Fish on net: Acoustic Doppler telemetry and remote monitoring of individual fish in aquaculture

Abstract

The two main contributions of this thesis are the Internet of Fish (IoF) concept and a novel fish swimming speed measurement principle. The IoF concept is a reliable communication protocol which could relay acoustic telemetry data over long distances at very low power consumption in real-time. The speed computation algorithm provides a novel and robust approach for measuring instantaneous swimming speed of individual fish by using Doppler analysis. The methods developed in this study were tested in commercial scale marine farms for Atlantic salmon (Salmo salar L.) production, however they could also be applied for other species farmed in marine environment and even in scientific studies of wild fish. Norway is the world’s largest producer of farmed Atlantic salmon and a global leader in marine farming. An important goal for the Norwegian farming industry is to have sustained growth with an improved fish welfare and environmental footprint. This could be achieved via novel technological solutions such as the Precision Fish Farming (PFF) concept. Whereas technology is innovating different aspects of farm management operations, monitoring fish underwater poses unique challenges due to lack of direct observations. This is further exacerbated by the recently growing number of more exposed farming sites. Acoustic biotelemetry has been reliably used for individual fish monitoring in the underwater environment. Basic building blocks of an acoustic telemetry system are a transmitter tag and one or more matched receivers for receiving and decoding telemetry data sent by the tag. Commercially available telemetry receivers are normally logging receivers and provide no real-time support to the telemetry data. Cabled and existing wireless or cellular protocols are often used to address the problem of real-time support. However, such solutions suffer from the issues of limited coverage area and offer poor energy efficiency, respectively. This was addressed by establishing the IoF concept in this study. The IoF provides long range, low power real-time support to the telemetry receivers. The IoF concept was realised by developing a dedicated surface communication module and was also extended for real-time fish positioning. A Quality of Service (QoS) of more than 90% proved the IoF concept as a reliable communication protocol. Fish swimming is an important indicator of fish behaviour, growth and energy expenditure. It becomes more relevant for assessing fish welfare at exposed farming sites where fish might face strong currents. Currently, there exists no solution for quantifying swimming speed of individual free-ranging fish. A novel method for measuring free-ranging individual fish swimming speed using Doppler analysis was developed and demonstrated in a commercial scale fish farm. The method is elegant in the sense that the speed measurement can be piggybacked onto the existing Pulse Position Modulation (PPM) signal sent by a tag. In essence, this means that the new speed measurement data value could be extracted from the existing acoustic carrier wave without significantly modifying the telemetry system. Although requiring significant signal processing capacity in the acoustic receiver, it remains much easier to expand a receiver with additional resources with respect to computational capacity and energy. The proposed speed measurement algorithm was tested via a series of experiments ranging from emulated motions in a lab to a marine farm with fish tagged with acoustic transmitter inside a fully stocked commercial sea cage. A relative rms error of less than 10% of the overall speed range was achieved in all the experimental stages, affirming that the proposed method is promising and could be used for in-situ swimming speed measurement of an individual free-ranging fish.