SLAC-SD - Sea and Land Surface-Atmosphere-Climate Interactions within the Framework of Statistical Downscaling

SLAC-SD. Relevance of sea and land surface-atmosphere-climate interactions within the framework of statistical downscaling for Europe and the Mediterranean area.

Start date: 01.01.2015
Duration: 3 Years
Funded by: DFG (Deutsche Forschungsgemeinschaft)
Local project leader: PD Dr. Elke Hertig Dr. Karin Romberg


Land-atmosphere and ocean-atmosphere interactions are of considerable importance in the global climate. Major challenges in current climate research are the impact of sea surface and land processes on climate variability and change as well as on extreme events. Within this context it is important to assess the dynamic and physical mechanisms by which the sea and land surfaces influence atmospheric processes and climate. Up to now most studies concentrated on numerical model studies to address the atmospheric and climatic response to slowly varying states of the ocean and land surfaces. However, with the coarse resolution and still relatively simple physical parameterizations, current Earth System Models (ESMs) can only provide limited information on local to regional scales. Thus, it is important to downscale the large-scale model output to smaller scales in order to provide fine-scale climate change information. In this project the aim is to explicitly introduce sea surface and land surface characteristics as predictors in the framework of statistical downscaling. In this context it is essential to analyse these variables in the observational data sets as well as in the ESMs. Using multivariate statistical analysis, the general representation of Soil Moisture (SM) and Sea Surface Temperatures (SSTs) and the specific SM/SST-atmosphere-climate interactions in and between the observations and the ESM data are investigated. Subsequent analyses are devoted to the transferability and to the modifications of the observed predictor-predictand relationships under the assumption of future climate change. This will lead to a better understanding of the processes involved and to an assessment of the potential for improvement of statistical downscaling models through the inclusion of SM and/or SSTs as additional predictors. Analyses will concentrate on temperature and precipitation in the European/ Mediterranean area. The selection of this domain makes it possible to capture potential climate change induced spatial shifts of sea and land surface-atmosphere-climate interactions. Besides mean values, temperature and precipitation variability and extremes will be addressed. Finally, the project aims to contribute to an enhanced understanding of land and sea surface-atmosphere-climate interactions in observations and model data, to evaluate and improve the performance of statistical downscaling techniques and thus to reduce uncertainties of climate change projections on regional to local scales.