Regional data-driven weather modeling with a global stretched-grid

TL;DR

This paper introduces a regional data-driven weather forecasting model using graph neural networks with a stretched-grid architecture, achieving high-resolution forecasts over specific regions by leveraging extensive historical data.

Contribution

The novel stretched-grid architecture combined with graph neural networks enables high-resolution regional weather predictions within a global framework, improving accuracy over existing models.

Findings

Outperforms MEPS ensemble mean for 2 m temperature forecasts.

Produces competitive precipitation and wind speed predictions.

Underestimates extreme weather events.

Abstract

A data-driven model (DDM) suitable for regional weather forecasting applications is presented. The model extends the Artificial Intelligence Forecasting System by introducing a stretched-grid architecture that dedicates higher resolution over a regional area of interest and maintains a lower resolution elsewhere on the globe. The model is based on graph neural networks, which naturally affords arbitrary multi-resolution grid configurations. The model is applied to short-range weather prediction for the Nordics, producing forecasts at 2.5 km spatial and 6 h temporal resolution. The model is pre-trained on 43 years of global ERA5 data at 31 km resolution and is further refined using 3.3 years of 2.5 km resolution operational analyses from the MetCoOp Ensemble Prediction System (MEPS). The performance of the model is evaluated using surface observations from measurement stations across Norway and is compared to short-range weather forecasts from MEPS. The DDM outperforms both the control run and the ensemble mean of MEPS for 2 m temperature. The model also produces competitive precipitation and wind speed forecasts, but is shown to underestimate extreme events.