Deep Learning‐Based Super‐Resolution Climate Simulator‐Emulator Framework for Urban Heat Studies.

This proof‐of‐concept study couples machine learning and physical modeling paradigms to develop a computationally efficient simulator‐emulator framework for generating super‐resolution (<250 m) urban climate information, that is required by many sectors. To this end, a regional climate model/simulator is applied over the city of Montreal, for the summers of 2019 and 2020, at 2.5 km (LR) and 250 m (HR) resolutions, which are used to train and validate the proposed super‐resolution deep learning (DL) model/emulator. The DL model uses an efficient sub‐pixel convolution layer to generate HR information from LR data, with adversarial training applied to improve physical consistency. The DL model reduces temperature errors significantly over urbanized areas present in the LR simulation, while also demonstrating considerable skill in capturing the magnitude and location of heat stress indicators. These results portray the value of the innovative simulator‐emulator framework, that can be extended to other seasons/periods, variables and regions. Plain Language Summary: One of the major barriers in undertaking super‐resolution (<250 m) urban climate simulations to generate climate and climate change information at high spatial and temporal resolutions, as required by many sectors, is their high computational cost. New approaches are therefore required to overcome this barrier. This paper makes use of the unique opportunity to couple machine learning and physical modeling to develop a computationally efficient simulator‐emulator framework to generate super‐resolution climate information. The trained deep neural network model generates high‐resolution urban climate data from low‐resolution (LR) inputs, considering also the urban morphology fields and inter‐variable relationships to improve output realism. The developed framework, applied to urban heat‐related variables, demonstrates the high potential of this approach as it captures well the magnitude and location of heat stress indicators. The generic nature of the developed framework makes it even more promising as it can be applied to other climate variables, periods and regions. Key Points: Innovative simulator‐emulator framework proposed to generate super‐resolution climate informationNew framework applied at city‐scale demonstrates considerable skill for urban heat‐related variablesThe generic nature of the framework enables adaptability to other climate variables and regions [ABSTRACT FROM AUTHOR]