Closing the marine plastic mass budget

Plastic waste is causing a high amount of pollution in the ocean. A lot of plastic pollution remains afloat, and accumulates in the subtropical gyres, in the so called `floating garbage patches’. It is not well known where most of the marine plastic pollution actually comes from: is it for example originating from rivers, coastlines, or from fishing activity? Furthermore, it is relatively unknown which processes `remove’ plastic pollution from the surface of the ocean over time. A part of the plastic pollution ends up on the coastlines, starts to sink down after some time (e.g. under the influence of biofouling), or fragments into smaller particles. Current estimates of the amount of plastic pollution that is floating in the ocean are much lower than the estimated amount of plastic pollution entering the ocean every year. This thesis is written within the TOPIOS project (Tracking Of Plastics In Our Seas), where we attempt to resolve this discrepancy. In order to estimate the marine plastic mass budget, we combine numerical simulations of the ocean with observational measurements of plastic concentrations. One way to do this is by using statistical models. We use a regression method (random forests) to estimate the total amount of litter on the Dutch North Sea coastline. Using a beach cleanup dataset from The North Sea Foundation, we train a model to predict the amount of beach litter as a function of various environmental variables (e.g. conditions related to the weather, ocean currents, tides). This model is used to extrapolate the amount of litter along the entire Dutch North Sea coastline (about 16,500-31,200 kilograms in total). Observational data of plastic concentrations can also be used to calibrate unknown parameters in numerical models of the ocean currents. We do one such analysis for the Mediterranean Sea, and a more extended analysis for plastic pollution on a global scale. One important conclusion is that the total amount of plastic pollution in the global ocean is likely much higher than previously expected: roughly 3,000,000 tonnes (3 billion kilograms). Previous research, with estimated quantities of roughly 250,000 tonnes, mainly used net measurements to calibrate the numerical models. These kind of net measurements are likely to miss larger objects, which contribute to most of the marine plastic mass. In our analysis we additionally use data of larger plastic objects from visual observations (both from the open ocean and from beaches), and net measurements from the water column for the model calibration. We estimate that a relatively large part of marine plastics sinks down over time (for example due to growth of algae). Fragmentation of plastics plays an important role: most of the plastic items found in the ocean are small fragments originating from larger objects. We estimate that beaches contain about 2% of the global marine plastic mass. We estimate that the input of plastic pollution is much lower than previously expected, and that residence times of plastic items in the marine environment are higher than previously anticipated.