Evaluating the Usage of Multi-frequency Backscatter Data as an Additional Tool for Seafloor Characterization
|Title||Evaluating the Usage of Multi-frequency Backscatter Data as an Additional Tool for Seafloor Characterization|
|Degree and Program||Master of Science|
|Degree||Earth Sciences/Ocean Mapping|
|Number of Pages||161|
|University||University of New Hampshire|
|Keywords||EM 2040; Frequency dependency; Multibeam backscatter; Multifrequency backscatter; Seafloor characterization; Sonar equation|
A reliable understanding regarding the seafloor characteristics can have innumerable application in a variety of fields of knowledge, such as ocean mapping and military fields. In the last decades, studies associating backscatter intensity to seafloor characterization have increased, based on the principle that different types of seabed may provide, a priori, different reflectivity responses patterns. Those differences in intensity levels can be used to attempt seafloor classification.
This thesis proposes to evaluate the potential usage of multi-frequency backscatter as an additional tool for seafloor characterization. Modern multibeam systems can provide high-resolution bathymetry and backscatter data. The echosounder used to collect data for this research was a Kongsberg EM 2040, which can operate using three different center frequencies (200, 300 and 400 kHz) at a time. The backscatter data, which was collected using all three available frequencies, was investigated under two different perspectives. The first one consists of interpreting how backscattering strength curves may vary when the same frequency is used to ensonify distinct types of substrates. This approach can be used to establish a connection between acoustic wavelength and intensity levels, and the result is supposed to be useful in seafloor characterization. The second perspective consists of verifying the existence of any frequency dependency, such as seafloor roughness or sediment volume contribution, when the same type of seabed is ensonified with different frequencies.
Beyond that, some of the corrections that had been applied to the raw data during the data acquisition process were compared to post-processing models, which, a priori, might be more accurate, in order to evaluate, under some circumstances, if approximations made by the acquisition software could impact the final result.