Estimation of geoacoustic parameters from narrowband data using a search-optimization technique

Geoacoustic parameters were estimated for vertical array data from the matc hed-field inversion benchmark data sets. Separate inversions were performed for narrowband data at 25 Hz, 50 Hz and 75 Hz, using a matching function c onsisting of the incoherent sum of the Bartlett outputs for the five vertic al arrays at-ranges of 1, 2, 3, 4 and 5 km. Parameter estimation was perfor med using a parabolic equation sound propagation algorithm to generate the replica fields, and a search-optimization technique to obtain estimates of the optimized parameter values. This technique involved an initial search s tage in which the parameter space was sampled, and a second optimization st age in which each of a specified number of the best matches found in the se arch stage was used as the starting point for optimization. This approach p rovided multiple independent estimates of the geoacoustic parameters, and a llowed assessment of the non-uniqueness of the problem and the sensitivity of the matching function to the individual parameters. A method was develop ed to combine the results for several frequencies to estimate the parameter s. It used a weighted average with weights computed on the basis of the rel ative sensitivities at those frequencies; these sensitivities were estimate d by the root-mean-square (RMS) gradient observed during the optimizations. Strong interdependencies among the parameters were found in the analysis, particularly between the sediment thickness and the sound speed at the bott om of the sediment. For the single-frequency matching function used here, i t was observed that the inversion problems were ill-posed in that sets of p arameter values from a wide region of the parameter space gave essentially perfect matches. The consistency of the parameter estimates was greatly imp roved by including a regularization term in the matching function. Regulari zed search-optimization provided an efficient method for estimating an effe ctive geoacoustic model for acoustic field prediction.