Probabilistic weather forecasting with machine learning
2024/12/04 by Ilan Price, Álvaro Sánchez‐González, Ferran Alet +9 · 3 voices · 153 citations
Earth and Planetary Sciences · Environmental Science · #Meteorological Phenomena and Simulations #Climate variability and models #Tropical and Extratropical Cyclones Research
paper · doi:10.1038/s41586-024-08252-9
Abstract
Abstract Weather forecasts are fundamentally uncertain, so predicting the range of probable weather scenarios is crucial for important decisions, from warning the public about hazardous weather to planning renewable energy use. Traditionally, weather forecasts have been based on numerical weather prediction (NWP) 1 , which relies on physics-based simulations of the atmosphere. Recent advances in machine learning (ML)-based weather prediction (MLWP) have produced ML-based models with less forecast error than single NWP simulations 2,3 . However, these advances have focused primarily on single, deterministic forecasts that fail to represent uncertainty and estimate risk. Overall, MLWP has remained less accurate and reliable than state-of-the-art NWP ensemble forecasts. Here we introduce GenCast, a probabilistic weather model with greater skill and speed than the top operational medium-range weather forecast in the world, ENS, the ensemble forecast of the European Centre for Medium-Range Weather Forecasts 4 . GenCast is an ML weather prediction method, trained on decades of reanalysis data. GenCast generates an ensemble of stochastic 15-day global forecasts, at 12-h steps and 0.25° latitude–longitude resolution, for more than 80 surface and atmospheric variables, in 8 min. It has greater skill than ENS on 97.2% of 1,320 targets we evaluated and better predicts extreme weather, tropical cyclone tracks and wind power production. This work helps open the next chapter in operational weather forecasting, in which crucial weather-dependent decisions are made more accurately and efficiently.
Citations
Cited by
- Inductive Risk of AI Hype
- scPortrait integrates single-cell images into multimodal modeling
- Community Research Earth Digital Intelligence Twin: a scalable framework for AI-driven Earth System Modeling
- Can AI weather models predict out-of-distribution gray swan tropical cyclones?
- MAPCast: A Convection Allowing MPAS Emulator for Ensemble-based Background Error Covariance Estimation Toward Multi-Scale Data Assimilation
- GraphCast Skill and Systematic Biases in Indian Summer Monsoon Forecasts: Evaluation Against ERA5 and IMERG
- Explainable quantum-compressed machine learning for complex fluid flows
- Flexible generation of daily Earth system model projections across radiative forcing scenarios
- Native Extrapolation Awareness in Flow-Based Conditional Generation
- Spatial Generalization Tests for Machine Learning-based Weather Models to Assess Physical Consistency
- Theory-to-Practice Gap for Neural Networks and Neural Operators
- Apeliotes: A Diffusion-Based Modeling Framework for km-scale Multi-Level Atmospheric Fields
- BG4Sea: Biogeochemical Seasonal Forecastability via Progressive Information Scaling
- The Effect of Stochasticity in Score-Based Diffusion Sampling: a KL Divergence Analysis
- Diffusion models recover accurate mixture weights despite score function insensitivity
- Training-Free Bayesian Filtering with Generative Emulators
- Transformers for dynamical systems learn transfer operators in-context
- HiRO-ACE: Fast and skillful AI emulation and downscaling trained on a 3 km global storm-resolving model
- Controllable Probabilistic Forecasting with Stochastic Decomposition Layers
- Numerical models outperform AI weather forecasts of record-breaking extremes
- Panda: A pretrained forecast model for chaotic dynamics
- Erwin: A Tree-based Hierarchical Transformer for Large-scale Physical Systems
- Physics‐informed ensemble forecasts in data‐driven models
- Long-Range Distillation: Distilling 10,000 Years of Simulated Climate into Long Timestep AI Weather Models
- Integrating GNSS-Derived Zenith Wet Delay into a Weather Foundation Model Improves Precipitation Forecasting
- Zhinv: Real-time hub-height wind field reconstruction using only local sparse observations
- AIFS-CRPS: Ensemble forecasting using a model trained with a loss function based on the Continuous Ranked Probability Score
- Autoregressive One-Step Generative Modeling for Dynamical System Forecasting
- Field-Space Attention for Structure-Preserving Earth System Transformers
- Simulation-Driven Railway Delay Prediction: An Imitation Learning Approach
- Skillful Subseasonal-to-Seasonal Forecasting of Extreme Events with a Multi-Sphere Coupled Probabilistic Model
- Generative Urban Flow Modeling: From Geometry to Airflow with Graph Diffusion
- Reliable Statistical Guarantees for Conformal Predictors with Small Datasets
- VLCs: Managing Parallelism with Virtualized Libraries
- DAISI: Data Assimilation with Inverse Sampling using Stochastic Interpolants
- High-Resolution Probabilistic Data-Driven Weather Modeling with a Stretched-Grid
- Climatological benchmarking of AI-generated tropical cyclones
- Validity in machine learning for extreme event attribution
- Large-Scale In-Game Outcome Forecasting for Match, Team and Players in Football using an Axial Transformer Neural Network
- On the Predictive Skill of Artificial Intelligence-based Weather Models for Extreme Events using Uncertainty Quantification
- Towards Open-Ended Visual Scientific Discovery with Sparse Autoencoders
- FreqFlow: Long-term forecasting using lightweight flow matching
- SWR-Viz: AI-assisted Interactive Visual Analytics Framework for Ship Weather Routing
- A Space-Time Transformer for Precipitation Nowcasting
- When is a System Discoverable from Data? Discovery Requires Chaos
- Generalizable Insights for Graph Transformers in Theory and Practice
- AIA Forecaster: Technical Report
- Nowcast3D: Reliable precipitation nowcasting via gray-box learning
- Efficient probabilistic surrogate modeling techniques for partially-observed large-scale dynamical systems
- Diffusion Models Bridge Deep Learning and Physics in ENSO Forecasting
- Towards Multi-Fidelity Scaling Laws of Neural Surrogates in CFD
- Hydra: Dual Exponentiated Memory for Multivariate Time Series Analysis
- Sensitivity Analysis for Climate Science with Generative Flow Models
- AI-boosted rare event sampling to characterize extreme weather
- Using data assimilation tools to dissect GraphDOP
- Efficient Generative AI Boosts Probabilistic Forecasting of Sudden Stratospheric Warmings
- Predictability of Storms in an Idealized Climate Revealed by Machine Learning
- Uncertainty Quantification for Regression: A Unified Framework based on kernel scores
- Can Deep Generative Models Reproduce Non-Stationary Gaussian Random Fields?
- Hierarchical Testing of a Hybrid Machine Learning‐Physics Global Atmosphere Model
- Extreme Weather Bench: A framework and benchmark for evaluation of high-impact weather
- The Rise of AI in Weather and Climate Information and its Impact on Global Inequality
- Evaluating Extreme Precipitation Forecasts: A Threshold-Weighted, Spatial Verification Approach for Comparing an AI Weather Prediction Model Against a High-Resolution NWP Model
- Forecasting Arctic Temperatures with Temporally Dependent Data Using Quantile Gradient Boosting and Adaptive Conformal Prediction Regions
- Revealing the Potential of Learnable Perturbation Ensemble Forecast Model for Tropical Cyclone Prediction
- The Benchmarking Epistemology: Construct Validity for Evaluating Machine Learning Models
- LLMComp: A Language Modeling Paradigm for Error-Bounded Scientific Data Compression (Technical Report)
- CSU-PCAST: A Dual-Branch Transformer Framework for medium-range ensemble Precipitation Forecasting
- Cross-Domain Long-Term Forecasting: Radiation Dose from Sparse Neutron Sensor via Spatio-Temporal Operator Network
- Benchmarking Probabilistic Time Series Forecasting Models on Neural Activity
- Quantile Regression, Variational Autoencoders, and Diffusion Models for Uncertainty Quantification: A Spatial Analysis of Sub-seasonal Wind Speed Prediction
- Operator Flow Matching for Timeseries Forecasting
- An Operational Deep Learning System for Satellite-Based High-Resolution Global Nowcasting
- Assessing the Geographic Generalization and Physical Consistency of Generative Models for Climate Downscaling
- Progressive multi-fidelity learning with neural networks for physical system predictions
- On Foundation Models for Temporal Point Processes to Accelerate Scientific Discovery
- Cross-Scale Reservoir Computing for large spatio-temporal forecasting and modeling
- DAWP: A framework for global observation forecasting via Data Assimilation and Weather Prediction in satellite observation space
- Adversarial Attacks on Downstream Weather Forecasting Models: Application to Tropical Cyclone Trajectory Prediction
- Evaluating the Shanghai Typhoon Model against State-of-the-Art Machine-Learning Weather Prediction Models: A Case Study for Typhoon Danas (2025)
- Quantifying Very Extreme Precipitation and Temperature Using Huge Ensembles Generated by Machine Learning-based Climate Model Emulators
- Control-Augmented Autoregressive Diffusion for Data Assimilation
- Climate Model Tuning with Online Synchronization-Based Parameter Estimation
- DANRA: The Kilometer-Scale Danish Regional Atmospheric Reanalysis
- Improved probabilistic regression using diffusion models
- Benchmarking atmospheric circulation variability in an AI emulator, ACE2, and a hybrid model, NeuralGCM
- Zephyrus: An Agentic Framework for Weather Science
- Riemannian Consistency Model
- Probability calibration for precipitation nowcasting
- Diffusion Modeling of the Three-Dimensional Magnetic Field in the Sun's Corona
- EnScale: Temporally-consistent multivariate generative downscaling via proper scoring rules
- Assessing the risk of future Dunkelflaute events for Germany using generative deep learning
- A Synergistic Approach: Dynamics–AI Ensemble in Tropical Cyclone Forecasting
- Using deep learning to generate key variables in global mitigation scenarios
- Graph-based Neural Space Weather Forecasting
- From Observation to Insight: Mechanistic World Models and the Quest for Autonomous Discovery
- Can AI Weather Models Predict Beyond Two Weeks? A Quantitative Benchmark and Analysis of Long Rollouts
- Navigating Artificial Intelligence With Human Stupidity… and Vice Versa
- Global Forecasting of Tropical Cyclone Intensity Using Neural Weather Models
- Training-Free Data Assimilation with GenCast
- Technical overview and architecture of the FastNet Machine Learning weather prediction model, version 1.0
- FastNet: Improving the physical consistency of machine-learning weather prediction models through loss function design
- STCast: Adaptive Boundary Alignment for Global and Regional Weather Forecasting
- From Prediction to Understanding: Will AI Foundation Models Transform Brain Science?
- A data-driven global ocean forecasting model with sub-daily and eddy-resolving resolution
- DPSformer: A long-tail-aware model for improving heavy rainfall prediction
- Data-Efficient Ensemble Weather Forecasting with Diffusion Models
- Breaking the Statistical Similarity Trap in Extreme Convection Detection
- Knowledge-data fusion framework for frequency security assessment in low-inertia power systems
- Seasonal forecasting using the GenCast probabilistic machine learning model
- Data-driven solar forecasting enables near-optimal economic decisions
- Finetuning AI Foundation Models to Develop Subgrid-Scale Parameterizations: A Case Study on Atmospheric Gravity Waves
- Scale-Adaptive Generative Flows for Multiscale Scientific Data
- MAUSAM: An Observations-focused assessment of Global AI Weather Prediction Models During the South Asian Monsoon
- RAMS: Residual-based adversarial-gradient moving sample method for scientific machine learning in solving partial differential equations
- Self-organized learning emerges from coherent coupling of critical neurons
- Post-processing of ensemble photovoltaic power forecasts with distributional and quantile regression methods
- From forecast skill to economic value: sub-hourly wildfire potential forecasting across Australian regions
- Huracan: A skillful end-to-end data-driven system for ensemble data assimilation and weather prediction
- Learning from nature: insights into GraphDOP's representations of the Earth System
- Forecasting Extreme Day and Night Heat in Paris
- High-Accuracy Temporal Prediction via Experimental Quantum Reservoir Computing in Correlated Spins
- CERA: A Framework for Improved Generalization of Machine Learning Models to Changed Climates
- Adapting LLMs to Time Series Forecasting via Temporal Heterogeneity Modeling and Semantic Alignment
- MeteorPred: A Meteorological Multimodal Large Model and Dataset for Severe Weather Event Prediction
- Resilience metrics to guide back-up investments in the power system during extreme weather
- SolarSeer: Ultrafast and accurate 24-hour solar irradiance forecasts outperforming numerical weather prediction across the USA
- Flow Matching for Probabilistic Learning of Dynamical Systems from Missing or Noisy Data
- PnP-DA: Towards Principled Plug-and-Play Integration of Variational Data Assimilation and Generative Models
- Data-driven global ocean model resolving ocean-atmosphere coupling dynamics
- End-to-end data-driven weather prediction. [europepmc]
- An interpretable machine learning model for seasonal precipitation forecasting. [europepmc]
- Selected Topics in Time Series Forecasting: Statistical Models vs. Machine Learning. [europepmc]
- Global data-driven prediction of fire activity. [europepmc]
- Can AI weather models predict out-of-distribution gray swan tropical cyclones? [europepmc]
- Learning dynamical systems with hit-and-run random feature maps. [europepmc]
- AI-Y: An AI Checklist for Population Ethics Across the Global Context. [europepmc]
- Multi-model approach to understand and predict past and future dengue epidemic dynamics. [europepmc]
- Addressing the data imbalance issue in machine learning modeling of rare and disruptive outage events. [europepmc]
- AI for atmosphere-ocean sciences: advancements, challenges and ways forward. [europepmc]
- Toward AI foundation models for epidemics: Promise, challenges, and paths forward. [europepmc]
- Seasonal forecasting using the GenCast probabilistic machine learning model. [europepmc]
- Who's afraid of synthetic data? Hybrid approaches to deliver medical digital twins. [europepmc]
- Weather forecasts become more important for reducing mortality as the climate warms. [europepmc]
- Quantum-informed machine learning for predicting spatiotemporal chaos with practical quantum advantage. [europepmc]
- ArchesWeatherGen: Skillful and compute-efficient probabilistic weather forecasting with machine learning. [europepmc]
- The application of large language models in meteorology graduate research: current status, impact, and prospects. [europepmc]
- Frequency-adaptive deep learning for multi-horizon weather forecasting in environmental monitoring applications. [europepmc]
- Physics-based models outperform AI weather forecasts of record-breaking extremes. [europepmc]
- Data-driven global ocean model resolving atmospherically forced ocean dynamics. [europepmc]
- Generative machine learning for multivariate angular simulation. [europepmc]
- Skillful subseasonal Indian Ocean marine heatwave forecasts using a neural network [europepmc]
- scPortrait integrates single-cell images into multimodal modeling [europepmc]
- Multi-model approach to understand and predict past and future dengue epidemic dynamics [europepmc]
- Forecasting the eddying ocean with a deep neural network. [europepmc]
Discussions
- DeepMind AI weather forecaster beats world-class system (this time with Bayesian statistics) [lemmy, 28 points, 4 comments]
- The so awaited GenCast #weather forecasting paper is out (Ilan Price et al.) and it is pretty amazing. I have seen first results on this year's @climformatics.bsky.social , but reading the article is [bsky, 10 points, 1 comments]
- Price, I., Sanchez-Gonzalez, A., Alet, F. et al. Probabilistic weather forecasting with machine learning. Nature (2024). doi.org/10.1038/s415... [bsky, 2 points, 0 comments]
Related