Adaptation at High Altitudes: A Comparative Analysis of Body Size and Proportions in Ancient Tibetan and Lowland Chinese Populations

Summary High altitude environments such as the Tibetan Plateau present significant challenges to human survival. While modern highland indigenous people show unique genetic and physiological adaptations, the skeletal expression and growth outcomes in ancient Tibetan groups remain less well understood. This study integrates osteometric methods and statistical modelling to investigate body size and proportions from five highland Tibetan sites (~2,700–900 BP), alongside comparative samples from two lowland assemblages. By incorporating broader archaeological limb bone data and contemporary anthropometric datasets from China into Bayesian hierarchical models, it further examines how climate, altitude, and socio-cultural factors have shaped morphological variation across space and time. Part I: Methodological Explorations (Chapters 3–5) This section addresses challenges in analysing fragmented and commingled remains from Tibetan sites. Chapter 3 assesses a 3D virtual method for re-associating hip joints, highlighting its potential but also its limitations due to post-depositional damage and reference mismatches. When re-association is not feasible, single-bone sex estimation offers a viable alternative. Discriminant functions based on modern Chinese data perform well for lower limbs, while the SexEst tool—primarily based on Western references—shows better accuracy for upper limbs but inconsistencies for lower limbs, likely due to inter-population morphological differences. Chapter 4 introduces an ‘aggregate’ method for estimating group-level limb proportions from commingled elements. Instead of relying on complete skeletons, this approach calculates the brachial and crural indices using sample means of relevant measurements. Validation against a global dataset (>2,000 individuals) and simulation tests indicate that the method yields reliable estimates when sample sizes are adequate (≥30 upper or ≥50 lower limb bones) and intra-sample variation remains low (standard deviation of index values < 0.02). This offers a practical solution for quantifying limb proportions in mixed assemblages where individual-level associations are not possible. Chapter 5 evaluates equations commonly used to estimate stature and body mass, comparing mathematical and mechanical methods against anatomical and morphological benchmarks from well-preserved individuals, specifically to assess their performance on the archaeological samples analysed in this study. Stature equations based on South Korean and Arctic samples yield the most accurate estimates for lowland and highland groups, respectively. For body mass, European equations by Ruff et al. (2012) provide the closest approximations across both contexts. Together, these methodological explorations highlight the potential of incomplete remains and establish a foundation for the analyses in the second half of the thesis. Part II: Spatial and Temporal Variation in Body Form (Chapters 6–8) The second part of the thesis applies the methods developed in Part I to investigate spatial and temporal variation in body size and proportions among ancient and modern Chinese populations. Chapter 6 compares skeletal remains from five highland Tibetan sites (~2,700–900 BP) with contemporaneous lowland groups. Highland individuals appear shorter, lighter, and have relatively shorter tibiae, patterns possibly linked to high-altitude stress. Some regional variation is noted, with individuals from the northeastern plateau showing relatively larger body size than those from the central and southern areas, perhaps reflecting differences in ecology, subsistence, and population history. Chapter 7 expands the analysis to over 3,000 archaeological individuals from Holocene China, drawing on both published sources and original data. Using Bayesian Generalised Additive Mixed Models (GAMMs), it explores the effects of climate, altitude, chronology, and spatial structure on long bone dimensions. Results suggest a gradual decline in stature from the Early Neolithic to Late Iron Age, especially among females, possibly reflecting agricultural intensification and rising inequality. In contrast, femoral head diameter (a proxy for body mass) remained relatively elevated, perhaps due to canalisation or gene flow from cold-adapted northern populations. Climatic variables, particularly minimum precipitation, are positively associated with limb length, while spatial residuals imply unmeasured genetic or cultural factors also have a measurable influence. Chapter 8 turns to modern data, using Bayesian GAMMs to analyse anthropometric traits (stature, body mass, relative sitting height, and body breadth) in over 90,000 adults from 213 Chinese groups spanning the mid-20th to early 21st centuries, aiming to characterise recent spatiotemporal variation in body form and explore its associations with environmental and cultural factors. Results show modest increases in stature but more pronounced gains in body mass, likely reflecting economic development, urbanisation, and improved living conditions. Maximum precipitation emerges as the most consistent climatic correlate of body size, while temperature and altitude show weaker associations. Linguistic affiliation accounts for a small but non-negligible proportion of variance, suggesting that shared ancestry or cultural norms may modestly shape body form. Notably, the modern Sherpa group remains smaller than other Tibetan Plateau groups, perhaps reflecting variation in workload, adaptation, or migration history. Overall, this thesis demonstrates the value of combining methodological innovation with comparative analysis to investigate how ecological and social factors have influenced patterns of human variability and adaptability across time and space, especially the high-altitude environments. Findings from both highland and lowland contexts, across archaeological and modern datasets, indicate that morphological variation cannot be attributed solely to either adaptive responses or developmental constraints. Instead, it reflects the overlapping effects of environmental stress, growth plasticity, cultural practices, and population history—factors that remain difficult to disentangle and warrant further investigation.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2025

Author dc:creator

• Cao, Doudou

Advisor dc:contributor.advisor

• Pomeroy, emma

• Body form
• High-altitude adaptation
• Tibetan archaeology
• Chinese populations

Licence

• https://www.repository.cam.ac.uk/bitstreams/0048d745-e231-4095-9fa0-c24a3b9e6688/download
• http://purl.org/NET/rdflicense/allrightsreserved

Language dc:language

eng

DOI dc:identifier.doi

https://doi.org/10.17863/CAM.123688

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/393308