Search for a command to run...
Approaches to invert multibeam echosounder survey data for sediment properties typically rely on correlations between bathymetry, the backscattered signal, and in-situ ground-truth measurements. The difficulty with these approaches is that the seafloor roughness and volume heterogeneities which drive the acoustic scattering process are themselves impacted not just by sediment composition but also by hydrodynamics and bioturbation. As a result, there is often not a one-to-one correlation between seafloor scattering strength and the sediment type. To overcome this shortcoming and to eliminate the need for in-situ ground truth measurements, a physics-based sonar inversion algorithm has been developed which relies on high-frequency acoustic models of seafloor scattering to invert directly for geoacoustic parameters. The inversion has been evaluated through two field tests in Sequim Bay, a shallow, protected bay along the northern coast of the Olympic peninsula in Washington State. The first field test in 2019 focused on collecting data over short, 200-m-long survey tracks at the center of which extensive ground-truth data was collected which could be used to assess the inversion performance. The goal of the second field test was to collect data along survey lines which densely-sampled the northern portion of the bay where there was significant variation in sediment type. In this paper, we discuss the results of this second field test and efforts to visualize the inversion output and its associated uncertainties.