Matched-field inversion using the down hill simplex algorithm

Abstract

In recent years inversion techniques have been proposed to effectively and efficiently determine geophysical parameters. The focus of this matched-field inversion (MFI) study is to determine the geoacoustic and geometric parameters for a North-East Pacific shallow water data set. MFI correlates modeled data, called replicas, with the measured data and uses a search algorithm to find model values which maximize the correlation function. In our MFI the parabolic equation (PE) is used to compute the replicas to account for mode coupling in the range-dependent environment. The first stage of the optimization algorithm employs a random search to determine N+1 parameter sets with the best correlations, where N is the number of parameters being determined. Next the N+1 sets are the input to the downhill simplex algorithm. The algorithm is shown to perform well for simulated vertical line array data for search parameter ranges representative of the environment in the shallow North-East Pacific Ocean. The inversion technique was then applied to measured data obtained in the North-East Pacific on the continental shelf using towed CW and broadband sources and a vertical array of hydrophones. The geometric parameters of the experiment and dominant geoacoustic parameters were successfully recovered.