This PhD project is funded by the Swiss National Science Foundation (SNSF) and it will be conducted in the climate section of MeteoSwiss (Zurich Airport). The supervision of the project is shared between Dr. Christoph Frei (MeteoSwiss, topical supervision) and Prof. Heini Wernli (ETH Zürich, main PhD supervisor). The ideal starting date is October/November 2021 or soon thereafter.
Climate variables are commonly measured at weather stations. Deriving a complete spatial field from these local measurements is a challenging problem that needs to combine knowledge of atmospheric processes with modern statistical methods. “Spatial climate analysis” develops climate datasets on a high-resolution grid and over several decades into the past. The datasets are widely used for modelling environmental processes (e.g., in hydrology and ecology), for monitoring climate change, and for evaluating numerical weather prediction models.
This PhD project tackles the major limitation that current datasets are restricted to the daily time resolution. The candidate will conceive, implement and test novel methods that allow advancing into the sub-daily time scale in order to resolve the diurnal cycle of climate variables. To this end, methods of spatio-temporal statistical modelling will be exploited, which allow integrating data over space and time. The project builds on theoretical progress in spatio-temporal modelling over the past years. The focus is on surface air temperature and the challenges posed by complex topography.
We are looking for a highly motivated and curious PhD candidate with a MSc degree in applied statistics, environmental / earth science or physics. A solid background in statistical concepts and modelling is required, ideally combined with experience in climate data analysis and knowledge in meteorology, atmospheric physics and observations. The candidate should have a passion in conducting complex data analyses, creativity in combining climatology with statistics, an interest in the practical utility of own developments, and good programming skills (ideally in R). Very good spoken and written English language skills are required for the presentation of results at international conferences, and for the reading and writing of scientific papers.