Clustering of Floating Tracers in the Ocean

该项目的主要动机是海洋塑料污染以及了解和减轻污染的必要性。海洋环流的特点是复杂的中尺度和亚中尺度涡旋运动（即涡旋），其长度尺度从几公里到数百公里，时间尺度从几天到几个月，这些现象为涉及许多基础和实际方面的活跃研究范围提供了依据。这些漩涡影响了漂浮在海洋表面的许多示踪剂和物质：塑料污染（包括垃圾和微塑料）、石油泄漏、马尾藻、浮游生物、冰和残骸碎片。我们主要关注的是塑料污染，这是最大的环境威胁之一。由于大规模的海洋环流、漩涡和较小尺度的流动特征，漂浮物质以一种复杂而鲜为人知的方式不断移动、重新分布和聚集。这里，聚类指的是材料属性在结构和时间演化的空间格局中积累和聚集的过程。这个过程实际上很重要，因为不同的浮动和中性浮动示踪剂，包括营养物和生物量，聚集方式不同，这对它们的相互作用有深远的影响；例如，污染集群与海洋生物之间的正相关关系增强了对后者的不利影响。在这个项目中，我们的目标是了解集群的一些基础知识。准确预测漂浮示踪剂的分布和传输途径有许多实际需要，但也涉及许多理论方面，这些方面构成了该项目的研究设计和目标。这些方面可以归纳为以下几点：(1)大多数海洋塑料和石油是漂浮的，而且塑料也是由有限大小的颗粒组成的——这导致这些示踪剂具有显著的非被动和非拉格朗日性质。因此，他们在被动示踪剂浓度或拉格朗日粒子集合方面的常见近似值具有系统误差和偏差。该项目将估计后者，并探索浮动示踪剂更准确的近似值。(2)示踪剂的时空变化主要是由中尺度和亚中尺度涡旋场驱动的，这些涡旋场在大气环流模式和野外观测中通常是未解析或欠解析的。因此，由于缺乏（本身昂贵的）原位观测，以及在常规环流模式中对中尺度和亚中尺度涡旋进行强力解析所需的巨大计算负担，阻碍了对浮动示踪剂预测理解的进展。正如项目中概述的那样，这将对所涉及的机制和示踪剂行为的理论理解置于研究前沿。(3)聚类过程在很大程度上是由多尺度和微弱的地表速度辐散引起的，由于它是由不同的物理过程引起和控制的，这些物理过程的贡献鲜为人知，相互反馈复杂，因此很难观测和建模。以一种系统的方式，该项目旨在整理出底层物理过程的群集贡献。(4)就第(1)至(3)项而言，本计划亦旨在发展有效的聚类分析方法，以及实际应用的聚类数学模型。预计该项目的理论结果将对现有和即将开展的观测项目以及许多研究海洋污染和海洋物质运输的研究人员有用。已经考虑了与相关工作组和任务的联系，并获得了协调全球海洋塑料污染研究的埃里克·范·塞比尔的支持。与EPSRC的战略主题和研究领域保持一致：流体动力学，数学物理，非线性系统，生活与环境变化。

The Project is mostly motivated by marine plastic pollution and the need to understand and mitigate it. Ocean circulation is characterized by complicated mesoscale and submesoscale eddying motions (i.e., eddies) acting on length scales from a few to hundreds of kilometres, and on timescales from days to many months - these phenomena feed an active research scope involving many fundamental and practical aspects. The eddies affect many tracers and substances floating on the ocean surface: plastic pollution (including litter and microplastic), oil spills, sargassum, plankton, ice and wreckage debris. Our main focus is on the plastic pollution, which is one of the largest environmental threats. In a complicated and poorly understood way, the floating materials are constantly moved around, redistributed and clustered by the large-scale ocean circulation, eddies and smaller-scale flow features. Here, by clustering we refer to the process of accumulation and aggregation of material properties in structured and temporally evolving spatial patterns. This process is practically important, because different floating and neutrally buoyant tracers, including nutrients and biomass, cluster differently - this has profound effect on their interactions; e.g., positive correlations between clusters of pollution and marine life have enhanced adverse effect on the latter. In the Project we aim to understand some fundamentals of the clustering. There are many practical needs for accurate predictions of distributions and transport pathways of the floating tracers, but there are also many theoretical aspects involved, which shape up the research design and aims of the Project. These aspects can be grouped as the following: (1) Most of the marine plastic and oil is floating, and the plastic is also made up of finite-size particles - this results in significant non-passive and non-Lagrangian properties of these tracers. As a result, their common approximations in terms of passive-tracer concentrations or ensembles of Lagrangian particles have systematic errors and biases. The Project will estimate the latter and explore more accurate approximations for floating tracers. (2) Most of the spatio-temporal variability of the tracers is driven by the mesoscale and sub-mesoscale eddy fields, which are routinely either unresolved or underresolved in general circulation models and field observations. Thus, progress in predictive understanding of floating tracers is hindered by the lack of (inherently expensive) in situ observations and by the enormous computational burden required for brute-force resolution of mesoscale and submesoscale eddies in operating general circulation models. This puts theoretical understanding of the involved mechanisms and tracer behaviours at the research frontline, as outlined in the Project. (3) To a large degree the clustering process is induced by the multiscale and weak, surface velocity divergence, that is difficult to observe and model, because it is induced and controlled by different physical processes with poorly understood contributions and complicated mutual feedbacks. In a systematic way the Project aims to sort out clustering contributions of the underlying physical processes. (4) In connection with (1)-(3), the Project also aims to develop efficient methodologies for clustering analyses, as well as mathematical models of clustering for practical applications. It is expected that theoretical results of the Project will be useful for the existing and upcoming observational programs, as well as for many researchers dealing with marine pollution and material transport in the ocean. Links to relevant working groups and missions have been thought out, and support from Erik van Sebille, who coordinates the global research on marine plastic pollution, has been secured. Alignment to EPSRC's strategic theme and research areas: fluid dynamics, mathematical physics, nonlinear systems, living with environmental change.