An overview is presented of the GLENS project, a community-wide effort enabling analyses of global and regional changes from stratospheric aerosol geoengineering in the presence of internal climate variability. CESM1(WACCM) STRATOSPHERIC AEROSOL GEOENGINEERING LARGE ENSEMBLE PROJECTSimone TilmeS, Jadwiga H. RicHTeR, Ben KRaviTz, douglaS g. macmaRTin, micHael J. millS, iSla R. SimpSon, anne S. glanville, JoHn T. FaSullo, adam S. pHillipS, Jean-FRancoiS lamaRque, JoSepH TRiBBia, Jim edwaRdS, SHeRi micKelSon, and SiddHaRTHa gHoSH S olar geoengineering using stratospheric sulfate aerosols has been discussed as a potential means of deliberately offsetting some of the effects of climate change (Crutzen 2006). Various model studies have demonstrated that reducing incoming solar radiation globally can offset the increase in global average surface temperature associated with increasing greenhouse gases (e.g., Kravitz et al. 2013). Despite the stabilization of global surface temperature, these simulations show significant changes in atmospheric conditions with global solar reductions or stratospheric sulfur or aerosol injections. Side effects in these simulations include "overcooling" of the tropics and "undercooling" of the poles, leading to continued Arctic summer sea ice loss (e.g., Moore et al. 2014;Tilmes et al. 2016). Additionally, the slowing of the hydrological cycle (e.g., Schmidt et al. 2012) and the potentially uneven cooling between the two hemispheres resulting from solar geoengineering can lead to shifts in precipitation patterns (Haywood et al. 2013; Jones et al. 2017) and reductions in monsoon precipitation (Tilmes et al. 2013). Many available model results to date are based on an artificial design intended to explore the impact of large forcing effects through global solar dimming. For other experiments, only a few ensemble members are performed, making it difficult to identify the robustness of regional climate effects.Simulations of stratospheric sulfate aerosol geoengineering inject sulfur dioxide (SO 2 ) into the stratosphere that oxidizes to form sulfate aerosols or they use direct injections of sulfate aerosols. These experiments require model capabilities beyond those in solar reduction simulations. The stratospheric aerosol distribution resulting from such injections depends on the model's aerosol microphysical scheme, as well as interactions with chemical, dynamical, and radiative processes (Pitari et al. 2014;Mills et al. 2017). Aerosol size and sedimentation are increased with the injection amount and the efficiency of the sulfates to affect the top of the atmosphere radiative imbalance is reduced (Niemeier and Timmreck 2015;Kleinschmitt et al. 2017). The warming of the tropical stratosphere in response to the enhanced aerosol burden results in circulation changes in the stratosphere with potential effects 2361NOVEMBER 2018 AMERICAN METEOROLOGICAL SOCIETY | on the quasi-biennial oscillation (QBO; Aquila et al. 2014), as well as impacts on the tropospheric circulation (Richter et al. 2018). Chan...

We describe the first simulations of stratospheric sulfate aerosol geoengineering using multiple injection locations to meet multiple simultaneous surface temperature objectives. Simulations were performed using CESM1(WACCM), a coupled atmosphere‐ocean general circulation model with fully interactive stratospheric chemistry, dynamics (including an internally generated quasi‐biennial oscillation), and a sophisticated treatment of sulfate aerosol formation, microphysical growth, and deposition. The objectives are defined as maintaining three temperature features at their 2020 levels against a background of the RCP8.5 scenario over the period 2020–2099. These objectives are met using a feedback mechanism in which the rate of sulfur dioxide injection at each of the four locations is adjusted independently every year of simulation. Even in the presence of uncertainties, nonlinearities, and variability, the objectives are met, predominantly by SO2 injection at 30°N and 30°S. By the last year of simulation, the feedback algorithm calls for a total injection rate of 51 Tg SO2 per year. The injections are not in the tropics, which results in a greater degree of linearity of the surface climate response with injection amount than has been found in many previous studies using injection at the equator. Because the objectives are defined in terms of annual mean temperature, the required geongineering results in “overcooling” during summer and “undercooling” during winter. The hydrological cycle is also suppressed as compared to the reference values corresponding to the year 2020. The demonstration we describe in this study is an important step toward understanding what geoengineering can do and what it cannot do.

We present new insights into the evolution and interactions of stratospheric aerosol using an updated version of the Whole Atmosphere Community Climate Model (WACCM). Improved horizontal resolution, dynamics, and chemistry now produce an internally generated quasi‐biennial oscillation and significant improvements to stratospheric temperatures and ozone compared to observations. We present a validation of WACCM column ozone and climate calculations against observations. The prognostic treatment of stratospheric sulfate aerosols accurately represents the evolution of stratospheric aerosol optical depth and perturbations to solar and longwave radiation following the June 1991 eruption of Mount Pinatubo. We confirm the inclusion of interactive OH chemistry as an important factor in the formation and initial distribution of aerosol following large inputs of sulfur dioxide (SO2) to the stratosphere. We calculate that depletion of OH levels within the dense SO2 cloud in the first weeks following the Pinatubo eruption significantly prolonged the average initial e‐folding decay time for SO2 oxidation to 47 days. Previous observational and model studies showing a 30 day decay time have not accounted for the large (30–55%) losses of SO2 on ash and ice within 7–9 days posteruption and have not correctly accounted for OH depletion. We examine the variability of aerosol evolution in free‐running climate simulations due to meteorology, with comparison to simulations nudged with specified dynamics. We assess calculated impacts of volcanic aerosols on ozone loss with comparisons to observations. The completeness of the chemistry, dynamics, and aerosol microphysics in WACCM qualify it for studies of stratospheric sulfate aerosol geoengineering.

Injection of SO2 into the stratosphere has been proposed as a method to, in part, counteract anthropogenic climate change. So far, most studies investigated injections at the equator or in a region in the tropics. Here we use Community Earth System Model version 1 Whole Atmosphere Community Climate Model (CESM1(WACCM)) to explore the impact of continuous single grid point SO2 injections at seven different latitudes and two altitudes in the stratosphere on aerosol distribution and climate. For each of the 14 locations, 3 different constant SO2 emission rates were tested to identify linearity in aerosol burden, aerosol optical depth, and climate effects. We found that injections at 15°N and 15°S and at 25 km altitude have equal or greater effect on radiation and surface temperature than injections at the equator. Nonequatorial injections transport SO2 and sulfate aerosols more efficiently into middle and high latitudes and result in particles of smaller effective radius and larger aerosol burden in middle and high latitudes. Injections at 15°S produce the largest increase in global average aerosol optical depth and increase the change in radiative forcing per Tg SO2/yr by about 15% compared to equatorial injections. High‐altitude injections at 15°N produce the largest reduction in global average temperature of 0.2° per Tg S/yr for the last 7 years of a 10 year experiment. Injections at higher altitude are generally more efficient at reducing surface temperature, with the exception of large equatorial injections of at least 12 Tg SO2/yr. These findings have important implications for designing a strategy to counteract global climate change.

Geoengineering methods could potentially offset aspects of greenhouse gas‐driven climate change. However, before embarking on any such strategy, a comprehensive understanding of its impacts must be obtained. Here, a 20‐member ensemble of simulations with the Community Earth System Model with the Whole Atmosphere Community Climate Model as its atmospheric component is used to investigate the projected hydroclimate changes that occur when greenhouse gas‐driven warming, under a high emissions scenario, is offset with stratospheric aerosol geoengineering. Notable features of the late 21st century hydroclimate response, relative to present day, include a reduction in precipitation in the Indian summer monsoon, over much of Africa, Amazonia and southern Chile and a wintertime precipitation reduction over the Mediterranean. Over most of these regions, the soil desiccation that occurs with global warming is, however, largely offset by the geoengineering. A notable exception is India, where soil desiccation and an approximate doubling of the likelihood of monsoon failures occurs. The role of stratospheric heating in the simulated hydroclimate change is determined through additional experiments where the aerosol‐induced stratospheric heating is imposed as a temperature tendency, within the same model, under present day conditions. Stratospheric heating is found to play a key role in many aspects of projected hydroclimate change, resulting in a general wet‐get‐drier, dry‐get‐wetter pattern in the tropics and extratropical precipitation changes through midlatitude circulation shifts. While a rather extreme geoengineering scenario has been considered, many, but not all, of the precipitation features scale linearly with the offset global warming.

Abstract. Understanding the climate impacts of solar geoengineering is essential for evaluating its benefits and risks. Most previous simulations have prescribed a particular strategy and evaluated its modeled effects. Here we turn this approach around by first choosing example climate objectives and then designing a strategy to meet those objectives in climate models.There are four essential criteria for designing a strategy: (i) an explicit specification of the objectives, (ii) defining what climate forcing agents to modify so the objectives are met, (iii) a method for managing uncertainties, and (iv) independent verification of the strategy in an evaluation model.We demonstrate this design perspective through two multi-objective examples. First, changes in Arctic temperature and the position of tropical precipitation due to CO 2 increases are offset by adjusting high-latitude insolation in each hemisphere independently. Second, three different latitude-dependent patterns of insolation are modified to offset CO 2 -induced changes in global mean temperature, interhemispheric temperature asymmetry, and the Equator-to-pole temperature gradient. In both examples, the "design" and "evaluation" models are state-of-the-art fully coupled atmosphere-ocean general circulation models.

By injecting different amounts of SO2 at multiple different latitudes, the spatial pattern of aerosol optical depth (AOD) can be partially controlled. This leads to the ability to influence the climate response to geoengineering with stratospheric aerosols, providing the potential for design. We use simulations from the fully coupled whole‐atmosphere chemistry climate model CESM1(WACCM) to demonstrate that by appropriately combining injection at just four different locations, 30°S, 15°S, 15°N, and 30°N, then three spatial degrees of freedom of AOD can be achieved: an approximately spatially uniform AOD distribution, the relative difference in AOD between Northern and Southern Hemispheres, and the relative AOD in high versus low latitudes. For forcing levels that yield 1–2°C cooling, the AOD and surface temperature response are sufficiently linear in this model so that the response to different combinations of injection at different latitudes can be estimated from single‐latitude injection simulations; nonlinearities associated with both aerosol growth and changes to stratospheric circulation will be increasingly important at higher forcing levels. Optimized injection at multiple locations is predicted to improve compensation of CO2‐forced climate change relative to a case using only equatorial aerosol injection (which overcools the tropics relative to high latitudes). The additional degrees of freedom can be used, for example, to balance the interhemispheric temperature gradient and the equator to pole temperature gradient in addition to the global mean temperature. Further research is needed to better quantify the impacts of these strategies on changes to long‐term temperature, precipitation, and other climate parameters.

Injections of sulfur dioxide into the stratosphere are among several proposed methods of solar radiation management. Such injections could cool the Earth's climate. However, they would significantly alter the dynamics of the stratosphere. We explore here the stratospheric dynamical response to sulfur dioxide injections ∼5 km above the tropopause at multiple latitudes (equator, 15°S, 15°N, 30°S and 30°N) using a fully coupled Earth system model, Community Earth System Model, version 1, with the Whole Atmosphere Community Climate Model as its atmospheric component (CESM1(WACCM)). We find that in all simulations, the tropical lower stratosphere warms primarily between 30°S and 30°N, regardless of injection latitude. The quasi‐biennial oscillation (QBO) of the tropical zonal wind is altered by the various sulfur dioxide injections. In a simulation with a 12 Tg yr−1 equatorial injection, and with fully interactive chemistry, the QBO period lengthens to ∼3.5 years but never completely disappears. However, in a simulation with specified (or noninteractive) chemical fields, including O3 and prescribed aerosols taken from the interactive simulation, the oscillation is virtually lost. In addition, we find that geoengineering does not always lengthen the QBO. We further demonstrate that the QBO period changes from 24 to 12–17 months in simulations with sulfur dioxide injections placed poleward of the equator. Our study points to the importance of understanding and verifying of the complex interactions between aerosols, atmospheric dynamics, and atmospheric chemistry as well as understanding the effects of sulfur dioxide injections placed away from the Equator on the QBO.