Influence of turbidity and passage rate on the efficiency of an infrared counter to enumerate and measure riverine fish.
The ability to observe fish in their natural habitat is very difficult. Researchers seeking to obtain information on fish behaviour are generally limited to capture-dependent studies, where fish are physically-handled. Handling fish may impacts on their natural behaviour, but it is unsure to what degree this occurs. Researchers overcome this issue by using electronic fish counting technology such as sonar or underwater cameras, but many techniques are yet to be assessed in Australia.
Infrared fish counters are now commonly used to assess fish passage efficiency within fishways. This technology relies on fish passing through an infrared beam, and transmitting information back to a datalogger or computer. There is no need to handle the fish, which is advantageous, but red light is unable to penetrate very far into water. The systems can therefore be limited by high turbidity and increased fish passage rates. The impact of these on optimal system performance is unclear. Trials to determine the influence of these variables on fish count accuracy and size determination were undertaken under controlled laboratory conditions.
Silver perch (Bidyanus bidyanus) were passed through the infrared counter under turbidity values ranging between 0 and 100 Nephelometric Turbidity Units (NTU). The effect of turbidity was insignificant when compared to the inability of the infrared counted to deal with multiple fish. There were no significant differences in the accuracy of fish counts whether passing single or multiple fish through the unit. This suggested that the unit was often underestimating fish numbers. When multiple fish were simultaneously passed through the counter, the software detected them as a single fish and overestimated fish length. These issues may be overcome with further software development to cater for slow moving fish.