Meet the experts: Destination Earth (DestinE)
Meet some of the experts from ECMWF and partner organisations who are helping to build Destination Earth (DestinE) with our latest video series. Discover the different components of the initiative through the people behind them, learn more about the expertise driving Europe’s digital twins of the Earth and what excites them about beeing part of this European initiative.
Developing digital twins of the Earth within the European Commission’s DestinE initiative is a truly European effort. Together with partners across Europe, ECMWF is building the Climate Change Adaptation Digital Twin (Climate DT), the Weather-Induced Extremes Digital Twin (Extremes DT) and the Digital Twin Engine, while advancing machine learning (ML) and artificial intelligence (AI) capabilities for DestinE. This work brings together experts from a wide range of disciplines, all contributing their knowledge and expertise to different parts of the initiative.
AI and machine learning are an important part of this work. ECMWF and its partners are scaling up ML-driven models and developing AI-based software solutions to further enhance DestinE’s capabilities. A key area of this work is the development of ML-based Earth system components, including for hydrology, ocean waves, sea ice, the ocean and land surface. These developments explore how machine learning can complement existing approaches to modelling different parts of the Earth system. DestinE’s ML activities are closely linked and complementary to the wider AI and ML efforts led by ECMWF together with its Member States. Underpinning these advances is access to Europe’s world-class high-performance computing infrastructure. By leveraging the computing capabilities provided by the European High Performance Computing Joint Undertaking (EuroHPC JU), DestinE brings together cutting-edge Earth system modelling, artificial intelligence and high-performance computing at an unprecedented scale.
Behind all these developments are scientists, software engineers and technical experts working together across organisations and countries. In this video series, meet some of the people behind DestinE and discover their work. They explain what they are contributing to the initiative, why their work matters and what excites them about working together to build a digital twins of the Earth system for Europe.
Irina Sandu
– Deputy Director General and Director of Science and Innovation at ECMWF, former Director for Destination Earth at ECMWF
Irina Sandu introduces DestinE and explains how the initiative is pushing the boundaries of science by combining advanced Earth system modelling, artificial intelligence and EuroHPC infrastructure.
Sebastian Milinski
– Senior Scientist, ECMWF
Sebastian Milinski leads the development of the Climate Change Adaptation Digital Twin (Climate DT) at ECMWF. Working with partner institutions across Europe, he is helping to build an operational system that produces global climate projections at kilometre-scale resolution to better support climate change adaptation.
Sophie Buurman
– Data Scientist, Royal Netherlands Meteorological Institute (KNMI)
Sophie Buurman works on ensemble regional modelling within DestinE, including the development of a regional multi-domain model. Her work contributes to advancing regional modelling capabilities and exploring how the Earth system can be represented at increasingly high levels of detail.
Matthew Chantry
– Strategic Lead for Machine Learning at ECMWF
As Strategic Lead for Machine Learning at ECMWF, Matthew Chantry oversees and coordinates machine learning activities within DestinE. He explains how the initiative is pushing the frontier of AI-based modelling across different parts of the Earth system and across timescales, from short-range weather prediction to sub-seasonal and climate timescales.
Ana Prieto Nemesio
– Former Team Lead for the ML Engineering Team, ECMWF
Ana Prieto Nemesio develops AI tooling for DestinE, with a particular focus on Anemoi – the open-source machine learning framework jointly developed by ECMWF and several of its Member States. Anemoi provides the tools needed to build, train and run data-driven models and underpins the development of ML-based Earth system components within DestinE.
Cathal O’Brien
– Computational Scientist, ECMWF
Computational Scientist Cathal O’Brien works on optimising Anemoi, the open-source software framework developed by ECMWF and several national meteorological services which underpins the development of ML-based Earth system models.
Sara Hahner
– Machine Learning Scientist, ECMWF
One of the key areas being explored within DestinE is the development of ML-based models for different components of the Earth system. Machine Learning Scientist Sara Hahner is working on a new AI-based wave model designed to emulate ocean wave dynamics. Her work explores how machine learning can contribute to modelling the ocean waves more efficiently, with potential applications in areas including maritime safety, renewable energy and coastal climate adaptation.
Fernando Iglesias Suarez
– Climate Scientist, Predictia
Bringing different ML-based Earth system components together requires them to work as part of an integrated system. Climate Scientist Fernando Iglesias-Suarez from Predictia is part of the team developing a machine learning climate emulator for DestinE. The emulator aims to replicate key physical properties of the Climate DT using machine learning, supporting the exploration and integration of AI-based approaches within DestinE’s evolving Earth system modelling capabilities.
Harrison Cook
– Research Software Engineer for Machine Learning, ECMWF
Harrison Cook works on the inference capabilities of Anemoi, helping ensure that machine learning models developed using the framework can move from research into operational use. He also contributes to Earthkit, a set of open-source tools developed by ECMWF and designed to make accessing, processing and analysing weather and climate data easier.
Jakob Schlör
– Machine Learning Scientist, ECMWF
Jakob Schlör works on data-driven sub-seasonal weather forecasting, extending predictions beyond the two-week range of medium-range weather forecasts. Improving predictions at these longer timescales has the potential to support earlier preparedness for extreme weather and decision-making in sectors including agriculture, energy and disaster management – areas that are particularly relevant to the goals of DestinE’s Digital Twins.
Mariana Clare
– Machine Learning Scientist, ECMWF
ECMWF’s Artificial Intelligence Forecasting System (AIFS) is ECMWF operational machine learning model which will also feed in the AI Earth System Model developed in DestinE. Machine Learning Scientist Mariana Clare works on advancing AIFS, with a particular focus on improving the accuracy and physical consistency of its forecasts.