WeatherGNN: Exploiting Meteo- and Spatial-Dependencies for Local Numerical Weather Prediction Bias-Correction
This addresses bias-correction for local weather forecasting, which is incremental as it builds on existing data-driven methods by incorporating domain knowledge and dependencies.
The paper tackles biases in local numerical weather prediction by proposing WeatherGNN, a method using graph neural networks to exploit meteorological and spatial dependencies, achieving state-of-the-art performance with a 4.75% average improvement in RMSE over baselines.
Due to insufficient local area information, numerical weather prediction (NWP) may yield biases for specific areas. Previous studies correct biases mainly by employing handcrafted features or applying data-driven methods intuitively, overlooking the complicated dependencies between weather factors and between areas. To address this issue, we propose WeatherGNN, a local NWP bias-correction method that utilizes Graph Neural Networks (GNNs) to exploit meteorological dependencies and spatial dependencies under the guidance of domain knowledge. Specifically, we introduce a factor GNN to capture area-specific meteorological dependencies adaptively based on spatial heterogeneity and a fast hierarchical GNN to capture dynamic spatial dependencies efficiently guided by Tobler's first and second laws of geography. Our experimental results on two real-world datasets demonstrate that WeatherGNN achieves the state-of-the-art performance, outperforming the best baseline with an average of 4.75 \% on RMSE.