Summer temperature, growing season length, and precipitation reconstructions from near-annual proxy-records in northern Sweden
|Location||International Geological Congress,oslo 2008|
|Author||Barnekow, Lena۱; Finsinger, Walter۲; Loader, Neil J.۳; Schoning, Kristian۴; Wagner-Cremer, Friederike۵|
|Holding Date||27 September 2008|
To place recent climate change in a longer term context, several studies have developed millennial length, annually resolved reconstructions of northern hemispheric temperatures. There is, however, the need for complementary investigations at smaller scales with the expressed aim of reconstructing other climate parameters and not just temperature. Here we report on a multi-proxy study involving pollen, leaf cuticles of Betula nana, and testate amoebae that were analysed in a peat profile from Kiruna (northern Sweden). The aim was to compare these proxy records with instrumental meteorological records in order to explore their potential for climate-reconstruction purposes. Multiple AMS radiocarbon dating and the use of the post atomic-bomb-test period (AD 1961-2002) provided a robust temporal control including an exceptionally high (near annual) resolution in the uppermost part of the profile. The pollen spectra analysed show strong correlation between summer temperature and pollen accumulation rates (PAR) for Pinus sylvestris, Picea abies, and Betula spp. (excluding B. nana). The strongest correlation was observed between PAR of P. abies pollen and July and August mean temperatures (r2adjusted=0.53; n=36).
A training-set for growing season length (GSL) determination is based on B. nana leaves, annually collected between AD 1998 and 2007. For this calibration period, the sinuosity of epidermal cells measured on the leaf cuticles is strongly correlated to the thermal sum >5°C (r2=0.63; n=19), which is a measure for GSL. Sub-fossil leaf macro-remains collected from the Kiruna peat profile exhibit, indeed, a high sinuosity in years that had extended GSL according to the (historical-) instrumental records.
The testate amoebae assemblages show a good correspondence with annual precipitation and a transfer function for the relationship between testate amoebae assemblages and precipitation was constructed with r2=0.69 (jacked knifed). Testate amoebae derived water table changes as well show a good correlation with annual precipitation.
The results show that it is possible to provide a direct link to meteorological variables and expand the knowledge to study palaeoclimate variations. The pollen data support the results from modern pollen monitoring and demonstrate that the summer temperature signal is preserved within the peat archive. High precision and accuracy of the dating and subsampling are essential to be able to establish relationship between proxies and meteorological variables. Whilst care should be taken to minimise the influence of signal integration between levels (reflecting variability in the topography of the mire surface and the effects of sub-sampling), this study shows the strengths and possibilities of (high resolution) analysis of the wide range of environmental