Development and application of an electrical conductivity transfer function, using diatoms from lakes in the Urals, Russia

• DOI: 10.1007/s10933-019-00106-z
• Publication Date: 2020-01-02T13:02:44Z
• Authors: Anna Valer’evna M...
• Keywords: 13. Climate action; 15. Life on land; 01 natural sciences; 6. Clean water; 0105 earth and related environmental sciences

Despite a relatively large number of paleoenvironmental and archaeological studies in the Southern Urals, Russia, Holocene moisture dynamics in the region remain a subject of debate. Electrical conductivity (EC) of lake waters in the Southern Urals is climate-dependent and associated with effective moisture. I developed a transfer function using a 72-lake regional diatom dataset and inferred past EC using diatoms in sediment cores from three lakes. The diatom-inferred EC inferences were compared with pollen records to validate the climate interpretation and explore the possible influence of vegetation on lake hydrochemistry. Multivariate analysis showed a clear response of diatom communities to electrical conductivity across a range from 55 to 3780 μS cm−1. The best electrical conductivity inference model (r2boot = 0.78, RMSEPboot = 0.21 log10 μS cm−1), suitable for quantitative reconstructions, was developed using weighted averaging with classical deshrinking. Application of the transfer function to sediment cores from mountain lakes in the forest and forest-steppe regions of the Southern Urals revealed relations between EC and variations in pollen contents of herbs, pine (for the forest-steppe lake) and spruce (for the forest lakes). Although a decrease in EC may be attributable to coniferous forest spread in the watersheds, it is more likely that climate fluctuations caused the coincident shifts in vegetation and lake-water hydrochemistry. Forest lakes were marked by a declining trend in EC since the beginning of the Holocene (11.9–11.5 cal ka BP), which was independent of the larch and birch forest distribution (11.5–10.0 cal ka BP) and was interrupted by minor climate-driven EC increases. Electrical conductivity of lake water was higher than present in the period between 6.5 and 3.8 cal ka BP in all three study lakes. Forest lakes were characterized by gradual increases and decreases in EC, with maximum values occurring ca. 5.0–4.0 cal ka BP. Diatoms in the core from the forest-steppe lake record variations in EC through time. In particular, the record shows highest EC values at 5.3 cal ka BP, a decline in EC at 4.1–3.8 cal ka BP, a slight increase about 3.6 cal ka BP and a substantial decrease since 2.8 cal ka BP.