Work

On the suitability of a convolutional neural network based RCM-emulator for fine spatio-temporal precipitation

Year: 2024

Type: article

Abstract: High resolution climate models are necessary to capture local precipitation but are too expensive to explore the uncertainties associated with future projections. To solve this resolution-uncertainty dilemma, Doury et al (2022) proposed a neural network based RCM-emulator for the near-surface temperature, at a daily and 12km-resolution. It uses existing RCM simulations to learn the relationship between low-resolution predictors and high resolution surface variables. When trained the emulator can be applied to any GCM simulation to produce ensembles of high resolution emulated simulations. This study assess the suitability of applying the RCM-emulator for precipitation thanks to a novel asymmetric loss function targeting to reproduce the entire precipitation distribution over any grid point. In perfect model evaluation, the resulting emulator shows striking ability to reproduce the RCM original series with an excellent spatio-temporal correlation. In particular, a very good behaviour is obtained for the two tails of the distribution, measured by the number of dry days and the 99th quantile. Moreover, it creates consistent precipitation objects with a slight lack of precision. The emulator quality holds for all simulations of the same RCM, with any driving GCM, ensuring transferability of the tool to GCMs never downscaled by the RCM. A first showcase of downscaling GCM simulations showed that the RCMemulator brings significant added-value with respect to the GCM as it produces adequate high resolution spatial structure and extremes’ intensity. Nevertheless, further work is needed to understand the differences that occur with the RCM and establish a relevant evaluation framework for GCM applications.

Source: Climate Dynamics

DOI: 10.1007/s00382-024-07350-8

Authors: Antoine Doury, Samuel Somot, Sebastien Gadat