Seabed characterization with a small towed array - simulation of system design facors

This report studies design factors for a small activated towed-array system for use in seabed characterization. By seabed characterization is here meant the estimation of geo-acoustic parameters including sound speed, density, and attenuation profiles, and layer thicknesses of the upper few meters of the seabed. The effect of system design factors such as array length (16–48 m), number of hydrophones (3–33), system height above the seabed (10–20 m) and acoustic source frequency content (0.5–4 kHz) are studied by use of simulated data for a modeled activated towed-array system. The signal processing technique applied is matched-field inversion of hydrophone data. A Bayesian inverse method based on Markov-chain Monte Carlo sampling for posterior probability densities is used, which allows for quantified comparisons of effects of individual system factors on the geo-acoustic information content of data. Based on the simulations in this report for a typical Continental Shelf type seabed, we find that a critical design factor is the distance between the acoustic source and the array, which preferably should be on the order of 30 to 60 m (measured to the closest array element). However, shorter distance does not necessarily preclude meaningful information content. It is shown that the information content decreases with reduced array length and with reduced number of array elements, with a length of 32 m and 33 elements an apparent practical lower limit for meaningful information content of data. The results from and methods applied in this report can be further used in design of small systems for seabed characterization, e.g., an activated towed array system for an autonomous underwater vehicle.