Acoustic sensing of temperature changes in a strongly range-dependent ocean

Abstract

The inverse problem of inferring information about the intervening ocean can be solved by a comparison of the computed arrival times and measured patterns of arrivals. This approach is very effective if in the process of sound signal propagation the modal energy exchange is negligible, the mode propagation can be considered as adiabatic and each mode has a clear amplitude and phase. In the range-dependent ocean, drastic `repopulation' of the modes takes place due to intense energy exchange. The energy distribution between modes changes, and the order of modal arrivals varies. For the strong-mode-coupling case a fixed mode travel time does not have a monotonic relationship with the temperature variation. To obtain monotonic dependence of the sound signal parameters upon the environmental characteristics, specific approaches to the forward problem solution have to be used; two of them - the collective arrival time and the cumulative sum - are considered in this paper. They consists of defining trends in the signal arrival pattern caused by the temperature changes, and computing average parameters of the arriving signal ensemble.