Abrupt reduction in shipping emission as an inadvertent geoengineering termination shock produces substantial radiative warming

Human activities have significantly warmed the Earth's atmosphere by increasing greenhouse gas concentrations, trapping thermal radiation and creating positive climate forcing. Simultaneously, increased anthropogenic aerosol concentrations have a cooling effect, either directly by reflecting solar radiation or indirectly by interacting with clouds. The magnitude of this aerosol cooling effect is crucial for accurately estimating climate sensitivity to greenhouse gas forcing and predicting future warming. The effectiveness of anthropogenic aerosols also directly relates to solar radiation modification (SRM) geoengineering schemes, which aim to temporarily cool the planet by increasing solar radiation reflection. These schemes are not a solution to greenhouse gas-induced warming and have uncertain consequences beyond the intended cooling. Marine cloud brightening (MCB) is one SRM method, involving seeding marine low clouds with aerosols to increase their reflectivity. Early satellite observations revealed ship tracks—linear features of brighter clouds downwind of ships—as examples of small-scale, inadvertent MCB experiments. Ship emissions add aerosols, increasing cloud droplet numbers and reflectivity. However, aerosols can also alter cloud liquid water path (LWP) and cloud fraction (CF), significantly influencing solar radiation reflection. Global shipping significantly impacts clouds, making shipping emissions a long-running, inadvertent MCB experiment. The International Maritime Organization (IMO) 2020 regulations, which reduced maximum sulfur content in international shipping fuel from 3.5% to 0.5%, represent a sudden termination of this inadvertent geoengineering experiment, potentially leading to global climate impacts by reducing the cloud-brightening effect of ship emissions. This study investigates the impact of this abrupt change.

The literature extensively explores the effects of aerosols on climate, including their cooling influence and the implications for climate sensitivity (Forster et al., 2021; Bellouin et al., 2020). Geoengineering proposals, such as MCB, have been analyzed regarding their potential efficacy and risks (Crutzen, 2006; National Academies of Sciences, Engineering, and Medicine, 2021; Robock et al., 2009; Jones et al., 2009; Bala et al., 2011; Latham et al., 2012; Wood, 2021). Studies have examined the impact of ship emissions on clouds, documenting the formation of ship tracks and their influence on cloud reflectivity (Conover, 1966; Coakley et al., 1987; Twomey, 1977; Albrecht, 1989; Wang et al., 2011; Christensen & Stephens, 2011; Toll et al., 2019). Previous research has quantified the radiative forcing from shipping emissions (Capaldo et al., 1999; Lauer et al., 2007; Peters et al., 2012; Partanen et al., 2013; Gettelman et al., 2023), providing a context for evaluating the impact of the IMO 2020 regulations. The effects of the IMO 2020 regulations on ship tracks and cloud properties have been studied using remote sensing data (Yuan et al., 2022; Diamond, 2023; Diamond et al., 2020), highlighting the potential for unintended consequences of pollution reduction measures.

This study uses a combination of satellite observations and a chemical transport model to quantify the radiative forcing from the IMO 2020 regulations and assess their climate impacts. The NASA GEOS-GOCART model simulated the impact of IMO 2020 on maritime aerosol concentrations. The model's reduction in aerosol optical depth (AOD) due to decreased SO2 emissions from shipping is shown, revealing peak reductions in specific regions. The ratio between the AOD change due to IMO 2020 and the pre-industrial to present-day change was calculated, highlighting regions where ship emissions are a significant component of total aerosols. The model also simulated changes in cloud droplet number concentration (Nd). These Nd changes, along with satellite observations, were used to estimate radiative forcing, considering both the Twomey effect and adjustments in LWP and CF. The LWP and CF adjustments followed functional forms derived from a large dataset of ship tracks, depending on background cloud Nd, SST, inversion strength, and background low cloud fraction. The radiative forcing was calculated using a methodology that accounts for the Twomey effect and the changes in cloud liquid water path and cloud fraction (Yuan et al., 2023). The resulting patterns of annual mean radiative forcing are shown. A simple energy balance model was used to estimate the expected transient temperature increase due to the IMO 2020-induced warming, ignoring deep ocean heat uptake in the short term. The model was used to estimate the expected warming trajectory in the 2020s. Regional variations in radiative forcing were considered. The study compared the IMO 2020-induced radiative forcing with observed changes in planetary heat uptake and top-of-atmosphere (TOA) absorbed solar radiation to assess its impact on the energy balance. The study also assessed the seasonal variations of the forcing and examined the contributions of different factors such as background Na, CF, and ΔNa to seasonal variations in the North Atlantic. Uncertainty analysis is included by comparing the results to a recent observational study in the South Atlantic. Deep learning was used to estimate Nd.

The study found that the IMO 2020 regulations caused a global ocean-averaged radiative forcing of +0.2 ± 0.11 Wm⁻². The Twomey effect contributed 40%, while cloud fraction adjustment contributed 60%, with LWP adjustment being near neutral. Regional variations were significant, with the strongest forcing in the North Atlantic (~1.4 Wm⁻²). This forcing translates to ~0.16 K of warming within 7 years, more than doubling the average warming rate since 1980 (0.24 K/decade versus 0.19 K/decade). The IMO 2020 effect accounts for approximately 80% of the observed increase in planetary heat uptake since 2020. Strong hemispheric contrast in forcing was observed, with higher warming in the northern hemisphere. The study found that the seasonal variations are dominated by the seasonal variations of the solar radiation. The change of cloud droplet concentration contributed the most to the seasonal variations.

The findings demonstrate that the IMO 2020 regulations, intended to improve air quality, have had a significant, albeit unintended, warming effect on the climate. The magnitude of the radiative forcing exceeds the National Academy of Sciences' recommended threshold for outdoor SRM experiments. While the study shows that MCB might be effective in slowing warming, the significant spatial and temporal heterogeneity underscores the challenges in managing such a geoengineering scheme. The hemispheric contrast observed highlights the potential for unintended consequences on regional climates. The study's results emphasize the need for careful consideration of spatiotemporal heterogeneity in future geoengineering efforts. The trade-off between air quality improvements and potential warming from reduced aerosol pollution is an important policy consideration.

The IMO 2020 regulations represent a significant inadvertent geoengineering experiment, resulting in substantial warming through a reverse MCB. The study highlights the complexity of aerosol-cloud interactions and the potential unintended consequences of pollution control measures. Further research is necessary to fully understand the long-term impacts of the IMO 2020 regulations and to develop effective and safe geoengineering strategies. The spatial and temporal heterogeneity observed underscores the need for detailed regional climate modeling to mitigate potential adverse effects. The trade-off between air quality benefits and warming effects needs more attention from policy makers.

The study acknowledges several sources of uncertainty. The estimation of Na change relies on modeling, not direct observations. While LWP and cloud fraction adjustments are robust based on a large dataset, they have inherent limitations. The comparison with the South Atlantic observational study provides a degree of cross-validation but doesn't cover the global scale. Feedback processes are not explicitly considered. The study notes that while the global-scale impact of IMO2020 may take time to be clearly observed, regional changes, such as in the North Atlantic, might be detectable sooner.