Transferring climate change physical knowledge

TL;DR

This paper demonstrates that transfer learning with machine learning can significantly reduce uncertainties in climate projections by effectively combining Earth system model data and observations, leading to more reliable regional temperature forecasts.

Contribution

The study introduces a novel transfer learning approach that merges climate model simulations and observations to reduce projection uncertainties and improve regional climate predictions.

Findings

Uncertainty in climate projections reduced by over 50%.

Improved regional temperature pattern accuracy.

Narrower projection uncertainty compared to existing methods.

Abstract

Precise and reliable climate projections are required for climate adaptation and mitigation, but Earth system models still exhibit great uncertainties. Several approaches have been developed to reduce the spread of climate projections and feedbacks, yet those methods cannot capture the nonlinear complexity inherent in the climate system. Using a Transfer Learning approach, we show that Machine Learning can be used to optimally leverage and merge the knowledge gained from global temperature maps simulated by Earth system models and observed in the historical period to reduce the spread of global surface air temperature fields projected in the 21st century. We reach an uncertainty reduction of more than 50% with respect to state-of-the-art approaches while giving evidence that our method provides improved regional temperature patterns together with narrower projections uncertainty, urgently required for climate adaptation.