ClimateBench: A benchmark dataset for data-driven climate projections

3rd NOAA Workshop on Leveraging AI in Environmental Sciences

Many different emission pathways exist that are compatible with the Paris climate agreement, and many more are possible 33 that miss that target. While some of the most complex Earth System Models have simulated a small selection of Shared 34 Socioeconomic Pathways, it is impractical to use these expensive models to fully explore the space of possibilities. Such 35 explorations therefore mostly rely on one-dimensional impulse response models, or simple pattern scaling approaches to 36 approximate the physical climate response to a given scenario. Here we present ClimateBench - a benchmarking framework 37 based on a suite of CMIP, AerChemMIP and DAMIP simulations performed by a full complexity Earth System Model, and 38 a set of baseline machine learning models that emulate its response to a variety of forcers. These emulators can predict 39 annual mean global distributions of temperature, diurnal temperature range and precipitation (including extreme 40 precipitation) given a wide range of emissions and concentrations of carbon dioxide, methane and aerosols, allowing them to 41 efficiently probe previously unexplored scenarios. We discuss the accuracy and interpretability of these emulators and 42 consider their robustness to physical constraints such as total energy conservation. Future opportunities incorporating such 43 physical constraints directly in the machine learning models and using the emulators for detection and attribution studies are 44 also discussed. This opens a wide range of opportunities to improve prediction, robustness and mathematical tractability. We 45 hope that by laying out the principles of climate model emulation with clear examples and metrics we encourage engagement 46 from statisticians and machine learning specialists keen to tackle this important and demanding challenge.

Full Text: https://www.essoar.org/doi/10.1002/essoar.10509765.2

How to Cite: Watson-Parris, D., Rao, Y., Olivié, D., Seland, Ø., Nowack, P. J., Camps-Valls, G., Stier, P., Bouabid, S., Dewey, M., Fons, E., & et al. (2021). ClimateBench: A benchmark dataset for data-driven climate projections. Earth and Space Science Open Archive, 33. https://doi.org/10.1002/essoar.10509765.2