GraphDOP: Towards skilful data-driven medium-range weather forecasts learnt and initialised directly from observations
This work addresses the need for accurate medium-range weather forecasts by developing a novel end-to-end data-driven approach that could reduce reliance on traditional physics-based models, though it appears incremental as it builds on existing data-driven methods.
The paper tackles the problem of medium-range weather forecasting by introducing GraphDOP, a data-driven system trained and initialized directly from Earth System observations without physics-based inputs, achieving skillful predictions of weather parameters up to five days into the future.
We introduce GraphDOP, a new data-driven, end-to-end forecast system developed at the European Centre for Medium-Range Weather Forecasts (ECMWF) that is trained and initialised exclusively from Earth System observations, with no physics-based (re)analysis inputs or feedbacks. GraphDOP learns the correlations between observed quantities - such as brightness temperatures from polar orbiters and geostationary satellites - and geophysical quantities of interest (that are measured by conventional observations), to form a coherent latent representation of Earth System state dynamics and physical processes, and is capable of producing skilful predictions of relevant weather parameters up to five days into the future.