A failed human expansion out of Africa 100,000 years ago

A failed human expansion out of Africa 100,000 years ago | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article A failed human expansion out of Africa 100,000 years ago Jeffrey Rose, Roman Garba, Lee Arnold, Amir Beshkani, Nuno Bicho, and 11 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8751643/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Early modern humans expanded out of Africa multiple times before 70,000 years ago, yet none of these populations are ancestral to living non-Africans. What caused these dispersals to fail? Here we investigate one such expansion using new evidence from southern Arabia. We present robust chronologies from four stratified sites in Dhofar, Oman, associated with Nubian Levallois technology—a distinctive stone tool tradition of African origin. The ages constrain this industry to 109-95 thousand years ago, synchronous with Nubian occurrences in the eastern Mediterranean and coinciding with the estimated timing of early modern human gene flow into Neanderthals. Integrating our chronology with palaeoenvironmental records, we show that the southern Arabian population disappeared abruptly with the onset of aridification 95 thousand years ago, with no technological continuity into later industries. These results demonstrate that early human expansions beyond Africa were climate-dependent and demographically fragile, and that the global establishment of our species was preceded by repeated failures. Biological sciences/Evolution/Archaeology Earth and environmental sciences/Climate sciences/Palaeoclimate Biological sciences/Genetics/Genetic hybridization Figures Figure 1 Figure 2 Figure 3 Figure 3 Figure 4 Introduction Early modern humans expanded out of Africa multiple times before 70,000 years ago, yet none of these populations are ancestral to living non-Africans. What caused these dispersals to fail? Here we investigate one such expansion using new evidence from southern Arabia. We present robust chronologies from four stratified sites in Dhofar, Oman, associated with Nubian Levallois technology—a distinctive stone tool tradition of African origin. The ages constrain this industry to 109-95 thousand years ago, synchronous with Nubian occurrences in the eastern Mediterranean and coinciding with the estimated timing of early modern human gene flow into Neanderthals. Integrating our chronology with palaeoenvironmental records, we show that the southern Arabian population disappeared abruptly with the onset of aridification 95 thousand years ago, with no technological continuity into later industries. These results demonstrate that early human expansions beyond Africa were climate-dependent and demographically fragile, and that the global establishment of our species was preceded by repeated failures. Genomic studies consistently demonstrate that all present-day non-Africans descend from a single ancestral population that expanded out of Africa between 70 and 50 thousand years ago (ka) 1,2 . Archaeological and fossil evidence, however, document the presence of anatomically modern humans in Southwest Asia during earlier intervals within Marine Isotope Stages 6 and 5 (MIS 6 and 5; 191–71 ka) 3-5 . Although these earlier populations left no detectable genetic legacy in living humans, recent analyses of archaic genomes reveal that modern humans contributed genetic material to Neanderthals on at least two occasions 6,7 , the most recent around 120-100 ka. What happened to these early dispersals? Did they represent sustained colonisations that were later replaced, or brief incursions that failed to establish viable populations? Palaeoclimate modelling suggests that dispersal corridors into Eurasia were intermittently available during early MIS 5, but that climatic windows closed between ~96 and 78 ka, effectively isolating any populations that had reached Southwest Asia 8 . The fate of these early human pioneers, and the reason for their genetic disappearance, remains a central puzzle in human prehistory. One potential archaeological signature of an MIS 5 dispersal is Nubian Levallois stone tool technology—a distinctive prepared core reduction strategy first identified in the Nile Valley 9 and subsequently documented across Northeast Africa 10,11 . This technology, characterized by a standardized system of core preparation to produce predetermined pointed flakes, has also been found outside Africa in the eastern Mediterranean 12-14 and the Arabian Peninsula 15-17 . Its presence in these regions has been interpreted by some as evidence for population movement during MIS 5 15 , although whether this reflects dispersal, diffusion, or convergent evolution remains debated 18,19 . In southern Arabia, the Dhofar region presents a compelling case, where over 250 Middle Palaeolithic sites have been mapped across the Nejd plateau (Fig. 1), nearly all characterized by Nubian Levallois technology 20,21 . This technological homogeneity distinguishes Dhofar from the eastern Mediterranean and central/northern Arabia, where Nubian cores co-occur with other Levallois variants 13,14,16,17 , suggesting a distinct cultural signature in the southern part of the Peninsula. Despite its potential significance for understanding early human dispersals, the chronology of the Dhofar Nubian industry remains poorly constrained. Two quartz optically stimulated luminescence (OSL) ages of ~106 and ~107 ka from the site of Aybut Auwal established an MIS 5 association 15 , but without additional dating, the duration, fate, and broader context of this industry could not be evaluated. Here we present quartz OSL ages from three new stratified Dhofar Nubian sites—TH505 (Aybut Hills 5), TH584 (Aybut Hills 6), and TH571 (Upper Amut 1)—providing the first comprehensive chronology for this industry in southern Arabia. Results Sites and lithic assemblages The studied sites are located on the Nejd plateau in the Dhofar Governorate of southern Oman, an arid limestone tableland incised by deep wadis that expose chert-rich limestone beds. At TH505 and TH584, artefact-bearing deposits consist of colluvial slope debris (angular limestone and chert clasts in a sandy-silt matrix) with evidence of post-depositional gypsum induration. In contrast, TH571 is situated in an endorheic basin and contains artefacts within a fluvial gravel layer overlain by compacted aeolian loess, indicating accumulation during an earlier humid phase prior to loess deposition. The assemblages comprise 655 lithic artefacts from stratified and surface contexts (Fig. S8). All artefacts were manufactured on locally outcropping chert. Nubian Levallois are the most frequent core type at each site (Fig. S9; Tables S6-S8). Alternative Levallois strategies (centripetal, convergent) are rare or absent—a pattern characteristic of Nubian assemblages in both Dhofar and the Negev desert 12 , 20 – 22 . The low percentage of retouched tools (4%; n = 28), proximity to raw material outcrops, and predominance of cores confirm the workshop character of these sites. Stratified contexts at all three localities contain diagnostic Nubian Levallois cores and products (Extended Data Fig. 1 ), demonstrating that the dated sediments directly bracket Dhofar Nubian occupation rather than later reoccupation events. Analysis of post-depositional surface modifications on the assemblage from TH584 shows that bleaching intensity increased significantly with stratigraphic depth (Cochran-Armitage trend test, p < 0.001). Surface artefacts exhibit minimal bleaching while buried contexts demonstrate moderate to heavy bleaching (Fig. S10; Table S11). This pattern confirms stratigraphic integrity and indicates that buried artefacts were rapidly interred following discard. Moreover, buried specimens retain sharp, undamaged edges consistent with primary deposition (Extended Data Fig. 1 ), in contrast to the high degree of edge damage characteristic of Dhofar Nubian surface assemblages 23 . OSL chronology Deposition of artefact units at the three sites is constrained by a total of eight OSL samples, yielding ages that are tightly clustered between 107 and 95 ka (Table 1 ; Fig. 2 ). The stratified deposits of Aybut Hills 5 (TH505) and Aybut Hills 6 (TH584) have weighted mean OSL ages of 106.4 ± 4.5 ka (n = 2) and 99.6 ± 2.5 ka (n = 5), respectively, while Unit B of Upper Amut 1 (TH571) is constrained to 104.5 ± 7.1 ka (n = 1). The single-grain OSL equivalent dose (D e ) datasets are consistent with those reported for well-bleached and unmixed sediments (e.g., ref. 24; overdispersion values = 25–32%; Fig. S7), and the ages are statistically indistinguishable from the multiple-grain OSL ages of 106 ± 9 ka and 107 ± 9 ka reported for the nearby site of Aybut Auwal 15 . Table 1 OSL analytical data from excavated Dhofar Nubian sites . All samples were measured using single-grain quartz (Qz) optically stimulated luminescence (OSL). Ages calculated using the central age model (CAM). Grain sizes in microns (µm); dose rates and equivalent doses (D e ) in Grays (Gy); ages in thousands of years (ka) before CE 2023. Uncertainties are 1σ. Further details of D e and dose rate datasets are provided in Tables S2-S4. Site Lab ID Field ID Depth (cm) Grain size (µm) Dose rate (Gy/ka) ± De (Gy) ± Age (ka) ± TH505 OMANRG-4 AYB3-1 24 90–180 0.67 0.03 71.0 1.8 106.2 6.3 OMANRG-5 AYB3-2 40 90–180 0.91 0.04 96.9 2.8 106.5 6.4 TH584 OMANRG-9 AYB2-1 21 90–180 0.91 0.04 87.0 2.7 95.3 5.7 OMANRG-1 AYB2-3 21 90–300 0.71 0.04 72.1 2.5 100.8 6.4 OMANRG-3 AYB2-5 32 90–125 0.82 0.04 81.7 1.9 99.9 5.5 OMANRG-10 AYB2-2 46 90–180 0.67 0.03 67.9 1.5 101.3 6.0 OMANRG-2 AYB2-4 46 90–125 0.72 0.04 72.5 1.5 100.6 5.7 TH571 OMANRG-6 AMU1-1 47 90–300 0.70 0.03 73.6 3.3 104.5 7.1 OxCal Bayesian modelling of the pooled OSL dataset (n = 10) produces a combined age range of 109.2–95.1 ka (95.4% credible interval) and a mean age of 102.1 ± 3.4 (1σ) for the Dhofar Nubian industry (Extended Data Fig. 2 ). The modelled age constrains this industry to a ~ 14,000-year window corresponding closely to MIS 5c (105 − 93 ka). Palaeovegetation Phytolith analysis of sediment samples from Aybut Auwal and Aybut Hills 6 provides direct evidence of local vegetation coinciding with Dhofar Nubian occupation (Fig. S11; Table S12). Grass (Poaceae) phytoliths dominate the assemblage, with limited palm and woody indicators. The phytolith assemblage indicates that the landscape was open and dominated by grasses, reflecting savanna or grassland conditions rather than the hyperarid desert seen today. The limited presence of palm and woody phytoliths suggests that the trees were likely restricted to watercourses and springs or locally favourable microhabitats during intervals of increased moisture. These results support regional palaeoclimate reconstructions that indicate enhanced precipitation and more hospitable environments in southern Arabia during MIS 5c 25–30 , providing a plausible context for human occupation at these sites. Discussion Our OSL chronology establishes that the Dhofar Nubian industry was restricted to a narrow temporal window corresponding to MIS 5c, between approximately 109 and 95 ka (Bayesian modelled 95.4% credible interval). This finding has several implications for understanding the nature and fate of this population. First, the Dhofar Nubian industry is contemporaneous with Nubian assemblages in the eastern Mediterranean (Fig. 3 ). At Dimona South and Ein Ziq in the Negev desert, stratified deposits with Nubian Levallois technology have been dated between 116 and 82 ka 12,22 . The connection between these geographically distant occurrences extends beyond chronological overlap. Refitting analyses at Dimona South and in Dhofar document the same operational schema for Nubian Levallois core reduction 12 , 31 , and the same conspicuous absence of other Levallois variants. This pattern supports the interpretation that Nubian technology spread rapidly across Southwest Asia during MIS 5, whether through population movement, cultural transmission, or some combination of both. Second, the Dhofar ages align chronologically with the later phases of Nubian technology in Northeast Africa. In the Nile Valley, the Nubian Complex spans a considerable time depth, with early manifestations at Sai Island dated to sometime after mid-MIS 6 (< 150 ka) 32 and occupations at Sodmein Cave 33 , 34 and Taramsa 35 throughout MIS 5. Consequently, the Dhofar Nubian industry is thought to represent the cultural footprint of this technological tradition, appearing in southern Arabia at a later stage of the Nubian Complex's temporal range. The chronological coherence of assemblages with Nubian technology in Northeast Africa, the eastern Mediterranean, and southern Arabia renders convergent technological invention an increasingly untenable alternative. Independent populations arriving at the same highly specific reduction strategy within a narrow temporal window across three adjacent regions is less parsimonious than population dispersal under favourable climatic conditions. During the MIS 5c humid interval, the Nejd plateau was a habitable savanna with reliable groundwater. Speleothem records from southern Arabian caves document active growth, indicating sufficient moisture to sustain cave dripwater systems 25 . Palaeolake deposits in the interior of Oman record high water stands around 100 ka 27 , and marine sediment cores from the Gulf of Aden show reduced aeolian input consistent with increased vegetation cover 30 . Our phytolith data provide site-level confirmation of these regional signals, demonstrating that grass-dominated landscapes existed in the immediate vicinity of Dhofar Nubian sites during occupation. The absence of archaeological material dated to earlier MIS 5e is notable given that this interval represents peak interglacial humidity across Arabia. Despite favourable conditions documented in marine and terrestrial archives, no MIS 5e sites have been identified in southern Arabia, or elsewhere on the Peninsula apart from Jebel Faya in the Hajar mountain foothills 36 . This pattern suggests that environmental conditions alone did not determine the timing of expansion; demographic growth in source regions like the Nile Valley may have been a necessary precondition. Sea level also presented a barrier. During MIS 5e, the Bab al Mandab strait remained approximately 30 km wide, however, when sea level dropped below − 50 m during MIS 5d (~ 115–105 ka), the crossing narrowed to less than 5 km via the Hanish Islands sill, with intervisibility between shorelines 37 , 38 . The Dhofar Nubian industry may thus represent an expansion via the southern dispersal route 39 , which crossed the Red Sea during the MIS 5d lowstand and spread across the interior as MIS 5c humid conditions transformed southern Arabia to grasslands. The Bayesian modelled end boundary for our OSL dataset (95.1 ka; 95.4% CI) places the terminal occupation of the Dhofar Nubian around the MIS 5c/5b boundary (~ 93 ka). Multiple lines of evidence suggest that this timing coincides with a pronounced aridification event across southern Arabia. The KL15 marine core record from the Gulf of Aden documents a sharp increase in aeolian dust flux at ~ 95 ka, signalling desertification 30 . Speleothem growth ceased in caves that had been active during MIS 5c 25 , and palaeolake systems contracted or disappeared as effective recharge ceased 26 , 27 . Essential for populations to survive the prolonged dry season, groundwater-fed springs once sustained by aquifer recharge were reduced to desiccated basins. The geographic isolation of Dhofar compounds the demographic vulnerability imposed by climate and hydrology. The Nejd plateau is bounded to the north by the hyperarid Rub' al Khali desert, which likely presented a formidable barrier, even during humid intervals, due to the sparse distribution of resources. To the east, systematic surveys in south-central Oman have recorded rare examples of Nubian sites despite extensive fieldwork 40 . The 300-km Jiddat al Harassis plain—devoid of drainages or springs—forms a significant biogeographic obstacle between Dhofar and regions to the northeast, with freshwater availability found to be the most likely limiting factor in the distribution of Nubian technology across raw material-rich landscapes of southern Arabia 41 . Without continued demographic influx from Africa, remnant populations in southern Arabia became susceptible to stochastic local extinction 8 . Based on the current dating evidence, we interpret its disappearance after ~ 95 ka as coinciding with environmental collapse and demographic fragmentation. The fate of the Dhofar Nubian population bears directly on whether early expansions contributed to the ancestry of later Eurasians. The subsequent Mudayyan industry in Dhofar shows no technological continuity with its predecessor. The Dhofar Nubian is characterized by large cores and preferential flaking; the Mudayyan, in contrast, exhibits systematic miniaturization, volumetric core configurations, and recurrent bidirectional flaking 21 . Our excavated assemblages fall unambiguously within Dhofar Nubian parameters (Fig. 4 ), and despite surveys documenting over 250 Middle Palaeolithic sites across southern Oman 20 , no assemblages show intermediate characteristics between these two industries. Taphonomic evidence supports temporal separation: Dhofar Nubian assemblages exhibit heavier patination and more advanced chemical dissolution than Mudayyan assemblages on the same landscapes 21 , 23 . The Dhofar Nubian appears, flourishes, and disappears as a cohesive package. This abrupt termination followed by a technologically disjunctive industry indicates demographic discontinuity, not in situ evolution. The pattern of ephemeral occupation followed by population collapse aligns with findings from northern Arabia, where multiple hominin dispersals over the past 400,000 years each produced distinct material culture, with no evidence for long-term continuity 42 . This does not mean the MIS 5 dispersal left no genetic trace. Recent analyses identified at least two episodes of gene flow from modern humans into Neanderthals, the more recent around 120 − 100 ka 6,7 . The Dhofar Nubian population, far from the known Neanderthal range, would not have contributed directly to this introgression. However, the existence of contemporaneous Nubian assemblages in the eastern Mediterranean 12 , 14 , 22 falls within the zone of overlap, raising the possibility that a related population was responsible. Under this scenario, the expansion signalled by Nubian technology split into two branches: one entering the eastern Mediterranean, encountering Neanderthals, and contributing genes before disappearing; the other crossing the Red Sea into southern Arabia, persisting during MIS 5c, and succumbing to environmental collapse. Neither contributed ancestry to living humans. These findings add to a growing body of genomic evidence that the successful colonisation of Eurasia occurred after 80 ka 1,2 . Earlier dispersals were demographically fragile and did not establish an enduring foothold outside Africa. Isolating the factors that allowed later populations to overcome the constraints of these earlier pulses—be they environmental, demographic, or cultural—is the next challenge in reconstructing our global prehistory. Methods Excavation procedures Sites were selected based on evidence of buried archaeological deposits, identified by bleached Nubian Levallois artefacts eroding from sedimentary contexts (TH505, TH584) or proximity to sedimentary basins with preservation potential (TH571). All excavations followed a 1×1 m grid system and proceeded by geologic unit. Sediments were excavated by trowel and dry-screened through 5 mm mesh. At TH584, artefacts were piece-plotted using a total station during the 2023 season; at other sites, artefacts were collected by square and geologic unit. Single-grain OSL dating Eight sediment samples from the three excavated sites were dated using optically stimulated luminescence (OSL) of quartz grains. The OSL dating samples were taken from cleaned sedimentary exposures using metal tubes, with additional bulk sediment collected from the surrounding few cm of each tube for beta dose rate determination and water content analysis. Purified coarse grain quartz extracts (90–125, 90–180 or 90–300 µm) were processed under safe light conditions (630 nm LEDs, < 0.15 µW/cm 2 power density at sample position) at Adelaide University using standard preparation procedures (e.g., ref. 43). Single-grain OSL measurements were made using the experimental apparatus, single-aliquot regenerative-dose (SAR) procedures, and quality assurance criteria published previously by Arnold et al. 44,45 , which are further detailed in the Supplementary Information. Between 800 and 2000 single-grain equivalent dose (D e ) measurements were made for each sample using the OSL SAR procedure shown in Table S1 , which yielded suitable dose-recovery test results for sample OMANRG-10 (Fig. S6). Environmental dose rates were evaluated using instrumental neutron activation analysis (INAA) of bulk sediment collected from the OSL sample position. U, Th and K concentrations were measured by INAA at the Nuclear Physics Institute of the Czech Academy of Sciences (Table S4). The dose rate calculations additionally take into account cosmic ray contributions, an assumed minor internal alpha dose rate, beta-dose attenuation and long-term sediment water content. Detailed OSL dating methods and results are available in Supplementary Information Section 2, including discussions of the D e distributions and statistical age models used to derive representative burial dose estimates for each sample. In general, the single-grain OSL D e distributions exhibit limited scatter and are characteristic of well-bleached, unmixed samples (e.g., refs. 44, 46–48), hence we have used the weighted mean D e values (calculated using the central age model; CAM) to derive final burial dose estimates (Table 1 ). Bayesian age modelling OxCal Bayesian modelling 49 has been used to calculate a combined age range for stratified Dhofar Nubian industry units preserved at the excavated sites. For this purpose, we have pooled all OSL ages (chronometric likelihoods) in direct association with Nubian deposits from the present study and Rose et al. 15 as a single, unordered regional Phase model with delineating start and end boundaries. The combined age range for the Nubian industry of Dhofar has been calculated from the modelled posterior probabilities of the start and end boundaries of this regional Phase model using the OxCal date function. The CQL code used to construct the Dhofar Nubian phase model is provided in Supplementary Information Section 2. The ten OSL dating likelihoods from sites TH505, TH584, TH571 and TH59 have been input into the model as calendar ages before CE 2023 (the year of sample collection for the dating samples from the current study), together with their associated 1σ uncertainty ranges. As the two OSL samples published by Rose et al. 15 were collected 13 years earlier than the OSL samples from the present study, these likelihoods have been adjusted to the CE 2023 common baseline prior to running the model by adding 13 years to the original age estimates. The Bayesian model was run using the general outlier function 50 with prior outlier probabilities of 5% assigned to all dating samples. Likelihood estimates that yielded posterior outlier probabilities > 5% were not excluded from the final model but were proportionally down-weighted in the iterative Markov Chain Monte Carlo runs 50 . The Bayesian modelling results for the Dhofar Nubian occupation are summarised in Extended Data Fig. 2 and Table S5. The model exhibits a median convergence integral of 99.4% for all individual posterior distributions, and none of the ten likelihoods included in the model are identified as major statistical outliers. All likelihoods exhibit posterior outlier probabilities equal to, or less than, the specified prior threshold of 5% (Table S5). Lithic analysis Lithic analysis focused on cores, the most technologically diagnostic component of these workshop assemblages. All cores and retouched tools were subject to full attribute analysis; debitage was sorted by category and counted. Nubian core attributes were classified following the technological criteria described by Usik et al. 31 , which established morphological parameters for distinguishing Nubian from other Levallois technologies. Metric attributes recorded for each core included maximum length, maximum width, midpoint thickness, platform dimensions, and median distal ridge angle on Nubian cores. Geometric mean (∛[length × width × thickness]) was calculated as a size-independent measure of core volume. Taphonomic condition of surface artefacts was recorded following the rubric described in Rose et al. 23 . Phytolith analysis Six sediment samples were processed from TH59 (n = 3) and TH584 (n = 3) at the La Trobe University Palaeosciences Lab. Fifteen grams of each sample were dissolved in 7% HCl to remove carbonates; the absence of organics precluded treatment with nitric acid or hydrogen peroxide. Samples were deflocculated in 5% Calgon, followed by density separation using sodium polytungstate (2.5 g/cm³). Ten microlitres of each processed sample were mounted on slides and examined at ×400 magnification using a Zeiss light microscope. Phytoliths were classified by morphotype and measured using a calibrated eyepiece graticule. Declarations Data availability All R code with raw lithic data, statistical analyses, and scripts for figure generation have been deposited in an OSF research compendium and can be accessed via https://osf.io/jt7ke/. Lithic, OSL, and phytolith data are available in Supplementary Information. Reporting summary Further information on research design is available in the Nature Research Reporting Summary linked to this paper. Acknowledgements We wish to thank the Ministry of Heritage and Tourism in Oman for permissions and logistical support. We are grateful to Rhianna Power (Adelaide University) for assistance with preparing and measuring the OSL dating samples. We thank Alan Cooper for his intellectual input regarding the conceptual framework of this study. Funding declarations This work was funded by a Czech Academy of Sciences Praemium Academiae awarded to V.C and Australian Research Council Discovery project DP230100152 awarded to M.M.-W., L.A. and R.G. Fieldwork was also funded in part by a private donation from A.E.M. J.I.R. is supported by the European Research Council under the European Union’s Horizon 2020 research and innovation programme awarded to N.B. (Advanced Grant, project DISPERSALS 101052761). INAA was supported by the long-term conceptual development project RVO 61389005, neutron irradiation was performed under CICRR infrastructure supported by MEYS project LM2023041. Author contributions J.I.R. and A.E.M. conceived and designed the study. J.I.R., R.G., A.B., D.C., R.D., E.S.H, A.A.Mahri, O.S., V.I.U., and A.E.M. conducted field surveys and excavations. D.C. and R.D. documented stratigraphic profiles and provided geological and sedimentological assessments. L.A. performed optically stimulated luminescence dating and Bayesian age modelling. J.K. conducted neutron activation analysis. M.M.-W. conducted phytolith analysis. J.I.R., O.S., and R.G. created GIS basemaps. V.I.U., A.E.M., O.S., E.S.H, and J.I.R. analysed lithic assemblages. J.I.R. performed statistical analyses, wrote code, drafted figures, and created online repository. J.I.R. wrote the manuscript with contributions from R.G., L.A., M.M.-W., O.S., N.B., R.D., and V.C. R.G., V.C., N.B., M.M.-W., and A.E.M. provided funding sources. R.G. organized fieldwork permits. A.A.Mahrooqi, A.A.Mahri, and R.G. provided administrative and logistical support. All authors reviewed and approved the final manuscript. Competing interests The authors declare no competing interests. Correspondence and requests for materials should be addressed to J. I. Rose or R. Garba. References Bergström A, Stringer C, Hajdinjak M, Scerri EML, Skoglund P (2021) Origins of modern human ancestry. Nature 590:229–237 Mallick S et al (2016) The Simons Genome Diversity Project: 300 genomes from 142 diverse populations. Nature 538:201–206 Grün R et al (2005) U-series and ESR analyses of bones and teeth relating to the human burials from Skhul. J Hum Evol 49:316–334 Valladas H et al (1988) Thermoluminescence dating of Mousterian 'proto-Cro-Magnon' remains from Israel and the origin of modern man. Nature 331:614–616 Hershkovitz I et al (2018) The earliest modern humans outside Africa. Science 359:456–459 Kuhlwilm M et al (2016) Ancient gene flow from early modern humans into Eastern Neanderthals. Nature 530:429–433 Li L, Comi TJ, Bierman RF, Akey JM (2024) Recurrent gene flow between Neanderthals and modern humans over the past 200,000 years. Science 385:eadi1768 Beyer RM, Krapp M, Eriksson A, Manica A (2021) Climatic windows for human migration out of Africa in the past 300,000 years. Nat Commun 12:4889 Marks AE (1968) The Mousterian industries of Nubia. In: Wendorf F (ed) The Prehistory of Nubia, vol 1. Southern Methodist University, pp 194–314 Van Peer P, Vermeersch PM (2000) The Nubian Complex and the dispersal of modern humans in North Africa. Stud Afr Archaeol 7:47–60 Hallinan E (2024) A synthesis and critical inventory of Nubian cores in Middle Stone Age and Middle Palaeolithic assemblages. J Open Archaeol Data 12:9 Oron M et al (2024) Nubian Levallois technology during MIS 5: refitted lithic sequences and OSL ages of Dimona South, Israel. J Paleolithic Archaeol 7:4 Goder-Goldberger M, Gubenko N, Hovers E (2016) Diffusion with modifications: Nubian assemblages in the central Negev highlands of Israel. Quat Int 408:121–139 Hilbert YH et al (2026) Nubian cores along the coast of the central Levantine corridor: Exploring variability of Levallois point production in Lebanon. J Paleolithic Archaeol 9:3 Rose JI et al (2011) The Nubian Complex of Dhofar, Oman: an African Middle Stone Age industry in southern Arabia. PLoS ONE 6:e28239 Hilbert YH, Crassard R, Charloux G, Loreto R (2016) Nubian technology in northern Arabia: Impact on interregional variability of Middle Paleolithic industries. Quat Int 408:68–84 Crassard R et al (2018) Middle Palaeolithic occupations in central Saudi Arabia during MIS 5 and MIS 7: new insights on the origins of the peopling of Arabia. Archaeol Anthropol Sci 11:3101–3120 Groucutt HS (2020) Culture and convergence: the curious case of the Nubian Complex. In: Groucutt HS (ed) Culture History and Convergent Evolution. Springer, pp 55–86 Hallinan E et al (2022) The nature of Nubian: developing current global perspectives on Nubian Levallois technology and the Nubian complex. Evol Anthropol 31:227–232 Rose JI et al (2023) Prehistoric settlement patterns in southern Oman. Paléorient 49:83–108 Rose JI et al (2025) Lithic miniaturization and technological innovation on the Southern Dispersal Route out of Africa. J Paleolithic Archaeol 8:40 Oron M et al (2025) Nubian Levallois cores from MIS 5 alluvial terraces in the Negev Desert: new insights into the Middle Paleolithic in the arid regions of the southern Levant. J Paleolithic Archaeol 8:4 Rose JI et al (2025) Mapping lateral stratigraphy at Palaeolithic surface sites: a case study from Dhofar, Oman. J Archaeol Sci 173:106117 Arnold LJ, Roberts RG (2009) Stochastic modelling of multi-grain equivalent dose (De) distributions: Implications for OSL dating of sediment mixtures. Quat Geochronol 4:204–230 Fleitmann D, Matter A (2009) The speleothem record of climate variability in Southern Arabia. C R Geosci 341:633–642 Rosenberg TM et al (2011) Humid periods in southern Arabia: windows of opportunity for modern human dispersal. Geology 39:1115–1118 Rosenberg TM et al (2012) Late Pleistocene palaeolake in the interior of Oman: a potential key area for the dispersal of anatomically modern humans out-of-Africa? J Quat Sci 27:13–16 Matter A, Neubert E, Preusser F, Rosenberg T, Al-Wagdani K (2015) Palaeo-environmental implications derived from lake and sabkha deposits of the southern Rub' al-Khali, Saudi Arabia and Oman. Quat Int 382:120–131 Nicholson SL et al (2020) Pluvial periods in Southern Arabia over the last 1.1 million-years. Quat Sci Rev 229:106112 Fischer ML et al (2024) Spatio-temporal variations of climate along possible African–Arabian routes of H. sapiens expansion. Quat Sci Adv 14:100174 Usik VI, Rose JI, Hilbert YH, Van Peer P, Marks AE (2013) Nubian Complex reduction strategies in Dhofar. Quat Int 300:244–266 Van Peer P et al (2003) The early to Middle Stone Age transition and the emergence of modern human behaviour at site 8-B-11, Sai Island, Sudan. J Hum Evol 45:187–193 Mercier N et al (1999) TL dates of burnt lithics from Sodmein Cave (Red Sea Mountains, Egypt). Quat Sci Rev 18:623–630 Schmidt C et al (2015) Multi-emission luminescence dating of heated chert from the Middle Stone Age sequence at Sodmein Cave (Red Sea Mountains, Egypt). J Archaeol Sci 63:94–103 Van Peer P, Vermeersch PM, Paulissen E (2010) Chert Quarrying, Lithic Technology and a Modern Human Burial at the Palaeolithic Site of Taramsa 1, Upper Egypt. Leuven University Armitage SJ et al (2011) The southern route 'out of Africa': evidence for an early expansion of modern humans into Arabia. Science 331:453–456 Lambeck K et al (2011) Sea level and shoreline reconstructions for the Red Sea: isostatic and tectonic considerations and implications for hominin migration out of Africa. Quat Sci Rev 30:3542–3574 Hill J et al (2022) Sea-level change, palaeotidal modelling and hominin dispersals: The case of the southern Red Sea. Quat Sci Rev 293:107719 Lahr MM, Foley R (1994) Multiple dispersals and modern human origins. Evol Anthropol 3:48–60 Chlachula D et al (2025) Evidence of Middle Palaeolithic human occupation in south-central Oman. Antiquity 99(404):e9 Eren MI et al (2025) Examining the distribution of Middle Paleolithic Nubian cores relative to chert quality in southern (Nejd, Dhofar) and south-central (Duqm, Al Wusta) Oman. Geoarchaeology 40(1), e22019 Groucutt HS et al (2021) Multiple hominin dispersals into Southwest Asia over the past 400,000 years. Nature 597:376–380 Arnold LJ et al (2024) Single-grain luminescence and combined U-series/ESR dating of the early Upper Palaeolithic Lagar Velho Rock Shelter, Leiria, Portugal. Quat Geochronol 83:101572 Arnold LJ et al (2016) OSL dating of individual quartz ‘supergrains’ from the Ancient Middle Palaeolithic site of Cuesta de la Bajada, Spain. Quat Geochronol 36:78–101 Arnold LJ et al (2022) Examining sediment infill dynamics at Naracoorte Cave megafauna sites using multiple luminescence dating signals. Quat Geochronol 70:101301 Bailey RM, Arnold LJ (2006) Statistical modelling of single grain quartz De distributions and an assessment of procedures for estimating burial dose. Quat Sci Rev 25:2475–2502 Arnold LJ, Bailey RM, Tucker GE (2007) Statistical treatment of fluvial dose distributions from southern Colorado arroyo deposits. Quat Geochronol 2:162–167 Arnold LJ et al (2008) Optical dating of perennially frozen deposits associated with preserved ancient plant and animal DNA in north-central Siberia. Quat Geochronol 3:114–136 Bronk Ramsey C (2009) Bayesian analysis of radiocarbon dates. Radiocarbon 51:337–360 Bronk Ramsey C (2009) Dealing with outliers and offsets in radiocarbon dating. Radiocarbon 51:1023–1045 Additional Declarations There is NO Competing Interest. Supplementary Files Roseetal.Failed5cexpansionSI2026.02.05.docx floatimage5.png Extended Data Fig. 1 | Selected diagnostic lithic artefacts from stratified contexts at study sites. (A) Cores from stratified deposits: 1, Nubian Levallois (TH505, Unit D); 2, Nubian Levallois (TH584, Unit A); 3, bidirectional (TH584, Unit B); 4, Exhausted Levallois (TH584, Unit B). (B) Nubian Levallois products: 5,7 (TH571, Unit B); 6,8,9 (TH505, Unit B), 10 (TH584, Unit B). Scale bar = 5 cm. floatimage6.png Extended Data Fig. 2 | Bayesian chronology of Dhofar Nubian sites . OxCal Bayesian phase model integrating all ten OSL ages from sites TH505, TH584, TH571 and TH59. The unmodelled age distributions for the OSL dating determinations (likelihoods) are shown as lighter blue probability distribution functions (PDFs). The modelled posterior distributions for the dating determinations and phase model boundaries are shown as darker blue and maroon PDFs, respectively. The grey PDF shows the combined age estimate for the Dhofar Nubian industry (109-95 ka, 95.4% CI), calculated using the date function. Unmodelled and modelled ages are shown on the calendar year timescale, and both are expressed in years before CE 2023. The white circles and associated uncertainty bars represent the mean ages and 1σ uncertainty ranges of the PDFs. The 68.3% and 95.4% ranges of the highest posterior probabilities are indicated by the horizontal bars underneath the PDFs. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8751643","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":586906291,"identity":"e271ca72-277d-46a2-8773-ee74b6bdee13","order_by":0,"name":"Jeffrey Rose","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAklEQVRIiWNgGAWjYBACxgbGBx9ADDYI3wYk1ngAvxZmwxlIWtLAYni1MDBAtUDBYTCJVwtzezNjc0XFYQY+Bt6HD39UnLdb234YaEuNTTROh/UcZmw8c+Yw0GHsxsY8Z24nbzuTCNRyLC23AZeWGfnHHza2gbSwsUkztt1ONjsA1MLYcBi3lvmPGRsb/0G0SP78dy7Z7PxDAlpmMAO1NEC0SPA2HLAzu0HIlp5kxsaGY+k8bMxszMY8x5ITzG4AbUnA4xfDdqD3G2qs5eTb2xgf/qixszc7n/7wwYcaG9xaIBLNPAzMEIFEsEACDuUgIA+h6uAC9ngUj4JRMApGwQgFAJ2VXba4Qw1XAAAAAElFTkSuQmCC","orcid":"","institution":"Universidade do Algarve","correspondingAuthor":true,"prefix":"","firstName":"Jeffrey","middleName":"","lastName":"Rose","suffix":""},{"id":586906292,"identity":"7801d0eb-3999-44f9-a307-b5b268faca44","order_by":1,"name":"Roman Garba","email":"","orcid":"","institution":"Czech Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Roman","middleName":"","lastName":"Garba","suffix":""},{"id":586906293,"identity":"2c389d9e-bbc7-432a-81cb-384ea617f3ec","order_by":2,"name":"Lee Arnold","email":"","orcid":"https://orcid.org/0000-0001-9603-3824","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Lee","middleName":"","lastName":"Arnold","suffix":""},{"id":586906294,"identity":"3845dd8b-e0c2-4dca-99c4-7379df985331","order_by":3,"name":"Amir Beshkani","email":"","orcid":"","institution":"National Museum of Natural History","correspondingAuthor":false,"prefix":"","firstName":"Amir","middleName":"","lastName":"Beshkani","suffix":""},{"id":586906295,"identity":"8f39f421-59e5-462f-bdcc-b294872b915e","order_by":4,"name":"Nuno Bicho","email":"","orcid":"https://orcid.org/0000-0001-9655-0549","institution":"Universidade do Algarve","correspondingAuthor":false,"prefix":"","firstName":"Nuno","middleName":"","lastName":"Bicho","suffix":""},{"id":586906296,"identity":"4866a830-a042-4bb6-a640-f1294b2ff43e","order_by":5,"name":"Viktor Černý","email":"","orcid":"https://orcid.org/0000-0003-1197-6634","institution":"Institute of Archaeology of the Academy of Sciences of the Czech Republic","correspondingAuthor":false,"prefix":"","firstName":"Viktor","middleName":"","lastName":"Černý","suffix":""},{"id":586906297,"identity":"7ba06f59-e429-42e3-98b8-13cc06508d9f","order_by":6,"name":"Dominik Chlachula","email":"","orcid":"","institution":"Czech Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Dominik","middleName":"","lastName":"Chlachula","suffix":""},{"id":586906298,"identity":"e42a2ee5-47f9-4a36-9938-360f5abb0a09","order_by":7,"name":"Rudolf Dellmour","email":"","orcid":"","institution":"University of Vienna","correspondingAuthor":false,"prefix":"","firstName":"Rudolf","middleName":"","lastName":"Dellmour","suffix":""},{"id":586906299,"identity":"a31fa5a3-8794-4e35-9f63-78b98da7fe4b","order_by":8,"name":"Emily Hallinan","email":"","orcid":"","institution":"Universidade do Algarve","correspondingAuthor":false,"prefix":"","firstName":"Emily","middleName":"","lastName":"Hallinan","suffix":""},{"id":586906300,"identity":"35545a4f-c468-48ea-b431-34c6172d61b4","order_by":9,"name":"Jan Kamenik","email":"","orcid":"https://orcid.org/0000-0001-7740-7683","institution":"Czech Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Jan","middleName":"","lastName":"Kamenik","suffix":""},{"id":586906301,"identity":"75c8e785-dda4-4c9d-b95d-c42efdf4993f","order_by":10,"name":"Ali Al Mahri","email":"","orcid":"","institution":"Sultan Qaboos University","correspondingAuthor":false,"prefix":"","firstName":"Ali","middleName":"Al","lastName":"Mahri","suffix":""},{"id":586906302,"identity":"dcd729ad-cd04-4fad-86ff-89464fea9c30","order_by":11,"name":"Al Al Mahrooqi","email":"","orcid":"","institution":"Ministry of Heritage and Tourism","correspondingAuthor":false,"prefix":"","firstName":"Al","middleName":"Al","lastName":"Mahrooqi","suffix":""},{"id":586906303,"identity":"009d59d4-8c4b-4960-8850-c1cfcf5619f5","order_by":12,"name":"Matthew Meredith-Williams","email":"","orcid":"https://orcid.org/0000-0001-5284-641X","institution":"Palaeoscience, Department of Archaeology and History, La Trobe University","correspondingAuthor":false,"prefix":"","firstName":"Matthew","middleName":"","lastName":"Meredith-Williams","suffix":""},{"id":586906304,"identity":"4749d8b7-16cc-4106-a0dd-a326d2512a80","order_by":13,"name":"Osama Samawi","email":"","orcid":"","institution":"Universidade do Algarve","correspondingAuthor":false,"prefix":"","firstName":"Osama","middleName":"","lastName":"Samawi","suffix":""},{"id":586906305,"identity":"69c4457d-14ce-43f5-8db6-257afbde66dc","order_by":14,"name":"Vitaly Usik","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Vitaly","middleName":"","lastName":"Usik","suffix":""},{"id":586906306,"identity":"2b82f39a-95e1-419b-b7a0-c505b9a85e6d","order_by":15,"name":"Anthony Marks","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Anthony","middleName":"","lastName":"Marks","suffix":""}],"badges":[],"createdAt":"2026-01-31 17:30:09","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-8751643/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8751643/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102826678,"identity":"cef5cae8-12b5-4001-b6ce-25ea741c6532","added_by":"auto","created_at":"2026-02-17 09:00:13","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2455551,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGeographic context of Dhofar Nubian sites.\u003c/strong\u003e (\u003cstrong\u003eA\u003c/strong\u003e) Inter-regional distribution of sites with Nubian Levallois technology in Northeast Africa and Southwest Asia. Bibliographic details with site locations are provided in Supplementary Information Section 6. White box indicates area depicted in Panel B. (\u003cstrong\u003eB\u003c/strong\u003e) Distribution of Nubian Levallois sites on the western Nejd plateau. (\u003cstrong\u003eC\u003c/strong\u003e) View north across the Nejd plateau showing the dissected terrain characteristic of the study area. Base map data source for A \u0026amp; B: GEBCO 2025 Grid (http://gebco.net), colour shaded-relief image generated using QGIS v.3.44.6 Solothurn (EPSG:3035).\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8751643/v1/5135c9f66adad7beb12bcaa7.png"},{"id":102826683,"identity":"76559cfd-4cda-4ab4-8255-572c3f54ba74","added_by":"auto","created_at":"2026-02-17 09:00:14","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":365327,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eStratigraphic context of Dhofar Nubian sites. \u003c/strong\u003eStratigraphic columns from excavated sites showing sedimentary units, artefact distributions, and OSL and phytolith sample locations. Ages reported as ka ± 1σ.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8751643/v1/f42de9746cebd7147bd4f2e0.png"},{"id":102962921,"identity":"ce4ad10e-4d7f-485d-9428-4ae2a9fc1193","added_by":"auto","created_at":"2026-02-19 04:12:10","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":474332,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eChronology of Nubian Levallois sites across Northeast Africa and Southwest Asia in relation to regional palaeoclimate and sea level.\u003c/strong\u003e (\u003cstrong\u003eA\u003c/strong\u003e) Numeric ages from Nubian sites in Northeast Africa plotted against ODP 967 δ¹⁸O, a proxy for Nile discharge. Green shading indicates humid anomalies (higher discharge); tan shading indicates arid anomalies, with colour intensity proportional to deviation from baseline. Marine Isotope Stage (MIS) boundaries are labelled at top. (\u003cstrong\u003eB\u003c/strong\u003e) ODP 967/968 humidity index with OSL ages from Nubian sites in the Negev and northern Arabia. Purple bar above curve indicates the estimated timing of H. sapiens to Neanderthal gene flow. (\u003cstrong\u003eC\u003c/strong\u003e) OSL ages from Dhofar Nubian sites plotted against the KL15 humidity index and Red Sea relative sea level. The Red Sea level secondary axis aligns with the humidity baseline at -50 m, which is the threshold for intervisibility across the Bab al Mandab strait at the Hanish Sill. Blue vertical bands highlight periods when both low sea level (\u0026lt;-50 m) and positive humidity anomalies occurred simultaneously, signalling optimal windows for crossing the Red Sea. Error bars show 1σ uncertainties; arrows indicate minimum or maximum age constraints; error bars with end caps indicate direct bracketing ages. Bibliographic details for all palaeoclimate and sea-level proxy data sources are provided in Supplementary Information Section 6.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8751643/v1/4fdb2aa31e605d02dfec426c.png"},{"id":102868843,"identity":"c95279d7-ab8c-4675-9dc2-ff423b69bcc9","added_by":"auto","created_at":"2026-02-17 17:25:08","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":474332,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eChronology of Nubian Levallois sites across Northeast Africa and Southwest Asia in relation to regional palaeoclimate and sea level.\u003c/strong\u003e (\u003cstrong\u003eA\u003c/strong\u003e) Numeric ages from Nubian sites in Northeast Africa plotted against ODP 967 δ¹⁸O, a proxy for Nile discharge. Green shading indicates humid anomalies (higher discharge); tan shading indicates arid anomalies, with colour intensity proportional to deviation from baseline. Marine Isotope Stage (MIS) boundaries are labelled at top. (\u003cstrong\u003eB\u003c/strong\u003e) ODP 967/968 humidity index with OSL ages from Nubian sites in the Negev and northern Arabia. Purple bar above curve indicates the estimated timing of H. sapiens to Neanderthal gene flow. (\u003cstrong\u003eC\u003c/strong\u003e) OSL ages from Dhofar Nubian sites plotted against the KL15 humidity index and Red Sea relative sea level. The Red Sea level secondary axis aligns with the humidity baseline at -50 m, which is the threshold for intervisibility across the Bab al Mandab strait at the Hanish Sill. Blue vertical bands highlight periods when both low sea level (\u0026lt;-50 m) and positive humidity anomalies occurred simultaneously, signalling optimal windows for crossing the Red Sea. Error bars show 1σ uncertainties; arrows indicate minimum or maximum age constraints; error bars with end caps indicate direct bracketing ages. Bibliographic details for all palaeoclimate and sea-level proxy data sources are provided in Supplementary Information Section 6.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8751643/v1/56af5de971108bd778fe5978.png"},{"id":102826680,"identity":"012ba8ff-bab3-4ee2-8994-bacd7eebef31","added_by":"auto","created_at":"2026-02-17 09:00:13","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":239641,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTechnological distinction between Dhofar Nubian and Mudayyan industries.\u003c/strong\u003e (\u003cstrong\u003eA\u003c/strong\u003e) Ternary plot of assemblage-level core reduction strategies showing proportions of Nubian, bidirectional, and unidirectional core types. \u003cstrong\u003e(B)\u003c/strong\u003e Bivariate plot of core dimensions (length vs thickness) for the same three core types. Density contours represent 6-bin two-dimensional kernel density estimations of the reference assemblages. Points represent individual specimens from the dated test sites.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8751643/v1/e8c2b637d0290ab7aeb18a81.png"},{"id":106959140,"identity":"a47457a7-25c2-4e4c-95d1-c64df758373e","added_by":"auto","created_at":"2026-04-15 08:48:36","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4825810,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8751643/v1/1260e2d9-94d5-449c-9ce0-ae014d6277b4.pdf"},{"id":102826685,"identity":"4a33bc44-4041-4799-9689-2aa558238085","added_by":"auto","created_at":"2026-02-17 09:00:14","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":19455144,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"Roseetal.Failed5cexpansionSI2026.02.05.docx","url":"https://assets-eu.researchsquare.com/files/rs-8751643/v1/07d54f1610a0773bf986695c.docx"},{"id":102826684,"identity":"84b00033-4306-41ab-b64f-d8cf04d2a2c5","added_by":"auto","created_at":"2026-02-17 09:00:14","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":2724882,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExtended Data Fig. 1 | Selected diagnostic lithic artefacts from stratified contexts at study sites.\u003c/strong\u003e (\u003cstrong\u003eA\u003c/strong\u003e) Cores from stratified deposits: 1, Nubian Levallois (TH505, Unit D); 2, Nubian Levallois (TH584, Unit A); 3, bidirectional (TH584, Unit B); 4, Exhausted Levallois (TH584, Unit B). (\u003cstrong\u003eB\u003c/strong\u003e) Nubian Levallois products: 5,7 (TH571, Unit B); 6,8,9 (TH505, Unit B), 10 (TH584, Unit B). Scale bar = 5 cm.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8751643/v1/025f60766b4a435f0fc264fa.png"},{"id":102826682,"identity":"a712fe55-68ce-4db6-b2b6-611d55fc6500","added_by":"auto","created_at":"2026-02-17 09:00:13","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":154762,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExtended Data Fig. 2 | Bayesian chronology of Dhofar Nubian sites\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e.\u003c/strong\u003e\u003c/em\u003e\u003cem\u003e \u003c/em\u003eOxCal Bayesian \u003cem\u003ephase\u003c/em\u003e model integrating all ten OSL ages from sites TH505, TH584, TH571 and TH59. The unmodelled age distributions for the OSL dating determinations (likelihoods) are shown as lighter blue probability distribution functions (PDFs). The modelled posterior distributions for the dating determinations and \u003cem\u003ephase\u003c/em\u003e model boundaries are shown as darker blue and maroon PDFs, respectively. The grey PDF shows the combined age estimate for the Dhofar Nubian industry (109-95 ka, 95.4% CI), calculated using the \u003cem\u003edate\u003c/em\u003efunction. Unmodelled and modelled ages are shown on the calendar year timescale, and both are expressed in years before CE 2023. The white circles and associated uncertainty bars represent the mean ages and 1σ uncertainty ranges of the PDFs. The 68.3% and 95.4% ranges of the highest posterior probabilities are indicated by the horizontal bars underneath the PDFs.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8751643/v1/2176c58534dc99fc20e57d7b.png"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"A failed human expansion out of Africa 100,000 years ago","fulltext":[{"header":"Introduction","content":"\u003cp\u003eEarly modern humans expanded out of Africa multiple times before 70,000 years ago, yet none of these populations are ancestral to living non-Africans. What caused these dispersals to fail? Here we investigate one such expansion using new evidence from southern Arabia. We present robust chronologies from four stratified sites in Dhofar, Oman, associated with Nubian Levallois technology—a distinctive stone tool tradition of African origin. The ages constrain this industry to 109-95 thousand years ago, synchronous with Nubian occurrences in the eastern Mediterranean and coinciding with the estimated timing of early modern human gene flow into Neanderthals. Integrating our chronology with palaeoenvironmental records, we show that the southern Arabian population disappeared abruptly with the onset of aridification 95 thousand years ago, with no technological continuity into later industries. These results demonstrate that early human expansions beyond Africa were climate-dependent and demographically fragile, and that the global establishment of our species was preceded by repeated failures.\u003c/p\u003e\n\u003cp\u003eGenomic studies consistently demonstrate that all present-day non-Africans descend from a single ancestral population that expanded out of Africa between 70 and 50 thousand years ago (ka)\u003csup\u003e1,2\u003c/sup\u003e. Archaeological and fossil evidence, however, document the presence of anatomically modern humans in Southwest Asia during earlier intervals within Marine Isotope Stages 6 and 5 (MIS 6 and 5; 191–71 ka)\u003csup\u003e3-5\u003c/sup\u003e. Although these earlier populations left no detectable genetic legacy in living humans, recent analyses of archaic genomes reveal that modern humans contributed genetic material to Neanderthals on at least two occasions\u003csup\u003e6,7\u003c/sup\u003e, the most recent around 120-100 ka. What happened to these early dispersals? Did they represent sustained colonisations that were later replaced, or brief incursions that failed to establish viable populations? Palaeoclimate modelling suggests that dispersal corridors into Eurasia were intermittently available during early MIS 5, but that climatic windows closed between ~96 and 78 ka, effectively isolating any populations that had reached Southwest Asia\u003csup\u003e8\u003c/sup\u003e. The fate of these early human pioneers, and the reason for their genetic disappearance, remains a central puzzle in human prehistory.\u003c/p\u003e\n\u003cp\u003eOne potential archaeological signature of an MIS 5 dispersal is Nubian Levallois stone tool technology—a distinctive prepared core reduction strategy first identified in the Nile Valley\u003csup\u003e9\u003c/sup\u003e and subsequently documented across Northeast Africa\u003csup\u003e10,11\u003c/sup\u003e. This technology, characterized by a standardized system of core preparation to produce predetermined pointed flakes, has also been found outside Africa in the eastern Mediterranean\u003csup\u003e12-14\u003c/sup\u003e and the Arabian Peninsula\u003csup\u003e15-17\u003c/sup\u003e. Its presence in these regions has been interpreted by some as evidence for population movement during MIS 5\u003csup\u003e15\u003c/sup\u003e, although whether this reflects dispersal, diffusion, or convergent evolution remains debated\u003csup\u003e18,19\u003c/sup\u003e. In southern Arabia, the Dhofar region presents a compelling case, where over 250 Middle Palaeolithic sites have been mapped across the Nejd plateau (Fig. 1), nearly all characterized by Nubian Levallois technology\u003csup\u003e20,21\u003c/sup\u003e. This technological homogeneity distinguishes Dhofar from the eastern Mediterranean and central/northern Arabia, where Nubian cores co-occur with other Levallois variants\u003csup\u003e13,14,16,17\u003c/sup\u003e, suggesting a distinct cultural signature in the southern part of the Peninsula. Despite its potential significance for understanding early human dispersals, the chronology of the Dhofar Nubian industry remains poorly constrained. Two quartz optically stimulated luminescence (OSL) ages of ~106 and ~107 ka from the site of Aybut Auwal established an MIS 5 association\u003csup\u003e15\u003c/sup\u003e, but without additional dating, the duration, fate, and broader context of this industry could not be evaluated. Here we present quartz OSL ages from three new stratified Dhofar Nubian sites—TH505 (Aybut Hills 5), TH584 (Aybut Hills 6), and TH571 (Upper Amut 1)—providing the first comprehensive chronology for this industry in southern Arabia.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003eSites and lithic assemblages\u003c/h2\u003e \u003cp\u003eThe studied sites are located on the Nejd plateau in the Dhofar Governorate of southern Oman, an arid limestone tableland incised by deep wadis that expose chert-rich limestone beds. At TH505 and TH584, artefact-bearing deposits consist of colluvial slope debris (angular limestone and chert clasts in a sandy-silt matrix) with evidence of post-depositional gypsum induration. In contrast, TH571 is situated in an endorheic basin and contains artefacts within a fluvial gravel layer overlain by compacted aeolian loess, indicating accumulation during an earlier humid phase prior to loess deposition.\u003c/p\u003e \u003cp\u003eThe assemblages comprise 655 lithic artefacts from stratified and surface contexts (Fig. S8). All artefacts were manufactured on locally outcropping chert. Nubian Levallois are the most frequent core type at each site (Fig. S9; Tables S6-S8). Alternative Levallois strategies (centripetal, convergent) are rare or absent\u0026mdash;a pattern characteristic of Nubian assemblages in both Dhofar and the Negev desert\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan additionalcitationids=\"CR21\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. The low percentage of retouched tools (4%; n\u0026thinsp;=\u0026thinsp;28), proximity to raw material outcrops, and predominance of cores confirm the workshop character of these sites. Stratified contexts at all three localities contain diagnostic Nubian Levallois cores and products (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), demonstrating that the dated sediments directly bracket Dhofar Nubian occupation rather than later reoccupation events.\u003c/p\u003e \u003cp\u003eAnalysis of post-depositional surface modifications on the assemblage from TH584 shows that bleaching intensity increased significantly with stratigraphic depth (Cochran-Armitage trend test, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Surface artefacts exhibit minimal bleaching while buried contexts demonstrate moderate to heavy bleaching (Fig. S10; Table S11). This pattern confirms stratigraphic integrity and indicates that buried artefacts were rapidly interred following discard. Moreover, buried specimens retain sharp, undamaged edges consistent with primary deposition (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), in contrast to the high degree of edge damage characteristic of Dhofar Nubian surface assemblages\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eOSL chronology\u003c/h2\u003e \u003cp\u003eDeposition of artefact units at the three sites is constrained by a total of eight OSL samples, yielding ages that are tightly clustered between 107 and 95 ka (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The stratified deposits of Aybut Hills 5 (TH505) and Aybut Hills 6 (TH584) have weighted mean OSL ages of 106.4\u0026thinsp;\u0026plusmn;\u0026thinsp;4.5 ka (n\u0026thinsp;=\u0026thinsp;2) and 99.6\u0026thinsp;\u0026plusmn;\u0026thinsp;2.5 ka (n\u0026thinsp;=\u0026thinsp;5), respectively, while Unit B of Upper Amut 1 (TH571) is constrained to 104.5\u0026thinsp;\u0026plusmn;\u0026thinsp;7.1 ka (n\u0026thinsp;=\u0026thinsp;1). The single-grain OSL equivalent dose (D\u003csub\u003ee\u003c/sub\u003e) datasets are consistent with those reported for well-bleached and unmixed sediments (e.g., ref. 24; overdispersion values\u0026thinsp;=\u0026thinsp;25\u0026ndash;32%; Fig. S7), and the ages are statistically indistinguishable from the multiple-grain OSL ages of 106\u0026thinsp;\u0026plusmn;\u0026thinsp;9 ka and 107\u0026thinsp;\u0026plusmn;\u0026thinsp;9 ka reported for the nearby site of Aybut Auwal\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cb\u003eOSL analytical data from excavated Dhofar Nubian sites\u003c/b\u003e. All samples were measured using single-grain quartz (Qz) optically stimulated luminescence (OSL). Ages calculated using the central age model (CAM). Grain sizes in microns (\u0026micro;m); dose rates and equivalent doses (D\u003csub\u003ee\u003c/sub\u003e) in Grays (Gy); ages in thousands of years (ka) before CE 2023. Uncertainties are 1σ. Further details of D\u003csub\u003ee\u003c/sub\u003e and dose rate datasets are provided in Tables S2-S4.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"11\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSite\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLab ID\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eField ID\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDepth (cm)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGrain size (\u0026micro;m)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDose rate (Gy/ka)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026plusmn;\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eDe (Gy)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003e\u0026plusmn;\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003eAge (ka)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c11\"\u003e \u003cp\u003e\u0026plusmn;\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTH505\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOMANRG-4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAYB3-1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e90\u0026ndash;180\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e71.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e1.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e106.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e6.3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOMANRG-5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAYB3-2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e90\u0026ndash;180\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.91\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e96.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e106.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e6.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTH584\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOMANRG-9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAYB2-1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e90\u0026ndash;180\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.91\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e87.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e95.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e5.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOMANRG-1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAYB2-3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e90\u0026ndash;300\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e72.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e2.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e100.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e6.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOMANRG-3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAYB2-5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e90\u0026ndash;125\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e81.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e1.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e99.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e5.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOMANRG-10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAYB2-2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e90\u0026ndash;180\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e67.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e1.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e101.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e6.0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOMANRG-2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAYB2-4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e90\u0026ndash;125\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e72.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e1.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e100.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e5.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTH571\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOMANRG-6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAMU1-1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e90\u0026ndash;300\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e73.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e3.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e104.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e \u003cp\u003e7.1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eOxCal Bayesian modelling of the pooled OSL dataset (n\u0026thinsp;=\u0026thinsp;10) produces a combined age range of 109.2\u0026ndash;95.1 ka (95.4% credible interval) and a mean age of 102.1\u0026thinsp;\u0026plusmn;\u0026thinsp;3.4 (1σ) for the Dhofar Nubian industry (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The modelled age constrains this industry to a\u0026thinsp;~\u0026thinsp;14,000-year window corresponding closely to MIS 5c (105\u0026thinsp;\u0026minus;\u0026thinsp;93 ka).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003ePalaeovegetation\u003c/h3\u003e\n\u003cp\u003ePhytolith analysis of sediment samples from Aybut Auwal and Aybut Hills 6 provides direct evidence of local vegetation coinciding with Dhofar Nubian occupation (Fig. S11; Table S12). Grass (Poaceae) phytoliths dominate the assemblage, with limited palm and woody indicators. The phytolith assemblage indicates that the landscape was open and dominated by grasses, reflecting savanna or grassland conditions rather than the hyperarid desert seen today. The limited presence of palm and woody phytoliths suggests that the trees were likely restricted to watercourses and springs or locally favourable microhabitats during intervals of increased moisture. These results support regional palaeoclimate reconstructions that indicate enhanced precipitation and more hospitable environments in southern Arabia during MIS 5c\u003csup\u003e25\u0026ndash;30\u003c/sup\u003e, providing a plausible context for human occupation at these sites.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eOur OSL chronology establishes that the Dhofar Nubian industry was restricted to a narrow temporal window corresponding to MIS 5c, between approximately 109 and 95 ka (Bayesian modelled 95.4% credible interval). This finding has several implications for understanding the nature and fate of this population.\u003c/p\u003e \u003cp\u003eFirst, the Dhofar Nubian industry is contemporaneous with Nubian assemblages in the eastern Mediterranean (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). At Dimona South and Ein Ziq in the Negev desert, stratified deposits with Nubian Levallois technology have been dated between 116 and 82 ka\u003csup\u003e12,22\u003c/sup\u003e. The connection between these geographically distant occurrences extends beyond chronological overlap. Refitting analyses at Dimona South and in Dhofar document the same operational schema for Nubian Levallois core reduction\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e, and the same conspicuous absence of other Levallois variants. This pattern supports the interpretation that Nubian technology spread rapidly across Southwest Asia during MIS 5, whether through population movement, cultural transmission, or some combination of both. Second, the Dhofar ages align chronologically with the later phases of Nubian technology in Northeast Africa. In the Nile Valley, the Nubian Complex spans a considerable time depth, with early manifestations at Sai Island dated to sometime after mid-MIS 6 (\u0026lt;\u0026thinsp;150 ka)\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e and occupations at Sodmein Cave\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e and Taramsa\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e throughout MIS 5. Consequently, the Dhofar Nubian industry is thought to represent the cultural footprint of this technological tradition, appearing in southern Arabia at a later stage of the Nubian Complex's temporal range. The chronological coherence of assemblages with Nubian technology in Northeast Africa, the eastern Mediterranean, and southern Arabia renders convergent technological invention an increasingly untenable alternative. Independent populations arriving at the same highly specific reduction strategy within a narrow temporal window across three adjacent regions is less parsimonious than population dispersal under favourable climatic conditions.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eDuring the MIS 5c humid interval, the Nejd plateau was a habitable savanna with reliable groundwater. Speleothem records from southern Arabian caves document active growth, indicating sufficient moisture to sustain cave dripwater systems\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. Palaeolake deposits in the interior of Oman record high water stands around 100 ka\u003csup\u003e27\u003c/sup\u003e, and marine sediment cores from the Gulf of Aden show reduced aeolian input consistent with increased vegetation cover\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. Our phytolith data provide site-level confirmation of these regional signals, demonstrating that grass-dominated landscapes existed in the immediate vicinity of Dhofar Nubian sites during occupation. The absence of archaeological material dated to earlier MIS 5e is notable given that this interval represents peak interglacial humidity across Arabia. Despite favourable conditions documented in marine and terrestrial archives, no MIS 5e sites have been identified in southern Arabia, or elsewhere on the Peninsula apart from Jebel Faya in the Hajar mountain foothills\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. This pattern suggests that environmental conditions alone did not determine the timing of expansion; demographic growth in source regions like the Nile Valley may have been a necessary precondition. Sea level also presented a barrier. During MIS 5e, the Bab al Mandab strait remained approximately 30 km wide, however, when sea level dropped below \u0026minus;\u0026thinsp;50 m during MIS 5d (~\u0026thinsp;115\u0026ndash;105 ka), the crossing narrowed to less than 5 km via the Hanish Islands sill, with intervisibility between shorelines\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. The Dhofar Nubian industry may thus represent an expansion via the southern dispersal route\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e, which crossed the Red Sea during the MIS 5d lowstand and spread across the interior as MIS 5c humid conditions transformed southern Arabia to grasslands.\u003c/p\u003e \u003cp\u003eThe Bayesian modelled end boundary for our OSL dataset (95.1 ka; 95.4% CI) places the terminal occupation of the Dhofar Nubian around the MIS 5c/5b boundary (~\u0026thinsp;93 ka). Multiple lines of evidence suggest that this timing coincides with a pronounced aridification event across southern Arabia. The KL15 marine core record from the Gulf of Aden documents a sharp increase in aeolian dust flux at ~\u0026thinsp;95 ka, signalling desertification\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. Speleothem growth ceased in caves that had been active during MIS 5c\u003csup\u003e25\u003c/sup\u003e, and palaeolake systems contracted or disappeared as effective recharge ceased\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Essential for populations to survive the prolonged dry season, groundwater-fed springs once sustained by aquifer recharge were reduced to desiccated basins. The geographic isolation of Dhofar compounds the demographic vulnerability imposed by climate and hydrology. The Nejd plateau is bounded to the north by the hyperarid Rub' al Khali desert, which likely presented a formidable barrier, even during humid intervals, due to the sparse distribution of resources. To the east, systematic surveys in south-central Oman have recorded rare examples of Nubian sites despite extensive fieldwork\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. The 300-km Jiddat al Harassis plain\u0026mdash;devoid of drainages or springs\u0026mdash;forms a significant biogeographic obstacle between Dhofar and regions to the northeast, with freshwater availability found to be the most likely limiting factor in the distribution of Nubian technology across raw material-rich landscapes of southern Arabia\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. Without continued demographic influx from Africa, remnant populations in southern Arabia became susceptible to stochastic local extinction\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. Based on the current dating evidence, we interpret its disappearance after ~\u0026thinsp;95 ka as coinciding with environmental collapse and demographic fragmentation.\u003c/p\u003e \u003cp\u003eThe fate of the Dhofar Nubian population bears directly on whether early expansions contributed to the ancestry of later Eurasians. The subsequent Mudayyan industry in Dhofar shows no technological continuity with its predecessor. The Dhofar Nubian is characterized by large cores and preferential flaking; the Mudayyan, in contrast, exhibits systematic miniaturization, volumetric core configurations, and recurrent bidirectional flaking\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. Our excavated assemblages fall unambiguously within Dhofar Nubian parameters (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e), and despite surveys documenting over 250 Middle Palaeolithic sites across southern Oman\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e, no assemblages show intermediate characteristics between these two industries. Taphonomic evidence supports temporal separation: Dhofar Nubian assemblages exhibit heavier patination and more advanced chemical dissolution than Mudayyan assemblages on the same landscapes\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. The Dhofar Nubian appears, flourishes, and disappears as a cohesive package. This abrupt termination followed by a technologically disjunctive industry indicates demographic discontinuity, not in situ evolution.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe pattern of ephemeral occupation followed by population collapse aligns with findings from northern Arabia, where multiple hominin dispersals over the past 400,000 years each produced distinct material culture, with no evidence for long-term continuity\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. This does not mean the MIS 5 dispersal left no genetic trace. Recent analyses identified at least two episodes of gene flow from modern humans into Neanderthals, the more recent around 120\u0026thinsp;\u0026minus;\u0026thinsp;100 ka\u003csup\u003e6,7\u003c/sup\u003e. The Dhofar Nubian population, far from the known Neanderthal range, would not have contributed directly to this introgression. However, the existence of contemporaneous Nubian assemblages in the eastern Mediterranean\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e falls within the zone of overlap, raising the possibility that a related population was responsible. Under this scenario, the expansion signalled by Nubian technology split into two branches: one entering the eastern Mediterranean, encountering Neanderthals, and contributing genes before disappearing; the other crossing the Red Sea into southern Arabia, persisting during MIS 5c, and succumbing to environmental collapse. Neither contributed ancestry to living humans. These findings add to a growing body of genomic evidence that the successful colonisation of Eurasia occurred after 80 ka\u003csup\u003e1,2\u003c/sup\u003e. Earlier dispersals were demographically fragile and did not establish an enduring foothold outside Africa. Isolating the factors that allowed later populations to overcome the constraints of these earlier pulses\u0026mdash;be they environmental, demographic, or cultural\u0026mdash;is the next challenge in reconstructing our global prehistory.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eExcavation procedures\u003c/h2\u003e \u003cp\u003eSites were selected based on evidence of buried archaeological deposits, identified by bleached Nubian Levallois artefacts eroding from sedimentary contexts (TH505, TH584) or proximity to sedimentary basins with preservation potential (TH571). All excavations followed a 1\u0026times;1 m grid system and proceeded by geologic unit. Sediments were excavated by trowel and dry-screened through 5 mm mesh. At TH584, artefacts were piece-plotted using a total station during the 2023 season; at other sites, artefacts were collected by square and geologic unit.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eSingle-grain OSL dating\u003c/h2\u003e \u003cp\u003eEight sediment samples from the three excavated sites were dated using optically stimulated luminescence (OSL) of quartz grains. The OSL dating samples were taken from cleaned sedimentary exposures using metal tubes, with additional bulk sediment collected from the surrounding few cm of each tube for beta dose rate determination and water content analysis. Purified coarse grain quartz extracts (90\u0026ndash;125, 90\u0026ndash;180 or 90\u0026ndash;300 \u0026micro;m) were processed under safe light conditions (630 nm LEDs, \u0026lt;\u0026thinsp;0.15 \u0026micro;W/cm\u003csup\u003e2\u003c/sup\u003e power density at sample position) at Adelaide University using standard preparation procedures (e.g., ref. 43). Single-grain OSL measurements were made using the experimental apparatus, single-aliquot regenerative-dose (SAR) procedures, and quality assurance criteria published previously by Arnold et al.\u003csup\u003e44,45\u003c/sup\u003e, which are further detailed in the Supplementary Information. Between 800 and 2000 single-grain equivalent dose (D\u003csub\u003ee\u003c/sub\u003e) measurements were made for each sample using the OSL SAR procedure shown in Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e, which yielded suitable dose-recovery test results for sample OMANRG-10 (Fig. S6).\u003c/p\u003e \u003cp\u003eEnvironmental dose rates were evaluated using instrumental neutron activation analysis (INAA) of bulk sediment collected from the OSL sample position. U, Th and K concentrations were measured by INAA at the Nuclear Physics Institute of the Czech Academy of Sciences (Table S4). The dose rate calculations additionally take into account cosmic ray contributions, an assumed minor internal alpha dose rate, beta-dose attenuation and long-term sediment water content. Detailed OSL dating methods and results are available in Supplementary Information Section 2, including discussions of the D\u003csub\u003ee\u003c/sub\u003e distributions and statistical age models used to derive representative burial dose estimates for each sample. In general, the single-grain OSL D\u003csub\u003ee\u003c/sub\u003e distributions exhibit limited scatter and are characteristic of well-bleached, unmixed samples (e.g., refs. 44, 46\u0026ndash;48), hence we have used the weighted mean D\u003csub\u003ee\u003c/sub\u003e values (calculated using the central age model; CAM) to derive final burial dose estimates (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eBayesian age modelling\u003c/h3\u003e\n\u003cp\u003eOxCal Bayesian modelling\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e has been used to calculate a combined age range for stratified Dhofar Nubian industry units preserved at the excavated sites. For this purpose, we have pooled all OSL ages (chronometric likelihoods) in direct association with Nubian deposits from the present study and Rose et al.\u003csup\u003e15\u003c/sup\u003e as a single, unordered regional \u003cem\u003ePhase\u003c/em\u003e model with delineating start and end boundaries. The combined age range for the Nubian industry of Dhofar has been calculated from the modelled posterior probabilities of the start and end boundaries of this regional \u003cem\u003ePhase\u003c/em\u003e model using the OxCal \u003cem\u003edate\u003c/em\u003e function. The CQL code used to construct the Dhofar Nubian \u003cem\u003ephase\u003c/em\u003e model is provided in Supplementary Information Section 2. The ten OSL dating likelihoods from sites TH505, TH584, TH571 and TH59 have been input into the model as calendar ages before CE 2023 (the year of sample collection for the dating samples from the current study), together with their associated 1σ uncertainty ranges. As the two OSL samples published by Rose et al.\u003csup\u003e15\u003c/sup\u003e were collected 13 years earlier than the OSL samples from the present study, these likelihoods have been adjusted to the CE 2023 common baseline prior to running the model by adding 13 years to the original age estimates. The Bayesian model was run using the general \u003cem\u003eoutlier\u003c/em\u003e function\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e with prior outlier probabilities of 5% assigned to all dating samples. Likelihood estimates that yielded posterior outlier probabilities\u0026thinsp;\u0026gt;\u0026thinsp;5% were not excluded from the final model but were proportionally down-weighted in the iterative Markov Chain Monte Carlo runs\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe Bayesian modelling results for the Dhofar Nubian occupation are summarised in Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and Table S5. The model exhibits a median convergence integral of 99.4% for all individual posterior distributions, and none of the ten likelihoods included in the model are identified as major statistical outliers. All likelihoods exhibit posterior outlier probabilities equal to, or less than, the specified prior threshold of 5% (Table S5).\u003c/p\u003e\n\u003ch3\u003eLithic analysis\u003c/h3\u003e\n\u003cp\u003eLithic analysis focused on cores, the most technologically diagnostic component of these workshop assemblages. All cores and retouched tools were subject to full attribute analysis; debitage was sorted by category and counted. Nubian core attributes were classified following the technological criteria described by Usik et al.\u003csup\u003e31\u003c/sup\u003e, which established morphological parameters for distinguishing Nubian from other Levallois technologies. Metric attributes recorded for each core included maximum length, maximum width, midpoint thickness, platform dimensions, and median distal ridge angle on Nubian cores. Geometric mean (∛[length \u0026times; width \u0026times; thickness]) was calculated as a size-independent measure of core volume. Taphonomic condition of surface artefacts was recorded following the rubric described in Rose et al.\u003csup\u003e23\u003c/sup\u003e.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003ePhytolith analysis\u003c/h2\u003e \u003cp\u003eSix sediment samples were processed from TH59 (n\u0026thinsp;=\u0026thinsp;3) and TH584 (n\u0026thinsp;=\u0026thinsp;3) at the La Trobe University Palaeosciences Lab. Fifteen grams of each sample were dissolved in 7% HCl to remove carbonates; the absence of organics precluded treatment with nitric acid or hydrogen peroxide. Samples were deflocculated in 5% Calgon, followed by density separation using sodium polytungstate (2.5 g/cm\u0026sup3;). Ten microlitres of each processed sample were mounted on slides and examined at \u0026times;400 magnification using a Zeiss light microscope. Phytoliths were classified by morphotype and measured using a calibrated eyepiece graticule.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eData availability\u003c/h2\u003e\n\u003cp\u003eAll R code with raw lithic data, statistical analyses, and scripts for figure generation have been deposited in an OSF research compendium and can be accessed via https://osf.io/jt7ke/. Lithic, OSL, and phytolith data are available in Supplementary Information.\u003c/p\u003e\n\u003ch2\u003eReporting summary\u003c/h2\u003e\n\u003cp\u003eFurther information on research design is available in the Nature Research Reporting Summary linked to this paper.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eAcknowledgements\u003c/h2\u003e\n\u003cp\u003eWe wish to thank the Ministry of Heritage and Tourism in Oman for permissions and logistical support. We are grateful to Rhianna Power (Adelaide University) for assistance with preparing and measuring the OSL dating samples. We thank Alan Cooper for his intellectual input regarding the conceptual framework of this study.\u003c/p\u003e\n\u003ch2\u003eFunding declarations\u003c/h2\u003e\n\u003cp\u003eThis work was funded by a Czech Academy of Sciences Praemium Academiae awarded to V.C and Australian Research Council Discovery project DP230100152 awarded to M.M.-W., L.A. and R.G. Fieldwork was also funded in part by a private donation from A.E.M. J.I.R. is supported by the European Research Council under the European Union\u0026rsquo;s Horizon 2020 research and innovation programme awarded to N.B. (Advanced Grant, project DISPERSALS 101052761). INAA was supported by the long-term conceptual development project RVO 61389005, neutron irradiation was performed under CICRR infrastructure supported by MEYS project LM2023041.\u003c/p\u003e\n\u003ch2\u003eAuthor contributions\u003c/h2\u003e\n\u003cp\u003eJ.I.R. and A.E.M. conceived and designed the study. J.I.R., R.G., A.B., D.C., R.D., E.S.H, A.A.Mahri, O.S., V.I.U., and A.E.M. conducted field surveys and excavations. D.C. and R.D. documented stratigraphic profiles and provided geological and sedimentological assessments. L.A. performed optically stimulated luminescence dating and Bayesian age modelling. J.K. conducted neutron activation analysis. M.M.-W. conducted phytolith analysis. J.I.R., O.S., and R.G. created GIS basemaps. V.I.U., A.E.M., O.S., E.S.H, and J.I.R. analysed lithic assemblages. J.I.R. performed statistical analyses, wrote code, drafted figures, and created online repository. J.I.R. wrote the manuscript with contributions from R.G., L.A., M.M.-W., O.S., N.B., R.D., and V.C. R.G., V.C., N.B., M.M.-W., and A.E.M. provided funding sources. R.G. organized fieldwork permits. A.A.Mahrooqi, A.A.Mahri, and R.G. provided administrative and logistical support. All authors reviewed and approved the final manuscript.\u003c/p\u003e\n\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorrespondence and requests for materials\u003c/strong\u003e should be addressed to J. I. Rose or R. Garba.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBergstr\u0026ouml;m A, Stringer C, Hajdinjak M, Scerri EML, Skoglund P (2021) Origins of modern human ancestry. Nature 590:229\u0026ndash;237\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMallick S et al (2016) The Simons Genome Diversity Project: 300 genomes from 142 diverse populations. Nature 538:201\u0026ndash;206\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGr\u0026uuml;n R et al (2005) U-series and ESR analyses of bones and teeth relating to the human burials from Skhul. J Hum Evol 49:316\u0026ndash;334\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eValladas H et al (1988) Thermoluminescence dating of Mousterian 'proto-Cro-Magnon' remains from Israel and the origin of modern man. Nature 331:614\u0026ndash;616\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHershkovitz I et al (2018) The earliest modern humans outside Africa. Science 359:456\u0026ndash;459\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuhlwilm M et al (2016) Ancient gene flow from early modern humans into Eastern Neanderthals. Nature 530:429\u0026ndash;433\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi L, Comi TJ, Bierman RF, Akey JM (2024) Recurrent gene flow between Neanderthals and modern humans over the past 200,000 years. Science 385:eadi1768\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBeyer RM, Krapp M, Eriksson A, Manica A (2021) Climatic windows for human migration out of Africa in the past 300,000 years. Nat Commun 12:4889\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarks AE (1968) The Mousterian industries of Nubia. In: Wendorf F (ed) The Prehistory of Nubia, vol 1. Southern Methodist University, pp 194\u0026ndash;314\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan Peer P, Vermeersch PM (2000) The Nubian Complex and the dispersal of modern humans in North Africa. Stud Afr Archaeol 7:47\u0026ndash;60\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHallinan E (2024) A synthesis and critical inventory of Nubian cores in Middle Stone Age and Middle Palaeolithic assemblages. J Open Archaeol Data 12:9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOron M et al (2024) Nubian Levallois technology during MIS 5: refitted lithic sequences and OSL ages of Dimona South, Israel. J Paleolithic Archaeol 7:4\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoder-Goldberger M, Gubenko N, Hovers E (2016) Diffusion with modifications: Nubian assemblages in the central Negev highlands of Israel. Quat Int 408:121\u0026ndash;139\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHilbert YH et al (2026) Nubian cores along the coast of the central Levantine corridor: Exploring variability of Levallois point production in Lebanon. J Paleolithic Archaeol 9:3\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRose JI et al (2011) The Nubian Complex of Dhofar, Oman: an African Middle Stone Age industry in southern Arabia. PLoS ONE 6:e28239\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHilbert YH, Crassard R, Charloux G, Loreto R (2016) Nubian technology in northern Arabia: Impact on interregional variability of Middle Paleolithic industries. Quat Int 408:68\u0026ndash;84\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCrassard R et al (2018) Middle Palaeolithic occupations in central Saudi Arabia during MIS 5 and MIS 7: new insights on the origins of the peopling of Arabia. Archaeol Anthropol Sci 11:3101\u0026ndash;3120\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGroucutt HS (2020) Culture and convergence: the curious case of the Nubian Complex. In: Groucutt HS (ed) Culture History and Convergent Evolution. Springer, pp 55\u0026ndash;86\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHallinan E et al (2022) The nature of Nubian: developing current global perspectives on Nubian Levallois technology and the Nubian complex. Evol Anthropol 31:227\u0026ndash;232\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRose JI et al (2023) Prehistoric settlement patterns in southern Oman. Pal\u0026eacute;orient 49:83\u0026ndash;108\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRose JI et al (2025) Lithic miniaturization and technological innovation on the Southern Dispersal Route out of Africa. J Paleolithic Archaeol 8:40\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOron M et al (2025) Nubian Levallois cores from MIS 5 alluvial terraces in the Negev Desert: new insights into the Middle Paleolithic in the arid regions of the southern Levant. J Paleolithic Archaeol 8:4\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRose JI et al (2025) Mapping lateral stratigraphy at Palaeolithic surface sites: a case study from Dhofar, Oman. J Archaeol Sci 173:106117\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArnold LJ, Roberts RG (2009) Stochastic modelling of multi-grain equivalent dose (De) distributions: Implications for OSL dating of sediment mixtures. Quat Geochronol 4:204\u0026ndash;230\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFleitmann D, Matter A (2009) The speleothem record of climate variability in Southern Arabia. C R Geosci 341:633\u0026ndash;642\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRosenberg TM et al (2011) Humid periods in southern Arabia: windows of opportunity for modern human dispersal. Geology 39:1115\u0026ndash;1118\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRosenberg TM et al (2012) Late Pleistocene palaeolake in the interior of Oman: a potential key area for the dispersal of anatomically modern humans out-of-Africa? J Quat Sci 27:13\u0026ndash;16\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMatter A, Neubert E, Preusser F, Rosenberg T, Al-Wagdani K (2015) Palaeo-environmental implications derived from lake and sabkha deposits of the southern Rub' al-Khali, Saudi Arabia and Oman. Quat Int 382:120\u0026ndash;131\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNicholson SL et al (2020) Pluvial periods in Southern Arabia over the last 1.1 million-years. Quat Sci Rev 229:106112\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFischer ML et al (2024) Spatio-temporal variations of climate along possible African\u0026ndash;Arabian routes of \u003cem\u003eH. sapiens\u003c/em\u003e expansion. Quat Sci Adv 14:100174\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUsik VI, Rose JI, Hilbert YH, Van Peer P, Marks AE (2013) Nubian Complex reduction strategies in Dhofar. Quat Int 300:244\u0026ndash;266\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan Peer P et al (2003) The early to Middle Stone Age transition and the emergence of modern human behaviour at site 8-B-11, Sai Island, Sudan. J Hum Evol 45:187\u0026ndash;193\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMercier N et al (1999) TL dates of burnt lithics from Sodmein Cave (Red Sea Mountains, Egypt). Quat Sci Rev 18:623\u0026ndash;630\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchmidt C et al (2015) Multi-emission luminescence dating of heated chert from the Middle Stone Age sequence at Sodmein Cave (Red Sea Mountains, Egypt). J Archaeol Sci 63:94\u0026ndash;103\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan Peer P, Vermeersch PM, Paulissen E (2010) Chert Quarrying, Lithic Technology and a Modern Human Burial at the Palaeolithic Site of Taramsa 1, Upper Egypt. Leuven University\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArmitage SJ et al (2011) The southern route 'out of Africa': evidence for an early expansion of modern humans into Arabia. Science 331:453\u0026ndash;456\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLambeck K et al (2011) Sea level and shoreline reconstructions for the Red Sea: isostatic and tectonic considerations and implications for hominin migration out of Africa. Quat Sci Rev 30:3542\u0026ndash;3574\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHill J et al (2022) Sea-level change, palaeotidal modelling and hominin dispersals: The case of the southern Red Sea. Quat Sci Rev 293:107719\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLahr MM, Foley R (1994) Multiple dispersals and modern human origins. Evol Anthropol 3:48\u0026ndash;60\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChlachula D et al (2025) Evidence of Middle Palaeolithic human occupation in south-central Oman. Antiquity 99(404):e9\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEren MI et al (2025) Examining the distribution of Middle Paleolithic Nubian cores relative to chert quality in southern (Nejd, Dhofar) and south-central (Duqm, Al Wusta) Oman. \u003cem\u003eGeoarchaeology\u003c/em\u003e 40(1), e22019\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGroucutt HS et al (2021) Multiple hominin dispersals into Southwest Asia over the past 400,000 years. Nature 597:376\u0026ndash;380\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArnold LJ et al (2024) Single-grain luminescence and combined U-series/ESR dating of the early Upper Palaeolithic Lagar Velho Rock Shelter, Leiria, Portugal. Quat Geochronol 83:101572\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArnold LJ et al (2016) OSL dating of individual quartz \u0026lsquo;supergrains\u0026rsquo; from the Ancient Middle Palaeolithic site of Cuesta de la Bajada, Spain. Quat Geochronol 36:78\u0026ndash;101\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArnold LJ et al (2022) Examining sediment infill dynamics at Naracoorte Cave megafauna sites using multiple luminescence dating signals. Quat Geochronol 70:101301\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBailey RM, Arnold LJ (2006) Statistical modelling of single grain quartz De distributions and an assessment of procedures for estimating burial dose. Quat Sci Rev 25:2475\u0026ndash;2502\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArnold LJ, Bailey RM, Tucker GE (2007) Statistical treatment of fluvial dose distributions from southern Colorado arroyo deposits. Quat Geochronol 2:162\u0026ndash;167\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArnold LJ et al (2008) Optical dating of perennially frozen deposits associated with preserved ancient plant and animal DNA in north-central Siberia. Quat Geochronol 3:114\u0026ndash;136\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBronk Ramsey C (2009) Bayesian analysis of radiocarbon dates. Radiocarbon 51:337\u0026ndash;360\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBronk Ramsey C (2009) Dealing with outliers and offsets in radiocarbon dating. Radiocarbon 51:1023\u0026ndash;1045\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8751643/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8751643/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Early modern humans expanded out of Africa multiple times before 70,000 years ago, yet none of these populations are ancestral to living non-Africans. What caused these dispersals to fail? Here we investigate one such expansion using new evidence from southern Arabia. We present robust chronologies from four stratified sites in Dhofar, Oman, associated with Nubian Levallois technology—a distinctive stone tool tradition of African origin. The ages constrain this industry to 109-95 thousand years ago, synchronous with Nubian occurrences in the eastern Mediterranean and coinciding with the estimated timing of early modern human gene flow into Neanderthals. Integrating our chronology with palaeoenvironmental records, we show that the southern Arabian population disappeared abruptly with the onset of aridification 95 thousand years ago, with no technological continuity into later industries. These results demonstrate that early human expansions beyond Africa were climate-dependent and demographically fragile, and that the global establishment of our species was preceded by repeated failures.","manuscriptTitle":"A failed human expansion out of Africa 100,000 years ago","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-17 09:00:08","doi":"10.21203/rs.3.rs-8751643/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"fac8b6c6-7f60-4241-be35-2061f85e81ef","owner":[],"postedDate":"February 17th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":62447085,"name":"Biological sciences/Evolution/Archaeology"},{"id":62447086,"name":"Earth and environmental sciences/Climate sciences/Palaeoclimate"},{"id":62447087,"name":"Biological sciences/Genetics/Genetic hybridization"}],"tags":[],"updatedAt":"2026-02-23T14:46:41+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-17 09:00:08","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8751643","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8751643","identity":"rs-8751643","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}