The Faroe Islands, a North Atlantic archipelago situated between Norway and Iceland, have long been a subject of debate regarding their initial human settlement. While the prevailing view points to Norse settlement around 800-900 CE, supported by archaeological findings, several lines of indirect evidence hint at a much earlier human presence. The discovery of charred barley grains dating back to 351-543 CE at the A Sondum archaeological site ignited this debate, suggesting pre-Norse human activity. Further indirect evidence includes mentions by an Irish monk in 825 CE suggesting settlements on northern islands for at least a century, Celtic place names and grave markings on the islands, and the genetic makeup of the modern Faroese population, which exhibits a significant asymmetry between paternal (primarily Scandinavian) and maternal (primarily British Isles) lineages. However, the interpretation of these clues remains inconclusive, lacking direct physical evidence to confirm pre-Norse settlement. This research aimed to address this gap by examining lake sediment cores for definitive markers of human activity and livestock introduction, providing a continuous environmental record to complement the often-fragmented archaeological data. The study focuses on Eiðisvatn, a lake near the Argisbrekka archaeological site, a Norse shieling (summer farm settlement). The environmental record, specifically focusing on pollen, plant macrofossils, and more recently, sedaDNA, has been previously investigated. While these studies indicated changes in vegetation consistent with human impact, dating uncertainties hindered a precise timeline for human arrival. Therefore, this study leverages a combination of advanced dating techniques (radiocarbon dating and tephrochronology) with the analysis of fecal biomarkers and sedaDNA to provide a more precise and conclusive timeline of human occupation of the Faroe Islands.

Previous research on the Faroe Islands' early settlement relied heavily on archaeological evidence, primarily dating to the Viking Age (800-1100 CE). The discovery of charred barley grains at A Sondum, dated to 351-543 CE, challenged this narrative, suggesting earlier human presence. Indirect evidence, including historical accounts mentioning hermits on northern islands, Celtic place names and grave markings, and the genetic asymmetry in the modern Faroese population (Scandinavian paternal lineage and British Isles maternal lineage), further supported the hypothesis of pre-Norse settlement. Paleoecological studies using pollen and plant macrofossils from lake and bog sediments provided insights into vegetation changes, with a reduction in woody plants attributed to human-induced changes such as grazing. However, challenges in dating these sediment cores limited the precision of these studies, preventing a clear confirmation of pre-Norse settlement. This study built on this existing research by employing more robust dating techniques and integrating multiple lines of molecular evidence, including fecal biomarkers and sedaDNA, to address the limitations of previous approaches and provide a more definitive answer.

This study utilized a multi-proxy approach combining advanced dating techniques, biomarker analysis, and sedaDNA metabarcoding to reconstruct the timeline of human arrival and environmental impact in the Faroe Islands. Lake sediment cores were collected from Eiðisvatn, a lake proximal to the Argisbrekka archaeological site. An age model was developed using 10 radiocarbon ages and 5 geochemically fingerprinted tephra layers, employing Bayesian age modeling software (Bacon) to account for uncertainties. Sediment samples were analyzed for fecal sterols (coprostanol and stigmastanol), which serve as unambiguous indicators of mammal presence, including humans and livestock. The concentrations were normalized to organic matter content (LOI) to mitigate the influence of increased sediment accumulation during periods of erosion. Sedimentary ancient DNA (sedaDNA) was extracted from selected samples and subjected to metabarcoding to identify plant and mammal DNA. The analysis included three replicate PCRs per metabarcode, with a minimum read threshold of five to minimize false positives. Plant sedaDNA data were analyzed to assess changes in plant community composition, indicating shifts in vegetation associated with human and livestock activity. The mammalian sedaDNA analysis aimed to identify the presence of livestock, such as sheep, providing further evidence of human introduction of these species. The results from the different analytical techniques were integrated to establish a chronological framework for human arrival and associated environmental changes.

The study revealed a clear increase in fecal biomarker concentrations (coprostanol and stigmastanol) at a sediment depth corresponding to approximately 500 CE (95% confidence interval 370–610 CE). This coincides precisely with the first detection of sheep DNA in the sediment samples. The analysis of plant sedaDNA shows a simultaneous shift in vegetation, with an increase in grasses and a decline in woody plants. These findings are consistent with the introduction of livestock and subsequent grazing-induced changes in the landscape. The age of the sedimentary horizon showing the co-occurrence of these molecular indicators (sheep DNA and increased fecal biomarkers) was rigorously constrained using a Bayesian age model based on multiple radiocarbon dates and tephra layers. The weighted mean age for the onset of human activity, as indicated by the first appearance of sheep DNA and increased fecal biomarkers, is estimated at 500 CE, with a 95% confidence interval of 370–610 CE. This firmly establishes that humans and their livestock were present in the Faroe Islands at least 300 years before the onset of the documented Norse settlement around 800 CE. Further corroboration was found in the observed changes in the plant community, supporting the interpretation of human-induced environmental changes at this time. The magnitude of the increase in stigmastanol was greater than that of coprostanol, indicating a higher contribution of sheep feces, possibly reflecting a predominance of sheep among introduced livestock.

The findings presented in this study provide strong and conclusive evidence for human occupation of the Faroe Islands significantly earlier than previously accepted. The co-occurrence of increased fecal biomarkers and the appearance of sheep DNA, supported by vegetation changes, offers a powerful combination of molecular evidence that strongly suggests human arrival and the introduction of livestock around 500 CE. This predates the generally accepted timeframe for Norse settlement by approximately 300 years. These findings are consistent with the previously discovered charred barley grains at A Sondum, offering further support for pre-Norse human activity. While the exact identity of these early settlers remains unknown, the results challenge the long-held view of Norse as the first inhabitants of the Faroe Islands. The study also highlights the importance of using multiple lines of evidence, including molecular data and detailed chronological control, in reconstructing human history in remote regions. The ecological consequences of early human settlement are also highlighted by the observed changes in vegetation, primarily the shift from woody to herbaceous plant communities as a result of grazing. This has implications for our understanding of Holocene vegetation development in the Faroe Islands and challenges previous assumptions that attributed vegetation changes solely to climate fluctuations.

This research provides irrefutable evidence for human and livestock presence in the Faroe Islands at least 300 years before the commonly accepted Norse arrival. The integration of advanced dating techniques with molecular data (fecal biomarkers and sedaDNA) yielded a compelling chronological framework for early human activity. Further research should focus on identifying the origins of these early settlers, potentially through additional sedaDNA analysis or investigations of other archaeological sites. This study underscores the power of using combined molecular and chronological approaches to uncover hidden chapters of human history.

While this study provides strong evidence for pre-Norse settlement, several limitations exist. First, the study focuses on a single lake sediment core. While the location was carefully selected for its proximity to a known archaeological site, further research is needed to confirm these findings at additional sites. Second, the identity of the early settlers remains unknown. Future work incorporating advanced genomic analysis of sedaDNA might shed light on their origin and relationship to later populations. Finally, the precise extent of their land use and impact on the ecosystem is not fully captured in this study. Future research could expand on these aspects.