The new generation of automated acoustic data collection platforms: analysis of vessel avoidance by fish and zooplankton patch structure

Tom Evans, Hannah Blair, Rudstam, Watkins, Suresh Sethi, Kayden Nasworthy (Cornell) Peter Esselman, David Warner, Dan Yule, Tim O’Brien, Mark DuFour, and others at USGS and cooperating agencies. (Funded by USGS) 

This project uses acoustic data collected with automated surface and underwater vehicles to evaluate bias in fisheries surveys to understand the distribution of fish and zooplankton on a whole-lake scale. The work relies upon Saildrones which are uncrewed quiet surface vessels powered by wind and solar that collect data continuously for over a month and a long range autonomous underwater vehicle (LRAUV). The primary objective of the saildrone data is to evaluate fish avoidance of traditional crewed surface vessels used in fisheries surveys. In 2024 we analyzed data collected in previous years and initiated the first whole-lake survey with an acoustic drone, which was conducted in Lake Superior. Two Saildones were deployed from Ashland, WI and were remotely controlled by Tom Evans at Cornell, successfully navigating the whole-lake over 42 days. In addition to avoidance of crewed surface vessels, Hannah Blair is leading the effort to analyze upwards looking acoustic data obtained with the LRAUV to explore the importance of the upper blind zone, water close to the surface that is invisible to traditional ship based acoustics. These projects will also help answer questions about whole-lake distribution patterns of fish and plankton, and allow exploration of diel migrations, with higher temporal and spatial resolution than has been possible. In 2025 we continue working with analyses of the large amount of data collected 2021 to 2024.  We are also working closely with Dr. Angus Galloway at the University of Guelph to develop an AI that will reduce the amount of processing time needed to analyze the acoustic data.