A surface wave inversion method incorporating interface constraints and its application in the assessment of ground vibrations caused by mining-induced seismicity

Objective In mining areas of western China, underlying interfaces with significant fluctuations affect seismic exploration accuracy while also posing challenges to the assessment of ground vibrations triggered by dynamic hazards such as mining-induced seismicity. To enhance the characterization of shallow velocity structures and to provide support for dynamic hazard prevention and control in mining areas, this study developed an inversion method incorporating prior information about geological interfaces.

Methods A dense array consisting of 489 single-component and 108 three-component seismometers was deployed within a mining area in Longkou Town, Junggar Banner, Inner Mongolia. Accordingly, continuous ambient noise data for about one month were collected. Subsequently, the thicknesses of the Quaternary sedimentary layers (Q₄ interfaces) were inverted using the horizontal-to-vertical spectral ratio (HVSR) method. Surface wave dispersion curves were extracted from both the cross-correlation functions of ambient noise and time-frequency analysis. Then, constrained inversion was conducted by integrating prior information about the Q₄ and coal seam interfaces. As a result, the 3D S-wave velocity structures under undulating interfaces were established. Finally, numerical simulations were performed using the spectral element method (SEM) and the SPECFEM3D (SEM3D) software to compare the seismic wavefield responses of the homogeneous, unconstrained, and proposed constrained models.

Results Compared to unconstrained inversion, the surface wave inversion method incorporating interface constraints exhibited a decrease of approximately 40% in travel time residual and faster convergence speeds. Furthermore, the proposed method effectively suppressed path distortions caused by interface undulations and significantly enhanced the imaging accuracy of shallow velocity structures. Numerical simulation results indicate that the sedimentary layers and intermontane basins strongly influence the distribution and attenuation of seismic wave energy, with the basin areas producing more prominent amplification effects on wavefields than the sedimentary layers.

Conclusions The proposed inversion method that incorporates interface constraints provides a technical approach for the fine-scale modeling of shallow velocity structures in mining areas with undulating interfaces. The findings of this study offer methodological and model support for the assessment, prevention, and control of ground vibrations triggered by dynamic hazards such as mining-induced seismicity.