Late Quaternary sea ice reconstructions in the Greenland Sea using sedimentary ancient DNA

Arctic sea ice is declining fast, heralding a change that will affect local communities and ecosystems, as well as local and global economy and politics. To predict future Arctic climate, climate models require input from, or need to be tested against, observations of past sea ice cover. Satellite and historical records only go back to the 1970s or around 1850, respectively, but beyond that we crucially lack tools that can quantitatively assess past Arctic sea ice variability. Sedimentary ancient DNA (sedaDNA) as a palaeoceanographic proxy is still in its infancy and its potential is not yet fully explored. However, emerging multidisciplinary studies suggest that, through its geochemical link with biological diversity, sedaDNA can become a valuable tool to deepen our understanding of palaeoceanography and sea ice history. We investigated a sediment core from the Greenland Sea for sedaDNA, dinoflagellate cysts and the sea ice biomarker IP25. Metabarcoding analysis using universal eukaryotic primers targeting the V7 hypervariable region of the small subunit ribosomal RNA gene (SSU) revealed a rich but distinct diversity in the eight sediment samples analyzed, representing the recent past and as far back as approximately 100 ka. The sedaDNA sample dated to ~34 ka was unique in the study interval with nearly-absent diatoms and a dominance of dinoflagellates, cercozoans and unclassified sequences. The same sample also yielded the sea-ice-associated dinoflagellate cyst Islandinium, the organic biomarker IP25 and high relative abundance of the sea-ice-associated dinoflagellate Polarella glacialis, all indicative of seasonal sea ice cover in this sample. Our goal is to fully develop the potential of sedaDNA to identify open ocean, seasonal sea ice or permanent sea ice environments, as well as to develop tools to quantify sea ice parameters in the geological record.