Convergence of cumulative SO₂ injections beyond 2070

Determine whether the cumulative sulfur dioxide injections required by the ARISE-SAI feedback-controlled stratospheric aerosol injection strategy under the SSP2-4.5 emissions scenario converge after 2070 between the two simulated deployments with identical 1.37°C above preindustrial temperature targets but different start dates (2035 versus 2045) in CESM2-WACCM6.

Background

The study compares two CESM2-WACCM6 ensembles of stratospheric aerosol injection (SAI) under SSP2-4.5 that target the same global mean temperature (~1.37°C above preindustrial) but begin in 2035 versus 2045 using the ARISE feedback strategy with four-latitude SO₂ injections controlling T0, T1, and T2.

While the delayed-start (2045) scenario requires about 20% higher injection rates and produces about 30% lower net radiative forcing to maintain the same surface temperature, the cumulative SO₂ injections for the two scenarios are similar at the end of the simulation horizon (~2070). Because aerosol microphysical responses are nonlinear and the simulations end in 2070, the authors explicitly note uncertainty about whether the cumulative injections would converge beyond this date, which bears on long-term program costs and impacts.

References

Interestingly, cumulative injections for the two scenarios are similar at the end of the simulations around 2070 (Figure 2B), but given the known nonlinearities and limited simulation duration, it is unclear whether they would converge past 2070. This has implications for the total long-term cost of an SAI program.

— Kicking the Can Down the Road: Understanding the Effects of Delaying the Deployment of Stratospheric Aerosol Injection  (2402.11992 - Brody et al., 2024) in Section III, Results and Discussion, discussion of Figure 2; p. 6