The causal pathways of the Indian-Atlantic interocean exchange

物理海洋学中的一个基本问题是不同海洋如何相互作用。回答这个问题对于我们理解全球海洋环流和海洋在气候系统中的作用至关重要。其中一个相互作用点是非洲南部的海域，那里是印度洋、南大西洋和南大洋的交汇点。虽然这一领域的大洋间交换具有全球重要性已是公认的事实，但人们对决定这一交换的存在和规模的基本过程却知之甚少。该项目将促进人们了解哪些进程负责印度洋-大西洋间的交换，以及它们如何以及在多大程度上协同工作以实现这种交换。它汇集了在非线性数据同化、海洋建模、观测数据集和非线性因果发现方面的最新发展，以促进这一理解。因此，该项目加强了我们对海洋环流的基本了解，但也进一步开发了可用于许多其他海洋问题、其他地球科学领域和其他领域的新工具。在绝地系统中实现的完全非线性数据同化方法在这个高度动荡的海洋区域进行了测试，这在世界上是独一无二的，科学家和作战中心都可以应用。系统的非线性因果发现是一个不断发展和令人兴奋的研究领域，值得成熟为一个标准的、校准良好的工具，超越更传统的线性因果发现方法。因果发现框架可以应用于许多突出的海洋学和更广泛的地学问题。此外，它还可以应用于纯模型研究，以比较因果路径并突出模型的缺陷。学生将接触到尖端数据同化、海洋建模和观测以及因果发现，从而培养出一位特别全面的初出茅庐的科学家。研究结果将通过会议报告和出版物广泛传播给海洋学、气候学和应用数学/工程界。许多过程被认为对非洲的大洋间交换S很重要，例如纳塔尔脉冲(阿古拉斯流的大型气旋性弯道)、莫桑比克涡旋、马达加斯加南端脱落的偶极子、回流区的正压-斜压混合不稳定，可能与Rossby波类似的盆地模式、印度尼西亚贯穿气流和南赤道流的变化、热力结构的大尺度变化，包括NADW流入产生的涡度、大尺度风扰动和气候耦合模式例如，印度洋偶极子。这个项目将检验这样一种假设，即大洋间的交换是由阿古拉什反反射区的紧密环流涡旋控制的，而该环流涡旋又受到上述物理过程的调制。这项研究的主要贡献将是使用一种新的完全非线性因果发现方法来解开和量化这些直接贡献和来自非线性相互作用的贡献。这种方法将这些非线性相互作用完全分解为两个过程相互作用、三个过程相互作用等。此外，它还允许量化任何未确定的控制过程的贡献的大小，指导对迄今未知的重要物理过程的研究。这种方法将被应用于处理南非周围海域长达30年的重新分析的时间序列，使用完全非线性的数据同化，结合所有可用的观测，以回答以下主要研究问题：·导致大洋间交换的物理过程是什么？·这些物理过程如何相互作用导致大洋间交换？该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

A fundamental question in physical oceanography is how different oceans interact. Answering this question is crucial for our understanding of the global ocean circulation and the role the ocean plays in the climate system. One of these interaction points is the ocean area south of Africa where the Indian, South Atlantic and Southern Oceans meet. While it is now well established that the interocean exchange in this area is of global importance, much less is known on what the underlying processes are that determine the existence and size of this exchange. This project will advance the understanding of which processes are responsible for the Indian-Atlantic inter-ocean exchange and how and to what extend they work in concert to achieve the exchange. It brings together state-of-the-art developments in nonlinear data assimilation, ocean modeling, observational data sets and nonlinear causal discovery to advance that understanding. As such the project enhances our fundamental understanding of the ocean circulation, but also further develops new tools that can be used in many other oceanographic problems, in other geoscience areas, and beyond. The fully nonlinear data assimilation method implemented in the JEDI system and tested on this highly turbulent ocean area will be unique in the world and can be applied by scientists and operational centers alike. Systematic nonlinear causal discovery is a growing and exciting research field that deserves maturing into a standard well- calibrated tool, beyond more traditional linear causal discovery methodologies. The causal discovery framework can be applied to many outstanding oceanographic and wider geoscience problems. Furthermore, it can be applied to pure modeling studies to compare causal pathways and highlight model deficiencies. A student will be exposed to cutting edge data assimilation, ocean modeling and observations, and causal discovery, leading to an exceptionally rounded budding scientist. Results of the study will be broadly communicated to the oceanographic, climate and applied mathematical/engineering communities through conference presentations and publicationsMany processes have been put forward as important for the interocean exchange S of Africa, such as Natal Pulses (large cyclonic meanders of the Agulhas Current), Mozambique Eddies, dipoles shed from the southern tip of Madagascar, mixed barotropic-baroclinic instabilities in the retroflection area, perhaps related to Rossby-wave like basin modes, variations in the Indonesian Throughflow and the South Equatorial Current, large-scale variations in the thermodynamic structure, including vorticity generation due to NADW inflow, large-scale wind perturbations, and coupled climate modes, e.g. the Indian Ocean Dipole. This project will test the hypothesis that the interocean exchange is controlled by the tight recirculation gyre in the Agulhas Retroflection area, which is in turn modulated by the physical processes mentioned. The main contribution of this research will be to unravel and quantify these direct contributions and the contributions from nonlinear interactions using a new fully nonlinear causal discovery methodology. This methodology provides for a complete decomposition of these nonlinear interactions into two-process interactions, three-process interactions etc. Furthermore, it allows to quantify the magnitude of the contributions of any unidentified controlling processes, directing research to hitherto unknown important physical processes. This methodology will be applied to process time series from a 30-year reanalysis of the ocean area around South Africa, using fully nonlinear data assimilation, incorporating all available observations to answer the following overarching research questions:• What are the physical processes that are responsible for the interocean exchange?• How do these physical processes interact to cause the interocean exchange?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.