Cloud Parameterization Frameworks

PI将继续他的对流云系统和行星边界层（PBL）的参数化发展。他的新PBL参数化将包括横滚环流对水平动量垂直输送的机械表示。该模型预测了轧辊的宽度和方向，并对扰动风场进行了诊断。该信息用于从非弹性压力方程中计算扰动压力场，进而用于计算预测垂直速度方差和其他统计数据所需的压力相关性。参数化将使用大涡模拟（LES）结果进行测试。PBL参数化将改变为使用显式PBL深度，其中大约有10层或更少的层。PI的目的是在科罗拉多州立大学的大气环流模式中测试这种参数化，并最终在社区气候系统模式中测试。为了预测深度，他必须将夹带速率参数化。他将开发一种夹带参数化方法，用于垂直分辨PBL，并预测垂直速度场的偏度等统计数据。从不同的角度对夹带问题进行新的审视，可能会提高对基础物理学的理解。一些重点将放在蒸发冷却在提高夹带速率和减少分数云量方面的作用。在他早期工作中使用的顶帽概率密度函数（PDF）将被更现实和灵活的“空间分布函数”所取代。最后，新的参数化将包括降水过程的表示。深对流的参数化将被改变，以放弃Arakawa-Schubert方法，该方法基于带云型谱的带羽云模式。在它的位置上，PI将在每层放置一个单一的对流上升气流，但内部结构已分解。随着模型垂直分辨率的增加，这将改善参数化的缩放。对流下降气流和层状云也将包括在这个框架中。PI对深对流及其伴随的层状云的修正参数化将利用一种新的云模式进行参数化。在Arakawa-Schubert参数化中，有许多云类型，每一种都有一个粗略理想化的内部结构。PI将用一个更真实的内部结构的单一“云类型”来取代它，包括垂直速度和热力学变量的联合变化（在给定的水平上）。对流下降气流将通过第二个PDF表示。云模式的PDF可以在一个统一的框架中包括对流云和层状云。事实上，PDF甚至可以包括诸如空气中水汽混合比的空间变异性。有研究表明，湿润的中尺度区域（周围空气干燥得多）可以为深积云的生长提供培育环境，而在缺乏这种中尺度湿润区域的情况下，深对流受到抑制。通过参数化晴空中水汽的中尺度变化，PI可以在大尺度模式的背景下探索这一想法。他的工作将以高分辨率云模型的结果为指导。为了完成参数化，他的新云模型将与预测闭合相结合。他将预测垂直速度作为高度函数的多个矩，以及热力学变量的多个矩。该研究将代表着PBL和深层对流参数化统一的一步。更广泛的影响：这项研究将为改进天气预报和改进气候变化模拟铺平道路。人们普遍认为，云参数化的缺陷是阻碍更可靠地模拟气候变化的最严重障碍之一。这项研究还将有助于培养从事大气科学工作的研究生和博士后研究人员。

The PI will continue his development of parameterizations of convective cloud systems and the planetary boundary layer (PBL). His new PBL parameterization will include a mechanistic representation of the vertical transport of horizontal momentum by roll circulations. The model predicts the width and orientation of the rolls, and the perturbation wind field is diagnosed. This information is used to compute the perturbation pressure field from the anelastic pressure equation, and this in turn is used to compute the pressure correlations needed to predict the vertical velocity variance and other statistics. The parameterization will be tested using large-eddy simulation (LES) results. The PBL parameterization will be altered to use an explicit PBL depth, with about 10 or fewer layers inside. The PI's intention is to test this parameterization in the Colorado State University general circulation model and eventually in the Community Climate System Model. In order to predict the depth, he must parameterize the entrainment rate. He will develop an entrainment parameterization that is designed for use with a vertically resolved PBL and predicted statistics such as the skewness of the vertical velocity field. This fresh look at the entrainment problem, from a different perspective, may lead to an improvement in the understanding of the basic physics. Some emphasis will be placed on the role of evaporative cooling in enhancing the entrainment rate and reducing the fractional cloudiness. The top-hat probability density function (PDF) used in his earlier work will be replaced by a more realistic and flexible "spatial distribution function." Finally, the new parameterization will include a representation of precipitation processes. The parameterization of deep convection will be altered to abandon the Arakawa-Schubert approach based on an entraining plume cloud model with a spectrum of cloud types. In its place the PI will put a single convective updraft at each level, but with a resolved internal structure. This will improve the scaling of the parameterization as the vertical resolution of the model is increased. Convective downdrafts and stratiform clouds will also be included in this framework. The PI's revised parameterization of deep convection and the attendant stratiform clouds will make use of a new cloud model for parameterization. In the Arakawa-Schubert parameterization there are many cloud types, each with a crudely idealized internal structure. The PI will replace this by a single "cloud type" with a more realistic internal structure, including joint variations (at a given level) of the vertical velocity and the thermodynamic variables. Convective downdrafts will be represented through a second PDF. The PDF of the cloud model can include both convective and stratiform clouds in a unified framework. In fact, the PDF can even include such things as spatial variability of the water vapor mixing ratio in clear air. It has been suggested that humid mesoscale regions (surrounded by much drier air) can provide nurturing environments for the growth of deep cumuli, and that in the absence of such mesoscale humid regions deep convection is suppressed. By parameterizing the mesoscale variability of water vapor in clear air, the PI can explore this idea in the context of a large-scale model. His work will be guided by the results obtained with high-resolution cloud models. To complete the parameterization, his new cloud model will be combined with a prognostic closure. He will predict multiple moments of the vertical velocity as functions of height, as well as multiple moments of the thermodynamic variables. This research will represent a step towards unification of the parameterizations of the PBL and deep convection.Broader Impacts:The research will pave the way for improved weather forecasts and improved simulations of climate change. Deficiencies in cloud parameterizations are widely acknowledged to be among the most serious obstacles standing in the way of more reliable simulations of climate change. The research will also contribute to the training of graduate students and postdoctoral researchers for careers in atmospheric science.