Late Holocene climatic variability reconstructed from incremental data from pines and pearl mussels - a multi-proxy comparison of air and subsurface temperatures

Dendrochronological and sclerochronological records are mean series of arboreal and molluscan increments that are correctly aligned in time by rigorous dating. These records of tree rings and annual shell-growth increments exhibit climate signals that can be used to reconstruct fluctuations and trends in past climates. Here we present a multi-proxy reconstruction of temperature histories using a combination of dendrochronological and sclerochronological evidence. Regional curve standardization (RCS) was used to remove the non-climatic variations from dendrochronological and sclerochronological series prior to palaeoclimatic interpretation. Conventional and signal-free methods of RCS were compared. It was found that the signal-free methods produced more reliable chronologies and systematically higher climate proxy correlations. Consequently, the temperature reconstructions were derived using the chronologies constructed by this method. Proxy-based histories of summer (July August) temperatures were reconstructed from AD 1767 onwards. The compound use of proxies resulted in reconstructions that were of higher quality than single-proxy reconstructions. Further improvement of reconstructions was accomplished by the inclusion of lagging increment values in the transfer functions. The final multi-proxy model explained 58% of the temperature variance over the instrumental period. The multi-proxy temperature reconstruction correlated well with the long records of instrumental temperatures from Tornedalen, St. Petersburg, Uppsala and Stockholm. Overall, the reconstruction for the past 250 years agreed reasonably well with borehole temperature reconstructions obtained in northern Finland. In general, this study demonstrates the benefits of the compound use of several proxies in reconstructing climate histories. In particular, the study emphasizes the so far largely unexploited advantages of multi-proxy data sets obtained by rigorously cross-dated incremental chronologies to produce more robust palaeoclimatic reconstructions.