Summer School on Data Assimilation in the context of Improving Earth System Modeling and Projections

Recent climate model developments, partly established through increased resolution, have led to substantial improvements in their skill to simulate the time-evolving, coupled Earth system as well as its subcomponents. However, regardless of their quality, climate models will always be expected to produce climate features and variability that will differ from the real world and will remain to be prone to biases. Further improvements therefore are necessary which are expected to arise from improved representation of physical processes realized through model-data fusion. This will create an unprecedented opportunity to better exploit a large array of Earth observations, from in situ measurements to weather radars and satellite observations, as the resolved scales of the models become closer to those of the observations. For this, climate data assimilation will be the central tool to bring models and observations into consistency, either by producing initial conditions, or by improving uncertain model parameters. 

In general terms, climate data assimilation must aim to enhance climate knowledge through improved ability to simulate and predict the real world by optimally combining Earth system models and all available global observations from different Earth system components and domains. For this, climate data assimilation needs to bring the simulations of climate models into consistency with the natural world as observed by the global climate observing system, and to produce a dynamically balanced climate estimate in support of initialized climate predictions, investigation of climate processes, and the identification and reduction of model bias. 

Various approaches are in use, ranging from filters – primarily ensemble filers - to whole-domain smothers usually realized through adjoint modeling. Their use typically concerns weather forecast and reanalysis in the atmosphere and the ocean. More recently machine learning (ML) gains attention. In the future, however, arguably the most important aspect of climate model data assimilation and ML in support of model improvement and to enhance predictive capabilities, might become optimizing model parameters to mitigate model biases and thereby improve the model’s skill in simulating the observed climate and enhance model predictability and skill for climate projections. 

A summer school will be held at Universität Hamburg during the week August 26 – August 30. The school will host up to 40 students from around the world. Registration is open until the maximum number is reached as first come first serve. 

A full program of the school will be circulated by the end of March, 2024. Details about the school, enrolled participants and lecturers and the evolving program can be found under: https://www.spp-sealevel.de/news/detailpage/summer-school-on

The school will consist of lectures on general and theoretical aspects of data assimilation combined with a multitude of hands-on experiments where students will gain first-hand experiences in performing data assimilation exercises. Attending the school will be free of charge and will include coffee breaks and lunch breaks. Travel to Hamburg and housing needs to be covered by the students or their supervisors.  

A central goal of this summer school on climate data assimilation is to provide students the insight into different data assimilation approaches and their use in weather forecasting and climate modeling and to provide skill to use the methods. The summer school will introduce the various methods and their application, highlighting advantages and complementarities of adjoint-based smoother approaches, ensemble-based filter approaches, or other ML approaches.Emphasis will also be placed on the computation of errors of the resulting estimates. Students will also be introduced to the rich data base available for climate data assimilation as well as their uncertainties. 

Interested students should register their interest by sending an email to Detlef Stammer (detlef.stammer@uni-hamburg.de) and Armin Köhl (armin.koehl@uni-hamburg.de), combined with a reference letter form their advisors. The letter should state the suitability of the student for an assimilation training school. It should also attest the availability of travel accommodation funds for the student to come to Hamburg. 

The School will be sponsored by WCPR/ESMO and will receive financial support from COPERNICUS, ESA and EUMETSAT.