Recently, researchers led by Dr. Xu Min from the South China Sea Institute of Oceanology, Chinese Academy of Sciences, in collaboration with their counterparts, utilized background noise data recorded by active-source seafloor seismometer profiles near the "Challenger Abyss" in the southern Mariana Trench to conduct shear wave velocity structure imaging. The findings were published in the "Journal of Geophysical Research: Solid Earth."
Dr. Zhang Yayun, the lead author of the paper and an assistant researcher at the South China Sea Institute of Oceanology, Chinese Academy of Sciences, explained that in traditional seismological studies, data from active-source seafloor seismometers are primarily used to detect longitudinal wave structural information of the oceanic lithosphere, while background noise signals are often underutilized. Background noise imaging typically requires long-duration data from land-based seismic arrays to obtain high-quality empirical Green's functions through stacking multiple cross-correlated waveforms, and then to invert for shear wave velocity. However, the recording duration of active-source seafloor seismometer data is usually only 1-3 weeks, making it extremely challenging to extract reliable surface wave signals from short-duration noise waveforms.
It was learned that Dr. Xu Min's team, in collaboration with Dr. Lin Jian, a specially appointed researcher at the South China Sea Institute of Oceanology, Chinese Academy of Sciences, Dr. Qiu Xuelin, a researcher, and Professor Yang Hongfeng from the Chinese University of Hong Kong, utilized innovative methods. They employed short-duration cross-correlation techniques to obtain a sufficient number of cross-correlated waveforms and used a method of selecting stacked high signal-to-noise ratio cross-correlated waveforms to effectively eliminate the influence of local unknown seismic sources, meeting the requirements of the diffuse wave field. Additionally, they improved the quality of the final Green's functions using time-frequency domain phase-weighted stacking techniques.
Applying this data processing workflow to active-source seafloor seismometer profiles in the southern Mariana Trench, researchers obtained high-quality Green's functions and successfully extracted surface wave dispersion data. Through the inverted velocity model, they revealed the structural characteristics of the region, particularly the low-velocity zone at the top of the mantle on the subducting slab. Estimation results of water content indicated that the degree of serpentinization in this region is significantly higher than that in the central and northern parts of the Mariana Trench, a finding consistent with results obtained from dynamic simulations and seismic relocations.
"We innovatively utilized short-duration active-source data recordings for successful passive-source seismic imaging," said Dr. Xu Min, the corresponding author of the paper. This study not only provides the possibility of reusing a large amount of existing active-source seafloor seismometer data but also demonstrates that this method remains effective even with inter-station spacings exceeding 400 kilometers. Moreover, this method is equally applicable to passive-source seafloor seismometer data, capable of obtaining high signal-to-noise ratio cross-correlated waveforms, providing a solid data foundation for more refined imaging work, such as waveform fitting.
The stacked results for channels L25-L27 within the 2-5 Hz frequency band at the station are as follows:
(a) Linear stacking of all cross-correlations.
(b) Linear stacking after selecting cross-correlations with high correlation. See figure provided by the respondent.
Different stacking methods for improving data quality. LS: Linear stacking, Selected LS: Selected linear stacking, Selected tf-PWS: Selected time-frequency phase-weighted stacking. Respondent-provided image.
Related paper information: https://doi.org/10.1029/2023JB027043
