ALBERT

Description: Here, we provide the raw pollen data archived in three Siberian lake sediment cores spanning the mid-Holocene to the present (7.6-0 cal ka BP), from northern typical tundra to southern open larch forest in the Omoloy region. There are three cores: 1. 14-OM-20B, Lat. / °: 70.53, Lon. / °: 132.91, Ele. / m a.s.l.: 52, Modern vegetation: open larch forest, Lake area / km2: 0.26, Maximal depth / m: 3.4 2. 14-OM-02B, Lat. / °: 70.72, Lon. / °: 132.67, Ele. / m a.s.l.: 58, Modern vegetation: forest tundra, Lake area / km2: 0.08, Maximal depth / m: 3.5 3. 14-OM-12A, Lat. / °: 70.96, Lon. / °: 132.57, Ele. / m a.s.l.: 60, Modern vegetation: tundra, Lake area / km2: 0.09, Maximal depth / m: 4.5 Three lake sediment cores, 14OM12A (33 cm long), 14OM02B (49.5 cm long) and 14OM20B (86 cm long), were recovered from three sites using a UWITEC gravity corer (6 cm internal diameter) equipped with a hammer tool in July 2014. From the three cores, 16 bulk organic carbon samples were selected because of the lack of macrofossil remains and radiocarbon dated using accelerator mass spectrometry (AMS) at Poznań radiocarbon laboratory of Adam Mickiewicz University, Poland. In addition, 30 freeze-dried samples per core at 0.25 or 0.5 cm intervals between 0 and 15 cm were analysed for 210Pb/137Cs at the Liverpool University Environmental Radioactivity Laboratory. In this project, we analyse pollen and sedaDNA (Liu et al., 2020; doi:10.5061/dryad.69p8cz900) from three lake sediment cores from the Omoloy region in north-eastern Siberia (northern Yakutia), which are currently surrounded by different vegetation types ranging from typical tundra to open larch forest. First, our aim is to compare sedaDNA with the pollen data to see whether both methods track the same pattern with respect to compositional changes and diversity changes across the northern Russian treeline zone or are complementary to each other. Second, we reconstruct the mid- to late-Holocene changes of vegetation composition along a north–south transect. Third, we use the sedaDNA data to reconstruct variations in species richness and relate this to vegetation and climate change.

Description: Arctic treelines are facing a strong temperature increase as a result of recent global warming, causing possible changes in forest extent, which will alter vegetation-climate feedbacks. However, the mode and strength of the response is rather unclear, as potential changes are happening in areas that are very remote and difficult to access, and empirical data are still largely lacking. Here, we assessed the current population structure and genetic differentiation of Larix Mill. tree stands within the northernmost latitudinal treeline reaching ~72° N in the southern lowlands of the Taymyr Peninsula (~100° E). We sampled 743 individuals belonging to different height classes (seedlings, saplings, trees) at eleven locations along a gradient from 'single tree' tundra over 'forest line' to 'dense forest' stands and conducted investigations applying eight highly polymorphic nuclear microsatellites. Results suggest a high diversity within subpopulations (HE=0.826-0.893), coupled, however, with heterozygote deficits in all subpopulations, but pronounced in 'forest line' stands. Overall, genetic differentiation of subpopulations is low (FST=0.005), indicating a region-wide high gene flow, although 'forest line' stands harbour few rare and private alleles, likely indicating greater local reproduction. 'Single tree' stands, located beyond the northern forest line, are currently not involved in treeline expansion, but show signs of a long-term refuge, namely asexual reproduction and change of growth-form from erect to creeping growth, possibly having persisted for thousands of years. The lack of differentiation between the subpopulations points to a sufficiently high dispersal potential, and thus a rapid northward migration of the Siberian arctic treeline under recent global warming seems potentially unconstrained, but observations show it to be unexpectedly slow.

Description: Reliable information of past vegetation changes are important to project future changes, especially for areas undergoing rapid transitioning such as the boreal treeline. The application of detailed sedDNA records has the potential to enhance our understanding of vegetation changes gained mainly from pollen studies of lake sediments. This study investigates sedDNA and pollen records from 31 lakes along a gradient of increasing larch forest cover in northern Siberia (Taymyr Peninsula) and compares them with vegetation field surveys within the lake's catchment. With respect to vegetation richness, sedDNA recorded 114 taxa, about half of them to species level, while pollen analyses identified 43 pollen taxa. Both approaches exceed the 31 taxa revealed by vegetation field surveys of 400 m**2 plots. From north to south, Larix percentages increase, as is consistently recorded by all three methods. Furthermore, tundra sites are separated from forested sites in the plots of the principal component analyses. Comparison of ordination results by Procrustes and Protest analyses yields a significant fit among all compared pairs of records. Despite the overall comparability of sedDNA and pollen analyses certain idiosyncrasies in the compositional signal are observed, such as high percentages of Alnus and Betula in all pollen spectra and high percentages of Salix in all sedDNA spectra. In conclusion, our results from the treeline show that sedDNA analyses perform better than pollen in recording site-specific richness (i.e. presence/absence of certain vegetation taxa in the direct vicinity of the lake) and perform as good as pollen in tracing regional vegetation composition.

Description: Pollen productivity estimates (PPE) are used to quantitatively reconstruct variations in vegetation within a specific distance of the sampled pollen archive. Here, for the first time, PPEs from Siberia are presented. The study area (Khatanga region, Krasnoyarsk territory, Russia) is located in the Siberian Sub-arctic where Larixis the sole forest-line forming tree taxon. Pollen spectra from two different sedimentary environments, namely terrestrial mosses (n=16) and lakes (n=15, median radius ~100 m) and their surrounding vegetation were investigated to extract PPEs. Our results indicate some differences in pollen spectra between moss and lake pollen. Larix and Cyperaceae for example obtained higher representation in the lacustrine than in terrestrial moss samples. This highlights that in calibration studies modern and fossil dataset should be of similar sedimentary origin. The results of the Extended R-Value model were applied to assess the relevant source area of pollen (RSAP) and to calculate the PPEs for both datasets. As expected, the RSAP of the moss samples was very small (about 10 m) compared to the lacustrine samples (about 25 km). Calculation of PPEs for the six most common taxa yielded generally similar results for both datasets. Relative to Poaceae (reference taxon, PPE=1) Betula nana-type (PPEmoss: 1.8, PPElake: 1.8) and Alnusfruticosa-type (PPEmoss: 6.4, PPElake: 2.9) were overrepresented while Cyperaceae (PPEmoss: 0.5, PPElake: 0.1), Ericaceae (PPEmoss: 0.3, PPElake 〈0.01), Salix (PPEmoss: 0.03, PPElake 〈0.01) and Larix (PPEmoss 〈0.01, PPElake: 0.2) were under-represented in the pollen spectra compared to the vegetation in the RSAP. The estimation for the dominant tree in the region, Larixgmelinii, is the first published result for this species, but need to be considered very preliminary. The inferred sequence from over- to under-representation is mostly consistent with results from Europe; however, still the absolute values show some differences. Gathering vegetation data was limited by flowering season and low resolute satellite imagery and accessibility of the remote location of our study area. Therefore, our estimate may serve as first reference to strengthen future vegetation reconstructions in this climate-sensitive region.

Description: The Arctic treeline ecotone is characterized by a steep vegetation gradient from arctic tundra to northerntaiga forests, which is thought to influence the water chemistry of thermokarst lakes in this region.Environmentally sensitive diatoms respond to such ecological changes in terms of variation in diatomdiversity and richness, which so far has only been documented by microscopic surveys. We applied nextgenerationsequencing to analyse the diatom composition of lake sediment DNA extracted from 32 lakesacross the treeline in the Khatanga region, Siberia, using a short fragment of the rbcL chloroplast gene as agenetic barcode. We compared diatom richness and diversity obtained from the genetic approach withdiatom counts from traditional microscopic analysis. Both datasets were employed to investigate diversityand relationships with environmental variables, using ordination methods. After effective filtering of theraw data, the two methods gave similar results for diatom richness and composition at the genus level(DNA 12 taxa; morphology 19 taxa), even though there was a much higher absolute number of sequencesobtained per genetic sample (median 50,278), compared with microscopic counts (median 426). Dissolvedorganic carbon explained the highest percentage of variance in both datasets (14.2 % DNA; 18.7 %morphology), reflecting the compositional turnover of diatom assemblages along the tundra-taigatransition. Differences between the two approaches are mostly a consequence of the filtering process ofgenetic data and limitations of genetic references in the database, which restricted the determination ofgenetically identified sequence types to the genus level. The morphological approach, however, allowedidentifications mostly to species level, which permits better ecological interpretation of the diatom data.Nevertheless, because of a rapidly increasing reference database, the genetic approach with sediment DNAwill, in the future, enable reliable investigations of diatom composition from lake sediments that will havepotential applications in both paleoecology and environmental monitoring.

Description: This data set describes times series of diatom species counts, organic matter measurements including total carbon TC, total organic carbon TOC, nitrogen N, stable carbon isotopes d13C, and mercury Hg concentrations in four short cores (PG2133, PG2139, PG2203, PG2208 from Lake Bolshoe Toko, eastern Siberia. These short core have been dated using 210Pb and 137Cs methods covering the transition zone of the industrialization until 2013 CE. Methods are provided in the associated paper (Biskaborn et al. 2021).