High-throughput information processing in fisheries and plankton acoustics

Active acoustic data collected by echosounding instruments are widely used to infer the spatiotemporal distributions and behaviors of fish and zooplankton in the ocean. The broad deployment of these instruments on ocean observing platforms has led to a rapid increase in water column sonar data over the past decade. These data hold great potential to link oceanographic and climate drivers to ecosystem responses at the mid- to high-trophic levels that directly support global food systems. In this presentation, we highlight our group′s efforts to accelerate information processing of active acoustic data by tackling the challenges of analysis scalability, consistency, reproducibility, and interoperability. Taking a multifaceted approach, we develop data-driven methods to extract echo information on commercially targeted species and ecologically relevant spatiotemporal patterns, build open-source software and workflows to leverage the ever-growing local and cloud computing resources, collaborate closely with survey missions to support operational priorities, and create and host open educational resources and workshops to support the broader community. We believe these efforts are essential for achieving high-throughput active acoustic information processing and enabling community-wide collaboration, which are key to distill both scientific knowledge and actionable insights to help us better understand and adapt to rapidly changing marine ecosystems.