mlsanityData to 7 Oct 2026

Machine learning methods for postprocessing ensemble forecasts of wind gusts: A systematic comparison

arXiv 2106.09512 ↗First posted 17 Jun 2021stat.ML cs.LG physics.ao-phcited by 34

Not on the map. Why?

Abstract (arXiv)

Postprocessing ensemble weather predictions to correct systematic errors has become a standard practice in research and operations. However, only few recent studies have focused on ensemble postprocessing of wind gust forecasts, despite its importance for severe weather warnings. Here, we provide a comprehensive review and systematic comparison of eight statistical and machine learning methods for probabilistic wind gust forecasting via ensemble postprocessing, that can be divided in three groups: State of the art postprocessing techniques from statistics (ensemble model output statistics (EMOS), member-by-member postprocessing, isotonic distributional regression), established machine learning methods (gradient-boosting extended EMOS, quantile regression forests) and neural network-based approaches (distributional regression network, Bernstein quantile network, histogram estimation network). The methods are systematically compared using six years of data from a high-resolution, convection-permitting ensemble prediction system that was run operationally at the German weather service, and hourly observations at 175 surface weather stations in Germany. While all postprocessing methods yield calibrated forecasts and are able to correct the systematic errors of the raw ensemble predictions, incorporating information from additional meteorological predictor variables beyond wind gusts leads to significant improvements in forecast skill. In particular, we propose a flexible framework of locally adaptive neural networks with different probabilistic forecast types as output, which not only significantly outperform all benchmark postprocessing methods but also learn physically consistent relations associated with the diurnal cycle, especially the evening transition of the planetary boundary layer.

Key facts. Cited by 34 papers in the map (data through 7 Oct 2026). Builds on 11 of its 88 references that are in the map. First posted on arXiv 17 Jun 2021.

Builds on

11 of 88 references.

References · 11

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2015ranger: A Fast Implementation of Random Forests for High Dimensional Data in C++ and Rcited by 171
  3. 2018Neural networks for post-processing ensemble weather forecastscited by 95
  4. 2019Isotonic Distributional Regressioncited by 28
  5. 2013Uncertainty Quantification in Complex Simulation Models Using Ensemble Copula Couplingcited by 21
  6. 2019Deep Distribution Regressioncited by 16
  7. 2014Log-normal distribution based EMOS models for probabilistic wind speed forecastingcited by 10
  8. 2016Combining predictive distributions for statistical post-processing of ensemble forecastscited by 9
  9. 2020Distributional (Single) Index Modelscited by 6
  10. 2020Truncated generalized extreme value distribution based EMOS model for calibration of wind speed ensemble forecastscited by 4
  11. 2021Calibration of wind speed ensemble forecasts for power generationcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 7; 2023: 6; 2024: 7; 2025: 12; 2026: 1.

34 papers.

Sort:

By area:

  1. 2022Generative machine learning methods for multivariate ensemble post-processingcited by 26
  2. 2024Uncertainty quantification for data-driven weather modelscited by 24
  3. 2023Precipitation nowcasting with generative diffusion modelscited by 22
  4. 2023Evidential Deep Learning: Enhancing Predictive Uncertainty Estimation for Earth System Science Applicationscited by 11
  5. 2025An Axiomatic Assessment of Entropy- and Variance-based Uncertainty Quantification in Regressioncited by 9
  6. 2021Self-Attentive Ensemble Transformer: Representing Ensemble Interactions in Neural Networks for Earth System Modelscited by 7
  7. 2023Postprocessing of Ensemble Weather Forecasts Using Permutation-invariant Neural Networkscited by 7
  8. 2022Aggregating distribution forecasts from deep ensemblescited by 5
  9. 2022Convolutional autoencoders for spatially-informed ensemble post-processingcited by 5
  10. 2022A two-step machine learning approach to statistical post-processing of weather forecasts for power generationcited by 4
  11. 2024Graph Neural Networks and Spatial Information Learning for Post-Processing Ensemble Weather Forecastscited by 4
  12. 2022Distributional regression and its evaluation with the CRPS: Bounds and convergence of the minimax riskcited by 3
22 more
  1. 2022Analysis, Characterization, Prediction and Attribution of Extreme Atmospheric Events with Machine Learning: a Reviewcited by 3
  2. 2023Extreme heatwave sampling and prediction with analog Markov chain and comparisons with deep learningcited by 3
  3. 2025Generating ensembles of spatially-coherent in-situ forecasts using flow matchingcited by 3
  4. 2024Improving Model Chain Approaches for Probabilistic Solar Energy Forecasting through Post-processing and Machine Learningcited by 2
  5. 2024Enhancing multivariate post-processed visibility predictions utilizing CAMS forecastscited by 2
  6. 2024Distributional Regression U-Nets for the Postprocessing of Precipitation Ensemble Forecastscited by 2
  7. 2024Self-attentive Transformer for Fast and Accurate Postprocessing of Temperature and Wind Speed Forecastscited by 2
  8. 2025Graph Neural Networks for Enhancing Ensemble Forecasts of Extreme Rainfallcited by 2
  9. 2023Ensemble weather forecast post-processing with a flexible probabilistic neural network approachcited by 1
  10. 2023Statistical post-processing of visibility ensemble forecastscited by 1
  11. 2024Machine learning-based probabilistic forecasting of solar irradiance in Chilecited by 1
  12. 2025Uncertainty Quantification of Wind Gust Predictions in the Northeast United States: An Evidential Neural Network and Explainable Artificial Intelligence Approachcited by 1
  13. 2025Learning low-dimensional representations of ensemble forecast fields using autoencoder-based methodscited by 1
  14. 2025Improving Predictions of Convective Storm Wind Gusts through Statistical Post-Processing of Neural Weather Modelscited by 1
  15. 2025Improving Medium Range Severe Weather Prediction through Transformer Post-processing of AI Weather Forecastscited by 1
  16. 2022Physics-constrained deep learning postprocessing of temperature and humiditycited by 0
  17. 2025Analogue Forecast System for Daily Precipitation Prediction Using Autoencoder Feature Extraction: Application in Hong Kongcited by 0
  18. 2025Bayesian Deep Learning for Convective Initiation Nowcasting Uncertainty Estimationcited by 0
  19. 2025CNN-based Surface Temperature Forecasts with Ensemble Numerical Weather Predictioncited by 0
  20. 2025Quantile Regression, Variational Autoencoders, and Diffusion Models for Uncertainty Quantification: A Spatial Analysis of Sub-seasonal Wind Speed Predictioncited by 0
  21. 2025Achieving Skilled and Reliable Daily Probabilistic Forecasts of Wind Power at Subseasonal-to-Seasonal Timescales over Francecited by 0
  22. 2026Improving the sharpness in neural network-based parametric post-processing of ensemble forecastscited by 0