Imagine trying to predict the weather while the ingredients of the air are changing. As we move away from burning coal and oil, the particles in our atmosphere shift from man-made pollution to natural sources, which changes how clouds form and rain falls. This work creates better digital maps and tools to predict these changes so we aren't surprised by extreme weather.
Current climate projections are unreliable because we don't understand how clouds react to the shift from man-made pollution to natural aerosols. This uncertainty makes it difficult for businesses to predict extreme weather and hydrological changes.
A set of state-of-the-art algorithms for satellite proxies, improved kilometer-scale climate models using machine learning, and validated data from Arctic and Mediterranean field campaigns.
If you are a crop insurance provider dealing with unpredictable extreme precipitation—this project developed improved kilometer-scale climate models that provide more accurate regional weather forecasts. This allows for better risk pricing and loss estimation in a changing climate.
If you are a satellite instrument manufacturer dealing with sensor calibration errors—this project developed state-of-the-art algorithms and validation data for upcoming satellite missions. This ensures your hardware delivers precise aerosol and cloud measurements.
If you are a grid operator dealing with volatile solar and wind output due to cloud cover—this project developed advanced machine learning and data assimilation tools to better constrain climate models. This leads to more reliable seasonal predictions for energy production.
Based on available project data, no pricing or commercial cost structure is mentioned as the project is a research-driven Horizon-RIA initiative.
The project focuses on improving kilometer-scale and large-scale climate models and satellite algorithms, which are inherently scalable for global weather and climate services.
Based on available project data, specific licensing terms are not provided; however, it is a consortium of 20 partners primarily from universities and research centers.
The project uses machine learning and data assimilation to improve current and next-generation climate models, suggesting integration via software updates to existing meteorological tools.
The project period runs from 2024-01-01 to 2027-12-31, with results emerging through 7 total deliverables.
The consortium is heavily weighted toward academic and public research, consisting of 11 universities and 8 research organizations across 12 countries. With 0% industry participation and 0 SMEs, the project is focused on fundamental scientific breakthroughs and model improvement rather than immediate commercialization. This suggests the primary output will be open-access data, algorithms, and peer-reviewed climate projections.
Contact Aarhus Universitet (DK) regarding climate model calibration
Contact us to find partners for implementing these climate models in your business strategy.
