Workshop on the future of coastal and estuarine modeling; June 2018; North Carolina State University

精确的数值模型对于提高我们对沿海和河口尺度的基本物理过程的理解至关重要，在这些尺度上，人类与海洋之间的联系最为明显，特别是在世界各地迅速扩张的沿海特大城市附近。由于计算能力的提高和数值方法的改进，包括非结构化网格/模型嵌套、数据同化和模型耦合，模型在过去二十年中取得了重大进展。此外，遥感和原位观测技术的进步使数据集大大扩大和更加准确，从而大大提高了评估模式性能和量化不确定性的能力。然而，尽管最近取得了进展，但在过去二十年中已经开发了大量的沿海和河口模型，并且缺乏统一的方法，如在区域建模中所见。本次研讨会将汇集模型使用者和开发人员，讨论如何改进沿海和河口尺度物理海洋学科学研究的建模工具。预算将为在美国工作的早期专业人员提供旅费和参与支持，并将重点吸引代表性不足的少数民族和妇女参加研讨会。早期职业科学家将与经验丰富的模型开发人员和用户在一个专注于讨论和协作的环境中进行互动，从而帮助促进一个更强大的社区，将继续提高沿海和河口模型的准确性和效率。这次讲习班的重点是发展对目前最先进的沿海和河口过程数值模拟的优点和局限性的理解，并就提高这种模式的准确性和可靠性所需的研究和发展的关键领域达成共识。研讨会的成果将包括对不同方法的优缺点的统一理解，例如有限体积与有限元方法，曲线与非结构化网格，海洋，大气，陆地（水文）模型的耦合，大涡模拟与RANS湍流建模，以及不同的模型嵌套和自适应网格策略。讨论将集中在这些方法以及强制策略、边界条件的实施、验证、测试和不确定性量化，以及为模型基准制定统一框架的潜力。研讨会的首要主题将是建模方法的统一，这些方法是理解模型用户和开发人员都感兴趣的新兴科学主题所必需的。研讨会的第二个重点将是评估社区如何更好地利用现有和未来的计算技术来提高模型性能，从而更好地解决河口和沿海海洋的多尺度动力学问题。不仅迫切需要开发高性能的计算策略来提高单个模型的性能，而且还需要实现集成方法以及开发数据同化和不确定性量化方法。

Accurate numerical models are essential for improving our understanding of fundamental physical processes at coastal and estuarine scales, where the connection between human populations and the ocean are most apparent, particularly near rapidly expanding coastal megacities around the world. Models have seen significant advances in the past two decades owing to increases in computational power and improved numerical methods including unstructured grids/model nesting, data assimilation, and model coupling. Furthermore, advances in remote-sensing and in-situ observational technologies have led to substantially larger and more accurate datasets which have significantly improved the ability to assess model performance and quantify uncertainty. However, despite recent advances, a multitude of coastal and estuarine models have been developed in the past two decades, and there has been a lack of a unified approach as seen in regional modeling. This workshop will bring together model users and developers to discuss ways to improve modeling tools for scientific research in physical oceanography at coastal and estuarine scales. The budget will provide travel and participation support to U.S.-based early-career professionals, and there will be an emphasis on attracting underrepresented minorities and women to attend the workshop. Early-career scientists will interact with experienced model developers and users in an environment that focuses on discussion and collaboration, thus helping to promote a stronger community that will continue to improve the accuracy and efficiency of coastal and estuarine models.This workshop will focus on developing an understanding of the advantages and limitations of the current state-of-the-art in numerical modeling of coastal and estuarine processes and to form a consensus on key areas of research and development needed to improve the accuracy and reliability of such models. Outcomes of the workshop will include the development of a unified understanding of the pros and cons of different approaches, such as finite-volume vs. finite-element methods, curvilinear vs. unstructured grids, couplings of ocean, atmosphere, land (hydrology) models, large-eddy simulation vs. RANS turbulence modeling, as well as of different model nesting and adaptive-grid strategies. Discussions will focus on these methods along with strategies for forcing, implementation of boundary conditions, validation, testing, and uncertainty quantification, with the potential for development of a unified framework for model benchmarking. The overarching theme of the workshop will be on the unification of modeling approaches that are needed to understand emerging science topics of interest to both model users and developers. A second focus of the workshop will be to assess how the community can better leverage existing and future computational technologies to improve model performance in better resolving multiscale dynamics in estuaries and the coastal ocean. Not only is there a pressing need for the development of high performance computational strategies for improvement of individual model performance, but also for the implementation of ensemble approaches and the development of methods for data assimilation and uncertainty quantification.