Ambient noise carries abundant subsurface structure information and attracts ever-increasing attention in the past decades. However, there are lots of interference factors in the ambient noise in the real world, making the noise difficult to be utilized in seismic interferometry. The paper performs shear-wave tomography on a very short recording of ocean ambient noise with interference. An adapted eigenvalue-based filter is adopted as a pre-processing method to deal with the strong, directional interference problem. Beamforming and the noise crosscorrelation analyses show that the filter works well on the noise recorded by the array. Directional energy is significantly suppressed and the background diffuse component of the noise is relatively enhanced. The shear-wave tomography shows a 4-layer subsurface structure of the area covered by the array, with relatively homogeneous distribution of the shear-wave velocity values in the top three layers and a complicated structure in the bottom layer. Moreover, 3 high-velocity zones can be recognized in the bottom layer. The result is compared with several other tomography results using different methods and data. It demonstrates that, although the ambient noise used in this paper is very short and severely contaminated, a reasonable tomography result can be obtained by applying the adapted eigenvalue-based filter. Since it is the first application of the adapted eigenvalue-based filter in seismic tomography using ambient noise, the paper proves the effectiveness of this technique and shows the potential of the technique in ambient noise processing and passive seismic interferometry.
Ambient noise field and propagation in an Arctic fjord Kongsfjorden, Svalbard
