Carbon dioxide and climate impulse response functions for the computation of greenhouse gas metrics: a multi-model analysis

Abstract

The responses of carbon dioxide (CO<sub>2</sub>) and other climate variables to an emission pulse of CO<sub>2</sub> into the atmosphere are often used to compute the Global Warming Potential (GWP) and Global Temperature change Potential (GTP), to characterize the response time scales of Earth System models, and to build reduced-form models. In this carbon cycle-climate model intercomparison project, which spans the full model hierarchy, we quantify responses to emission pulses of different magnitudes injected under different conditions. The CO<sub>2</sub> response shows the known rapid decline in the first few decades followed by a millennium-scale tail. For a 100 Gt C emission pulse, 24 ± 10% is still found in the atmosphere after 1000 yr; the ocean has absorbed 60 ± 18% and the land the remainder. The response in global mean surface air temperature is an increase by 0.19 ± 0.10 °C within the first twenty years; thereafter and until year 1000, temperature decreases only slightly, whereas ocean heat content and sea level continue to rise. Our best estimate for the Absolute Global Warming Potential, given by the time-integrated response in CO<sub>2</sub> at year 100 times its radiative efficiency, is 92.7 × 10<sup>−15</sup> yr W m<sup>−2</sup> per kg CO<sub>2</sub>. This value very likely (5 to 95% confidence) lies within the range of (70 to 115) × 10<sup>−15</sup> yr W m<sup>−2</sup> per kg CO<sub>2</sub>. Estimates for time-integrated response in CO<sub>2</sub> published in the IPCC First, Second, and Fourth Assessment and our multi-model best estimate all agree within 15%. The integrated CO<sub>2</sub> response is lower for pre-industrial conditions, compared to present day, and lower for smaller pulses than larger pulses. In contrast, the response in temperature, sea level and ocean heat content is less sensitive to these choices. Although, choices in pulse size, background concentration, and model lead to uncertainties, the most important and subjective choice to determine AGWP of CO<sub>2</sub> and GWP is the time horizon

The exponential eigenmodes of the carbon-climate system, and their implications for ratios of responses to forcings

“…There are nonlinearities in the coupling between these linear model components, especially through the dependence of CO 2 radiative forcing and the ocean-air and land-air carbon exchanges on CO 2 concentration. These are incorporated in many models of the coupled carbon-climate system through weakly nonlinear coupling between otherwise linear model components (Hasselmann et al, 1997;Petschel-Held et al, 1999;Hooss et al, 2001;Joos et al, 2001Joos et al, , 2012Raper et al, 2001; also the nonlinear model used here). Fully linear versions of such models can always be developed within a limited subspace around any given state of the Earth system, because a weakly nonlinear carbon-climate model can be linearised about that state.…”

Inhomogeneous Radiative Forcing of NF3

“…Here, a lifetime of 740 years is used in Robson et al [2], a lifetime of 500 years is used in IPCC AR5 and this work, and a lifetime of 509 years is used in Totterdill et al [10] Robson et al [2] used the fraction remaining formula of CO 2 included in Shine et al [36], and others have used the formula provided by Joos et al [37]. Similar to previous discussion in this paper regarding RE and IRE, the differences among Robson et al [2], IPCC AR5, and this work are small.…”

“…Additionally, there is a negative correlation coefficient of −0.88 between F c and integrated water content, taking away the effect of surface temperature. Huang et al [37] used a regression model to study the inhomogeneous radiative forcing of CO 2 . We use a similar method to study the inhomogeneous radiative forcing of NF 3 in clear sky conditions.…”

A critical view on scientific consensus building in life cycle impact assessment

“…This procedure, however, may give the false impression that harmonization reduces uncertainty. As an alternative strategy, Joos et al (2013) performed a multi-model analysis to provide carbon dioxide impulse response functions for the computation of global warming potentials (GWPs). Instead of selecting one single best model, they provided a best estimate on the basis of a multi-model mean of 15 climate models and the confidence range of the mean.…”