总结海水声速模型研究进展，将海水声速模型分为数学模型和数值模型。数学模型以经验公式、Munk模型、GDEM（Generalized Digital Environmental Model）为典型代表，数值模型以经验正交函数（Empirical Orthogonal Function，EOF）模型为主要类型，基于全球数据集得到广泛应用。数学模型具有良好的拓展性但过于理想；数值模型可达到较高的声速拟合精度，但也存在局限性，需要进一步发展。以南海声速模型为切入点，总结南海声速模型的研究需要和应用需求，认为南海海水声速模型将在高精度宏观三维声速模型、动态预报宏观声速模型的方向上不断深化应用领域，取得新的进展。
The research progress of seawater sound velocity models is summarized, the models are divided into mathematical models and numerical models. The mathematical models are represented by the empirical formula, Munk model and Generalized Digital Environmental Model (GDEM). The numerical models take Empirical Orthogonal Function (EOF) model as the main type and are widely used based on the global data set. The mathematical models are of good expansibility, but too ideal; while the numerical models can achieve high sound velocity fitting accuracy, but they also have limitations. Taken the sound velocity models in the South China Sea as the starting point, the research and application needs of the sound velocity models in the South China Sea are summarized, and it is believed that the sound velocity models in the South China Sea will continue to widen application and achieve new results in the direction of high-precision three-dimensional sound velocity models and dynamic prediction sound velocity models.