SAI Research Roadmap

PHASEFoundational knowledgeWhat can existing evidence tell us, and where does it fall short?Time2–3 yrsCost~$30M–$75M

PHASEAerosol microphysicsDoes data from small outdoor experiments validate the processes in our models?Time4–8 yrsCost~$70M–$150MEst. uncertainty reduction (cooling efficacy)~25%

PHASEStratospheric distributionDo sulfate aerosols mix and distribute, globally, in the way our models predict?Time4–11 yrsCost~$270M–$1.14BEst. uncertainty reduction (cooling efficacy)~66%

PHASEMonitored coolingAs cooling begins, do observed effects match predictions? Are any unexpected effects emerging?TimeOngoingCostTBD

ImpactsClimate response, impacts, and risks

AerosolsMicrophysics, chemistry, transport

ObservationsMonitoring systems, instruments, and data

EngineeringAircraft R&D, materials handling, operations

Aerosols

CLOUD Chamber studies

CLOUD chamber studies of H₂SO₄ nucleation at stratospheric conditions

Est. direct cost: $500K–$900K
Est. duration

Why it’s on the roadmap

Sulfate particles form when sulfuric acid (H₂SO₄) molecules cluster together and grow large enough to become stable, a process called nucleation. Extensive chamber experiments, most notably the Cosmics Leaving Outdoor Droplets (CLOUD) campaigns (Dias et al. 2017), have mapped how nucleation behaves under typical tropospheric conditions (Dunne et al., 2016).

Far fewer experiments have reproduced stratospheric conditions, where two differences likely matter. Water vapor is one of the two components of nucleation, and the stratosphere is extremely dry. Cosmic rays also produce ions at these altitudes, and those ions can stabilize newly formed clusters.

Models reflect this gap. Compared with the limited chamber studies run at stratospheric conditions, commonly used nucleation schemes overestimate new particle formation by 3–4 orders of magnitude (Yu et al., 2023). The same schemes overestimate background particle concentrations relative to in situ data by a smaller factor of 2–4 (Yu et al., 2023).

Chamber experiments on H₂SO₄ nucleation at low humidity, low temperature, and low pressure, run both with and without ions, will constrain these rates directly. Nucleation is one of several uncertainties in aerosol microphysics, but it sets the number of particles formed per unit of injected mass, and therefore the size distribution that governs how efficiently an SAI deployment scatters sunlight.

Tracking condensation and coagulation over longer timescales remains challenging in a chamber. Particle and gas losses to the chamber walls compete with the processes of interest, and they add uncertainty to both the measurements and the analysis that follows. Characterizing wall loss as a function of particle size and gas-phase species will reduce that uncertainty: correcting measured size distributions for the loss terms recovers coagulation and growth rates that can be compared against microphysics schemes.

Scope of work

This research activity funds a chamber campaign that reproduces stratospheric conditions and delivers a nucleation parameterization suitable for global aerosol models. The design follows the progression established by the CLOUD program, which moved from laboratory rate measurements through a fitted parameterization to implementation and evaluation in a global model (Dunne et al., 2016). Three phases will run in sequence, with parameterization work beginning before the measurement campaign concludes.

Phase 1: Chamber measurements — will measure binary H₂SO₄-H₂O nucleation rates across the stratospheric parameter space. Each condition will run both with and without ionization, at ion pair production rates matching galactic cosmic ray fluxes between 15 and 25 km. Facility selection forms part of the initial scope, since reaching stratospheric temperature and pressure simultaneously exceeds the operating envelope of most existing chambers.
1. Parameter space: temperatures of 190–230 K, pressures of 30–150 hPa, relative humidity below 5%, and H₂SO₄ concentrations spanning 10⁵–10⁹ cm⁻³.
2. Instrument suite: will resolve the smallest formed clusters, since the discrepancy with current schemes may originate below the detection limit of conventional counters. Chemical ionization mass spectrometry will measure gas-phase H₂SO₄; an APi-TOF will resolve neutral and charged cluster composition; a particle size magnifier paired with a nano-SMPS will recover size distributions from roughly 1 to 100 nm; and a neutral cluster and air ion spectrometer will track the charged fraction separately, isolating the ion-induced channel from the neutral channel in the fit.
3. Wall loss characterization: will run alongside the nucleation measurements rather than as a calibration afterthought. Size-resolved particle deposition rates will come from decay experiments with monodisperse seed particles, and gas-phase H₂SO₄ loss rates from decay experiments at each temperature and pressure setting. These loss terms enter the rate retrieval directly and set the uncertainty floor on the reported nucleation rates.
Phase 2: Parameterization — will fit the measured rates to a parameterization expressing nucleation rate as a function of H₂SO₄ concentration, temperature, relative humidity, and ion pair production rate, treating neutral and ion-induced channels as separate terms. The functional form will follow Dunne et al., 2016 so that the stratospheric fit can substitute directly into models already carrying the tropospheric version. Reported uncertainties will propagate the wall loss corrections from Phase 1.
Phase 3: Implementation and evaluation — will implement the parameterization in sectional and modal stratospheric aerosol schemes and evaluate the result against in situ size distribution measurements from balloon and aircraft platforms. The evaluation will test whether the revised scheme closes the factor of 2–4 background concentration bias identified by Yu et al. (2023). Rerunning standard SAI injection scenarios with the revised scheme will then quantify how much of the nucleation uncertainty propagates into predicted particle size distributions and forcing efficiency.

Open full page