Listening to the echoes of the seabed
Marine seabed exploration, based on air and satellite platforms, is often dependent on water conditions conducive to light penetration. New techniques using remote sensing technologies based on the reflection of acoustical pulses are now being investigated in order to determine the sediment properties of the seabed. This approach relates the sound characteristics, for example sound reflection and absorption, to the physical properties of the seabed, such as bulk density, grain size and porosity. The characteristics of an acoustic response, meaning the amplitude and shape of an acoustic signal reflected from the sea floor, determine the seabed class or type. The acoustic seabed classification relies on the geological and biological diversity of the seabed expressed in acoustic diversity of echoes. The extraction of characteristic features from echoes requires extensive data processing. Theoretical and numerical models and procedures have been developed in order to group together echoes with similar features and identify the interrelationships of acoustic parameters with the physical properties of the sediment. The relations between classes based on acoustic responses and the biological or geological differences of interest are validated by the acquisition of appropriate samples using divers or grab samplers. A group of researchers observed the scattered and reflected acoustic signals from the seabed and from sub-bottom layers after the transmission of high power acoustic signals. They also developed theoretical and numerical accurate acoustic wave propagation models putting in use parameter estimators, such as Maximum Likelihood Estimator. The models and procedures were validated through tank experiments (laboratory scale), using conventional equipment and geophysical and geotechnical instruments. During the validation process errors were determined in the models. The smallest model errors originated from an intrinsic attenuation in the frequency of fine sediments. Larger errors were found in models where many parameters are still unknown. Further development of these techniques will assist companies involved in ocean bottom exploration by providing more accurate seabed classification.
