Abstract This paper summarizes the findings of a workshop convened in the United States in 2018 to discuss methods in coastal and estuarine modeling and to propose key areas of research and development needed to improve their accuracy and reliability. The focus of this paper is on physical processes, and we provide an overview of the current state-of-the-art based on presentations and discussions at the meeting, which revolved around the four primary themes of parameterizations, numerical methods, in-situ and remote-sensing measurements, and high-performance computing. A primary outcome of the workshop was agreement on the need to reduce subjectivity and improve reproducibility in modeling of physical processes in the coastal ocean. Reduction of subjectivity can be accomplished through development of standards for benchmarks, grid generation, and validation, and reproducibility can be improved through development of standards for input/output, coupling and model nesting, and reporting. Subjectivity can also be reduced through more engagement with the applied mathematics and computer science communities to develop methods for robust parameter estimation and uncertainty quantification. Such engagement could be encouraged through more collaboration between the forward and inverse modeling communities and integration of more applied math and computer science into oceanography curricula. Another outcome of the workshop was agreement on the need to develop high-resolution models that scale on advanced HPC systems to resolve, rather than parameterize, processes with horizontal scales that range between the depth and the internal Rossby deformation scale. Unsurprisingly, more research is needed on parameterizations of processes at scales smaller than the depth, including parameterizations for drag (including bottom roughness, bedforms, vegetation and corals), wave breaking, and air–sea interactions under strong wind conditions. Other topics that require significantly more work to better parameterize include nearshore wave modeling, sediment transport modeling, and morphodynamics. Finally, it was agreed that coastal models should be considered as key infrastructure needed to support research, just like laboratory facilities, field instrumentation, and research vessels. This will require a shift in the way proposals related to coastal ocean modeling are reviewed and funded.

中文翻译：

---

沿海和河口建模的未来：研讨会的发现

摘要 本文总结了 2018 年在美国召开的研讨会的结果，该研讨会讨论了沿海和河口建模的方法，并提出了提高其准确性和可靠性所需的关键研究和开发领域。本文的重点是物理过程，我们根据会议上的演示和讨论提供了当前最先进技术的概述，这些讨论围绕参数化、数值方法、原位的四个主要主题展开。和遥感测量以及高性能计算。研讨会的一个主要成果是就减少主观性和提高沿海海洋物理过程建模的可重复性的必要性达成了一致。可以通过制定基准、网格生成、通过制定输入/输出、耦合和模型嵌套以及报告的标准，可以提高验证和再现性。通过更多地与应用数学和计算机科学界合作，开发用于稳健参数估计和不确定性量化的方法，也可以减少主观性。可以通过正向和逆向建模社区之间的更多合作以及将更多应用数学和计算机科学整合到海洋学课程中来鼓励这种参与。研讨会的另一个成果是同意需要开发可在高级 HPC 系统上扩展的高分辨率模型，以解决而不是参数化具有介于深度和内部 Rossby 变形尺度之间的水平尺度的过程。不出所料，需要对小于深度的尺度上的过程参数化进行更多研究，包括阻力参数化（包括底部粗糙度、床型、植被和珊瑚）、波浪破碎和强风条件下的海气相互作用。其他需要更多工作才能更好地参数化的主题包括近岸波浪建模、沉积物输送建模和形态动力学。最后，一致同意应将沿海模型视为支持研究所需的关键基础设施，就像实验室设施、现场仪器和研究船一样。这将需要改变与沿海海洋建模相关的提案的审查和资助方式。植被和珊瑚）、波浪破碎和强风条件下的海气相互作用。其他需要更多工作才能更好地参数化的主题包括近岸波浪建模、沉积物输送建模和形态动力学。最后，一致同意应将沿海模型视为支持研究所需的关键基础设施，就像实验室设施、现场仪器和研究船一样。这将需要改变与沿海海洋建模相关的提案的审查和资助方式。植被和珊瑚）、波浪破碎和强风条件下的海气相互作用。其他需要更多工作才能更好地参数化的主题包括近岸波浪建模、沉积物输送建模和形态动力学。最后，一致同意应将沿海模型视为支持研究所需的关键基础设施，就像实验室设施、现场仪器和研究船一样。这将需要改变与沿海海洋建模相关的提案的审查和资助方式。人们一致认为，沿海模型应被视为支持研究所需的关键基础设施，就像实验室设施、现场仪器和研究船一样。这将需要改变与沿海海洋建模相关的提案的审查和资助方式。人们一致认为，沿海模型应被视为支持研究所需的关键基础设施，就像实验室设施、现场仪器和研究船一样。这将需要改变与沿海海洋建模相关的提案的审查和资助方式。