Research Scientist Position: Machine Learning to Improve Climate Models at the DLR Institute of Atmospheric Physics
2024-06-18 near Munich
The Department Earth System Model Evaluation and Analysis of the Institute of Atmospheric Physics at the German Aerospace Center (DLR-IPA) invites applications for a research scientist position in the field of improving climate models with machine learning (ML) as part of the ERC Synergy Grant "Understanding and Modelling the Earth System with Machine Learning (USMILE)".
The scientist will work with Prof. Veronika Eyring, Head of the department and Professor of Climate Modelling at the University of Bremen, and her team as well as collaborating institutions to develop an ML-enhanced version of the Icosahedral nonhydrostatic (ICON) Earth system model, ICON-ML.
The tasks include the following
New developments and application of machine learning methods for ICON-ML
-
Development of ML-based parameterizations for ICON-ML
-
Implementation of these ML-based parameterizations in the ICON Earth system model
-
Extensive testing of the performance of the ML-based parameterization in ICON-ML
-
Tuning of the resulting ICON-ML version
Conceptual preparation, planning and realization as well as coordination of simulations with the ICON and ICON-ML models and their evaluation with the Earth System Model Evaluation Tool (ESMValTool)
-
Conceptual preparation and planning of climate model simulations with ICON and ICON-ML
-
Execution of the climate model simulations
-
Evaluation of the simulations with the ESMValTool for historical simulations and various scenarios for the future.
We are looking for candidates with a PhD in a scientific / mathematical field and extended experience in climate modelling and the development of machine learning methods, especially as applied to climate modelling.
At the DLR Institute of Atmospheric Physics we provide excellent facilities with opportunities to work with world-renowned experts in the field of Earth system modelling, Earth observations, and machine learning. The department develops an ML-enhanced version of the ICON model alongside an evaluation system (ESMValTool) that supports the comprehensive evaluation of Earth system models in comparison to observations and to other models participating in the Coupled Model Intercomparison Project (CMIP). The ultimate goal is to improve climate models and projections with machine learning and spaceborne Earth observations for actionable climate science and technology assessments in aeronautics, space, transport, and energy research. For further reference of our work, please see Veronika Eyrings’s publications and our Github repository.
Please submit your application including a letter of motivation, research proposal, curriculum vitae, publication list, documentation of academic degrees and certificates, and two letters of reference here.
