Premium
Development and Evaluation of an Acoustic Device to Estimate Size Distribution of Channel Catfish in Commercial Ponds
Author(s) -
Goodwiller Bradley T.,
Beecham Rachel V.,
Heffington J. D.,
Chambers James P.
Publication year - 2015
Publication title -
north american journal of aquaculture
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.432
H-Index - 41
eISSN - 1548-8454
pISSN - 1522-2055
DOI - 10.1080/15222055.2014.956197
Subject(s) - catfish , ictalurus , transducer , underwater , calibration , channel (broadcasting) , fishery , aquaculture , environmental science , acoustics , fish <actinopterygii> , marine engineering , hydrology (agriculture) , biology , mathematics , geology , engineering , statistics , oceanography , physics , geotechnical engineering , telecommunications
As one step in the continued effort to utilize acoustic methods and techniques to the betterment of catfish aquaculture, an acoustic “catfish sizer” was designed to determine the size distribution of Channel Catfish Ictalurus punctatus in commercial ponds. The catfish sizer employed a custom‐built 460 kHz piezoelectric transducer with a 20° beam angle. The sizer was built at the National Center for Physical Acoustics and tested in commercial catfish ponds in Mississippi and Arkansas in June and October. To test the system, fish were collected in seven different ponds using a cut seine. They were weighed and then sent through a pipe into the open pond. The transducer was mounted underwater at the end of the pipe. As the individual fish exited the pipe, they passed through the acoustic field created by the transducer and the echo detected by the transducer was recorded. In three of the seven ponds, calibration tests were run. For these calibration tests, the same process was used with the added step that a jet of bubbles was introduced in front of the transducer after every fifth fish. Analysis of the measured fish weights and acoustic data were used to produce a Gaussian prediction model. This model was applied to the data from the blind tests to predict the average weight and the SD of fish in each pond. The device and methodology worked well, with significant differences between the measured and calculated weights found in only one of the seven ponds tested.