Building Machine Learning Limited Area Models: Kilometer-Scale Weather Forecasting in Realistic Settings

2025-04-12

AI Summary

Machine learning model development and operational integration remain challenging for high-resolution (kilometer-scale) limited-area weather forecasting. Method: This study proposes the first end-to-end graph neural network framework tailored for local forecasting. It introduces a novel rectangular–triangular hybrid graph topology, multi-step rolling training, and a dynamic boundary forcing mechanism—enabling flexible ingestion of either reanalysis or operational model boundary conditions. A systematic evaluation is conducted on boundary width, graph topology design, and boundary integration strategies. Contribution/Results: The framework is validated operationally over complex terrain in Denmark and Switzerland. In Switzerland, the model outperforms conventional numerical weather prediction systems in near-surface temperature and wind speed forecasts while reducing computational cost substantially—demonstrating strong potential for real-world operational deployment.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Machine learning is revolutionizing global weather forecasting, with models that efficiently produce highly accurate forecasts. Apart from global forecasting there is also a large value in high-resolution regional weather forecasts, focusing on accurate simulations of the atmosphere for a limited area. Initial attempts have been made to use machine learning for such limited area scenarios, but these experiments do not consider realistic forecasting settings and do not investigate the many design choices involved. We present a framework for building kilometer-scale machine learning limited area models with boundary conditions imposed through a flexible boundary forcing method. This enables boundary conditions defined either from reanalysis or operational forecast data. Our approach employs specialized graph constructions with rectangular and triangular meshes, along with multi-step rollout training strategies to improve temporal consistency. We perform systematic evaluation of different design choices, including the boundary width, graph construction and boundary forcing integration. Models are evaluated across both a Danish and a Swiss domain, two regions that exhibit different orographical characteristics. Verification is performed against both gridded analysis data and in-situ observations, including a case study for the storm Ciara in February 2020. Both models achieve skillful predictions across a wide range of variables, with our Swiss model outperforming the numerical weather prediction baseline for key surface variables. With their substantially lower computational cost, our findings demonstrate great potential for machine learning limited area models in the future of regional weather forecasting.

Problem

• Develop kilometer-scale ML models for regional weather forecasting
• Evaluate design choices in boundary conditions and graph constructions
• Assess model performance against analysis data and real observations

Innovation

• Flexible boundary forcing method for conditions
• Specialized graph constructions with diverse meshes
• Multi-step rollout training for temporal consistency