TECTONICS, VOL. 9, NO.

6, PAGES 1369-1386, DECEMBER 1990

CRUSTAL SHORTENING IN THE

PALMYRIDE FOLD BELT, SYRIA, AND
IMPLICATIONS FOR MOVEMENT ALONG
THE DEAD SEA FAULT SYSTEM

Thomas A. Chaimov and Muawia Barazangi

Institute for the Study of the Continentsand

Department of Geological Sciences,Cornell
University, Ithaca, New York

Damen A1-Saad, Tarif Sawaf, and Ali Gebran

Syrian PetroleumCompany,Ministry of Petroleum

and Mineral Resources, Damascus

Abstract. The Palmyride fold belt is a to explain the discrepancybetweenthe 80 km of

northeast-trending, 400 by 100 km transpressive predictedshorteningand the only 20 km of
belt in central Syria embeddedin the northern shorteningmeasuredfrom restoredcross
Arabian platform, boundedto the north by the sections. Restoredcross sectionsoffer only
Aleppo plateauand to the southby the Rutbah minimum shorteningestimates,so the calculated
uplift. Palinspasticallyrestoredcrosssections 20 km may underestimateshortening. Second,
from three transectsacrossthe Palmyridefold belt evidenceof strike-slipdisplacementrecognizedin
demonstratea minimum NW-SE shorteningof minimum shorteningestimates,so the calculated
about 20% or 20 km across the southwestern 20 km may underestimateshortening. Second,
segmentof the belt, diminishingto 1-2 km in the evidenceof strike-slipdisplacementrecognizedin
northeast, close to the Euphrates graben system. the field and reportedin the literature,and
The cross sections are based on the 1:200,000 indicatedby new focal mechanismsolutionsof two
scalegeologicmap of Syria andpreviously recent earthquakesin the Palmyfides,indicates
unavailable seismic reflection and well data, all that some of the still "missing"displacementmay
providedby the Syrian PetroleumCompany. be distributedthroughoutcentral and northern
These resultsdiffer significantlyfrom those Syria as strike-slipmotion oblique to the relative
predictedby kinematicmodelsof Middle East northwardconvergenceof the Arabian plate on the
plate motions. In western Syria and eastern Eurasian plate. Alternatively, previous estimates
Lebanon the Palmyrides obliquely intersect (at of slip along the northernsegmentof the Dead
about 45) the roughly north-trendingDead Sea Sea transformfault systemmay be only minimum
transformfault system. The Dead Sea fault estimates. A final possibility is that the Dead
systemshowswell-documentedevidenceof 105 Sea transformfault in northwesternSyria has
km of left-lateral displacementsincemid-Tertiary beenactivefor only the past5-6 m.y. or so,
time south of its intersectionwith the Palmyrides, implying that it was either nonexistentor moved
yet only about 25 km of motionhas been only slightlybeforethe Pliocene. This would
documentednorth of thatjuncturein Lebanonand suggestthat thereis a total of only 45 km of N-S
westernSyria. Thus, kinematicmodelsof Middle convergenceto be found in centraland northern
East plate motionspredict 80 km of shorteningin Syria, about 25 km on the Dead Sea fault system
Syria, most of which shouldbe accommodatedin and about 20 km in the Palmyfides. This last
the Palmyride fold belt. Severalpossibilitiesexist possibilityrequiresthat the northernand southern
segmentsof the Dead Sea fault systemdeveloped
Copyright 1990 independentlyduringmost of the past 15-20 m.y.
by the AmericanGeophysicalUnion In light of the documentedbut unquantifiedstrike-
slip motion in the Palmyfides,it seemsreasonable
Papernumber90TC01071. that strike-slipmotion does accommodatea
0278-7407/90/90TC-01071 $10.00 significantportion of the convergencebetweenthe
1370 Chaimov et al.' Palmyride shorteningand Dead Sea Fault Movement

Arabian and Eurasianplates. It is likely, McBride et al., 1990]. Tertiary deformation

however, that one or more of the other proposed includesboth north and southvergentoblique
mechanismsalso accountsfor a componentof the reverse and thrust faults, reactivated normal
expected80 km of shortening. faults, broad as well as tight folds, and even Jura-
type box folds. The Palmyridescan be subdivided
INTRODUCTION into at least three distinctregionsor blocks,two
to the north of a major E-W trendingstrike-slip
Complex tectonicbelts surroundthe Syria fault, the Jharfault, and one more intenselyfolded
region (Figure la). To the northof Syria lies the block to the southof the fault (seeFigure lb).
B itlis suture and the East Anatolian fault of The broadfolds are observedprimarilyin the Bilas
southernTurkey, to the eastthe Zagrosfold belt and Bishri blocks to the north of the Jhar fault,
of Iran and Iraq. In Lebanonand westernSyria giving thoseregionsa more gently deformed
lies the left-lateral Dead Sea transform fault appearancethan the southwesternpart. McBride
system. These interrelatedtectonicelementsall et al. [ 1990] provide a thoroughoverviewof the
place constraintson Palmyrideevolution. geologicand tectoniccharacterof the belt and
Therefore,any tectonichistoryof the Palmyfide surroundingregions.
belt must fit in with the tectonic framework of the Palmyfide tectonicsare importantfor
whole region. understandingthe diffuse plate boundarybetween
The Palmyfidefold belt in centralSyria trends Asia and Arabia that extends for 300 km from the
northeastand is boundedto the north by the more Palmyfidesto the Bitlis suture. Many scientists
stableAleppo plateau and to the southby the have studiedMiddle East tectonics,especiallythe
Rutbahuplift, the northernmarginof the stable Dead Sea transformfault system,the Red Sea,
Arabian platform (Figure lb). Thickened and the Afar region [e.g., Quennell, 1958; Freund
Paleozoicand Mesozoic stratain the uplifted et al., 1970; Le Pichon and Francheteau, 1978;
blocksof the Palmyfidesindicatean inverted Hempton, 1987], but as yet little attentionhas
Upper Paleozoicand Mesozoicbasin,perhapsan beenfocusedon the detailedtectonicsof Syria. A
aulaogen [Ponikerov,1966; Beydoun,1981; few authors have, however, included the

o 38 4o 42 44 4 E

B..i'TLISSUTURE

7k sYR
IA

Faults
Seismic
/'"'Reflection
'*.-DEAD SEA / Transects
FAULTYSTEM 0 km 200 N
! I

Fig. l a. Generalizedgeographicand tectonicsettingof Syria showinglocationsof

transectsA, B, and C (see Plate 1 and Figures2 and 3) in the Palmyridefold belt.
Enlargementof boxedarea is shownin Figure lb.
Chaimovet al.' Palmyrideshorteningand Dead Sea Fault Movement 1371

Palmyridebelt in their studiesof the region[e.g. mountainbelt. Not only do balancedcross

Beydoun,1981;Lovelock,1984;Walley, 1988]. sectionsprovidea basisfrom whichshortening in
The SyrianPetroleumCompanyhas made the Palmyfidescan be estimated,but the estimate
availableto Cornell University seismicreflection itselfmay helpexplainan enigmaof Middle East
and well data from Syria. Thesedata have platetectonics,the discrepancy betweenleft-
facilitated a concerted effort to understand the lateral offsetsalong the northernversusthe
tectonicsof Syria and their relationto surrounding southernsegmentsof the Dead Sea transform
tectonic features. The present study concentrates fault system.
on datafrom thePalmyfideintracratonicfold belt
in central Syria. DEAD SEA TRANSFORM FAULT SYSTEM
Most balanced cross section studies are
conductedin fold and thrust belts near the margins A questionthat pervadesthe geologic
of continents in belts that formed as a direct result literatureof the Middle East region concernsthe
of platecollision. In thePalmyfidecasewe apply magnitudeof left-lateralmotionobservedacross
the methodof restoringbalancedcrosssectionsto the southernversusthe northern segmentsof the
estimateshorteningin an intracontinentalfold Dead Sea transformfault system. The Dead Sea
belt. The Palmyfides,wheregeologicaland fault system,which strikesNNE-SSW in Lebanon
geophysical dataareplentiful,providea unique and N-S in westernmostSyria, displays-105 km
opportunityto examinein detailthe subsurface offset on its southernsegment[Quennell,1958;
structureand tectonichistoryof an intracratonic Freund et al., 1970], south of Lebanon. This

37 E 380E 39 o E

1987BISH

35
N BILAS
BLOCKi SOUKHNE

..
,c
OUTH ........
Fig.5:%OUBAYAT

_,I.CHERRIFE ; Cretaceous
PALMYRA
Fig.
BLOCK
:""--"
Outcrop
7....Seismic
Faults
B '"Reflection
34
Transects
N
0 km 40
I I I

Fig. lb. Map viewlocationsof thethreecrosssections

A, B, andC andwellsusedin
the study. Cretaceous
outcropis shadedandreflectstheoccurrenceof structural
highs.Detailedlithologiccontactsareshownon thecrosssectionsin Plate1 and
Figures2 and3. Starsrepresent epicenters
of earthquakes
discussedin text,and
focal mechanismsof thesetwo eventsare shownon lower hemisphereprojections
with compressional
quadrantsshaded.
1372 Chaimovet al.: Palmyrideshorteningand Dead Sea Fault Movement

amountof offset is generallyaccepted[Courtillot terrainwhich often coincideswith exposuresof

et al., 1987]. As yet, however,no one has resistant Cretaceous carbonate rocks. Other
unequivocally documented morethan20-30 km of seismicdata problemsare few and minor.
left-lateral offset on the northern segment Explorationwells in Syria rangein depthfrom
[Quennell,1958; Trifonov et al., 1983; Vladimir very shallowto over 4 km. Well informationfrom
Trifonov,personalcommunication, 1989]. The the three wells usedin this study, the locationsof
questionarises,then, as to what hasbecomeof which appearin Figures1-3, consistsprimarily of
the missing80 km of displacement. formationtopspickedby Syrian Petroleum
The little-studiedPalmyfide belt in central Companygeologistsas well as sparse
Syria (Figure 1) strikesobliquely(about30-45) geophysicalwell logs. The interpretivecross
to the Dead Sea transformfault system,and its sectionspresented,however, are basedprimarily
intersection with the Dead Sea transform fault on the seismicreflection sections. Integral to the
systemdivides that systeminto its interpretationof the seismicsectionsand for
aforementionednorthern and southernsegments. geologicextrapolationwhere seismicdata were
It may hold the answerto the problemof not available was Ponikerov's [1963] 1:200,000
discrepantoffsetsacrossthe northernversusthe scalegeologicmap of Syria.
southernsegmentsof the transformfault system. Three -50 km long northwest-southeast
If the Palmyfidesalone accommodatedthe transectswhich crossmost of the Palmyfide belt
missing80 km solelyby thrustingand foldingone normal to strike were constructed from the
would expectto find evidencefor largescale seismicreflectionprofiles (seeFigure lb). The
shorteningin crosssectionsconstructedacross mostly high quality (to-3 sec) seismicsections
the belt. If, however,considerablestrike-slip were correlatedwith surfacegeologyand well
motion is invoked to accommodate some of the information. Several stratigraphicsequenceson
shorteningone would expect to observeless each time sectionwere convened to depth using
thrustingandfolding. Basedon his kinematic stackingvelocitiesand formationtopsfrom nearby
model of the northeastern Mediterranean wells. These units are, in ascendingorder,
Quennell [1958] assertedthat if the northern Paleozoic, Jurassic and Triassic, Cretaceous,
Dead Sea transformfault systemmoved Paleogene,and Neogene. The resultantcross
independentlyof the southernsegment,an sections were area balanced in order to maximize
importantassumptionon which he elaborates their "chanceof beingfight" [Dahlstrom,1969]
little, then 45 km of shorteningshouldbe apparent and simultaneouslyrestoredto an undeformed
in the southwesternPalmyfides,diminishingto state to estimate the shorteningaccommodated
zero 400 km along strike to the northeast.Recent by thrustfaultingandfolding.
acceleratedactivity in explorationfor At times, in areasof complexstructurewhere
hydrocarbons in Syria hasresultedin the stratigraphiccontrolwas minimal, several
collectionof much seismicand well data. Newly differentinterpretationsof the seismicdata
availableseismicdata in the regionprovidea seemedplausible,but only one interpretation
medium for new structuralinterpretationsand balanceduponrestoration.For example,over the
facilitate an estimateof shortening,thereby northwesternmost SW vergent anticline of
providinga quantitativetestof Quennell's transectA (see Plate 1) the Cretaceoussectionis
hypothesis. continuous.This placesstrictlimits on the
amountof shorteninginterpretedin the poorly
DATA AND METHOD OF ANALYSIS reflective Jurassic and Triassic section. Where an
unbalancedduplexstructuremight appearto fit
Seismic,well, and otherrelevantgeologic geometrically,the simpleramp/flat shownin Plate
informationwas providedfor this studyby the 1 balances.
SyrianPetroleumCompany. From nearly 1000km The cross sectionswere balancedonly as far
of seismicdata at Cornell University, three north as the Jhar fault, north of which
transectsthroughthe Palmyride belt were compressional structuresare lesspronounced.
selectedfor this study. The seismicdata were North of the steep,E-W strikingJharfault strike-
collectedandprocessedby a variety of oil slipduplexing[Woodcock,1986] is probablythe
companiesover many years,so their quality dominantmodeof uplift, wherelarge,relatively
varies. For the most part, however,the four to undeformedupliftedcrustalblocksareboundedby
five secondseismicprofiles,nearly all processed steepfaultsthat evincenegligibleshortening.
papercopies,imagewell the thick Mesozoicand This separationof deformationstylesis similarto
Cenozoicsedimentarysectionof centralSyria. the strainpartitioningdiscussedby Mount and
Data acquisitionand/orprocessingproblemswere Suppe[1987] andZobacket al. [1987] for the San
commonlyencounteredover unusuallyrugged Andreas system.
 1373
Chaimovet al.: Palmyrideshorteningand Dead Sea Fault Movement

RESULTS versusthe northeasternPalmyrides. The

southwestern,most intensely folded region
These new resultsenrich our understandingof exhibitstypical fold and thrustbelt structures,
Middle East tectonics,in particularthe tectonics with a low-angled6collementsurfaceprobablyin
of Syria. Theyprovideimportantinformation on the lower (basal?)Triassic. Clear evidenceof a
thetimingof deformation in Syriaandthe low-angledetachmentat this stratigraphiclevel is
seenon the seismicreflection data only from the
magnitude of shortening
in thePalmyridefold belt. southwestern region. It is this detachmentthat is
These data and interpretationsalso contrastthe interpretedto correspondto an evaporiticlayer.
very differentdeformational styleswithinthebelt, Figure4 showsthe bestseismicexampleof the
the southwesternthin-skinnedstyle of d6collement in the southwest.
deformationwhere a low-angle detachmentis The strataof probablePaleozoicage below the
observed versus the northeastern thicker-skinned d6collementappearslightlycurved. This
style where no detachmentis apparent. phenomenon hasbeensimplymodeledas velocity
pull up of a very gentlydippingflat layer. A
simplemodelbasedon a few majorreflections
Timingof PalmyrideDeformation was constructedusing the Sierra Geophysics
three dimensionalseismicmodelingpackage. The
A generalstatementaboutthe timingof relativelyundeformedPaleozoicstratacan be
compressional deformationin thePalmyfidescan modeledas a flat layer or, with a slight
be madebasedon the datapresentedin this perturbationin seismicvelocitiesas a very gently
study.While it is difficultto determinefrom this dippinghorizon. Velocitiesusedin themodeling
data setthe exactage of the onsetof compression were approximatedue to the poorlyconstrained
in the Palmyfides,it is accurateto saythat most stackingvelocitiesof the seismicsectionand the
Palmyrideshorteningtook placeafter the late dearthof sonicwell log information. Nonetheless,
Oligocene.This conclusionstemsfrom the preliminaryseismicmodelingdoesindicate
observationthat the thick Paleogene(Paleocene relatively undeformedsub-d6collement strata. In
throughOligocene)sectionthroughout the the future a detailedstudyof the overlying
Palmyridesvariesvery graduallyin thicknessin deformed Mesozoic and Cenozoic seismic
any givendip section(seefor example,Figure2), sequences will yield betterconstraints
on the
and it furthermore shows little evidence of internal
timingof deformationin thePalmyrides.
onlap,suggesting relativelyquiescenttectonics In the northeasternPalmyfides,where
duringdeposition.In contrast,all of the numerous shortening is aboutan orderof magnitudeless,a
smallNeogene(MioceneandPliocene)basinsin basalTriassicevaporiteis absentor not well
the regionexhibitmarkedonlap,and someeven
downlap,suggesting activeNeogenetectonism. developed.Figure5 showsan examplefrom the
northeast of an anticline whose deformation
involves almost the entire Phanerozoic column.
Shortening The Doubayat-2well penetratesto a total depth
of 3614 m (about 2.0 secondson the seismic
Palinspasticreconstructions
of crosssections sectionin Figure5) not far from the top of
based on seismic reflection data tied with well Precambrian basement and records no faults in the
informationand constrainedby surfacegeology section. Any detachmentlevel in the northeastern
yieldminimumcrustalshortening valuesmuch regionmustlie deeperthanthe Doubayatwell,
lower thanthe expected80 km. An estimateof which bottoms in the Ordovician.
minimum total shorteningfor the three sections In the southwestthe presenceof a d6collement
yieldedfrom southwest to northeast 20 km, 3 km, surfacein the lower Triassic has important
and about 1 km, respectively. Plate 1 and Figures implicationsfor hydrocarbon
explorationin the
2 and 3 showthe threecompositeline drawings Palmyfides.If a lowerTriassicevaporitehorizon
and their respectivedepth-convertedcross decouples the overlyingMesozoicandCenozoic
sections in the deformed and undeformed states. sediments from the Paleozoic section where
Their locationsare shownin Figure lb. hydrocarbon
potentialis good[Ponikerovet al.,
Stratigraphic
picks on the crosssectionsare 1967], structuresin the Paleozoicneednot mimic
basedon seismiccharacterand only limited well those in the Mesozoic and Cenozoic. Hence,
data and so are only approximate. particularcareshouldbe givento the processing
The presenceand absence of a detachment in of deeper(Paleozoic)seismicdata,especiallyin
Triassic (?) strata,possiblylocatedalong an the southwesternPalmyfideswhere a detachment
evaporitebed, likely controlsthe vastlydifferent clearlyexistsand Paleozoicstrataare decoupled
deformationalstylespresentin the southwestern from overlyingMesozoicand Cenozoicstrata.
 Depth (km)
o o od l-

I
oo

LLI a
i
i
i
i o:j

o
(D

I
I
I (_)

Wo_
oO'
LL
o ooo
-- o

ooo,
o

o(

o,

o,

E
o

z
o -- & o
o
Chaimov et al.' Palmyride shorteningand Dead Sea Fault Movement 1377

Depth (km)
o

o,

oO

Oo

o,

oo

oc

oo:, oo

o %,
%0
_

oo

o :oO LJ.J
ooo

OOoO ,

o oo o

oo

'_L :Oo?
1378 Chaimov et al.: Palmyride shorteningand Dead Sea Fault Movement

NW $

Fig. 4. An exampleof Lower Triassic(?) d6collementin the southwestern Palmyrides

decouplingrelatively undeformedPaleozoicrocksfrom overlyingMesozoicstrata. The
locationof this excerptfrom crosssectionA is shownin Figure lb. The seismic
manifestationof the Jharfault occupiesthe left-mostpart of the section.

Another importantimplicationof the DISCUSSION

interpretationof a detachmentis the necessityto
continuethe d6collementnorthwardbeyondthe A variety of possibilitiesexiststo explainthe
northern end of the cross section, that is, noah of obviousdiscrepancy betweenthe expected(80
the Jharfault. The interpretationas presentedin km) and observed(20 km) shorteningvaluesin
crosssectionA (Plate lb) requiresthis northward the Palmyridebelt. First, and leastlikely of the
extension of the d6collement. It is unknown possibilities,is that the interpretivesections
whether the dcollement continues at the same introducedin this studymay not reflect actual
stratigraphiclevel for hundredsof kilometersinto shorteningin Syria. It is well known that
southernTurkey where it meets the Bitlis suture, balancedand restoredcrosssectionsyield
whetherit rampsdown deeperand eventually conservativeestimatesof shortening[e.g.,
soles into the middle or lower crust, or whether it Hossack,1979]. It shouldbe kept in mind that
rampsup to the north, essentiallydecapitatingthe shorteningestimatesbasedon this data set, a
Aleppo plateau. smallportion of that which existsin the region,
Most of the Palmyfide shorteningis are minimum estimates. Second,strike-slipfaults
accommodated in this southwesternregion. 20 in the Aleppo plateauand in the Palmyridesmust
km of NW-SE shorteningin southwestern central accommodate at leastsomeof the "missing"60
Syria (near Damascus)is clearly documentedin km of expectedcrustalshortening,although,the
the new crosssections. But the magnitudeof apparentstability(lack of seismicity,subdued
shorteningpredictedby kinematicmodelswas topography,and monotonousPaleogeneoutcrop)
about 80 km. There still remain 60 km of crustal of the Aleppo plateauimpliesonly minor
shorteningunaccounted
for. tectonismthere. A third possibilityis that the
Chaimov et al.' Palmyride shorteningand Dead Sea Fault Movement 1379

NW SE

0.0

z - Top Pz
o 1.0

Doubayat-2well 2 KM

Fig. 5. An examplefrom the northeastern

Palmyfidesshowingno evidenceof a
d6collement surface. Paleozoic strata are involved in the deformation. The location of
this excerpt,part of crosssectionC, is shownin Figure lb.

estimatesof 25 km of slip alongthe northern implicit in constructingany balancedcrosssection

segmentof the Dead Sea transformfault system acrossa fold and thrustbelt, amongthem the
are too low. The elaboratemappingof the area assumptions of a regionald6collementhorizon,
arguesstrongly against such an assertion. two-dimensional deformation, and constant bed
Finally, the original apparent80 km discrepancy thickness.Many "rules"of classicfold and thrust
between northern and southern Dead Sea fault belts are violated by the Palmyfide belt. First, no
systemoffsets may be false. It is possiblethat obviousregionaldetachmentsurfaceimagedby
duringthe first 10-15 m.y. of motionon the Dead the seismic underlies the entire belt. In the Atlas
Sea transformfault (about 60 km of strike-slip mountainsof Tunisia, a possiblehomologto the
motion), the fault itself did not extend into what is Palmyfides,Snokeet al. [1988] also see no
now centralLebanonand northwesternSyria. In regionaldetachment,but ratherreactivatednormal
effect, this would mean that there really is no faults. If this were the casein the Palmyfides,
discrepancy,and that the 20-25 km of motionon estimatesof shorteningbasedon balancedcross
the northernextensionof the Dead Sea fault plus sections would be even lower than 20 km because
the 20-25 km shorteningin the Palmyfidesis all steepreactivatednormal faults cannot
that existsin northernSyria. Each of these accommodate much shortening.Seismicdata
possibilitieswill now be examinedin detail. provethe existenceof a detachmentsurfaceonly
in the southwesternPalmyrides.
Limitations of Balanced Sections Furthermore,both surfacemappingand
earthquakefocal mechanismsdocumentstrike slip
Dahlstrom [1969] and others[e.g., Elliot, 1983; in the mountainbelt. The mostprominentmapped
De Paor, 1988] detail the many assumptions fault in the Palmyfides,and the northernlimit of all
1380 Chaimovet al.: Palmyrideshorteningand Dead Sea Fault Movement

three crosssectionspresentedin this study,is restored. Strictly, the term retrodeformableis, in

the Jharfault, a 170 km long steepfault that reference to these cross sections,more accurate
strikesroughly E-W and separatesthe than balanced[Suppe, 1985]. While balanced
southwestern Palmyfideblock from the Bilas and crosssectionsin thrustbelts by definitionrestore
Bishriblocks. Judgingfrom its linearcontinuity to a minimumpossibleundeformed length
and steepdip, the Jharlikely has beena strike- [Hossack,1979] it is difficultto imaginethatthe
slip fault. Actual magnitudeof lateral offsets calculated20 km shortening in the moststrongly
acrossthe Jharfault, however,are extremely deformedsouthwestern regionof the Palmyfides
difficult to quantify. Associatedwith otherfaults, is a factorof fourlowerthanactualshortening.If
two recentmb-4.8 earthquakes, oneon October5, the Palmyfideswere truly shortened by the
1970 andoneon May 20, 1987,providedataon expected80 km, then this impliesthat the now
the presentday senseof motionin the 100 km wide fold belt was once 180 km wide, and
Palmyfides. Focal mechanismsfrom both events surelyalmost50% shorteningwould give rise to a
showthat strikeandreverseslip are still activein very prominentmountainbelt. But the Palmyfides
the Palmyfides. The focal mechanismsare based have on the averageonly a few hundredmetersof
on all availableup-m-datedataincludingthat relief. Becausegravitymodelingshowsthe crust
reportedin the InternationalSeismologicalCenter beneaththe Palmyfidesto be of typicalcontinental
andPreliminaryDeterminationof Epicenters thickness,about40 km [Bestet al., 1990], simple
bulletins.Additionalfirst motionpolarityreadings volumeffic calculations indicate that the crust
were obtained from World Wide Standardized wouldhaveto havebeenonly24 km thickpriorto
Seismic Network records. A focal mechanism of thepostulated 80 km of shortening.Continental
the 1970 eventwaspublishedin 1976 [Ben- crustthis thin wouldalmostcertainlybe
Menahem et al., 1976], but was basedon associated with shallow if not surface volcanics as
relativelylimited observations
from only a few are seen,for example,in the BasinandRange
seismic stations. Our new focal mechanisms Province of the western United States where the
indicatemoderatelysteep(N70) SSE-dipping crustis thin. With exposureof Jurassic
through
fault planeswith an obliquesenseof motionabout Holocenerocks,the conspicuous paucityof
equallydividedbetweenreverseand fight-lateral volcanicsin the Palmyfidesprecludessuch
strike-slip. The locationsof the earthquakesand markedshortening.Minor occurrences of volcanic
their associatedfault plane solutionsare shownin rockshave,however,beenreportedfrom a few
Figure lb. wells.
Surfacemappingat a scaleof 1:200,000shows Theselinesof evidence,the mutualconsistency
little if any evidenceof strikeslipin the study of the threecrosssections,the goodassumption
area, so strike-slipmotion, while it is a factorin of two-dimensional deformation in the study
Palmyfidedeformation, is notof majorconcernin region,andtheimprobability of onceverythin
this study. The sectionspresented in thispaper crust,all lendcredenceto the 20 km shortening
were constructedassumingtwo-dimensional estimate.
deformation.Three parallelcrosssectionswere
constructed alongstriketo allay concernsabout SidewaysEscapeof Northern Syria
this assumption.Sincethe three sectionsgive
mutuallyconsistentresults,that is, they do not Assumingthat the shorteningestimatedhere
show that shorteningvaries in a randommanner is correct suggeststhat someother mechanismis
alongstrike,then the assumption of two- responsiblefor accommodating the expected80
dimensionaldeformationheremay not be a bad km of shortening.Strike-slipmotionobliqueto the
one. Furthermore,if NE-SW strike slip were the crosssectionsmay accommodatethe remainderof
majorcontrolof the foldsin the Palmyfides,one the shortening,or at leasta portionof it. Both
might expectto seeunevenlythickenedand mappedfaults,the Jharfault, for example,and
thinnedbedsin a givensection. No stretched earthquakefocal mechanismssupportthis
strata have been observed. assertion.
Finally, deformationin the sedimentarysection Jacksonand McKenzie [1988] review two end
extendsbeyondthe southernpaleobasinmargin member ways in which the deformationbetween
as definedby the rapidthinningof Mesozoic two convergingplatestakesplace (seeFigure6).
strata,therebyfurthercomplicating the balancing Oneendmemberis simplefoldingandthrusting in
process.For example,in crosssectionC (Figure the directionof convergenceas in the caseof
3b) where the Jurassicand Triassic section thins someregionsin western North America. The
markedlyto the southindicatingthemarginof the othermodeinvolvesthe sidewayswedgingout or
Mesozoic basin, at least three small folds occur squirtingout of a relativelyrigid continentalblock
still farther south. as summarizedby Burke and SengOr[1986], for
With all thesecomplicationsin mind,the cross example,in the westernTurkey case. This
sectionswere balancedto the degreepossibleand secondmoderequiresstrike-slipmotionobliqueto
Chaimov et al.: Palmyride shorteningand Dead Sea Fault Movement 1381

a)EuAstA strike slip may be subtlydistributedthroughout

the belt. While numerousvertical (probably
I
strike-slip)faults dissectthe Aleppo plateauto
the north (Figure 7), the magnitudeof
displacementacrossthem is unknown. While
recentfault slip data indicatea fight-lateral
componentin the Palmyfides,Walley [1988]
offers a scenariowherebyfaults in the Palmyfides
comprisea left-lateral braidedstrike-slipsystem
that distributesthe excessshorteningthroughout
centraland northernSyria (seeFigure 8). Strike-
AABIA slip motionplays at leasta small role in shaping
the Palmyfidebelt, and may accountfor over 30
km of northwardArabianmotionin northernSyria
b)EURA
SIA [Walley, 1988]. The net senseand magnitudeof
strike-slipmotion in the Palmyfideshave yet to be
well determined,and a thoroughinvestigationof
strike-slipmotion in Syria is warranted.

Uncharted Offsets on the Dead Sea Fault

System

Another viable possibility is that the estimates

of about25 km of left-lateral slip on the northern
ARABIA segmentof the Dead Sea transformfault system
are too low. Based on his kinematic model of
Cenozoic easternMediterraneanplate motions
Fig. 6. Map views showingtwo end
Quennell[ 1984] reported30 km of displacement
memberways in which the zone between
alongone of the two major strandsof the northern
the Arabian and Eurasianplatescould Dead Sea transformfault. Similarly Trifonov et al.
respondto convergence:(a) thrustingand [ 1983] and V. Trifonov (personalcommunication,
foldingparallelto convergence;(b) strike 1989) documentonly 20-25 km of slip on the
slip obliqueto convergence
[modifiedfrom northern Dead Sea faults, based on offset mid-
Courtillot et al., 1987].
Miocene and youngervolcanicsand on a 0.5-0.6
cm/yr slip rate calculatedfrom carbon-dated
Quaternaryalluvium offsets. In particular,V.
the directionof maximum shorteningas well as Trifonov (personalcommunication,1989) cites
essentiallya free surface,for example,an offset of 5-6 m.y. old basalts. At a constantpost-
oceanic/continental plate boundary,throughwhich Miocene (post-6m.y.) rate of 0.5 cm/yr this is in
the continentalblock can escape. The northeast- goodagreementwith the total 25 km offset.
trendingPalmyridebelt strikesapproximately45 However, mostof the recognized105 km of left-
to the northward encroachment of the Arabian lateral motion on the southernsegmentof the
plate. An eastwardescapemodel at first looks Dead Sea transformfault occurredduring the past
deskablein the north Syrian setting,with left- 20 m.y. [Quennell, 1958 and many others].
lateral motion along the Dead Sea and East Hypothesizinga minimumpre-Pliocene(pre-6
Anatolian fault systemson the west and north, m.y.) slip rate of 0.5 cm/yr in the north,only 12
respectively,and possiblefight-lateralmotion in m.y. are requiredto accountfor the missing60 km
the Palmyrides[Quennell, 1984], but northern on the northernpart. In other words,if the
Syria has no free surfaceto the east through northernDead Sea fault systemhad been active
which it can escapeas westernTurkey does. for the entire 20 m.y. that the southernsegment
Indentationof the Euphratesgraben has, then large unrecognizedoffsetsexist. Such
northeastwardalong strike from the Palmyfides offsetswouldprobablybe difficult to map along
(see Figure la) is also not observed,so eastward the northernDead Sea faults where young
escapeof northernSyria as a singleblock from volcanicsand sedimentsobscuregeologic
Arabia is not likely. Furthermore,a N-S structure[Quennell, 1984] and where the Dead
componentof 60 km of shorteningwould be Sea systembifurcatesand becomescomplex.
manifestas over 80 km of strike slip in the NE- Miocene and youngeroffsetson the southern
trendingPalmyfides. There is no obvious Dead Sea faults are seen in Precambrian, Triassic,
evidencefor this amountof strike slip on any one and Jurassicrocks [Courtillot et al., 1987] that
fault in the Palmyrides,although,considerable strikeobliquelyto the fault. Theseoffsetsare
1382 Chaimov et al.' Palmyride shorteningand Dead Sea Fault Movement

NW SE

.0

t-
O
0

.0

.0

2 km

Fig. 7. An examplefrom the northernAleppoplateaushowinginterpretation of a

typical flower structuresuchas commonlyresultsfrom strike-slipmotion. The
locationof this excerptis shownin Figure la.
Chaimovet al.: Palmyfideshorteningand Dead Sea Fault Movement 1383

thusinherentlyeasierto map than would be Pre-PlioceneIndependenceof Northernand

offsetson the northernsegmentwhere strata SouthernDead Sea Fault Segments
strikeat a muchlower angleto the N-S trending
faults. Acceptinga maximumof 25 km of The final remainingmodelthatcan accountfor
recognizeddisplacement,a further60 km of the missing60 km is depictedin Figure9. Figure
displacementon the northernsegmentof the Dead 9 showsa scenarioin which slip alongthe
Sea fault systemmay still remain uncharted. northernDead Sea fault systemwas not initiated
before6 Ma. This implies that thereare indeed
only 45 km of convergence to accountfor in
northernSyria, 20-25 km in the Palmyfidesand
the acknowledged20-25 km on the northern
segmentof the Dead Sea transformfault. In this
last case,the northernand southernsegmentsof
the Dead Sea fault systemfunctioned
independently until the Early Pliocene(5-6 Ma.).
The Roum fault in southernLebanon, the seaward
splayoff the southernDead Seafault in Figure9,
may representthe extinct northward extensionof
the Miocene Dead Sea transform fault.
Paleomagneticdata from the Levant [Van
Dongenet al., 1967; Zijderveldand Van der Voo,
1973] bearout this tectonicmodel. Sparse
volcanicsamplesfrom thisregionindicateas much
as a 28 counterclockwise rotation of the Levant
relative to Africa since the Lower Cretaceous.
Sincethat time the Arabianplatformhasrotated
in the samesenseonly 7 relativeto Africa. The
excessrotation of the Levant, most of which can
Fig. 8. A possiblemodel that invokes be takenas post-Cretaceous basedon the timing
distributedstrikeslip to accommodate of the majorregionaltectonism,may be
excessshortening[modifiedfrom Walley, attributableto a poor statisticalsamplingof rocks
1988]. This modelrequiresa totalof about [Van Dongenet al., 1967] or it may in part
80 km of NE-SW strike-slipmotion. representa real counterclockwiserotation of

Fig. 9. Onepossiblekinematicmodelto explainthe apparent discrepancybetween

left-lateraloffsetsobservedon the segmentsof the northernversusthe southern
Dead Sea fault system. The Roumfault may have beenthe main northward
continuationof the Dead Sea systemduringthe Miocene.
1384 Chaimov et al.: Palmyride shorteningand Dead Sea Fault Movement

northernLebanon and northwesternSyria as seismicreflectiondata. The studyalsopresents

depictedin Figure 9. Certainlymorein-depth severalmodelsthat accountfor the apparent
palcomagneticstudiesneed to be completed discrepancyin observedleft-lateral offsetsacross
beforegreatconfidencecan be placedin sucha the northern versus the southern Dead Sea
model;however,the simplicityof sucha model transformfault systemas well as an examination
holdsconsiderableappeal. of theviability of eachmodel.
The Dead Sea systemshowsevidenceof over
Summary 105 km of left-lateral displacementon its southern
segmentyet only 25 km on its northernsegment,
In view of the four very differentandin some leavingan apparent80 km of crustalmovement
casesmutually exclusivemodelsproposed,some unaccountedfor. Palinspasticallyrestoredcross
discussionof the viability of eachis required. sectionsthroughthe Palmyfidefold belt in Syria
After examiningcloselyall the models,only two of showthat at least20 km of the "missing"80 km is
the four possibilitiesremain viable, that strike-slip accommodated by thrustingandfoldingin the
faults accommodatesomeshorteningor that the southwesternPalmyrides. At least four
apparent80 km of expectedshorteningis illusive, possibilitiescan explain the still elusive60 km of
and the northernand southernsegmentsof the crustalshorteningin northernArabia. The 20 km
Dead Sea systemdevelopedindependently. predictedby the retrodeformablecrosssections
First, the improbablemodelsmust be may be almosta factorof four lower than actual
discounted.Clearly the first possibility,that this shorteningin the Palmyfides,but, becausewe
balancedcrosssectionstudyis wrong, is estimatea present-dayN40 km thick crust,this
undesirable. For reasonsalready discussed, would necessitatethe existenceof unreasonably
includingmutually consistentresultsthroughout thin crustprior to shortening.The very rare
the 400 km-long belt, and confidencein the then three possibilitiesremain. (1) Distributed
assumptionof two-dimensionaldeformation,this strikeslip in northernSyria accountsfor 80
possibilityis unlikely. Equallyunlikelyis the kilometersof NE-SW strike-slipmotion,which
possibilitythat the reliable scientistswho for has not been observed. New focal mechanisms
years scrutinizedthe northernDead Sea fault supportobservationsof strike slip in the field, so
systemin aerial and satelliteimagesand in the someas yet undeterminedamountof strike slip
field [Ponikerov et al., 1963; Trifonov et al., 1983; must exist in the Palmyfides. (2) There remain
Quennell, 1984;] missed tens of kilometers of left- 60 kilometersof unchartedleft-lateral slip on the
lateral offset. Discountingthesetwo scenarios northernsegmentof the Dead Sea transform,a
leavesonly two viable modelsto explain the very unlikely possibility. (3) The northernand
shorteningmeasurements which are seemingly southernsegmentsof the Dead Sea fault system
inconsistentwith other observationsin the region. moved,at least duringMiocene time,
Strike slip surelyplaysa role in accommodating independently,as Quennell[1958] suggested.In
shorteningin the Palmyridesand northernSyria, other words, the northernsegmentof the Dead
but to what extent is unknown. Evidence of Sea transformfault systemis much youngerthan
strike-slipfaulting is readily observedin the field, the southernpart, and thereare only 45
but so is thrustfaulting. The magnitudeof strike kilometersof crustalconvergenceexpressedin
slip in the Palmyrides,while significant,remains northernSyria.
to be quantified. Strike slip mustexplainat least Documentedstrike-slipmotion in the
someof the observedenigma. Palmyfidesimmediatelylendssupportto the
The only otherviable modelof thoseproposed hypothesisthat strike-slipmotion accommodates
is that the main left-lateral movement on the at least someof the remainingdisplacement.In
northernsegmentof the Dead Sea fault system addition,an intriguingpossibilityintroduced
by
was initiatedonly after the southernsegment, Quennell[1958] is that the northernsegmentof
includingthe Roum fault, hadaccruedabout60 km the Dead Sea fault systemwas not coupledto the
of pre-Plioceneoffset. While palcomagneticdata southernsegmentprior to the Pliocene. This
supportthis model, thesedataunfortunatelyare would meanthe 80 km of "missing"displacement
few and so perhapsnot per se convincing. is illusive, and that there really has been a total of
Nevertheless,this model is intriguingand is only 45 km of N-S shortening in Syria. The truth
attractivein its simplicity. likely lies in somecombinationof thesetwo
models.
CONCLUSIONS
Acknowledgments.,Partial fundingfor this
This studypresentsa quantitativemeasureof researchwas providedby Amoco, ARCO, Exxon,
the magnitudeof shortening in the Palmyfidefold Marathon,and Unocal oil companies.Thanksare
belt in Syria,basedprimarilyon newly available due Bob Litak, Art Barnes,and especiallyJohn
Chaimov et al.: Palmyride shorteningand Dead Sea Fault Movement 1385

Best for their thought provoking discussions,and the Mediterranean and Middle East, Geophys.
R. Allmendinger,D. E. Karig, C. De Paor, and one J., 93, 45-73, 1988.
anonymousreviewer for their critical reviewsof Le Pichon,X., and J. Francheteau,A plate-tectonic
the manuscript. Our appreciationis extendedalso analysisof the Red Sea-Gulf of Aden area,
to Robin Adams for providingimportant Tectonophysics,46, 369-406, 1978.
seismologicaldata. Institute for the Study of the Lovelock, P. E. R., A review of the tectonics of the
Continentsat Cornell University contribution125. northernMiddle East region, Geol. Mag., 121,
577-587, 1984.
McBride, J., M. Barazangi, J. Best, D. A1-Saad,T.
Sawaf, M. A1-Otri, A. Gebran, Seismic
reflection structure of the intracratonic
REFERENCES
Palmyridefold-thrustbelt and surrounding
Arabian platform in Syria, Am. Assoc.Pet.
Ben-Menahem, A., A. Nur, and M. Vered, Geol. Bull., 74, 238-259, 1990.
Tectonics, seismicityand structureof the Afro- Mount, V. S., and J. Suppe, State of stressnear
Eurasianjunction-Thebreakingof an incoherent the San AndreasFault: Implicationsfor wrench
plate,Phys.Earth Planet. Inter., 12, 1-50, tectonics,Geology,15, 1143-1146, 1987.
1976.
Ponikerov, V. P. (Ed.), The Geological Map of
Best, J. A., M. Barazangi,D. A1-Saad,T. Sawaf, Syria, scale 1:200,000, Ministry of Industry,
A. Gebran, Bouguergravity trendsand crustal Damascus,Syria, 1963.
structureof Palmyfide mountainbelt and Ponikerov, V. P., V. G. Kazmin, I. A. Mikhailov,
surroundingnorthernArabianplatform in Syria, A. V. Razvaliayev, V. A. Krasheninnikov,V. V.
Geology,in press,1990. Kozlov, E. D. Soulidi-Kondratiyew,and V. A.
Beydoun,Z. R., Someopenquestionsrelatingto Faradzhev, Explanatory notes, The Geological
the petroleumprospects
of Lebanon,J. Pet. Map of Syria, scale1' 1,000,000, 111 pp.,
Geol., 3, 303-314, 1981. Ministry of Industry,Damascus,Syria, 1966.
Burke, K. and C. Seng6r,Tectonicescapein the Ponikerov, V. P., V. G. Kazmin, I. A. Mikhailov,
evolution of the continentalcrust, in Reflection A. V. Razvaliayev, V. A. Krasheninnikov,V. V.
Seismology:The ContinentalCrust, Kozlov, E. D. Soulidi-Kondratiyew,K. Ya.
GeodynamicsSeries,vol. 14, edited by M. Mikhailov, V. V. Kulakov, V. A. Faradzhev,
BarazangiandL. Brown,pp. 41-53, AGU, and K. M. Mirzayev, ExplanatorynotesPart I,
Washington,D.C., 1986. The GeologicalMap of Syria, scale1:500,000,
Courtillot,V., R. Armijo, P. Tapponnier, 230 pp., Ministry of Industry,Damascus,Syria,
Kinematicsof the Sinai triplejunctionanda 1967.
two-phasemodelof Arabia-Africarifting,in Quennell, A.M., The structuraland geomorphic
Continental Extensional Tectonics,Spec. Publ. evidence of the Dead Sea Rift, Q. J. Geol. Soc.
28, edited by M.P. Coward,J. F. Dewey, and London, 114, 1-24, 1958.
P. L. Hancock,pp. 559-573, GeologicalSociety Quennell, A.M., The western Arabian rift system,
of London 1987. in The GeologicalEvolutionof the Eastern
Dahlstrom, C. D. A., Balanced cross sections, Mediterranean, Spec.Publ. 17, editedby J. F.
Can. J. Earth Sci., 6, 743-757, 1969. Dixon and A. H. F. Robertson,pp. 775-788,
De Paor, C., Balanced sectionin thrust belts part GeologicalSocietyof London,1984.
I: Construction, Am. Assoc. Pet. Geol. Snoke, A. W., S. Schamel, and R. M. Karasek,
Bull.,72, 73-90, 1988. Structuralevolutionof Djebal Debadib anticline:
Elliott, D., The constructionof balanced cross- A clue to the regional tectonicstyle of the
sections, J. Struct. Geol., 5, 101, 1983. Tunisian Atlas, Tectonics, 7, 497-516, 1988.
Freund, R., Z. Garfunkel, I. Zak, M. Goldberg,T. Suppe,J., Principlesof StructuralGeology,
Weissbrod, and B. Derin, The shear along the Prentice Hall, Englewood Cliffs, New Jersey,
Dead Sea rift, Philos. Trans. R. Soc. London, 1985.
267, 107-130, 1970. Trifonov, V., A. Youssef, Y. al-Khair, and T. Zaza,
Hempton, M., Constraintson Arabian plate Using satelliteimagery to infer the tectonics
motion and extensionalhistory of the Red Sea, and the petroleumgeologyof Syria, report,p. 1-
Tectonics, 6, 687-705, 1987. 18, Syrian PetroleumCompany,Damascus,
Hossack, J. R., The use of balanced cross-sections Syria, 1983.
in the calculationof orogeniccontraction:a Van Dongen, P. G., R. Van der Voo, and T.
review, J. Geol. Soc. London, 136, 705-711, Raven, Paleomagnetismand the Alpine
1979. tectonicsof Eurasia,part III, paleomagnetic
Jackson,J. and D. McKenzie, The relationship research in the Central Lebanon Mountains and
betweenplate motions and seismicmoment in the Tartous area (Syria), Tectonophysics,4,
tensors, and the rates of active deformation in 35-53, 1967.
1386 Chaimovet al.: PalmyrideShorteningand Dead SeaFault Movement

Walley, C. D., A braidedstrike-slipmodel for the stressof the San Andreasfault system,
northern continuation of the Dead Sea Fault Science, 238, 1105-1111, 1987.
and its implicationsto Levantinetectonics,
Tectonophysics,145, 63-72, 1988.
Woodcock,N.H., Strike-slipduplexes,J. Struct. D. A1-Saad,A. Gebran, and T. Sawaf,Syrian
Geol., 8, 725-735, 1986. PetroleumCompany,Ministry of Petroleumand
Zijderveld,J. D. A. andR. Van der Voo, Mineral Resources,Damascus,Syrian Arab
Paleomagnetismin the Mediterraneanarea, in Republic
Implicationsof ContinentalDrift to the Earth M. Barazangi and T. A. Chaimov, Institute for
Sciences,vol. 1, editedby D. H. Tarling and S the Studyof the ContinentsandDepartmentof
K. Runcorn,pp. 133-161, Academic,SanDiego, GeologicalSciences,CornellUniversity,Ithaca,
Calif., 1973. New York 14853
Zoback, M.D., M. L. Zoback, V. S. Mount, J.
Suppe,J.P. Eaton,J. H. Healy, D.
Oppenheimer,P. Reasenberg, L. Jones,C. B. (ReceivedJanuary16, 1990;
Raleigh,I. G. Wong, O. Scotti, and C. revised May 9, 1990;
Wentworth, New evidence on the state of acceptedMay 10, 1990.)