The Peculiar Trajectory of Global Warming

Abstract

The large range in equilibrium climate sensitivity (the global average temperature change following a doubling of atmospheric CO 2) of 1.5°-4.5° reported in IPCC AR5 (IPCC, 2013) persists and is even wider (but see also assessment by Sherwood et al. (2020)) in the most recent Coupled Model Intercomparison Project, Phase 6 (CMIP6) simulations (Zelinka et al., 2020). This introduces uncertainty regarding the urgency of introducing measures to curb CO 2 emissions to limit global warming to specified targets, such as a global temperature increase relative to preindustrial times of no more than +1.5°C or +2°C (IPCC, 2018). The period since the late 1970s is arguably the best suited to provide observational constraints on the sensitivity of present climate due to both the availability of truly global observations provided by satellites, and to less uncertainty in atmospheric aerosol than in earlier decades (Anderson et al., 2003; Jimenez-de-la Cuesta & Mauritsen, 2019; Kiehl, 2007). However, it has been recognized that the cloud radiative feedback may have evolved unexpectedly, providing a strong negative feedback in particular over this best-observed period (

On the Effect of Historical SST Patterns on Radiative Feedback

“…These are merely some of the proposed hypotheses, and not meant to be an exhaustive list. But whatever the reason, the fact that AOGCMs rarely simulate this pattern (e.g., Dong et al., 2021; Fueglistaler & Silvers, 2021; Watanabe et al., 2021) is a concern, suggesting either that their unforced decadal variability is deficient, or that their forced response is biased, and in either case there is a serious systematic error which affects all AOGCMs. Moreover, each of the above interpretations imply different futures, and therefore untangling them is critical for informing both near‐term and long‐term climate projections.…”

Errors in Simple Climate Model Emulations of Past and Future Global Temperature Change

“…Climate model simulations show that climate feedbacks weaken through time in response to step‐forcings and changes in feedbacks are associated with changes in SST patterns (e.g., Dong et al., 2020; Dunne et al., 2020). Incorporating time‐varying feedbacks in EBM2, however, requires further research to distinguish forced changes in feedbacks from unforced climate noise and to explicitly link global feedback changes to variations in SST patterns (e.g., using SST anomalies for regions of tropical deep convection (Fueglistaler & Silvers, 2021).…”

“…Second, EBM2 could be developed to incorporate variations in climate feedbacks and the evolution of AOGCM pattern effects. Late twentieth‐century warming has been suppressed by changes in the observed sea surface temperature (SST) patterns and associated cloud feedbacks (Andrews et al., 2018; Dong et al., 2021; Fueglistaler & Silvers, 2021). Future warming could be affected by changes in the pattern effect (Zhou et al., 2021).…”

The El Niño–Southern Oscillation Pattern Effect

“…These biases could result in incorrect cloud-radiative feedbacks on coupled modes of variability, leading to errors in the timing or amplitude of these modes. Future research should assess the implications of cloud biases for coupled climate variability, and particularly for the known discrepancies between models and observations in Pacific SST variability over recent decades (Fueglistaler & Silvers, 2021).…”