Model-Based Sediment Characterization Using Multibeam Angular Backscatter Data

This paper discusses a model-based approach to sediment characterization using multibeam backscatter data collected offshore from Qingdao, China in 2018. In this experiment, the NORBIT wideband multibeam sonar was utilized to collect backscatter data, the angular response curves of which reflect sediment-type differences. For the synthetic analysis of experimental data, a high-frequency backscatter model based on small-slope approximation and volume-scattering theory was adapted to simulate the variation of the backscatter strength with the incident angle. Afterward, these generated model replicas were matched with experimental backscatter data using a simulated annealing optimization algorithm to estimate sediment parameters, such as sediment–water sound-speed ratio, density ratio, roughness spectral parameters, volume-scattering strength, and acoustics attenuation. The preliminary inversion results are basically consistent with the experimental data on sediment distribution and texture.