Uncertainties in Visual Observations of Floating Riverine Plastic.

Accurate and reliable monitoring data are crucial for the design of effective plastic pollution reduction and mitigation strategies. One common approach to monitor macroplastic (>0.5 cm) in rivers is the visual observation method, where floating plastics are counted from bridges to estimate plastic flux. However, this method lacks robust uncertainty analyses, resulting in unknown error margins and potentially suboptimal monitoring strategies. The goal of this study was to quantify these uncertainties. Three key design elements that contribute to uncertainty include (1) cross-sectional coverage, (2) observation time, and (3) observation frequency. Through a case study on the Dutch Rhine-Meuse delta, we show how these uncertainties can be quantified and that they can be used to make informed monitoring design decisions. We further demonstrate that the detection rate of true flux (recovery rate) is a key parameter to consider during uncertainty analyses. By integrating an uncertainty optimization step into the design process, the efficiency and effectiveness of monitoring protocols can be improved. These insights enhance data quality and reliability, ultimately supporting efforts to mitigate the environmental impacts of macroplastic pollution.