Sage Journals: Discover world-class research

Abstract

Migration is a frequent phenomenon in human history. Previous studies mainly used migration as a simple process to account for any cultural changes observed in migrant communities. Recent studies, however, have recognized that migration is embedded in both environmental and social contexts, thus making it necessary to study the consequence of migration on a case-by-case basis. To better understand the changes associated with migrational processes, this case study investigates the subsistence pattern of a Neolithic site (5300-4700 cal. BP), Liujiazhai, in Northwestern Sichuan Highland by employing microbotanical residue analysis on pottery vessels. Our results on millet phytolith contribute to the overall picture of millet agriculture in Yangshao and Majiayao migrant communities and enrich our understanding of how varying crop patterns in Neolithic western China are likely a consequence of migration. In addition, we suggest that Liujiazhai migrants adapted to the high-altitude environment by utilizing more local wild plant resources. This study shows that, although Yangshao migrants were still connected to their homeland in terms of material culture, relocating to the challenging environment in NW Sichuan requires adaptive strategies that diverged the Yangshao migrants, including Liujiazhai, from their home culture. Hence, this study exemplifies how migration is an agent of change.

Keywords

agriculture human-environment relation middle Holocene migration millet residue analysis western China

Introduction

Migration is an essential part of human history. From the dispersal of Neolithic cultures to the diaspora of ethnic communities, migration is such an omnipresent phenomenon in the archeological record Defined as “one-way residential relocation to a different ‘environment’ by at least one individual,” migration is more than the physical movement of people, but it entails a profound change evident in biology, socio-culture, and linguistics (Cabana and Clark, 2011). Thus, many have argued for a more rigorous framework to explore the explanatory power of migration and better understand cultural change as its consequence (Adams et al., 1978; Anthony, 1990, 1997; Burmeister, 2000; Champion, 1990). Studies since then have not only recognized that migration is an inherent social strategy (Harper et al., 2019; Hofmann, 2020; Leppard, 2014), but have also found that the impact of migration is multifold – in many cases, migration leads to adaptive changes in subsistence, agriculture, and technologies, and stimulates a stronger expression of identity, whereas in other cases it produces little impact (Clark, 2011; Conolly et al., 2008; Cruz Berrocal, 2012; Ivanova et al., 2018; Kreuz et al., 2005; Mazzucco et al., 2020). Burmeister (2000) proposed that migrants often form a new cultural identity that is distinct of both the home culture and the local culture. Nevertheless, Silliman (2005) cautioned that archeological studies shouldn’t take material culture as the direct representation of cultural identity, adding complexity to migration studies. Without doubt, migration is an agent of change, but the scope and extent of change is context specific. This thinking has been adopted by many postprocessual archeologists who argue for a case-by-case examination on migration because any changes in migrants’ culture are influenced by specific social and environmental contexts that they encounter during migration (Cabana, 2011).

Despite recent progress on migration studies, van Dommelen (2014) reckoned that there still exists a strong concern with finding the “hard” evidence of large population migration, and that “an archaeological understanding of migration as a ‘multilayered process’ is practically non-existent” (p. 479). To better articulate the consequence and the various dimensions of migration, we present a case study of a Neolithic site, Liujiazhai (5300–4700 cal. BP), located in the Northwest Sichuan highland in China. By employing microbotanical residue analysis on pottery vessels, we investigate how agriculture intersected with new landscape and led to changes in crop pattern and adaptations in subsistence strategies. Such study showcases how migrant society evolved in a new environmental setting as they moved further away from their homeland, also adding more information to our understanding of human-environment interaction in the Tibetan Plateau.

Archeological background

The Northwest Sichuan Highland is situated on the eastern margin of the Tibetan Plateau. The initial peopling of this extreme environment has received much attention as to understand the capacity of human adaption as well as the history of the Sino-Tibetan family (Aldenderfer, 2011; d’Alpoim Guedes and Aldenderfer, 2020; Lu, 2023). Upper Paleolithic sites (Piluo, Fulin, Shizishan, Yanyundong, etc.) found in western Sichuan show that a foraging population had already existed before the Neolithic (Shi, 2004; Sichuan Provincial Institute of Cultural Relics and Archaeology, 2021a; Sichuan Provincial Institute of Cultural Relics and Archaeology and School of Archaeology and Museology in Peking University, 2022). However, demography in western Sichuan during the Neolithic is much more complex. Both phylogenic and linguistic studies suggest that the modern-day Tibetans are rather a mix of a local population and a migrant population, possibly the Yangshao culture from the Yellow River valley, dating to 7800–4200 BP (He et al., 2021; Ning et al., 2020; Zhang et al., 2018; Zhao et al., 2009). In addition, genetic studies have found that low-altitude populations (the migrants) acquired adaptive alleles from the local highlanders, attesting to potential interactions between the two groups (He and Ding, 2013; Jeong et al., 2014).

The interactions between farming communities and the local foraging population still require more research, and the geographical location of NW Sichuan Highland offers a great opportunity to conduct such an investigation. So far, many middle and late Neolithic sites have been found in NW Sichuan. Among those sites, the Liujiazhai site (AMS date 5300–4700 cal. BP) is located in the upper reaches of the Dadu River, at an elevation of approximately 2630 m above sea level (Figure 1) (Sichuan Provincial Institute of Cultural Relics and Archaeology, 2021b; Sichuan Institute herein). The macrobotantical record at Liujiazhai consists mostly of broomcorn millet (Panicum miliaceum) and foxtail millet (Setaria italica), as well as chenopods and Asteraceae seeds (Sichuan Provincial Institute of Cultural Relics and Archaeology, 2022). While domesticated plants are present, animal remains only have a limited number of pig bones (wild or domesticated unclear) and are mainly from wild animals (Sichuan Institute, 2022).

Figure 1.

Map showing the locations of sites mentioned in this study and major rivers. Gray dashed lines mark the boundary of Qinghai, Sichuan, Gansu, and Shaanxi provinces. Arrows show possible migration routes suggested by Zhu (2019).

Migration background

It is proposed that the Neolithic migrants in NW Sichuan came from the Yangshao culture that flourished in the Yellow River valley beginning 7000 cal. BP. After broomcorn and foxtail millet were domesticated in northern China around 10,000 and 8000 cal. BP respectively (Lu et al., 2009b), Yangshao culture adopted dryland millet farming since the early phase (7000–6000 cal. BP) and involved millets in their cultural practices such as fermenting millets in amphoras to make alcoholic beverages (Liu, 2021; Liu et al, 2020). Red and painted pottery are distinct characteristics of Yangshao pottery style, and it became a symbol of cultural identity when Yangshao communities migrated out of the Yellow River valley. Starting 6000 cal. BP, the Yangshao migrants moved westward, settling in where is now Gansu and Qinghai, and formed new cultural variants such as the Shilingxia transitional phrase (5900–5200 cal. BP) and the Majiayao culture (5300–4500 cal. BP) (Ding 2010). Meanwhile, another wave of migration saw the emergence of Yangshao cultural sites in NW Sichuan (Figure 1), including Boxi and Yingpanshan (5300–4600 cal. BP) in the upper reaches of Min river, and Haxiu (5500–4700 cal. BP) and Liujiazhai (5300–4700 cal. BP) in the upper reaches of Dadu River (He, 2015; Jiang and Chen, 2003; Zhao and Chen, 2011). Though the radiocarbon date is not available for the Boxi site, it is dated to the middle Yangshao period, or the Miaodigou phase (6000– 4700 BP), based on its material culture (Jiang and Chen, 2003). Painted pottery and amphoras are ubiquitous in the material assemblages from these sites, although the majority are not painted pottery (He, 2015; Jiang and Chen, 2003; Ren, 2015). The provenance of those painted pottery vessels received much debate as it pertains to the development of pottery traditions, and potentially cultural identity, of the Yangshao and Majiayao migrants. Chemical analysis suggests that painted potteries in NW Sichuan were produced in Gansu and Qinghai, possibly acquired via trade, whereas the unpainted ones were made locally (Cui et al., 2011; Hung et al., 2010). Chen (2009), Ren et al. (2013), and Ren and Chen (2022), on the other hand, argued that vessel forms and decoration styles of some painted pottery from NW Sichuan are not found in Gansu and Qinghai, thus a local production of painted pottery remains likely. While the origin of painted pottery is open to future research, a cultural tie between the Yangshao/Majiayao culture and the NW Sichuan sites is still evident. Both macrobotanical record and starch residues show that the Yangshao/Majiayao migrants at Boxi, Haxiu, and Yingpanshan continued practicing millet cultivation (Liu et al., 2022; Sichuan Institute, 2022; Zhao and Chen, 2011). In addition, residue analyses on pottery samples from Boxi and Haxiu suggest they retained cultural traditions such as alcoholic fermentation and drinking rituals (Liu et al., 2022).

Environmental background in NW Sichuan

The Holocene amelioration provided foraging communities with the conditions to transition toward sedentism and agriculture. However, periodic climatic variations still occurred in the Holocene. A humid and warm climate from the early Holocene allowed forests to emerge and dominate the landscape in NW Sichuan. This trend continued until the Mid-Holocene optimum ended around 6000 cal. BP (An et al., 2004; Ran and Feng, 2013). According to paleoclimatic reconstruction, the Mid-Holocene precipitation in NW Sichuan was about 500–800 mm, and the average temperature was 2°C higher than today, categorizing a semi-arid environment (Wen et al., 2017; Zhou et al., 2021). Beginning in 5500–5300 cal. BP, the East Asia Summer Monsoon weakened, causing a series of hydrological recessions and declined seasonal precipitation in both Yellow River Valley and western China (An et al., 2004; Chen et al., 2015a; Huang et al., 2019; Jarvis, 1993; Zhang et al., 2023). This cold and dry period persisted until 4800 cal. BP when the climate became wetter again (Wen et al., 2017). The warm and wet Mid-Holocene period supported the Yangshao culture to expand, which is evidenced by the increasing settlement size (Gong, 2003). As a result, population growth and the subsequent population-resource imbalance could have played a role in motivating migration (Wu et al., 2018).

Materials and method

Field sampling

The field sampling took place at Sanxingdui Workstation in Sichuan, 2019, where all artifacts for this study were curated and stored (Figures 2 and 3). The vessel types and decoration styles of the 29 sherd samples are listed in Table 1 (full information of sampled sherds is in Supplemental Table S1, available online). We sampled 10 amphora sherds to determine if they had previously contained fermented beverages as it is a common practice in the Yangshao culture. To prevent contamination, we used new and sterile plastic bags, test tubes, pipettes, toothbrushes, and razor blades during the sampling process. First, all pottery samples were brushed and washed in running water to remove surface dust and particles. For small pottery sherds, we sonicated them in an ultrasonic bath, while larger sherds were brushed using an ultrasonic toothbrush. We placed each small sherd in a new polyvinyl bag with approximately 15 mL of distilled water and then placed each bag into the ultrasonic bath for 6 min. For larger sherds, we gently brushed the interior surface using an ultrasonic toothbrush while simultaneously adding distilled water using a pipette to rinse the surface. We used a new ultrasonic toothbrush and a new pipette for each sherd. After the ultrasonic bath or toothbrushing, each residue liquid sample was transferred to a new 15 mL test tube and then stored in a sealed plastic bag before laboratory analysis. In addition, we collected sediment samples from the exterior surfaces of Pot 1, Pot 5, Pot 10, and Pot 19 by using a clean razor blade, which served as control samples to compare with the residue samples obtained from the interior surface of these pots.

Figure 2.

Examples of sampled pottery sherds. (a) Pot 25, bottom of an unpainted basin. (b) Pot 4, rim of a painted amphora. (c) Pot 22, bottom of an unpainted jar with cord mark. (d) Pot 10, rim of an unpainted amphora. (e) Pot 9: body sherd of a painted bowl. (f) Pot 8, bottom of a painted basin. (g) Pot 24, rim of an unpainted amphora. (h) Pot 13, rim of an unpainted jar with cord mark. (i) Pot 19, bottom of an unpainted jar with cord mark.

Figure 3.

Examples of pottery types at Liujiazhai. (a) Jar with applique (2011TSE12:26). (b) Amphora (2011H116:107). (c) Basin (2011H116:106). (d) Bowl (2012 G1:34). (a–c) are from the 2011 Excavation Report (Sichuan Provincial Institute of Cultural Relics and Archaeology, 2021b) and (d) is from the 2012 Excavation Report (Sichuan Provincial Institute of Cultural Relics and Archaeology, 2022).

Table 1.

Quantity of sampled pottery sherds.

	Amphora (ping)	Bowl (bo)	Basin (pen)	Jar (guan)	Unidentified
Painted	6	1	2	1	1
Unpainted	4	2	1	11	0
Total quantity	10	3	3	12	1

Laboratory processing

The processing method followed the established method, as detailed in the Supporting Information, available online.

Starch and phytolith identification

All starch granules were counted and recorded. For phytolith, the entire slide is counted only when phytolith is fewer than 300. We identified starch and phytolith taxa by comparing them to our reference collection of more than 1000 economically important plant specimens from Asia and other published literature (Ge et al., 2018; Huan et al., 2015; Lu et al., 2009a; Zhang et al., 2011). Starch granules with cooking or heating damages were identified by comparing with the published literature (Babot, 2003; Henry et al., 2009; Wang et al., 2016). We followed the International Code for Phytolith Nomenclature 2.0 to describe phytolith morphotypes (Neumann et al., 2019).

Results

Starch residue analysis

In total, we found 351 starch granules, consisting of 272 identified starch granules and 79 unidentified ones (UNID), from 26 out of the 29 samples (Supplemental Table S2, available online; Figure 4). The identified starches (77.5% of the total count) are divided into eight types and detailed in the following (for modern starch reference and identification, see Supplemental Figure S1, available online). The UNID starch granules (22.5% of the total count) consist of gelatinized starch (n = 11), starches with unidentifiable damages (n = 19), unidentified starch (n = 44), and unidentified compound starches (n = 5). It is known that starch granules are subject to morphological alterations either caused by cooking or enzymatic damages. Gelatinization is the result of the heating process such as cooking, and it is characterized by swelling, distortion, and partial or complete loss of extinction cross (Williams et al., 1982). On the other hand, starches with unidentifiable damages show signs of cracks and pitting that result from various food processing and taphonomic processes (Takaya et al., 1978).

Figure 4.

Types of starch granules identified in this study. (a) Panicoideae. (b) Triticeae. (c) Job’s tears (Coix lacryma-jobi). (d) Unidentifiable USO. (e) Zingiberaceae (under DIC view). (f) Fern root. (g) Pueraria sp. (kudzu vine root). (h) Euryale ferox (foxnut). (i) Gelatinized. (j) UNID.

Type 1: Panicoideae, n = 108, size range: 5

69–16.839 μm; 30.8% of the total count; ubiquity = 0.66.

Type 1 starch granules are mostly spherical or polyhedral in shape, with hila located in the center and straight arms of extinction cross. Foxtail millet (Setaria italica) and broomcorn millet (Panicum miliaceum) most likely contribute to this type, with sizes ranging from 11.21 to 16.75 μm and 5.92 to 13 μm in our reference collection, respectively. The majority of Panicoideae starches (n = 65) are found in jars, with some are found in amphoras (n = 18) and bowls and basins (n = 18). Most of Type 1 starches (n = 83) are found in pure-clay cooking (jars) and serving vessels (bowls and basins), thus it is mainly associated with cooking and serving activities. Type 1 is the most abundant and ubiquitous type, which is consistent with the macrobotanical record of Liujizhai and millet cultivation in Yangshao culture.

Type 2: Triticeae, n = 69, size range: 17

704–31.3 μm; 19.7% of the total count; ubiquity = 0.62

Type 2 granules are lenticular or discoid in shape with a flat surface, with centric hila and visible lamella on some large granules, consistent with the Triticeae tribe. When the flat side of the granule is under observation, the extinction cross takes on a “+” shape, but when observed from the side, it becomes more like an “x.” The size of modern Triticeae granules can vary from 2.46 μm to larger than 20 μm. Agropyron, Elymus, and Leymus, which are commonly found in arid regions in China (Chen et al., 2013), are potential candidates for this type. Another possible source is Secale sp. (rye), as charred Secale sp. has been discovered at Yingpanshan (Zhao and Chen, 2011). However, due to the lack of more diagnostic features, the starch granules grouped here are referred to as Triticeae. The majority of Triticeae granules (n = 35) are found in jars with Panicoideae, while some (n = 18) are found in bowls and basins.

Type 3: Job’s tears (Coix lacryma-jobi), n = 47, size range: 17

759–25.322 μm; 13.4% of the total count; ubiquity = 0.48

Type 3 granules are spherical or polyhedral in shape, which are similar to Type 1 but show several distinctive features based on Liu et al. (2014)’s criteria for Job’s tears starch, which includes three features: (1) larger size, (2) higher eccentricity ratio of hilum, and (3) z-shape curved arms of extinction cross. Job’s tears typically grow in wetlands, distinguishing them from millets, which grow in different habitats. Although the history of Job’s tears cultivation or domestication remains unclear, its exploitation is evident in many nearby sites, including Haxiu, Boxi, Dadiwan, and the late Neolithic Baodun site in Sichuan (Guedes et al., 2013; Liu et al., 2022).

Type 4: Unidentifiable underground storage organs (USO), n = 39, size range: 8

415–15.45 μm; 11.4% of the total count; ubiquity = 0.48

Type 4 granules are round or elongate oval in shape with eccentric hila and bent arms of extinction cross, which are consistent with starches from USOs. USOs include tubers, roots, rhizomes, and bulbs, and their starch granules exhibit similar morphology but lack more diagnostic features for further identification. The relatively high ubiquity of USOs makes them an important food source at Liujiazhai, in addition to the previous three types.

Type 5: Zingiberaceae, n = 2, size range: 27

19–33.53 μm; 0.6% of the total count; ubiquity = 0.07.

Type 5 starch granules have elongate oval or nearly triangular shape, and it is generally larger than many other types. The hila of extinction cross is highly eccentric and is located extremely close to the edge. They show clear relief under the differential interference contrast (DIC) view, which is similar to the criterion set by Torrence and Barton (2006: Plate 27). In our modern reference collection, young ginger roots (Zingiber sp.) tend to have fan-shaped granules whereas mature ones have elongate granules. The starch granules grouped here generally match the description of ginger roots. The exploitation of ginger roots is evident at many sites in the Wei River valley (Liu, 2021; Wang et al., 2019), and it has long been used as medicine and food ingredients in China (Wang and Wang, 2005).

Type 6: Fern root, n = 2, size range: 11

545–30.071 μm; 0.6% of the total count; ubiquity = 0.07.

Type 6 starch granules exhibit various shapes, including round, semi-spherical, elongate oval, and facetted. The hila can be both centric and eccentric, and the extinction cross sometimes has bent or zig-zag arms. This type shows a high resemblance to the root of the eagle fern (Pteridium aquilinum) in our modern reference collection. Many fern species are distributed in Sichuan today, and their roots and leaves are frequently used as food and in medicine (Li, 1994). In addition, eagle fern is highly rich in proteins, carbohydrates, fats, and minerals (Qi et al., 2015). Thus ferns are great options for exploitation.

Type 7: Pueraria sp

(kudzu vine root), n = 2, size range: 9.631–15.105 μm; 0.6% of the total count; ubiquity = 0.07.

Type 7 starch granules are mostly facetted, with some sub-round or oval in shape. The hila is usually centric, and the extinction cross sometimes has uneven thickness with thicker arms toward the peripheries. This type is consistent with the morphological characteristics of kudzu vine granules in our modern reference collection. Kudzu produces edible starchy roots, and its fibrous vines have been used to make textiles and paper in ancient China. It is also a traditional medicine to treat fever and diarrhea (Wong et al., 2011).

Type 8: Euryale ferox (foxnut), n = 2, size range: 2

723–5.29 μm; 0.6% of the total count; ubiquity = 0.03.

Type 8 starch granules usually appear in a compound structure with a round or oval shape, while the individual starch granules are more polygonal. The hilum of the individual granule is mostly centric. This type is most consistent with the starch granules found in the roots of foxnut (Euryale ferox) in our modern reference. Foxnut is a wetland species that grows in ponds and lakes. It has also been identified at Boxi although the abundance and ubiquity are higher there (Liu et al., 2022, Supporting Information, available online).

In addition, the residue samples also revealed 11 gelatinized starch granules, most of which (n = 8) were from pure-clay vessels (jars, bowls, and basins) that were likely used for serving food. To conclude, cereals from Type 1 Panicoideae and Type 2 Triticeae account for most of the starch remains in Liujiazhai. In addition to these upland crops and grasses, wetland resources such as Job’s tears were also utilized, and to a lesser extent, foxnuts. USOs and other less ubiquitous wild plants such as fern roots and kudzu indicate reliance on local resources other than millets. The similarities of the starch remains found in Liujiazhai, Haxiu, and Boxi imply the exploitation of a similar set of local resources (Table 2). It is worth pointing out that Liujiazhai doesn’t have rice remains, unlike the other two sites. Whether this discrepancy implies different cultivation or subsistence strategies, or contrasting cultural practices, calls for further research.

Table 2.

Comparing starch residues found in the Liujiazhai, Haxiu, and Boxi sites.

	Liujiazhai	Haxiu	Boxi
Acorn		x	x
Bean			x
Foxnut	x		x
Fern root	x	x	x
Job’s tears	x	x	x
Kudzu vine root	x		x
Panicoideae	x	x	x
Rice		x	x
Silverweed		x
Triticeae	x	x	x
USO	x		x
Zingiberacea	x

Results of Haxiu and Boxi are from Liu et al. (2022). x means present. Silverweed is Potentilla anserina.

Phytolith results

We identified a total of 640 phytoliths, including 167 multicell and 473 single cell phytoliths (Supplemental Table S3, available online; Figure 5). Among the multicell phytoliths, 31 were identified as Ω -type phytoliths from broomcorn millets, 35 as η -type phytolith from foxtail millets, and 53 were classified as interdigitating type, most likely from Panicoideae inflorescences (Ge et al., 2020). Elongate forms, including El. dendriform and El. sinuate (n = 19), originating from grass husks or inflorescences, were also identified.

Figure 5.

(a) Ω -type phytolith. (b) η -type phytolith. (c) El. Sinuate. (d) Phragmite bulliform. (e) Common bulliform. (f) Cross. (g) Rondel. (h) Bilobate.

The most abundant single cell phytolith was rondel (n = 218) from Poaceae, with a significant presence in Pot 17 (n = 186) and low numbers in other samples. Hair cell was the second most abundant form (n = 114), commonly found in plants but not indicative of taxa. We also found phytolith types indicative of Panicoideae plants such as bilobate (n = 40), polylobate (n = 1), and cross (n = 10), which were consistent with the starch remains and suggesting that millets were the source of these phytoliths. In addition, Phragmites bulliform (n = 7) was also found, suggesting a minor presence of wetland resources. Other morphotypes include common bulliform (Poaceae; n = 34), trapeziform (likely Pooideae; n = 4), and elongates (grass husk, inflorescence, or leaves; n = 15).

Overall, the phytolith remains all come from Poaceae, including millets and other wild grasses. They are consistent with the starch remains and suggest additional use of wetland plants such as Phragmites. One difference between Liujiazhai and other sites is that Ω -type (broomcorn millet) and η -type (foxtail millet) are similarly abundant at Liujiazhai, whereas η -type phytolith is absent in Boxi samples (Liu et al., 2022, Supporting Information, available online). We hypothesize that this high abundance of broomcorn millet in the Boxi sample is due to alcohol fermentation. Broomcorn millet is a better ingredient for this process due to its higher amylose content and therefore requires more energy to gelatinize (Zhang and Sheng, 1995). The absence of η -type phytolith in Boxi samples might be the result of targeting broomcorn millet in alcohol production. On the other hand, Liujiazhai samples do not show evidence of fermentation-related starch, yeasts, or molds, and hence broomcorn millet phytolith is not more abundant than foxtail millet phytolith.

Control samples

We analyzed four control samples, which were extracted from the sediments adhering to the exterior surfaces of sherds. These samples showed an absence of starch granules, and only two samples revealed a few phytolith. Pot 5 control had three phytolith short cells (common bulliform, trapeziform, and Elongate irregular) and Pot 10 control had only one common bulliform. We determined that the minor presence of these phytoliths in the control samples suggests that the residue samples from the interior surfaces are associated with culinary practices rather than environmental contamination.

Discussion

Overall subsistence, millet agricultural, and local adaptation

Both starch and phytolith remains suggest that millets and local resources (such as fern roots and other USOs) characterize the subsistence pattern at Liujiazhai. The macrobotanical record of Liujiazhai shows exploitation of Chenopodium sp. and Brassica sp., as well as other weeds such as Digitaria sp. and Setaria sp. Thus, the overall subsistence at Liujiazhai has two main components: (1) millet cultivation, which is practiced in all Yangshao and Majiayao sites and (2) foraging local herbaceous plants and starchy tubers as a means of adapting to a new environment.

Both foxtail millet and broomcorn millet played significant roles as crops in the Yangshao culture of the Central Plain. While broomcorn millet was more dominant in the early periods, a shift in the crop pattern occurred around 5500 BP. Foxtail millet began to increase in quantity and eventually replaced broomcorn millet as the dominant crop (Bestel et al., 2018; Chen et al., 2020; He et al., 2022; Jia et al., 2013; Wang et al., 2018). This change can be attributed to three factors: population growth, environmental change, and the physiological characteristics of millets. Population likely expanded in the late Yangshao period (around 5500–5000 BP) as settlement size and density all increased compared to previous periods (Bi et al., 2013; Gong, 2003). Meanwhile, the weakened East Asia Summer Monsoon by 5500 BP resulted in reduced hydrological conditions and created an arid period (Chen et al., 2015a; Huang et al., 2000). The combined effect of population growth and declined hydrology likely motivated Yangshao communities to increase crop yields. Foxtail millet, as experimental studies have shown, is more productive and suffers less yield loss under water stress (Matsuura et al., 2012; Nematpour et al., 2019; Tadele, 2016). Therefore, the higher productivity of foxtail millet likely led to its intensification during a time of population growth and declining precipitation (Yang et al., 2022; Zhong et al., 2020).

However, millet farming among Yangshao migrants in western China show divergent patterns. In contrast to the general trend of increasing foxtail millet dominance in the Yangshao homeland, crop patterns in NW Sichuan, Gansu, and Qinghai show more local variations. On one hand, foxtail millet is more abundant than broomcorn millet at Yingpanshan, Shannashuzha, Dadiwan (Phase IV, Late Yangshao), and sites in the Northeastern Tibetan Plateau, as suggested by the macrobotanical record (Chen et al., 2015b; Gansu Provincial Institute of Cultural Relics and Archaeology, 2006; Hu, 2015; Zhao and Chen, 2011). On the other hand, broomcorn millet still outnumbers foxtail millet in the macrobotanical remains from the Lijiatai site (ca. 5400 cal. BP) and the Andaqiha site (ca. 5300 cal. BP) in Qinghai (Jia, 2012). Similarly, phytolith residue analysis on Haxiu, Boxi, and Dadiwan potteries (Phase III and IV, Middle to Late Yangshao) indicates that broomcorn millet is more abundant and more ubiquitous (Liu et al., 2022; Zhao and Liu, 2021). Our own phytolith analysis found an equal contribution of broomcorn ( Ω -type) and foxtail ( η -type) millets in food residues. While various taphonomic processes can lead to biased preservation of macrobotanical and phytolith remains, these lines of evidence suggest that crop patterns among Yangshao migrant communities exhibit significant variation.

This variation is likely resulted from two factors: local conditions at each migrant site and the physiological differences between broomcorn and foxtail millet. Firstly, as migration involves relocating to different environments, Yangshao migrants would counter varying local conditions that are significantly different from their homeland when they moved west and southwards to Gansu, Qinghai, and NW Sichuan. Elevation increases from Gansu to Qinghai and NW Sichuan as the latter two are located at the edges of the Tibetan Plateau (see Figure 1). Consequently, precipitation decreases with higher elevation. Moreover, numerous mountain ranges and basins in these three areas led to local microclimates (i.e. thermohydrology) and multiple vegetation types (i.e. dessert-steppe, forest, scrub, and meadow) (Chen et al., 1994; Li and Yang, 1989; Peng et al., 1989). Secondly, although foxtail millet has higher yields, broomcorn millet requires fewer nutrients and has a shorter growth cycle (Hunt et al., 2011). Thus, migrant communities could have developed their own crop patterns based on the suitability of each millet type to their specific local conditions. It is important to note that migration acts as an agent of change. When a small population migrants out of the home population, they adapt to the local conditions of their new settlement by altering subsistence patterns according to the adaptivity of their domesticates (see Ivanova et al., 2018; Kreuz et al., 2005). Therefore, the observed variation in crop patterns among Yangshao migrant communities, including Liujiazhai, shows how millet agriculture became diversified as a result of migration.

Furthermore, we propose that Liujiazhai migrants heavily exploited local wild resources, including plant roots (Starch Type 4–7) and wild Triticeae grasses, due to risky agricultural production and relative lack of food resources. d’Alpoim Guedes et al. (2015) argued that millet cultivation, especially broomcorn millet, would be extremely challenging in the Tibetan Plateau based on Growing Degree Days (GDD) modeling. Besides, the estimated Mid-Holocene mean annual precipitation (MAT) in the NW Sichuan highland is 500–800 mm, similar to the present-day MAT of 600–800 mm (Li and Yang, 1989; Zhou et al., 2011, 2021). This level of precipitation would not be ideal for millet cultivation since millet agriculture is nearly absent in most places in NW Sichuan today (Li and Yang, 1989). Thus, crop yields were likely very low. In addition to the difficulties in agricultural production, Liujiazhai had fewer resources to exploit compared to nearby sites. Macrobotanical analysis at the nearby Yingpanshan site reported a greater variety of plant taxa, including rye (Secale sp.), shiso mint (Perilla frutescens), wild grape (Vitis sp.), and woody and shrub species like Chinese Yew (Taxus chinensis), Chinese sumac (Rhus chinensis), peach (Prunus persica), and seaberry (Hippophae ehamnoides) (Zhao and Chen, 2011). In addition, faunal analysis of Yingpanshan and Liujiazhai shows that Yingpanshan consumed more domesticated pigs whereas Liujiazhai migrants relied more on wild animals (Luo et al., 2018). These differences in accessing plant and animal resources, coupled with low crop yields, suggest a challenging food environment in the mountainous highlands where Liujiazhai migrants exploited more wild resources to supplement their subsistence. Since the migrants likely interacted with the local highlanders as supported by genetic evidence (He and Ding, 2013; Jeong et al., 2014), we suggest that Liujiazhai migrants, like other migrant communities at Boxi and Haxiu (see Liu et al., 2022), integrated with the indigenous groups and embraced their subsistence practices with local plants and animals.

Another issue we need to address here is the absence of fermented beverages at Liujiazhai, considering evidence of fermentation has been observed at Haxiu and Boxi sites. This absence of evidence raises questions about potential changes in food-related practices. Liu et al. (2022) inferred that fermentation was practiced at Boxi and Haxiu based on the presence of rice starch granules, yeast, and the Monsacus (a type of fungi) qu starter, all of which are ingredients commonly used in fermentation. However, our study did not find such evidence in the 10 amphora sherds we sampled. It is possible that the absence of rice and Monascus may be due to insufficient sampling rather than a lack of fermented beverages at Liujiazhai. To draw conclusions on the continuity of this cultural tradition, more research on Neolithic NW Sichuan sites is required. The results of the current study have not yet reached a finalized statement about the practice of fermentation by the migrant communities when they settled in NW Sichuan.

Conclusion

In this study, starch and phytolith analyses were employed to investigate the subsistence economy of Liujiazhai migrants in NW Sichuan and explore the consequences of migration. Our findings indicate that the Majiayao migrants in Liujiazhai continued millet cultivation, similar to the Yangshao culture. Additionally, we propose that the Liujiazhai migrants adopted a different subsistence economy compared to their home culture, as evidenced by divergent crop patterns and the inclusion of local wild plants. These findings suggest active adaptation to the challenging landscape in NW Sichuan and potential integration with the local foraging population. Thus, our study exemplifies how migrating to a different microenvironment can lead to changes in agriculture and subsistence patterns.

While this study primarily focuses on subsistence, we acknowledge the need for further research to explore the cultural identity of migrant communities in NW Sichuan. As discussed earlier, material culture alone cannot serve as a direct representation of cultural identity. Indeed, our study reveals that the Liujiazhai migrants adopted a subsistence strategy diverged that from the Yangshao/Majiayao culture, despite their use of Majiayao-style pottery. Therefore, additional lines of evidence are necessary to determine the cultural identity of Liujiazhai and other migrant communities in NW Sichuan. This comprehensive approach will not only enhance our understanding of Sino-Tibetan history but also provide insights into the multifaceted dimensions of migration.

Supplemental Material

sj-docx-1-hol-10.1177_09596836231200438 – Supplemental material for Local adaptation and subsistence strategy of Yangshao migrants in Northwestern Sichuan in China during the Middle Neolithic (5300–4700 cal. BP)

Supplemental material, sj-docx-1-hol-10.1177_09596836231200438 for Local adaptation and subsistence strategy of Yangshao migrants in Northwestern Sichuan in China during the Middle Neolithic (5300–4700 cal. BP) by Yiyi Tang, Jiajing Wang, Li Liu and Wei Chen in The Holocene

Supplemental Material

sj-jpg-5-hol-10.1177_09596836231200438 – Supplemental material for Local adaptation and subsistence strategy of Yangshao migrants in Northwestern Sichuan in China during the Middle Neolithic (5300–4700 cal. BP)

Supplemental material, sj-jpg-5-hol-10.1177_09596836231200438 for Local adaptation and subsistence strategy of Yangshao migrants in Northwestern Sichuan in China during the Middle Neolithic (5300–4700 cal. BP) by Yiyi Tang, Jiajing Wang, Li Liu and Wei Chen in The Holocene

Supplemental Material

sj-xlsx-2-hol-10.1177_09596836231200438 – Supplemental material for Local adaptation and subsistence strategy of Yangshao migrants in Northwestern Sichuan in China during the Middle Neolithic (5300–4700 cal. BP)

Supplemental material, sj-xlsx-2-hol-10.1177_09596836231200438 for Local adaptation and subsistence strategy of Yangshao migrants in Northwestern Sichuan in China during the Middle Neolithic (5300–4700 cal. BP) by Yiyi Tang, Jiajing Wang, Li Liu and Wei Chen in The Holocene

Supplemental Material

sj-xlsx-3-hol-10.1177_09596836231200438 – Supplemental material for Local adaptation and subsistence strategy of Yangshao migrants in Northwestern Sichuan in China during the Middle Neolithic (5300–4700 cal. BP)

Supplemental material, sj-xlsx-3-hol-10.1177_09596836231200438 for Local adaptation and subsistence strategy of Yangshao migrants in Northwestern Sichuan in China during the Middle Neolithic (5300–4700 cal. BP) by Yiyi Tang, Jiajing Wang, Li Liu and Wei Chen in The Holocene

Supplemental Material

sj-xlsx-4-hol-10.1177_09596836231200438 – Supplemental material for Local adaptation and subsistence strategy of Yangshao migrants in Northwestern Sichuan in China during the Middle Neolithic (5300–4700 cal. BP)

Supplemental material, sj-xlsx-4-hol-10.1177_09596836231200438 for Local adaptation and subsistence strategy of Yangshao migrants in Northwestern Sichuan in China during the Middle Neolithic (5300–4700 cal. BP) by Yiyi Tang, Jiajing Wang, Li Liu and Wei Chen in The Holocene

Footnotes

Acknowledgements

We thank Yinzhi Cui from Northwest University (China) for her assistance in sample collection and processing.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work is supported by Burke Research Initiation Award at Dartmouth College and Min Kwaan Chinese Archaeology Fund at Stanford University.

ORCID iDs

Yiyi Tang

Jiajing Wang

Li Liu

Supplemental material

Supplemental material for this article is available online.

References

Adams

Van Gerven

Levy

(1978) The retreat from migrationism. Annual Review of Anthropology 7: 483–532.

Aldenderfer

(2011) Peopling the Tibetan plateau: Insights from archaeology. High Altitude Medicine & Biology 12(2): 141–147.

Feng

Tang

(2004) Environmental change and cultural response between 8000 and 4000 cal. yr BP in the western Loess Plateau, northwest China. Journal of Quaternary Science 19(6): 529–535.

Anthony

(1997) Prehistoric migration as social process. In: Chapman

Hamerow

(eds) Migration and Invasions in Archaeological Explanation. Oxford: Archaeopress, pp.21–32.

Anthony

(1990) Migration in archeology: The baby and the bathwater. American Anthropologist 92(4): 895–914.

Babot

(2003) Starch damages as an indicator of food processing. In: Hart

Wallis

(eds) Phytolith and Starch Research in the Australian-Pacific-Asian Regions: The State of the Art. Canberra: Pandanus Press, pp.69–81.

Bestel

Bao

Zhong

, et al. (2018) Wild plant use and multi-cropping at the early Neolithic Zhuzhai site in the middle Yellow River region, China. The Holocene 28(2): 195–207.

Liang

, et al. (2013) Spatial distribution of prehistoric settlement sites in Zhengzhou-Luoyang Region based on index model (jiyu zhishu moxing de zhengzhou-luoyang diqi shiqian juluo yizhi kongjian fenbu). Progress in Geography 32(10): 1454–1462.

Burmeister

(2000) Archaeology and migration. Current Anthropology 41(4): 539–567.

10.

Cabana

(2011) The problematic relationship between migration and culture change. In: Cabana

(ed.) Rethinking Anthropological Perspectives on Migration. Gainesville, FL: University Press of Florida, pp.16–28.

11.

Cabana

Clark

(2011) Introduction. Migration in Anthropology: Where we stand. In: Cabana

Clark

(eds) Rethinking Anthropological Perspectives on Migration. Gainesville, FL: University Press of Florida, pp.3–15.

12.

Champion

(1990) Migration revived. Danish Journal of Archaeology 9(1): 214–218.

13.

Chen

, et al. (2015a) East Asian summer monsoon precipitation variability since the last deglaciation. Scientific Reports 5(1): 1–11.

14.

Chen

Dong

Zhang

, et al. (2015b) Agriculture facilitated permanent human occupation of the Tibetan Plateau after 3600 B.P. Science 347(6219): 248–250.

15.

Chen

Peng

Huang

, et al. (1994) Vegetation characteristics and its distribution of Qilian Mountain region (qilishan diqu zhibei tezheng ji qi fenbu guilv). Acta Botanica Sinica 36(1): 63–72.

16.

Chen

(2009) Boxi, Yingpanshan, and Shawudu: Preliminary Analysis of Development of Neolithic cultures in Upper Min River (boxi yingpanshan ji shawudu – qianxi minjiang shangyou xinshiqi wenhua yanbian de jieduanxing). Archaeology and Cultural Relics (kaogu yu wenwu) 5: 65–70.

17.

Chen

Bun

Liu

, et al. (2013) Poaceae (Gramineae). In: Wu

Raven

Deyuan

(eds) Flora of China. Beijing: Science Press and St. Louis, MO: Missouri Botanical Garden, pp. 400–429. Available at: http://www.efloras.org/florataxon.aspx?flora_id=2&taxon_id=10711 (accessed 2 October 2023).

18.

Chen

Qiu

Liu

, et al. (2020) Human responses to climate change in the late prehistoric Western Loess Plateau, Northwest China. Radiocarbon 62(5): 1193–1207.

19.

Clark

(2011) Disappearance and diaspora. In: Cabana

Clark

(eds) Rethinking Anthropological Perspectives on Migration. Gainesville, FL: University Press of Florida, pp.84–108.

20.

Conolly

Colledge

Shennan

(2008) Founder effect, drift, and adaptive change in domestic crop use in early Neolithic Europe. Journal of Archaeological Science 35(10): 2797–2804.

21.

Cruz Berrocal

(2012) The Early Neolithic in the Iberian Peninsula and the Western Mediterranean: A review of the evidence on migration. Journal of World Prehistory 25(3–4): 123–156.

22.

Cui

Yang

(2011) Composition and origin of Neolithic pottery in Mao County, Sichuan (sichuan maoxian xinshiqi yizhi taoqi de chengfen fenxi ji laiyuan chutan). Cultural Relics 2: 79–85.

23.

Ding

(2010) Phrase, distribution, origin of Majiayao culture and its relation to adjacent cultures (majiayao wenhua de fenqi fenbu laiyuan ji qi yu zhoubian wenhua de guanxi). Ancient Civilization (8):36–87.

24.

d’Alpoim Guedes

Aldenderfer

(2020) The archaeology of the Early Tibetan Plateau: New research on the initial peopling through the Early Bronze Age. Journal of Archaeological Research 28(3): 339–392.

25.

d’Alpoim Guedes

Hein

, et al. (2015) Early evidence for the use of wheat and barley as staple crops on the margins of the Tibetan Plateau. Proceedings of the National Academy of Sciences of the United States of America 112(18): 5625–5630.

26.

Gansu Provincial Institute of Cultural Relics and Archaeology (2006) Dadiwan in Qin’an: Report on Excavations at a Neolithic Site (qinan dadiwan xinshiqi shidai yizhi faju baogao). Beijing: Cultural Relics Publishing House (wenwu chubanshe).

27.

Zhang

, et al. (2018) Phytolith analysis for the identification of Barnyard Millet (Echinochloa Sp.) and its implications. Archaeological and Anthropological Sciences 10(61–73): 1–13.

28.

Zhang

, et al. (2020) Phytoliths in inflorescence bracts: Preliminary results of an investigation on Common Panicoideae plants in China. Frontiers in Plant Science 10: 1736. DOI: 10.3389/fpls.2019.01736

29.

Gong

(2003) Preliminary study on settlement cluster and central settlements from Tianshui to Zhengzhou (tianshui zhi zhengzhou jian yangshao wenhua wanqi juluoqun yu Zhongxin juluo de chubu kaocha). Central China Cultural Relics 4: 28–38.

30.

Guedes

Jiang

, et al. (2013) Site of Baodun yields earliest evidence for the spread of rice and foxtail millet agriculture to south-west China. Antiquity 87: 758–771.

31.

Harper

Diachenko

Rassamakin

, et al. (2019) Ecological dimensions of population dynamics and subsistence in Neo-Eneolithic Eastern Europe. Journal of Anthropology and Archaeology 53: 92–101.

32.

Wang

Zou

, et al. (2021) Peopling history of the Tibetan Plateau and multiple waves of admixture of Tibetans inferred from both ancient and modern genome-wide data. Frontiers in Genetics 12: 725243. DOI: 10.3389/fgene.2021.725243

33.

(2015) Prehistoric culture and subsistence of the Haxiu site, Haerkang (maerkang haxiu yizhi shiqian wenhua yu shengye). Archaeology 5: 71–82.

34.

Ding

(2013) Stone industry of the Late Paleolithic Age in the Sichuan Basin (sichuan pendi jiushiqi shidai wanqi shiqi gongye). Huaxia Archaeology (huaxia kaogu) 4: 27–33.

35.

Zhang

, et al. (2022) Holocene spatiotemporal millet agricultural patterns in northern China: A dataset of archaeobotanical macroremains. Earth System Science Data 14(10): 4777–4791.

36.

Henry

Hudson

Piperno

(2009) Changes in starch grain morphologies from cooking. Journal of Archaeological Science 36(3): 915–922.

37.

Hofmann

(2020) Not going anywhere? Migration as a social practice in the early Neolithic Linearbandkeramik. Quaternary International 560-561: 228–239.

38.

Huang

Zhou

Pang

, et al. (2000) A regional aridity phase and its possible cultural impact during the Holocene megathermal in the Guanzhong Basin, China. The Holocene 10(1): 135–142.

39.

Huang

Zhu

, et al. (2019) Paleoenvironmental context of the evolution of the Baodun Culture at Chengdu Plain, Sichuan Province, China. The Holocene 29(11): 1731–1742.

40.

Huan

Wang

, et al. (2015) Bulliform phytolith research in wild and domesticated rice paddy soil in South China. PLoS One 10(10): e0141255.

41.

Hung

Cui

Wang

, et al. (2010) A provenance study of the Majiayao painted pottery found in Western Sichuan Province (chuanxi majiayao leixing caitao chuanyuan fenxi yu tantao). Southern Ethnoarchaeology 71: 1–58.

42.

Hunt

Campana

Lawes

, et al. (2011) Genetic diversity and phylogeography of broomcorn millet (Panicum miliaceum L.) across Eurasia. Molecular Ecology 20(22): 4756–4771.

43.

(2015) Research of charred botanical remains from Shannashuzhai Site in Gansu Province (gansu shanashuzha yizhi tanhua zhiwu yicun yanjiu). Master’s Thesis, Northwestern University, Evanston, IL.

44.

Ivanova

De Cupere

Ethier

, et al. (2018) Pioneer farming in southeast Europe during the early sixth millennium BC: Climate-related adaptations in the exploitation of plants and animals. PLoS One 13(5): e0197225.

45.

Jarvis

(1993) Pollen evidence of changing Holocene monsoon climate in Sichuan province, China. The Quaternary Research (Daiyonki-Kenkyu) 39: 325–337.

46.

Jeong

Alkorta-Aranburu

Basnyat

, et al. (2014) Admixture facilitates genetic adaptations to high altitude in Tibet. Nature Communications 5(1): 13281–13287.

47.

Jiang

Chen

(2003) Report on archaeological survey in Upper Min River in 2002 (2002 nian minjiang shangyou kaogu de shouhuo yu tansuo). Forum on Chinese Culture 4: 8–12.

48.

Jia

(2012) Cultural evolution process and plant remains during Neolithic-Bronze Age in Northeast Qinghai Province (qinghai sheng dongbeibu diqu xinshiqi - qingtong shidai wenhua yanbian guocheng yu zhiwu yicun yanjiu). PhD Dissertation, Lanzhou University, Lanzhou, China.

49.

Jia

Dong

, et al. (2013) The development of agriculture and its impact on cultural expansion during the late Neolithic in the Western Loess Plateau, China. The Holocene 23(1): 85–92.

50.

Kreuz

Marinova

Schäfer

, et al. (2005) A comparison of early neolithic crop and weed assemblages from the Linearbandkeramik and the Bulgarian Neolithic cultures: Differences and similarities. Vegetation History and Archaeobotany 14(4): 237–258.

51.

Leppard

(2014) Mobility and migration in the early Neolithic of the Mediterranean: Questions of motivation and mechanism. World Archaeology 46(4): 484–501.

52.

(1994) Medicinal fern resources in Sichuan and its comprehensive utilization (sichuan yaoyong juelei zhiwu zuyuan ji zonghe liyong). Zhongyaocai 2: 12–15.

53.

Yang

(1989) Research on biological resources and distribution in Sichuan (Sichuansheng shengwu ziyuan jiqi fenqu yanjiu). Resource Development and Protection 5(3): 64–77.

54.

Liu

(2021) Communal drinking rituals and social formations in the Yellow River valley of Neolithic China. Journal of Anthropology and Archaeology 63:101310. DOI: 10.1016/J.JAA.2021.101310

55.

Liu

Chen

Wang

, et al. (2022) Archaeological evidence for initial migration of Neolithic Proto Sino-Tibetan speakers from Yellow River valley to Tibetan Plateau. Proceedings of the National Academy of Sciences of the United States of America 119(51): e2212006119.

56.

Liu

Cui

(2014) Identification of starch granules using a two-step identification method. Journal of Archaeological Science 52: 421–427.

57.

Liu

Wang

Liu

(2020) The brewing function of the first amphorae in the Neolithic Yangshao culture, North China. Archaeological and Anthropological Sciences 12:118. DOI: 10.1007/s12520-020-01069-3

58.

Zhang

, et al. (2009a) Phytoliths analysis for the discrimination of foxtail millet (Setaria Italica) and common millet (Panicum miliaceum). PLoS One 4(2): e4448.

59.

Zhang

Liu

, et al. (2009b) Earliest domestication of common millet (Panicum miliaceum) in East Asia extended to 10,000 years ago. Proceedings of the National Academy of Sciences of the United States of America 106(18): 7367–7372.

60.

(2023) Local millet farming and permanent occupation on the Tibetan Plateau. Science China Earth Sciences 66: 430–434.

61.

Luo

Yao

Yuan

(2018) Subsistence economy of the Upper Yangtze River in the Pre-Qin period (changjiang shangyou diqu xianqin shiqi de shengye jingji. Southern Cultural Relics 4: 96–110.

62.

Matsuura

Tsuji

, et al. (2012) Effect of pre-and post-heading water deficit on growth and grain yield of four millets. Plant Production Science 15(4): 323–331.

63.

Mazzucco

Ibáñez

Capuzzo

, et al. (2020) Correction: Migration, adaptation, innovation: The spread of Neolithic harvesting technologies in the Mediterranean. PLoS One 15(7): e0232455–e0235874.

64.

Nematpour

Eshghizadeh

Zahedi

(2019) Drought-tolerance mechanisms in foxtail millet (Setaria italica) and proso millet (Panicum miliaceum) under different nitrogen supply and sowing dates. Crop and Pasture Science 70(5): 442–452.

65.

Neumann

Strömberg

CAE

Ball

, et al. (2019) International Code for Phytolith Nomenclature (ICPN) 2.0. Annals of Botany 124(2): 189–199.

66.

Ning

Wang

, et al. (2020) Ancient genomes from northern China suggest links between subsistence changes and human migration. Nature Communications 11(1): 1–9.

67.

Peng

Zhao

Chen

(1989) The vegetation in the Eastern Qinghai Province (qinghaisheng dongbu diqu de ziran zhibei). Acta Phytoecologica et Geobotanica Sinica 13(3): 250–257.

68.

Yang

Xiao

, et al. (2015) Nutrient values and bioactivities of the extracts from three fern species in China: A comparative assessment. Food & Function 6: 2918–2929.

69.

Ran

Feng

(2013) Holocene moisture variations across China and driving mechanisms: A synthesis of climatic records. Quaternary International 313–314: 179–193.

70.

Ren

(2015) Analysis of the Painted Pottery in Yingpanshan site (yingpanshan yizhi caitao shixi). Research of China’s Frontier Archaeology 18: 149–160.

71.

Ren

Chen

(2022) Preliminary analysis on late Yangshao sites in NW Sichuan Highlands (shilun chuanxibei gaoyuan yangshao shidai wanqi yicun). Archaeology 8: 84–97.

72.

Ren

Chen

Ren

(2013) New thoughts on the provenance of painted pottery from Western Sichuan (chuanxi caitao chandi laiyuan xinshuo jiantao). Sichuan Cultural Relics 2: 40–45.

73.

Shi

(2004) The Paleolithic cultures in migrating from the Yellow River Valley to West Sichuan Highland (cong jiushiqi wanqi wenhua yicun kan huanghe liuyu renqun xiang chuanxi gaoyuan de qianxi). Tibetan Studies 2: 33–39.

74.

Sichuan Provincial Institute of Cultural Relics and Archaeology (2021a) Preliminary report on 2019 paleolithic survey in Ganzi Tibetan autonomous prefecture, Sichuan (sichuan ganzi zangzu zizhizhou jiushiqi shidai yicun 2019 nian diaocha bagao). Sichuan Cultural Relics 6: 17–26.

75.

Sichuan Provincial Institute of Cultural Relics and Archaeology (2021b) Excavation report of Liujiazhai site in Jinchuan County, Sichuan, 2011 (sichuan xinchuanxian liujiazhai yizhi 2011 nian fajue jianbao). Archaeology 3–17.

76.

Sichuan Provincial Institute of Cultural Relics and Archaeology (2022) Excavation report of Lujiazhai site in Jinchuan County, Sichuan, 2012 (sichuan xinchuanxian liujiazhai yizhi 2012 nian fajue jianbao). Archaeology 4: 3–21.

77.

Sichuan Provincial Institute of Cultural Relics and Archaeology and School of Archaeology and Museology in Peking University (2022) Paleolithic age site at Piluo in Daocheng County, Sichuan (Sichuan daochengxian piluo jiushiqi shidai yizhi). Archaeology 7: 3–14.

78.

Silliman

(2005) Culture contact or colonialism? Challenges in the archaeology of Native North America. American Antiquity 70(1): 55–74.

79.

Tadele

(2016) Drought adaptation in millets. In Shanker AK and Shanker C (eds.) Abiotic and Biotic Stress in Plants - Recent Advances and Future Perspectives. London: InTech, pp.639–662. DOI: 10.5772/61929.

80.

Takaya

Sugimoto

Imo

Tominaga

Nakatani

Fuwa

(1978) Degradation of Starch Granules by alpha-Amylases of Fungi. Starch/Stärke 30(9): 289–293.

81.

Torrence

Barton

(eds) (2006) Ancient Starch Research, 1st edn. New York City, NY: Routledge.

82.

van Dommelen

(2014) Moving on: Archaeological perspectives on mobility and migration. World Archaeology 46(4): 477–483.

83.

Wang

, et al. (2018) Temporal changes of mixed millet and rice agriculture in Neolithic-Bronze Age Central Plain, China: Archaeobotanical evidence from the Zhuzhai site. The Holocene 28(5): 738–754.

84.

Wang

Liu

Ball

, et al. (2016) Revealing a 5,000-y-old beer recipe in China. Proceedings of the National Academy of Sciences of the United States of America 113(23): 6444–6448.

85.

Wang

Zhao

Wang

, et al. (2019) Plant exploitation of the first farmers in Northwest China: Microbotanical evidence from Dadiwan. Quaternary International 529: 3–9.

86.

Wang

(2005) Studies of commonly used traditional medicine-ginger (jiang de huaxue yaoli yanjiu jinzhan). China Journal of Chinese Materia Medica 30(20): 1569–1573.

87.

Wen

Chen

Feng

, et al. (2017) Mid-Late Holocene climatic changes recorded by loess deposits in the eastern margin of the Tibetan Plateau: Implication for human migrations. Quaternary International 441: 77–88.

88.

Williams

Bowler

Wycombe

(1982) Starch gelatinization: A morphological study of triticeae and other starches. The Star 34(7): 221–223.

89.

Wong

, et al. (2011) Kudzu root: Traditional uses and potential medicinal benefits in diabetes and cardiovascular diseases. Journal of Ethnopharmacology 134(3): 584–607.

90.

Zheng

Hou

, et al. (2018) The 5.5 cal ka BP climate event, population growth, circumscription and the emergence of the earliest complex societies in China. Science China Earth Sciences 61(2): 134–148.

91.

Yang

Wang

, et al. (2022) Shift in subsistence crop dominance from broomcorn millet to foxtail millet around 5500 BP in the western Loess Plateau. Frontiers in Plant Science 13: 13.

92.

Zhang

, et al. (2011) Phytolith analysis for differentiating between foxtail millet (Setaria Italica) and green foxtail (Setaria Viridis). PLoS One 6(5): e19726.

93.

Zhang

Storozum

Chen

, et al. (2023) Climatic shifts, geomorphic change, ancient routes of migration and adaption in southwestern China: Site formation processes at Luojiaba, Sichuan province. Geoarchaeology 38(3): 351–370.

94.

Zhang

Wang

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

95.

Zhang

Sheng

(1995) Nutritive composition of yellow rice wine and yellow millet wine (sumi huangjiu yu daomi huangjiu shumi huangjiu yingyang chengfen zhi bijiao). Journal of Nanjing Agricultural University 18(3): 124–127.

96.

Zhao

Kong

Wang

, et al. (2009) Mitochondrial genome evidence reveals successful Late Paleolithic settlement on the Tibetan Plateau. Proceedings for National Academy of Science 106(50): 21230–21235.

97.

Zhao

Liu

(2021) Alcohol production in Yangshao culture in Longdong region – Dadiwan site as an example (longing diqu yangshao wenhua niangjiu zhi fa chutan – Yi qinan dadiwanyizhi wei li). Cultural Relics of Central China 1: 49–63.

98.

Zhao

Chen

(2011) Macrobotanical results and analysis of Yingpanshan, Mao County, Sichuan (sichuan maoxian yingpanshan yizhi fuxuan jieguo yu fenxi). Southern Cultural Relics 3: 60–67.

99.

Zhong

Wang

, et al. (2020) Preliminary research of the farming production pattern in the Central Plain area during the Miaodigou period. Quaternary Sciences 40(2): 472–485.

100.

Zhou

Cen

, et al. (2011) Pattern of precipitation change in Sichuan in the past 40 years and its impact (sichuansheng jin 50 nian jiangshui de bianhua tezheng ji yingxiang). Acta Geographica Sinica 66(5): 619–630.

101.

Zhou

Wen

, et al. (2021) Holocene paleosols and paleoclimate for the arid upper Minjiang River valley in the Eastern Tibetan Plateau. CATENA 206: 105555.

102.

Zhu

(2019) Reconstruction and Evolution of Human Communication Route in Neolithic-Bronze Age of Qinghai-Tibet Plateau (qingzang gaoyuan xinshiqi – qingtong shidai renlei jiaoliu luxian de chongjian yu yanbian fenxi). Master’s Thesis, Qinghai Normal University, Qinghai, China.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.52 MB

5.25 MB

0.01 MB

0.02 MB