A Graph Neural Network Approach for Localized and High-Resolution Temperature Forecasting
Joud El-Shawa, Elham Bagheri, Sedef Akinli Kocak, Yalda Mohsenzadeh
TL;DR
This paper tackles the need for high-resolution, localized temperature forecasts to better protect vulnerable communities from heatwaves. It presents a graph neural network framework (GCN–GRU) that operates at 2.5 km resolution to forecast 2-meter temperature up to 48 hours, trained on NOAA URMA data for Southwestern Ontario. Results show that larger spatial contexts yield higher accuracy (Region C MAE ≈ 1.93°C, 48h MAE ≈ 2.93°C) and that a lighter 6-hour temporal window preserves much of the skill with reduced compute. An embedding-based input pathway (ClimateBERT + PCA) offers a standardized way to handle heterogeneous data and supports transferability to data-limited regions, enabling more equitable heatwave warnings and planning.
Abstract
Heatwaves are intensifying worldwide and are among the deadliest weather disasters. The burden falls disproportionately on marginalized populations and the Global South, where under-resourced health systems, exposure to urban heat islands, and the lack of adaptive infrastructure amplify risks. Yet current numerical weather prediction models often fail to capture micro-scale extremes, leaving the most vulnerable excluded from timely early warnings. We present a Graph Neural Network framework for localized, high-resolution temperature forecasting. By leveraging spatial learning and efficient computation, our approach generates forecasts at multiple horizons, up to 48 hours. For Southwestern Ontario, Canada, the model captures temperature patterns with a mean MAE of 1.93$^{\circ}$C across 1-48h forecasts and MAE@48h of 2.93$^{\circ}$C, evaluated using 24h input windows on the largest region. While demonstrated here in a data-rich context, this work lays the foundation for transfer learning approaches that could enable localized, equitable forecasts in data-limited regions of the Global South.