Abstract:
Over the past decade, the broader application of palaeoproteomics, particularly Zooarchaeology by Mass Spectrometry (ZooMS), has led to the discovery of new hominin fossils across various Palaeolithic sites. ZooMS, with its capacity to taxonomically identify faunal remains based on molecular features, has significantly expanded the scope of faunal research of the Pleistocene. Nonetheless, the workflow of ZooMS analysis requires further refinement, particularly in terms of chemical protocol applicability and data processing. It is also notable that the application of ZooMS has yet to be extended to East Asia.
This doctoral research pursued three main objectives. First, I aimed to assess the performance of various ZooMS protocols for screening projects and subsequently to determine the optimal protocol for samples of diverse preservation. Thus, I evaluated three existing ZooMS protocols and a new SP3 protocol. Through the analysis of 400 samples from seven sites, I concluded that the Acid-insoluble protocol consistently yielded optimal collagen extracts, while the non-destructive AmBic protocol proved effective only for well-preserved samples. The newly tested SP3 protocol exhibited inferior performance compared to other protocols.
The second objective was to enhance the efficiency of data processing for screening large numbers of bones using ZooMS. I introduced a semi-automated identification pipeline for ZooMS (QuickID-based) and tested it using a manually-identified dataset. Results demonstrated that QuickID accelerates data processing, ensuring consistent standards of identification and improving overall reproducibility.
The third objective involved applying the refined ZooMS workflow to Palaeolithic fauna collections from Europe and East Asia, then integrating ZooMS results and traditional zooarchaeological data. In the first ZooMS screening project, I analysed 287 bone fragments from Vogelherd Cave in the Swabian Jura, which provided insights regarding the undiagnostic small bones and resulted in the discovery of three new human bones. In the second project, I analysed 745 bones from Jinsitai Cave, an important Middle and Upper Palaeolithic site on the Mongolia Plateau in China. I uncovered two taxa - camel and ostrich - that had not previously been identified in the zooarchaeological study. The subsequent attempts on integration established a framework for combining ZooMS and zooarchaeological data in Palaeolithic contexts, aiding analysts in formulating research questions and making informed decisions regarding the use of ZooMS.
Additional ZooMS analyses of nearly 600 bones from four Chinese Palaeolithic sites yielded exciting insights into understanding how sedimentary context contributes to collagen preservation. More importantly, the results confirmed that ZooMS could be applied to very old (>100 ka) contexts with high success rates in some favourable conditions in southern China.
This dissertation encompasses refined analytical workflows for ZooMS, successful applications in European and East Asian Palaeolithic contexts, and insights into the integration of ZooMS and traditional zooarchaeology. These results have the potential to propel the field of palaeoproteomics forward and serve as a useful resource for future studies.
ZooMS