Partners
National Technical University of Athens
The Remote Sensing Lab of the School of Rural, Surveying and Geoinformatics Engineering (SRSE) of the National Technical University of Athens, leads and coordinates the MeDiTwin project. Harnessing the scientific leadership of each partner in AI for Earth Observation, emphasizing in physics-guided ML, explainable AI, and causality and cloud computing for climate emergencies, is responsible for consolidating and developing the main elements of this initiative.
Max Planck Institute for Biogeochemistry
The Model-Data Integration Group of the Max Planck Institute for Biogeochemistry (MPG) is a leading research group focused on global biogeochemical cycles and their role in the climate system. Combining observational studies on soil carbon, plant growth, and vegetation-atmosphere interactions with global-scale modelling, MPG advances our understanding of Earth’s biogeochemical processes. In collaboration with universities and research institutes, MPG is developing AI-driven methods, including deep learning and hybrid modelling, to predict and analyse climate extremes and their impacts. In the MeDiTwin project, MPG is responsible for predicting climate extremes and their impacts in the Mediterranean, focusing on enhancing the modelling of wildfire CO₂ emissions.
University of Valencia
The University of Valencia’s Image and Signal Processing group of the Image Processing Laboratory (IPL) contributes to the MeDiTwin project by developing AI methods for understanding complex Earth and climate systems. ISP’s expertise includes imaging from satellite data, remote sensing, and causal modelling, with research focused on climate extremes like droughts, wildfires, and floods. Leading the project’s work on climate modelling, UVEG is developing physics-guided causal models and explainable AI (xAI) to understand and improve extreme weather predictions and assess cascading impacts of climate events.
European Centre for Medium-Range Weather Forecasts
The European Centre for Medium-Range Weather Forecasts (ECMWF) is an international organization specializing in global numerical weather predictions and climate data services. Known for its advanced high-performance computing (HPC) facilities and vast data repository, ECMWF leads in producing high-quality weather forecasts and climate reanalysis for Europe and beyond. In the MeDiTwin project, ECMWF is leading the creation of the Mediterranean digital twin prototype, over the cloud/High-Performance Computing (HPC) to simulate climate extremes and enhance the current predictive accuracy.