Study On The Infrasound Characteristics Of Typical Rock Landslides In The Three Gorges Reservoir Area Based On Physical Model Test

China boasts a vast and intricate landscape,with mountains occupying approximately two-thirds of the country’s land area.Among various mountain-related disasters,landslides stand out for their suddenness and devastating impact.The Three Gorges reservoir area,situated at the confluence of the second and third stages of China’s landscape,presents treacherous mountains,deep canyons,and frequent rainstorms.Following the construction of the Three Gorges Dam,the cyclic water level fluctuations of the Yangtze River have caused an annual subsidence of up to 30 meters,leading to the deterioration of the reservoir bank’s geotechnical structure.Consequently,numerous ancient landslides have resurfaced,posing a significant threat to the safety of mountain residents and waterways,while also impeding the development of the Yangtze River economic belt.Therefore,conducting research on landslide monitoring and early warning systems holds immense importance in safeguarding the lives and property of individuals.The essence of landslide instability damage lies in the fact that the sliding force exceeds the resistance to sliding.Effectively monitoring and warning about landslides can be achieved through the detection of abnormal information such as force and displacement.However,these methods suffer from limitations in terms of monitoring range and cost,posing significant challenges in the monitoring and early warning of hidden and hazardous rocky landslides.Previous research indicates that the process of rock sliding generates distinct infrasound signals,making infrasound a potential technical means for rock slide monitoring.Exploring the relationship between landslide infrasound and mechanical parameters could simplify the current landslide monitoring and early warning system,thereby providing more time for affected individuals to evacuate.Nevertheless,the characteristics and mechanical mechanisms of infrasound generated by rocky landslides are currently unclear,which hinders the application and widespread adoption of this technology.In this study,we focused on investigating a representative rocky landslide,known as the lotus root pond landslide,in the Three Gorges reservoir area.Initially,comprehensive data regarding the landslide were collected,including conducting field geological surveys and obtaining rock samples.Subsequently,we analyzed the background and formation process of the lotus root pond landslide.To further our research,we developed and established an indoor physical model test system for rocky landslides,enabling us to simultaneously monitor infrasound and mechanical data throughout the entire landslide process.We employed various signal processing methods,such as automatic extraction of infrasound events,joint time-frequency domain analysis,wavelet transform,and HilbertYellow transform,to analyze the characteristics of infrasound signals at different stages of the landslide test process.Moreover,from a mechanical perspective,we investigated the response patterns of infrasound and mechanical characteristic parameters of the rock column in the locking section under different conditions of size,lithology,and saturation.Furthermore,we explored the response patterns of infrasound characteristic parameters of rocky landslides.The primary contributions and innovations of this study can be summarized as follows:(1)The physical geography and socio-economic profile of the Three Gorges reservoir area were introduced through field geological survey.The background of the root pond landslide was analyzed in terms of topography,stratigraphic lithology,geological structure,bank structure,seismic activity,rainfall,reservoir water rise and fall,and artificial activity.The formation process of the landslide was summarized,laying the foundation for subsequent indoor rocky landslide physical model testing.(2)The indoor rocky landslide physical model testing system was independently designed and built,combining data from the preliminary investigation.It was based on the lithological characteristics,structural characteristics,and engineering geomechanical properties of the landslide,along with the simplification of the landslide structure and material zoning.Monitoring instruments including high-sensitivity infrasound sensor arrays,mechanical sensors,and high-resolution video cameras were deployed to achieve simultaneous monitoring of the entire landslide process.The infrasound mechanics data was combined to divide the landslide test process into four stages: uniform deformation,accelerated deformation,slow sliding,and rapid sliding.Detailed explanations of the acoustical phenomena and characteristics were provided for each stage.It was proven that all stages of the landslide generate infrasound signals,but their dominant frequency bands and energy distribution differ.The infrasound events are most prominent during the accelerated deformation and rapid sliding stages of the landslide.(3)Signal processing methods such as time domain feature analysis,frequency domain feature analysis,wavelet transform,and Hilbert-Yellow transform were utilized to analyze and summarize the background noise and infrasound characteristics of each stage of the landslide.In terms of frequency and energy distribution,it was observed that while some frequencies of infrasound overlap in each stage of the landslide,they can be differentiated based on infrasound pressure,dominant frequency,and energy distribution.In terms of waveform characteristics,the infrasound waveforms during the landslide deformation stage exhibit sharper characteristics compared to those in the sliding stage.The infrasound waveforms in the uniform deformation and slow sliding stages are tilted to the right,whereas those in the accelerated deformation and fast sliding stages are tilted to the left.The infrasound signals in the accelerated deformation phase are the most skewed.The entire landslide process involves multi-band coupling,and the dominant frequency band of the infrasound signal shifts from 1.56 Hz to 12.25 Hz along with the deformation and sliding of the landslide,indicating that the frequency of the infrasound signal evolves with different processes.(4)The response law of rocky landslide infrasound characteristic parameters was investigated.From a mechanical perspective,the laws governing the mechanical parameters of the test group under different conditions were summarized,and the corresponding time-frequency domain and energy characteristics of the infrasound signal were analyzed.The size,lithology,and saturation of the locking section were found to influence the dominant frequency band and energy distribution of the landslide infrasound signal.The study’s findings provide a reference for rocky landslide infrasound monitoring and early warning systems.