Author(s): Joel R. Wittmann; Levi Graf; Vittorio Ferrero; Lea Kuster; Anais Wigger; Giulia Zambotto; Frederic De Schaetzen; Sandro Giacomuzzi; Roman Stocker; Luiz G. M. Silva

Linked Author(s): Luiz G Martins da Silva

Keywords: Fish larvae; Macroinvertebrates; Hydropeaking; Artificial intelligence

Abstract: Drift net studies are essential to understand the dispersal of early life-stages of fish and macroinvertebrates in rivers, but they are limited by a trade-off between temporal resolution and sampling effort. The Riverine Organism Drift Imager (RODI) overcomes this limitation by continuously recording drifting organisms, effectively decoupling the temporal resolution from the sampling effort. In this study, nine RODIs were deployed in a hydropeaking Swiss alpine River for 10 weeks to evaluate their technical reliability and demonstrate the detection of drifting organisms with a temporal resolution of one second. The RODIs recorded 77.6% of the deployment period. Sixteen classes of drifting objects were identified, including fish and insects, with a strong class imbalance. Ongoing work is refining and automating the classification process. Significantly more brown trout fry (Salmo trutta) drifted at night during a 72-hour case study, illustrating how high-resolution data obtained with RODI can improve our understanding of drift.

DOI: https://doi.org/10.64697/iahr.proc.ise2026.abs.179

Year: 2026