US and EU begin planning for the development of a coordinated Digital Twin to better understand and mitigate the impact of tropical cyclones in the Caribbean

Under the framework of the US - EU Trade and Technology Council (TTC) 2023 Administrative Agreement on AI and the general collaboration on ‘Tech/AI for the Public Good’, multiple US agencies and the European Commission’s Destination Earth (DestinE) initiative have begun the development of an ecosystem of interoperable digital twin (DT) replicas of Earth systems. As extreme weather, like tropical cyclones (TCs), and other climate events, become more frequent and destructive, DTs of earth systems can help address event mitigation and better inform decision makers. By leveraging AI/ML technology to fuse multi-platform earth observations with accelerated large multiscale simulations these DTs will drive better modeling of events in highly complex climate systems, including prediction and simulation that can support disaster preparedness and societal outcomes.

The impact of TCs in terms of economic loss and human suffering can be significant, particularly in Caribbean island nations. Through this US and EU partnership the intent is to provide a working tropical cyclone toolkit to access global environmental data with advanced local end-user focus. This initiative aims to develop a suite of integrated DTs that will enable users to rapidly run hypothetical what-if scenarios to improve our understanding, prediction, mitigation, and adaptation to these extreme storms. First, however, there is a need to further develop the interoperability of DT’s and address critical topics including uncertainty propagation, risk communication, and semantic interoperability frameworks that may impact the success of these types of international multidisciplinary projects. A special focus will be on the transparency and accountability of forecasts and impact assessments and how this can be community to local decision makers.

With NOAA and ECMWF as technical leads, US and EU scientists have begun working together to set up a community of practice where they aim to improve methods, set up common standards, enhance interoperability, foster data exchange, and establish best practices. They are also planning a series of advanced workshops focusing on data exchanges/standards for interoperability of emerging and future digital twins and the use of novel AI methods. In the past year, an international team of scientists have begun exchanging use cases, methods, approaches, and lessons learned. This presentation will present some of the group’s initial work including a comparison of DT and AI-based TC model outputs as well as plans for supporting local decision makers in the Caribbean region and the possible development of a regional Digital Twin for TC impacts.