Graph Neural Networks and Spatial Information Learning for Post-Processing Ensemble Weather Forecasts
This work addresses the problem of limited spatial information sharing in weather forecast post-processing for meteorologists, representing an incremental advancement over existing neural network approaches.
The paper tackled systematic errors in ensemble weather forecasts by proposing a graph neural network architecture that leverages spatial structures through attention mechanisms, resulting in substantial improvements over a competitive neural network-based method for 2-m temperature forecasts over Europe.
Ensemble forecasts from numerical weather prediction models show systematic errors that require correction via post-processing. While there has been substantial progress in flexible neural network-based post-processing methods over the past years, most station-based approaches still treat every input data point separately which limits the capabilities for leveraging spatial structures in the forecast errors. In order to improve information sharing across locations, we propose a graph neural network architecture for ensemble post-processing, which represents the station locations as nodes on a graph and utilizes an attention mechanism to identify relevant predictive information from neighboring locations. In a case study on 2-m temperature forecasts over Europe, the graph neural network model shows substantial improvements over a highly competitive neural network-based post-processing method.