声速剖面EOF重构的实测数据采样深度研究

经验正交函数（experiential orthogonal functions，EOF）是重构声速剖面（sound speed profile，SSP）的一种有效方法，利用部分实测数据结合历史剖面资料可以重构当前位置的声速剖面。针对实测数据的采样深度难以确定这一问题，本文介绍了一种基于历史声速剖面资料的实测数据采样深度选取方法，根据EOF空间函数的方差贡献率确定数据量，进而采用EOF算法重构全海深声速剖面。实验结果表明：采用该方法得到的数据重构的声速剖面与实测声速剖面具有较好的一致性，基于常梯度声线跟踪法得到的水深数据能够满足0.25%倍水深限差，有效波束比达到了100%，为实际测量作业中声剖数据的采样深度提供了参考。

Research on the sampling depth in the EOF reconstruction of the sound speed profile

Reconstructing the whole sea depth sound speed profile (SSP) at the current location can be done by performing empirical orthogonal functions (EOFs) by means of measured data and historical SSP data. Determination of the sampling depth of data is difficult. With this in mind, we introduce a way to choose the sampling depth of data according to the historical SSP data, determine the amount of data based on the variance contribution rate of the EOF spatial function, and then use the EOF algorithm in reconstructing the whole sea depth SSP. We find that the reconstructed SSP by the EOF method agrees well with the measured SSP. We also find that the depth data obtained by constant gradient sound ray tracing can meet the 0.25% depth limit set out by the National Oceanic and Atmospheric Administration and that the effective beam ratio has reached 100%, which provides a reference for the acquisition depth of the SSP data in the actual measurement operations and improves the operation efficiency.