Progress 09/01/01 to 10/31/04
Outputs
The results of the Phase II research and prototype development activities demonstrated that the low frequency acoustic method is a viable technique in inferring fish biomass information in a high-density cultured environment. This method complemented with a multi-transducer acoustic fish sizing system provides the necessary tools for aquaculture farmers to manage their inventory and reduce their production costs. The basic concept behind the low frequency acoustic fish biomass system is to monitor the shift in the fundamental acoustic modes of the water column of the fish tank/pond as fish were added to the medium. The linear correlation between the shift in the frequency and the fish density is the defining relationship that enables us to use low frequency acoustic techniques to obtain fish inventory information in high density cultured environments where other techniques had failed. Additionally, the correlation between the fish length and the high-frequency
scattered duration enables us to obtain fish size distribution in real time in any cultured settings. A theoretical model and experimental verification using fish and balloons provided the technical basis for developing low frequency acoustic prototype systems. Several desktop-PC based prototype systems equipped with National Instrument (NI) A/D boards were developed. Extensive LabView computer codes were developed to control, acquire, analyze the data and display useful information on the computer screens. The integrated electronic units consist of eight-channel filters/amplifiers (up to eight inputs), electric impedance units, and the A/D boxes. Each integrated electronic unit has enough space to house the fish sizing unit. Mackie 1400 power amplifiers were selected to power the acoustic transducers. One of the systems is currently installed in a fish tank at Kent Sea Tech farm and is used to obtain fish biomass information on a periodic basis. As part of the development phase, an
extensive research was conducted in order to select a proper acoustic transducer that is economical, provides the necessary power, and functions in the frequency range of interest. Two transducers appropriate for tanks and ponds were selected. The fish sizing portion of the system consist of transducers with 450 kHz peak frequency. Separate LabView codes were developed for this portion that interacts with the main programs. The functionality and robustness of these prototype systems were examined by conducting several controlled experiments in fish tanks/ponds and a model fish pen. The results obtained from these experiments reveal that proper usage of the low-frequency technique results in an accurate estimation of fish inventory in tanks. We have obtained limited information from fish ponds that also suggest this technique can be applied to fish pond environment. The data obtained from model sea pen, however, did not provide conclusive evidence on the applicability of this
technique. The changes in the acoustical characteristics of the model fish pen suggest that a modified version of this technique may work for sea pens. We are currently exploring this possibility.
Impacts
The rapid increase in aquaculture activities has generated a great need for methods of obtaining accurate inventories of fish crops and quantifying the effects of production practices on fish behavior. The classical hydroacoustic methods of counting fish use a simplistic model of sound scattering from a school/swarm of fish. This pulse-echo technique is based on classical sonar technology and is applicable only if the population density of fish is low (few fish per m3). To be able to overcome these known deficiencies in conventional hydroacoustic techniques for better management of fish crops, a novel low frequency acoustic technique to obtain both fish size and biomass in intensive aquaculture environments is developed. This low-frequency approach utilizes a system consisting of underwater acoustic transducers, hydrophones, and an electronic unit, which broadcasts sound waves in a frequency range of interest to the acoustic transducers. The hydrophones receive the
acoustical characteristics of the basin and searches for its fundamental mode of the water column containing the fish. The changes in this mode from its base line (no fish) indicate the fish density information. We have demonstrated that the low frequency acoustic method complemented with a dual-transducer acoustic fish sizing system is a viable technique in inferring fish biomass information in a high-density cultured fish environment.
Publications
- Ali R. Kolaini, 2005, "Acoustical characteristics of water basins containing fish," To be submitted to the Journal of Acoustical Society of America.
|
Progress 10/01/01 to 09/30/02
Outputs
The low-frequency acoustic system under development consists of an underwater acoustic transducer, a hydrophone(s), and an electronic unit that broadcasts sound waves in the frequency range of interest to the transducer and transfers the data from the output of the hydrophone(s) to a PC computer. User-friendly software has been developed that controls the device and acquires signals from the output of the hydrophone(s), analyzes the data, and displays the results in the form of fish density, fish size, and other useful parameters. The electronic hardware of the device also performs proper amplification and filtration of the acoustic signals. To achieve the fish size and biomass using this system the following steps are performed: 1. The electronic controller unit sends a white noise or swept sine waves to the underwater transducer in the frequency range of interest. 2. The hydrophone(s) picks up the sound and the water basin's response and down loads the data into the
computer via the electronic unit. 3. The developed algorithm communicates with the controller and receives the output of the hydrophone and performs the following steps: a) Calculates the average sound spectrum in the frequency domain. b) Identifies the cut-off frequency of the medium from the spectrum. A special routine is developed to search for the peaks in the spectrum. c) Compares the cut-off frequencies of the pond/tank with fish and without fish. The power spectrum of the empty pond/tank (no fish or minimum number of fish) is obtained every time fish are harvested from the tank/pond and is stored in the computer's memory to be used as a basis for the comparison with the cut-off frequency of the basin with fish in it. d) Finds the resonance frequency of the bladders using the acoustic spectra. e) Calculates the fish size and length from the resonance frequency of the fish. f) Measures (estimates or predicts) the size distribution of fish. g) From the size distributions and
cut-off frequencies, the computer program performs an inverse problem to find the fish biomass. The developed model will be used to achieve this. 4. Displays the information in the form of graphs on the PC monitor. 5. Steps 1-4 are repeated periodically to provide up-to-date information. The system has been tested in three farms (tilapia farm with indoor tanks, Stripped Bass farm with out door tanks, and catfish ponds). The algorithm to control the system, to acquire data and analyze data, and to display information on a monitor is completed. We are currently combining developed breadboard system into a compact robust system.
Impacts
The acoustic system has wide spread applicability to tanks, ponds and cages. This technique has the large advantage of making estimate of fish biomass based upon the entire culture system and while the technique may estimate high or low, it will be predicting based upon a measurement of the entire fish population and thus errors are relative to the whole and not a selective sample. This is of particular interest to cage operators where a tear or break in the cage can lead to an eventual loss of the entire population before the rupture is diagnosed. The invention's estimates of fish size and distribution are important also as a management tool to allow proper feeding. Fish are fed according to biomass and over or under-estimates of fish biomass and size distribution will cause improper non-optimum feeding procedures. This then results in direct economic loss when over feeding and associated drops in water quality or environmental quality and if underfeeding is
occurring, fish quality is compromised, i.e. there is no premium market for skinny fish! The economics of this device relate directly to the above discussion. Savings can be on the order of a few percent for optimized feeding protocols to large savings by preventing catastrophic fish losses due to escape or predation. Given the expected costs of the technology, one would expect one or two year paybacks for the investment made. Also, accurate monitoring of fish populations may identify the presence of a predator that could be eliminated.
Publications
- We are currently preparing two publications on this topic: Ali R. Kolaini, "Acoustical modes of tanks, ponds, and sea pens," to be submitted to the Journal of Acoustical Society of America, 2003. Ali R. Kolaini and Douglas Minchew, "Low-frequency acoustic techniques to obtain fish biomass," to be submitted to the Journal of Aquaculture, 2003.
|
|