The performance of 76 kHz positional acoustic telemetry is challenged by acoustic conditions in the tailrace of a hydroelectric dam.

Background: Positional acoustic telemetry permits high-resolution (2D and 3D) tracking of tagged fish based on the speed of sound and time at which acoustic signals are detected at independent receivers. Such positional data has important implications for understanding fish behavior in hydropower tailraces, where movement and fishway passage may be influenced by complex hydraulic conditions. However, environmental features common to these settings-such as rocks, vegetation, bubbles, and suspended particles-can reflect, scatter, or absorb acoustic signals, contributing to spatial and temporal variation in positioning error. Therefore, validating system performance in tailrace environments is important for study design and interpretation. We evaluated the performance of a 76 kHz positional acoustic telemetry system (LOTEK MAP) in the tailrace (0-500 m downstream) of a large (672 MW generating capacity) hydroelectric dam, an environment characterized by reflective surfaces, complex hydraulics, and ambient noise from power generation.

Methods: We quantified positional accuracy (i.e., Euclidean distance from measured deployment positions) and position efficiency (i.e., position estimates/total transmissions) for six stationary transmitters centered within two hydrophone arrays installed 0-150 m and 200-500 m downstream from the power station. Further, we used generalized linear models to explore how accuracy and position efficiency varied as a function of transmitter location, power station discharge (which represented variation in the acoustic environment at each hydrophone), and the number of receivers that contributed to position estimates.

Results: The array nearest the power station (i.e., 0-150 m downstream) yielded no positional data. Performance improved marginally for the second array (i.e., 200-500 m downstream), but position efficiency was ultimately low (1%), variable across transmitter locations, and negatively correlated with discharge. When position estimates were obtained, positional accuracy was moderate (overall mean = 3 ± 3 m) and increased with the number of contributing receivers, but also varied in direction and magnitude as a function of transmitter location and discharge conditions.

Conclusions: Performance of the telemetry system was limited by environmental conditions that influenced signal propagation at complex spatial and temporal scales and would ultimately challenge the interpretation of positional data for free-swimming fish in the tailrace. The telemetry system is likely not suitable for studies that require frequent and reliable positions in hydropower dam tailrace environments.

Supplementary information: The online version contains supplementary material available at 10.1186/s40317-025-00420-6.