New approach to assessing age uncertainties – The 2300-year varve chronology from Eklutna Lake, Alaska (USA). (1st January 2019)

• Record Type: Journal Article
• Authors: Fortin, David; Praet, Nore; McKay, Nicholas P.; Kaufman, Darrell S.; Jensen, Britta J.L.; Haeussler, Peter J.; Buchanan, Casey; De Batist, Marc
• Is Part Of: Quaternary science reviews. Volume 203(2019)
• Page Start: 90
• Page End: 101
• Publication Date: 2019-01-01
• Subjects: Varve -- Geochronology -- Age model -- Tephra -- Alaska; Geology, Stratigraphic -- Quaternary -- Periodicals; Stratigraphie -- Quaternaire -- Périodiques; 551.79
• DOI: 10.1016/j.quascirev.2018.10.018
• Languages: English
• ISSNs: 0277-3791

Abstract

Abstract: Developing robust chronological frameworks of lacustrine sediment is central to reconstructing past environmental changes. We present varve chronologies from five sites extending back 2300 years from Eklutna Lake, in the Chugach Mountains of south-central Alaska. The chronologies are built from image analysis of high-resolution photographs and CT scans of sediment cores. The age uncertainty of each record is tested by three methods. We first present varve chronologies from individual sites and reconcile the difference in varve delimitation from two observers. The varve chronologies from each site are then compared to each other using a series of marker beds that can be traced across the lake basin. Finally, using a new Bayesian probabilistic model, we develop age models that incorporate information regarding age uncertainty from the multiple-observer method and the age distribution of marker layers from multiple cores. To evaluate the accuracy of the Bayesian model output, we used seven radiocarbon ages from terrestrial macrofossils and four tephra layers traceable across the core sites. The major-element geochemistry of the tephra layers and their ages are presented here for the first time. The Bayesian age model offers a new approach to quantifying age uncertainty in inter-correlated cores of varved sediment. Highlights: New cores from Eklutna Lake contain the longest continuous varve chronology published from Alaska, extending back 2300 years. Bayesian probabilistic model integrates uncertainties from multiple observers and counts between marker beds in multiple cores. Analyzing more cores from the same lake increases the apparent age uncertainty, but improves understanding of likely errors. Four tephra layers are now precisely dated; new geochemical data indicate Redoubt and Augustine source volcanoes.