Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

The Northern Route for Human dispersal in Central and Northeast Asia: New evidence from the site of Tolbor-16, Mongolia



The fossil record suggests that at least two major human dispersals occurred across the Eurasian steppe during the Late Pleistocene. Neanderthals and Modern Humans moved eastward into Central Asia, a region intermittently occupied by the enigmatic Denisovans. Genetic data indicates that the Denisovans interbred with Neanderthals near the Altai Mountains (South Siberia) but where and when they met H. sapiens is yet to be determined. Here we present archaeological evidence that document the timing and environmental context of a third long-distance population movement in Central Asia, during a temperate climatic event around 45,000 years ago. The early occurrence of the Initial Upper Palaeolithic, a techno-complex whose sudden appearance coincides with the first occurrence of H. sapiens in the Eurasian steppes, establishes an essential archaeological link between the Siberian Altai and Northwestern China . Such connection between regions provides empirical ground to discuss contacts between local and exogenous populations in Central and Northeast Asia during the Late Pleistocene.


Although models for H. sapiens’ early dispersals out of Africa emphasize a southern route to Asia1,2,3,4,5, Neanderthal and Modern Human (MH) fossils in Siberia6,7,8,9 suggest that at least two other dispersals took place across the Eurasian steppe north of the Asian high mountains. Given the size of the area considered, human fossils are few but recent studies have suggested that a major change in the regional archaeological record could be indicative of a large-scale human dispersal event. Known as the Initial Upper Palaeolithic (IUP), it refers to the sudden appearance in contiguous regions of a specific blade technology sometimes associated with bone tools and ornaments10,11,12,13,14,15,16,17. How old these assemblages are, and how long the phenomenon lasts are still controversial questions, and little is known about the timing and environmental context of these population movements. Here we present new data following excavations of an archaeological site located in a low altitude pass, the Ikh-Tolborin-Gol, which connects Siberia with northern Mongolia. Our results document the early occurrence of the Upper Paleolithic in Mongolia and provide a chronological reference for population movement across Northeast Asia during the Late Pleistocene.

Site Description

The Tolbor-16 (N49 13.619 E102 55.383) (T16 hereafter) site is located in the Northern Hangai Mountains, along the western flank of the Tolbor River valley (1169 m asl), 13 km south of the confluence with the Selenga River18 (Fig. 1). From 2011 to 2016, we excavated five pits for archaeological and geoarchaeological investigations (SI Section 1). Pits 1 and 4 are the largest excavated areas, whereas Pits 2, 3 and 5 are smaller trenches that provide stratigraphic and chronological control. Six main archaeological horizons dating from the MIS3 to the Holocene have been documented at the site. The material relevant to the earliest human dispersals derives from the lowermost archeological horizon (AH6) in Pits 1 and 4.

Figure 1

Above (right): Schematic map of the main IUP sites in Central Europe (Bohunician) (A), in the Middle-East (Emirean) (B) and in Asia (D); and the Ust-Ishim human femur (C). Above (left): View on the T16 site and the Western flank of Tolbor Valley. Below: Detailed map of the potential IUP sites in the region. 1, Luotuoshi; 2, Ush-Bulak; 3, Malo Yaloman, Cave; 4, Kara-Bom, 5, Ust-Karakol-1 and Denisova Cave; 6, Kara-Tenesh; 7, Derbina sites; 8, Arembovski; 9, Makarovo-IV; 10, Khotyk; 11, Barun-Alan sites; 12, Kamenka and Varvarina Gora; 13, Tolbaga; 14, Podzvonkaya; 15, Egiin-Gol sites (Dörölj 1–2); 16, Tolbor-16 and Tolbor-4; 17, Tsatsyn Ereg; 18, Mojlt’yn-Am; 19, Chiken sites; 20, Tsagan-Agui; 21, Shuiddonggou 1; 22, Khanzat-1; 23, Khavsgayt (and Salkhit); 24, Rashaan Khad; 25, Otson Tsokhio16,17,18,75,76,77; Geo-atlas background map).


The T16 sediments consist of loess, reworked loess and soliflucted laminar silt, with gravel and cobbles. The sites sit at the mouth of a small canyon, which seems to have been a source for the sediment that accumulated through a complex interplay of colluviation, eolian sedimentation, and solifluction. Sedimentological data indicate a low-energy depositional environment (see SI Section 2). In most of the exposed sections at T16, the stratigraphy divides into three main units: the Holocene soil complex (unit 1) comprising one or more chernozem or kastanozem type soils (FAO classification system), an underlying layer of loess and reworked loess (unit 2), and a soliflucted diamict of laminar silt with gravel and cobbles (unit 3), which extends to a depth of 2.5 m. In Pit 4, at least five superposed solifluction lobes were identified within unit 3 (sub-units 3a, 3b, 3c, and 3d); archaeological material is restricted to the upper 2 m of the section and is present in units 1 and 2 and in solifluction lobes 3a, b and c.

Six archaeological horizons (AH) were identified within the three stratigraphic units at Pit 4 (Fig. 2). The soil complex contains a few artefacts, and these are designated AH1. AH2 occurs within primary loess at the top of unit 2, and AH3, in reworked sediments lower within unit 2, consists of intrusive material from the underlying unit (probably AH4-AH6). Slope change – resulting in the truncation of unit 3 and its partial incorporation by solifluction into unit 2 – is responsible for the mixed nature of AH3. AH4 and AH5 sit within lobes 3a and 3b, and AH6 - with which this paper is primarily concerned - lies at the bottom of the sequence in unit 3 c. A similar succession occurs in Pit 1 (c. 4 m to the northeast), although it also contains additional sedimentary units (SI Section 2) not preserved in pit 4.

Figure 2

From left to right: (a) North stratigraphic section at T16 Pit 4, showing the three main stratigraphic units and solifluction lobes 3b-e (3a was restricted to the westernmost part of the section and was not sampled); (b) Standard deviation of the fine (<2 mm) fraction by laser particle size analysis (top axis); (c) Organic matter content relative to mineral matter by loss-on-ignition (%, bottom axis); (d) Gravel content (wt%, top axis); (e) Calcium carbonate content relative to mineral matter by loss-on-ignition (%, hollow dots, top axis) within the sediments at pit 4. Well-sorted primary loess has a low standard deviation; redeposition and soil development affect this (see SI Section 2); the distance between (b,c) (gray area) is therefore a rough proxy for climate, with climatic amelioration indicated where the distance is greater. Gravel is present in the sampled sediments through gravitational input; the proportion of gravel increases during prolonged surface exposure or slow sediment accumulation. Carbonate content increases when evaporation is high relative to precipitation. Archaeological horizons (AH), modelled age of the boundaries between events (SI Section 4), and climatic interpretation using NGRIP as a comparison are also indicated.


We used three geochronological methods (polymineral post-IR IRSL, Quartz OSL and radiocarbon) to determine the antiquity of the site. For luminescence, we ran paired samples collected from Pits 1, 2 and 3 to allow an overarching chronological framework for sediment deposition (SI Section 3). Measurements obtained on both quartz and polymineral (feldspar-dominated) fine-grained aliquots provide independent age control; with one exception (EVA-1438), all paired ages fall within 2-sigma error. For radiocarbon, we collected bone samples from cultural layers AH2 through AH5 in Pits 1 and 4 (SI Section 4) to obtain radiocarbon dates on the animal species associated with the archaeological deposits. Due to the bone preservation, cut-marks, tool use or manufacture are not identifiable, but the spatial distribution of bones tightly coincides with the archaeological occupations. Deposits without stone tools contain few to no bones.

The results indicate that the T16 sequence dates to the Late Pleistocene, spanning Marine oxygen Isotope Stage (MIS) 3 and MIS2. AH2 is constrained to MIS2; AH3 and AH5 dates between to 35.1–38.5 ka. The material corresponding to the earliest human occupation is associated with AH6 and dated in Pit 1 and Pit 4 to 42.5–45.6 ka. Stratigraphically, AH3 and AH6 provide a temporal bracket for AH4, an assemblage located in a disturbed deposit that probably includes material of different ages. Although the archaeological sequence is concentrated in the two first meters of deposits, the age of the laminar sediment accumulation below AH6 is still to be determined.

Palaeoenvironmental conditions

The deposition of the 2.5 vertical meters of laminar silt in pit 4 (the first meter hosting AH 4, 5, and 6) is the result of episodic loess deposition, slopewash, carbonate crust formation, and pedogenesis. It records a different precipitation regime than today, with cyclic – both seasonal and millennial-scale – variation in moisture availability (see SI Section 2). Periodic climatic downturns during the deposition of unit 3 result in episodes of solifluction, confirming that northern Mongolia witnessed multiple fluctuations in moisture and temperature during MIS319,20. The high organic matter content of sediments associated with AH6 is consistent with a period of milder, relatively moist climate at the time of the occupation(s). A climatic deterioration follows the deposition of AH6; an episode of solifluction, which takes place together with a reduction in colluviation, the formation of a thick carbonate crust in Pit 1, and a possible episode of loess accumulation in Pit 4, is evidence for a period of cold and aridity. The overlying soliflucted slopewash deposits contain a lithic assemblage (AH5) of different character.

The deposition of the laminar sediment coincides with a period for which regional ecological (and other proxy) records indicate relatively moist conditions19,20,21,22,23,24,25; the dominance of colluviation over eolian deposition in much of the section supports moist conditions. At the same time, the cyclic development of thin carbonate crusts, normally an indicator of aridity, becomes important before and during the deposition of the IUP sediments (sub-units 3e, d, c); this may relate to strong seasonality of either insolation21 or precipitation patterns. Abundant root cell calcite throughout the studied section indicates the presence of grass, and, although reworked, strongly humic and calcareous crumbs within the redeposited material are consistent with steppe vegetation; episodic decreases in the mean particle size of the silt size fraction within the reworked sediments (SI Section 2) may relate to denser vegetation cover at certain times during the hypothesized warm events recorded in sub-units 3e, d, and c and a concomitant reduction in the energy of reworking. The transition to eolian deposition is difficult to date owing to truncation, but 14C dates from Pits 1 and 4 suggest a shift before the MIS3-2 transition; the timing of this change, while coinciding with global cooling and aridification, links the section to climate records from the south and east (eg Salawusu26, Guliya ice core27, Hulu cave28) where pronounced cooling is recorded from an earlier date than in the North Atlantic, and to regional climate records, which suggest a change to arid conditions at this time19,20,23,24,25.

In Pit 1 and Pit 4, bones occur in low concentrations within all AHs. They are visibly more numerous at the top of the laminated sediments of unit 3, within AH5 and AH6. By contrast, we found no bones associated with the archaeologically sterile sediments below (SI Section 4). The degree of fragmentation combined with surface alteration is a significant obstacle to the identification of taxa or anthropogenic modifications. Zooarchaeology by Mass Spectrometry (ZooMS)29,30 screening is in progress to identify species on bones from Pit 4 (SI Section 6) while preliminary results agree with previous zooarchaeological identifications18. They indicate the occurrence of Bos sp., Caprinae sp. and Equus sp., Felid sp. and Elephantidea sp. Bos sp. is associated with human occupation in solifluction lobes 3a trough 3c for which three different age boundaries are estimated using radiocarbon dates and Bayesian modelling: 35.1–38.5 ka; 36.6–42.5 ka and 42.5–45.6 (dating the AHs described above) (SI Section 4). Most of these taxa are common in an Pleistocene open steppe or a taiga environment but they also occur northward, in sites where tundra environments predominate (e.g. Transbaikal)31. According to a pollen record from the region, steppe and taiga environments alternate during the cold and warm climate fluctuations in the MIS325.

Lithic assemblage from AH6

The oldest archaeological assemblage, AH6, occurs within stratigraphic Unit 3c in Pit1 and in Pit4. The assemblage described here is a sample selected for its stratigraphic integrity and its clear association with the chronological data (SI Section 5). It contains 826 lithic artifacts with 362 piece-plotted artifacts (>20 mm) such as 19 cores and preforms, and 287 blanks, of which 91 are retouched tools. It also includes 463 smaller fragments (<20 mm). The artifacts are made of cryptocrystalline materials of medium to fine grain. The origin of the raw material is mostly local and it outcrops throughout the valley in sub-vertical bands.

The assemblage mainly consists in blades and artifacts linked with their production. Two specific patterns of core reduction dominate the sample. A method known elsewhere as asymmetrical (or oblique) core reduction14 leads to the production of large (and sometimes massive) blades, detached from two opposed platforms. The core reduction is initiated by the production of a crest while the frequent maintenance is documented by the removal of neo-crest and side (debordant) blades (Fig. 3A). The striking platform is regularly reshaped by the removal of core tablets. While some of the large blades are retouched into tools, the thickest ones are turned into cores using the burin-core method, to produce small blades/bladelets32 (Fig. 3B) (SI Section 5). The combination of the asymmetrical and the burin-core methods into an intricate system provides the basis for a solid technological definition of the IUP14,17, and specifically the Asian variant thereof.

Figure 3

Reduction sequence model for the blade production in AH6. Asymmetrical reduction (A) produces large blades (Ab) (SI5) used as tools and thick technical blades (Ac); some of the thickest blades are turned into cores to produce small blades/bladelets using the burin-core (B1, B2; B3) or the truncated-facetted methods (B3).

As it is often the case with IUP, some features in the assemblage are reminiscent of the Middle Palaeolithic (MP) (SI Section 5). For instance, a few artifacts resemble Levallois cores but a close look at their size and geometry indicates that they are blade cores pushed to exhaustion. In the sample studied, the MP features seem to reflect the site function (expedient tools, exhausted cores), the technique (mineral hammer) or the method (e.g. debordant blade) in use during the blade production. Overall, all the specific features of the assemblage are perfectly in line with definition of the IUP in Asia14,16.


The definition of the IUP in Central and Northeast Asia is mostly based on the detailed study of sites from the Gorny Altai (e.g. Kara-Bom)10,11,12,13,14,15,16,17 dated as early as 45–47 ka32,33,34. Eastward, where the Selenga River reaches Lake Baikal, a group of sites shows the same lithic technology (SI Section 5). At Kamenka, the IUP is potentially old but with problematic dating inconsistencies ranging 45.8 ka to 28.1 ka35,36. There are other sites between the Altai and Lake Baikal that probably belong to this technocomplex but for which confirmation is needed (e.g. Makarovo-V, dated to >39 ka uncal BP)37. In the same drainage system, T16 (along with Tolbor-4) yielded strong evidence for the occurrence of the IUP about 450 km upstream from Lake Baikal. With dates ranging between >41.5 to 31.5 ka uncal BP, the chronology is still uncertain38,39,40. The results presented here confirm that the IUP reaches Northeast Asia by 45 ka. We also illustrate geological processes (e.g. solifluction) that, if not identified, could generate artificial noise in the chronological signal (SI Sections 2 and 5).

The AH6 assemblage from T16 confirms that the IUP occurs in the Selenga River basin at roughly the same time it appears in the Altai. With such striking similarities between coeval assemblages, the most parsimonious way to explain such an archaeological consistency between contiguous regions is to consider the Asian IUP is a relatively united phenomenon – or techno-complex14,41 – that dispersed across Siberia to reach Mongolia around 45 ka10,11,12,13,14,15,16,17,42, moving south and east to reach Northwestern China43,44,45 perhaps as far as the Tibetan Plateau46.

The ages associated with the IUP at T16 and the modelled age boundaries for Unit 3c correspond to a short-lived period of milder climatic conditions ca. 45 ka (Fig. 4). This temperate episode is known in the northern hemisphere marine record as Greenland Interstadial 12 (GI12)47. Its environmental impact is well-documented in the terrestrial record of the Eurasian steppe48 and following a climate crisis during which Neanderthal populations reduced significantly, GI12 is thought to have contributed to the repopulation of Europe49. It also coincides with the appearance of the Emireo-Bohunician50,51, a techno-complex known as the Western variant of IUP17 that would correspond to human expansions from the Levant into Central Europe13,15. Although the impact of GI12 on the landscape is more variable eastward, it may have had profound demographic impact in the region. During this time period, hunter-gatherers ventured into challenging environments such as the tundra of Western Siberia8, one of the world’s largest unbroken lowlands (below 100 m asl), or up to 72° latitude North, well within the Arctic Circle52.

Figure 4

Bayesian model of T16 radiocarbon dates and the mean 14C date of Ust-Ishim H. sapiens femur. The IUP assemblage is associated with Unit 3c, and later UP assemblages are associated with Unit 3b and Unit 2/3a. The calibrated date of the earliest H. sapiens fossil in Northeast Asia, Ust-Ishim (Fig. 1) is in red. Dates are calibrated using IntCal1378; the model and boundaries were calculated using OxCal 4.379 including a General t-type Outlier Model80. The results are aligned with the NGRIP δ18O climate record.

Based on our chronological data, the particularly harsh climatic event which appears to conclude the IUP in the valley – evidenced by an episode of solifluction, with a reduction in overall sedimentation rates, patchy loess accumulation, and the formation of a thick calcium carbonate crust in Pit 1 (SI Section 2) – is tentatively correlated with Heinrich Event 4 (HE4). Originally identified in the Atlantic Ocean53, the HE4 event (ca. 40 ka) is well documented in Asia28,54. In fragile eco-systems such as that at T16, where the long-term accumulation of sheet erosion deposits suggests sparser vegetation cover outside of climatic optima, millennial-scale cold events such as HE4 leading to changes in annual temperature or precipitation may accelerate catastrophic changes in vegetation and carrying capacity55. The position of the Tolbor sites at the transition zone between westerlies-dominated Siberia and monsoon-dominated southern- and eastern Inner Asia (and within the present-day core of the Siberian high which drives the Asian winter monsoon) makes them vulnerable to the complex interplay of several linked atmospheric systems in the past. Hence, the sedimentary record at T16 can be connected to both North Atlantic and East Asian climate records. The IUP here coincides with a period of warmer, moister, and probably more strongly seasonal conditions, which possibly changed the carrying capacity of the landscape and enabled human expansion through the area. The subsequent cold arid event appears to see off the IUP populations, suggesting that both global and regionally specific fluctuations in temperature and moisture collectively drove presence and absence here.

In the absence of a direct association between archaeological assemblages and identifiable human remains, at least two main hypotheses must be formulated. The first posits that the makers of the IUP in Asia are early H. sapiens. The dates from T16 are identical to those obtained from the human femur from Ust-Ishim, currently the oldest known representative of our species in the Eurasian steppe. The latter indicate that a dispersal event took place during the GI12 (Fig. 4)8. AH6 is significantly older than the age estimates of 35 ka for the Salkhit skull from in east Mongolia56 (Fig. 1). We note that the latter is equivalent to the lower chronological boundary of AH3. Also, we find more support for this hypothesis in the archaeological record of the Selenga River Basin. Where preservation allows for it, the earliest examples of beads and bone tools known in the region are associated with the IUP16,57. These objects have been almost exclusively associated with our species while they keep being produced in later phases of the Upper Paleolithic and beyond18,57,58,59.

The alternative hypothesis is that the IUP makers were local populations other than H. sapiens. Archaeological, fossil and genetic data suggest that Neanderthals crossed Central Asia several times6,7,9, potentially following the same inland route as H. sapiens up to the Altai. To date, there are no identifiable Neanderthal remains east of the Northwest Altai and there are simply no other human taxa securely dated to the GI12 in Siberia, Mongolia or Northwestern China. A potential chronological overlap is not excluded6, but Neanderthal remains found in cave sites such as Okladnikov or Chagirskaya are associated with Mousterian assemblages distinct from the IUP60. Despite recent efforts, the archaeological associations between fossil taxa, dates and archaeological components of layer 11 in Denisova Cave are yet to be firmly established7,61,62. The timing, distribution and the archaeology of the hominins known as the Denisovans is even less clear. Genetic diversity between specimens from the eponymous site suggests a long but probably discontinuous occupation of the Altai63, including encounters with Neanderthals and episodes of gene flow between with the two populations9. Most of these events- and the only direct date available for Denisovan specimen - pre-date (and sometime greatly64) the first occurrence of IUP in the Altai. While gene flow events are documented between Denisovans and H. sapiens, it is notable that age estimates ca. 45 ka for an early encounter65 are consistent with the dates presented here for AH6 (Fig. 4).

To summarize, the data at hand suggests that H. sapiens are more likely to be the makers of the IUP. The sudden appearance of this technology in Mongolia takes place in GI12, a temperate phase contemporaneous with the expansion of H. sapiens, in Western and Northern Siberia. Other scenarios cannot be rejected but they are less parsimonious, and they simply lack convincing alternatives regarding the archaeology of H. sapiens populations. Either way, the population movements illustrated by the dispersal of the IUP occur in a general context of climatic instability associated with MIS3 in Mongolia, and more generally in Central and Northeast Asia66. Such variations in the higher latitude of the Eurasian steppe imply different challenges (eg. continental climate, aridity) than those met along the southern route into the Indian subcontinent67. An inland route to Asia borne out by the archaeological and fossil record is, however, not exclusive of other dispersal models (e.g. Southern or ‘coastal’ Route) while it remains consistent with the available genetic data. The genome of an aboriginal Australian shows that deeply rooted populations have a shared ancestry coming from at least two early H. sapiens dispersals out of Africa into Asia68. Meanwhile, extant populations from continental East-Asia69 and in Melanesia70,71 show evidence of gene flow from Denisovan individuals from at least two gene flow events65. The timing might still be uncertain, but the dates presented here for the dispersal of IUP are consistent with the age estimates for an early H. sapiens – Denisovan encounter, distinct from the other gene flow events that would take place much later in Southeast Asia (or Melanesia)65. After subsequent mutations, the genes passed along from Denisovans around 45 ka may have been decisive for the survival of our species in extreme environments (Tibetan populations’ adaptation to hypoxia of high altitude)72,73. Finally, it is notable that East-Asian and Melanesian populations do also show evidence of gene flow coming from Neanderthal individuals65. Granted that the Pleistocene population dynamics were probably complex, a model integrating dispersals of several taxa, such as Neanderthal and H. sapiens across the continent is a parsimonious complement to a strict coastal migration scenario74, 42.


The results obtained at T16 indicate that the IUP techno-complex occurs in Western front of East Asia as early as 45 ka. Contemporaneous with skeletal evidence of H. sapiens, the IUP in Tolbor establishes an essential archaeological link between Siberia and Northern China. Together with evidence from these adjacent regions, the material found in AH6 suggest that a widespread behavioral shift took place along the Eurasian steppe belt during a period of climatic instability. The latter included numerous milder and wetter phases that potentially increases the carrying capacity of the landscape, thereby facilitating human expansions in these regions. Although causality links remain to be established, the timing of the IUP and H. sapiens dispersal appears to be coeval with the millennial-scale GI12 temperate climate event.


We excavated the material presented here between 2011 and 2015 and analyzed it in 2016–2017. The emphasis is on the lowest archaeological assemblage (AH6) in Pit4 and Pit1. We used a total station to map the anthropogenic features but also to locate in place artifacts bigger than 2 cm in length. When possible, we recorded orientation and inclination of the artifacts. Within stratigraphic units, we excavated sediments by arbitrary spits of 3–5 cm thickness and dry-sieved using a 4- and 2-mm mesh. We recorded sample locations with the total station. Geological analyses include lithostratographic descriptions, analyses of bulk samples and micromorphology. In Pit 1 and Pit 4, we collected bulk samples every 5 cm (SI 2) from the ground surface and for micromorphology, sediment blocks at the interface between units or sub-units. More information can be found in the the Supplementary information.


  1. 1.

    Lahr, M. M. & Foley, R. Multiple dispersals and modern human origins. Evol Anthropol 3(2), 48–60 (1994).

    Article  Google Scholar 

  2. 2.

    Stringer, C. Palaeoanthropology. Coasting out of Africa. Nature 405(6782), 24–25 (2000).

    CAS  Article  Google Scholar 

  3. 3.

    Field, J. S. & Lahr, M. M. Assessment of the Southern Dispersal: GIS-Based Analyses of Potential Routes at Oxygen Isotopic Stage 4. Journal of World Prehistory 19(1), 1–45 (2006).

    Article  Google Scholar 

  4. 4.

    Petraglia, M. D., Haslam, M., Fuller, D. Q., Boivin, N. & Clarkson, C. Out of Africa: New hypotheses and evidence for the dispersal of Homo sapiens along the Indian Ocean rim. Ann Hum Biol 37(3), 288–311 (2010).

    Article  Google Scholar 

  5. 5.

    Mellars, P., Gori, K. C., Carr, M., Soares, P. A. & Richards, M. B. Genetic and archaeological perspectives on the initial modern human colonization of southern Asia. Proc Natl Acad Sci USA 110(26), 10699–10704 (2013).

    ADS  CAS  Article  Google Scholar 

  6. 6.

    Krause, J. et al. Neanderthals in central Asia and Siberia. Nature 449, 902–904 (2007).

    ADS  CAS  Article  Google Scholar 

  7. 7.

    Prüfer, K. et al. The complete genome sequence of a Neanderthal from the Altai Mountains. Nature 505, 43–49 (2013).

    ADS  Article  Google Scholar 

  8. 8.

    Fu, Q. et al. Genome sequence of a 45,000-year-old modern human from western Siberia. Nature 514, 445–449 (2014).

    ADS  CAS  Article  Google Scholar 

  9. 9.

    Slon, V. et al. The genome of the offspring of a Neanderthal mother and a Denisovan father. Nature 561, 113–116 (2018).

    ADS  CAS  Article  Google Scholar 

  10. 10.

    Goebel, T. The Middle to Upper Paleolithic transition in Siberia [PhD dissertation]. Department of Anthropology, University of Alaska Fairbanks (1993).

  11. 11.

    Brantingham, P. J., Krivoshapkin, A. I., Jinzeng, L. & Tserendagva, Y. The Initial Upper Paleolithic in Northeast Asia. Curr Anthropol 42(5), 735–47 (2001).

    Article  Google Scholar 

  12. 12.

    Goebel, T. The Early Upper Paleolithic of Siberia. In The Early Upper Paleolithic Beyond Western Europe. eds Brantingham, P. J., Kuhn, S. L. & Kerry, K. W. (Los Angeles: University of California Press), 162–195 (2004).

  13. 13.

    Hoffecker, J. F. The spread of modern humans in Europe. Proc Natl Acad Sci USA 106(38), 16040–16045 (2009).

    ADS  CAS  Article  Google Scholar 

  14. 14.

    Zwyns, N. Laminar technology and the onset of Upper Paleolithic in the Altai (Siberia) (Leiden Univ Press, Leiden, The Netherlands) (2012).

  15. 15.

    Hublin, J.-J. The earliest modern human colonization of Europe. Proc Natl Acad Sci USA 109(34), 13471–13472 (2012).

    ADS  CAS  Article  Google Scholar 

  16. 16.

    Rybin, E. P. Tools, beads, and migrations: Specific cultural traits in the Initial Upper Paleolithic of Southern Siberia and Central Asia. Quat Int 347, 39–52 (2014).

    Article  Google Scholar 

  17. 17.

    Kuhn, S. L. & Zwyns, N. Rethinking the Initial Upper Paleolithic. Quat Int 347, 29–38 (2014).

    Article  Google Scholar 

  18. 18.

    Zwyns, N. et al. The open-air site of Tolbor 16 (Northern Mongolia): preliminary results and perspectives. Quat Int 347, 53–65 (2014).

    Article  Google Scholar 

  19. 19.

    Gillespie, A. R., Burke, R. M., Komatsu, G. & Bayasgalan, A. Late Pleistocene glaciers in Darhad basin, northern Mongolia. Quat Res 69, 169–187 (2008).

    Article  Google Scholar 

  20. 20.

    Rother, H., Lehmkuhl, F., Fink, D. & Nottebaum, V. Surface exposure dating reveals MIS-3 glacial maximum in the Khangai Mountains of Mongolia. Quat Res 82, 297–308 (2014).

    Article  Google Scholar 

  21. 21.

    Batbaatar, J., Gillespie, A. R., Fink, D., Matmon, A. & Fujioka, T. Asynchronous glaciations in arid continental climate. Quat Sci Rev 182, 1–19 (2018).

    ADS  Article  Google Scholar 

  22. 22.

    Lehmkuhl, F., Klinge, M., Rother, H. & Hülle, D. Distribution and timing of Holocene and late Pleistocene glacier fluctuations in western Mongolia. Annals of Glaciology 57(71), 169–178 (2016).

    ADS  Article  Google Scholar 

  23. 23.

    Herzschuh, U. Palaeo-moisture evolution in monsoonal Central Asia during the last 50,000 years. Quaternary Science Reviews 25, 163–178 (2006).

    ADS  Article  Google Scholar 

  24. 24.

    Yu, K. et al. Late Quaternary environments in the Gobi Desert of Mongolia: Vegetation, hydrological, and palaeoclimate evolution. Palaeogeography, Palaeoclimatology, Palaeoecology 514, 77–91 (2019).

    ADS  Article  Google Scholar 

  25. 25.

    Ma, Y. et al. Vegetation changes and associated climate variations during the past ~38,000 years reconstructed from the Shaamar eolian-paleosol section, northern Mongolia. Quaternary International 311, 25–35 (2013).

    ADS  Article  Google Scholar 

  26. 26.

    Wen, X., Li, B., Zheng, Y., Zhang, D. D. & Ye, J. Climate variability in the Salawusu River valley of the Ordos Plateau (Inner Mongolia, China) during Marine Isotope Stage 3. Journal of Quaternary Science 24(1), 61–74 (2008).

    ADS  Article  Google Scholar 

  27. 27.

    Yao, T. Climate variation since the Last Interglaciation recorded in the Guliya ice core. Science in China (series D) 40(6), 662–668 (1997).

    ADS  CAS  Article  Google Scholar 

  28. 28.

    Wang, Y. J. A high-resolution absolute-dated late Pleistocene Monsoon record from Hulu Cave, China. Science 294(5550), 2345–8 (2001).

    ADS  CAS  Article  Google Scholar 

  29. 29.

    Buckley, M., Collins, M., Thomas‐Oates, J. & Wilson, J. C. Species identification by analysis of bone collagen using matrix‐assisted laser desorption/ionisation time‐of‐flight mass spectrometry. Rapid Commun. Mass Spectrom. 23, 3843–3854 (2009).

    CAS  Article  Google Scholar 

  30. 30.

    Welker, F. et al. Palaeoproteomic evidence identifies archaic hominins associated with the Châtelperronian at the Grotte du Renne. Proc. Natl Acad. Sci. USA 113, 1162–1167 (2016).

    Article  Google Scholar 

  31. 31.

    Vasiliev, S. A. Faunal exploitation, subsistence practices and Pleistocene extinctions in Paleolithic Siberia. Deinsea 9, 513–556 (2003).

    Google Scholar 

  32. 32.

    Zwyns, N., Rybin, E. P., Hublin, J.-J. & Derevianko, A. P. Burin-core technology & laminar reduction sequence in the initial Upper Paleolithic from Kara-Bom (Gorny-Altai, Siberia). Quat Int 259, 33–47 (2012).

    Article  Google Scholar 

  33. 33.

    Derevianko, A. P., Petrin, V. T. & Rybin, E. P. The Kara-Bom site and the characteristics of the Middle-Upper Paleolithic transition in the Altai. Archaeol Ethnol Anthropol Eurasia 2(2), 33–52 (2000).

    Google Scholar 

  34. 34.

    Goebel, T., Derevianko, A. P. & Petrin, V. T. Dating the Middle-to-Upper-Paleolithic transition at Kara-Bom. Curr Anthropol 34(4), 452–8 (1993).

    Article  Google Scholar 

  35. 35.

    Zwyns, N. & Lbova, L. V. The Initial Upper Paleolithic of Kamenka site, Zabaikal region (Siberia): A closer look at the blade technology. Archaeol Res Asia 17, 24–29 (2019).

    Article  Google Scholar 

  36. 36.

    Higham, T. & Lbova, L. New perspectives on the chronology of the Initial Upper Palaeolithic of the Transbaikal, Russia. Session XXXIX Perspective on the Prehistory of Central Asia, 18th UISPP World Congress, Paris, June 2018 SE (book of abstract), 32 (2018).

  37. 37.

    Goebel, T. & Aksenov, M. Accelerator radiocarbon dating of the Initial Upper Palaeolithic in Southeast Siberia. Antiquity 69(263), 349–57 (1995).

    Article  Google Scholar 

  38. 38.

    Derevianko, A. P. et al. The technology of early Upper Paleolithic lithic reduction in Northern Mongolia: The Tolbor-4 site. Archaeol Ethnol Anthropol Eurasia 29(1), 16–38 (2007).

    Article  Google Scholar 

  39. 39.

    Gladyshev, S. A., Olsen, J. W., Tabarev, A. V. & Kuzmin, Y. V. Chronology and periodization of Upper Paleolithic sites in Mongolia. Archaeol Ethnol Anthropol Eurasia 38(3), 33–40 (2010).

    Article  Google Scholar 

  40. 40.

    Gladyshev, S. A. Radiocarbon dating of Paleolithic sites in the Ikh-Tulberiin-Gol River Valley, Northern Mongolia. Vestnik 12(5), 44–48 (2013).

    Google Scholar 

  41. 41.

    Bar-Yosef, O. & Belfer-Cohen, A. Following Pleistocene road signs of human dispersals across Eurasia. Quat Int 285, 30–43 (2011).

    Article  Google Scholar 

  42. 42.

    Goebel, T. The overland dispersal of modern humans to eastern Asia: an alternative, northern route from Africa. Emergence and Diversity of Modern Human Behavior in Paleolithic Asia eds Kaifu, Y., Izuho, M., Goebel, T., Sato, H. & Ono, A. (Texas A&M University Press, College Station), 437–452 (2015).

  43. 43.

    Brantingham, P. J., Gao, X., Madsen, D. B., Bettinger, R. L. & Elston, R. G. The Initial Upper Paleolithic at Shuidonggou, northwestern China. The Early Upper Paleolithic Beyond Western Europe. eds Brantingham, P. J., Kuhn, S. L. & Kerry, K. W. (Los Angeles: University of California Press), 233–241 (2004).

  44. 44.

    Li, F., Kuhn, S. L., Gao, X. & Chen, F. Re-examination of the dates of large blade technology in China: a comparison of Shuidonggou locality 1 and locality 2. J Hum Evol 64, 161–68 (2013).

    Article  Google Scholar 

  45. 45.

    Peng, F., Wang, H. & Gao, X. Blade production of Shuidonggou locality 1 (northwest China): a technological perspective. Quat Int 347, 12–20 (2014).

    Article  Google Scholar 

  46. 46.

    Zhang, X. L. et al. The earliest human occupation of the high-altitude Tibetan Plateau 40 thousand to 30 thousand years ago. Science 362(6418), 1049–1051 (2018).

    ADS  CAS  Article  Google Scholar 

  47. 47.

    Svensson, A. et al. A 60,000-year Greenland stratigraphic ice core chronology. Clim Past 4, 47–57 (2008).

    Article  Google Scholar 

  48. 48.

    Haesaerts, P., Borziac, I. & Chekha, V. Climatic signature and radiocarbon chronology of Middle and Late Pleniglacial loess from Eurasia: comparison with the marine and Greenland records. Radiocarbon 51(1), 201–318 (2009).

    Article  Google Scholar 

  49. 49.

    Müller, U. C., Pross, J., Tzedakis, P. C., Gamble, C. & Kotthoff, U. The role of climate in the spread of modern humans into Europe. Quat Sci Rev 30, 273–279 (2011).

    ADS  Article  Google Scholar 

  50. 50.

    Svoboda, J. A. Continuities, discontinuities, and interactions in Early Upper Paleolithic technologies: a view from the Middle Danube. The Early Upper Paleolithic Beyond Western Europe. eds Brantingham PJ, Kuhn SL, Kerry KW (Los Angeles: University of California Press), 30–49 (2004).

  51. 51.

    Škrdla, P. The Bohunician in Moravia and adjoining regions. Archaeol Ethnol Anthropol Eurasia 41(3), 2–13 (2013).

    Article  Google Scholar 

  52. 52.

    Pitulko, V. V. et al. Early human presence in the Arctic: Evidence from 45,000-year-old mammoth remains. Science 351(6270), 260–263 (2016).

    ADS  CAS  Article  Google Scholar 

  53. 53.

    Heinrich, H. Origin and consequences of cyclic ice rafting in the Northeast Atlantic Ocean during the past 130,000 years. Quat Res 29(2), 142–152 (1988).

    Article  Google Scholar 

  54. 54.

    Evans, M. E., Rutter, N. W., Catto, N., Chlachula, J. & Nyvlt, D. Magnetoclimatology: teleconnection between the Siberian loess record and North Atlantic Heinrich events. Geology 31(6), 537–540 (2003).

    ADS  Article  Google Scholar 

  55. 55.

    Begzsurena, S., Ellis, J. E., Ojima, D. S., Coughenourb, M. B. & Chuluun, T. Livestock responses to droughts and severe winter weather in the Gobi Three Beauty National Park, Mongolia. Journal of Arid Environments 59(4), 785–796 (2003).

    ADS  Article  Google Scholar 

  56. 56.

    Devièse, T. et al. Compound-specific radiocarbon dating and mitochondrial DNA analysis of the Pleistocene hominin from Salkhit Mongolia. Nature Communications 10, 274 (2019).

    ADS  Article  Google Scholar 

  57. 57.

    Lbova, L. Evidence of Modern Human Behavior in the Baikal Zone during the Early Upper Paleolithic Period. Bulletin of the Indo-Pacific Prehistory Association 30, 9–13 (2011).

    Article  Google Scholar 

  58. 58.

    Jaubert, J. et al. Le Paléolithique supérieur ancien de Mongolie: Dörölj 1 (Egiïn Gol). Analogies avec les données de l’Altaï et de Sibérie. Secrétariat général du Congrès (éd.), Section 6. Le Paléolithique supérieur/The Upper Paleolithic. Sessions générales et posters / General Sessions and Posters, Actes du XIVe Congrès UISPP, Université de Liège, 2–8 sept. 2001, Archaeopress, pp. 225–241., BAR International Series, 1240 (2004).

  59. 59.

    Janz, L., Elston, R. G. & Burr, G. S. Dating North Asian surface assemblages with ostrich eggshell: implications for palaeoecology and extirpation. J. Arch Sci 36(9), 1982–1989 (2009).

    Article  Google Scholar 

  60. 60.

    Derevianko, A. P., Markin, S. V. & Shunkov, M. V. The Sibiryachikha Facies of the Middle Paleolithic of the Altai. Archaeol Ethnol Anthropol Eurasia 41(1), 89–103 (2013).

    Article  Google Scholar 

  61. 61.

    Douka, K. et al. Age estimates for hominin fossils and the onset of the Upper Palaeolithic at Denisova Cave. Nature 565, 640–644 (2019).

    ADS  CAS  Article  Google Scholar 

  62. 62.

    Jacobs, Z. et al. Timing of archaic hominin occupation of Denisova Cave in southern Siberia. Nature 565(7741), 594–599 (2019).

    ADS  CAS  Article  Google Scholar 

  63. 63.

    Slon, V. et al. A fourth Denisovan individual. Sci Adv 3(7), e1700186 (2017).

    ADS  Article  Google Scholar 

  64. 64.

    Chen, F. et al. A late Middle Pleistocene Denisovan mandible from the Tibetan Plateau. Nature 569(7756), 409–412 (2019).

    ADS  CAS  Article  Google Scholar 

  65. 65.

    Jacobs, G., Hudjashov, G., Saag, L., Kusuma, P. & Darusallam, C. Multiple Deeply Divergent Denisovan Ancestries in Papuans. Cell 177(4), 1010–1021.e32 (2019).

    CAS  Article  Google Scholar 

  66. 66.

    Fitzsimmons, K. E. et al. Loess accumulation in the Tian Shan piedmont: implications for palaeoenvironmental change in arid Central Asia. Quat Int 469, 30–43 (2018).

    Article  Google Scholar 

  67. 67.

    Hoffecker, J. F. Modern Humans: Their African Origin and Global Dispersal (New York: Columbia University Press) (2017).

  68. 68.

    Rasmussen, M. et al. An Aboriginal Australian Genome Reveals Separate Human Dispersals into Asia. Science 334, 94–7 (2011).

    ADS  CAS  Article  Google Scholar 

  69. 69.

    Skoglund, P. & Jakobsson, M. Archaic human ancestry in East Asia. Proc Natl Acad Sci USA 108, 18301–18306 (2011).

    ADS  CAS  Article  Google Scholar 

  70. 70.

    Reich, D. et al. Genetic history of an archaic hominin group from Denisova Cave in Siberia. Nature 468, 1053–1060 (2010).

    ADS  CAS  Article  Google Scholar 

  71. 71.

    Reich, D. et al. Denisova admixture and the first modern human dispersals into Southeast Asia and oceania. Am J Hum Genet 89(4), 516–28 (2011).

    CAS  Article  Google Scholar 

  72. 72.

    Huerta-Sanchez, E. Altitude adaptation in Tibetans caused by introgression of Denisovan-like DNA. Nature 512, 194–197 (2014).

    ADS  CAS  Article  Google Scholar 

  73. 73.

    Zhang, X. et al. High Altitude Adaptation and the Evolutionary History of EPAS1 Gene in Tibetan Population. American. The 88th Annual Meeting of the American Association of Physical Anthropologists – Podium abstract, (2019).

  74. 74.

    Weaver, T. Tracing the paths of modern humans from Africa. Proc Natl Acad Sci USA 111(20), 7170–7171 (2014).

    ADS  CAS  Article  Google Scholar 

  75. 75.

    Izuho, M., Tsogtbatar, B., Yamaoka, T., Hayashi, K. & Enkhtor, A. Preliminary Report of Geoarchaeological Investigation at Khanzat-1, Eastern Mongolia. In: Current Archaeological Research in Mongolia (eds. an Bemmann, Hermann Parzinger, Ernst Pohl, Damdinsüren Tseveendorzh), Vor- und Frühgeschichtliche Archäologie Rheinische Friedrich-Wilhelms-Universität Bonn, 31–45 (2009).

  76. 76.

    Odsuren, D., Bukhchuluun, D., Janz, L. & Rosen, A. Preliminary results of ongoing research at Stone Age site Zaraa uul). In: Archaeological Research, Preservation, and Protection in Eastern Mongolia. Mongolian Academy of Sciences, Ulaanbaatar, 234–248 (2017).

  77. 77.

    Gunchinsuren, B., Bazargur, D., Yi, S. & Lee, J. The study results of Stone Age site Rashaan khad, Studia Archaeologica, 225–229 (2017).

  78. 78.

    Reimer, P. J. et al. Intcal09 and Marine09 radiocarbon age calibration curves, 0–50,000 years cal BP. Radiocarbon 51(4), 1111–1150 (2009).

    CAS  Article  Google Scholar 

  79. 79.

    Bronk Ramsey, C. Bayesian analysis of radiocarbon dates. Radiocarbon 51(1), 337–360 (2009).

    Article  Google Scholar 

  80. 80.

    Bronk Ramsey, C. Dealing with outliers and offsets in radiocarbon dating. Radiocarbon 51(3), 1023–1045 (2009).

    Article  Google Scholar 

Download references


The authors are grateful to the Mongol authorities and the Governor of Bulgan Aimag, to the Leakey Foundation (2011–2016) and the Max Planck Society (2011–2016) for their continuous support throughout the excavation of the Tolbor-16 site. Additional support was provided by the Hellman Foundation (2015), the National Scientific Foundation (2016) and the UC-Davis Academic Senate (2013). We thank all the (many) participants to the field seasons, the staff and the interns who helped us along the way. We are grateful to Damdinsuren Tseveendorj, Sergey A. Gladyshev, Andrei V. Tabarev, Shannon P. McPherron and John W. Olsen for their help in setting up the project and fieldwork. We thank Clementine Defontaine, Yuri Demidenko, Peng Fei, Jovan Galfi, Masami Izuho, Lisa Janz, Nathaniel Kitchel, Alexander Kuban, Steven L. Kuhn, Feng Li, Timothee Libois, Sam Lin, Philip R. Nigst, Marcel Otte, Morgan Roussel, John R. Stewart, Nicolas Teyssandier, Timothy Weaver, Teresa Steele, Giulia Gallo, Corey Johnson and Peiqi Zhang, for their help in fieldwork and for scientific discussions. FW is supported by the VILLUM Fonden (#17649) and MSCA-IF (#795569). A special thanks goes to Bernd Kromer who did run our samples in the Mannheim AMS facility.

Author information




N.Z. designed the research; N.Z., C.H.P., B.T., S.T., D.F., A.G. and T.D. performed research; N.Z., C.H.P., S.T., T.D., K.E.F., J.C.G. and F.W. contributed new reagents/analytic tools; N.Z., C.H.P., B.T., S.T., D.F., A.G., L.G., O.D., K.E.F., T.D., N.D., F.W., J.B.N., R.S.B., A.F.A., K.N.S., A.M.K., E.P.R., G.B. and J.-J.H. analyzed data; N.Z. and C.H.P. wrote the paper.

Corresponding author

Correspondence to Nicolas Zwyns.

Ethics declarations

Competing Interests

The authors declare no competing interests.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Zwyns, N., Paine, C.H., Tsedendorj, B. et al. The Northern Route for Human dispersal in Central and Northeast Asia: New evidence from the site of Tolbor-16, Mongolia. Sci Rep 9, 11759 (2019).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing