Reanalysis-based Global Radiative Response to Sea Surface Temperature Patterns: Evaluating the Ai2 Climate Emulator

TL;DR

This paper uses Green's function simulations with the Ai2 Climate Emulator to estimate the radiative response to sea surface temperature changes, evaluating the emulator's physical accuracy and limitations.

Contribution

It introduces a Green's function approach to assess machine learning climate emulators' physical fidelity and compares emulator responses with traditional models.

Findings

ACE2-ERA5 underestimates historical radiative response

Green's function experiments evaluate emulator performance

Emulators can be tested for out-of-distribution predictions

Abstract

The sensitivity of the radiative flux at the top of the atmosphere to surface temperature perturbations cannot be directly observed. The relationship between sea surface temperature (SST) and top-of-atmosphere radiation can be estimated with Green's function simulations by locally perturbing the sea surface temperature boundary conditions in atmospheric climate models. We perform such simulations with the Ai2 Climate Emulator (ACE), a machine learning-based emulator trained on ERA5 reanalysis data (ACE2-ERA5). This produces a sensitivity map of the top-of-atmosphere radiative response to surface warming that aligns with our physical understanding of radiative feedbacks. However, ACE2-ERA5 likely underestimates the radiative response to historical warming. We compare to two additional versions of ACE and traditional climate models. We argue that Green's function experiments can be used to evaluate the performance and limitations of machine learning-based climate emulators by examining if causal physical relationships are correctly represented and testing their capability for out-of-distribution predictions.