Fundamental limits and trade-offs of stratospheric aerosol geoengineering

减少二氧化碳和其他温室气体（GHG）的净排放对于任何应对气候变化的措施都至关重要，但减少的速度可能不够快，无法避免重大的气候影响。平流层气溶胶地球工程的模式预估表明，它可以减少一些气候影响，因此可能成为全面气候变化战略的一个附加要素。然而，目前的知识不足以支持明智的决策。评估地球工程的一个关键问题是，地球工程在多大程度上能够很好地管理温室气体增加造成的气候反应，其基本限制或权衡是什么？也就是说，地球工程能做什么，不能做什么？以最近的研究为基础，这个项目将解决这个基本问题。这项研究的基本动机是了解减少未来气候影响的潜在选择。需要更好的信息来支持围绕部署的未来决策，并支持制定这些决策所需的治理能力的开发。这项研究将使我们能够比以前更全面地了解部署地球工程的影响，通过生成模拟，捕获更全面的部署选项集，而不仅仅是一两个；此外，还将评估不同目标能够或不能同时实现的程度。该项目将生成一组气候模型模拟，每个模型对哪些气候目标相对于其他目标优先考虑做出不同的选择，并利用这些选择来确定潜在的权衡（相互排斥的目标集）和边界（哪些目标可以实现，哪些目标不可实现）。在整个过程中，研究小组将聘请政策和治理专家，研究可能激励不同行为者的气候目标的潜在范围，以及确定的权衡对治理的影响。所有可能的策略从未被探索过，部分原因是在可用自由度空间上的优化——主要是气溶胶注入的纬度和季节——由于不确定性和非线性相互作用（来自微物理和气溶胶加热引起的平流层环流变化）而变得复杂，并且由于组合计算的复杂性而变得复杂。为了应对这些挑战，研究团队将结合三项创新。首先，本研究的关键促成因素是对设计空间“大小”的初步评估；有多少个有用的独立自由度？这减少了组合问题。其次，通过将了解平流层气溶胶光学深度（AOD）的空间和季节分布所需的模拟与评估特定气溶胶分布的气候响应所需的模拟分离，可以减少计算负担。AOD可能很短，但需要一个完整的平流层模型，而AOD需要几十年的模拟，但不需要精确的平流层。第三，非线性和不确定性可以通过调整注入速率的反馈来管理；这样就可以根据指定的目标而不是指定的注入速率来比较模拟。研究小组将设计一套模拟，分别满足不同的目标，并共同跨越可能的结果空间。由此，评估和可视化权衡的关键工具是通过帕累托最优曲面：策略及其响应如何作为优化标准的函数而变化。尽管模拟将侧重于理解物理科学的权衡，但社会和治理维度在理解哪些目标可能最重要，哪些策略在政治上不可行方面发挥着关键作用，从而以气候建模未揭示的方式限制了空间。因此，研究团队将始终与治理专家进行交互，以确保研究为政策提供信息。模拟也将提供给更广泛的国际社会。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Reducing net emissions of CO2 and other greenhouse gas (GHG) is essential to any response to climate change, but may not occur fast enough to avoid significant climate impacts. Model projections of stratospheric aerosol geoengineering suggest that it could reduce some climate impacts, and thus might potentially become an additional element of a comprehensive climate change strategy. However, current knowledge is insufficient to support informed decisions. A critical question in evaluating geoengineering is what are the fundamental limits or trade-offs in how well geoengineering can manage the climate response from increased GHG? That is, what can geoengineering do, and what can it not do? Building on recent research, this project will address this essential question. The fundamental motivation for this research is to understand a potential option to reduce future climate impacts. Better information is needed both to support future decisions around deployment, and support the development of governance capacity that will be needed to make these decisions. This research will enable a more complete view of the impacts of deploying geoengineering than has previously been possible, by generating simulations that capture a more comprehensive set of deployment options rather than just one or two; and furthermore will assess the extent to which different objectives can or cannot be simultaneously met.This project will generate a set of climate model simulations that each make different choices for which climate goals to prioritize relative to others, and use this to identify potential tradeoffs (sets of objectives that are mutually exclusive) and boundaries (which objectives are achievable and which are not). Throughout this process, the research team will engage policy and governance experts, regarding the potential range of climate goals that might motivate different actors, and on the governance implications of identified trade-offs. The full range of possible strategies has never been explored, in part because optimization over the space of available degrees of freedom – primarily latitudes and seasons of aerosol injection – is complicated by uncertainty and nonlinear interactions (from both microphysics and aerosol-heating-induced changes in stratospheric circulation), and compounded by combinatorial computational complexity. To address these challenges, the research team will combine three innovations. First, the key enabler to this research is an initial assessment on the “size” of the design space; how many usefully-independent degrees of freedom are there? This reduces the combinatorial problem. Second, the computational burden can be reduced by separating the simulations needed to understand the spatial- and seasonal- distribution of stratospheric aerosol optical depth (AOD), which can be short but require a complete stratosphere model, from those needed to assess the climate response to a specified aerosol distribution, which require multi-decadal simulations but not an accurate stratosphere. And third, nonlinearities and uncertainty can be managed through feedback that adjusts injection rates; this enables comparing simulations based on specified objectives rather than specified injection rates. The research team will design a suite of simulations that individually meet different objectives and collectively span the space of possible outcomes. From this, the key tool in evaluating and visualizing trade-offs is through Pareto-optimal surfaces: how do strategies and their responses change as a function of the optimization criteria. Although the simulations will be focused on understanding physical science tradeoffs, social and governance dimensions play a critical role in understanding which objectives may be most important to achieve or which strategies are simply politically infeasible, thus limiting the space in ways not revealed by climate modeling. Therefore, the research team will interface with governance experts throughout to ensure research informs policy. Simulations will also be made available to the wider international community.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.