https://iopscience.iop.org/article/10.1088/2752-5295/ae7c24
Authors: Natalia Okanikova, Alistair Duffey, Jared Farley, Peter Irvine, Harold Heorton and Michel Tsamados
Accepted Manuscript online 11 June 2026
Abstract
Stratospheric aerosol injection (SAI) is a proposed method of reducing global warming by dispersing sunlight-reflecting particles in the stratosphere. SAI in high northern latitudes would face lower logistical barriers than low-latitude SAI and has been suggested as a means of counteracting rapid Arctic warming. A central risk of Arctic-only SAI is the potential impact on tropical precipitation via changes to the Interhemispheric Temperature Difference (IHTD). However, as the Northern hemisphere is currently warming faster than the Southern, some Arctic-only SAI may be consistent with maintaining a constant IHTD. Here, we quantify the Arctic-only (60° N) injection magnitude consistent with stabilising IHTD at 2035 levels across 36 CMIP6 models, using the Climate Intervention Dynamical Emulator (CIDER). Notably, CESM2-WACCM and UKESM1, the two most widely used models in SAI research, sit at opposite extremes of the ensemble, with respectively low and high IHTD futures. In the UKESM1 model, IHTD-stabilising SAI halves Arctic warming between 2035 and 2100. In the median model, IHTD remains approximately constant under injection rates that increase linearly from zero in 2035 to 4.1 Mt SO₂/yr by 2100, at a rate of 0.73 Mt SO₂/decade. Arctic-only SAI with magnitude of several Mt SO2/year would likely reduce rather than exacerbate the IHTD disturbance over the 21st century, but the implications for tropical precipitation remain uncertain, as IHTD proves an imperfect proxy for tropical rainfall latitude.
Source: IOP Science