More than 80 years ago, the use of sonar technology to detect fish underwater was demonstrated. Year after year, technological improvements were made and sonar became more frequently used in research and fisheries. With the development of calibration techniques, quantitative assessments became possible and non-invasive fisheries research could be performed. Species identification, abundance estimates, distribution patterns and fish behaviour are all part of the list of possibilities.Even though oceans and seas are interconnected, they all have their own characteristics. Many studies regarding the use of echosounder tools in fisheries research have been performed on many different locations across the Atlantic and Pacific Ocean and the North and Baltic Sea. Within the North Sea, however, the most southern part of the North Sea remains quite untouched. This sea is characterized by its high flow rates and often rough conditions. In the Belgian part of this southern North Sea (BPNS), the use of sonar is mainly focussed on bathymetry, in which seabed profiling and detection of bottom structures can be done.In the current state of global overexploitation of fish stocks, more accurate and extensive fish stock assessments are required. This is where sonar comes up as a possible tool to provide this additional insights. Many different sonar systems are available on the market and an exploration on their capabilities to detect fish in the BPNS was requested. In this thesis, three sonar tools have been studied regarding their performances by the use of the available literature, preceded by an extensive report on the general principles of sonar tools . In addition to this, water column recordings have been studied, which were collected during the demonstrations on these tools on the research vessel RV Simon Stevin. The goal was to investigate if fish can be detected and visualized and to interpret the results to formulate a conclusion which is relevant for biological fisheries research. These recordings have been processed by the use of software, which is specifically designed to process water column recordings from sonar tools. Several variables were obtained during the processing, which also were interpreted.The differences between the locations in the BPNS, which were sampled by the use of a sonar tool, immediately became clear. During every survey, different weather and sea conditions were applicable. Additional to this, the followed protocol, in terms of vessel speed during the recording and time duration of the recording, was not equal for all water column recordings. All the surveys were also performed during different seasons and different times of the day. This created severe implications for the further post-processing of these data. Nonetheless, the differences in detection capabilities between the two demonstrated sonar tools were examined and this difference was not proven by the analysis.Because of the wide variety in working methods in the post-processing software packages, these differences have been elaborately discussed as well, to provide certain guidelines for continuation of this work. Because of this, during the post-processing, sources of noise in the data were further examined. This which showed that turbulence appears to account for a substantial part of the total amount of detected targets. This was the case for both sonar tools. These turbulence zones were shown to be present in both surface layers and the layers above the seabed. The difficult settings in the post-processing software package and limited previous knowledge, made that the detected targets could not be confirmed to be fish purely.As a last part, the conclusion was made on the explorative study on the three compared sonar tools, in which is stated that the IXBlue SeaPix 3D sonar appears to be the most promising tool to use for fish detection. However, these findings are not assisted by data results, as no demonstration with this tool was done on the research vessel RV Simon Stevin. |
