Summary Sedimentary DNA (sedDNA) has recently emerged as a new proxy for reconstructing past vegetation, but its taphonomy, source area and representation biases need better assessment. We investigated how sedDNA in recent sediments of two small Scottish lakes reflects a major vegetation change, using well‐documented 20th Century plantations of exotic conifers as an experimental system.We used next‐generation sequencing to barcode sedDNA retrieved from subrecent lake sediments. For comparison, pollen was analysed from the same samples.The sedDNA record contains 73 taxa (mainly genus or species), all but one of which are present in the study area. Pollen and sedDNA shared 35% of taxa, which partly reflects a difference in source area. More aquatic taxa were recorded in sedDNA, whereas taxa assumed to be of regional rather than local origin were recorded only as pollen.The chronology of the sediments and planting records are well aligned, and sedDNA of exotic conifers appears in high quantities with the establishment of plantations around the lakes. SedDNA recorded other changes in local vegetation that accompanied afforestation. There were no signs of DNA leaching in the sediments or DNA originating from pollen.

Information on past land cover in terms of absolute areas of different landscape units (forest, open land, pasture land, cultivated land, etc.) at local to regional scales is needed to test hypotheses and answer questions related to climate change (e.g. feedbacks effects of land-cover change), archaeological research, and nature conservancy (e.g. management strategy). The palaeoecological technique best suited to achieve quantitative reconstruction of past vegetation is pollen analysis. A simulation approach developed by Sugita (the computer model POLLSCAPE) which uses models based on the Communicated by J. Dearing.

Environmental DNA retrieved from modern soils (eDNA) and late-Quaternary palaeosols and sediments (aDNA and sedaDNA) promises insight into the composition of present and past terrestrial biotic communities, but few studies address the spatial relationship between recovered eDNA and contributing organisms. Svalbard's vascular plant flora is well known, and a cold climate enhances preservation of eDNA in soils. Thus, Svalbard plant communities are excellent systems for addressing the representation of plant eDNA in soil samples. In two valleys in the inner fjord region of Spitsbergen, we carried out detailed vegetation surveys of circular plots up to a 4-m radius. One or three near-surface soil samples from each plot were used for extraction and metabarcoding of soil-derived eDNA. Use of PCR replicates and appropriate filtering, plus a relevant reference metabarcode catalogue, provided taxon lists that reflected the local flora. There was high concordance between taxa recorded in plot vegetation and those in the eDNA, but floristic diversity was under-sampled, even at the scale of a 1-m radius plot. Most detected taxa grew within <0.5-1.0 m of the sampling point. Taxa present in vegetation but not in eDNA tended to occur further from the sampling point, and most had above-ground cover of <5%. Soil-derived eDNA provides a highly local floristic signal, and this spatial constraint should be considered in sampling designs. For palaeoecological or archaeological studies, multiple samples from a given soil horizon that are spatially distributed across the area of interest are likely to provide the most complete picture of species presence.

No abstract

No abstract