Collaborative Research: Numerical Modeling of the Internal-Wave Cascade and Submesoscale Lateral Dispersion in the Ocean

横向搅动是基本过程之一，沿着平流和底界（垂直）混合，决定着海洋中水团特性、营养物和溶解气体的分布和归宿。海洋环流模式（OGCM）解决了尺度大于~30 km的搅拌，区域模式解决了尺度为~1 km的搅拌，但必须将亚中尺度（1至10 km）和细尺度（10 m至1 km）过程的等密（横向）搅拌参数化。示踪剂释放实验一致发现，在1-10公里尺度的等密度扩散系数是一个数量级大于内波剪切色散的预测。该项目将对内波和涡旋模式在10米至10公里量级的次中尺度横向搅动中的作用以及能量对耗散垂直尺度的级联提供更全面和准确的评估，以解决这一矛盾。内波和涡模在观测上很难区分。然而，明智地使用数值模型可以通过隔离不同的物理来区分它们的影响。该项目方法的一个主要优点是，所有上述过程/动力学可以在一个单一的模式的背景下进行探索，使我们能够无缝地探索内部波级联和横向分散的不同机制，在广泛的参数制度和强迫条件。拟议的模拟将阐明基本的物理和告知我们的理解，这些过程如何在海洋中工作，以便提供一个订单一次参数化的次中尺度等密扩散系数的OGCM。在这项研究中开发的参数化将通过同行评审的出版物和会议介绍提供。两名研究生将接受高分辨率数值建模和数据分析技术的培训。还将维护一个公共项目网站，我们的模拟的初始条件和参数存档，并应要求与其他有兴趣合作的研究人员共享。所提出的方法将使用Boussinesq伪谱模型来研究横向色散和内波能量级联。具体目标是确定（1）内波叶栅中的波/波和波/涡相互作用对小垂直尺度和湍流产生的作用，（2）等密度搅拌中内波场的波/波和波/涡失相，（3）内波剪切弥散中的湍流不稳定性，以及（4）涡模剪切弥散和搅拌。模式结果也将用于检验（5）细尺度滚降的物理学和（6）不稳定切变事件的统计（Richardson数小于四分之一）。模拟有和没有细尺度位涡产生的内波破碎将被用来隔离的涡模逆叶栅的作用。被动示踪剂受到diapycnal混合和非扩散拉格朗日粒子将被用来区分剪切分散和搅拌。模拟将运行不同的浮力频率，科里奥利频率，内波频谱能量水平和频谱形状，探索级联，细尺度滚降，理查森数统计，diapycnal扩散和等密度扩散的基本参数依赖性。OGCM和区域模式的一阶次中尺度水平扩散系数将产生。将测试用于将内波场维持在统计稳定状态的不同强迫（表面风、潮汐）。夏季在马尾藻海收集的剖面浮标和染料数据将用于指导模型初始化和评估结果的真实性。在可能的情况下，将确定模拟弥散特征的差异，并用于确定观测数据集中的搅拌机制。

Lateral stirring is among the fundamental processes, along with advection and diapycnal (vertical) mixing that determines the distribution and fate of water-mass properties, nutrients and dissolved gases in the ocean. Ocean general circulation models (OGCMs) resolve stirring on scales larger than ~30 km, and regional models on scales ~1 km, but isopycnal (lateral) stirring by submesoscale (1 to 10 km) and finescale (10 m to 1 km) processes must be parameterized. Tracer-release experiments consistently find isopycnal diffusivities at scales of 1-10 km to be an order of magnitude larger than predictions for internal-wave shear dispersion. This project will provide a more complete and accurate assessment of the roles of internal waves and vortical mode in lateral stirring at the submesoscale of order 10 m to 10 km), and the cascade of energy to dissipative vertical scales, to resolve this paradox. Internal waves and vortical mode are difficult to distinguish observationally. However, judicious use of a numerical model can tease apart their influences by isolating different physics. A major advantage of the approach of this project is that all of the above processes/dynamics can be explored in the context of a single model, allowing us to seamlessly explore the internal-wave cascade and lateral dispersion by the different mechanisms across a wide range of parameter regimes and forcing conditions. The proposed simulations will elucidate the underlying physics and inform our understanding of how these processes work in the ocean so as to provide an order-one parameterization of submesoscale isopycnal diffusivities for OGCMs. Parameterizations developed in this study will be made available through peer-reviewed publications and conference presentations. Two graduate students will be trained in high resolution numerical modeling and data analysis techniques. A public project website will also be maintained, with initial conditions and parameters for our simulations archived and shared on request with other investigators interested in collaboration.The proposed approach will use a Boussinesq pseudo-spectral model to investigate lateral dispersion and the internal-wave energy cascade. Specific goals are to determine the roles of (1) wave/wave and wave/vortex interactions in the internal-wave cascade to small vertical scales and turbulence production, (2) wave/wave and wave/vortex de-phasing of the internal wave field in isopycnal stirring, (3) turbulent intermittency in internal-wave shear dispersion, and (4) vortical-mode shear dispersion and stirring. Model results will also be used to examine (5) the physics of the finescale roll-off and (6) the statistics of unstable shear events (Richardson number less than a quarter). Simulations with and without finescale potential vorticity production by internal-wave breaking will be used to isolate the roles of the vortical-mode inverse cascade. Passive tracers subject to diapycnal mixing and non-diffusive Lagrangian particles will be used to distinguish between shear dispersion and stirring. The simulations will be run with varying buoyancy frequency, Coriolis frequency, internal-wave spectral energy level and frequency spectral shapes to explore fundamental parameter dependences of the cascade, finescale roll-off, Richardson Number statistics, diapycnal diffusion and isopycnal diffusion. First-order submesoscale horizontal diffusivities for OGCMs and regional models will result. Different forcings (surface wind, tidal) for maintaining the internal-wave field in a statistically steady state will be tested. Profiling float and dye data collected in the Sargasso Sea during summer will be used to guide model initialization and assess the realism of the results. Where possible, differences in modelled dispersion characteristics will be identified and used to identify stirring mechanisms in the observational data sets.