The interaction between feedback controlled SAI forcing and internal (unforced) modes of climate variability is one possible mechanism here, but changes could also be induced by other mechanisms, such as stratospheric heating or residual changes in meridional or zonal temperature gradients under SAI. ENSO is picked for the metric as a representative example, but the uncertainty includes other modes such as the North Atlantic Oscillation (NAO), Pacific Decadal Oscillation (PDO) and Southern Annular Mode (SAM).
Metric
SAI detectably changes the mean-state, period or amplitude of ENSO on a 50-year timescale at 0.5°C cooling, relative to a reference world with the same global mean temperature but neither higher CO2 nor SAI.
Uncertainty
There has been little research addressing this uncertainty. Feedback control responds to unforced modes of variability (MacMartin et al 2014, Connolly et al., 2024), potentially increasing variability (Diao et al., 2023) if that is not accounted for. Rezaei et al. (2023) find that SAI mostly offsets changes in various modes of variability predicted under warming, with some exceptions. Given studies thus far have not found large changes, under scenarios with significantly greater signal than our 0.5C of cooling, we estimate the likelihood of our metric as low.
Decision relevance
ENSO and other large-scale modes of variability are important drivers of regional climate impacts, such as extreme weather. This uncertainty represents one major pathway by which regional residual changes under SAI might be mediated.