Comment on gmd-2020-412

Abstract

This study develops a neural network-based approach for emulating high-resolution modeled precipitation data with comparable statistical properties but at greatly reduced computational cost. The key idea is to use combination of low-and highresolution simulations to train a neural network to map from the former to the latter. Specifically, we define two types of CNNs, one that stacks variables directly and one that encodes each variable before stacking, and we train each CNN type both with a conventional loss function, such as mean square error (MSE), and with a conditional generative adversarial network (CGAN), for a total of four CNN variants. We compare the four new CNN-derived high-resolution precipitation results with precipitation generated from original high resolution simulations, a bilinear interpolater and the state-of-the-art CNN-based super-resolution (SR) technique. Results show that the SR technique produces results similar to those of the bilinear interpolator with smoother spatial and temporal distributions and smaller data variabilities and extremes than the original high resolution simulations.While the new CNNs trained by MSE generate better results over some regions than the interpolator and SR technique do, their predictions are still not as close as the original high resolution simulations. The CNNs trained by CGAN generate more realistic and physically reasonable results, better capturing not only data variability in time and space but also extremes such as intense and long-lasting storms. The new proposed CNN-based downscaling approach can downscale precipitation from 50 km to 12 km in 14 min for 30 years once the network is trained (training takes 4 hours using 1 GPU), while the conventional dynamical downscaling would take 1 month using 600 CPU cores to generate simulations at the resolution of 12 km over contiguous United States. IntroductionEarth system models (ESMs) integrate the interactions of atmospheric, land, ocean, ice, and biosphere and generate principal data products used across many disciplines to characterize the likely impacts and uncertainties of climate change (Heavens et al., 2013;Stouffer et al., 2017). The computationally demanding nature of ESMs, however, limits their spatial resolution mostly to between 100 and 300 km. Such resolutions are not sufficient to resolve critical physical processes such as clouds, which play a key role in determining the Earth's climate by transporting heat and moisture, reflecting and absorbing radiation, and producing rain. Moreover, ESMs cannot assess stakeholder-relevant local impacts of significant changes in the attributes of these processes (at scales of 1-10 km; Gutowski Jr et al., 2020). Higher-resolution simulations covering the entire globe 1