For the reverberation experiments with a near surface source or receiver, surface reverberation arriving first after source signal is not influenced by bottom reverberation due to time-delay difference of acoustic multipath in deep water. A surface reverberation model is proposed in the paper. The Green' s function of sound propagation is described by the ray theory, and first-order small-slope approximation is employed for the surface scattering from full angle. The effect of bubbles scattering is also considered to get the surface reverberation theory. The reverberation model is verified by comparing the simulation results with the experimental data. The measured data show that the decaying rate of surface reverberation intensity decreases with the frequency increasing. Numerical calculation demonstrates that the frequency dependence is caused by the positive correlation between scattering strength and frequency, and the surface reverberation at low frequency for low sea state is dominated by the scattering from rough air-sea interface. Moreover, the reverberation data from experiment show that the surface reverberation is not sensitive to the change of receiving depth. A method of inversion for the two surface-wave spectral parameters in the model is achieved based on the reverberation model. The inversion results verify that the spectral parameters of rough surface can be obtained from surface reverberation data on the premise of the wind speed parameter. As a result, the scattering properties of rough interface will be obtained.

Acoustical properties of the sea bottom can be described with a geoacoustic (GA) model comprising several layers characterized by the values of GA parameters, such as sound speed, density, sound absorption, and thickness. As the parameter values are difficult to measure directly, they are often estimated using inversion methods. Although GA inversion is the most economical method to infer the seabed acoustical parameter values, the results are far from unique owing to the ambiguities between GA parameter values thus obtained. Numerical simulations show that the model-free parameters (P, Q) can describe the reflection characteristics of the bottom at small grazing angles. Since the bottom reflection properties are important relevant information required for propagation modeling, directly inverting the P and Q values instead of inverting the GA parameter values will not depend on details of the assumed bottom model and provides significant benefits. In this study, data from an acoustic propagation experiment in the South China Sea are inverted for the seabed GA parameter values. Based on the inverted results, it was concluded that different sets of “effective GA parameters” can be constructed with values that produce the same reflective characteristics, i.e., the same (P, Q) values.

The frequency shift of multipath sound rays induced by the motion of a sound source in an ocean waveguide environment is a crucial factor affecting the detection capabilities of both active and passive sonar systems, as well as the quality of underwater communication. Therefore, investigating the sound field characteristics of a moving sound source in the SOFAR channel is of significant importance. By comparing the spectra of continuous-wave (CW) signals with pulse widths of 1 s and 15 s received by a vertical array in SOFAR channel, it was observed that the sound field of the moving source exhibits a stable spectral splitting characteristic. Two frequency shift bright lines in the vertical direction were identified, corresponding to two sets of sound ray paths. One set of sound ray paths corresponds to the direct sound and the first surface-reflected sound, and the other set of sound ray paths corresponds to the first seabed-reflected sound and the first surface- and seabed-reflected sound. This study revealed that the spectral splitting of the moving sound source’s sound field displays a distribution trend in a depth direction similar to that of the multipath delay structure. A multipath sound ray frequency shift calculation model, based on ray theory, was developed to explain and predict the vertical distribution pattern of spectral splitting in the sound field of a moving sound source. By combining the model with measured data, it was found that the spectral splitting arises from the frequency shift differences corresponding to multipath sound ray paths. Additionally, the frequency shifts for the D&S and B&SB ray paths are generally proportional to the cosine values of the initial grazing angles of the sound waves at the emission source and the cosine values of the horizontal azimuthal angle between the source motion direction and the receiver.

在浅海, 尤其是负梯度声速剖面和海面较为平静的浅海波导, 海底界面反向散射是浅海混响的主要来源. 经验散射模型只适用于分析浅海混响平均强度衰减特性, 而基于物理机理建立的反向散射模型克服了这一缺陷, 但同时也引入了其受地声模型约束的问题. 本文结合了海底反射系数的三参数模型, 对浅海远场海底反向散射模型进行了简化, 以减少地声模型的输入参数. 理论分析了海底反射系数的相移参数可以描述海底对声场的散射作用, 无需任何海底地声参数的先验知识. 通过对海底反向散射模型近似简化, 结果表明在临界角附近和甚小掠射角范围内的海底粗糙界面反向散射模型的角度特性和强度特性受海底沉积层的影响不同: 在临界角附近, 海底反向散射的角度特性受海底反射系数的相移参数加权,而其散射系数则近似与相移参数无关; 对于甚小掠射角, 海底反向散射的角度特性近似与海底反射系数的相移参数无关, 其散射系数则近似与相移参数的4次方成正比.