Palaeoenvironmental reconstructions from modern and ancient plant DNA

Abstract: Palaeoecological studies on lake sediments and peat archives have provided fundamental knowledge about past environments, nevertheless, a lot remains to be learned. In this thesis, I focussed on plant ancient DNA extracted from sedimentary archives (sedaDNA), in combination with DNA from living trees with the aims of: (1) investigating different extraction methods and levels of inhibition in sediments, (2) investigating postglacial recolonisation history of Norway spruce in Fennoscandia, (3) comparing different sequencing and bioinformatic approaches to investigate past flora changes using sedaDNA and investigating past flora change in contrasting environments (southern Italy versus southern Sweden), and (4) communicating my work more broadly using art.Based on seven original methodological case studies investigating enzyme inhibition, the effect of freeze-thaw and oxygen exposure on microbial composition, and the effects of DNA extraction methods on biological diversity, we formulated guidelines and recommendations for future sedaDNA analyses. We analysed DNA traces of spruce in multiple sediments from northern Europe and confirmed that the species was present in central Sweden already at the beginning of the Holocene and, more controversially, at the end of the last glaciation in southern Sweden. Moreover, genetic ancestry of spruce trees still growing today in central Sweden suggested eastern origin of early spruce colonisers. We further demonstrated that, when sedaDNA reads are mapped to a large and high-quality nuclear plant DNA database, shotgun metagenomic analysis outperforms metabarcoding for reconstructing flora in degraded samples. However, metabarcoding remains a powerful tool to investigate past flora, because our metabarcoding analysis on southern Italian lake sediment, showed woody taxa replacement in the Holocene in good agreement with previous pollen analysis. Despite several methodological issues which complicated our floristic interpretations, at this site we also recovered a rich herbaceous flora and many aquatics. Similar methodological issues and complications, however, were also observed at higher latitudes in southern Sweden during the transition from full Glacial to the Holocene, indicating that sedaDNA from temperate environments is a comparable source of information for floral reconstruction as from high-latitude lakes. Finally, during the work of my thesis, we produced five original paintings through a long collaboration with the scientist and artist Jill Pelto, which I used in this thesis and on many other occasions to illustrate and communicate our work to a broader audience.