A 7500-year record of plant physiology as palaeoenvironmental proxy from tree-ring δ13C and growth rates - the CARATE project

This contribution presents a recently launched 4-year research project, "CARATE", aiming to produce new climatic and plant physiological records at high temporal resolution by the synthesis of annually/decadally resolved tree-ring archives of growth-rates and cellulose δ13C values from high-latitude continental Europe. The project mainly relies on a supra-long pinewood chronology from Finnish Lapland covering the mid and late Holocene times continuously through the millennia since 5634 B.C. until the present-day (Eronen et al. 2002, Helama et al. 2008). The chronology provides a highly sensitive, absolutely dated proxy record of past environment and climate variability at highest possible resolution that can be calibrated directly with instrumental records of environmental variables. While growth rate records are invaluable tools for paleoclimate research at high frequencies, tree-ring δ13C compositions have the potential to portray the high-to-low-frequency climate signals per se, without complications from time-series analysis. By combining isotope and growth rate information, we aim to reconstruct regional high- and low-frequency climate variability over the past 7500 years with increased reliability, and explore the climate forcings behind the observed variations. The project was started by studying the strength of the common climatic signal both within- and between-sites, and possible associations to tree-physiological parameters. For this purpose, a set of 70 living pine trees (Pinus sylvestris) , growing in proximity of the subfossil pinewood collection sites in western and eastern Lapland, were cored for analysis of growth rates and δ13C values. α-cellulose was extracted from decadal samples corresponding to the time period 1970-2010 including both early- and latewood. The δ13C time series show a consistent response to regional environmental forcing over the entire study area, with no discernable differences between western and eastern Lapland. Within a single site, a typical (e.g. Leavitt 2010) between-tree variability of 1-3‰ in δ13C values is observed. The strength of the within-site signal is demonstrated by high correlations (>0.87) between the δ13C chronologies from individual trees (n=8-10 per site). Mean δ13C time series calculated for each site are also highly correlated (>0.91) with each other. In the following months, the δ13C responses will be studied in terms of tree-physiological variables measured on site, and selected trees are sampled in annual resolution for the construction of a calibration chronology. References: Helama S., et al. 2008, Norwegian Journal of Geography 62, 271-277. Eronen M., et al. 2002, The Holocene 12, 673-680. Leavitt, S.W. 2010. Science of the Total Environment 408, 5244-5253.