Articles | Volume 20, issue 4
Atmos. Chem. Phys., 20, 2303–2317, 2020
https://doi.org/10.5194/acp-20-2303-2020
Atmos. Chem. Phys., 20, 2303–2317, 2020
https://doi.org/10.5194/acp-20-2303-2020

Technical note 26 Feb 2020

Technical note | 26 Feb 2020

Technical note: Deep learning for creating surrogate models of precipitation in Earth system models

Theodore Weber et al.

Model code and software

A Deep Neural Network approach for estimating precipitation fields in Earth System Models T. Weber, A. Corotan, B. Hutchinson, B. Kravitz, and R. P. Link https://github.com/hutchresearch/deep_climate_emulator

Download
Short summary
Climate model emulators can save computer time but are less accurate than full climate models. We use neural networks to build emulators of precipitation, trained on existing climate model runs. By doing so, we can capture nonlinearities and how the past state of a model (to some degree) shapes the future state. Our emulator outperforms a persistence forecast of precipitation.
Altmetrics
Final-revised paper
Preprint