Summary

Observing fish in their natural environment is usually difficult, especially in turbid water or at night. Despite many recent advances in technology, traditional sampling methods generally require scientists to interact with the fish in some way to obtain biological information. Disturbing or handling fish may alter their behaviour however, so it is difficult to know how ‘natural’ such observations are.

Recently developed sonar systems are currently being assessed in North America and their non-invasive application to fish migration studies is extremely promising. One such device, the Dual-Frequency Identification Sonar (DIDSON), uses sound-distorting lenses to create high quality video images. When operating in high frequency mode, the DIDSON creates images from sound beams that can show the outline, shape and even the fin details of target fish. In addition, the unit’s software can count and measure fish automatically. With such features, this technology should allow the direct observation of fish behaviours such as spawning, feeding and migration even in very low visibility conditions. No assessment of this technology for fisheries-based applications has been undertaken outside North America and our study provides the first assessment of a DIDSON unit in Australia.

The results show that the DIDSON is a powerful tool for observing freshwater fish. When used with traditional trapping equipment, the DIDSON consistently provided additional data on fish behaviour that could not be otherwise obtained. For example, in fishways along the Murray River, the DIDSON demonstrated that many more fish were approaching and entering the fishways than were actually trapped. In many cases, these fish were actively avoiding traps. The DIDSON also provided useful observations of non-migratory activity by fish and previously unknown information about other aquatic animals such as freshwater shrimps. Predatory birds and fish were observed to use fishways primarily to hunt prey.

The study demonstrated that DIDSON has enormous potential to improve our scientific understanding of fish migrations and behaviour in Australian aquatic systems. The DIDSON had several characteristics that could enhance monitoring programs, including directly observing behaviour, species identification, counting, recording fish length, estimating sampling gear efficiency and permitting direct observation of fish in turbid conditions or at night. It is recommended that the capabilities of this technology should be developed to enhance existing research programs and to provide a new sampling tool for fisheries scientists.