Institute of Atmospheric Physics
Area of research:
Scientific / postdoctoral posts
The Institute of Atmospheric Physics develops innovative methods for the evaluation and analysis of Earth system models in comparison to observations with the aim of better understand and project the Earth system.
The evaluation and ensemble analysis of Earth system models is crucial for model improvements and a prerequisite for reliable climate projections of the 21st century to be used as guidelines for climate policy.
Together with many international partners, the institute is leading the development of an open-source software tool to take the evaluation of complex Earth System Models with observations to the next level (ESMValTool). The goal of this internationally well-recognized effort is to improve comprehensive and routine evaluation of Earth System models participating in the Coupled Model Intercomparison Project (CMIP). Together with partners from Jena, Valencia and New York, the institute has recently been awarded a renowned and prestigious European Reserach Council (ERC) Synergy Grant for the project „Understanding and Modelling the Earth System with Machine Learning" (USMILE). Within this project it is planned to close gaps in the understanding of small scale physical and biological processes such as clouds, stomata and microbes within the climate system. The aim of USMILE is to better understand and model the changes and feedbacks of these processes and their impacts on the Earth’s ecosystems with machine learning.
Within the framework of the research environment outlined above, the position involves the following tasks:
Technical development and adaption of machine learning techniques for the analysis of Earth system data monitoring of new developments within the field of Artificial Intelligence and machine learning techniques and providing regular updates to the department advising the climate scientists of the department on the selection of machine learning techniques for specific scientific applications programming work to further develop and adjust these machine learning techniques for analyses of Earth system data exemplary application of machine learning techniques to Earth system data technical support of department members with implementation and development of machine learning techniques on different high performance computing systems Technical development of the ESMValTool with focus on user support re-formatting of ESA CCI satellite data sets to a CMOR compliant format adding and re-formatting existing and new observational data sets and diagnostics in the ESMValTool support of the ESMValTool development cycles, software engineering and error analysis including debugging development and performance of regular tests on different computing platforms, and performance optimization based on the test results automatization of quality control tests, and tests of the ESMValTool integrity and functionality preparation of the technical ESMValTool documentation