Improved Representation of Cloud-Aerosol Interactions in the Community Earth System Model: A New Sectional Cloud Model that Interacts with Modal and Sectional Aerosol Models

云通过将阳光反射到太空并阻挡向外的红外辐射，对地球的能量平衡有着深远的影响。因此，地球的气候对所有过程都很敏感，这些过程统称为云微物理学，控制着云粒子的形成和增长以及通过蒸发和降水去除它们。 例如，随着气候变暖，云中的冰粒丰度减少，有利于液滴，这使得云更具反射性，从而对变暖产生抵消作用（负反馈）。 液态云也往往持续更长时间，因为它们在产生降水方面的效率较低，这可能进一步增强从冰粒到水滴转变的负反馈。气候对云微物理的敏感性给气候研究带来了挑战，特别是因为气候模型必须代表整个全球气候系统，而大部分微物理发生在不到一毫米的距离上。 全球模式使用参数化来表示云微物理学的整体效应，但这些参数化必然是粗糙的，因为需要执行长时间和计算密集型的模拟。 微物理参数化的一个问题是，它们使用非物理参数来调整云的行为，这些参数对重要的气候系统行为有直接影响，例如全球温度对温室气体增加的敏感性。 一个恰当的例子是自动转换，这是云粒子相互作用形成降水过程的参数表示。自动转换根据外部强加的阈值标准将云的冻结和液态水的一部分概括地转换为雨滴或雪花。 自动转换阈值的选择影响云的寿命，从而影响大气层顶的能量平衡，从而使非物理阈值对模拟的气候产生巨大的影响。这里所做的工作发展了一种替代的云微物理模式，其中自动转换和其他一步近似被一个更详细的公式所取代，在这个公式中，云粒子用尺寸分布来表示，这意味着该模型将液滴和冰粒划分到一组尺寸箱（也称为尺寸分布的部分）中，并跟踪每个箱中的粒子丰度。 然后通过箱之间的相互作用来表示微观物理，例如，如果小液滴随着水蒸气在其上冷凝而变大，则它们从用于小液滴的箱转移到用于较大液滴的箱。 该方案的一个优点是，云微物理的更明确的表示消除了许多非物理参数中发现的简单的方案。 云微物理模式是基于CARMA（Community Aerosol and Radiation Model for Atmospheres）的，它使用一个尺寸箱来表示气溶胶的化学和微物理。 在这里，箱方案适于表示液体云滴和云冰的微观物理，具有表示通过冻结和解冻的水滴箱和冰颗粒箱之间的水的转移的能力。 用于表示气溶胶和云的CARMA版本分别称为CARMA-aerosol和CARMA-cloud，它们一起用于表示气溶胶对云粒子的影响。 该奖项继续在以前的支持下开发CARMA，最近通过AGS-1640903开发，该模型用于解决云物理和气候动力学中的几个问题。 特别是，该模式被用来考虑云微物理学对气候变化的影响，通过模拟，其中二氧化碳浓度瞬间增加一倍，一个标准的方式来评估模拟气候对温室气体增加的敏感性。 模拟的动机来自于最新一代气候模式中发现的气候敏感性增加，这些气候模式有助于气候模式相互比较项目（CMIP），这被归因于云微物理学的变化。这项工作通过努力提高对云微物理学在气候变化中作用的理解而具有社会意义。 云微物理学是一个特别令人关注的问题，因为在用于决策支持的未来气候变化模型预测中，云经常被称为最大的不确定性来源。 这项工作也有利于世界各地使用和开发CESM的研究人员。 该项目通过开发适合课堂使用的CARMA云模型的独立版本以及通过对研究生的支持和培训而具有教育价值。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估而被认为值得支持。

Clouds have a profound effect on the energy balance of the Earth, by reflecting sunlight to space and blocking outgoing infrared radiation. Earth's climate is thus sensitive to all the processes, collectively referred to as cloud microphysics, that govern the formation and growth of cloud particles and their removal through evaporation and precipitation. For example, as climate warms the abundance of ice particles in clouds decreases in favor of liquid droplets, which makes the clouds more reflective and thus has a counteracting effect on the warming (a negative feedback). Liquid clouds also tend to last longer as they are less effective in generating precipitation, which could further enhance the negative feedback of the shift from ice particles to droplets.The sensitivity of climate to cloud microphysics poses a challenge for climate research, particularly as climate models must represent the full global climate system while much of the microphysics takes place over distances less than a millimeter. Global models use parameterizations to represent the bulk effects of cloud microphysics but these parameterizations are necessarily crude given the need to perform long and computationally intensive simulations. One concern with microphysics parameterizations is that they use nonphysical parameters to adjust the behavior of the clouds, and these parameters have a direct effect on important climate system behaviors such as the sensitivity of global temperature to greenhouse gas increases. A case in point is autoconversion, a parametric representation of the processes through which cloud particles interact to form precipitation. Autoconversion summarily converts some portion of a cloud's frozen and liquid water into raindrops or snowflakes according to externally imposed threshold criteria. The choice of threshold values for autoconversion affects cloud lifetimes and thus affects the top-of-atmosphere energy balance, thus giving the nonphysical thresholds an outsized effect on the simulated climate.Work performed here develops an alternative cloud microphysics model in which autoconversion and other one-step approximations are replaced by a more detailed formulation in which cloud particles are represented in terms of a size distribution, meaning the model partitions droplets and ice particles into a set of size bins, also referred to as sections of the size distribution, and keeps track of the abundance of particles in each bin. Microphysics is then represented through interactions between bins, for instance if small droplets grow bigger as water vapor condenses on them they are transferred from a bin for small droplets to a bin for larger ones. An advantage of the scheme is that the more explicit representation of cloud microphsyics eliminates many of the nonphysical parameters found in simpler schemes. The scheme is too computationally intensive for use in century-scale climate simulations but is practical for decadal simulations and can be used to inform development of simpler fast schemes.The cloud microphysics model is based on CARMA, the Community Aerosol and Radiation Model for Atmospheres, which uses a size bin scheme to represent the chemistry and microphysics of aerosols. Here the bin scheme is adapted to represent the microphysics of liquid cloud droplets and cloud ice, with the ability to represent the transfer of water between droplet bins and ice particle bins through freezing and thawing. The versions of CARMA used to represent aerosols and clouds are referred to as CARMA-aerosol and CARMA-cloud, respectively, and they are used together to represent the effects of aerosols on cloud particles. This award continues development of CARMA under previous support, most recently through AGS-1640903.Once developed, the model is used to address several issues in cloud physics and climate dynamics. In particular the model is used to consider the effect of cloud microphysics on climate change through simulations in which carbon dioxide concentration is instantaneously doubled, a standard way to assess the sensitivity of simulated climate to greenhouse gas increases. Motivation for the simulations comes from the increased climate sensitivity found in the latest generation of climate models contributing to the Climate Model Intercomparison Project (CMIP), which has been ascribed to changes in cloud microphysics.The work has societal relevance through its effort to improve understanding of the role of cloud microphysics in climate change. Cloud microphysics is a particular concern as clouds are frequently called out as the greatest source of uncertainty in model projections of future climate change used for decision support. The work also benefits the worldwide community of researchers who use and develop CESM. The project has educational value through the development of a stand-alone version of CARMA-cloud model which is suitable for classroom use, and through the support and training of a graduate student.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.