Terminal Pleistocene and Early Holocene archaeology and stratigraphy of the southern Nejd, Oman

lable at ScienceDirect

Quaternary International xxx (2015) 1e14

Contents lists avai

Quaternary International

journal homepage: www.elsevier .com/locate/quaint

Terminal Pleistocene and Early Holocene archaeology and stratigraphyof the southern Nejd, Oman

Y.H. Hilbert a, *, A. Parton b, M.W. Morley c, d, L.P. Linnenlucke c, Z. Jacobs c,L. Clark-Balzan b, R.G. Roberts c, C.S. Galletti e, J.-L. Schwenninger b, J.I. Rose f

a CNRS, UMR 5133 ‘Arch�eorient’, Maison de l'Orient et de la M�editerran�ee, Lyon, Franceb School of Archaeology, Research Laboratory for Archaeology and the History of Art, University of Oxford, Oxford, UKc Centre for Archaeological Science, School of Earth and Environmental Sciences, University of Wollongong, NSW 2522, Australiad Department of Anthropology and Geography, Oxford Brookes University, Oxford, UKe School of Geographical Science and Urban Planning, Arizona State University, Tempe, AZ, USAf Ronin Institute, Montclair, NJ, USA

a r t i c l e i n f o

Article history:Available online xxx

Keywords:Late PalaeolithicGeoarchaeologyLithic technologyOSL datingSouth Arabia

* Corresponding author.E-mail address: [email protected] (Y.H. Hi

http://dx.doi.org/10.1016/j.quaint.2015.02.0531040-6182/© 2015 Elsevier Ltd and INQUA. All rights

Please cite this article in press as: Hilbert, YNejd, Oman, Quaternary International (2015

a b s t r a c t

Archaeological, geoarchaeological and geochronological research at the Terminal Pleistocene/Early Ho-locene sites of al-Hatab, Ghazal and Khamseen Rockshelters help to elucidate our understanding of theLate Palaeolithic occupation of the South Arabian Highlands. The stone tools found at these sites areattributed to the Nejd Leptolithic tradition; an autochthonous lithic culture found across Dhofar (Oman).The artifact-bearing deposits excavated within these rockshelters have been chronologically constrainedby optically stimulated luminescence dating, while sedimentological analyses help to establish siteformation processes. Together, these methods provide a chronological anchor for some Nejd Leptolithicarchaeological findspots across Dhofar. The archaeological evidence from surface and stratified sitesindicates a technological continuum across the shifting climatic regimes of the Terminal Pleistocene/Early Holocene. This technological continuity points to possible population persistence within one of theposited South Arabian refugia.

© 2015 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction

It has been previously suggested that the hyperarid conditionsin Arabia during Marine Isotope Stage (MIS) 2 (ca. 30e10 ka)affected a discontinuity between Palaeolithic and Neolithic pop-ulations (e.g. Drechsler, 2007; Uerpmann et al., 2009; Bretzke et al.,2013). Archaeological and genetic data from the southern portion ofthe Peninsula indicate, however, that not only was there a TerminalPleistocene human occupation in this area, but that distinct humangroups may have coexisted in the region during the Early Holocene(Al-Abri et al., 2012; Hilbert et al., 2012; Hilbert, 2013, 2014; Roseet al., 2013; Zarins, 2013). Consequently, as our understanding ofthe South Arabian Palaeolithic grows, so does the need to developlocal palaeoclimatic sequences that provide environmental con-texts for the various lithic industries found across these territories.At present, the palaeoenvironmental resolution for much of thisvast region, which encompasses all territories south of (and

lbert).

reserved.

.H., et al., Terminal Pleistocen), http://dx.doi.org/10.1016/j.

including) the Rub' al-Khali remains coarse, as does our knowledgeof the Terminal Pleistocene/Early Holocene human occupation ofthis area (e.g. Maher, 2009; Rose and Usik, 2009). The majority ofsites attributed to the Late Palaeolithic (LP) are undated surfacesites. The term LP was first used to describe the pre-Neolithicoccupation of Arabia by Drechsler (2009) and is here consideredto represent the time period between 15 and 8 ka. This term wasdeliberately chosen for use in Dhofar over “Epipalaeolithic” and“Mesolithic” given the specific techno-typological connotationsassociated with these names (Maher, 2009); both are characterizedby bladelet production and microlithic technology, elements thatare absent in the pre-Neolithic archaeological record of the Nejd(“plateau” in Arabic) (Hilbert et al., 2012; Hilbert, 2014).

To augment the available archaeological and palaeoclimaticdata, this paper describes a series of local environmental proxyrecords and associated archaeological remains from Dhofar. Thedated sites al-Hatab, Ghazal, and Khamseen Rockshelters are usedto assess climatic fluctuations within this region during the Ter-minal Pleistocene/Early Holocene (TP/EH).

e and Early Holocene archaeology and stratigraphy of the southernquaint.2015.02.053

mailto:[email protected]

www.sciencedirect.com/science/journal/10406182

http://www.elsevier.com/locate/quaint

http://dx.doi.org/10.1016/j.quaint.2015.02.053



Y.H. Hilbert et al. / Quaternary International xxx (2015) 1e142

2. Dhofar: contextual and environmental background

The Governorate of Dhofar is partitioned into six distinctecological zones: the coastal plain; the summit grasslands; theseaward slopes and the southern draining wadis; the northerndraining wadis on the northern face of the escarpment; the dryplateau; and the interior desert (Fig. 1) (Miller and Morris, 1988).These landscapes and their geological context will be describedhere in a northesouth transect.

Northern Dhofar is classified as open desert, comprising flatgravel plains and the Rub' al-Khali sand sea, where sand dunes riseover 100 m in height and are superimposed over limestone plateaulandscape features. Vegetation across the open desert is generallysparse. The dry plateau, also known as the Nejd, is a rocky, mediumrelief landscape. Elevation increases markedly to the south where itborders the watershed divide of the Dhofar Mountains. Northwarddraining wadis have carved long, river valleys that debouch fromthe Dhofar Mountains to the open desert in the north (Fig. 2).Although these valleys today are highly ephemeral and largely dry,the relict fluvial geomorphology suggests former perennial flowsand wide, meandering channels. However, dense stands of vege-tation can be found in wadi basins where moisture is directedfollowing precipitation events, and in the southernmost areaswhere the monsoon rain shadow provides cooling winds duringthe summer. The Nejd dips gently in elevation towards the

Fig. 1. Ecological zones of Dhofar. Map showing the different eco-zones and their distributioet al., 2011, Fig. 3).

Please cite this article in press as: Hilbert, Y.H., et al., Terminal PleistocenNejd, Oman, Quaternary International (2015), http://dx.doi.org/10.1016/j.

northeast, stretching 250 km from the Yemen border to the Jiddatal-Harassis gravel plain, which marks the eastern edge of theplateau. The dry plateau is composed of two horizontally bedded,Paleocene to Late Eocene, shallow-marine carbonate shelf units, theUpper and the Lower Hadhramaut Group. These consist of alter-nating limestone and dolomitic chalk beds, containing severalchert-bearing layers (Lepvrier et al., 2002).

The summit grasslands and seaward slopes form a partitionbetween the monsoon-affected areas and the dry plateau. Grassesare the most common land cover on the summit, but trees andshrubs of the Somali-Masai phytogeographic zone are also found(Gallagher, 1977; Gazanfar and Fisher, 1998; Gazanfar, 1999). Theseaward slopes are situated at the northern extent of summermonsoon rainfall and as a result are heavily vegetated. Forests arecommon on the seaward slopes and include characteristic treespecies such as Anogeissus dhofarica and Commiphora spp. With theexception of a small area to the south-west of Jebel Qara, whichcomprises lacustrine and shallow marine limestone of the DhofarGroup, the mountain range is formed exclusively from rocks of theHadhramaut Group. The coastal plain is mostly flat and covered byfluvial/alluvial sediment coming from the slopes of the DhofarMountains. In areas where the basement geology is exposed car-bonate and conglomeratic deposits (Fars Group) are found, whilstto the east between Mirbat and Hasik, crystalline basement rocksare apparent (Platel et al., 1992; Lepvrier et al., 2002). Vegetation

n across Dhofar. Eco-zones are based on elevation, vegetation and hydrology (after Rose


Fig. 2. UmmalKhashabpanorama.Panoramicphotographsofdifferent terraintypes foundacross theUmmalKhashab.Typicalare thesteepescarpments, lowinselbergsand tablemountainsas well as themeandering wadi systems. a) southwest of Jebel Ardif; b) primary drainages leading to Wadi Dawkah at the foot of the northern facing slopes of the Dhofar escarpment.

Y.H. Hilbert et al. / Quaternary International xxx (2015) 1e14 3

across the coastal plain tends to be concentrated in the foothills atthe base of the seaward slopes where moisture availability ishighest, as well as in wadi courses draining into the ocean.

The palaeoclimatic variability of southern Arabia is a function ofthe latitudinal variability of the Intertropical Convergence Zone(ITCZ) and associated Indian Ocean Summer Monsoon (IOSM)rainfall belt. Fluctuations in these systems are inextricably linked tochanges in northern hemispheric insolation and mid-high latitudeglacial boundary conditions. Peak humid phases are generallyassociated with eccentricity-paced interglacial periods (e.g. Neffet al., 2001; Fleitmann et al., 2011), however, recent evidencesuggests that the Peninsula may exhibit a more spatially andtemporally heterogeneous climate driven by insolation maxima(Parton et al., in this issue). With respect to temporal-spatial cli-matic variability at the onset of the Holocene, a comparison ofevidence from speleothems in southern Oman and Yemen withpalaeolakes from southeast Arabia suggests that the northwardmigration of the monsoon rainfall belt may have taken over ~1500years to move from southern to northern extremes (Parker, 2009).Holocene palaeolake records from Yemen (Davies, 2006; L�ezineet al., 2007) indicate that the onset of humidity began at ca.11 ka, while speleothem records from southern Oman record anabrupt increase in precipitation at ca. 10.5 ka (Fleitmann et al.,2007; Fleitmann and Matter, 2009; Preusser, 2009). Furthernorth, records from northern Oman indicate that the onset of hu-midity occurred at ca. 9.6 ka while in the Gulf Region, lake forma-tion began at ca. 8.5 ka (e.g., Parker et al., 2004, 2006; Preston et al.,2012). Findings from excavations conducted throughout thesouthern Nejd indicate a later onset of pluvial conditions at 8.7 ka(Cremaschi and Negrino, 2005), which are clearly in contrast withspeleothem data from the same region. This age, however, isderived from a single, short-lived depositional event and, as such,may reflect the role of topographic constraints in the activation ofdrainage systems. Additionally, speleothems require rainfall be-tween ~300 and 350 mm per year for growth (e.g., Vaks et al., 2010;Fleitmann et al., 2011) and as such, there remains a broad range ofprecipitation rates between these values and current conditionsthat may have enabled the activation of smaller, ephemeral chan-nels not recorded in the speleothem archive.


Within the study region, incursions of monsoon rainfall north ofthe current orographic barrier located along the Dhofar Escarpmentwould have led to greatly enhanced seasonal, possibly perennial,streamflow during humid phases (Parker and Rose, 2008).Conversely, during insolation minima, the southwardly-displacedITCZ would have meant that and monsoon rainfall belt no longerbreached the Dhofar Escarpment and penetrated the desertinterior.

3. Methods

In the course of geoarchaeological and geochronological in-vestigations, sediment samples were collected for optically stimu-lated luminescence (OSL) dating (see online supportingmaterial fordetails). OSL dating provides an estimate of the time since grains ofquartz or feldspar were last exposed to sunlight (Huntley et al.,1985; Aitken, 1998; Lian and Roberts, 2006). The burial age isestimated by dividing the equivalent dose (De, a measure of theradiation energy absorbed by the grains during the period of burial)by the environmental dose rate (the rate of supply of ionizing ra-diation to the grains over the same period). The De is determinedfrom laboratorymeasurements of the OSL signals, and the dose rateis calculated from laboratory and field measurements of the envi-ronmental radioactivity, plus an estimate of the contribution fromcosmic rays. In this study, we estimated the De values from the OSLsignals emitted by 3e5 grains of quartz measured simultaneously.This approach has also been employed by others (e.g., Featherset al., 2010) and is sometimes referred to as ‘micro-hole’ or‘pseudo’ single-grain OSL dating (e.g., Berger, 2011; Arnold et al.,2012). It represents a trade-off between the benefits of ‘true’single-grain dating, in which each grain is measured individually(Roberts et al., 1998; Jacobs and Roberts, 2007; Duller, 2008), andincreasing the probability of each set of 3e5 grains containing aluminescent grain, as quartz grains are typically very dim. Unde-sirable averaging effects can arise from measuring multiple lumi-nescent grains simultaneously (Arnold and Roberts, 2009; Arnoldet al., 2012), but for samples with relatively few bright grains(such as those studied here) pseudo single-grain analysis


Fig. 3. Late Palaeolithic sites in Dhofar. Elevation based map of Dhofar showing the location of: a) Late Palaeolithic sites recorded during the Dhofar Archaeological Project as well asthe location of al-Hatab, Ghazal and Khamseen Rockshelter (red circle); b) a close up of the immediate area surrounding the al-Hatab, Ghazal and Khamseen Rockshelter. (Forinterpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)


represents a satisfactory compromise between the advantages oftrue single-grain dating and the efficient use of instrument time.

To gain information regarding site formation processes and toreconstruct landscape dynamics at the sites, sediment samples


were collected from stratigraphic profiles at al-Hatab andKhamseen Rockshelters. Loss-on-ignition (LOI) estimates of organicmatter (LOIorg) and carbonate content (LOIcarb) were measuredfollowing standard procedures set out by Heiri et al. (2001). Grain


Table 1OSL ages from the stratified sites from the Umm al Khashab area of Dhofar.

Site Sample nr. Depth (cm) Layer Archaeology OSL age (ka)

Al-Hatab TH29.1 26 GH3 ? 2.7 ± 0.2Al-Hatab TH29.2 69 GH8 Sterile 19.9 ± 1.3Al-Hatab AH-OSL1 50 Unit B LP 12.8 ± 1.0Al-Hatab AH-OSL2 65 Unit C Sterile 12.6 ± 1.5Ghazal TH47.4 20 GH2 LP 7.2 ± 0.5Ghazal TH47.1 35 GH3 Sterile 7.3 ± 0.5Ghazal TH47.3 40 GH5 Sterile 8.6 ± 0.6Ghazal TH47.2 57 GH5 Sterile 8.2 ± 0.7Khamseen TH50.4 83 GH4B Neolithic 6.3 ± 0.5Khamseen TH50.3 104 GH5A LP 7.1 ± 0.6Khamseen TH50.2 118 GH5A LP 8.7 ± 0.7Khamseen TH50.1 165 GH5B LP 9.7 ± 0.7


size distribution of samples involved air-drying sediment andgently disaggregating particles in deionized water before analysisusing a Malvern Mastersizer 2000. Mass specific, low frequencymagnetic susceptibility values (clf) were obtained from air-driedsamples using a Bartington MS2C sensor at 0.1 SI sensitivity, asdescribed by Dearing (1999).

4. Stratified sites in the Wadi Dawkah area (Umm al Khashab)

Numerous surface and buried Late Palaeolithic sites have beendocumented throughout every region of Dhofar over the past de-cades (e.g., Phillips, 1955; Pullar, 1974; Charpentier, 1996; Zarins,2001; Cremaschi and Negrino, 2002, 2005; Rose, 2006; Rose andUsik, 2009; Newton and Zarins, 2010; Hilbert et al., 2012; Hilbert,2013, 2014; Cremaschi et al., 2015). Buried sites are most oftenfound in Dhofar in close proximity to fluvial and erosion gullysystems, rockshelters and close to piedmont situations.

Al-Hatab, Ghazal and Khamseen rockshelters are located on thedry plateau north of the Dhofar Mountains, in the vicinity of WadiDawkah (Fig. 3), in an area locally known as the Umm al Khashab(literal translation “mother of wood”, intended to mean “the placewith plentiful wood”). These sites share similar geomorphologicalsettings: a low limestone rockshelter, a pronounced talus slope, anda sheltered location conducive to the preservation of sedimentsfrom erosion. Landscape configurations such as these are commonacross the Umm al Khashab, and specifically along the length ofWadi Dawkah where the landscape is characterized by low scarpslopes. The caprock of these scarps is composed of laminated cal-carinitic limestone, which is more resilient to erosion than thesofter homogeneous dolomitic chalk forming the lower portion ofthe scarps. Fluvial downcutting and redeposition of the erosionaldebris has created a complex network of small drainage channels,complex basins and box canyons. OSL ages for the deposits exca-vated at each of these LP sites are discussed in Section 5 and listedin Table 1, with full supporting procedural details and data providedin the online supplementary material.

4.1. Al-Hatab (OM.JA.TH.29)

Al-Hatab (17.313417�N, 54.061050�E) is a small collapsed rock-shelter on the interior slopes of Jebel Qara. The rockshelter itself hasbeen stripped of sediments, while two low-gradient talus slopessituated on both sides of the depression contain stratified LatePalaeolithic archaeological deposits. The talus on the northwardfacing slope is incised by a meandering gully feeding into a tribu-tary ofWadi Dawkah; the opposite slope remains undisturbed. Twosmall excavated sondages yielded artifacts attributed to the NejdLeptolithic. Some general aspects of the archaeology, preliminarydating of Sondage 1, and sedimentology have already been pub-lished (Rose, 2006; Rose and Usik, 2009; Hilbert et al., 2012). In thisstudy, we present a revised OSL chronology of Sondage 1 andadditional OSL dates and sedimentological data from Sondage 2.

The al-Hatab depositional sequence from Sondage 2 comprises atotal of eight geological horizons (GH) characterized by minero-genic silt- and gravel-dominated lithological units. The strata areprimarily categorized by the ratio of coarse components to finematrix and the sporadic presence of anthropogenic inclusions(Fig. 4). GH8 is at the base of the sequence and comprises anaccumulation of well-cemented medium course sand, containingoccasional coarse components. The sediment units GH7 to 5 showwell-developed imbrication of gravel clasts at the base, which fine-up towards the top of GH5. The sediment matrix within these unitsis composed of sand and fine gravels. Grain size decreases towardGH4, the sediment from this unit being composed of silt and very


fine sands. GH3 is characterized by the increased deposition ofpoorly-sorted coarser-grain material. GH2 and GH1 contain potteryand charcoal remains recorded in GH2B and, to a lesser extent, inGH2A. The charcoal in GH2B has been reworked into overlyingunits, confirmed by peaks in magnetic susceptibility. A rise inmagnetic susceptibility seen throughout the upper units confirmsin-situ weathering following the deposition of these layers.

From Sondage 2, an OSL sample was taken from GH8 that yiel-ded a numeric age estimate of 19.9 ± 1.3 ka. In Sondage 1, two OSLsamples were previously taken from the base of LP artifact-bearing“Unit B” and the top of underlying archaeologically sterile Unit C(Rose and Usik, 2009). Reanalysis of these samples has produceddates an age of 12.8 ± 1.0 ka for Unit B, while the Unit C sampleyielded a numeric age of 12.6 ± 1.5 ka. Given the error marginsassociated with these age estimates, it is likely that they reflectcoeval deposition. While it is not possible to directly correlate thesedimentary units between Sondages 1 and 2, we infer that both LPartifact-bearing units (GH7 to 5 in Sondage 2 and Unit B in Sondage1) were deposited at the same time. Near the top of the sequence,GH3 produced a numeric age estimate of 2.7 ± 0.2 ka.

The characteristics of basal GH8 are consistent with a largelystable landscape, with small-scale, localized colluvial processesdriving sedimentation (Fig. 5). The interface between GH8 and GH7marks a distinct shift in depositional environment at the site. Thesediments associated with GH7, GH6 and GH5 are indicative ofgully erosion, evidenced by field observations and supported by theanalytical data. This may be compatible with phases of activationand de-activation of regional drainage systems. The grain size datashows fluctuations in water flow/sediment deposition within theseminor channels. Low carbonate values reflect a decrease in land-scape stability, with material removed from the substrate throughdissolution. The shift to more humid conditions at this time iscoincident with the presence of human activity at the site, as evi-denced by the presence of LP artifacts. GH4 appears to mark acessation in drainage activation at the site, while GH3 may show abrief increase in humidity and run-off.

4.2. Ghazal Rockshelter (OM.JA.TH.47)

Ghazal Rockshelter (17.314483�N, 54.056617�E) is locatedapproximately 500 m to the west of al-Hatab, within a network ofdrainage gullies and small tributaries that dissect the surroundingscabland. A total of 16 m2 were excavated to approximately 50 cmdepth, yielding a total of 425 LP artifacts from two distinctarchaeological horizons. Six geological horizons have been identi-fied at Ghazal based on stratigraphy recorded in the northern andeastern sections (Fig. 6). Artifacts from archaeological level 1 aredistributed vertically through the 20 cm of unit GH2, while level 2artifacts have been found exclusively at the contact of units GH5


Fig. 4. Al-Hatab: a) Topographic map of the site showing the location of the two test-pits; b) photograph of the east section from sondage 2; c) east section of Sondage 2 showinglocation of OSL samples and respective stratigraphic/geological horizons (GH); d) west section of Sondage 1 showing the location of OSL samples.

Please cite this article in press as: Hilbert, Y.H., et al., Terminal Pleistocene and Early Holocene archaeology and stratigraphy of the southernNejd, Oman, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.02.053


and GH4. Both level 1 and 2 assemblages are marked by charac-teristic Nejd Leptolithic core reduction (Hilbert, 2014). Blade cores,blades, flakes, cortical pieces and some tools were found within thedeposit. Technological and typological analyses of these artifactsare described elsewhere (Hilbert et al., 2012). Here we present newOSL age estimates for the sand layers bracketing the archaeologicallevels at the site.

The sedimentary sequence at Ghazal comprises alternatingsands and course limestone clasts. Unit GH5 is distributed evenly atthe base of the sequence and consists of medium-to fine-grainedpale yellow sand without inclusions. Unit GH4 contains artifactsfrom archaeological level 2, and is composed of moderatelycemented, angular limestone debris and pale yellow, medium-tofine-grained sand and silt. Artifacts show little to no edge dam-age. Unworked chert nodules and geofacts were also found in thislayer. GH3 is distributed unevenly on top of unit GH4, within andbeneath a roof collapse at the brow of the rockshelter. Thisarchaeologically sterile unit separates archaeological level 2 fromsuperimposed archaeological level 1 and is characterized by lightbrown silts and small angular limestone clasts. Unit GH1 is thesurface of the site, consisting of limestone debris and chert geofacts.

Four OSL samples were taken from Ghazal Rockshelter. UnitGH2, which contains the level 1 archaeological horizon, produced anumeric age of 7.2 ± 0.5 ka, while a sample taken from the un-derlying, culturally sterile GH3measured 7.3 ± 0.5 ka. At the base of

Fig. 5. Al-Hatab Multi proxy sediment record. Sedimentology (granulometry of <2 mm fraccarbonate content) from al-Hatab.


the sequence, culturally sterile unit GH5 yielded an age of8.5 ± 0.4 ka; this estimate represents the weighted mean forsamples TH47-2 and TH47-3, both of which were collected fromunit GH5. These data indicate that GH4, which contains the lowerLP archaeological horizon at the site, was deposited after 8.5 ± 0.4(GH5) and before 7.3 ± 0.5 ka (GH3). The OSL chronology at GhazalRockshelter suggests that the sediments accumulated over a shorttime span, of only 0.2e2.3 ka duration (95% confidence). The con-dition of the artifacts is indicative of rapid burial in a relatively lowenergy environment.

4.3. Khamseen Rockshelter (OM.JA.TH.50)

Khamseen Rockshelter (17.313517�N, 54.042111�E) was discov-ered in 2010, with test excavations revealing a rich succession ofoccupation layers spanning much of the Early to Late Holocene. Thesite is situated within an approximately 50 m long and 30 m widesemi-circular depression in the limestone scarp. As was the casewith all of the other rockshelters found in the Umm al Khashabregion of the Nejd, Khamseen is devoid of sediment fill, except fortalus slopes that converge towards the center of the depression. Atotal of nine GH were identified in the eastern profile of Test-pit 1(Fig. 7). These layers reveal a distinct set of lithic technologies andtool types, including 766 LP artifacts from the basal horizon (GH5Aand 5B). The LP assemblage consists of blade cores, blades, flakes

tion) and multiproxy data (magnetic susceptibility values, and silt, organic carbon and



and a toolkit containing of pedunculated point fragments, end-scrapers and piercers. Initial OSL estimates, radiocarbon ages, andobservations concerning the early to mid-Holocene archaeologicalrecord are described elsewhere (Hilbert et al., 2012; Hilbert, 2013).Here, we present revised OSL age estimates and sedimentologicalanalyses from the site.

At the base of the sequence, units GH5B and GH5A are charac-terized by coarse and gravel-rich sediments; the courser sedimentcomponents possibly associated with the physical weathering ofthe host bedrock. A coarser, near symmetrical aeolian sand andconcomitant decrease in silt content and magnetic susceptibilitywas recorded at the base of GH4B, while the top of this unit shows apeak in magnetic susceptibility values and silt content, with min-erogenic fines being washed into the site. Similarly, an influx ofaeolianmaterial and decreased susceptibility values are recorded atthe base GH4A, followed by an increase in the deposition of silts. Inthe upper portion of the sequence, units GH1 to GH3 yieldedmodern to sub-recent archaeological materials, hence, are outsidethe chronological focus of this paper and have been omitted fromthis study.

A series of OSL samples were taken from the Khamseen strati-graphic profile in Test-pit 1. The sample from basal unit GH5B

Fig. 6. Ghazal Rockshelter: a) Topographic map of the site showing the location and extphotograph of the west section between squares C1eC3; d) schematic image of the west secgeological horizons (GH); e) schematic image of the west section between squares C1eCschematic image of the north section between squares A2eB2 showing the stratigraphic/g


produced an age of 9.7 ± 0.7 ka, while overlying unit GH5A yieldedages of 7.1 ± 0.6 ka and 8.7 ± 0.7 ka. Sample TH50-2, from the lowerpart of unit GH5A, is significantly (>2s) older than the age ofsample TH50-3, thus, calculation of a weighted mean age from thedates does not provide a meaningful chronological control. In unitGH4b, the sample produced an age of 6.3 ± 0.5 ka. This layeroverlies a hearth, from which ash returned a radiocarbon age of6845 ± 105 cal BP (Hilbert et al., 2012); as such, shows a corre-spondence between the independent OSL and radiocarbon chro-nologies at Khamseen.

GH5B appear to record an increase in moisture availabilitypromoting the physical and chemical breakdown of the bedrockand resulting in colluvial deposition at the site. In GH5A, the in-crease in magnetic susceptibility values and silt content may indi-cate the influx of minerogenic fines via these colluvial processes.The deposition of aeolian sands at the bottom of GH4Bmay indicatea short-lived arid event within the Early to Mid-Holocene humidphase (Fig. 8) at ~6.3 ka. This age corresponds well with other ev-idence for the widespread aridification of Arabia at this time (e.g.Preston et al., 2012). A return to more humid conditions in theupper part of GH4B is reflected in the peak in magnetic suscepti-bility values and silt content.

ent of the excavation b) photograph of the west section between squares B1eB3; c)tion between squares B1eB3 showing the location of the OSL sample and stratigraphic/3 showing the location of the OSL sample and stratigraphic/geological horizons; f)

eological horizons.



5. Discussion

One of the central considerations for reconstructing the de-mographic history of Arabia hinges upon whether human groupscould endure extreme climatic downturns. To what extant was theArabian Peninsula depopulated (i.e., tabula rasa), and to whatextant could human groups survive arid phases within environ-mental refugia (e.g., Rose, 2007; Rose and Petraglia, 2009;Uerpmann et al., 2009)? Potential refugia have been suggestedthroughout the Peninsula, including the Red Sea basin (Bailey,2009), the western Yemeni Highlands (Fedele, 2009), the Hadra-mawt in Yemen (Crassard, 2008a) and the Gulf “Oasis” (Rose, 2010),In this paper, we consider this question as it relates to the region ofDhofar across the Terminal Pleistocene and Early Holocene periods.Given the climate history of the Nejd, it is unlikely this regionserved as a primary refugium; however, its southerly location andextant hydrology suggest that this may have been one of the firstplaces in Arabia to become ameliorated after the arid conditionsduring MIS 2.

The oldest dated evidence for post-Glacial human occupation inDhofar comes from units GH5 to GH7 from Sondage 2 and Unit Bfrom Sondage 1 at al-Hatab, suggesting that hunteregatherer

Fig. 7. Khamseen Rockshelter: a) Topographic map of the site showing the location of the tweast section from test-pit 1; d) schematic image of the east section from test-pit 1 showin


populations had returned to the region by ~13 ka. Artifacts fromthis early stage of the Nejd Leptolithic tradition are characterizedtechnologically by a predominance of single platform parallel bladeand flake cores. Tools include pedunculated points, bi-pointedelongated bifaces, thick elongated ovoid bifaces, partially unifacialpoints and an abundance of endscrapers and burins (mostly ontruncation) (Fig. 9). The combination of hard hammer blade tech-nology and such tool types has been recorded throughout southernand central Oman (Rose and Usik, 2009; Jagher et al., 2011) andcentral and eastern Yemen (Amirkhanov, 1994, 2006; Crassard,2009). No parallels are known from beyond these areas, makingthis specific techno/typological package unique to South Arabia.

Archaeological deposits dated to the later stages of the NejdLeptolithic tradition were excavated at Ghazal and KhamseenRockshelters. These assemblages are composed of unidirectionalsingle platform cores, which produced blades and occasional flakes.The tool kit includes pedunculated points, endscrapers, piercers,burins on truncation, unifacial cutting implements (pseudo-backedknifes) and a regionally specific form dubbed “trifaces” (Fig. 10)(Hilbert, 2014). Core exploitation follows one of three differentreduction modalities, characterized by different forms of unidi-rectional semi-tournant reduction (Hilbert et al., 2012). As

o test-pits; b) photograph of the site during excavation in 2010; c) photograph of theg the stratigraphic/geological horizons (GH) and location of OSL samples.



previously described (Hilbert et al., 2012), there are clear parallelsbetween these modalities and the Wa'shah method in Yemen(Crassard, 2008a, 2008b). Survey and excavation activities under-taken by Cremaschi and Negrino (2002) on the Jebel Qara area ofDhofar have discovered an additional eight “EpipalaeolithiceearlyNeolithic” sites with similar technological and typological featuresas those found at Ghazal and Khamseen.

Stratigraphic records from the LP sites presented in this paperprovide a record of climatically driven landscape change againstwhich the archaeological record of the regionmay be set. The lowerunit (GH8) at al-Hatab accumulated from approximately 20 ka to13 ka. The overlying unit (GH7) reveals a succession of ephemeralfluvial activation events occurring sometime around 13 ka. Theassemblages excavated from GH7 to GH5 (and archaeological UnitB) fall chronologically near the end of the Pleistocene, at a timewhen monsoon rainfall in southern Arabia was not sufficient tocause speleothem growth, but was capable of activating drainagesystems north of the Dhofar escarpment.

The onset of wetter conditions, marking the beginning of theHolocene Climatic Optimum, have been detected in the lower unitsat Khamseen (GH5A and GH5B), around 10 ka. The OSL ages andgeoarchaeological analyses at Khamseen suggest slow and possiblycontinuous sedimentation processes across the lower part of thesequence. In contrast, the sequence and chronology at Ghazal in-dicates a faster and intermittent depositional record. The lower unit(GH5) at Ghazal is dated to 8.5 ± 0.4 ka and represents a period of

Fig. 8. Khamseen multi proxy sediment record. Sedimentology (granulometry of <2 mm fracarbonate content) from TH.50.


aeolian influx to the site. This age appears incongruous with otherpalaeoclimatic evidence from Arabia at this time, which demon-strate widespread humidity was in place across most of thePeninsula. While there is some evidence for a brief climaticdownturn at ca. 8.2 ka (e.g. Parker et al., 2006), the absence ofsimilar aeolian deposits dating to the same age at Khamseen and al-Hatab may indicate that unit GH5 at Ghazal represents anephemeral, localized occurrence. Subsequently, Unit GH4, brack-eted between approximately 9 ka and 7 ka, contains archaeologicallevel 2, formed during a phase of colluvial deposition in a higherenergy environment.

All of the LP assemblages considered in this study share closetechnological and typological affinities (e.g., blade technology, bu-rins, endscrapers, and pedunculated points), as such belong to theNejd Leptolithic tradition. The earlier assemblage from al-Hatab isdistinguished by the presence of bifacial foliate tools, while thelater assemblages excavated from Ghazal and Khamseen Rock-shelter are absent of these forms, and also exhibit the Wa'shahmethod of blade production. This pattern suggests there may havebeen minor technological and typological changes over time. Fromthese data, we argue there was cultural continuity across the Ter-minal PleistoceneeEarly Holocene boundary in Dhofar. It ispossible that such continuity was at least partly facilitated by theclose proximity of the region to the latitudinal position of themonsoon rainfall belt. Corroborating this assertion, genetic evi-dence indicates that portions of the modern South Arabian gene

ction) and multiproxy data (magnetic susceptibility values, and silt, organic carbon and



pool are derived from a Near Eastern population that underwentsubstantial growth within Arabia ca. 13 ka ago (�Cerný et al., 2011;Al-Abri et al., 2012; Rose et al., 2013).

6. Conclusions

The geoarchaeological and chronological data presented herereveal a complex depositional history marked by phases of local-ized aeolian sedimentation and widespread colluvial and fluvial

Fig. 9. Hatabian artifacts from al-Hatab. 1 and 2: tanged points; 3: distal fragment of unifablade; 9 and 11: burin on straight truncation; 10: burin on convex truncation; 12 and 13: bucores.


deposition spanning the Terminal Pleistocene and Early Holocene.Both the fluvial and colluvial units identified at these sites suggestthat conditions were wetter than present and that the southernNejd was habitable during this period. While the availablearchaeological data are insufficient to establish whether a ‘tabularasa’ event occurred during the Last Glacial Maximum in Arabia,our findings indicate that the Nejd Leptolithic tradition arose from apopulation that was present in Dhofar prior to the Holocene Cli-matic Optimum. The absence of matching Terminal Pleistocene

cial point; 4e6: partially retouched points; 7: laterally retouched blade; 8: truncatedrin on concave oblique truncation; 14e16 bifaces; 17e19 single platform unidirectional


Fig. 10. Khashabian artifacts from sites across the Nejd. 1e3: tanged points (from Khamseen, Jebel Eva and TH.200 respectively); 4, 6 and 9: partially retouched points (FromKhamseen, SJ.51 and TH.200 respectively); 5 and 8: pseudo backed knifes (from Jebel Eva and TH.200 respectively); 7: triface made on large blade (Wadi Haluf 1); 10: burin ontruncation (TH.34); 11: dihedral burin (TH.34); 12: endscraper on flake (Khamseen); 13: refitting of six flakes on alternate platform core, numbers show the order of removals andarrows the directionality of the detaching blows (Ghazal).


Please cite this article in press as: Hilbert, Y.H., et al., Terminal Pleistocene and Early Holocene archaeology and stratigraphy of the southernNejd, Oman, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.02.053


assemblages outside of South Arabia may suggest that the NejdLeptolithic tradition developed locally or nearby in Yemen.

Acknowledgements

Research in Dhofar was undertaken under the auspices of theMinistry of Heritage and Culture of Oman. Funding for excavationand analysis was provided by a UK Arts and Humanities ResearchCouncil Early Career Research grant (AH/H033912/1) awarded to J.I.Rose. OSL ages were made possible through an Australian ResearchCouncil Discovery Project grant and Australian ProfessorialFellowship awarded to R.G. Roberts (DP0880675) and a DiscoveryProject grant and Queen Elizabeth II Fellowship awarded to Z. Ja-cobs (DP1092843). A Fyssen Foundation grant awarded to Y.H.Hilbert provided funding for the publication of these results. We areparticularly grateful to the other members of the Dhofar Archaeo-logical Project field team: J.M. Geiling, A. Beshkani, MohamadJaboob and V.I. Usik. We extend our gratitude to R�emy Crassard andthree anonymous reviewers for their comments on thismanuscript.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.quaint.2015.02.053.

References

Aitken, M.J., 1998. An Introduction to Optical Dating. Oxford University Press,Oxford.

Al-Abri, A., Podgorna, E., Rose, J.I., Pereira, L., Mulligan, C., Silva, N., Bayoumi, R.,Soares, P., �Cerný, V., 2012. Pleistocene-Holocene boundary in Southern Arabiafrom the perspective of human mtDNA variation. American Journal of PhysicalAnthropology 149, 291e298.

Amirkhanov, H., 1994. Research on the Palaeolithic and Neolithic of Hadramaut andMahra. Arabian Archaeology and Epigraphy 5, 217e228.

Amirkhanov, H., 2006. Stone Age of South Arabia. Nauka, Moscow (in Russian).Arnold, L.J., Roberts, R.G., 2009. Stochastic modelling of multi-grain equivalent dose

(De) distributions: implications for OSL dating of sediment mixtures. Quater-nary Geochronology 4, 204e230.

Arnold, L.J., Demuro, M., Navazo Ruiz, M., 2012. Empirical insights into multi-grainaveraging effects from ‘pseudo’ single-grain OSL measurements. RadiationMeasurements 47, 652e658.

Bailey, G., 2009. The Red Sea, coastal landscapes, and hominin dispersals. In:Petraglia, M.D., Rose, J.I. (Eds.), Evolution of Human Populations in Arabia:Paleoenvironments, Prehistory and Genetics. Springer Academic Publishers,Dordrecht, pp. 15e37.

Berger, G.W., 2011. Surmounting luminescence age overestimation in Alaska-margin Arctic Ocean sediments by use of ‘micro-hole’ quartz dating. Quater-nary Science Reviews 30, 1750e1769.

Bretzke, K., Armitage, S.J., Parker, A.G., Walkington, H., Uerpmann, H.-P., 2013. Theenvironmental context of Paleolithic settlement at Jebel Faya, Emirate Sharjah,UAE. Quaternary International 300, 83e93.

�Cerný, V., Mulligan, C., Fernandes, V., Silva, N., Alshamali, F., Non, A., Harich, N.,Cherni, L., El Gaaied, A., Al-Meeri, A., Pereira, L., 2011. Internal diversification ofmitochondrial haplogroup R0a reveals post-last glacial maximum demographicexpansions in South Arabia. Molecular Biology and Evolution 28, 71e78.

Charpentier, V., 1996. Entre sables du Rub'al Khali et mer d'Arabie, Pr�ehistoirer�ecente du Dhofar et d'Oman: Les industries �a point de “fasad”. Proceedings ofthe Seminar of Arabian Studies 26, 1e14.

Crassard, R., 2008a. La pr�ehistoire du Y�emen. Diffusions et diversit�es locales, �atravers l'�etude d'industries lithiques du Hadramawt. In: British ArchaeologicalReports (International Series), vol. 1842. Archaeopress, Oxford.

Crassard, R., 2008b. The “Wa'shah” method: an original laminar debitage fromHadramawt, Yemen. Proceedings of the Seminar for Arabian Studies 38, 3e14.

Crassard, R., 2009. Modalities and characteristics of human occupations in Yemenduring the Early/Mid-Holocene. Comptes Rendus Geoscience 341, 713e725.

Cremaschi, M., Negrino, F., 2002. The Frankincense of Sumhuram: palae-oenvironmental and prehistorical background. In: Avanzini, A. (Ed.), Khor RoriReport 1. Universit�a di Pisa, Pisa, pp. 325e363.

Cremaschi, M., Negrino, F., 2005. Evidence for an abrupt climatic change at 8700 14Cyr B.P. in rockshelters and caves of Gebel Qara (Dhofar-Oman): palae-oenvironmental implications. Geoarchaeology 20, 559e579.

Cremaschi, M., Zerboni, A., Charpentier, V., Crassard, R., Isola, I., Regattieri, E.,Zanchetta, G., 2015. Early-Middle Holocene environmental changes and pre-


Neolithic human occupations as recorded in the cavities of Jebel Qara (Dho-far, southern Sultanate of Oman). Quaternary International. http://dx.doi.org/10.1016/j.quaint.2014.12.058.

Davies, C.P., 2006. Holocene paleoclimates of southern Arabia from lacustrine de-posits of the Dhamar highlands, Yemen. Quaternary Research 66, 454e464.

Dearing, J., 1999. Environmental Magnetic Susceptibility: Using the MS2 System.British Library and Publication Data.

Drechsler, P., 2007. Spreading the Neolithic over the Arabian Peninsula. Proceedingsof the Seminar for Arabian Studies 37, 93e109.

Drechsler, p., 2009. The Dispersal of the Neolithic over the Arabian Peninsula. In:British Archaeological Reports (International Series), vol. 1969. Archaeopress,Oxford.

Duller, G.A.T., 2008. Single-grain optical dating of Quaternary sediments: whyaliquot size matters in luminescence dating. Boreas 37, 589e612.

Feathers, J., Kipnis, R., Pil�o, L., Arroyo-Kalin, M., Coblentz, D., 2010. How old is Luzia?Luminescence dating and stratigraphic integrity at Lapa Vermelha, Lagoa Santa,Brazil. Geoarchaeology 25, 395e436.

Fedele, F., 2009. Early Holocene in the Highlands: data on the peopling of theEastern Yemen Plateau, with a note on the Pleistocene evidence. In:Petraglia, M.D., Rose, J.I. (Eds.), Evolution of Human Populations in Arabia:Paleoenvironments, Prehistory and Genetics. Springer Academic Publishers,Dordrecht, pp. 215e236.

Fleitmann, D., Matter, A., 2009. The speleothem record of climate variability inSouthern Arabian. Comptes Rendus Geosciences 341, 633e642.

Fleitmann, D., Burns, S., Mangini, A., Mundelsee, M., Kramers, J., Villa, I., Neff, U., Al-Subbary, A., Beuttner, A., Hippler, D., Matter, A., 2007. Holocene ITCZ and Indianmonsoon dynamics recorded in stalagmites from Oman and Yemen (Socotra).Quaternary Science Reviews 26, 170e188.

Fleitmann, D., Burns, S., Pekala, M., Mangini, A., Al-Subbary, A., Al-Aowah, M.,Kramers, J., Matter, A., 2011. Holocene and Pleistocene pluvial periods in Yemen,southern Arabia. Quaternary Science Reviews 30, 783e787.

Gallagher, M., 1977. The Oman flora and fauna survey, 1977 (Dhofar). Journal ofOman Studies 3, 9e12.

Ghazanfar, S.A., 1999. Present flora as an example of palaeoclimate: examples fromthe Arabian Peninsula. In: Singhvi, A., Derbyshire, E. (Eds.), PaleoenvironmentalReconstruction in Arid Lands. Balkema, Rotterdam, pp. 263e275.

Ghazanfar, S.A., Fisher, M., 1998. Vegetation of the Arabian Peninsula. Kluwer Ac-ademic Publishers, Dordrecht.

Heiri, O., Lotter, A.F., Lemcke, G., 2001. Loss on ignition as a method for estimatingorganic and carbonate content in sediments: reproducibility and comparabilityof results. Journal of Paleolimnolgy 25, 101e110.

Hilbert, Y.H., 2013. Khamseen rock shelter and the Late Palaeolithic-Neolithictransition in Dhofar. Arabian Archaeology and Epigraphy 24, 51e58.

Hilbert, Y.H., 2014. Khashabian: a Late Paleolithic Industry from Dhofar, SouthernOman. In: British Archaeological Reports (International Series), vol. 2601.Archaeopress, Oxford.

Hilbert, Y.H., Rose, J.I., Roberts, R.G., 2012. Late Palaeolithic core-reduction strategiesin Dhofar, Oman. Proceedings of the Seminar for Arabian Studies 42, 101e118.

Huntley, D.J., Godfrey-Smith, D.I., Thewalt, M.L.W., 1985. Optical dating of sedi-ments. Nature 313, 105e107.

Jacobs, Z., Roberts, R.G., 2007. Advances in optically stimulated luminescence datingof individual grains of quartz from archeological deposits. Evolutionary An-thropology 16, 210e223.

Jagher, R., Pümpin, C., Wegmüller, F., Winet, I., 2011. Central Oman palaeolithicsurvey, report of the 2007 season. The Journal of Oman Studies 17, 15e50.

Lepvrier, C., Fournier, M., B�erard, T., Roger, J., 2002. Cenozoic extension in coastalDhofar (southern Oman): implications on the oblique rifting of the Gulf ofAden. Tectonophysics 357, 279e293.

L�ezine, A., Tiercelin, J., Robert, C., Sali�ege, F., Cleuziou, S., Inizan, M., Braemer, F.,2007. Centennial to millennial-scale variability of the Indian monsoon duringthe Early Holocene from a sediment, pollen and isotope record from thedesert of Yemen. Palaeogeography, Palaeoclimatology, Palaeoecology 243,235e249.

Lian, O.B., Roberts, R.G., 2006. Dating the Quaternary: progress in luminescencedating of sediments. Quaternary Science Reviews 25, 2449e2468.

Maher, L.A., 2009. The late Pleistocene of Arabia in relation to the Levant. In:Petraglia, M.D., Rose, J.I. (Eds.), Evolution of Human Populations in Arabia:Paleoenvironments, Prehistory and Genetics. Springer Academic Publishers,Dordrecht, pp. 187e202.

Miller, A.G., Morris, M., 1988. Plants of Dhofar, the Southern Region of Oman,Traditional, Economic and Medicinal Uses. Office of the Adviser for Conserva-tion of the Environment, Diwan of Royal Court, Sultanate of Oman.

Neff, U., Burns, S., Mangini, A., Mudelsee, M., Fleitmann, D., Matter, A., 2001. Strongcoherence between solar variability and the monsoon in Oman between 9 and6 kyr ago. Nature 411, 290e293.

Newton, L.S., Zarins, J., 2010. Preliminary results of the Dhofar archaeological sur-vey. Proceedings of the Seminar of Arabian Studies 40, 247e265.

Parker, A.G., 2009. Pleistocene climate change in Arabia: developing a frameworkfor hominid dispersal over the last 350 KYR. In: Petraglia, M.D., Rose, J.I. (Eds.),Evolution of Human Populations in Arabia: Paleoenvironments, Prehistory andGenetics. Springer Academic Publishers, Dordrecht, pp. 39e49.

Parker, A.G., Rose, J.I., 2008. Climate change and human origins in southern Arabia.Proceedings of the Seminar for Arabian Studies 38, 25e42.




http://refhub.elsevier.com/S1040-6182(15)00160-3/sref1




















































































































































































Parker, A.G., Eckersley, L., Smith, M.M., Goudie, A.S., 2004. Holocene vegetationdynamics in the northeastern Rub' al Khali desert, Arabian Peninsula d apollen, phytolith and carbon isotope understanding the evolution of the Ho-locene pluvial phase study. Journal of Quaternary Science 19, 665e676.

Parker, A.G., Goudie, A.S., Stokes, S., White, K., 2006. A record of Holocene climatechange from lake geochemical analyses in southeastern Arabia. QuaternaryResearch 66, 465e476.

Parton, A., White, T.S., Parker, A.G., Breeze, P.S., Jennings, R., Groucutt, H.S.,Petraglia, M.D., 2015. Orbital-scale climate variability in Arabia as a potentialmotor for human dispersals. Quaternary International (in this issue).

Phillips, W., 1955. Qataban and Sheba: Exploring the Ancient Kingdoms on theBiblical Spice Routes of Arabia. Brace and Company, New York.

Platel, J., Roger, J., Petters, T., Mercolli, I., Kramers, J., Le-M�etour, J., 1992. GeologicalMap of Salalah, Explanatory Notes. Ministry of Petroleum and Minerals, Sul-tanat of Oman.

Preston, G.W., Parker, A.G., Walkington, H., Leng, M.J., Hodson, M.J., 2012. Fromnomadic herder-hunters to sedentary farmers: the relationship betweenclimate change and ancient subsistence strategies in south-eastern Arabia.Journal of Arid Environments 86, 122e130.

Preusser, F., 2009. Chronology of the impact of Quaternary climate change oncontinental environments in the Arabian Peninsula. Comptes Rendus Geo-science 341, 621e632.

Pullar, J., 1974. Harvard archaeological survey in Oman, 1973: flint sites in Oman.Arabian Seminar 4, 33e48.

Roberts, R.G., Bird, M., Olley, J., Galbraith, R., Lawson, E., Laslett, G., Yoshida, H.,Jones, R., Fullagar, R., Jacobsen, G., Hua, Q., 1998. Optical and radiocarbon datingat Jinmium rock shelter in northern Australia. Nature 393, 358e362.

Rose, J.I., 2006. Among Arabian Sands: Defining the Palaeolithic of Southern Arabia(Ph.D dissertation). Southern Methodist University, Dallas (Unpublished).

Rose, J.I., 2007. The Arabian Corridor Migration Model: archaeological evidence forhominin dispersals into Oman during the Middle and Upper Pleistocene. Pro-ceedings of the Seminar for Arabian Studies 37.


Rose, J.I., 2010. New light on human prehistory in the Arabo-Persian Gulf oasis.Current Anthropology 51 (6), 849e883.

Rose, J.I., Praglia, M.D., 2009. Tracking the origin and evolution of human pop-ulations in Arabia. In: Petraglia, M.D., Rose, J.I. (Eds.), Evolution of HumanPopulations in Arabia: Paleoenvironments, Prehistory and Genetics. SpringerAcademic Publishers, Dordrecht, pp. 1e12.

Rose, J.I., Usik, V.I., 2009. The “Upper Paleolithic” of South Arabia. In:Petraglia, M.D., Rose, J.I. (Eds.), Evolution of Human Populations in Arabia:Paleoenvironments, Prehistory and Genetics. Springer Academic Publishers,Dordrecht, pp. 169e185.

Rose, J.I., Usik, V.I., Marks, A.E., Hilbert, Y.H., Galletti, C.S., Parton, A., Geiling, J.M.,Cerny, V., Morley, M.W., Roberts, R.G., 2011. The Nubian complex of Dhofar,Oman: an African middle stone age industry in southern Arabia. PLoS One 6(11), e28239. http://dx.doi.org/10.1371/journal.pone/0028239.

Rose, J.I., �Cerný, V., Bayoumi, R., 2013. Tabula rasa or refugia? Using genetic data toassess the peopling of Arabia. Arabian Archaeology and Epigraphy 24, 95e101.

Uerpmann, H.-P., Potts, D., Uerpmann, M., 2009. Holocene (RE-) occupation ofEastern Arabia. In: Petraglia, M.D., Rose, J.I. (Eds.), Evolution of Human Pop-ulations in Arabia: Paleoenvironments, Prehistory and Genetics. Springer Aca-demic Publishers, Dordrecht, pp. 205e214.

Vaks, A., Bar-Matthews, M., Matthews, A., Ayalon, A., Frumkin, A., 2010. Middle-LateQuaternary paleoclimate of northern margins of the Saharan-Arabian desert:reconstruction from speleothems of Negev desert, Israel. Quaternary ScienceReviews 29, 2647e2662.

Zarins, J., 2001. The Land of Incense: Archaeological Work in the Governorate ofDhofar, Sultanate of Oman, 1990e1995. Sultan Qaboos University Publications,Muscat.

Zarins, J., 2013. Hailat Araka and the South Arabian Neolithic. Arabian Archaeologyand Epigraphy 24, 109e117.























































http://dx.doi.org/10.1371/journal.pone/0028239






















Terminal Pleistocene and Early Holocene archaeology and stratigraphy of the southern Nejd, Oman

Documents