Abstract. Recent advances in deep convolutional neural network (CNN)-based super resolution can be used to downscale atmospheric chemistry simulations with substantially higher accuracy than conventional downscaling methods. This work both demonstrates the downscaling capabilities of modern CNN-based single image super resolution and video super-resolution schemes and develops modifications to these schemes to ensure they are appropriate for use with physical science data. The CNN-based video super-resolution schemes in particular incur only 39 % to 54 % of the grid-cell-level error of interpolation schemes and generate outputs with extremely realistic small-scale variability based on multiple perceptual quality metrics while performing a large (8×10) increase in resolution in the spatial dimensions. Methods are introduced to strictly enforce physical conservation laws within CNNs, perform large and asymmetric resolution changes between common model grid resolutions, account for non-uniform grid-cell areas, super-resolve lognormally distributed datasets, and leverage additional inputs such as high-resolution climatologies and model state variables. High-resolution chemistry simulations are critical for modeling regional air quality and for understanding future climate, and CNN-based downscaling has the potential to generate these high-resolution simulations and ensembles at a fraction of the computational cost.
Geiss, Andrew, et al. "Downscaling atmospheric chemistry simulations with physically consistent deep learning." Geoscientific Model Development (Online), vol. 15, no. 17, Sep. 2022. https://doi.org/10.5194/gmd-15-6677-2022
Geiss, Andrew, Silva, Sam J., & Hardin, Joseph C. (2022). Downscaling atmospheric chemistry simulations with physically consistent deep learning. Geoscientific Model Development (Online), 15(17). https://doi.org/10.5194/gmd-15-6677-2022
Geiss, Andrew, Silva, Sam J., and Hardin, Joseph C., "Downscaling atmospheric chemistry simulations with physically consistent deep learning," Geoscientific Model Development (Online) 15, no. 17 (2022), https://doi.org/10.5194/gmd-15-6677-2022
@article{osti_1885494,
author = {Geiss, Andrew and Silva, Sam J. and Hardin, Joseph C.},
title = {Downscaling atmospheric chemistry simulations with physically consistent deep learning},
annote = {Abstract. Recent advances in deep convolutional neural network (CNN)-based super resolution can be used to downscale atmospheric chemistry simulations with substantially higher accuracy than conventional downscaling methods. This work both demonstrates the downscaling capabilities of modern CNN-based single image super resolution and video super-resolution schemes and develops modifications to these schemes to ensure they are appropriate for use with physical science data. The CNN-based video super-resolution schemes in particular incur only 39 % to 54 % of the grid-cell-level error of interpolation schemes and generate outputs with extremely realistic small-scale variability based on multiple perceptual quality metrics while performing a large (8×10) increase in resolution in the spatial dimensions. Methods are introduced to strictly enforce physical conservation laws within CNNs, perform large and asymmetric resolution changes between common model grid resolutions, account for non-uniform grid-cell areas, super-resolve lognormally distributed datasets, and leverage additional inputs such as high-resolution climatologies and model state variables. High-resolution chemistry simulations are critical for modeling regional air quality and for understanding future climate, and CNN-based downscaling has the potential to generate these high-resolution simulations and ensembles at a fraction of the computational cost.},
doi = {10.5194/gmd-15-6677-2022},
url = {https://www.osti.gov/biblio/1885494},
journal = {Geoscientific Model Development (Online)},
issn = {ISSN 1991-9603},
number = {17},
volume = {15},
place = {Germany},
publisher = {Copernicus GmbH},
year = {2022},
month = {09}}
Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part IIIhttps://doi.org/10.1007/978-3-319-24574-4_28
Twenty-Seventh International Joint Conference on Artificial Intelligence {IJCAI-18}, Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligencehttps://doi.org/10.24963/ijcai.2018/759