★ WeatherBench 2: A benchmark for the next generation of data-driven global weather models
WeatherBench 2 provides an open evaluation framework, training and ground-truth data, and baselines for data-driven global 1-14 day forecasts.
Benchmarks, datasets & evaluation
Benchmarks, ML-ready datasets, verification methods and model intercomparisons.
Search and filter this topic Plain list on GitHub
★ Highlights
★ The rise of data-driven weather forecasting
Compares PanguWeather with ECMWF IFS and finds comparable skill for global and extreme metrics, with forecast smoothness and bias drift as drawbacks.
★ The EUPPBench postprocessing benchmark dataset v1.0
Releases EUPPBench, a time-aligned forecast-observation dataset with evaluation methods and a first benchmark of 2 m temperature postprocessing.
Deutscher Wetterdienst (DWD)ECMWFKarlsruhe Institute of Technology (KIT)Met OfficeMétéo-France (incl. CNRM)MeteoSwisscode
★ ClimateBench v1.0: A Benchmark for Data‐Driven Climate Projections
Introduces ClimateBench, a CMIP-based benchmark with baseline ML emulators predicting temperature and precipitation responses to emissions scenarios.
ETH ZurichMET Norway (Norwegian Meteorological Institute)University of Oxford
★ WeatherBench: A benchmark dataset for data-driven weather forecasting
WeatherBench introduces an ERA5-derived benchmark with standard metrics and baselines (linear regression, deep learning, physical models) for data-driven medium-range global weather forecasting.
ECMWFTechnical University of Munich (TUM)University of Washingtoncode
★ WeatherBench: A Benchmark Data Set for Data-Driven Weather Forecasting
WeatherBench provides an ERA5-derived benchmark with metrics and baselines for data-driven global medium-range forecasting, enabling direct model comparison.
ECMWFUniversity of WashingtonTechnical University of Munich (TUM)code
All papers (52)
Extreme Weather Bench: A framework and benchmark for evaluation of high-impact weather
Extreme Weather Bench is an open community benchmark of case studies, observations and impact metrics for verifying AI and NWP models on high-impact weather.
Recommendations for Comprehensive and Independent Evaluation of Machine Learning-Based Earth System Models
Five recommendations for standardized, independent evaluation of ML-based Earth system models, stressing physical consistency for climate projections without historical analogues.
Colorado State University (incl. CIRA)Lawrence Berkeley National Laboratory / NERSCNSF NCAR / UCAROak Ridge National LaboratoryUniversity of California, Berkeley
WeatherReal: A Benchmark Based on In-Situ Observations for Evaluating Weather Models
Introduces WeatherReal, a benchmark from global in-situ surface observations showing reanalysis-trained AI models diverge from station data for temperature, wind and precipitation.
ChaosBench: A Multi-Channel, Physics-Based Benchmark for Subseasonal-to-Seasonal Climate Prediction
ChaosBench is a physics-based S2S benchmark showing weather-scale AI models collapse to climatology, with ensembles and physics constraints suggested to extend skill.
Intercomparison of Deep Learning Architectures for the Prediction of Precipitation Fields With a Focus on Extremes
Compares deep learning architectures for European precipitation, including extremes; a two-level U-Net matches deeper versions at lower cost, with layer-wise relevance attribution.
★ WeatherBench 2: A benchmark for the next generation of data-driven global weather models
WeatherBench 2 provides an open evaluation framework, training and ground-truth data, and baselines for data-driven global 1-14 day forecasts.
★ The rise of data-driven weather forecasting
Compares PanguWeather with ECMWF IFS and finds comparable skill for global and extreme metrics, with forecast smoothness and bias drift as drawbacks.
★ The EUPPBench postprocessing benchmark dataset v1.0
Releases EUPPBench, a time-aligned forecast-observation dataset with evaluation methods and a first benchmark of 2 m temperature postprocessing.
Deutscher Wetterdienst (DWD)ECMWFKarlsruhe Institute of Technology (KIT)Met OfficeMétéo-France (incl. CNRM)MeteoSwisscode
SuperBench: A Super-Resolution Benchmark Dataset for Scientific Machine Learning
SuperBench provides a super-resolution benchmark across fluids, cosmology and weather data, finding that deep SR methods miss fine-scale physics; weather is one of three domains.
Verification against in-situ observations for Data-Driven Weather Prediction
Releases a quality-controlled NOAA MADIS in-situ observation dataset to verify data-driven weather prediction models against real-world operational observations rather than reanalysis.
Tropical Cyclone Wind Speed Estimation: A Large Scale Training Data Set and Community Benchmarking
Curates an hourly-interpolated satellite image and wind-speed dataset for tropical cyclones, plus a community competition whose winners improved on the benchmark by about 50%.
LSDSSIMR: Large-Scale Dust Storm Database Based on Satellite Images and Meteorological Reanalysis Data
Releases LSDSSIMR, a 2020–2022 dataset of FY-4A satellite dust images at 15-minute, 4 km resolution, paired with reanalysis for dust storm prediction.
A climate index collection based on model data
Open dataset of climate indices from 1000-year Earth system model control runs, with ML benchmarks predicting Sahel rainfall and ENSO.
ClimateLearn: Benchmarking Machine Learning for Weather and Climate Modeling
Open-source PyTorch library standardizing ERA5 and CMIP6 pipelines, deep models and evaluation for ML weather forecasting and climate downscaling.
NADBenchmarks -- a compilation of Benchmark Datasets for Machine Learning Tasks related to Natural Disasters
Compiles and proposes a searchable web platform of ML benchmark datasets for natural disasters, organized by disaster management cycle. Mostly a catalogue, low novelty.
An archived dataset from the ECMWF Ensemble Prediction System for probabilistic solar power forecasting
Archived ECMWF ensemble dataset for 2017 to 2020 over Europe and North America supports probabilistic solar forecasting, with calibration and ML-versus-physics case studies.
Deep learning for quality control of surface physiographic fields using satellite Earth observations
A neural network trained on ERA5 and MODIS skin temperature quality-checks IFS physiographic upgrades, improving land-surface temperature prediction by 0.37 K for updated lakes.
★ ClimateBench v1.0: A Benchmark for Data‐Driven Climate Projections
Introduces ClimateBench, a CMIP-based benchmark with baseline ML emulators predicting temperature and precipitation responses to emissions scenarios.
ETH ZurichMET Norway (Norwegian Meteorological Institute)University of Oxford
AutoML for Climate Change: A Call to Action
Benchmarks off-the-shelf AutoML libraries on climate modelling and wind power forecasting; they fail to beat human-designed models, motivating spatiotemporal AutoML search spaces.
Efficient Climate Simulation via Machine Learning Method
NeuroClim provides a hybrid climate modeling platform, open dataset, and stability and approximation metrics for learning climate simulation components.
Evaluating the Dark Sky Weather Forecast Data Over Florida and Georgia
Evaluates Dark Sky daily forecasts against 124 Florida and Georgia stations; finds good yes/no rain detection (POD 0.91–1.0) but poor rainfall amounts.
Multi-source and heterogeneous marine hydrometeorology spatio-temporal data analysis with machine learning: a survey
Survey of machine learning for multi-source marine hydrometeorological spatio-temporal data; no abstract available, so scope is inferred from the title.
Challenges and Benchmark Datasets for Machine Learning in the Atmospheric Sciences: Definition, Status, and Outlook
Provides a recipe for building atmospheric-science ML benchmark datasets and lists domain-specific challenges and needed benchmarks for weather, climate and air quality.
ECMWFForschungszentrum Jülich / Jülich Supercomputing CentreNSF NCAR / UCARNVIDIAUniversity of Oklahoma (AI2ES)
Benchmark Dataset for Precipitation Forecasting by Post-Processing the Numerical Weather Prediction
KoMet benchmark pairs GDAPS-KIM NWP outputs with Korean station observations to test deep post-processing of precipitation, with open-source code.
AI-Based Climate Change Forecasting Models
Compares six AI approaches across four climate tasks, claiming hybrid physics-ML best at 6-month temperature RMSE; claims are unverified and methodology thin.
GLObal Building heights for Urban Studies (UT-GLOBUS) for city- and street- scale urban simulations: Development and first applications
UT-GLOBUS uses machine learning on spaceborne altimetry to map building heights for 1200+ cities, improving WRF-Urban intra-urban air temperature RMSE by 55% over a table-based approach.
WeatherBench Probability: A benchmark dataset for probabilistic medium-range weather forecasting along with deep learning baseline models
Extends WeatherBench with probabilistic metrics, an IFS ensemble baseline and three probabilistic ML models; Monte Carlo dropout underestimates uncertainty and no ML model matches IFS.
Can we integrate spatial verification methods into neural-network loss functions for atmospheric science?
Spatially enhanced loss functions using neighbourhood and spectral filters improve neural network convection forecasts, with spectral-filter models giving better-calibrated probabilities than pixelwise training.
Colorado State University (incl. CIRA)NOAA (incl. GFDL, PSL, GSL, NCEP/EMC)
QpefBD: A Benchmark Dataset Applied to Machine Learning for Minute-Scale Quantitative Precipitation Estimation and Forecasting
Introduces QpefBD, a minute-scale radar-based benchmark dataset for machine learning precipitation estimation and forecasting (abstract not provided; classified from title).
Machine Learning in Weather Prediction and Climate Analyses—Applications and Perspectives
Bibliometric review of about 500 papers on ML for weather prediction and climate analysis, mapping topics, methods, variables and countries since 2018.
In Search of Micromodels: Black-box Evaluation of Spatio-temporal Models in Normal and Extreme Conditions
Black-box evaluation of NWP temperature and precipitation outputs under normal and extreme conditions to find scenarios for lightweight data-driven micromodels on Edge devices.
A spatiotemporal stochastic climate model for benchmarking causal discovery methods for teleconnections
A stochastic vector-autoregressive climate model generates gridded teleconnection benchmarks for causal discovery, plus a new grid-level causal method outperforming existing approaches by orders of magnitude.
ENS-10: A Dataset For Post-Processing Ensemble Weather Forecasts
Introduces ENS-10, a 20-year, 10-member perturbed NWP ensemble dataset at 0.5° for ML post-processing, with baselines for 48-hour correction and extremes.
ClimART: A Benchmark Dataset for Emulating Atmospheric Radiative Transfer in Weather and Climate Models
ClimART offers over 10 million radiative transfer samples from climate conditions to benchmark neural emulators, showing shortcomings of prior datasets and architectures.
SubseasonalClimateUSA: A Dataset for Subseasonal Forecasting and Benchmarking
SubseasonalClimateUSA is a curated dataset and benchmark for 2-6 week US forecasts, comparing dynamical models with ten ML and deep learning methods.
Towards Representation Learning for Atmospheric Dynamics
AtmoDist is a self-supervised learned metric for atmospheric dynamics, trained to predict temporal distance between shuffled wind fields, and it improves GAN super-resolution of vorticity over MSE-based baselines.
RAIN-F+: The Data-Driven Precipitation Prediction Model for Integrated Weather Observations
RAIN-F+ fuses radar, station and satellite data over Korea and provides U-Net benchmarks, showing fusion beats radar-only precipitation nowcasts.
IowaRain: A Statewide Rain Event Dataset Based on Weather Radars and Quantitative Precipitation Estimation
Releases a 2016-2019 statewide Iowa rainfall event dataset built from NEXRAD radar with quantitative precipitation estimation, intended for ML and physical modeling.
CIRA Guide to Custom Loss Functions for Neural Networks in Environmental Sciences -- Version 1
A practical guide to writing custom loss functions for environmental and meteorological neural networks, including spatial verification scores like the fractions skill score and physical constraints.
EarthNet2021: A large-scale dataset and challenge for Earth surface forecasting as a guided video prediction task
EarthNet2021 dataset and challenge forecasts Sentinel-2 surface imagery conditioned on mesoscale future weather, enabling model intercomparison at 20 m resolution.
Identification and inter‐comparison of appropriate long‐term precipitation datasets using decision tree model and statistical matrix over China
A fast-frugal decision tree ranks seven long-term precipitation datasets against 813 China gauge grids, placing APHRODITE first; machine learning is a light evaluation tool.
RainNet: A Large-Scale Imagery Dataset and Benchmark for Spatial Precipitation Downscaling
Releases RainNet, a 62,400-pair precipitation downscaling dataset with 17 years of hourly maps and benchmarks 14 models, including deep learning baselines.
RainBench: Towards Global Precipitation Forecasting from Satellite Imagery
RainBench is a multimodal global benchmark combining simulated satellite data, ERA5 and IMERG precipitation, with the PyRain library and medium-range baselines.
RainBench: Enabling Data-Driven Precipitation Forecasting on a Global Scale
RainBench provides a global ML-ready dataset of precipitation and atmospheric variables for data-driven precipitation forecasting (no abstract; based on title).
EarthNet2021: A novel large-scale dataset and challenge for forecasting localized climate impacts
Introduces EarthNet2021, a 32,000-sample dataset and challenge for forecasting 20 m Sentinel-2 imagery conditioned on 1.28 km weather forecasts, with a new EarthNetScore ranking.
Boosting Weather Prediction with Machine Learning
Eos article introducing WeatherBench, an ERA5-based benchmark dataset for evaluating data-driven AI weather forecasting against standard baselines.
CloudCast: A Satellite-Based Dataset and Baseline for Forecasting Clouds
Releases CloudCast, a 70,080-frame satellite dataset of 10 cloud types over Europe at 3 km and 15-min resolution, with ConvLSTM, GAN and optical-flow video prediction baselines.
Texture‐based classification of high‐resolution precipitation forecasts with machine‐learning methods
Random forests classify continuous versus intermittent precipitation textures in AROME high-resolution forecasts, outperforming other classifiers and transferring to gridded observations for verification.
★ WeatherBench: A benchmark dataset for data-driven weather forecasting
WeatherBench introduces an ERA5-derived benchmark with standard metrics and baselines (linear regression, deep learning, physical models) for data-driven medium-range global weather forecasting.
ECMWFTechnical University of Munich (TUM)University of Washingtoncode
★ WeatherBench: A Benchmark Data Set for Data-Driven Weather Forecasting
WeatherBench provides an ERA5-derived benchmark with metrics and baselines for data-driven global medium-range forecasting, enabling direct model comparison.
ECMWFUniversity of WashingtonTechnical University of Munich (TUM)code
Cumulo: A Dataset for Learning Cloud Classes
Cumulo is a global dataset pairing one year of 1km MODIS imagery with CloudSat cloud labels, with a flow-based baseline for cloud classification.
Scientific Visualization of Climate Simulation Data for Deep Convolutional Neural Network
Conference abstract on visualizing climate simulation data as input for deep convolutional networks; no abstract details or results are given.
Also relevant (80)
CLIMATEAGENT: Multi-Agent Orchestration for Complex Climate Data Science Workflows
A multi-agent LLM framework completes climate data workflows with 100% task completion on an 85-task benchmark, beating a GPT-5 baseline.
Zephyrus: An Agentic Framework for Weather Science
Zephyrus is an LLM agent with a weather-data environment and ZephyrusBench, outperforming text-only baselines by up to 44 points in correctness.
Do data-driven models beat numerical models in forecasting weather extremes? A comparison of IFS HRES, Pangu-Weather, and GraphCast
Comparison of Pangu-Weather, GraphCast and IFS HRES finds data-driven models mostly beat IFS on global RMSE for 1-10 day forecasts and compete on temperature and wind extremes.
WeatherQA: Can Multimodal Language Models Reason about Severe Weather?
WeatherQA is a multimodal benchmark of 8,000+ image-text pairs for severe convective weather reasoning; GPT-4o shows a substantial gap to human forecasters.
LLMs & agentsAllen Institute for AI (Ai2, ACE)University of Washingtoncode
Validating Deep Learning Weather Forecast Models on Recent High-Impact Extreme Events
Compares GraphCast, Pangu-Weather, FourCastNet and HRES on three extreme events, finding ML models weaker on heatwave aggregates but better on a winter storm.
Enhancing Regional Climate Downscaling through Advances in Machine Learning
Review of machine learning for empirical climate downscaling, covering out-of-distribution generalization, extremes, and a proposed evaluation framework for downscaling algorithms.
Do AI models produce better weather forecasts than physics-based models? A quantitative evaluation case study of Storm Ciarán
Case study of Storm Ciarán finds FourCastNet, Pangu, and GraphCast capture synoptic structure but underestimate peak winds and miss fine warning-relevant features.
Sim2Real for Environmental Neural Processes
Sim2Real pre-trains a ConvCNP on reanalysis and fine-tunes on German weather stations, outperforming reanalysis-only or station-only training for surface air temperature interpolation.
Observations & remote sensingMicrosoft Research (AI for Science)University of Cambridge (Aardvark)
A Distributed Approach to Meteorological Predictions: Addressing Data Imbalance in Precipitation Prediction Models through Federated Learning and GANs
Applies SMOTE and GAN augmentation with federated learning to classify rare weather events in imbalanced tabular station data; no headline numbers given.
CCClim - A machine-learning powered cloud class climatology
Machine-learning-derived 1° global climatology of eight WMO cloud types from ESA Cloud_cci retrievals, almost gapless from 1982 to 2016.
Retrieval of surface solar irradiance from satellite imagery using machine learning: pitfalls and perspectives
MLP retrieval of surface solar irradiance from MSG imagery can beat a physical method but depends strongly on training-set similarity, highlighting pitfalls.
Finding the Perfect Fit: Applying Regression Models to ClimateBench v1.0
Evaluates Gaussian process, SVR and kernel ridge regression emulators on ClimateBench; GP performs best but is computationally expensive.
Connecting Large‐Scale Meteorological Patterns to Extratropical Cyclones in CMIP6 Climate Models Using Self‐Organizing Maps
Self-organizing maps link large-scale patterns to extratropical cyclones and show CMIP6 models produce too few cyclones and too zonal patterns over northeastern US.
Comparison of machine learning statistical downscaling and regional climate models for temperature, precipitation, wind speed, humidity and radiation over Europe under present conditions
Compares a machine-learning statistical downscaling model with six EURO-CORDEX regional climate models over Europe; the ML model scores better on most metrics before bias correction.
The skill of statistical downscaling in future climate with high-resolution climate models as pseudo-reality
Evaluates 21 statistical downscaling methods, including ML, against a 4 km pseudo-reality for Belgian extreme rainfall, recommending an ensemble of downscaling approaches.
Hierarchical Autoencoder-based Lossy Compression for Large-scale High-resolution Scientific Data
Autoencoder lossy compression reaches ratio 140 on benchmarks and 200 on 500 years of CESM climate output with negligible reconstruction error.
Comparison of a novel machine learning approach with dynamical downscaling for Australian precipitation
An MLP with a novel loss downscales Australian precipitation from 12–48 km to 5 km, outperforming dynamical downscaling on extremes with 15–85% lower bias at 12 km.
TCDetect: A New Method of Detecting the Presence of Tropical Cyclones Using Deep Learning
Lightweight deep learning detects tropical cyclone presence during simulations for online data reduction, reaching 92% recall on held-out ERA-Interim years.
Scaling Spherical CNNs
Scales spherical CNNs to large problems, reaching state of the art on QM9 and competitive results on several weather forecasting tasks; weather is a secondary target.
Mesogeos: A multi-purpose dataset for data-driven wildfire modeling in the Mediterranean
Mesogeos is a 1 km daily Mediterranean wildfire datacube with meteorology and ignition records, plus ML tracks for danger forecasting and burned-area estimation.
Probing the Skill of Random Forest Emulators for Physical Parameterizations Via a Hierarchy of Simple CAM6 Configurations
Random forest emulators of CAM6 physics skill declines as model complexity rises across dry, moist, and extra-convection configurations, indicating limits for RF physics emulation.
Trustworthy Artificial Intelligence for Environmental Sciences: An Innovative Approach for Summer School
Describes an AI2ES summer school teaching trustworthy and explainable AI for severe storms, tropical cyclones and space weather to 677 participants.
Explainability, uncertainty & physicsNOAA (incl. GFDL, PSL, GSL, NCEP/EMC)NSF NCAR / UCARUniversity of WashingtonColorado State University (incl. CIRA)University of Oklahoma (AI2ES)
chatClimate: Grounding Conversational AI in Climate Science
GPT-4 grounded with IPCC AR6 content via retrieval; expert-scored answers show the hybrid chatClimate more accurate than GPT-4 alone.
Inductive biases in deep learning models for weather prediction
Reviews and analyses inductive biases (data, objectives, losses, architectures, optimisation) of deep-learning weather prediction models to guide more efficient, probabilistic designs.
Creating and Evaluating Uncertainty Estimates with Neural Networks for Environmental-Science Applications
Introduces six neural network uncertainty methods and evaluation scores for environmental science, demonstrated on dewpoint profiles and Himawari-8 convection prediction.
Explainability, uncertainty & physicsColorado State University (incl. CIRA)NOAA (incl. GFDL, PSL, GSL, NCEP/EMC)
Finding the right XAI method -- A Guide for the Evaluation and Ranking of Explainable AI Methods in Climate Science
Proposes a framework to evaluate XAI methods for climate-science networks on robustness, faithfulness, randomization, complexity and localization, ranking methods on a temperature-map task.
Identifying climate models based on their daily output using machine learning
Machine learning classifiers identify individual days of CMIP6 and ICON-Sapphire output versus observations, showing model differences appear at daily timescales.
Evaluation of statistical climate reconstruction methods based on pseudoproxy experiments using linear and machine-learning methods
Pseudoproxy tests compare Bi-LSTM with linear PCR and CCA for millennial NH and NAE summer temperature reconstruction; Bi-LSTM is less skilful but captures extreme cooling.
Evaluation of Deep-Learning-Based Very Short-Term Rainfall Forecasts in South Korea
Evaluates deep-learning very-short-term rainfall forecasts over South Korea; abstract not provided, so specific skill numbers are unknown.
Outcomes of the WMO Prize Challenge to Improve Subseasonal to Seasonal Predictions Using Artificial Intelligence
WMO prize challenge winners used AI to post-process ECMWF S2S forecasts, beating bias-corrected references for weeks 3-6 precipitation and temperature.
Sub-seasonal to seasonalColumbia University (LEAP, Lamont-Doherty)Environment and Climate Change Canada (ECCC)ECMWFMax Planck Institute for MeteorologySwiss Data Science Center (SDSC)
Using machine learning to identify novel hydroclimate states
Unsupervised ML on tree-ring PDSI drought atlases finds outlier drought patterns in 80% of 21st-century years, showing unusual hydroclimate states.
Climate analysis & attributionNASAColumbia University (LEAP, Lamont-Doherty)
Compressing multidimensional weather and climate data into neural networks
Coordinate-based neural networks compress weather and climate grids by 300x to 3000x, beating SZ3 on error, and a 790x compressed WeatherBench loader raises RMSE by under 2%.
Repeatable high-resolution statistical downscaling through deep learning
U-Net and U-Net++ deep learning perfect-prognosis downscaling produces daily 1 km precipitation for Saxony with improved skill and full reproducibility.
AICCA: AI-driven Cloud Classification Atlas
A rotation-invariant autoencoder with clustering yields AICCA, a 22-year atlas of 42 unsupervised cloud classes from 198 million MODIS image patches.
Observations & remote sensingArgonne National LaboratoryUniversity of Chicago
Machine learning applications for weather and climate need greater focus on extremes
Position review arguing ML weather and climate studies must evaluate and improve performance on rare, high-return-period extremes, which remain poorly assessed.
Intercomparison of deep learning architectures for the prediction of precipitation fields
Compares deep learning architectures for European precipitation fields; U-Net performs best, and a single-level U-Net matches deeper versions on extremes.
Machine-learned cloud classes from satellite data for process-oriented climate model evaluation
Deep-learning cloud-type classifier trained on MODIS and CloudSat labels is applied to coarse satellite data to enable process-oriented evaluation of clouds in climate models.
Observations & remote sensingColumbia University (LEAP, Lamont-Doherty)
Enhancing classification performance of binary class imbalanced Data for Weather Forecasting using Machine Learning Classifiers
Compares logistic regression, tree ensembles and boosting for next-day rainfall classification on ten years of Australian station data with imbalance handling.
Deep Learning augmented Weather Prediction based on Cubic Sphere for global precipitation forecasting
Dataset of inputs and targets for deep-learning-augmented cubic-sphere weather prediction for global precipitation forecasting.
Learning from Precipitation Events in the Wider Domain to Improve the Performance of a Deep Learning–Based Precipitation Nowcasting Model
Training a deep learning radar nowcaster on all of Japan and then transfer learning to regions improves heavy-precipitation skill over locally trained models.
Current Training and Validation Weaknesses in Classification-Based Radiation Fog Nowcast Using Machine Learning Algorithms
XGBoost radiation fog nowcasts are evaluated with persistence and guessing baselines, showing high validation scores can hide operational failure and that blockwise CV inflates skill.
Synthetic Photovoltaic and Wind Power Forecasting Data
Open dataset of synthetic hourly PV and wind power for 393 German plants over 500 days, built from ICON-EU weather inputs, with baseline forecasts.
Evaluation of statistical downscaling methods for climate change projections over Spain: Present conditions with imperfect predictors (global climate model experiment)
Compares five statistical downscaling methods, including ANN, SVM and kernel ridge, over 18 CMIP5 GCMs for Spain, quantifying errors from imperfect GCM predictors.
Exploring Pathways to More Accurate Machine Learning Emulation of Atmospheric Radiative Transfer
A bidirectional RNN emulates shortwave radiation with fewer parameters than feed-forward nets, achieving under 1% flux error and a 4-fold CPU speedup.
Earthformer: Exploring Space-Time Transformers for Earth System Forecasting
Earthformer is a space-time transformer with cuboid attention that reaches state-of-the-art precipitation nowcasting and ENSO forecasting on real benchmarks.
Next Day Wildfire Spread: A Machine Learning Data Set to Predict Wildfire Spreading from Remote-Sensing Data
Curated nearly decade-long US wildfire spread dataset combining satellite fire data with weather and fuel variables, plus a neural network baseline for next-day spread.
Toward Foundation Models for Earth Monitoring: Proposal for a Climate Change Benchmark
Proposes a climate-change-oriented benchmark to develop foundation models for Earth observation imagery, with a call for community collaboration on evaluation.
Synthetic weather radar using hybrid quantum-classical machine learning
Hybrid quantum-classical CNNs generate synthetic weather radar from satellite and NWP inputs, proposed as a benchmark for quantum advantage. No quantitative results given in the abstract.
Evaluation of statistical downscaling methods for climate change projections over Spain: Future conditions with pseudo reality (transferability experiment)
Intercomparison of five statistical downscaling methods, including ANN, SVM and kernel ridge regression, reveals transferability problems for future climate over Spain.
Evaluating Predictive Uncertainty and Robustness to Distributional Shift Using Real World Data
Proposes uncertainty and robustness-to-shift metrics for regression, evaluated with baselines on the Shifts weather prediction dataset under distributional shift.
Model Evaluation and Intercomparison of Marine Warm Low Cloud Fractions With Neural Network Ensembles
Neural network ensembles compare marine low cloud fraction relationships in E3SM at two vertical resolutions against ERA5-CERES, showing shifts in meteorological drivers dominate differences.
Climate analysis & attributionNOAA (incl. GFDL, PSL, GSL, NCEP/EMC)
Exploring the tropical Pacific manifold in models and observations
Manifold learning characterizes the tropical Pacific climate attractor, revealing model biases that diverge between historical and warming-scenario climate model simulations.
Learning and Dynamical Models for Sub-seasonal Climate Forecasting: Comparison and Collaboration
Comparison of modern ML models against SubX dynamical models for US sub-seasonal forecasting finds ML wins on average, and feeding dynamical forecasts into ML improves skill.
A Benchmark to Test Generalization Capabilities of Deep Learning Methods to Classify Severe Convective Storms in a Changing Climate
CNN trained on present-day WRF storms classifies strongly rotating thunderstorms in end-of-century climate simulations with skill, and explanations reveal learned physical features.
An Evaluation of Machine Learning and Deep Learning Models for Drought Prediction using Weather Data
Benchmarks 16 classical and 16 deep learning models predicting drought levels from 90 days of 18 meteorological indicators, with no single model best across metrics.
A Comparative Study on the Skill of CMIP6 Models to Preserve Daily Spatial Patterns of Monsoon Rainfall Over India
A probabilistic graphical model evaluates how well CMIP6 models reproduce daily spatial monsoon rainfall patterns and clusters over India.
Towards a Collective Agenda on AI for Earth Science Data Analysis
Position paper proposing research agendas for AI in Earth science and remote sensing to move beyond disciplinary comfort zones.
Towards implementing artificial intelligence post-processing in weather and climate: proposed actions from the Oxford 2019 workshop
Oxford workshop position on AI post-processing for weather and climate, releasing five archived ML-ready datasets with baseline notebooks to encourage trustworthy, FAIR and physically informed methods.
Post-processing & calibrationThe Alan Turing InstituteKarlsruhe Institute of Technology (KIT)Met OfficeNSF NCAR / UCAR
Storm-scale Ensemble-based Severe Weather Guidance: Development of an Object-based Verification Framework and Applications of Machine Learning
Object-based verification of storm-scale mesocyclone probabilities, plus evaluation of ML calibrations of Warn-on-Forecast severe weather guidance.
Copula-based synthetic data augmentation for machine-learning emulators
Copula-generated synthetic data augments training of a neural emulator of longwave radiation, reducing mean absolute error by up to 62% in a toy setting.
Meta-modeling strategy for data-driven forecasting
Meta-modeling uses cheap low-fidelity models to skip costly high-fidelity evaluations when predicting temperature fields, giving an adaptive training strategy for geophysical emulation.
CNN Profiler on Polar Coordinate Images for Tropical Cyclone Structure Analysis
CNN on polar-coordinate satellite images profiles full tropical cyclone structure parameters and wind profiles, with a new labeled benchmark dataset as baseline.
HECT: High-Dimensional Ensemble Consistency Testing for Climate Models
Uses probabilistic classifiers, including tree-based models and neural networks, for a goodness-of-fit test checking CESM climate simulation ensemble consistency after code changes.
ExGAN: Adversarial Generation of Extreme Samples
GAN combined with extreme value theory generates realistic extreme samples, evaluated on US precipitation, with constant-time generation at a given extremeness probability.
Data-driven medium-range weather prediction with a Resnet pretrained on climate simulations: A new model for WeatherBench
A residual CNN pretrained on climate model output and fine-tuned on reanalysis gives WeatherBench 5-day forecasts that outperform prior submissions and rival a physical baseline.
Precipitation Merging Based on the Triple Collocation Method Across Mainland China
Triple collocation merges CMORPH, PERSIANN and ERA5 precipitation over China, improving on each input product against rain-gauge data; the method is statistical rather than ML.
Unsupervised Severe Weather Detection Via Joint Representation Learning Over Textual and Weather Data
Unsupervised representation learning over text and weather data improves clustering for detecting severe weather events such as windstorms, floods and tornado outbreaks.
A review of radar-based nowcasting of precipitation and applicable machine learning techniques
Reviews radar-based precipitation nowcasting methods from environmental science and the machine learning techniques applicable to them, arguing for closer collaboration between the two communities.
Configuration and intercomparison of deep learning neural models for statistical downscaling
Systematic VALUE-based comparison of CNN downscaling against linear benchmarks over Europe finds CNNs help mainly for precipitation and temperature extremes, including under warm-period extrapolation.
A Causal Inference Model Based on Random Forests to Identify the Effect of Soil Moisture on Precipitation
Random-forest-based Granger causal framework isolates soil moisture to precipitation feedback over the US and uses it to benchmark Earth system models for land-atmosphere coupling.
Causal networks for climate model evaluation and constrained projections
Causal networks from sea level pressure fingerprints objectively evaluate climate models and better constrain precipitation projections than storm-track metrics.
Image Labeler: Label Earth Science Images for Machine Learning
Web-based Image Labeler tool speeds labeling of Earth science images, such as hurricane imagery, to build shareable datasets for deep learning.
Benchmarking Tropical Cyclone Rapid Intensification with Satellite Images and Attention-based Deep Models
Benchmark of satellite images for tropical cyclone rapid intensification, with an attention-based spatiotemporal model showing deep learning feasibility for RI prediction.
Gauge Equivariant Convolutional Networks and the Icosahedral CNN
Gauge-equivariant CNNs on the icosahedron improve segmentation of global climate patterns, alongside omnidirectional images, with a single conv2d implementation.
Machine Learning techniques to evaluate reproducibility of climate model simulations on hybrid architectures.
AGU abstract-less entry applying machine learning to assess reproducibility of climate model simulations across hybrid HPC architectures.
A test case for application of convolutional neural networks to spatio-temporal climate data: Re-identifying clustered weather patterns
Uses unsupervised clustering to auto-label North American weather regimes in climate model output, then trains CNNs to re-identify them, reaching 90–94% accuracy.
Intercomparison of machine learning methods for statistical downscaling: the case of daily and extreme precipitation
Intercomparison of machine learning methods for statistical downscaling of daily and extreme precipitation; abstract unavailable, classification based on title.
Cluster-based analysis of multi-model climate ensembles
Applies density-based clustering to choose grid-cell model subsets for multi-model means of tropospheric ozone, improving agreement with satellite climatology in chemistry-climate models.
Machine Learning Predictions of a Multiresolution Climate Model Ensemble
Random forests trained on a multiresolution perturbed-parameter GCM ensemble predict high-resolution TOA flux and precipitation using few high-resolution runs, cutting simulation cost.
Deep Learning for Precipitation Nowcasting: A Benchmark and A New Model
TrajGRU learns location-variant recurrent connections for radar nowcasting and is benchmarked on a Hong Kong Observatory dataset, outperforming ConvLSTM.