Computer simulation of tidal signal prediction in long range ocean acoustic travel-time measurements

When using long-range sound travel-time measurements to monitor the ocean temperature changes due to the global warming,tidal effects must be corrected from the data.This paper presents the computer simulation of tidal signal prediction for the ATOC project.On the basis of a linear model of the tidal signal,and using measurements taken within a specified sampling Period and at 4 hour intervals,a pseudo-inverse method is used to predict the travel-time changes attributed to the aggregate effect of barotropic tides along the sound path.In order to obtain an adequate prediction accuracy,a sampling Period of no less than 18 months is recommended,so that all the major tidal constituentS can be estimated,and some closely spaced frequency components are separable.The linear model should include as many constituents as possible to minimize the prediction error.This is feasible because in modeling the tidal signal,the only parameter needed for each constituent is the frequency,and the frequencies of the astronomical components are known to a high precision,while the non-astronomical components are trivial in this application.Quantization is one of the major sources of prediction errors.Quantization errors can be effectively reduced by means of a multipath averaging procedure.