--- id: ai-for-weather-forecasting title: "AI for Weather Forecasting: Data-Driven Numerical Weather Prediction and Nowcasting" schema_type: article category: ai language: en confidence: high last_verified: "2026-05-24" created_date: "2026-05-24" generation_method: ai_assisted ai_models: - claude-4.5-sonnet derived_from_human_seed: true conflict_of_interest: none_declared is_live_document: false data_period: static completeness: 0.85 atomic_facts: - id: af-ai-for-weather-forecasting-1 statement: >- Google DeepMind's GraphCast (Science, November 2023) demonstrated that a graph neural network trained on 39 years of ERA5 reanalysis data can produce 10-day global weather forecasts in under 1 minute on a single TPU -- matching or exceeding the ECMWF Integrated Forecasting System (IFS) on 90% of 1,380 verification targets, while being 1,000x computationally cheaper. This represents a paradigm shift from physics-based numerical weather prediction (NWP) to data-driven forecasting. source_title: GraphCast, Lam et al., Science (2023) -- doi:10.1126/science.adi2336 / ECMWF ERA5 reanalysis / Google DeepMind source_url: https://www.science.org/doi/10.1126/science.adi2336 confidence: high - id: af-ai-for-weather-forecasting-2 statement: >- Huawei's Pangu-Weather (Nature, July 2023, Bi et al.) and NVIDIA's FourCastNet (2022) demonstrated similar capabilities using 3D transformer and Fourier neural operator architectures respectively. By 2025-2026, ECMWF has integrated AI models (AIFS -- Artificial Intelligence Forecasting System) into operational ensemble forecasting, running alongside traditional physics-based IFS to produce calibrated, probabilistic forecasts that combine the strengths of both approaches. source_title: Pangu-Weather, Bi et al., Nature (2023) / NVIDIA FourCastNet (2022) / ECMWF AIFS (2024-2026) -- operational AI weather forecasting source_url: https://www.nature.com/articles/s41586-023-06185-3 confidence: high primary_sources: - id: ps-ai-for-weather-forecasting-1 title: Learning skillful medium-range global weather forecasting (GraphCast) type: academic_paper year: 2023 institution: Science / Google DeepMind doi: 10.1126/science.adi2336 url: https://www.science.org/doi/10.1126/science.adi2336 - id: ps-ai-for-weather-forecasting-2 title: Accurate medium-range global weather forecasting with 3D neural networks (Pangu-Weather) type: academic_paper year: 2023 institution: Nature / Huawei Cloud doi: 10.1038/s41586-023-06185-3 url: https://www.nature.com/articles/s41586-023-06185-3 known_gaps: - Extreme weather event prediction -- hurricanes, tornadoes, flash floods with sufficient lead time - Probabilistic AI weather forecasting with well-calibrated uncertainty estimates disputed_statements: [] secondary_sources: - title: Learning Skillful Medium-Range Global Weather Forecasting (GraphCast) type: journal_article year: 2023 authors: - Lam, Remi - Sanchez-Gonzalez, Alvaro - Willson, Matthew - Wirnsberger, Peter - Fortunato, Meire - et al. institution: Google DeepMind / Science url: https://doi.org/10.1126/science.adi2336 - title: Probabilistic Weather Forecasting with Machine Learning (GenCast) type: journal_article year: 2024 authors: - Price, Ilan - Sanchez-Gonzalez, Alvaro - Alet, Ferran - et al. institution: Google DeepMind / Nature url: https://www.nature.com/articles/s41586-024-08252-9 - title: Accurate Medium-Range Global Weather Forecasting with 3D Neural Networks (Pangu-Weather) type: journal_article year: 2023 authors: - Bi, Kaifeng - Xie, Lingxi - Zhang, Hengheng - Chen, Xin - Gu, Xiaotao - Tian, Qi institution: Huawei Cloud / Nature url: https://www.nature.com/articles/s41586-023-06185-3 - title: "AI for Climate and Weather: A Comprehensive Survey on Deep Learning Methods" type: survey_paper year: 2024 authors: - multiple institution: arXiv / WIREs Climate Change url: https://doi.org/10.1002/wcc.890 updated: "2026-05-24" --- ## TL;DR AI is revolutionizing weather forecasting -- models like GraphCast, Pangu-Weather, and FourCastNet can produce 10-day global forecasts in seconds that rival the accuracy of traditional supercomputer-based physics simulations that took hours. By 2025, the European Centre for Medium-Range Weather Forecasts is running AI forecasts operationally alongside classical models, marking the beginning of the data-driven weather prediction era. ## Core Explanation Traditional Numerical Weather Prediction (NWP): solve the Navier-Stokes equations and atmospheric physics PDEs on a 3D grid covering the Earth using finite-difference or spectral methods. ECMWF IFS runs on ~1 million CPU cores, producing a 10-day forecast in 1-2 hours. It works well but is computationally expensive and limited by our incomplete understanding of sub-grid physics (cloud formation, turbulence). Data-driven approach: train a neural network on decades of historical weather data (ERA5 reanalysis -- 39 years, 0.25-degree resolution, hourly). The model learns the mapping from current atmospheric state to future state directly from data, bypassing explicit physics. Key architectural insight (GraphCast): the Earth is represented as a multi-scale icosahedral mesh (graph), with nodes at 0.25-degree resolution. The processor applies learned message-passing across edges to simulate atmospheric dynamics. Training: multi-step -- predict 6-hour steps, backpropagate through 6 steps to learn longer-range dynamics. Inference: autoregressive rollout. ## Detailed Analysis Architecture comparison: GraphCast (Google) -- GNN on icosahedral mesh, 36.7M parameters, 1-minute forecast on TPUv4. Pangu-Weather (Huawei) -- 3D Earth-Specific Transformer (3DEST), processes atmosphere as 3D grid with 13 pressure levels, 64M parameters. FourCastNet (NVIDIA) -- Adaptive Fourier Neural Operator (AFNO), learns in frequency domain via FFT, 74M parameters. Performance: GraphCast outperforms IFS on 90% of targets for deterministic forecasts. For extreme events: GraphCast predicted Hurricane Lee making landfall in Nova Scotia 9 days in advance (vs. 6 days for IFS). ECMWF AIFS (2024-2026): the operational transition. ECMWF now runs GraphCast and other AI models as part of its ensemble prediction system, producing 50 perturbed forecasts for probabilistic prediction. AI models excel at medium-range (1-10 day) deterministic forecasts; physics-based models remain essential for ensembles, uncertainty quantification, and rare events. The future is hybrid -- AI for fast, accurate deterministic prediction + physics for calibration and extremes.