The overall radiative forcing from an SO₂ injection — the net change in how much energy reaches Earth's surface — depends on both the aerosol size distribution and how long the aerosols remain in the stratosphere. Lifetime in turn depends on both size (larger particles fall out of the stratosphere faster) and transport. The spatial distribution of aerosols across the stratosphere also shapes the surface climate response, and that distribution likewise depends on stratospheric transport. Validating model predictions of aerosol growth, transport, and lifetime is thus essential to fully understand radiative forcing and surface climate response.
The aerosol microphysics experiment will test physical understanding of aerosol microphysics on timescales of days to weeks. But even at that point, much of the injected SO₂ will still be unreacted gas, so new particle formation may continue afterward. Condensation and coagulation will certainly continue to reshape the size distribution, the aerosols will spread across an entire hemisphere, and as they grow larger, they will begin to fall out. These months-scale processes ultimately determine the net radiative forcing from an SO₂ injection, and the aerosol microphysics experiment in Phase 1 cannot probe them.
A larger-scale experiment is therefore necessary to test how well models match the evolution of aerosol size and spatial distribution over months to a year — the interplay between aerosol microphysics and hemispheric stratospheric transport. To remain detectable over this duration, the amount released must be large enough to stand out against background sulfate aerosol concentrations even after dilution. Current estimates suggest a total injection of 25,000 tonnes of SO₂ in one hemisphere over the course of one season, with daily flights on the order of 250 tonnes of SO₂ per day.
A release at this scale would not be enough to affect climate, but would be measurable from satellite and in situ observations for many months. It remains small compared to not-infrequent volcanic eruptions: one of the most recent sulfur-rich eruptions, the 2024 Ruang eruption, released roughly 300,000 tonnes in a single event — about an order of magnitude larger than what would be needed here. And, while visible by satellite instruments, this had no measurable climatic effects. This experiment would show how aerosols grow and spread over their full stratospheric lifetime, and the local radiative response and its evolution could be detected directly. Repeating individual plume sampling as in the aerosol microphysics experiment would additionally reduce uncertainty in how new particle formation changes as background aerosol concentrations rise.
Because the capabilities required for an aerosol evolution and transport experiment — aircraft, storage and dispersal systems, ground infrastructure — approach those required for deployment itself, a decision to proceed would likely indicate that deployment is under serious consideration, contingent on results. Proceeding is itself contingent on Phase 1: if the details of microphysics and plume dynamics show that models have overestimated the effectiveness of SAI with sulfates, an aerosol evolution and transport experiment may not go forward. Many research activities on impacts and risks must therefore be completed before an aerosol evolution and transport experiment, and repeated after it; this sequencing shapes where those activities sit on the roadmap.
