Reconstructing temperature variations at high elevation lake sites in Europe during the instrumental period

Abstract

The longest instrumental climate records that exist anywhere in the world are those from lowland Europe. In contrast, instrumental climate records from European upland areas are generally short and scarce. The study of mountain climate is nevertheless important, since alpine areas represent some of the most pristine and unpolluted environments on Earth, and provide a variety of palaeoclimate proxy records (e.g. mountain lake sediments and alpine glaciers). Furthermore, air-temperature anomalies at mountain sites in the European Alps show an amplification of the climate signal with respect to nearby lowland stations (BENISTON et al. 1997), emphasizing the value of mountain areas for the study of climate change. Our aim here is to construct a statistical model to provide transfer functions from lowland to upland climates. This is made possible by the high correlation we find between lowland and upland temperature records (after removing seasonal effects). Our statistical model, which is based on stepwise multiple linear regression, provides the basis for the climate reconstruction at several remote mountain lakes in Europe now being studied as part of the MOLAR project. Climatic forcing is important for high altitude lakes (SOMMARUGA-WOGRATH et al. 1997). In particular, climate has an important effect on the duration of ice cover (e.g. RUOSTEENOJA 1986) and growing season (MELA 1996). Both ice cover and growing season condition the seasonal evolution of the lakes chemically (e.g. pH), physically (e.g. lake stratification and circulation) and ecologically (e.g. productivity). This paper describes the use of reconstructed surface air temperatures to derive long series (1781–1997) of estimates of ice-cover and growing season duration for five alpine lakes in Europe.