Predicting Weather Uncertainty with Deep Convnets
Abstract
Modern weather forecast models perform uncertainty quantification using ensemble prediction systems, which collect nonparametric statistics based on multiple perturbed simulations. To provide accurate estimation, dozens of such computationally intensive simulations must be run. We show that deep neural networks can be used on a small set of numerical weather simulations to estimate the spread of a weather forecast, significantly reducing computational cost. To train the system, we both modify the 3D U-Net architecture and explore models that incorporate temporal data. Our models serve as a starting point to improve uncertainty quantification in current real-time weather forecasting systems, which is vital for predicting extreme events.
Cite
@article{arxiv.1911.00630,
title = {Predicting Weather Uncertainty with Deep Convnets},
author = {Peter Grönquist and Tal Ben-Nun and Nikoli Dryden and Peter Dueben and Luca Lavarini and Shigang Li and Torsten Hoefler},
journal= {arXiv preprint arXiv:1911.00630},
year = {2019}
}
Comments
Poster presentation at NeurIPS2019 "Machine Learning and the Physical Sciences" Workshop