Print

Earth System Model Evaluation and Analysis

Head of department: Prof. Dr. habil. Veronika Eyring

The department 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 guide-lines for climate policy. To improve routine and comprehensive evaluation of climate models, the department leads the development of the Earth System Model Evaluation Tool (ESMValTool). A key focus of the department is the development and application of machine learning methods to improve understanding and modelling of the Earth System.

Our research topics are:

• Analysis of Earth system model simulations in combination with observations to better understand the processes of climate change and the climate system.
• Development of ‘Emergent Constraints‘ to constrain key climate feedbacks in the Earth System with observations
• Constraining uncertainties in multi-model climate projections with observations
• Identification of systematic biases in Earth system models and recommendations for model improvements
• Understanding and modelling the Earth System with Machine Learning, including the development of machine learning based parametrizations for climate models
• Development and application of innovative methods to analyse large volumes of data
• Development of a diagnostic tool for efficient routine evaluation of ESMs with observations
• Contributions to international Model Intercomparison Projects (coordination, model simulations, and analysis), in particular to the Coupled Model Intercomparison Project (CMIP).

Our main tools and data are:

• The ESMValTool for routine and improved evaluation of Earth System models with observations
• Machine learning methods to improve climate models and their analysis
• Simulations with the Icosahedral Nonhydrostatic (ICON) climate model and with the global atmosphere-chemistry model EMAC (ECHAM/MESSy Atmospheric Chemistry)
• Simulations with Earth System models participating in international Model Intercomparison Projects (in particular CMIP)
• Observations and meteorological reanalyses from Observations for Climate Model Intercomparisons (obs4MIPs), Analysis for Model Intercomparison Projects (ana4MIPs) and other sources

Close collaboration exists with the Department of “Climate Modelling” of the University of Bremen (Chair: Prof Veronika Eyring), with the ”Climate Informatics Group” of the DLR Institute of Data Sciences, as well as with the co-PIs of the European Research Council (ERC) Synergy Grant “Understanding and Modelling the Earth System with Machine Learning (USMILE)”: Prof. Markus Reichstein (Max Planck Institute for Biogeochemistry, Jena), Prof. Gustau Camps-Valls (University of Valencia, Spain), and Prof. Pierre Gentine (Columbia University, New York, USA). The department is strongly linked to international research activities within the World Climate Research Programme (WCRP), with substantial contributions particularly to CMIP, and contributes regularly to international climate and ozone assessments of the Intergovernmental Panel on Climate Change (IPCC) and the World Meteorological Organization (WMO).