An ML model trained only on harmonized gridded observations achieves competitive medium-range weather forecast skill with the IFS for several upper-air and surface headline scores when verified against observations.
hub Canonical reference
AIFS -- ECMWF's data-driven forecasting system
Canonical reference. 83% of citing Pith papers cite this work as background.
33
Pith papers citing it
70
external citations · Pith
Background
83% of classified citations
abstract
Machine learning-based weather forecasting models have quickly emerged as a promising methodology for accurate medium-range global weather forecasting. Here, we introduce the Artificial Intelligence Forecasting System (AIFS), a data driven forecast model developed by the European Centre for Medium-Range Weather Forecasts (ECMWF). AIFS is based on a graph neural network (GNN) encoder and decoder, and a sliding window transformer processor, and is trained on ECMWF's ERA5 re-analysis and ECMWF's operational numerical weather prediction (NWP) analyses. It has a flexible and modular design and supports several levels of parallelism to enable training on high-resolution input data. AIFS forecast skill is assessed by comparing its forecasts to NWP analyses and direct observational data. We show that AIFS produces highly skilled forecasts for upper-air variables, surface weather parameters and tropical cyclone tracks. AIFS is run four times daily alongside ECMWF's physics-based NWP model and forecasts are available to the public under ECMWF's open data policy.
hub tools
citation-role summary
background 5
method 1
citation-polarity summary
representative citing papers
Observation-only machine learning can generate multi-decade global atmospheric reanalyses with large-scale skill near ERA5 and surface errors between ERA-Interim and ERA5, in a single day of compute.
AI weather models may simulate the atmosphere via particle positions in latent space whose updates follow gradient flow on a learned free energy functional rather than conventional physical equations.
OceanCBM is the first concept bottleneck model for spatiotemporal ocean prediction that uses mixed supervision on physical concepts and a free concept to deliver consistent mechanistic representations for mixed layer heat content forecasts.
QuadNorm uses quadrature-based moments instead of uniform averaging in normalization layers, achieving O(h²) consistency across resolutions and better cross-resolution transfer in neural operators.
GPROF-IR is a CNN-based retrieval that uses temporal context in geostationary IR observations to produce precipitation estimates with lower error than prior IR methods and climatological consistency with PMW retrievals for integration into IMERG V08.
Cast3 translates NWP principles into a data-driven model using cubed-sphere grids, super-ensembles, and generative nudging to achieve state-of-the-art ensemble predictions that outperform baselines.
Two new global-domain smoothing methods enable spatial verification scores like FSS on high-resolution global precipitation forecasts while handling grid area variability and missing data.
Hybrid LSTM-ViT model using mesonet surface data and profiler vertical profiles improves HRRR forecast error prediction for precipitation, wind speed, and temperature, with roughly twofold skill gain for precipitation over baseline LSTM.
NTK-UQ produces 31-37% sharper 90% prediction intervals than split conformal prediction for extreme weather forecasts, with adaptive scaling via architecture-dependent eigenvalue truncation and ICA decomposition of last-layer features.
ObsCast produces skillful short-term high-resolution weather analyses and forecasts over the contiguous US and Europe using only observational data, outperforming operational NWP without relying on NWP-derived data for training or inference.
RealBench is a benchmark for data-driven weather forecasting that enforces operational conditions via a 2025 OOD test set, operational analysis, in-situ observations, and event-specific extreme metrics to expose gaps versus reanalysis-based evaluation.
AirCast-SR is a latent consistency diffusion model that super-resolves GraphCast forecasts to 1 km hourly resolution over eight surface variables with near-zero bias and preserved fine-scale spectral power.
SwAIther-Precip uses lead-time-conditioned U-Net bias correction followed by diffusion-based generative downscaling to reduce CRPS by 48% and achieve ~4 km effective resolution from 0.25° AIFS forecasts.
Njord introduces a probabilistic GNN model using latent variables and adaptive K-means meshes for ensemble ocean forecasting with uncertainty estimates on global and regional domains.
Diffusion model climate emulators provide probability density estimates that allow likelihood calculations and odds-ratio-based importance sampling for extreme events such as tropical cyclones.
A GraphCast-based ocean emulator achieves skillful 10-15 day forecasts, with a Mahalanobis loss that accounts for variable correlations improving performance over MSE and acting as a statistical-dynamical regularizer.
A generative 3D Gaussian splatting model with scale-aware attention enables unified arbitrary-resolution forecasting and downscaling of 87 atmospheric variables.
Baguan-solar integrates Baguan weather foundation model forecasts with geostationary satellite data via a decoupled two-stage multimodal framework to deliver kilometer-scale 24-hour solar irradiance predictions, cutting RMSE by 16% versus baselines over East Asia.
GraphCSVAE is a new probabilistic framework that builds graph representations from satellite data to model and audit spatiotemporal changes in physical vulnerability using categorical inference and expert priors.
LSTM and 1D CNN emulators replicate a 1D marine biogeochemistry model at daily resolution, remain stable over decades, reproduce spring bloom timing years ahead, and outperform the parent model on reanalysis-driven forecasts for key variables.
Otter Weather is a spatiotemporal model that outperforms NWP baselines by 9.6% at 24h lead with under 3.5 A100-days training and extends efficiency gains to probabilistic forecasting via CRPS.
Instability-guided perturbations in the Aurora AI model can induce downstream shifts in an atmospheric river's moisture transport, potentially lowering landfall intensity in a California case study.
Probabilistic bias correction doubles ECMWF AI subseasonal skill and improves debiased dynamical forecasts for most pressure, temperature, and precipitation targets, winning ECMWF’s 2025 real-time competition.
citing papers explorer
-
AIFS-DOP: End-to-End Medium-Range Weather Prediction from Observations Alone with Machine Learning
An ML model trained only on harmonized gridded observations achieves competitive medium-range weather forecast skill with the IFS for several upper-air and surface headline scores when verified against observations.
-
Global reanalysis from observations alone with machine learning
Observation-only machine learning can generate multi-decade global atmospheric reanalyses with large-scale skill near ERA5 and surface errors between ERA-Interim and ERA5, in a single day of compute.
-
The physics of AI weather models
AI weather models may simulate the atmosphere via particle positions in latent space whose updates follow gradient flow on a learned free energy functional rather than conventional physical equations.
-
OceanCBM: A Concept Bottleneck Model for Mechanistic Interpretability in Ocean Forecasting
OceanCBM is the first concept bottleneck model for spatiotemporal ocean prediction that uses mixed supervision on physical concepts and a free concept to deliver consistent mechanistic representations for mixed layer heat content forecasts.
-
QuadNorm: Resolution-Robust Normalization for Neural Operators
QuadNorm uses quadrature-based moments instead of uniform averaging in normalization layers, achieving O(h²) consistency across resolutions and better cross-resolution transfer in neural operators.
-
GPROF-IR: An Improved Single-Channel Infrared Precipitation Retrieval for Merged Satellite Precipitation Products
GPROF-IR is a CNN-based retrieval that uses temporal context in geostationary IR observations to produce precipitation estimates with lower error than prior IR methods and climatological consistency with PMW retrievals for integration into IMERG V08.
-
Cast3: Translating numerical weather prediction principles into data-driven forecasting
Cast3 translates NWP principles into a data-driven model using cubed-sphere grids, super-ensembles, and generative nudging to achieve state-of-the-art ensemble predictions that outperform baselines.
-
Smoothing and spatial verification of global fields
Two new global-domain smoothing methods enable spatial verification scores like FSS on high-resolution global precipitation forecasts while handling grid area variability and missing data.
-
A Hybrid LSTM--Vision Transformer Architecture for Predicting HRRR Forecast Errors
Hybrid LSTM-ViT model using mesonet surface data and profiler vertical profiles improves HRRR forecast error prediction for precipitation, wind speed, and temperature, with roughly twofold skill gain for precipitation over baseline LSTM.
-
Scalable Uncertainty Quantification for Extreme Weather Forecasting via Empirical Neural Tangent Kernels
NTK-UQ produces 31-37% sharper 90% prediction intervals than split conformal prediction for extreme weather forecasts, with adaptive scaling via architecture-dependent eigenvalue truncation and ICA decomposition of last-layer features.
-
Skillful high-resolution weather forecasting independent of physical models
ObsCast produces skillful short-term high-resolution weather analyses and forecasts over the contiguous US and Europe using only observational data, outperforming operational NWP without relying on NWP-derived data for training or inference.
-
RealBench: Benchmarking Data-Driven Numerical Weather Forecasting Under Operational Conditions and Extreme Event Challenges
RealBench is a benchmark for data-driven weather forecasting that enforces operational conditions via a 2025 OOD test set, operational analysis, in-situ observations, and event-specific extreme metrics to expose gaps versus reanalysis-based evaluation.
-
AirCast-SR: A Foundation Model for Kilometer-Scale Atmospheric Super-Resolution via Latent Consistency Diffusion
AirCast-SR is a latent consistency diffusion model that super-resolves GraphCast forecasts to 1 km hourly resolution over eight surface variables with near-zero bias and preserved fine-scale spectral power.
-
SwAIther-Precip: Lead-Time-Aware Bias Correction Enables Kilometer-Scale Downscaling of Global AI Precipitation Forecasts over Switzerland
SwAIther-Precip uses lead-time-conditioned U-Net bias correction followed by diffusion-based generative downscaling to reduce CRPS by 48% and achieve ~4 km effective resolution from 0.25° AIFS forecasts.
-
Njord: A Probabilistic Graph Neural Network for Ensemble Ocean Forecasting
Njord introduces a probabilistic GNN model using latent variables and adaptive K-means meshes for ensemble ocean forecasting with uncertainty estimates on global and regional domains.
-
Towards accurate extreme event likelihoods from diffusion model climate emulators
Diffusion model climate emulators provide probability density estimates that allow likelihood calculations and odds-ratio-based importance sampling for extreme events such as tropical cyclones.
-
Skillful Global Ocean Emulation and the Role of Correlation-Aware Loss
A GraphCast-based ocean emulator achieves skillful 10-15 day forecasts, with a Mahalanobis loss that accounts for variable correlations improving performance over MSE and acting as a statistical-dynamical regularizer.
-
Generative 3D Gaussian Splatting for Arbitrary-ResolutionAtmospheric Downscaling and Forecasting
A generative 3D Gaussian splatting model with scale-aware attention enables unified arbitrary-resolution forecasting and downscaling of 87 atmospheric variables.
-
Integrating Weather Foundation Model and Satellite to Enable Fine-Grained Solar Irradiance Forecasting
Baguan-solar integrates Baguan weather foundation model forecasts with geostationary satellite data via a decoupled two-stage multimodal framework to deliver kilometer-scale 24-hour solar irradiance predictions, cutting RMSE by 16% versus baselines over East Asia.
-
GraphCSVAE: Graph Categorical Structured Variational Autoencoder for Spatiotemporal Auditing of Physical Vulnerability Towards Sustainable Post-Disaster Risk Reduction
GraphCSVAE is a new probabilistic framework that builds graph representations from satellite data to model and audit spatiotemporal changes in physical vulnerability using categorical inference and expert priors.
-
Deep learning model emulators for marine biogeochemistry forecasting from days to decades
LSTM and 1D CNN emulators replicate a 1D marine biogeochemistry model at daily resolution, remain stable over decades, reproduce spring bloom timing years ahead, and outperform the parent model on reanalysis-driven forecasts for key variables.
-
Otter Weather: Skillful and Computationally Efficient Medium-Range Weather Forecasting
Otter Weather is a spatiotemporal model that outperforms NWP baselines by 9.6% at 24h lead with under 3.5 A100-days training and extends efficiency gains to probabilistic forecasting via CRPS.
-
Instability-Aware Steering of an Extreme Atmospheric River in an AI Weather Foundation Model
Instability-guided perturbations in the Aurora AI model can induce downstream shifts in an atmospheric river's moisture transport, potentially lowering landfall intensity in a California case study.
-
Enhancing AI and Dynamical Subseasonal Forecasts with Probabilistic Bias Correction
Probabilistic bias correction doubles ECMWF AI subseasonal skill and improves debiased dynamical forecasts for most pressure, temperature, and precipitation targets, winning ECMWF’s 2025 real-time competition.
-
Regimes of Scale in AI Meteorology
AI/ML weather tools face integration challenges from mismatched 'regimes of scale' in how data and models are organized compared to traditional meteorology practices.
-
Event-Aware Loss Design for Forecasting of Convective Precipitation and Lightning
A multi-task Patch-cGAN with lightning-derived spatial loss weighting improves post-processed forecasts of intense precipitation and lightning occurrence over the Korean Peninsula in summer 2025.
-
Performance Evaluation of GraphCast for Medium-Range Weather Forecasting over Brazil
GraphCast shows regime-dependent skill versus ECMWF HRES in Brazil, underperforming on winter baroclinic systems in medium range but gaining in extended range and summer moisture transport.
-
Prediction of Drought and Flash Drought in Africa at the Seasonal-to-Subseasonal Scale using the Community Research Earth Digital Intelligence Twin Framework
DroughtFormer predicts soil moisture, vegetation health, and related variables in Africa with skill out to 90 days that matches or exceeds climatology for most targets, but shows lower accuracy for precipitation and flash drought indices.
-
Benchmarking atmospheric circulation variability in an AI emulator, ACE2, and a hybrid model, NeuralGCM
ACE2 emulator and NeuralGCM hybrid capture tropical wave and eddy spectra but fail to reproduce QBO (~28-month) and SAM propagation (~150-day) timescales.
-
Modelling convective cell occurrence in proximity to cold fronts using extreme gradient boosting
An XGBoost model reproduces convective cell frequency near cold fronts with high skill but underestimates counts at the surface front, depending most on CAPE and time of day.
-
Towards a Foundation Model for the Martian Atmosphere
The paper reviews data sources, physical models, downstream applications, and AI techniques to outline considerations for building a foundation model for the Martian atmosphere.
-
Earth Science Foundation Models: From Perception to Reasoning and Discovery
A review of Earth science foundation models covering capability evolution from perception to discovery, applications across atmosphere/hydrosphere/lithosphere/biosphere/anthroposphere/cryosphere, over 200 datasets, and key challenges.
-
Toward Artificial Intelligence Enabled Earth System Coupling
AI methods can strengthen cross-domain interactions and support more coherent multi-component representations in Earth system models.