Collaborative Proposal: Harnessing simulation data to characterize transition layer mixing rates and mechanisms

协作提案：利用模拟数据来表征过渡层混合速率和机制

• 批准号: 2122867
• 负责人: Eric D'Asaro
• 金额: $ 8.86万
• 依托单位: University of Washington
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-09-01 至 2024-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2122867&HistoricalAwards=false
• 关键词: Collaborative; Proposal; Harnessing; simulation; data

This project will examine entrainment processes in the ocean transition layer (TL), the vertical region between the lower part of the turbulent near-surface mixed layer and the stably stratified ocean interior. The project will use a novel combination of approaches, including further analysis of existing high-resolution profile data through the upper ocean layers, high-resolution modeling of entrainment and internal wave processes in the transition layer, and machine learning techniques. These processes help determine ocean mixed layer depth and temperature and ultimately mediate air-sea exchange effects on the global ocean, yet they remain poorly understood or represented in numerical model parameterizations. Outcomes will be directly relevant to improving physical and biogeochemical ocean models. Additionally, the methods developed will be applicable to interpreting and analyzing data from a variety of geophysical flows, and the analysis scripts will be made publicly available. The work will support an early career investigator, the training of a graduate student, contributions to local outreach and educational programs, and will form the basis for a project for a graduate student in the WHOI summer program in Geophysical Fluid Dynamics. A suite of high-resolution direct numerical simulations will be generated covering a range of expected TL mechanisms, including Kelvin-Helmholtz and Holmboe instabilities and interfacial waves. Using standard fluid dynamical analyses, including characterization of the linear instabilities and a detailed analysis of the flow energetics, this comprehensive library of flow fields will be used to determine how well a given stratified mixing event can be characterized from limited measurements. The simulations will then be used as training data for a neural network-based flow classification method, allowing for input profiles of temperature and velocity to be classified in terms of the underlying waves or instabilities. After the classification method is validated using simulation data, it will be applied to the observations, allowing for identification of the relevant mechanisms driving entrainment in the TL. Knowledge of the mixing associated with each mechanism can thus be used to describe the mixing efficiency and turbulent fluxes in the observational record.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.

该项目将研究海洋过渡层（TL）中的夹带过程，该过渡层是湍流近地表混合层下部与稳定分层海洋内部之间的垂直区域。该项目将采用新颖的方法组合，包括进一步分析上层海洋层的现有高分辨率剖面数据、过渡层夹带和内波过程的高分辨率建模以及机器学习技术。这些过程有助于确定海洋混合层深度和温度，并最终调节全球海洋的海气交换效应，但它们在数值模型参数化中仍然知之甚少或代表不足。结果将与改进物理和生物地球化学海洋模型直接相关。此外，开发的方法将适用于解释和分析来自各种地球物理流的数据，并且分析脚本将公开。这项工作将支持早期职业调查员、研究生培训、对当地推广和教育项目的贡献，并将构成 WHOI 地球物理流体动力学夏季项目研究生项目的基础。将生成一套高分辨率直接数值模拟，涵盖一系列预期的 TL 机制，包括开尔文-亥姆霍兹和 Holmboe 不稳定性和界面波。使用标准流体动力学分析，包括线性不稳定性的表征和流动能量学的详细分析，这个综合流场库将用于确定通过有限的测量来表征给定分层混合事件的程度。然后，模拟将用作基于神经网络的流分类方法的训练数据，允许根据底层波浪或不稳定性对温度和速度的输入轮廓进行分类。使用模拟数据验证分类方法后，它将应用于观察，从而识别驱动 TL 中夹带的相关机制。因此，与每种机制相关的混合知识可用于描述观测记录中的混合效率和湍流通量。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力优点和更广泛的影响审查标准进行评估，被认为值得支持。