From coastlines to the deep sea

Plastic is one of the most commonly used materials in the world today, with polymers such as polypropylene and polyethylene used for applications such as packaging, textiles and commercial fishery. However, not all plastic objects are properly disposed of at the end of their useful lives, and today plastic pollution is ubiquitous throughout the environment and particularly in the ocean. Here plastic pollution can cause harm in a variety of ways, such as entangling marine wildlife, leaching chemical compounds into the water and causing economic damage by deterring tourism at commercial beaches. However, it is difficult to understand the full scope of these threats, in part because the fate of plastic once it enters the ocean is poorly understood. Plastic debris has been found everywhere from on coastlines to the open ocean, and from the ocean surface down to the deep ocean on the seafloor, but it is not always clear what physical processes contribute to the observed distribution of plastic objects. Numerical models can provide insight into plastic debris transport by modeling various transport scenarios, but physical processes such as plastic beaching and resuspension, vertical transport and fragmentation are not always completely represented or even included at all. This thesis investigates the transport of plastic debris in the global ocean by means of Lagrangian particle transport scenarios, where plastic debris is represented by virtual particles. Using circulation and other oceanographic data from oceanic general circulation model (OGCM) reanalysis products, particles trajectories are calculated that provide insight into the distribution and pathways of plastic debris in marine environments. By modifying the model setup in the various scenarios, the influence of di↵erent physical processes on plastic transport is investigated. Chapter 1 provides a general overview of marine plastic pollution, where this is broadly split into insights gained from the observational record and from modeling studies. The current understanding of the relative distribution of plastic debris on a global scale is described, as well as the current knowledge of how this is influenced by various physical processes. This also highlights current knowledge gaps such as the role of coastal and vertical transport processes, which are investigated in later chapters of this thesis. While the specific model frameworks are described in each subsequent chapter in this thesis, chapter 2 provides a general overview of Lagrangian ocean modeling and particularly the Parcels modeling framework. In addition, while all the details of OGCMs and ocean reanalysis products are beyond the scope of this thesis, a general overview is given of various OGCM features that are relevant to the work described in the following chapters. Chapter 3 investigates plastic debris beaching and resuspension on a global scale. The spatial and temporal resolutions of OGCMs are insufficient to resolve the physical processes that contribute to debris beaching and resuspension, and stochastic parametrizations are introduced to represent plastic beaching and resuspension within a large-scale modeling framework. Coastlines and coastal waters are globally shown to hold at least 77% of all positively buoyant plastic debris, with the spatial distribution of beached plastic being strongly influenced by the model input scenario. As such, coastal dynamics are shown to play a more prominent role in global-scale plastic transport than previously thought. Chapter 4 describes various parametrizations for modelling the wind-driven vertical turbulent mixing of buoyant particles within the surface mixed layer. Ocean reanalysis products generally do not provide turbulence data fields, but turbulent vertical transport is an important driving process in the full three-dimensional distribution of plastic in the ocean. The modeled vertical microplastic concentration profiles correspond reasonably well with field observations, and the parametrizations are numerically stable with an integration timestep #t = 30 seconds. This makes it computationally feasible to apply the paramatrizations in large-scale three-dimensional modeling frameworks. Chapter 5 examines the influence of particle size on the three-dimensional transport of microplastic debris in the Mediterranean Sea. The distribution of plastic in beached, coastal waters and open waters reservoirs is strongly a↵ected by the particle size, with smaller particles being more likely to reach open water. Smaller particles are also mixed farther below the ocean surface up to depths of 3000 m. Fragmentation is shown to be a slow process over timescales of years to decades, with ocean-based fragmentation likely being negligible compared with beach-based fragmentation processes. Therefore, while fragmentation was not shown to strongly influence the particle size distribution over the course of 3 years, over longer timescales it can play an important in the gradual plastic mass transfer to o↵shore and subsurface waters. Finally, chapter 6 provides a general overview and discussion of the main results described in chapters 3 - 5, and outlines future possible research directions.