全球变化知识资源中心

Lake surface temperature (LST) is a key characteristic of lakes, shaping the ecological properties of these inland water bodies and their environment. This study aims at establishing a long-term, high-quality, monthly LST dataset within the European Alps reaching back to 1880, which is provided to the scientific community for further research. Therefore monthly temperature records from Austrian lakes covering a period of about six decades are digitized from hydrological yearbooks. Clustering techniques (rotated empirical orthogonal functions and hierarchical cluster analysis) are used to identify groups of lakes signified by inner similarity and outer separation. These are not only used for an overall quality assessment, but also provide optimal starting conditions for the application of a homogenization procedure, warranting homogeneous LST data from 1950 onwards. LST reconstructions back to 1880 are derived from atmospheric covariates (provided by HISTALP) via sets of transfer-functions. Applied transfer-functions have passed a selection process ensuring mathematical, physical and quality requirements. They are selected from about 160 million candidates according to skill, which is determined through a comprehensive assessment based on validation experiments and several performance measures. Results show overall high skill following a seasonal cycle. From 1880 to 1950 LSTs feature generally slight increases accompanied by a succession of climate states. These are in alignment with outstanding climate periods and sustained, far-reaching repercussions triggered by significant events, which are known from historical documents. LST developments throughout the second half of the twentieth century up to date are characterized by a decline until the mid-1980s, indicating the impact of industrial aerosols. This behaviour is superseded by a steep increase, revealing the gradual unmasking of the anthropogenic greenhouse effect by the continuous reduction of aerosol loads in the atmosphere. The latter substantiates man-made climate change, whose prove is based on atmospheric variables by different data pertaining to the hydrosphere. Potential research hypotheses corresponding to various fields of science that may be investigated by using the here established LST dataset are considered. We hope that these data sets and associated findings will make a contribution to the broader research community.