Horizontal ocean current tomography with iterative model weighting constraint

In an inverse problem of ocean acoustic current mapping, prior information about the unknown current fields is usually required to better determine model parameters due to finite number of the acoustic rays. An iterative inversion method is applied to tomographic reconstruction of horizontal ocean current field to improve the estimate of current fields while prior information of the current distribution is absent. The method solves for the model parameters iteratively with an updated model weighting constraint from the previous iterate. The performance of the method is evaluated via numerical simulations using current fields generated from various sizes of checkerboard stream functions. Compared to a damped least square method, the iterative method considerably improves the current mapping. The reconstructed current field using the iterative method yields the fractional error less than 0.2 for most of the checkerboard stream functions. The iterative method reduces the influence of prior information on horizontal ocean current fields and increase the accuracy of the current mapping results in the coastal seas.