Seismic signal characteristics and interpretation of the 2020 "6.17" Danba landslide dam failure hazard chain process

Landslide dam failures have devastating capability to cause significant hazards to human lives and infrastructure along their flooding paths. Recent studies have explored such hazards based on environmental seismology to analyze the process of evolution using the time-frequency characteristics of seismic signals. However, most research has focused on reconstructing individual processes (e.g., debris flows, landslides, and floods). Previous studies on a comprehensive analysis of the entire hazard chain process have been limited. Challenges lie in the integration of seismic signals with different energy levels during different hazard stages. To address this issue, post-hazard surveys, remote sensing, and seismic signals were obtained in this study to comprehensively analyze the evolution process of the Danba landslide dam failure hazard chain of June 17, 2020, in China. We developed weak-signal processing using a band-pass filter, empirical mode decomposition, fast Fourier transform, and short-time Fourier transform to process and analyze the seismic data and accurately extract the signals for the debris flow, landslide, flood, and noise attenuation in the hazard chain. The evolution process of each stage in the hazard chain was then interpreted. The debris flow, landslide, barrier lake bursting, and flood routing were analyzed to determine their evolution modes. In addition, the reactivation of ancient landslides downstream caused by flood erosion was also identified, with stage explanations. This method provides new ideas for interpreting the hazard chain processes and evolution modes of landslide dam failure hazard chains and theoretical guidance for future hazard early warning and mitigation.