Reconstructing High Resolution ESM Data Through a Novel Fast Super Resolution Convolutional Neural Network (FSRCNN)

• Published in: Geophysical research letters Vol. 49; no. 4
• Main Authors: Passarella, Linsey S., Mahajan, Salil, Pal, Anikesh, Norman, Matthew R.
• Format: Journal Article
• Language: English
• Published: Washington John Wiley & Sons, Inc 28.02.2022; American Geophysical Union
• Subjects: Artificial neural networks; Climate change; Climatic data; Computer applications; Control simulation; Convolution; Deep learning; Dimensions; Environmental assessment; GEOSCIENCES; Global climate; High resolution; Image reconstruction; Image resolution; Local climates; Methods; Neural networks; Resolution; Surface temperature
• ISSN: 0094-8276; 1944-8007
• DOI: 10.1029/2021GL097571

We present the first application of a fast super resolution convolutional neural network (FSRCNN) based approach for downscaling earth system model (ESM) simulations. Unlike other SR approaches, FSRCNN uses the same input feature dimensions as the low resolution input. This allows it to have smaller convolution layers, avoiding over‐smoothing, and reducing computational costs. We adapt the FSRCNN to improve reconstruction on ESM data, we term the FSRCNN‐ESM. We use high‐resolution (∼0.25°) monthly averaged model output of five surface variables over North America from the US Department of Energy's Energy Exascale Earth System Model's control simulation. These high‐resolution and corresponding coarsened low‐resolution (∼1°) pairs of images are used to train the FSRCNN‐ESM and evaluate its use as a downscaling approach. We find that FSRCNN‐ESM outperforms FSRCNN and other super‐resolution methods in reconstructing high resolution images producing finer spatial scale features with better accuracy for surface temperature, surface radiative fluxes, and precipitation. Plain Language Summary High resolution global climate data is computationally expensive to run but crucial for assessing climate change effects at local and regional scales. Here, we adapt a new deep learning technique, called fast super‐resolution convolutional neural network, to remap climate data from low resolution to high resolution grids. This approach is faster and more accurate for statistical downscaling climate data compared to other prevalent methods. Key Points We present a fast super resolution convolutional neural network (FSRCNN) based approach for downscaling gridded earth system model (ESM) data FSRCNN‐ESM's reconstruction of high resolution spatial patterns improves upon both traditional and machine learning downscaling methods The FSRCNN is computationally less expensive to train than other machine learning downscaling methods