Uncertainty quantification for data-driven weather models

arXiv:2403.13458v1 Announce Type: cross
Abstract: Artificial intelligence (AI)-based data-driven weather forecasting models have experienced rapid progress over the last years. Recent studies, with models trained on reanalysis data, achieve impressive results and demonstrate substantial improvements over state-of-the-art physics-based numerical weather prediction models across a range of variables and evaluation metrics. Beyond improved predictions, the main advantages of data-driven weather models are their substantially lower computational costs and the faster generation of forecasts, once a model has been trained. However, most …

abstract art artificial artificial intelligence arxiv beyond data data-driven evaluation evaluation metrics forecasting improvements intelligence metrics numerical numerical weather prediction physics physics.ao-ph prediction prediction models predictions progress quantification results stat.ap state stat.ml studies type uncertainty variables weather weather forecasting weather prediction