Generative machine learning methods for multivariate ensemble post-processing
This work addresses the challenge of accurately modeling multivariate dependencies in weather forecasting for meteorologists and climate scientists, representing an incremental advance by introducing a nonparametric, data-driven approach to overcome limitations of existing two-step methods.
The authors tackled the problem of systematic errors in multivariate ensemble weather forecasts by proposing a generative machine learning method for post-processing, which directly samples from the forecast distribution using a neural network and shows significant improvements in spatial dependencies over state-of-the-art methods in case studies on temperature and wind speed forecasting in Germany.
Ensemble weather forecasts based on multiple runs of numerical weather prediction models typically show systematic errors and require post-processing to obtain reliable forecasts. Accurately modeling multivariate dependencies is crucial in many practical applications, and various approaches to multivariate post-processing have been proposed where ensemble predictions are first post-processed separately in each margin and multivariate dependencies are then restored via copulas. These two-step methods share common key limitations, in particular the difficulty to include additional predictors in modeling the dependencies. We propose a novel multivariate post-processing method based on generative machine learning to address these challenges. In this new class of nonparametric data-driven distributional regression models, samples from the multivariate forecast distribution are directly obtained as output of a generative neural network. The generative model is trained by optimizing a proper scoring rule which measures the discrepancy between the generated and observed data, conditional on exogenous input variables. Our method does not require parametric assumptions on univariate distributions or multivariate dependencies and allows for incorporating arbitrary predictors. In two case studies on multivariate temperature and wind speed forecasting at weather stations over Germany, our generative model shows significant improvements over state-of-the-art methods and particularly improves the representation of spatial dependencies.