ESTIMATION OF OCEAN-BOTTOM PROPERTIES BY MATCHED-FIELD INVERSION OF ACOUSTIC FIELD DATA

Properties of the ocean bottom, including bathymetry and geoacoustic p arameters such as sound speed, density and attenuation, are required f or reliable acoustic propagation modeling and matched-field source loc alization. Bottom properties can be difficult to measure directly; how ever, inversion procedures hold much promise for estimating their valu es from acoustic field measurements. The inversion can be formulated a s an optimization problem by assuming a discrete model of unknown para meters and a bounded search space for each parameter. The optimization then involves finding the set of parameter values which minimizes the mismatch between the measured acoustic field and modeled replica fiel ds. Since the number of possible models can be extremely large, we hav e employed the method of simulated annealing which provides an efficie nt optimization that avoids becoming trapped in suboptimal solutions. The matching fields are computed using a normal mode model. In inversi ons for range-dependent parameters, the adiabatic approximation is emp loyed: this allows mode values to be pre-computed for a grid of parame ter values and stored in ''look-up'' tables for fast reference, which greatly improves computational efficiency. Synthetic inversion example s are presented for realistic range-independent and range-dependent en vironments.