Multiple Voronoi Partition Improves Multimodal Dispersion Imaging From Ambient Noise: A Case Study of LASSO Dense Array

Recent studies on the frequency-Bessel transform (F-J) method demonstrate the ability of the array-based method to extract higher-mode surface wave dispersion curves from ambient noise, which provides a new opportunity to reveal more accurate underground structures. However, problems with the subarray selection for three-dimensional imaging remain. On the one hand, we need the subarray to be large enough such that the F-J method can capture high-quality dispersion curves, and on the other hand, we want the subarray to be small enough to maintain a sufficient horizontal resolution. To solve this problem, we propose a strategy that randomly and repeatedly partitions the subarray based on Voronoi diagrams. We call this the FJ-VoroTomo method. The FJ-VoroTomo method does not require tedious parameter tuning and can measure high-quality dispersion while maintaining horizontal resolution. More importantly, this method can quantitatively analyze the uncertainty of the measured dispersion curves. In this work, we use the noise cross-correlation functions from the LArge-n Seismic Survey in Oklahoma array as an example to demonstrate the dispersion and its uncertainty obtained by the FJ-VoroTomo method and evaluate the three-dimensional S wave velocity structure beneath this dense array. We hope that by using this strategy, the multimodal surface wave method can be more feasible in dense arrays that are currently widely used.