EN中文

Journal of Geographical Sciences >

2025 , Vol. 35 >Issue 8: 1733 - 1742

DOI: https://doi.org/10.1007/s11442-025-2310-y

Special Issue: Human, Civilization Evolution and Environmental Interaction

Chronology of the urban evolution of Ancient Loulan City

  • XU Deke , 1, 2 ,
  • LI Chang 1, 3 ,
  • JIN Yingyu 1, 3, 4 ,
  • DENG Zhenhua 1, 5 ,
  • LI Hao 1, 2 ,
  • XU Bing 1, 2 ,
  • SUN Xiaohong 6 ,
  • FENG Jing 7 ,
  • JIAO Yingxin 7 ,
  • QIN Xiaoguang 1, 2 ,
  • ZHANG Jianping 1, 2 ,
  • WU Naiqin 1 ,
  • LU Houyuan 1, 2, 3
Expand
  • 1. State Key Laboratory of Lithospheric and Environmental Coevolution, Institute of Geology and Geophysics, CAS, Beijing 100029, China
  • 2. Institute of Earth Science, CAS, Beijing 100029, China
  • 3. School of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China
  • 4. Beijing Jianhua Experimental Yizhuang School, Beijing 100176, China
  • 5. School of Archaeology and Museology, Peking University, Beijing 100871, China
  • 6. Geological Research Institute of China Chemical Geology and Mine Bureau, Beijing 100013, China
  • 7. Loulan Museum of Ruoqiang County, Bayingol Mongolian Autonomous Prefecture 841800, Xinjiang, China

Xu Deke (1980-), PhD and Associate Professor, specialized in paleoecology and paleoclimate. E-mail:

Received date: 2024-07-12

  Accepted date: 2024-10-12

  Online published: 2025-09-04

Supported by

Strategy Priority Research Program (Category B) of Chinese Academy of Sciences(XDB0710000)

National Key Research and Development Program of China(2022YFF0801502)

National Natural Science Foundation of China(42071103)

National Natural Science Foundation of China(42471175)

National Natural Science Foundation of China(42242104)

Fold

Abstract

Understanding the historical development of civilization in the Western Regions of China necessitates a detailed chronology and an in-depth analysis of the developmental dynamics of Ancient Loulan City (LA). However, systematic chronological investigations of specific archaeological sites within LA remain scarce, leaving the construction and occupation dates of many sites, along with the urban extent and development phases, unclear. To address these gaps, we established the Loulan Radiocarbon Dating Database and applied the summed probability distribution (SPD) method to reconstruct the chronology of individual sites and the urban development trajectory of LA. Our findings reveal the following: (1) Between ~500 BC and ~200 BC, only site units LA-I and LA-VIII existed, representing a nascent village phase; (2) From ~200 BC to ~AD 100, the construction of LA-IV marked the transition to an urban phase, as evidenced by abundant plant and animal remains that indicate a blend of agricultural and pastoral practices; (3) From ~AD 100 to ~AD 400, the remaining site units were constructed, occupied, and utilized, signifying the urban phase. This period saw the emergence of complex social stratification, with roles such as monks, soldiers, officials, and blacksmiths shaping the city’s structure. These chronological insights provide a new understanding of LA’s urban evolution, offering critical evidence for its socio-economic transformation.

Key words： Loulan Radiocarbon Dating Database; 14C; summed probability distribution; Western Regions; the ancient Silk Road

Cite this article

XU Deke , LI Chang , JIN Yingyu , DENG Zhenhua , LI Hao , XU Bing , SUN Xiaohong , FENG Jing , JIAO Yingxin , QIN Xiaoguang , ZHANG Jianping , WU Naiqin , LU Houyuan . Chronology of the urban evolution of Ancient Loulan City[J]. Journal of Geographical Sciences, 2025 , 35(8) : 1733 -1742 . DOI: 10.1007/s11442-025-2310-y

1 Introduction

Ancient Loulan City (LA), strategically located along the ancient Silk Road, served as a crucial hub for military, political, and cultural exchanges between Eastern and Western civilizations (Si, 1993). Its prominence spanned from the Eastern Han Dynasty (AD 25-AD 220) to the Former Liang Dynasty (AD 318-AD 376) (Chen, 2014; Hou, 2001; 2002). Investigating the detailed chronology and developmental process of LA is pivotal for understanding the historical evolution of civilizations in the Western Regions of China (Xiyu) and their cultural exchanges (Xu et al., 2017; Li et al., 2020; Li et al., 2021; Xu et al., 2023).
Current knowledge of LA’s chronology and development derives primarily from three evidence types:
(1) Indirect historical records and travelogues: The earliest known reference to LA appears in the Records of the Grand Historian, documenting its submission to Xiongnu rule in 176 BC (Si, 1993). Western Han forces later occupied LA in 77 BC, integrating it into the Western Regions Protectorate (Si, 1993). By AD 492, the Gaoche invasion led to LA’s decline (Xiao, 2011), and it was abandoned before AD 645 (Fa, 2018).
(2) Direct archaeological evidence: Documents excavated by Sven Hedin, Marc Stein, and Zuicho Tachibana in the early 20th century—written in Chinese, Sogdian, and Kharosthian—highlight frequent interactions among diverse ethnic groups (Sven, 1898; Giles, 1924; Zuicho, 1994). Wooden slips and paper documents reveal LA served as the administrative headquarters for the “Chief Official of the Western Regions” between AD 252 and AD 330 (Hou, 2001; 2002; Chen, 2014).
(3) Radiocarbon dating of sites: Chronological studies began with Hou’s publication of the first radiocarbon dating results for the LA-IV site in 2001 (Hou, 2001). Subsequent studies evaluated eight chronologies for the pagoda (LA-X) and three-room houses (LA-II), dating them between AD 100 and AD 230 (Lü et al., 2010). Later systematic sampling yielded comprehensive radiocarbon results (Xu et al., 2017; Li et al., 2019; Xu et al., 2023), showing human activity beginning around 350 BC and peaking between AD 160 and AD 340 (Xu et al., 2023). LA was ultimately abandoned between AD 560 and AD 600 (Xu et al., 2017; 2023). Significant plant and animal remains—such as foxtail millet, broomcorn millet, barley, and camel and sheep dung—date to the peak period (AD 160 to AD 340). These findings underscore the inhabitants’ reliance on mixed agriculture and pastoralism during this vibrant phase of LA’s history (Xu et al., 2023).
Analyses of historical documents, combined with prior research, have yielded detailed insights into the development of the urban landscape, population dynamics, and subsistence patterns of LA. However, the absence of systematic chronological studies targeting individual archaeological site units has left uncertainties regarding the construction and occupation dates of these units, as well as the overall extent and phases of urban development. Since the late 20th century, the application of radiocarbon (14C) probability density function analysis has advanced chronological studies in archaeology (Crema and Bevan, 2021). This method is based on the premise that the quantity and diversity of datable materials, such as architectural remains and occupation traces, directly correlate with the precision of chronological data. An increase in these materials enhances the accuracy of estimates regarding a site’s construction, occupation duration, and associated human activities (Crema et al., 2016). This technique has been extensively applied to archaeological research across regions, including Europe (Shennan, 2013), Asia (Wang et al., 2014), Australia (Williams, 2012), and the Americas (Chaput et al., 2015).
In this study, we utilized the Loulan Radiocarbon Dating Database, comprising 70 samples from architectural remains, crop residues, and animal remains. High-precision 14C probability density analyses were conducted on various units at the LA site. The results identified three developmental stages of LA spanning ~500 BC to ~AD 400: a village phase (~500 BC-200 BC), a town phase (~200 BC-AD 100), and a city phase (~AD 100-AD 400). These stages were accompanied by a progressive specialization of professions. Initially, the inhabitants depended on agriculture and animal husbandry. As the settlement evolved into a city, new professions, including monks, soldiers, officials, and blacksmiths, emerged. This diversification of occupations refined the social division of labor, reflecting a transition toward a complex urban society.

2 Natural environmental context of Ancient Loulan City

The Tarim Basin, located in arid Central Asia, is bordered by prominent natural barriers, including the Altun Mountains, the Western Kunlun Mountains, the Pamir Plateau, and the Tianshan Mountains. Glacial runoff from these mountain ranges feeds the Tarim River and its tributary, the Kongque River, ultimately converging at the basin’s lowest point, Lop Nur (Xia, 2014). Situated on Lop Nur’s western shore within the Tarim Basin, LA (40.516206°N, 89.917769°E) occupies a near-square area, with city walls measuring approximately 327-333 m on each side, enclosing a total area of ~0.108 km2. The site includes diverse architectural remains, such as city walls, three-room houses, pagodas, and building clusters (Figure 1).
Figure 1 Location and overview of Ancient Loulan City (c. City plan based on the drawings of Aurel Stein (Giles, 1924). Green, yellow, and red building remains represent structures constructed during Phases 1, 2, and 3, respectively.)
In the early 1900s, Aurel Stein meticulously cataloged the city’s structures (Giles, 1924). A river running northwest-southeast bisects the city into two sections. The northeastern section contains buildings like LA-I, LA-VIII, and LA-X, while the southwestern section is dominated by structures such as LA-II, LA-III, LA-V, LA-VI-ii, LA-VI, and LA-VII (Figure 1). The pagoda, designated as LA-X, is situated near platform LA-I (~60 m southeast) and the collapsed house LA-VIII (~80 m southeast). Administrative offices, known as the three-room houses (LA-II), are located near a large wooden frame (LA-III). A refuse pile (LA-VI-ii) lies ~35 m west of LA-II, adjacent to a cluster of large houses (LA-IV). Other significant structures include the small house LA-V (~10 m west of LA-II), the formal residence LA-VI near LA-V, and a group of modest dwellings (LA-VII), located ~90 m south of LA-II.

3 Methods

3.1 Site survey and collection of dating samples

In September 2015 and September 2016, our research team conducted systematic surveys and sample collections at the LA site (Figure 2). Chronological samples were collected from architectural remains across the following units: LA-I (the platform), LA-II (the three-room houses), LA-III (the wooden frame), LA-IV (the large houses), LA-V (the small house), LA-VI (the formal house), LA-VI-ii (the refuse pile), LA-VII (the modest houses), LA-VIII (the collapsed house), LA-X (the pagoda), and the city wall. The dating materials included a diverse array of organic samples, such as stems and branches of Populus euphratica, Tamarix, and Phragmites; camel and sheep dung; crop seeds; and bran. This diversity allowed for a comprehensive reconstruction of the construction and habitation timelines of the site, offering nuanced insights into its historical development.
Figure 2 Units of sampling sites in Ancient Loulan City
A total of 61 samples were collected and sent to Peking University, the Institute of Geology and Geophysics (Chinese Academy of Sciences), the Institute of Earth Environment (Chinese Academy of Sciences), and Beta Analytic (USA) for radiocarbon dating. Among these, 21 radiocarbon dating results are reported here for the first time, while the remaining 40 have been previously published (Xu et al., 2017; Li et al., 2019; Xu et al., 2023). Integrating nine additional results from earlier studies, we have established the updated Loulan Radiocarbon Dating Database, which now comprises a total of 70 radiocarbon (14C) dating results (Table S1) (Hou, 2001; Lü et al., 2010).
Table S1 Loulan radiocarbon dating database
14CID Source 14CAge 14CSD Lab code Phase code SiteID Material Material species
1 Hou et al.,
2001		 1860 75 Null LAIV F4 Wood NULL
2 Lu H et al.,
2010		 1820 35 BA081861 LAX(Stupa) LL-3 Grass NULL
3 Lu H et al.,
2010		 1790 35 BA081862 LAX(Stupa) LL-5 Charcoal NULL
4 Lu H et al.,
2010		 1857 39 XLLQ1669 LAX(Stupa) N5 Chinese Tamarisk NULL
5 Lu H et al.,
2010		 1930 55 XLLQ1728 LAX(Stupa) N6 Chinese Tamarisk NULL
6 Lu H et al.,
2010		 1825 40 BA081863 LAII
(Three Rooms)		 LL5-1 Poplar Populus euphratica
7 Lu H et al.,
2010		 1930 120 XLLQ1672 LAII
(Three Rooms)		 N1 Poplar Populus euphratica
8 Lu H et al.,
2010		 1930 115 XLLQ1670 LAII
(Three Rooms)		 N4 Poplar Populus euphratica
9 Lu H et al.,
2010		 1915 100 XLLQ167? LAII
(Three Rooms)		 N3 Camel excrement NULL
10 Xu B
et al., 2017		 2327 20 CN47 LAVIII L15-2-02 Tamarisk bark NULL
11 Xu B
et al., 2017		 2284 20 CN45 LAI L15-2-01 Tamarisk twig NULL
12 Xu B
et al., 2017		 1918 20 CN290 LAII
(Three Rooms)		 L15-9 Tamarisk twigs NULL
13 Xu B
et al., 2017		 1874 30 CN55 LAX (Stupa) L15-10 Straw NULL
14 Xu B
et al., 2017		 1863 25 CN49 LAII
(Three Rooms)		 L15-3 Reed NULL
15 Xu B
et al., 2017		 1859 30 CN57 LAVI L15-C4 Camel excrement NULL
16 Xu B
et al., 2017		 1848 25 CN71 South City
Wall		 L15-8 Tamarisk bark NULL
17 Xu B
et al., 2017		 1860 25 CN59 LAI L15-5 Reed NULL
18 Xu B
et al., 2017		 1849 25 CN52 LAV L15-18 Reed NULL
19 Xu B
et al., 2017		 1839 20 CN42 LAVIii L14-7 Reed NULL
20 Xu B
et al., 2017		 1840 40 CN53 LAII
(Three Rooms)		 L15-13 Reed NULL
21 Xu B
et al., 2017		 1826 25 CN51 LAII
(Three Rooms)		 L15-4 Reed NULL
22 Xu B
et al., 2017		 1804 25 CN60 LAI L15-11 Straw NULL
23 Xu B
et al., 2017		 1800 25 CN46 LAII
(Three Rooms)		 L15-1 Reed NULL
24 Xu B
et al., 2017		 1790 20 CN56 LAVII L15-17 Straw NULL
25 Xu B
et al., 2017		 1786 25 CN65 LAV L15-7 Tamarisk twigs NULL
26 Xu B
et al., 2017		 1771 25 CN67 LAVII L14-6 Reed NULL
27 Xu B
et al., 2017		 1748 35 CN62 LAIV L15-6 Tamarisk bark NULL
28 Xu B
et al., 2017		 1682 50 CN72 LAVII L14-5 Reed NULL
29 Xu B
et al., 2017		 1703 25 CN63 LAVI L14-C4 Reed NULL
30 Xu B
et al., 2017		 1678 35 CN50 LAVIii L14-C3 Reed NULL
31 Xu B
et al., 2017		 1626 35 CN64 LAVI L14-3 Reed NULL
32 Li K et al.,
2019		 1820 30 BA494205 LAX (Stupa) LA03 Poplar Populus euphratica
33 Li K et al.,
2019		 1880 30 BA494206 LAX(Stupa) LA04 Poplar Populus euphratica
34 Li K et al.,
2019		 1800 30 BA494207 LAX(Stupa) LA06 Poplar Populus euphratica
35 Xu D
et al., 2023		 1840 30 BA425113 LAVIII LA-I-S02 Seed and
Bran		 Panicum miliaceum & Setaria italica
36 Xu D
et al., 2023		 1750 30 BA622012 LAVIII LA-I-S06 Seed and
Bran		 Panicum miliaceum & Setaria italica
37 Xu D
et al., 2023		 1730 30 BA622013 LAVIII LA-I-S11 Seed and
Bran		 Panicum miliaceum & Setaria italica
38 Xu D
et al., 2023		 1760 30 BA425114 LAVIII LA-I-S14 Seed and
Bran		 Panicum miliaceum & Setaria italica
39 Xu D
et al., 2023		 1880 30 BA622014 LAVIII LA-I-5-10 Seed and
Bran		 Panicum miliaceum & Setaria italica
40 Xu D
et al., 2023		 1760 30 BA622015 LAVIII LA-I-20-25 Seed and
Bran		 Panicum miliaceum & Setaria italica
41 Xu D
et al., 2023		 1780 30 BA622016 LAV LA-V-0-10 Seed and
Bran		 Panicum miliaceum & Setaria italica
42 Xu D
et al., 2023		 1780 30 BA622017 LAV LA-V-30-40 Seed and
Bran		 Panicum miliaceum & Setaria italica
43 Xu D
et al., 2023		 1860 30 BA622018 LAVI LA-VI-10-20 Seed and
Bran		 Panicum miliaceum & Setaria italica
44 Xu D
et al., 2023		 1880 30 BA622019 LAVI LA-VI-30-40 Seed and
Bran		 Panicum miliaceum & Setaria italica
45 Xu D
et al., 2023		 1780 30 BA622020 LAVIii LA-VI-ii-1 Seed and
Bran		 Panicum miliaceum & Setaria italica
46 Xu D
et al., 2023		 1680 30 BA622021 LAVIii LA-VI-ii-4 Seed and
Bran		 Panicum miliaceum & Setaria italica
47 Xu D
et al., 2023		 1910 30 BA425111 LAVIii LA-VI-ii-07 Seed and
Bran		 Panicum miliaceum & Setaria italica
48 Xu D
et al., 2023		 1720 30 BA425112 LAII
(Three Rooms)		 LA-II-1 Seed and
Bran		 Panicum miliaceum & Setaria italica
49 This study 1950 25 BA221280 LAII
(Three Rooms)		 2016LA-2 10-15cm Seed and
Bran		 Panicum miliaceum & Setaria italica
50 This study 3110 20 BA221281 LAII
(Three Rooms)		 2016LA-2 25-30cm Seed and
Bran		 Panicum miliaceum & Setaria italica
51 This study 1745 20 BA221282 LAII
(Three Rooms)		 2016LA-2 40-45cm Seed and
Bran		 Panicum miliaceum & Setaria italica
52 This study 1845 20 BA221283 LAII
(Three Rooms)		 2016LA-2 60-65cm Seed and
Bran		 Panicum miliaceum & Setaria italica
53 This study 1815 20 BA221284 LAII
(Three Rooms)		 2015LA-II HP middle Seed and
Bran		 Panicum miliaceum & Setaria italica
54 This study 1780 30 BA221285 LAIII 2016LA-III Seed and
Bran		 Panicum miliaceum & Setaria italica
55 This study 2005 25 BA221286 LAIV 2016LA-IV-2 sheep
excrement		 NULL
56 This study 2095 20 BA221287 LAIV 2016LA-IV-3 sheep
excrement		 NULL
57 This study 2065 25 BA221288 LAIV 2016LA-IV-4 sheep
excrement		 NULL
58 This study 1840 20 BA221289 LAIV 2015LA-IV-02 Seed and
Bran		 Panicum miliaceum & Setaria italica
59 This study 1885 25 BA221290 LAIV 2015LA-IV-03 Seed and
Bran		 Panicum miliaceum & Setaria italica
60 This study 1975 20 BA221291 LAIV 2015LA-IV-05 Seed and
Bran		 Panicum miliaceum & Setaria italica
61 This study 1965 20 BA221292 LAIV 2015LA-IV-06 Seed and
Bran		 Panicum miliaceum & Setaria italica
62 This study 1840 20 BA221293 LAIV 2015LA-IV-08 Seed and
Bran		 Panicum miliaceum & Setaria italica
63 This study 2005 25 BA221294 LAIV 2015LA-IV-09 Seed and
Bran		 Panicum miliaceum & Setaria italica
64 This study 2030 20 BA221295 LAIV 2015LA-IV-10 Seed and
Bran		 Panicum miliaceum & Setaria italica
65 This study 1825 25 BA221296 LAVIii 2016LA-VI-ii South Seed and
Bran		 Panicum miliaceum & Setaria italica
66 This study 1895 20 BA221297 LAVIii 2016LA-VI-ii North Seed and
Bran		 Panicum miliaceum & Setaria italica
67 This study 1900 20 BA221298 LAVII 2016LA-VII-1 Northeast Seed and
Bran		 Panicum miliaceum & Setaria italica
68 This study 1940 20 BA221299 LAVII 2016LA-VII-2 Southeast Seed and
Bran		 Panicum miliaceum & Setaria italica
69 This study 1840 20 BA221300 LAVII 2016LA-VII-3 North Seed and
Bran		 Panicum miliaceum & Setaria italica
70 This study 1540 20 BA221301 LAVII 2016LA-VII-4 West Seed and
Bran		 Panicum miliaceum & Setaria italica

3.2 14C probability density functions

The “rcarbon” package, a versatile software tool in the R programming environment (Crema and Bevan, 2021), facilitates functions such as 14C age calibration, data binning and merging, data visualization, and statistical analysis of summed probability density (SPD) functions (Crema and Bevan, 2021).
Constructing an SPD involves two key steps:
1) Correcting individual radiocarbon dates to produce probability density distribution curves for each sample.
2) Summing the probability densities of all samples to generate a composite probability density curve for a given site.
The SPD outcome is influenced by the morphology, dating precision, and calibration profile of the SPD curve for each sample (Crema and Bevan, 2021).
However, uneven archaeological sampling across sites can introduce biases, such as “false peaks” and “pseudo valleys” in the SPD data (Bevan et al., 2017). To mitigate these distortions, oversampled sites must aggregate samples with similar ages within fixed temporal intervals (typically 50-200 years) (Crema and Bevan, 2021). The chi-square test determines the statistical validity of this amalgamation. If the test is significant, the combined age is incorporated into the overall SPD; otherwise, individual dates are treated independently in subsequent calculations (Wang et al., 2014). To standardize analysis across diverse samples, the SPD area weights for each sample are adjusted before summation. This ensures equal treatment of all test ages, with the cumulative SPD calculated through standardized area superposition (Crema and Bevan, 2021).

4 Summed probability distribution of the sites in Ancient Loulan City

Based on the timing of SPD peaks at various archaeological sites in the LA city, we can divide the data into three distinct phases (Figure 3):
Figure 3 Summed probability distribution of Ancient Loulan City
Phase 1 (~500 BC to ~200 BC): Both LA-I and LA-VIII show their first SPD peaks, reaching values of 0.05 and 0.005, respectively.
Phase 2 (~200 BC to ~AD 100): LA-IV shows a single SPD peak, reaching a value of 0.031.
Phase 3 (~AD 100 to ~AD 400): The following units all show SPD peaks: LA-I (0.010), LA-II (0.045), LA-III (0.008), LA-V (0.019), LA-VI (0.016), LA-VI-II (0.024), LA-VII (0.190), LA-VIII (0.024), LA-X (0.340), and the city wall (0.009). The respective median ages of LA-I, LA-II, LA-III, LA-V, LA-VI, LA-VI-II, LA-VII, LA-VIII, LA-X, and the city wall are AD 225, AD 222, AD 306, AD 7, AD 246, AD 181, AD 219, AD 228, AD 264, AD 223, and AD 214.
Between ~AD 400 and ~AD 560, the SPD for all the archaeological units exhibits a gradual decreasing trend.

5 Process of the urban development of Ancient Loulan City

During the period from ~500 BC to ~200 BC (Phase 1), two archaeological units, namely LA-I and LA-VIII, emerged at the LA site, indicating an early village phase. This finding is consistent with those of previous studies (Xu et al., 2017; 2023). The SPD curves for all units show a gradual decline, suggesting reduced activity across sites. During Phase 1, archaeological evidence indicates the emergence of early village settlements at LA-I and LA-VIII. This phase is supported by two radiocarbon-dated samples of building materials (Table S1), but a lack of additional evidence precludes definitive conclusions about subsistence patterns (Figures 3 and 4). Furthermore, the absence of Han Dynasty wooden tablets or multilingual documents suggests that LA was not a significant center for information exchange or cultural interaction during this period (Chen, 2014).
Figure 4 Three phases of the development of Ancient Loulan City and the social division of labor among its inhabitants
From approximately 200 BC to AD 100 (Phase 2), the development of the LA-IV site marks a transition from a village to a burgeoning town. This progression, as noted by previous researchers, has been challenging to fully document due to the absence of detailed subdivisions of the various units within the LA site. However, dating materials from this period are relatively abundant. Alongside architectural remains, evidence of grain storage and animal dung from camels and sheep suggests that the inhabitants relied heavily on agriculture and animal husbandry (Figures 3, 4, and Table S1).
Between ~AD 100 and ~AD 400 (Phase 3), the construction, occupation, and utilization of all major architectural units—including LA-II, LA-III, LA-V, LA-VI-ii, LA-VI, LA-VII, LA-X, and the city wall—took place (Figures 3 and 4). This period is characterized by the second and highest peaks in SPDs observed at LA-I and LA-VIII, which may indicate significant repairs, extensions, or reinforcements of these units. Alternatively, these peaks could reflect increased population density or heightened mobility within the site. The completion of these functional structures signifies that LA had evolved into a fully developed urban center, aligning with earlier findings that LA reached its demographic and urbanization zenith during this time (Xu et al., 2017; 2023).
Dating materials from this phase are notably diverse, encompassing architectural elements such as poplar, tamarisk, and reeds, as well as animal dung (sheep and camel) and crop residues (millet seeds and bran). The community continued to depend primarily on animal husbandry and agriculture.
Our findings, along with prior research (Hou, 1988a; Chen, 2014), reveal abundant charcoal and iron slag within the stratigraphic layers of LA-VI-ii (the refuse pile) and LA-VII, indicating possible iron-smelting activities. During Aurel Stein’s 1906 excavations at LA-I, nine Han Dynasty wooden tablets and paper fragments were unearthed, including one dated to AD 330 (Giles, 1924). Additional discoveries included wool and silk textile fragments and Han Dynasty coins (Giles, 1924). Notably, three significant structures—LA-II (comprising three-room houses and an administrative office), LA-X (a pagoda), and the city walls—were constructed during this period, reflecting the city’s administrative, religious, and military significance. The LA-II site yielded numerous wooden documents dated between AD 252 and AD 310 (Chen, 2014). Among these, the “Li Bai documents”, authored by the Chief Secretary of the Western Regions during the Former Liang Dynasty, are dated to AD 328 (Zuicho, 1994). On the northern side of LA-II, four additional Han Dynasty wooden tablets were discovered, one dated to AD 263 (Chen, 2014). These historical records corroborate the estimated SPD peak for LA-II, occurring between ~AD 80 and ~AD 350. Although LA-III has only a single radiocarbon date with a peak around AD 280, this aligns with the date of AD 270 recorded on wooden documents from the site. The SPD peak for LA-VI-ii is similarly situated between ~AD 100 and ~AD 370. Over 200 written documents recovered from this site, dating between AD 264 and AD 312 (Chen, 2014), further support these findings.
The presence of diverse functional structures, abundant plant and animal remains, and evidence of religious, political, military, and metallurgical activities underscores a phase of professional specialization and urban complexity. Radiocarbon dating, archaeological evidence, and historical records (Hou, 1988a; 1988b; 2001; 2002; Chen, 2014) collectively highlight LA’s prosperity, suggesting it became a critical urban center in the Western Regions during this era. The city facilitated economic and cultural exchanges among various ethnic groups (Sven, 1898; Giles, 1924; Hou, 1988b; Zuicho, 1994), supporting activities such as agriculture, animal husbandry, industrial production, trade, and administration. Consequently, LA emerged as a pivotal hub in the Western Regions.
A comparative analysis of paleoclimate records from the Tianshan and West Kunlun Mountains and the southern margin of the Tibetan Plateau reveals that the origin, growth, and prosperity of LA were intrinsically tied to increased moisture from the Westerlies and the Indian monsoon (Xu et al., 2017; 2023). Enhanced precipitation in these mountainous regions boosted glacial meltwater production, leading to greater flows in the Tarim and Kongque Rivers. These abundant water resources sustained agriculture, livestock, and a growing human population (Li et al., 2020; Xu et al., 2017; 2023).
After ~AD 400, the SPDs for all archaeological units steadily declined, indicating a cessation of large-scale urban construction at LA. This aligns with the consensus that human activity in LA diminished during this period (Xu et al., 2017; 2023). Wooden slips unearthed at the LA site date the latest documentary evidence to ~AD 330 (Chen, 2014), corroborating this decline. Historical records indicate that the final reference to LA occurs around ~AD 399, when the monk Faxian (Fa, 2018) passed through Shanshan (ancient LA) on his westward journey to obtain sacred sutras (Fa, 2018). During his 30-day stay, Faxian described the city in his chronicle, A Record of Buddhist Kingdoms, as barren and inhospitable despite the presence of numerous monks (Fa, 2018). By AD 645, the pilgrim Xuanzang confirmed that LA had been abandoned for an extended period (Xuan and Bian, 2006).
Three key factors likely contributed to the cessation of construction and the subsequent decline LA after ~AD 400:
1) Effects of climate change: Reduced glacial meltwater from surrounding mountains led to a critical water shortage (Xu et al., 2017; 2023).
2) Human overexploitation: Intensive use of water resources caused their depletion (Mischke et al., 2017; Li et al., 2020).
3) Conflict and Famine: Wars and food scarcity exacerbated the city's decline (Yuan and Zhao, 1997; Xu et al., 2017).

6 Conclusions

To reconstruct the SPDs for various archaeological units, we developed the Loulan Radiocarbon Dating Database. Our findings divide LA’s urban development into three distinct phases:
(1) Village Phase (~500 BC to ~200 BC): During this period, only site units LA-I and LA-VIII were constructed, reflecting a small village settlement.
(2) Town Phase (~200 BC to ~AD 100): The construction of site unit LA-IV indicates the transition to a town, with inhabitants engaged in agriculture and animal husbandry.
(3) Urban Phase (~AD 100 to ~AD 400): This phase saw the construction and occupation of all remaining site units. Key features, such as the pagoda (LA-X), city walls, government offices (LA-II, including the “three-rooms”), and iron-smelting remains (sites LA-VI-ii and LA-VII), highlight diverse urban functions, including religious practices, military defense, administration, and industrial activities. The emergence of specialized occupations such as monks, soldiers, officials, and blacksmiths reflects an increasingly complex social division of labor.
These three developmental stages illustrate the evolution of occupational specialization in LA from simple to sophisticated systems. The chronological evidence presented here offers valuable insights into the urbanization process of LA, paving the way for further in-depth investigations.

This research contributes to the Diverse K (a working group of PAGES) initiative. We acknowledge the use of the “Refrigeration storage for high value sediment samples” facility at Beijing National Observatory of Space Environment, Institute of Geology and Geophysics, Chinese Academy of Sciences.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
Bevan A, Colledge S, Fuller D et al., 2017. Holocene fluctuations in human population demonstrate repeated links to food production and climate. Proceedings of the National Academy of Sciences of the United States of America, 114(49): E10524-E10531. doi: 10.1073/pnas.1709190114.

[2]
Chaput M A, Kriesche B, Betts M et al., 2015. Spatiotemporal distribution of Holocene populations in North America. Proceedings of the National Academy of Sciences of the United States of America, 112(39): 12127-12132. doi: 10.1073/pnas.1505657112.

[3]
Chen X, 2014. Archaeology of Loulan. Lanzhou: Lanzhou University, 443 pp. (in Chinese)

[4]
Crema E R, Bevan A, 2021. Inference from large sets of radiocarbon dates: Software and methods. Radiocarbon, 63(1): 23-39. doi: 10.1017/rdc.2020.95.

[5]
Crema E R, Habu J, Kobayashi K et al., 2016. Summed probability distribution of 14C dates suggests regional divergences in the population dynamics of the Jomon Period in eastern Japan. PLOS ONE, 11(4): e0154809. doi: 10.1371/journal.pone.0154809.

[6]
Fa H, 2018. Fa-Hein’s Record of Buddhistic Kingdoms. Taiyuan: Sanjin, 228 pp. (in Chinese)

[7]
Giles L, 1924. Serindia: Detailed report of explorations in Central Asia and Westernmost China, carried out and described under the orders of H.M. Indian Government by Aurel Stein, K.C.I.E., Indian Archæological Survey. With 345 illustrations from photographs, 59 plans, 175 plates, and 96 maps. pp. 1,580. Vols. I-III, text; Vol. IV, plates; Vol. V, maps. 13 × 10. Oxford: Clarendon Press, 1921. Journal of the Royal Asiatic Society, 56(1): 141-146. doi: 10.1017/S0035869X00095277.

[8]
Hou C, 1988a. A brief report on the investigation and excavation of the ancient city site of Loulan. Cultural Relics, (7): 1-22. (in Chinese)

[9]
Hou C, 1988b. A study of the newly discovered wooden slips and paper documents from Loulan. Cultural Relics, (7): 40-50. (in Chinese)

[10]
Hou C, 2001. The contrete evidence for the seat of Jurisdiction of Clarklik under superior chief of Western Region during Wei and Jin Dynasty: Argument on question of charklik (I). Dunhuang Research, (4): 105-111, 183. (in Chinese)

[11]
Hou C, 2002. A critical study of Loulan research: Argument on question of charklik (2). Cultural Relics, (1): 66-72. (in Chinese)

[12]
Li K, Qin X, Zhang L et al., 2019. Oasis landscape of the ancient Loulan on the west bank of Lake Lop Nur, Northwest China, inferred from vegetation utilization for architecture. The Holocene, 29 (6): 1030-1044. doi: 10.1177/0959683619831423.

[13]
Li Y, Storozum M J, Jia X et al., 2020. Reconceptualizing water history of Chinese Central Asia: Hydraulic modeling of the early 1st mill. AD irrigation system at Mohuchahangoukou-4 (MGK4), Xinjiang, China. Journal of Archaeological Science: Reports, 33: 102534. doi: 10.1016/j.jasrep.2020.102534.

[14]
Li Y, Storozum M, Li H et al., 2021. Architectural connections between western Central Asia and China: New investigations at Haermodun (cal AD 90-321), a fortified circular settlement in Xinjiang, China. Antiquity, 95(380): e10. doi: 10.15184/aqy.2021.17.

[15]
Lü H, Xia X, Liu J et al., 2010. A preliminary study of chronology for a newly-discovered ancient city and five archaeological sites in Lop Nor, China. Chinese Science Bulletin, 55(1): 63-71. doi: 10.1007/s11434-009-0586-4.

[16]
Mischke S, Liu C, Zhang J et al., 2017. The world’s earliest Aral-Sea type disaster: the decline of the Loulan Kingdom in the Tarim Basin. Scientific Reports, 7(1): 43102. doi: 10.1038/srep43102.

[17]
Shennan S, 2013. Demographic continuities and discontinuities in Neolithic Europe: Evidence, methods and implications. Journal of Archaeological Method and Theory, 20(2): 300-311. doi: 10.1007/s10816-012-9154-3.

[18]
Si M, 1993. Records of the Grand Historian (Shiji), New York: Columbia University Press.

[19]
Sven H. 1898. Through Asia. London: Methuen & Co., 714 pp.

[20]
Wang C, Lu H, Zhang J, et al., 2014. Prehistoric demographic fluctuations in China inferred from radiocarbon data and their linkage with climate change over the past 50,000 years. Quaternary Science Reviews, 98: 45-59. doi: 10.1016/j.quascirev.2014.05.015.

[21]
Williams A N, 2012. The use of summed radiocarbon probability distributions in archaeology: A review of methods. Journal of Archaeological Science, 39(3): 578-589. doi: https://doi.org/10.1016/j.jas.2011.07.014.

[22]
Xia X. 2014. Lop Nor China. Beijing: Science Press, 752 pp. (in Chinese)

[23]
Xiao Z, 2011. Book of the Southern Qi Dynasty. Beijing: Zhonghua Book Company, 199 pp. (in Chinese)

[24]
Xu B, Gu Z, Qin X et al., 2017. Radiocarbon dating the ancient city of Loulan. Radiocarbon, 59(4): 1215-1226. doi: 10.1017/RDC.2017.21.

[25]
Xu D, Li C, Jin Y et al., 2023. Relationship between the rise and fall of Loulan Ancient City and centennial-scale climate events and cycles. Frontiers of Earth Science, 17(4): 1070-1080. doi: 10.1007/s11707-023-1091-9.

[26]
Xuan Z, Bian J, 2006. The Great Tang Dynasty Record of the Western Regions (W.M. Keck Foundation Series). Moraga: BDK America.

[27]
Yuan G, Zhao Z, 1997. Relationship between the rise-fall of ancient Loulan city and environmental changes, Xinjiang. Arid Land Geography, 20(3): 7-12. (in Chinese)

[28]
Zuicho T, 1994. Central Asian Expedition. Urumqi: Xinjiang People’s Publishing House, 166 pp. (in Chinese)

Options
Outlines

/

〈 〉