Graph Neural Networks and Spatial Information Learning for Post-Processing Ensemble Weather Forecasts

TL;DR

This paper introduces a graph neural network approach for post-processing ensemble weather forecasts, effectively leveraging spatial information across locations to improve temperature prediction accuracy over Europe.

Contribution

It presents a novel GNN architecture with attention mechanisms for spatially-aware post-processing of ensemble weather forecasts, outperforming existing neural network methods.

Findings

Significant accuracy improvements in temperature forecasts.

Effective utilization of spatial relationships between locations.

Outperforms traditional neural network-based post-processing methods.

Abstract

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.