TY - JOUR
T1 - Data-driven global weather predictions at high resolutions
AU - Taylor, John A.
AU - Larraondo, Pablo
AU - de Supinski, Bronis R.
N1 - Publisher Copyright:
© The Author(s) 2021.
PY - 2022/3
Y1 - 2022/3
N2 - Society has benefited enormously from the continuous advancement in numerical weather prediction that has occurred over many decades driven by a combination of outstanding scientific, computational and technological breakthroughs. Here, we demonstrate that data-driven methods are now positioned to contribute to the next wave of major advances in atmospheric science. We show that data-driven models can predict important meteorological quantities of interest to society such as global high resolution precipitation fields (0.25°) and can deliver accurate forecasts of the future state of the atmosphere without prior knowledge of the laws of physics and chemistry. We also show how these data-driven methods can be scaled to run on supercomputers with up to 1024 modern graphics processing units and beyond resulting in rapid training of data-driven models, thus supporting a cycle of rapid research and innovation. Taken together, these two results illustrate the significant potential of data-driven methods to advance atmospheric science and operational weather forecasting.
AB - Society has benefited enormously from the continuous advancement in numerical weather prediction that has occurred over many decades driven by a combination of outstanding scientific, computational and technological breakthroughs. Here, we demonstrate that data-driven methods are now positioned to contribute to the next wave of major advances in atmospheric science. We show that data-driven models can predict important meteorological quantities of interest to society such as global high resolution precipitation fields (0.25°) and can deliver accurate forecasts of the future state of the atmosphere without prior knowledge of the laws of physics and chemistry. We also show how these data-driven methods can be scaled to run on supercomputers with up to 1024 modern graphics processing units and beyond resulting in rapid training of data-driven models, thus supporting a cycle of rapid research and innovation. Taken together, these two results illustrate the significant potential of data-driven methods to advance atmospheric science and operational weather forecasting.
KW - Unet
KW - Weather prediction
KW - data-driven modelling
KW - deep neural networks
KW - scalable neural networks
UR - http://www.scopus.com/inward/record.url?scp=85113143890&partnerID=8YFLogxK
U2 - 10.1177/10943420211039818
DO - 10.1177/10943420211039818
M3 - Article
SN - 1094-3420
VL - 36
SP - 130
EP - 140
JO - International Journal of High Performance Computing Applications
JF - International Journal of High Performance Computing Applications
IS - 2
ER -