Your role
The scientist recruited for this role will be responsible for developing and maintaining the variational and ensemble based ECMWF data assimilation systems. The successful candidate is expected to take an active role in exploring and implementing new scientific ideas and technologies to advance DA science and their application to ECMWF Earth Data Assimilation system. A specific focus of the role is towards the continuous development of the ECMWF variational ensemble DA system and its extension to coupled Earth system components. The objective is to fully characterise the uncertainties in the assimilation cycle and use this information to improve the accuracy of the analyses and the skill of the forecasts. This development work will take place using both established variational/optimal estimation technologies and emerging machine learning methodologies.
Together with algorithmic developments, the role involves coding them into the ECMWF Integrated Forecasting System on a High Performance Parallel Computing infrastructure. The successful candidate will embrace the technical complexities of the job and be alert to the opportunities of the rapidly evolving computing infrastructure.
The scientists will be based in the Data Assimilation Methodologies team within the ESAS Section.
About the Earth System Assimilation/DA Methodologies Team
The Earth System Assimilation Section (ESAS) forms part of ECMWF’s Research Department. It develops and maintains state-of-the-art data assimilation techniques and infrastructure to bring together information from the forecast model and the global satellite and in-situ observation network to support the ECMWF numerical prediction systems. Activity covers all components of the Earth System (atmosphere, land, ocean and cryosphere) with the primary focus of improving the accuracy of weather forecasts. The techniques and infrastructure developed in ESAS are also being applied for environmental monitoring and prediction (e.g. atmospheric composition) and the generation of climate reference datasets (reanalyses).
Inside ESAS, the Data Assimilation Methodologies (DA) Team maintains and continuously develops the variational and ensemble-based assimilation infrastructure that is common to all the data assimilation activities at ECMWF. Increasingly, Machine Learning technologies are being integrated into the standard DA development workflows.