Inversion study of the coastal ambient noise measurement near Qingdao

摘要

During the past decades, inversion of ocean environment parameters by ocean ambient noise has gradually made great progress. It has been suggested that the time-domain Green's function between two receivers can possibly be extracted by using ocean ambient noise. The correct location of each element of an hydrophone array can be calibrated according to information about Green's function. With the beam-forming technique, a vertical array can be used as a passive fathometer to measure layered sea bottom information. For conventional ambient noise data processing, the down-going and up-going noise signals are applied in the process of cross-correlating, and then the effect of the horizontal-propagating noise can be reduced by a beamforming method. But in some cases, especially in shallow water, interference signals such as the noise from shipping traffic is too great to reduce. So, it is necessary to perform some additional modified data processing in order to achieve better results. In this paper, ambient noise measurements near Qingdao are analyzed, and the minimum variance distortionless response (MVDR) adaptive beamforming technique is applied to extract more clearly layered sea bottom information, and the layered sea bottom information at depths of 29m, 31m and 34m become more distinct. Empirical orthogonal function (EOF) analysis is applied to separate different characteristic noise signals, then by combining some higher order EOF components, the vertically propagating noise signal can be enhanced and those horizontally propagating components are removed well. Results show that combined MVDR and EOF processing can effectively optimize the sea bottom inversion of coastal ambient noise measurement. The signal-to-noise ratio (SNR) can be effectively improved, by about 2dB on average, and the maximum SNR improvement is about 4-5dB.