Lebanon and SW Syria):
Geological Society of London Special Publication 290, p. 285–303, doi:
10.1144/290.10.
Grootes, P.M., Nadeau, M.J., and Rieck, A., 2004,
14
C-AMS at the Leibniz-
Labor: Radiometric dating and isotope research: Nuclear Instruments and
Methods in Physics Research, ser. B, v. 223–224, p. 55–61.
Grootes, P.M., Nadeau, M.J., Roth, S., Andersen, N., Huels, M., Meghraoui,
M., and Sbeinati, R., 2006, 1000 years of usage: The life story of a Roman
Aqueduct provides tectonic information: Eos (Transactions, American
Geophysical Union), v. 87, no. 52, Fall Meeting Suppl., abstract T13B-
0500.
Grünthal, G., 1998, European Macroseismic Scale, 1998: Cahiers du Centre
Européen de Géodynamique et de Seismologie: Luxembourg, Conseil de
l’Europe, 230 p.
Guidoboni, E., Comastri, A., and Traina, G., 1994, Catalogue of Ancient Earth-
quakes in the Mediterranean Area up to the 10th Century: Bologna, ING-
SGA (Istituto Nazionale di Geofi sica–Storia Geofi sica Ambiente), 504 p.
Guidoboni, E., Bernardini, F., and Comastri, A., 2004a, The 1138–1139 and
1156–1159 destructive seismic crisis in Syria, south-eastern Turkey
and northern Lebanon: Journal of Seismology, v. 8, p. 105–127, doi:
10.1023/B:JOSE.0000009502.58351.06.
Guidoboni, E., Bernardini, F., Comastri, A., and Boschi, E., 2004b, The
large earthquake on 29 June 1170 (Syria, Lebanon, and central south-
ern Turkey): Journal of Geophysical Research, v. 109, p. B07304, doi:
10.1029/2003JB002523.
Ibn Al-Athir, Ezz Ad-Din (AH 491–541, A.D. 1097–1146), 1982, Al-Kamil fi
Al-Tarikh (The Complete History), v. 8, 9, 10, 11, 12: Beirut, Dar Sader.
Karabacak, V., Altunel, E., Meghraoui, M., and Akyüz, S., 2010, Field evi-
dences from northern Dead Sea fault zone (south Turkey): New fi ndings
for the initiation age and slip rate: Tectonophysics, v. 480, p. 172–182,
doi: 10.1016/j.tecto.2009.10.001.
Klinger, Y., Avouac, J.P., Abou Karaki, N., Dorbath, L., Bourles, D., and Reyss,
J.L., 2000, Slip rate on the Dead Sea transform in northern Araba Val-
ley (Jordan): Geophysical Journal International, v. 142, p. 755–768, doi:
10.1046/j.1365-246x.2000.00165.x.
Klinger, Y., Sieh, K., Altunel, E., Akoglu, A., Barka, A., Dawson, T., Gonza-
lez, T., Meltzner, A., and Rockwell, T., 2003, Paleoseismic evidence of
characteristic slip on the western segment of the North Anatolian fault,
Turkey: Bulletin of the Seismological Society of America, v. 93, p. 2317–
2332, doi: 10.1785/0120010270.
Le Beon, M., Klinger, Y., Qader Amrat, A., Agnon, A., Dorbath, L., Baer, G.,
Ruegg, J.-C., Charade, O., and Mayyas, O., 2008, Slip rate and locking
depth from GPS profi les across the southern Dead Sea transform: Journal
of Geophysical Research, v. 113, p. B11403, doi: 10.1029/2007JB005280.
Marco, S., 2008, Recognition of earthquake-related damage in archaeological
sites—3 examples from the Dead Sea fault zone: Tectonophysics, v. 453,
p. 148–156, doi: 10.1016/j.tecto.2007.04.011.
Marco, S., Hartal, M., Hazan, N., Lev, L., and Stein, M., 2003, Archaeology,
history, and geology of the A.D. 749 earthquake, Dead Sea Transform:
Geology, v. 31, p. 665–668, doi: doi:10.1130/G19516.1.
Marco, S., Rockwell, T.K., Heimann, A., Frieslander, U., and Agnon, A., 2005,
Late Holocene activity of the Dead Sea transform revealed in 3D paleo-
seismic trenches on the Jordan Gorge segment: Earth and Planetary Sci-
ence Letters, v. 234, p. 189–205, doi: 10.1016/j.epsl.2005.01.017.
McClusky, S., Reilinger, R., Mahmoud, S., Ben, D., and Tealeb, A., 2003,
GPS constraints on Africa (Nubia) and Arabia plate motions: Geophysi-
cal Journal International, v. 155, p. 126–138, doi: 10.1046/j.1365-246X
.2003.02023.x.
Meghraoui, M., Gomez, F., Sbeinati, R., Van der Woerd, J., Mouty, M., Darkal,
A., Radwan, Y., Layyous, I., Najjar, H., Darawcheh, R., Hijazi, F., Al-
Ghazzi, R., and Barazangi, M., 2003, Evidence for 830 years of seismic
quiescence from paleoseismology, archeoseismology and historical seis-
micity along the Dead Sea fault in Syria: Earth and Planetary Science
Letters, v. 210, p. 35–52, doi: 10.1016/S0012-821X(03)00144-4.
Mouty, M., and Sbeinati, R., 1988, Historical Earthquake Catalogue for Syria
and Adjacent Areas: Damascus, Atomic Energy Commission of Syria,
Internal Report, 70 p.
Nemer, T., and Meghraoui, M., 2006, Evidence of coseismic ruptures along
the Roum fault (Lebanon): A possible source for the A.D. 1837 earth-
quake: Journal of Structural Geology, v. 28, p. 1483–1495, doi: 10.1016/
j.jsg.2006.03.038.
Nemer, T., Meghraoui, M., and Khair, K., 2008, The Rachaya-Serghaya fault
system (Lebanon): Evidence of coseismic ruptures, and the A.D. 1759
earthquake sequence: Journal of Geophysical Research, v. 113, B05312,
doi: 10.1029/2007JB005090.
Niemi, T., Zhang, H., Atallah, M., and Harrison, J.B.J., 2001, Late Pleisto-
cene and Holocene slip rate of the northern Wadi Araba fault, Dead
Sea transform, Jordan: Journal of Seismology, v. 5, p. 449–474, doi:
10.1023/A:1011487912054.
Nur, A., and Cline, E.H., 2000, Poseidon’s Horses: Plate tectonics and earth-
quake storms in the late Bronze Age Aegean and Eastern Mediterra-
nean: Journal of Archaeological Science, v. 27, p. 43–63, doi: 10.1006/
jasc.1999.0431.
Quennell, A.M., 1984, The western Arabia rift system, in Dixon, J.E., and Rob-
ertson, A.H.F., eds., The Geological Evolution of the Eastern Mediterra-
nean: Geological Society of London Special Publication 17, p. 775–788.
Reilinger, R., McClusky, S., Vernant, P., Lawrence, S., Ergintav, S., Cakmak, R.,
Ozener, H., Kadirov, F., Guliev, I., Stepanyan, R., Nadariya, M., Hahubia,
G., Mahmoud, S., Sakr, K., ArRajehi, A., Paradissis, D., Al-Aydrus, A.,
Prilepin, M., Guseva, T., Evren, E., Dmitrotsa, A., Filikov, S.V., Gomez,
F., Al-Ghazzi, R., and Karam, G., 2006, GPS constraints on continental
deformation in the Africa–Arabia–Eurasia continental collision zone and
implications for dynamics of plate interactions: Journal of Geophysical
Research, v. 111, doi: 10.1029/ 2005JB004051.
Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Bertrand,
C.J.H., Blackwell, P.G., Buck, C.E., Burr, G.E., Cutler, K.B., Damon,
P.E., Edwards, R.L., Fairbanks, R.G., Friedrich, M., Guilderson, T.P.,
Hogg, A.G., Hughen, K.A., Kromer, B., McCormac, G., Manning, S.,
Bronk Ramsey, C., Reimer, R.W., Remmele, S., Southon, J.R., Stuiver,
M., Talamo, S., Taylor, F.W., van der Plicht, J., and Weyhenmeyer, C.E.,
2004, IntCal04 Terrestrial radiocarbon age calibration, 26–0 ka BP:
Radiocarbon, v. 46, p. 1029–1058.

Timing of earthquake ruptures at the Al Harif Roman aqueduct (Dead Sea fault, Syria) 267
Rockwell, T., Ragona, D., Seitz, G., Langridge, R., Aksoy, M.E., Ucarkus, G.,
Ferry, M., Meltzner, A.J., Klinger, Y., Meghraoui, M., Satir, D., Barka, A.,
and Akbalik, B., 2009, Paleoseismology of the North Anatolian fault near
the Marmara Sea: Implications for fault segmentation and seismic hazard,
in Reicherter, K., Michetti, A.M., and Silva, P.G., eds., Palaeoseismol-
ogy: Historical and Prehistorical Records of Earthquake Ground Effects
for Seismic Hazard Assessment: Geological Society of London Special
Publication 316, p. 31–54, doi: 10.1144/SP316.3.
Salamon, A., Hofstetter, A., Garfunkel, Z., and Ron, H., 2003, Seismotec-
tonics of the Sinai subplate—The Eastern Mediterranean region: Geo-
physical Journal International, v. 155, p. 149–173, doi: 10.1046/j.1365
-246X.2003.02017.x.
Sbeinati, M.R., Darawcheh, R., and Mouty, M., 2005, The historical earth-
quakes of Syria: An analysis of large and moderate earthquakes from
1365 B.C. to 1900 A.D.: Annals of Geophysics, v. 48, p. 347–435.
Shimazaki, K., and Nakata, T., 1980, Time-predictable recurrence model for
large earthquakes: Geophysical Research Letters, v. 7, p. 279–282, doi:
10.1029/GL007i004p00279.
Stiros, S., and Jones, R.E., eds., 1996, Archaeoseismology: British School of
Athens, Greece, Fitch Laboratory Occasional Paper 7, 268 p.
Trifonov, V.G., Tribukhin, V.M., Adzhamyan, Z., Dshallad, S., El-Khair, Y., and
Ayed, K., 1991, Levant fault zone in northwest Syria: Geotectonics, v. 25,
p. 145–154.
Wdowinski, S., Bock, Y., Baer, G., Prawirodirdjo, L., Bechor, N., Naaman, S.,
Knafo, R., Forrai, Y., and Melzer, Y., 2004, GPS measurements of cur-
rent crustal movements along the Dead Sea fault: Journal of Geophysical
Research, v. 109, B05403, doi: 10.1029/2003JB002640.
Weldon, R., Scharer, K., Fumal, T., and Biasi, G., 2004, Wrightwood and
the earthquake cycle: What a long recurrence record tells us about
how faults work: GSA Today, v. 14, no. 9, p. 4–10, doi: 10.1130/
1052-5173(2004)014<4:WATECW>2.0.CO;2.
Wells, D.L., and Coppersmith, K.J., 1994, New empirical relationships among
magnitude, rupture length, rupture width, rupture area, and surface dis-
placement: Bulletin of the Seismological Society of America, v. 84,
p. 974–1002.
Wesnousky, S.G., 2006, Predicting the endpoints of earthquake ruptures:
Nature, v. 444, p. 358–360, doi: 10.1038/nature05275.
Wessel, P., and Smith, H.F., 1998, New, improved version of the Generic Map-
ping Tools released: Eos (Transactions, American Geophysical Union),
v. 79, p. 579.
Westaway, R., 2004, Kinematic consistency between the Dead Sea fault zone
and the Neogene and Quaternary left-lateral faulting in SE Turkey: Tecto-
nophysics, v. 391, p. 203–237, doi: 10.1016/j.tecto.2004.07.014.
Yeats, R., Sieh, K., and Allen, C., 1997, The Geology of Earthquakes: Oxford,
UK, Oxford University Press, 568 p.
Zilberman, E., Amit, R., Porat, N., Enzel, Y., and Avner, U., 2005, Surface rup-
tures induced by the devastating 1068 A.D. earthquake in the southern
Arava valley, Dead Sea Rift, Israel: Tectonophysics, v. 408, p. 79–99, doi:
10.1016/j.tecto.2005.05.030.
MANUSCRIPT ACCEPTED BY THE SOCIETY 19 MAY 2010
Printed in the USA