摘要: The vast and complex oceans that are optically opaque acoustically transparent, enabling characterization of physical biological bodies processes sea using sound as a premier tool. Lack direct observations vertical profiles velocimeters and/or temperature salinity, from which speed can be calculated, limits specifications investigation temporal spatial variabilities the three-dimensional structure in oceans. In this study, authors demonstrate estimation (SSPs) surface an artificial neural network (ANN) method. Surface mooring central Arabian Sea used proxy to satellite observations. ANN-estimated SSPs had root-mean-square error 1.16 m/s coefficient determination 0.98. About 76% (93%) estimates lie within plusmn1 (plusmn2 m/s) obtained situ salinity