A localization method is developed using time differences of eigenray arrivals in a two-dimensional range-dependent underwater multiple reflection environment. This fast and robust method applies a perturbation theory to a horizontal range and a travel time which are the ray solutions given as a functional of both horizontal slowness and the sound speed profile. The linearized equations for estimating the horizontal range and depth of an unknown sound source are derived in a Cartesian coordinate system. The method’s only constraint is the applicability of the ray theory. That implies that a low-frequency sound in a shallow water environment is no more applicable than a high-frequency sound to the method because it violates the ray approximation. Identification of three eigenrays provides the passive range and depth estimation of underwater sources without a priori information on sub-bottom properties. Experiments with an impulsive source were conducted at coastal areas in Suruga Bay, Japan, to validate the patented method. The bathymetry gradually deepens toward the location of the receiver around1 km from the source. Four arrivals were detected by the use of the Wigner distribution and identified as the corresponding eigenrays. The estimated range and depth of the source are in an excellent agreement with the measurements.

Many kinds of hydroacoustic devices are used in the field of ocean development of resources and observations. The effects of sound reflection from the sea surface are often evaluated in estimating the performance of these devices and in analyzing acoustic data obtained in actual seas. When sea surface waves are sufficiently calm, the assumption of specular reflection of sound waves is valid for the estimation of acoustic propagation characteristics. On the other hand, actual surface waves change randomly over time because of wind and gravity, and consequently reflected sound waves also fluctuate randomly. In this study, we evaluated the variability characteristics of reflected sound waves from the sea surface by acoustic simulation using the finite-difference-time-domain method. Our results clarify the effect of the wave height and wavelength at the sea surface on the variability characteristics of reflected sound waves.