Winter-to-summer seasonal migration of microlithic human activities on the Qinghai-Tibet Plateau

Abstract

The Qinghai-Tibet Plateau (QTP) has become a valuable site for investigation of adaptive regimes of prehistoric humans to extreme environments. At present most studies have focused solely on a single site. Using a more integrated approach that covers the complete scope of the plateau is needed to better understand the expansion logic of prehistoric humans moving towards the plateau. Here, we conducted accelerator mass spectrometry ¹⁴C dating of two microlithic sites. Canxiongashuo (CXGS) and Shalongka (SLK), which are located at the inner and marginal areas of the QTP, respectively. By using geographic information system, literature, and natural environmental factors, we constructed a model for the relationship between traveling distance and time, and we also used these factors to construct a plateau environmental index. The results indicated that the ages of the CXGS and SLK sites are 8.4–7.5 cal. ka BP and 8.4–6.2 cal. ka BP, respectively. Combining the archaeological evidence and literature, hunter-gatherers may have seasonal migration activities at low altitude in winter and high altitude in summer in order to make full use of natural resources. Our model of relationship between traveling distance and time shows that hunter-gatherers in CXGS site was active on the plateau all year-round at approximately 8.3 cal. ka BP. According to EI and archaeological remains, we propose that SLK site was a winter camp of prehistoric hunter-gatherers. Taken together, we determined 8.4–6.0 cal. ka BP as a transitional period from the Paleolithic to Neolithic Ages, and winter camps of hunter-gatherers evolved into settlements in the Neolithic Age.

Full text



Scientic Reports | (2020) 10:11659 | 
www.nature.com/scientificreports
Winter‑to‑summer seasonal
migration of microlithic human
activities on the Qinghai‑Tibet
Plateau
Guangliang Hou1,2*, Jingyi Gao2, Youcheng Chen3, Changjun Xu4, Zhuoma Lancuo1,
Yongming Xiao5, Linhai Cai5 & Yuanhong He6
The Qinghai‑Tibet Plateau (QTP) has become a valuable site for investigation of adaptive regimes of
prehistoric humans to extreme environments. At present most studies have focused solely on a single
site. Using a more integrated approach that covers the complete scope of the plateau is needed to
better understand the expansion logic of prehistoric humans moving towards the plateau. Here, we
conducted accelerator mass spectrometry 14C dating of two microlithic sites. Canxiongashuo (CXGS)
and Shalongka (SLK), which are located at the inner and marginal areas of the QTP, respectively. By
using geographic information system, literature, and natural environmental factors, we constructed
a model for the relationship between traveling distance and time, and we also used these factors to
construct a plateau environmental index. The results indicated that the ages of the CXGS and SLK sites
are 8.4–7.5 cal. ka BP and 8.4–6.2 cal. ka BP, respectively. Combining the archaeological evidence and
literature, hunter‑gatherers may have seasonal migration activities at low altitude in winter and high
altitude in summer in order to make full use of natural resources. Our model of relationship between
traveling distance and time shows that hunter‑gatherers in CXGS site was active on the plateau all
year‑round at approximately 8.3 cal. ka BP. According to EI and archaeological remains, we propose
that SLK site was a winter camp of prehistoric hunter‑gatherers. Taken together, we determined
8.4–6.0 cal. ka BP as a transitional period from the Paleolithic to Neolithic Ages, and winter camps of
hunter‑gatherers evolved into settlements in the Neolithic Age.
ere has been an increasing attention on the study of the activity patterns of hunter-gatherers during the transi-
tional period between the Paleolithic and Neolithic Ages1–3, and it is generally considered that prehistoric human
during that time are high in mobility. e Qinghai-Tibet Plateau (QTP) with its vulnerable environmental, is
extremely sensitive to climate change4. Furthermore, scarce vegetation cover, low temperature, and dicient in
oxygen atmosphere during the winter time, resulted in extremely harsh environment for prehistoric humans to
settle5,6. erefore, in academic community there is an increasing interest on the prehistoric huaman’s adapta-
tion to the QTP and a striking progress has been made in the study of the prehistoric human–environment
relationships on the QTP7–10. ese studies provide insights to help us comprehensively understand the process
of human-land relationship. Nevertheless, there are controversial opinions on the prehistoric humans’ occupa-
tion of the plateau due to the lack of surveys and excavations11,12.
Previous studies suggest that hunter-gatherers did not occupy the plateau during the winter until to the
Neolithic age. It was not until 3.6cal. ka BP with the support of animal domistication and barley cultivation that
prehistoric humans could occupy the plateau all year round13. However, according to the studies of Chusang
site there are some scholars believe that hunter-gatherers had occupied the plateau all through the year since the
early Holocene14. Nevertheless, this conclusion has been questioned owing to the controversy over its travel cost
modelling of hunter-gatherers11. At present, many studies showed that the life style of hunter-gatherers before
the Neolithic Age was dominated by high mobility15. It is likely that they migrated to lower elevation regions in
OPEN
       School of Geographic Science, Qinghai
            
            
           Department of
 *
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2020) 10:11659 | 
www.nature.com/scientificreports/
order to avoid extremely harsh winter environments, while they would have moved to higher elevation, regions
to hunt in the summer6,11. us, to determine whether the prehistoric humans had occupied the QTP in winter
was the resolution for this dispute. is is also a key step for prehistoric human to adapt the extreme environ-
ment condition of the plateau.
Over the past several decades, many microlithic sites of early-mid Holocene have been discovered on the QTP.
Research shows that microlithic technologies were used by prehistoric humans3, who mainly hunted small- and
medium-sized mammals, such as antelope, deer, and marmot, which had dominated the plateau during that
period16,17. ese hunter-gatherer activities had two characteristics: rst, that their activities were relatively scat-
tered, with long distances and frequent migrations; second, their camps were characterized by seasonality and
randomness. ere were additional sites with rich remains and multiple functions, such as those at the Heima
River, Yantaidong, Jiangxigou 2, and Layihai, which may have been repeatedly and seasonally used for a long
time as the sites had properties of central camps18,19. ese ndings suggest that large groups of hunter-gathers
oen came to Qinghai Lake Basin for long-term or seasonal living.
is seasonal migration pattern is also very common on the QTP even today3. In particular, this traditional
nomadic seasonal migration between winter and summer camp is very popular today in the Qinghai Lake Basin
(Supplementary Fig.1). e vertical zonality of QTP leads to seasonal migration of wildlife between high and
low elevations. Wild donkey is an excellent example. In winter, they tend to go from highland to low-elevation
plain, while in summer, they go from low-elevation plain to highland. us, people in the Qinghai Lake Basin
are also deeply inuenced by the seasonal migration of wildlife and they perform regular seasonal migration
pattern each year. More specically, in winter, in the lowlands along the lake are the tribal settlements, because
the highlands are extremely cold and the snow is deep, and it is dicult for livestock to get food, while during
summer, people return to the river valley regions and move to the high elevation mountains. e low elevation
pastures on the lakeside are reserved for their winter fodder. is seasonal migration has some typical char-
acteristics: there are relatively xed camps and routes across the four seasons and a one-way length is within
200km. Notably, this migration has a seasonal property, and most of these routes are along rivers. e nomadic
seasonal migration along the Haergai River can serve as an example. From November to April of the following
year, people camped at Nalongwaerma, which is located at the ground level near the Qinghai Lake, was used as
a winter camp, owing to its low elevation and rich hay. In the following April, people migrate to the spring camp,
Qingdama, which has higher elevation. From June to July, they arrive at their summer camp, Mori in the river
valley of the upper reaches of the Datong River. Finally, from September to October, they move to their autumn
Figure1. e stratigraphic lithology of sedimentary sequences and the positions of AMS 14C data and
microliths in (a) CXGS and (b) SLK sites.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2020) 10:11659 | 
www.nature.com/scientificreports/
camp, Adasi. In November, they return back to their winter camp at Nalongwaerma. e entire one-way travel
distance is approximately 110km, which is repeated each year (Supplementary Fig.1).
e above mentioned subsistence strategy of seasonal migration in the Qinghai Lake Basin is highly consist-
ent with the ecological and phenological changes of the QTP20. Previous studies have suggested that the growth
period of vegetation on the QTP starts from the end of April to the beginning of May, with the maturity period
at the end of July to the end of August. At the beginning of September growth starts to decline, and ends at the
middle-to-late October. Finally, the dormancy period starts in the middle of November and ends the following
April. Based on the archaeological evidence and living data of modern residents16–19,21, this seasonal migration
pattern seems to have obvious changes from the Last Deglacial to modern times, which is a relatively xed life
tradition. Given these ndings, it can be used as a reference for understanding the patterns of prehistoric human
activities.
In this paper, our research period is focused at 8.5–6.0cal. ka BP of the early-mid Holocene. e accelerator
mass spectrometry (AMS)14C dating was employed in the Canxiongashuo (CXGS) (inner of the plateau) and
Shalongka (SLK) (margin of the plateau) sites section. We used geographic information system (GIS) to estab-
lish a model for the relationship between traveling distance and time as well as the plateau environmental index
(EI). e objective of the study was to provide chronological information and activity characteristics of CXGS
and SLK sites, and determine whether the humans of the CXGS site occupied the plateau all year-round and
the nature of the SLK site camp. On this base, this study further elucidated the distribution area of the winter
camps on the QTP and the characteristics of the prehistoric human activities during the transitional period from
Paleolithic to Neolithic Age.
Results and discussion
Age. CXGS section was approximately 105cm thick and stone artifacts were found in a layer of 95–20cm
deep with most of them were found in layers of 70–65 and 75–70cm deep. e AMS 14C dating results of CXGS
section indicates stratigraphic deposition since the early-late Holocene, but the calibration ages of charcoal were
notably inverted in 95–70cm depth, which is common in the stratum aected by human activities. Basically, the
calibration ages of the charcoal were mainly concentrated across two periods: (1) 8.4–7.8cal. ka BP, the results
of four charcoal ages occurred at a depth of 95–65cm, and (2) 2.3–2.2cal. ka BP, the results of two charcoal ages
occurred in the stratum at a depth of 50–21cm. In addition, the rst occupancy of microlithic artifacts was at
approximately 8.4cal. ka BP and continued until 7.5cal. ka BP. e number of microlith signicantly intensied
at around 7.8cal. ka BP (Fig.1a, Table1).
SLK section was approximately 415cm thick. According to the unearthed cultural relics (pottery shards,
stone artifact), the bottom-up cultural sequence was 413–304cm as the microlithic culture layer, 280–250cm
as the Yangshao culture layer (6.0–5.0cal. ka BP), 110–88cm as the Qijia culture layer (4.3–3.6cal. ka BP) and
88–78cm as the Kayue culture layer (3.6–2.6cal. ka BP). In addition, each cultural layer was mixed with dier-
ent thicknesses of uvial deposits. e dating results of the SLK section indicates stratigraphic deposition since
the early-late Holocene, and had a deep section with several culture layers. ese layers were also interactively
deposited with uvial deposit-silt layers (Fig.1b, Table1). e age-depth correlation was reasonably good and
relatively continuous. In general, this stratigraphic deposition from 8.4cal. ka BP. e period of microlithic
hunter-gatherers was found at 8.4–6.2cal. ka BP, and the peak of microlithic artifacts occurred in the stratum
approximately 7.6cal. ka BP. Furthermore, there were regular column holes in the stratum, which were speculated
by the archaeological excavators to be closely related to architectural remains such as sheds.
By comparing the dating results and unearthed stone artifacts of the CXGS and SLK sites, it was determined
that the prehistoric humans of the both sites all used microlithic technology. Also, these two sites were all dated
from approximately 8.4cal. ka BP and signicantly increased at 7.8–7.6cal. ka BP. Taken together, although the
Table 1. e AMS 14C dates from the CXGS and SLK sections.
Section Lab number Depth (cm) Dating material R adiocarbon age (year BP) C alibrated age (2σ: cal. year BP)
CXGS
BA 151,596 21 Charcoal 2,330 ± 20 2,346 ± 50
BA 151,597 45–50 Charcoal 2,210 ± 25 2,235 ± 84
BA 151,598 65–70 Charcoal 7,055 ± 25 7,894 ± 53
BA 151,599 75–80 Charcoal 7,765 ± 30 8,565 ± 28
BA 151,600 85–90 Charcoal 6,985 ± 25 7,821 ± 32
BA 151,601 95 Charcoal 7,570 ± 25 8,383 ± 30
SLK
BA 161,048 98 Charcoal 1940 ± 25 1884 ± 59
BA 161,049 183 Charcoal 5,390 ± 45 6,230 ± 60
BA 161,051 275 Charcoal 6,605 ± 30 7,480 ± 44
BA 161,052 297 Charcoal 6,865 ± 30 7,692 ± 69
BA 161,054 317 Charcoal 6,760 ± 30 7,621 ± 45
BA 161,055 321 Charcoal 6,875 ± 30 7,723 ± 67
BA 161,056 348 Charcoal 6,970 ± 30 7,758 ± 80
BA 161,057 393 Charcoal 7,640 ± 30 8,432 ± 48
BA 161,058 408 Charcoal 7,645 ± 30 8,435 ± 49
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2020) 10:11659 | 
www.nature.com/scientificreports/
two sites were located in the inner and margin of the plateau, respectively, the nature and age of the microlithic
hunter-gatherers at the two sites were basically similar.
Traveling time in CXGS site. Based on the model of the relationship between traveling distance and time,
we obtained analysis results of the relationship between traveling distance and time from the margin to the inner
QTP (Fig.2). ese results showed that the time from the margin of the plateau to the other known microlithic
sites on the plateau was less than 45 days15–18. However, it should be noted that the time from the margin of
the plateau to the SLK site was less than 15days but the time from the margin of the plateau to CXGS site was
approximately 60days (Fig.2). As discussed above, the time for hunter-gatherers to arrive at the inner plateau
during one year should not exceed 60days. us, based on the behavioral patterns of the hunter-gatherers
presented above, if any hunter-gatherers entered an area that took ≥ 60days, they would be unable to leave the
plateau. However, if hunter-gatherers from the margin of the plateau went to the CXGS site at the inner of the
plateau, it would take approximately 60days. e study suggests that due to the extreme environment (e.g.,
hypoxic conditions), travelling speed is aected not only by slope, but also by other factors such as elevation.
Moreover, the decrease of maximal oxygen consumption (VO2max) and anaerobic threshold (AT) in the plateau
environment results in decreased physical activity and labor22. In general, for every 1,000m increase in eleva-
tion, labor endurance decreases by half. erefore, the traveling speed in the plateau should be gradually lowered
with an increase in elevation, resulting in increased travel time. In addition, hunter-gatherers in the plateau were
mainly engaged in traveling, in which search of food and hunting activities played signicant roles along the way.
Additionally, during 8.5–6.0cal. ka BP, there were no domesticated animals in the plateau to be used as transport
tools, so hunter-gatherers in the plateau were slower than either those at lower elevations or those using ani-
mals for travel (e.g. documented sixteenth–twentieth century). erefore, in the 60days of traveling, the actual
traveling distance was shorter than that calculated by the model. From the above analysis of simulation results,
if the microlithic hunter-gatherers reached the CXGS site from the margin of the plateau in spring of a certain
year during 8.3–7.5cal. ka BP, it was dicult for them to get out of the plateau in autumn. In other words, both
in summer and winter, the hunter-gatherers were active on the plateau and had not le the plateau that is, they
occupied the plateau all year-round.
Plateau environmental index and human activities. According to the above analysis, prehistoric
hunter-gatherers in the inner of the plateau during early-late Holocene lived on the plateau year-round, speci-
cally, they settled on the high elevation plateau in summer and low elevation plateau in winter. Here, the key
question is which low elevation areas were suitable for the microlithic hunter-gatherer as winter camps? e
research site of SLK in the study would be one of the alternative sites. ere are two main reasons, rst, this site
Figure2. e map representing the relationship between distance and traveling time (days) from the margin to
the inner of the plateau.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2020) 10:11659 | 
www.nature.com/scientificreports/
had low elevation, high winter temperature, shallow snow, short snow-cover time, and as well as relatively supe-
rior environmental conditions. Furthermore, the deposition of the cultural layer in the microlith of the SLK sec-
tion was not only thick, but continuous, in which many stone artifacts such as microliths were found. Especially,
there was an emergence of several regular column holes in the culture layer in the SLK section’s microlith at a
depth of 318cm. Such column holes have been considered to be obvious signals of housing remains, indicating
that prehistoric hunter-gatherers lived there for a long time, and there once was a relatively xed seasonal camp
with a long activity time and repeated use. us, we can infer that this site could be used as a winter camp for the
microlithic hunter-gatherers. Also, it is implied that any site with environment conditions similar to the SLK site
in prehistoric times may be used as a winter camp. In order to determine other natural environment areas similar
to the SLK site, the plateau EI was constructed in this study (Fig.3a). e results show that the EI of SLK site was
6.8. us, we dened areas with environmental indices of ≥ 6.8 as suitable areas for winter activity. It can be seen
from Fig.3 that the areas suitable for winter activity is mainly distributed in a few areas, such as Luoyu of the Yar-
lung Zangbo River, and the river valley areas of Guide and Minhe county at the upper reaches of the Yellow River.
On this basis, this study discusses the environmental index of Neolithic, Bronze Age and modern human
settlements, as well as the distribution area of human activities related to it. During the middle Holocene
(5.6–4.0cal. ka BP), Neolithic Age cultures such as the Majiayao, Zongri, and Karuo were distributed within
the plateau23,24 (Fig.3b). Also, Fig.3a or 3b show where these sites are located. Among these sites, houses have
been commonly found, along with cultural relics such as ground stones and potteries, indicating that prehis-
toric hunter-gatherers had settled on the plateau. In particular, the Zongri site is mainly distributed at the upper
reaches of the Yellow River, while the Karuo site is located at the upper reaches of the Lancang River on the inner
plateau24–26. e EI of both two sites were approximately 5.3. Given this, areas with EI between 5.3 and 6.8 have
Figure3. EI and settlements in the QTP (a) EI in QTP; (b) EI and Neolithic sites; (c) EI and modern
settlements.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2020) 10:11659 | 
www.nature.com/scientificreports/
been dened as areas for winter activity. Basically, such areas have been considered to be classic examples of set-
tled areas on the plateau during the Neolithic Age, which mainly include the Shannan and river valley under the
Qushui of Yarlung Zangbo River, the river valleys of three parallel rivers, Nujiang, Lancang and Jinsha Rivers,
and the river valley area under Maqu of the upper reaches of the Yellow River. e Kayue culture (3.6–2.6cal. ka
BP) played an important role in the Bronze culture in the northeast of the QTP13,24,27. e introduction of a mixed
economy of barley and animal husbandry nally enabled prehistoric human to permanently occupy on the inner
part of the plateau13. is culture was mainly distributed in areas with an EI of ≥ 4.2; thus, we have dened areas
with EI between 4.2 and 5.3 as transitional area of prehistoric hunter-gatherer activite area. Interestingly, the EI
of modern settlements was found to be more than 1.8, while in the areas with EI of less than 1.8, there are less
human activities, so they were dened as weakly active areas. According to the distribution pattern of Neolithic,
Bronze Age and modern human, the areas with EI between 1.8 and 4.2 may be the areas where the prehistoric
hunter gatherers active in summer. erefore, it can be seen from Fig.3 that the suitable areas for winter camps
were mainly distributed in a few areas, such as the Luoyu region of the Yarlung Zangbo River, and the river val-
leys of the upper reaches of Yellow River in Guide County; And the prehistoric hunter gatherers were mainly
active during the winter Shannan valley and the river valley under the Qushui of the Yarlung Zangbo River, the
river valleys of three parallel rivers, Nujiang, Lancang, and Jinsha Rivers, and the river valley areas under the
Maqu River of the upper reaches of the Yellow River. ese analyses showed that suitable areas for winter camp
and areas that were occupied by prehistoric hunter gatherers during the winter were basically river valley areas
with relatively low elevation and good climate conditions.
It is worth noting that if only the environmental conditions are considered, the SLK site in this study may also
be a winter camp for the prehistoric humans at the CXGS site (Fig.4), however, if the long distance between the
two sites, the large energy consumption, and the risk along the way are also considered, the SLK site is unlikely to
be the winter camp for the prehistoric humans at the CXGS site. If the CXGS site is regarded as a summer camp,
the corresponding winter camp is possibly distributed in the river valley of the lower reaches of the Tongtian
River with warm and wet environment. More importantly, these two camps have a shorter distance, which needs
approximately 15–35days for prehistoric humans to arrive at one camp from another. In summary, prehistoric
humans are likely to follow the “up and down the plateau” seasonal migration pattern along river valleys with
better environmental conditions.
Evolution and possible mechanisms from the microlith to Neolithic Ages. Some studies have
revealed that the competitive pressure of the agricultural population in the Loess Plateau resulted in the appear-
Figure4. e topographic map of the QTP and the potential route between the SLK and CXGS sites. It is worth
noting that the potential route is the shortest path based on slope simulation. Inset shows the location of the
QTP in china.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2020) 10:11659 | 
www.nature.com/scientificreports/
ance of specialized epi-Paleolithic blade and bladelet technologies on the high plateau aer 8.2cal. ka BP28. is
may indicate more permanent occupation and new full-time residents on the plateau. Based on the above discus-
sion, the CXGS site of the inner plateau and the SLK site of the marginal plateau are likely to be the classic exam-
ples of summer and winter camps related to the activities of microlithic hunter-gatherers on the plateau, which
also support the view that aer 8.2cal. ka BP, there was emergence of hunter-gatherers active on the plateau all
year-round. In addition, the history of human development indicates that in dierent periods, human activities
developed characteristics of inheritance and continuity. e research period of this study was the transitional
stage between the Paleolithic and Neolithic Ages on the plateau and the human behavior pattern that was the
transitional stage from hunting and gathering to agriculture, from mobility to settling, and from seasonal camps
to settled camps. ese have clear signs of transition, inheritance, and continuity. Although we found a sedimen-
tary layer mixed with proluvial and uvial deposits in SLK section, the more obvious feature is that the Yangshao
culture layer superimposed on the microlith culture layer, and the column holes representing human settlement
behavior were found in the microlith culture layer (Fig.1b), indicating that SLK site is probably evolved from
the winter camp of microlithic hunter-gatherers to the settlement of Neolithic Age, and its potential distribution
area is probably the suitable area for winter activities based on the plateau EI.
At 8.0cal. ka BP, hunter-gatherers emerged were active on the plateau all year-round. is was likely owing
due to the competition and population pressure of the agricultural people of lower elevations29–31, the applicabil-
ity of microlithic technology and relatively superior natural environment conditions32. Specically, the Dadiwan
site of the Neolithic Age, which was dominated by agricultural cultivation, appeared in this period on the Loess
Plateau (adjacent to the northeast of the QTP)33. Subsequently, hunter-gatherers were squeezed out of this area
by the development of the Neolithic culture. In order to maintain the tradition of hunter-gatherers, prehistoric
humans had to spread into the QTP all year-round on a large scale6. Due to the applicability of microlithic tech-
nology in the harsh environment, a large number of microlithic sites were found in the QTP during the middle
Holocene, such as Jiangxigou 234 and Yeniugou35. Additionally, quantitatively reconstructed precipitation based
on pollen data from Lake Luanhaizi36, Lake Yidun37 and Lake Qinghai38 on the eastern QTP indicated that there
was more precipitation in the early and middle Holocene39, less aeolian deposits weakly40, paleosols were formed
mainly during 9.5–4.0ka BP41. With the above background, QTP is likely to have seen the emergence of hunter-
gatherers all year-round in the early and middle Holocene.
Surely, due to the complexity and particularity of human activities on the plateau, traveling speed and the time
the hunter-gathers spent on the plateau depended not only on the slope, but also on related factors like eleva-
tion, topography, underlying surface, and weather. For example, hunter-gatherers were oen confronted with
glaciers, sand, swamps, and other underlying diculties to pass. Large rivers and storms also became obstacles
to traveling. erefore, the actual traveling distance would be less than that calculated by our model regarding
the relationship between distance and time. In other words, the hunter-gatherers’ traveling distance was less than
the distance calculated by the model within the same time (60days) spent on traveling.
Conclusion
e CXGS site in the inner plateau and the SLK site on the margin of the plateau are all the remains of micro-
lithic hunter-gatherers. e age of these microlithic hunter-gatherers was approximately 8.3–7.5cal. ka BP and
occurred in the early and middle Holocene. e age of the activities at the two sites is consistent with the relics
excavated from the sites.
e microlithic sites on the plateau exhibited characteristics of dispersion, randomness, and seasonality. ere
should be seasonal migratory activities of microlithic hunter-gatherers between winter and summer camps.
It means hunter-gatherers were active at low-elevation regions in the winter and at high-elevation regions in
the summer. More specically, hunter-gatherers moved to their high elevation summer camp from April to
May, and returned to their low elevation winter camp from September to October. Furthermore, this seasonal
migration between winter and summer needed to happen within no more than 60days every year. According to
our model regarding the relationship between distance and time at CXGS site, at 8.0cal. ka BP, the microlithic
hunter-gatherers did not leave the plateau in the winter. EI and archaeological evidence revealed that SLK site
was a suitable winter camp for hunter-gatherers. Furthermore, the winter camp and distribution area of hunter-
gatherers likely later evolved into a Neolithic settlement and distribution area of these sites.
Materials and methods
Study areas. e CXGS site (33°48′17″N, 96°02′36″E, 4,030m a.s.l (above sea level)) is located on the rst
terrace of the Tongtian River in the center of the QTP (Supplementary Figs.2a, 4). e accumulated temperature
of ≥ 0°C that is measured throughout the year (from the beginning of the farming period until to the start of
grass growth) is 1,009°C. is accumulated temperature can only develop independently to husbandry pro-
duction (Table2). ere are only 20 frost-free days throughout the year when the daily minimum temperature
is > 0°C; these occur from July 24 to August 13. e vegetation is mainly alpine meadow, alpine grassland,
and shrub. e CXGS site was rst investigated and excavated in detail by the Institute of Cultural Relics and
Archaeology of Qinghai Province. e CXGS site is located in the inner plateau and it is a large-scale site (1 km2)
with the properties of a central camp. And a core area of approximately 15,000 m2. ere are a large number of
microblades, microblade cores, and other microlithic artifacts scattered on the surface of the site. Wedge-shaped
microblade cores have long been considered a classic example of microlithic artifacts; notably, microblades and
akes also appeared at the site. Some studies have noticed that microblade technology in the CXGS site and
"Layihai technology", which was found in the Yellow River Valley at the northeast margin of the plateau, exhib-
ited a high degree of similarity in their microblade cores. is similarity is in terms of material selection, preform
production, platform rejuvenation, and microblade detaching42, and indicates that central plateau had a close
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2020) 10:11659 | 
www.nature.com/scientificreports/
relationship with the plateau margin in their individual microlithic technologies. In the core area of the site, the
Institute of Cultural Relics and Archaeology of Qinghai Province and other organizations established quadrats
to excavate.
e SLK site (36° 02′ 52″ N, 101° 57′ 72″ E, 2076m a.s.l.) is located on the second terrace of the Yellow River
in the northeastern region of the QTP (Fig.4, Supplementary Fig.2c). e SLK site has been considered to be a
suitable site for planting industry, owing to its accumulated temperature of ≥ 0°C throughout the year. ere are
approximately 186 frost-free days throughout the year when the daily minimum temperature is > 0°C; these occur
from April 16 to October 19. In the past, the dominant vegetation was temperate forest and grassland; now, the
vegetation is mainly cultivated plants, including spring wheat, broad bean and pea, and deciduous broadleaved
trees such as poplar and willow (Table2). e SLK site was rst investigated and excavated in detail by the Insti-
tute of Cultural Relics and Archaeology of Qinghai Province. e Institute of Cultural Relics and Archaeology
of Qinghai Province, which has carried out archaeological investigation, and found the area is approximately
24,000 m2, with complex cultural properties and continuous cultural sequences.
e CXGS site presents an extreme environment with low temperatures all year round. e winter features
with deep and long term snow cover. In addition, severe disasters from snow and ice are prone to occur in the
area. When compared with the CXGS site, the SLK site has much better conditions in the winter. ere are no
prominent meteorological disasters, making it a more suitable site to spend the winter time. However, there is a
little dierence in precipitation between the two sites.
Sampling and dating. To obtain detailed ages of the prehistoric human activities in the CXGS and SLK
sites, based on previous reports43,44, six pieces of charcoal were collected at depths of 21cm, 45–50cm, 65–70cm,
75–80cm, 85–90cm, and 95cm in the stratum of quadrats of the CXGS site, and nine pieces of charcoal were
collected at depths of 98cm, 183cm, 250cm, 275cm, 297cm, 310cm, 317cm, 321cm, 348cm, 393cm, and
408cm in the stratum of quadrats of the SLK site (Fig.1b, Table1). e charcoal samples collected from the
above two sites were sent to the Quaternary Dating Laboratory of Peking University for AMS 14C dating. e
AMS 14C dates were further converted to calendar year values by applying the IntCal 13 Calibration Curves using
the Calib 6.1.0 program45,46.
Data. e data used in this study were obtained mostly from the national geoscience data sharing platform
(https ://www.geoda ta.cn), including the scope and boundary of the QTP47, China Digital Elevation Model
(DEM) of 1km resolution (2000), China 1:250,000 data sets for hydrographic net classication of grades 1, 3, 4
and 5 (2002), China lake database of 1:100,000 (2000). Data regarding the accumulated temperature ≥ 0°C used
the database of the data level of temperature and humidity, which serves as the background level of Chinese
ecological environment. is work was originally completed by the Institute of Agricultural Natural Resources
and Agricultural Regionalization of Chinese Academy of Agricultural Sciences.
Modeling the relationship between traveling distance and time. ere were no domesticated ani-
mals during the period of time covered in this study, so there was no reliable animal power used during migra-
tion. As a result, travel relied solely on walking. Studies have shown that people will adjust their speed according
to slope of the route in order to expend the lowest amount of energy on travel. us the size of a slope is a decisive
factor for travel speed48,49. Traveling speed (v) and slope (p) have the following relationship50:
en the time “t” required for traveling 1m can be obtained from the following equation:
(1)
V=5.1 ·e
−2.25
|
tg
(
p·π/180
)
+0.07
|
(2)
t=3600/(1000 ·V)
Table 2. Meteorological and industrial comparison between the CXGS and SLK sites. e CXGS and SLK sites
data were from Zhiduo and Jianzha county meteorological stations whose distance were about 16 and 8km
during, respectively, 1960 and 1980.
Socioeconomic indicators CXGS SLK
Altitude (m a.s.l) 4,030 2076
Mean annual temperature (°C) −1.7 7.7
Mean January temperature (°C) −12.6 −6.3
≥ 0°C accumulated temperature (°C) 1,009 3,296
Mean annual precipitation (mm) 387 354
Mean January precipitation (mm) 1.8 2.1
Frost-free period of the year (day) 20 186
Mean annual snow depth (cm) 20 5
Mean annual snow days (day) 83 18
Industry Planting industry Animal husbandry
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2020) 10:11659 | 
www.nature.com/scientificreports/
In Arc GIS 10.3, slope analysis was used to convert DEM of 1km resolution of plateau into slope. In this
study, raster calculator and formulae (1) and (2) were used to make cost raster data “t” for traveling distance in
one second, taking CXGS site and “t” as the target layer and cost raster. e cost distance tool was then used to
calculate the raster for the relationship between time and distance. Given this formula, the time spent going from
the margin of the plateau to the inner of the plateau can be obtained. Similarly, the travel time from the margin
of the plateau to the CXGS site can also be obtained, which was the cumulative time spent traveling from the
nearest margin of the plateau to the CXGS site, in unit of “s”.
e key issue here is how much time did hunter-gatherers spend traveling each day? By referring to past
historical records of traveling on the plateau (occurred between A.D. 1,620–1937, and each travel period was
within one year)51–53, we attempted to answer this question. Specically, the historical records counted the starting
place, destination, route and time spent by past plateau travelers (Table3). It is worth noting that, in comparison
with the microlithic hunter-gatherers, past travelers using animal-powered travel had abundance of supplies for
their journey and typically travelled with a clear direction. Contrastingly, constant search for food during the
migration plays an important role in the travel of microlithic hunter-gatherers. As a result, unnecessary travel
(in terms of their nal destination) would be a large by-product of their nal route. Previous traveler records
indicate that a hunter’s eective traveling time was approximately 4h per day. In this paper, eective traveling
time refers to the time from the starting place to the destination of traveling, without counting non-traveling
time for other purposes. On this basis, the model of relationship between traveling distance and time was used to
calculate the time from starting place to destination of the past travelers (Table3). e comparison of historical
records and simulation results showed that actual traveling time of past travelers was more than the simulated
time, implying the average eective traveling time of past travelers on the plateau was no more than 4h per day,
with most traveling occurring in spring, summer, and autumn, while little traveling in winter. us, according
to the above discussion, we assumed that there were seasonal camps for hunter-gatherers. ey marched from
winter camps to spring camps from April to May of each year, and were active at high elevation summer camps
from June to August; and then they returned back to winter camps from September to October, and spend winter
at low elevation winter camps from November to the next March. Overall, the time for departure and return was
approximately 60days, respectively each year, and the eective traveling time for daily was 4h.
Constructed EI. To better elucidate the comprehensive natural environment conditions of the QTP, this
study selected the vegetation type, elevation, river classication (river and lake), accumulated temperature
of ≥ 0°C, and longitude indicators as geographical and environmental factors based on the activity character-
istics of the microlithic hunter-gatherers (Table4). Using the reclassication and raster calculator tools in the
spatial analysis of Arc GIS, we constructed an EI of the QTP to comprehensively characterize the natural envi-
ronment of the QTP. Using this approach, a higher index would indicate better environmental conditions that
are more suitable for human survival and living. e specic construction method is as follows:
Previous studies have shown that a mixture of forest-steppe was the dominated vegetation in which hunter-
gatherers existed, followed by steppe54. Due to the vegetation density (forest) or relative lack of vegetation
(desert), we concluded the sites that would not be conducive to human activities. e former would be dicult for
prehistoric humans to conduct productive activities and transportation, while the latter would be more dicult
to provide the necessary resources to support human societies owing to its inherent low biological productivity.
Accordingly, the highest evaluation of temperature forests and steppe was 9, each type of temperature steppe was
approximately 6–8, broad-leaved and coniferous forests were approximately 4–7, shrub was 6, meadow vegeta-
tion was 5, alpine vegetation and desert were approximately 1–2, and swamp and non-vegetation regions were 0.
In general, an elevation of 1,600–2,400m a.s.l may result in a human response to the relatively low atmos-
pheric oxygen. With an elevation above 3,000–3,600m a.s.l, humans would have obvious signs of hypoxia.
However, if humans are in an adaptive elevation ladder and have enough time, they will gradually adapt to these
Table 3. e list of the traveling time spent by past travelers of the plateau. ese events occurred in A.D.
1,620–1937, and each travel period was within 1 year.
Figure and event Place of departure Place of arrival Traveling route Travel time (days) Simulated time (days) References
Ma hetian investigates in Yushu Xining Yushu Tang-Tibet ancient road 41 40.6 43
Missionary, Jean Grueber, into Tibet Xining Lhasa Xining-Qaidam Basin-Lhasa 90 89.2 44
Missionary, Jean Grueber, out of Tibet Lhasa Nyalam Not specied 30 23 44
Missionary, Francois Marley, into
Tibet Kathmandu Lhasa Passing through Nyalam -Dingri 60 34 44
Missionary, Dominic, into Tibet Kathmandu Lhasa Passing through Nyalam 57 34 44
Missionary, Dominic, out of Tibet Lhasa Kathmandu Passing through Nyalam 45 34 44
Missionary, Desideri, into Tibet Leh Lhasa Passing through Qiangtang 223 96 44
Missionary, Desideri, into Tibet Saga Lhasa Not specied 49 42 44
Missionary, Pedro Cabral, into Tibet Bhutan Shigatse Not specied 33 12.5 44
e Northern route of Qing army’s
entering Tibet Xining Lhasa Passing through the sources of the
Yellow River and the Yangtze River 135 89.2 45
e northern route of Qing army’s
entering Tibet Chengdu Lhasa Passing through Litang, Batang,
Changdu 122 102 45
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2020) 10:11659 | 
www.nature.com/scientificreports/
hypoxic plateau conditions. When the elevation is more than 4,500m a.s.l and the atmospheric pressure is close
to one-half of the sea level, humans will suer from obvious hypoxemia and signicant, negative physiological
responses. When the elevation is higher than 5,500m a.s.l, humans will suer from a severe decline in function,
with some damage being irreparable55. According to elevation, values from 0–9 were evaluated (Table4), namely,
elevation classication. With this ranking, lower evaluation numbers indicate elevations that are more dicult
for humans to adapt to and survive in.
Rivers were divided into four grades 1, 3, 4 and 5 with each river used as a buer area according to 5km or
10km distance from the main river. In general, the higher the grade is, the higher the evaluation of the region
near the river.
e accumulated temperature of ≥ 0°C plays an important role in characterizing the climate of the plateau.
Specically, plateau regions with accumulated temperatures ≥ 6,500°C belong to tropical zones, < 500°C belong
to frigid zones, and approximately 1,500–4,200°C represent temperate zones. At present, temperature crops are
grown in areas of ≥ 2000°C, while chimonophilous crops are exclusively grown in areas of < 2000°C56. Accord-
ingly, the higher the accumulated temperature is, the higher the evaluation is.
Longitude is an important factor for prehistoric human to spread on the plateau. Since Last Glacial Maximum
(LGM), the general direction of prehistoric human expansion to the QTP was a process going from east to west
with the characteristics that the eastern sites have the early age results and the western sites have the late age
results. e reasons for this expansion pattern are as follows. First, some scholars have proposed a “three-step
jumping” model for human beings on their march into the plateau57. In this model, the rst step was the LGM
with an age range of 24.0–16.0cal. ka BP. During this time, human activities occurred in the grassland at an
elevation below 3,000m. e second step was from 15.0–11.2cal. ka BP during the Last Deglacial period, when
hunter-gatherers expanded to regions with elevations of 3,000–4,000m a.s.l. During this expansion, they le
behind short-term camp sites, which were used to search for small- and medium-sized animals. e third and
nal step occurred during the early and middle Holocene when humans spread to high elevation regions above
4,000m a.s.l. Second, genetic studies have revealed that approximately 98% of the maternal genetic component
of modern Tibetans can be traced back to the northern Chinese who prehistorically moved into the QTP6,58.
Tibetans have been recognized as Mongolians from East Asia, so there should be a view that people in northern
Table 4. Classication and evaluation of geographical factors. Due to the diversity of vegetation types, only
typical representatives were selected for “Typical types of vegetation” column heading. “River” Column
heading represents grades 1, 3, 4, and 5 rivers, respectively; 5km, 7.5km, and 10km represent buer areas
0–5km, 5–7.5km, and 5–10km from river, respectively.
Typical types of
vegetation Preset value Altitude (m) Preset value River Preset value ≥ 0°C accumulated
temperature (°C) Preset value L ongitude (°E) Preset value
Temperate deciduous
shruband, etc 9 < 1,600 9 Grade 1, 5km 9 ≥ 6,500 9 101.8–105.0 9
Temperate grass and
forb meadow, etc 8 1,600–2000 8 Grade 1, 7.5km 8 5,300–6,500 8 98.6–101.8 8
Temperate coniferous
forests, subtropical
and tropical mountain
coniferous forests,
temperate deciduous
small-leaf forests, etc
7 2000–2,400 7 Grade 1, 10km 7 4,200–5,300 7 95.4–98.6 7
Subtropical coniferous
forests, broadleaf
mixed forests, temper-
ate deciduous broad-
leaf forests, etc
6 2,400–2,800 6 Grade 3, 5km 6 3,500–4,200 6 92.2–95.4 6
Cold-temperate and
temperate mountain
coniferous forests,
alpine meadow, etc
5 2,800–3,200 5 Grade 3, 10km 5 2000–3,500 5 89.0–92.2 5
Subtropical evergreen
broadleaf forests, etc 4 3,200–3,600 4 Grade 4, 5km 4 1,500–2000 4 85.8–89.0 4
Subtropical evergreen
broadleaf forests, etc 3 3,600–4,100 3 Grade 4, 10km 3 1,000–1,500 3 82.6–85.8 3
Tropical rainforests,
etc 2 4,100–4,600 2 Grade 5, 5km 2 800–1,000 2 79.4–82.6 2
Dwarf trees desert,
shrub desert, cushion
subshrub Alpine
desert, alpine dwarf
semi-shrub alpine
desert, alpine sparse
vegetation, etc
1 4,600–5,500 1 Grade 5, 10km 1 500–800 1 76.2–79.4 1
Alpine bog, desert,
bare land, snow-
capped land, saline
soil, etc
0 > 5,500 0 > 10km 0 < 500 0 73.0–76.2 0
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2020) 10:11659 | 
www.nature.com/scientificreports/
China are expanded from east to west across the plateau58. However, the possibility of people originating from
other directions and entering the plateau cannot be excluded, owing to the limitation of scale and dierences
in what time they may have entered the plateau59. Given this, these complexities have been omitted from the
current discussion. us, we can conclude that the higher the longitude, the earlier the prehistoric age of human
activity, and the higher the evaluation value.
Based on the above classication criteria and their application to the Analytic Hierarchy Process (AHP), each
factor was evaluated, weighted, and the following relationship was constructed:
In the formula (3), “I” represents the plateau EI, “H” is the elevation classication, “R” is the river classi-
cation, “P” is the vegetation classication, and “T” is classication of accumulated temperature of ≥ 0°C. e
higher the EI is, the more suitable it is for human survival and occupation. It should be pointed out that we used
modern geographical factors to construct this EI, which are—to some extent—dierent from past environmental
and geographical factors, but the changes of the natural environment are systematic and continuous, and the
present natural environment is the inheritance of historical evolution, therefore the present natural environ-
ment has inseparable connections to its past form. us, the simulation used in this study plays an important
role in such work.
Data availability
All data needed for the evaluation of this paper are present in the paper or in supplementary information.
Received: 5 February 2020; Accepted: 25 June 2020
References
1. Hou, G. L. et al. e earliest prehistoric pottery in the Qinghai-Tibetan Plateau and its archaeological implications. Quat. Geo-
chronol. 30, 431–437 (2015).
2. Osorio, D. et al. Hunter-gatherer mobility strategies in the high Andes of Northern Chile during the Late Pleistocene-Early Holo-
cene transition (ca. 11,500–9500 CAL BP). J. Field Archaeol. 42, 228–240 (2017).
3. Brantingham, P. J. et al. Peopling of the northern Tibetan plateau. World Archaeol. 38, 387–414 (2006).
4. An, Z. S. et al. Interplay between the Westerlies and Asian monsoon recorded in Lake Qinghai sediments since 32 ka. Sci. Rep. 2,
619 (2012).
5. Shi, P. J. et al. Factors contribution to oxygen concentration in Qinghai-Tibetan Plateau. Chin. Sci. Bull. 64, 715–724 (2019) (in
Chinese).
6. Zhang, D. J. et al. History and possible mechanisms of prehistoric human migration to the Tibetan Plateau. Sci. China Earth Sci.
59, 1765–1778 (2016).
7. Chen, F. H. et al. A late Middle Pleistocene Denisovan mandible from the Tibetan Plateau. Nature 569, 409–412 (2019).
8. Zhang, X. L. et al. e earliest human occupation of the high-altitude Tibetan Plateau 40 thousand to 30 thousand years ago. Sci-
ence 362, 1049–1051 (2018).
9. Gao, J. Y. et al. Prehistoric human activity and its environmental background in Lake Donggi Cona basin, northeastern Tibetan
Plateau. e Holocene 30, 657–671 (2020).
10. Wei, H. C. et al. Using coprophilous fungi to reconstruct the history of pastoralism in the Qinghai Lake Basin, Northeastern
Qinghai-Tibetan Plateau. Prog. Phys. Geophys. 44, 70–93 (2020).
11. Zhang, D. J. et al. Comment on “Permanent human occupation of the central Tibetan Plateau in the early Holocene”. Science 357,
13 (2017).
12. D’Alpoim Guedes, J., Bocinsky, K. R. & Butler, E. E. Comment on Agriculture facilitated permanent human occupation of the
Tibetan Plateau aer 3600 B.P. Science 348, 872 (2015).
13. Chen, F. H. et al. Agriculture facilitated permanent human occupation of the Tibetan Plateau aer 3600 BP. Science 347, 248–250
(2015).
14. Meyer, M. C. et al. Permanent human occupation of the central Tibetan Plateau in the Early Holocene. Science 355, 64–67 (2017).
15. Wang, J. et al. Subsistence strategies of prehistoric hunter-gatherers on the Tibetan Plateau during the Last Deglaciation. Sci. China
Earth Sci. 63, 395–404(2019).
16. Madsen, D. B. et al. e late upper Paleolithic occupation of the northern Tibetan Plateau margin. J. Archaeol. Sci. 33, 1433–1444
(2006).
17. Gai, P. & Wang, G. D. Excavation report on a Mesolithic site at Layihai, Upper Yellow River. Acta Anthropol. Sin. 2, 49–59 (1983)
(in Chinese).
18. Yi, M. J. et al. A preliminary report on investigations in 2009 of some prehistoric sites in the Tibetan Plateau Marginal Region.
Acta Anthropol. Sin. 30, 124–136 (2011) (in Chinese).
19. Rhode, D. et al. Mind the gaps: Testing for hiatuses in regional radiocarbon date sequences. J. Archaeol. Sci. 52, 567–577 (2014).
20. Kong, D. D. et al. Vegetation phenology change in Tibetan Plateau from 1982 to 2013 and its related eteorological factors. Acta
Geogr. Sin. 72, 39–52 (2017) ( in Chinese).
21. Huang, X. Z. et al. Long-term herbivore population dynamics in the northeastern Qinghai-Tibetan Plateau and its implications for
early human impacts (Palaeobot. Palyno, Rev, 2020). https ://doi.org/10.1016/j.revpa lbo.2020.10417 1.
22. Wu, T. Y. & Kayser, B. High altitude adaptation in Tibetans. High Alt. Med. Biol. 7, 193–208 (2006).
23. Ren, L. L. et al. Foraging and farming: Archaeobotanical and zooarchaeological evidence for Neolithic exchange on the Tibetan
Plateau. Antiquity 94, 1–16. https ://doi.org/10.15184 /aqy.2020.35 (2020).
24. Xie, D. J. Prehistorical archaeology of Gansu and Qinghai provinces, China (Cultural Relices Press, Beijing, 2002).
25. Chen, H. H., Ge, S. B. & Li, G. L. Discuss to the nature of Zongri relic. Archaeol. 19, 23–27 (1998) (in Chinese).
26. Chen, H. H., Ge, S. B. & Wang, G. S. Investigation of Zongri cultural site in the Hainan Prefecture, Qinghai Province. Western
Archaeol. 2, 16–40 (2007) (in Chinese).
27. Huang, X. Z. et al. Early human impacts on vegetation on the northeastern Qinghai-Tibetan Plateau during the middle to late
Holocene. Prog. Phys. Geoph. 41, 286–301 (2017).
28. Brantingham, P. J. & Gao, X. Peopling of the northern Tibetan plateau. World Archaeol. 38, 387–414 (2006).
29. Dong, G. H. et al. Agricultural intensication and its impact on environment during Neolithic Age in northern China. Chin. Sci.
Bull. 61, 2913–2925 (2016).
(3)
I=0.3 ∗H+0.21 ∗R+0.245 ∗P+0.209 ∗T
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2020) 10:11659 | 
www.nature.com/scientificreports/
30. Dong, G. H. et al. Evolution of human–environmental interactions in China from the Late Paleolithic to the Bronze Age. Prog.
Phys. Geophys. 10, 2. https ://doi.org/10.1177/03091 33319 87680 2 (2019).
31 . Li, X. Q. et al. Holocene agriculture in the Guanzhong Basin in NW China indicated by pollen and charcoal evidence. e Holocene
19, 1213–1220 (2009).
32. Yi, M. J., Gao, X., Li, F. & Chen, F. Y. Rethinking the origin of microblade technology: A chronological and ecological perspective.
Quat. Int. 400, 130–139 (2016).
33. Zhang, D. J. et al. Archaeological records of Dadiwan in the past 60 ka and the origin of millet agriculture. Chin. Sci. Bull. 55,
883–890 (2010).
34. Rhode, D. et al. Epipaleolithic/early Neolithic settlements at Qinghai Lake, western China. J. Archaeol. Sci. 34, 600–612 (2007).
35. Tang, H. S., Zhou, C. L., Li, Y. Q. & Liang, Z. A new discovery of microlithic information at the entrance to the northern Qinghai-
Tibet Plateau of the Kunlun Mountains of Qinghai. Chin. Sci. Bull. 58, 247–253 (2013) (in Chinese).
36. Herzschuh, U. et al. A modern pollen-climate calibration set based on lake sediments from the Tibetan Plateau and its application
to a Late Quaternary pollen record from the Qilian Mountains. J. Biogeogr. 37, 752–766 (2010).
37. Shen, C. M., Liu, K. B., Tang, L. Y. & Overpeck, J. T. Quantitative relationships between modern pollen rain and climate in the
Tibetan Plateau. Rev. Palaeobot. Palynol. 140, 61–77 (2006).
38 . Li, J. Y. et al. Quantitative precipitation estimates for the northeastern Qinghai-Tibetan Plateau over the last 18,000years. J. Geophys.
Res. Atmos. 122, 429–435 (2017).
39. Chen, F. H. et al. Holocene moisture and East Asian summer monsoon evolution in the northeastern Tibetan Plateau recorded by
Lake Qinghai and its environs: A review of conicting proxies. Quat. Sci. Rev. 154, 111–129 (2016).
40. ChongYi, E. et al. High resolution OSL dating of Aeolian activity at Qinghai Lake, Northeast Tibetan Plateau. Catena 183, 104180
(2019).
41. Lu, R. J. et al. Holocene aeolian activity and climatic change in Qinghai Lake basin, northeastern Qinghai-Tibetan Plateau. Pal-
aeogeogr. Palaeoclimatol. Palaeoecol. 430, 1–10 (2015).
42. Han, F. et al. A study on the microblade technology from the Deng’equ valley in South Qinghai Plateau. Acta Anthropol. Sin. 37,
53–69 (2018) (in Chinese).
43. Dong, G. H. et al. Spatial and temporal variety of prehistoric human settlement and its inuencing factors in the upper Yellow
River valley, Qinghai Province China. J. Archaeol. Sci. 40, 2538–2546 (2013).
44. He, Y. H., Cai, L. H., Li, P. & Du, W. Trace of the early ancestors of the southern Qinghai Tibet Plateau: A surney of the microlithic
remains in Tongtian River Basin. China Cult. Relics Newsp. 8, 21–27 (2014) (in Chinese).
45. Reimer, P. J. et al. Calib radiocarbon calibration program. Radiocarbon 51, 1111–1150 (2009).
46. Stuiver, M. & Braziunas, T. F. Modeling atmospheric 14C inuences and 14C ages of marine samples to 10,000 BC. Radiocarbon 35,
137–189 (1993).
47. Zhang, Y. L., Li, B. Y. & Zheng, D. A discussion on the boundary and area of the Tibetan Plateau in China. Geogr. Res. 21, 1–8
(2002) (in Chinese).
48. Magyarisáska, Z. & Stefan, D. Determining minimum hiking time using DEM. Geoge. Inform. Syst. 4, 124–129 (2012).
49. Wang, L. & Wu, H. DEM based quantitative site catchment denition and statistics on retrievable farmlands. Remot. Sens. Technol.
Appl. 32, 281–288 (2017) (in Chinese).
50. Pingel, T. J. Modeling slope as a contributor to route selection in mountainous areas. Carto. Geogr. Inform. Sci. 37, 137–148 (2010).
51. Ma, H. T. e investigation of Gansu Qinghai Tibet region (Gansu People’s Publishing, Housing, 2003) (in Chinese).
52. Wu, K. M. History of the activities of early missionaries in Tibet (China Tibetology Press, Beijing, 1992) (in Chinese).
53. Deng, R. L. e passage of the Qing army into Tibet in 1720. Ethnic Study 1, 85–91 (2000) (in Chinese).
54. Chen, S. Q. Modeling hunter–gatherers of China. Acta Anthropol. Sin. 25, 42–55 (2006) (in Chinese).
55. Wu, T. Y. Challenges to human being under hypoxia environment on plateau. J. Med. Res. 35, 1–3 (2006) (in Chinese).
56. Dai, J. X. e Climate of Tibet Plateau (China Meteorological Press, Beijing, 1990) (in Chinese).
57. Brantingham, P. J. et al. A short chronology for the peopling of the Tibetan Plateau. Dev. Quat. Sci. 9, 129–150 (2007).
58. Zhao, M. et al. Mitochondrial genome evidence reveals successful Late Paleolithic settlement on the Tibetan Plateau. Proc. Natl
Acad. Sci. USA 106, 21230–21235 (2009).
59. Zhang, D. J., Xia, H. & Chen, F. H. Early human occupation of the Tibetan Plateau. Sci. Bull. 63, 1598–1600 (2018).
Acknowledgements
is research was supported by the National Natural Science Foundation of China (Grant no. 41761018), the
Strategic Priority Research Program of the Chinese Academy of Sciences, Pan-ird Pole Environment Study for
a Green Silk Road (Pan-TPE) (Grant no. XDA2004010101) and the Second Tibetan Plateau Scientic Expedi-
tion and Research Program(STEP) (Grant no. 2019QZKK0601). We thank Qingbo Wang, Long Yang, Fangfang
Wang and Xiaoqing Hou for their help in the eld. We are grateful for the correcting the English and constructive
suggestions provided by two anonymous reviewers, editors, Rong Jiang and Zhuoma Wende who greatly helped
improve this manuscript.
Author contributions
G.H. designed the study and supervised the eld work. G.H., J.G., Y.X., L.C. and Y.H. conducted the archaeologi-
cal investigations. Y.C. and Y.H. analyzed the stone artifacts. G.H., C.X. and Z.L. analyzed the data. G.H. draed
the initial manuscript, which was revised by G.H. and J.G. All authors approved the nal version.
Competing interests
e authors declare no competing interests.
Additional information
Supplementary information is available for this paper at https ://doi.org/10.1038/s4159 8-020-68518 -w.
Correspondence and requests for materials should be addressed to G.H.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2020) 10:11659 | 
www.nature.com/scientificreports/
Open Access is 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. e 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 http://creat iveco mmons .org/licen ses/by/4.0/.
© e Author(s) 2020
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com