1,2,3,1,5,1- 1University of Naples Federico II, Department of Civil Engineering and Environmental (DICEA), Napoli, Italy (khimcathleen.saddi@unina.it)
- 2Consiglio di Classe di Scienze, Tecnologie e Società, Scuola Universitaria Superiore IUSS Pavia, Pavia, Italy
- 3Department of Civil Engineering and Architecture, Ateneo de Naga University, Naga, Philippines
- 4UNESCO Chair in Water Resources Management and Culture, University for Foreigners of Perugia, Perugia, Italy
- 5Department for Innovation in Biological, Agro-food and Forest systems, University of Tuscia, Viterbo, Italy
- 6Dipartimento di Informatica - Scienza e Ingegneria, Università di Bologna, Italy
Rivers are key conduits of plastic debris from land to sea, with transport often amplified during rainfall-driven high-flow events While large rivers can sometimes be monitored from space, narrow rivers and small basins—such as the Sarno River—require high-resolution, in situ approaches capable of resolving rapid, event-scale dynamics.
This study consists of two main parts: 1) long-term, fixed-site monitoring, and 2) dense citizen-science observations to characterize river plastic transport under contrasting flow conditions. A low-cost RGB camera system was installed at a representative Sarno cross-section and operated continuously for one year (November 2023–October 2024), acquiring time-lapse imagery at 15 s intervals. Hydrometeorological forcing was reconstructed using ERA5-Land precipitation (hourly to monthly products) over the upstream and downstream portions of the basin to identify and contextualize low- and high-flow periods. Plastic items were detected and counted from the imagery using a YOLO-based model trained with a self-/weakly supervised strategy, implemented in two configurations: a single-class detector (plastic vs. background) and a multi-class detector (13 classes) to better differentiate plastic categories and support source/process interpretation. Analyses were performed on dates with complete camera data, spanning both low-flow conditions and rainfall-driven events.
To complement the fixed-site record and capture network-scale variability during anticipated high-rainfall events, we deployed the RiverWatch app to collect geotagged images of plastic presence across multiple locations along the Sarno river network. By coupling continuous, cross-section-scale detection with event-focused, spatially distributed citizen observations, this work demonstrates a scalable pathway to quantify plastic transport dynamics in small rivers and to support monitoring strategies that are inclusive, low-cost, and transferable.
Keywords: riverine plastic, image-based monitoring, YOLO, long-term monitoring, citizen science, Sarno River
How to cite: Saddi, K. C., Miglino, D., Moe, A. C., Proietti, G., Biscarini, C., Tauro, F., Poggi, M., and Manfreda, S.: Revealing River Plastic Transport under different flow conditions: Long-term camera monitoring and Citizen Science in the Sarno River (Southern Italy), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18438, https://doi.org/10.5194/egusphere-egu26-18438, 2026.
