Collaborative Research: PetaApps: Enabling Multiscale Modeling of Turbulent Clouds on Petascale Computers

合作研究：PetaApps：在千万亿级计算机上实现湍流云的多尺度建模

• 批准号: 0904534
• 负责人: Lian-Ping Wang
• 金额: $ 106.45万
• 依托单位: University of Delaware
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2009
• 资助国家: 美国
• 起止时间: 2009-09-01 至 2013-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0904534&HistoricalAwards=false
• 关键词: Collaborative; Research; PetaApps; Enabling; Multiscale

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5). Reliable weather and climate prediction at local-to-global scales ultimately depends on a full and quantitatively accurate understanding of microphysical processes governing interactions of individual cloud droplets and their interplay with cloud-scale dynamics. Of those mechanisms known to affect the growth of individual droplets and evolution of cloud droplet populations within clouds, turbulence induced collision-coalescence appears to be among the most complex. Previous direct numerical simulation methods capable of resolving individual droplets have traditionally been confined to consideration of exceedingly small volumes (of order a cubic meter), as well as being restricted to relatively laminar (non-turbulent) flows. These methods have thus been incapable of addressing the impact of turbulent eddies, which typically have dimensions of 10-100 m, on cloud droplet-scale processes such as condensation, coalescence and mixing. Even very high resolution "Large Eddy Simulation (LES)" models, which have only recently achieved grid spacings routinely approaching 10 m, must approximate cloud microphysics via simplifying (and inherently limited) parameterization techniques. As such, the millions to billions of individual droplets comprised by even a small cloud cannot be individually represented and tracked via these traditional approaches.Application of advanced petascale computing methods by this investigative team will allow turbulence-to-cloud-scale (i.e. multiscale) cloud dynamics and microphysics to be treated in a seamless fashion absent unjustified simplifying assumptions. This approach will lead to a more robust understanding of shallow ice-free convective clouds, such as subtropical stratocumulus and trade-wind cumulus, which are known to play a critical role in the global climate system. Computer codes developed under this effort will be designed for ease of portability, ensuring their availability for broad use throughout the relevant research community.Broader Impact: This collaborative effort will allow advanced engineering fluid-mechanics research tools and state-of-the-art petascale computational methods to be effectively merged and applied to critical atmospheric processes within a unified numerical simulation framework. This approach will address several open questions that have faced the cloud-physics research community for many years, and will ultimately contribute to improved weather and climate forecasts. Multidisciplinary education will be enhanced through collaborative involvement of multiple graduate and undergraduate students, as well as a Postdoctoral scholar.

该奖项是根据2009年美国复苏和再投资法案（公法111-5）资助的。从局部到全球尺度的可靠天气和气候预测最终取决于对控制单个云滴相互作用及其与云尺度动力学相互作用的微物理过程的全面和定量准确的理解。 在已知的影响单个云滴的生长和云中云滴群演化的机制中，湍流诱导的碰撞聚结似乎是最复杂的机制之一。 先前能够分辨单个液滴的直接数值模拟方法传统上限于考虑极小的体积（立方米量级），以及限于相对层流（非湍流）流动。 因此，这些方法无法解决湍流涡旋的影响，湍流涡旋通常具有10-100米的尺寸，对云滴尺度的过程，如凝结，合并和混合。 即使是非常高分辨率的“大涡模拟（LES）”模式，最近才实现网格间距经常接近10米，必须近似云微物理通过简化（和固有的限制）参数化技术。 因此，即使是一个很小的云，也不能通过这些传统方法单独表示和跟踪数百万到数十亿个单独的液滴。这个研究小组应用先进的千万亿次计算方法，将使云尺度（即多尺度）的云动力学和微观物理学能够以无缝的方式处理，而不需要不合理的简化假设。 这种方法将导致对浅层无冰对流云的更强有力的理解，如亚热带层积云和信风积云，这些云在全球气候系统中发挥着关键作用。 在这项工作下开发的计算机代码将被设计为易于移植，确保其可用性，在整个相关的研究社区广泛使用。更广泛的影响：这一合作努力将允许先进的工程流体力学研究工具和国家的最先进的千万亿次计算方法有效地合并，并在一个统一的数值模拟框架内应用于关键的大气过程。 这种方法将解决云物理研究界多年来面临的几个悬而未决的问题，并最终有助于改善天气和气候预报。 多学科教育将通过多个研究生和本科生的合作参与，以及博士后学者得到加强。