Improving Statistical Postprocessing for Extreme Wind Speeds using Tuned Weighted Scoring Rules

2025-03-10

AI Summary

Conventional statistical post-processing methods for numerical weather prediction exhibit suboptimal performance in probabilistic forecasting of extreme wind speeds. Method: This paper proposes a wCRPS-optimized statistical post-processing framework, featuring a novel multi-objective hyperparameter optimization algorithm that automatically learns wind-speed-dependent weighting functions (e.g., shifted Gaussian CDF), thereby overcoming the traditional body–tail trade-off limitation. It integrates convolutional neural networks (CNNs) with flexible distributional assumptions—including truncated normal, log-normal, generalized extreme value (GEV), and adaptive mixture models. Contribution/Results: The framework achieves a 12% improvement in tail-region CRPS for extreme wind speeds at the 48-hour lead time, while preserving accuracy over the bulk of the distribution. We further demonstrate, for the first time, that CNN-based architectures jointly enhance both routine and extreme wind speed forecast skill. The learned weighting functions are model-specific, whereas distributional choices exert only marginal influence on overall performance.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Natural Language Processing: Learning & Optimization for NLP
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Recent statistical postprocessing methods for wind speed forecasts have incorporated linear models and neural networks to produce more skillful probabilistic forecasts in the low-to-medium wind speed range. At the same time, these methods struggle in the high-to-extreme wind speed range. In this work, we aim to increase the performance in this range by training using a weighted version of the continuous ranked probability score (wCRPS). We develop an approach using shifted Gaussian cdf weight functions, whose parameters are tuned using a multi-objective hyperparameter tuning algorithm that balances performance on low and high wind speed ranges. We explore this approach for both linear models and convolutional neural network models combined with various parametric distributions, namely the truncated normal, log-normal, and generalized extreme value distributions, as well as adaptive mixtures. We apply these methods to forecasts from KNMI's deterministic Harmonie-Arome numerical weather prediction model to obtain probabilistic wind speed forecasts in the Netherlands for 48 hours ahead. For linear models we observe that even with a tuned weight function, training using the wCRPS produces a strong body-tail trade-off, where increased performance on extremes comes at the price of lower performance for the bulk of the distribution. For the best models using convolutional neural networks, we find that using a tuned weight function the performance on extremes can be increased without a significant deterioration in performance on the bulk. The best-performing weight function is shown to be model-specific. Finally, the choice of distribution has no significant impact on the performance of our models.

Problem

• Enhance extreme wind speed forecast accuracy using weighted scoring rules.
• Balance performance between low and high wind speeds via tuned weight functions.
• Evaluate linear and neural network models with various parametric distributions.

Innovation

• Weighted CRPS for extreme wind speed training
• Multi-objective tuning balances speed ranges
• Model-specific weight functions optimize performance