Centuries of thermal sea-level rise due to anthropogenic emissions of short-lived greenhouse gases

Abstract

Mitigation of anthropogenic greenhouse gases with short lifetimes (order of a year to decades) can contribute to limiting warming, but less attention has been paid to their impacts on longer-term sea-level rise. We show that short-lived greenhouse gases contribute to sea-level rise through thermal expansion (TSLR) over much longer time scales than their atmospheric lifetimes. For example, at least half of the TSLR due to increases in methane is expected to remain present for more than 200 y, even if anthropogenic emissions cease altogether, despite the 10-y atmospheric lifetime of this gas. Chlorofluorocarbons and hydrochlorofluorocarbons have already been phased out under the Montreal Protocol due to concerns about ozone depletion and provide an illustration of how emission reductions avoid multiple centuries of future TSLR. We examine the "world avoided" by the Montreal Protocol by showing that if these gases had instead been eliminated in 2050, additional TSLR of up to about 14 cm would be expected in the 21st century, with continuing contributions lasting more than 500 y. Emissions of the hydrofluorocarbon substitutes in the next half-century would also contribute to centuries of future TSLR. Consideration of the time scales of reversibility of TSLR due to short-lived substances provides insights into physical processes: sea-level rise is often assumed to follow air temperature, but this assumption holds only for TSLR when temperatures are increasing. We present a more complete formulation that is accurate even when atmospheric temperatures are stable or decreasing due to reductions in short-lived gases or net radiative forcing.

Keywords: Montreal Protocol; climate change; greenhouse gases; reversibility; sea-level rise.

Fig. 1.

Climate response computed with the UVic ESCM for a scenario with emissions of CO₂, CH₄, N₂O, and HCs (including CFCs, HFCs, HCFCs, and perfluorocarbons) following RCP8.5 to year 2050 and zero anthropogenic emissions thereafter. (A) Atmospheric CO₂ concentration. (B) Total RF. (C) SAT anomaly relative to year 1800. (D) Ocean thermal expansion relative to year 1800. GHGs are changed sequentially in the model simulations to isolate the contributions of each gas.

Fig. S1.

Linearity of SAT (A) and thermal sea-level-rise (B) responses associated with CH₄ emissions. CH₄-induced responses are diagnosed from simulations with CH₄ RF applied in isolation or from differences between simulations with different combinations of GHGs. Results are shown for a scenario with emissions following RCP8.5 to year 2050, with zero anthropogenic emissions thereafter (ZE2050). TSLR is slightly larger and takes slightly longer to reverse if CH₄ is emitted in isolation, than if it is emitted simultaneously with CO₂, which has a much larger RF (Fig. S2). This difference in TSLR is due to differences in ocean heat uptake at low and high RF: when only CH₄ is emitted, the ocean is less stratified and more heat is taken up, resulting in a slightly lower SAT and slightly larger thermal expansion.

Fig. 2.

Climate response for scenarios with high (Velders-high), medium (RCP8.5), and low (Phasedown) HC (includes HFC, HCFC, CFC, and perfluorocarbon) RF to 2050 and exponentially declining RF thereafter. Results are also shown for two world-avoided scenarios with RF of ozone-depleting substances increasing at 4% per year (the increase rate before implementation of the Montreal Protocol) to 2015 and 2050 and zero emissions thereafter. The response to HC forcing is calculated as the difference between CO₂ + N₂O + CH₄ + HC and CO₂ + N₂O + CH₄ simulations. (A) HC RF. (B) SAT difference. (C) Ocean thermal expansion difference.

Fig. 3.

Climate response computed with the UVic ESCM for scenarios with CH₄ emissions following RCP8.5 to year 2050, 2100, 2150, and zero anthropogenic emissions (ZE) thereafter. Variables are calculated as differences between CO₂ + CH₄ and CO₂-only simulations and are aligned at the time emissions are set to zero (which results in a shift by 50 and 100 y in the ZE2100 and ZE2150 scenarios, respectively). (A) CH₄ RF. (B) SAT anomaly relative to year 1800. (C) Ocean thermal expansion relative to year 1800.

Fig. S2.

Climate response computed with the UVic ESCM for a scenario with emissions of CO₂ and CH₄, following RCP8.5 to year 2050, 2100, 2150, and with zero anthropogenic emissions (ZE) thereafter. (A) Atmospheric CO₂ concentration. (B) SAT anomaly relative to year 1800. (C) Ocean thermal expansion relative to year 1800. (D) Total RF. (E) Rate of SAT change. (F) Rate of ocean thermal expansion. Solid lines refer CO₂-only simulations; dashed lines refer to CO₂ + CH₄ simulations.

Fig. S3.

Comparison of SAT (Top) and ocean-temperature (Bottom) response for the UVic ESCM (dashed lines) and the HadGEM2 ESM (solid lines). Results are shown for idealized scenarios with a 1% per year increase in atmospheric CO₂ (“ramp up”), followed by a 1% per year CO₂ decrease (“ramp down”) from 2× CO₂, 3× CO₂, and 4× CO₂ (1, 2). For better comparison of the response time scales involved, temperature anomalies are normalized to the peak value in the respective model.

Fig. 4.

Energy balance terms and rate of ocean thermal expansion for scenarios with CO₂ (A and B) and CH₄ (C and D) emissions following RCP 8.5 to year 2100 and zero anthropogenic emissions thereafter. The simple energy balance model terms are shown by the blue curves in B and D, which are scaled for comparison with the full model calculation (red curves). The climate response to CH₄ forcing is calculated as the difference between CO₂ + CH₄ and CO₂-only simulations. In these calculations, the climate-feedback parameter $\lambda =\beta /\alpha$ is set to 1.0 W/m²/K, the value diagnosed in the standard configuration of the UVic ESCM.