Multivariate Estimations of Equilibrium Climate Sensitivity from Short Transient Warming Simulations

TL;DR

This paper introduces a multivariate linear regression approach that improves estimates of equilibrium climate sensitivity by capturing multiple climate feedback modes from short transient warming simulations.

Contribution

The paper presents an extension to existing extrapolation techniques by incorporating multiple observables, enabling better long-term climate sensitivity estimates from short-term data.

Findings

Enhanced accuracy in climate sensitivity estimation.

Captures multiple climate feedback eigenmodes.

Improves extrapolation from transient to equilibrium states.

Abstract

One of the most used metrics to gauge the effects of climate change is the equilibrium climate sensitivity, defined as the long-term (equilibrium) temperature increase resulting from instantaneous doubling of atmospheric CO$_2$. Since global climate models cannot be fully equilibrated in practice, extrapolation techniques are used to estimate the equilibrium state from transient warming simulations. Because of the abundance of climate feedbacks - spanning a wide range of temporal scales - it is hard to extract long-term behaviour from short-time series; predominantly used techniques are only capable of detecting the single most dominant eigenmode, thus hampering their ability to give accurate long-term estimates. Here, we present an extension to those methods by incorporating data from multiple observables in a multi-component linear regression model. This way, not only the dominant but also the next-dominant eigenmodes of the climate system are captured, leading to better long-term estimates from short, non-equilibrated time series.