Varve series cross-correlation and palaeosecular variation from two clay varve basins in western Estonia

Work on clastic varves has advanced to further our understanding of Pleistocene climate including processes of deglaciation. In this study we provide a correlation of two varve chronologies representing different basins (Pärnu and Vigala) in western Estonia of the eastern Baltic Sea. Via regional correlations of palaeomagnetic secular variation we date the deglaciation of coastal Estonia. The decay of Late Weichselian ice from Estonian territory between about 14.7 and 12.7 cal ka BP was followed by extensive proglacial bodies of water, which developed in accordance to the receding ice margin and opening of new drainage routes. This is reflected in a wide distribution of clastic varved clays. Using visual and numerical methods we correlate two varve sequences from the Pärnu basin representing previously established 584-varve-year chronology and a new 532-varve-year sequence from the Vigala basin. Our analysis indicates that varve formation in the ice-proximal Vigala basin commenced approximately 140 years later than in the more distal Pärnu basin, yielding a combined local chronology of about 670 varve years. Natural remanent magnetization relative declination measured from the studied three varve records reveal an initial continuous eastward shift of approximately 90° during the first 400 varve years, after which values stabilize. Regional correlation of these palaeo secular variation records dates the clay accumulation in western Estonia to between ca 14.0 and 13.3 cal ka BP. More precisely, our data estimates the ages for the Pandivere-Neva and the Palivere ice-marginal complexes in western Estonia to ca 13.9–13.8 cal ka BP and younger than 13.3 cal ka BP, respectively, both thus, immediately predating Younger Dryas Salpausselkä ice-marginal complex in southern Finland. These results provide an independent age estimate for deglaciation of coastal Estonia and are expected to contribute to regional deglaciation models.